Gyn&obs&pat&zhe&fad&qui&she&guo&vol 2&1 ST
Gyn&obs&pat&zhe&fad&qui&she&guo&vol 2&1 ST
Gyn&obs&pat&zhe&fad&qui&she&guo&vol 2&1 ST
Oluwole Fadare
Charles Matthew Quick
Danhua Shen
Donghui Guo
Editors
123
Gynecologic and Obstetric Pathology, Volume 2
Wenxin Zheng
Oluwole Fadare
Charles Matthew Quick
Danhua Shen
Donghui Guo
Editors
Donghui Guo
Department of Obstetrics and Gynecology
Tianjin Gynecologic and Obstetrics Central
Hospital
Tianjin
China
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore
Foreword
The field of gynecologic and obstetric pathology is at a cross-roads. In the past decade we have
begun to witness the departure of venerated contributors to this discipline and now we are
being inundated with new information about pathogenesis that informs diagnostic expectations
and patient outcome. The old model in which the next generation sits at the knee of the learned
and patiently awaits their turn at the helm is rapidly fading. Succession is now not simply
achieved by learning the old language but by speaking a new one.
The textbook Gynecologic and Obstetrics Pathology, edited by Drs. Zheng, Fadare, and
Quick is emblematic of the sea change. We’ve all heard the joke about resorting to one’s grand-
child to solve a computer conundrum. How many of us turn to our younger colleagues to
interpret emerging genomic information in the management of gynecologic cancer? An appre-
ciation of such talent is crucial to our evolution as well as that of our discipline.
The senior editor in this project, Dr. Wenxin Zheng, has a long track record of innovation.
With this has come a facility to recognize the most talented young clinician-investigators and
to recruit them into this new textbook. Drs. Fadare and Quick as well as the younger chapter
authors are well on their way, having already put us on notice that by their dedication, creativ-
ity and their role in discovery. Their input is what will keep this and subsequent editions at the
forefront of pathology texts dedicated to women’s health.
Energy and intellect drive discovery but experience is essential to provide a needed perspec-
tive when this information is transmitted to the practicing pathologist. The editors wisely bal-
ance the list of talented newcomers with recognized experts in the field. Together they provide
the finer details of diagnosis and differential diagnosis while eliciting the nuances relevant to
clinical management.
In the current world, where discoveries and their impact on practice can become global
almost instantaneously, one does not need to travel far to realize that expertise in obstetric and
gynecologic pathology is intercontinental. In recognizing this, Zheng et al. will also provide an
edition written in Chinese, bringing this message to pathologists (and their patients) in coun-
tries where the language is read and spoken. To my knowledge, this book will be the first of its
kind to accomplish this, creating a truly international presence that will place this first edition
among the leading texts in the field. The editors and authors are to be commended for their
contribution and I look forward to their success in opening a new chapter (and book!) in the
history of the pathology of the female reproductive tract.
Christopher P. Crum
Division of Women’s and Perinatal Pathology
Harvard Medical School, Brigham and Women’s Hospital
Boston, MA, USA
v
Preface
Pathology of the female genital tract is complex, and encompasses a wide spectrum of neoplas-
tic and non-neoplastic diseases of the gonads, reproductive ducts, secondary müllerian system,
and external genitalia. Clinical practitioners in this discipline must therefore familiarize them-
selves with a broad spectrum of pathology, including skin-like diseases of the vulva, a myriad
of peritoneal diseases, as well as conventional diseases of other female genital tract organs.
This field progresses in a vibrant and dynamic academic environment in which diagnostic
concepts continually evolve as our understanding of various disease processes improves over
time. The contemporary gynecologic pathologist is in a unique position to recognize and define
morphologic correlates to newly defined genomic profiles and individual gene mutations,
assess whether they are likely to have diagnostic or prognostic significance for a given patient
and/or her family, and broadly participate in the push towards increasingly personalized cancer
care. These exciting trends notwithstanding, it remains true that definitive pathologic classifi-
cation of gynecologic disease is still primarily based on the traditional pillars of surgical
pathology, including gross pathology, morphologic assessment buttressed by immunopheno-
typic analysis where needed, and careful clinicopathologic correlation.
This book is envisioned as a “bridge” that acknowledges both of the aforementioned reali-
ties. It is designed to provide a broad coverage of diagnostic gynecologic and obstetric pathol-
ogy, inclusive of both neoplastic and non-neoplastic diseases. The book is neither a dense and
comprehensive treatise on every disease process nor is it a dry listing of relevant “facts” about
each entity. Rather, it is best conceptualized as a large scale aggregation of the most up to date
information in gynecologic pathology, all presented in a concise and narrative manner that is
designed to be easily accessible to the general practitioner, specialist and student alike. An
overt effort has been made to discuss each topic in a way that is maximally relevant to the
diagnostic surgical pathologist, such that by reading any section should substantially increase
the reader’s confidence that the most germane clinicopathologic information on that entity has
been reviewed before a diagnostic decision is made.
The material is organized into 36 chapters, representing the full spectrum of diagnostic
gynecologic pathology. In addition to chapters on traditional topics in gynecologic pathology,
there are individual chapters on site-specific carcinogenesis, gynecologic cytology, intra-
operative consultation, endometriosis and development/maldevelopment of the female repro-
ductive system, among others. Additionally, in a departure from most current texts, there are
stand-alone chapters to provide intensive coverage of some traditionally under-covered topics,
including melanocytic lesions of the female genital tract, non-neoplastic diseases of the endo-
metrium, and vulvovaginal soft tissue lesions. Entities with a significant diagnostic component
are presented, where feasible, divided into the following subsections: Clinical features, Gross
findings, Microscopic findings, Differential diagnosis, Biomarkers, and Genetic features. As
expected, not all entities or chapters lend themselves to this specific format, but most chapters
are broadly structured based on these general themes. Microscopic findings are lavishly illus-
trated, and numerous tables help summarize pertinent points for easy reference. The overall
objective of each chapter is to integrate traditional pathologic features, clinical features, where
applicable, and current paradigms in disease classification into a format that can be readily
applied in routine practice. These chapters are authored by over 50 physicians, most of whom
vii
viii Preface
are experienced subspecialty practitioners of clinical gynecologic pathology from around the
world, and without whose expertise, dedication, and diligence this work would not have been
possible. It is the sincere hope of the editors that all of those who are interested in gynecologic
pathology—diagnostic pathologists, students, residents and investigators—will find this book
tremendously useful.
The understanding of gynecologic disease involves pathologists and researchers from
across oceans and all over the world, and to that end an exciting feature of this book is that it
is written with a direct linkage to the second edition of the book “Gynecologic and Obstetric
Pathology”. The latter book is in Chinese, and is published by Science Press, Beijing, China.
The current text is published in both English and Chinese, representing a collaborative effort
by both publishers: Springer and Science Press. Although the titles and the number of chapters
are identical in the two books, the authors of the chapters are different. Additionally, while the
chapter outlines and some of the contents overlap, the two books do not represent a direct
translation from one to the other. Rather, they are best considered complementary “sister”
books. This results from Dr. Wenxin Zheng serving as the first editor-in-chief for both books.
Considering this special and close relationship, the co-editors for the Chinese book, Drs.
Danhua Shen and Donghui Guo, are listed as co-editors of the English version of this book,
while Drs. Fadare and Quick are also listed as co-editors on the second edition of the Chinese
book.
To my dear wife Wenda and my beloved family (Yuxin, Genfu, and Deshun) for their constant
love and endless support! In memory of my parents Maoguan Zheng and Jinxian Wang as well
as my ultimate mentor Dr. Stuart C. Lauchlan.
Wenxin Zheng
To my wife Abby for her love, encouragement, and inspiration, and to our beloved children
Nathaniel, Darrell and Olivia for (mostly) putting up with the occasional absences that were
required to do this.
Oluwole Fadare
To my beloved family members for their continuous support and care, and to all my colleagues
who participated in editing this outstanding book. I am honored to be an editor for this presti-
gious book.
Danhua Shen
To my beloved family members for their constant support and care, and to Drs. Zhaoai Kong
and Song Lin whose past instruction and training have been and continues to be invaluable.
Donghui Guo
The Editors would like to sincerely thank Dr. Christopher P. Crum, M.D. for serving as a senior
consulting adviser for this book. Dr. Crum has made significant contributions to the field of
gynecologic pathology and has trained innumerable residents and fellows, including many
who have served as an author for this book. The quality of this work is in part attributable to
his years of dedicated teaching and research in the field of gynecologic pathology.
ix
Contents
xi
xii Contents
xiii
xiv About the Editors
xvii
xviii Contributors
Jefferson Terry Children’s and Women’s Health Centre of British Columbia, Vancouver, BC,
Canada
University of British Columbia, Vancouver, BC, Canada
Yan Wang Department of Pathology, HKU-Shenzhen Hospital, Shenzhen, China
Jing Zhang Department of Pathology, Xijing Hospital, The Fourth Military Medical
University, Shaanxi, People’s Republic of China
Wenxin Zheng Departments of Pathology, Obstetrics and Gynecology, University of Texas
Southwestern Medical Center, Dallas, TX, USA
Uterine Mesenchymal Lesions
1
Brooke E. Howitt and Marisa R. Nucci
This chapter will cover the pathology of uterine mesenchy- 1.2.1 Endometrial Stromal Nodule
mal tumors, including endometrial stromal neoplasms,
undifferentiated uterine sarcomas, uterine tumors resem- 1.2.1.1 Definition
bling ovarian sex cord-stromal tumors (UTROSCT), smooth Endometrial stromal nodule (ESN) is defined as an endome-
muscle tumors, perivascular epithelioid cell tumor trial stromal neoplasm with no or minimal myometrial inva-
(PEComa), Mullerian adenosarcoma, and inflammatory sion and no vascular invasion.
myofibroblastic tumor, as well as some less common
1.2.1.2 Clinical Features
Endometrial stromal nodules (ESN) are benign neoplasms
B. E. Howitt that occur across a very wide age range but are most fre-
Department of Pathology, Stanford University Medical Center, quently encountered in women in their fifth to sixth decades
Stanford, CA, USA
[3–6].
e-mail: [email protected]
M. R. Nucci (*)
1.2.1.3 Gross Findings
Division of Women’s and Perinatal Pathology, Department of
Pathology, Brigham and Women’s Hospital, Boston, MA, USA ESN may be located at the submucosal layer of the uterine
e-mail: [email protected] wall, project into the endometrial cavity as an exophytic
1.2.1.4 Microscopic Findings Fig. 1.3 Endometrial stromal nodule. Microscopically, endometrial
Histologically, ESN are well-circumscribed but unencapsu- stromal nodules typically are sharply demarcated from the surrounding
myometrium without infiltrative borders
lated tumors resembling the nonneoplastic stroma of prolif-
erative endometrium, composed of cells with uniform round
to ovoid/fusiform nuclei that have scant to moderate amounts ESN no vascular invasion is present, and any irregular foci at
of eosinophilic to amphophilic cytoplasm. These cells appear the myometrial interface must be <3 in number, and each of
to whorl around a prominent vascular component, which these foci should extend <3 mm from the main mass [3].
resembles spiral arterioles of nonneoplastic endometrium ESN may demonstrate variant morphology, including smooth
(Fig. 1.2). The vessels are typically evenly spaced and uni- muscle and sex cord-like differentiation, making the diagno-
form in caliber throughout the neoplasm. In a minority of sis more challenging [6, 9, 10] (discussed in more detail in
cases, larger, thick-walled vessels may be focally present and the low-grade endometrial stromal sarcoma (LGESS) variant
are usually located at the periphery. ESN is characterized by morphology section).
sharp circumscription between the tumor nodule and sur- Occasionally, some endometrial stromal tumors are pre-
rounding tissue (endometrium or myometrium) (Fig. 1.3). dominantly well-circumscribed but exhibit greater than
Sometimes there may be some slight irregularity to the bor- 3 mm extension into the surrounding myometrium yet also
der in the form of small lobulated or fingerlike extensions lack the typical overt myometrial permeation seen in endo-
into the surrounding myometrium (Fig. 1.4) [3]. However in metrial stromal sarcoma. The term “endometrial stromal
1 Uterine Mesenchymal Lesions 3
mass or uterine enlargement may be palpated. In about one lymphovascular invasion, in contrast to ESN (Fig. 1.6).
third of cases, clinical presentation may be related to signs or Hyaline bands and plaques are commonly encountered in
symptoms related to metastatic disease. LGESS but are not a specific finding (Fig. 1.7a). Foamy his-
tiocytes, either singly or in clusters, may be seen in LGESS
1.2.2.3 Gross Findings (Fig. 1.7b). Less commonly encountered morphologic fea-
LGESS may present as an intracavitary or intramural uterine tures include extensive myxoid change and a prominent
mass and can appear well-circumscribed or with overt myo- fibrous stroma.
metrial infiltration with tonguelike extensions and wormlike
intravascular tumor plugs (Fig. 1.5) [7]. The cut surface is 1.2.2.5 Biomarkers
generally tan to yellow and hemorrhage may be apparent. CD10 was recognized as a potential marker of endometrial
Necrosis is less commonly identified grossly. stromal differentiation based on its high expression in LGESS
[15–18]. It is now well known that CD10 is expressed in
1.2.2.4 Microscopic Findings and Histologic endometrial stromal cells including those in eutopic endome-
Grading trium, adenomyosis, and endometriosis in addition to ESN
Histologically, the tumor cells resemble nonneoplastic and LGESS [18, 19]. CD10 is typically strongly and diffusely
proliferative-phase endometrial stroma and appear virtually positive in nonneoplastic and neoplastic endometrial stroma
identical to ESN. However, LGESS by definition exhibits (Fig. 1.8a); however, some endometrial stromal tumors may
prominent fingerlike penetration of the myometrium and/or be negative for this marker [20–22]. Another marker of
Fig. 1.5 Low-grade endometrial stromal sarcoma, gross appearance. Fig. 1.6 Low-grade endometrial stromal sarcoma. From low-power
Endometrial stromal sarcoma is a tan to yellow mass with nodular microscopic examination, the permeative fingerlike pattern of infiltra-
“wormlike” growth through the myometrium tion into the myometrium is apparent
a b
Fig. 1.7 Low-grade endometrial stromal sarcoma histologic features. also be present in small numbers or large aggregates in endometrial
(a) Endometrial stromal neoplasms frequently have hyaline plaques or stromal neoplasms
bands, but this is not specific to this entity. (b) Foamy histiocytes may
1 Uterine Mesenchymal Lesions 5
a b
Fig. 1.8 Low-grade endometrial stromal sarcoma immunohistochem- plasms. (c) Desmin is typically negative in endometrial stromal neo-
istry. (a) CD10 is almost always positive in LGESS (as well as ESN). plasms; however, it may be positive in areas of smooth muscle
(b) The hormone receptors estrogen receptor and progesterone receptor differentiation
are typically strongly positive in low-grade endometrial stromal neo-
e ndometrial stromal differentiation is the novel marker inter- Other less common variant features include epithelioid mor-
feron-induced transmembrane protein 1 (IFITM1) which has phology, clear or granular cytoplasm, bizarre atypia, pseudo-
shown to be a sensitive and specific marker for endometrial papillary growth, ossification, osteoclast-like giant cells, and
stromal tumors and is also highly expressed in nonneoplastic adipocytic metaplasia [27–32].
endometrial stroma [23]. Endometrial stromal tumors are
generally positive for ER and PR (Fig. 1.8b) and negative for 1.2.2.7 Endometrial Stromal Tumor
smooth muscle markers (Fig. 1.8c); however desmin positiv- with Endometrioid Glands
ity has been documented in otherwise conventional appearing Uncommonly, LGESS may have foci of endometrioid glandu-
ESTs, while caldesmon appears to be specific for smooth lar differentiation [30, 33]. Although divergent differentiation
muscle differentiation. p53 is generally wild type in is the most likely explanation for their presence, some exam-
LGESS. Nuclear beta-catenin is frequently expressed in both ples may represent entrapped nonneoplastic endometrial glands
ESN and LGESS [24]. (Fig. 1.9). In general, this type of differentiation in LGESS is
focal and the main differential diagnosis is the distinction from
1.2.2.6 Variants of Endometrial Stromal Tumors adenomyosis, particularly gland-poor adenomyosis, which
Both ESN and LGESS may exhibit a wide range of altered only rarely forms a distinct grossly apparent mass [34].
differentiation, including smooth muscle, sex cord-like, and
epithelial (endometrioid-type glands); very rarely skeletal 1.2.2.8 Endometrial Stromal Tumor with Smooth
muscle differentiation may be seen [4, 9, 10, 25–27]. Muscle Differentiation
Extensive stromal hyalinization secondary to increased col- ESN/LGESS with smooth muscle differentiation are not
lagenous matrix production may impart a fibroblastic appear- common but are a source of confusion in the recognition and
ance. However, in general, other areas with typical diagnosis of endometrial stromal tumors [10, 26]. If the
endometrial stromal morphology are usually present. In smooth muscle component comprises more than 30% of the
addition, the characteristic vascular component and arrange- tumor, then they are considered by some to be mixed endo-
ment of the tumor cells around them are maintained [28, 29]. metrial stromal-smooth muscle tumors [10]; however, many
6 B. E. Howitt and M. R. Nucci
prefer to use the terminology ESN with smooth muscle dif- 1.2.2.9 Endometrial Stromal Tumor with Sex
ferentiation or LGESS with smooth muscle differentiation. Cord-Like Elements
Historically these tumors may have been called “stromomy- LGESS/ESN may contain variable amounts of sex cord-like
oma.” Histologic features of smooth muscle differentiation elements, typically resembling ovarian granulosa cell or
include typical smooth muscle morphology reminiscent of Sertoli cell tumors (Fig. 1.11a, b). This finding is seen in up
that seen in leiomyomata (Fig. 1.10a), nodules with central
prominent hyalinization, or irregular islands that can be
either discrete or merge imperceptibly with the areas charac- a
teristic of stromal differentiation (Fig. 1.10b). These tumors
have been shown to be of endometrial stromal derivation by
molecular analyses; therefore, determination of benign ver-
sus malignant should be made using the criteria for endome-
trial stromal tumors [10]. Immunohistochemically, areas of
smooth muscle differentiation within an endometrial stromal
tumor will stain identically to smooth muscle tumors, while
areas of conventional endometrial stromal appearance should
be negative for caldesmon but may be variably positive for
desmin. CD10 is not entirely specific and may be expressed
in the smooth muscle component.
a b
Fig. 1.11 Low-grade endometrial stromal sarcoma with sex cord differentiation. Endometrial stromal tumors may have variant sex cord differen-
tiation (a). Typically, this is in the form of cords and trabeculae resembling granulosa cell or Sertoli cell tumor (b)
1 Uterine Mesenchymal Lesions 7
to 60% of endometrial stromal tumors and may coexist with be confused with uterine smooth muscle tumors [10, 20, 27,
smooth muscle differentiation. The presence of sex cord-like 28, 51]. Conversely, uterine smooth muscle tumors may
differentiation in ESN and LGESS does not appear to have mimic endometrial stromal tumors, particularly when the
an impact on clinical behavior. Histopathologic criteria for former is markedly cellular (e.g., highly cellular leiomyoma)
the distinction between a stromal nodule and stromal sar- or has prominent vascular invasion (e.g., intravascular leio-
coma, as outlined previously, are applied regardless of the myomatosis, IVL) [8, 52–55]. In general, the morphologic
presence of sex cord-like elements. Immunohistochemically, appearance of tumor cells and the growth pattern within the
the sex cord-like elements will stain for typical sex cord myometrium can distinguish LGESS from leiomyosarcoma.
markers (e.g., inhibin, calretinin, CD99) only in the areas of In contrast to leiomyosarcoma, areas of smooth muscle dif-
sex cord differentiation. The background endometrial stro- ferentiation in LGESS tend to be bland and will not exhibit
mal tumor will typically not be positive for these markers. the degree of cellularity and nuclear pleomorphism that is
frequently encountered in leiomyosarcoma. In cases in
1.2.2.10 Genetic Profile which there is prominent lymphatic or vascular permeation
Chromosomal rearrangements involving chromosomes 6, 7, by leiomyosarcoma, morphologic features such as the pres-
and 17 are the most frequent cytogenetic abnormalities that ence of fascicular growth even in the intravascular compo-
have been reported in both LGESS and ESN, with JAZF1- nent help facilitate its recognition as a malignant smooth
SUZ12 being the most common gene fusion identified muscle tumor. In addition, a panel of antibodies including
(reported frequency ranges from 25% to >90% depending on caldesmon, desmin, and CD10 may be helpful in difficult
the study design and tumor morphology) [32, 35–41]. This cases, provided one is aware of the potential pitfalls. ESN/
gene fusion reflects the chromosomal translocation t(7;17) LGESS are at most only focally positive for smooth muscle
(p15;q21) or related variant translocations frequently actin and desmin in most cases; however, a subset of mor-
observed in LGESS via conventional cytogenetics or phologically typical cases may show more extensive expres-
FISH. In ESN, one study has demonstrated that the non- sion of these markers [9, 10, 28, 56–58]. In contrast,
rearranged JAZF1 allele is transcriptionally active in ESN, caldesmon is a more specific marker of smooth muscle dif-
but in ESS it appears to be silenced [42], suggesting that epi- ferentiation than desmin and may be useful in this differen-
genetic changes play a role in LGESS pathogenesis. Another tial [56, 59]. In the distinction between intravenous
translocation involving JAZF1, t(6;7)(p21;p15), resulting in leiomyomatosis and LGESS, the histologic features sugges-
a JAZF1-PHF1 gene fusion, is present in up to 28% of ESS tive of smooth muscle in the former including vasculature
[41, 43–45]. LGESS with PHF1 rearrangement is enriched and clefting may be helpful. Additionally, the nuclei of
for sex cord-like differentiation [43, 45] but also may show smooth muscle tumors tend to be blunted (“cigar-shaped”)
myxoid morphology, smooth muscle differentiation, or typi- rather than delicately ovoid or fusiform as in endometrial
cal morphology [46]. In tumors lacking JAZF1 abnormali- stromal tumors. One pitfall to keep in mind is that caldesmon
ties, PHF1 has been also been found to be recurrently may, in some cases of highly cellular smooth muscle neo-
involved in another chromosomal translocation and resultant plasms, including cellular IVL, only show patchy or focal
gene fusion with MEAF6 [47, 48]. Other partners of PHF1 positivity [56]. One study has shown that IFITM1 may be a
described in LGESS include EPC1, BRD8, and EPC2 [49]. more specific marker of endometrial stromal differentiation
A subset of LGESS with conventional karyotyping have no in the distinction from smooth muscle tumors [60]. Similarly,
evidence of chromosomal rearrangements as well as no evi- nuclear beta-catenin staining is supportive of EST diagnosis
dence of JAZF1 or PHF1 gene fusions by RT-PCR or FISH, over a smooth muscle tumor in one study [24]. If there is a
suggesting that some of the molecular alterations in these question diagnostically, FISH for JAZF1 rearrangement may
tumors have not yet been discovered or may be too small to be performed.
detect with these methods. In addition, FISH studies of Endometrial stromal tumors with sex cord-like differen-
mixed endometrial stromal-smooth muscle tumors show evi- tiation must be distinguished from uterine tumors resem-
dence for the JAZF1 gene rearrangement in both the endo- bling ovarian sex cord tumor (UTROSCT). In the initial
metrial stromal component and smooth muscle component, description of UTROSCT, all endometrial stromal tumors
supporting the concept that these tumors are of endometrial with sex cord elements were divided into group I or group II
stromal derivation [38, 40]. Rarely, LGESS can show MDM2 based on the percent of sex cord elements present in the
amplification by FISH as well as MDM2 protein expression tumor, with group I containing only focal sex cord elements
by immunohistochemistry [50]. and group II a predominance [61]. Since that time, it has
been recognized that group I represents endometrial stromal
1.2.2.11 Differential Diagnosis tumors with sex cord-like elements while group II represents
Endometrial stromal tumors with smooth muscle differentia- UTROSCT (see later section). The finding of conventional
tion or those that have a fibrous or myxoid appearance may endometrial stromal neoplasia is the distinguishing feature
8 B. E. Howitt and M. R. Nucci
a b
Fig. 1.13 High-grade endometrial stromal sarcoma, microscopic fea- stromal neoplasms, the vasculature in high-grade endometrial stromal
tures. (a) Though permeative fingerlike infiltration is often seen, sarcoma is characterized by thin-walled, delicate arborizing vessels.
destructive myometrial invasion (pictured here) may also be identified Additionally the tumor cells are epithelioid with vesicular nuclei
in high-grade stromal sarcoma. (b) In contrast to low-grade endometrial
a b
Fig. 1.14 High-grade endometrial sarcoma (YWHAE-rearranged) is strongly and diffusely positive for cyclin D1 (a) and BCOR (b)
contrast, the high-grade component is typically negative for worth noting that other biologically related genomic altera-
CD10, ER, and PR and shows strong and diffuse positivity tions have been recently discovered in high-grade uterine
(>70% tumor cell nuclei) for cyclin D1 (Fig. 1.14a), which sarcomas (see BCOR-altered sarcomas below). Although
may be used as a marker for YWHAE rearrangement [32, 70, most HGESS are positive for CD117 (c-kit), they lack hot
72], although this is not entirely sensitive or specific [73, 74]. spot mutations in CKIT [75].
Additionally, CD117 (c-kit) and BCOR (Fig. 1.14b) may be
positive in HGESS [75, 76]. 1.2.3.7 Differential Diagnosis
Given that HGESS has an epithelioid appearance, it may be
1.2.3.6 Genetic Profile confused with an undifferentiated carcinoma, particularly in
HGESS has the characteristic karyotypic abnormality limited sampling such as a biopsy or curettage. One pitfall to
t(10;17)(q22;p13) associated with a YWHAE-NUTM2A/B keep in mind is that endometrial undifferentiated carcinoma
fusion [67, 68]. YWHAE rearrangements have not been found can also show strong and diffuse positivity for cyclin D1;
in other gynecologic tumors, and FISH or RT-PCR studies however, they will usually show focal positivity for EMA, a
may serve as a useful adjunct to the histologic diagnosis [72, broad-spectrum cytokeratin. Undifferentiated carcinomas
77]. FISH analysis on FFPE tissue for the YWHAE- may also be positive for CD10 [73]. Identification of a well-
NUTM2A/B rearrangement is becoming more widely avail- differentiated adenocarcinoma component is helpful in
able and is feasible to use in routine clinical practice. It is establishing a diagnosis of undifferentiated carcinoma
10 B. E. Howitt and M. R. Nucci
a b
Fig. 1.15 BCOR-altered sarcoma. (a) The gross appearance of a BCOR- frequently demonstrate a permeative, fingerlike pattern of myometrial inva-
fusion sarcoma is similar to other uterine sarcomas with a fleshy, heteroge- sion. (c) Histologically, a myxoid stroma is typically prominent in sarco-
neous cut surface and foci of hemorrhage and necrosis. (b) These sarcomas mas with BCOR ITD, with hyperchromatic spindled tumor cells
a b
Fig. 1.16 BCOR-altered sarcoma, immunohistochemistry. (a) CD10 is be negative or only weakly focally positive for CD10. (b) Cyclin D1 is
strongly and diffusely positive in BCOR-fusion sarcomas as pictured strongly and diffusely positive in BCOR-altered sarcomas
here; however, those with internal tandem duplications of BCOR tend to
12 B. E. Howitt and M. R. Nucci
1.2.5.1 Definition
Undifferentiated uterine sarcoma (UUS) is a heterogeneous
group of tumors and likely represents dedifferentiated forms
of specific uterine sarcomas (adenosarcoma, endometrial
stromal sarcoma, carcinosarcoma, leiomyosarcoma, etc.).
Some consider this entity as “undifferentiated endometrial
sarcoma”; however, we prefer the term UUS, as some of these
tumors are likely not of endometrial stromal derivation.
f requently have abnormal overexpression of p53 in contrast are subgroups of UUS with varying prognoses [88].
to HGESS [84]. Treatment options include complete debulking surgery as
well as consideration of radiation therapy for local control
1.2.5.6 Differential Diagnosis and chemotherapy for systemic control.
The diagnosis of UUS is essentially a diagnosis of exclusion
and should only be considered after excluding an undifferen-
tiated carcinoma, carcinosarcoma, leiomyosarcoma, and 1.2.6 Diagnostic Considerations
high-grade endometrial sarcoma; all of which may have from Curettage Specimens
areas of morphologic overlap. A diagnosis of UUS should
only be made on a complete excision specimen, as biopsies A suspected endometrial stromal neoplasm identified in a
may lack the diagnostic features of other uterine sarcomas/ curettage or biopsy specimen must be distinguished from
carcinomas. Extensive tumor sampling, immunohistochemi- potential mimics, including nonneoplastic aglandular func-
cal staining, and in some cases molecular testing are required tionalis, endometrial basalis, as well as benign neoplasms
to confidently diagnose UUS. such as endometrial polyp and malignant neoplasms most
UUS can be distinguished from undifferentiated carci- notably adenosarcoma. Multiple fragments of aglandular
noma and carcinosarcoma by identifying areas diagnostic of cellular endometrial-type stroma containing spiral arteriole-
an epithelial component, either morphologically or immuno- like vessels are suggestive of a stromal neoplasm (Fig. 1.18);
histochemically (EMA and/or cytokeratin positivity). in one study, curettage of a stromal neoplasm typically pro-
Similarly, leiomyosarcoma (LMS) may demonstrate a vari- duces fragments of aglandular stroma measuring ≥5 mm
ety of patterns and identifying a better differentiated area of [89]. Distinction of endometrial stromal neoplasms from
the tumor is key – LMS should contain, at least focally, fragments of basalis is made by virtue of the presence of an
strong positivity for one or more smooth muscle markers orderly component of glands in the latter. Strips of aglandu-
(SMA, desmin, caldesmon). The distinction of UUS from lar functionalis, usually associated with submucosal leio-
HGESS was discussed previously in the HGESS section. In myomata (Fig. 1.19) or occasionally progestin therapy, tend
all of these cases, adequate sampling and careful evaluation to be less cellular and show features of compression or reac-
of all tumor morphologies is necessary in making a diagnosis tive surface changes when related to submucosal leiomy-
of UUS. As molecular studies are more widely available, oma. Fragments of cellular endometrial polyps usually
these should be considered to exclude known translocation- exhibit other features of polyp, including large thick-walled
associated entities including YWHAE, JAZF1, and BCOR vessels and abnormal glandular architecture. Adenosarcoma
FISH as well as examining for BCOR internal tandem also may exhibit a cellular stroma that is CD10 positive but
duplications. typically has an epithelial component with glandular cuff-
ing, albeit sometimes subtle. Appreciation of a more spin-
1.2.5.7 Genetic Profile dled atypical stroma without the characteristic vascular
UUS tend to be cytogenetically complex, particularly when network facilitates this distinction.
histologically pleomorphic [85]. As mentioned above, previ-
ously termed “UUS with uniform morphology” likely repre-
sent various forms of HGESS and rarely may harbor JAZF1
translocation [83]. TP53 mutations are not uncommon in
UUS, in contrast to endometrial stromal neoplasms [83],
suggesting that those UUS with TP53 mutation may have no
relationship with ESS or that they have acquired a secondary
TP53 mutation. One study interrogated for KIT, EGFR, and
PDGFR hot spot mutations as well as amplification of EGFR
in UUS; none of the tumors included in their study had any
molecular aberration in these genes [86]. In an array CGH
study on endometrial sarcomas, a large number of copy num-
ber alterations in UUS are described, including gain of 7p in
a subset [85].
1.2.5.8 Management and Outcomes Fig. 1.18 Recognizing endometrial stromal neoplasia in an endome-
trial curettage. Multiple large fragments or thick strips of pure endome-
UUS is an aggressive tumor with a dismal prognosis. The trial stromal tissue without any glands is concerning for an endometrial
mean overall survival is less than 2 years [87]. In one study, stromal neoplasm. The finding of sex cord elements, as demonstrated
one third of patients did survive >5 years, suggesting there here, is additionally supportive of a neoplasm
14 B. E. Howitt and M. R. Nucci
a b
Fig. 1.20 Histologic features of uterine tumor resembling ovarian sex grooves. (b) Though usually well-circumscribed, UTROSCT can also
cord tumor (UTROSCT). (a) The granulosa cell-like pattern is com- have infiltrative borders. (c) Rarely, brisk mitotic activity may be
mon, with trabeculae and nests of uniform ovoid cells with nuclear appreciated
a b
Fig. 1.21 Immunohistochemical markers in uterine tumor resembling fusely positive in UTROSCT, similar to sex cord tumors of the ovary.
ovarian sex cord tumor (UTROSCT). (a) Steroidogenic factor 1 (SF-1) (b) Calretinin is typically strongly positive in UTROSCT but is a less
is a specific marker for sex cord differentiation and is strongly and dif- specific marker of sex cord differentiation
16 B. E. Howitt and M. R. Nucci
displaying sex cord differentiation, most notably endometrial Treatment is surgical, most commonly hysterectomy, but in
stromal neoplasms and less commonly adenosarcoma [100]. cases where fertility preservation is desired, a more conser-
The most important distinction is from LGESS, which is vative approach has been advocated with successful out-
malignant and associated with a much higher recurrence rate comes [105, 106]. In the rare examples of UTROSCT with
than UTROSCT. worrisome histologic features such as necrosis and brisk
ESN and LGESS may exhibit morphologic features of mitotic rate, we favor regarding these tumors as having a
ovarian sex cord-stromal tumors (discussed in the endome- higher likelihood for malignant behavior, though no studies
trial stromal tumors section). These elements may be present have established histologic criteria for malignancy in
focally, most often in the form of inter-anastomosing trabec- UTROSCT. As such, they are uniformly regarded as neo-
ulae, cords, and less commonly tubules. Uncommonly, they plasms with uncertain malignant potential.
may also be the predominant pattern in which the stromal
element is less conspicuous, particularly in a biopsy speci-
men. Whether UTROSCT also represent a form of endome- 1.3 Uterine Smooth Muscle Tumors
trial stromal neoplasia is a subject of debate although recent
molecular evidence suggests they may not be of endometrial Overview
stromal derivation (see genetic profile section below). Other Smooth muscle tumors are the most common uterine neo-
frequent entities considered in the differential diagnosis with plasms and include benign leiomyoma and variants, intrave-
UTROSCT are an epithelioid smooth muscle tumor and nous leiomyomatosis, smooth muscle tumors of uncertain
PEComa. Immunohistochemistry should be helpful in distin- malignant potential, and leiomyosarcoma.
guishing an epithelioid smooth muscle tumor from
UTROSCT, as smooth muscle tumors are positive for SMA
and desmin, and occasionally cytokeratins, but should be 1.3.1 Leiomyoma
negative for sex cord markers (inhibin, SF1, CD99, WT1,
melanA, etc.). PEComa demonstrates dual positivity for 1.3.1.1 Definition
smooth muscle and melanocytic markers (MelanA, HMB45) Leiomyoma is a benign smooth muscle tumor.
but should be negative for SF-1, WT1, and CD99. As indi-
vidual examples of UTROSCT are not always positive for all 1.3.1.2 Clinical Features
of these markers, a large panel of immunohistochemical Although more than 75% of women have leiomyoma(s) in
stains may be helpful in making the diagnosis. Less likely to their uterus, only about 25% are symptomatic [107–109].
be confused with UTROSCT is an endometrioid carcinoma Symptoms related to uterine leiomyoma include abnormal
with extensive sex cord-like differentiation, either primary in uterine bleeding, pelvic pain or pressure, and infertility or
the endometrium or metastatic from the ovary. Identifying recurrent pregnancy loss [108, 110, 111]. The type and sever-
typical adenocarcinoma morphology is helpful in this dis- ity of symptoms associated with leiomyomata is broadly
tinction, as well as the lack of smooth muscle marker and sex related to tumor size and location within the uterine wall
cord marker positivity in endometrioid carcinoma. [108]. The myometrium can conceptually be divided into
three zones: submucosal, intramural, and subserosal.
1.2.7.7 Genetic Profile Submucosal leiomyomata, as well as larger intramural
UTROSCT have been thought to represent endometrial stro- tumors that distort the endometrial cavity, may cause abnor-
mal neoplasms which show predominantly or entirely sex mal uterine bleeding and reproductive problems. Subserosal
cord differentiation; however, they do not harbor JAZF1 or or deeper leiomyomata are less likely to cause uterine bleed-
PHF1 rearrangements [101, 102]. Thus, many now believe ing but are more likely to be associated with pelvic pain or
that these may not represent endometrial stromal neoplasms. pressure.
Similarly, they do not harbor FOXL2 or DICER1 mutations
characteristic of the ovarian sex cord-stromal tumors, adult- 1.3.1.3 Gross Findings
type granulosa cell tumor and Sertoli-Leydig cell tumor, Leiomyomas may be essentially microscopic or very large
respectively [96, 103]. No recurrent or specific molecular (>20 cm) and solitary or numerous. The usual leiomyoma is
alterations have been described in UTROSCT, however the grossly well circumscribed and has a firm, rubbery texture
karyotype of one case has been reported, which revealed two (Fig. 1.22). Leiomyoma typically bulges out upon sectioning
balanced translocations t(X;6)(p22.3,q23.2) and t(4;18) because of increased intratumoral pressure. The cut surfaces
(q21.1;q21.3) [104]. are white or slightly pink and the bands of neoplastic smooth
muscle are often whorled, giving the impression that the
1.2.7.8 Management and Outcomes smooth muscle bundles are wrapped around a central core.
UTROSCT is generally a benign tumor with excellent prog- Lastly, there should be minimal variation in the appearance
nosis, but rare cases have been reported to recur [61, 91, 93]. of the cut surface. Minute foci of hemorrhage may be seen,
1 Uterine Mesenchymal Lesions 17
a b
Fig. 1.25 Histologic features of leiomyoma with hydropic change. (a) more difficult. (b) From low-power magnification, alternating hyper-
Prominent edema is present in hydropic leiomyoma, which may mimic and hypocellular regions may be apparent
myxoid stroma, and make the recognition of a smooth muscle neoplasm
Hydropic foci may be microscopic and patchy (Fig. 1.25a). sound energy transfer results in microscopically bland coag-
Microscopic foci of hydropic degeneration may be worri- ulative necrosis lacking atypical ghost cells.
some at lower magnifications because they often have more
irregular contours and may have sharp borders (Fig. 1.25b). 1.3.1.6 Genetic Features of Leiomyoma
Careful attention at higher magnification often resolves this Leiomyoma is characterized by recurrent point mutations
by noting the fluid-filled intercellular spaces and the absence and small deletions as well as characteristic chromosomal
of truly necrotic cells. When more extensive, the wet, slip- translocations. The karyotype of leiomyoma is either normal
pery cut surface of hydropic degeneration may often be (60%) or noncomplex with one or more chromosomal trans-
described as “myxoid.” Such tumors should be histologically locations (40%) [119–124]. The gene most commonly
examined to exclude a myxoid leiomyosarcoma by recogniz- mutated in leiomyoma is MED12, with heterozygous MED12
ing their edematous nature and lack of other malignant fea- mutations found in up to 70–80% of uterine leiomyomas
tures. In contrast to myxoid leiomyosarcoma, hydropic [125–130]. The vast majority of these are located in exon 2
leiomyomata do not infiltrate into the fascicles of adjacent at the codon 44 position. There is no evidence to date that
myometrium. germline MED12 mutations play a role in the development
In addition to the naturally occurring patterns of degen- of leiomyoma. HMGA1 or HMGA2 overexpression is com-
eration, we now frequently encounter iatrogenic necrosis and mon in leiomyoma, reflecting chromosomal translocations
other changes related to attempted conservative (nonsurgi- involving 6p21 (HMGA1) and 12q14 (HMGA2). HMGA2
cal) management of uterine fibroids. A number of new thera- overexpression is found in 10–25% of uterine leiomyomas
peutic strategies to noninvasively ablate uterine leiomyomata and tends to be mutually exclusive with MED12 mutations
have been devised. The most widespread of these new proce- [127]. When considering uterine leiomyomas lacking
dures is uterine artery embolization [118]. In this technique, MED12 mutation, HMGA2 overexpression is found in ~40%
a catheter is threaded into the uterine artery and spheres of of tumors. Other recurrent molecular aberrations in leiomy-
synthetic polymer are injected. Because the blood supply to oma include COL4A5/6 deletions, which are mutually exclu-
leiomyomata is larger than to the surrounding myometrium, sive with MED12, HMGA2, and FH alterations [131].
the tumors tend to be more vulnerable to ischemia. Of note, Leiomyomas lack TP53 mutations [128, 131].
embolization material can frequently be seen distending and Recurrent chromosomal aberrations in leiomyoma
occluding vessels within and near the tumor. Another nonin- include (in addition to HMGA2/HMGA1 as described above)
vasive therapy for uterine leiomyomata is MRI-guided ther- 13q, 1p36, 10q22, 7q deletion, trisomy 12, 1p deletion, and
mal ablation by focused application of ultrasound energy. Of monosomy 22 [122, 132–137]. Interestingly, 7q deletions
interest to pathologists, the gross appearance of a leiomyoma have been identified in both MED12-mutated and HMGA2-
early after this type of treatment has some features grossly altered leiomyoma, suggesting that loss of 7q may be impor-
concerning for malignancy. Specifically, the borders of soni- tant for progression rather than initiation of tumorigenesis.
cated and nonsonicated tissues may be geographic and well-
defined, and sonicated tissues may be variegated in color and 1.3.1.7 Clinical Management and Outcome
show extensive hemorrhage. In distinction to geographic Leiomyomata are the most common indication for hysterec-
tumor necrosis, thermal effect resulting from focused ultra- tomy [108, 138]. However, leiomyomata may be managed
1 Uterine Mesenchymal Lesions 19
expectantly if associated with no or minimal symptoms. mitoses per 10 high-power fields; such tumors are classified
Myomectomy (resection with uterine conservation) is the as “leiomyomata with increased mitotic activity” [2, 149–
most widely employed hysterectomy alternative for women 152]. This diagnostic term is useful in denoting the special
who wish to preserve fertility [139]. Tumor size and location nature of these tumors, as well as communicating the absence
play an important role in determining which technique is most of other worrisome features. Ki-67 is a proliferation marker
appropriate. When compared with hysterectomy, the alterna- that may be helpful in the diagnosis of leiomyoma with
tives are associated with several unique risks, namely, the risks increased mitotic activity. One way in which Ki-67 immuno-
of tumor “recurrence” and uterine rupture during pregnancy histochemistry might be useful is in excluding pyknotic
following myomectomy [108]. The risk of symptomatic recur- nuclei as a mimic of mitoses. It should be noted that prolif-
rence is more likely due to the growth of a second crop of erative “hot spots” may be seen beneath eroded and attenu-
leiomyomata and requires a second operation in a subset of ated mucosal surfaces of submucosal leiomyoma or adjacent
cases [108]. Morcellation with powered devices may result in to ischemic changes. Mitotically active leiomyoma is not
peritoneal dissemination of benign and malignant smooth known to have any clinical significance with regard to recur-
muscle tumors [140, 141]. In fact, studies have shown that the rence or progression to malignancy. Similarly, there are no
rate of unexpected sarcoma after a laparoscopic morcellation known molecular or cytogenetic alterations specific to mitot-
procedure may be far higher than once thought and that dis- ically active leiomyoma. Atypical mitotic figures reflect
semination of leiomyosarcoma by morcellation occurs in genomic instability and as such should be considered a wor-
nearly two thirds of cases, some resulting in mortality [141]. risome finding. Detection of atypical mitoses in a smooth
Uterine artery embolization as described previously muscle tumor that would otherwise be classified as a leio-
results in symptomatic improvement for most patients, but myoma with increased mitotic activity should raise suspi-
this technique has been associated with adverse outcomes cion for a more ominous diagnosis as these are rarely if ever
ranging from postprocedure fevers to amenorrhea, uterine encountered in benign leiomyomata. Mitotically active leio-
rupture, endomyometritis, and fatal sepsis [142–144]. myomas appear to have similar molecular features as con-
Uterine artery embolization also has been implicated as a ventional leiomyoma, with frequent MED12 mutations, but
factor delaying the diagnosis of uterine sarcomas [145]. infrequent abnormal expression of HMGA2 or FH [128,
Medical therapy, in the form of androgenic steroids, 153].
gonadotropin-releasing hormone (GnRH) agonists, or selec-
tive estrogen receptor modulators, may also be attempted in 1.3.2.2 Cellular Leiomyoma
the clinical management of symptomatic fibroids, with vari- Leiomyomata are often more cellular than their surrounding
able success rates [146, 147]. myometrium. The term cellular leiomyoma is reserved for
notably cellular tumors. Although hypercellularity is neces-
sary to classify a tumor as this benign variant, it is quite
1.3.2 Variants of Leiomyoma subjective and difficult to quantitate just how cellular a tumor
must be to make a diagnosis of cellular leiomyoma. On
According to the 2014 WHO classification, there are a num- inspection of tissue sections without the microscope, the
ber of distinct morphologic variants of leiomyoma, including heavy hematoxylin staining of a cellular leiomyoma often
cellular leiomyoma, leiomyoma with bizarre nuclei, mitoti- will stand out, reflecting the increased number of nuclei as
cally active leiomyoma, hydropic leiomyoma, apoplectic well as the relative lack of extracellular matrix deposition.
leiomyoma, epithelioid leiomyoma, myxoid leiomyoma, and One practical approach is to classify only those tumors with
dissecting (cotyledonoid) leiomyoma, some with specific an extreme or eye-catching degree of cellularity as cellular
genetic alterations. Additionally, there are leiomyomas with leiomyoma. Another approach is to reserve the term “cellular
other cellular elements such as lipoleiomyoma and angi- leiomyoma” for cases in which an endometrial stromal neo-
oleiomyoma (or “vascular leiomyoma”). Each of these will plasm was at least considered in the microscopic differential
be discussed below with relevant clinicopathologic and diagnosis. Few studies have examined the molecular altera-
genetic information included with each variant. tions specifically in cellular leiomyoma. From the published
data available for review, they lack TP53 mutations [128],
1.3.2.1 Mitotically Active Leiomyoma only 6% have PTEN deletions, and only 9–14% have MED12
Both nonneoplastic myometrium and leiomyoma demon- mutations [128, 154], suggesting the pathogenesis of cellular
strate variation in mitotic rate depending on the phase of leiomyoma may differ from conventional leiomyoma. 1p
menstrual cycle [113]. Some investigators have noted that deletion was present in 23% of cellular leiomyomas in one
exogenous progesterone increases the mitotic rate in leio- study [155] and a subset have been reported to have 10q22
myomata [148]. Occasionally, leiomyomas may be histolog- rearrangements, which have also been described in conven-
ically unremarkable but demonstrate mitotic activity up to 15 tional leiomyoma [156].
20 B. E. Howitt and M. R. Nucci
similarly well circumscribed, without significant cytologic there is some molecular evidence suggesting other mecha-
atypia or necrosis. Mitoses should number <2 per 10 high- nisms. For example, rearrangements of 12q15 and aberrant
power fields and lack atypical forms. As the amount of myx- expression of HMGA2 have been described, which are simi-
oid matrix may vary, conventional areas of leiomyoma are lar genetic aberrations that have been described in lipoma,
often seen. The principal differential diagnostic consider- suggesting a common mechanism for adipocytic differentia-
ations include distinction from myxoid leiomyosarcoma, tion [168, 169]. Rare examples of liposarcoma arising from
inflammatory myofibroblastic tumor, and myxoid change lipoleiomyoma have been described; in these cases, the
associated with degeneration. The diagnosis of myxoid leio- tumors showed features of liposarcoma including pleomor-
myosarcoma should be considered in any myxoid smooth phic hypercellular areas and lipoblasts [170].
muscle tumor with cytologic atypia, infiltrative margins,
and/or a mitotic rate of >2 per 10 high-power fields (see 1.3.2.8 Leiomyoma with Other Cellular
myxoid leiomyosarcoma discussion below). Inflammatory Components: Angioleiomyoma
myofibroblastic tumor can also have a myxoid matrix but is (“Vascular” Leiomyoma)
distinguished from myxoid leiomyoma by more often having Leiomyomas may sometimes have increased numbers of
an irregular interface with the myometrium, its often promi- vascular channels and these have been variably described as
nent lymphoplasmacytic infiltrate, and its characteristic “tis- “vascular” leiomyoma or angioleiomyoma [171, 172]. The
sue culture”-like appearance with cells having open vesicular vascular component can be composed of large-caliber ves-
nuclei. In addition, inflammatory myofibroblastic tumor sels (“venous”-like) or numerous smaller vessels with peri-
typically shows diffuse positivity for ALK corresponding to vascular concentric smooth muscle proliferations
ALK rearrangements. Alk immunohistochemistry should be (“myopericytoma”-like). When the tumor has large-caliber
considered before rendering a diagnosis of myxoid leiomy- vessels, the main distinction is from an arteriovenous malfor-
oma, as it has been reportedly misdiagnosed in the literature mation, which would not be well circumscribed. When the
[163]. A more frequent diagnostic consideration is the dis- vasculature is myopericytic-like, an endometrial stromal
tinction from leiomyoma with degenerative changes. In this tumor may be entertained; however, an angioleiomyoma
scenario, the myxoid change is typically focal or multifocal would show positivity for smooth muscle markers, including
and tends to occur immediately adjacent to ischemic necro- desmin and caldesmon.
sis or around blood vessels. Appreciating the association
with degenerative changes is helpful in this distinction. 1.3.2.9 Leiomyoma with Special Growth
Patterns: Diffuse Leiomyomatosis
1.3.2.6 Dissecting (“Cotyledonoid”) Leiomyoma Diffuse leiomyomatosis, also known as intrauterine leiomy-
Dissecting leiomyoma is an uncommon lesion which also omatosis, is an unusual entity characterized by a large num-
has been referred to as “cotyledonoid” due to its gross resem- ber (typically >100) of leiomyomata which are confined to
blance to placental tissue when the tumor shows extrauterine the uterus [173]. Typically, on gross examination, there are
extension. This unusual entity typically occurs in innumerable masses which enlarge and distort the uterus and
reproductive-aged women and is characterized by tongues of often appear confluent on cut section. Microscopically, the
benign-appearing smooth muscle neoplasia, with varying tumors tend to be more cellular than conventional leiomy-
degrees of hydropic change, dissecting the surrounding myo- oma and show indistinct borders with merging of the adja-
metrium with occasional involvement of the broad ligament cent tumor masses. Nuclear atypia, necrosis, and vascular
and pelvis [164–166]. There is no cytologic atypia, necrosis, involvement are absent. This is a clinically benign entity of
or significant mitotic activity. When there is intravascular uncertain etiology although molecular studies have shown
extension, this tumor overlaps with intravenous leiomyoma- the tumorous masses to be polyclonal suggesting that patients
tosis (see below). with this disease are prone to leiomyoma development [174].
or bleeding typically predominates. If extension into the vena aggressive biologic property. Despite this malignant charac-
cava is significant, symptoms such as syncope and dyspnea teristic, the intravascular growth is usually indolent. In most
may be prominent. Upon gross examination, wormlike tumor cases, surgical intervention is sufficient treatment.
projections may be seen in the parametria of hysterectomy Gonadotropin-releasing hormone agonist (e.g., leuprolide)
specimens (Fig. 1.27a). This pattern of intravascular growth may be used as short-term therapy. Molecularly, IVL appears
also may be seen in endometrial stromal sarcoma as well as to have similar cytogenetic alterations commonly seen in con-
uterine leiomyosarcoma [55]; however, the appearance of ventional leiomyoma such as t(12;14). Regional losses on
IVL is otherwise identical to that of typical leiomyoma, with chromosomes 22q and 1p and gains on chromosomes 12q
a firm, white to tan rubbery cut surface. Similarly, the micro- were the most common alterations in one study [179]. MED12
scopic appearance in most cases of IVL is that of a typical mutations have not been documented in IVL [179].
uterine leiomyoma but uncommonly may exhibit the same Karyotypes of two cases of IVL have shown the presence of
spectrum of benign variant morphologies as seen in leiomy- a derivative chromosome, der(14)t(12;14)(q15;q24), which is
oma [52] (Fig. 1.27b). Atypical histology and adhesion to the also frequently found in typical uterine leiomyomata [180].
vessel wall may be associated with a more aggressive clinical Presumably, the ability of IVL to grow within venous spaces
course [178]. Intravenous extension might be considered an reflects some additional unknown genetic alterations.
1.3.3 Leiomyosarcoma
1.3.3.5 Biomarkers
Immunohistochemically, LMS are usually positive for smooth
muscle markers including smooth muscle actin (SMA), des-
min, and caldesmon; however, it is not unusual to lose expres-
sion for one or more of these, particularly in morphologically
Fig. 1.31 Conventional leiomyosarcoma, microscopic appearance. high-grade or poorly differentiated tumors. Of note, LMS may
Leiomyosarcoma often has at least focally marked pleomorphism and demonstrate keratin and/or EMA positivity [20, 197, 198].
multinucleated tumor giant cells Once the line of smooth muscle differentiation is established,
additional stains have been proposed to aid in classification of
demonstrate diffuse cytologic atypia that is moderate to the smooth muscle tumor, including p16, p53, ER, PR, ki67,
severe. Multinucleated tumor cells are present in over half of stathmin, fascin, and IMP-3. p16 and p53 are overexpressed in
cases (Fig. 1.31). Tumor cell necrosis is defined as an abrupt most LMS, and proliferation as indicated by ki67 index is also
transition between viable tumor and necrotic areas, without significantly higher in LMS [199–205]. Complete absence of
an intervening area of hyalinization/inflammation or other p16 staining may also be seen in LMS [206]. Stathmin is not a
ischemia-related histopathology (Fig. 1.32). Tumor cell specific marker for LMS, but the presence of only weak stath-
necrosis and abundant mitotic activity is usually, but not min staining in a smooth muscle tumor should prompt special
always, present. If tumor cell necrosis is present in a smooth consideration for a benign variant as all LMS were diffusely
muscle neoplasm with at least diffusely moderate nuclear strongly positive for stathmin in one study [207]. Fascin and
atypia, then the mitotic index is not critical for the diagnosis IMP-3 are two additional markers showing increased expres-
of leiomyosarcoma. If, however, tumor cell necrosis is sion in LMS and STUMP compared to leiomyoma [208, 209].
absent, then both significant diffuse moderate to severe Approximately one third of LMS express ER/PR, which may
nuclear atypia as well as mitoses numbering at least 10 per have prognostic and/or therapeutic implications (see
10 HPFs is required for the diagnosis of LMS [195]. Tumors “Management and Outcomes” section below).
1 Uterine Mesenchymal Lesions 25
Immunohistochemistry should be performed selectively on [245–247]. Ovarian preservation has been considered in pre-
the blocks of tumor containing fascicular growth as staining menopausal patients with early-stage leiomyosarcoma but
more poorly differentiated/undifferentiated tumor foci will appears to yield little benefit [192, 241]. Intraperitoneal mor-
be of little use. cellation of an unsuspected LMS results in increased risk of
Inflammatory myofibroblastic tumor (IMT) may be con- recurrence and shorter time to recurrence [248]. Most studies
sidered in the differential diagnosis of LMS, though more have found that adjuvant chemotherapy or radiotherapy has
commonly in myxoid variants and LMS with less severe minimal, if any, effect on survival [192, 237, 240] though
cytologic atypia as well as smooth muscle tumors of uncer- some recent studies have shown benefit of chemotherapy in
tain malignant potential (STUMP; discussed later). The pres- metastatic LMS [192, 247]. Aromatase inhibitors may pro-
ence of a prominent lymphoplasmacytic inflammatory cell long progression-free survival for some patients with
component and myxoid stroma should prompt consideration ER-positive leiomyosarcoma [249–251]. mTOR inhibition
of IMT. Positive Alk immunohistochemistry and/or break- has been proposed in uterine LMS as activated AKT/mTOR
apart FISH can confirm the diagnosis [230]. pathway proteins are highly expressed [252–254]. With the
Gastrointestinal stromal tumor/sarcoma may rarely be advent of widely available next-generation sequencing and
encountered in the uterus and as there may be significant other molecular testing on tumor specimens, it is likely that
morphologic overlap with smooth muscle tumors including a personalized approach to the treatment of LMS may prove
LMS, a panel of desmin, CD34, c-kit (CD117), and DOG-1 more beneficial than standard chemotherapy. One recent
will aid in classification. Of note, both DOG-1 and c-kit may example of this is palbociclib, a CDK4/6 inhibitor, which has
be expressed strongly in LMS, but without associated KIT been shown to be effective for LMS harboring CDKN2A
mutation [231–233]. alterations, which is encountered in LMS in up to 20% of
Rhabdomyosarcoma, particularly those with predomi- cases [255].
nantly spindled morphology, may be confused with LMS,
although the latter is overwhelmingly more common in the 1.3.3.9 Variants of Leiomyosarcoma:
uterus [234]. Morphologic features to raise concern for rhab- Diagnostic Considerations
domyosarcoma include densely eosinophilic cytoplasm and and Molecular Features
presence of cross-striations. Immunohistochemistry for
caldesmon and myogenin or other skeletal muscle-specific Epithelioid Leiomyosarcoma
marker will distinguish between smooth muscle and skeletal Epithelioid leiomyosarcoma (Fig. 1.33) shows epithelioid
muscle differentiation. differentiation in the form of round to polygonal cells with
ample eosinophilic cytoplasm (Fig. 1.34); conventional spin-
1.3.3.8 Management and Outcomes dled areas may be seen, and in these cases, a leiomyosar-
LMS is an aggressive tumor, with high rates of both local coma, mixed epithelioid and spindle types, can be considered.
recurrence and distant metastasis. The 5-year survival rate Similar to conventional spindled leiomyosarcoma, features
for LMS is 25–50%, albeit a bit higher in stage I disease indicative of malignancy include large tumor size, nuclear
[235–237]. No study has shown a definite prognostic impor-
tance of the morphologic grade once the diagnostic threshold
for leiomyosarcoma is met. Factors that affect outcome
include patient age and stage at presentation, and within the
early-stage cohort, vascular invasion, mitotic count, and
tumor circumscription have also been shown to have prog-
nostic significance in some studies [238–241]. Progesterone
receptor (PR) expression appears to be a favorable prognos-
tic indicator in low-stage LMS [242]. Similarly, androgen
receptor (AR) expression may also be a positive prognostic
indicator, particularly when co-expressed with ER and/or PR
[243]. High levels of IMP-3 expression in LMS may be asso-
ciated with worse disease-specific survival [244].
The treatment of choice for leiomyosarcoma is total
abdominal hysterectomy and bilateral salpingo-
oophorectomy. The likelihood of nodal disease is quite low
in the absence of more obvious disease, suggesting that
Fig. 1.33 Epithelioid leiomyosarcoma, gross appearance. Similar to
upfront lymph node dissection has little prognostic or other uterine sarcomas, epithelioid leiomyosarcoma has a fleshy, multi-
therapeutic role in the management of leiomyosarcoma nodular appearance with hemorrhage and necrosis
1 Uterine Mesenchymal Lesions 27
Myxoid Leiomyosarcoma
Similar to conventional LMS, myxoid LMS occurs in late
reproductive or postmenopausal women and most present
with abnormal uterine bleeding or pain and stage I disease;
however it differs from conventional LMS in its gross and
microscopic characteristics [258, 259]. Grossly, the cut sur-
face is typically gelatinous and myxoid, although it may also
have deceptively well-circumscribed borders with the sur-
rounding myometrium similar to conventional LMS. Necrosis
and hemorrhage may be seen. Histologically, the tumor cells
show diffuse or fascicular growth, but the low-power appear-
ance may be hypocellular in comparison to the surrounding
myometrium due to the abundant myxoid matrix separating
the neoplastic cells. The tumor cells are spindled to stellate
Fig. 1.35 Epithelioid leiomyosarcoma, immunohistochemistry. and have variable amounts of eosinophilic cytoplasm with
Smooth muscle actin (SMA) (shown here) is positive in epithelioid
leiomyosarcoma, as is desmin and caldesmon; however, the extent and variably sized but typically hyperchromatic nuclei (Fig. 1.36).
intensity of positivity can be variable The mitotic rate is variable and is sometimes low which may
in part be due to the abundant myxoid matrix and associated
hypocellularity. As such, the diagnostic criteria for malig-
atypia, necrosis, mitotic rate, and vascular invasion. Tumors nancy in myxoid LMS differ from conventional LMS in that
with significant cytologic atypia and a mitotic rate ≥5 per 10 the threshold for the mitotic rate is much lower at 2 per 10
high-power fields are classified as malignant [160]. high-power fields. Thus, in a tumor with moderate to marked
Epithelioid leiomyosarcomas are typically positive for cytologic atypia, the mitotic rate needs only reach 2 per 10
smooth muscle markers, including smooth muscle actin, des- high-power fields in order to diagnose the tumor as malig-
min, and h-caldesmon, although the expression can be vari- nant. In one study, all myxoid LMS had infiltrative borders,
able (Fig. 1.35). Of note, some may show focal HMB-45 most commonly in the form of a markedly irregular and
expression, and in these instances, the possibility of a angulated interface; though in a minority of cases, tongue-
PEComa should be considered [256]; other characteristic like permeative growth or only focal destructive myometrial
areas of PEComatous differentiation should be identified and invasion was seen [258]. Immunohistochemically, these
co-expression of other melanocytic markers such as melanA tumors are typically positive for smooth muscle markers
and cathepsin K (which should be diffuse in PEComa) should although the degree of expression may be less than that of
be evaluated [257]. No studies have specifically evaluated tumors of the conventional type. They typically are negative
the molecular features of epithelioid LMS. The differential for ALK and approximately half show loss of p16 and
28 B. E. Howitt and M. R. Nucci
1.3.4 S
mooth Muscle Tumors of Uncertain
Malignant Potential
Fig. 1.36 Myxoid leiomyosarcoma, microscopic appearance. Myxoid Smooth muscle tumors of uncertain malignant potential
leiomyosarcoma is composed of hyperchromatic spindled to stellate (STUMP) are not well defined secondary to poor reproduc-
tumor cells set in a myxoid stroma imparting a somewhat hypocellular
appearance. Mitotic activity is present
ibility agreement on histomorphologic classification, but
with this caveat in mind, one study found that 11% of
STUMPs harbor MED12 mutations, similar to the fre-
a bnormal p53 staining patterns which correlate with TP53 quency in LMS [128]. Likewise, approximately one third
mutations or deletions [206]. No studies to date have specifi- of STUMPs have PTEN deletions [128]. STUMPs with any
cally evaluated the molecular features of myxoid LMS. degree of myxoid features should be stained immunohisto-
Distinction of myxoid LMS from conventional LMS is chemically for Alk or examined with FISH analysis for ALK
somewhat subjective; the finding of a grossly gelatinous rearrangement, given that a recent study found that a sig-
mass or a tumor in which >50% shows a prominent myxoid nificant minority of STUMPs were reclassified as
matrix is considered sufficient in diagnosing a leiomyosar- inflammatory myofibroblastic tumor (IMT) after Alk
coma as of the myxoid rather than the conventional subtype immunohistochemical staining [260]. Our practical
[258]. The differential diagnosis also includes myxoid leio- approach to the consideration of whether a tumor should be
myoma, hydropic leiomyoma, inflammatory myofibroblas- considered of uncertain malignant potential (or of uncertain
tic tumor, and BCOR-altered high-grade sarcoma. Myxoid biologic behavior) is the following. Tumors with significant
leiomyoma is a very uncommon tumor and should only be cytologic atypia and a mitotic rate that is near the threshold
diagnosed when a myxoid smooth muscle tumor is well cir- (i.e., 8 or 9 mitoses per 10 high-power fields) should be
cumscribed, lacks cytologic atypia, and shows no or only considered of uncertain malignant potential as it seems
rare mitoses. Hydropic leiomyoma typically has a watery unlikely there is significant biologic difference in tumors
rather than gelatinous cut surface which corresponds micro- with such close mitotic rates; as a practical point, in any
scopically to edema fluid which appears “pink” rather than situation in which the mitotic rate is close to the threshold,
“bluish”; in addition, it is well circumscribed, lacks atypia, extensive sampling and exhaustive mitotic counts should be
and usually shows minimal to no mitoses. The distinction performed (every block of tissue; at least 100 high-power
from inflammatory myofibroblastic tumor is more chal- fields counted). In some cases, a proliferation marker can
lenging as both can be myxoid and have infiltrative mar- be used to identify “hot spots.” A stand-alone diagnosis of
gins, and indeed these tumors have been misclassified as “STUMP” should not be used in a biopsy/curettage as the
myxoid LMS in the past [258], most likely because they fragments may not be large enough to provide a formal
can have leiomyoma-like areas as well as these only being mitotic count; in this scenario, “atypical smooth muscle
relatively recently recognized as occurring in the gyneco- tumor representing at least STUMP” can be rendered.
logic tract. They tend to have a “fasciitis-like” appearance STUMP can also be considered in epithelioid or myxoid
being composed of cells with enlarged nuclei, vesicular tumors in which the mitotic rate is also near the threshold.
chromatin, and prominent nucleoli. This appearance in Recently, additional histologic parameters for the diagnosis
combination with the finding a lymphoplasmacytic infil- of STUMP have been proposed which include atypical
trate sprinkled throughout and aggregating at the periphery mitoses, infiltrative margins, and vascular involvement
of the mass are clues to the diagnosis. Inflammatory myofi- [261]. In this study, the risk of adverse outcome was higher
broblastic tumors can be positive for desmin and SMA but (36.4%) than in previously published reports (range
also are positive in most cases for ALK, which corresponds 7–26.7%) suggesting that use of more stringent criteria can
to rearrangements of the ALK gene. In addition, they exclude some patients from further follow-up.
1 Uterine Mesenchymal Lesions 29
1.3.5 A
typical Leiomyoma (“Leiomyoma One subtype of atypical leiomyoma that deserves specific
with Bizarre Nuclei”) mention is FH-deficient atypical leiomyoma, a finding which
is not uncommon; in one study ~38% of atypical leiomyoma
Atypical leiomyoma is characterized by significant cyto- were FH deficient [264]. Hereditary leiomyomatosis and
logic atypia (readily seen from 4× objective) (Fig. 1.37a) renal cell carcinoma syndrome (HLRCC; formerly known as
but lacks both tumor cell necrosis and significant mitotic “Reed syndrome”) is characterized by numerous uterine and/
activity [195]. Nuclear atypia in atypical leiomyoma or cutaneous leiomyomata, often presenting at very young
comprises some combination of nuclear enlargement, mul- ages, as well as renal cell carcinoma [265]. It has also been
tinucleation, hyperchromasia, coarse chromatin, or promi- suggested that HLRCC is associated with uterine LMS; how-
nent nucleoli (Fig. 1.37b). The atypia may be focal or ever, some of these “leiomyosarcomas” may actually repre-
diffuse. Though in the past these tumors have variably sent atypical leiomyoma. There is only one large study from
been named atypical leiomyoma, symplastic leiomyoma, Finland that reports LMS occurring in the setting of FH
and pleomorphic leiomyoma, it has been proposed that germline mutation [266], and this did not include pathology
these be termed “leiomyoma with bizarre nuclei,” in the review.
most recent WHO [2]; however, many of the molecular Histologically, FH-deficient atypical leiomyomata are
alterations present in these tumors overlap with those characterized by diffuse nuclear atypia (in contrast to atypi-
found in LMS. Specifically, 12% have TP53 mutations, cal leiomyomas with focal or scattered foci of severe nuclear
10% have MED12 mutations, and 24% have PTEN dele- atypia), epithelioid nuclei with very prominent nucleoli
tion [128]. Additionally, the miRNA profile of atypical (characteristically cherry red or “orangeophilic”) with peri-
leiomyoma was more similar to LMS/STUMP than to con- nucleolar clearing, intracellular and extracellular aggregates
ventional leiomyoma or cellular leiomyoma [128]. of eosinophilic material (Fig. 1.38a), as well as staghorn-
Histologically, the distinction of atypical leiomyoma from shaped intratumoral vasculature (Fig. 1.38b) [264, 267–270];
LMS is the absolute absence of tumor cell necrosis and a however, these features are not specific nor sensitive for
mitotic index no higher than ordinary leiomyomata. Bell HLRCC-associated or FH-deficient atypical leiomyomata
et al. have suggested that atypical leiomyoma have no [264, 271, 272]. Patients with HLRCC have germline muta-
more than 10 mitotic figures per 10 HPFs [195]. While one tions in the fumarate hydratase gene, FH, located on 1q42.1,
study has shown no evidence of recurrence [262], another and the tumors from these patients have accumulated a sec-
has shown that atypical leiomyomas have a low risk (<2%) ondary somatic inactivation of FH, resulting in complete loss
of extrauterine/intraabdominal recurrence [263]. of protein function and expression, which can be demon-
Therefore, careful follow-up is required or consideration strated with FH immunohistochemical staining (loss of
of hysterectomy for a diagnosis of atypical leiomyoma in staining in tumor cells) (Fig. 1.38c) [267]. Of note, loss of
which there was incomplete excision. FH can also be found in non-syndromic leiomyomata [268,
a b
Fig. 1.37 Atypical leiomyoma (“leiomyoma with bizarre nuclei”). (a) significant mitotic activity. (b) The nuclear atypia in atypical leiomy-
Atypical leiomyoma is characterized by fascicular growth of spindled oma is characterized by nuclear enlargement, hyperchromasia, and
cells with abundant eosinophilic cytoplasm and nuclear atypia that is multinucleation
readily identifiable from low power, but without tumor cell necrosis or
30 B. E. Howitt and M. R. Nucci
a c
Fig. 1.38 FH-deficient leiomyoma. (a) Histologic features of vasculature is thought to be one of the more specific histologic features
FH-deficient leiomyoma include large eosinophilic or orangeophilic of these tumors. (c) Loss of FH protein expression by immunohisto-
nucleoli and prominent perinucleolar clearing and extracellular and/or chemistry is seen in the majority. Note the vessels serving as positive
intracellular aggregates of eosinophilic material. (b) Staghorn-shaped internal controls with retained cytoplasmic expression of FH
271, 273, 274]. FH inactivation is thought to account for less FH immunohistochemistry. It is now recognized that loss of
than 2% of all uterine leiomyomas and is mutually exclusive FH by immunohistochemistry is not entirely specific for the
with both MED12 and HMGA2 alterations [153, 274]. germline FH mutation (HLRCC). Likewise, it is also possi-
Recognition of this FH-deficient leiomyomata is important ble that the mutated protein may be nonfunctional but stable
because they may indicate that the patient has and thus detectable by immunohistochemistry. In summary,
HLRCC. Clinical recognition of HLRCC may facilitate can- recognition of an atypical leiomyoma with the distinctive
cer screening, especially for an aggressive form of renal cell morphologic features suggestive of FH-deficiency should
carcinoma. Immunohistochemistry may be helpful in rein- prompt confirmation of FH loss by immunohistochemistry, if
forcing the diagnosis of an atypical leiomyoma with available. In addition, it is reasonable to recommend referral
FH-deficiency. Specifically, immunohistochemistry for to a medical geneticist in the appropriate clinical setting.
fumarate hydratase (FH) reveals staining loss in the majority
of HLRCC-associated tumors. FH staining is normally dif-
fusely cytoplasmic in a granular pattern but tumors with FH 1.4 Mixed Epithelial and Stromal Tumors
loss have essentially absent staining. Adjacent myometrium
and endothelial cells serve as useful internal controls. Thus, Overview
presence of staining is a “negative” or normal result while Mixed epithelial-stromal tumors of the uterus encountered in
lack of staining is an abnormal result, which supports con- practice include primarily adenosarcoma. Even though they
cern for the diagnosis of HLRCC in atypical leiomyomas. are classified as a mixed tumor, uterine adenosarcoma has
There is a risk of both false negative and positive results with been shown to represent a mesenchymal neoplasm as the
1 Uterine Mesenchymal Lesions 31
epithelial component does not harbor clonal somatic altera- 1.4.2.3 Gross Findings
tions [275]. Given this information, true mixed epithelial and Adenosarcomas are typically bulky, soft, polypoid masses
stromal tumors may not exist in the uterus. Adenofibroma that fill the endometrial cavity (Fig. 1.39) and may prolapse
and carcinofibroma exist in theory yet rarely (if ever) occur through the cervical os. Papillary projections on broad stalks
in practice. Carcinosarcoma, which is composed of malig- may be seen within cystic spaces. They most commonly
nant epithelial and stromal components, is not an uncom- involve the uterine fundus but may also occasionally arise in
monly encountered tumor but is now understood to represent the lower uterine segment or cervix.
a form of metaplastic epithelial malignancy rather than a true
mixed epithelial-stromal tumor and thus is discussed in detail 1.4.2.4 Microscopic Features
in Chap. 18 (Vol. 1). Histologically, adenosarcoma is characterized by phyllodi-
form architecture in the form of dilated, cleft-like glands
with intraglandular polypoid projections (Fig. 1.40). The
1.4.1 Adenofibroma glandular component is lined by benign epithelium, which
may show a range of appearances from bland to atypical,
Adenofibroma is by definition a mixed tumor with both with metaplasia frequently present, particularly squamous
benign epithelial and stromal components. Histologically it or tubal metaplasia (Fig. 1.41a, b). Additional characteris-
is characterized by phyllodes-like architecture but lacks tic histologic features include periglandular cuffing by
any evidence of periglandular cuffing and stromal cytologic hypercellular stroma, stromal mitotic activity (at least 1–2
atypia, and mitotic activity should be <2 per 10 HPFs [276]. mitoses per 10 high-power fields), and stromal atypia
Adenofibroma of the uterus is exceedingly rare, and the (Fig. 1.42a, b). Mitotic activity and stromal atypia can be
diagnosis should be made with extreme caution as many quite variable in adenosarcoma, with some cases exhibiting
(including these authors) believe these represent especially little of either, which likely leads to both its under-recogni-
low-grade forms of adenosarcoma; indeed some “adenofi- tion as well as its overdiagnosis [277, 287]. The mesenchy-
bromas” with low mitotic counts and lacking significant mal component of uterine adenosarcoma is more commonly
stromal cytologic atypia have been reported to be associ- homologous (with a stromal or fibrous appearance); how-
ated with tumor recurrence and/or metastasis [277]. In ever, heterologous differentiation, most commonly in the
practice this diagnosis should never be made in biopsy or
curettage specimens.
1.4.2.1 Definition
Adenosarcoma is a biphasic tumor composed of a benign
Mullerian glandular (epithelial) component and a malignant
stromal (mesenchymal) component.
1.4.2.5 Biomarkers
Currently there is no specific immunohistochemical bio-
marker for adenosarcoma. CD10, ER, and PR are commonly
positive, particularly in morphologically low-grade adeno-
sarcomas and those lacking sarcomatous overgrowth [22,
290–294]. WT-1 is another marker frequently positive in
adenosarcoma; however, in contrast to CD10, ER, and PR,
Fig. 1.41 Adenosarcoma, epithelial component. (a) The epithelial WT-1 is also often positive in areas of sarcomatous over-
component of adenosarcoma is typically bland but may also show some growth [294]. However, in most instances, immunohisto-
atypia. (b) Epithelial metaplasia, commonly in the form of squamous
differentiation, is a typical finding in adenosarcoma chemistry is not necessary to establish the diagnosis of
adenosarcoma, and only in occasional circumstances is
immunohistochemistry helpful in the differential diagnosis.
form of rhabdomyosarcomatous differentiation, also may Of note, though most adenosarcomas demonstrate normal
be seen, but most commonly in the setting of sarcomatous (wild-type) p53 staining in the form of patchy or weak
overgrowth. Sarcomatous overgrowth is defined as having a expression, occasional cases of adenosarcoma, usually those
pure sarcoma without any epithelial elements present in with a morphologically high-grade sarcomatous component,
1 Uterine Mesenchymal Lesions 33
may display strong diffuse or completely absent (null) abnor- diagnosis of adenosarcoma, it is our opinion that these
mal p53 expression [277, 295–297]. lesions should be classified as “endometrial polyp with
unusual features” with a comment stating that although diag-
1.4.2.6 Molecular Features nostic features of adenosarcoma are not identified, close
Few studies have examined the molecular features of adeno- clinical follow-up with sampling of any “recurrent” polyps is
sarcoma, but TP53 mutations are uncommon and when pres- recommended [287].
ent are almost always associated with sarcomatous Carcinosarcoma can be distinguished from adenosarcoma
overgrowth, a poor prognostic indicator [275, 295, 297, 298]. by the presence of a malignant epithelial component in addi-
High-level copy number gains of MYBL1 are seen in a subset tion to the malignant mesenchymal component.
of adenosarcomas, most often associated with sarcomatous Carcinosarcoma usually lacks the typical architectural fea-
overgrowth. Low-level amplification of MDM2 has also been tures of adenosarcoma, though occasionally carcinosarco-
described, seemingly unrelated to the presence of sarcoma- mas with adenosarcoma-like growth have been encountered
tous overgrowth. ATRX mutations were identified in a subset in our practice. Examining the epithelium at high magnifica-
of adenosarcoma, which may be associated with loss of tion is helpful. Additionally, a collision tumor (adenosar-
expression of ATRX by IHC. DICER1 mutations have been coma and carcinoma) could be considered.
identified in MA with rhabdomyosarcomatous differentia- Occasionally LGESS may contain benign-appearing
tion [275]. Few gene fusions have been described in MA, but endometrioid glands [30, 33]; usually the glandular elements
small numbers of cases with NCOA2/3-expressed gene are focal, but they may occasionally be more extensively
fusions have been reported [275]. Cytogenetically, one series present. Morphologic features including the lack of periglan-
evaluated the karyotypes obtained on adenosarcomas both dular stromal cuffing, subepithelial condensation, and
with and without sarcomatous overgrowth and found that characteristic leaflike glandular contours will aid in this dis-
markedly complex karyotypes were only seen in adenosar- tinction. Immunohistochemistry is of little value as both
coma with sarcomatous overgrowth and metastatic tumors LGESS and adenosarcoma are typically positive for CD10,
[299]. Additionally, abnormalities of chromosome 8 were ER, and PR. Molecular analysis may be useful as most
present in the majority of cases (71%), characterized by LGESS harbor abnormalities of JAZF1 on chromosome 7.
either gaining an extra copy of chromosome 8 or by rear- Similar to LGESS, other uterine sarcomas (typically high
rangement at 8q13 [299]. grade) may demonstrate infiltration through or around pre-
existing benign endometrial glands, which can also pose a
1.4.2.7 Differential Diagnosis diagnostic challenge. Identifying typical areas of low-grade
The differential diagnosis of adenosarcoma varies widely, adenosarcoma without sarcomatous overgrowth will aid in
with some adenosarcomas being difficult to distinguish from making the diagnosis of adenosarcoma. Extensive sampling
benign uterine polyps, uterine polyps with unusual features, of the tumor is often helpful. In the case of a high-grade sar-
and so-called adenofibroma, while at the other end of the coma with focal intratumoral benign-appearing glands, but
spectrum, clearly malignant adenosarcomas must be distin- lacking phyllodes-like or broad papillary leaflike architec-
guished from other types of uterine malignancies including ture and/or periglandular stromal cuffing, the diagnosis of
carcinosarcoma and other uterine sarcomas with a glandular adenosarcoma cannot be confirmed.
component or infiltrative growth through pre-existing glands.
At the very low-grade end of the spectrum of adenosar- 1.4.2.8 Clinical Management and Outcome
coma morphologies, it can be challenging to distinguish Patients with uterine adenosarcoma typically present with
from a benign endometrial polyp, which can exhibit unusual disease confined to the uterus, with less than 5% of cases
features including highly cellular stroma, stromal atypia, and associated with metastasis at the time of diagnosis [282].
some architectural features of adenosarcoma [287, 300]. Adenosarcoma is associated with better survival in compari-
Endometrial polyps with stromal cell atypia in the form of son to other types of uterine sarcoma, particularly carcinosar-
multinucleation, symplastic, or degenerative hyperchroma- coma and leiomyosarcoma; however, the presence of
sia can occur, similar to stromal polyps seen at other sites in sarcomatous overgrowth is associated with a poor prognosis
the female genital tract [300]; however, mitotic activity similar to other high-grade uterine malignancies [301].
should be absent in the atypical cells. In these instances, a Similarly, once there is a recurrence, the overall prognosis is
diagnosis of endometrial polyp can be made without further poor [283]. In the largest clinicopathologic series of Mullerian
qualification. However, in polyps with unusual architecture adenosarcoma, recurrence occurred in approximately one
(including subtle phyllodes-like glandular appearance and fourth of patients, most commonly in the vagina, pelvis, or
intraglandular polypoid projections of stroma), subepithelial abdomen [278]; however, smaller studies have found recur-
stromal condensation reminiscent of what is seen in adeno- rence rates as high as 45% [280, 281, 297, 302]. The 5-year
sarcoma, yet overall the features are insufficient for the survival rates for early-stage disease range from 60 to 87%,
34 B. E. Howitt and M. R. Nucci
and the variation is likely related to innate differences in study may be found in the fundus or upper endocervix and may be
cohorts, follow-up times, and size of studies. When consider- associated with subsequent development of, or co-existence
ing stage III–IV, 5-year survival rates are much lower (15– of, endometrioid intraepithelial neoplasia and low-grade
48%) [284, 303]. Histologic features that are repeatedly endometrioid adenocarcinoma.
associated with poor prognosis include the presence of sarco- Grossly, APAs may present as sessile polyps or be exophytic and
matous overgrowth and deep myometrial invasion in the pri- pedunculated; less commonly they can be endophytic extending
mary tumor [278, 283, 285]. Clinical features associated with deep into the myometrium. Microscopically, APA is composed of
adverse outcome include increased patient age (i.e., post- varying amounts of myomatous or “myofibromatous” stroma as
menopausal status) [282]. Patients with uterine adenosar- well as a glandular component; the glands appear loosely clustered
coma are treated by hysterectomy, typically with bilateral imparting a lobulated appearance from low-power microscopic
salpingo-oophorectomy. Lymphadenectomy is not indicated examination and have haphazard, irregular branching [304]. The
in the absence of specific clinical suspicion as the rate of glands display an irregular distribution within the stroma and exhibit
lymph node metastasis is extremely low. Although the epithe- mild to moderate cytologic atypia. A characteristic finding is (often
lial and stromal component of Mullerian adenosarcoma can extensive) morular metaplasia (Fig. 1.43). On occasion, the architec-
be positive for estrogen and progesterone receptor (suggest- tural glandular complexity may parallel that seen in low-grade endo-
ing tumoral hormonal sensitivity) [293], the impact of bilat- metrioid adenocarcinoma, and indeed the degree of architectural
eral oophorectomy on outcome, particularly in women who complexity in APA correlates with recurrence risk as well as myoin-
wish to maintain their fertility, is not clear. Radiotherapy or vasive carcinoma in follow-up. Immunohistochemically, the myofi-
chemotherapy is also considered in patients with inoperable bromatous stroma is usually positive for SMA and/or desmin.
or metastatic disease [283]. Radiation therapy has not been Recent studies have demonstrated abnormalities in PTEN, CTNNB1
shown to have significant benefit in patients with uterine ade- (beta-catenin), HNF-1B, mTOR, and mismatch repair proteins that
nosarcoma. Interestingly, one large study found that adjuvant are similar to those seen in endometrioid adenocarcinoma [305]. As
radiotherapy was associated with decreased overall survival a result, recent recommendations suggest treating these lesions simi-
[282]. The possible benefit of chemotherapy is unknown, as larly to endometrioid intraepithelial neoplasia/atypical hyperplasia
many of the large studies either excluded adenosarcoma or [306]. In general, hysterectomy is advised; however, conservative
had such small numbers of patients with adenosarcoma that excision has been performed with preservation of fertility [307].
no conclusion could be made although independent case The differential diagnosis of APA includes endometrioid
reports of response to various chemotherapeutic agents exist. intraepithelial neoplasia/atypical hyperplasia, myoinvasion
in well-differentiated adenocarcinoma, adenomyoma, and
adenomyomatous polyp. In curettage specimens, the diagno-
1.4.3 Carcinofibroma sis is more challenging; APAs may be confused with endo-
metrial polyps with fibrotic stroma or even endometrial
Carcinofibroma exists in theory, composed of a malignant carcinoma with myoinvasion [308]. Myoinvasion in low-
epithelial component and a benign (but neoplastic) stromal grade adenocarcinoma may be extremely difficult to distin-
component. However, in practice these are virtually never guish from APA, and unfortunately, immunohistochemistry
encountered or diagnosed.
1.4.4 Carcinosarcoma
a b
Fig. 1.45 PEComa, microscopic features. (a) Low-power magnifica- lioid and spindle cells with clear to granular cytoplasm, pleomorphism,
tion of PEComa demonstrating an ESS-like pattern of invasion and and scattered intranuclear pseudo-inclusions
lymphovascular invasion. (b) Higher-power magnificent reveals epithe-
a b
c d
Fig. 1.46 PEComa, immunohistochemistry. (a) This PEComa is predominantly epithelioid in morphology. (b) Desmin is strongly positive. (c)
HMB-45 demonstrates multifocal, patchy positivity. (d) MelanA is multifocally positive
1.5.1.5 Biomarkers factor (MiTF)) markers but are negative for PAX8 and kera-
PEComas express both smooth muscle (such as SMA, des- tins [257, 309] (Fig. 1.46a–d). Importantly, the degree of
min, and caldesmon) and melanocytic (HMB-45, MelanA, positivity for either the smooth muscle markers or melano-
CathepsinK, and microphthalmia-associated transcription cytic markers can vary greatly, and even focal positivity
1 Uterine Mesenchymal Lesions 37
should be interpreted as a positive result. PEComas harbor- Other translocations have also been recently reported in
ing TFE3 rearrangements have a slightly different morphol- small numbers in PEComa (one case each: HTR4-ST3GAL1
ogy, with typically purely epithelioid cells with cleared and RASSF1-PDZRN3) [315]. Array comparative genomic
cytoplasm and nested architecture, and have diminished hybridization studies have identified a number of recurrent
expression of SMA and desmin [311]. losses/gains, most notably loss of chromosomes 19, 16p,
17p, 1p, and 18p and gains of X, 12q, 3q, 5, and 2q [320].
1.5.1.6 Genetic Profile
Most studies interrogating the molecular alterations in 1.5.1.7 Differential Diagnosis
PEComa include tumors from various anatomic sites, Perhaps the most common, and arguably one of the most
including the uterus, so the molecular features discussed challenging, differential diagnostic considerations for
are not specific only to uterine PEComas, but rather PEComa is that of smooth muscle tumors, particularly the
PEComas of any anatomic site. The most well-character- epithelioid variants and when clear cell morphology is pres-
ized alterations are those resulting in inactivation of TSC2 ent, as they may have overlapping immunoprofiles [51, 256,
(16p13.3) or less commonly TSC1 (9q34) due to the asso- 309, 321–329]. While HMB45 may be positive in some epi-
ciation of PEC tumors (angiomyolipoma and lymphangi- thelioid smooth muscle tumors, MelanA positivity is more
oleiomyomatosis) with the genetic disease tuberous specific for PEComa and may be helpful in this differential
sclerosis complex. TSC1/2 are involved in many cell cycle diagnosis.
regulatory pathways, including the mTOR pathway. Loss Uterine tumor resembling ovarian sex cord tumor
of TSC1/2 results in increased activation of mTOR [312]; (UTROSCT) may also be considered in the differential diag-
thus many have proposed using mTOR inhibitors in these nosis with PEComa. As discussed in the UTROSCT section,
tumors. While any inactivating genetic hit in TSC1/2 (muta- both tumors are commonly positive for smooth muscle and
tion, deletion, copy number loss) may contribute to the melanocytic markers; however, sex cord markers SF-1 and
pathogenesis, not all PEComas have inactivation of TSC1/2. WT-1 are generally negative in PEComa.
As some PEComas lack inactivation of TSC2 and the resul- Occasionally, LGESS may exhibit an epithelioid cyto-
tant activation of the mTOR pathway, some have suggested morphology [27], which may raise the possibility of
that mTOR pathway activation be confirmed before treating PEComa; however, the latter has more prominent thick-
a PEComa patient with an mTOR inhibitor [310]. TFE3, a walled vessels and lacks the characteristic spiral arteriolar
member of the MIT/TFE family of transcription factors, is vessels of LGESS. Immunohistochemistry for HMB-45 and
rearranged in a small subset of PEComas and appears to be CD10 will help in this distinction; however, HMB-45 posi-
mutually exclusive with TSC1/2 altered PEComa [311, tivity alone does not equate with the diagnosis of PEComa,
313, 314]. PEComas harboring TFE3 rearrangements have as it also been reported in several cases of uterine leiomyo-
a slightly different morphology, with purely epithelioid sarcoma with both conventional and clear cell areas, leading
cells with cleared cytoplasm and nested architecture, and some to question the veracity of PEComa as an entity in the
immunohistochemically show no or minimal expression of uterus [94, 323].
SMA and desmin [311]. TFE3-rearranged PEComas lack Alveolar soft part sarcoma (ASPS) only rarely involves
TSC2 inactivation, suggesting that this subset of PEComa the gynecologic tract including the uterus [330] but may also
has an alternate pathogenesis and may not be amenable to be considered in the differential diagnosis with PEComa
mTOR inhibition therapy [315, 316]. One PEComa has given the similarity in morphologic growth patterns, as ASPS
been reported to have TFE3 amplification rather than rear- is composed of nests of epithelioid cells with abundant
rangement [317]. The TFE3 translocation partners docu- eosinophilic cytoplasm and large nuclei with prominent
mented in alveolar soft part sarcoma and Xp11 translocation nucleoli (Fig. 1.47). Immunohistochemically, ASPS is TFE3
renal cell carcinoma have not been found in PEComa to positive, reflecting the characteristic ASPSCR1-TFE3 gene
date. Recently, multiple groups have identified PSF as the fusion, but is typically negative for SMA, desmin, HMB-45,
most frequent translocation partner resulting in a SFPQ/ and MelanA.
PSF-TFE3 gene fusion and one case harbored a DVL2-
TFE3 gene fusion [315, 318, 319]. 1.5.1.8 Management and Outcomes
RAD51B (14q24) translocations (resulting in RAD51B- Histologic parameters shown to be useful in predicting malig-
RRAGB or RAD51B-OPHN1 gene fusions) have also been nant behavior include tumor size >5 cm, high-grade nuclear
identified in a small minority (8%) of uterine PEComas in atypia, infiltrative borders, tumor cell necrosis, lymphovascu-
one study [315]. In one of these cases, TSC2 and TP53 muta- lar invasion, and mitotic activity. In one study by Folpe et al.
tions were also identified, suggesting that these PEComas, that included PEComas from all anatomic sites, they con-
unlike the TFE3-associated PEComas, likely have a shared cluded that having any two of these morphologic criteria
pathogenesis with the TSC2-inactivated PEComas. qualified the PEComa to be “malignant” [331]. A subsequent
Interestingly, it was reported that the RAD51B translocation study by Schoolmeester et al. looking specifically at prognos-
PEComas were initially diagnosed as leiomyosarcoma. tic histologic parameters for uterine PEComa found that
38 B. E. Howitt and M. R. Nucci
a b
c d
Fig. 1.48 Inflammatory myofibroblastic tumor (IMT), microscopic most notably lymphocytes and plasma cells, at least focally, are identi-
appearance. (a) The characteristic morphology and one that is most fied in the vast majority. (c) The fasciitis-like pattern is also a common
readily identifiable as IMT is the myxoid pattern with abundant myxoid morphologic finding in IMT. (d) Collagenous stroma may also be
stroma imparting a hypocellular appearance. (b) Inflammatory cells, present
moderate, but occasionally may be severe [340]. Mitotic rate immunohistochemistry, in part due to the morphologic over-
can vary, from <1 to >20 per 10 HPFs in the largest series of lap and inability to reliably distinguish IMT from the much
uterine IMT [230, 340]. more common uterine smooth muscle tumors [341].
1.5.3 Adenomatoid Tumor gous component) most commonly Mullerian adenosarcoma and
carsinosarcoma and are not included in this discussion. In gen-
Adenomatoid tumors are tumors of mesothelial origin that eral, rhabdomyosarcoma is separated into three histologic
typically involve the serosal surfaces of uterus or (more com- types – embryonal, alveolar, and pleomorphic. In the female
monly) the fallopian tube. In addition to their characteristic genital tract, embryonal rhabdomyosarcoma (ERMS) is uncom-
anatomic location, the gross appearance of adenomatoid mon and most frequently occurs in the uterine cervix but has also
tumors are firm, rubbery tumors with white or gray cut sur- been reported in the uterine corpus [234, 354–359]. When pri-
faces. The border with the adjacent myometrium may not bulge mary in the cervix, it typically occurs in children or young
out or be as defined as typical leiomyomata. Microscopically, women and often presents as a polypoid, grape-like mass; in this
the mesothelial proliferation forms gland-like spaces nestled scenario it is referred to as “botryoid rhabdomyosarcoma” which
between smooth muscle fascicles (Fig. 1.50), but the mesothe- just refers to the subset of ERMS that occur at a mucosal site and
lial cells are not infrequently inconspicuous. Consequently, the grow in this distinctive fashion. ERMS of the uterine corpus
gland-like spaces may look more like venules or even adipose occurs over a wide age range, but patients are typically older than
or signet ring cells. The mesothelial origin of these cells can be those presenting with cervical primaries. Patients typically pres-
confirmed by their expression of cytokeratins, WT-1 [352], ent with vaginal bleeding or abdominal distension or pain. On
D2-40, and calretinin, but not of CD-31 or CD-34 [353]. It is gross examination, the tumor also typically projects into endo-
presumed that the mesothelial cells are neoplastic and the metrial cavity with a “grape-like” appearance and has a gray-
smooth muscle cells are hyperplastic. As adenomatoid tumors white fleshy to glistening cut surface (“botryoid subtype”) but
are usually benign, under 2 cm in size, and an incidental finding can also be an intramural mass. Necrosis and hemorrhage may
in women in the years before menopause, they have no clinical be seen. Histologically, uterine ERMS has similar features to that
impact apart from their mimicry of leiomyomata and other of soft tissue and other sites; namely, a tumor composed of small,
benign tumors. Distinction from malignant mesothelioma is round to spindle-shaped “undifferentiated” blue cells intermin-
not a common diagnostic dilemma as mesothelioma demon- gled with variable numbers of rhabdomyoblasts with brightly
strates a highly infiltrative growth pattern. eosinophilic cytoplasmic processes that may contain cross-stria-
tions (also termed “tadpole” or strap cells). The botryoid subtype
typically shows alternating hypocellular and hypercellular
1.5.4 Rhabdomyosarcoma primitive-appearing spindled cells (Fig. 1.51a) that frequently
condense underneath surface epithelium to form a “cambium”
Rhabdomyosarcomas are malignant tumors showing skeletal layer (Fig. 1.51b, c). Tumor cells may also demonstrate striation
muscle differentiation by morphology and immunohistochemi- (“strap” cells) and islands of cartilaginous differentiation may be
cal expression; of note, rhabdomyosarcomatous differentiation present (Fig. 1.50d). Immunohistochemistry shows positivity for
can be seen as a component of other tumor types (as a heterolo- skeletal muscle markers, although it may be patchy. This tumor
may be difficult to recognize, particularly in older women, and
there can be morphologic overlap with adenosarcoma as well as
a poorly sampled carcinosarcoma, both of which may com-
monly have rhabdomyosarcomatous differentiation. The only
well-characterized molecular alteration in ERMS is DICER1
mutation, which may occur as either germline or somatic inacti-
vation [359–364]. DICER1 is involved in miRNA processing
and its inactivation is likely a key step in the pathogenesis of
ERMS. No studies to date have evaluated for the presence of
DICER1 mutations in other tumors demonstrating rhabdomyo-
sarcomatous differentiation aside from the aforementioned ade-
nosarcoma [275]. ERMS in adults, even if of the botryoid
subtype, appears behave more aggressively with a worse overall
survival [355].
Alveolar and pleomorphic rhabdomyosarcoma are rare in
the uterus [355, 365–369]. The former is characterized by
rounded undifferentiated cells growing in a nested/alveolar
pattern surrounded by fibrous septa and admixed with vari-
Fig. 1.50 Adenomatoid tumor. Adenomatoid tumor is characterized by
attenuated flattened to cuboidal mesothelial cells forming cyst-like
able numbers of wreath-like multinucleate cells and rhabdo-
spaces within myometrium. In this example, the mesothelial compo- myoblasts. Most alveolar rhabdomyosarcomas express
nent is conspicuous PAX3-FKHR or PAX7-FKHR gene fusions with the latter
42 B. E. Howitt and M. R. Nucci
a b
c d
Fig. 1.51 Rhabdomyosarcoma, embryonal type. (a) Embryonal rhab- underneath nonneoplastic surface epithelium. (c) In some cases, the
domyosarcoma with alternating hypocellular and hypercellular areas, cambium layer can be very subtle, particularly in areas of hemorrhage.
with hyperchromatic, primitive-appearing spindled cells in the hyper- (d) Cartilaginous differentiation, as shown here, may be seen in embry-
cellular foci. (b) The classic histologic feature of embryonal rhabdo- onal rhabdomyosarcoma
myosarcoma is the “cambium” layer in which the tumor cells condense
possibly associated with a more favorable prognosis [370]. present at a wide age range (34–81 years). Tumors range in
Pleomorphic rhabdomyosarcoma is a high-grade sarcoma size from 3.5 to 12.5 cm and grossly appear as a well-
composed of pleomorphic cells which may show features of circumscribed firm yellow-tan nodule. Histologic appear-
ERMS or alveolar subtype elsewhere and may best be con- ance is similar to SFT described elsewhere with bland spindle
sidered “anaplastic” rhabdomyosarcoma in this context. cells arranged in a “patternless pattern” and alternating
hypo- and hypercellular areas, as well as the characteristic
staghorn or hemangiopericytoma-like thin-walled branching
1.5.5 Solitary Fibrous Tumor (SFT) vasculature. An adipocytic component or myxoid change
may be present in a minority. Also similar to SFTs reported
Solitary fibrous tumor rarely occurs in the uterus, but is elsewhere, histologic features are not entirely helpful in
worth mentioning as the clinical and imaging impression is determining malignant potential; indeed at least one case in
often that of leiomyoma, and thus pathologic recognition of the largest series that metastasized to the lung demonstrated
this entity is essential as SFT are considered tumors of uncer- low mitotic activity, nuclear atypia, or necrosis, although the
tain malignant potential, even when no worrisome histologic tumor size was 10 cm [371, 372]. Immunohistochemically,
features are present. Within the female genital tract, the uter- SFT are positive for CD34 and STAT6 but negative for the
ine corpus is the most common location [371] and patients smooth muscle markers SMA, desmin, and caldesmon.
1 Uterine Mesenchymal Lesions 43
immunohistochemical comparison of 34 cases. Mod Pathol. 36. Micci F, et al. Cytogenetic and molecular genetic analyses of
2001;14(5):465–71. endometrial stromal sarcoma: nonrandom involvement of chro-
18. Toki T, et al. CD10 is a marker for normal and neoplastic endome- mosome arms 6p and 7p and confirmation of JAZF1/JJAZ1 gene
trial stromal cells. Int J Gynecol Pathol. 2002;21(1):41–7. fusion in t(7;17). Cancer Genet Cytogenet. 2003;144(2):119–24.
19. Sumathi VP, McCluggage WG. CD10 is useful in demonstrating 37. Hrzenjak A, et al. JAZF1/JJAZ1 gene fusion in endometrial
endometrial stroma at ectopic sites and in confirming a diagnosis stromal sarcomas: molecular analysis by reverse transcriptase-
of endometriosis. J Clin Pathol. 2002;55(5):391–2. polymerase chain reaction optimized for paraffin-embedded tis-
20. Oliva E, et al. An immunohistochemical analysis of endometrial sue. J Mol Diagn. 2005;7(3):388–95.
stromal and smooth muscle tumors of the uterus: a study of 54 38. Nucci MR, et al. Molecular analysis of the JAZF1-JJAZ1 gene
cases emphasizing the importance of using a panel because of fusion by RT-PCR and fluorescence in situ hybridization in endo-
overlap in immunoreactivity for individual antibodies. Am J Surg metrial stromal neoplasms. Am J Surg Pathol. 2007;31(1):65–70.
Pathol. 2002;26(4):403–12. 39. Huang HY, Ladanyi M, Soslow RA. Molecular detection of
21. Zhu XQ, et al. Immunohistochemical markers in differential diag- JAZF1-JJAZ1 gene fusion in endometrial stromal neoplasms with
nosis of endometrial stromal sarcoma and cellular leiomyoma. classic and variant histology: evidence for genetic heterogeneity.
Gynecol Oncol. 2004;92(1):71–9. Am J Surg Pathol. 2004;28(2):224–32.
22. Abeler VM, Nenodovic M. Diagnostic immunohistochemistry 40. Oliva E, et al. High frequency of JAZF1-JJAZ1 gene fusion in
in uterine sarcomas: a study of 397 cases. Int J Gynecol Pathol. endometrial stromal tumors with smooth muscle differentiation by
2011;30(3):236–43. interphase FISH detection. Am J Surg Pathol. 2007;31(8):1277–84.
23. Parra-Herran CE, et al. Targeted development of specific bio- 41. Chiang S, et al. Frequency of known gene rearrangements in endo-
markers of endometrial stromal cell differentiation using bio- metrial stromal tumors. Am J Surg Pathol. 2011;35(9):1364–72.
informatics: the IFITM1 model. Mod Pathol. 2014;27(4): 42. Li H, et al. A neoplastic gene fusion mimics trans-splicing of
569–79. RNAs in normal human cells. Science. 2008;321(5894):1357–61.
24. Jung CK, et al. Diagnostic use of nuclear beta-catenin expression 43. Micci F, et al. Consistent rearrangement of chromosomal band 6p21
for the assessment of endometrial stromal tumors. Mod Pathol. with generation of fusion genes JAZF1/PHF1 and EPC1/PHF1 in
2008;21(6):756–63. endometrial stromal sarcoma. Cancer Res. 2006;66(1):107–12.
25. Lloreta J, Prat J. Endometrial stromal nodule with smooth and 44. Panagopoulos I, Mertens F, Griffin CA. An endometrial stromal
skeletal muscle components simulating stromal sarcoma. Int J sarcoma cell line with the JAZF1/PHF1 chimera. Cancer Genet
Gynecol Pathol. 1992;11(4):293–8. Cytogenet. 2008;185(2):74–7.
26. Clement PB. The pathology of uterine smooth muscle tumors and 45. D’Angelo E, et al. Endometrial stromal sarcomas with sex cord
mixed endometrial stromal-smooth muscle tumors: a selective differentiation are associated with PHF1 rearrangement. Am J
review with emphasis on recent advances. Int J Gynecol Pathol. Surg Pathol. 2013;37(4):514–21.
2000;19(1):39–55. 46. Ali RH, et al. Molecular characterization of a population-based
27. Oliva E, Clement PB, Young RH. Epithelioid endometrial and series of endometrial stromal sarcomas in Kuwait. Hum Pathol.
endometrioid stromal tumors: a report of four cases emphasizing 2014;45(12):2453–62.
their distinction from epithelioid smooth muscle tumors and other 47. Panagopoulos I, et al. Novel fusion of MYST/Esa1-associated
oxyphilic uterine and extrauterine tumors. Int J Gynecol Pathol. factor 6 and PHF1 in endometrial stromal sarcoma. PLoS One.
2002;21(1):48–55. 2012;7(6):e39354.
28. Oliva E, et al. Myxoid and fibrous endometrial stromal tumors 48. Micci F, et al. MEAF6/PHF1 is a recurrent gene fusion in endome-
of the uterus: a report of 10 cases. Int J Gynecol Pathol. trial stromal sarcoma. Cancer Lett. 2014;347(1):75–8.
1999;18(4):310–9. 49. Brunetti M, et al. Identification of an EPC2-PHF1 fusion tran-
29. Yilmaz A, Rush DS, Soslow RA. Endometrial stromal sarcomas script in low-grade endometrial stromal sarcoma. Oncotarget.
with unusual histologic features: a report of 24 primary and meta- 2018;9(27):19203–8.
static tumors emphasizing fibroblastic and smooth muscle differ- 50. Schoolmeester JK, et al. Analysis of MDM2 amplification in 43
entiation. Am J Surg Pathol. 2002;26(9):1142–50. endometrial stromal tumors: a potential diagnostic pitfall. Int J
30. Clement PB, Scully RE. Endometrial stromal sarcomas of the Gynecol Pathol. 2015;34(6):576–83.
uterus with extensive endometrioid glandular differentiation: a 51. Oliva E. Practical issues in uterine pathology from banal to bewil-
report of three cases that caused problems in differential diagno- dering: the remarkable spectrum of smooth muscle neoplasia.
sis. Int J Gynecol Pathol. 1992;11(3):163–73. Mod Pathol. 2016;29(Suppl 1):S104–20.
31. Kim HS, et al. Fibromyxoid variant of endometrial stromal 52. Clement PB, Young RH, Scully RE. Intravenous leiomyomatosis of
sarcoma with atypical bizarre nuclei. Int J Clin Exp Pathol. the uterus. A clinicopathological analysis of 16 cases with unusual
2015;8(3):3316–21. histologic features. Am J Surg Pathol. 1988;12(12):932–45.
32. Stewart CJ, et al. Evaluation of fluorescence in-situ hybridiza- 53. Nogales FF, et al. Uterine intravascular leiomyomatosis: an update
tion in monomorphic endometrial stromal neoplasms and their and report of seven cases. Int J Gynecol Pathol. 1987;6(4):331–9.
histological mimics: a review of 49 cases. Histopathology. 54. Mulvany NJ, et al. Intravenous leiomyomatosis of the uterus:
2014;65(4):473–82. a clinicopathologic study of 22 cases. Int J Gynecol Pathol.
33. McCluggage WG, Ganesan R, Herrington CS. Endometrial 1994;13(1):1–9.
stromal sarcomas with extensive endometrioid glandular dif- 55. Coard KC, Fletcher HM. Leiomyosarcoma of the uterus with a
ferentiation: report of a series with emphasis on the potential florid intravascular component (“intravenous leiomyosarcomato-
for misdiagnosis and discussion of the differential diagnosis. sis”). Int J Gynecol Pathol. 2002;21(2):182–5.
Histopathology. 2009;54(3):365–73. 56. Nucci MR, et al. h-Caldesmon expression effectively distin-
34. Goldblum JR, Clement PB, Hart WR. Adenomyosis with sparse guishes endometrial stromal tumors from uterine smooth muscle
glands. A potential mimic of low-grade endometrial stromal sar- tumors. Am J Surg Pathol. 2001;25(4):455–63.
coma. Am J Clin Pathol. 1995;103(2):218–23. 57. Abrams J, Talcott J, Corson JM. Pulmonary metastases in patients
35. Koontz JI, et al. Frequent fusion of the JAZF1 and JJAZ1 genes with low-grade endometrial stromal sarcoma. Clinicopathologic
in endometrial stromal tumors. Proc Natl Acad Sci U S A. findings with immunohistochemical characterization. Am J Surg
2001;98(11):6348–53. Pathol. 1989;13(2):133–40.
1 Uterine Mesenchymal Lesions 45
58. Franquemont DW, Frierson HF Jr, Mills SE. An immunohisto- 79. Lewis N, et al. ZC3H7B-BCOR high-grade endometrial stromal
chemical study of normal endometrial stroma and endometrial sarcomas: a report of 17 cases of a newly defined entity. Mod
stromal neoplasms. Evidence for smooth muscle differentiation. Pathol. 2018;31(4):674–84.
Am J Surg Pathol. 1991;15(9):861–70. 80. Hoang LN, et al. Novel high-grade endometrial stromal sarcoma:
59. Rush DS, et al. h-Caldesmon, a novel smooth muscle-specific a morphologic mimicker of myxoid leiomyosarcoma. Am J Surg
antibody, distinguishes between cellular leiomyoma and endome- Pathol. 2017;41(1):12–24.
trial stromal sarcoma. Am J Surg Pathol. 2001;25(2):253–8. 81. Mansor S, et al. ZC3H7B-BCOR-rearranged endometrial
60. Busca A, et al. IFITM1 outperforms CD10 in differentiating low- stromal sarcomas: a distinct subset merits its own classifica-
grade endometrial stromal sarcomas from smooth muscle neo- tion? Int J Gynecol Pathol. 2018; https://doi.org/10.1097/
plasms of the uterus. Int J Gynecol Pathol. 2018;37(4):372–8. PGP.0000000000000523.
61. Clement PB, Scully RE. Uterine tumors resembling ovarian sex- 82. Panagopoulos I, et al. Fusion of the ZC3H7B and BCOR genes
cord tumors. A clinicopathologic analysis of fourteen cases. Am J in endometrial stromal sarcomas carrying an X;22-translocation.
Clin Pathol. 1976;66(3):512–25. Genes Chromosomes Cancer. 2013;52(7):610–8.
62. Seagle BL, et al. Low-grade and high-grade endometrial stro- 83. Kurihara S, et al. Endometrial stromal sarcomas and related
mal sarcoma: a National Cancer Database study. Gynecol Oncol. high- grade sarcomas: immunohistochemical and molecu-
2017;146(2):254–62. lar genetic study of 31 cases. Am J Surg Pathol. 2008;32(8):
63. Gadducci A, et al. Endometrial stromal sarcoma: analysis of treat- 1228–38.
ment failures and survival. Gynecol Oncol. 1996;63(2):247–53. 84. Jakate K, et al. Endometrial sarcomas: an immunohistochemical
64. Aubry MC, et al. Endometrial stromal sarcoma metastatic to and JAZF1 re-arrangement study in low-grade and undifferenti-
the lung: a detailed analysis of 16 patients. Am J Surg Pathol. ated tumors. Mod Pathol. 2013;26(1):95–105.
2002;26(4):440–9. 85. Halbwedl I, et al. Chromosomal alterations in low-grade endome-
65. Rauh-Hain JA, et al. Endometrial stromal sarcoma: a clinicopatho- trial stromal sarcoma and undifferentiated endometrial sarcoma as
logic study of 29 patients. J Reprod Med. 2014;59(11–12):547–52. detected by comparative genomic hybridization. Gynecol Oncol.
66. Lee CH, Nucci MR. Endometrial stromal sarcoma—the new 2005;97(2):582–7.
genetic paradigm. Histopathology. 2015;67(1):1–19. 86. Sardinha R, et al. Endometrial stromal tumors: immunohisto-
67. Lee CH, et al. 14-3-3 fusion oncogenes in high-grade endome- chemical and molecular analysis of potential targets of tyrosine
trial stromal sarcoma. Proc Natl Acad Sci U S A. 2012;109(3): kinase inhibitors. Clin Sarcoma Res. 2013;3(1):3.
929–34. 87. Tanner EJ, et al. High grade undifferentiated uterine sarcoma:
68. Lee CH, et al. The clinicopathologic features of YWHAE-FAM22 surgery, treatment, and survival outcomes. Gynecol Oncol.
endometrial stromal sarcomas: a histologically high-grade and 2012;127(1):27–31.
clinically aggressive tumor. Am J Surg Pathol. 2012;36(5):641–53. 88. Hardell E, et al. Validation of a mitotic index cutoff as a prognostic
69. Sciallis AP, et al. High-grade endometrial stromal sarcomas: marker in undifferentiated uterine sarcomas. Am J Surg Pathol.
a clinicopathologic study of a group of tumors with heterog- 2017;41(9):1231–7.
enous morphologic and genetic features. Am J Surg Pathol. 89. Stemme S, Ghaderi M, Carlson JW. Diagnosis of endometrial
2014;38(9):1161–72. stromal tumors: a clinicopathologic study of 25 biopsy speci-
70. Lee CH, et al. Cyclin D1 as a diagnostic immunomarker for endo- mens with identification of problematic areas. Am J Clin Pathol.
metrial stromal sarcoma with YWHAE-FAM22 rearrangement. 2014;141(1):133–9.
Am J Surg Pathol. 2012;36(10):1562–70. 90. de Leval L, et al. Diverse phenotypic profile of uterine tumors
71. Aisagbonhi O, et al. YWHAE rearrangement in a purely conven- resembling ovarian sex cord tumors: an immunohistochemical
tional low-grade endometrial stromal sarcoma that transformed study of 12 cases. Am J Surg Pathol. 2010;34(12):1749–61.
over time to high-grade sarcoma: importance of molecular testing. 91. Bakula-Zalewska E, et al. Uterine tumors resembling ovarian sex
Int J Gynecol Pathol. 2018;37:441–7. cord tumors, a clinicopathologic study of six cases. Ann Diagn
72. Croce S, et al. YWHAE rearrangement identified by FISH and Pathol. 2014;18(6):329–32.
RT-PCR in endometrial stromal sarcomas: genetic and pathologi- 92. Irving JA, Carinelli S, Prat J. Uterine tumors resembling ovarian
cal correlations. Mod Pathol. 2013;26(10):1390–400. sex cord tumors are polyphenotypic neoplasms with true sex cord
73. Shah VI, McCluggage WG. Cyclin D1 does not distinguish differentiation. Mod Pathol. 2006;19(1):17–24.
YWHAE-NUTM2 high-grade endometrial stromal sarcoma 93. Blake EA, et al. Clinical characteristics and outcomes of uterine
from undifferentiated endometrial carcinoma. Am J Surg Pathol. tumors resembling ovarian sex-cord tumors (UTROSCT): a sys-
2015;39(5):722–4. tematic review of literature. Eur J Obstet Gynecol Reprod Biol.
74. Marino-Enriquez A, et al. BCOR internal tandem dupli- 2014;181:163–70.
cation in high-grade uterine sarcomas. Am J Surg Pathol. 94. Hurrell DP, McCluggage WG. Uterine tumour resembling
2018;42(3):335–41. ovarian sex cord tumour is an immunohistochemically poly-
75. Lee CH, et al. Frequent expression of KIT in endometrial stro- phenotypic neoplasm which exhibits coexpression of epithe-
mal sarcoma with YWHAE genetic rearrangement. Mod Pathol. lial, myoid and sex cord markers. J Clin Pathol. 2007;60(10):
2014;27(5):751–7. 1148–54.
76. Chiang S, et al. BCOR is a robust diagnostic immunohistochemi- 95. Stewart CJ, Crook M, Tan A. SF1 immunohistochemistry is
cal marker of genetically diverse high-grade endometrial stromal useful in differentiating uterine tumours resembling sex cord-
sarcoma, including tumors exhibiting variant morphology. Mod stromal tumours from potential histological mimics. Pathology.
Pathol. 2017;30(9):1251–61. 2016;48(5):434–40.
77. Isphording A, et al. YWHAE-FAM22 endometrial stromal sar- 96. Croce S, et al. Uterine tumor resembling ovarian sex cord tumor
coma: diagnosis by reverse transcription-polymerase chain reac- (UTROSCT) commonly exhibits positivity with sex cord markers
tion in formalin-fixed, paraffin-embedded tumor. Hum Pathol. FOXL2 and SF-1 but lacks FOXL2 and DICER1 mutations. Int J
2013;44(5):837–43. Gynecol Pathol. 2016;35(4):301–8.
78. Hemming ML, et al. YWHAE-rearranged high-grade endometrial 97. Baker RJ, et al. Inhibin and CD99 (MIC2) expression in uterine
stromal sarcoma: two-center case series and response to chemo- stromal neoplasms with sex-cord-like elements. Hum Pathol.
therapy. Gynecol Oncol. 2017;145(3):531–5. 1999;30(6):671–9.
46 B. E. Howitt and M. R. Nucci
98. McCluggage WG. Uterine tumours resembling ovarian sex cord 122. Pandis N, et al. Chromosome analysis of 96 uterine leiomyomas.
tumours: immunohistochemical evidence for true sex cord differ- Cancer Genet Cytogenet. 1991;55(1):11–8.
entiation. Histopathology. 1999;34(4):375–6. 123. Rein MS, et al. Cytogenetic abnormalities in uterine leiomyomata.
99. Krishnamurthy S, et al. Uterine tumors resembling ovarian sex- Obstet Gynecol. 1991;77(6):923–6.
cord tumors have an immunophenotype consistent with true sex- 124. Fletcher JA, et al. Chromosome aberrations in uterine smooth
cord differentiation. Am J Surg Pathol. 1998;22(9):1078–82. muscle tumors: potential diagnostic relevance of cytogenetic
100. Mohammadizadeh F, et al. Extensive overgrowth of sex cord-like instability. Cancer Res. 1990;50(13):4092–7.
differentiation in uterine mullerian adenosarcoma: a rare and chal- 125. Makinen N, et al. MED12, the mediator complex subunit 12 gene,
lenging entity. Int J Gynecol Pathol. 2016;35(2):153–61. is mutated at high frequency in uterine leiomyomas. Science.
101. Staats PN, et al. Uterine tumors resembling ovarian sex cord 2011;334(6053):252–5.
tumors (UTROSCT) lack the JAZF1-JJAZ1 translocation fre- 126. Schwetye KE, Pfeifer JD, Duncavage EJ. MED12 exon 2 muta-
quently seen in endometrial stromal tumors. Am J Surg Pathol. tions in uterine and extrauterine smooth muscle tumors. Hum
2009;33(8):1206–12. Pathol. 2014;45(1):65–70.
102. Nucci MRSJ, Sukov W, Oliva E. Uterine tumors resembling ovar- 127. Bertsch E, et al. MED12 and HMGA2 mutations: two indepen-
ian sex cord tumor (UTROSCT) lack rearrangement of PHF1 by dent genetic events in uterine leiomyoma and leiomyosarcoma.
FISH. Mod Pathol. 2014;27:298A. Mod Pathol. 2014;27(8):1144–53.
103. Chiang S, et al. FOXL2 mutation is absent in uterine tumors 128. Zhang Q, et al. Molecular analyses of 6 different types of uterine
resembling ovarian sex cord tumors. Am J Surg Pathol. smooth muscle tumors: emphasis in atypical leiomyoma. Cancer.
2015;39(5):618–23. 2014;120(20):3165–77.
104. Wang J, et al. Uterine tumor resembling ovarian sex cord 129. Markowski DN, et al. MED12 mutations occurring in benign
tumor: report of a case with t(X;6)(p22.3;q23.1) and t(4,;18) and malignant mammalian smooth muscle tumors. Genes
(q21.1;q21.3). Diagn Mol Pathol. 2003;12(3):174–80. Chromosomes Cancer. 2013;52(3):297–304.
105. Liu CY, et al. Clinical experience of uterine tumors resembling 130. Matsubara A, et al. Prevalence of MED12 mutations in uter-
ovarian sex cord tumors: a clinicopathological analysis of 6 cases. ine and extrauterine smooth muscle tumours. Histopathology.
Int J Clin Exp Pathol. 2015;8(4):4158–64. 2013;62(4):657–61.
106. Watrowski R, et al. Hysteroscopic treatment of uterine tumor 131. Mehine M, et al. Characterization of uterine leiomyomas by
resembling ovarian sex cord-like tumor (UTROSCT). Gynecol whole-genome sequencing. N Engl J Med. 2013;369(1):43–53.
Endocrinol. 2015;31(11):856–9. 132. Nilbert M, et al. Characteristic chromosome abnormalities,
107. Marshall LM, et al. Variation in the incidence of uterine leio- including rearrangements of 6p, del(7q), +12, and t(12;14), in 44
myoma among premenopausal women by age and race. Obstet uterine leiomyomas. Hum Genet. 1990;85(6):605–11.
Gynecol. 1997;90(6):967–73. 133. Nibert M, Heim S. Uterine leiomyoma cytogenetics. Genes
108. Stewart EA. Uterine fibroids. Lancet. 2001;357(9252):293–8. Chromosomes Cancer. 1990;2(1):3–13.
109. Buttram VC Jr. Uterine leiomyomata—aetiology, symptomatol- 134. Levy B, Mukherjee T, Hirschhorn K. Molecular cytogenetic
ogy and management. Prog Clin Biol Res. 1986;225:275–96. analysis of uterine leiomyoma and leiomyosarcoma by com-
110. Stewart EA, Nowak RA. Leiomyoma-related bleeding: a classic parative genomic hybridization. Cancer Genet Cytogenet.
hypothesis updated for the molecular era. Hum Reprod Update. 2000;121(1):1–8.
1996;2(4):295–306. 135. Xing YP, Powell WL, Morton CC. The del(7q) subgroup in uter-
111. Stovall DW, et al. Uterine leiomyomas reduce the efficacy of ine leiomyomata: genetic and biologic characteristics. Further evi-
assisted reproduction cycles: results of a matched follow-up study. dence for the secondary nature of cytogenetic abnormalities in the
Hum Reprod. 1998;13(1):192–7. pathobiology of uterine leiomyomata. Cancer Genet Cytogenet.
112. Bourlev V, et al. Different proliferative and apoptotic activity in 1997;98(1):69–74.
peripheral versus central parts of human uterine leiomyomas. 136. Ozisik YY, et al. Deletion 7q22 in uterine leiomyoma. A cytoge-
Gynecol Obstet Investig. 2003;55(4):199–204. netic review. Cancer Genet Cytogenet. 1993;71(1):1–6.
113. Kawaguchi K, et al. Mitotic activity in uterine leiomyomas during 137. Sargent MS, et al. Translocations in 7q22 define a critical region in
the menstrual cycle. Am J Obstet Gynecol. 1989;160(3):637–41. uterine leiomyomata. Cancer Genet Cytogenet. 1994;77(1):65–8.
114. Maluf HM, Gersell DJ. Uterine leiomyomas with high content of 138. Lepine LA, et al. Hysterectomy surveillance—United States,
mast cells. Arch Pathol Lab Med. 1994;118(7):712–4. 1980–1993. MMWR CDC Surveill Summ. 1997;46(4):1–15.
115. Orii A, et al. Mast cells in smooth muscle tumors of the uterus. Int 139. Practice Committee of the American Society for Reproductive
J Gynecol Pathol. 1998;17(4):336–42. Medicine. Electronic address, A.a.o. and M. Practice Committee
116. Ferry JA, Harris NL, Scully RE. Uterine leiomyomas with lym- of the American Society for Reproductive. Removal of myomas in
phoid infiltration simulating lymphoma. A report of seven cases. asymptomatic patients to improve fertility and/or reduce miscar-
Int J Gynecol Pathol. 1989;8(3):263–70. riage rate: a guideline. Fertil Steril. 2017;108(3):416–25.
117. Chuang SS, et al. Uterine leiomyoma with massive lymphocytic 140. Einstein MH, et al. Management of uterine malignancy found
infiltration simulating malignant lymphoma. A case report with incidentally after supracervical hysterectomy or uterine mor-
immunohistochemical study showing that the infiltrating lympho- cellation for presumed benign disease. Int J Gynecol Cancer.
cytes are cytotoxic T cells. Pathol Res Pract. 2001;197(2):135–8. 2008;18(5):1065–70.
118. Fonseca MCM, et al. Uterine artery embolization and surgical 141. Seidman MA, et al. Peritoneal dissemination complicating
methods for the treatment of symptomatic uterine leiomyomas: a morcellation of uterine mesenchymal neoplasms. PLoS One.
systemic review and meta-analysis followed by indirect treatment 2012;7(11):e50058.
comparison. Clin Ther. 2017;39(7):1438–1455.e2. 142. You JH, Sahota DS, Yuen PM. Uterine artery embolization, hys-
119. Mark J, et al. Chromosomal patterns in human benign uterine leio- terectomy, or myomectomy for symptomatic uterine fibroids: a
myomas. Cancer Genet Cytogenet. 1990;44(1):1–13. cost-utility analysis. Fertil Steril. 2009;91(2):580–8.
120. Meloni AM, et al. Uterine leiomyomas: cytogenetic and histologic 143. de Blok S, et al. Fatal sepsis after uterine artery embolization with
profile. Obstet Gynecol. 1992;80(2):209–17. microspheres. J Vasc Interv Radiol. 2003;14(6):779–83.
121. Nilbert M, et al. Karyotypic rearrangements in 20 uterine leio- 144. Vashisht A, et al. Fatal septicaemia after fibroid embolisation.
myomas. Cytogenet Cell Genet. 1988;49(4):300–4. Lancet. 1999;354(9175):307–8.
1 Uterine Mesenchymal Lesions 47
145. Dover RW, Ferrier AJ, Torode HW. Sarcomas and the conservative 169. Hu J, Surti U, Tobon H. Cytogenetic analysis of a uterine lipoleio-
management of uterine fibroids: a cause for concern? Aust N Z J myoma. Cancer Genet Cytogenet. 1992;62(2):200–2.
Obstet Gynaecol. 2000;40(3):308–12. 170. McDonald AG, et al. Liposarcoma arising in uterine lipoleiomy-
146. Vilos GA, et al. The management of uterine leiomyomas. J Obstet oma: a report of 3 cases and review of the literature. Am J Surg
Gynaecol Can. 2015;37(2):157–78. Pathol. 2011;35(2):221–7.
147. Singh SS, et al. The past, present, and future of selective proges- 171. Gupta M, et al. Angioleiomyoma of uterus: a clinicopathologic
terone receptor modulators in the management of uterine fibroids. study of 6 cases. Int J Surg Pathol. 2018;26(1):18–23.
Am J Obstet Gynecol. 2018;218(6):563–72. e1 172. Matsuyama A, Hisaoka M, Hashimoto H. Angioleiomyoma: a
148. Tiltman AJ. The effect of progestins on the mitotic activity of uter- clinicopathologic and immunohistochemical reappraisal with spe-
ine fibromyomas. Int J Gynecol Pathol. 1985;4(2):89–96. cial reference to the correlation with myopericytoma. Hum Pathol.
149. Prayson RA, Hart WR. Mitotically active leiomyomas of the 2007;38(4):645–51.
uterus. Am J Clin Pathol. 1992;97(1):14–20. 173. Mulvany NJ, Ostor AG, Ross I. Diffuse leiomyomatosis of the
150. O'Connor DM, Norris HJ. Mitotically active leiomyomas of the uterus. Histopathology. 1995;27(2):175–9.
uterus. Hum Pathol. 1990;21(2):223–7. 174. Baschinsky DY, et al. Diffuse leiomyomatosis of the uterus: a case
151. Dgani R, et al. Clinical-pathological study of uterine leiomyo- report with clonality analysis. Hum Pathol. 2000;31(11):1429–32.
mas with high mitotic activity. Acta Obstet Gynecol Scand. 175. Ordulu Z, et al. Intravenous leiomyomatosis: an unusual inter-
1998;77(1):74–7. mediate between benign and malignant uterine smooth muscle
152. Perrone T, Dehner LP. Prognostically favorable “mitotically tumors. Mod Pathol. 2016;29(5):500–10.
active” smooth-muscle tumors of the uterus. A clinicopathologic 176. Norris HJ, Parmley T. Mesenchymal tumors of the uterus.
study of ten cases. Am J Surg Pathol. 1988;12(1):1–8. V. Intravenous leiomyomatosis. A clinical and pathologic study of
153. Makinen N, et al. Characterization of MED12, HMGA2, and FH 14 cases. Cancer. 1975;36(6):2164–78.
alterations reveals molecular variability in uterine smooth muscle 177. Kir G, et al. Estrogen and progesterone expression of vessel walls
tumors. Mol Cancer. 2017;16(1):101. with intravascular leiomyomatosis; discussion of histogenesis.
154. Makinen N, et al. MED12 exon 2 mutations in histopatho- Eur J Gynaecol Oncol. 2004;25(3):362–6.
logical uterine leiomyoma variants. Eur J Hum Genet. 178. Lam PM, et al. Intravenous leiomyomatosis with atypical histologic
2013;21(11):1300–3. features: a case report. Int J Gynecol Cancer. 2003;13(1):83–7.
155. Hodge JC, et al. Uterine cellular leiomyomata with chromosome 179. Buza N, et al. Recurrent chromosomal aberrations in intravenous
1p deletions represent a distinct entity. Am J Obstet Gynecol. leiomyomatosis of the uterus: high-resolution array comparative
2014;210(6):572.e1–7. genomic hybridization study. Hum Pathol. 2014;45(9):1885–92.
156. Ozisik YY, et al. Involvement of 10q22 in leiomyoma. Cancer 180. Dal Cin P, et al. Intravenous leiomyomatosis is characterized
Genet Cytogenet. 1993;69(2):132–5. by a der(14)t(12;14)(q15;q24). Genes Chromosomes Cancer.
157. Hodge JC, et al. Molecular and cytogenetic characteriza- 2003;36(2):205–6.
tion of plexiform leiomyomata provide further evidence for 181. Wu RC, et al. Massively parallel sequencing and genome-wide
genetic heterogeneity underlying uterine fibroids. Am J Pathol. copy number analysis revealed a clonal relationship in benign
2008;172(5):1403–10. metastasizing leiomyoma. Oncotarget. 2017;8(29):47547–54.
158. Seidman JD, Thomas RM. Multiple plexiform tumorlets of the 182. Patton KT, et al. Benign metastasizing leiomyoma: clonality,
uterus. Arch Pathol Lab Med. 1993;117(12):1255–6. telomere length and clinicopathologic analysis. Mod Pathol.
159. Kaminski PF, Tavassoli FA. Plexiform tumorlet: a clinical and 2006;19(1):130–40.
pathologic study of 15 cases with ultrastructural observations. Int 183. Tietze L, et al. Benign metastasizing leiomyoma: a cytoge-
J Gynecol Pathol. 1984;3(2):124–34. netically balanced but clonal disease. Hum Pathol. 2000;31(1):
160. Prayson RA, Goldblum JR, Hart WR. Epithelioid smooth-muscle 126–8.
tumors of the uterus: a clinicopathologic study of 18 patients. Am 184. Nucci MR, et al. Distinctive cytogenetic profile in benign metas-
J Surg Pathol. 1997;21(4):383–91. tasizing leiomyoma: pathogenetic implications. Am J Surg Pathol.
161. Jones MW, Norris HJ. Clinicopathologic study of 28 uter- 2007;31(5):737–43.
ine leiomyosarcomas with metastasis. Int J Gynecol Pathol. 185. Lin J, Song X, Liu C. Pelvic intravascular leiomyomatosis asso-
1995;14(3):243–9. ciated with benign pulmonary metastasizing leiomyoma: clinico-
162. Karaiskos C, et al. Cytogenetic findings in uterine epithelioid leio- pathologic, clonality, and copy number variance analysis. Int J
myomas. Cancer Genet Cytogenet. 1995;80(2):103–6. Gynecol Pathol. 2014;33(2):140–5.
163. Haimes JD, et al. Uterine inflammatory myofibroblastic tumors 186. Canzonieri V, et al. Leiomyomatosis with vascular invasion. A
frequently harbor ALK fusions with IGFBP5 and THBS1. Am J unified pathogenesis regarding leiomyoma with vascular micro-
Surg Pathol. 2017;41(6):773–80. invasion, benign metastasizing leiomyoma and intravenous leio-
164. Roth LM, Reed RJ, Sternberg WH. Cotyledonoid dissecting leio- myomatosis. Virchows Arch. 1994;425(5):541–5.
myoma of the uterus. The Sternberg tumor. Am J Surg Pathol. 187. Quade BJ, et al. Disseminated peritoneal leiomyomatosis.
1996;20(12):1455–61. Clonality analysis by X chromosome inactivation and cytoge-
165. Roth LM, Reed RJ. Cotyledonoid leiomyoma of the uterus: report netics of a clinically benign smooth muscle proliferation. Am J
of a case. Int J Gynecol Pathol. 2000;19(3):272–5. Pathol. 1997;150(6):2153–66.
166. Smith CC, et al. Cotyledonoid dissecting leiomyoma of the uterus: 188. Bisceglia M, et al. Selected case from the Arkadi M. Rywlin
a review of clinical, pathological, and radiological features. Int J International Pathology Slide Series: Leiomyomatosis peritonea-
Surg Pathol. 2012;20(4):330–41. lis disseminata: report of 3 cases with extensive review of the lit-
167. Shintaku M. Lipoleiomyomatous tumors of the uterus: a hetero- erature. Adv Anat Pathol. 2014;21(3):201–15.
geneous group? Histopathological study of five cases. Pathol Int. 189. Hardman WJ 3rd, Majmudar B. Leiomyomatosis peritonealis dis-
1996;46(7):498–502. seminata: clinicopathologic analysis of five cases. South Med J.
168. Pedeutour F, et al. Dysregulation of HMGIC in a uterine lipo- 1996;89(3):291–4.
leiomyoma with a complex rearrangement including chromo- 190. Al-Talib A, Tulandi T. Pathophysiology and possible iatrogenic
somes 7, 12, and 14. Genes Chromosomes Cancer. 2000;27(2): cause of leiomyomatosis peritonealis disseminata. Gynecol Obstet
209–15. Investig. 2010;69(4):239–44.
48 B. E. Howitt and M. R. Nucci
191. Ordulu Z, et al. Disseminated peritoneal leiomyomatosis after 212. Cuppens T, et al. Integrated genome analysis of uterine leiomyo-
laparoscopic supracervical hysterectomy with characteristic sarcoma to identify novel driver genes and targetable pathways.
molecular cytogenetic findings of uterine leiomyoma. Genes Int J Cancer. 2018;142(6):1230–43.
Chromosomes Cancer. 2010;49(12):1152–60. 213. Chudasama P, et al. Integrative genomic and transcriptomic analy-
192. Giuntoli RL 2nd, et al. Retrospective review of 208 patients sis of leiomyosarcoma. Nat Commun. 2018;9(1):144.
with leiomyosarcoma of the uterus: prognostic indicators, sur- 214. Wang Z, et al. Survival of patients with metastatic leiomyosar-
gical management, and adjuvant therapy. Gynecol Oncol. coma: the MD Anderson Clinical Center for targeted therapy
2003;89(3):460–9. experience. Cancer Med. 2016;5(12):3437–44.
193. Mittal K, Joutovsky A. Areas with benign morphologic and immu- 215. Makinen N, et al. Exome sequencing of uterine leiomyosarcomas
nohistochemical features are associated with some uterine leio- identifies frequent mutations in TP53, ATRX, and MED12. PLoS
myosarcomas. Gynecol Oncol. 2007;104(2):362–5. Genet. 2016;12(2):e1005850.
194. Mittal KR, et al. Molecular and immunohistochemical evidence 216. Agaram NP, et al. Targeted exome sequencing profiles genetic
for the origin of uterine leiomyosarcomas from associated leio- alterations in leiomyosarcoma. Genes Chromosomes Cancer.
myoma and symplastic leiomyoma-like areas. Mod Pathol. 2016;55(2):124–30.
2009;22(10):1303–11. 217. Yang CY, et al. Targeted next-generation sequencing of cancer
195. Bell SW, Kempson RL, Hendrickson MR. Problematic uterine genes identified frequent TP53 and ATRX mutations in leiomyo-
smooth muscle neoplasms. A clinicopathologic study of 213 sarcoma. Am J Transl Res. 2015;7(10):2072–81.
cases. Am J Surg Pathol. 1994;18(6):535–58. 218. Ravegnini G, et al. MED12 mutations in leiomyosarcoma and
196. Pautier P, et al. Analysis of clinicopathologic prognostic factors extrauterine leiomyoma. Mod Pathol. 2013;26(5):743–9.
for 157 uterine sarcomas and evaluation of a grading score vali- 219. Liau JY, et al. Leiomyosarcoma with alternative lengthening of
dated for soft tissue sarcoma. Cancer. 2000;88(6):1425–31. telomeres is associated with aggressive histologic features, loss of
197. Iwata J, Fletcher CD. Immunohistochemical detection of cyto- ATRX expression, and poor clinical outcome. Am J Surg Pathol.
keratin and epithelial membrane antigen in leiomyosarcoma: a 2015;39(2):236–44.
systematic study of 100 cases. Pathol Int. 2000;50(1):7–14. 220. de Graaff MA, et al. Mediator complex subunit 12 exon 2 muta-
198. Brown DC, et al. Cytokeratin expression in smooth muscle and tion analysis in different subtypes of smooth muscle tumors con-
smooth muscle tumours. Histopathology. 1987;11(5):477–86. firms genetic heterogeneity. Hum Pathol. 2013;44(8):1597–604.
199. Gannon BR, Manduch M, Childs TJ. Differential immunoreac- 221. Perot G, et al. MED12 alterations in both human benign and malig-
tivity of p16 in leiomyosarcomas and leiomyoma variants. Int J nant uterine soft tissue tumors. PLoS One. 2012;7(6):e40015.
Gynecol Pathol. 2008;27(1):68–73. 222. Kampjarvi K, et al. Somatic MED12 mutations in uterine leiomyo-
200. Hakverdi S, et al. Immunohistochemical analysis of p16 expres- sarcoma and colorectal cancer. Br J Cancer. 2012;107(10):1761–5.
sion in uterine smooth muscle tumors. Eur J Gynaecol Oncol. 223. de Graaff MA, et al. A translocation t(6;14) in two cases of
2011;32(5):513–5. leiomyosarcoma: molecular cytogenetic and array-based com-
201. Bodner-Adler B, et al. Expression of p16 protein in patients with parative genomic hybridization characterization. Cancer Genet.
uterine smooth muscle tumors: an immunohistochemical analysis. 2015;208(11):537–44.
Gynecol Oncol. 2005;96(1):62–6. 224. Liau JY, et al. Comprehensive screening of alternative lengthening
202. Chen L, Yang B. Immunohistochemical analysis of p16, p53, and of telomeres phenotype and loss of ATRX expression in sarcomas.
Ki-67 expression in uterine smooth muscle tumors. Int J Gynecol Mod Pathol. 2015;28(12):1545–54.
Pathol. 2008;27(3):326–32. 225. Slatter TL, et al. Loss of ATRX and DAXX expression identifies
203. Lee CH, et al. A panel of antibodies to determine site of ori- poor prognosis for smooth muscle tumours of uncertain malignant
gin and malignancy in smooth muscle tumors. Mod Pathol. potential and early stage uterine leiomyosarcoma. J Pathol Clin
2009;22(12):1519–31. Res. 2015;1(2):95–105.
204. Liang Y, et al. Diagnostic value of progesterone receptor, p16, p53 226. Zang Y, et al. Identification of key genes and pathways in uter-
and pHH3 expression in uterine atypical leiomyoma. Int J Clin ine leiomyosarcoma through bioinformatics analysis. Oncol Lett.
Exp Pathol. 2015;8(6):7196–202. 2018;15(6):9361–8.
205. O'Neill CJ, et al. Uterine leiomyosarcomas are characterized by 227. An Y, et al. Distinct molecular subtypes of uterine leiomyosar-
high p16, p53 and MIB1 expression in comparison with usual coma respond differently to chemotherapy treatment. BMC
leiomyomas, leiomyoma variants and smooth muscle tumours of Cancer. 2017;17(1):639.
uncertain malignant potential. Histopathology. 2007;50(7):851–8. 228. Lim D, et al. Interobserver variability in the interpretation of
206. Schaefer IM, et al. Abnormal p53 and p16 staining patterns distin- tumor cell necrosis in uterine leiomyosarcoma. Am J Surg Pathol.
guish uterine leiomyosarcoma from inflammatory myofibroblastic 2013;37(5):650–8.
tumour. Histopathology. 2017;70(7):1138–46. 229. Zhang Q, et al. The selected biomarker analysis in 5 types of uter-
207. Allen MM, et al. An immunohistochemical analysis of stathmin 1 ine smooth muscle tumors. Hum Pathol. 2018;76:17–27.
expression in uterine smooth muscle tumors: differential expres- 230. Parra-Herran C, et al. Inflammatory myofibroblastic tumor of
sion in leiomyosarcomas and leiomyomas. Int J Clin Exp Pathol. the uterus: clinical and pathologic review of 10 cases includ-
2015;8(3):2795–801. ing a subset with aggressive clinical course. Am J Surg Pathol.
208. Kefeli M, et al. Fascin expression in uterine smooth muscle 2015;39(2):157–68.
tumors. Int J Gynecol Pathol. 2009;28(4):328–33. 231. Raspollini MR, et al. c-Kit expression in patients with uterine leio-
209. Cornejo K, Shi M, Jiang Z. Oncofetal protein IMP3: a use- myosarcomas: a potential alternative therapeutic treatment. Clin
ful diagnostic biomarker for leiomyosarcoma. Hum Pathol. Cancer Res. 2004;10(10):3500–3.
2012;43(10):1567–72. 232. Wang L, et al. The proto-oncogene c-kit is expressed in leiomyo-
210. Quade BJ, et al. Frequent loss of heterozygosity for chromosome sarcomas of the uterus. Gynecol Oncol. 2003;90(2):402–6.
10 in uterine leiomyosarcoma in contrast to leiomyoma. Am J 233. Sah SP, McCluggage WG. DOG1 immunoreactivity in uterine
Pathol. 1999;154(3):945–50. leiomyosarcomas. J Clin Pathol. 2013;66(1):40–3.
211. Hu J, et al. Genomic alterations in uterine leiomyosarcomas: 234. McCluggage WG, et al. Rhabdomyosarcoma of the uterus: report
potential markers for clinical diagnosis and prognosis. Genes of two cases, including one of the spindle cell variant. Int J
Chromosomes Cancer. 2001;31(2):117–24. Gynecol Cancer. 2002;12(1):128–32.
1 Uterine Mesenchymal Lesions 49
235. Lusby K, et al. Uterine leiomyosarcoma management, outcome, 256. Silva EG, et al. Uterine epithelioid leiomyosarcomas with clear
and associated molecular biomarkers: a single institution’s experi- cells: reactivity with HMB-45 and the concept of PEComa. Am J
ence. Ann Surg Oncol. 2013;20(7):2364–72. Surg Pathol. 2004;28(2):244–9.
236. Garcia C, et al. Clinical outcomes and prognostic markers in uter- 257. Schoolmeester JK, et al. Perivascular epithelioid cell neoplasm
ine leiomyosarcoma: a population-based cohort. Int J Gynecol (PEComa) of the gynecologic tract: clinicopathologic and immu-
Cancer. 2015;25(4):622–8. nohistochemical characterization of 16 cases. Am J Surg Pathol.
237. Pellanda AF, et al. Outcome and prognostic factors in 110 con- 2014;38(2):176–88.
secutive patients with primary uterine leiomyosarcoma: a 258. Parra-Herran C, et al. Myxoid leiomyosarcoma of the uterus: a
Rare Cancer Network study. Chin J Cancer Res. 2017;29(6): clinicopathologic analysis of 30 cases and review of the literature
521–32. with reappraisal of its distinction from other uterine myxoid mes-
238. Pelmus M, et al. Prognostic factors in early-stage leiomyosarcoma enchymal neoplasms. Am J Surg Pathol. 2016;40(3):285–301.
of the uterus. Int J Gynecol Cancer. 2009;19(3):385–90. 259. Burch DM, Tavassoli FA. Myxoid leiomyosarcoma of the uterus.
239. Mayerhofer K, et al. Leiomyosarcoma of the uterus: a clini- Histopathology. 2011;59(6):1144–55.
copathologic multicenter study of 71 cases. Gynecol Oncol. 260. Devereaux KA, Kunder CA, Longacre TA. ALK-rearranged
1999;74(2):196–201. tumors are highly enriched in the STUMP subcategory of uter-
240. Bodner K, et al. Evaluating prognostic parameters in women with ine tumors. Am J Surg Pathol. 2018; https://doi.org/10.1097/
uterine leiomyosarcoma. A clinicopathologic study. J Reprod PAS.0000000000001083.
Med. 2003;48(2):95–100. 261. Gupta M, et al. Predictors of adverse outcome in uterine smooth
241. Kapp DS, Shin JY, Chan JK. Prognostic factors and survival muscle tumours of uncertain malignant potential (STUMP): a
in 1396 patients with uterine leiomyosarcomas: emphasis clinicopathological analysis of 22 cases with a proposal for the
on impact of lymphadenectomy and oophorectomy. Cancer. inclusion of additional histological parameters. Histopathology.
2008;112(4):820–30. 2018;73(2):284–98.
242. Davidson B, et al. Progesterone receptor expression is an indepen- 262. Croce S, Young RH, Oliva E. Uterine leiomyomas with bizarre
dent prognosticator in FIGO stage I uterine leiomyosarcoma. Am nuclei: a clinicopathologic study of 59 cases. Am J Surg Pathol.
J Clin Pathol. 2016;145(4):449–58. 2014;38(10):1330–9.
243. Baek MH, et al. Androgen receptor as a prognostic biomarker and 263. Ly A, et al. Atypical leiomyomas of the uterus: a clinicopathologic
therapeutic target in uterine leiomyosarcoma. J Gynecol Oncol. study of 51 cases. Am J Surg Pathol. 2013;37(5):643–9.
2018;29(3):e30. 264. Miettinen M, et al. Fumarase-deficient uterine leiomyomas: an
244. Yasutake N, et al. Insulin-like growth factor II messenger RNA- immunohistochemical, molecular genetic, and clinicopathologic
binding protein-3 is an independent prognostic factor in uterine study of 86 cases. Am J Surg Pathol. 2016;40(12):1661–9.
leiomyosarcoma. Histopathology. 2018;72(5):739–48. 265. Tomlinson IP, et al. Germline mutations in FH predispose to domi-
245. Goff BA, et al. Uterine leiomyosarcoma and endometrial stromal nantly inherited uterine fibroids, skin leiomyomata and papillary
sarcoma: lymph node metastases and sites of recurrence. Gynecol renal cell cancer. Nat Genet. 2002;30(4):406–10.
Oncol. 1993;50(1):105–9. 266. Lehtonen HJ, et al. Increased risk of cancer in patients with
246. Leitao MM, et al. Incidence of lymph node and ovarian fumarate hydratase germline mutation. J Med Genet. 2006;43(6):
metastases in leiomyosarcoma of the uterus. Gynecol Oncol. 523–6.
2003;91(1):209–12. 267. Joseph NM, et al. Morphology and immunohistochemistry for
247. Seagle BL, et al. Prognosis and treatment of uterine leiomyo- 2SC and FH aid in detection of fumarate hydratase gene aber-
sarcoma: a National Cancer Database study. Gynecol Oncol. rations in uterine leiomyomas from young patients. Am J Surg
2017;145(1):61–70. Pathol. 2015;39(11):1529–39.
248. George S, et al. Retrospective cohort study evaluating the impact 268. Bennett JA, et al. Leiomyoma with bizarre nuclei: a morphologi-
of intraperitoneal morcellation on outcomes of localized uterine cal, immunohistochemical and molecular analysis of 31 cases.
leiomyosarcoma. Cancer. 2014;120(20):3154–8. Mod Pathol. 2017;30(10):1476–88.
249. Thanopoulou E, et al. Treatment of hormone positive uterine 269. Reyes C, et al. Uterine smooth muscle tumors with features sug-
leiomyosarcoma with aromatase inhibitors. Clin Sarcoma Res. gesting fumarate hydratase aberration: detailed morphologic
2014;4:5. analysis and correlation with S-(2-succino)-cysteine immunohis-
250. George S, et al. Phase 2 trial of aromatase inhibition with letrozole tochemistry. Mod Pathol. 2014;27(7):1020–7.
in patients with uterine leiomyosarcomas expressing estrogen and/ 270. Sanz-Ortega J, et al. Morphologic and molecular characteristics
or progesterone receptors. Cancer. 2014;120(5):738–43. of uterine leiomyomas in hereditary leiomyomatosis and renal
251. O’Cearbhaill R, et al. Treatment of advanced uterine leio- cancer (HLRCC) syndrome. Am J Surg Pathol. 2013;37(1):
myosarcoma with aromatase inhibitors. Gynecol Oncol. 74–80.
2010;116(3):424–9. 271. Harrison WJ, et al. Fumarate hydratase-deficient uterine leiomyo-
252. Cuppens T, et al. Potential targets’ analysis reveals dual PI3K/ mas occur in both the syndromic and sporadic settings. Am J Surg
mTOR pathway inhibition as a promising therapeutic strategy Pathol. 2016;40(5):599–607.
for uterine leiomyosarcomas—an ENITEC Group Initiative. Clin 272. Alsolami S, et al. Current morphologic criteria perform poorly in
Cancer Res. 2017;23(5):1274–85. identifying hereditary leiomyomatosis and renal cell carcinoma
253. Dhingra S, et al. Constitutive activation with overexpression of the syndrome-associated uterine leiomyomas. Int J Gynecol Pathol.
mTORC2-phospholipase D1 pathway in uterine leiomyosarcoma 2014;33(6):560–7.
and STUMP: morphoproteomic analysis with therapeutic implica- 273. Zhang Q, et al. Fumarate hydratase mutations and altera-
tions. Int J Clin Exp Pathol. 2010;4(2):134–46. tions in leiomyoma with bizarre nuclei. Int J Gynecol Pathol.
254. Brewer Savannah KJ, et al. Dual targeting of mTOR and aurora-A 2018;37(5):421–30.
kinase for the treatment of uterine leiomyosarcoma. Clin Cancer 274. Lehtonen R, et al. Biallelic inactivation of fumarate hydratase
Res. 2012;18(17):4633–45. (FH) occurs in nonsyndromic uterine leiomyomas but is rare in
255. Elvin JA, et al. Clinical benefit in response to palbociclib treat- other tumors. Am J Pathol. 2004;164(1):17–22.
ment in refractory uterine leiomyosarcomas with a common 275. Piscuoglio S, et al. Uterine adenosarcomas are mesenchymal neo-
CDKN2A alteration. Oncologist. 2017;22(4):416–21. plasms. J Pathol. 2016;238(3):381–8.
50 B. E. Howitt and M. R. Nucci
276. Zaloudek CJ, Norris HJ. Adenofibroma and adenosarcoma 297. Blom R, Guerrieri C. Adenosarcoma of the uterus: a clinicopatho-
of the uterus: a clinicopathologic study of 35 cases. Cancer. logic, DNA flow cytometric, p53 and mdm-2 analysis of 11 cases.
1981;48(2):354–66. Int J Gynecol Cancer. 1999;9(1):37–43.
277. Gallardo A, Prat J. Mullerian adenosarcoma: a clinicopatho- 298. Howitt BE, et al. Targeted genomic analysis of Mullerian adeno-
logic and immunohistochemical study of 55 cases challenging sarcoma. J Pathol. 2015;235(1):37–49.
the existence of adenofibroma. Am J Surg Pathol. 2009;33(2): 299. Howitt BE, et al. Involvement of Chromosome 8 in Mullerian
278–88. adenosarcoma. Int J Gynecol Pathol. 2017;36(1):24–30.
278. Clement PB, Scully RE. Mullerian adenosarcoma of the uterus: a 300. Tai LH, Tavassoli FA. Endometrial polyps with atypical (bizarre)
clinicopathologic analysis of 100 cases with a review of the litera- stromal cells. Am J Surg Pathol. 2002;26(4):505–9.
ture. Hum Pathol. 1990;21(4):363–81. 301. Krivak TC, et al. Uterine adenosarcoma with sarcomatous over-
279. McCluggage WG. Mullerian adenosarcoma of the female genital growth versus uterine carcinosarcoma: comparison of treatment
tract. Adv Anat Pathol. 2010;17(2):122–9. and survival. Gynecol Oncol. 2001;83(1):89–94.
280. Carroll A, et al. Uterine adenosarcoma: an analysis on manage- 302. Kaku T, et al. Adenosarcoma of the uterus: a gynecologic oncol-
ment, outcomes, and risk factors for recurrence. Gynecol Oncol. ogy group clinicopathologic study of 31 cases. Int J Gynecol
2014;135(3):455–61. Pathol. 1992;11(2):75–88.
281. Nathenson MJ, et al. Uterine adenosarcoma: a review. Curr Oncol 303. Tanner EJ, et al. Management of uterine adenosarcomas
Rep. 2016;18(11):68. with and without sarcomatous overgrowth. Gynecol Oncol.
282. Seagle BL, et al. Survival of women with Mullerian adeno- 2013;129(1):140–4.
sarcoma: a National Cancer Data Base study. Gynecol Oncol. 304. Longacre TA, et al. Atypical polypoid adenomyofibromas (atypi-
2016;143(3):636–41. cal polypoid adenomyomas) of the uterus. A clinicopathologic
283. Verschraegen CF, et al. Clinicopathologic analysis of mullerian study of 55 cases. Am J Surg Pathol. 1996;20(1):1–20.
adenosarcoma: the M.D. Anderson Cancer Center experience. 305. Nemejcova K, et al. Atypical polypoid adenomyoma of the uterus:
Oncol Rep. 1998;5(4):939–44. an immunohistochemical and molecular study of 21 cases. Am J
284. Arend R, et al. Long-term outcome and natural history of uterine Surg Pathol. 2015;39(8):1148–55.
adenosarcomas. Gynecol Oncol. 2010;119(2):305–8. 306. McCluggage WG. A practical approach to the diagnosis of mixed
285. Clement PB. Mullerian adenosarcomas of the uterus with sarco- epithelial and mesenchymal tumours of the uterus. Mod Pathol.
matous overgrowth. A clinicopathological analysis of 10 cases. 2016;29(Suppl 1):S78–91.
Am J Surg Pathol. 1989;13(1):28–38. 307. Matsumoto T, et al. Clinical management of atypical polypoid
286. Clement PB, Oliva E, Young RH. Mullerian adenosarcoma of the adenomyoma of the uterus. A clinicopathological review of 29
uterine corpus associated with tamoxifen therapy: a report of six cases. Gynecol Oncol. 2013;129(1):54–7.
cases and a review of tamoxifen-associated endometrial lesions. 308. Soslow RA, et al. Atypical polypoid adenomyofibroma (APA)
Int J Gynecol Pathol. 1996;15(3):222–9. versus well-differentiated endometrial carcinoma with prominent
287. Howitt BE, Quade BJ, Nucci MR. Uterine polyps with features stromal matrix: an immunohistochemical study. Int J Gynecol
overlapping with those of Mullerian adenosarcoma: a clinicopath- Pathol. 1996;15(3):209–16.
ologic analysis of 29 cases emphasizing their likely benign nature. 309. Vang R, Kempson RL. Perivascular epithelioid cell tumor
Am J Surg Pathol. 2015;39(1):116–26. (‘PEComa’) of the uterus: a subset of HMB-45-positive epithe-
288. Clement PB, Scully RE. Mullerian adenosarcomas of the uterus lioid mesenchymal neoplasms with an uncertain relationship to
with sex cord-like elements. A clinicopathologic analysis of eight pure smooth muscle tumors. Am J Surg Pathol. 2002;26(1):1–13.
cases. Am J Clin Pathol. 1989;91(6):664–72. 310. Conlon N, Soslow RA, Murali R. Perivascular epithelioid
289. Stolnicu S, et al. The impact on survival of an extensive sex cord- tumours (PEComas) of the gynaecological tract. J Clin Pathol.
like component in Mullerian adenosarcomas: a study comprising 2015;68(6):418–26.
6 cases. Int J Gynecol Pathol. 2016;35(2):147–52. 311. Schoolmeester JK, et al. TFE3 translocation-associated perivascu-
290. Amant F, et al. Immunohistochemical expression of CD10 lar epithelioid cell neoplasm (PEComa) of the gynecologic tract:
antigen in uterine adenosarcoma. Int J Gynecol Cancer. morphology, immunophenotype, differential diagnosis. Am J Surg
2004;14(6):1118–21. Pathol. 2015;39(3):394–404.
291. Mikami Y, et al. Expression of CD10 in malignant mullerian 312. Kenerson H, et al. Activation of the mTOR pathway in sporadic
mixed tumors and adenosarcomas: an immunohistochemical angiomyolipomas and other perivascular epithelioid cell neo-
study. Mod Pathol. 2002;15(9):923–30. plasms. Hum Pathol. 2007;38(9):1361–71.
292. Amant F, et al. Immunohistochemical determination of hor- 313. Shen Q, et al. Perivascular epithelioid cell tumor (PEComa)
mone receptors in uterine adenosarcomas. Gynecol Oncol. with TFE3 gene rearrangement: clinicopathological, immu-
2003;88(3):463–4. nohistochemical, and molecular features. Virchows Arch.
293. Amant F, et al. Immunohistochemical determination of estrogen 2014;465(5):607–13.
and progesterone receptor positivity in uterine adenosarcoma. 314. Liu F, et al. Malignant perivascular epithelioid cell tumor
Gynecol Oncol. 2004;93(3):680–5. (PEComa) of cervix with TFE3 gene rearrangement: a case report.
294. Soslow RA, Ali A, Oliva E. Mullerian adenosarcomas: an Int J Clin Exp Pathol. 2014;7(9):6409–14.
immunophenotypic analysis of 35 cases. Am J Surg Pathol. 315. Agaram NP, et al. Dichotomy of genetic abnormalities in
2008;32(7):1013–21. PEComas with therapeutic implications. Am J Surg Pathol.
295. Hodgson A, et al. High-grade Mullerian adenosarcoma: genomic 2015;39(6):813–25.
and clinicopathologic characterization of a distinct neoplasm with 316. Malinowska I, et al. Perivascular epithelioid cell tumors (PEComas)
prevalent TP53 pathway alterations and aggressive behavior. Am J harboring TFE3 gene rearrangements lack the TSC2 alterations
Surg Pathol. 2017;41(11):1513–22. characteristic of conventional PEComas: further evidence for a
296. Swisher EM, et al. The expression of epidermal growth factor biological distinction. Am J Surg Pathol. 2012;36(5):783–4.
receptor, HER-2/Neu, p53, and Ki-67 antigen in uterine malignant 317. Argani P, et al. Melanotic Xp11 translocation renal cancers: a
mixed mesodermal tumors and adenosarcoma. Gynecol Oncol. distinctive neoplasm with overlapping features of PEComa, car-
1996;60(1):81–8. cinoma, and melanoma. Am J Surg Pathol. 2009;33(4):609–19.
1 Uterine Mesenchymal Lesions 51
318. Rao Q, et al. PSF/SFPQ is a very common gene fusion partner 337. Benson C, et al. A retrospective study of patients with malig-
in TFE3 rearrangement-associated perivascular epithelioid cell nant PEComa receiving treatment with sirolimus or temsiroli-
tumors (PEComas) and melanotic Xp11 translocation renal can- mus: the Royal Marsden Hospital experience. Anticancer Res.
cers: clinicopathologic, immunohistochemical, and molecular 2014;34(7):3663–8.
characteristics suggesting classification as a distinct entity. Am J 338. Coffin CM, et al. Extrapulmonary inflammatory myofibroblas-
Surg Pathol. 2015;39(9):1181–96. tic tumor (inflammatory pseudotumor). A clinicopathologic
319. Tanaka M, et al. Perivascular epithelioid cell tumor with SFPQ/ and immunohistochemical study of 84 cases. Am J Surg Pathol.
PSF-TFE3 gene fusion in a patient with advanced neuroblastoma. 1995;19(8):859–72.
Am J Surg Pathol. 2009;33(9):1416–20. 339. Rabban JT, et al. Inflammatory myofibroblastic tumor of the
320. Pan CC, et al. Comparative genomic hybridization study of peri- uterus: a clinicopathologic study of 6 cases emphasizing distinc-
vascular epithelioid cell tumor: molecular genetic evidence of tion from aggressive mesenchymal tumors. Am J Surg Pathol.
perivascular epithelioid cell tumor as a distinctive neoplasm. Hum 2005;29(10):1348–55.
Pathol. 2006;37(5):606–12. 340. Bennett JA, et al. Inflammatory myofibroblastic tumor of the
321. Silva EG, et al. A uterine leiomyosarcoma that became positive for uterus: a clinicopathological, immunohistochemical, and molec-
HMB45 in the metastasis. Ann Diagn Pathol. 2005;9(1):43–5. ular analysis of 13 cases highlighting their broad morphologic
322. Silva EG, et al. Uterine leiomyosarcoma with clear cell areas. Int spectrum. Mod Pathol. 2017;30(10):1489–503.
J Gynecol Pathol. 1995;14(2):174–8. 341. Fuehrer NE, et al. ALK-1 protein expression and ALK gene
323. Simpson KW, Albores-Saavedra J. HMB-45 reactivity in rearrangements aid in the diagnosis of inflammatory myofibro-
conventional uterine leiomyosarcomas. Am J Surg Pathol. blastic tumors of the female genital tract. Arch Pathol Lab Med.
2007;31(1):95–8. 2012;136(6):623–6.
324. Ruco LP, et al. Epithelioid lymphangioleiomyomatosis-like 342. Yamamoto H, et al. ALK, ROS1 and NTRK3 gene rearrange-
tumour of the uterus in a patient without tuberous sclerosis: a ments in inflammatory myofibroblastic tumours. Histopathology.
lesion mimicking epithelioid leiomyosarcoma. Histopathology. 2016;69(1):72–83.
1998;33(1):91–3. 343. Cessna MH, et al. Expression of ALK1 and p80 in inflammatory
325. Michal M, Zamecnik M. Hyalinized uterine mesenchymal neo- myofibroblastic tumor and its mesenchymal mimics: a study of
plasms with HMB-45-positive epithelioid cells: epithelioid leio- 135 cases. Mod Pathol. 2002;15(9):931–8.
myomas or angiomyolipomas? Report of four cases. Int J Surg 344. Takahashi A, et al. Anaplastic lymphoma kinase-negative uterine
Pathol. 2000;8(4):323–8. inflammatory myofibroblastic tumor containing the ETV6-NTRK3
326. Kwon BS, et al. Two cases of perivascular epithelioid cell tumor fusion gene: a case report. J Int Med Res. 2018;46:3498–503.
of the uterus: clinical, radiological and pathological diagnostic 345. Alassiri AH, et al. ETV6-NTRK3 is expressed in a subset of ALK-
challenge. Eur J Med Res. 2017;22(1):7. negative inflammatory myofibroblastic tumors. Am J Surg Pathol.
327. Mills AM, Longacre TA. Smooth muscle tumors of the female 2016;40(8):1051–61.
genital tract. Surg Pathol Clin. 2009;2(4):625–77. 346. Antonescu CR, et al. Molecular characterization of inflamma-
328. Fadare O. Uterine perivascular epithelioid cell tumors (PEComas) tory myofibroblastic tumors with frequent ALK and ROS1 gene
and epithelioid smooth muscle neoplasms. Arch Pathol Lab Med. fusions and rare novel RET rearrangement. Am J Surg Pathol.
2008;132(11):1714. 2015;39(7):957–67.
329. Fadare O. Perivascular epithelioid cell tumors (PEComas) 347. Hornick JL, et al. Expression of ROS1 predicts ROS1 gene rear-
and smooth muscle tumors of the uterus. Am J Surg Pathol. rangement in inflammatory myofibroblastic tumors. Mod Pathol.
2007;31(9):1454–5. author reply 1455–6 2015;28(5):732–9.
330. Schoolmeester JK, et al. Alveolar soft part sarcoma of the female 348. Subbiah V, et al. STUMP un“stumped”: anti-tumor response to ana-
genital tract: a morphologic, immunohistochemical, and molecu- plastic lymphoma kinase (ALK) inhibitor based targeted therapy in
lar cytogenetic study of 10 cases with emphasis on its distinction uterine inflammatory myofibroblastic tumor with myxoid features
from morphologic mimics. Am J Surg Pathol. 2017;41(5):622–32. harboring DCTN1-ALK fusion. J Hematol Oncol. 2015;8:66.
331. Folpe AL, et al. Perivascular epithelioid cell neoplasms of 349. Marino-Enriquez A, et al. Epithelioid inflammatory myofibroblas-
soft tissue and gynecologic origin: a clinicopathologic study tic sarcoma: an aggressive intra-abdominal variant of inflamma-
of 26 cases and review of the literature. Am J Surg Pathol. tory myofibroblastic tumor with nuclear membrane or perinuclear
2005;29(12):1558–75. ALK. Am J Surg Pathol. 2011;35(1):135–44.
332. Wagner AJ, et al. Clinical activity of mTOR inhibition with siro- 350. Butrynski JE, et al. Crizotinib in ALK-rearranged inflammatory
limus in malignant perivascular epithelioid cell tumors: targeting myofibroblastic tumor. N Engl J Med. 2010;363(18):1727–33.
the pathogenic activation of mTORC1 in tumors. J Clin Oncol. 351. Mosse YP, et al. Targeting ALK with Crizotinib in pediatric
2010;28(5):835–40. anaplastic large cell lymphoma and inflammatory myofibro-
333. Starbuck KD, et al. Treatment of advanced malignant uterine blastic tumor: a children’s Oncology Group study. J Clin Oncol.
perivascular epithelioid cell tumor with mTOR inhibitors: single- 2017;35(28):3215–21.
institution experience and review of the literature. Anticancer Res. 352. Schwartz EJ, Longacre TA. Adenomatoid tumors of the female
2016;36(11):6161–4. and male genital tracts express WT1. Int J Gynecol Pathol.
334. Gao F, et al. Combination targeted therapy of VEGFR inhibitor, 2004;23(2):123–8.
sorafenib, with an mTOR inhibitor, sirolimus induced a remark- 353. Sangoi AR, et al. Adenomatoid tumors of the female and male
able response of rapid progressive uterine PEComa. Cancer Biol genital tracts: a clinicopathological and immunohistochemical
Ther. 2016;17(6):595–8. study of 44 cases. Mod Pathol. 2009;22(9):1228–35.
335. Dickson MA, et al. Extrarenal perivascular epithelioid cell tumors 354. Ditto A, et al. Embryonal rhabdomyosarcoma of the uterine cervix
(PEComas) respond to mTOR inhibition: clinical and molecular in adults: a case report and literature review. J Low Genit Tract
correlates. Int J Cancer. 2013;132(7):1711–7. Dis. 2013;17(4):e12–7.
336. Ghosh I, et al. Metastatic perivascular epithelioid cell tumor 355. Ferguson SE, et al. Clinicopathologic features of rhabdomyo-
responding to mammalian target of rapamycin inhibition. Indian J sarcoma of gynecologic origin in adults. Am J Surg Pathol.
Med Paediatr Oncol. 2014;35(1):99–102. 2007;31(3):382–9.
52 B. E. Howitt and M. R. Nucci
356. Daya DA, Scully RE. Sarcoma botryoides of the uterine cervix in 368. Rivasi F, et al. Alveolar rhabdomyosarcoma of the uterine
young women: a clinicopathological study of 13 cases. Gynecol cervix. A case report confirmed by FKHR break-apart rear-
Oncol. 1988;29(3):290–304. rangement using a fluorescence in situ hybridization probe on
357. Ghaemmaghami F, Karimi Zarchi M, Ghasemi M. Lower genital paraffin-embedded tissues. Int J Gynecol Pathol. 2008;27(3):
tract rhabdomyosarcoma: case series and literature review. Arch 442–6.
Gynecol Obstet. 2008;278(1):65–9. 369. Ordi J, Stamatakos MD, Tavassoli FA. Pure pleomorphic
358. Li RF, et al. Embryonal rhabdomyosarcoma (botryoid type) of the rhabdomyosarcomas of the uterus. Int J Gynecol Pathol.
uterine corpus and cervix in adult women: report of a case series 1997;16(4):369–77.
and review of the literature. Am J Surg Pathol. 2013;37(3):344–55. 370. Sorensen PH, et al. PAX3-FKHR and PAX7-FKHR gene
359. Dehner LP, Jarzembowski JA, Hill DA. Embryonal rhabdomyo- fusions are prognostic indicators in alveolar rhabdomyosar-
sarcoma of the uterine cervix: a report of 14 cases and a discus- coma: a report from the children's oncology group. J Clin Oncol.
sion of its unusual clinicopathological associations. Mod Pathol. 2002;20(11):2672–9.
2012;25(4):602–14. 371. Yang EJ, et al. Solitary fibrous tumour of the female genital
360. de Kock L, et al. Sequencing of DICER1 in sarcomas identifies tract: a clinicopathological analysis of 25 cases. Histopathology.
biallelic somatic DICER1 mutations in an adult-onset embryonal 2018;72(5):749–59.
rhabdomyosarcoma. Br J Cancer. 2017;116(12):1621–6. 372. Strickland KC, et al. Solitary fibrous tumor of the uterus present-
361. Heravi-Moussavi A, et al. Recurrent somatic DICER1 mutations in ing with lung metastases: a case report. Int J Gynecol Pathol.
nonepithelial ovarian cancers. N Engl J Med. 2012;366(3):234–42. 2016;35(1):25–9.
362. Foulkes WD, et al. Extending the phenotypes associated with 373. Schammel DP, Tavassoli FA. Uterine angiosarcomas: a morpho-
DICER1 mutations. Hum Mutat. 2011;32(12):1381–4. logic and immunohistochemical study of four cases. Am J Surg
363. Schultz KAP, et al. DICER1 and associated conditions: identifica- Pathol. 1998;22(2):246–50.
tion of at-risk individuals and recommended surveillance strate- 374. Cardinale L, et al. Angiosarcoma of the uterus: report of 2 new
gies. Clin Cancer Res. 2018;24(10):2251–61. cases with deviant clinicopathologic features and review of the
364. Doros L, et al. DICER1 mutations in embryonal rhabdomyosarco- literature. Ann Diagn Pathol. 2008;12(3):217–21.
mas from children with and without familial PPB-tumor predispo- 375. Harris NL, Scully RE. Malignant lymphoma and granulocytic sar-
sition syndrome. Pediatr Blood Cancer. 2012;59(3):558–60. coma of the uterus and vagina. A clinicopathologic analysis of 27
365. Kodet R, et al. Childhood rhabdomyosarcoma with ana- cases. Cancer. 1984;53(11):2530–45.
plastic (pleomorphic) features. A report of the Intergroup 376. Hirschowitz L, et al. Intravascular adenomyomatosis: expanding
Rhabdomyosarcoma Study. Am J Surg Pathol. 1993;17(5):443–53. the morphologic spectrum of intravascular leiomyomatosis. Am J
366. Fukunaga M. Pure alveolar rhabdomyosarcoma of the uterine cor- Surg Pathol. 2013;37(9):1395–400.
pus. Pathol Int. 2011;61(6):377–81. 377. Meenakshi M, McCluggage WG. Vascular involvement in adeno-
367. Fadare O, et al. Pleomorphic rhabdomyosarcoma of the uterine myosis: report of a large series of a common phenomenon with
corpus: a clinicopathologic study of 4 cases and a review of the observations on the pathogenesis of adenomyosis. Int J Gynecol
literature. Int J Gynecol Pathol. 2010;29(2):122–34. Pathol. 2010;29(2):117–21.
Fallopian Tube
2
David L. Kolin and Brooke E. Howitt
Abstract Keywords
The fallopian tube contains a muscular wall and plicae Fallopian tube · Serous carcinoma · Serous tubal
lined by ciliated, tubal-type epithelium. While primary intraepithelial carcinoma · Ectopic pregnancy · Pelvic
tumors of the fallopian tube are unusual, benign, border- inflammatory disease
line, and malignant tumors occur, many of which have
similar counterparts in the endometrium and ovary. Over
the past two decades, there has been a realization that the
fimbriae are the source of many (and some believe all) 2.1 Introduction
cases of “ovarian” high-grade serous carcinoma. The spec-
trum of serous tubal neoplasia includes p53 signatures, The fallopian tube has garnered increased attention over the
serous tubal intraepithelial carcinoma, and high- grade past 15 years as it has become implicated as the site of origin
serous carcinoma. These entities have unique morphologic, of many pelvic high-grade serous carcinomas. High-grade
immunophenotypic, and genetic features, with varied clini- serous carcinoma of the fallopian tube was likely underap-
cal significance. The fallopian tube is a frequent site of preciated in the past, as a result of under-sampling of the
metastatic disease from not only the ovary and uterus but fallopian tube and lack of recognition of in situ lesions. Now
may also contain metastases from distant sites (most often that serous tubal intraepithelial carcinoma (STIC) is recog-
the gastrointestinal tract and breast). Metastases may be nized as the precursor of many, if not all, high-grade serous
located within the serosa, mucosa, muscular wall, or intra- carcinomas, more tumors are being classified as fallopian
vascular spaces. The fallopian tube can manifest several tube primaries. With the introduction of the Sectioning and
metaplasias, such as mucinous and transitional, as well as Extensively Examining the Fimbria (SEE-FIM) protocol for
reactive phenomena, including pseudocarcinomatous grossing fallopian tubes, the ability to detect small intramu-
hyperplasia. Sexually transmitted diseases are the most cosal lesions in the fimbria has increased significantly. Since
common causes of infectious salpingitis, which may result the fallopian tube is now accepted to be the site of origin of
in pelvic inflammatory disease and ectopic pregnancy. many BRCA1- and BRCA2-related “ovarian” malignancies,
Because the fimbriae are a site of serous carcinogenesis, the the frequency of prophylactic salpingectomies has increased
SEE-FIM grossing protocol was developed to extensively significantly in recent years. It therefore behooves the pathol-
sample the fimbria in risk-reducing salpingectomies and is ogist to familiarize themselves with the spectrum of intraepi-
now applied in many specimens containing salpingecto- thelial lesions which are commonly encountered in routine
mies, including those from low-risk women. At a mini- salpingectomy specimens.
mum, entirely submitting and examining the distal fallopian This chapter begins with a review of fallopian tube anat-
tube are generally advised. omy and histology. Tubal neoplasias are described, with an
emphasis on both benign and malignant tumors which are
unique to the fallopian tube. Nonneoplastic disorders are
D. L. Kolin then addressed, including inflammatory conditions and vari-
Department of Pathology, Brigham and Women’s Hospital,
ous metaplasias which may be observed and are important to
Boston, MA, USA
recognize as they may mimic neoplastic processes. Special
B. E. Howitt (*)
attention is given to tubal ectopic pregnancies. Finally, sug-
Department of Pathology, Stanford University Medical Center,
Stanford, CA, USA gested grossing protocols are given for the most commonly
e-mail: [email protected] encountered tubal specimens.
2.3.1 Overview
2.3.2.3 Serous Tubal Intraepithelial Carcinoma proliferation index (Fig. 2.4b, c) [11]. There is no accepted
(STIC) or validated Ki-67 cutoff for STIC lesions, although most
Serous tubal intraepithelial carcinoma (STIC) is a precursor STICs show greater than 75% staining [11]. Positive staining
lesion of high-grade serous carcinoma found in the fallopian with stathmin 1 and p16 (strong and diffuse) is also support-
tube. Approximately 90% of STICs are located within the ive of STIC (Fig. 2.4d).
fimbria and with the remainder in the more proximal tube STIC is by definition intraepithelial and thus “noninva-
[11, 21]. STICs are seen in 5–8% of risk-reducing salpingo- sive”; however, similar to serous endometrial intraepithelial
oophorectomy specimens [10, 22]. They are also identified carcinoma, it may disseminate in the peritoneum or less
in the setting of a tubal or ovarian HGSC: 61% of ovarian, commonly metastasize to lymph nodes before progressing to
67% of tubal, and 22–50% of so-called primary peritoneal an invasive tubal carcinoma [26]. Given this metastatic
HGSC are associated with a STIC [21, 23]. It is uncommon potential, STIC is usually staged as Tla (AJCC) or IA
to detect a STIC incidentally (i.e., in non-BRCA1/2 carriers (FIGO), even in the absence of invasion [27].
without HGSC) in a salpingectomy specimen (0.1–0.8%) One may encounter serous tubal intraepithelial prolifera-
[24, 25]. tions that do not fully satisfy criteria for a STIC. Occasionally,
STICs are characterized by increased nuclear-to- one is seen adjacent to a high-grade serous carcinoma with-
cytoplasmic ratio, loss of polarity, mitotic activity, and epi- out an identified STIC [28]. This suggests that invasion may
thelial stratification (Fig. 2.4a). Immunohistochemically, have occurred directly from the intraepithelial proliferation
they show abnormal p53 staining and increased Ki-67 or that the carcinoma overgrew a pre-existing STIC.
a b
c d
Fig. 2.4 Serous tubal intraepithelial carcinoma (STIC) is characterized tubal epithelium STIC shows increased Ki-67 proliferative activity (b),
by increased nuclear-to-cytoplasmic ratio, mitotic activity, loss of abnormal p53 staining pattern (c), and diffuse positivity for p16 (d)
polarity, and prominent nucleoli (a). Compared to the adjacent benign
2 Fallopian Tube 57
There is little evidence to guide treatment of patients carcinoma, while delaying premature menopause. It has
diagnosed with a STIC in the absence of an invasive malig- been speculated that cells from a STIC may exfoliate or
nancy. A small but significant fraction of these patients detach from the fallopian tube and later result in advanced
(4–11%) will go on to develop so-called “primary perito- pelvic carcinoma (“precursor escape”) [28]. This mecha-
neal” serous carcinoma at a later time [22, 29, 30]. Peritoneal nism has been postulated to explain some cases of pelvic
washings may be positive in up to 15% of patients with an high-grade serous carcinoma that arise following prophy-
incidentally detected STIC [31], but the prognostic signifi- lactic salpingo-oophorectomies.
cance of positive washings is unclear. Given the potential for
extratubal metastases, patients are usually observed with
serial serum CA-125 and pelvic ultrasounds, with consider- 2.3.4 D
efinition of Primary Organ Site
ation of BRCA1/2 germline testing if not already performed of Cancers Involving Tube, Ovary,
[30, 32]. Some have advocated for staging, including lymph- and Peritoneum
adenectomy, following a STIC diagnosis [26, 30]. The role
for prophylactic chemotherapy, if any, is unclear [29]. There is some controversy regarding how to assign a pri-
mary site to high-grade serous carcinomas that involve the
tube, ovary, and/or peritoneum. Traditional clinicopatho-
2.3.3 T
ubal Fimbria as the Main Source logic parameters which have been used to differentiate pri-
of Pelvic Serous Carcinoma mary ovarian carcinoma from metastatic disease, such as
bilaterality and surface involvement, are less useful in
Historically, the ovarian surface epithelium or cortical inclu- determining the site of origin of serous carcinoma [21].
sion cysts were thought to be the site of origin of ovarian Features which favor a tubal primary include unilateral
high-grade serous carcinoma. A new paradigm emerged tubal disease, a spectrum of neoplasia in the tube including
when examination of risk-reducing salpingo-oophorectomy STIC, and the absence of advanced (e.g., omental) disease
specimens in BRCA1 and BRCA2 mutation carriers identi- spread [28].
fied early tubal carcinomas, and it was postulated that the Recent consensus guidelines have suggested an algorithm
fallopian tube may be the source of ovarian serous carcinoma for assigning the primary site of serous carcinoma (Table 2.2).
[9–13]. Incidentally discovered STICs in patients without They state that when the tube is involved by carcinoma (by
disease in the ovary added additional evidence that HGSC STIC and/or invasive carcinoma), the primary organ site
arises in the tube [33–35]. Next-generation sequencing of should be specified as the fallopian tube, even if there is a
synchronous STICs and ovarian HGSC have found identical greater burden of disease present in one of the ovaries [27,
mutations in TP53, further bolstering the evidence for STICs 40]. Furthermore, high-grade serous carcinoma should only
as HGSC precursors [36].
There has been some debate about whether all pelvic (i.e.,
Table 2.2 Guidelines for assigning a primary site for carcinomas
tubal, ovarian, and peritoneal) HGSC arises from the fallo- involving the tube, ovary, and peritoneum
pian tube or if there may be a second, fallopian tube-
Primary site Criteria
independent, pathway. For example, STIC is seen less
Fallopian STIC present, with or without ovarian or peritoneal
frequently in BRCA1/2 carriers with HGSC than those with- tube disease OR
out germline mutations, raising the possibility of two differ- Invasive mucosal carcinoma in tube, with or without
ent mechanisms for oncogenesis in these populations [37]. ovarian or peritoneal disease OR
Fallopian tube incorporated in tubo-ovarian mass
Integrated genomic analyses of cases of HGSC both with
Ovary Ovarian carcinoma (even microscopic), in the absence
and without identified STICs have shown similar copy num- of a STIC or tubal mucosal carcinoma AND
ber alterations and mRNA expression profiles between the Tubes must be fully evaluated with SEE-FIM to
two groups, providing evidence that both share a common exclude a tubal primary
origin (putatively the fallopian tube) [38]. Details of the cell Primary No STIC or invasive carcinoma in tubes or ovaries,
peritoneal after examining with SEE-FIM and submitting
of origin of ovarian serous carcinomas are described in Chap.
ovaries in toto
4 of this book. N.B. Do not diagnose “primary peritoneal” in the
Given the high lifetime risk of ovarian carcinoma, pro- setting of neo-adjuvant chemotherapy; these cases
phylactic bilateral salpingo-oophorectomies are recom- should be assigned a site of origin of “tubo-ovarian,”
even if there is no residual disease at these sites
mended for BRCA1/2 carriers. Since the fimbria is now
Tubo- HGSC diagnosed on cytology or omental biopsy OR
recognized as the site of origin for the majority of “ovar- ovarian Tumors treated with neo-adjuvant chemotherapy in
ian” carcinoma, prophylactic salpingectomies are some- which there is peritoneal disease, but no remaining
times performed first, with delayed oophorectomies [39]. tumor in the tubes or ovaries
This gives some of the benefits of risk reduction for ovarian Adapted from ref. [27]
58 D. L. Kolin and B. E. Howitt
be assigned a primary site of the ovary if both tubes have Table 2.3 WHO classification of fallopian tube tumors
been examined using the SEE-FIM protocol and tubal neo- Epithelial tumors and cysts
plasia has been excluded. Primary peritoneal serous carci- Hydatid cyst
noma should only be diagnosed when the tubes and ovaries Benign epithelial tumors
have been examined microscopically and are disease-free. Papilloma
Serous adenofibroma
Neoadjuvant therapy for HGSC is common and can greatly
Epithelial precursor lesion: serous tubal intraepithelial carcinoma
reduce tumor burden and may obscure the site of origin.
Epithelial borderline lesion: serous borderline tumor
Consequently, in the setting of treated disease, a tubo-ovarian Malignant epithelial tumors
primary should be assumed even if no residual disease Low-grade serous carcinoma
remains in the tubes or ovaries. High-grade serous carcinoma
The stage-adjusted outcomes of tubal, ovarian, and pri- Endometrioid carcinoma
mary peritoneal high-grade serous carcinoma are similar, Undifferentiated carcinoma
suggesting that these diagnoses could be grouped together as Mucinous carcinoma
“tubo-ovarian” or “pelvic” high-grade serous carcinoma Transitional carcinoma
[41]. Others argue that because the ovaries are frequently Clear cell carcinoma
Tumor-like lesions
involved by disease, the terminology “HGSC of ovary”
Tubal hyperplasia
should be retained [42]. The incidence of fallopian tube car-
Tubo-ovarian abscess
cinoma will likely increase as cases that may have previously Salpingitis isthmica nodosa
been classified as primary ovarian carcinoma are now recog- Metaplastic papillary tumor
nized to have likely originated from the f allopian tube. Placental site nodule
Mucinous metaplasia
Endometriosis
2.4 Tumors of the Fallopian Tube Endosalpingiosis
Mixed epithelial-mesenchymal tumors
2.4.1 Overview Adenosarcoma
Carcinosarcoma
Mesenchymal tumors
The tube is a site of a wide variety of neoplasia, including
Leiomyoma
benign and malignant epithelial and mesenchymal tumors. Leiomyosarcoma
Historically, the fallopian tube was viewed as an unusual site Mesothelial tumors
for neoplasia in the female genital tract. However, with the Adenomatoid tumor
recognition that many so-called ovarian carcinomas arise Germ cell tumors
from the fimbria, the relative incidence in tubal carcinoma Mature teratoma
has increased. In this chapter, attention will be given to those Immature teratoma
lesions which are common or unique to the tube. The WHO Lymphoid and myeloid tumors
classification of fallopian tube tumors is presented in Modified slightly from ref. [40]
Table 2.3. Many tumors that occur in the fallopian tube have
more common, histologically identical, counterparts in the 2.4.2.2 Borderline Tumors
ovary and endometrium; those lesions are discussed in the Serous borderline tumors of the fallopian tube are rare and
relevant chapters. have a morphologic appearance similar to those of the ovary
[45, 46]. They usually present with abdominal pain and can
range in size from 2 to 23 cm [45]. Most are treated conser-
2.4.2 Epithelial Tumors vatively, with excellent outcomes [45, 46]. Both endometri-
oid and mucinous borderline tumors of the fallopian tube
2.4.2.1 Benign Tumors have also been reported [47, 48].
a b
c d
Fig. 2.6 (a) High-grade serous carcinoma is composed of nests of sion of p16 (b), and abnormal p53 staining pattern (c). In contrast to
pleomorphic, hyperchromatic cells. Mitotic activity is abundant, and endometrial serous carcinoma, which is usually WT-1 negative, tubo-
calcifications may be present. (b) The tumor shows diffuse overexpres- ovarian serous carcinoma is positive for WT-1 (d)
a b
Fig. 2.7 Carcinosarcoma is a biphasic malignant tumor. (a) A malignant epithelial proliferation involves the tubal fimbria. (b) Elsewhere, there is
a malignant stromal component (chondrosarcoma, in this case)
2 Fallopian Tube 61
worse prognosis and should be reported if present. There is 2.4.4 Other Tubal Tumors
no minimum amount of sarcomatous differentiation required
to qualify as a carcinosarcoma, and the combination of any 2.4.4.1 Benign Mesenchymal Tumors
amount of sarcoma with a carcinoma should be diagnosed as Leiomyomas are uncommon in the fallopian tube, but are the
a carcinosarcoma. most common mesenchymal tumor of this site. They are
identical histologically to those that arise in the myome-
Other Tumors trium. Rare cases of cavernous and capillary hemangiomas
Primary fallopian tube mucinous, endometrioid, and of the fallopian tube have been described [62, 63].
clear cell carcinomas have been reported but are extremely
rare [57–59]. Gestational trophoblastic disease (partial 2.4.4.2 Malignant Mesenchymal Tumors
mole, complete mole, and gestational choriocarcinoma), Primary sarcomas of the fallopian tube are extremely rare
identical to intrauterine disease, rarely arises in ectopic [64]. Leiomyosarcomas of the tube have a morphology and
pregnancies [60]. immunophenotype which mimics those of uterine
leiomyosarcomas. Criteria for malignancy are the same as in
the uterus. Their prognosis is generally poor [65, 66]. There
2.4.3 M
ixed Epithelial and Mesenchymal are single case reports describing biphasic synovial sarcoma
Tumors [67], well-differentiated liposarcoma [68], and chondrosar-
coma [69] of the fallopian tube. Embryonal, alveolar, and
2.4.3.1 Serous Adenofibroma pleomorphic rhabdomyosarcomas of the fallopian tube, not
Serous adenofibromas of the tube are benign biphasic tumors arising in an adenosarcoma or carcinosarcoma, have also
composed tubal epithelium overlying a fibromatous stromal been reported [70–72].
nodule (Fig. 2.8). They are small and often incidentally
found during examination of salpingectomies for other indi- 2.4.4.3 Adenomatoid Tumor
cations. They are seen in 10% of women with either BRCA1/2 Adenomatoid tumors are the most common benign tumor of
mutations or a strong family history of breast/ovarian carci- the fallopian tube. They are mesothelial-derived prolifera-
noma but only in 2.5% of non-high-risk women [25]. tions that may occur in the uterus, tube, and ovary. Grossly,
they are solid tan-white nodules, usually smaller than 2 cm.
2.4.3.2 Adenosarcoma In the fallopian tube, they are well-circumscribed and usu-
Adenosarcomas are malignant neoplasms composed of ally lack the infiltrative border seen in other organs [73].
benign epithelium and sarcomatous stroma. Primary fallo- Histologically, they may show a variety of architectures
pian tube adenosarcomas are extremely rare and share the including angiomatoid, tubular, or solid (Fig. 2.9).
same morphology with leaf-like architecture and periglan- A lipoblast-like or signet-ring cell morphology is frequently
dular stromal condensation as those in the endometrium or identified [73]. Cytologic atypia in the form of bizarre or
cervix [61]. symplastic-type atypia (i.e., enlarged, hyperchromatic nuclei
a b
Fig. 2.8 (a) Serous adenofibromas may form small nodules on the fallopian tube, typically on the fimbria. (b) They are benign biphasic tumors
composed of a fibromatous stromal component and serous epithelial component
62 D. L. Kolin and B. E. Howitt
a b
Fig. 2.9 (a, b) Adenomatoid tumors of the tube form well-circumscribed nodules and are often composed of small tubules
with smudgy chromatin) may be seen in adenomatoid tumors tube often show an aberrant expression, which can mimic
and does not affect the excellent prognosis. They display a the staining pattern of a STIC [77]. Recent phylogenetic
mesothelial immunophenotype and are positive for pankera- analysis of cases of synchronous STICs and ovarian high-
tin, calretinin, WT-1, and D2-40 [73, 74]. They are more grade serous carcinoma have shown that the STIC, usually
likely to occur in immunosuppressed patients post-renal presumed to be the precursor lesion, in some cases actually
transplant, but the mechanism for this phenomenon is unclear represents an intramucosal metastases from the ovarian
[75]. The differential diagnosis includes mesothelioma, sal- tumor [80].
pingitis isthmica nodosa, and metastatic adenocarcinoma.
Mesothelioma is larger, is often symptomatic, and, by defini-
tion, has infiltrative growth with invasion into adipose tissue 2.5 Nonneoplastic Disorders
or adjacent organs. Salpingitis isthmica nodosa is lined by of the Fallopian Tube
tubal epithelium and is negative for mesothelial-specific
markers such as D2-40 and calretinin. 2.5.1 T
ubal Epithelial Metaplasia
and Hyperplasia
2.4.4.4 Metastatic Tumors
The fallopian tube is an uncommon site of metastatic dis- 2.5.1.1 Mucinous Metaplasia
ease. In one series of 287 fallopian tubes from 145 patients Mucinous metaplasia is found in approximately 3% of fal-
removed for various reasons, metastatic carcinoma was lopian tubes [25]. Although rare, it is significant because it
identified in 1.4% of tubes [6]. Metastatic tumors in the tube may accompany mucinous lesions elsewhere in the female
may be present in lymphovascular spaces or in intraluminal, genital tract and be associated with Peutz-Jeghers syndrome.
mucosal, submucosal, muscular, or serosal compartments of Mucinous metaplasia of the tube is associated with muci-
the tube (Fig. 2.10) [76–78]. By far the most common type nous cystadenomas, mucinous ovarian tumors of low malig-
of metastatic malignancy is adenocarcinoma (87%), nant potential, and lobular endocervical glandular hyperplasia
although lymphoma, neuroendocrine tumors, and mesothe- [81, 82]. In patients with Peutz-Jeghers syndrome, the muci-
lioma may also metastasize to the tube [77]. Excluding the nous epithelium is positive for MUC6 by immunohistochem-
uterus and ovary, the most common primary sites of meta- istry, suggesting pyloric gland differentiation [81].
static carcinoma are the appendix, colon, stomach, biliary Metastases to the tubal epithelium may occasionally show
system, and breast [76, 77]. Metastatic mucosal disease may mucinous differentiation with bland cytology, so care should
histologically mimic a STIC, so care should be taken when be taken when diagnosing mucinous metaplasia in a patient
diagnosing these lesions in patients with a synchronous pri- with a history of adenocarcinoma elsewhere [77].
mary or history of a previous malignancy [36, 77]. Of note,
tubal mucosal metastases from both endometrioid and 2.5.1.2 Transitional Metaplasia and Walthard
serous endometrial carcinoma can histologically mimic Cell Rest
STICs [36, 79]. Immunohistochemistry for p53, p16, and Walthard cell rests are small, solid, or cystic nests of transi-
WT-1 may be helpful, depending on the context. Caution tional epithelium (Fig. 2.11). They are seen in 5% of salpin-
should be used with p53 and p16, since metastases to the gectomy specimens and are of no significance [6].
2 Fallopian Tube 63
a b
c d
Fig. 2.10 (a, b) Metastatic breast carcinoma NOS to the fallopian tube nantly in lymphovascular spaces. (d) Metastatic lobular carcinoma of
forms a mural nodule composted of epithelioid cells. (c) In this meta- the breast is composed of inconspicuous signet-ring cells
static gastric carcinoma to the fallopian tube, tumor is present predomi-
a b
Fig. 2.11 Walthard cell rests may be solid or cystic (a) and are composed of benign transitional epithelium (b)
64 D. L. Kolin and B. E. Howitt
2.5.1.3 Metaplastic Papillary Tumor clusters from papillary tubal hyperplasia may shed from the
Metaplastic papillary tumor is a rare lesion of the fallopian tube and seed the ovarian or peritoneal surface, where they
tube. Twelve cases have been reported to date, and it is become cortical inclusion cysts or endosalpingiosis [61].
almost always found during pregnancy or in the immediate These deposits may then acquire KRAS or BRAF mutations
postpartum period [83, 84]. It is usually not grossly visible. and develop into borderline serous tumors. Others have not
Microscopically, the tumor shows papillary architecture with found an association between papillary hyperplasia and bor-
edematous papillary cores with sparse lymphocytic inflam- derline tumors [85]. The lack of diagnostic criteria and inter-
mation. Cytologically, the cells are columnar and non- observer variability complicate this diagnosis.
ciliated, with enlarged nuclei and occasional prominent
nucleoli. Epithelial budding and pseudostratification can be 2.5.1.7 Pseudocarcinomatous Hyperplasia
seen. Very infrequent mitotic activity is permitted [83, 84]. It
is uncertain if the tumor represents a metaplasia or is a true Definition
neoplasm. The differential diagnoses include other papillary Pseudocarcinomatous hyperplasia is an exuberant prolifera-
lesions of the tube, including a papilloma, papillary hyper- tion of the tubal epithelium which may be mistaken patho-
plasia, and a borderline tumor. logically for carcinoma. It usually occurs between the ages
of 17–40 (mean 29 years), much younger than most serous
2.5.1.4 Other Metaplasias carcinomas. It may either present as an adnexal mass or be
Oncocytic (eosinophilic) metaplasia is incidental and seen in discovered incidentally [86]. Half of cases are seen in asso-
4% of fallopian tubes [6]. Clear intraepithelial vacuoles, ciation with pelvic inflammatory disease [86].
which are PAS and mucicarmine negative, are thought to
represent a degenerative change seen in older patients [6]. Pathological Findings
Grossly, the tube may be enlarged with a thickened wall.
2.5.1.5 Tubal Mucosal (Epithelial) Hyperplasia Pyosalpinx, a tubo-ovarian abscess, or hydrosalpinx may
Inflammatory conditions of the tube can induce a florid also be present. Microscopically, plical fusion results in the
hyperplasia of the tubal epithelium, with complex architec- epithelium forming cribriform and pseudoglandular struc-
ture which may mimic a neoplasm. Hyperplasia can be dif- tures (Fig. 2.13). Invagination of the epithelium into the
ferentiated from neoplasia based on the lack of significant stroma may induce a stromal response which resembles des-
atypia and mitotic activity and the presence of an inflamma- moplasia [86]. Cytologically, the cells may show moderate
tory infiltrate. nuclear atypia, including prominent nucleoli and focal loss
of polarity [86]. However, there is usually only infrequent
2.5.1.6 Papillary Tubal Hyperplasia mitotic activity, with no atypical forms, and cells often main-
Papillary tubal hyperplasia refers to a specific form of epi- tain a normal nuclear-to-cytoplasmic ratio. Marked chronic
thelial hyperplasia, in which the epithelium forms papillary inflammation, occasionally associated with an acute infil-
tufts with detached papillae within the lumen (Fig. 2.12). It trate, is always present. Psammoma bodies are often seen, as
is often, but not always, associated with psammoma bodies is overlying mesothelial hyperplasia. Papillary structures
or chronic salpingitis. It has been suggested that epithelial may even be present in tubal lymphatics [86].
a b
Fig. 2.12 (a, b) Papillary tubal hyperplasia is characterized by abundant detached papillae within the tubal lumen
2 Fallopian Tube 65
developing world but is rarely seen in North America [57]. It Pelvic Inflammatory Disease
can induce an extensive inflammatory infiltrate with fibrosis, Xanthogranulomatous salpingitis may occur after longstand-
granulomatous reaction, and destruction of normal tubal ing PID and is characterized by foamy macrophages in the
architecture (Fig. 2.16). wall of the tube (Fig. 2.18). Pelvic tuberculosis may clini-
Noninfectious causes of granulomatous salpingitis cally present as pelvic inflammatory disease and should be
include Crohn disease, sarcoidosis, and foreign body reac- considered in immunocompromised patients [85]. Rare cases
tions (Fig. 2.17). It may be difficult to determine a specific of salpingitis can also be caused by Enterobius vermicularis
etiology without adequate history. and pelvic coccidioidomycosis [91, 92].
Fig. 2.15 Tuberculosis infection of the tube shows non-necrotizing Fig. 2.16 Fibrous tissue surrounds calcified Schistosoma eggs in the
granulomatous inflammation fallopian tube
a b
Fig. 2.17 (a) Foreign bodies may elicit a giant cell reaction. (b) The foreign material can often be visualized with polarization
2 Fallopian Tube 67
a b
Fig. 2.18 (a, b) Xanthogranulomatous salpingitis is associated with PID and is characterized by a mixed inflammatory infiltrate, including foamy
macrophages
Fig. 2.19 Neutrophils fill the tubal lumen and infiltrate the lamina pro- Fig. 2.20 Acute salpingitis may eventually result in plical fusion,
pria in acute salpingitis termed follicular salpingitis
pyosalpinx. As the acute salpingitis resolves, the plicae may 2.6.3 Tubal Ectopic Pregnancy
become fused and lamina propria fibrotic, called follicular
salpingitis (Fig. 2.20). Ectopic pregnancies are those in which the embryo implants
outside the endometrium. The vast majority (~95%) occur in
Hydrosalpinx the fallopian tube and less frequently in the cervix, ovary,
As salpingitis resolves, the fimbria may fuse. The tube and abdominal cavity [93]. It occurs in 1% of natural preg-
secondarily becomes distended with clear serous fluid, a nancies and 2–3% of pregnancies which are a result of assis-
condition termed hydrosalpinx. It commonly mimics a tive reproductive technologies [94]. Most tubal ectopics are
serous cystadenoma both clinically and grossly. in the ampulla (80%), and less commonly in the isthmus
Microscopically, however, smooth muscle bundles are (12%), infundibulum, fimbria, or cornua [93].
present in the cystic wall in a hydrosalpinx, which differ-
entiate it from a cystadenoma (Fig. 2.21). The tubal-type 2.6.3.1 Etiology
epithelial lining is the same in both a hydrosalpinx and Tubal ectopic pregnancies are caused by pathologic processes
serous cystadenoma. which obstruct the tubal lumen and interfere with the transport
68 D. L. Kolin and B. E. Howitt
a b
Fig. 2.21 (a) A hydrosalpinx cystically distends the fallopian tube lumen. (b) The hydrosalpinx is lined by tubal-type epithelium, with a mus-
cular wall
Fig. 2.24 (a) At low power, the tube is distended with hemorrhage. (b) Fig. 2.25 Cells with abundant eosinophilic cytoplasm in tubal decidu-
Higher power shows chorionic villi with trophoblast adherent to the tube osis resemble the stromal cells in gestational endometrium
70 D. L. Kolin and B. E. Howitt
patients taking high-dose progestins. The high levels of pro- such as paratubal cysts or hydrosalpinx [103]. Grossly, the
gesterone during pregnancy also cause physiologic shifts in tube appears congested, with possible ischemic necrosis.
the histology of the tubal epithelium, as described earlier. Microscopically, tubal blood vessels may be dilated and
associated with hemorrhage and necrosis. Treatment options
include laparoscopic reduction of the torsion or salpingec-
2.6.6 Endometriosis tomy. In resected specimens, the tubes may show congestion
and ischemic changes. Untreated cases may progress to
There is tubal involvement in 6% of patients with endome- peritonitis.
triosis [100]. Clinically, it may cause infertility if the
endometriosis is mass-forming and occludes the tubal
2.6.7.2 Tubal Prolapse
lumen. Histologically, the diagnosis is made if two of the Prolapse of the fallopian tube through the vaginal vault may
following three features are noted: Müllerian glands (usu- occur after either vaginal or abdominal hysterectomies. Most
ally endometrioid), endometrial-type stroma, and hemosid- cases present symptomatically with vaginal discharge, pelvic
erin-laden macrophages (Fig. 2.26). If only glands are pain, or postcoital bleeding [104]. Predisposing factors are
present, endosalpingiosis should be considered. In a preg- those which impede healing of the vault, such as infection or
nant patient, deciduosis manifests as only decidualized vaginal cuff hematoma [104]. It is more common in premeno-
endometrial stroma without glands. Salpingitis is present in pausal women and can occur at any time post-hysterectomy
66% of patients with endometriosis and may be seen even if (reports range from the immediate postoperative period to
the fallopian tube itself is not involved with endometriosis 32 years after surgery) [104]. Instead of involving the vaginal
[101]. Pseudoxanthomatous salpingiosis is frequently seen vault, the tube may rarely prolapse through an abdominal cae-
with tubal endometriosis (Fig. 2.2). In so-called polypoid sarian section wound [105].
endometriosis, the endometriosis forms a mass and may be Histologically, prolapsed tube shows distortions from its
clinically mistaken for a neoplasm. usual appearance, with a polypoid shape, tubal epithelium,
Occasionally the proximal fallopian tube mucosa is colo- and an increase in stroma (Fig. 2.27). Clinically, the pro-
nized by endometrial glands and stroma, a process termed lapsed tube resembles granulation tissue under colposcopy
endometrialization [102]. This process is associated with and may be misdiagnosed as “adenocarcinoma” pathologi-
tubal ligation. cally as well if tubal epithelium is not appreciated [106]. On
cervical cytology, prolapsed fallopian tube shows columnar
cells which may be mistaken for malignancy if with signifi-
2.6.7 Other Nonneoplastic Disorders cant reactive atypia [107].
a b
n oninfectious in origin. They are often small lesions inciden- identical to those found in the ovary and testis (Fig. 2.28)
tally noted in resection specimens, but they may also present [109, 110]. Hilus cell heterotopia in the tube can be associ-
as a mass. Similar to hydrosalpinx, paratubal cysts com- ated with ipsilateral hilus cell hyperplasia [110]. They
monly contain smooth muscle fibers in the cystic wall. occasionally show brown lipofuscin pigment or Reinke
crystals [109]. Most cases form discrete, circumscribed
2.6.7.4 Tubal Submucosal Hilus Cells nodules; however, others may display an infiltrative, single-
Heterotopic hilus cells, or Leydig cell hyperplasia, are seen file morphology which mimics metastatic lobular breast
in less than 1% of fallopian tubes and may involve the carcinoma [110].
endosalpinx or mesosalpinx [108]. Clinically, they may be
associated with androgenic manifestations. Microscopically, 2.6.7.5 Salpingoliths
they are composed of epithelioid cells with abundant, Salpingoliths are calcifications of the fallopian tube that may
eosinophilic cytoplasm and prominent nucleoli and are
be intraluminal, epithelial, or in the lamina propria, with an
associated overlying epithelial proliferation (Fig. 2.29)
[101]. Salpingoliths occur in about a third of tubes with acute
or chronic salpingitis or in cases with ovarian serous neopla-
sia, but in only 5% of cases without tumors or salpingitis
Fig. 2.27 Prolapsed fallopian tube may show fimbria lined by tubal-
type epithelium, as well as areas resembling granulation tissue. Fig. 2.28 Hilus cell hyperplasia in the tube is composed of cells identi-
Photomicrograph courtesy of Dr. C. Crum (Brigham and Women’s cal to ovarian hilus cells, with large, epithelioid cells with abundant
Hospital, Boston, MA) eosinophilic cytoplasm
a b
Fig. 2.29 Salpingoliths are calcifications in the tube and may be intraluminal (a) or within the tubal stroma (b)
72 D. L. Kolin and B. E. Howitt
a b
Fig. 2.31 (a) Radiograph of fallopian tube containing metal contraceptive coil (“Essure”). (b) Fallopian tube containing coil after opening
longitudinally
74 D. L. Kolin and B. E. Howitt
Am J Surg Pathol. 2015;39:357–64. https://doi.org/10.1097/ analysis consortium study. Lancet Oncol. 2013;14:853–62. https://
PAS.0000000000000353. doi.org/10.1016/S1470-2045(13)70253-5.
36. McDaniel AS, Stall JN, Hovelson DH, et al. Next-generation 54. Malpica A, Deavers MT, Lu K, et al. Grading ovarian serous
sequencing of tubal intraepithelial carcinomas. JAMA Oncol. carcinoma using a two-tier system. Am J Surg Pathol. 2004;28:
2015;1:1128. https://doi.org/10.1001/jamaoncol.2015.1618. 496–504.
37. Howitt BE, Hanamornroongruang S, Lin DI, et al. Evidence 55. Al-Hussaini M, Stockman A, Foster H, McCluggage WG. WT-1
for a dualistic model of high-grade serous carcinoma: BRCA assists in distinguishing ovarian from uterine serous carcinoma
mutation status, histology, and tubal intraepithelial carcinoma. and in distinguishing between serous and endometrioid ovarian
Am J Surg Pathol. 2015;39:287–93. https://doi.org/10.1097/ carcinoma. Histopathology. 2004;44:109–15.
PAS.0000000000000369. 56. Shen Y-M, Xie Y-P, Xu L, et al. Malignant mixed müllerian tumor
38. Ducie J, Dao F, Considine M, et al. Molecular analysis of high- of the fallopian tube: report of two cases and review of literature.
grade serous ovarian carcinoma with and without associated Arch Gynecol Obstet. 2010;281:1023–8. https://doi.org/10.1007/
serous tubal intra-epithelial carcinoma. Nat Commun. 2017;8:990. s00404-009-1331-6.
https://doi.org/10.1038/s41467-017-01217-9. 57. Wheal A, Jenkins R, Mikami Y, et al. Primary mucinous carci-
39. Kwon JS, Tinker A, Pansegrau G, et al. Prophylactic salpingec- noma of the fallopian tube. Int J Gynecol Pathol. 2017;36:393–9.
tomy and delayed oophorectomy as an alternative for BRCA https://doi.org/10.1097/PGP.0000000000000330.
mutation carriers. Obstet Gynecol. 2013;121:14–24. https://doi. 58. Malak M, Klam S. Primary fallopian tube clear cell adenocarci-
org/10.1097/AOG.0b013e3182783c2f. noma in pregnancy: case presentation and review of the litera-
40. Singh N, Gilks CB, Hirshowitz L, et al. Adopting a uniform ture. Case Rep Obstet Gynecol. 2015;2015:183243. https://doi.
approach to site assignment in tubo-ovarian high-grade serous car- org/10.1155/2015/183243.
cinoma: the time has come. Int J Gynecol Pathol. 2016;35:230–7. 59. Rabczyński J, Ziółkowski P. Primary endometrioid carcinoma
https://doi.org/10.1097/PGP.0000000000000270. of fallopian tube. Clinicomorphologic study. Pathol Oncol Res.
41. Moss EL, Evans T, Pearmain P, et al. Should all cases of high- 1999;5:61–6.
grade serous ovarian, tubal, and primary peritoneal carcino- 60. Muto MG, Lage JM, Berkowitz RS, et al. Gestational tropho-
mas be reclassified as tubo-ovarian serous carcinoma? Int J blastic disease of the fallopian tube. J Reprod Med. 1991;36:
Gynecol Cancer. 2015;25:1201–7. https://doi.org/10.1097/ 57–60.
IGC.0000000000000477. 61. Gallardo A, Prat J. Mullerian adenosarcoma. Am J
42. Prat J, Young RH. Letter to the editor on “Should All Cases of High- Surg Pathol. 2009;33:278–88. https://doi.org/10.1097/
Grade Serous Ovarian, Tubal, and Primary Peritoneal Carcinomas PAS.0b013e318181a80d.
Be Reclassified as Tubo-Ovarian Serous Carcinoma?” Int J 62. Atere-Roberts N, Angelopoulos L, Ghataura S, Vijayanand
Gynecol Cancer 2015;25. Int J Gynecol Cancer. 2016;26:226–7. R. Cavernous haemangioma of the fallopian tube presenting as
https://doi.org/10.1097/IGC.0000000000000617. appendicitis in an adolescent girl. J Obstet Gynaecol (Lahore).
43. Kaspersen P, Buhl L, Møller BR. Fallopian tube papilloma in 2010;30:70–1. https://doi.org/10.3109/01443610903315645.
a patient with primary sterility. Acta Obstet Gynecol Scand. 63. Katiyar R, Patne SCU, Bharti S, Jain M. Capillary hemangioma
1988;67:93–4. of the fallopian tube. J Clin Diagnostic Res. 2016;10:QD01–2.
44. Gisser SD. Obstructing fallopian tube papilloma. Int J Gynecol https://doi.org/10.7860/JCDR/2016/16774.7570.
Pathol. 1986;5:179–82. 64. Zagouri F, Dimopoulos M-A, Thomakos N, et al. Sarcomas
45. Choi SM, Kang WD, Choi HS, et al. Serous borderline tumor of of the fallopian tube: disentangling a rare entity. Onkologie.
the fallopian tube. Obstet Gynecol Sci. 2014;57:334. https://doi. 2011;34:132–8. https://doi.org/10.1159/000324792.
org/10.5468/ogs.2014.57.4.334. 65. Jacoby AF, Fuller AF, Thor AD, Muntz HG. Primary leiomyo-
46. Zheng W, Wolf S, Kramer EE, et al. Borderline papillary serous sarcoma of the fallopian tube. Gynecol Oncol. 1993;51:404–7.
tumour of the fallopian tube. Am J Surg Pathol. 1996;20: https://doi.org/10.1006/gyno.1993.1312.
30–5. 66. Ueda T, Emoto M, Fukuoka M, et al. Primary leiomyosarcoma of
47. Im HS, Kim JO, Lee SJ, et al. Borderline mucinous tumor aris- the fallopian tube. Int J Clin Oncol. 2010;15:206–9. https://doi.
ing in a paratubal cyst: a case report. Eur J Gynaecol Oncol. org/10.1007/s10147-010-0027-z.
2011;32:206–7. 67. Mitsuhashi A, Nagai Y, Suzuka K, et al. Primary synovial sar-
48. Salamon C, Tornos C, Chi DS. Borderline endometrioid tumor coma in fallopian tube: case report and literature review. Int
arising in a paratubal cyst: a case report. Gynecol Oncol. J Gynecol Pathol. 2007;26:34–7. https://doi.org/10.1097/01.
2005;97:263–5. https://doi.org/10.1016/j.ygyno.2004.12.032. pgp.0000225841.13880.3a.
49. Prat J. Abridged republication of FIGO’s staging classification 68. Oishi H, Moriyama S, Kotera K, et al. First case of liposarcoma
for cancer of the ovary, fallopian tube, and peritoneum. Cancer. arising from the fallopian tube: case report and review of the
2015;121:3452–4. https://doi.org/10.1002/cncr.29524. literature. J Obstet Gynaecol Res. 2008;34:713–6. https://doi.
50. Riska A, Sund R, Pukkala E, et al. Parity, tubal sterilization, org/10.1111/j.1447-0756.2008.00913.x.
hysterectomy and risk of primary fallopian tube carcinoma in 69. Pickel H, Thalhammer M. Chondrosarcoma of the fallopian tube.
Finland, 1975–2004. Int J Cancer. 2007;120:1351–4. https://doi. Geburtshilfe Frauenheilkd. 1971;31:1243–8.
org/10.1002/ijc.22491. 70. Buchwalter CL, Jenison EL, Fromm M, et al. Pure embryo-
51. Koskela-Niska V, Riska A, Lyytinen H, et al. Primary fallo- nal rhabdomyosarcoma of the fallopian tube. Gynecol Oncol.
pian tube carcinoma risk in users of postmenopausal hormone 1997;67:95–101. https://doi.org/10.1006/gyno.1997.4833.
therapy in Finland. Gynecol Oncol. 2012;126:241–4. https://doi. 71. Ferguson SE, Gerald W, Barakat RR, et al. Clinicopathologic
org/10.1016/j.ygyno.2012.04.045. features of rhabdomyosarcoma of gynecologic origin in adults.
52. Inal MM, Hanhan M, PIlanci B, Tinar S. Fallopian tube malig- Am J Surg Pathol. 2007;31:382–9. https://doi.org/10.1097/01.
nancies: experience of Social Security Agency Aegean Maternity pas.0000213352.87885.75.
Hospital. Int J Gynecol Cancer. 2004;14:595–9. https://doi. 72. Shahin NA, Alqaisy A, Zheng W. Primary alveolar rhabdomyosar-
org/10.1111/j.1048-891X.2004.14404.x. coma of fallopian tube masquerading as a unilateral adnexal mass:
53. Sieh W, Köbel M, Longacre TA, et al. Hormone-receptor A case report and literature review. Indian J Pathol Microbiol.
expression and ovarian cancer survival: an ovarian tumor tissue 2015;58:521. https://doi.org/10.4103/0377-4929.168884.
76 D. L. Kolin and B. E. Howitt
73. Sangoi AR, McKenney JK, Schwartz EJ, et al. Adenomatoid 90. Evans DT. Actinomyces israelii in the female genital tract: a
tumors of the female and male genital tracts: a clinico- review. Genitourin Med. 1993;69:54–9.
pathological and immunohistochemical study of 44 cases. 91. Smith G, Hoover S, Klotz SA, Sobonya R. Abdominal and pelvic
Mod Pathol. 2009;22:1228–35. https://doi.org/10.1038/ coccidioidomycosis. Am J Med Sci. 2011;341:308–11. https://doi.
modpathol.2009.90. org/10.1097/MAJ.0b013e3182058736.
74. Terada T. An immunohistochemical study of adenomatoid tumors 92. Erhan Y, Zekioğlu O, Ozdemir N, Sen S. Unilateral salpin-
of the uterus and fallopian tube. Appl Immunohistochem Mol gitis due to enterobius vermicularis. Int J Gynecol Pathol.
Morphol. 2012;20:173–6. 2000;19:188–9.
75. Mizutani T, Yamamuro O, Kato N, et al. Renal transplantation- 93. Hacker NF, Gambone JC, Hobel CJ. Hacker & Moore’s essentials
related risk factors for the development of uterine adenoma- of obstetrics and Gynecology. Philadelphia: Elsevier; 2015.
toid tumors. Gynecol Oncol Rep. 2016;17:96–8. https://doi. 94. Walker JJ. Ectopic pregnancy. Clin Obstet Gynecol. 2007;50:89–
org/10.1016/j.gore.2016.05.003. 99. https://doi.org/10.1097/GRF.0b013e31802f4f79.
76. Stewart CJR, Leung YC, Whitehouse A. Fallopian tube metas- 95. Ankum WM, Mol BW, Van der Veen F, Bossuyt PM. Risk fac-
tases of non-gynaecological origin: a series of 20 cases empha- tors for ectopic pregnancy: a meta-analysis. Fertil Steril.
sizing patterns of involvement including intra- epithelial 1996;65:1093–9.
spread. Histopathology. 2012;60:E106–14. https://doi. 96. Li C, Zhao W-H, Zhu Q, et al. Risk factors for ectopic pregnancy:
org/10.1111/j.1365-2559.2012.04194.x. a multi-center case-control study. BMC Pregnancy Childbirth.
77. Rabban JT, Vohra P, Zaloudek CJ. Nongynecologic metastases to 2015;15:187. https://doi.org/10.1186/s12884-015-0613-1.
fallopian tube mucosa. Am J Surg Pathol. 2015;39:35–51. https:// 97. Green LK, Kott ML. Histopathologic findings in ectopic tubal
doi.org/10.1097/PAS.0000000000000293. pregnancy. Int J Gynecol Pathol. 1989;8:255–62.
78. Na K, Kim H-S. Clinicopathological characteristics of fallo- 98. Baergen RN, Rutgers J, Young RH. Extrauterine lesions of inter-
pian tube metastases from primary endometrial, cervical, and mediate trophoblast. Int J Gynecol Pathol. 2003;22:362–7. https://
nongynecological malignancies: a single institutional experi- doi.org/10.1097/01.pgp.0000092132.88121.d1.
ence. Virchows Arch. 2017;471:363–73. https://doi.org/10.1007/ 99. Felix AS, Brinton LA, McMeekin DS, et al. Relationships of tubal
s00428-017-2186-z. ligation to endometrial carcinoma stage and mortality in the NRG
79. Kommoss F, Faruqi A, Gilks CB, et al. Uterine serous carcinomas oncology/gynecologic oncology group 210 trial. J Natl Cancer
frequently metastasize to the fallopian tube and can mimic serous Inst. 2015;107 https://doi.org/10.1093/jnci/djv158.
tubal intraepithelial carcinoma. Am J Surg Pathol. 2017;41:161– 100. Jenkins S, Olive DL, Haney AF. Endometriosis: pathogenetic
70. https://doi.org/10.1097/PAS.0000000000000757. implications of the anatomic distribution. Obstet Gynecol.
80. Eckert MA, Pan S, Hernandez KM, et al. Genomics of ovar- 1986;67:335–8.
ian cancer progression reveals diverse metastatic trajectories 101. Seidman JD, Sherman ME, Bell KA, et al. Salpingitis, salpingo-
including intraepithelial metastasis to the fallopian tube. Cancer liths, and serous tumors of the ovaries: is there a connection? Int J
Discov. 2016;6:1342–51. https://doi.org/10.1158/2159-8290. Gynecol Pathol. 2002;21:101–7.
CD-16-0607. 102. Mutter GL, Prat J. Pathology of the female reproductive tract. 3rd
81. Kato N, Sugawara M, Maeda K, et al. Pyloric gland metaplasia/ ed. Philadelphia: Elsevier; 2014.
differentiation in multiple organ systems in a patient with Peutz- 103. Casey RK, Damle LF, Gomez-Lobo V. Isolated fallopian tube tor-
Jegher’s syndrome. Pathol Int. 2011;61:369–72. https://doi. sion in pediatric and adolescent females: a retrospective review
org/10.1111/j.1440-1827.2011.02670.x. of 15 cases at a single institution. J Pediatr Adolesc Gynecol.
82. Seidman JD. Mucinous lesions of the fallopian tube. A report of 2013;26:189–92. https://doi.org/10.1016/j.jpag.2013.02.010.
seven cases. Am J Surg Pathol. 1994;18:1205–12. 104. Ouldamer L, Caille A, Body G. Fallopian tube prolapse after
83. Jang MI, Sung J-Y, Kim J-Y, Kim H-S. Clinicopathological char- hysterectomy: a systematic review. PLoS One. 2013;8:e76543.
acteristics of metaplastic papillary tumor of the fallopian tube. https://doi.org/10.1371/journal.pone.0076543.
Anticancer Res. 2017;37:3693–701. 105. Goyal LD, Maheshwari S, Kaur S, Kaur H. Prolapse of fallopian
84. Saffos RO, Rhatigan RM, Scully RE. Metaplastic papillary tumor tube through abdominal wound after caesarean section mimicking
of the fallopian tube—a distinctive lesion of pregnancy. Am J Clin scar endometriosis: a case report. J Med Case Rep. 2015;9:280.
Pathol. 1980;74:232–6. https://doi.org/10.1186/s13256-015-0769-3.
85. Wolsky RJ, Price MA, Zaloudek CJ, Rabban JT. Mucosal pro- 106. Song YS, Kang JS, Park MH. Fallopian tube prolapse misdi-
liferations in completely examined fallopian tubes accom- agnosed as vault granulation tissue: a report of three cases.
panying ovarian low-grade serous tumors: Neoplastic Pathol Res Pract. 2005;201:819–22. https://doi.org/10.1016/j.
precursor lesions or normal variants of benign mucosa? Int prp.2005.09.001.
J Gynecol Pathol. 2018;37:262–74. https://doi.org/10.1097/ 107. Nasir N, Desai M, Marshall J, Gupta N. Prolapsed fallopian tube:
PGP.0000000000000410. cytological findings in a ThinPrep liquid based cytology vagi-
86. Cheung AN, Young RH, Scully RE. Pseudocarcinomatous hyper- nal vault sample. Diagn Cytopathol. 2013;41:146–9. https://doi.
plasia of the fallopian tube associated with salpingitis. A report of org/10.1002/dc.21742.
14 cases. Am J Surg Pathol. 1994;18:1125–30. 108. Honoré LH, O’Hara KE. Ovarian hilus cell heterotopia. Obstet
87. Majmudar B, Henderson PH, Semple E. Salpingitis isthmica Gynecol. 1979;53:461–4.
nodosa: a high-risk factor for tubal pregnancy. Obstet Gynecol. 109. Lewis JD. Hilus-cell hyperplasia of ovaries and tubes; report of a
1983;62:73–8. case. Obstet Gynecol. 1964;24:728–31.
88. Jenkins CS, Williams SR, Schmidt GE. Salpingitis isth- 110. Hirschowitz L, Salmons N, Ganesan R. Ovarian hilus cell het-
mica nodosa: a review of the literature, discussion of clinical erotopia. Int J Gynecol Pathol. 2011;30:46–52. https://doi.
significance, and consideration of patient management. Fertil
org/10.1097/PGP.0b013e3181eaa1ff.
Steril. 1993;60:599–607. 111. Horn L-C, Angermann K, Hentschel B, et al. Frequency of pap-
89. Chappell CA, Wiesenfeld HC. Pathogenesis, diagnosis, and man- illary tubal hyperplasia (PTH), salpingoliths and transition from
agement of severe pelvic inflammatory disease and tuboovar- adenoma to borderline ovarian tumors (BOT): a systematic analy-
ian abscess. Clin Obstet Gynecol. 2012;55:893–903. https://doi. sis of 74 BOT with different histologic types. Pathol Res Pract.
org/10.1097/GRF.0b013e3182714681. 2017;213:305–9. https://doi.org/10.1016/j.prp.2017.02.001.
2 Fallopian Tube 77
112. Fadare O, Khabele D. Salpingo-oophorectomy specimens for 114. Rabban JT, Krasik E, Chen L-M, et al. Multistep level sec-
endometrial cancer staging: a comparative analysis of represen- tions to detect occult fallopian tube carcinoma in risk-reducing
tative sampling versus whole tissue processing. Hum Pathol. salpingo-oophorectomies from women with BRCA mutations:
2013;44:643–50. https://doi.org/10.1016/j.humpath.2012.07.015. implications for defining an optimal specimen dissection proto-
113. Mahe E, Tang S, Deb P, et al. Do deeper sections increase the col. Am J Surg Pathol. 2009;33:1878–85. https://doi.org/10.1097/
frequency of detection of serous tubal intraepithelial carci- PAS.0b013e3181bc6059.
noma (STIC) in the “sectioning and extensively examining the 115. Dhruva SS, Ross JS, Gariepy AM. Revisiting essure—toward safe
FIMbriated end” (SEE-FIM) protocol? Int J Gynecol Pathol. and effective sterilization. N Engl J Med. 2015;373:e17. https://
2013;32:353–7. https://doi.org/10.1097/PGP.0b013e318264ae09. doi.org/10.1056/NEJMp1510514.
Benign Diseases of the Ovary
3
David Suster, Martina Z. Liu, and Douglas I. Lin
Fig. 3.6 Wolffian remnants (epoophoron) Fig. 3.8 Multiple primordial follicles in neonatal ovary
A second cell population that can be identified in the hilus become atretic if not selected for ovulation. At birth,
is the epithelial lining constituting the rete ovarii, which thousands of primordial follicles are present in the neo-
resembles the male rete testis counterpart. Morphologically, natal ovary (Fig. 3.8); and, throughout life, their numbers
this cell population ranges from cuboidal to columnar and is steadily decrease through folliculogenesis and atresia. In
arranged in a network of clefts, tubules, and small papilla the reproductive age ovary, primordial, primary, and sec-
(Fig. 3.5). The rete ovarii stains positively for EMA, cyto- ondary follicles are found in the superficial ovarian cor-
keratin, and Pax8 and is negative for Pax2 and Gata3 [8, 9]. tex (Fig. 3.9). Primordial follicles are characterized by a
The rete ovarii may be in close proximity to mesonephric primary oocyte surrounded by a single layer of flattened,
tubule remnants (i.e., epoophoron) (Fig. 3.6) and Walthard’s mitotically inactive, granulosa cells (Fig. 3.10). In the
nests (i.e., transitional cell metaplasia) (Fig. 3.7). first stage of maturation from primordial to primary fol-
licle, the flattened, mitotically inert granulosa layer
assumes a cuboidal to columnar shape, becomes mitoti-
3.1.2 Follicular Development and Atresia cally active, and increases in size (Fig. 3.11). Proliferation
of the granulosa cells results in their stratification with
Follicular development (folliculogenesis) is a continuous up to five concentric layers of granulosa cells around the
process wherein primordial follicles are either selected oocyte and progression to a secondary or pre-antral
and undergo maturation in preparation for ovulation or follicle (Fig. 3.12).
82 D. Suster et al.
Fig. 3.9 Primordial, primary, and secondary follicles in the cortex Fig. 3.12 Secondary (pre-antral) follicle
Fig. 3.10 Primordial follicles in reproductive age ovary Fig. 3.13 Mature (Graafian) follicle. From left to right: antrum, granu-
losa cell layer, theca cell layers, and ovarian cortical stroma
3.1.3 H
ormones and Peptides Secreted
by the Ovary
Fig. 3.14 Mature (Graafian) follicle
The ovary produces a variety of hormones and peptides that
are secreted into the blood in both physiologic and patho-
logical conditions. We will focus on hormones and peptides
that are produced by the benign ovary and that are physio-
logically and clinically useful. Factors that are associated
with neoplasia, such as elevated CA125 and germ cell tumor
markers, are discussed elsewhere in the book. Briefly,
regarding CA125, in addition to a variety of cancers, a num-
ber of benign ovarian or gynecological conditions may be
associated with CA125 elevation, such as endometriosis,
fibroids, pelvic inflammatory disease, menses, pregnancy,
and peritonitis [12].
The ovary is a site of steroid hormone production in both
physiological and pathological conditions. Numerous stud-
ies have demonstrated the extra-adrenal steroidogenic poten-
tial and gonadotropin responsiveness of ovarian sex cord
stromal cells. During ovulatory cycles, there is an intimate
Fig. 3.15 Preovulatory follicle. Oocyte with surrounding zona pellu- physiologic interrelationship between the hypothalamo-
cida and layer of granulosa cells floating in antrum fluid pituitary-ovarian-endometrial axis. The interplay of rising
a b
Fig. 3.16 Corpus luteum low (a) and high (b) power
84 D. Suster et al.
a b
a b
The ovarian sex cord stromal cells also produce nonste- early follicular phase inhibin B levels, are all correlated
roidal hormones, such as glycoproteins, inhibin A and B, with ovarian reserve and are predictive of reproductive
which are secreted into the bloodstream and negatively regu- potential [17]. AMH is produced by the granulosa cells of
late FSH synthesis by the pituitary [14]. The main cellular small growing follicles. During follicular maturation, AMH
sources of inhibin B are the growing follicle granulosa cells physiologically inhibits initial follicle recruitment and sub-
and granulosa cell tumors, while inhibin A is mainly pro- sequent FSH-dependent selection and growth of pre-antral
duced by the corpus luteum and the placenta [15]. For this and small antral follicles. As follicles mature, AMH remains
reason, in infertile women who are undergoing assisted highly expressed in cumulus cells [18]. For these reasons,
reproduction therapies, inhibin B may be a useful marker of AMH levels reflect the number of antral and pre-antral fol-
ovarian reserve, a term that describes the remaining pool of licles present in the ovaries, and the levels of circulating
oocytes or a woman’s reproductive potential [16], since low AMH in blood have been used in various clinical scenarios
inhibin B levels have been shown to predict a poor ovulatory and applications. For instance, circulating AMH levels
response [15]. In addition, serum inhibin B levels are helpful have been suggested to predict the ovarian response to
in the diagnosis and management of granulosa cells tumors. hyperstimulation of the ovaries for in vitro fertilization
Both adult and juvenile types of granulosa cells tumors may (IVF) and egg/oocyte retrieval and the timing of meno-
present with elevation of serum inhibin B in either a primary pause. Low AMH levels have also been proposed to indi-
or a recurrent tumor setting. cate iatrogenic damage to the ovarian oocyte reserve, and
Besides inhibin B, another marker of ovarian reserve is AMH has also been used as a surrogate marker for antral
the anti-Müllerian hormone (AMH). AMH levels, as well follicle count in the diagnosis of polycystic ovary syndrome
as chronologic age, early follicular phase FSH levels, and (PCOS) [19].
86 D. Suster et al.
3.2.1.2 Pathology
Grossly, the corpus luteum of pregnancy is a single, yellow,
cerebriform nodule, often with hemorrhagic center.
Microscopically, wedge-shaped invaginations imparting a
cerebriform, undulating architecture are seen at low power
magnification (Figs. 3.20 and 3.21). The surrounding ovarian
stroma may undergo deciduosis or ectopic decidua (Fig. 3.21,
top). On high power, the lesion is composed of luteinized
granulosa cells with eosinophilic cytoplasm and central
nuclei with prominent nucleoli (Fig. 3.22). In contrast to the
nonpregnant corpus luteum, pink hyaline globules and calci-
fications are features specific of the corpus luteum of preg-
nancy (Fig. 3.22). The main differential diagnosis includes
pregnancy luteoma [21]. Features favoring corpus luteum of Fig. 3.22 Corpus luteum of pregnancy, high power hyaline globules
3 Benign Diseases of the Ovary 87
3.2.2 P
regnancy Luteoma (Theca-Lutein
Hyperplasia of Pregnancy)
3.2.2.2 Pathology
Grossly, pregnancy luteomas are discrete or multinodular
brown or red/hemorrhagic solid nodules ranging from 1.0 cm
up to 20 cm that replace the ovarian parenchyma (Fig. 3.23).
They may be bilateral in up to a third of cases [22, 23].
Microscopically, the nodules are typically composed of a
Fig. 3.25 Luteoma of pregnancy, high power: solid growth of eosino-
philic cells with round nuclei, prominent nucleoli, and delicate blood
vessels
3.2.3.2 Pathology
Hyperreactio luteinalis is characterized by bilateral ovarian
enlargement up to 15 cm or more in greatest dimension. The
ovarian surface is lobulated, and sectioning reveals numer-
ous cysts ranging from 1 to 3 cm in size (Fig. 3.27) that are
remarkable for a smooth, yellow lining. The cyst content is
Fig. 3.26 Luteoma of pregnancy, high power: typical follicle-like clear watery to hemorrhagic. Morphologically, the cysts are
spaces containing colloid-like secretory secretion lined by luteinized granulosa and theca interna cells often
with edema or features of torsion of the cyst wall and ovarian
stroma (Fig. 3.28).
luteinized granulosa cell tumor, Leydig cell tumor, and meta-
static carcinoma or melanoma.
3.2.4 L
arge Solitary Luteinized Follicular Cyst
of Pregnancy
3.2.4.2 Pathology
The gross appearance is of a large unilocular cyst, up to 55 cm
in size (average diameter of 25 cm), filled with either serous
or mucoid fluid [28]. The cyst wall is typically thin, measur-
ing less than 0.5 cm [31], which, upon histological examina-
tion, is composed of up to ten layers of luteinized cells with
eosinophilic to vacuolated cytoplasm lining the cyst wall.
Distinct granulosa and theca layers are typically not seen.
Instead, the luteinized cells lining the cyst wall are uniform
with small, round nuclei, smooth nuclear contours, and small
Fig. 3.30 Large solitary luteinized follicle cyst of pregnancy with
nucleoli (Fig. 3.29). However, foci of scattered cells may focal nuclear atypia
show marked nuclear enlargement, hyperchromasia, and
bizarre and pleomorphic nuclei (Fig. 3.30) [28]. Importantly,
hyaline bodies or calcified foci that are frequently seen in cor- 3.2.5 Ectopic Decidua (Deciduosis)
pus luteum of pregnancy are not present. Mitosis, apoptosis,
and necrosis are typically absent. Nests of luteinized cells, 3.2.5.1 Clinical Features
similar in appearance to those lining the cyst, can be identi- The development of ectopic decidua is considered a physi-
fied within the stroma of the cyst wall, which should not be ologic phenomenon of pregnancy and has been reported in
mistaken for invasion. A reticulin stain can highlight scattered almost all ovaries at term [32]. During pregnancy, ectopic
fibrils surrounding the nests of luteinized cells [29]. decidua may also be encountered in diverse locations such
as endocervical stroma, uterine serosa, fallopian tubes,
3.2.4.3 Differential Diagnosis omentum, appendix, small and large intestine, and urinary
The major malignant differential diagnoses that should be bladder [33]. Involvement of the appendix may give rise to
excluded include unilocular cystic granulosa cell tumors of signs and symptoms of acute appendicitis during pregnancy
either adult or juvenile types. Other benign differential diag- [34]. Ectopic decidua may also occur in early gestation, as
noses include corpus luteum of pregnancy, pregnancy lute- well as in nonpregnant women with progestin treatment,
oma, and hyperreactio luteinalis, which can be distinguished with trophoblastic disease, and with a hormone-producing
based on the gross and histological features. ovarian or extra-ovarian lesion [33]. Deciduosis is a benign
90 D. Suster et al.
process that regresses in the postpartum period without any 3.2.5.3 Differential Diagnosis
major complications. The differential diagnosis includes a luteinized sex cord stro-
mal tumor or a metastatic neoplasm (melanoma, deciduoid
3.2.5.2 Pathology mesothelioma, carcinoma), especially when encountered in
Gross lesions are often not seen but, if present, may mimic non-gynecological sites such as the omentum [36].
a neoplasm as gray to white nodules, polyps, or sheets on
the ovarian surface or extra-ovarian sites [32].
Microscopically, ectopic decidua typically involves the 3.2.6 Pregnancy-Associated Granulosa
ovarian cortex or areas of periovarian adhesions (Fig. 3.31). Cell Proliferation
Morphologically, deciduosis resembles the decidualized
endometrial stroma counterpart of the gestational endome- 3.2.6.1 Clinical Features
trium, and it is characterized by epithelioid cells with abun- Pregnancy-associated granulosa cell proliferations are
dant clear to pink cytoplasm and bland, round nuclei benign, incidental findings in ovaries removed during preg-
(Fig. 3.32). Mitotic activity is absent, but focal nuclear nancy. They are thought to occur due to the FSH-like activity
pleomorphism, hyperchromasia, prominent nucleoli, and of HCG and the hormonal milieu of pregnancy. Follow-up
necrosis may be seen [35]. data on these lesions is scant, with only ten reported cases in
the literature; however they are considered benign nonneo-
plastic lesions of the ovary [37, 38].
3.2.6.2 Pathology
Because they are incidental findings, gross appearance has
not been described. Microscopically, they are characterized
by multifocal, microscopic foci of solid, non-cystic, intrafol-
licular granulosa cell proliferation, typically measuring
<1 mm in diameter, but have been reported to be up to 5 mm
in greatest dimension. They are spatially located within the
center of atretic ovarian follicles and are surrounded by a
theca layer. The classic architecture and cytological features
of granulosa cells are seen, including solid, insular, microfol-
licular, or trabecular patterns, high nuclear-cytoplasmic ratio,
and oval nuclei with nuclear grooves. Two other patterns of
proliferation have also been reported: (1) microscopic,
<5 mm, nodules of luteinized granulosa cells with round,
Fig. 3.31 Ectopic decidua involving the ovarian cortex
non-grooved nuclei, and (2) a sertoliform tubular pattern with
vacuolated cytoplasm, mimicking testicular sertoli tubules.
3.2.7.2 Pathology
Similar to tubal ectopic pregnancy, pathological diagnosis of Fig. 3.34 Ovarian ectopic pregnancy with placental chorionic villi
primary ovarian pregnancy requires gross or histological evi-
dence of embryonic implantation in the ovary. Grossly, ovar- products of conception including chorionic villus, implanta-
ian pregnancy may appear as a hemorrhagic cyst or mass that tion site, isolated trophoblast, and embryonic sac with or
may mimic a neoplasm. Gross identification of a gestational without fetal tissue [43]. When the morphologic evidence is
sac, placental villi, or an embryo may be possible depending equivocal in the ovary, ancillary immunohistochemistry for
on the gestational age. Microscopically, identification of staining β-hCG, human placental lactogen, and inhibin to
implantation site trophoblasts (Fig. 3.33), immature placen- highlight trophoblasts can be used to aid the diagnosis [43].
tal chorionic villi (Fig. 3.34), or embryonic tissue within
ovarian parenchyma is required for the diagnosis.
3.3 varian Disorders Associated
O
3.2.7.3 Differential Diagnosis with Infertility
The differential diagnosis includes hemorrhagic cysts, such
as the corpus luteum, a luteinized follicular cyst, a germ cell 3.3.1 P
remature Ovarian Failure
tumor, and a ruptured tubal ectopic pregnancy. Distinction with Accelerated Follicle Depletion
from a fallopian tube pregnancy requires that the ipsilateral
fallopian tube is intact or with no evidence of pregnancy 3.3.1.1 Clinical Features
after extensive histologic examination and in addition to the Premature ovarian failure (POF) or primary ovarian
identification, within the affected ovary, of any evidence of insufficiency (POI) is defined as at least 4 months of
92 D. Suster et al.
a menorrhea and FSH levels of >40 mIU/mL on two sepa- thus limiting the accuracy of the test. A patient with dimin-
rate occasions, at least 1 month apart, in a woman under ished ovarian reserve may have few to no follicles present
40 years old. Presenting symptoms include irregular men- by biopsy, but these findings are known to have a poor cor-
ses, which may precede definitive diagnosis by several relation with the presence of follicles elsewhere in the
years. In addition, patients may suffer from infertility or ovary owing to heterogeneous distribution. In addition, the
menopausal-type symptoms secondary to low estrogen procedure is invasive, and rarely, pregnancy has occurred in
levels. The most common causes of POF are chemotherapy the absence of follicles on biopsy. Therefore ovarian biopsy
and radiotherapy treatments for malignancy; however, is not necessary or recommended as a means of diagnosing
most cases of POF with accelerated follicle depletion and POF [16].
diminished ovarian follicle reserve have no known etiol-
ogy [44]. A small proportion of cases may be attributed to
an underlying genetic or autoimmune (discussed elsewhere 3.3.2 Autoimmune Oophoritis
in this chapter) disorder. Among the known genetic altera-
tions that have been linked to POF, those related to the X 3.3.2.1 Clinical Features
chromosome are the most common, with a smaller propor- Autoimmune oophoritis is a rare cause of premature ovar-
tion attributed to somatic genes. Variations in the X chro- ian failure (POF) with accelerated follicle depletion.
mosome include Turner syndrome (mosaic or complete Patients affected by autoimmune oophoritis present in early
monosomy), trisomy X syndrome, fragile X syndrome, stages with menstrual irregularities and decreased fertility,
and deletions in either the long or short arm of the X chro- ultimately leading to sustained amenorrhea before the age
mosome. Somatic genetic alterations include galactose- of 40 with associated high FSH and low estrogen levels
mia, 17-alpha-hydroxylase deficiency, and mutations in [47]. Autoimmune oophoritis is most commonly associated
the FSH receptor, LH receptor, inhibin, and FOXL2 genes. with autoimmune Addison’s disease in the setting of auto-
Other known causes of POF include tobacco smoking, immune polyendocrine syndromes (APS) type 1 and type
infectious oophoritis, and pelvic surgery for benign dis- 2. Outside of this association, autoimmune POF may be
ease. Diagnostic testing should include FSH, LH, prolac- seen with autoimmune thyroiditis or type 1 diabetes melli-
tin, estradiol levels, and markers of ovarian follicle reserve tus most commonly and, rarely, with other non-endocrine
such as inhibin B and/or anti-Müllerian hormone (AMH) autoimmune disorders. Autoimmune POF has been attrib-
levels [45]. Pregnancy and autoimmune oophoritis should uted to steroid-cell autoantibodies, also termed ovary and
be excluded, and karyotyping or other genetic testing adrenal autoantibodies, which are directed against
should be considered, especially if there is a familial his- steroidogenic enzymes: 17 alpha-hydroxylase and cyto-
tory of POF (present in approximately 20–30% of cases). chrome P450-side chain cleavage enzymes [48].
Noninvasive testing to determine ovarian reserve include Immunofluorescence studies show that these autoantibod-
clomiphene citrate challenge, FSH, inhibin B, AMH lev- ies localize to the theca cells of antral (secondary and ter-
els, ovarian volume, and antral follicle count [45]. tiary) follicles but spare granulosa cells and the primary
Management of premature ovarian failure includes hor- follicles. These autoantibodies incite an immune response
mone replacement therapy and fertility treatments to leading to the destruction of theca cells and a decrease in
induce ovarian hyperstimulation for egg retrieval in cases estrogen secondary to lack of substrate. Thus, FSH and LH
of diminished ovarian reserve or utilizing donor oocytes levels increase, stimulating production of inhibin B by the
when there is complete follicle depletion. remaining granulosa cells. This pattern of concurrent
increase in FSH and inhibin B is specific for autoimmune
3.3.1.2 Pathology oophoritis and is not seen in other causes of POF. Serologic
Grossly, the ovaries of patients diagnosed with POF may detection of steroid-cell autoantibodies is the preferred
appear unremarkable or be distorted by cystic follicles or method of diagnosis [49]. Ultrasound may show ovarian
corpora lutea. Ovarian biopsy has been suggested as a enlargement or multiple cystic follicles but is nonspecific.
method for determining ovarian reserve by documenting Biopsy is rarely performed because the expected inflamma-
the number and type of follicles present. Histological eval- tory infiltrate is almost never seen in the absence of sero-
uation of the cortical tissue by ovarian biopsy should logic evidence of autoimmune POF. Management includes
include quantification of all follicles in various states of hormone replacement and fertility treatments utilizing
maturation including corpora lutea and atretic forms [46]. donor oocytes. A safe and effective immunosuppressive
However, ovarian biopsy is no longer recommended as a regimen has not been developed or validated for use early
means of determining the ovarian reserve because follicle in the course of autoimmune oophoritis to theoretically
distribution within the ovarian cortex is heterogeneous, restore ovarian function.
3 Benign Diseases of the Ovary 93
3.3.2.2 Pathology
Ovaries affected by autoimmune oophoritis may appear
grossly unremarkable or may be mildly enlarged. Rarely, the
ovary may be markedly enlarged with numerous cystic
follicles [50]. Histological features include a dense mono-
nuclear infiltrate composed of T and B lymphocytes and
plasma cells involving the theca interna of antral follicles,
while primary follicles and granulosa cells are spared
(Figs. 3.35, 3.36, 3.37, and 3.38) [51]. Rarely, an eosino-
philic infiltrate accompanies the mononuclear cells in the
destruction of the antral follicles (Fig. 3.39) [52]. Later in
disease progression, diminished numbers of follicles remain.
As with any inflammatory process, oophoritis secondary to
infection must be ruled out. Correlation between the histo-
logical and clinical serologic findings is essential in the diag-
nosis of autoimmune oophoritis. Fig. 3.37 Autoimmune oophoritis with destruction of theca layer
Fig. 3.35 Autoimmune oophoritis, low power Fig. 3.38 Autoimmune oophoritis with CD3 immunostain highlight-
ing T cells in theca layer
3.3.3 P
rimary Ovarian Insufficiency Without
Follicle Depletion (Resistant Ovarian
Syndrome)
3.4.1.2 Pathology
Grossly, ovaries are small and shrunken in size, typically
<2 cm, with smooth or gyriform surface (Fig. 3.44), and they
may be difficult to identify if adhesions are present.
Microscopically, there is cortical stromal atrophy, which is
characterized by a thin cortex and minimal amounts of
medullary stroma (Figs. 3.45 and 3.46). In the medulla and
hilar junction, multiple corpora albicantia and vessels may Fig. 3.46 Atrophy, medium power
be prominent.
3.4.2.2 Pathology
Macroscopically, both ovaries may be normal in size or
enlarged, and a grossly identifiable mass is typically absent.
Fig. 3.44 Atrophic ovary, gross When enlarged, the cortex and/or medulla may be expanded
3 Benign Diseases of the Ovary 97
3.4.3.2 Pathology
Macroscopically, by definition, a lesion is not grossly
Fig. 3.48 Stromal hyperplasia, high power visible. Microscopically, there are ill-defined clusters or
98 D. Suster et al.
Fig. 3.51 Hilus (Leydig) cell hyperplasia, low power Fig. 3.54 Hilus (Leydig) cell hyperplasia, inhibin immunostain
Fig. 3.52 Hilus (Leydig) cell hyperplasia, high power Fig. 3.55 Hilus (Leydig) cell hyperplasia, calretinin immunostain
3 Benign Diseases of the Ovary 99
3.5.1.2 Pathology
Grossly, an ovarian abscess is characterized by a cystic mass
with central areas of necrosis and hemorrhage and areas of Fig. 3.57 Acute oophoritis
100 D. Suster et al.
Fig. 3.58 Acute oophoritis with abscess formation, low power Fig. 3.61 Xanthogranulomatous oophoritis, high power
3.5.2 Actinomycosis
3.6.1 E
ndosalpingiosis, Cortical Inclusion
Cysts, and Simple Cysts
3.6.1.2 Pathology
Endosalpingiosis and cortical inclusion cysts are grossly
recognizable as cystic spaces within the ovarian cortex that
range in size from microscopic to up to 1 cm. The cyst con-
tent is watery, and the lining is smooth. Histologically, the
cyst lining of endosalpingiosis consists of a single layer of
fallopian tube-like epithelium, which is remarkable for cili-
ated, secretory, and intercalated cells (Figs. 3.68 and 3.69)
that are either Pax8 (secretory cells) or tubulin positive (cili-
ated cells) and calretinin negative by immunohistochemis-
try. In contrast, cortical inclusion cysts display a flattened
lining composed of mesothelial-like ovarian surface epithe-
lial cells (Fig. 3.70) that are positive for calretinin and nega-
tive for Pax8 [91]. Papillary projections of the epithelium
and associated fibromatous stroma are absent from both cyst
types. Rare cases of mass-forming endosalpingiosis have Fig. 3.70 Cortical inclusion simple cyst
been reported and consist of aggregates of thin-walled,
layer of mucinous epithelium with bland, basally located atretic follicular cysts, which are cysts derived from follicles
nuclei. By definition, significant cytological nuclear atypia or prior to ovulation. Alternatively, commonly encountered
architectural complexity is absent in cystadenomas. For addi- cysts are derived from corpus luteum after ovulation, such as
tional details regarding epithelial ovarian lesions and distinc- hemorrhagic corpus luteum, corpus luteal cysts, and corpus
tion between cystadenoma with borderline tumors, see also albicans cysts.
Chaps. 5–7.
3.6.4.2 Non-luteinized Follicle Cyst
Pathology
Fig. 3.72 Non-luteinized follicular cyst: high power view demon- Luteinized cysts may be divided into granulosa-lutein cysts
strates normal granular cell layer four to five cells thick with surround- and theca-lutein cysts and are characterized by prominent
ing theca cell layer luteinization of the granulosa cell layer, the theca cell layer,
or sometimes both. The luteinization appears as larger cells
with increased amounts of eosinophilic cytoplasm; these
changes mimic those seen in the corpus luteum. Grossly,
these cysts are composed of large unilocular or multilocular
smooth-walled cysts that can grow in size up to 30 cm.
Granulosa-lutein cysts show luteinization of the granulosa
cell layer where the granulosa cells are characterized by an
increased amount of eosinophilic cytoplasm (Figs. 3.74 and
3.75). The distinction between granulosa and theca cell lay-
ers in these cases may be difficult to identify. Theca-lutein
cysts may also show an expansion of the theca cell layer,
characterized by large eosinophilic theca cells with a rela-
tively normal appearing overlying granulosa cell layer
(Fig. 3.76). The differential diagnosis includes other types
of ovarian cysts, such as endometriotic cysts and cystic
ovarian neoplasms.
Clinical Features
Variable degrees of luteinization may be seen in follicular Fig. 3.74 Luteinized follicular cyst: prominent eosinophilic, granular
cysts. Luteinized follicular cysts may develop under the cell layer
106 D. Suster et al.
Pathology
Grossly, these cysts may resemble their follicular counter-
parts. Microscopically, atretic cysts are characterized by
absence or degeneration of the granulosa cell layer and are
lined by an attenuated/degenerating cyst wall lining or nor-
mal ovarian stroma (Figs. 3.77 and 3.78). Within this stroma,
single or small groups of luteinized theca cells may be iden-
tified. The central cavity is often either compressed or
replaced by a central zone of loose connective tissue
(Fig. 3.79). Atretic cysts may histologically be indistinguish-
able from involuting follicles or small follicular cysts.
Fig. 3.77 Early follicle with loss of antrum and early degenerative
changes
Clinical
Atretic follicular cysts may be found incidentally on oopho-
rectomy specimens and clinically are not associated with any
significant pathology. Atresia refers to the process of granu-
losa cell apoptosis under the influence of hormonal control
by multiple ligand-receptor pathways including the tumor
necrosis factor alpha, Fas, and other pathways. Both un-
luteinized and luteinized follicular cysts may undergo atretic Fig. 3.78 Atretic follicle cyst with loss of granular cell layers and
changes [10, 11]. attenuation of wall lining
3 Benign Diseases of the Ovary 107
Pathology
Grossly, the hemorrhagic corpus luteum is as a unilocular
cystic structure within the ovarian cortical parenchyma with
a yellow-orange cyst wall lining and contains a dark red cav-
ity composed of hemorrhagic material (Fig. 3.80). These
cysts may be multiple and, like the corpus luteum, may vary
in size. Microscopically, the cyst wall lining appears as a
convoluted layer of luteinized granulosa cells with an outer
layer of theca cells. The central cystic cavity is often filled
with red blood cells which may extravasate into the cyst wall
(Figs. 3.81 and 3.82). Small thin-walled vessels may
Clinical Features
Following ovulation and prior to the formation of the true
corpus luteum, the corpus hemorrhagicum (“bleeding cor-
pus luteum”) may be formed when blood vessels supplying
the mature follicle rupture. The hemorrhagic corpus luteum
is composed of a cyst-like space with a central area of hem- Fig. 3.80 Hemorrhagic corpus luteum, gross
orrhage [98]. Corpus hemorrhagicum cysts are one of the
most commonly encountered types of ovarian cysts. They
are more often unilateral than bilateral and are encountered
in patients just after the ovulation period. These cysts can
be seen with increased frequency in patients undergoing
ovulation therapy for pregnancy, in various bleeding disor-
ders, or in patients who are on anticoagulation therapy [99,
100]. Clinically, patients with these cysts have variable pre-
sentations, often being asymptomatic or presenting with
abdominal pain [101]. In some cases, rupture of the cysts
can present as hemoperitoneum with signs of acute abdo-
men [102]. The differential diagnosis includes other com-
mon causes of abdominal pain in a female such as ovarian
torsion, ovarian cyst rupture, acute appendicitis, ectopic
pregnancy, infection, and malignancy. Most corpus hemor-
rhagicum cysts will spontaneously regress; however, some
cysts may require treatment. Excision is usually curative;
however, in some cases, an oophorectomy is required to Fig. 3.81 Hemorrhagic corpus luteal cyst showing large central cavity
stop the bleeding [103]. with blood and red blood cell extravasation into cyst wall
108 D. Suster et al.
Pathology
Grossly, corpus luteum cysts may range in size from 3 cm to
greater than 10 cm and contain clear fluid or blood. The cyst
wall lining is usually thin, convoluted, and yellow in color.
Microscopically, corpus luteum cysts resemble the normal
corpus luteum with increased convolutions of the cyst wall
which is composed of luteinized granulosa cells and theca
cells (Figs. 3.84 and 3.85). A small zone of vessels may also
Fig. 3.82 Hemorrhagic corpus luteal cyst: convoluted luteinized cyst be identified within the cyst wall (Fig. 3.85). The granulosa
wall lining involved by prominent hemorrhage cells are fully luteinized and are present singly or in clusters
depending on the stage of the cyst (Figs. 3.85 and 3.86).
Smaller peripheral clusters of theca-lutein cells may be iden-
tified residing within the cyst wall. Evidence of fresh hemor-
rhage, fibrin deposition, and sometimes organizing blood
Clinical
Corpus luteum cysts are derived from corpora lutea which
form in the postovulatory period of the menstrual cycle.
Corpus luteum cysts will typically involute spontaneously
if fertilization does not occur but may sometimes persist if
Fig. 3.85 Corpus luteum cyst, medium power: Convoluted luteinized
they become filled with fluid or blood. Clinically they are cyst wall lining with focal areas of red blood cell extravasation and
nearly all asymptomatic. Similar to the normal corpora small vessels visible within the cyst wall
3 Benign Diseases of the Ovary 109
Fig. 3.86 Corpus luteum cyst, high power: fully luteinized cyst wall
lining
Fig. 3.87 Corpora albicans cyst with a hyalinized convoluted cyst wall
collections may be identified in association with a fibrous and central edematous cystic cavity
tissue reaction and hemosiderin-laden macrophages. The
cyst wall linings may become attenuated or obscured, and
the convolutions of the cyst wall may be lost when there is a
large amount of hemorrhage or fluid. The differential diag-
nosis includes luteinized follicular cysts as well as endome-
triotic cysts.
Clinical
If pregnancy does not occur following the rupture of a mature
follicle during ovulation, the corpus luteum undergoes invo-
lution to create the corpora albicans. The corpora albicans
essentially represents an avascular scar, which is the remnant
of the reabsorbed mature ovarian follicle. The corpora albi-
cans is a solid, hyalinized structure; however, in some cases,
Fig. 3.88 Corpora albicans with cystic degeneration of the central
it may undergo cystic change with a central cavity filled with portion
clear fluid. Unlike follicular cysts, luteinized cysts, and cor-
pus luteal cysts, corpora albicans cysts tend to be small and
remain less than 1 cm in size. Clinically, these are nonfunc-
tional cysts, with few to no clinical signs or associated
pathology.
Pathology
Microscopically, corpora albicans cysts have a convo-
luted cyst wall lining composed of a thick band of acel-
lular hyalinized material similar to the normal corpora
albicans (Fig. 3.87). The cyst wall may have a thin layer
of fibrous tissue, a remnant of the degenerated corpus
luteum. The central cavity is often cystically dilated and
may be empty or filled with a faintly opaque edematous
liquid (Figs. 3.88 and 3.89). While calcifications may be
present in association with the corpora albicans, they are
less likely to appear in association with a corpora albi- Fig. 3.89 Corpora albicans cyst. High power view shows a hyalinized
cans cyst [105]. corpora albicans cyst wall lining with edematous clear liquid filling the
central cavity
110 D. Suster et al.
3.7.1.2 Pathology
Grossly, the ovaries are often enlarged to a size of 8–12 cm;
however they can grow to as large as 35 cm [106, 107, 109,
110]. The enlarged ovarian mass may exhibit a tan-white col-
oration and a tense external surface. On cut sections, the
stroma is white-gray in color with a weeping clear protein-
aceous fluid infiltrating the ovarian stroma. Microscopically,
the ovarian stroma becomes hypocellular and is replaced by
edematous clear fluid imparting a myxoid appearance. This
edema is more often located within the inner cortex, while the
periphery of the ovary tends to be more cellular (Figs. 3.90
and 3.91). The edematous fluid displays apart normal struc-
tures which may still be identified within the area of edema Fig. 3.92 Massive ovarian edema. Residual corpora albicans within
(Fig. 3.92). The vasculature and lymphatic vessels are often edematous changes
3 Benign Diseases of the Ovary 111
3.7.2.2 Pathology
The underlying pathophysiology of ovarian torsion involves
torsion of the ovarian pedicle leading to reduced venous
return, stromal edema, internal hemorrhage, and infarction.
For this reason, grossly, the ovary may appear enlarged,
swollen, and replaced by a hemorrhagic or infarcted mass
(Figs. 3.95 and 3.96). Microscopically, partial or intermit- Fig. 3.99 Calcified foreign embolization material for uterine fibroids
in ovarian hilar vessels
tent ovarian torsion may result in massive ovarian edema,
which was previously described above in this chapter, or
hemorrhage separating non-infarcted normal structures 3.8 Incidental Findings
(Fig. 3.97). Alternatively, full torsion ultimately results in
hemorrhagic infarction of the ovary and of the underlying 3.8.1 Cortical Fibromatosis
cyst or neoplasm, if one is present (Fig. 3.98). Due to exten-
sive infarction, a definitive pathological diagnosis and clas- Prominent areas of cortical fibrosis, which are character-
sification of cyst or neoplasm may not always be possible; ized by replacement of ovarian cortical stroma by areas of
however, they are most often benign. The differential diag- abundant collagen deposition, can be seen in peri- and post-
nosis of torsion includes ovarian abscess and necrosis due menopausal ovaries as part of aging-related changes [1].
to ovarian vein thrombosis or due to ovarian artery occlu- Focal areas of cortical fibromatosis may also be seen in
sion following uterine artery embolization for leiomyomata reproductive age ovaries in the setting of polycystic ova-
[117–119]. Identification of thrombosis or foreign emboli- ries, cortical fibromas, or massive edema [120]. Cortical
zation material is required to make the latter diagnosis fibromatosis may morphologically resemble a small
(Fig. 3.99). fibroma (Figs. 3.100 and 3.101); however, the distinction
3 Benign Diseases of the Ovary 113
Fig. 3.100 Cortical fibromatosis, low power Fig. 3.102 Benign mesothelial proliferation within ovarian surface
adhesions and hemosiderin deposition
3.9.3 S
econdary Ovarian Involvement
by Systemic Disorders
Fig. 3.111 Ovary with non-necrotizing granuloma due to polarizable
Rarely, the ovary is a site of secondary involvement by sys- foreign material. The patient had a history of abdominal myomectomy
years prior to oophorectomy
temic disorders, and ovarian involvement is usually an inci-
dental finding in women with systemic disease. For
instance, systemic inflammatory disorders such as sarcoid- scleroderma; or (3) amyloid deposition in the ovaries with
osis, Crohn’s disease, or vasculitis have been reported in systemic amyloidosis [137, 139, 140]. Clinical history and
the literature to secondarily involve the ovary [133–137]. In pathological correlation are helpful in identifying second-
addition, systemic amyloidosis and systemic storage disor- ary involvement by these disorders. The main differential
ders involving the ovary have also been reported [138, diagnosis of non-necrotizing granulomas in the ovary is a
139]. Patients may present with a tumor-like ovarian mass, foreign body reaction to exogenous material (Figs. 3.110
uterine bleeding, and constitutional symptoms such as and 3.111), often suture material introduced from a previ-
fever. Microscopically, features in the ovary are similar to ous operative procedure, to keratin from a teratoma or to
morphological changes seen in other organ systems. This is refractile crystalline material of uncertain origin [140]. In
illustrated by the presence of non-necrotizing granulomas the differential of ovarian necrotizing granulomas due to
in the ovary in cases of secondary involvement by either systemic disease, an infectious oophoritis process with
(1) systemic sarcoidosis or Crohn’s disease; (2) vasculitis abscess formation and necrotic pseudoxanthomatous nodu-
involving ovarian hilar or adnexal vessels in cases of lar inflammation due to endometriosis must also be
systemic giant cell arteritis, polyarteritis nodosa, and
excluded [141].
3 Benign Diseases of the Ovary 117
43. Wang Y, Chen H, Zhao M, Fadare O, Zheng W. Primary ovarian 63. Zhang X-L, Zhang C-W, Xu P, et al. SNP rs2470152 in CYP19 is
pregnancy. Int J Gynecol Pathol. 2018; https://doi.org/10.1097/ correlated to aromatase activity in Chinese polycystic ovary syn-
PGP.0000000000000482. drome patients. Mol Med Rep. 2012;5:245–9.
44. Nippita TA, Baber RJ. Premature ovarian failure: a review. 64. Carvalho LML, Ferreira CN, Candido AL, Reis FM, Sóter MO,
Climacteric. 2007;10:11–22. Sales MF, Silva IFO, Nunes FFC, Gomes KB. Metformin reduces
45. Sun W, Stegmann BJ, Henne M, Catherino WH, Segars total microparticles and microparticles-expressing tissue factor in
JH. A new approach to ovarian reserve testing. Fertil Steril. women with polycystic ovary syndrome. Arch Gynecol Obstet.
2008;90:2196–202. 2017;296:617–21.
46. Massin N, Gougeon A, Meduri G, et al. Significance of ovarian 65. Api M. Could transvaginal, ultrasound-guided ovarian interstitial
histology in the management of patients presenting a premature laser treatment replace laparoscopic ovarian drilling in women
ovarian failure. Hum Reprod. 2004;19:2555–60. with polycystic ovary syndrome resistant to clomiphene citrate?
47. Silva CA, Yamakami LYS, Aikawa NE, Araujo DB, Carvalho JF, Fertil Steril. 2009;92:2039–40.
Bonfá E. Autoimmune primary ovarian insufficiency. Autoimmun 66. Charalampakis V, Tahrani AA, Helmy A, Gupta JK, Singhal
Rev. 2014;13:427–30. R. Polycystic ovary syndrome and endometrial hyperplasia: an
48. Welt CK, Falorni A, Taylor AE, Martin KA, Hall JE. Selective overview of the role of bariatric surgery in female fertility. Eur J
theca cell dysfunction in autoimmune oophoritis results in multi- Obstet Gynecol Reprod Biol. 2016;207:220–6.
follicular development, decreased estradiol, and elevated inhibin 67. Lepine S, Jo J, Metwally M, Cheong YC. Ovarian surgery for
B levels. J Clin Endocrinol Metab. 2005;90:3069–76. symptom relief in women with polycystic ovary syndrome.
49. La Marca A, Brozzetti A, Sighinolfi G, Marzotti S, Volpe A, Cochrane Database Syst Rev. 2017;11:CD009526.
Falorni A. Primary ovarian insufficiency: autoimmune causes. 68. Pasquali R, Diamanti-Kandarakis E, Gambineri A. Management
Curr Opin Obstet Gynecol. 2010;22:1. of endocrine disease: secondary polycystic ovary syndrome: theo-
50. Bannatyne P, Russell P, Shearman RP. Autoimmune oophoritis: retical and practical aspects. Eur J Endocrinol. 2016;175:R157–69.
a clinicopathologic assessment of 12 cases. Int J Gynecol Pathol. 69. Ozkan S, Murk W, Arici A. Endometriosis and infertility. Ann N Y
1990;9:191–207. Acad Sci. 2008;1127:92–100.
51. Sedmak DD, Hart WR, Tubbs RR. Autoimmune oophoritis: a his- 70. Prentice L, Stewart A, Mohiuddin S, Johnson NP. What is endo-
topathologic study of involved ovaries with immunologic charac- salpingiosis? Fertil Steril. 2012;98:942–7.
terization of the mononuclear cell infiltrate. Int J Gynecol Pathol. 71. Pavlik EJ, DePriest PD, Gallion HH, Ueland FR, Reedy MB,
1987;6:73–81. Kryscio RJ, van Nagell JR. Ovarian volume related to age.
52. Lewis J. Eosinophilic perifolliculitis: a variant of autoimmune Gynecol Oncol. 2000;77:410–2.
oophoritis? Int J Gynecol Pathol. 1993;12:360–4. 72. Karam K, Hajj S. Hyperthecosis syndrome. Clinical, endocri-
53. Aittomäki K, Lucena JL, Pakarinen P, et al. Mutation in the nologic and histologic findings. Acta Obstet Gynecol Scand.
follicle-
stimulating hormone receptor gene causes hereditary 1979;58:73–9.
hypergonadotropic ovarian failure. Cell. 1995;82:959–68. 73. Sasano H, Fukunaga M, Rojas M, Silverberg SG. Hyperthecosis
54. Lourenço D, Brauner R, Lin L, et al. Mutations in NR5A1 asso- of the ovary. Clinicopathologic study of 19 cases with immuno-
ciated with ovarian insufficiency. N Engl J Med. 2009;360: histochemical analysis of steroidogenic enzymes. Int J Gynecol
1200–10. Pathol. 1989;8:311–20.
55. Levine MA, Downs RW, Moses AM, Breslau NA, Marx SJ, 74. Snowden JA, Harkin PJ, Thornton JG, Wells M. Morphometric
Lasker RD, Rizzoli RE, Aurbach GD, Spiegel AM. Resistance to assessment of ovarian stromal proliferation—a clinicopathologi-
multiple hormones in patients with pseudohypoparathyroidism. cal study. Histopathology. 1989;14:369–79.
Association with deficient activity of guanine nucleotide regula- 75. Rabban JT, Barnes M, Chen L-M, Powell CB, Crawford B,
tory protein. Am J Med. 1983;74:545–56. Zaloudek CJ. Ovarian pathology in risk-reducing salpingo-
56. Aittomäki K, Herva R, Stenman UH, Juntunen K, Ylöstalo P, oophorectomies from women with BRCA mutations, emphasizing
Hovatta O, de la Chapelle A. Clinical features of primary ovarian the differential diagnosis of occult primary and metastatic carci-
failure caused by a point mutation in the follicle-stimulating hor- noma. Am J Surg Pathol. 2009;33:1125–36.
mone receptor gene. J Clin Endocrinol Metab. 1996;81:3722–6. 76. Brunham RC, Gottlieb SL, Paavonen J. Pelvic inflammatory dis-
57. Rogenhofer N, Pavlik R, Jeschke U, Wypior G, Ochsenkühn R, ease. N Engl J Med. 2015;372:2039–48.
Thaler CJ. Effective ovarian stimulation in a patient with resistant 77. Wiesenfeld HC, Hillier SL, Meyn LA, Amortegui AJ, Sweet
ovary syndrome and antigonadotrophin antibodies. Am J Reprod RL. Subclinical pelvic inflammatory disease and infertility. Obstet
Immunol. 2015;73:185–91. Gynecol. 2012;120:37–43.
58. Koninckx PR, Brosens IA. The “gonadotropin-resistant ovary” 78. Furuya M, Murakami T, Sato O, Kikuchi K, Tanaka S, Shimizu
syndrome as a cause of secondary amenorrhea and infertility. M, Yoshiki T. Pseudoxanthomatous and xanthogranulomatous sal-
Fertil Steril. 1977;28:926–31. pingitis of the fallopian tube: a report of four cases and a literature
59. Russell P, Bannatyne P, Shearman RP, Fraser IS, Corbett review. Int J Gynecol Pathol. 2002;21:56–9.
P. Premature hypergonadotropic ovarian failure: clinicopathologi- 79. Kim SH, Kim SH, Yang DM, Kim KA. Unusual causes of tubo-
cal study of 19 cases. Int J Gynecol Pathol. 1982;1:185–201. ovarian abscess: CT and MR imaging findings. Radiographics.
60. Sheehan MT. Polycystic ovarian syndrome: diagnosis and man- 2004;24:1575–89.
agement. Clin Med Res. 2004;2:13–27. 80. Burkman R, Schlesselman S, McCaffrey L, Gupta PK, Spence
61. Strauss JF. Some new thoughts on the pathophysiology and M. The relationship of genital tract actinomycetes and the devel-
genetics of polycystic ovary syndrome. Ann N Y Acad Sci. opment of pelvic inflammatory disease. Am J Obstet Gynecol.
2003;997:42–8. 1982;143:585–9.
62. Varanasi LC, Subasinghe A, Jayasinghe YL, Callegari ET, 81. Evans DT. Actinomyces israelii in the female genital tract: a
Garland SM, Gorelik A, Wark JD. Polycystic ovarian syndrome: review. Genitourin Med. 1993;69:54–9.
prevalence and impact on the wellbeing of Australian women aged 82. Boyle DP, McCluggage WG. Combined actinomycotic and pseu-
16-29 years. Aust New Zeal J Obstet Gynaecol. 2017; https://doi. doactinomycotic radiate granules in the female genital tract:
org/10.1111/ajo.12730. description of a series of cases. J Clin Pathol. 2009;62:1123–6.
3 Benign Diseases of the Ovary 119
83. Nogales-Ortiz F, Tarancón I, Nogales FF. The pathology of 104. Lee MS, Moon MH, Woo H, Sung CK, Jeon HW, Lee
female genital tuberculosis. A 31-year study of 1436 cases. Obstet TS. Ruptured corpus luteal cyst: prediction of clinical outcomes
Gynecol. 1979;53:422–8. with CT. Korean J Radiol. 2017;18:607–14.
84. Chow TWP, Lim BK, Vallipuram S. The masquerades of female 105. Millet J, Much M, Gunabushanam G, Buza N, Schwartz PE,
pelvic tuberculosis: case reports and review of literature on Scoutt LM. Large ovarian calcifications from an unresorbed cor-
clinical presentations and diagnosis. J Obstet Gynaecol Res. pus albicans. J Ultrasound Med. 2012;31:1465–8.
2002;28:203–10. 106. Harrison BT, Berg RE, Mittal K. Massive ovarian edema associ-
85. Mahdavi A, Malviya VK, Herschman BR. Peritoneal tuberculo- ated with a broad ligament leiomyoma: a case report and review.
sis disguised as ovarian cancer: an emerging clinical challenge. Int J Gynecol Pathol. 2014;33:418–22.
Gynecol Oncol. 2002;84:167–70. 107. Harke AB, Sigamani K, Thukkaram C, Ramamurthy M, Sekar
86. Bhanothu V, Theophilus JP, Rozati R. Use of endo-ovarian tissue M. Massive ovarian oedema—a case report. J Clin Diagn Res.
biopsy and pelvic aspirated fluid for the diagnosis of female geni- 2016;10:ED03–4.
tal tuberculosis by conventional versus molecular methods. PLoS 108. Dahmoush H, Anupindi SA, Pawel BR, Chauvin NA.
One. 2014;9:e98005. Multimodality imaging findings of massive ovarian edema in chil-
87. Morrison JC, Givens JR, Wiser WL, Fish SA. Mumps oophoritis: dren. Pediatr Radiol. 2017;47:576–83.
a cause of premature menopause. Fertil Steril. 1975;26:655–9. 109. Machairiotis N, Stylianaki A, Kouroutou P, et al. Massive ovar-
88. Subietas A, Deppisch LM, Astarloa J. Cytomegalovirus oophori- ian oedema: a misleading clinical entity. Diagn Pathol. 2016;
tis: ovarian cortical necrosis. Hum Pathol. 1977;8:285–92. 11:18.
89. Tutschka BG, Lauchlan SC. Endosalpingiosis. Obstet Gynecol. 110. Praveen R, Pallavi V, Rajashekar K, Usha A, Umadevi K, Bafna
1980;55:57S–60S. U. A clinical update on massive ovarian oedema - a pseudotu-
90. Kaneda S, Fujii S, Nosaka K, Inoue C, Tanabe Y, Matsuki T, mour? Ecancermedicalscience. 2013;7:318.
Ogawa T. MR imaging findings of mass-forming endosalpingiosis 111. Callen AL, Illangasekare T, Poder L. Massive ovarian edema, due
in both ovaries: a case report. Abdom Imaging. 2015;40:471–4. to adjacent appendicitis. Emerg Radiol. 2017;24:215–8.
91. Li J, Abushahin N, Pang S, Xiang L, Chambers SK, Fadare O, 112. De Cecio R, Cantile M, Fortunato N, De Chiara A, Losito NS,
Kong B, Zheng W. Tubal origin of “ovarian” low-grade serous Franco R, Botti G. Massive edema of ovary with cytogenetic
carcinoma. Mod Pathol. 2011;24:1488–99. alteration of chromosome 12q13-15 in adolescent patient: a case
92. Banet N, Kurman RJ. Two types of ovarian cortical inclusion report. J Ovarian Res. 2013;6:13.
cysts. Int J Gynecol Pathol. 2015;34:3–8. 113. Sasaki KJ, Miller CE. Adnexal torsion: review of the literature. J
93. Esselen KM, Ng S-K, Hua Y, White M, Jimenez CA, Welch WR, Minim Invasive Gynecol. 2014;21:196–202.
Drapkin R, Berkowitz RS, Ng S-W. Endosalpingiosis as it relates 114. Childress KJ, Dietrich JE. Pediatric ovarian torsion. Surg Clin
to tubal, ovarian and serous neoplastic tissues: an immunohisto- North Am. 2017;97:209–21.
chemical study of tubal and Müllerian antigens. Gynecol Oncol. 115. Asfour V, Varma R, Menon P. Clinical risk factors for ovarian tor-
2014;132:316–21. sion. J Obstet Gynaecol. 2015;35:721–5.
94. Valle RF, Sciarra JJ. Endometriosis: treatment strategies. Ann N Y 116. Spinelli C, Piscioneri J, Strambi S. Adnexal torsion in adoles-
Acad Sci. 2003;997:229–39. cents. Curr Opin Obstet Gynecol. 2015;27:320–5.
95. Cohen I, Figer A, Tepper R, Shapira J, Altaras MM, Yigael D, 117. González-Bosquet E. Ovarian vein thrombosis. Risk factors, diag-
Beyth Y. Ovarian overstimulation and cystic formation in pre- nosis and treatment. Medicina (B Aires). 2009;69:347–9.
menopausal tamoxifen exposure: comparison between tamoxifen- 118. Kominiarek MA, Hibbard JU. Postpartum ovarian vein thrombo-
treated and nontreated breast cancer patients. Gynecol Oncol. sis: an update. Obstet Gynecol Surv. 2006;61:337–42.
1999;72:202–7. 119. Payne JF, Robboy SJ, Haney AF. Embolic microspheres within
96. Bolli P, Schädelin S, Holland-Cunz S, Zimmermann P. Ovarian ovarian arterial vasculature after uterine artery embolization.
torsion in children: development of a predictive score. Medicine Obstet Gynecol. 2002;100:883–6.
(Baltimore). 2017;96:e8299. 120. Young RH, Scully RE. Fibromatosis and massive edema of the
97. Ron-El R, Herman A, Golan A, Raziel A, Soffer Y, Caspi E. Follicle ovary, possibly related entities: a report of 14 cases of fibro-
cyst formation following long-acting gonadotropin-releasing hor- matosis and 11 cases of massive edema. Int J Gynecol Pathol.
mone analog administration. Fertil Steril. 1989;52:1063–6. 1984;3:153–78.
98. Morales C, García-Pardo L, Reymundo C, Bellido C, Sánchez- 121. Lee S, Illei PB, Han JS, Epstein JI. Florid mesothelial hyperplasia
Criado JE, Gaytán F. Different patterns of structural luteoly- of the tunica vaginalis mimicking malignant mesothelioma. Am J
sis in the human corpus luteum of menstruation. Hum Reprod. Surg Pathol. 2014;38:54–9.
2000;15:2119–28. 122. Clement PB, Young RH. Florid mesothelial hyperplasia associated
99. Beyth Y, Klein Z, Tepper R, Weinstein S, Aviram R. Hemorrhagic with ovarian tumors: a potential source of error in tumor diagnosis
corpus luteum is associated with ovarian edema. J Pediatr Adolesc and staging. Int J Gynecol Pathol. 1993;12:51–8.
Gynecol. 2006;19:325–7. 123. McCluggage WG, Young RH. Non-neoplastic granulosa cells
100. Yamakami LYS, de Araujo DB, Silva CA, Baracat EC, de Carvalho within ovarian vascular channels: a rare potential diagnostic pit-
JF. Severe hemorrhagic corpus luteum complicating anticoagula- fall. J Clin Pathol. 2004;57:151–4.
tion in antiphospholipid syndrome. Lupus. 2011;20:523–6. 124. Vydianath B, Ganesan R, McCluggage WG. Displaced granulosa
101. Hatipoglu S, Hatipoglu F, Abdullayev R. Acute right lower cells in the fallopian tube mimicking small cell carcinoma. J Clin
abdominal pain in women of reproductive age: clinical clues. Pathol. 2008;61:1323–5.
World J Gastroenterol. 2014;20:4043–9. 125. Duncan DL, Rambally BS, Lininger RA, DiFurio MJ. Displaced
102. Suh DS, Han SE, Yun KY, Lee NK, Kim KH, Yoon MS. Ruptured granulosa cells in the fallopian tube mistaken for metastatic granu-
hemorrhagic corpus luteum cyst in an undescended ovary: a rare losa cell tumor. Int J Gynecol Pathol. 2013;32:35–7.
cause of acute abdomen. J Pediatr Adolesc Gynecol. 2016;29:e21–4. 126. Kho RM, Abrao MS. Ovarian remnant syndrome. Curr Opin
103. Spinelli C, Di Giacomo M, Mucci N, Massart F. Hemorrhagic Obstet Gynecol. 2012;24:210–4.
corpus luteum cysts: an unusual problem for pediatric surgeons. J 127. Magtibay PM, Magrina JF. Ovarian remnant syndrome. Clin
Pediatr Adolesc Gynecol. 2009;22:163–7. Obstet Gynecol. 2006;49:526–34.
120 D. Suster et al.
128. Pai SA, Desai SB, Borges AM. Uteruslike masses of the ovary 135. Gawron I, Chmura L, Adamek D, Skotniczny K, Pityński K,
associated with breast cancer and raised serum CA 125. Am J Surg Ludwin A. Laparoscopic intervention in acute Crohn’s disease
Pathol. 1998;22:333–7. involving an ovary. J Minim Invasive Gynecol. 2017;24:1040–5.
129. Takeda A, Imoto S, Mori M, Yamada J, Nakamura H. Uterus-like 136. Pereira R, de Carvalho J, de Medeiros A, Silva C, Capelozzi V,
mass of ovarian ligament: image diagnosis and management by Bonfá E. Ovarian necrotizing vasculitis in a patient with lupus.
laparoendoscopic single-site surgery. J Obstet Gynaecol Res. Lupus. 2009;18:1313–5.
2011;37:1895–9. 137. Bell DA, Mondschein M, Scully RE. Giant cell arteritis of the
130. Pueblitz-Peredo S, Luévano-Flores E, Rincón-Taracena R, Ochoa- female genital tract. A report of three cases. Am J Surg Pathol.
Carrillo FJ. Uterus like mass of the ovary: endomyometriosis or 1986;10:696–701.
congenital malformation? A case with a discussion of histogen- 138. Dincsoy HP, Rolfes DB, McGraw CA, Schubert WK. Cholesterol
esis. Arch Pathol Lab Med. 1985;109:361–4. ester storage disease and mesenteric lipodystrophy. Am J Clin
131. Shutter J. Uterus-like ovarian mass presenting near menarche. Int Pathol. 1984;81:263–9.
J Gynecol Pathol. 2005;24:382–4. 139. Mount SL, Eltabbakh GH, Hardin NJ. Beta-2 microglobulin amy-
132. Fukunaga M. Smooth muscle metaplasia in ovarian endometrio- loidosis presenting as bilateral ovarian masses: a case report and
sis. Histopathology. 2000;36:348–52. review of the literature. Am J Surg Pathol. 2002;26:130–3.
133. Zurkova M, Turkova M, Tichy T, Pilka R, Kolek V, Kriegova 140. McCluggage WG, Allen DC. Ovarian granulomas: a report of 32
E. Sarcoidosis of female reproductive organs in a postmenopausal cases. J Clin Pathol. 1997;50:324–7.
woman. Menopause. 2015;22:549–53. 141. Clement PB, Young RH, Scully RE. Necrotic pseudoxanthoma-
134. Parveen A, Elliott H, Howells R. Sarcoidosis of the ovary. J Obstet tous nodules of ovary and peritoneum in endometriosis. Am J Surg
Gynaecol (Lahore). 2004;24:465. Pathol. 1988;12:390–7.
Ovarian Epithelial Carcinogenesis
4
Jing Zhang, Elvio G. Silva, Anil K. Sood, and Jinsong Liu
© Science Press & Springer Nature Singapore Pte Ltd. 2019 121
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_4
122 J. Zhang et al.
Therefore, these two theories have become less popular in 3. The unknown origin of OEI: Some epithelial cells in OEI
recent years. On the other hand, the dualistic model of ovarian exhibit the differentiation toward fallopian tube epithe-
carcinogenesis is among most discussed model in past decade. lium, rather than OSE.
However, ovarian cancer can be potentially have multiple cell 4. Different histomorphology: There are no obvious histo-
origins, and the malignant transformation may be achieved via logical transitions between mesothelial cells and
formation of tetraploidy or polyploidy as an intermediate stage Müllerian epithelium.
and followed by amitosis rather mitosis to generate genetically
aberrant stem cells for cancer initiation. In order to better under- In addition, the histological observations on the ovaries
stand the overall landscape of ovarian carcinogenesis, we will from the carriers with hereditary BRCA mutation or contra-
discuss each of cell origins and mechanisms involved in tumor lateral normal ovary of sporadic EOC reveal that the mor-
progression in light of the most recent research progress. phological changes (hyperplastic papillae on the ovarian
surface, the increase and dilatation of cortical inclusion
cysts, and mild cell atypia) are not sufficient to achieve the
4.1.1 The Ovarian Surface Epithelial Cells diagnostic criteria for precancerous lesions. For all of the
abovementioned reasons, this theory has become less popu-
This theory is based on the speculation that all ovarian epithe- lar in the past two decades.
lial tumors originate from OSE. The OSE represents mesothe-
lial-like cells and is configured as a single layer of stable
epithelium without prominent differentiated characteristics in 4.1.2 The Secondary Müllerian System
the general state. They possess two potentials that differentiate
into mesenchymal cells or epithelial cells, called epithelial- At the early stage of embryogenesis, the somatic epithe-
mesenchymal transition (EMT) or reversal of this process lium and its subepithelial mesenchyme are derived from
named mesenchymal-epithelial transition (MET). As one of the Müllerian ducts. During the embryonic development,
the normal physiological functions, EMT/MET plays a vital the distal parts of the two Müllerian ducts (also known as
role in regulating the OSE repair process after ovarian ovula- the primary Müllerian system) fuse to form the uterus, cer-
tion [2]. The imbalance of EMT/MET is suggested to be a vix, and proximal one third of the vagina, while the proxi-
possible mechanism in the initiation of EOC. In this hypothe- mal Müllerian ducts remain separated to become the two
sis, ovarian epithelial inclusion (OEI)/inclusion cyst (OIC) is fallopian tubes [5]. The mucosal epithelium lining on those
formed via OSE invagination. These OEIs may transform into sites is called the Müllerian epithelium. Indeed, the ovaries
Müllerian epithelial cells via metaplasia and give rise to differ- do not belong to the Müllerian system because the gonads
ent histological types (e.g., serous, endometrioid, clear cell, and reproductive tract are developed separately in embry-
mucinous and transitional cells) under the influence of local onic stage.
factors (such as steroid hormones). Their morphologies are In view of the similarity between ovarian epithelial tumors
parallel to the mucosa of fallopian tube, endometrium, gastro- and Müllerian epithelium, Lauchlan [6] put forward the sec-
intestinal tract or endocervix, and bladder, respectively. These ond Müllerian system theory in 1972, which is that the coe-
OEIs with Müllerian phenotype further gain the ability for lothelium has an ability to transform into the Müllerian
malignant transformation and to progress to corresponding epithelium. Taking this ability into consideration, he further
EOCs (serous carcinoma, endometrioid carcinoma, mucinous suggested that OSE, OEI, and all extraovarian Müllerian-
carcinoma, or other subtype) with lineage infidelity via an type epithelial tissues adjacent to fallopian tube and pelvic
abnormal regulation of homeobox (HOX) gene [3]. In this cavity are part of the second Müllerian system. The second
process of tumor transformation, both OEIs and cancer cells Müllerian system includes endometriosis, endosalpingiosis,
show the changes that cells gradually lose the characteristics and endocervicosis, which are collectively referred to as
of mesenchymal cells and obtain different Müllerian epithelial Müllerianosis. With his comprehensive understanding of the
differentiated features, including the expression of specific morphologic features of these lesions, Dr. Lauchlan thought
epithelial marker (E-cadherin) in the differentiated stage [4]. that these three lesions can change into each other through
However, the theory has been challenged in the several metaplasia and thus that all epithelial tumors in the ovary and
aspects: pelvic cavity can be potentially derived from the second
Müllerian system.
1. Different cells of origin: OSE belongs to the coelothelium Similar to OSE metaplasia theory described above, the
(mesothelial cells), rather than Müllerian epithelium. The secondary Müllerian system theory cannot entirely account
histotypes of EOCs primarily show differentiation toward for other observations on ovarian carcinogenesis and accord-
Müllerian, rather than mesothelial cells. ingly has become less popular in recent years. However, it
2. Discrepancy in immunophenotype: OSE does not express remains a potential cell of origin for extraovarian or pelvic
common EOC markers (such as PAX-8), but highly Müllerian epithelial tumors, especially low-grade lesions
express calretinin and the other mesothelial markers. (type I) as described below.
4 Ovarian Epithelial Carcinogenesis 123
4.1.3 The Dualistic Model Table 4.1 The comparison between type I and type II epithelial ovar-
ian carcinomas [9]
In recent years, data have accumulated on the histological Type I epithelial ovarian Type II epithelial
carcinomas ovarian carcinomas
observation and molecular genetic levels demonstrated that
Histologic features
EOCs are heterogeneous diseases with several histological
Histological Low-grade serous High-grade serous
subtypes with different cells of origin, pathogenesis, and types carcinoma, carcinoma,
clinical biological features. Kurman and colleagues pro- endometrioid undifferentiated
posed the “dualistic model” of ovarian carcinogenesis, i.e., carcinoma, clear cell carcinoma, malignant
carcinoma, mixed Müllerian
EOC could be classified as type I and type II tumors based on
seromucinous tumors
their distinct set of clinicopathologic features [7–9]. carcinoma, mucinous
Type I EOCs include low-grade serous carcinoma, endo- carcinoma, malignant
metrioid carcinoma, clear cell carcinoma, seromucinous car- Brenner tumors
cinoma, mucinous carcinoma, and malignant Brenner Tumor grade Low-grade (except High-grade
clear cell carcinoma)
tumors. In the genesis, these carcinomas often follow a
Proliferation Usually low Usually high
sequential pattern of evolution from benign to borderline to activity
malignant tumors. Clinically, these tumors grow slowly and Clinical features
most of them have an indolent biological behavior; most are FIGO stage Usually early stage Usually advanced
confined to one ovary at presentation. All histological sub- (FIGO I) stage (FIGO III and
types of type I EOCs are low-grade tumors, and the progno- FIGO IV)
Clinical process Slow and indolent Rapid and aggressive
sis is relatively good, with the exception of clear cell
Response to General Good (but late
carcinoma. The mortality rate accounts for only 10% of all chemotherapy recurrence)
EOCs. The tumor genomes are relatively stable at the molec- Progress course Benign to borderline to Mostly from serous
ular genetic level, although there are different genotypes in malignant tumors tubal intraepithelial
different histological types (see Table 4.1) [9]. carcinoma
Type II EOCs include high-grade serous carcinoma, Early screening Feasible Difficult
Prognosis Relatively good Relatively poor
undifferentiated carcinoma, and malignant mixed Müllerian
Genetic features
tumors (MMMT, also called carcinosarcoma). Clinically,
Chromosomal Low High
these tumors are highly aggressive and rapidly progressive. instability
All histological subtypes of type II EOCs belong to high- Common gene KRAS, BRAF, PTEN, TP53, BRCA1/2
grade tumors. More than 75% of cases are diagnosed at mutation ARID1A, PIK3CA,
advanced stage (FIGO III and IV) with extensive dissemina- CTNNB1, ERBB2,
PPP2R1A
tion. The prognosis is poor and the mortality rate accounts
Deficiency of Rare Common
for 90% of all EOCs. In the molecular genetics, these tumors homologous
have highly unstable genomes and prone to have the amplifi- recombination
cation or deletion of DNA copy numbers. Among them, repair proteins
TP53 gene mutation is most common (>95% of high-grade
serous carcinoma) (Table 4.1) [9].
could be the origin of high-grade serous carcinoma, and both
stromal-epithelial interaction and steroid hormones play critical
4.1.4 The Fere Ex Nihilo Model roles in tumorigenesis and progression [11].
The stem cells are defined as a subgroup of cells with dif-
The dualistic model of ovarian carcinogenesis helps in our ferentiation potential and the ability for self-renewal. Under
understanding of EOCs and may even provide a frame work to certain conditions, these cells can differentiate into a variety of
guide clinical decision-making. However, this model may be functional cells. It has been reported that both OSE and ovar-
too simplified for such a highly heterogeneous group of dis- ian cancer cell are capable of expressing stem cell markers
eases. For example, even among high-grade serous carcinomas, such as SOX2 [12], CD133 [13], and NANOG [14]. The small
the tumors have different clinical biological behaviors [10]. In populations of stem cells, which possess the features of stem
particular, Silva raised several questions that argue against fal- cell or mimic stem cells, are predominantly located in hilar
lopian tube theory for pelvic serous carcinoma (see Sect. OSE in mouse model. The hilum OSE are cycling slowly and
4.2.1.2). Toward this end, Silva put forward the fere ex nihilo express stem cell markers ALDH1, LGR5, LEF1, CD133, and
model (or out of nothing) from unremarkable primitive or early CK6B [15]. Therefore, those OSE cells or other stromal cell
epithelial or mesenchymal stem cells: this model hypothesizes types that express stem cell markers can potentially be the cell
that uncommitted or stem cells from the mesenchyme could be of origin of benign and malignant ovarian tumors [16, 17].
a potential source of transformation for both benign and malig- Evidence supporting this view includes the development
nant tumors. During this process, stromal fibroblasts, via MET, of benign epithelial neoplasms of the ovaries from guinea
124 J. Zhang et al.
pigs following treatment of testosterone and estrogenic hor- of mitotic spindle. There are two kinds of endoreplication.
mones [18, 19]. Inflammation is a well-known risk factor for The first form is called the endocycle, which consists of
ovarian cancer, as may occur in ovulation, which causes the alternating DNA synthesis (S) phases and gap (G) phases
rupture of OSE and a repair process, resulting in pelvic foci of without chromosome segregation during a mitotic (M) phase
inflammatory microenvironment. Using SV40 T/t antigen to or cell division (cytokinesis). The developmentally con-
disable to ovarian surface epithelial cells together with onco- trolled endocycle results in cells with a single polyploid
gene RAS, Liu’s laboratory has successfully been able to nucleus and no feature of mitosis to support specific need of
transform the normal ovarian epithelial cells into high-grade development [35–39]. Another form of endoreplication is
Müllerian carcinoma, which is associated with massive known as endomitosis, in which cells execute an abortive
upregulation of inflammatory cytokine (e.g., interleukin-1β, mitosis that does not result in fully separate sister chromatids
interleukin-6, and interleukin-8) [20]. Through the chemo- or cell division, followed by subsequent re-entering of S
kines and cytokines secreted by cancer cells, for example, phase to generate multinucleated cells [36–39]. Endocycle
Gro-1 can induce stromal fibroblast senescence [21]. CXCR2 and endomitosis can be mixed, and the distinction between
and interleukin-1β promote tumor cell proliferation [22, 23]. the two forms may be context development depending on
The close interaction among tumor cells, stromal cells, and specific type of development.
inflammatory tumor microenvironment provides a favorable Polyploid giant cancer cells (PGCCs), characterized by a
condition for tumor cell recruitment, adhesion, migration, sur- single, giant nucleus or multinucleated cells, are commonly
vival, and colonization [24, 25]. The fact that OSE can be trans- found in tumor tissues with high-grade carcinoma or after
formed into high-grade Müllerian carcinoma provides further treatment (such as chemotherapy, radiotherapy, or targeted
support that the OSEs can be also the potential cells of origin for therapy) [40–42]. PGCCs have a distinct advantage over regu-
high-grade carcinoma. lar cancer cells in dealing with stresses (e.g., hypoxia, starva-
tion, temperature, pH, and diet conditions in physiologic
stresses; drugs and radiation in pathologic stresses) and repro-
4.1.5 Somatic Blastomere Model duction [35–39]. Increasing DNA content by endoreplication
is a widely utilized effective mechanism to sustain the mass
The above theories described the cell of origin for ovarian production of proteins and high metabolic activity necessary
cancer. However, it remains unknown how the cells are trans- for tumor growth. Following endoreplication, cancer cells
formed into carcinoma. The most accepted paradigm in car- may thus arrest the mitotic cell cycle and allow the cell to sur-
cinogenesis is an accumulation of genetic mutations or vive during mitotic catastrophe or genotoxic stresses and enter
aneuploidy [26, 27]. However, these theories at the individ- endoreplication cell cycle to form PGCCs [30, 43–45].
ual gene or individual chromosomal levels cannot entirely Accumulating evidence suggests that PGCCs may have
account for enormous genomic and epigenetic changes played a fundamental role in tumor initiation. They may have
detected by The Cancer Genome Atlas (TCGA) projects in hijacked normal embryonic developmental program to facili-
high-grade carcinomas [28]; it has been argued that somatic tate the generation of new diploid cancer initiating cells in
mutation theory may be wrong for most cancer [29]. In addi- response to oncogenic and therapeutic stress [30, 45]. Our
tion, all of the four models described above fail to consider laboratory has provided the first experimental evidence that
the level of differentiation as reflected in the level of malig- normal ovarian or fallopian tubal epithelial cells and cancer
nancy in different histotypes. Recently, we have proposed a cells can undergo endoreplication [42, 46]. This process can
unified somatic blastomere model to explain the origin of all lead to genomic instability and dedifferentiation into more
cancers and disease relapse [30]. This model is based on primitive state, to facilitate the reprogramming and emer-
long-term and puzzling observation that early blastomere gence of new cancer initiating cells. We have shown that this
stage embryo is highly chaotic [31] with high frequency of multistep reprogramming process includes four distinct but
polyploidy [32, 33]. The endoreplication and cell fusion are overlapping process including initiation, self-renewal, termi-
required for development from blastomere to compaction/ nation, and stability and facilitating normal or cancer cells to
morula stage embryo, which is required for the first differen- be reprogrammed to cancer cells or resistant cancer cells
tiation event to become trophoblasts and inner cell mass fol- [45]. The mitotic apparatus for well-known mitotic division
lowing fertilization [34]. Unlike the mitotic cell cycle, which is shut down with activation of senescence program in the
involves several distinct phases including DNA synthesis (S) giant cell cycle; emergence of new tumor initiating cells is
and distribution of replicated DNAs to two identical daugh- largely from amitotic separation from giant mother cells
ter cells via mitosis (M) with the intervening gap phase (G), including budding, splitting, and branching, the more primi-
endoreplication represents a specific process in which tive mode of cell division used in fungi [42, 45]. Unexpectedly,
nuclear membrane does not break down while the genome is endoreplication of ovarian cancer cells recapitulates the divi-
replicated twice or multiple times without cell division and sion and growth pattern of blastomere stage embryo to form
subsequently separated into daughter cells without formation compaction and morulae-like embryonic cell types, which is
4 Ovarian Epithelial Carcinogenesis 125
associated with massive mitotic and cytokinesis failure and tumor). The closer toward the embryonic stage, the higher
genomic instability [30]. Moreover, the endoreplicating cells developmental potential to allow the tumor to behave in
recapitulate the morphology and spatial and time-dependent malignant manner [30]. Thus, it is possible that high-grade
activation of embryonic reprogramming factor OCT4/ serous carcinoma or the other EOCs, particularly those can-
NANOG/SOX2, capable of differentiation into three germ cers with marked nuclear atypia, may be achieved via the
layers and develop into malignant germ cell tumors [30]. giant cell cycle-mediated reprogramming following the re-
Formation of tetraploidy or polyploidy is a common feature differentiation and followed by developmental arrest [30, 42,
at the border of normal epithelial cells and mesenchymal and 45]. This model also offers a sensible explanation why high-
high grade carcinoma, which is associated with activation of grade serous carcinoma is usually detected in late stage with
senescence and dedifferentiation program and stem cell acti- wide dissemination in the peritoneal cavity. The schematic
vation, supporting the generality of our model to other types diagram on how epithelial or mesenchymal cell is trans-
of cancer [47]. Further, formation of polyploidy appears to formed into cancer cells via the giant cell cycle is shown in
be a major mechanism in response to starvation or mitotic Fig. 4.1. The details on the role of the PGCCs and the giant
insult in Drosophila [48]. Subsequent generation of intestinal cell cycle in tumor initiation and disease relapse can be found
stem cells from polyploidy cells is associated amitosis, a in recent review by Liu [47].
primitive form of cell division without using mitotic spindle
[48]. Taken together, the above data together provide a previ-
ously appreciated mechanism via formation of polyploid 4.2 he Cell Origin and Molecular Genetic
T
cells for generating genetically altered daughter stem cells in Profiles
response to acute or chronic insults using primitive mode of
cell division for cancer initiation [47]. During the past decade, it has become clear that some high-
Our model also explains the level of differentiation grade serous carcinomas arise from the mucosal epithelium
observed in the ovarian cancer. Depending on the level of of fallopian tube fimbria [49–51]. In addition, significant
stem cell arrested at the specific developmental hierarchy progress has also been made in the cell origin of the other
during organ development, the tumors could behave in subtypes of EOCs. These progresses will help out our under-
benign or low-grade type of malignancy such as cystade- standing on the origin of EOC and offer a new potential strat-
noma or borderline tumor or high-grade carcinoma (type II egy for treatment and early screening.
a b1 b2 c d
Fig. 4.1 A schematic diagram of somatic blastomere-like model for facilitate the genomic reorganization and reprogramming. With the ter-
how normal fallopian tubal and ovarian epithelial cells are transformed mination of the giant cell cycle via endomitosis to form early tumor
into high-grade serous carcinoma. Normal fallopian tubal cells (or ovar- papillae (c), multiple tumor papilla with different genetic and epigenetic
ian epithelial cells or stromal fibroblasts) (a) the nucleus starts endorep- changes together generated via amitotic endoreplication division. In this
lication due to mitotic/cytokinesis failure; (b) endoreplicating cells grow process, the clone(s) with advantageous p53 mutations achieve the sta-
autonomously (self-renewal, b1 and b2) and lead to genomic chaos and ble tumor expansion and develop into high-grade serous carcinoma (d)
126 J. Zhang et al.
4.2.1 Ovarian Serous Carcinoma directly detach and adhere to the ovarian surface, it is possible
that the tubal OEIs may be formed via the shedding and
Serous carcinoma is the most common histological subtype invagination of tubal mucosal epithelium into the ovarian cor-
of ovarian epithelial tumors. It represents a heterogeneous tex. Interestingly, the different proportional distribution of
group of tumors at both morphology and molecular genetics mixed ciliated and secretory cells has been found in the tubal
and exhibits a typical dualistic model of ovarian carcinogen- OSEs, tube OEIs, serous cystadenomas, and borderline
esis. In 2004, based on the systematic analysis of histology tumors, with a remarkably increasing secretory/ciliated cell
and clinical behavior with more than 10-year follow-up, ratio in the low-grade serous carcinoma, suggesting that tubal
Malpica et al. [40] at MD Anderson Cancer Center first epithelial cells could be the potential origin of these tumors.
proposed a two-tier system for grading ovarian serous carci- In the 2014 WHO classification, ovarian borderline serous
noma. Namely, ovarian serous carcinoma is classified as tumors are further divided into two types, namely, serous
low-grade and high-grade. This grading system is currently borderline tumor/atypical proliferative serous tumor and
widely used and is accepted by the WHO classification of borderline serous tumor-micropapillary variant/noninvasive
gynecological tumors [52, 53]. Although they are in the same low-grade serous carcinoma. Both of them have the same
category of serous carcinoma, low-grade serous carcinoma is KRAS mutation rate (about 50% of cases); however, there is
significantly different from high-grade serous carcinoma in a more analogous genetic profile between borderline serous
morphology, genotyping, and biological behavior. tumor-micropapillary variants and low-grade serous carci-
noma in comparison with serous borderline tumor, which
4.2.1.1 Low-Grade Serous Carcinoma suggests that borderline serous tumor-micropapillary variant
Low-grade serous carcinoma has typical features of type I may be an intermediate lesion in the development of border-
EOCs. The progression from benign to borderline to malig- line serous tumor to low-grade serous carcinoma [52]. On
nant process can be observed histologically. The epithelial the other hand, development of low-grade serous carcinoma
cells lining on the serous cystadenoma have the same histo- from serous borderline tumor, regardless conventional type
logical and immunophenotype as OEI cells. The areas of or micropapillary variants, is time-dependent [56]. With
benign serous components present in almost all borderline increasing follow-up time, conventional serous borderline
tumors. Likewise, the transition from borderline to malig- tumor will also develop into low-grade serous carcinoma. It
nant components also exists in the majority of low-grade remains to be determined whether such further subclassifica-
serous carcinoma. As the precursor lesion represents an tion is clinically meaningful or potentially misleading as the
important clue for tumor origin, the abovementioned patho- term of atypical proliferative tumor neglects the low malig-
logical findings suggest that the ovarian serous cystadenoma, nant potential of conventional serous borderline tumor.
borderline tumor, or low-grade carcinoma may develop via a Genetic and genomic profiles: The most common molecular
sequential progression process from OEIs. genetic alterations in low-grade serous carcinoma are KRAS,
Cells of origin: While it is well-known that the fallopian BRAF, or ERBB2 mutations. These three gene mutations can
tube can serve as a precursor lesion for high-grade serous promote the transduction of growth signals into the nucleus,
carcinoma, recent evidence suggests that these epithelial resulting in uncontrollable cell proliferation and malignant
cells can be also potential source of low-grade lesions includ- transformation via sustained activation of the downstream
ing cystadenoma, serous borderline tumor, and low-grade MAPK kinase signaling pathway [57, 58]. Accumulating evi-
serous carcinoma. Due to close anatomic relationship dence points toward that there are mutually exclusive relation-
between the fallopian tubal fimbria and ovaries, ovarian rup- ships among KRAS, BRAF, or ERBB2 mutations with only one
ture caused by periodic ovulation may provide an opportu- gene mutation exists at an individual tumor in most cases. The
nity for implantation of the epithelial cells of fallopian tube mutation rate accounts for about 2/3 of the borderline tumors
on the ovary. Through morphological and immunohisto- and low-grade serous carcinoma, of which 33.3% of the bor-
chemical comparisons among OSEs, OEIs, tubal epithelium, derline tumors and low-grade serous carcinomas have KRAS
serous cystadenomas, serous borderline tumors, and low- mutations at codon 12 and 13, 33.3% of serous borderline
grade serous carcinomas, Zheng and Kong et al. [23, 54, 55] tumors and a small number of low-grade serous carcinomas
found two types of OEIs including mesothelial type (cal- show BRAF mutation at codon 600, whereas ERBB2 mutation
retinin+/PAX8−/tubulin–) and tubal type (calretinin–/ is less than 5% of entire tumors. KRAS or BRAF mutation is
PAX8+/tubulin+). The tubal OEIs account to 78% of all OEIs considered to be an early event in low-grade serous carcinomas
and have a significantly higher proliferative index, which may because they also exist in serous cystadenomas adjacent to bor-
further proliferate and progress to ovarian serous cystade- derline serous tumors [59]. Compared with BRAF mutation,
noma or borderline tumors. Conversely, mesothelial OEIs serous borderline tumors with KRAS mutation have a greater
may not develop to tumor because of its low proliferative rate. potential to progress to low-grade serous carcinoma. In fact,
As the mucosal epithelium of fallopian tubal fimbria can BRAF mutation in advanced low-grade serous carcinoma is
4 Ovarian Epithelial Carcinogenesis 127
a b c d f
Fig. 4.2 The illustration of progression of ovarian low-grade serous grade serous carcinoma (f), with/without the stage of borderline serous
carcinoma and associated changes in molecular genetics. The mucosal tumor-micropapillary variants/noninvasive low-grade serous carcinoma
epithelial cells (a) of fallopian tubal fimbria may shed and adhere to the (e). During the period of tumorigenesis and progression, the KRAS/
ovarian surface and then form the tubal OEIs (b, yellow arrow). The BRAF/ERBB2 mutation rate gradually increases. And the multiple chro-
progression process is from serous cystadenoma (c) to borderline serous mosomal allelic imbalances also promote the formation of low-grade
tumor/atypical proliferative serous tumor (d) and eventually to low- serous carcinoma simultaneously
very rare [60]. In addition, the low-grade serous carcinomas are ing evidence accumulated in the literatures that provide the
more likely to have an allele imbalance of 1p, 5q, 8p, 18q, 22q, support for the origin of high-grade serous carcinoma. (1) In
and Xp chromosomes than borderline tumors [61]. The hetero- the specimens of prophylactic bilateral salpingectomy from
zygosity deletion of ch1p36 and the heterozygosity/homozy- the patients with hereditary BRCA1/2 mutation carriers,
gous deletion of ch9p21 often are found in low-grade serous there are the occult precancerous lesions that are p53 signa-
carcinoma. Given that there are tumor suppressor genes (such ture and STICs, but no malignant ovarian tumors. The p53
as miR-34a) in ch1p36 region and CDKN2A/B (encoding signature is arbitrarily defined as more than 12 successive
tumor suppressor protein p14, p16, and p15) in ch9p21 region, secretory cells with benign morphological features exhibit-
these deletions may lead to the uncontrollable cell growth in ing strongly positive expression of p53 immunohistochemi-
borderline tumors and eventually progress to low-grade serous cal staining and less than 10% of Ki-67 cell proliferation
carcinomas [62] (Fig. 4.2). index [65]. As for STIC, its cytological features are enlarged,
polymorphic, and hyperchromatic nuclei with nucleoli and
4.2.1.2 High-Grade Serous Carcinoma mitotic figures, high ratio of nucleus to cytoplasm, multi-
High-grade serous carcinoma is the most common EOC. Its layered cells, or the lack of cellular polarity. The immunohis-
incidence is about 60–80% of all EOCs. More than 75% of tochemical staining of p53 exhibits strongly positive or
tumors are in advance stage with extensive abdominopelvic totally negative expression, and Ki-67 cell proliferation
dissemination at the diagnosis [63]. However, there are usu- index is more than 10% [51]. (2) There are tubal p53 signa-
ally no morphologically recognizable precursor lesions in ture and/or STICs in 50–60% of cases with the sporadic
ovarian tissues. Recent studies showed that some ovarian ovarian and/or peritoneal high-grade serous carcinoma [62].
high-grade serous carcinomas may not originate in ovarian (3) Similar to ovarian high-grade serous carcinomas, there
tissues but in the seeding from serous epithelial tumor cells are the overexpression of p53 protein and the mutation of
in the mucosal epithelium of fallopian tubal fimbria as sec- TP53 gene both in p53 signature and STICs. The TP53 muta-
ondary tumors. The spectrum of tumorigenesis and progres- tion rate gradually increases with the process of tumor pro-
sion is from tubal secretory cell expansion [64] or secretory gression from p53 signature to STICs to high-grade serous
cell outgrowths to tubal p53 signature, to tubal serous tubal carcinoma. (4) The evidence that there is the same TP53
intraepithelial carcinoma (STIC)/noninvasive high-grade mutation site in concurrent p53 signature, STICs, and ovar-
serous carcinoma, to shedding and implantation on the ovary, ian high-grade serous carcinoma supports the notion that
and eventually to ovarian high-grade serous carcinoma. high-grade serous carcinoma arises from the clonal prolifer-
Evidence supporting the fallopian tubal epithelial origin ation of p53 signature cells [66, 67]. (5) The genetic profile
as the main source of pelvic serous carcinoma: The follow- of high-grade serous carcinoma is close to tubal epithelial
128 J. Zhang et al.
cells [68]. (6) STICs have shorter telomeres compare with 5. No STICs in the fallopian tube are identified in 50%
concurrent ovarian high-grade serous carcinoma, while telo- serous carcinoma; it will be very difficult to unify a the-
mere shortening is known to be an early molecular event of ory while no precursor lesions are identified in 50% of
tumorigenesis [69]. (7) In the models of genetically modified cases of the same cancer.
mice, the secretory cells in the mucous of fallopian tube can 6. The main molecular evidence of clonality of the same
transform to malignant lesion due to Tp53, Pten, and Brca TP53 mutation should be treated with caution as it has
mutations, resulting in STIC and ovarian high-grade serous been shown to be an unreliable indicator of metastasis in
carcinoma [70, 71]. (8) A recent study confirmed that the many other tumors.
salpingectomy is an effective method to reduce EOC risk in
the general population based on the analysis of large clinical Ovarian serous carcinoma as multicellular cells of origin:
database between 1973 and 2009 [72]. As mentioned above, there are approximately half of the
Based on above data, the secretory cells in the mucosa of cases of ovarian high-grade serous carcinoma without con-
fallopian tube have been proposed to be the cell origin of current STIC, suggesting that these tumors may be derived
high-grade serous carcinoma via the following process [73]: from other cell origins [75, 76]. One possible source of cell
the secretory cells have DNA damages that cannot be nor- origin is the OEIs within ovarian cortex; some OEIs may be
mally repaired due to the stimulation of a variety of DNA formed by direct shedding and implantation of normal fal-
toxicity factors, resulting in the accumulation of DNA dam- lopian tubal epithelial cells into the ovarian cortex [55, 77],
ages. Because of the pressure of survival, there are a series of which undergo a series of molecular genetics and morpho-
molecular genetic alterations in the damaged cells, such as logical change including TP53 mutations, and eventually
adaptive TP53 mutations, which lead to uncontrollable cell develop into high-grade serous carcinomas [78]. In addition,
growth, over-proliferation, secretory cell outgrowths, and it is documented that a small number of low-grade ovarian
then p53 signature. Some cells with p53 signature can serous carcinoma can progress into high-grade serous carci-
develop to STIC directly or through tubal dysplasia to form noma or that borderline serous tumors can progress to high-
STIC [65, 74]. Due to the close contact between the fallopian grade serous carcinoma after tumor recurrence. The incidence
tubal fimbria and the ovarian surface, and the loose adhesion by this pathway is likely to be low, since less than 3% of
between STIC cells, it is possible the tumor cells shed and high-grade serous carcinomas show concurrent serous bor-
implant on the ovarian surface and ultimately form “ovarian” derline tumor, low-grade carcinoma, and high-grade compo-
high-grade serous carcinoma, a possibility that is more plau- nents in same tumor. These low-grade serous carcinomas
sible than ovarian metastasis via lymphovascular invasion. may progress to high-grade carcinoma possibly via TP53
Recent studies that there are tumor cells in the intraperito- mutations [79]. In addition, some high-grade serous carci-
neal washings in some patients with STIC also provide a noma has been observed to arise from adenofibroma; and the
direct evidence for this disseminated implantation [51]. fibroma components cannot be explained by tubal implanta-
Evidence against fallopian tubal epithelial cells as main tion theory. Based on the clinical and pathologic observation
source of pelvic serous carcinoma: Despite of prevalent view and arguable evidence for fallopian tube theory, in his alter-
of fimbria of fallopian tube as a major source of high-grade native view article, Silva proposed that the fere ex nihilo
serous carcinoma discussed above, there are several impor- model including the multicentric cell origin under the influ-
tant clinical and pathological observations of pelvic serous ence of environmental factors (such as hormones) may be the
carcinoma that cannot be clearly explained by fallopian tube responsible for development of various ovarian carcinomas
origin [11]: (see Sect. 1.1.4). However, regardless of the cells of origin,
the transformation process can be achieved via the giant cell
1. Most of high-grade serous carcinomas are at stages III cycle described above (Fig. 4.1).
and IV, while 70% the fallopian tubal carcinoma are stage Subclassification of high-grade serous carcinoma.
I or II. Through the morphological analysis of high-grade serous
2. There is no direct evidence that the cells from small tubal carcinoma-associated BRCA1/2 mutation, Soslow et al. [80]
intraepithelial carcinoma can travel all the way against found that these tumors usually present with solid pattern,
gravity to the abdomen rather to the pelvis, which is endometrioid carcinoma- and transitional cell carcinoma-
against the law of gravity. like feature, which is characterized by “SET” (solid, pseudo
3. Most STICs are noninvasive, while most high-grade
endometrioid, transitional cell carcinoma-like). BRCA1-
serous carcinomas are highly invasive; it is difficult to associated cancer is associated with highly cellular prolifera-
rationalize that the noninvasive precursor carcinomas tive activity, tumor-infiltrating lymphocytes, and geographic
give rise to 70% invasive carcinoma in peritoneal cavity. or comedo necrosis. Compared with the classic high-grade
4. Many high-grade serous carcinomas are located intra- serous carcinoma whose histological patterns are papillary,
ovarian stromal, not on the surface. glandular, cribriform, and solid area with slit-like space, the
4 Ovarian Epithelial Carcinogenesis 129
SET subtype is more common in younger women, and the ular genetic alterations also participate in the transformation
tumor cells are more sensitive to chemotherapy due to defi- process. Norquist et al. [88] found that the loss of BRCA1/2
ciencies of homologous recombination. Moreover, the allele exists in STICs, but not in p53 signature, suggesting
patients have better prognosis [80, 81]. In view of the above- that both TP53 mutation and BRCA1/2 deletion are the key
mentioned different clinicopathological characteristics, it is events in early carcinogenesis. Kim et al. [71] reported that
suggested that high-grade serous carcinoma should be ovarian/fallopian tubal high-grade serous carcinoma cannot
divided into classic subtype and SET subtype [9]. be directly induced by Tp53 mutation alone but by both Tp53
Based on the results of TCGA genome analysis, in 2013 and Pten mutation. Drapkin and Dinulescu et al. [70] also
Verhaak et al. [82] presented a molecular subtype of high- confirmed that the combination of Brca1/2, Tp53, and Pten
grade serous carcinoma associated with the prognosis of mutations in tubal epithelial cells leads to STIC and high-
patients. That is classification of ovarian cancer (CLOVAR). grade serous carcinoma, whereas only Tp53 mutation cannot
The tumors were classified into differentiated, proliferative, induce tumor formation. The simultaneous loss of functions
immunoreactive, and mesenchymal type. Among these sub- leads to a decrease in cell genome stability, a series of onco-
types, the prognosis of patients with immunoreactive type is gene activation and/or tumor suppressor gene inactivation,
the best, and the prognosis is the worst in patients with mes- and leads to multiple chromosomal breaks and deletions and
enchymal type. In 2016 Murakami et al. [83] further associ- the formation of aneuploidy or polyploidy, which is condu-
ated the molecular typing with tumor morphology: the cive to avoiding immune surveillance, over-proliferation, and
mesenchymal type often presents a significant desmoplastic the progression to high-grade serous carcinoma from STIC.
reaction; the immunoreactive type usually manifests as The prevalent view of carcinogenesis of high-grade serous
plenty of lymphocyte infiltration in the tumor tissues; the carcinoma and its main molecular genetic alterations is
proliferative type frequently has the solid growth pattern; shown in Fig. 4.3. Because of conflicting evidence and dif-
and the differentiation type generally exhibits the papillary ferent views that is for or against the tubal theory, we ask
pattern. However, this molecular subtype needs be further readers of this chapter to take the views from different
confirmed by more studies before used clinically. authors with great caution and make your own observation
Common genetic and genomic alterations: The most during the daily practice and decide which theory will best fit
prominent genetic features of high-grade serous carcinoma with what you observe, rather than blindly believe what are
are genomic instability (the abnormalities of many DNA described by a particular author, journal, or academic group.
copy number and structure) and TP53 mutations. According We want to point out that health discussion and expression of
to genomic TCGA project, TP53 mutation is found in almost different opinions should greatly help us to clarify controver-
all tumors tissues (96%) through a sequenced genomic anal- sial points and move field to next level. As Einstein puts it,
ysis on 489 cases of high-grade serous carcinoma [28]. The “blind belief in authority is the greatest enemy of truth.”
TP53 mutation rate is as high as 57% even in precancerous
lesions, p53 signature [84]. TP53 mutation is not only an
initial event of high-grade serous carcinogenesis but also 4.2.2 T
he Other Ovarian Epithelial
involved in tumor progression. In addition, nearly half of the Carcinomas
high-grade serous carcinomas display BRCA1 (17q21.31)
and BRCA2 (13q13.1) inactivating mutations (including 4.2.2.1 Endometrioid Carcinoma, Clear Cell
germline mutation, somatic mutation, or promoter methyla- Carcinoma, and Seromucinous Carcinoma
tion). Further, common amplification of CCNE1, NOTCH3, Ovarian endometrioid and clear cell carcinoma account for
PIKCA3, and AKT, and the inactivation of RB and NF1 has about 25% of all EOCs, which are the most common tumors
also been observed in some tumors [28, 85]. Alterations in after serous carcinoma. Seromucinous carcinoma (also
DNA copy number have been observed in early lesions like known as endocervical-type mucinous or mixed epithelial
STICs [86]. High-grade serous carcinoma with BRCA1/2 carcinomas of Müllerian type) is very rare. It is composed
mutation is characterized by numerous alterations in DNA predominately of serous and endocervical-type mucinous
copy number but without CCNE1 amplification, a very com- epithelium, foci of clear cells, and uncommon area of endo-
mon event in primary and refractory tumors [81]. The minor- metrioid and squamous differentiation. All of three tumors
ity of high-grade serous carcinomas with inherited mutations belong to type I EOCs and are believed that at least 1/3 of the
also has germline deletion mutations in BARD1, BRIP1, cases originate in endometriosis [89, 90]. These tumors usu-
MRE11, NBN, RAD51C, RAD51, and PALB2 genes involved ally have endometriosis, endometrial benign, and/or border-
in the signaling pathway of Fanconi anemia [87]. line tumors in the background lesions.
Although TP53 mutation is a common event in high-grade Most endometrial carcinomas are low-grade (FIGO grade
serous carcinoma, the mutation alone in TP53 is not sufficient 1), whose clinical manifestations are consistent with the fea-
to induce malignant transformation, suggesting other molec- tures of type I EOCs. Only a few endometrial carcinomas
130 J. Zhang et al.
a
most cases <3% cases KRAS,
TP53 BRAF,
mutation
Tubal epithelim ERBB2
b g
mutation
?
f
1p, 5q, 8p,
BRCA1/2 18q, 22q and
inactivation
Xp allelic
c h imbalances
few cases?
DNA copy
number
e
amplification
or deletion TP53 mutation
d
Fig. 4.3 The progression of high-serous serous carcinoma and asso- (f). Few ovarian high-grade serous carcinomas may be derived from
ciated molecular genetic alterations. The normal mucosa epithelium the OEIs. The TP53 and BRCA1/2 mutations and numerous amplifica-
of fallopian tubal fimbria (a) undergoes p53 signature (b), serous tions and deletions of DNA copy number involve in above two path-
tubal intraepithelial carcinoma (c), serous tubal invasive carcinoma ways. Less than 3% of high-grade serous carcinomas are originate in
(d), and ovarian high-grade serous carcinoma via the shedding and direct progression of low-grade serous carcinoma (h) or the recur-
implantation in ovary (e). The normal tubal epithelium may seed on rence of borderline serous tumor (g). This process may also be related
the ovarian surface during ovulation process and form the tubal OEIs to TP53 mutations
belong to intermediate-grade or high-grade (FIGO grade 2 or OEI cells exhibit the endometrioid morphological changes. In
3). In these tumors, it is frequently found the component of combination of the recent views that some OEIs are derive
low-grade carcinoma or the presence of some molecular from the seeding of tubal epithelial cells [54, 90], it is possi-
genetic alterations similar to concurrent low-grade carci- ble that fallopian tube may contribute the histogenesis of
noma, which suggests that these intermediate-grade or high- ovarian endometriosis. Lately, Yuan et al. [94] found that 18
grade components may be transformed via dedifferentiation cases (56%) showed a high level of FMO3 and a low level of
of low-grade carcinoma [52]. Few high-grade endometrioid DMBT1 expression in 32 ovarian endometriosis cases by a
carcinomas have the same clinical features and genetic alter- gene differential array analysis, which is similar to fallopian
ations (such as TP53 mutation) as those of high-grade serous tubal profile. This result suggests that approximately 60% of
carcinomas. Recent evidence supports the notion that these the ovarian endometriosis may be derived from the fallopian
tumors are more suitable for being classified to a variant of tube, whereas about 40% of the cases may be of endometrial
high-grade serous carcinomas [80, 90, 91]. origin. Using a fallopian tube- specific promoter OVGP1
Endometriosis is an estrogen-dependent inflammatory dis- gene, Wu et al. [95] developed an animal model with specifi-
ease. Ovary is the most common organ involved by endome- cally biallelic inactivation of Apc and Pten in oviductal or
triosis. Thus far its origin is still in dispute. Its major theory is ovarian epithelium using Ovgp1-iCreERT2 mice and found
the reflux menstruation and the coelomic metaplasia [92]. In that there was the formation of endometrioid carcinoma-like
2005, Zheng et al. [93] found the earliest morphological malignant epithelial tumors in the fallopian tubes or ovaries,
changes of endometriosis and named it “initial endometrio- respectively. On the basis of the morphology and global
sis.” Its histological feature is a transition from minimal to genetic profiles, the oviduct- derived tumors more closely
mature endometriosis in normal ovarian tissue. This change resemble human ovarian endometrioid carcinomas than do
presents both in the ovarian epithelium and in the cortical OSE-derived poorly differentiated ovarian carcinoma. The
OEI, and the latter is most common. That is, some ovarian abovementioned results provide strongly experimental evi-
4 Ovarian Epithelial Carcinogenesis 131
dence that tubal epithelial cells may be one of the potential homolog 1) or hMSH2 (human mutS homolog 2) protein.
origins of ovarian endometrioid carcinoma. KRAS and BRAF mutations are rare, with an incidence of
Recent data suggest that clear cell carcinoma may arise less than 7% [9, 52].
from ciliated cells in the endometrium (both eutopic endo- Compared with endometrioid carcinomas, clear cell car-
metrium and endometriosis) and fallopian tubes, whereas cinomas rarely have abnormal Wnt/β-catenin signaling path-
endometrioid carcinoma may arise from secretory cells way. The most common genetic alteration is ARID1A
rather than histotype-specific mutations. The cystathionine inactivation mutation. It is documented that the expression
gamma-lyase (CTH) and methylenetetrahydrofolate dehy- status of the SWI/SNF complex serves as an independent
drogenase 1 (MTHFD1) can serve as specific markers for prognostic factor. The loss of one or multiple SWI/SNF com-
ciliated and secretory cells, respectively [96]. plex subunits demonstrates aggressive behaviors and poor
Common genetic and genomic alterations: At the genetic prognosis [83]. Clear cell carcinoma also has a high PIK3CA-
level, several unique genes have been found in the endome- activated mutation (40%) and PTEN deletion mutation
trioid carcinoma. The most frequently mutated gene is (10%). More than 70% PIK3CA mutation occurs simultane-
ARID1A tumor suppressor gene. Its encoded protein ously with ARID1A mutation [52]. Clear cell carcinoma
BAF250a can bind to the AT-rich DNA sequence and is a key shows similar frequency of MSI [104, 105] as endometrioid
component of the switch/sucrose non-fermentable (SWI/ adenocarcinoma except that MSI is mainly manifested as
SNF) chromosome remodeling complex. The function of hMSH2 germline mutation. Furthermore, clear cell carci-
this complex is to mobilize nucleosomes, regulate gene noma has a unique epigenetic alteration, which involves in
expression and chromosome kinetics, and participate in the the methylation of multiple gene promoters in α-estrogen
regulation of a variety of cell processes (e.g., development, receptor alpha (ERα) and hepatocyte nuclear factor 1 (HNF-
differentiation, proliferation, DNA repair, and tumor sup- 1) signal pathways [106]. The changes in cell metabolism
pression). Mutations in ARID1A have been detected in induced by HNF-1β expression are conducive to tumorigen-
46–57% of cases of clear cell carcinoma, 30% in cases of esis and cell survival [107].
endometrioid carcinoma, and 33% in cases of seromucinous Seromucinous tumors are very rare. Morphologically, sero-
carcinoma; on other hand, there are no detected ARID1A mucinous tumors in addition to serous and endocervical-type
mutations or deletions in serous and mucinous carcinoma mucinous epithelium contain endometrioid, undifferentiated,
and malignant Brenner tumor, suggesting that these muta- and squamous-type epithelium with frequent expression of ER
tions are histotype-specific. Interestingly, and unexpectedly, and PR, lack of expression of CK20 and CDX2, and infre-
while it is widely believed that endometriosis is a precursor quent expression of WT1. This tumor is also frequently asso-
lesion of endometrioid, clear cell, and seromucinous carci- ciated with endometriosis-like endometrioid or clear cell
noma. These mutations are also found in endometriosis tissues carcinoma. Although seromucinous carcinomas were intro-
adjacent to cancer tissues [97], suggesting that these com- duced as a new entity in 2014 WHO classification, it has been
monly believed cancer-causing mutations are actually not the argued that its name may have a serious flaws that obscure that
drivers of malignant transformation. ARID1A may be an early nature of neoplasms [90]. Due to its rarity, there are few reports
event of endometriosis progression to cancer [98–100]. In about its molecular genetics except ARID1A inactivation
addition to ARID1A gene, endometrioid, clear cell, and sero- mutation. In 2017, through the study of 32 tumors diagnosed
mucinous carcinomas have other special genetic alterations. It as seromucinous carcinomas from 2 medical centers, the
is now clear that sole ARID1A inactivation is not sufficient to authors found that these tumors had KRAS (70%), PIK3CA
induce tumorigenesis. The deletion of both Arid1a and Pten (37%), PTEN (19%), and ARID1A (16%) mutations. 30% of
leads to endometrioid and undifferentiated carcinoma in vivo cases harbored concurrent KRAS and PIK3CA mutation [108].
studies [101]. And the combinations of deletions of ARID1A The immunophenotype and molecular genetics of seromuci-
and PIK3CA can cause clear cell-like tumors via the activating nous carcinoma overlapped predominantly with endometrioid
inflammation signaling pathways [102, 103]. and low-grade serous carcinoma. After integrating morphol-
The genetic alterations in the endometrioid carcinoma ogy, immunophenotype, and genotyping, 32 cases of seromu-
appear to be different from that of clear cell carcinoma. cinous carcinomas were reclassify to endometrioid (23 cases),
Mutations in PI3K/PTEN and Wnt/β-catenin signaling path- low-grade serous (8 cases), and mucinous carcinoma (1 case).
ways are frequently detected, including CTNNB1 (15–40%), It has been proposed to reclassify this group of tumors as
PIK3CA (20%) and PTEN (14–21%). CTNNB1 mutation is “mixed Müllerian tumors” which can be subcategorized as
associated with squamous differentiation, low-grade tumors, “mixed Müllerian cystadenomas,” “mixed Müllerian atypical
and good prognosis. PIK3CA and PTEN mutations can occur proliferative (borderline) tumors,” and “mixed Müllerian car-
simultaneously. 13–20% cases of endometrial carcinoma cinomas.” Therefore, seromucinous carcinoma may be due to
possess microsatellite instability (MSI), which usually mani- the heterogeneity and lineage infidelity of different histotypes
fest as the deletion expression of hMLH1 (human mutL and may not be a real distinct subtype of EOCs [108].
132 J. Zhang et al.
4.2.2.2 Mucinous Carcinoma and Malignant mucinous tumor may develop from a Brenner tumor in
Brenner Tumor which the component of Brenner tumor is compressed and
According to the 2014 WHO classification, ovarian muci- obliterated by an expanding mucinous neoplasm. The tran-
nous carcinoma refers to the gastrointestinal-type. With fur- sitional cell nests in Brenner tumors frequently contain
ther studies on clinical pathology and molecular genetics, it mucinous cells, prompting that the overgrowth of these
is clear that the vast majority (more than 90%) of ovarian mucinous epithelial cells eventually lead to the occurrence
mucinous carcinoma (especially advanced tumors) are sec- of mucinous tumors.
ondary, mostly from the gastrointestinal tract (especially col- Mucinous carcinomas commonly possess the activation
orectum and appendix). The other common metastatic sites of RAS/MEK signaling pathways. More than 90% of cases
include pancreas, bile duct, gallbladder, endocervix, bladder, present KRAS, BRAF, and/or ERRB2 mutation. Among
etc. The real primary mucinous carcinomas are less than 5% them the most common is KRAS-activated somatic cell
of EOCs [7, 109]. mutation (64.5%) [116]. The amplification of ERRB2 is
The size of primary ovarian mucinous carcinoma is often about 15% to 20%. KRAS mutation is thought to be an early
large (more than 10 cm). Histologically, there are significant event of mucinous carcinogenesis because KRAS mutation
heterogeneity changes in tumor tissue, such as the presence of is found in mucinous cystadenomas and borderline tumors
components of benign mucinous cystadenoma/adenofibroma, adjacent to cancer tissues [52]. Moreover, TP53 mutation
borderline tumor, and carcinoma. The vast majority of tumors rates in mucinous carcinomas and borderline tumors are
are low-grade, which fully reflects the tumorigenesis of type 56.8% and 11.5%, respectively [116]. About 21% cases of
I EOCs. Clinically, the patients with stage I ovarian mucinous mucinous carcinomas had an inactivated mutation of RNF3
tumors (borderline with/without intraepithelial carcinoma or tumor suppressor gene, similar to pancreatic mucinous
microinvasion, and invasive carcinoma) have excellent prog- tumors [117].
nosis, whose 5-year survival rate is more than 90%. Advanced Brenner tumors are composed of urethral/transitional epithe-
ovarian mucinous carcinomas are uncommon and have a poor lium nested around with fibromatous stroma. The vast majority
prognosis. The overall survival rate of these patients is the of tumors are benign. The malignant tumors are very rare, and
same as that of metastatic adenocarcinoma [110]. around 80% of tumors are confined to the ovary (FIGO stage I),
Compared with metastatic mucinous carcinoma, most pri- with a typical progression process of type I EOCs, i.e., from
mary tumors are confined to unilateral ovaries, suggesting that benign to borderline to malignant Brenner tumors.
its precursor lesions should be located in the ovaries. Its origin is still unknown because the morphology of
Accumulating molecular genetic evidence suggests that at Brenner tumors is totally different from Müllerian epithe-
least a subset of mucinous tumors (including mucinous cyst- lium and lacks the expression of Müllerian immunohisto-
adenomas, borderline tumors, and mucinous carcinomas) may chemical markers (PAX-8 and PAX-2). Some scholars
arise from mucous epithelium within mature cystic teratomas believe that its origin is independent of ovarian OEI [8].
(germ cell origin) and Brenner tumor (non-germ cell origin). Many tissue or cell types, including transitional cell metapla-
The cell origin of teratomas is believed to be postmeiotic sia from OSE, mesonephric remnant, the rete ovarii, muci-
ovarian germ cells [111]. There are mucinous tumors in nous tumor, fallopian tube, and teratoma, have been presumed
about 2–11% of ovarian mature cystic teratoma. Around to be the origin of Brenner tumor [113].
3–8% mucinous tumors are associated with teratomas. Using Walthard cell nests consist of transitional epithelium and
short tandem repeat analysis, genotypical concordance usually exist at/near the tuboperitoneal junction and fre-
between the teratomas and coexisting mucinous tumors are quently coexist with Brenner tumors. It has long been con-
found, which provide the evidence that the cell origin of sidered the origin of Brenner tumors because they display
some mucinous tumors is germ cells [111, 112]. epithelia with the same morphology and immunophenotype
Mucinous tumors are often associated with Brenner (e.g., GATA3 and p63 positive, germline marker SALL4
tumors. Seidman et al. [113] found that 20% of mucinous negative) as Brenner tumors [8, 118]. Roma et al. [118]
tumors have Brenner tumor components and 16% of found that 43% of patients with Brenner tumors had Walthard
Brenner tumors contain concurrent mucinous tumors. In cell nests in the soft tissue surround fallopian tubes/ovaries.
the molecular genetics, there is 12q14-21 amplification Through a morphologic and immunohistochemical analysis
both in ovarian mucinous carcinoma and coexistence of in Brenner tumors, Kuhn et al. [119] found that tubal secre-
Brenner tumors using an analysis of comparative genomic tory cells, transitional metaplasia, Walthard cell nests, and
hybridization [114]. It is also reported that all two compo- the epithelial component of Brenner tumors shared a similar
nents in most mixed Brenner and mucinous tumors have a immunophenotype, consistently expressing AKR1C3 (an
concordant X-chromosome inactivation pattern via a enzyme involved in androgen biosynthesis) and androgen
human androgen receptor gene assay [115]. These results receptor, but not calretinin. The stromal cells that closely
confirm that the components of Brenner and mucinous surrounded the epithelial nests showed strong expression of
tumors share a same clonal relationship. Some mucinous calretinin, inhibin, and steroidogenic factor 1 (markers of
tumors originate from the Brenner tumor and the pure steroidogenic cells) in the majority of tumors. There were
4 Ovarian Epithelial Carcinogenesis 133
small groups of cilia in transitional metaplasia and Walthard half of the cases have deficiencies of mismatch repair proteins.
cell nests, multifocal stretches of cilia and/or ciliated vacu- It is not clear whether the tumor is an independent histological
oles in benign tumors, and well-developed cilia in atypical subtype or a poorly differentiated variant of high-grade serous
proliferative tumors. These findings suggest a tubal origin of or endometrioid carcinoma. MMMTs have significant biphasic
Brenner tumors via transitional metaplasia and Walthard cell differentiation of high-grade carcinoma and sarcoma elements.
nests under the effect of androgenic stimulation. It is believe that these tumors belong to high-grade carcinoma
Malignant Brenner tumors are low-grade EOCs with the with sarcomatoid differentiation due to the same clonality of
activation of PI3K/AKT signaling pathway. PIK3CA mutation these two elements. Its most common molecular genetics alter-
has been reported in an individual case [120]. In benign Brenner ations are TP53 mutation and CDKN2A amplification [9].
tumors, the positive rate of p16 immunohistochemistry is 92%, In summary, many of EOCs come from extraovarian tis-
while completely negative expression of p16 is found in the sues, and the ovaries could be secondary involvement; a sig-
borderline/atypical proliferative Brenner tumor. The data from nificant number of ovarian cancer may also arise from
fluorescence in situ hybridization confirm that there is loss of multifocal origin of different type of Müllerian epithelial
heterozygosity of CDKN2A gene (encoding p16 protein), sug- cells. Recent studies have provided increasing evidence that
gesting that CDKN2A deletion may play a role in progression a number of EOCs may arise directly or indirectly from tubal
from benign to borderline Brenner tumor [9]. epithelial cells except that a part of mucinous carcinoma and
endometriosis-associated EOCs (endometrioid, clear cell,
4.2.2.3 Undifferentiated Carcinoma and seromucinous carcinoma) derive from ovarian germ
and Malignant Mixed Müllerian Tumor cells and glandular epithelial cells of endometriosis, respec-
Both of these histotypes are rare high-grade and aggressive tively (Fig. 4.4). However, readers should consider these
malignant tumors. Undifferentiated carcinomas lack the fea- models only as hypotheses. As more evidence accumulates
tures of high-grade serous or endometrioid carcinoma. Nearly in the future, these models will be revisited and updated.
Fig. 4.4 The proposed models of ovarian epithelial carcinogenesis. them are derived from OEIs. And less than 3% of cases are from the
Normal mucosa epithelial cells of fallopian tubal fimbria shed and progression of low-grade serous carcinomas or the recurrence of bor-
implant on the ovarian surface. The OEIs are formed via the epithelial derline serous tumors. Malignant Brenner tumors progress from benign
invagination during ovulation process and then undergo metaplasia. Brenner tumors that may derive from the Walthard cell nests, although
Endometrioid, clear cell, and seromucinous carcinoma originate in the cell origin of Walthard cell nests is unclear. Mucinous carcinomas
endometriotic glands and OEIs. Low-grade serous carcinoma arises originate in mucosal epithelium within the mature cystic teratomas and
from OEIs. High-grade serous carcinoma mostly comes from the seed- benign Brenner tumors
ing of serous tubal epithelial carcinoma (STIC). A very small number of
134 J. Zhang et al.
high-grade serous carcinoma with BRCA mutation in 2014, 9. Kurman RJ, Shih Ie M. The dualistic model of ovarian car-
cinogenesis: revisited, revised, and expanded. Am J Pathol.
the application of olaparib has been used in patients with 2016;186:733–47.
BRCA mutations. In addition, several the other PARP inhibi- 10. Espinosa I, Catasus L, Canet B, et al. Gene expression analysis
tors are also at the various stages of clinical trials [138]. The identifies two groups of ovarian high-grade serous carcinomas
therapeutic strategies of platinum-based medicines in combi- with different prognosis. Mod Pathol. 2011;24:846–54.
11. Silva EG. The origin of epithelial neoplasms of the ovary: an alter-
nation with PARP inhibitors may prolong disease-free sur- native view. Adv Anat Pathol. 2016;23:50–7.
vival of the patients with BRCA mutation-related EOCs 12. Zhang J, Chang DY, Mercado-Uribe I, et al. Sex-determining
[139]. On the other hand, current immunotherapies have region Y-box 2 expression predicts poor prognosis in human ovar-
been largely ineffective in advanced and current ovarian cancer ian carcinoma. Hum Pathol. 2012;43:1405–12.
13. Zhang J, Guo X, Chang DY, et al. CD133 expression asso-
patients [140]. Therefore, great challenges lie ahead why the ciated with poor prognosis in ovarian cancer. Mod Pathol.
ovarian cancers are largely unresponsive for commonly used 2012;25:456–64.
check point inhibitor for immunotherapy. Finally, if our hypoth- 14. Kenda Suster N, Smrkolj S, Virant-Klun I. Putative stem cells and
esis of somatic blastomere origin is correct, treatment should epithelial-mesenchymal transition revealed in sections of ovarian
tumor in patients with serous ovarian carcinoma using immuno-
be focused on preventing formation of PGCCs and redifferen- histochemistry for vimentin and pluripotency-related markers. J
tiating the PGCCs toward benign lineages, which will offer a Ovarian Res. 2017;10:11.
totally new paradigm for cancer prevention and therapy. It is 15. Flesken-Nikitin A, Hwang CI, Cheng CY, et al. Ovarian surface
expected that more and more efficient therapeutic strategies epithelium at the junction area contains a cancer-prone stem cell
niche. Nature. 2013;495:241–5.
will be developed to clinical treatment as we gain additional 16. Virant-Klun I, Kenda-Suster N, Smrkolj S. Small putative
insight into the mechanism of ovarian carcinogenesis. NANOG, SOX2, and SSEA-4-positive stem cells resembling very
small embryonic-like stem cells in sections of ovarian tissue in
Acknowledgments Jinsong Liu is supported by MD Anderson Cancer patients with ovarian cancer. J Ovarian Res. 2016;9:12.
Center SPORE and Moonshot Program in Ovarian Cancer. Jing Zhang 17. Ng A, Barker N. Ovary and fimbrial stem cells: biology, niche and
is supported by National Natural Science Foundation of China (NSFC, cancer origins. Nat Rev Mol Cell Biol. 2015;16:625–38.
No. 81372783 and 81572545). Author’s affiliation: 1. Department of 18. Silva EG, Tornos C, Deavers M, et al. Induction of epithelial neo-
Pathology, Xijing hospital, The Fourth Military Medical University, plasms in the ovaries of Guinea pigs by estrogenic stimulation.
Xi’an, People’s Republic of China 710032; 2. Department of Pathology, Gynecol Oncol. 1998;71:240–6.
3. Gynecologic Oncology and Reproductive Medicine, The University 19. Silva EG, Tornos C, Fritsche HA Jr, et al. The induction of
of Texas MD Anderson Cancer Center, Unit 1362, P.O. Box 301439, benign epithelial neoplasms of the ovaries of Guinea pigs by
Houston, TX 77230-1439, USA. Department of Correspondence: jliu@ testosterone stimulation: a potential animal model. Mod Pathol.
mdanderson.org 1997;10:879–83.
20. Liu J, Yang G, Thompson-Lanza JA, et al. A genetically defined
model for human ovarian cancer. Cancer Res. 2004;64:1655–63.
21. Yang G, Rosen DG, Zhang Z, et al. The chemokine growth-
regulated oncogene 1 (Gro-1) links RAS signaling to the senes-
References cence of stromal fibroblasts and ovarian tumorigenesis. Proc Natl
Acad Sci U S A. 2006;103:16472–7.
1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA 22. Yang G, Rosen DG, Liu G, et al. CXCR2 promotes ovarian cancer
Cancer J Clin. 2016;66:7–30. growth through dysregulated cell cycle, diminished apoptosis, and
2. Ahmed N, Thompson EW, Quinn MA. Epithelial-mesenchymal enhanced angiogenesis. Clin Cancer Res. 2010;16:3875–86.
interconversions in normal ovarian surface epithelium and 23. Schauer IG, Zhang J, Xing Z, et al. Interleukin-1beta promotes
ovarian carcinomas: an exception to the norm. J Cell Physiol. ovarian tumorigenesis through a p53/NF-kappaB-mediated
2007;213:581–8. inflammatory response in stromal fibroblasts. Neoplasia.
3. Cheng W, Liu J, Yoshida H, et al. Lineage infidelity of epithelial 2013;15:409–20.
ovarian cancers is controlled by HOX genes that specify regional 24. Zhang J, Liu J. Tumor stroma as targets for cancer therapy.
identity in the reproductive tract. Nat Med. 2005;11:531–7. Pharmacol Ther. 2013;137:200–15.
4. De Santis G, Miotti S, Mazzi M, et al. E-cadherin directly con- 25. Cardenas C, Alvero AB, Yun BS, et al. Redefining the origin and
tributes to PI3K/AKT activation by engaging the PI3K-p85 regu- evolution of ovarian cancer: a hormonal connection. Endocr Relat
latory subunit to adherens junctions of ovarian carcinoma cells. Cancer. 2016;23:R411–22.
Oncogene. 2009;28:1206–17. 26. Hanahan D, Weinberg RA. Hallmarks of cancer: the next genera-
5. Katre R, Morani AK, Prasad SR, et al. Tumors and pseudotu- tion. Cell. 2011;144:646–74.
mors of the secondary mullerian system: review with empha- 27. Vogelstein B, Kinzler KW. Cancer genes and the pathways they
sis on cross-sectional imaging findings. AJR Am J Roentgenol. control. Nat Med. 2004;10:789–99.
2010;195:1452–9. 28. Kurman RJ, Vang R, Junge J, et al. Papillary tubal hyperplasia:
6. Lauchlan SC. The secondary Mullerian system. Obstet Gynecol the putative precursor of ovarian atypical proliferative (borderline)
Surv. 1972;27:133–46. serous tumors, noninvasive implants, and endosalpingiosis. Am J
7. Shih Ie M, Kurman RJ. Ovarian tumorigenesis: a proposed model Surg Pathol. 2011;35:1605–14.
based on morphological and molecular genetic analysis. Am J 29. Brucher BL, Jamall IS. Somatic mutation theory - why it’s wrong
Pathol. 2004;164:1511–8. for most cancers. Cell Physiol Biochem. 2016;38:1663–80.
8. Kurman RJ, Shih Ie M. The origin and pathogenesis of epithe- 30. Niu N, Mercado-Uribe I, Liu J. Dedifferentiation into blastomere-
lial ovarian cancer: a proposed unifying theory. Am J Surg Pathol. like cancer stem cells via formation of polyploid giant cancer
2010;34:433–43. cells. Oncogene. 2017;36:4887–900.
4 Ovarian Epithelial Carcinogenesis 137
31. Vanneste E, Voet T, Le Caignec C, et al. Chromosome instabil- 54. Li J, Abushahin N, Pang S, et al. Tubal origin of ‘ovarian’ low-
ity is common in human cleavage-stage embryos. Nat Med. grade serous carcinoma. Mod Pathol. 2011;24:1488–99.
2009;15:577–83. 55. Li J, Fadare O, Xiang L, et al. Ovarian serous carcinoma: recent con-
32. Boyers SP, Diamond MP, Lavy G, et al. The effect of polyploidy cepts on its origin and carcinogenesis. J Hematol Oncol. 2012;5:8.
on embryo cleavage after in vitro fertilization in humans. Fertil 56. Silva EG, Gershenson DM, Malpica A, et al. The recurrence and
Steril. 1987;48:624–7. the overall survival rates of ovarian serous borderline neoplasms
33. Chavez SL, Loewke KE, Han J, et al. Dynamic blastomere behav- with noninvasive implants is time dependent. Am J Surg Pathol.
iour reflects human embryo ploidy by the four-cell stage. Nat 2006;30:1367–71.
Commun. 2012;3:1251. 57. Mayr D, Hirschmann A, Lohrs U, et al. KRAS and BRAF muta-
34. Niakan KK, Han J, Pedersen RA, et al. Human pre-implantation tions in ovarian tumors: a comprehensive study of invasive car-
embryo development. Development. 2012;139:829–41. cinomas, borderline tumors and extraovarian implants. Gynecol
35. Davoli T, de Lange T. The causes and consequences of polyploidy Oncol. 2006;103:883–7.
in normal development and cancer. Annu Rev Cell Dev Biol. 58. Singer G, Oldt R 3rd, Cohen Y, et al. Mutations in BRAF and
2011;27:585–610. KRAS characterize the development of low-grade ovarian serous
36. Fox DT, Duronio RJ. Endoreplication and polyploidy: insights carcinoma. J Natl Cancer Inst. 2003;95:484–6.
into development and disease. Development. 2013;140:3–12. 59. Ho CL, Kurman RJ, Dehari R, et al. Mutations of BRAF and
37. Orr-Weaver TL. When bigger is better: the role of polyploidy in KRAS precede the development of ovarian serous borderline
organogenesis. Trends Genet. 2015;31:307–15. tumors. Cancer Res. 2004;64:6915–8.
38. Zielke N, Edgar BA, DePamphilis ML. Endoreplication. Cold 60. Wong KK, Tsang YT, Deavers MT, et al. BRAF mutation is rare
Spring Harb Perspect Biol. 2013;5:a012948. in advanced-stage low-grade ovarian serous carcinomas. Am J
39. Lee HO, Davidson JM, Duronio RJ. Endoreplication: polyploidy Pathol. 2010;177:1611–7.
with purpose. Genes Dev. 2009;23:2461–77. 61. Singer G, Kurman RJ, Chang HW, et al. Diverse tumorigenic path-
40. Malpica A, Deavers MT, Lu K, et al. Grading ovarian serous carci- ways in ovarian serous carcinoma. Am J Pathol. 2002;160:1223–8.
noma using a two-tier system. Am J Surg Pathol. 2004;28:496–504. 62. Kuo KT, Guan B, Feng Y, et al. Analysis of DNA copy number
41. Sharma S, Zeng JY, Zhuang CM, et al. Small-molecule inhibitor alterations in ovarian serous tumors identifies new molecular
BMS-777607 induces breast cancer cell polyploidy with increased genetic changes in low-grade and high-grade carcinomas. Cancer
resistance to cytotoxic chemotherapy agents. Mol Cancer Ther. Res. 2009;69:4036–42.
2013;12:725–36. 63. Vang R, Shih Ie M, Kurman RJ. Ovarian low-grade and high-
42. Zhang S, Mercado-Uribe I, Xing Z, et al. Generation of cancer grade serous carcinoma: pathogenesis, clinicopathologic and
stem-like cells through the formation of polyploid giant cancer molecular biologic features, and diagnostic problems. Adv Anat
cells. Oncogene. 2014;33:116–28. Pathol. 2009;16:267–82.
43. Chen S, Stout JR, Dharmaiah S, et al. Transient endoreplication 64. Zhai Y, Kuick R, Tipton C, et al. Arid1a inactivation in an Apc- and
down-regulates the kinesin-14 HSET and contributes to genomic Pten-defective mouse ovarian cancer model enhances epithelial
instability. Mol Biol Cell. 2016;27:2911–23. differentiation and prolongs survival. J Pathol. 2016;238:21–30.
44. Erenpreisa J, Kalejs M, Cragg MS. Mitotic catastrophe and endo- 65. Lee Y, Miron A, Drapkin R, et al. A candidate precursor to serous
mitosis in tumour cells: an evolutionary key to a molecular solu- carcinoma that originates in the distal fallopian tube. J Pathol.
tion. Cell Biol Int. 2005;29:1012–8. 2007;211:26–35.
45. Niu N, Zhang J, Zhang N, et al. Linking genomic reorganization to 66. Kuhn E, Kurman RJ, Vang R, et al. TP53 mutations in serous tubal
tumor initiation via the giant cell cycle. Oncogene. 2016;5:e281. intraepithelial carcinoma and concurrent pelvic high-grade serous
46. Zhang S, Mercado-Uribe I, Sood A, et al. Coevolution of neo- carcinoma--evidence supporting the clonal relationship of the two
plastic epithelial cells and multilineage stroma via polyploid giant lesions. J Pathol. 2012;226:421–6.
cells during immortalization and transformation of mullerian epi- 67. O'Shannessy DJ, Jackson SM, Twine NC, et al. Gene expression
thelial cells. Genes Cancer. 2016;7:60–72. analyses support fallopian tube epithelium as the cell of origin of
47. Liu J. The dualistic origin of human tumors. epithelial ovarian cancer. Int J Mol Sci. 2013;14:13687–703.
Semin Cancer Biol. 2018. pii: S1044-579X(18)30023-3. https:// 68. Marquez RT, Baggerly KA, Patterson AP, et al. Patterns of gene
doi.org/10.1016/j.semcancer.2018.07.004. expression in different histotypes of epithelial ovarian cancer
48. Lucchetta EM, Ohlstein B. Amitosis of polyploid cells regener- correlate with those in normal fallopian tube, endometrium, and
ates functional stem cells in the drosophila intestine. Cell Stem colon. Clin Cancer Res. 2005;11:6116–26.
Cell. 2017;20:609–620.e606. 69. Kuhn E, Meeker A, Wang TL, et al. Shortened telomeres in serous
49. Crum CP, Drapkin R, Miron A, et al. The distal fallopian tube: tubal intraepithelial carcinoma: an early event in ovarian high-
a new model for pelvic serous carcinogenesis. Curr Opin Obstet grade serous carcinogenesis. Am J Surg Pathol. 2010;34:829–36.
Gynecol. 2007;19:3–9. 70. Perets R, Wyant GA, Muto KW, et al. Transformation of the fallo-
50. Levanon K, Crum C, Drapkin R. New insights into the pathogen- pian tube secretory epithelium leads to high-grade serous ovarian
esis of serous ovarian cancer and its clinical impact. J Clin Oncol. cancer in Brca;Tp53;Pten models. Cancer Cell. 2013;24:751–65.
2008;26:5284–93. 71. Kim J, Coffey DM, Ma L, et al. The ovary is an alternative
51. Callahan MJ, Crum CP, Medeiros F, et al. Primary fallopian tube site of origin for high-grade serous ovarian cancer in mice.
malignancies in BRCA-positive women undergoing surgery for Endocrinology. 2015;156:1975–81.
ovarian cancer risk reduction. J Clin Oncol. 2007;25:3985–90. 72. Falconer H, Yin L, Gronberg H, et al. Ovarian cancer risk after sal-
52. Kurman RJ, International Agency for Research on Cancer., World pingectomy: a nationwide population-based study. J Natl Cancer
Health Organization. WHO classification of tumours of female Inst. 2015;107. pii: dju410
reproductive organs. Lyon: International Agency for Research on 73. Levanon K, Ng V, Piao HY, et al. Primary ex vivo cultures of
Cancer; 2014. human fallopian tube epithelium as a model for serous ovarian
53. Malpica A, Deavers MT, Tornos C, et al. Interobserver and intrao- carcinogenesis. Oncogene. 2010;29:1103–13.
bserver variability of a two-tier system for grading ovarian serous 74. Jarboe E, Folkins A, Nucci MR, et al. Serous carcinogenesis in the
carcinoma. Am J Surg Pathol. 2007;31:1168–74. fallopian tube: a descriptive classification. Int J Gynecol Pathol.
2008;27:1–9.
138 J. Zhang et al.
75. Kindelberger DW, Lee Y, Miron A, et al. Intraepithelial carcinoma 96. Cochrane DR, Tessier-Cloutier B, Lawrence KM, et al. Clear cell
of the fimbria and pelvic serous carcinoma: evidence for a causal and endometrioid carcinomas: are their differences attributable to
relationship. Am J Surg Pathol. 2007;31:161–9. distinct cells of origin? J Pathol. 2017;243:26–36.
76. Przybycin CG, Kurman RJ, Ronnett BM, et al. Are all pelvic 97. Anglesio MS, Papadopoulos N, Ayhan A, et al. Cancer-associated
(nonuterine) serous carcinomas of tubal origin? Am J Surg Pathol. mutations in endometriosis without cancer. N Engl J Med.
2010;34:1407–16. 2017;376:1835–48.
77. Singh N, Gilks CB, Wilkinson N, et al. The secondary Mullerian 98. Wiegand KC, Shah SP, Al-Agha OM, et al. ARID1A mutations
system, field effect, BRCA, and tubal fimbria: our evolving under- in endometriosis-associated ovarian carcinomas. N Engl J Med.
standing of the origin of tubo-ovarian high-grade serous carci- 2010;363:1532–43.
noma and why assignment of primary site matters. Pathology. 99. Jones S, Wang TL, Shih Ie M, et al. Frequent mutations of chro-
2015;47:423–31. matin remodeling gene ARID1A in ovarian clear cell carcinoma.
78. Banet N, Kurman RJ. Two types of ovarian cortical inclusion Science. 2010;330:228–31.
cysts: proposed origin and possible role in ovarian serous carcino- 100. Wu CH, Mao TL, Vang R, et al. Endocervical-type mucinous
genesis. Int J Gynecol. 2015;34:3–8. borderline tumors are related to endometrioid tumors based on
79. Boyd C, McCluggage WG. Low-grade ovarian serous neoplasms mutation and loss of expression of ARID1A. Int J Gynecol Pathol.
(low-grade serous carcinoma and serous borderline tumor) associ- 2012;31:297–303.
ated with high-grade serous carcinoma or undifferentiated carci- 101. Guan B, Rahmanto YS, Wu RC, et al. Roles of deletion of Arid1a,
noma: report of a series of cases of an unusual phenomenon. Am J a tumor suppressor, in mouse ovarian tumorigenesis. J Natl Cancer
Surg Pathol. 2012;36:368–75. Inst. 2014;106. pii: dju146
80. Soslow RA, Han G, Park KJ, et al. Morphologic patterns associ- 102. Chandler RL, Damrauer JS, Raab JR, et al. Coexistent ARID1A-
ated with BRCA1 and BRCA2 genotype in ovarian carcinoma. PIK3CA mutations promote ovarian clear-cell tumorigenesis
Mod Pathol. 2012;25:625–36. through pro-tumorigenic inflammatory cytokine signalling. Nat
81. Patch AM, Christie EL, Etemadmoghadam D, et al. Whole- Commun. 2015;6:6118.
genome characterization of chemoresistant ovarian cancer. 103. Chandler RL, Raab JR, Vernon M, et al. Global gene expression
Nature. 2015;521:489–94. profiling of a mouse model of ovarian clear cell carcinoma caused
82. Verhaak RG, Tamayo P, Yang JY, et al. Prognostically relevant by ARID1A and PIK3CA mutations implicates a role for inflam-
gene signatures of high-grade serous ovarian carcinoma. J Clin matory cytokine signaling. Genom Data. 2015;5:329–32.
Invest. 2013;123:517–25. 104. Cai KQ, Albarracin C, Rosen D, et al. Microsatellite instability
83. Abou-Taleb H, Yamaguchi K, Matsumura N, et al. Comprehensive and alteration of the expression of hMLH1 and hMSH2 in ovarian
assessment of the expression of the SWI/SNF complex defines clear cell carcinoma. Hum Pathol. 2004;35:552–9.
two distinct prognostic subtypes of ovarian clear cell carcinoma. 105. Liu J, Albarracin CT, Chang KH, et al. Microsatellite instability
Oncotarget. 2016;7:54758–70. and expression of hMLH1 and hMSH2 proteins in ovarian endo-
84. Bossuyt V, Medeiros F, Drapkin R, et al. Adenofibroma of the metrioid cancer. Mod Pathol. 2004;17:75–80.
fimbria: a common entity that is indistinguishable from ovarian 106. Yamaguchi K, Huang Z, Matsumura N, et al. Epigenetic deter-
adenofibroma. Int J Gynecol Pathol. 2008;27:390–7. minants of ovarian clear cell carcinoma biology. Int J Cancer.
85. Nakayama K, Nakayama N, Jinawath N, et al. Amplicon profiles 2014;135:585–97.
in ovarian serous carcinomas. Int J Cancer. 2007;120:2613–7. 107. Amano Y, Mandai M, Yamaguchi K, et al. Metabolic alterations
86. Salvador S, Rempel A, Soslow RA, et al. Chromosomal instabil- caused by HNF1beta expression in ovarian clear cell carcinoma
ity in fallopian tube precursor lesions of serous carcinoma and contribute to cell survival. Oncotarget. 2015;6:26002–17.
frequent monoclonality of synchronous ovarian and fallopian tube 108. Rambau PF, McIntyre JB, Taylor J, et al. Morphologic reproduc-
mucosal serous carcinoma. Gynecol Oncol. 2008;110:408–17. ibility, genotyping, and immunohistochemical profiling do not
87. Walsh T, Casadei S, Lee MK, et al. Mutations in 12 genes for support a category of Seromucinous carcinoma of the ovary. Am J
inherited ovarian, fallopian tube, and peritoneal carcinoma identi- Surg Pathol. 2017;41:685–95.
fied by massively parallel sequencing. Proc Natl Acad Sci U S A. 109. Ronnett BM, Shmookler BM, Sugarbaker PH, et al. Pseudomyxoma
2011;108:18032–7. peritonei: new concepts in diagnosis, origin, nomenclature, and
88. Norquist BM, Garcia RL, Allison KH, et al. The molecular relationship to mucinous borderline (low malignant potential)
pathogenesis of hereditary ovarian carcinoma: alterations in the tumors of the ovary. Anat Pathol. 1997;2:197–226.
tubal epithelium of women with BRCA1 and BRCA2 mutations. 110. Zaino RJ, Brady MF, Lele SM, et al. Advanced stage mucinous
Cancer. 2010;116:5261–71. adenocarcinoma of the ovary is both rare and highly lethal: a
89. Deligdisch L, Penault-Llorca F, Schlosshauer P, et al. Stage I ovar- gynecologic oncology group study. Cancer. 2011;117:554–62.
ian carcinoma: different clinical pathologic patterns. Fertil Steril. 111. Snir OL, Buza N, Hui P. Mucinous epithelial tumours aris-
2007;88:906–10. ing from ovarian mature teratomas: a tissue genotyping study.
90. Kurman RJ, Shih Ie M. Seromucinous tumors of the ovary. What’s Histopathology. 2016;69:383–92.
in a name? Int J Gynecol Pathol. 2016;35:78–81. 112. Fujii K, Yamashita Y, Yamamoto T, et al. Ovarian mucinous
91. McCluggage WG. My approach to and thoughts on the typing of tumors arising from mature cystic teratomas–a molecular genetic
ovarian carcinomas. J Clin Pathol. 2008;61:152–63. approach for understanding the cellular origin. Hum Pathol.
92. Bulun SE. Endometriosis. N Engl J Med. 2009;360:268–79. 2014;45:717–24.
93. Zheng W, Li N, Wang J, et al. Initial endometriosis showing direct 113. Seidman JD, Khedmati F. Exploring the histogenesis of ovarian
morphologic evidence of metaplasia in the pathogenesis of ovar- mucinous and transitional cell (Brenner) neoplasms and their
ian endometriosis. Int J Gynecol Pathol. 2005;24:164–72. relationship with Walthard cell nests: a study of 120 tumors. Arch
94. Yuan Z, Wang L, Wang Y, et al. Tubal origin of ovarian endome- Pathol Lab Med. 2008;132:1753–60.
triosis. Mod Pathol. 2014;27:1154–62. 114. Pejovic T, Burki N, Odunsi K, et al. Well-differentiated muci-
95. Wu R, Zhai Y, Kuick R, et al. Impact of oviductal versus ovarian nous carcinoma of the ovary and a coexisting Brenner tumor
epithelial cell of origin on ovarian endometrioid carcinoma pheno- both exhibit amplification of 12q14-21 by comparative genomic
type in the mouse. J Pathol. 2016;240:341–51. hybridization. Gynecol Oncol. 1999;74:134–7.
4 Ovarian Epithelial Carcinogenesis 139
115. Wang Y, Wu RC, Shwartz LE, et al. Clonality analysis of com- 127. Wang Y, Sundfeldt K, Mateoiu C, et al. Diagnostic potential of
bined Brenner and mucinous tumours of the ovary reveals their tumor DNA from ovarian cyst fluid. eLife. 2016;5. pii: e15175
monoclonal origin. J Pathol. 2015;237:146–51. 128. Schenberg T, Mitchell G. Prophylactic bilateral salpingectomy as
116. Mackenzie R, Kommoss S, Winterhoff BJ, et al. Targeted deep a prevention strategy in women at high-risk of ovarian cancer: a
sequencing of mucinous ovarian tumors reveals multiple overlap- mini-review. Front Oncol. 2014;4:21.
ping RAS-pathway activating mutations in borderline and cancer- 129. Purdie DM, Bain CJ, Siskind V, et al. Ovulation and risk of epithe-
ous neoplasms. BMC Cancer. 2015;15:415. lial ovarian cancer. Int J Cancer. 2003;104:228–32.
117. Ryland GL, Hunter SM, Doyle MA, et al. RNF43 is a tumour sup- 130. Yang-Hartwich Y, Gurrea-Soteras M, Sumi N, et al. Ovulation and
pressor gene mutated in mucinous tumours of the ovary. J Pathol. extra-ovarian origin of ovarian cancer. Sci Rep. 2014;4:6116.
2013;229:469–76. 131. Beral V, Doll R, Hermon C, et al. Ovarian cancer and oral contra-
118. Roma AA, Masand RP. Ovarian Brenner tumors and Walthard ceptives: collaborative reanalysis of data from 45 epidemiological
nests: a histologic and immunohistochemical study. Hum Pathol. studies including 23,257 women with ovarian cancer and 87,303
2014;45:2417–22. controls. Lancet. 2008;371:303–14.
119. Kuhn E, Ayhan A, Shih Ie M, et al. Ovarian Brenner tumour: a mor- 132. Fathalla MF. Non-hormonal interruption of incessant ovulation as
phologic and immunohistochemical analysis suggesting an origin a potential approach for ovarian cancer prevention. Int J Gynaecol
from fallopian tube epithelium. Eur J Cancer. 2013;49:3839–49. Obstet. 2016;132:356–8.
120. Cuatrecasas M, Catasus L, Palacios J, et al. Transitional cell 133. Ciccone MA, Maoz A, Casabar JK, et al. Clinical outcome of
tumors of the ovary: a comparative clinicopathologic, immunohis- treatment with serine-threonine kinase inhibitors in recurrent epi-
tochemical, and molecular genetic analysis of Brenner tumors and thelial ovarian cancer: a systematic review of literature. Expert
transitional cell carcinomas. Am J Surg Pathol. 2009;33:556–67. Opin Investig Drugs. 2016;25:781–96.
121. Buys SS, Partridge E, Black A, et al. Effect of screening on ovar- 134. Spreafico A, Oza AM, Clarke BA, et al. Genotype-matched treat-
ian cancer mortality: the prostate, lung, colorectal and ovarian ment for patients with advanced type I epithelial ovarian cancer
(PLCO) cancer screening randomized controlled trial. JAMA. (EOC). Gynecol Oncol. 2017;144:250–5.
2011;305:2295–303. 135. Bitler BG, Aird KM, Garipov A, et al. Synthetic lethality by tar-
122. Bristow RE, Tomacruz RS, Armstrong DK, et al. Survival effect geting EZH2 methyltransferase activity in ARID1A-mutated can-
of maximal cytoreductive surgery for advanced ovarian carci- cers. Nat Med. 2015;21:231–8.
noma during the platinum era: a meta-analysis. J Clin Oncol. 136. Bitler BG, Aird KM, Zhang R. Epigenetic synthetic lethality in
2002;20:1248–59. ovarian clear cell carcinoma: EZH2 and ARID1A mutations. Mol
123. Medeiros F, Muto MG, Lee Y, et al. The tubal fimbria is a pre- Cell Oncol. 2016;3:e1032476.
ferred site for early adenocarcinoma in women with familial ovar- 137. Shen J, Peng Y, Wei L, et al. ARID1A deficiency impairs the
ian cancer syndrome. Am J Surg Pathol. 2006;30:230–6. DNA damage checkpoint and sensitizes cells to PARP inhibitors.
124. Fishman DA, Cohen L, Blank SV, et al. The role of ultrasound Cancer Discov. 2015;5:752–67.
evaluation in the detection of early-stage epithelial ovarian cancer. 138. George A, Kaye S, Banerjee S. Delivering widespread BRCA test-
Am J Obstet Gynecol. 2005;192:1214–21; discussion 1221–1212 ing and PARP inhibition to patients with ovarian cancer. Nat Rev
125. Tate TH, Baggett B, Rice PF, et al. Multispectral fluorescence Clin Oncol. 2017;14:284–96.
imaging of human ovarian and fallopian tube tissue for early-stage 139. Gonzalez Martin A. Progress in PARP inhibitors beyond BRCA
cancer detection. J Biomed Opt. 2016;21:56005. mutant recurrent ovarian cancer? Lancet Oncol. 2017;18:8–9.
126. Grabowska-Derlatka L, Derlatka P, Szeszkowski W, et al. Diffusion- 140. Alipour S, Zoghi S, Khalili N, et al. Specific immunother-
weighted imaging of small peritoneal implants in “potentially” apy in ovarian cancer: a systematic review. Immunotherapy.
early-stage ovarian cancer. Biomed Res Int. 2016;2016:9254742. 2016;8:1193–204.
Serous Neoplasms of the Ovary
5
Preetha Ramalingam
© Science Press & Springer Nature Singapore Pte Ltd. 2019 141
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_5
142 P. Ramalingam
line tumors (SBT) [3]. The majority of LGSCs are associ- formed. Malignant serous tumors must be graded in all
ated with SBTs, and both have BRAF and KRAS mutations cases unless there are limitations that preclude accurate
[4–6]. Serous cystadenomas and borderline tumors are grading. The two-tier grading system proposed by Malpica
thought to arise from inclusion cysts in the ovarian paren- et al. is recommended, and tumors must be classified as
chyma as they are histologically and immunophenotypically either LGSC or HGSC [6, 13].
similar, and cystadenomas are distinguished from inclusion
cysts based on an arbitrary 1.0 cm cutoff. Inclusion cysts
from ovaries of patients with SBT have been shown to have 5.2 Benign Serous Tumors
higher levels of epithelial cell aneusomy when compared to
those from ovaries without neoplastic disease [7]. It is now 5.2.1 General Features
known that the inclusion cysts in the ovarian parenchyma,
i.e., endosalpingiosis, are derived from fallopian tube epi- Benign serous tumors typically occur in women from 40 to
thelium and are immunophenotypically different than the 60 years of age. Patients may either be asymptomatic or
ovarian surface epithelium. Inclusion cysts are positive for present with abdominal pain secondary to a large mass or
PAX-8 and negative for calretinin, while the converse is true torsion. Benign serous tumors can range in size from
for ovarian surface epithelium which is mesothelial derived 1.0 cm to up to 30.0 cm but are usually <15.0 cm. Cysts
and positive for calretinin and negative for PAX-8 [3, 8]. A less than 1.0 cm are best reported as inclusion cysts, which
study using tubulin to label ciliated cells showed a progres- are also called as endosalpingiosis. The outer surface is
sive decrease in ciliated cells from normal tubal epithelium, smooth and typically no papillary excrescences are identi-
ovarian inclusion cysts, and SBT to complete absence in fied in the inner wall. Cystadenofibromas may have some
LGSC [8]. HGSC arising from the fallopian tube is thought papillary areas which are more polypoid and edematous/
to be derived from a clonal expansion of tubal secretory fibrotic rather than truly papillary as seen in borderline
cells [9]. Similarly LGSC may represent a clonal expansion tumors (Fig. 5.1).
of secretory cells which acquire BRAF and KRAS rather
than p53 mutations [8].
Serous neoplasms comprise approximately 30% of all
ovarian tumors [10, 11]. About 60–65% are benign, 10% are
borderline tumors, and the remainder are malignant tumors
(25–30%) [10, 11]. The epithelium of the benign tumors
closely resembles that of the fallopian tube, but with border-
line and malignant tumors, the papillary nature, hierarchical
tufting, and psammoma bodies assist with the diagnosis.
Benign serous tumors are termed cystadenoma, adenofi-
broma, or cystadenofibroma depending on the extent of cys-
tic and fibrous component present in the tumor. Tumors that
have epithelial proliferation greater than that expected in
benign serous tumors are classified as SBTs. The 2014
WHO classification of tumors of female reproductive organs
recommends using this terminology for tumors that show
>10% of hierarchical tufting. In our institution, cases with
typical features of borderline tumor, even if less than 10%,
are called focal SBT, especially in postmenopausal women
who can have associated peritoneal implants [12]. Using
this terminology requires careful discussion with the sur-
Fig. 5.1 Gross image of serous cystadenofibroma with papillary areas
geon to ensure that the peritoneal cavity is examined at the that are smooth and polypoid unlike in borderline tumors which have
time of frozen section and limited staging biopsies per- papillary fronds and excrescences
5 Serous Neoplasms of the Ovary 143
a b
Fig. 5.2 (a) Serous cystadenoma showing a fibrous cyst wall lined by a single layer of epithelium without epithelial proliferation or tufting.
(b) Higher-power image showing cuboidal cells with no cytologic atypia and cilia (arrows)
144 P. Ramalingam
a b
Fig. 5.5 (a) Serous cystadenoma with epithelial proliferation on the cystadenofibroma with intracystic proliferation comprising <10%
ovarian surface best designated as serous borderline tumor even though which can be called either focal serous borderline tumor or serous cyst-
it comprises <10% due to the risk of peritoneal involvement. (b) Serous adenofibroma with epithelial proliferation
WHO classification of tumors of female reproductive organs variable involvement by papillary excrescences. The papillae
recommends use of either SBTs or atypical proliferative are grossly tan-yellow soft, friable, and edematous
serous tumors but not serous tumor of low malignant poten- (Fig. 5.7b). The cyst contents are typically serous; however,
tial (SLMP). However, several experts in the field do not mucinous fluid can also be present. Solid fibromatous areas
recommend the use of atypical proliferative serous tumors. may be present within the cyst wall which may grossly have
The concern with use of this terminology is that it may not the appearance of cystadenoma but microscopically could
effectively communicate to the clinician the malignant show areas consistent with borderline tumor; hence sampling
potential for these tumors, albeit in a small proportion of of these areas, in the absence of typical papillary excres-
cases [19, 21]. A recent population-based study has shown cences, is important.
that tumors designated as atypical proliferative serous
tumors have a 5-year absolute serous carcinoma risk of
0.9% but at 20 years was 3.7%. The authors also found that 5.5.2 Histologic Features
the risk of LGSC was significantly higher in patients with
APST than in the general population [22]. The latter finding The histologic appearance of SBT is characteristic with the
reiterates findings of prior reports that patients with APST presence of complex arborizing papillary that show hierar-
do not always have a “benign course” and the term SBT is chical tufting. The papillary emanating from the large edem-
more reflective of the biological behavior of these tumors atous/fibrotic papillae become progressively smaller
[21, 23–25]. Considering atypical proliferative serous resulting in detached papillary epithelial tufts (Fig. 5.8a).
tumors as benign tumors could potentially give a false The lining is either low cuboidal or stratified with at least
understanding of the clinical course of these tumors, which focal presence of ciliated cells (Fig. 5.8b). Cells with abun-
is time dependent, as shown by the proponents of this termi- dant eosinophilic cytoplasm can be seen in a subset of
nology and others [21, 22, 26]. We and many others recom- SBT. Intracytoplasmic mucin may be present in an otherwise
mend the use of SBT. typical SBT, and its focal presence does not warrant a diag-
nosis of a seromucinous borderline tumor. The lining epithe-
lium shows only mild to moderate atypia, and the presence
5.5 Pathological Findings of marked atypia is inconsistent with a diagnosis of
SBT. Tumors with a low-power appearance of SBT that is
5.5.1 Gross Findings intracystic without evidence of “invasion” that have marked
cytologic atypia must be designated as HGSC (Fig. 5.9a, b).
Grossly the tumors can range from 1.0 cm to 35.0 cm with an Therefore, high-power examination of these tumors to be
average size of 11.0 cm. The tumors may be intracystic with certain that the nuclei do not show atypia beyond what is
or without surface involvement or be purely surface tumors expected for SBT is recommended for all cases. Also of note
(Fig. 5.7a). The cyst may be unilocular or multilocular with is that focal HGSC can be present in an otherwise typical
a b
Fig. 5.7 (a) Serous borderline tumor involving surface and parenchyma of both ovaries displaying numerous papillary excrescences. (b) Cut
surface of ovary involved by serous borderline tumor showing delicate edematous papillae without evidence of hemorrhage or necrosis
146 P. Ramalingam
a b
Fig. 5.8 (a) Typical serous borderline tumor showing prominent epithelial proliferation arising from larger fibrotic papillae with hierarchical tuft-
ing. (b) High power of papillae showing cuboidal to columnar cells with minimal cytologic atypia and cilia
a b
Fig. 5.9 (a) Serous tumor resembling serous borderline tumor on low power showing characteristic hierarchical tufting and no invasion. (b)
Higher-power image of the tumor cells with marked atypia consistent with high-grade serous carcinoma
SBT, and careful examination must be performed to exclude 5.5.4 Differential Diagnosis
this possibility [27].
The differential diagnosis for SBTs includes serous cystad-
enofibroma with epithelial proliferation and HGSC, both of
5.5.3 Biomarkers which have been discussed earlier in this chapter. Another
consideration is clear cell carcinoma (CCC), as SBT can
Immunohistochemical stains are rarely necessary to make a show cytoplasmic clearing. The presence of papillary, solid,
diagnosis of SBT. However, they are positive for CK7, and tubulocystic patterns, hobnailing, and marked cytologic
MOC-31, PAX-8, WT-1, and ER/PR [28–31]. In cases where atypia would favor CCC. SBT will be positive for WT-1 and
a HGSC is to be excluded, diffuse staining for p53 and p16 CCC for napsin-A and HNF-1β, if further confirmation is
would favor this diagnosis over SBT [32]. necessary.
5 Serous Neoplasms of the Ovary 147
a b
Fig. 5.10 (a) Micropapillary serous borderline tumor showing slender typically five times long as they are wide, lined by cells with mild to
papillae lacking fibrovascular cores arising from larger edematous moderate atypia and the absence of prominent hierarchical tufting
papillae, imparting a “Medusa head” appearance. (b) The papillae are
148 P. Ramalingam
a b
Fig. 5.13 (a) Serous borderline tumor with microinvasion. The glands (arrowhead). (b) Microinvasion showing both single cells and small
are lined by cells with abundant eosinophilic cytoplasm (arrow), and papillae surrounded by clear spaces with no stromal reaction. The foci
similar cells are seen in the stroma without associated stromal reaction are small and measure <5.0mm
p sammoma bodies may be present in the stroma. Isolated with desmoplastic stromal reaction invading the surrounding
single cells with clear spaces may be present and does not fibrous/adipose tissue (Fig. 5.18). While the criteria seem to
warrant a diagnosis of invasive implant (Fig. 5.17). As a rule, be well established, their application can be quite challeng-
the epithelial stromal ratio is quite low in desmoplastic nonin- ing. There is much variation in interpretation of equivocal
vasive implants. Desmoplastic implants appear as firm implants in which either the epithelium is composed of papil-
plaques or nodules that have a stuck on appearance on the lae surrounded by clear spaces without clear invasion into the
omental surface or in the invagination of the fat lobules. surrounding adipose tissue (Fig. 5.19a) or when the epithelial
Invasive implants have been shown to behave like LGSC, stromal ratio is high within an otherwise desmoplastic nonin-
and the 2014 WHO classification of tumors of female repro- vasive implant (Fig. 5.19b). The former used to be termed
ductive organs recommends using the latter terminology [14]. early invasion, but this terminology has fallen out of favor
Invasive implants are characterized by small micropapillae [41]. Based on some studies, it appears that peritoneal
surrounded by clear spaces or irregular haphazard glands implants that have characteristic features of invasive implants,
especially micropapillary or cribriform pattern, that do not
invade the adjacent adipose tissue (Fig. 5.20) behave similar
to invasive implants/LGSC and should be designated as such
[46]. Review of cases from several small and large academic
centers and community pathologists, as part of our clinical
practice, has shown that these criteria are not uniformly
applied and there is significant interobserver variability in the
classification of implants even among experts [19, 46, 47].
This is particularly the case for implants that are “picked” off
the surface of peritoneal surfaces in which there is no associ-
ated adipose tissue to evaluate the presence or absence of
invasion. It may be impossible to classify an implant as inva-
sive or noninvasive, and “indeterminate for invasion” may be
used in select cases. In these challenging cases, especially, if
this finding is isolated in a well-staged patient, a conservative
approach is recommended. This brings up another issue on
whether the extent/quantity of invasive implant/LGSC involv-
ing the peritoneal cavity at the time of original diagnosis has
Fig. 5.15 Noninvasive peritoneal implant showing epithelial prolifera-
an impact outcome. There is no established data on this issue
tion resembling serous borderline tumor within cystically lined space and this merits further investigation.
a b
Fig. 5.16 (a) Desmoplastic noninvasive implant showing a prolifera- reactive stromal cells with scattered epithelial nests (arrows). The gland
tion of stromal cells with a granulation tissue-like appearance that is to stroma ratio is low
stuck on the omental surface. (b) Higher power shows hemorrhage and
5 Serous Neoplasms of the Ovary 151
Fig. 5.17 Desmoplastic noninvasive implants with isolated nests sur- Fig. 5.18 Invasive peritoneal implant showing marked proliferation of
rounded by clear spaces (arrows) with overall low gland to stroma ratio, papillae surrounded by halos with associated desmoplastic stromal
which does not warrant a diagnosis of invasive implant reaction and infiltration into the adjacent adipose tissue
a b
Fig. 5.19 (a) Implant with small nests surrounded by clear spaces reaction present within an adhesion. The degree of epithelial prolifera-
(arrows) within an adhesion and no invasion into adjacent adipose tis- tion may be considered sufficient for a diagnosis of invasive implant
sue. (b) Implant with marked epithelial proliferation with focal stroma
arise from preexisting endosalpingiosis and do not represent desmoplastic reaction. When single or clusters of epithelial
involvement by the ovarian tumor. Involvement of lymph cells with abundant eosinophilic cytoplasm that resemble cells
nodes by SBT is not an independent adverse prognostic indi- seen in microinvasion (Fig. 5.22) are present, they can mimic
cator and hence is reported as “involvement” rather than intranodal mesothelial cells. Using a limited panel of immu-
“metastatic carcinoma” [42, 48, 49]. nostains including calretinin for mesothelial cells and MOC-
Histologically the tumor can form small clusters of epithe- 31 or BER-EP4, PAX-8, and ER for SBTs can allow the
lial cells surrounded by clear spaces, micropapillae within cys- distinction. The only pattern of significance is that of a non-
tic spaces, or foci with cribriform architecture, usually within intracystic aggregate of tumor cells with or without micropap-
the lymph node sinuses or in the parenchyma (Fig. 5.21a, b). illary pattern that measures >1.0 mm, with associated
The nuclei show mild to moderate atypia, and there is often no desmoplasia (Fig. 5.23), which is consistent with LGSC. This
Fig. 5.20 Implants with a micropapillary pattern present in fibrous Fig. 5.22 Serous borderline tumor resembling single cell pattern
septae within adipose tissue without invasion of the underlying tissue. (arrows) of microinvasion resembling mesothelial cells
Some studies have shown that these lesions behave similar to unequivocal
invasive implants and should be considered as such
a b
Fig. 5.21 (a) Serous borderline tumor involving lymph node paren- line tumor in lymph node sinuses (arrows) with associated calcification
chyma showing epithelial proliferation and tufting that is in excess of and adjacent endosalpingiosis
that acceptable for endosalpingiosis. (b) Minute foci of serous border-
5 Serous Neoplasms of the Ovary 153
Fig. 5.23 Metastatic low-grade serous carcinoma in a lymph node 5.6.7 Management and Outcome
showing a focus of tumor with haphazard glands eliciting a desmoplas-
tic stromal reaction (arrow). Adjacent focus of serous borderline tumor
is present (asterisk) Early reports on survival of patients with stage I of SBTs
reported 99–100% survival rates, but the follow-up was
short. As our understanding of the disease course of SBTs
can be seen in the lymph node in association with ovarian SBT has evolved, these tumors could recur several years or some-
or LGSC. Distinction of SBT from endosalpingiosis is impor- times decades after initial diagnosis. Hence later reports
tant. Endosalpingiosis can have intraglandular papillae, usu- included much longer follow-up, and while the survival is
ally with a fibrovascular core; however, hierarchical tufting is still excellent, it is estimated to be approximately 95–98%.
absent. This is discussed in further detail in the peritoneal Recurrences occur in about 10% of cases and can either
serous tumors section, later in this chapter. recur as SBT or LGSC. However, a minority of patients who
recur as LGSC may have an adverse outcome, resulting in
disease-related death; hence, long-term follow-up is essen-
5.6.6 Genetic Studies tial for these patients. Given the long interval between diag-
nosis and recurrence, these tumors could represent
The pathogenesis of SBT, serous cystadenofibroma, and independent primaries or disease progression.
LGSC has been studied extensively, and there is sufficient Patients with stage II and III disease with noninvasive
evidence that this represents a spectrum of the same disease, implants are at increased risk for recurrence on long-term
both histologically and molecularly. Serous cystadenofibro- follow-up. Earlier reports of excellent overall survival were
mas are not infrequently seen in the background of SBT, and based on a 5-year follow-up which is too short for these
the latter is seen in almost two-thirds of LGSC. Studies have tumors. Regardless, the relapse rate (20–50%) is still much
shown that SBT has BRAF mutations in 23–48% and KRAS less than for patients with invasive implants (>50%).
mutations in 17–39.5% of cases. Interestingly in all the cases The treatment of SBT, especially in young women who
studied, KRAS and BRAF mutations were mutually exclu- want fertility preservation, is surgical resection with surgical
sive. BRAF mutations have been reported only in serous staging including pelvic washing, omentectomy, and perito-
cystadenofibroma associated with SBT. neal biopsies being recommended. There is some debate on
KRAS, BRAF, and ERBB2 are upstream regulators of the the merits of complete staging including pelvic lymphade-
mitogen-activated protein kinase (MAPK) pathway, and nectomy. Earlier studies have shown that lymph node
mutations in these genes result in tumor proliferation. The involvement is the only site of recurrence in about 20% of
incidence of BRAF mutations in SBT is much lower when patients with SBT emphasizing the importance of surgical
compared to LGSC; however, KRAS mutations seem to have staging [51]. Recent study compared the outcomes of patients
a similar incidence in both tumors. These findings seem to with complete staging versus patients without any surgical
suggest that unlike KRAS, BRAF mutations are rarely staging. Only 6% of patients in the first group had lymph
involved in progression of SBT to LGSC. Furthermore, node involvement, and the relapse rates were not signifi-
BRAF mutations have been identified in nonrecurrent SBTs cantly different in the two groups. Hence, currently routine
suggesting a protective role for this gene. In a recent study, lymphadenectomy is not recommended for patients with
KRAS mutations have been identified in 60% of invasive SBT, except in cases with lymphadenopathy [52].
154 P. Ramalingam
5.7 Serous Carcinoma Several risk factors have been postulated for HGSC such as
late age of menarche, parity, and early menopause; however,
5.7.1 Overview the most important risk factor is a family history of ovarian
cancer [70]. Patients with HGSCs are typically postmeno-
Serous carcinomas, particularly HGSC, comprise the majority pausal (mean age of 63 years) and present with symptoms
of all epithelial ovarian cancers [53]. Several studies have now related to disseminated peritoneal disease such as nausea,
shown that HGSC and LGSC represent two distinctly different vomiting, abdominal pain and bloating, constipation, and uri-
disease entities based on clinical, morphologic, immunopheno- nary frequency; however, their non-specific nature delays
typic, and molecular characteristics [54–57]. Though a small diagnosis in some patients [71]. Laboratory testing typically
subset of LGSC may be associated with or “progress” to shows an elevated serum CA-125 usually >500 U/mL; how-
HGSC, the majority develop through a distinct pathway that is ever, this marker is fairly non-specific. Radiologic examina-
different than HGSC. LGSC has been shown to arise in a step- tion shows either unilateral or frequently bilateral complex
wise fashion from inclusion cysts to serous cystadenoma to ovarian masses, with omental and peritoneal lesions.
SBT [8]. However, the precursor for HGSC in the ovary has
been a subject of controversy. Traditionally it was thought to
arise from the ovarian surface mesothelium which presumably 5.8.2 Pathologic Features
underwent tubal-type metaplasia; however, no putative precur-
sor lesion has been identified. An alternative theory that has 5.8.2.1 Gross Features
emerged is that majority of ovarian HGSCs arise from the tubal The gross appearance of HGSC is variable. The tumor may
epithelium of the fimbrial end of the fallopian tube [58–62]. In have solid, hemorrhagic, and necrotic appearance (Fig. 5.24)
asymptomatic BRCA-positive patients who have undergone as well as cystic and papillary components. The tumor size is
risk-reducing bilateral salpingo-oophorectomy, early serous highly variable and ranges from small papillary tumors
carcinoma has been more frequently identified in the fallopian involving the surface to large tumors with parenchymal and
tube than in the ovary. Majority (80%) of these early fallopian surface involvement; mean size is 8.0 cm.
tube cancers are associated with serous tubal intraepithelial
carcinoma (STIC). Furthermore, STIC has been reported to be 5.8.2.2 Histologic Features
present in up to 40–60% of patients with advanced-stage ovar- Histologically, these are serous neoplasms that can have
ian cancer [63, 64]. STIC is cytologically similar to HGSC but varying architectural patterns including papillary, microcys-
is limited to the tubal epithelium without invasion and harbors tic, micropapillary, glandular, and solid patterns but are char-
TP53 mutations [65]. Studies have shown that in disseminated acterized by marked nuclear atypia (Fig. 5.25a–d).
ovarian cancer associated with STIC, both lesions show identi- In some cases HGSC can mimic either endometrioid carci-
cal TP53 mutations supporting a clonal relationship [61, 65]. noma or clear cell carcinoma. The glandular lumen, unlike in
Telomere studies in matched STIC and HGSC have shown endometrioid carcinoma, is slit like with irregular rather than
shorter telomeres in majority of STIC when compared to both smooth borders. A subset of HGSC can have large papillary
normal tubal epithelium and HGSC [66]. These findings sug- fronds mimicking urothelial carcinoma (Fig. 5.26). These
gest that STIC is likely an earlier lesion than HGSC [66].
However, in about 15–30% of HGSC, no lesion is identified in
the fallopian tube, despite extensive sampling using the SEE-
FIM protocol [2, 59, 67]. A recent multicenter study performed
genomic analysis of advanced-stage HGSC with and without
STICs and found no significant differences in the RNA
sequence and miRNA data between the two cohorts suggesting
a common origin for all HGSCs [68]. Having said that, the pos-
sibility of multifocal disease arising in a background of meta-
plasia to tubal-type epithelium must also be considered [2].
a b
c d
Fig. 5.25 High-grade serous carcinoma with varying histologic patterns. (a) Papillary pattern. (b) Microcystic and cribriform pattern. (c)
Micropapillary pattern. (d) Glandular pattern
5.8.3 Pathologic Grading equivocal, the case may be signed out as “serous carcinoma”
with a comment detailing the challenge with grading.
Ovarian serous carcinomas were traditionally graded as well,
moderate and poorly differentiated carcinomas. In 2004 Malpica
and colleagues proposed a two-tier grading system to distinguish 5.8.4 Biomarkers
LGSC from HGSC. Cytologic features including marked nuclear
atypia, pleomorphism, hyperchromasia, and prominent nucleoli, HGSC is positive for epithelial markers such as MOC-31,
with concurrent increase in mitotic activity, >12/10 hpfs, are BER-EP4, and CK7. These tumors are positive for PAX-8,
characteristic of HGSC [6, 13]. This grading system has been WT-1, and hormone receptors ER and PR and are usually dif-
found to be highly reproducible and is recommended by the fusely positive for p16 (Fig. 5.27a–d). Majority of HGSC show
2014 WHO classification of tumors of the female reproductive diffuse staining for ER/PR; particularly ER tends to be consis-
organs. The mitotic activity is the secondary criterion and is par- tently expressed in these tumors. This is in contrast to ER/PR
ticularly not reliable in patients who have had prior chemother- expression in uterine serous carcinoma which is quite low
apy, as it can be quite low. Similarly nuclear atypia can be quite when compared to endometrial endometrioid adenocarcinoma
prominent in patients who have had prior therapy; however, [73]. p53 can show two patterns of staining (Fig. 5.28a, b), one
marked pleomorphism with bizarre atypical cells is more typical in which the tumor cells show diffuse, strong nuclear staining
of HGSC. Of note, an attempt must be made to grade all serous in >60–75% of the tumor cells (mutant phenotype) and the
carcinomas, which is usually possible in most cases. In small other in which there is complete loss of p53 expression (null
core biopsies and post therapy, if grading is challenging, immu- phenotype). The former corresponds to a missense mutation in
nohistochemical stains for p53 and Ki-67 may be used (specifics TP53 while the latter to a nonsense mutation that results in a
discussed later). In rare cases where the histologic findings are truncated version of the protein which is not detected by the
a b
c d
Fig. 5.27 (a) High-grade serous carcinoma positive for MOC31; (b) PAX-8, (c) WT-1, (d) estrogen receptor
5 Serous Neoplasms of the Ovary 157
p53 antibody. In the null phenotype pattern, close examination study that were reported to be negative were histologically
to ensure that the internal control is working is necessary to not compatible with a diagnosis of HGSC based on review by
definitively conclude that there is complete loss of expression seven gynecologic pathologists. In cases that lack TP53 muta-
of p53. Another pattern of staining described in a recent study tion, p53 dysfunction is associated with a copy number gain
of HGSC is that of diffuse cytoplasmic staining in the tumor in MDM2 or MDM4.This results in the typical immunohisto-
cells in the absence of strong nuclear staining [74]. In this chemical pattern described above. Almost 50% of HGSC are
study, four cases of HGSC that showed diffuse cytoplasmic associated with either germline or somatic mutations of
staining were associated with loss-of-function (either indel or BRCA1 and BRCA2 genes. Another characteristic feature is
splicing) mutations of TP53. Loss-of-function mutations are the presence of numerous DNA copy number changes.
typically associated with complete absence of p53 expression,
the so-called null phenotype, and in this small cohort, it appears
that cytoplasmic staining may be another manifestation of this 5.8.6 Management and Outcome
type of mutation. Caution must be use when interpreting com-
plete absence or cytoplasmic staining in tumors with morpho- The standard approach for management of patients with
logic features of HGSC. The immunophenotype of common HGSC typically includes primary surgery followed by adju-
epithelial tumors and mesothelioma is summarized in Table 5.1. vant chemotherapy with platinum-based drug carboplatin
and taxane such as paclitaxcel (Taxol). The extent of residual
disease after primary surgery correlates with response to
5.8.5 Genetic Profile subsequent therapy [76]. In patients who cannot be optimally
debulked, neoadjuvant chemotherapy is recommended fol-
HGSC is associated with TP53 mutations in 96% of HGSCs, lowed by interval debulking after three cycles. In such
per the Cancer Genome Atlas [75]. Recent studies have shown patients preoperative laparoscopic examination to evaluate if
that the majority (71%) of cases in the Cancer Genome Atlas they can be optimally cytoreduced (no tumor nodules
a b
Fig. 5.28 (a) High-grade serous carcinoma showing diffuse strong staining for p53 (mutated phenotype) (b) and complete absence of p53 staining
(null phenotype)
Fig. 5.32 Macropapillary pattern of low-grade serous carcinoma Fig. 5.34 Low grade serous carcinoma with low grade cytology and
showing numerous large edematous papillae surrounded by clear numerous psammoma bodies
spaces. When not so prominent, they can be mistaken for tangential
sectioning of the serous borderline tumor component
Therefore correlation with morphology and p53 staining is
more important.
5.9.3 Biomarkers
Similar to HGSC, LGSCs are positive for PAX-8, WT-1, 5.9.4 Genetic Studies
estrogen receptor (ER), and progesterone receptor (PR).
When distinction of LGSC from HGSC is morphologically LGSC is typically associated with either KRAS or BRAF
challenging, immunohistochemical stains for p16, p53, and mutations in about 35% and 33% of cases, respectively,
Ki-67 may be used. LGSCs show patchy staining for p16, which are also seen in SBTs [86, 87]. KRAS mutations at
wild-type staining for p53 (Fig. 5.36a, b), and low Ki-67 codons 12 and 13 and BRAF mutations at codon 599 are
proliferation index, while HGSCs show diffuse staining for mutually exclusive in that only one of the two mutations is
p16 and p53 and elevated Ki-67 proliferation index [83–85]. present in a given tumor [87]. LGSCs have a fewer number
These markers are also helpful in detecting focal HGSC in of somatic mutations [88], and methylation profiling shows
an otherwise predominantly LGSC. In some cases p16 that they are more closely related to SBT than HGSC [89].
staining may show aberrant overexpression in LGSC Subsequent studies have shown that BRAF mutations are
(Fig. 5.37a, b) or patchy expression in HGSC (Fig. 5.37c, d). much less frequent in advanced-stage LGSC (2–6%) [4, 90].
160 P. Ramalingam
a b
Fig. 5.36 Low grade serous carcinoma, with patchy staining for (a) p16, and (b) wild-type staining for p53
5 Serous Neoplasms of the Ovary 161
a b
c d
Fig. 5.37 (a) Example of low-grade serous carcinoma with banal cytology showing (b) aberrant overexpression of p16. (c) Case of high grade
serous carcinoma with marked cytologic atypia showing (d) patchy staining for p16
appearance with numerous psammoma bodies mimicking squamous differentiation, lack of WT-1 expression, and
LGSC from low power; however, atypia and mitotic activity wild-type staining for p53 favor endometrioid adenocarci-
are usually apparent on high power; hence careful examina- noma, though a small subset of high-grade endometrioid
tion is necessary in all cases. Additionally immunohisto- adenocarcinoma can be diffusely p53 positive [96].
chemical stains for p16, p53, and Ki-67 may be employed in
this differential as these markers are overexpressed in
HGSC, while they show patchy or wild-type expression in 5.10.3 Clear Cell Carcinoma
LGSC. In some cases especially core biopsies, the distinc-
tion may be challenging to make and must be stated as such Clear cell carcinoma can closely mimic HGSC especially
in the report. since both tumors can have papillary pattern and cytoplasmic
clearing (Fig. 5.38a–d). The presence of tubulocystic pattern,
papillae with hyalinized fibrovascular cores, and hobnailing
5.10.2 Endometrioid Adenocarcinoma support a diagnosis of CCC over HGSC. In challenging cases,
use of immunostains such as ER, WT-1, HNF-1β, and napsin-
Endometrioid adenocarcinoma typically lacks the degree of A can be helpful in the differential diagnosis. CCC is typically
nuclear atypia seen in HGSC. The glands do not have intrag- positive for HNF-1β and napsin-A and negative for ER and
landular tufting and papillae seen in HGSC. The presence of WT-1, while HGSC shows the opposite staining pattern [97].
162 P. Ramalingam
a b
c d
Fig 5.38 (a) High-grade serous carcinoma showing cytoplasmic clearing and (b) hobnailing mimicking clear cell carcinoma. Tumor cells are
positive for WT-1 (c) and negative for napsin-A (d)
a b
a b
c d
Fig. 5.40 (a and b) Malignant mesothelioma showing positive staining for (a) WT1 and (b) calretinin and negative staining for (c) MOC31 and
(d) estrogen receptor
164 P. Ramalingam
noted that both GCDFP-15 and mammaglobin are not very 5.11.2 Seromucinous Cystadenoma/
sensitive markers and a negative staining does not exclude a Adenofibroma
breast primary. On the other hand, both these markers can
show staining in a small subset of ovarian carcinomas [98]. Seromucinous cystadenoma/adenofibromas are benign
GATA3 is currently the most sensitive and specific marker tumors that are composed of a mixture of epithelial cells
for breast carcinomas and should be included in the immu- which comprises at least 10% of the lesion. They present
nohistochemical panel when a breast primary is in the dif- in women in the reproductive age group and are detected
ferential diagnosis. PAX-8 is expressed in most HGSC; either incidentally or due to symptoms related to an ovar-
however, we have encountered a few cases of breast carci- ian mass. The tumors are usually unilocular and grossly
noma that express this marker; therefore, using a panel of indistinguishable from typical serous cystadenomas. The
stains rather than reliance on any one marker is cyst may contain serous or mucinous fluid. Histologically,
recommended. the epithelial components are primarily serous and muci-
Other tumors with a well-described micropapillary pat- nous; however, endometrioid, transitional-like, and squa-
tern such as lung and bladder carcinomas can mimic either mous cells may be present. In adenofibromas, the stroma
HGSC or LGSC. For lung adenocarcinomas, the use of PAX- is bland and fibrous. They are treated with unilateral
8, WT-1, ER, and TTF-1 may facilitate the distinction. salpingo-oophorectomy.
Aberrant TTF-1 staining has been reported in a small subset
of ovarian serous carcinomas; hence, a panel of markers
must be employed in this differential diagnosis. Metastatic 5.11.3 Seromucinous Borderline Tumor
urothelial carcinoma with micropapillary pattern can be dis-
tinguished from serous carcinomas using CK7, CK20, These tumors are noninvasive proliferative tumors that
GATA3, ER, and WT-1. were previously considered to be a subtype of mucinous
tumors and were designated as endocervical-type mucinous
borderline tumor [99–101]. In the 2014 WHO classification
5.11 Seromucinous Tumors of tumors of female reproductive organs, they are catego-
rized as a separate group of tumors [14]. The patients are
5.11.1 Overview typically in the reproductive age group and present with
non-specific symptoms related to an ovarian mass. They are
Seromucinous tumors are a new category of tumors in the associated with endometriosis in a third of the cases.
2014 WHO classification of tumors of female reproductive Majority of cases are unilateral; however, up to 30% can be
organs [14]. They are divided into benign, borderline and bilateral [99]. The tumors have an average size of 8–10 cm.
carcinomas, similar to other epithelial tumors of the ovary. The cysts are invariably unilocular with intracystic papil-
The term seromucinous tumor was first coined by Shappell lary excrescences. The cyst wall may contain hemorrhage,
in 2002 [99]. In their series, there was no clinical difference and histologically these usually represent areas of
between cases of endocervical-like mucinous borderline endometriosis.
5 Serous Neoplasms of the Ovary 165
a b
c d
Fig. 5.43 (a) Seromucinous borderline tumor showing areas with dant cytoplasm, i.e., indifferent cells (arrow) and squamous differentia-
prominent cytoplasmic mucin resembling endocervical glands. (b) tion (asterisk), are present. (d) Intracellular neutrophilic infiltrate is
Other areas show distinct tubal-type morphology. (c) Cells with abun- frequently seen
166 P. Ramalingam
a b
Fig. 5.44 (a) Seromucinous carcinoma with endocervical-type (arrow) and squamous differentiation (asterisk). (b) Another example showing
tubal and mucinous differentiation. The tumor was WT-1 positive (not shown) and best classified as low-grade serous carcinoma
5 Serous Neoplasms of the Ovary 167
The prognosis of these tumors in the few reported studies sis with miliary granular appearance or cyst-like lesions with
has been favorable with most patients presenting with stage I psammomatous calcification [108]. As mentioned previ-
disease. Advanced-stage disease has been associated with ously, by definition, the ovaries are either uninvolved or
worse prognosis. show only minimal surface involvement. Grossly the omen-
tum shows adhesions, plaques, and granular lesions.
Histologically, the tumors are similar to noninvasive epithe-
5.11.7 Management and Outcome lial (Fig. 5.45) and desmoplastic implants. Endosalpingiosis
is usually present and can have some intraglandular prolif-
The majority of the tumors both borderline and carcinoma eration (Fig. 5.46a); however, hierarchical tufting when pres-
are stage I; however, peritoneal implants, usually desmoplas- ent warrants a diagnosis of SBT (Fig. 5.46b). The lesions
tic noninvasive, and lymph node involvement can be seen in may be present in the fibrous septae separating adipose tissue
a minority of cases (~10%). The treatment is similar to SBTs nodules; however, no invasion is identified. Adequate sam-
and consists of unilateral or bilateral salpingo-oophorectomy pling of the omental tissue is recommended to rule out areas
with staging biopsies. The prognosis of seromucinous bor- of LGSC.
derline tumors is excellent in the reported cases thus far, even
in the presence of peritoneal implants [99].
a b
Fig. 5.46 (a) Areas of endosalpingiosis with intraluminal papillae without epithelial tufting, which is not sufficient for a diagnosis of serous
borderline tumor. (b) Serous borderline tumor partially involving a gland of endosalpingiosis (arrow)
168 P. Ramalingam
The prognosis for patients with peritoneal SBT is favorable The assessment of peritoneal serous tumors should include
similar to patients with ovarian SBT with noninvasive distinction of LGSC from HGSC using criteria and immuno-
implants. However, similar to their ovarian counterparts, histochemical stains previously described. Other entities
there is a small risk for progression to LGSC; hence, long- such as peritoneal mesothelioma should also be excluded
term follow-up is recommended for these patients. (previously discussed).
The Gynecologic Oncology Group (GOG) established crite- The genetic profile of primary peritoneal HGSC is similar to
ria to distinguish ovarian serous from primary peritoneal other pelvic HGSCs with predominance of TP53 mutations
serous carcinoma in 1993 [109]. Per the GOG criteria, (1) [68, 112]. KRAS and BRAF mutations have been identified in
both ovaries should be normal in size or enlarged by a benign peritoneal LGSC similar to ovarian primaries [87, 113].
process; (2) involvement of extraovarian sites should be
larger than that involving the surface of both ovaries; and (3)
the tumor in the ovarian surface and superficial cortical 5.12.7 Management and Outcome
stroma or within the ovarian parenchyma must measure
5 × 5 mm. Per the College of American Pathologists, a diag- The management and outcome for peritoneal serous carci-
nosis of primary peritoneal serous carcinoma is made in the noma are similar to advanced-stage ovarian carcinoma,
presence of normal bilateral ovaries and fallopian tubes by either HGSC or LGSC. If the patients can be optimally deb-
gross and microscopic examination or when they are enlarged ulked, then surgery followed by adjuvant therapy is the
by benign disease. Hence both ovaries and fallopian tubes mainstay of treatment. In patients who are not surgical can-
must be entirely submitted for histologic evaluation. The didates, neoadjuvant therapy with platinum-based drug car-
presence of either serous tubal intraepithelial carcinoma or boplatin and taxane such as paclitaxcel (Taxol), as used for
tubal mucosal serous carcinoma, regardless of the size, is ovarian serous carcinoma, is recommended.
consistent with a fallopian tube primary. The presence of
serous carcinoma in the fimbrial end, rather than the ovary, in
patients with BRCA mutations and demonstration of clonal
relationship between STIC and extratubal tumor have References
resulted in many investigators proposing that the precursor
1. Maeda D, Ota S, Takazawa Y, Ohashi K, Mori M, Imamura T,
lesion of the majority of “ovarian” HGSC originates in the et al. Mucosal carcinoma of the fallopian tube coexists with ovar-
fallopian tube [61, 65, 110, 111]. The presence of minimal ian cancer of serous subtype only: a study of Japanese cases.
ovarian surface involvement (<5 × 5 mm) is acceptable. The Virchows Arch. 2010;457(5):597–608.
2. Tang S, Onuma K, Deb P, Wang E, Lytwyn A, Sur M, et al.
diagnosis of primary peritoneal serous carcinoma should be Frequency of serous tubal intraepithelial carcinoma in various
made only at the time of primary surgery, prior to neoadju- gynecologic malignancies: a study of 300 consecutive cases. Int J
vant chemotherapy. Of note, while an attempt at accurate site Gynecol Pathol. 2012;31(2):103–10.
designation, i.e., ovary versus fallopian tube versus perito- 3. Kurman RJ, Shih Ie M. The origin and pathogenesis of epithe-
lial ovarian cancer: a proposed unifying theory. Am J Surg Pathol.
neum, must be made in all cases, from a clinical standpoint, 2010;34(3):433–43.
it does not affect either treatment or outcome of these tumors. 4. Grisham RN, Iyer G, Garg K, Delair D, Hyman DM, Zhou Q,
Peritoneal serous carcinoma typically occurs in post- et al. BRAF mutation is associated with early stage disease and
menopausal women with a mean age 62 years for HGSC and improved outcome in patients with low-grade serous ovarian can-
cer. Cancer. 2013;119(3):548–54.
57 years for LGSC. Histologically, the tumors resemble 5. Ho CL, Kurman RJ, Dehari R, Wang TL, Shih Ie M. Mutations
ovarian LGSC or HGSC, and diagnosis is made using crite- of BRAF and KRAS precede the development of ovarian serous
ria established for the ovarian counterparts. borderline tumors. Cancer Res. 2004;64(19):6915–8.
6. Malpica A, Deavers MT, Lu K, Bodurka DC, Atkinson EN,
Gershenson DM, et al. Grading ovarian serous carcinoma using a
two-tier system. Am J Surg Pathol. 2004;28(4):496–504.
5.12.4 Biomarkers 7. Korner M, Burckhardt E, Mazzucchelli L. Different proportions of
aneusomic cells in ovarian inclusion cysts associated with serous
Immunohistochemical stains used for the diagnosis of perito- borderline tumours and serous high-grade carcinomas support dif-
ferent pathogenetic pathways. J Pathol. 2005;207(1):20–6.
neal serous carcinoma are similar to its ovarian counterpart, 8. Li J, Abushahin N, Pang S, Xiang L, Chambers SK, Fadare O,
which has been discussed earlier. It is recommended to add epi- et al. Tubal origin of ‘ovarian’ low-grade serous carcinoma. Mod
thelial and mesothelial markers in the panel, when appropriate. Pathol. 2011;24(11):1488–99.
5 Serous Neoplasms of the Ovary 169
9. Chen EY, Mehra K, Mehrad M, Ning G, Miron A, Mutter GL, 27. Boyd C, McCluggage WG. Low-grade ovarian serous neoplasms
et al. Secretory cell outgrowth, PAX2 and serous carcinogenesis (low-grade serous carcinoma and serous borderline tumor) associ-
in the Fallopian tube. J Pathol. 2010;222(1):110–6. ated with high-grade serous carcinoma or undifferentiated carci-
10. Katsube Y, Berg JW, Silverberg SG. Epidemiologic pathology of noma: report of a series of cases of an unusual phenomenon. Am J
ovarian tumors: a histopathologic review of primary ovarian neo- Surg Pathol. 2012;36(3):368–75.
plasms diagnosed in the Denver Standard Metropolitan Statistical 28. Laury AR, Hornick JL, Perets R, Krane JF, Corson J, Drapkin
Area, 1 July-31 December 1969 and 1 July-31 December 1979. Int R, et al. PAX8 reliably distinguishes ovarian serous tumors from
J Gynecol Pathol. 1982;1(1):3–16. malignant mesothelioma. Am J Surg Pathol. 2010;34(5):627–35.
11. Koonings PP, Campbell K, Mishell DR Jr, Grimes DA. Relative 29. Laury AR, Perets R, Piao H, Krane JF, Barletta JA, French C, et al.
frequency of primary ovarian neoplasms: a 10-year review. Obstet A comprehensive analysis of PAX8 expression in human epithe-
Gynecol. 1989;74(6):921–6. lial tumors. Am J Surg Pathol. 2011;35(6):816–26.
12. Ramalingam PDM, Malpica A. Serous tumor of low malignant 30. Barcena C, Oliva E. WT1 expression in the female genital tract.
potential of the ovary-is the 10% cut off reliable? Mod Pathol. Adv Anat Pathol. 2011;18(6):454–65.
2012;25(S2):294A. 31. Escobar J, Klimowicz AC, Dean M, Chu P, Nation JG, Nelson GS,
13. Malpica A, Deavers MT, Tornos C, Kurman RJ, Soslow R, et al. Quantification of ER/PR expression in ovarian low-grade
Seidman JD, et al. Interobserver and intraobserver variability of a serous carcinoma. Gynecol Oncol. 2013;128(2):371–6.
two-tier system for grading ovarian serous carcinoma. Am J Surg 32. Altman AD, Nelson GS, Ghatage P, McIntyre JB, Capper D, Chu
Pathol. 2007;31(8):1168–74. P, et al. The diagnostic utility of TP53 and CDKN2A to distin-
14. Kurman RJ Carcangiu ML, Herrington CS, et al. WHO guish ovarian high-grade serous carcinoma from low-grade serous
Classification of Tumours of Female Repreductive Organs. ovarian tumors. Mod Pathol. 2013;26(9):1255–63.
International Agency for Resarch on Cancer. 2014. 33. Burks RT, Sherman ME, Kurman RJ. Micropapillary serous car-
15. Bell DA, Scully RE. Ovarian serous borderline tumors with cinoma of the ovary. A distinctive low-grade carcinoma related to
stromal microinvasion: a report of 21 cases. Hum Pathol. serous borderline tumors. Am J Surg Pathol. 1996;20(11):1319–30.
1990;21(4):397–403. 34. du Bois A, Ewald-Riegler N, de Gregorio N, Reuss A, Mahner
16. de Nictolis M, Montironi R, Tommasoni S, Carinelli S, Ojeda B, S, Fotopoulou C, et al. Borderline tumours of the ovary: a cohort
Matias-Guiu X, et al. Serous borderline tumors of the ovary. A study of the Arbeitsgmeinschaft Gynakologische Onkologie
clinicopathologic, immunohistochemical, and quantitative study (AGO) Study Group. Eur J Cancer. 2013;49(8):1905–14.
of 44 cases. Cancer. 1992;70(1):152–60. 35. Seidman JD, Ronnett BM, Kurman RJ. Pathology of borderline
17. Eichhorn JH, Bell DA, Young RH, Scully RE. Ovarian serous (low malignant potential) ovarian tumours. Best Pract Res Clin
borderline tumors with micropapillary and cribriform patterns: a Obstet Gynaecol. 2002;16(4):499–512.
study of 40 cases and comparison with 44 cases without these pat- 36. Sangoi AR, McKenney JK, Dadras SS, Longacre TA. Lymphatic
terns. Am J Surg Pathol. 1999;23(4):397–409. vascular invasion in ovarian serous tumors of low malignant
18. Kurman RJ, Seidman JD, Shih IM. Serous borderline tumours of potential with stromal microinvasion: a case control study. Am J
the ovary. Histopathology. 2005;47(3):310–5. Surg Pathol. 2008;32(2):261–8.
19. Prat J, De Nictolis M. Serous borderline tumors of the ovary: a 37. Maniar KP, Wang Y, Visvanathan K, Shih Ie M, Kurman
long-term follow-up study of 137 cases, including 18 with a RJ. Evaluation of microinvasion and lymph node involvement in
micropapillary pattern and 20 with microinvasion. Am J Surg ovarian serous borderline/atypical proliferative serous tumors: a
Pathol. 2002;26(9):1111–28. morphologic and immunohistochemical analysis of 37 cases. Am
20. Bell DA, Longacre TA, Prat J, Kohn EC, Soslow RA, Ellenson J Surg Pathol. 2014;38(6):743–55.
LH, et al. Serous borderline (low malignant potential, atypi- 38. McKenney JK, Balzer BL, Longacre TA. Patterns of stromal
cal proliferative) ovarian tumors: workshop perspectives. Hum invasion in ovarian serous tumors of low malignant potential
Pathol. 2004;35(8):934–48. (borderline tumors): a reevaluation of the concept of stromal
21. Silva EG, Gershenson DM, Malpica A, Deavers M. The recur- microinvasion. Am J Surg Pathol. 2006;30(10):1209–21.
rence and the overall survival rates of ovarian serous borderline 39. Kennedy AW, Hart WR. Ovarian papillary serous tumors of low
neoplasms with noninvasive implants is time dependent. Am J malignant potential (serous borderline tumors). A long-term
Surg Pathol. 2006;30(11):1367–71. follow-up study, including patients with microinvasion, lymph
22. Hannibal CG, Vang R, Junge J, Frederiksen K, Kurman RJ, Kjaer node metastasis, and transformation to invasive serous carcinoma.
SK. A nationwide study of ovarian serous borderline tumors in Cancer. 1996;78(2):278–86.
Denmark 1978-2002. Risk of recurrence, and development of 40. Kraus JA, Seidman JD. The relationship between papillary infarc-
ovarian serous carcinoma. Gynecol Oncol. 2017;144(1):174–80. tion and microinvasion in ovarian atypical proliferative (“bor-
23. Gershenson DM, Silva EG, Tortolero-Luna G, Levenback C, derline”) serous and seromucinous tumors. Int J Gynecol Pathol.
Morris M, Tornos C. Serous borderline tumors of the ovary with 2010;29(4):303–9.
noninvasive peritoneal implants. Cancer. 1998;83(10):2157–63. 41. Gershenson DM, Silva EG, Levy L, Burke TW, Wolf JK, Tornos
24. Silverberg SG, Bell DA, Kurman RJ, Seidman JD, Prat J, Ronnett C. Ovarian serous borderline tumors with invasive peritoneal
BM, et al. Borderline ovarian tumors: key points and workshop implants. Cancer. 1998;82(6):1096–103.
summary. Hum Pathol. 2004;35(8):910–7. 42. Seidman JD, Kurman RJ. Ovarian serous borderline tumors: a
25. Deavers MT, Gershenson DM, Tortolero-Luna G, Malpica A, Lu critical review of the literature with emphasis on prognostic indi-
KH, Silva EG. Micropapillary and cribriform patterns in ovarian cators. Hum Pathol. 2000;31(5):539–57.
serous tumors of low malignant potential: a study of 99 advanced 43. Segal GH, Hart WR. Ovarian serous tumors of low malignant
stage cases. Am J Surg Pathol. 2002;26(9):1129–41. potential (serous borderline tumors). The relationship of exo-
26. Vang R, Hannibal CG, Junge J, Frederiksen K, Kjaer SK, Kurman phytic surface tumor to peritoneal “implants”. Am J Surg Pathol.
RJ. Long-term behavior of serous borderline tumors subdivided 1992;16(6):577–83.
into atypical proliferative tumors and noninvasive low-grade car- 44. Michael H, Roth LM. Invasive and noninvasive implants in
cinomas: a population-based clinicopathologic study of 942 cases. ovarian serous tumors of low malignant potential. Cancer.
Am J Surg Pathol. 2017;41(6):725–37. 1986;57(6):1240–7.
170 P. Ramalingam
45. Bell DA, Weinstock MA, Scully RE. Peritoneal implants of ovar- 64. Lee Y, Miron A, Drapkin R, Nucci MR, Medeiros F, Saleemuddin
ian serous borderline tumors. Histologic features and prognosis. A, et al. A candidate precursor to serous carcinoma that originates
Cancer. 1988;62(10):2212–22. in the distal fallopian tube. J Pathol. 2007;211(1):26–35.
46. Bell KA, Smith Sehdev AE, Kurman RJ. Refined diagnostic cri- 65. Kindelberger DW, Lee Y, Miron A, Hirsch MS, Feltmate C,
teria for implants associated with ovarian atypical proliferative Medeiros F, et al. Intraepithelial carcinoma of the fimbria and pel-
serous tumors (borderline) and micropapillary serous carcinomas. vic serous carcinoma: evidence for a causal relationship. Am J
Am J Surg Pathol. 2001;25(4):419–32. Surg Pathol. 2007;31(2):161–9.
47. McKenney JK, Gilks CB, Kalloger S, Longacre TA. Classification 66. Kuhn E, Meeker A, Wang TL, Sehdev AS, Kurman RJ, Shih Ie
of extraovarian implants in patients with ovarian serous borderline M. Shortened telomeres in serous tubal intraepithelial carcinoma:
tumors (tumors of low malignant potential) based on clinical out- an early event in ovarian high-grade serous carcinogenesis. Am J
come. Am J Surg Pathol. 2016;40(9):1155–64. Surg Pathol. 2010;34(6):829–36.
48. Djordjevic B, Clement-Kruzel S, Atkinson NE, Malpica A. Nodal 67. Salvador S, Rempel A, Soslow RA, Gilks B, Huntsman D, Miller
endosalpingiosis in ovarian serous tumors of low malignant poten- D. Chromosomal instability in fallopian tube precursor lesions
tial with lymph node involvement: a case for a precursor lesion. of serous carcinoma and frequent monoclonality of synchronous
Am J Surg Pathol. 2010;34(10):1442–8. ovarian and fallopian tube mucosal serous carcinoma. Gynecol
49. Longacre TA, McKenney JK, Tazelaar HD, Kempson RL, Oncol. 2008;110(3):408–17.
Hendrickson MR. Ovarian serous tumors of low malignant poten- 68. Ducie J, Dao F, Considine M, Olvera N, Shaw PA, Kurman RJ,
tial (borderline tumors): outcome-based study of 276 patients et al. Molecular analysis of high-grade serous ovarian carcinoma
with long-term (> or =5-year) follow-up. Am J Surg Pathol. with and without associated serous tubal intra-epithelial carci-
2005;29(6):707–23. noma. Nat Commun. 2017;8(1):990.
50. Zuo T, Wong S, Buza N, Hui P. KRAS mutation of extraovar- 69. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin
ian implants of serous borderline tumor: prognostic indicator for DM. Estimates of worldwide burden of cancer in 2008:
adverse clinical outcome. Mod Pathol. 2018;31(2):350–7. GLOBOCAN 2008. Int J Cancer. 2010;127(12):2893–917.
51. Djordjevic B, Malpica A. Lymph node involvement in ovarian 70. Chen S, Parmigiani G. Meta-analysis of BRCA1 and
serous tumors of low malignant potential: a clinicopathologic BRCA2 penetrance. J Clin Oncol Off J Am Soc Clin Oncol.
study of thirty-six cases. Am J Surg Pathol. 2010;34(1):1–9. 2007;25(11):1329–33.
52. Gershenson DM. Management of borderline ovarian tumours. 71. Goff BA, Mandel LS, Melancon CH, Muntz HG. Frequency of
Best Pract Res Clin Obstet Gynaecol. 2017;41:49–59. symptoms of ovarian cancer in women presenting to primary care
53. Seidman JD, Horkayne-Szakaly I, Haiba M, Boice CR, Kurman clinics. JAMA. 2004;291(22):2705–12.
RJ, Ronnett BM. The histologic type and stage distribution of 72. Takeuchi T, Ohishi Y, Imamura H, Aman M, Shida K, Kobayashi
ovarian carcinomas of surface epithelial origin. Int J Gynecol H, et al. Ovarian transitional cell carcinoma represents a poorly
Pathol. 2004;23(1):41–4. differentiated form of high-grade serous or endometrioid adeno-
54. Diaz-Padilla I, Malpica AL, Minig L, Chiva LM, Gershenson DM, carcinoma. Am J Surg Pathol. 2013;37(7):1091–9.
Gonzalez-Martin A. Ovarian low-grade serous carcinoma: a com- 73. Darvishian F, Hummer AJ, Thaler HT, Bhargava R, Linkov I,
prehensive update. Gynecol Oncol. 2012;126(2):279–85. Asher M, et al. Serous endometrial cancers that mimic endome-
55. McCluggage WG. My approach to and thoughts on the typing of trioid adenocarcinomas: a clinicopathologic and immunohisto-
ovarian carcinomas. J Clin Pathol. 2008;61(2):152–63. chemical study of a group of problematic cases. Am J Surg Pathol.
56. Soslow RA. Histologic subtypes of ovarian carcinoma: an over- 2004;28(12):1568–78.
view. Int J Gynecol Pathol. 2008;27(2):161–74. 74. Kobel M, Piskorz AM, Lee S, Lui S, LePage C, Marass F, et al.
57. Vang R, Shih Ie M, Kurman RJ. Ovarian low-grade and high- Optimized p53 immunohistochemistry is an accurate predic-
grade serous carcinoma: pathogenesis, clinicopathologic and tor of TP53 mutation in ovarian carcinoma. J Pathol Clin Res.
molecular biologic features, and diagnostic problems. Adv Anat 2016;2(4):247–58.
Pathol. 2009;16(5):267–82. 75. Ahmed AA, Etemadmoghadam D, Temple J, Lynch AG, Riad M,
58. Herrington CS, McCluggage WG. The emerging role of the distal Sharma R, et al. Driver mutations in TP53 are ubiquitous in high
Fallopian tube and p53 in pelvic serous carcinogenesis. J Pathol. grade serous carcinoma of the ovary. J Pathol. 2010;221(1):49–56.
2010;220(1):5–6. 76. Elattar A, Bryant A, Winter-Roach BA, Hatem M, Naik R. Optimal
59. Przybycin CG, Kurman RJ, Ronnett BM, Shih Ie M, Vang R. Are primary surgical treatment for advanced epithelial ovarian cancer.
all pelvic (nonuterine) serous carcinomas of tubal origin? Am J Cochrane Database Syst Rev. 2011;8:CD007565.
Surg Pathol. 2010;34(10):1407–16. 77. Nick AM, Coleman RL, Ramirez PT, Sood AK. A framework for
60. Zeppernick F, Meinhold-Heerlein I, Shih IM. Precursors of ovar- a personalized surgical approach to ovarian cancer. Nat Rev Clin
ian cancer in the fallopian tube: serous tubal intraepithelial carci- Oncol. 2015;12(4):239–45.
noma–an update. J Obstet Gynaecol Res. 2015;41(1):6–11. 78. Aletti GD, Dowdy SC, Gostout BS, Jones MB, Stanhope CR,
61. Kuhn E, Kurman RJ, Vang R, Sehdev AS, Han G, Soslow R, Wilson TO, et al. Aggressive surgical effort and improved
et al. TP53 mutations in serous tubal intraepithelial carcinoma survival in advanced-stage ovarian cancer. Obstet Gynecol.
and concurrent pelvic high-grade serous carcinoma–evidence 2006;107(1):77–85.
supporting the clonal relationship of the two lesions. J Pathol. 79. Chi DS, Eisenhauer EL, Lang J, Huh J, Haddad L, Abu-Rustum
2012;226(3):421–6. NR, et al. What is the optimal goal of primary cytoreductive sur-
62. Vang R, Visvanathan K, Gross A, Maambo E, Gupta M, gery for bulky stage IIIC epithelial ovarian carcinoma (EOC)?
Kuhn E, et al. Validation of an algorithm for the diagnosis of Gynecol Oncol. 2006;103(2):559–64.
serous tubal intraepithelial carcinoma. Int J Gynecol Pathol. 80. Berns EM, Bowtell DD. The changing view of high-grade serous
2012;31(3):243–53. ovarian cancer. Cancer Res. 2012;72(11):2701–4.
63. Crum CP, Drapkin R, Miron A, Ince TA, Muto M, Kindelberger 81. Gershenson DM, Sun CC, Lu KH, Coleman RL, Sood AK,
DW, et al. The distal fallopian tube: a new model for pelvic serous Malpica A, et al. Clinical behavior of stage II-IV low-grade serous
carcinogenesis. Curr Opin Obstet Gynecol. 2007;19(1):3–9. carcinoma of the ovary. Obstet Gynecol. 2006;108(2):361–8.
5 Serous Neoplasms of the Ovary 171
82. Malpica A, Deavers MT. Ovarian low-grade serous carcinoma 98. Gown AM, Fulton RS, Kandalaft PL. Markers of metastatic carci-
involving the cervix mimicking a cervical primary. Int J Gynecol noma of breast origin. Histopathology. 2016;68(1):86–95.
Pathol. 2011;30(6):613–9. 99. Shappell HW, Riopel MA, Smith Sehdev AE, Ronnett BM, Kurman
83. O’Neill CJ, McBride HA, Connolly LE, Deavers MT, Malpica RJ. Diagnostic criteria and behavior of ovarian seromucinous
A, McCluggage WG. High-grade ovarian serous carcinoma (endocervical-type mucinous and mixed cell-type) tumors: atypi-
exhibits significantly higher p16 expression than low-grade cal proliferative (borderline) tumors, intraepithelial, microinvasive,
serous carcinoma and serous borderline tumour. Histopathology. and invasive carcinomas. Am J Surg Pathol. 2002;26(12):1529–41.
2007;50(6):773–9. 100. Lee KR, Nucci MR. Ovarian mucinous and mixed epithelial
84. Phillips V, Kelly P, McCluggage WG. Increased p16 expression carcinomas of mullerian (endocervical-like) type: a
in high-grade serous and undifferentiated carcinoma compared clinicopathologic analysis of four cases of an uncommon
with other morphologic types of ovarian carcinoma. Int J Gynecol variant associated with endometriosis. Int J Gynecol Pathol.
Pathol. 2009;28(2):179–86. 2003;22(1):42–51.
85. Vang R, Gown AM, Farinola M, Barry TS, Wheeler DT, 101. Wu CH, Mao TL, Vang R, Ayhan A, Wang TL, Kurman RJ, et al.
Yemelyanova A, et al. p16 expression in primary ovarian Endocervical-type mucinous borderline tumors are related to
mucinous and endometrioid tumors and metastatic adenocar- endometrioid tumors based on mutation and loss of expression of
cinomas in the ovary: utility for identification of metastatic ARID1A. Int J Gynecol Pathol. 2012;31(4):297–303.
HPV-related endocervical adenocarcinomas. Am J Surg Pathol. 102. Dube V, Roy M, Plante M, Renaud MC, Tetu B. Mucinous ovar-
2007;31(5):653–63. ian tumors of Mullerian-type: an analysis of 17 cases including
86. Sieben NL, Macropoulos P, Roemen GM, Kolkman-Uljee SM, borderline tumors and intraepithelial, microinvasive, and invasive
Jan Fleuren G, Houmadi R, et al. In ovarian neoplasms, BRAF, but carcinomas. Int J Gynecol Pathol. 2005;24(2):138–46.
not KRAS, mutations are restricted to low-grade serous tumours. J 103. Rodriguez IM, Irving JA, Prat J. Endocervical-like mucinous bor-
Pathol. 2004;202(3):336–40. derline tumors of the ovary: a clinicopathologic analysis of 31
87. Singer G, Oldt R 3rd, Cohen Y, Wang BG, Sidransky D, Kurman cases. Am J Surg Pathol. 2004;28(10):1311–8.
RJ, et al. Mutations in BRAF and KRAS characterize the develop- 104. Taylor J, McCluggage WG. Ovarian seromucinous carcinoma:
ment of low-grade ovarian serous carcinoma. J Natl Cancer Inst. report of a series of a newly categorized and uncommon neo-
2003;95(6):484–6. plasm. Am J Surg Pathol. 2015;39(7):983–92.
88. Jones S, Wang TL, Kurman RJ, Nakayama K, Velculescu 105. Rambau PF, McIntyre JB, Taylor J, Lee S, Ogilvie T, Sienko A,
VE, Vogelstein B, et al. Low-grade serous carcinomas of the et al. Morphologic reproducibility, genotyping, and immunohis-
ovary contain very few point mutations. J Pathol. 2012;226(3): tochemical profiling do not support a category of seromucinous
413–20. carcinoma of the ovary. Am J Surg Pathol. 2017;41(5):685–95.
89. Shih Ie M, Chen L, Wang CC, Gu J, Davidson B, Cope L, et al. 106. Biscotti CV, Hart WR. Peritoneal serous micropapillomatosis of
Distinct DNA methylation profiles in ovarian serous neoplasms low malignant potential (serous borderline tumors of the perito-
and their implications in ovarian carcinogenesis. Am J Obstet neum). A clinicopathologic study of 17 cases. Am J Surg Pathol.
Gynecol. 2010;203(6):584.e1–22. 1992;16(5):467–75.
90. Wong KK, Tsang YT, Deavers MT, Mok SC, Zu Z, Sun C, et al. 107. Bell DA, Scully RE. Serous borderline tumors of the peritoneum.
BRAF mutation is rare in advanced-stage low-grade ovarian Am J Surg Pathol. 1990;14(3):230–9.
serous carcinomas. Am J Pathol. 2010;177(4):1611–7. 108. Go HS, Hong HS, Kim JW, Woo JY. CT appearance of primary
91. Gershenson DM. Low-grade serous carcinoma of the ovary or peritoneal serous borderline tumour: a rare epithelial tumour of
peritoneum. Ann Oncol. 2016;27(Suppl 1):i45–i9. the peritoneum. Br J Radiol. 2012;85(1009):e22–5.
92. Gershenson DM. The life and times of low-grade serous carci- 109. Bloss JD, Liao SY, Buller RE, Manetta A, Berman ML, McMeekin
noma of the ovary. Am Soc Clin Oncol Educ Book. 2013; https:// S, et al. Extraovarian peritoneal serous papillary carcinoma: a
doi.org/10.1200/EdBook_AM.2013.33.e195. case-control retrospective comparison to papillary adenocarci-
93. Gershenson DM, Bodurka DC, Coleman RL, Lu KH, Malpica A, noma of the ovary. Gynecol Oncol. 1993;50(3):347–51.
Sun CC. Hormonal maintenance therapy for women with low- 110. Colgan TJ, Murphy J, Cole DE, Narod S, Rosen B. Occult car-
grade serous cancer of the ovary or peritoneum. J Clin Oncol Off cinoma in prophylactic oophorectomy specimens: prevalence
J Am Soc Clin Oncol. 2017;35(10):1103–11. and association with BRCA germline mutation status. Am J Surg
94. Gershenson DM, Sun CC, Iyer RB, Malpica AL, Kavanagh JJ, Pathol. 2001;25(10):1283–9.
Bodurka DC, et al. Hormonal therapy for recurrent low-grade 111. Gilks CB, Irving J, Kobel M, Lee C, Singh N, Wilkinson N,
serous carcinoma of the ovary or peritoneum. Gynecol Oncol. et al. Incidental nonuterine high-grade serous carcinomas arise
2012;125(3):661–6. in the fallopian tube in most cases: further evidence for the
95. Farley J, Brady WE, Vathipadiekal V, Lankes HA, Coleman R, tubal origin of high-grade serous carcinomas. Am J Surg Pathol.
Morgan MA, et al. Selumetinib in women with recurrent low- 2015;39(3):357–64.
grade serous carcinoma of the ovary or peritoneum: an open-label, 112. Schorge JO, Muto MG, Lee SJ, Huang LW, Welch WR, Bell
single-arm, phase 2 study. Lancet Oncol. 2013;14(2):134–40. DA, et al. BRCA1-related papillary serous carcinoma of the
96. Okuda T, Otsuka J, Sekizawa A, Saito H, Makino R, Kushima M, peritoneum has a unique molecular pathogenesis. Cancer Res.
et al. p53 mutations and overexpression affect prognosis of ovar- 2000;60(5):1361–4.
ian endometrioid cancer but not clear cell cancer. Gynecol Oncol. 113. Singer G, Stohr R, Cope L, Dehari R, Hartmann A, Cao DF, et al.
2003;88(3):318–25. Patterns of p53 mutations separate ovarian serous borderline
97. Kobel M, Kalloger SE, Carrick J, Huntsman D, Asad H, Oliva E, tumors and low- and high-grade carcinomas and provide support
et al. A limited panel of immunomarkers can reliably distinguish for a new model of ovarian carcinogenesis: a mutational analy-
between clear cell and high-grade serous carcinoma of the ovary. sis with immunohistochemical correlation. Am J Surg Pathol.
Am J Surg Pathol. 2009;33(1):14–21. 2005;29(2):218–24.
Ovarian Endometrioid and Clear-Cell
Tumors 6
Jennifer Katzenberg and Andres A. Roma
a b
Fig. 6.1 (a) Medium magnification of endometrioma composed of almost devoid of endometrial type stroma, but with persistent vessels
endometrial type epithelium and focal superficial dense stroma with and hemosiderin deposits. (d) Endometriosis with atypia consistent
vessels. (b) High power of endometriosis. (c) Area of endometriosis with atypical endometriosis
© Science Press & Springer Nature Singapore Pte Ltd. 2019 173
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_6
174 J. Katzenberg and A. A. Roma
c d
Fig. 6.1 (continued)
a b
Fig. 6.3 (a) Frozen section of an endometrioid adenofibroma composed of thick fibrotic, polypoid, bulbous papillary structures lined by cuboidal
epithelium. (b) Adjacent area with an abundance of endometrioid-type glands within a polypoid papillary structure
often unilateral with nonspecific symptoms; pain can result triosis, the main differential diagnosis is with a borderline
from torsion or bleeding [11, 13, 14]. tumor or even carcinoma. Endometriosis lacks the dense
fibromatous stroma, has more cellular endometrial type
Pathologic Findings stroma, and evidence of recent or remote hemorrhage.
Grossly, these tumors display substantial variability in size, Benign serous or mucinous cysts have a different epithe-
and may exceed 15 cm in greatest dimension in some cases. lial lining, displaying ciliated epithelium and intracyto-
Adenofibromas have a solid, white, firm cut surface and are plasmic mucin, respectively. Clear-cell and mucinous
usually unilateral; they may display a smooth fibroma-like cystadenofibromas are rare [10, 11]. Borderline tumors,
external surface or show a cauliflower-like appearance with described later, have more epithelial proliferation and
papillary excrescences [10, 11]. Papillary structures are thick atypia, while carcinomas have definitive features of
and fibrotic, which can be used to differentiate these tumors malignancy, including invasion, glandular crowding, and
from serous borderline tumors, which often have more significant atypia. In particular, serous borderline tumors
edematous papillae (Fig. 6.3a). Tumors have a predominance can be confused with endometrioid cystadenofibromas
of fibrous stroma and widely spaced glands within the with papillary features; however, the papillae are more
stroma, which is hypocellular in comparison to normal endo- fibrotic in endometrioid cystadenofibromas and serous
metrium (Fig. 6.3b), and can be associated with an endome- borderline tumors harbor abundant epithelial prolifera-
triotic cyst or endometriosis. Alternatively, they may form a tion, mild atypia, and cell detachment (i.e., detached pap-
nodule within the wall of an otherwise cystic/multicystic illary fragments devoid of stroma).
lesion. Stromal cells are typically spindled with scant cyto-
plasm and tapering nuclei. The stromal component domi- Management and Outcomes
nates in adenofibroma while the cystic component is more These are benign tumors and the prognosis is excellent; most
conspicuous in cystadenofibroma. The epithelial component tumors are treated with simple resection without staging [10,
is characterized by endometrial type epithelium, sometimes 11].
simple, cuboidal, or cylindrical, but most frequently strati-
fied inactive or proliferative-type endometrium [10, 11].
Other endometrial phases can be present, such as secretory 6.2.2 Borderline Endometrioid Tumors
patterns, or may be atrophic or metaplastic. There is usually
mild to absent atypia or mitosis, and no periglandular stro- The category of borderline tumor is well defined for serous
mal cuffing. Dystrophic stromal calcifications can be and mucinous tumors; however, there is less agreement in
present. the specific criteria that determine an endometrioid tumor to
be borderline [10]. Borderline ovarian tumors have more epi-
Differential Diagnosis thelial proliferation than is seen in their benign counterparts,
While it is important to differentiate from other benign and display variable nuclear atypia; however, in contrast to
ovarian adenofibromatous or cystic lesions and endome- carcinomas, there are no definitive features of malignancy,
176 J. Katzenberg and A. A. Roma
a b
Fig. 6.4 (a) Low power of endometrioid tumor with clustered, yet mildly atypical, endometrioid glands within a fibromatous stroma, features of
borderline tumor (courtesy Dr. Wenxin Zheng). (b) High-power magnification demonstrating mild atypia (courtesy Dr. Wenxin Zheng)
including destructive stromal invasion, and their prognosis is 6.2.2.4 Management and Outcomes
much better than that of carcinomas [16]. Endometrioid borderline ovarian tumors carry a good prog-
nosis. Most tumors are stage I; therefore surgical staging is
6.2.2.1 Clinical Features not necessary in most cases; however, in most series,
While endometrioid carcinomas are common ovarian tumors, patients were treated with total abdominal hysterectomy
borderline or atypical proliferative tumors (as they are also and bilateral salpingo-oophorectomy. Rare recurrences
known) are rare [17]. In one of the largest studies to date, were recorded, but malignant behavior has yet to be
Bell et al. reported 33 tumors. Patient age ranged from 24 to reported [11, 17, 18].
85 years with a mean of 46 years [17]. Presentation included
a mass in half the patients, pain in few, and incidental finding
in the remaining patients. Most tumors were unilateral, solid, 6.2.3 Ovarian Endometrioid Carcinomas
and cystic. More than 60% are associated with endometriosis
[10, 11, 17]. These tumors are malignant epithelial tumors that resemble
endometrioid carcinoma of the uterine corpus at the morpho-
6.2.2.2 Pathologic Findings logic level.
Tumors may display a predominant fibromatous back-
ground or may show an intracystic component. Predominant 6.2.3.1 Clinical Features
glandular and papillary patterns have also been described Ovarian endometrioid carcinoma accounts for less than 5%
[17]. of all ovarian tumors, and between 10 and 20% of all ovarian
Similarly to adenofibromas, the majority of these carcinomas [10, 11, 19]. They primarily present in post-
tumors contain abundant glands surrounded by a dense menopausal women with average age of 56 years. Most are
fibromatous stroma (Fig. 6.4a, b). The glands in these low grade and/or present at early stage, and more than 60%
tumors demonstrate moderate architectural complexity of cases arise from endometriotic cysts or have evidence of
and are more closely packed. As reported previously, most endometriosis elsewhere in the patient [10, 11, 19].
cases display grade 1 nuclei, while a third contain grade 2
nuclei [17]. The presence of extensive necrosis was found 6.2.3.2 Pathologic Findings
in one-third of cases, but predominantly the ischemic or The majority of tumors are cystic or multicystic, with varying
infarct type [17]. degrees of solid growth, either interposed between cysts or
The category of borderline or atypical proliferative projecting into cysts [17, 20]. Some tumors can be completely
tumor with intraepithelial carcinoma was described in solid with hemorrhage and necrosis. Other tumors can arise
tumors with diffuse, marked, cytologic atypia evidenced from an endometriotic cyst and might include areas of atypi-
by nuclear enlargement and pleomorphism with large, cal endometriosis (Fig. 6.5a, b); intracystic tumors can pres-
prominent nucleoli [17]. Its significance is uncertain ent as a polypoid nodule projecting into the lumen of a
since none of these patients have recurred or died of thick-walled endometriotic “chocolate” cyst (Fig. 6.6). Size is
disease. variable with a mean tumor size of 14–15 cm (range, 4–30 cm)
and tumors are uncommonly bilateral [17, 20]. Necrosis and
6.2.2.3 Differential Diagnosis hemorrhage are seen in less than half of the cases.
The main differential diagnosis is endometrioid adenocarci- The diagnosis of carcinoma is based on one of the two
noma, which is further detailed in the following section. morphologic features. When the tumor displays destructive,
6 Ovarian Endometrioid and Clear-Cell Tumors 177
a b
Fig. 6.5 (a, b) Low, medium, and high power of endometriotic cyst with an atypical lining. Note the simple lining of cells with large, pleomorphic
nuclei and hobnail features. These findings are not sufficient for a diagnosis of carcinoma
b
infiltrative growth the diagnosis is evident; these tumors har-
bor irregular, and in some cases incomplete or fragmented,
glands within a desmoplastic stroma (Fig. 6.7a, b).
Additionally, as reported by Bell et al., tumors with conflu-
ent glandular growth exceeding 5 mm in dimension are also
designated as carcinomas. This is the most common pattern
seen, at least in one series (Fig. 6.8a, b) [17, 20]. The glandu-
lar proliferation consists of crowded, back-to-back
endometrioid glands with moderate-to-severe cytologic
atypia, often with stratified epithelium, as well as glands
with extensive branching and budding, with concomitant
exclusion of intervening stroma or glands with substantial
internal complexity (bridging and cribriform growth pattern)
(Fig. 6.9a–h) [17, 20]. Nearly all tumors in one series dem-
onstrated background areas of borderline tumor (noninvasive Fig. 6.7 (a, b) Areas of stromal invasion with desmoplastic stroma and
component) or an adenofibromatous background [20]. incomplete glands
178 J. Katzenberg and A. A. Roma
a b
Fig. 6.8 (a) Adenocarcinoma composed of endometrioid glands with growth smaller than 5 mm; in isolation this focus would be insufficient
confluent growth and minimal intervening stroma exceeding 5 mm in to warrant a diagnosis of carcinoma
dimension. (b) An image from the same tumor showing confluent
a b
c d
Fig. 6.9 (a, b) Endometrioid adenocarcinoma composed of clustered carcinoma. Notice the lack of high-grade atypia associated with
glands and solid growth. (c–f) Complex glands with intraglandular pap- clear-cell or high-grade serous carcinoma
illary features and necrosis. (g–h) solid growth in endometrioid adeno-
6 Ovarian Endometrioid and Clear-Cell Tumors 179
e f
g h
Fig. 6.9 (continued)
a b
Fig. 6.11 (a–c) “Endometrioid carcinoma resembling sex cord stromal tumor” or the “sertoliform variant” composed of tubules that are fused in
areas. This growth pattern resembles that seen in a Sertoli cell tumor
a b
c d
Fig. 6.12 (a) High-power detail of endometrioid adenocarcinoma. (b) PAX8 is positive in tumor cell nuclei. (c) Cytokeratin 7 is positive in tumor
cell cytoplasm. (d) Cytokeratin 20 is negative in tumor cell cytoplasm
synchronous ovarian carcinomas, and may involve altera- risk is much smaller than for uterine cancer [34]. In one study,
tions in TP53 [28, 29]. loss of MMR protein function is more common in endometrio-
Similar molecular genetic profiles have been described in sis-associated carcinomas (10%) when compared to high-grade
endometriosis, endometrial borderline/atypical proliferative serous carcinomas (1%) [33]. Specifically, MMR function was
tumor, and ovarian endometrioid carcinoma, which supports lost in endometrioid (2/29 cases), clear-cell (1/27 cases),
endometriosis as a precursor lesion [23–31]. Mutations of undifferentiated (1/8 cases), and mixed carcinomas with an
the tumor-suppressor gene AT-rich interactive domain- endometrioid, clear cell, and/or undifferentiated component
containing protein 1A (ARID1A) are common to endometri- (3/5 cases) [33]. As expected tumors with abnormal MMR sta-
oid and clear-cell ovarian carcinomas; Wiegand et al. tus demonstrated MSI using a polymerase chain reaction-
reported mutations in ARID1A in 30% of endometrioid car- based assay evaluating five mononucleotide repeat markers
cinomas and none of the studied 76 high-grade serous ovar- [34]. The following features, commonly seen in endometrial
ian carcinomas [32]. carcinomas with abnormal MMR expression, were not seen in
Finally, microsatellite instability (MSI), or DNA mismatch ovarian tumors: tumor-infiltrating lymphocytes, peritumoral
repair (MMR) protein loss, in these tumors has been described lymphocytes, or variable morphology. Among seven patients
[33, 34]. Ovarian cancer, particularly endometrioid adenocar- with tumors with abnormal MMR/MSI, five (71%) were alive
cinoma, is also associated with Lynch syndrome, although the without disease at the conclusion of the study [33].
182 J. Katzenberg and A. A. Roma
a b
Fig. 6.13 Frozen section slides of an ovarian tumor in a patient with a tures of colorectal carcinoma. (c) Garland pattern of invasion composed
history of colorectal carcinoma. (a) Low power showing two distinct of large cysts with cribriform architecture and dirty necrosis
patterns of metastasis. (b) Destructive invasive pattern with classic fea-
6 Ovarian Endometrioid and Clear-Cell Tumors 183
6.2.4 S
pecific Issues Associated with Ovarian 6.2.4.2 Endometrioid Carcinoma Arising
Endometrioid Tumors in Endometriosis
Sampson first described the association between endometri-
6.2.4.1 Endometrioid Carcinoma Involving Both osis and endometrial carcinoma in the 1920s and since then
Ovary and Endometrium most studies have shown increased risks for ovarian cancer
Coincidence of both primary ovarian and endometrial carci- in women with endometriosis [8, 41–44]. In a systematic
noma is a relatively common event occurring in about 5–10% review, the relative risk for clear-cell or endometrioid ovar-
of cases; both tumors are of endometrioid type in the majority ian carcinomas was approximately twofold for women with
of cases [11, 38]. Usually, the tumors are well differentiated a history of endometriosis [45]. According to a recent
and have similar morphologic features. Often it is quite dif- population-based study in Denmark, women with endome-
ficult to determine if the tumors represent synchronous pri- triosis had statistically significantly increased risk for ovar-
maries or metastasis based on clinical or morphologic ian (standardized incidence ratios (SIR) 1.55; 95% confidence
features alone. Accurate diagnosis as independent primaries interval (CI): 1.35–1.77) and endometrial cancer (SIR 2.13;
or metastases is necessary for appropriate staging and treat- 95% CI: 1.77–2.55) [46]. For ovarian cancer, statistically
ment as there is a substantial change in grade between the significantly increased risk were observed after 1–4 years
two scenarios. (SIR 1.51; 95% CI: 1.00–2.18) and 5–9 years of follow-up
Synchronous tumors are diagnosed if most of the fol- (SIR 1.78; 95% CI: 1.30–2.37) but not ≥10 years after the
lowing features are present: histological dissimilarity of first diagnosis of endometriosis [46]. Also, for ovarian can-
tumor type, absent or only superficial myoinvasion of the cer, increased risk were observed only for endometrioid (SIR
endometrial tumor, absence of vascular invasion, and/or 1.64; 95% CI: 1.09–2.37) and clear-cell (SIR 3.21; 95% CI:
a b
Fig. 6.14 (a, b) Endometrial endometrioid tumor with pushing borders, cystic ovarian endometrioid tumor with divergent papillary and glandu-
confined to the endometrium. (b) Note the presence of benign endome- lar morphology indicative of synchronous primary
trial glands at the base of the endometrial tumor. (c, d) A concurrent
184 J. Katzenberg and A. A. Roma
c d
Fig. 6.14 (continued)
a b
c d
Fig. 6.15 (a) Invasive endometrial adenocarcinoma with lymphovascular invasion. (b, c) Ovarian tumor with similar morphologic features, albeit
with areas of necrosis. (d) The tumor also invaded the fallopian tube indicative of metastatic spread
6 Ovarian Endometrioid and Clear-Cell Tumors 185
a b
c d
Fig. 6.16 (a, b) Well-differentiated polypoid endometrioid adenocarcinoma arising from areas of endometriosis. (c, d) High-grade carcinoma
arising from endometriosis; other foci showed classic features of clear-cell carcinoma
2.01–4.85) tumors [46]. An increased risk for endometrioid 6.3 Ovarian Tumors with a Sarcomatous
tumors was observed ≥10 years after a diagnosis of endome- Component
triosis. For clear-cell tumors, statistically significantly
increased risk was observed 5–9 years and ≥10 years after a 6.3.1 Adenosarcoma
diagnosis of endometriosis [46].
While there are few clinical signs that can be concerning Müllerian adenosarcoma of the ovary is rare [11]. It has simi-
for malignant transformation, there is no definitive way to lar morphologic features to uterine adenosarcoma. In one of
risk stratify patients [47]. the largest series, Eichhorn and collaborators reported 40
In their analysis of benign ovarian endometrioid tumors cases [48]. Mean age at presentation was 54 years (range
and well-differentiated endometrioid carcinomas, Bell and 30–84 years). All patients underwent oophorectomy and in
Kurman found frequent coexistence of endometriosis, benign 85% hysterectomy as well. Contralateral oophorectomy was
endometrioid neoplasms such as endometrioid adenofibroma performed in 28% of patients; all but one patient had unilat-
or endometrioid tumor of borderline or low malignant poten- eral disease [48]. Grossly, most tumors were predominantly
tial, and well-differentiated endometrioid carcinoma solid with numerous small cysts. Microscopically, tumors
(Fig. 6.16a–d) [17]. These findings, and other similar molecu- were composed of benign-appearing, and occasionally
lar abnormalities seen in endometrioid tumors and adjacent malignant-appearing, glands set in sarcomatous stroma; sar-
endometriosis, support a process of endometriosis-carcinoma comatous overgrowth was seen in 25% of cases [48]. Most
progression. glands displayed endometrioid epithelium, while squamous,
186 J. Katzenberg and A. A. Roma
a b
Fig. 6.17 (a) Adenosarcoma with a classic phyllodes pattern of growth and dense, cellular stroma. (b) High power showing dense stroma and
benign epithelium; note the presence of stromal mitosis
a b
tubal, hobnail, clear-cell, or even mucinous-type epithelium Abnormal uterine bleeding was not as common as in
was also seen [48]. Cystic dilatation of the glands was com- endometrial tumors, and ovarian adenosarcomas more likely
mon and accounted for some of the cysts seen on gross presented with a pelvic mass [48]. Ovarian adenosarcoma
examination; leaflike stromal projections were common demonstrated a poorer prognosis when compared to the
(Fig. 6.17a, b). In most cases, the stromal component resem- endometrial subtype [48].
bled endometrial stroma, but could also be fibromatous or an
admixture of both. Periglandular cuffing was identified in all
tumors (Fig. 6.18a, b) [48]. 6.3.2 Endometrioid Stromal Sarcoma
Twenty of the 32 patients with available follow-up
recurred between 3 months and 6.6 years after diagnosis Endometrial stromal sarcoma is a mesenchymal, spindle-cell
(mean 2.6 years). Three patients developed lung metastases, neoplasm that predominantly arises from the endometrium
and two died of disease [48]. [49]. Extrauterine tumors are quite rare [49, 50]. Intrauterine
Compared with data from endometrial adenosarcoma, and extrauterine tumors share similar morphologic features,
ovarian tumors appear to present at younger age, with about including bland, uniform spindle cells resembling
half of patients presenting at or before 50 years of age. proliferative-type endometrial stroma with admixed spiral,
6 Ovarian Endometrioid and Clear-Cell Tumors 187
a b
Fig. 6.19 (a, b) Endometrial stromal sarcoma composed of bland, uniform, spindled cells resembling proliferative-type endometrium and
admixed spiral, arteriole-like vessels
a b
c d
e f
Fig. 6.20 (a) Biphasic neoplasm with malignant glands and spindled stroma. (b, c) Rhabdomyosarcomatous component of carcinosarcoma; note
the prominent rhabdomyoblastic differentiation. (d–f) Chondrosarcomatous component in carcinosarcoma
6 Ovarian Endometrioid and Clear-Cell Tumors 189
6.5 Ovarian Clear-Cell Tumors tal finding, a palpable pelvic mass, or with nonspecific symp-
toms such as vaginal bleeding [10, 62, 65–67]. While the
6.5.1 Overview pathogenic relationship between adenofibromas and clear-
cell carcinoma is unclear, there is some histomorphologic and
A model of ovarian carcinogenesis has been proposed that molecular evidence that suggests an adenofibroma-carcinoma
divides these diverse tumors into two groups designated, type sequence [65, 66, 68–70]. Studies have also suggested that
I and type II [60, 61]. The prototypic type I tumor is low- there is an alternate pathogenic mechanism in endometriosis-
grade serous carcinoma; type I tumors are generally confined associated clear-cell carcinoma compared to those with an
to the ovary at diagnosis (stage I) and develop in a stepwise adenofibromatous background [68–71]. Some adenofibromas
fashion from a well-established precursor lesion, such as a are associated with endometriosis, so it is difficult to defi-
borderline tumor [60, 61]. In addition to low-grade serous nitely claim that there are two different tumorigenic path-
carcinomas, endometrioid and mucinous carcinomas are con- ways; nevertheless, no study to date has shown malignant
sidered type I tumors. Molecular genetic alterations that are behavior in clear-cell adenofibromas which are considered
commonly encountered in type I tumors are KRAS, BRAF, benign [11].
PTEN, and/or CTNNB1 and/or microsatellite instability. Type
II tumors, on the other hand, are highly aggressive, have a 6.6.1.2 Pathological Findings
high frequency of TP53 mutations, and do not arise from pre- Clear-cell cystadenomas/adenofibromas are generally
cursor lesions. In other words, they arise de novo. The patho- smooth, lobulated, solid masses raging in size from 3 to
genesis of ovarian clear-cell carcinoma, however, does not fit 16 cm; the cut surface is gray-white with small, closely
neatly into one of these two categories [60–62]. In terms of packed cysts imparting a spongelike appearance [11, 65–67].
the precursor lesions, histologic and epidemiologic analyses Microscopically, they consist of orderly, widely spaced
have shown a close relationship between endometriosis and glands of various sizes lined by 1–2 layers of flat, cuboidal to
ovarian clear-cell carcinoma [63, 64]. While the current 2014 columnar epithelium. This epithelium is typically lined with
WHO subdivides all ovarian clear-cell tumors into benign, clear cells with or without cytoplasmic hobnailing and rarely
borderline, and malignant types, the criteria for distinguish- eosinophilic cells embedded in a fibromatous stroma
ing these categories are not well established, and their bio- (Fig. 6.21a–c). The stroma is composed of bundles of
logic behavior remains uncertain [11, 65–67]. spindle-shaped cells with plump elongated nuclei, resem-
bling ovarian stroma. There is little to no nuclear atypia, and
mitotic figures are rare. Because of the rarity of these tumors,
6.6 Benign Clear-Cell Tumors it is recommended that they be sampled extensively, or sub-
mitted entirely, to exclude areas of atypia or invasive carci-
6.6.1 Clear-Cell Cystadenoma/Adenofibroma noma [65].
a b
Fig. 6.21 (a–c) Different magnifications of a clear-cell adenofibroma, composed of a tubular epithelial proliferation of cuboidal epithelium with
clear cytoplasm devoid of cytologic atypia set in a dense stroma
190 J. Katzenberg and A. A. Roma
a b
Fig. 6.22 (a–c) Clear-cell borderline tumor composed of a tubular proliferation of cuboidal epithelium with clear cytoplasm and larger cells with
prominent nucleoli set in a fibromatous stroma. Notably absent is the presence of destructive invasion
a b
Fig. 6.23 (a, b) A clear-cell proliferation with prominent atypia that was seen adjacent to classic areas of clear-cell carcinoma that raises the pos-
sibility of a borderline tumor transitioning into carcinoma
192 J. Katzenberg and A. A. Roma
a b
c d
Fig. 6.24 (a–d) Yolk sac tumor with an endodermal sinus and microcystic pattern of growth resembling Müllerian clear-cell carcinoma
Histologically, yolk sac tumors are often heterogeneous, and diagnosis is 53–55 years (range 19–82). The presentation
extensive sampling will often reveal more classic features, is similar to borderline adenofibroma with abdominal
such as Schiller-Duval bodies or hyaline globules. They are swelling, a pelvic mass, symptoms of endometriosis,
positive for alpha-fetoprotein, glypican-3, and SALL-4, and symptoms related to paraneoplastic syndromes, or non-
CK7 and EMA are negative [11, 76]. specific symptoms [10, 11, 65, 66, 71, 79, 80]. Ovarian
clear-cell carcinomas are associated with several paraneo-
6.7.1.5 Management and Outcomes plastic syndromes including hypercalcemia, venous
Since these tumors are rare, their behavior is not well under- thromboembolism, subacute cerebellar degeneration, and
stood. They are thought to carry a good prognosis, unless bilateral diffuse uveal melanocytic proliferation [79–81].
associated with areas of malignant transformation [10, 11]. A study by Matsuura et al. demonstrated that a thrombo-
embolic event was noted in 27.3% of patients with clear-
cell carcinoma, compared to 6.8% of patients with other
6.8 Ovarian Clear-Cell Adenocarcinoma epithelial ovarian cancers [81]. As previously mentioned,
50–70% of these tumors are associated with endometrio-
6.8.1 Clinical Features sis, but a small percentage can arise from borderline/atyp-
ical proliferative clear-cell tumors [68–71]. Additionally,
Ovarian clear-cell carcinoma comprises 5–25% of all they have also been associated with Lynch syndrome
ovarian carcinomas [10, 11, 77, 78]. The average age at [82–85].
6 Ovarian Endometrioid and Clear-Cell Tumors 193
a b
Fig. 6.26 (a–d) Low power of clear-cell carcinoma with papillary (a), tubulocystic (b, c), and solid patterns (d)
194 J. Katzenberg and A. A. Roma
c d
Fig. 6.26 (continued)
a b
c d
Fig. 6.27 (a–f) High-power magnification showing clear-cell morphology, nuclear atypia and pleomorphism, hobnail features, and eosinophilic/
oncocytic change in clear-cell carcinoma
6 Ovarian Endometrioid and Clear-Cell Tumors 195
e f
Fig. 6.27 (continued)
By immunohistochemistry, clear-cell carcinomas are pos- PTEN, another tumor-suppressor gene located at chromo-
itive for CK7, EMA, PAX-8, napsin A, and HNF-1β; estrogen some 10q23.3, has been implicated in the sequential progres-
and progesterone receptors are usually negative but can be sion from endometrial cysts to clear-cell carcinoma [89].
focally positive. WT1 is negative and most tumors are P53 PTEN encodes a phosphatase that dephosphorylates
wild type [11, 76, 86, 87]. phosphatidylinositol-3,4,5-triphosphate (PIP3), a phospho-
lipid that promotes cell growth and survival; its inactivation
leads to phosphorylation and activation of PIP3. In a study
6.8.3 Genetic Profile by Sato et al., somatic mutations in PTEN were seen in
approximately 8% of ovarian clear-cell carcinomas [89].
Ovarian clear-cell carcinomas share features of both type Studies have shown that one-third of ovarian clear-cell
I and type II tumors. Like type I tumors, most are diag- carcinomas have PIK3CA-activating mutations [72, 88, 90].
nosed at a low stage and are associated with precursor While this gene is mutated in other tumor types, ovarian
lesions including endometriosis, clear-cell adenofibro- clear-cell carcinoma has the highest frequency of PIK3CA
mas, and clear-cell borderline tumors [60–64, 66, 68–71]. mutations. From a therapeutic perspective, this finding is sig-
Studies by Fukunaga et al. and Ogawa et al. have shown nificant because new PI3K-targeting drugs are being devel-
that clear-cell carcinoma arises from endometriosis in oped. Most PIK3CA mutations occur in ARID1A-deficient
50–70% of cases [63, 64]. Like type II tumors, they tend carcinomas, suggesting that they might work together to pro-
to be high grade and behave aggressively when diagnosed mote tumorigenesis [72, 88, 90].
at a higher stage [62]. The carcinogenesis of ovarian Lynch syndrome is associated with ovarian clear-cell car-
clear-cell carcinoma has been hypothesized to be com- cinoma [82–84, 91]. In one of the largest studies, abnormal
prised of two different pathways: the adenofibroma- MMR expression by immunohistochemistry was identified in
carcinoma sequence and an endometriosis-carcinoma 6% of tumors and included deficiencies in MSH2/MSH6 (3),
sequence [60–62]. MLH1/PMS2 (1), MSH6 (1), and PMS2 (1) [84]. Patients
ARID1A, a tumor-suppressor gene, is commonly mutated were younger, with a mean age of 40 (range 31–48), which
in ovarian clear-cell carcinomas (46–57%) resulting in loss contrasted with a mean of 53.2 years (range 28–82) for the
of its protein product, BAF250a [72]. Studies have shown overall cohort of 109 tumors [84]. Tumors with diffuse intra-
that the BAF250a loss is seen more often in the endometriosis- tumoral stromal inflammation and peritumoral lymphocytes
carcinoma sequence. Additionally, mutations in ARID1A are were more frequently associated with MMR loss [84].
also seen in adjacent endometriotic lesions and in adjacent Unlike high-grade serous carcinoma, ovarian clear-cell
proliferative/atypical clear-cell adenofibromas, supporting a carcinomas have a lower frequency of BRCA1 and BRCA2
causal relationship [72, 88]. mutation [92, 93].
196 J. Katzenberg and A. A. Roma
a b
Fig. 6.28 (a, b) Cystic high-grade serous carcinoma with micropapillary structures and significant cytologic atypia devoid of clear-cell change
6 Ovarian Endometrioid and Clear-Cell Tumors 197
marker has low sensitivity and specificity [101]. Glypican-3 is carcinomas (Fig. 6.30a–d). Metastatic clear-cell renal carcino-
expressed in the ample majority of yolk sac tumors (94–100%) mas are typically positive for CA-IX, CD10, and renal cell
but less frequently in clear-cell carcinomas (17%) [101]. CK7 carcinoma antigen, whereas ovarian clear-cell tumors express
expression in yolk sac tumors is variable (0–82%) but has Napsin A, CK7, and p504S; PAX8 and HNF1β do not help
been reported universally in clear-cell carcinomas [101, 102] distinguish between these tumors [103].
while EMA has been reported negative in yolk sac tumors and
widely expressed in carcinomas [102].
Lastly, metastatic tumors with clear cells can mimic pri- 6.8.5 Management and Outcomes
mary Müllerian carcinomas. In a recent study, Fadare and col-
laborators reviewed a large series of clear-cell renal carcinomas Ovarian clear-cell carcinomas are more likely to be detected
metastatic to the gynecological tract [103]. Although clear-cell at an early stage than other high-grade ovarian cancers, and
renal carcinomas and primary Müllerian clear-cell carcinomas if confined to the ovary have a good prognosis [104]. When
displayed extensive morphologic overlap, an increased mitotic presenting as advanced stage, the prognosis is very poor, and
index and prominent alveolar pattern were more frequently the tumors are usually resistant to standard treatment [104].
seen in metastatic tumors. Diffuse hobnail cells, hyaline glob- Cytoreductive surgery is recommended for patients with
ules, predominant tubulocystic pattern, or presence of a papil- advanced-stage (II or above) disease; hence, development of
lary growth pattern were more frequently seen in ovarian novel treatments based on molecular characteristics is
a b
c d
Fig. 6.29 (a, b) Dysgerminoma composed of large, clear cells, grow- (d) Placental alkaline phosphatase immunostain showing diffuse cyto-
ing in sheets, nests, and cords, separated by thick fibrous bands. (c) plasmic expression
Large cells with prominent nucleoli, characteristic of dysgerminoma.
198 J. Katzenberg and A. A. Roma
a b
c d
Fig. 6.30 (a–d) Clear-cell renal cell carcinoma composed of cystic and solid patterns composed of clear cells devoid of significant atypia
14. Roth LM, Czernobilsky B, Langley FA. Ovarian endometrioid 32. Wiegand KC, Shah SP, Al-Agha OM, et al. ARID1A mutations
adenofibromatous and cystadenofibromatous tumors: benign, pro- in endometriosis-associated ovarian carcinomas. N Engl J Med.
liferating, and malignant. Cancer. 1981;48:1838–45. 2010;363:1532–43.
15. McCluggage WG, et al. Benign, borderline, and malignant endo- 33. Lu FI, Gilks CB, Mulligan AM, Ryan P, Allo G, Sy K, Shaw PA,
metrioid neoplasia arising in endometriosis in association with Pollett A, Clarke BA. Prevalence of loss of expression of DNA
tamoxifen therapy. Int J Gynecol Pathol. 2000;19(3):276–9. mismatch repair proteins in primary epithelial ovarian tumors. Int
16. Prat J. Pathology of borderline and invasive cancers. Best Pract J Gynecol Pathol. 2012;31(6):524–31.
Res Clin Obstet Gynaecol. 2017;41:15–30. 34. Aysal A, Karnezis A, Medhi I, Grenert JP, Zaloudek CJ, Rabban
17. Bell KA, Kurman RJ. A clinicopathologic analysis of atypi- JT. Ovarian endometrioid adenocarcinoma: incidence and clinical
cal proliferative (borderline) tumors and well-differentiated significance of the morphologic and immunohistochemical mark-
endometrioid adenocarcinomas of the ovary. Am J Surg Pathol. ers of mismatch repair protein defects and tumor microsatellite
2000;24(11):1465–79. instability. Am J Surg Pathol. 2012;36(2):163–72.
18. Uzan C, Berretta R, Rolla M, Gouy S, Fauvet R, Darai E, Duvillard 35. Lewis MR, Deavers MT, Silva EG, Malpica A. Ovarian involve-
P, Morice P. Management and prognosis of endometrioid border- ment by metastatic colorectal adenocarcinoma: still a diagnostic
line tumors of the ovary. Surg Oncol. 2012;21(3):178–84. challenge. Am J Surg Pathol. 2006;30(2):177–84.
19. Prat J. Ovarian carcinomas: five distinct diseases with differ- 36. Judson K, McCormick C, Vang R, Yemelyanova AV, Wu LS,
ent origins, genetic alterations, and clinicopathological features. Bristow RE, Ronnett BM. Women with undiagnosed colorectal
Virchows Arch. 2012;460(3):237–49. adenocarcinomas presenting with ovarian metastases: clinico-
20. Chen S, Leitao MM, Tornos C, Soslow RA. Invasion patterns in pathologic features and comparison with women having known
stage I endometrioid and mucinous ovarian carcinomas: a clini- colorectal adenocarcinomas and ovarian involvement. Int J
copathologic analysis emphasizing favorable outcomes in carci- Gynecol Pathol. 2008;27(2):182–90.
nomas without destructive stromal invasion and the occasional 37. Bouchard-Fortier G, Panzarella T, Rosen B, Chapman W, Gien
malignant course of carcinomas with limited destructive stromal LT. Endometrioid carcinoma of the ovary: outcomes compared to
invasion. Mod Pathol. 2005;18(7):903–11. serous carcinoma after 10 years of follow-up. J Obstet Gynaecol
21. Berezowski K, Stastny JF, Kornstein MJ. Cytokeratins 7 and 20 Can. 2017;39(1):34–41.
and carcinoembryonic antigen in ovarian and colonic carcinoma. 38. Singh N. Synchronous tumours of the female genital tract.
Mod Pathol. 1996;9:426–9. Histopathology. 2010;56(3):277–85.
22. Nonaka D, Chiriboga L, Soslow RA. Expression of pax8 as a use- 39. Irving JA, Catasús L, Gallardo A, Bussaglia E, Romero M, Matias
ful marker in distinguishing ovarian carcinomas from mammary Guiu X, Prat J. Synchronous endometrioid carcinomas of the uter-
carcinomas. Am J Surg Pathol. 2008;32(10):1566–71. ine corpus and ovary: alterations in the beta-catenin (CTNNB1)
23. Prefumo F, Venturini PL, Fulcheri E. Analysis of p53 and c-erb-2 pathway are associated with independent primary tumors and
expression in ovarian endometrioid carcinomas arising in endo- favorable prognosis. Hum Pathol. 2005;36:605–19.
metriosis. Int J Gynecol Pathol. 2002;22:83–8. 40. Valtcheva N, Lang FM, Noske A, Samartzis EP, Schmidt AM,
24. Kolasa IK, Rembiszewska A, Janiec-Jankowska A, Dansonka- Bellini E, Fink D, Moch H, Rechsteiner M, Dedes KJ, Wild
Mieszkowska A, Lewandowska AM, Konopka B, Kupryjanczyk PJ. Tracking the origin of simultaneous endometrial and ovar-
J. Pten mutation, expression and loh at its locus in ovarian carcino- ian cancer by next-generation sequencing—a case report. BMC
mas. Relation to tp53, k-ras and brca1 mutations. Gynecol Oncol. Cancer. 2017;17(1):66.
2006;103:692–7. 41. Pearce CL, Templeman C, Rossing MA, Lee A, Near AM, Webb
25. Palacios J, Gamallo C. Mutations in the beta-catenin gene PM, Nagle CM, Doherty JA, Cushing-Haugen KL, Wicklund
(ctnnb1) in endometrioid ovarian carcinomas. Cancer Res. KG, Chang-Claude J, Hein R, Lurie G, Wilkens LR, Carney
1998;58:1344–7. ME, Goodman MT, Moysich K, Kjaer SK, Hogdall E, Jensen A,
26. Catasús L, Bussaglia E, Rodrguez I, Gallardo A, Pons C, Irving Goode EL, Fridley BL, Larson MC, Schildkraut JM, Palmieri RT,
JA, Prat J. Molecular genetic alterations in endometrioid carci- Cramer DW, Terry KL, Vitonis AF, Titus LJ, Ziogas A, Brewster
nomas of the ovary: similar frequency of beta-catenin abnor- W, Anton-Culver H, Gentry-Maharaj A, Ramus SJ, Anderson
malities but lower rate of microsatellite instability and PTEN AR, Brueggmann D, Fasching PA, Gayther SA, Huntsman DG,
alterations than in uterine endometrioid carcinomas. Hum Pathol. Menon U, Ness RB, Pike MC, Risch H, Wu AH, Berchuck A,
2004;35(11):1360–8. Ovarian Cancer Association Consortium. Association between
27. Cho KR, Shih IE. Ovarian cancer. Annu Rev Pathol. endometriosis and risk of histological subtypes of ovarian can-
2009;4:287–313. cer: a pooled analysis of case-control studies. Lancet Oncol.
28. Worley MJ, Welch WR, Berkowitz RS, Ng SW. Endometriosis- 2012;13:385–94.
associated ovarian cancer: a review of pathogenesis. Int J Mol Sci. 42. Heidemann LN, Hartwell D, Heidemann CH, Jochumsen
2013;14:5367–79. KM. The relation between endometriosis and ovarian cancer—a
29. Ali-Fehmi R, Khalifeh I, Bandyopadhyay S, et al. Patterns of review. Acta Obstet Gynecol Scand. 2014;93:20–31.
loss of heterozygosity at 10q23.3 and microsatellite instability 43. Kvaskoff M, Mu F, Terry KL, Harris HR, Poole EM, Farland L,
in endometriosis, atypical endometriosis, and ovarian carcinoma et al. Endometriosis: a high-risk population for major chronic dis-
arising in association with endometriosis. Int J Gynecol Pathol. eases? Hum Reprod Update. 2015;21:500–16.
2006;25:223–9. 44. Kim HS, Kim TH, Chung HH, Song YS. Risk and prognosis of
30. Sainz de la Cuesta R, Izquierdo M, Canamero M, et al. Increased ovarian cancer in women with endometriosis: a meta-analysis. Br
prevalence of p53 overexpression from typical endometriosis to J Cancer. 2014;110:1878–90.
atypical endometriosis and ovarian cancer associated with endo- 45. Kvaskoff M, Mu F, Terry KL, et al. Endometriosis: a high-risk
metriosis. Eur J Obstet Gynecol Reprod Biol. 2004;15:87–93. population for major chronic diseases? Hum Reprod Update.
31. Willner J, Wurz K, Allison KH, Galic V, Garcia RL, Goff BA, 2015;21:500–16.
Swisher EM. Alternate molecular genetic pathways in ovar- 46. Mogensen JB, Kjær SK, Mellemkjær L, Jensen A. Endometriosis
ian carcinomas of common histological types. Hum Pathol. and risks for ovarian, endometrial and breast cancers: a nation-
2007;38:607–13. wide cohort study. Gynecol Oncol. 2016;143(1):87–92.
200 J. Katzenberg and A. A. Roma
47. Wei JJ, William J, Bulun S. Endometriosis and ovarian cancer: 66. Roth LM, Langley FA, Fox H, et al. Ovarian clear cell ade-
a review of clinical, pathologic, and molecular aspects. Int J nofibromatous tumors: benign, of low malignant potential,
Gynecol Pathol. 2011;30(6):553–68. and associated with invasive clear cell carcinoma. Cancer.
48. Eichhorn JH, Young RH, Clement PB, Scully RE. Mesodermal 1984;53:1156–63.
(müllerian) adenosarcoma of the ovary: a clinicopathologic analy- 67. Kennedy AW, Biscotti CV, Hart WR, et al. Ovarian clear cell ade-
sis of 40 cases and a review of the literature. Am J Surg Pathol. nocarcinoma. Gynecol Oncol. 1989;32:342–9.
2002;26(10):1243–58. 68. Yamamoto S, Tsuda H, Takano M, Hase K, Tamai S, Matsubara
49. Amador-Ortiz C, Roma AA, Huettner PC, Becker N, Pfeifer O. Clear-cell adenofibroma can be a clonal precursor for clear-cell
JD. JAZF1 and JJAZ1 gene fusion in primary extrauterine endo- adenocarcinoma of the ovary: a possible alternative ovarian clear-
metrial stromal sarcoma. Hum Pathol. 2011;42(7):939–46. cell carcinogenic pathway. J Pathol. 2008;216:103–10.
50. Oliva E, Egger JF, Young RH. Primary endometrioid stromal 69. Yamamoto S, Tsuda H, Suzuki K, et al. An allelotype analysis
sarcoma of the ovary: a clinicopathologic study of 27 cases with indicating the presence of two distinct ovarian clear-cell car-
morphologic and behavioral features similar to those of uter- cinogenic pathways: endometriosis-associated pathway vs.
ine low-grade endometrial stromal sarcoma. Am J Surg Pathol. clear-cell adenofibroma- associated pathway. Virchows Arch.
2014;38(3):305–15. 2009;455:261–70.
51. Young RH, Prat J, Scully RE. Endometrioid stromal sarcomas 70. Yamamoto S, Tsuda H, Yoshikawa T, et al. Clear cell adeno-
of the ovary. A clinicopathologic analysis of 23 cases. Cancer. carcinoma associated with clear cell adenofibromatous com-
1984;53:1143–55. ponents: a subgroup of ovarian clear cell adenocarcinoma with
52. Back JA, Choi MG, Ju UC, Kang WD, Kim SM. A case of distinct clinicopathological characteristics. Am J Surg Pathol.
advanced-stage endometrial stromal sarcoma of the ovary arising 2007;31:999–1006.
from endometriosis. Obstet Gynecol Sci. 2016;59(4):323–7. 71. Veras E, Mao TL, Ayhan A, Ueda S, Lai H, Hayran M, Shih IM,
53. Kilzieh R, Rakislova N, Torné A, Salvador R, Nadal A, Ordi J, Kurman RJ. Cystic and adenofibromatous clear cell carcinomas of
Saco A. Endometrial stromal sarcoma arising in colorectal endo- the ovary: distinctive tumors that differ in their pathogenesis and
metriosis. Int J Gynecol Pathol. 2017;36(5):433–7. behavior: a clinicopathologic analysis of 122 cases. Am J Surg
54. Berton-Rigaud D, Devouassoux-Shisheboran M, Ledermann Pathol. 2009;33:844–53.
JA, Leitao MM, Powell MA, Poveda A, Beale P, Glasspool 72. Wiegand KC, Shah SP, Al-Agha OM, Zhao Y, Tse K, Zeng T, et al.
RM, Creutzberg CL, Harter P, Kim JW, Reed NS, Ray-Coquard ARID1A mutations in endometriosis-associated ovarian carcino-
I. Gynecologic Cancer InterGroup (GCIG) consensus review mas. N Engl J Med. 2010;363:1532–43.
for uterine and ovarian carcinosarcoma. Int J Gynecol Cancer. 73. Cho I, Lim S-C. Clear cell adenocarcinoma arising from adenofi-
2014;24(9 Suppl 3):S55–60. broma in a patient with endometriosis of the ovary. J Pathol Transl
55. Roma AA, Malpica A, Deavers MT. Malignant melanoma arising Med. 2016;50(2):155–9.
in an ovarian carcinosarcoma: case report and review of the litera- 74. Zirker TA, Silva EG, Morris M, et al. Immunohistochemical dif-
ture. Int J Gynecol Pathol. 2011;30(2):158–62. ferentiation of clear-cell carcinoma of the female genital tract and
56. De Jong RA, Nijman HW, Wilbrandi TF, et al. Molecular mark- endodermal sinus tumor with the use of alphafetoprotein and Leu-
ers and clinical behavior of uterine carcinosarcomas: focus on the M1. Am J Clin Pathol. 1989;91:511–4.
epithelial component. Mol Pathol. 2011;24:1368–79. 75. Klemi PJ, Meurman L, Gronroos M, et al. Clear cell (mesoneph-
57. Growdon W, Roussel B, Scialabba V, et al. Tissue-specific signa- roid) tumors of the ovary with characteristics resembling endoder-
tures of activating PI3CA and RAS mutations in carcinosarcomas mal sinus tumor. Int J Gynecol Pathol. 1982;1:95–100.
of gynecologic origin. Gynecol Oncol. 2011;121:212–7. 76. Ramalingam P, Malpica A, Silva EG, et al. The use of cytokera-
58. Garg G, Shah J, Kumar S, et al. Ovarian and uterine carcinosarco- tin 7 and EMA in differentiating ovarian yolk sac tumors from
mas: a comparative analysis of prognostic variables and survival endometrioid and clear cell carcinomas. Am J Surg Pathol.
outcomes. Int J Gynecol Cancer. 2010;20:888–94. 2004;28:1499–505.
59. Kanis MJ, Kolev V, Getrajdman J, Zakashansky K, Cohen C, 77. Anglesio MS, Carey MS, Kobel M, Mackay H, Huntsman
Rahaman J. Carcinosarcoma of the ovary: a single institution DG. Clear cell carcinoma of the ovary: a report from the first
experience and review of the literature. Eur J Gynaecol Oncol. ovarian clear cell symposium, June 24th, 2010. Gynecol Oncol.
2016;37(1):75–9. 2011;121:407–15.
60. Kurman RJ, Craig JM. Endometrioid and clear cell carcinoma of 78. Bennett JA, Dong F, Young RH, Oliva E. Clear cell carcinoma
the ovary. Cancer. 1972;29:1653–64. of the ovary: evaluation of prognostic parameters based on
61. Kurman RJ, Shih I-M. Pathogenesis of ovarian cancer. Lessons a clinicopathological analysis of 100 cases. Histopathology.
from morphology and molecular biology and their clinical impli- 2015;66(6):808–15.
cations. Int J Gynecol Pathol. 2008;27(2):151–60. 79. Goldstein BH, Birk CL, Van Houten M, Veve R, Brown JV,
62. Zhao C, LS-F W, Barner R. Pathogenesis of ovarian clear cell Rettenmaier MA, Micha JP. Ovarian cancer and late onset
Adenofibroma, atypical proliferative (borderline) tumor, and car- paraneoplastic cerebellar degeneration. Arch Gynecol Obstet.
cinoma: clinicopathologic features of tumors with endometriosis 2009;280:99–101.
or adenofibromatous components support two related pathways of 80. Lewin S, Dezube D, Guddati A, Mittal K, Muggia F, Klein
tumor development. J Cancer. 2011;2:94–106. P. Paraneoplastic hypercalcemia in clear cell ovarian adenocarci-
63. Ogawa S, Kaku T, Amada S, Kobayashi H, Hirakawa T, Ariyoshi noma. Ecancermedicalscience. 2012;6:271.
K, et al. Ovarian endometriosis associated with ovarian carci- 81. Matsuura Y, Robertson G, Marsden DE, Kim S-N, Gebski
noma: a clinicopathological and immunohistochemical study. V, Hacker NF. Thromboembolic complications in patients
Gynecol Oncol. 2000;77:298–304. with clear cell carcinoma of the ovary. Gynecol Oncol.
64. Fukunaga M, Nomura K, Ishikawa E, Ushigome S. Ovarian atypi- 2007;104(2):406–10.
cal endometriosis: its close association with malignant epithelial 82. Ueda H, Watanabe Y, Nakai H, Hemmi H, Koi M, Hoshiai
tumours. Histopathology. 1997;30:249–55. H. Microsatellite status and immunohistochemical fea-
65. Bell DA, Scully RE. Benign and borderline clear cell adenofibro- tures of ovarian clear-cell carcinoma. Anticancer Res.
mas of the ovary. Cancer. 1985;56:2922–31. 2005;25(4):2785–8.
6 Ovarian Endometrioid and Clear-Cell Tumors 201
83. Vierkoetter KR, Ayabe AR, VanDrunen M, Ahn HJ, Shimizu DM, 95. Sangoi AR, Soslow RA, Teng NN, Longacre TA. Ovarian clear
Terada KY. Lynch Syndrome in patients with clear cell and endo- cell carcinoma with papillary features: a potential mimic of
metrioid cancers of the ovary. Gynecol Oncol. 2014;135(1):81–4. serous tumor of low malignant potential. Am J Surg Pathol.
84. Bennett JA, Morales-Oyarvide V, Campbell S, Longacre TA, Oliva 2008;32(2):269–74.
E. Mismatch repair protein expression in clear cell carcinoma of 96. Acs G, Pasha T, Zhang PJ. WT1 is differentially expressed in
the ovary: incidence and morphologic associations in 109 cases. serous, endometrioid, clear cell, and mucinous carcinomas of the
Am J Surg Pathol. 2016;40(5):656–63. peritoneum, fallopian tube, ovary, and endometrium. Int J Gynecol
85. Willis BC, Sloan EA, Atkins KA, Stoler MH, Mills AM. Mismatch Pathol. 2004;23:110–8.
repair status and PD-L1 expression in clear cell carcinomas of the 97. Silva EG, Young RH. Endometrioid neoplasms with clear cells: a
ovary and endometrium. Mod Pathol. 2017;30(11):1622–32. report of 21 cases in which the alteration is not of typical secretory
86. Yamashita Y, Nagasaka T, Naiki-Ito A, Sato S, Suzuki S, Toyokuni type. Am J Surg Pathol. 2007;31(8):1203–8.
S, Ito M, Takahashi S. Napsin A is a specific marker for ovarian 98. Offman SL, Longacre TA. Clear cell carcinoma of the female gen-
clear cell adenocarcinoma. Mod Pathol. 2015;28(1):111–7. ital tract (not everything is as clear as it seems). Adv Anat Pathol.
87. DeLair D, Oliva E, Köbel M, Macias A, Gilks CB, Soslow 2012;19(5):296–312.
RA. Morphologic spectrum of immunohistochemically character- 99. Ulbright TM, Young RH. Seminoma with tubular, microcystic,
ized clear cell carcinoma of the ovary: a study of 155 cases. Am J and related patterns: a study of 28 cases of unusual morphologic
Surg Pathol. 2011;35(1):36–44. variants that often cause confusion with yolk sac tumor. Am J Surg
88. Nishikimi K, Kiyokawa T, Tate S, Iwamoto M, Shozu M. ARID1A Pathol. 2005;29(4):500–5.
expression in ovarian clear cell carcinoma with an adenofibroma- 100. Cao D, Guo S, Allan RW, et al. SALL4 is a novel sensitive and
tous component. Histopathology. 2015;67(6):866–71. specific marker of ovarian primitive germ cell tumors and is par-
89. Sato N, Tsunoda H, Nishida M, et al. Loss of heterozygosity on ticularly useful in distinguishing yolk sac tumor from clear cell
10q23.3 and mutation of the tumor suppressor gene PTEN in carcinoma. Am J Surg Pathol. 2009;33:894–904.
benign endometrial cyst of the ovary: possible sequence progres- 101. Esheba GE, Pate LL, Longacre TA. Oncofetal protein glypican-3
sion from benign endometrial cyst to endometrioid carcinoma and distinguishes yolk sac tumor from clear cell carcinoma of the
clear cell carcinoma of the ovary. Cancer Res. 2000;60:7052–6. ovary. Am J Surg Pathol. 2008;32:600–7.
90. Kuo KT, Mao TL, Jones S, et al. Frequent activating muta- 102. Ramalingam P, Malpica A, Silva EG, Gershenson DM, Liu JL,
tions of PIK3CA in ovarian clear cell carcinoma. Am J Pathol. Deavers MT. The use of cytokeratin 7 and EMA in differentiating
2009;174(5):1597–601. ovarian yolk sac tumors from endometrioid and clear cell carcino-
91. Cai KQ, Albarracin C, Rosen D, Zhong R, Zheng W, Luthra R, mas. Am J Surg Pathol. 2004;28:1499–505.
Broaddus R, Liu J. Microsatellite instability and alteration of the 103. Fadare O, Desouki MM, Gwin K, Hanley KZ, Jarboe EA, Liang
expression of hMLH1 and hMSH2 in ovarian clear cell carci- SX, Quick CM, Rawish KR, Roma AA, Zheng W, Hecht JL,
noma. Hum Pathol. 2004;35:552–9. Parkash V, Osunkoya AO. Clear cell renal cell carcinoma meta-
92. Okamoto A, Glasspool RM, Mabuchi S, Matsumura N, Nomura static to the gynecologic tract: a clinicopathologic analysis of 17
H, Itamochi H, et al. Gynecologic Cancer InterGroup (GCIG) con- cases. Int J Gynecol Pathol. 2018;37(6):525–35.
sensus review for clear cell carcinoma of the ovary. Int J Gynecol 104. Okamoto A, Glasspool RM, Mabuchi S, Matsumura N, Nomura
Cancer. 2014;24(Suppl 3):S20–5. H, Itamochi H, Takano M, Takano T, Susumu N, Aoki D, Konishi
93. Alsop K, Fereday S, Meldrum C, deFazio A, Emmanuel C, I, Covens A, Ledermann J, Mezzanzanica D, Steer C, Millan
George J, et al. BRCA mutation frequency and patterns of treat- D, McNeish IA, Pfisterer J, Kang S, Gladieff L, Bryce J, Oza
ment response in BRCA mutation-positive women with ovarian A. Gynecologic Cancer InterGroup (GCIG) consensus review for
cancer: a report from the Australian Ovarian Cancer Study Group. clear cell carcinoma of the ovary. Int J Gynecol Cancer. 2014;24(9
J Clin Oncol. 2012;30:2654–63. Suppl 3):S20–5.
94. Han G, Gilks CB, Leung S, et al. Mixed ovarian epithelial carci- 105. Mabuchi S, Sugiyama T, Kimura T. Clear cell carcinoma of the
nomas with clear cell and serous components are variants of high- ovary: molecular insights and future therapeutic perspectives. J
grade serous carcinoma. Am J Surg Pathol. 2008;32:955–64. Gynecol Oncol. 2016;27(3):e31.
Ovarian Mucinous, Brenner Tumors,
and Other Epithelial Tumors 7
Cathleen Matrai, Taylor M. Jenkins, Esther Baranov,
and Lauren E. Schwartz
© Science Press & Springer Nature Singapore Pte Ltd. 2019 203
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_7
204 C. Matrai et al.
a b
c d
Fig. 7.1 Mucinous cystadenoma. (a) The tumors are typically large in Microscopically, the tumor is typically composed of glands and cysts
size with a smooth external capsule. (b) Cysts may be unilocular or lined by a simple, non-stratified epithelium. (d) On high power, the
multilocular, with thick viscous contents and a glistening surface. (c) cells are bland, with basally located nuclei and abundant apical mucin
7.2 Benign Mucinous Tumors/Mucinous in size from a few centimeters to greater than 30 cm, with rare
Cystadenomas reports up to 70 kg [9]. On gross examination, the external
surface of the tumor is typically smooth, with a thick tan-white
7.2.1 Clinical Features capsule. Mucinous cystadenomas are usually multicystic, and
typically have multiple smooth-walled cysts of varying sizes
Mucinous cystadenomas comprise 10–15% of all benign with thick, viscous contents (Fig. 7.1). These features can help
ovarian neoplasms and approximately 80% of mucinous distinguish them from serous tumors grossly, which tend to be
ovarian neoplasms [1, 2, 5]. The vast majority, >95%, are smaller and generally less complex, though this is not uni-
unilateral. While these tumors occur in women of all ages, formly the case. Any papillary or solid areas should be sam-
they are most common in women in the third to fifth decades pled thoroughly, as these features are more common in
of life [1, 6]. borderline/APMT and mucinous carcinoma as well as some
special categories of mucinous tumors (i.e., mural nodules).
7.2.5 Differential Diagnosis worse prognosis [23]. Occasionally, these tumors may cause
obstructive symptoms, usually due to large size [2, 9, 25].
The differential diagnosis of benign mucinous tumors
includes:
7.3 Mucinous Borderline Tumors/Atypical
• Other cystic ovarian lesions: Proliferative Mucinous Tumors
–– Serous lesions, benign and borderline
–– Endometrioid lesions, benign and borderline 7.3.1 Clinical Features
• Metastatic mucinous tumors
Mucinous borderline tumors (MBT) comprise approximately
The main differential for mucinous cystadenomas is other 35–45% of tumors in the intermediate epithelial category [4,
cystic lesions and their borderline counterparts, especially 21]. They are less commonly bilateral than their serous coun-
those of serous and endometrioid type. The possibility of a terparts, with only approximately 7% showing involvement of
metastatic mucinous tumor from outside the ovary mimick- both ovaries [26]. While bilaterality does occur, one should be
ing a mucinous cystadenoma should always be considered prudent in confirming the site of origin, as this is a more com-
since some metastatic adenocarcinoma glandular epithelia mon feature observed in extraovarian primaries. MBT can
may have a bland-looking appearance, but this would be far occur in patients of all ages, though they are most common
less likely in the case of an otherwise unremarkable muci- between the fourth and sixth decades of life [16, 27, 28]. These
nous cystadenoma lacking in clinical concern for metastasis. tumors comprise approximately 67% of mucinous tumors that
Differentiating primary mucinous tumors from metastatic do not fall into the “benign” category [2]. In the past, MBT
lesions to the ovary will be discussed below with the differ- have also been referred to as “mucinous tumors of low malig-
ential diagnosis of mucinous carcinomas, given its higher nant potential,” though this term has fallen out of favor in recent
level of relevance to these tumors. Extensive sampling, two years. Many of the tumors initially placed in this group were
sections per cm of tumor, is recommended for mucinous characterized based upon the criteria that have now been estab-
tumors in order to accurately differentiate a mucinous cyst- lished to be associated with non-ovarian primaries, and recent
adenoma from other primary mucinous neoplasms. work has been extremely valuable in further establishing
their diagnostic features as well as prognostic implications.
Solid or firm areas can be seen, which may represent a minor growth, they are best placed in the borderline category. In
fibromatous component, densely packed small cysts, or reac- order to make this diagnosis, proliferation should involve at
tive stroma, though these areas should always be thoroughly least 10% of the tumor.
sampled to rule out malignancy [29]. Grossly, they tend to Microscopically, MBTs show a higher degree of prolifer-
have fewer loculations than seromucinous tumors, and may ation than their cystadenomatous counterparts as seen in the
contain intracystic papillae [30]. form of papillary proliferation, epithelial stratification, tuft-
ing, and villoglandular formation observed in at least 10% of
the tumor epithelial volume (Fig. 7.3). It is not uncommon to
7.3.3 Microscopic Findings find areas of associated cystadenoma [29]. Many tumors
show epithelial stratification within the range of 1–3 layers,
As with their serous counterparts, when mucinous tumors though a significant number contain foci with beyond three
show increased epithelial proliferation and mild nuclear layers of stratification. Epithelial stratification beyond even
atypia but lack destructive stromal invasion or expansile one layer should trigger close evaluation for a borderline
a b
c d
Fig. 7.3 Mucinous borderline tumor. (a) Mucinous borderline tumor extensive epithelial proliferation within glandular and cystic spaces
with background cystadenoma. Epithelial proliferation as seen in the with no associated destructive or infiltrative growth. The epithelium
center of the image comprising >10% of the tumor epithelial volume in shows papillary proliferation and stratification of the epithelium. (c)
this case. Typical mucinous cystadenomatous areas can be seen in the Mucinous borderline tumor. There is mild cytologic atypia and promi-
upper and lower portions of the image. If areas of proliferation amount nent epithelial stratification with fusion of papillae and few detached
to less than 10%, the tumor is classified as mucinous cystadenoma with cell clusters. (d) Mucinous borderline tumor with fusion of papillae and
focal epithelial proliferation. (b) Mucinous borderline tumor. There is focal cribriforming. There are few, scattered mitotic figures
208 C. Matrai et al.
component. Papillary structures may be single or branched, most common abnormality as illustrated by targeted deep
with thin fibrous cores or cores lacking in stroma. It is not sequencing rates of up to 92% [32, 33]. PIK3CA, p53 muta-
uncommon to see fusion of papillae and/or detached cell tion, and ERBB2 amplification have also been noted, though
clusters, the former of which may impart the appearance of these latter two categories to a lesser extent than seen in
small cribriform foci, not to be confused with the expansile mucinous carcinomas (11.5% vs. 56.8% and 6.2% vs. 18.8%,
growth pattern of carcinoma (Fig. 7.3). These findings may respectively) [32, 34]. MBT show lower rates of coding
be focal or multifocal, and are typically concentrated toward mutations than carcinoma and slightly higher rates of coding
the tips of papillae and centers of glandular structures. mutations than cystadenomas [34].
Tangential sectioning may further emphasize the architec-
tural complexity (Fig. 7.3).
The nuclei in MBTs are generally larger than those seen 7.3.6 Differential Diagnosis
in cystadenoma and atypia is usually mild to moderate. The
proliferative rate is typically low, and shows predominantly The differential diagnosis of borderline mucinous tumors
basal staining with Ki67 immunohistochemical stains, with includes:
decreasing staining seen toward the tips of papillae [4].
However, this diagnosis is predominantly made on histology, • Other cystic ovarian lesions:
and Ki67 staining is not routinely employed to discriminate –– Serous lesions: Benign, borderline, and malignant
between borderline tumors and its mimics. While some –– Endometrioid lesions: Benign, borderline, and
mitotic activity is typically present, numerous mitotic figures malignant
should prompt the observer to search for a more nefarious • Metastatic mucinous tumors
lesion, including intraepithelial carcinoma. Microscopic foci • Sex cord stromal tumors
of invasion may be seen (see Special Issues below), but by
definition there is no destructive stromal invasion in The main differential for mucinous borderline tumors is
MBT. Pools of acellular mucin (pseudomyxoma ovarii) are other cystic lesions and their borderline counterparts espe-
not uncommon, and frequently show associated mucin gran- cially those of serous and endometrioid type. Further, the
ulomas, findings which may mimic invasive disease. These possibility of a metastatic mucinous tumor from outside the
foci may be devoid of epithelium or there may be fragmented ovary mimicking a mucinous borderline tumor must be
free-floating epithelium suspended within mucin pools. The considered. Differentiating primary mucinous tumors from
ruptured gland may be visible in the adjacent tissue. In our metastatic lesions to the ovary will be discussed below in
experience, this and the presence of associated inflamma- the part of the chapter discussing the differential diagnosis
tion, foreign-body reaction, and edema can be helpful in dis- of mucinous carcinomas. Extensive sampling, two sections
tinguishing gland rupture and mucin granuloma from per cm of tumor, is recommended of mucinous borderline
microinvasion. tumors to rule out a carcinoma component and to better
evaluate for the possibility of a metastatic process
(Table 7.1).
7.3.4 Biomarkers
While the gold standard for treating MBT patients 7.4.2 Mucinous Borderline Tumor
includes resection with comprehensive staging, in actual- with Microinvasion
ity, the procedure performed strongly factors in patient
age, reproductive status, and desire to preserve fertility. As mentioned above, mucinous borderline tumors may be
While these tumors historically have recognized potential associated with microinvasion, which according to current
for spread, more recent studies have shown them to be classification criteria is defined as an area (or areas) of stro-
quite well behaved, with very low rates of extraovarian mal invasion measuring less than 5 mm in greatest linear
disease [36, 37]. In fact, most studies have shown that extent [21]. Multiple proposals have been put forth regarding
patients who underwent conservative surgery without standardized criteria for microinvasive foci, including 3 mm
complete staging did not have higher rates of recurrence in linear extent or 10mm2 in area, and there has generally
[38]. There has been a report of fertility-sparing surgery been no agreed-upon definition over the last years. However,
which showed higher rates of invasive recurrence than tumors with less than 5 mm of invasion have demonstrated
serous tumors, though this was only in one single-center an excellent prognosis, and this is the current cutoff as rec-
study and requires follow-up confirmatory studies [39]. ommended by the 2014 WHO [21]. Multiple foci of microin-
Microinvasion in these tumors typically does not portend vasive carcinoma may be seen but no single focus should
a worse prognosis [4, 21, 40]. However, there remain measure greater than 5 mm in order to qualify for this diag-
areas in need of further studies, such as whether or not the nosis. There is no limit to the number of foci identified in a
number of microinvasive foci or the cytology of the micro- single tumor, and geographically distinct foci should not be
invasive foci (low- or high-grade nuclei) has particular combined for an additive measurement. Microinvasive foci
significance [40]. are comprised of single cells, clusters, and/or small foci of
confluent or cribriform glandular growth within the stroma
(Fig. 7.5) [4, 21]. Most studies suggest that the rate of micro-
7.4 pecial Issues with Mucinous
S invasion falls in the 10–18% range, though figures up to 29%
Borderline Tumors have been described [4, 21, 40, 42]. Mucin granulomas with
associated epithelium may be a diagnostic pitfall in these
7.4.1 Mucinous Borderline Tumor cases, and may likely be the source of these higher reported
with Intraepithelial Carcinoma rates in the literature [40, 43].
While a great deal is now known about these tumors,
While epithelial stratification and architectural complexity there is still much that remains to be elucidated, includ-
are frequent features of typical MBT, marked nuclear atypia ing whether the presence of microinvasive disease in
is not. Intraepithelial carcinoma (IEC), as defined by the mucinous tumors with concurrent IEC has special signifi-
presence of high-grade nuclear atypia which differs from the cance [40]. Some authors have referred to mucinous
background epithelium, has been reported in up to 40–55% tumors with IEC and microinvasion as “microinvasive
of mucinous borderline tumors [4]. Historically, the defini- intraepithelial mucinous carcinoma” [40]. For prognosti-
tion of intraepithelial carcinoma in mucinous tumors has cation purposes, a cutoff of 10% tumor involvement has
been variable and has included the presence of both nuclear been proposed [40].
atypia and architectural complexity [40]. However, recent
proposals have suggested only grade 3 nuclear atypia or
frankly malignant cells be considered diagnostic for intraepi- 7.4.3 Mucinous Borderline Tumor
thelial carcinoma, and some follow-up studies have shown with Pseudomyxoma Peritonei
recurrence rates to be higher in this group while no differ-
ence was detected based upon architectural complexity [40]. Pseudomyxoma peritonei (PMP) is a rare clinical entity
These areas are typically cytologically distinct from the characterized by mucinous ascites and/or mucin deposits
background (a very helpful feature), with pleomorphism, within the peritoneal cavity. It was initially thought to
hyperchromasia, loss of polarity, and often increased mitotic represent the mucinous parallel of tumor implants associated
activity (Fig. 7.4). with serous borderline tumors. PMP typically has a pro-
Overall, most studies have not shown decreased survival tracted clinical course with multiple recurrences and obstruc-
based upon the presence of intraepithelial carcinoma, and the tive symptoms, with progressive disease and death being
prognosis is generally regarded as excellent [41]. MBT with seen in up to half of patients [44, 45]. It has been reported as
intraepithelial carcinoma is more commonly reported at fre- an incidental finding in approximately 2 of every 10,000
quencies of 30–40% [41], though it has been reported in up laparotomies [46]. Histologically, mucoid deposits may con-
to 40–55% of mucinous borderline tumors [40]. tain variable amounts of typically low-grade or cytologically
210 C. Matrai et al.
a b
Fig. 7.4 Mucinous borderline tumor with intraepithelial carcinoma. can be seen. (b) There is cytologic atypia and notably increased mitotic
(a) Both architectural complexity and high-grade nuclear atypia (right activity. (c). Marked cytologic atypia is seen
side of image) distinct from non-atypical epithelium (left side of image)
bland mucinous epithelium. While this entity has historically 7.4.4 Mucinous Tumors with Mural Nodules
been controversial in terms of its origin, numerous clinico-
pathologic and molecular studies have now concluded that Mural nodules are rare nodules that occur singly or multiply
the vast majority of PMP cases have an extraovarian origin, in mucinous ovarian tumors of all types, with the first series
most commonly the appendix. The majority of these tumors reported by Prat and Scully in 1979 [51]. There are three
are now termed “low-grade appendiceal mucinous neo- main types: (1) those composed of reactive sarcoma-like tis-
plasms” [45, 47, 48]. The high rates of bilateral ovarian sue, termed sarcoma-like mural nodule (SLMN); (2) true
involvement, concurrent appendiceal neoplasm with similar sarcoma; and (3) anaplastic carcinoma [51–53]. These nod-
histologic features, and progressive molecular derangements ules vary widely in size and may be sharply demarcated from
including identical KRAS mutations and loss of heterozy- the surrounding tissue in the cases of SLMN and less well
gosity in ovarian as compared to appendiceal tumors favor circumscribed in the latter two categories. Patients with
this conclusion [45, 49]. The rare cases of PMP of true ovar- SLMNs are typically younger than those with sarcoma/ana-
ian origin are most commonly seen in association with tumor plastic carcinoma, with a mean age of approximately
rupture, specifically in cases of mucinous tumors arising in 36 years [53]. The origin of SLMNs is unclear, and it is not
mature cystic teratomas. These tumors are morphologically known whether this entity represents a reactive or neoplastic
and immunohistochemically similar to primary gastrointesti- process.
nal tumors and very likely arise from gut elements of the Microscopically, the histologic appearance of SLMNs is
teratoma [50]. varied and tumors may consist of multinucleated osteoclast-
7 Ovarian Mucinous, Brenner Tumors, and Other Epithelial Tumors 211
a b
Fig. 7.5 Mucinous borderline tumor with microinvasion. (a) There is a derline tumors with microinvasion. The invasive cells show cytologic
haphazard arrangement of infiltrative small glands, cell clusters, and atypia and are arranged as single cells and nests within reactive stroma.
single cells with areas of associated extracellular mucin within the (c) Haphazard arrangement of angulated glands underlying noninvasive
stroma. The invasive foci measure less than 5 mm. (b) Mucinous bor- tumor. There is surrounding desmoplastic response
and epulis-like giant cells, pleomorphic spindle cells, and/or upon this component. While early studies illustrated the poor
histiocytes/inflammation. They may contain conspicuous behavior of tumors with malignant nodules with fatality rates
mitoses but do not show invasion into the surrounding tissue approaching 50%, few recent studies have shown that if a
or capillary lymphatic spaces. Sarcomatous nodules are most tumor is stage 1a, there may not be an adverse effect on the
commonly composed of undifferentiated sarcoma, though outcome [53]. However, given the limited information avail-
many types have been reported including fibrosarcoma, able, these tumors should be regarded with caution when
rhabdosarcoma, and more recently osteosarcoma [54]. encountered in clinical practice.
Anaplastic carcinomatous nodules show large atypical cells
with irregular nuclei and prominent nucleoli that are positive
for cytokeratins and EMA, a feature which can be very help- 7.5 Mucinous Carcinoma
ful in distinguishing more challenging cases from SLMNs.
The growth pattern may be rhabdoid, spindled, and/or Primary mucinous carcinoma of the ovary is very rare, repre-
pleomorphic. senting <5% of all primary ovarian carcinomas [55].
Thorough sampling and accurate diagnosis are crucial, as Although primary mucinous carcinoma of the ovary is rare,
SLMNs do not have an adverse effect on prognosis [21]. metastatic adenocarcinomas that often mimic primary ovar-
When an anaplastic carcinomatous or true sarcomatous com- ian mucinous carcinoma are common, and differentiating
ponent is identified, the tumor should be classified based these entities is of the utmost importance to the practice of
212 C. Matrai et al.
gynecologic pathology. In this part of the chapter, primary 7.5.3 Pathological Findings
mucinous carcinoma is discussed with an emphasis on dif-
ferentiating this neoplasm from the various metastatic ade- For the diagnosis of carcinoma to be rendered for a primary
nocarcinomas that have been reported to involve the ovary. mucinous neoplasm of the ovary one of the two, or both,
invasive patterns must be noted. The two invasive patterns
described in mucinous carcinoma are the confluent glandular
7.5.1 Clinical Features or expansile pattern and the destructive stromal invasive pat-
tern. The confluent glandular/expansile pattern denotes
Patients with primary mucinous carcinoma of the ovary architecturally complex glandular formations with minimal
tend to present at a younger age than those with serous to no intervening stroma (Fig. 7.6) [59, 60]. In other words,
tumors. The mean ages reported in the literature vary, but differentiating a borderline mucinous neoplasm from a muci-
generally note the majority of patients to be in their 40s nous carcinoma with confluent glandular/expansile invasion
[56, 57]. Patients generally present with increasing abdom- is similar to differentiating atypical endometrioid hyperpla-
inal girth, as is common for the other mucinous neoplasms sia from endometrioid carcinoma. In contrast to the glandu-
of the ovary [57]. The majority of patients diagnosed with lar/expansile pattern of invasion, destructive stromal invasion
primary mucinous carcinoma of the ovary are found to be is characterized by altered glands infiltrating altered stroma
of low stage with advance-stage disease noted in only very often leading to an inflammatory response. The malignant
rare cases [58]. glands noted in destructive invasion often exhibit altered
cytologic and architectural features compared to adjacent
noninvasive components of the same neoplasm (Fig. 7.7).
7.5.2 Gross Findings The altered features may include rounded glandular contours
with stroma now evident between glands and single cells and
At gross inspection mucinous carcinomas appear similar to cytologic changes including increasing cytoplasmic eosino-
borderline mucinous tumors, in that they tend to be large, philia combined with decreased intracellular mucin [60].
unilateral, and multicystic. The surface of the ovary should
be examined in detail and any nodules noted, as surface
involvement will affect staging. Often the surface of the neo- 7.5.4 Biomarkers
plasm will be smooth with the neoplasm confined to the
ovary. Rupture should also be noted. On cut section the neo- Limited studies are available discussing the role of biomark-
plasm is usually solid and cystic. Mucinous material should ers in the diagnosis and management of primary mucinous
be exuded from the neoplasm similar to other mucinous neo- carcinoma. In a study of patients diagnosed with mucinous
plasms. Microscopic examination, rather than gross exami- ovarian lesions preoperative CA19.9 and CA125 levels were
nation, is critical to determine a mucinous borderline tumor shown to be elevated in patients with carcinoma as compared
from a mucinous carcinoma. to patients found to have benign mucinous neoplasms at the
time of resection [61]. Further, in patients with normal The main differential diagnosis of primary mucinous
CA125, CA19.9 has been shown to be elevated in patients ovarian carcinoma is a borderline mucinous ovarian tumor as
with borderline and malignant mucinous ovarian neoplasms described above and metastatic carcinoma involving the
as compared to benign neoplasms [62]. Another tumor ovary. Metastatic lesions from a variety of organs have been
marker that is often elevated in patients with mucinous ovar- reported to involve the ovary and mimic primary ovarian
ian carcinoma is carcinoembryonic antigen (CEA) [2]. All mucinous neoplasms. The most commonly reported meta-
three markers CA19.9, CA125, and CEA can be used to fol- static carcinomas to the ovary are of gastrointestinal origin.
low patients postoperatively; however, none have proven These include carcinomas from the colon, stomach, small
effective as a screening test for primary mucinous ovarian bowel, and pancreaticobiliary tract. Further, extraovarian
carcinoma. gynecologic primaries, such as those from the endocervix,
have also been noted to metastasize to the ovary and mimic
primary ovarian mucinous carcinoma [59] (Table 7.2).
7.5.5 Molecular Features In the workup of a mucinous carcinoma involving the
ovary clinical, morphologic, and immunohistochemical fea-
A variety of molecular alterations have been reported in pri- tures of the neoplasm are important to consider when deter-
mary ovarian mucinous carcinomas, with mutations in KRAS mining if the neoplasm is primary to the ovary or metastatic.
being the most common [33, 63]. Other reported molecular A series of algorithms described in the literature conclude
alterations include HER2 amplification, aberrant signaling that a good place to start when evaluating mucinous neo-
involving the WNT pathway, and TP53 mutations [63, 64]. plasms is considering size and laterality. A general rule sug-
KRAS mutations and HER2 amplification have been found to gests that bilateral small tumors (<10–13 cm) are highly
be nearly mutually exclusive [64], while TP53 mutations suspicious for metastasis while unilateral large (>10–13 cm)
may occur in the context of other molecular alterations [63]. tumors are more suggestive of primary ovarian lesions [59,
65]. Important to note is that metastatic colorectal carcinomas
as well as metastatic endocervical carcinomas have been
7.5.6 Differential Diagnosis noted to not follow these general rules [59].
A series of studies have described a variety of immuno-
The differential diagnosis for mucinous carcinomas of the histochemical markers (Table 7.3) that can be helpful in dis-
ovary is relatively broad and includes: criminating primary mucinous carcinomas of the ovary from
metastatic lesions. The two most discussed cytokeratin stains
• Metastatic carcinoma involving ovary: used in the workup of mucinous lesions of the ovary are CK7
–– Upper and lower tract gastrointestinal primaries: and CK20 [31]. In general, CK7 diffuse positivity suggests a
∘∘ Colonic carcinoma primary ovarian lesion, but must be interpreted with caution
∘∘ Pancreatic carcinoma as metastatic lesions from the upper gastrointestinal prima-
–– Endocervical carcinoma ries can also show diffuse positivity. As far as CK20, diffuse
• Borderline mucinous ovarian tumors positivity suggests a lower gastrointestinal primary, but can
Table 7.2 Primary ovarian mucinous neoplasms versus metastatic lesions involving the ovary
Entity Gross features Histopathologic features Immunohistochemistry
Primary ovarian Tend to be unilateral Often identified within the stroma of the ovary with CK 7 expression greater than CK 20
mucinous and large with mean well-differentiated mucinous epithelium with variable expression, occasionally PAX8
neoplasm size >20 cm atypia. Carcinomas show two patterns of invasion positive
(confluent/expansile and destructive stromal). Often lack
surface involvement of the ovary
Metastatic Tend to be bilateral Can be noted to involve the surface of the ovary, Appendix/Lower GI : CK20 positive,
carcinomas and smaller superficial cortex, and stroma CK7 negative, SATB2 positive
involving the (<10–13 cm) Upper GI: CK7>CD20 expression,
ovary PAX8 negative
Pancreas: CK7>CK20, Can see loss
of DPC4
Endocervix: p16 diffuse and
HPV-ISH positive if HPV related,
Non-HPV related lesions difficult to
determine by IHC
Note: Important to consider all features together including clinical history as this is a very difficult differential diagnosis in many instances
214 C. Matrai et al.
Table 7.3 Immunohistochemical stains to consider in the workup of mary ovarian carcinomas tend to overlap. An
mucinous lesions involving the ovary
immunohistochemical stain for DPC4 has been shown to be
CK7 Compare with CK20 expression CK7>CK20 → Favor helpful. DPC4 staining is lost in many pancreatic and biliary
primary ovarian over upper GI tract
tract adenocarcinomas, whereas expression is retained in
CK20 Compare with CK7 expression CK20 > CK7 → Favor
ovarian primaries and metastases from other sites [68].
lower GI tract over primary ovarian
PAX8 Positivity favors primary ovarian Additional immunohistochemical stains that have been
CDX2 Not very helpful, positivity supports mucinous type explored in the workup of ovarian mucinous neoplasms are
SATB2 Positivity suggests lower GI tract estrogen receptor (ER) and progesterone receptor (PR).
DPC4 Loss of staining suggests pancreatic primary Unfortunately, these markers have proved for the most part
p16 Diffuse positivity suggests possible HPV-related cervical unhelpful since all primary mucinous carcinomas and the
carcinoma primary strong majority of metastatic mucinous carcinomas were
found to be negative for both markers. The only exception
be seen in some primary ovarian mucinous tumors [31]. was borderline seromucinous tumors, which were noted to
Critical to the interpretation of these markers is the compari- variably express ER [69]. ER and PR are helpful if an endo-
son of CK7 and CK20 expression. metrioid ovarian primary or metastasis is in the differential
PAX8, a commonly used Mullerian marker, is of some as these types of tumors should express ER and PR.
use in the workup of mucinous neoplasms of the ovary. Lastly, if a metastatic lesion from the cervix is under con-
Although positivity is found in the minority of primary ovar- sideration, p16 staining may be helpful. P16 expression should
ian mucinous neoplasms, when present, PAX8 positivity is be strong and diffuse in metastatic usual-type endocervical
strongly suggestive of a primary ovarian neoplasm, espe- adenocarcinomas to the ovary as these are invariably HPV
cially if metastasis from other Mullerian organs has been related [70]. When p16 is equivocal, HPV in situ hybridization
eliminated [66]. can be used to aid the diagnosis. Non-HPV-related endocervi-
CDX2, a transcription factor involved in intestinal differen- cal carcinomas can be extremely difficult to differentiate from
tiation, has been suggested as a helpful marker in distinguishing primary ovarian mucinous neoplasms. For this differential the
primary mucinous ovarian carcinoma from metastasis given size and laterality algorithm is of great importance, as well as
that CDX2 is positive in a large majority of colorectal carcino- correlation with history and examination of the cervix.
mas; however, expression has also been shown in ovarian muci- In summary, a variety of factors should be considered
nous carcinomas as well as appendiceal, pancreaticobiliary, and when differentiating a primary mucinous ovarian carcinoma
gastric carcinomas. As a result, it is important to interpret CDX2 or other mucinous neoplasms from a metastasis. The most
staining in the context of CK7 and CK20 staining realizing that important factors are size of the lesion(s) and laterality as
its expression or lack thereof may not be any more helpful than well as clinical history. Immunohistochemistry may prove
the comparison of CK7 to CK20 expression [31]. helpful in various situations, but should be interpreted with
SATB2 is a relatively new marker of lower intestinal dif- caution as the patterns of expression often overlap in the pri-
ferentiation that has been shown to be helpful in discriminat- mary and metastatic lesions (Table 7.1).
ing primary ovarian mucinous carcinomas from metastatic
lower gastrointestinal primaries. Like CDX2, it is suggested
that SATB2 staining be interpreted in the context of CK7 and 7.5.7 Management and Outcomes
CK20 staining. Unlike CDX2, SATB2 has been shown to be
negative in cervical metastasis and pancreaticobiliary metas- Surgery is the mainstay of treatment for primary ovarian car-
tases. Further, in a recent study SATB2 was noted to be nega- cinomas of the ovary [57]. Fortunately, the majority of muci-
tive in all primary mucinous carcinomas of the ovary except nous ovarian carcinomas are found to be confined to the
for those arising from a teratoma, further supporting its use ovary at diagnosis and can be removed in their entirety.
in the workup of mucinous lesions of the ovary [67]. Patients with successful surgery often have an excellent
The differential diagnosis of a metastatic pancreaticobili- prognosis with greater than 90% being disease free at 5 years
ary carcinoma to the ovary versus a primary ovarian muci- [57]. Mucinous carcinomas tend to be less responsive to che-
nous carcinoma can be especially challenging, although motherapy, especially platinum-based therapies, than other
applying the algorithm of size and laterality can be extremely ovarian epithelial neoplasms and as a result patients with
helpful. Many pancreaticobiliary metastases simulate bor- advanced disease often have a poor prognosis [57]. Currently
derline neoplasms and cystadenomas as they tend to have a there is not a specific primary chemotherapy regime recom-
simplistic architecture [68]. Further, many of these metasta- mended for primary mucinous ovarian carcinoma. The use of
ses may present at the same time if not before the main pan- gastrointestinal regimes is being explored and future studies
creaticobiliary primary. Many of the immunohistochemical may reveal an ideal regimen [57]. Further, futures studies
stains discussed above, CK7, CK20, and CDX2, are of lim- may also reveal the benefits of more targeted therapy in the
ited value as the patterns seen in pancreaticobiliary and pri- emerging era of personalized medicine.
7 Ovarian Mucinous, Brenner Tumors, and Other Epithelial Tumors 215
Fig. 7.8 Walthard cell nest associated with fallopian tube at the tubo- 7.7 Benign Brenner Tumors
peritoneal junction
7.7.1 Clinical Features
7.6 Ovarian Brenner Tumors Benign Brenner tumors represent 5% of all benign ovarian
tumors and typically arise in adults in the fifth to seventh
Brenner tumors of the ovary represent approximately 2% of decades (mean age 50 years), but can occur in patients
all primary surface epithelial ovarian tumors [71]. Brenner younger than 30 or older than 80 [72]. Most patients are
tumors are further classified as benign, borderline, and asymptomatic and the tumors are found incidentally. If the
malignant, depending on the degree of architectural com- tumors become large the patient may experience abdominal
plexity, cellular proliferation, nuclear atypia, and stromal discomfort or distension. Brenner tumors may rarely have
invasion. Brenner tumors are predominantly benign (>90%) functioning stroma resulting in endocrine symptoms [21].
with rare transformation to borderline or malignant Brenner
tumors (<10%). Borderline Brenner tumors present at an
older age (mean age 60 years), are large, and are often symp- 7.7.2 Gross Findings
tomatic related to the size of the tumor. Malignant Brenner
tumors occur during the fifth to sixth decades (mean age Most benign Brenner tumors (>50%) are <2 cm but can rarely
55 years) with similar symptoms to borderline tumors [72] reach up to 10 cm or greater (<10%) [76]. They are frequently
(Table 7.4). diagnosed incidentally in ovaries removed for another reason.
Brenner tumors are thought to arise from Walthard cell Grossly they are solid, white-tan to yellow with a firm, rubbery,
nests which develop in the setting of transitional metaplasia homogenous cut surface, similar to a fibroma. Calcifications
at the tubo-peritoneal junction (Fig. 7.8) [73]. This junction and small cysts filled with mucinous material may be seen.
216 C. Matrai et al.
a b
Fig. 7.9 Benign Brenner tumor. (a) Lower power showing nests of bland transitional type epithelial cells in a background of paucicellular fibro-
matous stroma. (b) Higher power of nest within benign Brenner tumor highlighting the cytologic features of the neoplasm
Rarely the tumor may be predominantly cystic. Approximately nous component is PAX8 negative, suggesting that these
25% of benign Brenner tumors are associated with other tumors do not arise from Mullerian epithelium.
tumors, most commonly mucinous ovarian neoplasms [77].
Less than 10% of Brenner tumors are bilateral [72].
7.7.4 Biomarkers
7.7.3 Microscopic Findings Brenner tumors are positive for cytokeratins, EMA, WT1,
p63, S100, GATA3, uroplakin, and thrombomodulin [16,
Brenner tumors of the ovary have epithelial elements resem- 21]. Most exhibit CK7 positivity and CK20 negativity, simi-
bling urothelial transitional cell epithelium. They are composed lar to other ovarian primary surface epithelial tumors. In con-
of oval to irregular nests of bland transitional type epithelial trast, transitional cell tumors of the urinary tract (urothelial
cells in a background of paucicellular fibromatous stroma carcinomas) in most instances exhibit both CK7 and CK20
(Fig. 7.9). The nests are usually solid but may also contain positivity. Brenner tumors exhibit evidence of urothelial dif-
mucinous or eosinophilic material within a central cavity. ferentiation with GATA3, uroplakin, and thrombomodulin
Usually the cells have moderate to abundant amounts of ampho- expression. Benign Brenner tumors may have an endocrine
philic to clear cytoplasm. The nuclei are oval with fine chroma- cell component and are often immunopositive for chromo-
tin and characteristic nuclear grooves (“coffee bean” nuclei). granin A, serotonin, and neuron-specific enolase (NSE) [82]
Small nucleoli may be present. Epithelial atypia and mitoses are (Table 7.5).
rare. Focal or extensive calcification may be present (Fig. 7.9).
Approximately 25% of Brenner tumors are seen in asso-
ciation with other benign tumors of the ovary, such as muci- 7.7.5 Molecular Features
nous cystadenomas (most common), teratomas, and struma
ovarii. In cases with struma ovarii association, the Brenner Few molecular alterations have been described in benign
component may stain with thyroglobulin [78–80]. A recent Brenner tumors. KRAS mutations at codon 12 were identified
study showed that clonality analysis of combined Brenner in one study [83]. There has been one reported case of a
and mucinous tumors of the ovary demonstrated that these 12q14-21 amplification [84]. Benign Brenner tumors associ-
two morphologies share a monoclonal origin [81]. In the ated with mucinous cystadenomas have been reported to
combined Brenner and mucinous cystadenomas, the muci- have MYC amplification in both components [80].
7 Ovarian Mucinous, Brenner Tumors, and Other Epithelial Tumors 217
7.7.6 Differential Diagnosis enlarging pelvic mass and the tumors are usually unilateral
and confined to the ovary [74]. Borderline Brenner tumors
The differential diagnosis of benign Brenner tumors includes: are thought to arise from their benign counterparts [89].
a b
Fig. 7.11 Borderline Brenner tumor in combination with borderline mucinous tumor. (a) Low power showing the various components of the
tumor. (b) Higher power highlighting the transition between the borderline Brenner component and mucinous component
p16, Rb, and p53. P16 is usually negative in the epithelial pools of acellular mucin. Serous carcinomas usually have
component of borderline Brenner tumors despite being posi- more atypia, higher nuclear to cytoplasmic ratios, and hob-
tive in the majority (92%) of benign Brenner tumors [89]. nailing. Immunohistochemical stains will help differentiate
This may be associated with promoter hypermethylation and these lesions. It is important to keep in mind that borderline
a homozygous deletion in CDKN2A, the gene encoding p16 Brenner tumors commonly occur simultaneously with muci-
[89] (Table 7.5). nous or serous ovarian neoplasms (Fig. 7.11). Whenever a
borderline Brenner tumor is identified, a malignant Brenner
tumor must be excluded by ruling out stromal invasion.
7.8.5 Molecular Features Finally, adult granulosa cell tumors can frequently mimic
many other ovarian neoplasms, including Brenner tumors.
Homozygous deletions of CDKN2A (p16-encoding gene) They can both have elongated nuclei with nuclear grooves,
have been identified in the epithelial component of border- moderate amounts of cytoplasm, and solid growth patterns.
line Brenner tumors (retained in benign). PIK3CA somatic Patients with granulosa cell tumors frequently present with
mutations have been detected in the stromal component in hyperestrogenism and endometrial hyperplasia.
5% of benign Brenner tumors. Somatic mutations in KRAS Immunohistochemically, granulosa cell tumors are positive
and PIK3CA are identified in 29% of borderline Brenner for inhibin, calretinin, CD99, SMA, and keratins (dot-like
tumors. Loss of CDKN2A and to a lesser extent KRAS and positivity) and are negative for EMA and uroplakin [21, 91,
PIK3CA somatic mutations may play a role in the transition 92].
of a benign to a borderline Brenner tumor [80, 89].
Approximately 80–90% of patients present with stage I disease 7.9.5 Molecular Features
with a 90% survival and the remaining 10–20% present with
stage II–IV disease [93, 94]. These patients usually present EGFR-RAS-MAPK mutations, Exon 9 PIK3CA mutations,
with abdominal pain or an enlarging abdominal mass. Rarely, P16 LOH [20], and Ras driver mutations have all been
abnormal vaginal bleeding may be the initial presentation [94]. described [7]. Although malignant Brenner tumors may mor-
phologically resemble urothelial carcinoma, the pathogene-
sis is distinct. TERT promoter mutations, commonly present
7.9.2 Gross Findings in urothelial carcinoma, have not been identified in benign or
malignant Brenner tumors [21].
Malignant Brenner tumors are 16–20 cm in greatest dimen-
sion on average [94]. They are cystic and solid with papillary
or polypoid components. A gritty texture on cut surface may 7.9.6 Differential Diagnosis
represent prominent calcifications.
The differential diagnosis of malignant Brenner tumors
includes:
7.9.3 Microscopic Findings
• Transitional-like high-grade serous carcinoma
Microscopically, malignant Brenner tumors have an exuber- • Serous ovarian cancer
antly proliferative transitional epithelium with cellular atypia • Metastatic urothelial carcinoma
and unequivocal stromal invasion in association with a • Metastatic squamous cell carcinoma
benign or borderline Brenner tumor component. The malig-
nant component consists of large, closely packed irregular Transitional-like high-grade serous carcinoma (TLHGSC)
nests of transitional epithelial cells infiltrating into the sur- is a variant of high-grade serous carcinoma of the ovary. This
rounding stroma. A desmoplastic stromal reaction will help entity was previously classified under the same umbrella as
identify the areas of invasion (Fig. 7.12). The epithelial cells Brenner tumors (transitional cell tumors of the ovary) and
demonstrate nuclear pleomorphism with hyperchromasia was called transitional cell carcinoma. These malignant
and numerous mitoses. Foci of necrosis are commonly seen. tumors have since been found to be related to high-grade
Calcifications are often prominent and squamous differentia- ovarian serous carcinoma and have been reclassified as such.
tion has been reported [21, 95]. TLHGSC is a distinct entity from the malignant Brenner
tumor [2, 20]. Histologically this tumor resembles a malig-
nant urothelial neoplasm but does not have a benign or bor-
7.9.4 Biomarkers derline Brenner tumor component. They frequently have a
papillary architecture with a multilayered transitional cell
Malignant Brenner tumors have a similar immunoprofile as epithelium and smooth luminal borders. A nested pattern
benign/borderline Brenner tumors; however, there is variable similar to that seen in malignant Brenner tumors can be seen
expression in the invasive components (Table 7.5). and may be a diagnostic pitfall. TLHGSC is usually strongly
a b
Fig. 7.12 Malignant Brenner tumor. (a) Low power highlighting the infiltrative nature of the glandular component of the tumor. (b) Higher power
highlighting the cytologic atypia associated with the malignant nature of this tumor
220 C. Matrai et al.
positive for p16, PAX8, and WT1 and shows p53 staining clinical importance, especially in benign and borderline
consistent with a p53 mutation, and is negative for EGFR, lesions. In the cases of mixed carcinoma, it may become
cyclin D1, and Ras. This staining pattern is the opposite for important to note and describe the component that composes
borderline and malignant Brenner tumors [21]. the minority of the carcinoma, especially if the carcinoma
Metastatic urothelial or metastatic squamous cell carcino- that composes the minority of the specimen is of higher
mas may mimic malignant Brenner tumors and immunohis- grade than the other component(s) or if metastases are identi-
tochemistry may be necessary for definitive diagnosis. In the fied that appear similar histologically to the minority compo-
setting of exuberant calcification, ovarian serous carcinoma/ nent. Immunohistochemical stains can be especially helpful
psammocarcinoma may also be a diagnostic pitfall; however, to better classify the components of mixed lesions; however,
immunohistochemistry can easily distinguish these entities stains must be interpreted with caution as the staining pat-
[96]. terns of many high-grade ovarian epithelial carcinomas over-
lap. It is our practice to note mixed histology and to describe
histologic and immunohistochemical findings. We find that
7.9.7 Management and Outcomes open communication with our clinicians in these particular
cases is of the utmost importance as this often assists the
Patients with malignant Brenner tumors most commonly oncologists in determining the best course of therapy.
present with stage I disease (55.4%). 14.4% of patients pres-
ent with stage II disease, 18% with stage III, and 12.2% with
stage IV disease [21, 93]. Approximately 5% of patients have 7.10.3 Undifferentiated Carcinoma
positive lymph nodes at diagnosis. Five-year disease-specific and Unclassified Adenocarcinoma
survival (DSS) of tumors confined to the ovary is 94.5%
compared to 51.3% for those with extraovarian spread [93]. In extremely rare circumstances a diagnosis of unclassified
Regional lymphatic spread is uncommon and thus regional adenocarcinoma may be rendered. This term should only be
lymphadenectomy as part of staging is unnecessary and does used in the extremely rare case that exhibits no clear charac-
not confer any improvement on survival. One case report teristics of any type of ovarian malignancy, but expresses
demonstrated transformation of a malignant Brenner tumor epithelial markers. Further, we feel that this diagnosis should
into a trabecular carcinoid [97]. not be rendered unless extensive sampling, an extensive
immunohistochemical workup, as well as extensive consul-
tation with experts in the field have been performed. It is
7.10 Other Epithelial Tumors especially important to consider metastatic lesions and non-
epithelial ovarian lesions as many of these may focally
7.10.1 Squamous Cell Carcinoma express epithelial stains. Hopefully, as ancillary studies con-
tinue to develop and become more common, the term unclas-
Very few pure squamous cell carcinomas (pSCC) primary to sified adenocarcinoma will disappear.
the ovary have been reported in the literature [98, 99]. The
majority of squamous cell carcinomas noted to involve the
ovary arise in association with teratomas, Brenner tumors, or 7.11 Other Rare Tumors
endometriosis, or represent metastatic disease [99]. It is very
important that all of these possible associations are elimi- 7.11.1 Small-Cell Carcinoma of the Ovary,
nated before a diagnosis of pSCC is considered. The origin Hypercalcemic Type (SCCOHT)
of pSCC is believed to be metaplasia of the surface lining of
the ovary [99]. Given the rarity of pSCC treatment guidelines Small-cell carcinoma of the ovary, hypercalcemic type
do not exist. Based on various case reports, it is believed that (SCCOHT), is a rare highly aggressive malignancy first
pSCC is very aggressive and should be managed aggres- described in 1979 by Robert E. Scully, and currently defined
sively [98, 100]. by deleterious genetic mutations in SMARCA4 and loss of
nuclear immunostaining for SMARCA4 (BRG1) [101].
SCCOHT was originally speculated to represent poorly dif-
7.10.2 Mixed Epithelial Tumors ferentiated carcinoma, although sex cord-stromal, germ cell,
and neuroendocrine origins have also been suggested [102,
Mixed epithelial tumors of the ovary are defined as ovarian 103]. While still characterized as a miscellaneous ovarian
tumors composed of two or more distinct epithelial compo- neoplasm, recent data suggests either a primitive germ cell or
nents, each representing >10% of the tumor. Fortunately, in mesenchymal origin for SCCOHT [104, 105]. Some have
most circumstances, the mixed nature of the lesion has no even suggested a change in nomenclature from SCCOHT to
7 Ovarian Mucinous, Brenner Tumors, and Other Epithelial Tumors 221
malignant rhabdoid tumor of the ovary (MRTO), given the contain characteristic follicle-like structures with eosino-
clinical, morphologic, and molecular similarities to malig- philic (rarely basophilic) fluid [106]. Roughly half of tumors
nant rhabdoid tumors [105]. contain foci or diffuse areas of neoplastic cells with moder-
Small-cell carcinoma of the ovary, hypercalcemic type ate to abundant eosinophilic cytoplasm, large pale nuclei,
(SCCOHT), affects primarily adolescent and young women, prominent nucleoli, and occasionally cytoplasmic hyaline
with a peak incidence in the second and third decades of life, globules. Cases with exclusively or predominantly such
although the age at diagnosis reported in the literature ranges tumor morphology are designated “large cell variant” of
between 14 months and 71 years of age [106, 107]. Patients SCCOHT [106]. Additional, but far more infrequent, mor-
most often present with abdominal pain or distention, but phologies include spindle cell or clear-cell components, mul-
may also present with a palpable mass, weight loss, nausea, tinucleate cells, or cystic glandular spaces lined by benign or
vomiting, and/or constipation [107]. Paraneoplastic hyper- malignant-appearing mucinous epithelium [106].
calcemia (elevated preoperative calcium) is present in The utility of immunohistochemistry in diagnosing
approximately 60% of cases, though the majority of these SCCOHT was revolutionized by the recent discovery of
patients are not symptomatic [106, 107]. Importantly, famil- disease-driving SMARCA4 mutations and the development
ial cases have been described with germline mutations in of an antibody against SMARCA4 (BRG1) [101]. Complete
SMARCA4 and autosomal dominant patterns of inheritance loss of nuclear staining with the SMARCA4 antibody is
[101, 108]. Patients with germline SMARCA4 mutations are found in greater than 95% of SCCOHT cases [105]. Most
often diagnosed at a younger age than patients with somatic cases also reveal diffuse nuclear immunoreactivity to WT1,
mutations [101]. with an antibody against the N-terminus of WT1 [103].
Gross examination typically reveals these tumors to be Other historically utilized immunostains include cytokera-
large (diameter 14.7 cm, range 6–30 cm), solid, fleshy, tins (focally positive, cytoplasmic), EMA (focally positive,
cream-colored masses with foci of hemorrhage, necrosis, membranous), calretinin (focally positive), CD10 (focally
and cystic degeneration, closely resembling ovarian lympho- positive), α-inhibin (negative), desmin (negative), and S100
mas or dysgerminomas [106, 107]. While typically unilat- (negative) [102, 103]. Interestingly, cases have been report-
eral, these tumors can rarely be bilateral, particularly in edly immunoreactive for parathyroid hormone-related pro-
familial cases with germline mutations in SMARCA4 [108]. tein (PTHRP), irrespective of preoperative calcium levels,
Microscopically, these tumors are predominantly com- but almost always negative for parathyroid hormone (PTH)
posed of diffuse sheets, or less commonly trabeculae, of [106].
monotonous small round blue cells with scant cytoplasm, SCCOHT is characterized by deleterious somatic or
ovoid to round hyperchromatic nuclei, and brisk mitotic germline mutations in the SMARCA4 gene [101]. SMARCA4
activity [106] (Fig. 7.13). There is often minimal tumor encodes the SMARCA4 (BRG1) protein, an enzymatic com-
stroma, though infrequently the stroma may be myxoid, ponent of the SWI/SNF chromatin-remodeling complex,
fibrous, or edematous [106]. Approximately 80% of tumors which is responsible for the control of downstream gene
expression [105]. Mutations in both gene alleles or loss of
heterozygosity is required for tumor formation [101].
Deleterious mutations in SMARCA4 have also been impli-
cated in rhabdoid tumor predisposition syndrome type 2
(RTPS2), a syndrome that predisposes patients to developing
malignant rhabdoid tumors [101]. In one study, approxi-
mately 50% of patients with known SCCOHT that were
tested for a germline mutation in SMARCA4 were found to
have a mutation, regardless of positive or negative family
history of disease [101]. This suggests a prominent role for
genetic testing and counseling in all patients with known or
suspected SCCOHT [105].
The differential diagnosis for poorly differentiated tumors
such as small-cell carcinoma of the ovary, hypercalcemic
type (SCCOHT), is broad, and includes adult or juvenile
granulosa cell tumors [105, 106]. Juvenile granulosa cell
tumors are also characterized by follicle-like spaces with
Fig. 7.13 Small-cell carcinoma of the ovary, hypercalcemic type.
Tumor composed of diffuse sheets and trabeculae, of monotonous small
eosinophilic material, but are not associated with hypercal-
round blue cells with scant cytoplasm and ovoid to round, hyperchro- cemia. Additionally, juvenile granulosa cell tumors are often
matic nuclei more pleomorphic appearing and will be negative for WT-1
222 C. Matrai et al.
and EMA, and positive for α-inhibin, and will have retained Among the reported cases of primary ovarian NSCNEC
SMARCA4 nuclear staining. Other malignancies on the dif- in the literature, age at diagnosis ranges between 18 and
ferential for SCCOHT include desmoplastic small round-cell 77 years of age with an average of 51 years of age [113, 114].
tumor (DSRCT), pulmonary-type small-cell carcinoma of Patients most commonly present with abdominal pain,
the ovary, endometrial stromal sarcoma (ESS), peripheral abdominal distention, ascites, and a palpable abdominal or
neuroectodermal tumor, neuroblastoma, rhabdomyosarcoma pelvic mass [112]. Most cases of primary ovarian NSCNEC
(RMS), and metastatic melanoma, lymphoma, or small-cell are associated with elevated CA-125 and/or CEA [111]. To
carcinoma of the lung [105]. Importantly, all other tumors on date, there are no known molecular mechanisms or genetic
the vast differential for SCCOHT should have retained profiles for primary ovarian NSCNEC.
nuclear positivity for the SMARCA4 antibody [105]. It Grossly, tumors are typically unilateral, partially cystic,
should be noted, however, that there are rare cases in the lit- and large, ranging in size from 5 to 30 cm, with an average
erature of SCCOHT with SMARCA4 gene mutations causing size of 15.3 cm [112]. Microscopically, NSCNEC are com-
retained SMARCA4 protein [101]. posed of intermediate to large atypical cells, with moderate
Clinically, SCCOHT is a highly aggressive neoplasm to abundant eosinophilic or granular cytoplasm, and round to
with a dismal prognosis [107, 109]. The long-term survival oval vesicular nuclei with coarsely stippled chromatin or
rate is 10–20% overall with a mean overall survival of prominent nucleoli, arranged in trabeculae, nests, or solid
14.9 months [109]. Slightly over half of patients are found to sheets [112]. Areas of confluent geographic necrosis and
have extraovarian disease in the abdomen and pelvis at diag- central comedo-like necrosis are common. Mitotic activity is
nosis [107]. The most important prognostic factor is tumor brisk with atypical mitoses present. Immunohistochemistry
stage at diagnosis, with a long-term survival rate of 33% and often reveals tumor cells positive for synaptophysin, chro-
a mean survival of 35.3 months for patients with early-stage mogranin A, neuron-specific enolase (NSE), CD56, EMA,
disease [109]. In the largest clinical analysis of patients to and pancytokeratin [113].
date, 33% of patients with stage IA disease were disease free Primary ovarian NSCNEC must be differentiated from
for an average of 5.7 years (range 1–13 years) postsurgery metastatic neuroendocrine carcinoma, primary or metastatic
[106]. Comparatively, no patients with stage IC disease were carcinoid tumor, small-cell carcinoma of pulmonary or
disease free at follow-up and 90% had died of disease within hypercalcemic type, and other non-neuroendocrine tumors
2 years postsurgery [106]. Factors associated with improved with neuroendocrine differentiation (such as sex cord-
prognosis include older age at diagnosis, normal preopera- stromal or germ cell tumors) [113]. Primary ovarian
tive serum calcium, lack of large-cell variant histology, NSCNEC can be distinguished from metastatic neuroendo-
tumor size less than 10 cm, and administration of adjuvant crine carcinoma as it is almost always unilateral and most
radiotherapy [106, 109]. Current management typically often associated with ovarian surface epithelial tumors, a
involves surgical resection, followed by chemotherapy or clue to its primary ovarian origin. Additionally, CA-125 is
radiotherapy [109]. More recent studies suggest that high- often elevated, pointing to an ovarian origin [111]. Primary
dose chemotherapy with autologous stem cell rescue (HDC- or metastatic carcinoid tumors involving the ovary are distin-
aSCR) offers the best chance at long-term survival [110]. guished by their smaller cell size, low mitotic rate, decreased
cellular atypia, and absence of necrosis. Small-cell carci-
noma of hypercalcemic type can be easily differentiated
7.11.2 Non-small, Neuroendocrine Carcinoma from NSCNEC with several immunostains including an
immunostain for SMARCA4 (see previous section).
Originally described by Collins et al. in 1991, primary ovar- Approximately two-thirds of patients with primary ovar-
ian non-small-cell neuroendocrine carcinoma (NSCNEC) of ian NSCNEC present with stage I disease, although out-
the ovary, also referred to as large-cell neuroendocrine carci- comes are variable [113]. Patients may survive up to 5.5 years
noma (LCNEC, LCNC) or undifferentiated carcinoma of with no evidence of disease or die of disease in as little as
non-small-cell neuroendocrine type, is an exceedingly rare 4 months despite aggressive treatment [112]. The remaining
diagnosis with less than 50 cases reported in the literature third of patients who present with stage III or IV disease
thus far [111]. The vast majority of primary ovarian NSCNEC often die of disease within months, although some have been
are admixed with a smaller component of ovarian surface epi- reported to survive up to 3 years with no evidence of disease
thelial neoplasms (benign, borderline, or malignant) or less [112, 113]. The majority of recent cases are treated with total
commonly a germ cell neoplasm, helping to differentiate abdominal hysterectomy (TAH) and bilateral salpingo-
these primary ovarian neuroendocrine carcinomas from meta- oophorectomy (BSO) with adjuvant chemotherapy [112].
static neuroendocrine carcinoma [112]. Even less commonly, Reported chemotherapy regimens are extremely variable,
NSCNEC may have pure neuroendocrine differentiation with although most documented regimens include platinum-based
no associated epithelial component [113]. chemotherapy [112]. Given the rarity of cases there is no
7 Ovarian Mucinous, Brenner Tumors, and Other Epithelial Tumors 223
Ovarian adenomatoid tumors, much like their counter- [128]. Tumor cells are classically positive for mesothelial
parts in the male and female genital tracts, are often found markers such as calretinin, CK5/6, and WT-1; however, recent
incidentally during autopsy or when the ovaries are removed studies have shown that adenomatoid tumors are most often
for other reasons; however, several cases in the literature positive for CAM5.2, calretinin, D2-40, and WT-1, with only
reportedly presented with symptoms [128]. The age at pre- a small percentage positive for CK5/6 [127].
sentation is wide, ranging from 23 to 79 years, with an aver- Until very recently, there have been no studies revealing a
age of 54 years of age [128]. Importantly, these benign molecular pathogenesis for ovarian adenomatoid tumors. One
mesothelial tumors are not associated with asbestos expo- recent study, however, performed next-generation sequencing
sure, while primary ovarian malignant mesotheliomas may on 31 adenomatoid tumors of the male and female genital
or may not be associated with asbestos exposure [129]. No tracts (7 epididymal tumors, 7 fallopian tube tumors, 17 uter-
clinical biomarkers have been associated with ovarian ade- ine tumors) and found a single heterozygous somatic mis-
nomatoid tumors or other adenomatoid tumors of the female sense mutation in the TRAF7 gene in all tumors [130]. These
genital tract to date [127]. somatic missense mutations were all clustered into one of the
On gross examination, these tumors are unilateral, pre- five locations at the C-terminus of the protein, with the most
dominantly located in the ovarian hilum and often quite common mutation of the female genital tract located at
small (ranging between 1 and 16 mm in size, average of p.S561R followed by p.H521R. A member of the tumor
7 mm), though sizes up to 8 cm have been reported [127, necrosis factor receptor-associated factor (TRAF) family, the
128]. Grossly, these lesions typically appear well circum- TRAF7 gene encodes an E3 ubiquitin ligase known to play a
scribed, solid, white to yellow, and firm [128]. role in regulating nuclear factor-kappa B (NF-kB). This study
Microscopically, adenomatoid tumors are often well cir- further revealed that mutations in TRAF7 drive aberrant acti-
cumscribed, though infiltrative patterns are reported [127, vation of the NF-kB signaling pathway in vitro, leading to
128]. A variety of architectural patterns have been described upregulation of L1 cell adhesion molecule (L1CAM), a
in the literature (adenoid, angiomatoid, glandular, microcys- known transcriptional target of NF-kB. While this study did
tic, trabecular, oncocytic, and solid), but all are distinctive for not include an ovarian adenomatoid tumor, the results suggest
oval, elongated, or slit-like tubules lined by flat to columnar that somatic missense mutations in TRAF7 may be a geneti-
cells with threadlike bridging strands that cross these tubular cally defining event for all adenomatoid tumors of the male
spaces [128] (Fig. 7.15). Tubules may be focally cystically and female genital tracts [130].
dilated and, while often empty, rarely may contain basophilic Ovarian adenomatoid tumors, due to their rarity, may
fluid [128]. Stroma may be scant to moderate. Tumor cells cause more diagnostic confusion than their tubal and uterine
often contain abundant dense eosinophilic cytoplasm to pale counterparts [128]. Architectural variants, particularly adeno-
vacuolated cytoplasm with bland nuclei, which lack cytologic matoid tumors with hyaline bodies, can mimic the reticular or
atypia and mitotic activity. Tumor cells often at least focally liposarcoma-like pattern of yolk sac tumor. While hyaline
contain small to large intracytoplasmic vacuoles, mimicking bodies are a characteristic finding in yolk sac tumors they are
signet-ring cells or lipoblast-like cells [127, 128]. Rarely, nonspecific and can occur in a variety of ovarian tumors
cells may contain PAS-positive eosinophilic hyaline bodies [128]. Adenomatoid tumors can be differentiated from yolk
sac tumors by their bland nuclei lacking primitive cytology.
The multicystic pattern of adenomatoid tumor can mimic a
multilocular peritoneal inclusion cyst (benign cystic mesothe-
lioma), though this would occur as an ill-defined collection of
cysts on the ovarian serosa rather than a well-circumscribed
mass in the ovarian hilum. The vacuolar pattern of adenoma-
toid tumor could mimic foci of a female adnexal tumor of
Wolffian origin (FATWO), although adenomatoid-like areas
of FATWOs are typically admixed with other architectural
patterns, allowing for distinction between the two.
Additionally, epithelioid hemangiomas and lymphangiomas
may be on the differential, though these could be distin-
guished with immunohistochemical stains [127, 128].
Adenomatoid tumors of the ovary, as well as those of the
male and female genital tract, display a benign clinical
course requiring no additional therapy after resection [130].
In fact, the majority of patients with ovarian adenomatoid
Fig. 7.15 Adenomatoid tumor tumors are found to harbor these neoplasms following resec-
7 Ovarian Mucinous, Brenner Tumors, and Other Epithelial Tumors 225
tion of the ovary for other reasons. Several recent studies carcinoma [137]. While there are no specific biomarkers
have revealed that long-term follow-up (on average 8 years) associated with DSRCT, serum CA-125 levels are often ele-
for adenomatoid tumors of the female genital tract, including vated in primary ovarian DSRCT [140]. Elevated inhibin has
those in the ovary, reveals no evidence of recurrence follow- been reported in rare cases [137]. Imaging and laparoscopy
ing resection [127]. typically reveal widespread intraperitoneal involvement with
one to several large masses and numerous smaller peritoneal
tumor deposits [136]. Ovarian involvement may be unilateral
7.11.6 Ovarian Mesothelioma or bilateral (approximately 50% are bilateral), and the most
common metastatic sites are liver, lymph nodes, lung, and
Mesothelioma arising within the peritoneal cavity to involve bone marrow [140].
the ovary and mimic a primary ovarian neoplasm is rare, but Gross examination reveals tumors to be large, with an
has been reported and usually affects middle-age and elderly average size of 11 cm (range 5–20 cm) [137]. Multiple large
patients [129, 131, 132]. Rare mesotheliomas confined to the tumor nodules are often present with multifocal hemorrhage
ovary have also been reported [129]. Morphologically meso- and necrosis. Microscopic examination reveals nests or
theliomas involving the ovary tend to be of the epithelioid sheets of monotonous, round to ovoid, small round blue cells
type, but some can contain a sarcomatous component [129]. with scant cytoplasm in an abundant desmoplastic stroma
Mesothelioma should be considered in the differential diag- [141]. Architectural variants include rosette-like, tubular,
nosis of a variety of ovarian neoplasms including low-grade glandular, or trabecular architecture [142]. Nuclei are hyper-
serous carcinoma or borderline serous tumors with implants. chromatic with inconspicuous nucleoli, brisk mitotic activ-
Of utmost importance is the consideration of the distribution ity, and focal necrosis. Additionally, tumor cells may be
of the lesion and immunohistochemical profile. markedly pleomorphic, appear spindled, contain abundant
Recommended immunohistochemical profiles to evaluate eosinophilic cytoplasm (rhabdoid appearance), or exhibit
for mesothelioma involve a variety of markers. Calretinin clear-cell or signet-ring features [141, 142].
and D2-40 are especially recommended given that they are DSRCT is classically defined by co-expression of epithe-
usually negative in ovarian epithelial neoplasms, while posi- lial, mesenchymal, and neural immunohistochemical mark-
tive in mesothelioma. While WT-1 is often used to evaluate ers; however, no single stain is specific or sensitive for
for mesothelioma in the pleural cavity, WT-1 is unhelpful DSRCT and thus immunostains must be interpreted with
when considering an ovarian neoplasm as WT-1 is positive in caution. Characteristically positive immunostains include
the majority of ovarian epithelial primaries. Additional stains cytokeratins (CAM 5.2, AE1–3, PanCK, but not CK5/6 or
to consider include PAX8, estrogen, and progesterone recep- CK20), EMA (membranous staining pattern), desmin (peri-
tors, which are usually negative in mesothelioma; however, nuclear dot-like staining pattern), and neuron-specific eno-
one must interpret these stains with caution given recent lase (NSE) [142]. Positive immunostaining for WT-1 (with
reports of positivity in rare cases of mesothelioma [133, an antibody directed against the C terminus) is seen in over
134]. A recent study has described the usefulness of the two-thirds of cases and INI1 expression is typically retained
marker BAP-1 in the differential diagnosis of mesothelioma [142]. Few cases may exhibit positive staining for CD99,
and serous carcinoma as loss of BAP-1 staining strongly smooth muscle actin (SMA), muscle-specific actin, chromo-
supports the diagnosis of mesothelioma [135]. granin, synaptophysin, CD56, neurofilament, and S100
[142]. Immunostains for myogenin, MyoD1, myoglobin,
glial fibrillary acidic protein (GFAP), and HMB-45 are clas-
7.11.7 Desmoplastic Small Round-Cell Tumor sically negative [142].
DSRCTs are genetically characterized by a t(11;22)
Desmoplastic small round-cell tumor (DSRCT) is a rare, (p13;q12) gene translocation leading to an EWS-WT1 gene
highly aggressive, intra-abdominal neoplasm with a strong fusion product [142]. This translocation fuses the Ewing sar-
male predominance (male-to-female ratio of 4:1) [136]. Of coma (EWS, also referred to as EWSR1) gene with the Wilms’
the approximately 850 cases described thus far in the litera- tumor-suppressor (WT1) gene, creating a functional chimeric
ture, less than 20 have been reported to arise in the ovary and fusion protein that is thought to act as a transcriptional acti-
involve the peritoneal cavity of adolescent women [137, vator on several downstream targets. Fluorescent in situ
138]. Primary ovarian DSRCT presents in the second and hybridization (FISH) with break apart probes for EWS is
third decades of life, with a reported age at presentation diagnostic of DSRCT; however, an EWS-WT1 gene fusion is
ranging from 6 to 31 years of age [137, 139]. not always present, and EWS gene rearrangements are pos-
Patients typically present with abdominal distention, sible [142].
intractable abdominal and pelvic pain, nausea, vomiting, Making a definitive diagnosis of DSRCT based on mor-
constipation, and weight loss, mimicking epithelial ovarian phologic features alone is not typically possible given the
226 C. Matrai et al.
expansive differential diagnosis for “small round blue cell” there is no known molecular mechanism characterizing
tumors. This differential includes undifferentiated or poorly FATWOs to date [143].
differentiated carcinoma, small-cell carcinoma of hypercal- On gross examination, FATWOs are unilateral, well cir-
cemic type, immature teratoma, poorly differentiated Sertoli- cumscribed, or encapsulated, solid to partially cystic masses
Leydig cell tumor, juvenile granulosa cell tumors, ranging from 0.8 to 25 cm in greatest diameter [145]. The cut
neuroblastoma, Ewing sarcoma, rhabdomyosarcoma, primi- surface is typically rubbery, tan-yellow to gray, and focally
tive neuroectodermal tumor, lymphoma, neuroblastoma, and hemorrhagic or necrotic. They occur throughout the broad
hepatoblastoma [141, 142]. ligament, often sparing the ovaries, but can rarely present as
The reported life expectancy after a diagnosis of ovarian ovarian tumors [145]. Rare bilateral ovarian FATWOs are
DSRCT ranges from 4 to 42 months [140]. The current stan- documented in the literature [146].
dard of care is aggressive surgical resection and adjuvant Histologically, FATWOs present with a variety of archi-
high-dose chemotherapy, followed by either whole abdomi- tectural patterns mimicking both surface epithelial tumors
nopelvic intensity-modulated radiation therapy or hyperther- and sex cord-stromal tumors. FATWOs are composed of epi-
mic intraperitoneal chemotherapy (HIPEC) [142]. Currently, thelial cells growing in tubular, sievelike (retiform), multi-
the most widely accepted adjuvant chemotherapy for DRSCT cystic, or diffuse (solid) patterns with variable amounts of
is the “P6 protocol,” which consists of four cycles of cyclo- fibrous stroma, often with a single architectural pattern pre-
phosphamide, doxorubicin, and vincristine (CDV), with dominating (Fig. 7.16) [144, 145]. The tubular pattern of
alternating ifosfamide and etoposide in between cycles architecture is characterized by closely packed hollow
[137]. Patients with chemoresponsive disease and optimal tubules with compressed lumens lined by cuboidal to colum-
surgical cytoreduction appear to have the best long-term out- nar epithelial cells. The sievelike (retiform) pattern is charac-
comes and improved survival. Overall, despite often respond- terized by solid areas separated by various-sized cysts filled
ing well to chemotherapy, tumors recur rapidly and long-term with eosinophilic secretions and lined by low cuboidal, occa-
survival is dismal [140]. sionally hobnailed, epithelial cells. The diffuse (solid) pat-
tern is characterized by sheets of oval to spindle-shaped cells
with scant eosinophilic cytoplasm. Tumor cells are typically
7.11.8 Female Adnexal Tumors of Wolffian cytologically bland, with fine even chromatin, inconspicuous
Origin nucleoli, and low mitotic activity (<1/10 high-power fields)
[144, 145].
First reported in 1973 by Kariminejad and Scully, female Immunohistochemistry can be confusing due to lack of
adnexal tumors of probable Wolffian origin (FATWO) are specificity or sensitivity of any one immunostain, including
rare epithelial ovarian tumors with fewer than 100 cases doc- those that have been shown to be positive in mesonephric
umented in the literature thus far [143]. Previously called remnants [126, 144]. More recently, studies have shown
“Wolffian adnexal tumor” and “retiform Wolffian adenoma,” PAX8 to be negative in virtually all cases of FATWO, which
FATWOs are thought to be derived from mesonephric
(Wolffian) duct remnants [144]. During male embryogene-
sis, mesonephric ducts and tubules form the vas deferens and
other male reproductive structures. In female embryogene-
sis, the mesonephric ducts and tubules degenerate, leaving
remnant mesonephric inclusions throughout the broad liga-
ment, extending from the hilum of the ovary along the meso-
salpinx to the lateral uterus and lateral walls of the vagina
(Gartner’s ducts) [145]. FATWOs exclusively arise in these
locations and share histologic and immunologic characteris-
tics with mesonephric remnants, leading to the hypothesis
that these tumors are of mesonephric (Wolffian) origin [144].
FATWOs either present with abdominal swelling and
pain, pelvic pain, and vaginal bleeding or are incidentally
discovered during pelvic examination or laparotomy in
approximately half of patients [145]. Age at presentation
ranges from 15 to 87 years of age, with a median age of
50 years old [145, 146]. No biomarkers are reported to be
Fig. 7.16 Female adnexal tumor of Wolffian origin composed of epi-
associated with FATWOs, and elevated serum CA-125 has thelial cells growing in tubular, sievelike (retiform), and diffuse (solid)
not been documented in the literature [145]. In addition, patterns with a minimal amount of intervening stroma
7 Ovarian Mucinous, Brenner Tumors, and Other Epithelial Tumors 227
may emerge as a useful immunohistochemical stain for diag- 3. Kurman RJ, Shih Ie M. The dualistic model of ovarian car-
cinogenesis: revisited, revised, and expanded. Am J Pathol.
nosing ovarian FATWOs [126]. 2016;186(4):733–47.
The differential diagnosis for FATWOs is exceedingly 4. Hauptmann S, et al. Ovarian borderline tumors in the 2014 WHO
broad, given the variety of histological presentations [145]. classification: evolving concepts and diagnostic criteria. Virchows
FATWOs are most often confused with sex cord-stromal Arch. 2017;470(2):125–42.
5. Hart WR. Mucinous tumors of the ovary: a review. Int J Gynecol
tumors such as Sertoli cell and Sertoli-Leydig cell tumors Pathol. 2005;24(1):4–25.
due to the tubular pattern of architecture. Well-differentiated 6. Milea A, et al. Retinoblastoma pathway deregulatory mechanisms
endometrioid adenocarcinoma and clear-cell carcinoma are determine clinical outcome in high-grade serous ovarian carci-
also often on the differential due to the occasionally glandu- noma. Mod Pathol. 2014;27(7):991–1001.
7. Hart WR, Norris HJ. Borderline and malignant mucinous tumors
lar appearance of FATWO with a predominantly tubular pat- of the ovary. Histologic criteria and clinical behavior. Cancer.
tern of large tubules. However, lack of intraluminal mucin 1973;31(5):1031–45.
argues against this diagnosis. Cysts lined with flattened 8. Han G, Soslow RA. Nonserous ovarian epithelial tumors. Surg
hobnail-like cells may suggest clear-cell carcinoma; how- Pathol Clin. 2011;4(1):397–459.
9. Tait DL, Miller DS. Multicystic ovarian tumor weighing 156 lbs:
ever, no papillary areas or intraluminal mucin also argue the second largest tumor in Texas. Tex Med. 1997;93(1):89–91.
against this. The location of the tumor is often the biggest 10. Seidman JD, Khedmati F. Exploring the histogenesis of ovarian
clue, but it is important to note that these tumors can occur mucinous and transitional cell (Brenner) neoplasms and their
within the ovary, and thus careful morphologic diagnosis is relationship with Walthard cell nests: a study of 120 tumors. Arch
Pathol Lab Med. 2008;132(11):1753–60.
crucial [144, 145]. 11. Czernobilsky B, Borenstein R, Lancet M. Cystadenofibroma of
The majority of FATWOs behave benignly with a reported the ovary. A clinicopathologic study of 34 cases and comparison
rate of recurrence between 10 and 20%, suggesting that these with serous cystadenoma. Cancer. 1974;34(6):1971–81.
tumors are of low malignant potential rather than uniformly 12. Kao GF, Norris HJ. Unusual cystadenofibromas: endometrioid,
mucinous, and clear cell types. Obstet Gynecol. 1979;54(6):729–36.
benign [145, 146]. It should be noted, however, that recur- 13. Randrianjafisamindrakotroka NS, Gasser B, Philippe E. [The
rences often occur in patients not initially treated with total malignant potential of adenofibroma and cystadenofibroma of
abdominal hysterectomy and bilateral salpingo-oophorectomy the ovary and mesovarium. 118 cases including 13 proliferative
[145]. Tumors are reported to recur within 13–96 months fol- and 5 carcinomatous]. J Gynecol Obstet Biol Reprod (Paris).
1993;22(1):33–8.
lowing resection (median of 48 months), and are often not 14. Lee DH. A case of mucinous cystadenofibroma of the ovary. Case
responsive to chemotherapy or radiation [145]. Malignancy is Rep Obstet Gynecol. 2014;2014:130530.
currently determined by clinical behavior (recurrence or 15. Roma AA, Masand RP. Different staining patterns of ovarian
metastases) rather than by pathological features, although Brenner tumor and the associated mucinous tumor. Ann Diagn
Pathol. 2015;19(1):29–32.
hypercellularity, cellular pleomorphism, and increased 16. Kondi-Pafiti A, et al. Clinicopathological features and immu-
mitotic activity are associated with increased risk for recur- noprofile of 30 cases of Brenner ovarian tumors. Arch Gynecol
rence [145, 146]. Surgical resection is recommended for both Obstet. 2012;285(6):1699–702.
primary and recurrent FATWOs, with a questionable role for 17. Rambau PF, et al. Morphologic reproducibility, genotyping,
and Immunohistochemical profiling do not support a category
adjuvant chemotherapy or radiotherapy [146]. of seromucinous carcinoma of the ovary. Am J Surg Pathol.
2017;41(5):685–95.
18. Li XS, Sun J, He XL. Expression of c-myc and mutation of the
7.12 Conclusion KRAS gene in patients with ovarian mucinous tumors. Genet Mol
Res. 2015;14(3):10752–9.
19. Auner V, et al. KRAS mutation analysis in ovarian samples using
In this chapter we covered a wide variety of ovarian lesions. a high sensitivity biochip assay. BMC Cancer. 2009;9:111.
While the majority of the chapter focused on mucinous neo- 20. Lee YJ, et al. Multipoint Kras oncogene mutations potentially
plasms of the ovary including mucinous cystadenomas, indicate mucinous carcinoma on the entire spectrum of mucinous
ovarian neoplasms. Oncotarget. 2016;7(50):82097–103.
mucinous borderline tumors, and mucinous carcinomas, we 21. Robert J. Kurman, Carcangiu, M.L., C. Simon Herrington, Robert
also discussed Brenner tumors and provided detailed infor- H. Young, WHO classification of tumours of female reproductive
mation on a wide-range of rare ovarian neoplasms, with organs. 4th ed. 2014, Lyon: IARC Press.
which every pathologist should be familiar. 22. Kamel RM. A massive ovarian mucinous cystadenoma: a case
report. Reprod Biol Endocrinol. 2010;8:24.
23. Mizrachi Y, et al. Intraoperative rupture of benign mucinous
cystadenoma does not increase its recurrence rate. Arch Gynecol
References Obstet. 2015;291(5):1135–9.
24. Ben-Ami I, et al. Does intraoperative spillage of benign ovarian
1. Chen VW, et al. Pathology and classification of ovarian tumors. mucinous cystadenoma increase its recurrence rate? Am J Obstet
Cancer. 2003;97(10 Suppl):2631–42. Gynecol. 2010;202(2):142 e1–5.
2. Brown J, Frumovitz M. Mucinous tumors of the ovary: cur- 25. Chao A, et al. Abdominal compartment syndrome secondary to
rent thoughts on diagnosis and management. Curr Oncol Rep. ovarian mucinous cystadenoma. Obstet Gynecol. 2004;104(5 Pt
2014;16(6):389. 2):1180–2.
228 C. Matrai et al.
26. Boger-Megiddo I, Weiss NS. Histologic subtypes and later- 45. Panarelli NC, Yantiss RK. Mucinous neoplasms of the appendix
ality of primary epithelial ovarian tumors. Gynecol Oncol. and peritoneum. Arch Pathol Lab Med. 2011;135(10):1261–8.
2005;97(1):80–3. 46. Ramaswamy V. Pathology of mucinous appendiceal tumors and
27. Lee KR, Scully RE. Mucinous tumors of the ovary: a clinico- pseudomyxoma peritonei. Indian J Surg Oncol. 2016;7(2):258–67.
pathologic study of 196 borderline tumors (of intestinal type) and 47. Young RH, Gilks CB, Scully RE. Mucinous tumors of the
carcinomas, including an evaluation of 11 cases with ‘pseudo- appendix associated with mucinous tumors of the ovary and
myxoma peritonei’. Am J Surg Pathol. 2000;24(11):1447–64. pseudomyxoma peritonei. A clinicopathological analysis of 22
28. Riopel MA, Ronnett BM, Kurman RJ. Evaluation of diagnostic cases supporting an origin in the appendix. Am J Surg Pathol.
criteria and behavior of ovarian intestinal-type mucinous tumors: 1991;15(5):415–29.
atypical proliferative (borderline) tumors and intraepithelial, 48. Ronnett BM, et al. Disseminated peritoneal adenomucinosis and
microinvasive, invasive, and metastatic carcinomas. Am J Surg peritoneal mucinous carcinomatosis. A clinicopathologic analysis
Pathol. 1999;23(6):617–35. of 109 cases with emphasis on distinguishing pathologic features,
29. Hart WR. Borderline epithelial tumors of the ovary. Mod Pathol. site of origin, prognosis, and relationship to “pseudomyxoma peri-
2005;18(Suppl 2):S33–50. tonei”. Am J Surg Pathol. 1995;19(12):1390–408.
30. Kurman RJ, Shih Ie M. Seromucinous tumors of the ovary. What’s 49. Cuatrecasas M, Matias-Guiu X, Prat J. Synchronous mucinous
in a name? Int J Gynecol Pathol. 2016;35(1):78–81. tumors of the appendix and the ovary associated with pseudo-
31. Vang R, et al. Immunohistochemical expression of CDX2 in myxoma peritonei. A clinicopathologic study of six cases with
primary ovarian mucinous tumors and metastatic mucinous comparative analysis of c-Ki-ras mutations. Am J Surg Pathol.
carcinomas involving the ovary: comparison with CK20 and 1996;20(6):739–46.
correlation with coordinate expression of CK7. Mod Pathol. 50. Ronnett BM, Seidman JD. Mucinous tumors arising in ovarian
2006;19(11):1421–8. mature cystic teratomas: relationship to the clinical syndrome of
32. Mackenzie R, et al. Targeted deep sequencing of mucinous ovar- pseudomyxoma peritonei. Am J Surg Pathol. 2003;27(5):650–7.
ian tumors reveals multiple overlapping RAS-pathway activating 51. Prat J, Scully RE. Ovarian mucinous tumors with sarcoma-
mutations in borderline and cancerous neoplasms. BMC Cancer. like mural nodules: a report of seven cases. Cancer.
2015;15:415. 1979;44(4):1332–44.
33. Cuatrecasas M, et al. K-ras mutations in mucinous ovarian tumors: 52. Prat J, Young RH, Scully RE. Ovarian mucinous tumors with foci
a clinicopathologic and molecular study of 95 cases. Cancer. of anaplastic carcinoma. Cancer. 1982;50(2):300–4.
1997;79(8):1581–6. 53. Provenza C, Young RH, Prat J. Anaplastic carcinoma in mucinous
34. Ryland GL, et al. Mutational landscape of mucinous ovarian car- ovarian tumors: a clinicopathologic study of 34 cases emphasizing
cinoma and its neoplastic precursors. Genome Med. 2015;7(1):87. the crucial impact of stage on prognosis, their histologic spectrum,
35. Ronnett BM, et al. Mucinous borderline ovarian tumors: points and overlap with sarcomalike mural nodules. Am J Surg Pathol.
of general agreement and persistent controversies regarding 2008;32(3):383–9.
nomenclature, diagnostic criteria, and behavior. Hum Pathol. 54. McFarland M, et al. Osteosarcoma as malignant mural nodule in
2004;35(8):949–60. ovarian mucinous neoplasms of intestinal type: report of 2 cases.
36. De Decker K, et al. Staging procedures in patients with mucinous Int J Gynecol Pathol. 2015;34(4):369–73.
borderline tumors of the ovary do not reveal peritoneal or omental 55. Balaton A, et al. Primary leiomyosarcoma of the ovary: a his-
disease. Gynecol Oncol. 2017;144(2):285–9. tological and immunocytochemical study. Gynecol Oncol.
37. Dewilde K, et al. Staging with unilateral Salpingo-oophorectomy 1987;28(1):116–20.
and expert pathological review result in no recurrences in a 56. WHO classification of tumours of femal reproductive organs. 4th
series of 81 intestinal-type mucinous borderline ovarian tumors. ed. Lyon: International Agency for Research on Cancer.
Gynecol Obstet Investig. 2018;83(1):65–9. 57. Ledermann JA, et al. Gynecologic Cancer InterGroup (GCIG)
38. Gungorduk K, et al. The impact of surgical staging on the prog- consensus review for mucinous ovarian carcinoma. Int J Gynecol
nosis of mucinous borderline tumors of the ovaries: a multicenter Cancer. 2014;24(9 Suppl 3):S14–9.
study. Anticancer Res. 2017;37(10):5609–16. 58. Zaino RJ, et al. Advanced stage mucinous adenocarcinoma of
39. Uzan C, et al. Influence of histological subtypes on the risk of an the ovary is both rare and highly lethal: a Gynecologic Oncology
invasive recurrence in a large series of stage I borderline ovar- Group study. Cancer. 2011;117(3):554–62.
ian tumor including 191 conservative treatments. Ann Oncol. 59. Yemelyanova AV, et al. Distinction of primary and metastatic
2014;25(7):1312–9. mucinous tumors involving the ovary: analysis of size and later-
40. Khunamornpong S, et al. Mucinous tumor of low malignant poten- ality data by primary site with reevaluation of an algorithm for
tial (“borderline” or “atypical proliferative” tumor) of the ovary: a tumor classification. Am J Surg Pathol. 2008;32(1):128–38.
study of 171 cases with the assessment of intraepithelial carcinoma 60. Chen S, et al. Invasion patterns in stage I endometrioid and muci-
and microinvasion. Int J Gynecol Pathol. 2011;30(3):218–30. nous ovarian carcinomas: a clinicopathologic analysis empha-
41. Vasconcelos I, Darb-Esfahani S, Sehouli J. Serous and mucinous sizing favorable outcomes in carcinomas without destructive
borderline ovarian tumours: differences in clinical presentation, stromal invasion and the occasional malignant course of carci-
high-risk histopathological features, and lethal recurrence rates. nomas with limited destructive stromal invasion. Mod Pathol.
BJOG. 2016;123(4):498–508. 2005;18(7):903–11.
42. Khunamornpong S, Russell P, Dalrymple JC. Proliferating (LMP) 61. Santotoribio JD, et al. Serum lactate dehydrogenase in combina-
mucinous tumors of the ovaries with microinvasion: morphologic tion with free-to-total serum prostate specific antigen ratio for
assessment of 13 cases. Int J Gynecol Pathol. 1999;18(3):238–46. diagnosis of prostate cancer. Clin Lab. 2014;60(6):1055–8.
43. Kim KR, et al. Is stromal microinvasion in primary mucinous 62. Cho HY, Kyung MS. Serum CA19-9 as a predictor of malignancy
ovarian tumors with “mucin granuloma” true invasion? Am J Surg in primary ovarian mucinous tumors: a matched case-control
Pathol. 2007;31(4):546–54. study. Med Sci Monit. 2014;20:1334–9.
44. Rouzbahman M, Chetty R. Mucinous tumours of appendix and 63. Kelemen LE, Kobel M. Mucinous carcinomas of the ovary and
ovary: an overview and evaluation of current practice. J Clin colorectum: different organ, same dilemma. Lancet Oncol.
Pathol. 2014;67(3):193–7. 2011;12(11):1071–80.
7 Ovarian Mucinous, Brenner Tumors, and Other Epithelial Tumors 229
64. Anglesio MS, et al. Molecular characterization of mucinous ovar- of 12q14-21 by comparative genomic hybridization. Gynecol
ian tumours supports a stratified treatment approach with HER2 Oncol. 1999;74(1):134–7.
targeting in 19% of carcinomas. J Pathol. 2013;229(1):111–20. 85. Ogawa K, Johansson SL, Cohen SM. Immunohistochemical anal-
65. Seidman JD, Kurman RJ, Ronnett BM. Primary and metastatic ysis of uroplakins, urothelial specific proteins, in ovarian Brenner
mucinous adenocarcinomas in the ovaries: incidence in routine tumors, normal tissues, and benign and neoplastic lesions of the
practice with a new approach to improve intraoperative diagnosis. female genital tract. Am J Pathol. 1999;155(4):1047–50.
Am J Surg Pathol. 2003;27(7):985–93. 86. Egan AJ, Russell P. Transitional (urothelial) cell metaplasia of the
66. Hu A, et al. Differentiating primary and extragenital metastatic fallopian tube mucosa: morphological assessment of three cases.
mucinous ovarian tumours: an algorithm combining PAX8 with Int J Gynecol Pathol. 1996;15(1):72–6.
tumour size and laterality. J Clin Pathol. 2015;68(7):522–8. 87. Rabban JT, et al. Transitional cell metaplasia of fallopian tube fim-
67. Perez Montiel D, et al. The value of SATB2 in the differential briae: a potential mimic of early tubal carcinoma in risk reduction
diagnosis of intestinal-type mucinous tumors of the ovary: pri- salpingo-oophorectomies from women with BRCA mutations.
mary vs metastatic. Ann Diagn Pathol. 2015;19(4):249–52. Am J Surg Pathol. 2009;33(1):111–9.
68. Meriden Z, et al. Ovarian metastases of pancreaticobiliary tract 88. Chia CC, Huang SC. A borderline ovarian Brenner tumor mimics
adenocarcinomas: analysis of 35 cases, with emphasis on the abil- uterine fibroids. Taiwan J Obstet Gynecol. 2011;50(1):103–5.
ity of metastases to simulate primary ovarian mucinous tumors. 89. Kuhn E, et al. The pathogenesis of atypical proliferative Brenner
Am J Surg Pathol. 2011;35(2):276–88. tumor: an immunohistochemical and molecular genetic analysis.
69. Vang R, et al. Immunohistochemistry for estrogen and progester- Mod Pathol. 2014;27(2):231–7.
one receptors in the distinction of primary and metastatic muci- 90. De Cecio R, et al. Borderline Brenner tumor of the ovary: a case
nous tumors in the ovary: an analysis of 124 cases. Mod Pathol. report with immunohistochemical and molecular study. J Ovarian
2006;19(1):97–105. Res. 2014;7:101.
70. McCluggage WG. Immunohistochemistry in the distinction 91. Movahedi-Lankarani S, Kurman RJ. Calretinin, a more sensitive
between primary and metastatic ovarian mucinous neoplasms. J but less specific marker than alpha-inhibin for ovarian sex cord-
Clin Pathol. 2012;65(7):596–600. stromal neoplasms: an immunohistochemical study of 215 cases.
71. Arnogiannaki N, et al. Proliferative Brenner tumor of the ovary. Am J Surg Pathol. 2002;26(11):1477–83.
Clinicopathological study of two cases and review of the litera- 92. Gordon MD, et al. CD99, keratin, and vimentin staining of sex
ture. Eur J Gynaecol Oncol. 2011;32(5):576–8. cord-stromal tumors, normal ovary, and testis. Mod Pathol.
72. Fox H, Agrawal K, Langley FA. The Brenner tumour of the ovary. 1998;11(8):769–73.
A clinicopathological study of 54 cases. J Obstet Gynaecol Br 93. Nasioudis D, et al. Malignant Brenner tumors of the ovary; a
Commonw. 1972;79(7):661–5. population-based analysis. Gynecol Oncol. 2016;142(1):44–9.
73. Kuhn E, et al. Ovarian Brenner tumour: a morphologic and immu- 94. Austin RM, Norris HJ. Malignant Brenner tumor and transitional
nohistochemical analysis suggesting an origin from fallopian tube cell carcinoma of the ovary: a comparison. Int J Gynecol Pathol.
epithelium. Eur J Cancer. 2013;49(18):3839–49. 1987;6(1):29–39.
74. Seidman JD, et al. The fallopian tube-peritoneal junction: a 95. Badve S, et al. Malignant Brenner tumor mimicking a pri-
potential site of carcinogenesis. Int J Gynecol Pathol. 2011; mary squamous cell carcinoma of the cervix. Gynecol Oncol.
30(1):4–11. 1999;74(3):487–90.
75. Ali RH, et al. Transitional cell carcinoma of the ovary is related 96. Vemavarapu L, Alatassi H, Moghadamfalahi M. Unusual presen-
to high-grade serous carcinoma and is distinct from malignant tation of benign cystic brenner tumor with exuberant psammoma-
Brenner tumor. Int J Gynecol Pathol. 2012;31(6):499–506. tous calcifications. Int J Surg Pathol. 2011;19(1):120–2.
76. Ruggiero S, et al. A singular observation of a giant benign Brenner 97. Roth LM, Goheen MP, Broshears JR. Malignant Brenner tumor of
tumor of the ovary. Arch Gynecol Obstet. 2011;284(2):513–6. the ovary with transformation to trabecular carcinoid: an immu-
77. Burg J, et al. Mature cystic teratoma of the ovary with struma and nocytochemical and electron microscopic study. Int J Gynecol
benign Brenner tumor: a case report with immunohistochemical Pathol. 2012;31(1):91–7.
characterization. Int J Gynecol Pathol. 2002;21(1):74–7. 98. Srivastava H, et al. Pure primary squamous cell carcinoma of
78. Takeuchi K, et al. Coexistence of brenner tumor and struma ovarii: ovary—a rare case report. J Clin Diagn Res. 2017;11(5):QD01–2.
case report. Eur J Gynaecol Oncol. 2005;26(1):109–10. 99. Pins MR, et al. Primary squamous cell carcinoma of the ovary.
79. Yoshida M, et al. Coexisting Brenner tumor and struma ovarii in Report of 37 cases. Am J Surg Pathol. 1996;20(7):823–33.
the right ovary: case report and review of the literature. Pathol Int. 100. Park JW, Bae JW. Pure primary ovarian squamous cell carci-
2004;54(10):793–7. noma: a case report and review of the literature. Oncol Lett.
80. Tafe LJ, et al. Molecular genetic analysis of ovarian Brenner 2015;9(1):321–3.
Tumors and associated mucinous epithelial neoplasms. High 101. Witkowski L, et al. Germline and somatic SMARCA4 mutations
variant concordance and identification of mutually exclusive characterize small cell carcinoma of the ovary, hypercalcemic
RAS driver mutations and MYC amplification. Am J Pathol. type. Nat Genet. 2014;46(5):438.
2016;186(3):671–7. 102. Aguirre P, Thor AD, Scully RE. Ovarian small cell carcinoma.
81. Wang Y, et al. Clonality analysis of combined Brenner and muci- Histogenetic considerations based on immunohistochemical and
nous tumours of the ovary reveals their monoclonal origin. J other findings. Am J Clin Pathol. 1989;92(2):140–9.
Pathol. 2015;237(2):146–51. 103. McCluggage W, et al. An immunohistochemical analysis of ovar-
82. Santini D, et al. Brenner tumor of the ovary: a correlative his- ian small cell carcinoma of hypercalcemic type. Int J Gynecol
tologic, histochemical, immunohistochemical, and ultrastructural Pathol. 2004;23(4):330–6.
investigation. Hum Pathol. 1989;20(8):787–95. 104. McCluggage WG, et al. Clinical, morphological and immuno-
83. Cuatrecasas M, et al. K-ras mutations in nonmucinous ovarian histochemical evidence that small-cell carcinoma of the ovary of
epithelial tumors: a molecular analysis and clinicopathologic hypercalcaemic type (SCCOHT) may be a primitive germ-cell
study of 144 patients. Cancer. 1998;82(6):1088–95. neoplasm. Histopathology. 2017;70(7):1147–54.
84. Pejovic T, et al. Well-differentiated mucinous carcinoma of the 105. Witkowski L, et al. Small-cell carcinoma of the ovary of hypercal-
ovary and a coexisting Brenner tumor both exhibit amplification cemic type (malignant rhabdoid tumor of the ovary): a review with
230 C. Matrai et al.
recent developments on pathogenesis. 2016: United States. Surg 127. Sangoi AR, et al. Adenomatoid tumors of the female and male
Pathol Clin. 2016;9(2):215–26. genital tracts: a clinicopathological and immunohistochemical
106. Young RH, Oliva E, Scully RE. Small-cell carcinoma of the ovary, study of 44 cases. Mod Pathol. 2009;22(9):1228–35.
hypercalcaemic type. A clinicopathological analysis of 150 cases. 128. Young R, Silva E, Scully R. Ovarian and Juxtaovarian
Am J Surg Pathol. 1994;18(11):1102–16. Adenomatoid Tumors: a report of six cases. Int J Gynecol Pathol.
107. Estel R, et al. Small cell carcinoma of the ovary of the hypercal- 1991;10(4):364–71.
caemic type: an analysis of clinical and prognostic aspects of a 129. Clement PB, Young RH, Scully RE. Malignant mesotheliomas
rare disease on the basis of cases published in the literature. Arch presenting as ovarian masses. A report of nine cases, includ-
Gynecol Obstet. 2011;284:1277–82. ing two primary ovarian mesotheliomas. Am J Surg Pathol.
108. Lavrut PM, et al. Small cell carcinoma of the ovary, hyper- 1996;20(9):1067–80.
calcemic type: report of a bilateral case in a teenager associ- 130. Goode B, et al. Adenomatoid tumors of the male and female
ated with SMARCA4 germline mutation. Pediatr Dev Pathol. genital tract are defined by TRAF7 mutations that drive aberrant
2016;19(1):56–60. NF-kB pathway activation. Mod Pathol. 2018;31(4):660–73.
109. Callegaro-Filho D, et al. Small cell carcinoma of the ovary- 131. Tangjitgamol S, et al. Huge peritoneal malignant mesothelioma
hypercalcemic type (SCCOHT): a review of 47 cases. Gynecol mimicking primary ovarian carcinoma. J Med Assoc Thail.
Oncol. 2016;140(1):53–7. 2013;96(1):107–11.
110. Witkowski L, et al. The influence of clinical and genetic factors on 132. Kapoor R, et al. A case report of peritoneal malignant mesothe-
patient outcome in small cell carcinoma of the ovary, hypercalce- lioma presenting as primary ovarian mass. J Cancer Res Ther.
mic type. Gynecol Oncol. 2016;141(3):454–60. 2015;11(3):654.
111. Agarwal L, Gupta B, Jain A. Pure large cell neuroendocrine carci- 133. Tandon R, et al. Immunohistochemistry in peritoneal mesothe-
noma of the ovary with metastasis to cervix: a rare case report and lioma: a single-center experience of 244 cases. Arch Pathol Lab
review of literature. J Clin Diagn Res. 2016;10(9):ED03. Med. 2018;142(2):236–42.
112. Veras E, et al. Ovarian nonsmall cell neuroendocrine carcinoma: 134. Chapel DB, et al. PAX8 expression in a subset of malignant peri-
a clinicopathologic and immunohistochemical study of 11 cases. toneal mesotheliomas and benign mesothelium has diagnostic
Am J Surg Pathol. 2007;31(5):774–82. implications in the differential diagnosis of ovarian serous carci-
113. Lindboe CF. Large cell neuroendocrine carcinoma of the ovary. noma. Am J Surg Pathol. 2017;41(12):1675–82.
APMIS. 2007;115(2):169. 135. Andrici J, et al. Loss of BAP1 expression is very rare in peritoneal
114. Kwon YS, Im KS, Choi DI. Ovarian large cell neuroendocrine and gynecologic serous adenocarcinomas and can be useful in the
carcinoma in the youngest woman. Eur J Gynaecol Oncol. differential diagnosis with abdominal mesothelioma. Hum Pathol.
2016;37(2):244. 2016;51:9–15.
115. Randolph LK, et al. Hepatoid carcinoma of the ovary: a case report 136. Ota S, et al. Desmoplastic small round cell tumor in the ovary:
and review of the literature. Gynecol Oncol Rep. 2015;13:64–7. report of two cases and literature review. J Obstet Gynaecol Res.
116. Ishikura H, Scully RE. Hepatoid carcinoma of the ovary. A newly 2010;36(2):430–4.
described tumor. Cancer. 1987;60(11):2775–84. 137. Nakayama J, et al. Desmoplastic small round cell tumor of the
117. Tochigi N, et al. Hepatoid carcinoma of the ovary: a report of three ovary: a rare but devastating disease in young women. Gynecol
cases admixed with a common surface epithelial carcinoma. Int J Oncol Case Rep. 2014;7:16–8.
Gynecol Pathol. 2003;22(3):266–71. 138. Mora J, et al. Desmoplastic small round cell tumor 20 years after
118. Pitman MB, et al. Hepatocyte paraffin 1 antibody does not dis- its discovery. Future Oncol. 2015;11(7):1071–81.
tinguish primary ovarian tumors with hepatoid differentiation 139. Neff RT, et al. Management of a rapidly enlarging new adnexal
from metastatic hepatocellular carcinoma. Int J Gynecol Pathol. mass: a rare case of desmoplastic small round cell tumor of the
2004;23(1):58–64. ovary arising in pregnancy. Gynecol Oncol Rep. 2016;17:23–5.
119. Scully RE, Young R, Clement PB. Tumors of the ovary, maldevel- 140. D'Ippolito G, Huizing MT, Tjalma WAA. Desmoplastic small
oped gonads, fallopian tube, and broad ligament. Vol. 23, 3rd ser. round cell tumor (DSRCT) arising in the ovary: report of a case
1998; Washington, DC: Armed Forces Institute of Pathology. diagnosed at an early stage and review of the literature. Eur J
120. Scurry JP, Brown RW, Jobling T. Combined ovarian serous pap- Gynaec Oncol. 2012;33(1):96.
illary and hepatoid carcinoma. Gynecol Oncol. 1996;63(1): 141. Young RH, et al. Ovarian involvement by the intra-abdominal des-
138–42. moplastic small round cell tumor with divergent differentiation: a
121. Sung JH, et al. Hepatoid carcinoma of the ovary without staining report of three cases. Hum Pathol. 1992;23(4):454–64.
for alpha-fetoprotein. Obstet Gynecol Sci. 2013;56(1):41–4. 142. Thway K, et al. Desmoplastic small round cell tumor: pathology,
122. Young RH, et al. Hepatocellular carcinoma metastatic to the genetics, and potential therapeutic strategies. Los Angeles: SAGE
ovary: a report of three cases discovered during life with discus- Publications; 2016. p. 672–84.
sion of the differential diagnosis of hepatoid tumors of the ovary. 143. Cossu A, et al. Female adnexal tumors of probable Wolffian origin
Hum Pathol. 1992;23(5):574–80. (FATWO): a case series with next-generation sequencing muta-
123. Rutgers JL, Scully RE. Cysts (cystadenomas) and tumors of the tion analysis. Int J Gynecol Pathol. 2017;36(6):575–81.
rete ovarii. Int J Gynecol Pathol. 1988;7(4):330–42. 144. Devouassoux-Shisheboran M, Silver SA, Tavassoli FA. Wolffian
124. Heatley MK. Adenomatous hyperplasia of the rete ovarii. adnexal tumor, so-called female adnexal tumor of probable
Histopathology. 2000;36(4):383–4. Wolffian origin (FATWO): immunohistochemical evidence in
125. Nogales FF, et al. Adenomas of the rete ovarii. Hum Pathol. support of a Wolffian origin. Hum Pathol. 1999;30(7):856.
1997;28(12):1428–33. 145. Ramirez PT, et al. Wolffian duct tumors: case reports and review
126. Goyal A, Masand RP, Roma AA. Value of PAX-8 and SF-1 immu- of the literature. Gynecol Oncol. 2002;86(2):225–30.
nohistochemistry in the distinction between female adnexal tumor 146. Hong S, et al. Malignant female adnexal tumor of probable
of probable Wolffian origin and its mimics. Int J Gynecol Pathol. Wolffian origin: case report and literature review. Int J Gynecol
2016;35(2):167–75. Pathol. 2017;00:1–7.
Germ Cell Tumors and Mixed Germ
Cell-Sex Cord-Stromal Tumors 8
of the Ovary
Abstract Keywords
Ovarian germ cell tumors (OGCTs) account for approxi- Mature cystic teratoma · Mature solid teratoma · Immature
mately 30% of all primary ovarian neoplasia, secondary teratoma · Homunculus (fetiform teratoma) · Teratoma
only to epithelial ovarian tumors. OGCTs occur mostly in with malignant transformation · Struma ovarii · Carcinoid
younger women and account for approximately 60% of Neuroectodermal type tumors · Sebaceous tumors ·
all ovarian tumors in women under 21 years of age. Most Dysgerminoma · Yolk sac tumor · Embryonal carcinoma ·
OGCTs are pure, and about 10% of OGCTs contain more Non-gestational choriocarcinoma · Polyembryoma ·
than one component. For these mixed tumors, the behav- Mixed malignant germ cell tumors · Gonadoblastoma ·
ior and prognosis are largely determined by the most Germ cell-sex cord-stromal tumor, unclassified
malignant component; therefore, thorough examination
and extensive specimen sampling are critical for identify-
ing and quantifying all components of the tumor. Germ
cell tumors generally mimic various stages of normal 8.1 Introduction
embryogenesis and have the potential to develop into any
normal tissue type; however, the haphazard distributions Ovarian germ cell tumors (OGCTs), which are believed to
and composition of various tissue types can be diagnosti- arise from primordial germ cells, account for approximately
cally challenging. Achieving the correct diagnosis 30% of all primary ovarian neoplasia, secondary only to epi-
depends on familiarity with this potentially haphazard thelial ovarian tumors. The vast majority of OGCTs (approx-
pattern of various tissue types, correct identification of all imately 95%) are mature cystic teratoma [1, 2]. OGCTs
tissue components, and recognition of foci of possible occur mostly in younger women and account for approxi-
malignant transformation. This chapter focuses on the mately 60% of all ovarian tumors in women under 21 years
clinicopathologic features of OGCTs and also briefly cov- of age, of which one-third are malignant germ cell tumors
ers mixed germ cell and sex cord-stromal tumors. [3]. In contrast, OGCTs account for only 6% of ovarian
tumors in postmenopausal women. Most OGCTs are pure
and only about 10% of OGCTs contain more than one com-
ponent. For these mixed tumors, the behavior and prognosis
of the tumor are largely determined by the most malignant
H. Chen
Department of Pathology, UTSouthwestern Medical Center, component; therefore, thorough examination and extensive
Dallas, TX, USA specimen sampling are critical for identifying and quantify-
C. M. Quick ing all components of tumor.
Department of Pathology, College of Medicine, University of Based on the anatomical location, age of unset, precursor
Arkansas for Medical Sciences, Little Rock, AR, USA lesions (cells of origin, genetic characteristics), and chromo-
O. Fadare somal makeup, human germ cell tumors can be roughly sub-
Department of Pathology, University of California San Diego, classified into the following groups [4]: type I, neonatal and
San Diego, CA, USA
infantile teratoma and yolk sac tumor; type II, dysgermi-
W. Zheng (*) noma and other germ cell tumors (occur mostly in adoles-
Departments of Pathology, Obstetrics and Gynecology,
University of Texas Southwestern Medical Center,
cents and young adults); type III, spermatocytic seminoma
Dallas, TX, USA (typically occurs in testis of elderly males); type IV, dermoid
e-mail: [email protected] cyst; and type V, hydatidiform mole of childbearing-age
© Science Press & Springer Nature Singapore Pte Ltd. 2019 231
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_8
232 H. Chen et al.
women. The pathogenesis of OGCTs is not yet fully under- patients can present with abdominal pain, bloating, and
stood. Mature cystic teratomas (MCT) contain diploid chro- abnormal vaginal bleeding, especially in larger and/or
mosomes with a normal 46, XX karyotype, and are believed ruptured tumors. Ovarian torsion occurs in 3–16% of
to originate from germ cells immediately after the first cell cases. Around 10% of cases are discovered during preg-
division of meiosis. Therefore, it is classified as a type IV nancy. These tumors can occasionally present with hir-
germ cell tumor using a previously described scheme [5]. sutism, which is associated with ovarian interstitial
However, recent studies demonstrate that MCTs may vary in luteinization. Teratoma-associated encephalitis is a rare
cell origin, with most MCTs originating from germ cells yet potentially fatal paraneoplastic syndrome that occurs
post-first cell division of meiosis., A minority of MCTs arise in young women, and is caused by antibodies to the
from primordial germ cells before meiosis occurs. In con- N-methyl-d-aspartic acid receptor (NMDAR). Tumors
trast to MCTs, neonatal and infantile yolk sac tumors, as causing this syndrome may display specific morphologi-
well as immature teratomas, likely originate from pluripotent cal features [8].
embryonic stem cells. OGCTs in elderly women can occur
synchronously with somatic type tumors, and are believed to 8.2.1.3 Macroscopic Features
originate from tumor stem cells. Overall, the pathogenesis of Mature teratomas are usually unilateral, but may be bilat-
OGCTs is different from that of male testicular germ cell eral in 10–15% cases. They are generally oval in shape
tumors which are mostly malignant and aneuploid, and are with a smooth external surface. They may be peduncu-
believed to originate mostly from primordial germ cells lated. The size of mature teratomas ranges from micro-
before meiosis. scopic up to 40 cm, with an average diameter of 8 cm [9].
Germ cell tumors generally mimic various stages of nor- On sectioning, the tumors are typically unilocular; how-
mal embryogenesis and have the potential to develop into ever, they may occasionally be multilocular. The cysts are
any normal tissue types; however, the haphazard distribu- usually filled with sebaceous material, hair, and occasion-
tions and composition of various tissue types can be diagnos- ally mucinous or serous fluid (Figs. 8.1 and 8.2). The cyst
tically challenging. Achieving the correct diagnosis depends is lined with skin, mucosa, or dark membranous tissue.
on the following three factors: (1) familiarity with this hap- The cyst wall varies in thickness, often with focal thicken-
hazard pattern of various tissue types; (2) correct identifica- ing or a nodular appearance. A solid nodule composed of
tion of all tissue components; and (3) correct identification of adipose tissue, teeth, and bone, protruding into the cyst
malignant transformation of benign/mature components. lumen, is colloquially called “Rokitansky’s protuberance.”
This chapter focuses on the clinicopathologic features of Other tissues such as gastrointestinal mucosa, white or
OGCTs and also briefly covers mixed germ cell and sex gray matter of the brain, and thyroid may be seen within
cord-stromal tumors. the large protuberance. The protuberance and areas of
thickened cyst wall are most likely to contain various
mature and immature tissues; the grossing of the specimen
8.2 Teratoma should focus on those areas.
8.2.1.1 Definition
This is a tumor composed exclusively of mature tissue
derived from two or three germ cell layers. These tumors are
usually cystic, but rarely can be solid. If the lining of cyst is
solely composed of mature skin and associated adnexal
structures, it may also be called a dermoid cyst.
a b
c d
Fig. 8.3 Mature cystic teratoma. The more common mature tissues include epidermis, sebaceous glands, hair follicles, and adipose tissue (a);
sweat glands and cartilage (b); mature glial tissue (c); and epithelium of the gastrointestinal and respiratory tracts (d)
234 H. Chen et al.
Table 8.1 Frequency of various tissue types identified in mature cystic adnexa. In contrast, mature cystic teratomas contain tis-
teratomas
sues derived from two or three embryonic layers; how-
Frequency Percentage Cell/tissue types ever, some use these terms interchangeably.
Common >66 Skin, hair, sebaceous glands, white fat, 2. Immature teratomas
brain tissue (children)
The diagnosis of an immature teratoma depends on the
Relatively 33–66 Sweat glands, brain tissue (adult),
common smooth muscle, peripheral neural tissue, identification of immature mesenchyme, which is almost
cartilage, bone, respiratory epithelia always composed of immature neuroectoderm. Extensive
Occasional 5–33 Teeth, gastrointestinal epithelia, sampling of the specimen, particularly the solid areas, is
thyroid, ependymal tissue, melanocytes required to rule out immature teratomas.
Rare <5 Prostatic gland, retina, mammary tissue,
3 . Homunculus (fetiform teratomas)
pituitary glands, choroid plexus,
ganglion, lung, liver, kidney, This rare form of mature teratoma contains mature tissue
cardiomyocytes, thymus, cerebellum, derived from three embryonic layers, and demonstrates
skeletal muscle, brown fat more defined organization causing it to resemble mal-
formed organs, and, rarely, a fetus.
mature cystic teratoma, in which presence of microscopic taining mucus or blood. The solid areas are gray-tan to pink,
foci of immature neural components is acceptable, this tumor soft, fleshy, and often hemorrhagic and necrotic (Fig. 8.4).
should be designated immature if microscopic, rare, primi- Patchy calcification is frequently seen. Gross analysis and
tive neuroectodermal components are present. As a result, sectioning should focus on the gray-tan, soft, fleshy areas,
extensive sampling of the specimen is required to rule out with special attention to areas of hemorrhage and necrosis,
immature teratomas. Gliomatosis peritonei usually consists as these are the areas in which immature tissue is most likely
of deposits of mature neuroglial tissue. Occasionally, other to be identified.
tissue types such as squamous epithelia or cartilage can be
seen within these foci. 8.2.3.4 Microscopic Features
Most of the tumors consist of a mixture of mature and immature
8.2.2.4 Differential Diagnosis embryonic components. Identification of immature embryonic
Grade 1 immature teratoma is the most important differential elements within tumors is the key for the diagnosis of immature
diagnosis. Identification of immature neuroectoderm within teratoma. The immature embryonic type tissue is usually neuro-
the tumor is the key. ectodermal and composed predominantly of small blue neuro-
blasts, primitive neuroepithelial rosettes, and tubules (Fig. 8.5).
8.2.2.5 Treatment and Prognosis The primitive neuroepithelial tubules and rosettes are lined by
Similar to mature cystic teratoma, these tumors have an
excellent prognosis; however, gliomatosis peritonei tends to
recur. Treatment consists of salpingo-oophorectomy with
complete resection of peritoneal glial implants.
8.2.3.1 Definition
This is a teratoma containing variable amounts of immature
(typically primitive neuroectodermal) tissue.
a crowded cells with a high N:C ratio and frequent mitotic figures
and apoptotic bodies. These foci may be pigmented. The lumi-
nal surface of the tubules displays a smooth membrane, which
is a helpful feature used to distinguish it from other nonneural
tubules (especially renal tubules). The relative amount of the
primitive neuroepithelial component is a critical factor in grad-
ing and determining the prognosis. In poorly differentiated, or
grade 3, immature teratoma, more primitive neuroepithelium is
present, which typically demonstrates increased nuclear atypia
and mitotic activity, as well as epithelial stratification of the epi-
thelial rosettes (Fig. 8.6). Islands of immature cartilage, bone,
skeletal muscle, and glandular structures set in a myxoid stroma
are often seen. Other embryonic endodermal elements such as
hepatic tissue, intestinal epithelium, and glomeruli (rare) may
b
also be present. The presence of the abovementioned immature
components alone is not sufficient for the diagnosis of immature
teratoma. In addition, one must not confuse retinal tissue, cere-
bellar tissue, or thymic tissue as primitive neuroepithelium.
Occasionally immature teratoma can be seen admixed with
other types of germ cell tumor, such as yolk sac tumor, chorio-
carcinoma, or embryonal carcinoma, with yolk sac tumor being
the most common (Fig. 8.7). If the largest focus of other germ
cell tumor types is less than 2 mm in diameter, they are found to
have no effect on the overall prognosis of the tumor [18].
Markedly elevated serum AFP level should alert the pathologist
to the presence of a yolk sac component, and if absent in the
Fig. 8.6 Immature teratoma. The neuroectodermal tubules are lined by
multiple layers of cells (a) with basophilic cytoplasm and overlapping initial sections additional tissue should be submitted. Of note,
nuclei. Mitotic activity is typically brisk (b) the presence of immature intestinal glandular tissue and liver
parenchyma may lead to increased levels of AFP in the absence
of a yolk sac component.
Table 8.2 Grading of ovarian immature teratomas (three-tiered Rare examples have been reported in which ovarian carci-
system)
nosarcoma may display mesenchymal components that
Grade Histological features resemble primitive neuroectodermal tissue or immature
1 Rare foci of immature neural tissue, less than 1 low-power tubules seen in an immature teratoma [21–24]. The term
field (4X) per slide
“teratoid carcinosarcoma” has been applied for these
2 Moderate foci of immature neural tissue, more than 1 but
less than 4 low-power fields per slide cases. The key to correctly identifying these unusual com-
3 Abundant foci of immature neural tissue, more than 4 lower binations is to sample the tumor well, and to characterize
power fields per slide (Fig. 8.8) each component accurately.
3. Mixed malignant germ cell tumor
Adequate sampling and careful histologic examination
8.2.3.5 Grading of Immature Teratomas can help to correctly identify other germ cell tumor types,
The grading of immature teratomas is based on the relative which is the key in making an accurate diagnosis and thus
quantities of the immature neural tissue. Norris et al. [16] determining the prognosis for the patient. Careful chart
first proposed a three-tiered system to grade immature tera- review may disclose clinical findings, such as elevated
tomas (shown in Table 8.2). This system has been further tumor markers like AFP that can serve as a clue to the
simplified into a two-tiered system: low grade (previously presence of altered differentiation.
grade 1) and high grade (previously grades 2 and 3). 4 . Mature cystic teratoma with rare, microscopic foci of
immature neural tissue
8.2.3.6 Immunohistochemistry Rare, microscopic foci of immature neural tissue (where
Neuroectodermal tissue is typically positive for NSE, S-100, the largest focus is less than 2 mm in diameter, and less
NF, synaptophysin, NGFR, and GFAP; immature neural tis- than four foci per tumor) are within the acceptable diag-
sue can express SALL4, LIN28, SOX2, and glypican-3, but nostic spectrum of a mature cystic teratoma. The presence
is usually weaker when compared with other germ cell of rare, microscopic immature neural tissue has no effect
tumors [19, 20]. Immature endodermal tissues such as intes- on prognosis. Yanai-Inbar and Scully studied 11 such
tinal type glands and hepatic tissue can express AFP, which cases, and found no recurrence of tumor [11]; however,
can lead to an elevated serum APF level. The Ki67 (MIB-1) extensive sampling of the tumor is the key to making the
proliferative index is usually high in primitive neuroepithe- correct diagnosis.
lial tubules or rosettes.
8.2.3.9 Treatment and Prognosis
8.2.3.7 Genetics Approximately one-third of patients have extra-ovarian
Immature teratomas typically exhibit fewer DNA changes spread of tumor at the time of diagnosis. Overall, the grade
than other germ cell tumors. No gain of 12p or i(12p) has and FIGO stage of the tumor determine treatment planning
been identified [10]. Grade 1–2 immature teratomas are usu- and clinical outcome. For patients with low-grade, low-stage
ally diploid, whereas most grade 3 tumors are aneuploid. (stage I–II) tumors, fertility-preserving surgical procedures
and close follow-up are recommended. Extra attention
8.2.3.8 Differential Diagnosis should be given to the contralateral ovary at the time of sur-
1. Malignant transformation of mature cystic teratoma gery, since approximately 10–15% of patients may have a
Mature cystic teratomas harboring malignancy are typi- mature teratoma in the contralateral ovary. For high-grade,
cally cystic, with a thickened cyst wall, or intraluminal low-stage (stage I–II) tumors, any stage III tumor, or recur-
protuberance, and filled with greasy sebaceous material rent tumor, adjuvant chemotherapy may be attempted. This
and hair. Microscopically, various patterns of squamous tumor is usually unresponsive to radiotherapy.
cell carcinoma or adenocarcinoma are present; however, It has been reported that surgical removal can lead to a
these lack the primitive neuroepithelial components seen cure in most children and adolescents regardless of tumor
in an immature teratoma. It is worth noting that malignant grade, and that chemotherapy can be given in recurrent cases
neuroendocrine carcinoma can have neuroepithelial [25]. In most recurrent cases in children, foci of yolk sac
tubule-like structures; therefore, the overall clinicopatho- tumor are typically present [15] as opposed to immature neu-
logical features of the tumor should be considered in ral tissue; therefore, identifying other germ cell components
order to make the correct diagnosis. is far more important than the pathological staging of tumor.
2. Carcinosarcoma of ovary (malignant mixed Müllerian Moreover, the following exceptions should be noted:
tumor)
These tumors characteristically contain both Müllerian- 1 . Extra-ovarian spread of immature teratomas, including
type adenocarcinoma, and a homogenous or heterogenous lymphovascular metastasis and peritoneal implants, has
sarcoma, such as rhabdomyosarcoma or chondrosarcoma. poor prognosis.
238 H. Chen et al.
2. Post-chemotherapy peritoneal glial implants and residual implants and separate grading of individual implants are
tumor are usually benign due to the elimination of the recommended. The following example diagnosis may be
immature tumor components by chemotherapy. Most utilized in these cases “Ovarian immature teratoma (grade
studies have shown that these tissues have no adverse 1), with peritoneal non-invasive implants (grade 0).”
effect on the overall prognosis of the patient.
3. Growing teratoma syndrome is defined by persistent
growth of tumor even after chemotherapy; however, 8.2.4 Homunculus (Fetiform Teratoma)
serum AFP levels may remain normal. Microscopically,
these tumors have a benign appearance. Since they are 8.2.4.1 Definition
resistant to chemotherapy, surgical removal is required. Fetiform teratoma (homunculus) is a term that has been
Complete removal of such tumors generally results in a given to a rare form of ovarian teratoma that resembles a
good long-term prognosis; however, rare cases of recur- malformed fetus, containing well-developed mature tissues
rence with other germ cell tumor types (such as yolk sac and malformed organ structures.
tumor) have been reported.
4. Gliomatosis peritonei is composed of peritoneal implants 8.2.4.2 Clinical Features
of mature glial tissue. It can occur de novo with the pri- It occurs in young females, around 20–35 years of age. It is
mary tumor or postsurgically. It is the most common type usually asymptomatic, occasionally presenting as a pelvic
of pelvic spread of immature teratoma, and occasionally mass. X-ray studies can show partially developed skull, man-
may be seen in patients with mature cystic or solid terato- dible, vertebral bodies and/or attached ribs, and pelvic bones.
mas [26]. In extremely rare situations, it can be seen in The pubic bone is the most common pelvic bone found in a
patients with no known history of a germ cell tumor under- homunculus. Rarely complete pelvic structures including a
going treatment with a ventricular peritoneal shunt [27]. In tailbone and femur can be seen.
addition to the peritoneal implants (Figs. 8.9 and 8.10),
lymphovascular, and even hematogenous spread can be 8.2.4.3 Macroscopic Features
seen in rare cases. Spontaneous or post-chemotherapeutic Typically this tumor is a large oval cystic adnexal mass, com-
maturation of the disseminated tumors occurs in approxi- monly containing fetus-like structures, recognizable skull
mately one-third of patients, leading to noninvasive (which may include cerebral and cerebellar tissue) facial bones,
implants. The exact pathogenesis of gliomatosis peritonei tissues resembling eyeballs, and a mandible with attached
is still unclear. Two theories have been proposed to explain teeth. The skull can be covered with skin and hair. A rudimen-
this phenomenon: direct seeding of teratoma or stem cell- tary torso including a rib cage which may contain pulmonary
derived peritoneal metaplasia. These glial cells are typi- tissue may be present. Other organs such as the intestines, and
cally positive for GFAP and SOX2. Glial implants are poorly formed upper or lower extremities, can also be seen.
considered grade 0, or noninvasive implants, and have Like a mature cystic teratoma, the cystic p ortion of this tumor
benign biological behavior [26]. Extensive sampling of can also be filled with greasy sebaceous material and hair.
Fig. 8.8 Immature teratoma, grade 3. There are many foci of immature Fig. 8.9 Immature teratoma, gliomatosis peritonei. There are innumer-
neuroepithelial tissue including many foci of neuroectodermal tubules able, military, gray nodules of glial implants on the congested perito-
(circled) neal surface
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 239
Fig. 8.10 Immature
teratoma, gliomatosis a b
peritonei. Typical peritoneal
glial implants (a). GFAP is
positive (b)
a b
Fig. 8.11 Teratoma with malignant transformation. Well-differentiated squamous cell carcinoma (a) and moderately differentiated adenocarci-
noma (b)
derline tumors were more common than squamous cell 8.3.1.2 Clinical Features
carcinoma [32]. Half of the squamous cell carcinomas are well Although only accounting for 2.5–5% of all ovarian tera-
differentiated. Squamous cell carcinoma can arise from the toma, struma ovarii is the most common type of monodermal
lining squamous epithelium or from squamous metaplasia of teratoma. Most patients are in their fifth decade [35]. Patients
the glandular epithelium (transition between squamous epi- can present with a pelvic mass or bloating, ascites (up to one-
thelium and glandular epithelium is commonly seen in mature third of patients), hyperthyroidism (5–15%), thyroid gland
cystic teratomas). Adenocarcinomas can arise from various hyperplasia or postsurgery compensatory thyroid gland
tissue types. Other rare types of malignancy include adeno- hyperplasia (5–15%), Meigs syndrome (triad of ascites,
squamous carcinoma, mucoepidermoid carcinoma, basal cell pleural effusion, and ovarian tumor), and peritoneal implants.
carcinoma, melanoma, and undifferentiated carcinoma. Significant elevation of the serum thyroglobulin level can be
Mucinous tumors arising from a teratoma can cause pseudo- seen in patients with a malignant struma ovarii [36]. A pre-
myxoma peritonei [33]. Sarcomas, including angiosarcoma, operational I131 abdominal imaging study or radionuclide
leiomyosarcoma, rhabdomyosarcoma, fibrosarcoma, osteo- angiography can help to identify the ovarian thyroid tissue.
sarcoma, chondrosarcoma, undifferentiated pleomorphic sar-
coma, among others have been described as well [34]. 8.3.1.3 Macroscopic Features
Struma ovarii are usually unilateral and usually range in size
8.2.5.5 Treatment and Prognosis from 0.5 to 10 cm (may be up to 20 cm). Grossly, it appears
The overall prognosis is poor and the outcome depends on similar to a goiter, with smooth or nodular outer surface. The
the tumor type and stage. The 5-year survival of FIGO stage cut surface is solid, gelatinous, and dark brown to beefy red
I–IV patients is 56%, 25%, 12%, and 0%, respectively. Of in color, often with multiple small cysts filled with gelati-
note, tumors with sebaceous differentiation usually have a nous or mucinous material. Some tumors can be predomi-
better prognosis, although this possibility requires more nately or entirely cystic, containing bluish gelatinous
extensive validation. Total hysterectomy and bilateral material (Fig. 8.12). Malignant struma ovarii is usually more
salpingo-oophorectomy is the first choice of treatment. than 10 cm in size, and associated with adhesion formation
and ascites.
and TTF-1. Cystic struma ovarii presents as dilated cysts Three to 10% of all struma ovarii are malignant [37] and
with thin fibrous septa. The cysts are lined with flat or low most are papillary carcinoma, followed by follicular carci-
cuboidal epithelium. Occasional tumors may demonstrate noma. Insular carcinoma (poorly differentiated), medullary
focal or diffuse adenomatous hyperplasia [37]. Follicular carcinoma, and undifferentiated carcinoma are rare [40]. The
adenomatous (crowded microfollicles/trabeculae and solid same diagnostic criteria of thyroid based papillary carcinoma
areas) or papillary hyperplastic changes that lack of nuclear are used in the diagnosis of ovarian papillary carcinoma,
features of papillary carcinoma are sometimes referred to as including clear nuclei (or ground glass nuclei), nuclear
hyperplastic struma ovarii, and are believed by some authors grooves, nuclear crowding, and nuclear pseudo-inclusions
to have malignant potential [39]. (Figs. 8.14 and 8.15). The diagnosis of follicular carcinoma
is difficult as the tumor lacks a capsule and is irregular in
shape. The capsular infiltration criteria often used in diagno-
sis of thyroid follicular carcinoma cannot be applied to the
ovarian variant of follicular carcinoma. The most reliable
finding used to make a diagnosis of ovarian follicular carci-
noma is lymphovascular invasion (Fig. 8.16) [37].
It is controversial whether the same diagnostic criteria for
thyroid carcinoma should be applied to the diagnosis of
malignant struma ovarii. This is complicated by the fact that
some morphologically benign strumas demonstrate
malignant behavior, such as extra-ovarian spread and recur-
rence. Robboy et al. [41] reported 27 cases of struma ovarii
with malignant behavior. Interestingly, only 12 cases met the
histologic criteria of thyroid carcinoma, while 15 cases
demonstrated benign morphology (13 cases of adenomatous
changes, 2 cases of benign thyroid tissue). The “benign”-
2 cm appearing tumors have been designated “highly differenti-
ated follicular carcinoma” [42]. This diagnosis can only be
made when extra-ovarian spread or metastasis is present. In
Fig. 8.12 Struma ovarii. The tumor has a capsule with multifocal cys-
tic and solid areas. There is dark brown and blue-green colloid material
rare cases, the spread or distant metastasis can occur as late
within the cysts as 17 years after treatment [43].
a b
Fig. 8.14 Ovarian thyroid papillary carcinoma, classical type. The cut surface is predominantly solid (a). The tumor shows a papillary growth
pattern (b), characteristic ground glass nuclei, nuclear grooves, and nuclear overlap (c)
8.3.2.2 Clinical Features nar cells are present at the periphery of islands, while tri-
Primary ovarian carcinoid is extremely rare. Insular carci- angular-shaped cells are located in the center. Tumor cells
noid is the most common type (26–53%), followed by may also form acinar, cribriform, and pseudo- rosette
strumal carcinoid (26–44%), with trabecular carcinoid as the structures. The acinar lumens contain eosinophilic secre-
third most common (23–29%). Mucinous (goblet cell) carci- tions, which may be condensed and calcified. Tumor cells
noid is exceptionally rare (1.5%) [49, 50]. The average age have abundant cytoplasm and many eosinophilic gran-
of patients is 53 (range 14–79) [49, 51, 52]. Most patients ules, which are more prominent at the periphery of the
present with an ovarian mass. Carcinoid syndrome can be islands. The cells have uniform, round nuclei with coarse
seen in up to 30% of insular carcinoid cases, 13% of trabecu- chromatin granules (salt and pepper nuclei), and conspic-
lar carcinoid cases, and 3.2% of strumal carcinoid cases. uous nucleoli. Mitotic figures are rare or absent.
25% of trabecular carcinoids produce YY polypeptides [53], 2. Trabecular carcinoid
which in turn can inhibit bowel movement and cause severe Trabecular variants resemble carcinoids of foregut or
constipation. The presence of elevated levels of serum sero- hindgut origin. The cords are composed of columnar cells
tonin (5-HT) and its metabolite 5-hydroxyindoleacetic acid which have abundant cytoplasm, oval nuclei, and argyro-
(5-HIAA) can be diagnostically useful. philic granules at both poles of the nucleus. Tumor cells
form parallel cords, trabeculae, or ribbons separated by
8.3.2.3 Macroscopic Features thin fibrous stroma. The elongated nuclei of tumor cells
Ovarian carcinoid is typically a unilateral tumor, and the are organized perpendicular to the long axis of cords or
mass is typically firm, yellow-tan, and solid. Hemorrhage ribbons (Fig. 8.19).
and necrosis are rare. Occasionally the tumor may present as 3 . Strumal carcinoid
a small firm nodule (usually less than 4 cm) protruding into Strumal carcinoid is composed of carcinoid tumor and
the lumen of a mature cystic teratoma or a mucinous cystic thyroid tissues in various proportions (Fig. 8.20).
tumor. Insular carcinoid usually has smooth capsule with an Approximately 50% of cases are trabecular carcinoids.
average size of 10 cm (3–28 cm). Grossly, carcinoid and Other types of carcinoid such as insular and mucinous
struma components may be separately recognized. carcinoid may also be seen. The thyroid component
resembles normal thyroid tissue or a follicular adenoma.
8.3.2.4 Microscopic Features Neuroendocrine cells can replace the original lining of the
1. Insular carcinoid thyroid-type follicles. Similar to thyroid medullary carci-
This tumor resembles a midgut carcinoid and is character- noma, various amounts of amyloid deposition in the
ized by nests or islands of tumor cells separated by fibrous stroma can be seen. One-third of cases show stromal
stroma (Figs. 8.17 and 8.18). Typically, palisading colum- luteinization.
Fig. 8.17 Insular carcinoid. The cells are arranged in acini (a) and have round nuclei and small eosinophilic granules in cytoplasm (b)
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 245
Fig. 8.18 Insular carcinoid. This tumor originates from a mature teratoma (mature teratoma marked by *) (a). The tumor cells are arranged in
acini and have round nuclei (b); synaptophysin (c) and chromogranin (d) are positive.
a b
Fig. 8.20 Strumal carcinoid. The cells show an infiltrative pattern within the tumor stroma (a); benign thyroid tissue (b)
and negative or weakly positive in gastric carcinoid [57]. 8.3.3.3 Macroscopic Features
Currently, the expression pattern of SATB2 in primary Tumors are usually unilateral, measuring 4–20 cm in diam-
appendiceal goblet cell carcinoid is still not well studied eter (average 14 cm) [58]. The cut surface is red-tan, fleshy,
due to the extreme rarity of the disease. and solid. Occasional tumors may also be cystic, with intra-
2. Sex cord-stromal tumor luminal papillary excrescences [60]. Hemorrhage and necro-
When associated teratomatous components are absent, sis may be prominent.
trabecular carcinoids need to be distinguished from sex
cord-stromal tumors. The Sertoli tubules in a well- 8.3.3.4 Microscopic Features
differentiated Sertoli-Leydig cell tumor can be some- Neuroectodermal tumors are morphologically identical to
times mistaken as trabecular carcinoid. A panel of their counterparts in the central or peripheral nervous sys-
immunostains can help in the diagnosis. Carcinoid cells tem. These tumors can be further subcategorized into three
express at least one neuroendocrine tumor marker groups: differentiated, primitive, and anaplastic [57, 59, 61,
(synaptophysin、chromogranin, Leu7); however, it is 62]. The differentiated group includes ependymoma, astro-
worth noting that both carcinoid and sex cord-stromal cytoma, and oligodendroglioma; the primitive group includes
tumors express CD56. In addition, sex cord-stromal Ewing sarcoma/PNET, neuroblastoma, ependymoblastoma,
tumors also express inhibin, WT-1, SF-1, and medulloblastoma, and medulloepithelioma; finally, the ana-
calretinin. plastic group is composed of glioblastoma multiforme
3 . Struma ovarii (Fig. 8.22) and anaplastic ependymoma (Fig. 8.23).
Typical struma may form trabecular structures, and often Ependymoma is occasionally associated with astrocytoma,
will coexist with strumal carcinoid. Immunostains can aid but only rarely associated with a teratoma, which is in con-
in the diagnosis. Strumal tissue expresses TTF-1 and thy- trast to the other types of differentiated tumors, which are
roglobulin, as opposed to carcinoid tumors which do not usually associated with mature and immature teratomas.
express either marker.
8.3.3.5 Immunohistochemistry
8.3.2.7 Treatment and Prognosis Ependymoma and astrocytoma are GFAP positive, whereas
Almost all ovarian primary trabecular and strumal carcinoids Ewing sarcoma/PNET is diffusely and strongly positive for
are stage I, and have an excellent prognosis. The 5-year sur- CD99. Oligodendroglioma can sometimes be confused with a
vival rate of insular carcinoid is 95%, and the 10-year sur- central neurocytoma; however, it is GFAP positive and synap-
vival rate is 88%. Total hysterectomy and bilateral tophysin negative, whereas central neurocytoma demonstrates
salpingo-oophorectomy is the first line of treatment. For the opposite staining pattern. The immunophenotypic com-
young patients who desire to preserve fertility, or patients parison between ovarian/peritoneal ependymoma and central
with trabecular carcinoid, unilateral salpingo-oophorectomy ependymoma is summarized in Table 8.3 [63]. Moreover,
and close follow-up can be offered. For patients with extra- ovarian/peritoneal ependymomas do not contain IDH1/2
ovarian disease, surgical resection of extra-ovarian tumor as mutations [59]. Taken together, the genetic and immunophe-
much as technically feasible, as well as chemotherapy,
should be considered.
8.3.3.1 Definition
These tumors are entirely, or predominantly, composed of
neuroectodermal tissue with similar morphology and differ-
entiation as neuroectodermal tumors of the central nervous
system. Less often these tumors may resemble peripheral
type neuroectodermal tumors such as Ewing sarcoma/primi-
tive neuroectodermal tumor.
a b
c d
Fig. 8.23 Anaplastic ependymoma. The tumor exhibits perivascular pseudorosettes (a), papillary (b), and solid growth with remarkable cellular
atypia (c). GFAP staining is diffusely positive (d)
Table 8.3 Immunophenotypical difference between ovarian/perito- tumors have a relatively good prognosis. When extra-ovarian
neal and central ependymomas spread occurs, the prognosis is poor.
Biomarkers Ovarian/peritoneal ependymoma Central ependymoma
CAM5.2 60%+ 10%+
CK18 100%+ 20%+ 8.3.4 Sebaceous Tumors
CK7 80%+ 10%+
34βE12 60%+ – 8.3.4.1 Definition
ER 100%+ 10%+ Sebaceous tumors are neoplasms that resemble cutaneous
PR 80%+ 20%+
sebaceous gland tumors, including sebaceous adenoma,
basal cell carcinoma with sebaceous differentiation, and
notypic evidence suggests that although these two tumors are sebaceous carcinoma. They may arise from dermoid cyst.
histologically similar, their pathogenesis may be divergent.
8.3.4.2 Clinical Features
8.3.3.6 Genetic Profiles Sebaceous tumors within a dermoid cyst are exceedingly
Ewing sarcoma/PNET may contain t(11;22)(q24;q12) [10]. rare, and typical occur in middle-aged to elderly women.
When they occur, they usually present as pelvic mass.
8.3.3.7 Treatment and Prognosis
Tumors in the differentiated group have a better prognosis 8.3.4.3 Macroscopic Features
than tumors in other groups; that being said when confined These tumors are predominantly cystic and are usually asso-
within the ovarian capsule, even primitive and anaplastic ciated with a mature cystic teratoma.
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 249
Fig. 8.25 Dysgerminoma.
The surface of the tumor is
a b
smooth and has a conspicuous
capsule. The position of
surgeon’s finger shows the
fallopian tube (a). The cut
surface is solid, fleshy, and
lobulated with focal
hemorrhage (b)
2 cm
250 H. Chen et al.
present (Fig. 8.25b). Such areas should be sampled exten- tumor cells are organized in insular, corded, diffuse, and
sively to identify other malignant germ cell tumor compo- pseudoglandular patterns. Nests of cells are intersected by
nents, which may have significant implications in the fibrous septa (alveolar pattern), typically containing lympho-
prognosis and treatment of the tumor. Of note, the presence cytes. Epithelioid histiocytes/granulomas with m ultinucleated
of small grainy calcifications may suggest the concurrent giant cells are present in approximately 20% of cases
presence of an underlying gonadoblastoma. The tumor may (Fig. 8.27). Syncytiotrophoblasts are present in 3–5% of
contain trophoblastic tissue, particularly in the areas of tumors (Fig. 8.28) [72], which may result in elevated serum
hemorrhage. β-hCG level. Mitotic figures are frequently seen. Necrosis
and hemorrhage are common. Atrophic calcification is occa-
8.4.1.4 Microscopic Features sionally seen in some tumors along with fibrotic and hyaline
Morphologically, dysgerminoma is identical to the semi- degeneration. Dysgerminoma is usually a pure disease; how-
noma of testis and germinoma of the thoracic region. The ever, it may occasionally be associated with gonadoblastoma
tumor is composed of a monotonous cell population. The or other germ cell tumors.
tumor cells are round or polygonal with abundant pale to
eosinophilic cytoplasm and discrete cell membranes. Tumor 8.4.1.5 Immunohistochemistry
cells have one large, rounded to squared-off nuclei with one Tumor cells composing a dysgerminoma are strongly posi-
or more prominent nucleoli. Monotonicity of cells and nuclei tive for PLAP, CD117/c-kit (Fig. 8.29a), and D2–40.
is a characteristic feature of dysgerminoma (Fig. 8.26). The Although CD117 membrane positivity is a characteristic fea-
a b
c d
Fig. 8.26 Dysgerminoma. Nests (a), cords (b), and microcysts (c) are separated by lymphocyte-rich fibrous connective tissue. The tumor cells are
round or polygonal, have abundant cytoplasm, and large, centered nuclei with prominent nucleoli (d)
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 251
a b
Fig. 8.27 Dysgerminoma. There are a large number of granulomatous changes in this tumor, essentially causing the tumor cells to become invis-
ible (a). SALL4 staining helps to demonstrate the scattered tumor cells (b)
Fig. 8.28 Dysgerminoma.
Syncytiotrophoblastic giant
a b
cells (a) that are positive for
β-hCG (b) are obvious in this
tumor
ture of dysgerminoma, it can also be seen in some yolk sac 8.4.1.6 Genetic Profiles
tumors (especially the solid variant) [73]. Tumor cell nuclei Approximately 25–50% of dysgerminomas contain c-KIT
are positive for OCT4 (Fig. 8.29b) and SALL4 (Fig. 8.29c) mutations, most commonly in exon 17, but not the exon 11,
[74–76]. LIN28 is another sensitive marker for dysgermi- which confers the susceptibility to imatinib therapy [10, 77].
noma (Fig. 8.29d) [20]. All three markers are also positive in Approximately 60% of tumors contain isochromosome 12p
embryonal carcinoma, and SALL4 and LIN28 are positive in and/or 12p overrepresentation [78].
yolk sac tumor as well. Syncytiotrophoblasts in dysgermi-
noma are strongly positive for β-hCG and CAM5.2, while 8.4.1.7 Differential Diagnosis
AE1/AE3 and CK7 are often focal and weakly positive. 1. Small-cell carcinoma, hypercalcemic type
CK20, EMA, HMWK, and CD30 should be negative. The gross appearance of the tumor and the age of patients
Commonly used immunohistochemical markers helpful in of hypercalcemic small-cell carcinoma are similar to
the workup of ovarian germ cell tumors are summarized in those of dysgerminoma. In addition, dysgerminoma may
Table 8.4. occasionally have the microcystic spaces similar to the
252 H. Chen et al.
a b
c d
Fig. 8.29 Dysgerminoma: common immunohistochemical markers. CD117 is positive along the cell membrane (a); OCT4 (b) and SALL4 (c) are
positive in the nuclei (b); and LIN28 is positive within the cytoplasm (d)
Table 8.4 Immunohistochemical markers for ovarian germ cell tumors and other ovarian malignancies
Biomarkers Dysgerminoma Embryonal carcinoma Yolk sac tumor Lymphoma Granulosa cell tumor Melanoma
CK −/+ + + − − −
SALL4 + + + −/+a −/+ −
LIN28 + + + − − −
OCT4 + + − −/+b − −
AFP − − +/− − − −
Glypican-3 − −/+ +/− − − −
CD117 + − −/+ − + −
CD30 − + − +/− − −
S-100 − − − − − +
LCA − − − + + −
MPO − − − − + −
Lymphoblastic lymphoma can be SALL4 positive
a
pseudofollicular spaces commonly seen in hypercalcemic composed of relatively large cells; when this occurs, the
small-cell carcinoma; however, small-cell carcinoma of cytoplasm is typically eosinophilic with or without rhab-
the hypercalcemic type is composed of small round cells doid inclusions. In contrast, the tumor cells of dysgermi-
with scant cytoplasm, hyperchromatic nuclei, and small noma have a centrally located round nucleus, with coarse
nucleoli. Occasionally, small-cell carcinoma may be chromatin, discrete nuclear membrane, and one or more
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 253
Table 8.5 The clinicopathological features of dysgerminoma and ever, lymphoma is frequently bilateral and can involve
embryonal carcinoma
both ovaries in approximately half of patients. Germ cell
Dysgerminoma Embryonal carcinoma (such as PLAP, D2-40, SALL4) and lymphoid markers
Clinical 10–15% cases are Unilateral can help to make the correct diagnosis (Table 8.3). It is
features bilateral
worth noting that the germ cell marker OCT4 can be
Rare endocrine Abnormal endocrine
manifestations manifestations in 60% of expressed in some large B-cell lymphomas [79].
cases 6. Gonadoblastoma
Histological Even distributed Severe atypia, variation in Gonadoblastomas contain germ cell as well as sex cord-
features monotonous nuclei nuclear size, and crowding stromal components. Calcification may be prominent.
of nuclei
More than 50% are associated with malignant germ cell
Monotonous cells Prominent pleomorphic cells
Lymphocytic Stromal edema, occasional
tumors, most commonly a dysgerminoma. Careful mor-
infiltration in the small spindled fibroblasts phologic assessment, as well as judicious use of immu-
fibrous stroma nostains, can help to identify the minor sex cord-stromal
Histocytic granulomas No granulomas should be components, if present. Moreover, other features such as
are often seen present heterogeneity of cells, deposition of basement membrane-
Syncytiotrophoblastic Syncytiotrophoblastic giant
giant cells are present cells often seen
like eosinophilic material, and distinctive clinical sce-
in only 5% of cases nario seen with gonadoblastomas can aid in the diagnosis
[80, 81].
7 . Ovarian trophoblastic disease or neoplasia
prominent nucleoli. A panel of immunostains can help. Because of the elevated β-hCG level, trophoblastic dis-
Small-cell carcinoma, hypercalcemic type, lacks expres- ease often enters into the differential diagnosis with dys-
sion of germ cell markers such as OCT4, SALL4, PLAP, germinoma with syncytiotrophoblastic differentiation.
CD117, and D2-40. The presence of a gestational sac and chorionic villi can
2. Embryonal carcinoma essentially rule out dysgerminoma (and choriocarci-
The clinicopathological features of both dysgerminoma noma). Choriocarcinoma is comprised of a dual-cell pop-
and embryonal carcinoma are summarized in Table 8.5. ulation (cytotrophoblasts and syncytiotrophoblasts), as
Both tumors are diffusely positive for SALL4, OCT4, and described elsewhere.
LIN28. Dysgerminoma is positive for CD117 and D2-40,
but negative for CD30. In contrast, embryonal carcinoma 8.4.1.8 Treatment and Prognosis
usually shows the opposite staining pattern, with the Surgical resection of the tumor and complete staging is the
exception of papillary and glandular variants which may first line of treatment of dysgerminoma. Because the tumor
be positive for D2-40. is sensitive to chemoradiation, conservative unilateral
3. Yolk sac tumor salpingo-oophorectomy can be offered to young patients to
Yolk sac tumor can have clear cytoplasm and be positive preserve fertility if no capsular involvement or extra-ovarian
for CD117. Unlike the prominent monotonicity of cells spread is present. Chemotherapy is usually offered to higher
and nuclei commonly seen in dysgerminoma, yolk sac stage patients. Recurrence occurs in 25–35% of patients,
tumor exhibits greater cellular and nuclear variation. It usually within 2 years of diagnosis. Dysgerminoma has an
often contains hyaline bodies, and lacks lymphocytic excellent prognosis, with a >90% 5-year survival rate. The
infiltration and granuloma formation. A panel of immu- most prognostic factor is the size of the tumor (<10 cm).
nostains including OCT4, AFP, and glypican-3 can help Interestingly, nuclear atypia and mitotic figure count have no
to facilitate the diagnosis (Table 8.3). effect on the overall prognosis. For patients with an abnor-
4. Clear-cell carcinoma mal hormonal presentation, total hysterectomy and bilateral
Similar to dysgerminoma, clear-cell carcinoma also has salpingo-oophorectomy are recommended.
clear cytoplasm, and exhibits a solid or nested growth pat-
tern; however, it typically occurs in older patients.
Morphologically, it usually shows a characteristic tubulo- 8.4.2 Yolk Sac Tumor
cystic and papillary growth pattern, as well as endome-
triosis or an adenofibromatous background. Clear-cell 8.4.2.1 Definition
carcinoma is PAX-8, EMA, napsin A, and HNF-1-β posi- Yolk sac tumors have various distinctive morphologic pat-
tive, but is negative for germ cell tumor markers such as terns, which can differentiate into several endodermal struc-
PLAP, CD117, D2-40, OCT4, and SALL4. tures including primitive gut and mesenchyme, derivatives of
5. Lymphoma extra-embryonal tissue (secondary yolk sac and allantois),
Diffuse large B-cell lymphoma should be distinguished and embryonal somatic tissues (intestine and liver).
from dysgerminoma as both can form nest or cords; how- Historically, it has been called “endodermal sinus tumor,”
254 H. Chen et al.
which is not recommend anymore. The majority of yolk sac sac tumors are stage I. Elevated serum AFP can be found in
tumors originate from undifferentiated or pluripotent germ almost all cases, and readings may reach 1000 ng/mL or
cells. A minority of tumors may originate from somatic epi- more; therefore, an elevated serum AFP level should alert
thelial neoplasia through a process of neometaplasia/retrodif pathologists and clinicians to the possibility of this tumor.
ferentiation [82]. Rare cases may derive from
gonadoblastoma. 8.4.2.3 Macroscopic Features
Almost all yolk sac tumors are unilateral, most commonly
8.4.2.2 Clinical Features within the right ovary. Indeed, if both ovaries are involved, it
Yolk sac tumors account for approximately 20% of malig- is strongly suggestive of metastasis. The tumor is usually
nant germ cell tumors, secondary only to dysgerminoma. large, with an average diameter of 15 cm (range 5–35 cm)
The majority of tumors occur in children, adolescents, and [83], and is frequently encapsulated. The cut surface is grey-
young women. About half of these tumors occur after men- yellow, soft, solid, and fleshy with frequent necrosis and
arche, with an average age of 18 years [82]. Patients often hemorrhage (Fig. 8.30). Occasionally, the tumor can be cys-
present with abdominal pain and a pelvic mass; torsion and tic with a honeycomb-like appearance [12, 84]. It is not
rupture occur in approximately 10% of cases. 70% of yolk unusual for the contralateral ovary to contain a mature cystic
teratoma [85].
3. Endodermal sinus pattern myxoid stroma are characteristic of this pattern. It can
The presence of Schiller-Duval (S-D) bodies is charac- be further subdivided into two groups: endometrioid-
teristic of this pattern. Classic S-D bodies are papillae like glands, often with prominent sub-nuclear vacuoles
with a fibro-vascular center which is mantled by cuboi- (Figs. 8.34 and 8.35), and glands similar to primitive
dal, or low columnar, hobnailing tumor cells (Fig. 8.32a,
b). Although S-D bodies are diagnostic for yolk sac
tumor, they are only present in about one third of tumors.
Sometimes anastomosing channels and tubules can form
labyrinth-like structures, which is also known as the fes-
tooning variant of a S-D body (Fig. 8.32c, d). It is impor-
tant to recognize the festooned variant of an S-D body,
especially when classic S-D bodies are not present.
4. Myxoid pattern
This pattern is composed of tumor cells arranged in
glands and cords which are scattered within large
amount of myxoid material (Fig. 8.33).
5. Alveolar-glandular pattern
This pattern contains variable amounts of glandular or
alveolar structures lined by flat or cuboidal tumor cells,
surrounded by myxoid stroma.
6. Glandular pattern
Fig. 8.33 Yolk sac tumor: myxoid pattern. This tumor demonstrates a
The glandular pattern can be seen in approximately one prominent glandular pattern and a large amount of myxoid material in
third of yolk sac tumors. Scattered, simple glands within the tumor stroma
a b
c d
Fig. 8.32 Yolk sac tumor: endodermal sinus pattern. Classical S-D bodies (a, b) and variants of S-D bodies (c, d)
256 H. Chen et al.
a b
c d
Fig. 8.35 Yolk sac tumor. The tumor is composed of microcystic (lower) and endometrioid (upper) patterns (a). AFP is focally positive (b).
Glypican-3 is negative (c). SALL4 is diffusely positive (d)
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 257
a b
Fig. 8.36 Yolk sac tumor: polyvesicular pattern. This tumor has cysts, or vesicles, of varying sizes (a). SALL4 is diffusely positive (b)
Fig. 8.37 Yolk sac tumor: papillary pattern. This tumor has prominent
fibrovascular stalks covered by atypical cells b
Fig. 8.38 Yolk sac tumor: parietal pattern. There is a large number of
8.4.2.5 Immunohistochemistry
extracellular amorphous eosinophilic basement membrane-like depos-
Yolk sac tumors are AFP, glypican-3, SALL4, and LIN28 its around tumor cells and nests of tumor cells
positive (Fig. 8.41). AFP positivity is characteristic of yolk
sac tumor; however its sensitivity is relatively low (only
258 H. Chen et al.
83%), and often it is only focally positive [19]. Tumor cells (28%), and be focally positive for CK7 and EMA. Thus, a
that compose the parietal pattern are usually AFP negative. panel including AFP, glypican-3, CK7, and EMA has
Glypican-3 is positive in 70% of cases. SALL4 and LIN28 only a limited utility in separating yolk sac tumor and
are positive in almost 100% of yolk sac tumors [19, 20], and clear-cell carcinoma. For this purpose the best markers
OCT4 and D2-40 are usually negative. Tumor, especially the are SALL4 and LIN28 [19, 20]. Although clear-cell carci-
solid pattern, can express CD117 [73]. Pan-cytokeratin is noma can be occasionally positive for SALL4 and LIN4,
usually diffusely positive. CK7 and EMA are usually nega- the positivity is usually focal (<10% cells).
tive or only focally positive (i.e., only expressed in <5% 2. Embryonal carcinoma
cells). Tumors deriving from a somatic epithelial tumor are Yolk sac tumor and embryonal carcinoma often coexist.
positive for CK7 and EMA [82]. Embryonal carcinoma can be solid and have clear cyto-
plasm; the tumor cells have larger nuclei, prominent
8.4.2.6 Differential Diagnosis nucleoli, and severe nuclear atypia. No Schiller-Duval
1. Clear-cell carcinoma bodies, hyaline bodies, basement membrane-like mate-
The clinicopathological features of yolk sac tumor and rial, hepatoid cells, or intestinal differentiation is seen in
clear-cell carcinoma are summarized in Table 8.6. Clear- embryonal carcinoma. Embryonal carcinoma is positive
cell carcinoma can express AFP (35%) and glypican-3 for OCT4 and CD30, whereas yolk sac tumor shows the
opposite staining pattern. Unlike yolk sac tumor, embryo-
nal carcinoma is negative for AFP and is rarely glypican-3
positive. SALL4 and LIN28 are strongly positive in both
tumors.
3. Dysgerminoma
Monotonous cells and nuclei are characteristic of dysger-
minoma. Dysgerminoma usually lacks papillae, and hya-
line bodies are rarely seen. OCT4 and CD117 are
positive.
4. Other tumors
Other tumors including endometrioid carcinoma,
juvenile-type granulosa cell tumor, hepatoid adenocarci-
noma, and polyembryoma may have histologic features
that can overlap with those seen in yolk sac tumor.
Fortunately, these features are often focal, and careful
examination of these tumors often reveals more character-
istic histology. In difficult cases application of immunos-
Fig. 8.39 Yolk sac tumor: solid pattern. This tumor shows demon-
strates a solid growth pattern. The tumor cell cytoplasm is clear tains can help to resolve these differential diagnoses.
a b
Fig. 8.40 Yolk sac tumor: hyaline globules. Hyaline globules (a, arrow) are conspicuous both within the cytoplasm and stroma. They are positive
for PAS-D (b)
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 259
a b
c d
Fig. 8.41 Yolk sac tumor: alveolar-glandular pattern. (a) Immunohistochemical staining is positive for AFP (b), SALL4 (c), and LIN28 (d)
Table 8.6 Clinicopathological features of yolk sac tumor and clear- 8.4.2.7 Treatment and Prognosis
cell carcinoma
Yolk sac tumor is a highly aggressive tumor. Local spread or
Yolk sac tumor Clear-cell carcinoma distant metastasis may occur even at early stage. At the time
Clinical Most cases occur in Older women, of diagnosis, local invasion or peritoneal spread is present in
features women <30 years. Only mostly in their 50s
rarely occurs in patients (40–70 years) approximately 30% of patients. Even when tumor is grossly
older than 40 or in confined to the ovary, occult metastasis can often be seen.
postmenopausal women Lymphatic metastasis is often seen in pelvic and periaortic
Macroscopic Grey-yellow, solid with Solid or cystic with lymph nodes. Hematogenous dissemination to lung and liver
features areas of necrosis and yellow, yellow-tan,
is the most common form of metastatic spread. This tumor is
hemorrhage, and areas of or brown papillary
myxoid (gelatinous) excrescences. sensitive to chemotherapy, and overall survival is more than
change Evidence of 80%. Tumors with a parietal, hepatoid, or glandular (mainly
endometriosis may intestinal) pattern are less sensitive to chemotherapy, and
be present
have a relatively poor prognosis. Yolk sac tumor is resistant
Characteristic Schiller-Duval bodies; Broad-based
morphologic hobnail cells; solid areas papillae lined by to radiation therapy.
features often composed of small hobnail cells with
primitive cells with clear cytoplasm;
frequent mitosis rare mitotic figures; 8.4.3 Embryonal Carcinoma
Polyvesicular areas cysts are lined by
composed of various clear or hobnail
shaped cysts; cysts are lined cells 8.4.3.1 Definition
by mesothelial-like Embryonal carcinoma is a rare type of germ cell tumor,
epithelium, and filled with exhibiting rudimentary epithelial differentiation. This tumor
PAS-positive material
is morphologically identical to its testicular counterpart.
260 H. Chen et al.
Fig. 8.42 Embryonal carcinoma. The tumor has distinct “epithelial” Fig. 8.43 Embryonal carcinoma. The primitive tumor cells in this
appearance with irregular silt-like glands lined with primitive tumor example are growing a solid pattern and have characteristic large vesic-
cells ular nuclei with prominent nucleoli
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 261
a b
Fig. 8.44 Embryonic carcinoma. The tumor cells are positive for CD30 in a membranous pattern (a) and OCT4 in the nuclei (b)
carcinoma is positive for OCT4 and CD30, whereas yolk is composed of a mixture of embryonal carcinoma and
sac tumor is negative for both. Unlike yolk sac tumor, yolk sac tumor.
embryonal carcinoma is negative for AFP and is rarely
glypican-3 positive. SALL4 and LIN28 are strongly posi- 8.4.3.8 Treatment and Prognosis
tive in both tumors. Embryonal carcinoma is a highly aggressive cancer, and is
3. Dysgerminoma usually treated with surgical removal and chemotherapy.
Monotonicity of nuclei and cells is a characteristic feature Fortunately, this tumor is sensitive to chemotherapy; how-
of dysgerminoma. Other characteristic histologic fea- ever it is resistant to radiation therapy. Patients with extra-
tures, such as fibrous septa, lymphocytic infiltrates, and ovarian spread have a poor prognosis.
histiocytic granuloma, are usually absent in embryonal
carcinoma. In addition, dysgerminoma generally lacks
papillary or glandular structures, which are commonly 8.4.4 Non-gestational Choriocarcinoma
seen in embryonal carcinoma. A panel of immunostains
including CD117 (negative in embryonal), SOX2, and 8.4.4.1 Definition
CD30 (both positive in embryonal) can help to make the A malignant neoplasm of germ cell origin composed of cyto-
correct diagnosis. trophoblastic and syncytiotrophoblastic cells.
4 . Choriocarcinoma
Choriocarcinoma should be distinguished from embryo- 8.4.4.2 Clinical Features
nal carcinoma with a prominent syncytiotrophoblastic Pure, non-gestational choriocarcinoma is an exceedingly
component. OCT4 and CD30 are diffusely positive in rare and highly aggressive tumor accounting for only 1% of
embryonal carcinoma, and are negative in the cytotropho- malignant germ cell tumors [90]. Pure tumors are most com-
blast/intermediate trophoblastic components of chorio- mon in children or young women before puberty, but occa-
carcinoma. In addition, SALL4 and LIN28 are diffusely sionally may occur in postmenopausal women. Rare cases
positive in embryonal carcinoma, and negative, or only are associated with gonadal dysplasia (46XY, 45XO/46XY)
weakly positive, in the cytotrophoblast/intermediate tro- [91, 92]. Much more commonly seen than the pure form, this
phoblastic components of choriocarcinoma. tumor is frequently a component of a mixed malignant germ
5 . Polyembryoma cell tumor. Most patients present with pseudo-puberty, vagi-
Polyembryoma is a tumor that contains innumerable nal bleeding, or signs mimicking an ectopic pregnancy
embryoid bodies composed of two cavities separated by (young women). Elevated serum or urine β-hCG is almost
embryonic germ disc. The embryonic disc is composed of always present [93].
a band of primitive cells similar to those seen in embryo-
nal carcinoma along with a smaller band composed of 8.4.4.3 Macroscopic Features
flattened yolk sac-type cells that lie underneath. This Choriocarcinoma is usually a large (often >6 cm), unilateral
tumor is considered a special type of germ cell tumor that tumor. The cut surface is grey-tan, solid, or solid and cystic,
262 H. Chen et al.
and displays abundant hemorrhage and necrosis. Luteinized a previous pregnancy. Features such as bilateral ovarian
nodules or cysts may be seen in the adjacent ovarian tissue. involvement and multi-organ metastasis are helpful and
may provide a hint to the correct diagnosis.
8.4.4.4 Microscopic Features 3 . Primary ovarian pregnancy-associated choriocarcinoma
Morphologically, non-gestational choriocarcinoma is identi- In the extraordinarily uncommon setting of a choriocarci-
cal to gestational choriocarcinoma. The tumor is composed of noma arising in an ovarian pregnancy, the history of a pre-
an intimate combination of cytotrophoblast and syncytiotro- vious ovarian pregnancy may be the only clue to the
phoblast (Fig. 8.45). The cytotrophoblast and syncytiotropho- origin of the malignancy.
blasts are admixed in a distinctive plexiform pattern, with the
latter located at the periphery, surrounding clusters of cytotro- 8.4.4.7 Treatment and Prognosis
phoblasts. Cytotrophoblastic cells are mononucleated with Non-gestational, pure, choriocarcinoma has a poor prognosis,
vesicular nuclei, abundant clear cytoplasm, and well-defined which is often worse than seen in gestational choriocarcinoma.
cell borders. Syncytiotrophoblast cells are multinucleated with Unilateral salpingo-oophorectomy combined with cisplatin-
dark nuclei and abundant, vacuolated, amphophilic cytoplasm. based multi-reagent chemotherapy has shown promising
Hemorrhage and necrosis are often prominent (Fig. 8.46). results.
8.4.4.5 Immunohistochemistry
Syncytiotrophoblastic cells are positive for β-hCG, whereas 8.4.5 Polyembryomas
cytotrophoblast cells are negative. The tumor cells are posi-
tive for cytokeratin, but negative for EMA. SALL4 and 8.4.5.1 Definition
LIN28 may be focally positive. Polyembryoma is an exceedingly rare malignant germ cell
tumor composed entirely (or predominantly) of embryoid bod-
8.4.4.6 Differential Diagnosis ies resembling an embryo at 15–16 days of gestation [94, 95].
1. Other germ cell tumors (such as dysgerminoma, So far, no pure polyembryoma has been reported as all reported
embryonal carcinoma) with syncytiotrophoblastic cases have consisted of polyembryoma as the dominant compo-
component nent or have been found coexisting with other germ cell tumors
Other germ cell tumors may contain syncytiotrophoblas- (mainly immature teratoma, occasionally embryonal carci-
tic cells; however, they lack cytotrophoblastic cells. noma, yolk sac tumor, dysgerminoma, choriocarcinoma, etc.)
Furthermore, when germ cell tumors contain syncytiotro- [96, 97]. In the 2014 WHO classification of ovarian germ cell
phoblasts they are usually scattered and not present in the tumors, it is classified as a variant of immature teratoma.
same quantity as those seen in choriocarcinoma.
2. Ovarian metastatic gestational choriocarcinoma 8.4.5.2 Clinical Features
On occasion a primary tumor of the uterus may metasta- Most of these tumors have occurred in children and young
size to the ovary. In this case there is usually a history of women. Reported patients commonly present with a pelvic
Fig. 8.45 Choriocarcinoma. These tumors are defined by their admix- Fig. 8.46 Choriocarcinoma. Large areas of hemorrhagic necrosis are
ture of conspicuous cytotrophoblastic and syncytiotrophoblastic cells common
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 263
mass and occasional endocrine abnormalities, such as pre- 8.4.5.4 Microscopic Features
cocious puberty in children. Some of these tumors can The defining histologic feature of this tumor is the presence
secrete AFP [98], which may be indicative of a YST of numerous embryoid bodies (Fig. 8.48). Well-developed
component. embryoid bodies are composed of two cavities (recapitulat-
ing amniotic and yolk sac cavities) separated by embryonic
8.4.5.3 Macroscopic Features germ disc. Less developed embryoid bodies may only con-
Polyembryoma is usually a unilateral, large (commonly tain a poorly formed bilayer of germ discs. The embryoid
diameter >10 cm), encapsulated mass with a smooth or nod- bodies are surrounded by an abundant edematous to myxoid
ular surface. The cut surface is grey-tan and solid with stroma with prominent blood vessels. Minor teratomatous
honeycomb-like cystic areas (Fig. 8.47). Areas of hemor- components are almost always present, although they typi-
rhage and necrosis are common. cally occupy <10% of the tumor volume. Scattered syncytio-
trophoblasts may be seen.
Embryonic Amniotic
germ disc cavity
a b
Fig. 8.48 Polyembryoma. The tumor has multiple embryoid bodies (a). A well-differentiated embryoid body contains blastoderm, and an amni-
otic and primitive yolk sac cavity (b)
264 H. Chen et al.
malignant germ cell tumors. Surgery combined with chemo- ous tumor components. Extensive sampling of these
therapy is the recommended treatment. tumors, focusing on grossly heterogeneous areas of the
tumor, is critical in order to identify all of the tumor
components.
8.4.6 Mixed Malignant Germ Cell Tumors
8.4.6.4 Microscopic Features
8.4.6.1 Definition The morphologic features of each individual component are
Mixed malignant germ cell tumors (MGCTs) are defined by identical to those seen in their respective pure forms. The
the presence of two or more types of malignant germ cell most common components include dysgerminoma, yolk sac
components within a tumor. The most common combination tumor, and immature teratoma. Embryonal carcinoma and
is dysgerminoma and yolk sac tumor [99]. Mature teratoma choriocarcinoma are relatively rare. The most common
with one malignant germ cell component, malignant transfor- combination is dysgerminoma and yolk sac tumor
mation of mature teratoma, gonadoblastoma, and mixed germ (Fig. 8.49); other common combinations include dysgermi-
cell-sex cord-stromal tumors are not included in this entity. noma and embryonal carcinoma (Fig. 8.50), yolk sac tumor
and embryonal carcinoma (Fig. 8.51), and immature tera-
8.4.6.2 Clinical Features toma and yolk sac tumor. The various components may be
MGCTs account for 8–20% of all ovarian malignant germ intimately admixed or form relatively discrete zones within
cell tumors [88]. They occur mostly in adolescents (average each tumor.
age of 16 years), and very rarely in postmenopausal women
[100]. Patients may present with abdominal pain, a pelvic/ 8.4.6.5 Treatment and Prognosis
abdominal mass, or bloating. Some patients manifest symp- The size and composition of the tumor are the most impor-
toms of endocrine abnormalities. tant prognostic factors. When the tumor is <10 cm, the prog-
nosis is usually good [101]. Surgery combined with
8.4.6.3 Macroscopic Features chemotherapy is the recommended treatment. The choice of
MGCTs are usually large (average diameter 15 cm) and chemotherapy depends on the relative composition of the
unilateral. The cut surface varies depending on the vari- malignant components of the tumor.
2 cm
Fig. 8.49 Mixed germ cell tumor. Different areas of differentiation are
grossly identifiable. In this case, the upper left area contains yolk sac
tumor and the lower right area is composed of dysgerminoma
Fig. 8.50 Mixed germ cell tumor (dysgerminoma and embryonal car-
cinoma). The left area is composed of dysgerminoma, and the right area
is embryonal carcinoma which displays significant nuclear atypia
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 265
a b
c d
Fig. 8.51 Mixed germ cell tumor (yolk sac tumor and embryonal car- large nuclei and prominent nucleoli (a, b). Immunohistochemical stain-
cinoma). In this example the yolk sac tumor is composed of cells with ing exhibits OCT4 expression in the embryonal carcinoma (c), and
small nuclei, microcystic change, and myxoid stroma, while the embry- SALL4 expression in the yolk sac tumor and embryonal carcinoma (d)
onal carcinoma component has a glandular architecture and cells with
8.5 Germ Cell-Sex Cord-Stromal Tumors considered to be clinically benign, and is often considered as
an “in situ” malignant germ cell tumor due to its high rate
Germ cell-sex cord-stromal tumors are composed of germ (60%) of associated malignant transformation.
cell and sex cord components, and can be subcategorized
into two groups: (1) gonadoblastoma, including gonadoblas- 8.5.1.2 Clinical Features
toma with a malignant germ cell tumor, and (2) germ cell-sex Gonadoblastomas occur mostly in patients <15 years of age
cord-stromal tumors, unclassified. Tumors of the first group (average age 18 years, range 1–38). They arise almost exclu-
arise almost exclusively in dysgenic gonads, whereas tumors sively in the setting of dysgenetic gonads. Clinical presenta-
of second group are not associated with dysgenic gonads. tions include amenorrhea or gynecologic tract developmental
abnormalities. Virilization of a phenotypic female is the most
common presentation (45%), followed by hypospadias in
8.5.1 Gonadoblastoma, Including phenotypic male (20%) [102]. Abdominal swelling or pain
Gonadoblastoma/Malignant Germ Cell may be present, and is usually due to malignant transforma-
Tumor tion of tumor.
streak gonads. The tumors are usually small (<3 cm). The cut toma” [80]; however, this terminology is controversial [81].
surface is solid, tan to yellow, soft to firm, and sometimes In fact, this diffuse growth pattern can be seen in most
gritty. When associated with malignant germ cell tumor, the gonadoblastomas in variable quantities. These tumors can be
tumor is usually larger with grossly appreciable areas of further subdivided into three patterns: (1) the solid or expans-
divergent differentiation. ile pattern, in which germ cells form large solid nests; (2) the
anastomosing pattern composed of aggregates of small nests;
8.5.1.4 Microscopic Features (3) and the corded, or trabecular, pattern. Sex cord-stromal
cells may be conspicuous in a diffuse-type tumor; however,
Classic-Type Gonadoblastoma immunostains may be needed to identify a component of sex
Gonadoblastoma is usually composed of discrete nests of cord-stromal tumor, and distinguish it from a dysgerminoma.
cells separated by fibrous stroma (Fig. 8.52). The nests con- The majority of gonadoblastomas with a (46 XY) genotype
tain primitive germ cells and sex cord cells, with the latter have a malignant germ cell tumor component, most com-
typically located at the periphery of the nest or around small monly dysgerminoma (57%), followed by yolk sac tumor
acini. The primitive germ cells are similar to the tumor cells (4%) and embryonal carcinoma (4%). A recent series of
that comprise dysgerminoma, with abundant, clear to fine gonadoblastoma demonstrated that the majority (12/14) of
granular cytoplasm, round vesicular nuclei and prominent gonadoblastomas with (46, XX) contained a malignant germ
nucleoli. Mitotic figures may be present (and may even be cell tumor as well (Fig. 8.53) [103].
brisk). Sex cord cells are usually located at the periphery of
the germ cell nest or around individual or clustered germ 8.5.1.5 Immunohistochemistry
cells. Small acini, or cysts, filled with amorphous eosino- Gonadoblastoma is composed of two cell types (Fig. 8.54a),
philic PAS-positive material (similar to Call-Exner bodies) including primitive germ cells that are positive for SALL4
may form. Variable amounts of basement membrane-like (Fig. 8.54b), LIN28, OCT4 (Fig. 8.54c), PLAP, CD117, and
material are present within the nests. Calcifications are com- D2–40 [19] and sex cord-stromal cells that are positive for
mon, ranging from small flecks to large mulberry-like aggre- WT-1 (Fig. 8.54d), inhibin, calretinin, SF-1, FOXL2, and
gates. Luteinized cells or Leydig-like cells are present in SOX9 [80]. Immunohistochemical stains may be helpful in
two-thirds of tumors; however, unlike Leydig cells, these differentiating these comingling components.
cells lack Reinke crystals. Tumor regression is common, and
presents as a decrease in cell mass and a relative increase in 8.5.1.6 Genetic Profiles
basement membrane-like material and calcification. Half of patients with this tumor have a 46,XY karyotype; 1/4
Occasionally, calcifications are the only histologic finding. of patients have a 45, XO/46, XY mosaic karyotype; other
less common karyotypes include 46,XX, 45,XO/46,XY, and
Diffuse-Type Gonadoblastoma 45,XO (Turner syndrome) [102]. The testis-specific protein
Occasionally, a gonadoblastoma may grow in a diffuse man- Y-encoded 1 (TSPY1)gene increases the susceptibility to
ner, in which it has been termed “dissecting gonadoblas- gonadoblastoma in patients with dysgenetic gonads [104].
a b
Fig. 8.52 Gonadoblastoma. This tumor is arranged in small round lumens with amorphous hyaline material (a) and tubules (b). The dysgermi-
noma- or seminoma-like cells are present within the tubules (a, right). There are conspicuous psammoma-like calcifications (a)
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 267
a b
c d
Fig. 8.54 Gonadoblastoma. These tumors consist of two cell types (a). The germ cells are positive for SALL4 (b) and OCT4 (c), while the sex
cord-stromal cells are positive for WT-1 (d)
268 H. Chen et al.
nested architecture, deposition of basement membrane- OCT4 and SALL4, whereas the sex cord-stromal component
like material, and calcification. Unlike gonadoblastoma, is usually negative for these markers, but positive for SF-1,
these mixed tumors are typically unilateral, and the inhibin, and calretinin [108].
patient will have a normal karyotype and developmentally
normal gonads. 8.5.2.5 Differential Diagnosis
1. Gonadoblastoma
8.5.1.8 Treatment and Prognosis These tumors display a distinct nested architecture, depo-
Gonadoblastoma is a clinically benign tumor when it is unas- sition of basement membrane-like material, and extensive
sociated with an overt malignancy; however, patients do mulberry-like calcifications. These features are character-
need a karyotype analysis. Most patients should undergo a istic of gonadoblastoma and are absent in unclassified
prophylactic bilateral gonadectomy to prevent the develop- mixed germ cell and sex cord-stromal tumors. In addition,
ment of a malignant germ cell tumor. The treatment and patients with gonadoblastoma have dysgenetic gonads
prognosis of a tumor with malignant transformation depend and abnormal karyotypes (Table 8.7).
on the type, size, and stage of the malignant components. 2. Dysgerminoma
When germ cells are the predominant component of the
tumor, mixed germ cell and sex cord-stromal tumors need to
8.5.2 M
ixed Germ Cell and Sex Cord-Stromal be distinguished from dysgerminomas. Careful examina-
Tumor, Unclassified tion, additional sections, and immunohistochemical stain-
ing may help to identify sex cord-stromal components.
8.5.2.1 Definition
Mixed germ cell and sex cord-stromal tumors are composed 8.5.2.6 Treatment and Prognosis
of a mixture of germ cells and sex cord-stroma components, Although these tumors are mostly benign, they do have
but they lack the distinctive features of a gonadoblastoma. malignant potential. Malignant transformation or metastasis
Namely, unlike their counterparts with gonadoblastoma, the may occur, but the risk is significantly lower than that
patients with these tumors are genetically and phenotypi- ascribed to gonadoblastoma. Conservative unilateral oopho-
cally normal. rectomy is the recommended treatment, and metastatic
tumors are sensitive to chemotherapy.
8.5.2.2 Clinical Features
Table 8.7 Comparison between gonadoblastoma and mixed germ cell
This is a rare tumor that most commonly occurs in infants and sex cord-stromal tumor, unclassified
and children younger than 10 years of age; however, rare
Mixed germ cell and
cases have been reported in postmenopausal women [105– sex cord-stromal
107]. Abdominal pain is the most common clinical presenta- Gonadoblastoma tumor, unclassified
tion. Some patients can present with an abdominal mass, and Clinical Most commonly occur in Almost all patients
a quarter of the patients can have precocious features young women 10–30 years are children
of age; majority of patients <10 years of age;
pseudo-puberty. have abnormal karyotypes, normal 46, XX
symptoms of virilization, karyotype; normal
8.5.2.3 Macroscopic Features amenorrhea, and other female phenotype
Mixed germ cell and sex cord-stromal tumors are typically endocrine abnormalities
unilateral, large tumors (diameter 7–18 cm). The cut surface Macroscopic Small tumor (usually <3 cm) Large tumor
features Bilateral in 40% of patients Almost always
is usually solid or solid and cystic, grey-tan to grey-yellow, Calcifications present in unilateral
and devoid of necrosis and calcification. 45% of tumors Calcification absent
Microscopic Distinct nested architecture Variable
8.5.2.4 Microscopic Features features Hyaline bodies and speckled architecture;
calcification common however there is a
These tumors are definitionally composed of a mixture of
Stromal luteinization is lack of nested
germ cells and sex cord-stromal components, but lack the common architecture
distinctive features of gonadoblastoma. As such, it is some- 60% tumors contain Hyaline bodies and
what of a diagnosis of exclusion. Germ cells may form cords, dysgerminoma or other speckled
malignant germ cell calcification
trabeculae, solid or hollow tubules (sometimes, similar to sex
components Rare stromal
cord tumor with annular tubules), cysts, or haphazard struc-
luteinization may
tures. Any or all of the abovementioned patterns can be seen be present
within one tumor. Nested architecture, deposition of base- Not associated with
ment membrane-like material, and calcification are typically malignant germ cell
absent. The germ cell component is typically positive for tumors
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 269
41. Robboy SJ, Shaco-Levy R, Peng RY, et al. Malignant struma ova- 61. Gorski JW, Taylor JS, Zhang J, Liu J, Jazaeri AA. Hormonal based
rii: an analysis of 88 cases, including 27 with extraovarian spread. treatment of ovarian anaplastic ependymoma with anastrozole.
Int J Gynecol Pathol. 2009;28:405–22. Gynecol Oncol Rep. 2017;20:93–6.
42. Roth LM, Karseladze AI. Highly differentiated follicular carci- 62. Ud Din N, Memon A, Aftab K, et al. Oligodendroglioma arising in
noma arising from struma ovarii: a report of 3 cases, a review the glial component of ovarian teratomas: a series of six cases and
of the literature, and a reassessment of so-called peritoneal stru- review of literature. J Clin Pathol. 2012;65:631–4.
mosis. Int J Gynecol Pathol. 2008;27:213–22. 63. Idowu MO, Rosenblum MK, Wei XJ, Edgar MA, Soslow
43. Cong H, Li T, Chen G, et al. Missed initial diagnosis of malignant RA. Ependymomas of the central nervous system and adult extra-
Struma Ovarii containing follicular thyroid carcinoma: metastatic axial ependymomas are morphologically and immunohistochemi-
pulmonary recurrence 17 yr after ovariectomy. Int J Gynecol cally distinct—a comparative study with assessment of ovarian
Pathol. 2015;34:445–9. carcinomas for expression of glial fibrillary acidic protein. Am J
44. Schmidt J, Derr V, Heinrich MC, et al. BRAF in papillary thy- Surg Pathol. 2008;32:710–8.
roid carcinoma of ovary (struma ovarii). Am J Surg Pathol. 64. Chumas JC, Scully RE. Sebaceous tumors arising in ovarian der-
2007;31:1337–43. moid cysts. Int J Gynecol Pathol. 1991;10:356–63.
45. Tan A, Stewart CJ, Garrett KL, Rye M, Cohen PA. Novel BRAF 65. Akhtar M, Young I, Brody H. Anterior pituitary component in
and KRAS mutations in papillary thyroid carcinoma arising in benign cystic ovarian teratomas. Report of three cases. Am J Clin
Struma Ovarii. Endocr Pathol. 2015;26:296–301. Pathol. 1975;64:14–9.
46. Boutross-Tadross O, Saleh R, Asa SL. Follicular variant papil- 66. Palmer PE, Bogojavlensky S, Bhan AK, Scully RE. Prolactinoma
lary thyroid carcinoma arising in struma ovarii. Endocr Pathol. in wall of ovarian dermoid cyst with hyperprolactinemia. Obstet
2007;18:182–6. Gynecol. 1990;75:540–3.
47. Leong A, Roche PJ, Paliouras M, et al. Coexistence of malignant 67. Al-Bazzaz S, Karamchandani J, Mocarski E, et al. Ectopic
struma ovarii and cervical papillary thyroid carcinoma. J Clin prolactin-producing pituitary adenoma in a benign ovarian cystic
Endocrinol Metab. 2013;98:4599–605. teratoma. Endocr Pathol. 2014;25:321–3.
48. Shaco-Levy R, Bean SM, Bentley RC, Robboy SJ. Natural history 68. Borup K, Leisgard Rasmussen K, Schierup L, Moller
of biologically malignant struma ovarii: analysis of 27 cases with JC. Amelanotic malignant melanoma arising in an ovarian der-
extraovarian spread. Int J Gynecol Pathol. 2010;29:212–27. moid cyst. Acta Obstet Gynecol Scand. 1992;71:242–4.
49. Davis KP, Hartmann LK, Keeney GL, Shapiro H. Primary ovarian 69. Shen X, Fan Y, Cao S. Primary malignant melanoma arising in an
carcinoid tumors. Gynecol Oncol. 1996;61:259–65. ovarian cystic teratoma. Melanoma Res. 2017;27:601–6.
50. Soga J, Osaka M, Yakuwa Y. Carcinoids of the ovary: an analysis 70. Kuroda N, Hirano K, Inui Y, et al. Compound melanocytic nevus
of 329 reported cases. J Exp Clin Cancer Res. 2000;19:271–80. arising in a mature cystic teratoma of the ovary. Pathol Int.
51. Baker PM, Oliva E, Young RH, Talerman A, Scully RE. Ovarian 2001;51:902–4.
mucinous carcinoids including some with a carcinomatous com- 71. Quero-Hernandez A, Estrada-Correa R, Tenorio-Rodriguez H,
ponent: a report of 17 cases. Am J Surg Pathol. 2001;25:557–68. Alvarez-Solis RM. [Malignant germ cell ovarian tumors: clini-
52. Shin JH, Bae JH, Lee A, et al. CK7, CK20, CDX2 and MUC2 cal characteristics, treatment and outcome] Cirugia y cirujanos.
Immunohistochemical staining used to distinguish metastatic 2007;75:81–5.
colorectal carcinoma involving ovary from primary ovarian muci- 72. Zaloudek CJ, Tavassoli FA, Norris HJ. Dysgerminoma with syncy-
nous adenocarcinoma. Jpn J Clin Oncol. 2010;40:208–13. tiotrophoblastic giant cells. A histologically and clinically distinc-
53. Bai X, Li N, Wang F, Li S, Yu Q. Primary ovarian trabecular car- tive subtype of dysgerminoma. Am J Surg Pathol. 1981;5:361–7.
cinoid tumor: a case report and literature review. Arch Gynecol 73. Kao CS, Idrees MT, Young RH, Ulbright TM. Solid pattern yolk
Obstet. 2010;282:407–11. sac tumor: a morphologic and immunohistochemical study of 52
54. Niu D, Li Z, Sun L, Cao D. Carcinoid arising from the teratoma- cases. Am J Surg Pathol. 2012;36:360–7.
tous bronchial mucosa in a mature cystic teratoma of the ovary: a 74. Sung MT, Wang M, MacLennan GT, et al. Histogenesis of sarco-
case report. Int J Gynecol Pathol. 2018;37(2):123–7. matoid urothelial carcinoma of the urinary bladder: evidence for
55. Rabban JT, Lerwill MF, McCluggage WG, Grenert JP, Zaloudek a common clonal origin with divergent differentiation. J Pathol.
CJ. Primary ovarian carcinoid tumors may express CDX-2: a 2007;211:420–30.
potential pitfall in distinction from metastatic intestinal carcinoid 75. Cheng L, Thomas A, Roth LM, et al. OCT4: a novel biomarker for
tumors involving the ovary. Int J Gynecol Pathol. 2009;28:41–8. dysgerminoma of the ovary. Am J Surg Pathol. 2004;28:1341–6.
56. Desouki MM, Lioyd J, Xu H, et al. CDX2 may be a useful 76. Chang MC, Vargas SO, Hornick JL, et al. Embryonic stem cell
marker to distinguish primary ovarian carcinoid from gas- transcription factors and D2-40 (podoplanin) as diagnostic
trointestinal metastatic carcinoids to the ovary. Hum Pathol. immunohistochemical markers in ovarian germ cell tumors. Int J
2013;44:2536–41. Gynecol Pathol. 2009;28:347–55.
57. Yang C, Sun L, Zhang L, et al. Diagnostic utility of SATB2 in met- 77. Hoei-Hansen CE, Kraggerud SM, Abeler VM, et al. Ovarian
astatic Krukenberg tumors of the ovary: an immunohistochemical dysgerminomas are characterised by frequent KIT mutations
study of 70 cases with comparison to CDX2, CK7, CK20, chromo- and abundant expression of pluripotency markers. Mol Cancer.
granin, and synaptophysin. Am J Surg Pathol. 2018;42(2):160–71. 2007;6:12.
58. Kleinman GM, Young RH, Scully RE. Primary neuroectoder- 78. Cossu-Rocca P, Zhang S, Roth LM, et al. Chromosome 12p abnor-
mal tumors of the ovary. A report of 25 cases. Am J Surg Pathol. malities in dysgerminoma of the ovary: a FISH analysis. Mod
1993;17:764–78. Pathol. 2006;19:611–5.
59. Liang L, Olar A, Niu N, et al. Primary glial and neuronal tumors 79. Williams AS, Shawwa A, Merrimen J, Dakin Hache K. Expression
of the ovary or peritoneum: a Clinicopathologic study of 11 cases. of OCT4 and SALL4 in diffuse large B-cell lymphoma: an analy-
Am J Surg Pathol. 2016;40:847–56. sis of 145 consecutive cases and testicular lymphomas. Am J Surg
60. Guerrieri C, Jarlsfelt I. Ependymoma of the ovary. A case report Pathol. 2016;40:950–7.
with immunohistochemical, ultrastructural, and DNA cytometric 80. Kao CS, Idrees MT, Young RH, Ulbright TM. “Dissecting
findings, as well as histogenetic considerations. Am J Surg Pathol. Gonadoblastoma” of Scully: a morphologic variant that often
1993;17:623–32. mimics germinoma. Am J Surg Pathol. 2016;40:1417–23.
8 Germ Cell Tumors and Mixed Germ Cell-Sex Cord-Stromal Tumors of the Ovary 271
81. Roth LM, Cheng L. Classical gonadoblastoma: its relationship 95. Young RH, Stall JN, Sevestre H. The Polyembryoma: one of the
to the ‘dissecting’ variant and undifferentiated gonadal tissue. most intriguing human neoplasms, with comments on the investi-
Histopathology. 2018;72(4):545–55. gator who brought it to light, Albert Peyron. Int J Gynecol Pathol.
82. McNamee T, Damato S, McCluggage WG. Yolk sac tumours of 2016;35:93–105.
the female genital tract in older adults derive commonly from 96. Nishida T, Oda T, Sugiyama T, et al. Ovarian mixed germ cell
somatic epithelial neoplasms: somatically derived yolk sac tumor comprising polyembryoma and choriocarcinoma. Eur J
tumours. Histopathology. 2016;69:739–51. Obstet Gynecol Reprod Biol. 1998;78:95–7.
83. Choi HJ, Moon MH, Kim SH, et al. Yolk sac tumor of the ovary: 97. King ME, Hubbell MJ, Talerman A. Mixed germ cell tumor of the
CT findings. Abdom Imaging. 2008;33:736–9. ovary with a prominent polyembryoma component. Int J Gynecol
84. Young RH. Ovarian tumors and tumor-like lesions in the first three Pathol. 1991;10:88–95.
decades. Semin Diagn Pathol. 2014;31:382–426. 98. Takeda A, Ishizuka T, Goto T, et al. Polyembryoma of ovary pro-
85. Vos A, Oosterhuis JW, de Jong B, et al. Karyotyping and DNA ducing alpha-fetoprotein and HCG: immunoperoxidase and elec-
flow cytometry of metastatic ovarian yolk sac tumor. Cancer tron microscopic study. Cancer. 1982;49:1878–89.
Genet Cytogenet. 1990;44:223–8. 99. Topuz S, Iyibozkurt AC, Akhan SE, et al. Malignant germ cell
86. Lee KH, Lee IH, Kim BG, et al. Clinicopathologic characteristics tumors of the ovary: a review of 41 cases and risk factors for recur-
of malignant germ cell tumors in the ovaries of Korean women: rence. Eur J Gynaecol Oncol. 2008;29:635–7.
a Korean Gynecologic Oncology Group Study. Int J Gynecol 100. Goyal LD, Kaur S, Kawatra K. Malignant mixed germ cell tumour
Cancer. 2009;19:84–7. of ovary--an unusual combination and review of literature. J
87. Ulbright TM, Roth LM, Brodhecker CA. Yolk sac differentiation Ovarian Res. 2014;7:91.
in germ cell tumors. A morphologic study of 50 cases with empha- 101. Smith HO, Berwick M, Verschraegen CF, et al. Incidence and
sis on hepatic, enteric, and parietal yolk sac features. Am J Surg survival rates for female malignant germ cell tumors. Obstet
Pathol. 1986;10:151–64. Gynecol. 2006;107:1075–85.
88. De Backer A, Madern GC, Oosterhuis JW, Hakvoort-Cammel FG, 102. Scully RE. Gonadoblastoma. A review of 74 cases. Cancer.
Hazebroek FW. Ovarian germ cell tumors in children: a clinical 1970;25:1340–56.
study of 66 patients. Pediatr Blood Cancer. 2006;46:459–64. 103. Gru AA, Williams ES, Cao D. Mixed gonadal germ cell tumor
89. Liu A, Cheng L, Du J, et al. Diagnostic utility of novel stem cell composed of a spermatocytic tumor-like component and germi-
markers SALL4, OCT4, NANOG, SOX2, UTF1, and TCL1 in noma arising in gonadoblastoma in a phenotypic woman with a
primary mediastinal germ cell tumors. Am J Surg Pathol. 46, XX peripheral karyotype: report of the first case. Am J Surg
2010;34:697–706. Pathol. 2017;41:1290–7.
90. Kurman RJ, Norris HJ. Malignant germ cell tumors of the ovary. 104. Tsuchiya K, Reijo R, Page DC, Disteche CM. Gonadoblastoma:
Hum Pathol. 1977;8:551–64. molecular definition of the susceptibility region on the Y chromo-
91. Chou HH, Lai CH, Wang PN, et al. Combination of high-dose some. Am J Hum Genet. 1995;57:1400–7.
chemotherapy, autologous bone marrow/peripheral blood stem 105. Lacson AG, Gillis DA, Shawwa A. Malignant mixed germ-cell-
cell transplantation, and thoracoscopic surgery in refractory non- sex cord-stromal tumors of the ovary associated with isosexual
gestational choriocarcinoma of a 45XO/46XY female: a case precocious puberty. Cancer. 1988;61:2122–33.
report. Gynecol Oncol. 1997;64:521–5. 106. De Backer A, Madern GC, Pieters R, et al. Influence of tumor site
92. Goswami D, Sharma K, Zutshi V, Tempe A, Nigam and histology on long-term survival in 193 children with extracra-
S. Nongestational pure ovarian choriocarcinoma with contralat- nial germ cell tumors. Eur J Pediatr Surg. 2008;18:1–6.
eral teratoma. Gynecol Oncol. 2001;80:262–6. 107. Jacobsen GK, Braendstrup O, Talerman A. Bilateral mixed germ
93. Lv L, Yang K, Wu H, Lou J, Peng Z. Pure choriocarcinoma of the cell sex-cord stroma tumour in a young adult woman. Case report.
ovary: a case report. J Gynecol Oncol. 2011;22:135–9. APMIS. 1991;23:132–7.
94. Simard LC. Polyembryonic embryoma of the ovary of partheno- 108. Roth LM, Cheng L. On the histogenesis of mixed germ cell-sex
genetic origin. Cancer. 1957;10:215–23. cord stromal tumour of the gonads. J Clin Pathol. 2017;70:222–7.
Sex Cord-Stromal Tumors of the Ovary
9
Mohamed Mokhtar Desouki
Abstract Keywords
Sex cord-stromal tumors (SCSTs) are rare neoplasms Sex cord stromal tumors · Granulosa cell tumors ·
derived from, or display differentiation towards primi- Sertoli-leydig cell tumors
tive sex cords or stromal cells. The tumors in this cate-
gory exhibit a great diversity of patterns with a long
array of mimickers, and accordingly may pose a diag-
nostic challenge. Some of the derivative cells are Abbreviations
involved in hormone production under physiologic con-
ditions; therefore, many tumors under this category pro- AGCT Adult granulosa cell tumor
duce excess hormones, which may lead to the EMA Epithelial membrane antigen
development of hormone- mediated syndromes. There FATWO Female adnexal tumor of probable Wolffian
are clear age-related differences in the incidence of sev- origin
eral tumors in this category, with juvenile granulosa cell GIST Gastrointestinal stromal tumor
tumors, for example, occurring primarily in the <20- GST Glutathione S-transferase
year age group, Sertoli-Leydig cell tumors predominat- JGCT Juvenile granulosa cell tumor
ing between menarche and 25 years, and adult granulosa LTSP Luteinized thecoma associated with sclerosing
cell tumors most commonly occurring between ages peritonitis
25–50 years. The correct diagnosis of an ovarian tumor MST Microcystic stromal tumor
is important in all patients, but is arguably particularly NOS Not otherwise specified
so for this category, since young patients represent a sig- PJS Peutz-Jeghers syndrome
nificant proportion of patients, and issues of fertility SCCHT Small cell carcinoma of hypercalcemic type
preservation abound. This makes awareness of the diag- SCST Sex cord-stromal tumor
nostic criteria and the differential diagnosis of the SCTAT Sex cord tumor with annular tubules
SCSTs an important task. The 2014 World Health SF-1 Steroidogenic factor 1
Organization classification of SCSTs categorized this SLCT Sertoli–Leydig cell tumor
group under three main categories: pure stromal tumors, SRST Signet ring stromal tumor
pure sex cord tumors, and mixed SCSTs. This chapter SST Sclerosing stromal tumor
highlights all SCSTs, with an emphasis on the clinico- WHO World Health Organization
pathologic characteristics and advances in molecular YST Yolk sac tumor
features of each entity, as well as a detailed discussion of
differential diagnostic considerations, and the utility and
limitations of immunohistochemical markers and ancil-
lary studies in their diagnostic work-up. 9.1 Histogenesis
© Science Press & Springer Nature Singapore Pte Ltd. 2019 273
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_9
274 M. M. Desouki
During embryogenesis, the primitive cells form subcoelomic androgenic symptoms. However, overlap does occur and
mesenchyme condensations that differentiate to the mature nonfunctioning tumors are common among this group as
cortical supporting stroma and the granulosa theca cells of well [1].
the developing follicles. The primitive cells have the ability
to differentiate towards the Sertoli and the Leydig cells of the
testis and the hilus of the ovary. 9.2 Incidence and Clinical Significance
Due to the wide range of differentiation, the tumors
develop from these cells are diverse and complex and contain Apart from fibroma, all other ovarian SCSTs are uncommon
mixture of more than one cell type in most instances. More neoplasms that are known to have a wide morphologic spec-
recently, it has been hypothesized that during development trum and which accordingly may be diagnostically challeng-
of the surface epithelium, some of the epithelial cell precur- ing. SCSTs constitute <10% of all ovarian tumors of which
sors (Gonadal Ridge Epithelial-Like) penetrate into the ovar- benign tumors e.g. fibromas and thecomas account for
ian stroma to form the granulosa cells surrounding the approximately 85% of all SCSTs. On the other hand, adult
follicles. The authors concluded that the ovarian surface epi- granulosa cell tumor (AGCT), Sertoli–Leydig cell tumor
thelial cells do penetrate into the ovary to form the granulosa (SLCT) and sclerosing stromal tumor (SST) account for
cells of follicles [2]. This may explain the morphologic simi- approximately 15% of SCSTs. The remaining tumors col-
larity between many tumors in this category and the more lectively are very rare, and constitute <1% of the SCSTs [3].
common epithelial tumors. Clinical information, including endocrine manifestations
As some of the mentioned cells are involved in hor- and the patient age are important parameters to clinically
mone production under physiologic conditions (e.g., investigate and communicate to the pathologist. The patient
estrogen, progesterone, androgen and corticosteroids), a age is especially significant since there are clear age-related
fraction of SCSTs are expected to produce hormones differences in the incidence of several tumors in this cate-
leading to the development of hormone mediated syn- gory: (a) Juvenile GCTs (JGCTs) occur primarily in the <20
dromes or isolated hormone production which may be the age group, (b) SLCTs predominate between menarche and
initial presenting symptom. Tumors derived from the 25 years, (c) fibrothecomas and AGCTs are most commonly
ovarian cells e.g., granulosa cells and theca cells produce occurring between 25–50-year age group, and (d) fibromas,
estrogenic symptoms and hormones developed from tes- steroid cell tumors and fibrosarcomas commonly emerge
ticular cell types e.g. Sertoli or Leydig cells produce after menopause [4–8] (Table 9.1).
Patients are either asymptomatic with the tumor discovered Table 9.2 Classification of ovarian sex cord-stromal tumors [3, 10]
incidentally on routine clinical examination or on pathologic I. Pure stromal tumors
examination for surgery performed for unrelated disease, or • Fibroma, cellular Fibroma and fibrosarcoma
present with abdominal pain, pelvic or abdominal mass, or • Thecoma (typical and luteinized associated with sclerosing
peritonitis)
endocrine related symptoms due to hyperestrinism (e.g., iso-
• Sclerosing stromal tumor
sexual precocity, abnormal uterine bleeding and endometrial
• Signet ring stromal tumor
hyperplasia or carcinoma) and hyperandrogenism (e.g., hirsut- • Microcystic stromal tumor
ism, acne, hoarse voice, irregular menstrual periods, male-pat- • Stromal tumor with minor sex cord elementsa
tern scalp hair and loss of normal female fat distribution) [3]. • Leydig cell tumor
A small subset of SCSTs have known associations with • Steroid cell tumor
specific syndromes, including Gorlin’s (fibroma), Meigs’ II. Pure sex cord tumors
(fibroma), Ollier’s disease (JGCT), DICER1 (SLCT, and • Adult granulosa cell tumor
JGCT), Maffucci’s (JGCT), and Peutz–Jeghers (sex cord • Juvenile granulosa cell tumor
tumor with annular tubules [SCTATs] and Sertoli cell tumor • Sertoli cell tumor
• Sex cord-stromal tumors with annular tubules
[SCT]) [5, 9] and therefore the initial presentation may be of
III. Mixed sex cord-stromal tumors
the extraovarian symptoms related to a particular syndrome
• Sertoli-Leydig cell tumors
(Table 9.1). • Sex cord-stromal tumors, not otherwise specified
The clinical behavior for SCSTs (Table 9.1) vary tremen- Not included in the 2014 WHO classification of SCSTs of the ovary
a
a b
Fig. 9.1 Alpha Inhibin and calretinin immunohistochemical stains in usually positive in most cases of sex cord stromal tumors with α-inhibin
an adult granulosa cell tumor. α-Inhibin (a) is a cytoplasmic marker is more specific and calretinin is more sensitive among the IHC
while calretinin (b) is a cytoplasmic/cell membrane marker. Both are markers
a b
Fig. 9.2 Steroidogenic factor 1 (SF-1) and WT1 immunohistochemical stains in representative cases of sex cord stromal tumors. Signet ring stromal
tumor (a) and adult granulosa cell tumor (b). Both markers are nuclear and are usually positive in most cases of sex cord stromal tumors
9 Sex Cord-Stromal Tumors of the Ovary 277
Table 9.4 Positivity rate of commonly used immunohistochemistry t hecomatous differentiation (fibrothecomas) constitute ~85%
markers in selected sex cord stromal tumors of SCSTs. Cellular fibromas constitute ~10% of this cate-
Tumor Inhibin Calretinin SF-1 WT1 CD99 FOXL2 CD10 gory [24]. Fibromas commonly affect women in the fifth
Fibroma 70 100 100 100 0 100 0 decade with a mean age of 48-years (range of 17–79 years).
Thecoma 100 100 100 100 0 60 70 The average age of women with cellular fibroma is 46-years
GCT 94 81 100 78 88 98 90
(range of 14–93 years). The most common presenting symp-
SCT 100 100 100 0 32 0 44
toms are pelvic mass and abdominal pain and occasionally
SLCT 100 89 100 100 50 50 44
SCST, 86 80 99 82 68 74 – symptoms related to urinary dysfunction (frequency, dysuria
NOS and nocturia), abnormal uterine bleeding, dyspareunia,
Numbers are the percent of positive cases pooled from different studies increasing abdominal girth and elevated CA-125 on routine
(see text for references) blood work [8, 23].
GCT granulosa cell tumor, SCT Sertoli cell tumor, SLCT Sertoli–Leydig
cell tumor, SCST NOS Sex cord stromal tumor not otherwise specified
9.5.1.3 Gross Findings
Fibromas are typically unilateral (9% bilateral) and range in
size from <1 cm to 16 cm (average of 3.8 cm). They typically
have smooth, sometimes lobulated outer surfaces, solid to
rubbery consistency and white to tan cut surfaces. Edema
and calcification are common gross findings. Infarction and
hemorrhage occasionally present. Fibromas associated with
Gorlin syndrome are usually (~75%) bilateral and more
commonly multinodular rather than smooth [9]. Cellular
fibromas usually reach large size compared to conventional
fibromas with an average size of 9 cm (range of 1–19 cm).
The outer surface may be smooth (34–36%), pedunculated/
polypoid (6–22%), solid, white-yellow or tan with cystic
areas [7, 8, 24].
9.5.1.1 Definition
Fibroma is a benign neoplasm composed of proliferation of
a pure population of fibroblasts that produce collagen.
Cellular fibroma is a fibroma with increased cellularity [23].
9.5.1.2 Clinical Presentations Fig. 9.5 Ovarian fibroma. Representative case of fibroma composed of
Fibromas represent 70% of all SCSTs (4% of all ovarian uniformly distributed, bland spindle-shaped cells arranged in a swirling
neoplasms), and in combination with tumors showing pattern with abundant intervening collagen matrix
278 M. M. Desouki
Fig. 9.6 Ovarian fibroma. The tumor is composed of uniform spindle- Fig. 9.8 Ovarian cellular fibroma. The tumor exhibits increased cellu-
shaped cells containing elongated nuclei with tapered ends. No cyto- larity with haphazard rather than the storiform growth pattern of con-
logical atypia or increased mitotic figures ventional fibroma with scant collagen matrix
Fig. 9.7 Ovarian fibroma. The tumor is composed of uniform bland Fig. 9.9 Ovarian cellular fibroma. The cells arrange in a herringbone
spindle cells with intervening collagen bundles in a swirling pattern pattern. The nuclei are bland, oval, with smooth nuclear contours, even
chromatin, and indistinct nucleoli with no atypia or increased mitotic
activity
findings [25]. The nuclei are elongated with tapered ends and
inconspicuous nucleoli. No cytological atypia is present, and 5%, foamy histiocytes in up to 3%, bizarre nuclei in up to 3%
mitotic figures are usually infrequent (Figs. 9.6 and 9.7). and calcifications in up to 6% of cases [8]. Cells arranged in
Calcification, intercellular edema, sex cord–like elements a herringbone pattern with prominent Verocay-like areas
and hyaline fibrosis may be focally present. The cellularity is were occasionally seen in cellular fibroma. The nuclei are
typically mild to moderate [7]. bland, oval, with smooth nuclear contours, even chromatin,
Cellular fibromas as the name implies tend to have and indistinct nucleoli with no more than mild atypia (mod-
increased cellularity. No consensus on how much cellularity erate atypia may be seen focally around areas of necrosis)
is reasonable to consider the fibroma a cellular rather than [8]. Mitotic figures varies with a mean of 1.5/10 HPF with
conventional one. It is our personal experience that when the most cases have 0–3 mitotic figures/10 HPF (Fig. 9.9).
fibroblasts are haphazardly arranged and there is no/minimal Mitotically active cellular fibroma is a tumor that exhibits
collagen matrix or edema, the term cellular fibroma is used high mitotic count (≥4/10 HPFs) with no more than mild
(Fig. 9.8). Necrosis with sharp demarcation from the adja- nuclear atypia at most. In these tumors the average mitotic
cent viable tumor tissue may be seen in up to 30%, minor sex count /10HPF is 6.7 (range of 4–19). The average age of the
cord-like elements in up to 9%, focal luteinized cells in up to patients is 41 years. The clinical presentation, gross and
9 Sex Cord-Stromal Tumors of the Ovary 279
9.5.1.5 Biomarkers
Fibromas, fibrosarcomas and fibrothecomas are infrequently
positive for α-inhibin (50–70%, variable intensity), calretinin
(25–100%, variable intensity), CD56 (100%, cytoplasmic
rather than membranous), WT1 (80–100%, weak and
patchy), SF-1 and FOXL2 (100%) and cytokeratin (20%,
weak). CD99, CK5/6, CD10, EMA and MART1 have been
reported as negative [13, 15, 16, 20] (Table 9.4).
9.5.2 Fibrosarcomas
9.5.2.1 Definition
A malignant fibroblastic tumor of the ovary [24].
Fig. 9.13 Primary ovarian fibrosarcoma. High power (×40) showing Fig. 9.14 Leiomyosarcoma. The tumor exhibits marked nuclear
increased, cellularity, moderate to severe atypia and prominent mitotic atypia, increased mitotic count and coagulative tumor necrosis.
figures Immunohistochemical markers with positive smooth muscle markers in
leiomyosarcomas and negative stains in fibrosarcoma are helpful in this
differential
9.5.2.5 Biomarkers
The diagnosis is based on morphology and exclusion of
other entities with specific lineage derivation and no specific
IHC panel for fibrosarcoma is recommended. Fibrosarcoma
has been reported positive for vimentin, while negative for
actin and desmin with 30–40% Ki-67 labeling index (a case
report) [34]. One of 8 cases has been reported weak positive
for CD10 [18] and 3 of 3 cases has been reported positive for
FOXL2 [20].
may contain heterologous differentiation (see SLCT section cases [6] especially the tumors previously categorized as
below). luteinized thecomas, which are reportedly more likely to
Diffuse adult granulosa cell tumor: AGCT exhibits be associated with androgenic rather than estrogenic
grooved nuclei and most probably will have other conven- manifestations [40].
tional GCT patterns (see GCT section below). LTSP occur in young patients (mean of 36 and median:
27 years) with wide age range from 10 months to 85 years.
9.5.2.8 Prognosis The lesions typically present with abdominal distension,
Fibrosarcoma has poor prognosis and all reported cases acute abdomen, ascites, bowel obstruction, nausea and vom-
died of disease [24]. Excision with or without postoperative iting [41].
chemo and radiotherapy are the current management
options. 9.5.3.3 Gross Findings
Thecomas are mostly (95%) unilateral tumors with firm,
smooth-surfaces and homogeneous yellow and lobulated cut
surfaces. The tumor may reach up to 22 cm with an average
Key Diagnostic Points dimension of 5 cm [6]. Most LTSPs are bilateral, polypoid,
Fibrosarcoma lobulated, or cerebriform with tan, gray, or pink, homoge-
Gross neous, cut surfaces with occasional hemorrhage [41].
• Unilateral, large, soft, tan-yellow, solid to cystic
with heterogeneous consistency. 9.5.3.4 Microscopic Findings
• Hemorrhage and necrosis are common. The neoplastic cells are fusiform and grow in a diffuse
pattern with alternating hyaline plaques or rarely in a nod-
Microscopic ular pattern. The cells have ill-defined membranes and
• Increased cellularity. pale gray cytoplasm. Calcifications and keloid-like scle-
• Spindle cells with storiform or herringbone growth rosis are common findings. Vesicular nuclei with delicate
pattern. membranes are usually appreciated with no nuclear atypia
• Moderate to severe atypia. and infrequent mitoses [6] (Figs. 9.16 and 9.17).
• Increased mitotic activity. Calcifications and bizarre nuclei resembling those in sym-
plastic leiomyoma do occur with smudgy chromatin and
Markers no increase in mitotic activity [42]. Pure forms composed
• Only useful in excluding other high grade malig- of theca cells only with no admixture of fibroblasts
nancies with spindle cells. (fibrothecomas) or granulosa cells (granulosa-theca cell
tumors) are extremely rare. The granulosa cell component
if any should not exceed 10% [40].
9.5.3.1 Definition
Thecomas are a group of pure ovarian stromal tumors origi-
nating from and resembling theca cells of the developing
ovarian follicles. If associated with sclerosing peritonitis, the
term luteinized thecoma is applicable [3].
a b
c d
Fig. 9.17 Ovarian thecoma. Low (a) and high (b) power captions of a tumor that is composed of plump to spindle cells with abundant pale cyto-
plasm and central nuclei. The tumor is positive of inhibin (c) and calretinin (d)
a b
c d
Fig. 9.19 Mitotically active ovarian thecoma. Low (a) and high (b) increased mitotic activity (arrow). The tumor cells are positive of
power captions of a mitotically active thecoma that is composed of inhibin (c) and calretinin (d)
plump to spindle cells with abundant cytoplasm and central nuclei with
a b
c d
Fig. 9.20 Ovarian luteinized thecoma. The tumor is composed of fascicles of spindled cells with abundant eosinophilic cytoplasm (a). The tumor
cells are positive of inhibin (b), calretinin (c) and reticulin histochemical stain surrounds individual cells (d)
a b
Fig. 9.21 Reticulin histochemical stain in fibrothecoma and adult granulosa cell tumor. A reticulin stain surround individual cells in fibrothecoma
(a) and nests of cells in adult granulosa cell tumor (b)
286 M. M. Desouki
a b
Fig. 9.24 ovarian sclerosing stromal tumor: Gross photograph of a sizable, circumscribed mass with lobulated outer surface (a) and solid and
smooth cut surfaces with gross calcifications (b)
288 M. M. Desouki
a b
Fig. 9.25 Ovarian sclerosing stromal tumor: The tumor is composed of alternating hypercellular areas separated by edematous and collagenous
hypocellular areas
a b
Fig. 9.26 Ovarian sclerosing stromal tumor: The tumor is composed amount of eosinophilic cytoplasm and vesicular nuclei with no nuclear
of alternating nodules of hypercellular areas separated by collagenous atypia and infrequent mitosis (b)
hypocellular areas (a). The cells are round to oval with moderate
a b
Fig. 9.27 Immunohistochemical markers in sclerosing stromal tumor: The same case depicted in Fig. 9.26. The lesional cells are positive for
inhibin (a), calretinin (b) and smooth muscle actin (c)
9 Sex Cord-Stromal Tumors of the Ovary 289
c IHC
• Positive: Vimentin (100%), SMA (100%), glutathi-
one S-transferase (100%) calretinin (100%), inhibin
(90%) and CD10 (90%).
• Negative: S100, Desmin, and CD34.
9.5.5.1 Definition
SRSTs are benign ovarian stromal neoplasms with prolifera-
tion of signet ring cells and variable amount of spindle-
shaped stromal cells [51, 52].
a b
Fig. 9.28 Ovarian signet ring stromal tumor: SRSTs usually exhibit cells with indented eccentric nuclei pushed against one pole of the cell
diffuse growth pattern. However, in this case, the tumor exhibits nodu- simulating signet rings (b). The nuclei are bland with no atypia or
lar growth pattern (a). The cells are bland which merge with vacuolated mitotic figures
a b
c d
Fig. 9.29 Immunohistochemical markers in signet ring stromal tumor: The same case (Fig. 9.28). The signet ring cells are positive for vimentin
(a) and CD 10 (b) while negative for inhibin (c) and WT1 (d)
9 Sex Cord-Stromal Tumors of the Ovary 291
Fig. 9.30 Mucicarmine stain in signet ring stromal tumor: The same Fig. 9.31 Signet ring carcinoma (Krukenberg tumor) metastatic to the
case (Fig. 9.28). The signet ring cells are negative for the mucicarmine ovary: sheets as well as individual cells dissecting through the hypercel-
histochemical stain lular stroma. The nuclei have mild atypia with no increase in mitotic
activity
9.5.6.1 Definition
A tumor that composed entirely of steroid secreting cells
(lutein, adrenal cortical and Leydig cells) with absent crys- Fig. 9.34 Ovarian steroid cell tumor: Gross photograph showing well-
tals of Reinke [56]. circumscribed yellow mass
9 Sex Cord-Stromal Tumors of the Ovary 293
Fig. 9.36 Ovarian steroid cell tumor: The tumor composed of a uni-
form cell population with finely granular eosinophilic cytoplasm and
centrally located nuclei with prominent nucleoli. The cells are sepa-
Fig. 9.35 Ovarian steroid cell tumor: Gross photograph showing lobu- rated by thin fibrous septa rich in vasculature
lated, solid gray-yellow cut surfaces
a b
Fig. 9.38 Ovarian steroid cell tumor: The tumor cells form nodules (a) separated by thin fibrous septa rich in vasculature (b)
a b
Fig. 9.39 Immunohistochemical markers in ovarian steroid cell tumor: The tumor cells are positive for inhibin (a), calretinin (b) and negative for
epithelial membrane antigen (c)
9 Sex Cord-Stromal Tumors of the Ovary 295
9.5.8.1 Definition
Microcystic stromal tumor (MST) of the ovary is a benign
tumor of probable stromal origin with distinctive conspicu-
ous microcystic change first described by Irving and Young
on 2009 [63]. Subsequent work has shown that they consis-
tently display alterations in in beta-catenin-APC pathway
and accordingly may represent a manifestation of Familial
Adenomatous Polyposis syndrome [64, 65].
a b
c d
Fig. 9.43 Leydig cell hyperplasia of the fallopian tube. The lesion is composed of small nodules with no organ distortion. The morphology is
identical to that of steroid/leydig cell tumors (a, b). The lesion is positive for inhibin (c) and calretinin (d)
9.5.8.3 Gross Findings bland, round, oval or spindle-shaped with fine chromatin and
The tumors are usually unilateral with a mean size of 8.7 cm indistinct nucleoli. Bizarre nuclei with smudgy chromatin
(range of 2–27 cm) and localized to the ovary. The tumors may be focally present. Mitotic rate is low (range of 0–2
are usually solid with cystic degenerations however; pure mitoses/10 HPFs) [63]. The diagnosis of MST is mainly by
solid or cystic tumors are rarely exist. The solid component exclusion of other known entities with the following features
are firm and tan or yellow in color. Small foci of hemorrhage (1) a predominant microcystic pattern with lobulated cellular
and/or necrosis may be present [63]. masses which are intervened with hyalinized fibrous stroma,
(2) absence of epithelial component, (3) absence of tera-
9.5.8.4 Microscopic Findings tomatous or other germ cell elements and (4) characteristic
The tumors consist of three main components; microcysts, IHC staining pattern (see below).
solid cellular regions, and fibrous stroma with variable
amount of each component. The microcysts predominate in 9.5.8.5 Biomarkers
most cases and are characterized by small round to oval The tumors are positive in all cases for beta-catenin (nuclear),
empty spaces sometimes coalesce to form large irregular cyclin D1, CD10, WT1, FOXL2, and vimentin and positive
spaces. Intracytoplasmic vacuoles are common findings. The in 60% of cases for SF-1, focally positive for cytokeratin in
solid areas are usually intersected by fibrous bands and hya- 25% of cases while negative for α-inhibin (very rarely posi-
line plaques simulating thecomas. The cytoplasm of the cells tive), calretinin (very rarely positive) and EMA by IHC
is finely granular, lightly eosinophilic and the nuclei are [63–65].
298 M. M. Desouki
9.5.8.6 Molecular Findings with clear vacuolated cells and eccentric nuclei, interspersed
Somatic missense point mutations of the CTNNB1 gene have with fibroblasts and lutein cells. α-Inhibin and calretinin are
been reported in 57% of ovarian MSTs. The mutations are typically positive [47].
heterozygous in exon 3, which will result in amino acid sub- Signet ring stromal tumor: Another tumor, which shares
stitutions at codons 32, 34, 35, and 37. The aberrant nuclear similarity with MST (see above). SRST is typically a unilat-
β-catenin expression by IHC detected in all cases may be the eral solid or solid-cystic tumor occurring in adult women
result of stabilizing CTNNB1 mutations in 57% of cases. without hormonal manifestations. Microscopically, the bland
Additionally, the dysregulation of the Wnt/B-catenin signet-ring cells are admixed with a diffuse fibroblastic
pathway may be involved in the tumorigenesis of MSTs and proliferation [51].
may be involved in activation of the β-catenin leading to Another tumors in the differential of MST include SLCT,
upregulation of cyclin D1 which has been reported positive JGCT, yolk sac tumor, monodermal teratomas (struma
in all cases tested by IHC [64, 65]. ovarii), carcinomas, Brenner tumor and FATWO [63]. These
tumors are discussed under different topics in this chapter.
9.5.8.7 Differential Diagnosis
Thecoma: women with MSTs are 2 decades younger than 9.5.8.8 Prognosis
those with thecomas [40]. Prominent hyaline plaques in a These tumors are thought to be benign based on limited evi-
tumor with microcysts prompted a diagnosis of thecoma. dence. Patients that have treated with simple excisions have
Estrogenic manifestations are common in thecomas. Grossly, shown with no evidence of disease at a mean of 4.3 years
the majority of MSTs are solid-cystic, whereas thecomas are follow up [63].
usually solid and yellow. Microscopically, both tumors con-
sist of bland cells with round to elongate nuclei and infre-
quent mitotic figures. However, thecomas are characterized
by conspicuous population of polygonal to fusiform cells Key Diagnostic Points
(rather than round) with pale cytoplasm. Bizarre nuclei are Microcystic Stromal Tumor
common in MSTs [66]. CD10 staining is faint weak to mod- Gross
erate in thecomas in contrast to the strong and diffuse stain- • Unilateral, medium sized (average of 8.7 cm), solid
ing in MST [18]. α-Inhibin and calretinin are usually negative with cystic formation, firm, tan or yellow.
in MSTs compared to positive pattern in thecomas [14].
Lutenized thecoma associated with sclerosing peritonitis: Microscopic
LTSP may show microcystic pattern. It occurs in young • Three components: Microcysts, solid cellular
women and is bilateral, composed of cellular proliferations regions and fibrous stroma.
of non-luteinized spindle cells, with numerous aggregates of • Microcysts predominate with lobulated solid cellu-
luteinized cells. The cells are diffusely positive for α-inhibin, lar masses, which are intervened with hyalinized
calretinin, and CD56 and the tumor is associated with scle- fibrous stroma.
rosing peritonitis [43]. • Microcysts are small round/oval empty spaces.
Steroid cell tumor with oxyphil cells: In contrast to MST, ste- • Intracytoplasmic vacuoles are common.
roid cell tumors have a frequent manifestation with hormonal • Finely granular, eosinophilic cytoplasm.
manifestations, especially androgenic. Occasionally MST may • Bland, round/oval or spindled nuclei with fine chro-
exhibit yellow cut surfaces on gross examination. matin and indistinct nucleoli.
Microscopically, steroid cell tumors often exhibit a diffuse pat- • Low mitotic rate.
tern with minimal to absent stroma in most cases. However, • No epithelial, teratomatous or germ cell elements.
NOS tumors consist of aggregates of cells interspersed among • Exclusion of other known entities with microcysts
fibrous and edematous stroma. The tumor cells in steroid cell should be performed.
tumors have rounded contour with moderately abundant eosino-
Markers
philic cytoplasm resembling cells of the MST, but in the former,
the nuclei are central with prominent nucleoli. CD10 is usually • Positive: CD10 (100%), vimentin (100%), cyclin
positive in steroid cell tumors, but unlike the MST, α-inhibin D1 (100%), WT1 (100%), FOXL2 (100%) nuclear
and calretinin are positive in most steroid cell tumors [18, 43]. beta-catenin (100%), SF-1 (80%) and cytokeratin
Sclerosing stromal tumor: Like that of MST, an SST is (25%, focal).
characteristically unilateral, solid, white or yellow often with • CTNNB1 gene somatic mutations (57%).
edema and cyst formation. Microcysts are not a striking fea- • Negative: EMA, inhibin (weak positive in 6% of
ture of SST while staghorn, thin-walled vessels are promi- cases), calretinin (weak positive in 6% of cases).
nent. The neoplastic cells of the SST are rounded to polygonal
9 Sex Cord-Stromal Tumors of the Ovary 299
9.5.9 Stromal Tumors with a Minor cigar shaped with inconspicuous nucleoli and no/rare
Sex-Cord Element mitotic figures [42].
a b
Fig. 9.44 Metastatic carcinoid tumor to the ovary: The tumor is composed of scattered small aggregates of cells in a fibromatous background
resembling sex-cord elements (a). The tumor is positive for chromogranin by immunohistochemistry (b)
300 M. M. Desouki
stromal tumor with a minor sex-cord element, carcinoids are SCSTs after fibromas. The tumor predominates in perimeno-
usually bilateral and present elsewhere (e.g. small intestine) pausal or early postmenopausal women with a median age of
[5] (see the differential diagnosis section under SCT below). 50–55 years [1, 4, 69]. The tumor may produce estrogen or
Stromal hyperplasia and hyperthecosis: see the differen- androgen with clinical effects including abnormal uterine
tial diagnosis section under thecoma above. bleeding due to hyperplasia. Associated endometrial carci-
noma have been reported to be ~ 5% of cases with AGCTs.
9.5.9.7 Prognosis The most common presenting symptom is postmenopausal
Surgical excision is curative. bleeding due to endometrial hyperplasia or adenocarcinoma
secondary to estrogen production by the AGCT [69]. Initial
presentation in perimenopausal women include menstrual
irregularity, menorrhagia, amenorrhea, infertility, virilizing
Key Diagnostic Points features or hirsutism, abdominal/pelvic pain which could be
Stromal Tumor with a Minor Sex-Cord Element severe (acute abdomen) due to tumor rupture or torsion and
Gross abdominal distention [68, 69].
• Similar to fibroma and thecoma with smooth outer 9.6.1.3 Gross Findings
surface. AGCTs are unilateral in 90% of cases. The tumor size
• Solid, firm, white, yellow or pink. ranges from microscopic incidental finding to a large tumor
• No hemorrhage or necrosis. filling the whole pelvis (average of 12.7 cm) [69]. The
Microscopic tumors are solid with variable cystic spaces and hemorrhage
(Figs. 9.45 and 9.46). Tumors with prominent theca ele-
• The stromal component is that of fibroma/thecoma ments are often yellow and firm. Rupture at the time of sur-
morphology (see above). gery is not uncommon [4].
• The sex cord element should not exceed 10% (gran-
ulosa cells or setoliform tubules or indifferent 9.6.1.4 Microscopic Findings
elements). AGCTs are composed of organoid or diffuse architecture
• The sex cord elements have rounded nuclei with with epithelioid and spindle cell components. Multiple pat-
prominent nucleoli and occasional crystals of terns are usually seen in a single tumor. The patterns include;
Reinke. diffuse, trabecular, insular, gland-like, corded, microfollicu-
• Sex cord elements may be scattered or localized in lar, macrofollicular, “watered silk”, pseudopapillary
few foci. (Fig. 9.47), nodular, sarcomatoid and systic [1].
Markers In the diffuse pattern, which is one of the most common
patterns, the cells arrange in sheets with scant cytoplasm
• Reticulin highlights individual cells in the spindle ‘small blue cell tumor’ (Figs. 9.48 and 9.49). Careful
cell component and groups of cells in the sex cord
component.
• The sex cord elements are positive for inhibin, cal-
retinin, CD99 and CD56.
9.6.1.1 Definition
AGCTs are ovarian tumors, which are derived from granu-
losa cells of the ovarian stroma that under physiologic condi-
tions produce estrogen [68].
Fig. 9.47 Adult granulosa cell tumor of the ovary. The characteristic Fig. 9.48 Adult granulosa cell tumor of the ovary. The tumor is com-
tumor cells form pseudo-papillary growth pattern (image courtesy of posed of diffuse growth pattern of small blue round cells with scant
Dr. Oluwole Fadare) cytoplasm
a b
Fig. 9.49 Adult granulosa cell tumor of the ovary. The tumor is composed of diffuse architecture with nuclear grooves (a). The tumor cells are
positive for inhibin (b)
302 M. M. Desouki
Fig. 9.50 Adult granulosa cell tumor of the ovary. Microfollicular pat- Fig. 9.51 Adult granulosa cell tumor of the ovary. Microfollicular pat-
tern with Call–Exner bodies characterized by arrangement of the granu- tern with vaguely formed Call–Exner bodies
losa cells around spaces filled with eosinophilic material with
degenerating nuclei
a b
Fig. 9.52 Ovarian adult granulosa cell tumor of the ovary. The tumor oval, haphazardly arranged nuclei with evenly dispersed chromatin and
is composed of diffuse architecture with micro and macro follicles (a). nuclear grooves (b)
The cells are monotonous with scant cytoplasm and uniform angular to
The cells in AGCT have scant cytoplasm, pale, uniform, inimal to exuberant. The stroma may be vascular, fibrous or
m
angular to oval, haphazardly arranged nuclei with evenly dis- edematous to hypercellular with o ccasional luteinized and
persed chromatin and nuclear grooves (Figs. 9.49 and 9.52). vacuolated cells. Leydig cells with crystals of Reinke and
Scattered atypical (bizarre) nuclei is rare and is independently hepatoid stroma could also occur in AGCTs [1, 71].
correlated with the biologic behavior of the tumor [66]. The
presence of the most famous feature (nuclear grooves), varies 9.6.1.5 Biomarkers
from tumor to tumor and may be inconspicuous. Mitotic rate AGCTs are positive for α-inhibin (94%), calretinin (81%)
also varies from tumor to tumor but usually low and most (Figs. 9.1, 9.2 and 9.49), WT1 (78%), SF-1 (100%), CD56
tumors exhibit low-grade features. Sometimes the cells are (100%), FOXL2 (98%) (Fig. 9.4), CD10 (90%), PR (100%)
luteinized with ample cytoplasm, which is pale or eosino- and CD99 (88%) by IHC [11–13, 18, 20] (Table 9.4).
philic simulating thecoma [1]. SLCT component may be Reticulin histochemical stain usually surrounds nests of cells
present in AGCT. The volume of the stroma varies from (Figs. 9.21 and 9.53).
9 Sex Cord-Stromal Tumors of the Ovary 303
a b
Fig. 9.54 Endometrioid adenocarcinoma with sarcomatoid growth and lacks nuclear grooves (a). The tumor cells are positive for the low
pattern. In contrast to adult granulosa cell tumor with sarcomatoid pat- molecular weight CAM5.2 (b)
tern, this tumor exhibits a higher nuclear grade, more mitotically active
a b
Fig. 9.55 Poorly differentiated ovarian endometrioid adenocarcinoma. In contrast to adult granulosa cell tumor, this tumor exhibits a higher
nuclear grade and lacks nuclear grooves (a, b)
Struma ovarii: The rounded structures may simulate the Metastatic breast carcinoma: Lobular and invasive mam-
microfollicles—Call–Exner bodies of the AGCT. However, mary carcinoma of no special types typically are bilateral
acini of struma usually contain colloid-like material and the and is EMA and GATA3 positive and α-inhibin and WT1
diversity of patterns in struma ovarii e.g. trabecular patterns negative [1] (Fig. 9.58).
is remarkable (Fig. 9.57). Immunostaining for thyroglobulin Metastatic malignant melanoma: History of an extraovarian
will aid to sort out this differential [83]. melanoma, bilateral ovarian involvement, melanin pigment,
Small cell carcinoma of the hypercalcemic type (SCCH): and positivity for melanoma markers will aid to segregate this
The presence of follicles in SCCHT and the small blue round devastating neoplasm from AGCT [1] (Fig. 9.59).
cells with scant cytoplasm characteristic of this high-grade Table 9.5 lists the utilization of commonly used IHC
tumor are similar to those of AGCT. The cells in AGCTs markers in the differential diagnosis of GCTs with few enti-
have regular arrangement and grooved nuclei and always ties mentioned above.
lack highly malignant features and mitotic rate compared to
SCCHTs [84]. α-inhibin stain is positive in AGCT and nega- 9.6.1.9 Prognosis
tive in SCCHT. (See the differential diagnosis section under Most patients (up to 91%) present at an early stage. However,
JGCTs below). recurrence tends to occur often many year after the initial
9 Sex Cord-Stromal Tumors of the Ovary 305
Fig. 9.56 Ovarian endometrioid adenocarcinoma. Conventional mor- Fig. 9.57 Struma ovarii: The rounded structures may simulate the
phology of endometrioid carcinoma present in poorly differentiated microfollicles of adult granulosa cell tumor. The acini of struma contain
cases of endometrioid carcinoma (this is another field in for the same colloid-like material
cases depicted in Fig. 9.55)
a b
Fig. 9.58 Metastatic breast carcinoma to the ovary: Invasive mammary carcinoma of no special type to the ovary (lower field) and coincidental
high-grade serous carcinoma (upper field) (a). The breast carcinoma is negative for WT-1 while the serous carcinoma is positive for the marker (b)
diagnosis in many cases (Fig. 9.60). The staging system used d ebulking, large tumor size (>15 cm), grossly solid tumors,
for AGCT is the same for ovarian epithelial tumors (FIGO reduced β-catenin expression, high mitotic rate, capsular
staging) [85]. Surgery is the main line of management for invasion with extraovarian spread and lymphovascular space
patients with AGCT. Staging usually includes an omentec- invasion [22, 69].
tomy, lymph node dissection, peritoneal biopsies and obtain- FOXL2 mutated tumors especially those with homozy-
ing peritoneal washings for cytological evaluation. For gous mutations have been reported to have significantly
postmenopausal women, a total hysterectomy and bilateral higher relapse rates and worse disease-free and overall sur-
salpingo-oophorectomy is the operation of choice and a vival than tumors with the wild-type FOXL2 gene [21]. The
more conservative approach is reasonable for younger high frequency of FOXL2 mutation in AGCTs encouraged
women with early stage disease to preserve fertility [46, 68]. scientists to suggest that it could be a therapeutic target in
Clinico-pathologic factors that have been reported to be this tumor. However, studies are still in its early stages and
associated with worse outcome include advanced stage at therapeutics designed to target FOXL2 may adversely affect
presentation, older age (>40 years), patients with abdominal regulatory activities of other important FOX transcription
symptoms, clinically palpable mass, incomplete tumor factors [86].
306 M. M. Desouki
a b
Fig. 9.59 Metastatic malignant melanoma to the ovary: The tumor cells look like an adult granulosa cell tumor (a). History of an extraovarian
melanoma, bilateral ovarian involvement, melanin pigment, and positivity for MART1 (b) will aid to solve this differential
Table 9.5 Utilization of immunohistochemistry in the differential diagnosis of granulosa cell tumors
Inhibin PR EMA Chromogranin EpCAM PAX8 FOXL2
Diagnosis (cytoplasm) (nucleus) (cytoplasm) (cytoplasm) (membrane) (nucleus) (nucleus)
Granulosa cell 94 100 1% 0 0 11 98
tumor
Endometrial 0 71 95% 0 97 90 0
carcinoma
Carcinoid 1 2 21% 90 93 20 0
Small cell 0 0 40 33 0 0 0
carcinoma
The numbers are the reported frequency of positive cases from multiple studies (see text for references)
a b
Fig. 9.60 Recurrent adult granulosa cell tumor in the peritoneal cavity. The patient has a history of primary ovarian AGCT 15 years prior to this
lesion. The tumor is composed of diffuse growth pattern (a). The tumor cells are positive for inhibin (b).
9 Sex Cord-Stromal Tumors of the Ovary 307
9.6.2.1 Definition
JGCTs are ovarian tumors derived from the granulosa cells
of the ovarian stroma which occur predominantly at young
age group [1, 72].
a b
Fig. 9.62 Juvenile granulosa cell tumor of the ovary. Foci of atypical nuclei with smudgy chromatin are present in this case (a). Diffuse prolifera-
tion of tumor cells among follicle-like spaces of different sizes and shapes. Reticulin stain surrounds nests of cells (b)
a b
Fig. 9.64 Small cell carcinoma of the hypercalcemic type: Low (a) and high (b) power captions of small cell carcinoma of the hypercalcemic
type. The tumor lacks the nodular growth pattern of JGCTs and the cells are frankly primitive
(see the differential diagnosis section under steroid cell stage [95]. In contrast to AGCTs, JGCTs are less likely to
tumor above) [57]. recur and if recurrence does occur, this usually happen within
Small cell carcinoma of the hypercalcemic type (SCCH): 3 years of initial diagnosis [4, 96].
Hypercalcemia (in 60% of SCCH) and the lack of estrogenic
manifestations may help in differentiating this tumor from
GCTs. Grossly, most cases of SCCH are bilateral and
advanced stage, as compared with GCTs, which are typically Key Diagnostic Points
confined to one ovary. Morphologically, SCCH and JGCT Juvenile Granulosa Cell Tumor
may both have high mitotic activity, and both may display Gross
follicles. However, the constituent cells of SCCH are notably • Similar to AGCT (see above).
primitive and lack abundant cytoplasm (with the exception
of the large cell variant) (Fig. 9.64). A subset of SCCH show Microscopic
mucinous glands, which would be uncharacteristic in • Nodular or diffuse proliferation in a myxoid and
JGCT. Furthermore, SCCHT lacks the nodular growth pat- edematous background separated by follicles.
tern of JGCTs. The cells of SCCH lack α-inhibin positivity • The follicle-like spaces contain proteinaceous
and have as a defining feature a loss of expression of material and lined by granulosa cells.
SMARCA4 (BRG1); the latter distinguishes SCCH from all • Rare/no call-Exner bodies.
histologic mimics [94]. • Abundant eosinophilic or vacuolated cytoplasm.
• Non-grooved large, round, hyperchromatic nuclei
9.6.2.8 Prognosis with variable atypia (minimal to severe).
The presence of gsp mutation may carry worse prognosis • High mitotic count.
since this mutation is usually associated with advanced stage
at initial presentation [88]. Other factors, which may indicate Markers
poor outcome include advanced stage at presentation and • As that of AGCT (see above) and positive for Fli-1
strong FOXL2 protein expression. Of note, the presence of by IHC.
cytological atypia or high mitotic rate does not affect the • Activating mutations of the trimeric G protein
clinical outcome. The overall survival for patients with stage (30%).
I is 97% [4, 21]. Excision is usually adequate for early stage • FOXL2 somatic gene mutation in only 10% of
disease and adjuvant chemotherapy may be considered for cases.
patients with recurrence or those who present at advanced
310 M. M. Desouki
9.6.3 Sertoli Cell Tumors c ollection of Leydig or granulosa cells could be identified and
is not a background to consider mixed SCST diagnosis.
9.6.3.1 Definition Malignant criteria include large tumors (>5 cm), >5
A tumor composed exclusively of Sertoli cells [97]. mitosis/10HPF, nuclear atypia and necrosis [97].
a b
c d
Fig. 9.65 Dysgerminoma: The tumor cells grow in a diffuse pattern and traversed by fibrous septa containing lymphocytes (a). The tumor cells
usually are positive for CD117 (b), SALL4 (c) and OCT 3/4 (d) stains
Markers
• Positive: Inhibin (82%), calretinin (50%), CD99 9.6.4.4 Microscopic Findings
(86%), WT1 (100%), SF-1 (100%), vimentin Patients with PJS: The SCTAT is bilateral in more than 50%
(94%), AE1/AE3 (65%), CAM5.2 (65%) and CD10 of cases. The tumors are characterized by sharply circum-
(44%). scribed, rounded nests with ring-shaped tubules containing
• Negative: EMA, chromogranin, and CK-7. hyalinized basement membrane-like material. The tubules
are distributed singly or in clusters in simple or communi-
cating architecture within the cortical or medullary stroma.
The nuclei are located at the periphery of the tubules and
9.6.4 Sex Cord Tumors with Annular Tubules around the hyaline material (Fig. 9.66). Calcification is
common in the ovarian stroma. The cells may contain large
9.6.4.1 Definition lipid vacuoles.
A rare SCST with a distinctive pattern of simple and com- Patients without PJS: The appearance of the tumor is
plex annular tubules [105]. similar to that in patients with the PJS. SCT pattern with
elongated tubules and pale cytoplasm occupying the central
9.6.4.2 Clinical Presentations portions is common. The neoplastic cells are columnar with
Patients with PJS present at an earlier age (mean of 27 years) ample pale cytoplasm. The nuclei are basally located, round
compared to a mean age of 34 years in patients without to oval with occasional grooves and single nucleoli. Nuclear
PJS. Most patients with PJS have gastrointestinal polyposis pleomorphism and mitotic figures are very rare even in
and muco-cutaneous melanin pigmentation with incidental cases with malignant behavior [106]. Moderate increase in
SCTAT on microscopic examination of the ovaries [105, mitotic count (7–10/10 HPF) has been reported in a malig-
106]. In contrast to tumors associated with PJS, abdominal/ nant case [107].
pelvic masses are palpable in ~50% of patients without PJS
[106]. Table 9.6 summarizes the characteristics of SCTAT in 9.6.4.5 Biomarkers
patients with and without PJS. Like most other SCSTs, SCTAT are positive for α-inhibin
(100%), calretinin (100%), WT1 (100%) and CD56 (100%).
9.6.4.3 Gross Findings AE1/AE3 has been reported positive in 67% of cases. CD10,
The SCTAT in patients with PJS may be seen grossly in only EMA, CK5/6 and Melan-A/MART-1 are negative in all
28% of cases as single or multiple small nodules. The tumors reported cases [15, 17–19].
are grossly yellow and relatively small (up to 3 cm). Ovarian
cysts and tumors other than the SCTAT are present in 25% of 9.6.4.6 Differential Diagnosis
cases. The SCTAT in patients without PJS are unilateral and The SCTAT has a specific morphology and usually is a
range in size from microscopic to 20 cm. Gross hemorrhage straightforward diagnosis if the pathologist is familiar with
and necrosis are rarely seen (Table 9.6) [105]. the entity.
9 Sex Cord-Stromal Tumors of the Ovary 313
a b
Fig. 9.66 Sex cord tumor with annular tubules: Low (a) and high (b) characteristic of SCTAT. The nuclei are located at the periphery of the
power captions of sharply circumscribed, rounded nests with ring- tubules and around the hyaline material
shaped tubules containing hyalinized basement membrane-like material
9.6.4.7 Prognosis
SCTAT in patients with PJS is benign with rare malignant • Elongated tubules and pale cytoplasm.
cases have been reported [108]. The patients may die of asso- • Columnar cells with ample pale cytoplasm.
ciated cervical adenoma malignum, which does metastasize. • Basally located round to oval nuclei with occasional
SCTAT is clinically malignant in ~15% of cases without PJS grooves and single nucleoli.
with half of them died of disease [106]. • Rare atypia and low mitotic count.
Markers
Key Diagnostic Points • Positive: Inhibin (100%), calretinin (100%), WT1
Sex Cord Tumor with Annular Tubules (100%), CD56 (100%) and AE1/AE3 (67%).
Gross • Negative: CD10, EMA, CK5/6 and MART-1.
Patients with PJS:
• Bilateral (>50%), multiple, small (up to 3 cm), yel-
low nodules.
9.7 Mixed Sex Cord-Stromal Tumors
Patients without PJS:
• Unilateral, large (up to 20 cm). 9.7.1 Sertoli–Leydig Cell Tumors
Microscopic 9.7.1.1 Definition
Patients with PJS: SLCTs are mixed SCSTs with formation of Sertoli cells,
• Circumscribed, round nests with ring-shaped Leydig cells and sometimes primitive gonadal stroma and
tubules containing hyalinized basement membrane- heterologous elements [109].
like material.
• The tubules are distributed singly or in clusters. 9.7.1.2 Clinical Presentations
• In the ovarian cortex or medulla. SLCTs predominate (75%) of patients younger than 30 years
• The nuclei are located at the periphery of the tubules and only 10% are >50 years with and average age of 25 years
and around the hyaline material. [110]. The retiform tumors occur at a young age (mean of
• Calcification is common in the ovarian stroma. 15 years) [111]. Androgenic symptoms e.g. virilism,
amenorrhea, deepening of voice, and clitoromegaly are the
Patients without PJS: main symptoms in approximately half of patients especially
• Similar to that in patients with the PJS. in well-differentiated tumors and less commonly in retiform
and tumors with heterologous elements. Other symptoms
314 M. M. Desouki
include abdominal swelling or pain [109]. In association perpendicular to the basement membranes with occasional
with cervical embryonal rhabdomyosarcoma, SCLTs have bizarre forms [99]. The Leydig cells are present throughout
been reported in 4 cases [112]. the tumor with variable density in a single or cluster pattern
(Figs. 9.68 and 9.69) [110]. Well-differentiated tumors are
9.7.1.3 Gross Findings almost always without heterologous or retiform elements
SLCT are usually (98.5%) unilateral tumors that range in [113].
size from microscopic to 51 cm (average 13.5 cm). SLCTs
tend to be solid, lobulated and yellow in color with or with- 9.7.1.6 Moderately Differentiated
out cyst formation. Some tumors are completely cystic espe- The characteristic morphology is the presence of alternating
cially those with retiform component with or without cellular areas separated by hypocellular edematous stroma in
papillary or polypoid excrescences resembling serous lobulated pattern. The neoplastic cells grow as sheets,
tumors. Higher-grade lesions especially those with mesen- tubules, nests, cords or microcysts. The cells arrange in an
chymal heterologous elements show gross hemorrhage and alveolar pattern with scant cytoplasm and small, round nuclei
necrosis [5] (Fig. 9.67). Tumors with mucinous heterologous which are sometimes bizarre [66]. Cysts with luminal eosin-
elements may be multicystic with mucinous contents similar
to mucinous tumors. SLCTs usually present at an early stage
[101, 109].
9.7.1.5 Well-Differentiated
These tumors are composed predominantly of tubules form-
ing either [1] compact lobules of round or oval small tubules,
which may be cystically dilated or [2] tubules, which infil-
trate between the intervening collagen bundles. The tubules
Fig. 9.68 Well-differentiated Sertoli-Leydig cell tumor. The neoplas-
usually have open lumina and rarely are solid with incon-
tic cells form tubules with elongated nuclei, which arrange perpendicu-
spicuous lumens. The tubules are lined by cells with lipid lar to the basement membranes. The Leydig cells are present throughout
rich or oxyphilic cytoplasm and elongated nuclei arranged the tumor (image courtesy of Dr. Oluwole Fadare)
a b
Fig. 9.67 Sertoli-Leydig cell tumor: Gross photo of an SLCT showing large, yellow tumor with smooth outer surface (a) and nodular cut surfaces
with areas of gross hemorrhage and necrosis (b)
9 Sex Cord-Stromal Tumors of the Ovary 315
a b
Fig. 9.69 Well-differentiated Sertoli-Leydig cell tumor with bizarre nuclei. The neoplastic nuclei are enlarged, hyperchromatic with smudgy
chromatin. However, no increase in mitotic activity. The Leydig cells are present throughout the tumor (a, b)
Fig. 9.70 Moderately differentiated sertoli-Leydig cell tumor. The Fig. 9.71 Moderately differentiated sertoli-Leydig cell tumor. Leydig
characteristic morphology is shown in this photomicrograph with alter- cells are present at the periphery of the hypercellular areas
nating cellular areas separated by hypocellular stroma in a lobulated
pattern. The cells have small, round nuclei and scant cytoplasm. Leydig
cells are scattered in the hypocellular areas
rate with rare to absent Leydig cells [1, 109] (Figs. 9.72
and 9.73).
ophilic material simulating struma ovarii and follicles may
9.7.1.8 Sertoli-Leydig Cell Tumor with Retiform
be present [1]. Leydig cells are usually rimming the nodules
and populating the hypocellular areas (Figs. 9.70 and 9.71). Pattern
The stroma may be fibrous or cellular and sometimes simu- These tumors represent 15% of the moderately and poorly
lating sarcoma with frequent edema containing Leydig cells differentiated SLCTs and as the name implies are character-
[109]. ized by growth pattern simulating the rete testis. The reti-
form pattern may be the pure pattern or more commonly
9.7.1.7 Poorly Differentiated occur within otherwise typical moderately and poorly dif-
These tumors are characterized by sarcomatoid growth ferentiated SLCTs with or without heterologous elements.
pattern with focal pattern of conventional SLCTs. They are Morphologically these tumors exhibit irregularly branched,
formed of solid sheets of poorly differentiated cells that elongated, narrow, tubules that include poorly formed tubules
range from spindle cells mimicking sarcoma to epithelial- and cysts with intraluminal polypoid projections lined by
like with high nuclear grade and invariably high mitotic epithelial cells with varying degrees of stratification and
316 M. M. Desouki
Fig. 9.72 Poorly Differentiated Sertoli-Leydig cell tumor. The tumor Fig. 9.74 Sertoli-Leydig cell tumor with heterologous mucinous element.
cells exhibit sarcomatoid growth pattern. They are formed of spindle The tumor exhibits the morphology of moderately differentiated tumor
cells mimicking sarcoma with high nuclear grade and absent tubule with alternating hypercellular and hypocellular stroma in a lobulated pat-
formation tern. Notice the presence of bland gastrointestinal mucinous element
9.7.1.10 Biomarkers
SLCTs stain strongly for α-inhibin (100%), calretinin (89%),
WT1 (100%), AE1/AE3 (100%), vimentin, CD56 (100%),
SF-1 (100%), CD99 (Sertoli component only; 50% of cases),
FOXL2 (Sertoli component only; 50%) and CD10 (44%)
while negative for CK5/6 and EMA [13, 15, 17, 18, 20].
Fig. 9.73 Poorly Differentiated Sertoli-Leydig cell tumor. The same SLCTs may exhibit focal staining for alpha fetoprotein [1]
tumor exhibits sarcomatoid (a) and epithelioid (b) growth patterns (Table 9.4).
9 Sex Cord-Stromal Tumors of the Ovary 317
Fig. 9.77 Endometrioid carcinoma with corded and hyalinized pat- Fig. 9.78 Carcinosarcoma. The growth pattern of the epithelial and
tern. The tumor mimics Sertoli Leydig cell tumor with no conventional mesenchymal components mimics sertoli-Leydig cell tumor
adenocarcinoma component in this field
Mucinous tumors: Because of the different patterns of elements have no specific prognostic significance whereas
SLCT and the relatively common association with mucinous the significance of mesenchymal elements is generally more
heterologous elements, mucinous tumors lend themselves in unfavorable. In general, tumor rupture, the presence of mes-
this differential. On microscopic examination, most heterol- enchymal heterologous elements and high stage at presenta-
ogous tumors contain classic foci of Sertoli cells between the tion carry a worse outcome [5, 101].
mucinous component. This emphasizes the requirement of
adequately sampling grossly mucinous neoplasms.
Serous tumors: SLCT with retiform pattern may be misdi- Key Diagnostic Points
agnosed as serous carcinoma. Clinical information including Sertoli Leydig Cell Tumor
the patient age is relevant to the differential diagnosis Gross
between SLCT and a serous neoplasm or even a malignant
mixed mesodermal tumor (carcinosarcoma). The latter usu- • Unilateral (98.5%), large (average of 13.5 cm),
ally occur at a different age range with very rare exceptions. solid, lobulated and yellow with or without cyst
Germ cell tumor: Differentiating a SLCT from a germ formation.
cell tumor including embryonal carcinoma or reticular- • Retiform tumors tend to be cystic.
microcystic pattern of YST rely on the presence of regions • High-grade tumors have gross hemorrhage and
diagnostic in one direction or the other [1, 93]. necrosis.
Ovarian edema: The presence of typical nodular pattern • Tumors with mucinous heterologous elements may
with cellular growth “blue nodules” separated by edematous be multicystic with mucinous contents.
stroma in SLCT is usually present (see the differential diag- Microscopic
nosis section under thecoma above).
Sarcomas and poorly differentiated carcinomas including • Types: Well-differentiated, moderately differenti-
carcinosarcoma: Important differential diagnostic consider- ated, poorly differentiated, retiform pattern and
ations especially to poorly differentiated SLCT that have with heterologous elements.
been discussed under the differential diagnosis sections of Well-differentiated:
other tumors above (Fig. 9.78).
• Small tubules with open lumina lined by uniform
lipid rich or oxyphilic cells.
9.7.1.13 Prognosis
• Elongated nuclei arranged perpendicular to the
The grade of SLCT drives the prognosis but overall the prog-
basement membranes.
nosis is favorable. Well-differentiated tumors have 100%
• Leydig cells are present throughout the tumor.
survival. Moderately differentiated tumors have 10% malig-
• No heterologous or retiform elements.
nant behavior. Retiform and poorly differentiated tumors
carry a worse outcome with recurrence within 2 years Moderately-differentiated:
especially in the peritoneal cavity. Mucinous heterologous
9 Sex Cord-Stromal Tumors of the Ovary 319
9.7.2 S
ex Cord-Stromal Tumors Not References
Otherwise Specified (Unclassified)
1. Young RH. Ovarian sex cord-stromal tumours and their mimics.
Pathology. 2018;50(1):5–15.
9.7.2.1 Definition 2. Hummitzsch K, Irving-Rodgers HF, Hatzirodos N, et al. A new
SCST the lacks specific features of a particular tumor type or model of development of the mammalian ovary and follicles. PLoS
with features intermediate between those of GCT and One. 2013;8(2):e55578.
320 M. M. Desouki
3. Kurman RJ, Carcangiu ML, Herrington CS, Young RH. WHO 22. Lim D, Oliva E. Ovarian sex cord-stromal tumours: an update in
Classification of Tumours of the Female Reproductive Organs recent molecular advances. Pathology. 2018;50(2):178–89.
(IARC WHO Classification of Tumours). 4th ed. Lyon, International 23. Dockerty M, Masson JC. Ovarian fibromas: a clinical and patho-
Agency for Research on Cancer (IARC); 2014. logic study of two hundred and eighty-three cases. Am J Obstet
4. Young RH, Dickersin GR, Scully RE. Juvenile granulosa cell tumor Gynecol. 1944;47:741–52.
of the ovary. A clinicopathological analysis of 125 cases. Am J Surg 24. Prat J, Scully RE. Cellular fibromas and fibrosarcomas of the
Pathol. 1984;8(8):575–96. ovary: a comparative clinicopathologic analysis of seventeen cases.
5. Young RH, Prat J, Scully RE. Ovarian Sertoli-Leydig cell tumors Cancer. 1981;47(11):2663–70.
with heterologous elements. I. Gastrointestinal epithelium and 25. Michal M, Kacerovska D, Mukensnabl P, et al. Ovarian fibromas
carcinoid: a clinicopathologic analysis of thirty-six cases. Cancer. with heavy deposition of hyaline globules: a diagnostic pitfall. Int J
1982;50(11):2448–56. Gynecol Pathol. 2009;28(4):356–61.
6. Burandt E, Young RH. Thecoma of the ovary: a report of 70 cases 26. Leung WY, Schwartz PE, Ng HT, Yang-Feng TL. Trisomy 12 in
emphasizing aspects of its histopathology different from those benign fibroma and granulosa cell tumor of the ovary. Gynecol
often portrayed and its differential diagnosis. Am J Surg Pathol. Oncol. 1990;38(1):28–31.
2014;38(8):1023–32. 27. Tsuji T, Catasus L, Prat J. Is loss of heterozygosity at 9q22.3
7. Young RH, Scully RE. Ovarian sex cord-stromal tumors. Problems (PTCH gene) and 19p13.3 (STK11 gene) involved in the patho-
in differential diagnosis. Pathol Annu. 1988;23(Pt 1):237–96. genesis of ovarian stromal tumors? Hum Pathol. 2005;36(7):
8. Irving JA, Alkushi A, Young RH, Clement PB. Cellular fibromas of 792–6.
the ovary: a study of 75 cases including 40 mitotically active tumors 28. Irving JA, McCluggage WG. Ovarian spindle cell lesions: a review
emphasizing their distinction from fibrosarcoma. Am J Surg Pathol. with emphasis on recent developments and differential diagnosis.
2006;30(8):929–38. Adv Anat Pathol. 2007;14(5):305–19.
9. Gorlin RJ. Nevoid basal-cell carcinoma syndrome. Medicine. 29. Young RH, Scully RE. Fibromatosis and massive edema of the
1987;66(2):98–113. ovary, possibly related entities: a report of 14 cases of fibro-
10. Tavassoli F, Devilee P, The International Agency for Research
matosis and 11 cases of massive edema. Int J Gynecol Pathol.
on Cancer. In: Tavassoli F, Devilee P, editors. World Health 1984;3(2):153–78.
Organization: tumours of the breast and female genital organs 30. Nielsen GP, Young RH. Fibromatosis of soft tissue type involving
(WHO/IARC Classification of Tumours). 1st ed. Lyon: IARCPress- the female genital tract: a report of two cases. Int J Gynecol Pathol.
WHO; 2003. 1997;16(4):383–6.
11. Costa MJ, Ames PF, Walls J, Roth LM. Inhibin immunohistochem- 31. Lerwill MF, Sung R, Oliva E, Prat J, Young RH. Smooth muscle
istry applied to ovarian neoplasms: a novel, effective, diagnostic tumors of the ovary: a clinicopathologic study of 54 cases empha-
tool. Hum Pathol. 1997;28(11):1247–54. sizing prognostic criteria, histologic variants, and differential diag-
12. McCluggage WG, Maxwell P. Immunohistochemical staining for nosis. Am J Surg Pathol. 2004;28(11):1436–51.
calretinin is useful in the diagnosis of ovarian sex cord-stromal 32. Irving JA, Lerwill MF, Young RH. Gastrointestinal stromal tumors
tumours. Histopathology. 2001;38(5):403–8. metastatic to the ovary: a report of five cases. Am J Surg Pathol.
13. Zhao C, Vinh TN, McManus K, Dabbs D, Barner R, Vang
2005;29(7):920–6.
R. Identification of the most sensitive and robust immunohis- 33. Lee HY, Ahmed Q. Fibrosarcoma of the ovary arising in a fibroth-
tochemical markers in different categories of ovarian sex cord- ecomatous tumor with minor sex cord elements. A case report and
stromal tumors. Am J Surg Pathol. 2009;33(3):354–66. review of the literature. Arch Pathol Lab Med. 2003;127(1):81–4.
14. McCluggage WG. Immunohistochemical and functional biomark- 34. Ozdemir O, Sari ME, Sen E, Ilgin BU, Guresci S, Atalay
ers of value in female genital tract lesions. Int J Gynecol Pathol. CR. Primary ovarian fibrosarcoma: a case report and review of the
2006;25(2):101–20. Epub 2006/04/25. literature. J Exp Ther Oncol. 2016;11(3):225–35.
15. Deavers MT, Malpica A, Liu J, Broaddus R, Silva EG. Ovarian 35. Kraemer BB, Silva EG, Sneige N. Fibrosarcoma of ovary. A new
sex cord-stromal tumors: an immunohistochemical study component in the nevoid basal-cell carcinoma syndrome. Am J
including a comparison of calretinin and inhibin. Mod Pathol. Surg Pathol. 1984;8(3):231–6.
2003;16(6):584–90. 36. Christman JE, Ballon SC. Ovarian fibrosarcoma associated with
16. He H, Luthringer DJ, Hui P, Lau SK, Weiss LM, Chu PG. Expression Maffucci's syndrome. Gynecol Oncol. 1990;37(2):290–1.
of CD56 and WT1 in ovarian stroma and ovarian stromal tumors. 37. Tsuji T, Kawauchi S, Utsunomiya T, Nagata Y, Tsuneyoshi
Am J Surg Pathol. 2008;32(6):884–90. M. Fibrosarcoma versus cellular fibroma of the ovary: a compara-
17. McCluggage WG, McKenna M, McBride HA. CD56 is a sensitive tive study of their proliferative activity and chromosome aberrations
and diagnostically useful immunohistochemical marker of ovarian using MIB-1 immunostaining, DNA flow cytometry, and fluores-
sex cord-stromal tumors. Int J Gynecol Pathol. 2007;26(3):322–7. cence in situ hybridization. Am J Surg Pathol. 1997;21(1):52–9.
18. Oliva E, Garcia-Miralles N, Vu Q, Young RH. CD10 expression 38. Young RH, Scully RE. Sarcomas metastatic to the ovary: a report of
in pure stromal and sex cord-stromal tumors of the ovary: an 21 cases. Int J Gynecol Pathol. 1990;9(3):231–52.
immunohistochemical analysis of 101 cases. Int J Gynecol Pathol. 39. Oliva E, Young RH, Amin MB, Clement PB. An immunohisto-
2007;26(4):359–67. chemical analysis of endometrial stromal and smooth muscle
19. Yao DX, Soslow RA, Hedvat CV, Leitao M, Baergen RN. Melan-A tumors of the uterus: a study of 54 cases emphasizing the impor-
(A103) and inhibin expression in ovarian neoplasms. Appl tance of using a panel because of overlap in immunoreactivity for
Immunohistochem Mol Morphol. 2003;11(3):244–9. individual antibodies. Am J Surg Pathol. 2002;26(4):403–12.
20. Al-Agha OM, Huwait HF, Chow C, et al. FOXL2 is a sensitive and 40. Bjorkholm E, Silfversward C. Theca-cell tumors. Clinical features
specific marker for sex cord-stromal tumors of the ovary. Am J Surg and prognosis. Acta Radiol Oncol. 1980;19(4):241–4.
Pathol. 2011;35(4):484–94. 41. Staats PN, McCluggage WG, Clement PB, Young RH. Luteinized
21. D’Angelo E, Mozos A, Nakayama D, et al. Prognostic significance thecomas (thecomatosis) of the type typically associated with
of FOXL2 mutation and mRNA expression in adult and juvenile sclerosing peritonitis: a clinical, histopathologic, and immu-
granulosa cell tumors of the ovary. Mod Pathol. 2011;24(10):1360– nohistochemical analysis of 27 cases. Am J Surg Pathol.
7. Epub 2011/05/31 2008;32(9):1273–90.
9 Sex Cord-Stromal Tumors of the Ovary 321
42. Young RH, Scully RE. Ovarian stromal tumors with minor sex 61. Jones MW, Harri R, Dabbs DJ, Carter GJ. Immunohistochemical
cord elements: a report of seven cases. Int J Gynecol Pathol. profile of steroid cell tumor of the ovary: a study of 14 cases and a
1983;2(3):227–34. review of the literature. Int J Gynecol Pathol. 2010;29(4):315–20.
43. Clement PB, Young RH, Hanna W, Scully RE. Sclerosing
62. Mehta JM, Miller JL, Cannon AJ, Mardekian SK, Kenyon LC,
peritonitis associated with luteinized thecomas of the ovary. Jabbour SA. Ovarian leydig cell hyperplasia: an unusual case of
A clinicopathological analysis of six cases. Am J Surg Pathol. virilization in a postmenopausal woman. Case Rep Endocrinol.
1994;18(1):1–13. 2014;2014:762745.
44. McCluggage WG, Staats PN, Gilks CB, Clement PB, Young
63. Irving JA, Young RH. Microcystic stromal tumor of the ovary:
RH. Luteinized thecomas (thecomatosis) associated with scle- report of 16 cases of a hitherto uncharacterized distinctive ovarian
rosing peritonitis exhibit positive staining with sex cord mark- neoplasm. Am J Surg Pathol. 2009;33(3):367–75.
ers steroidogenic factor-1 (SF-1) and FOXL2. Am J Surg Pathol. 64. McCluggage WG, Irving JA, Chong AS, et al. Ovarian microcystic
2013;37(9):1458–9. stromal tumors are characterized by alterations in the beta-catenin-
45. Fletcher JA, Gibas Z, Donovan K, et al. Ovarian granulosa-
APC pathway and may be an extracolonic manifestation of familial
stromal cell tumors are characterized by trisomy 12. Am J Pathol. adenomatous polyposis. Am J Surg Pathol. 2018;42(1):137–9.
1991;138(3):515–20. 65. Irving JA, Lee CH, Yip S, Oliva E, McCluggage WG, Young
46. Bjorkholm E, Pettersson F. Granulosa-cell and theca-cell tumors. RH. Microcystic stromal tumor: a distinctive ovarian sex cord-
The clinical picture and long-term outcome for the Radiumhemmet stromal neoplasm characterized by FOXL2, SF-1, WT-1, cyclin D1,
series. Acta Obstet Gynecol Scand. 1980;59(4):361–5. and beta-catenin nuclear expression and CTNNB1 mutations. Am J
47. Chalvardjian A, Scully RE. Sclerosing stromal tumors of the ovary. Surg Pathol. 2015;39(10):1420–6.
Cancer. 1973;31(3):664–70. 66. Young RH, Scully RE. Ovarian sex cord-stromal tumors with
48. Lam RM, Geittmann P. Sclerosing stromal tumor of the ovary. A bizarre nuclei: a clinicopathologic analysis of 17 cases. Int J
light, electron microscopic and enzyme histochemical study. Int J Gynecol Pathol. 1983;1(4):325–35.
Gynecol Pathol. 1988;7(3):280–90. 67. Byrne P, Vella EJ, Rollason T, Frampton J. Ovarian fibromatosis
49. Kawauchi S, Tsuji T, Kaku T, Kamura T, Nakano H, Tsuneyoshi with minor sex cord elements. Case report. Br J Obstet Gynaecol.
M. Sclerosing stromal tumor of the ovary: a clinicopathologic, 1989;96(2):245–8.
immunohistochemical, ultrastructural, and cytogenetic analy- 68. Stenwig JT, Hazekamp JT, Beecham JB. Granulosa cell tumors of
sis with special reference to its vasculature. Am J Surg Pathol. the ovary. A clinicopathological study of 118 cases with long-term
1998;22(1):83–92. follow-up. Gynecol Oncol. 1979;7(2):136–52.
50. Tiltman AJ, Haffajee Z. Sclerosing stromal tumors, thecomas,
69. Fox H, Agrawal K, Langley FA. A clinicopathologic study of 92
and fibromas of the ovary: an immunohistochemical profile. Int J cases of granulosa cell tumor of the ovary with special reference to
Gynecol Pathol. 1999;18(3):254–8. the factors influencing prognosis. Cancer. 1975;35(1):231–41.
51. Vang R, Bague S, Tavassoli FA, Prat J. Signet-ring stromal tumor 70. Young RH, Dudley AG, Scully RE. Granulosa cell, Sertoli-Leydig
of the ovary: clinicopathologic analysis and comparison with cell, and unclassified sex cord-stromal tumors associated with preg-
Krukenberg tumor. Int J Gynecol Pathol. 2004;23(1):45–51. nancy: a clinicopathological analysis of thirty-six cases. Gynecol
52. Cashell AW, Jerome WG, Flores E. Signet ring stromal tumor of the Oncol. 1984;18(2):181–205.
ovary occurring in conjunction with brenner tumor. Gynecol Oncol. 71. Ahmed E, Young RH, Scully RE. Adult granulosa cell tumor of the
2000;77(2):323–6. ovary with foci of hepatic cell differentiation: a report of four cases
53. Ramzy I. Signet-ring stromal tumor of ovary. Histochemical, light, and comparison with two cases of granulosa cell tumor with Leydig
and electron microscopic study. Cancer. 1976;38(1):166–72. cells. Am J Surg Pathol. 1999;23(9):1089–93.
54. Kiyokawa T, Young RH, Scully RE. Krukenberg tumors of the 72. Schumer ST, Cannistra SA. Granulosa cell tumor of the ovary. J
ovary: a clinicopathologic analysis of 120 cases with emphasis Clin Oncol. 2003;21(6):1180–9.
on their variable pathologic manifestations. Am J Surg Pathol. 73. Schmidt D, Ovitt CE, Anlag K, et al. The murine winged-helix tran-
2006;30(3):277–99. scription factor Foxl2 is required for granulosa cell differentiation
55. Baker PM, Oliva E, Young RH, Talerman A, Scully RE. Ovarian and ovary maintenance. Development. 2004;131(4):933–42.
mucinous carcinoids including some with a carcinomatous compo- 74. Moumne L, Batista F, Benayoun BA, et al. The mutations and
nent: a report of 17 cases. Am J Surg Pathol. 2001;25(5):557–68. potential targets of the forkhead transcription factor FOXL2. Mol
Epub 2001/05/09 Cell Endocrinol. 2008;282(1–2):2–11.
56. Seidman JD, Abbondanzo SL, Bratthauer GL. Lipid cell (steroid 75. Shah SP, Kobel M, Senz J, et al. Mutation of FOXL2 in granulosa-
cell) tumor of the ovary: immunophenotype with analysis of poten- cell tumors of the ovary. N Engl J Med. 2009;360(26):2719–29.
tial pitfall due to endogenous biotin-like activity. Int J Gynecol Epub 2009/06/12
Pathol. 1995;14(4):331–8. 76. Kommoss S, Anglesio MS, Mackenzie R, et al. FOXL2 molecular
57. Norris HJ, Taylor HB. Nodular theca-lutein hyperplasia of preg- testing in ovarian neoplasms: diagnostic approach and procedural
nancy (so-called “pregnancy luteoma”). A clinical and pathologic guidelines. Mod Pathol. 2013;26(6):860–7.
study of 15 cases. Am J Clin Pathol. 1967;47(5):557–66. 77. Goulvent T, Ray-Coquard I, Borel S, et al. DICER1 and FOXL2
58. Hayes MC, Scully RE. Ovarian steroid cell tumors (not otherwise mutations in ovarian sex cord-stromal tumours: a GINECO Group
specified). A clinicopathological analysis of 63 cases. Am J Surg study. Histopathology. 2016;68(2):279–85.
Pathol. 1987;11(11):835–45. 78. Farkkila A, McConechy MK, Yang W, et al. FOXL2 402C>G muta-
59. Zhang J, Young RH, Arseneau J, Scully RE. Ovarian stromal tumors tion can be identified in the circulating tumor DNA of patients with
containing lutein or Leydig cells (luteinized thecomas and stromal adult-type granulosa cell tumor. J Mol Diagn. 2017;19(1):126–36.
Leydig cell tumors)--a clinicopathological analysis of fifty cases. 79. Crisponi L, Deiana M, Loi A, et al. The putative forkhead transcrip-
Int J Gynecol Pathol. 1982;1(3):270–85. tion factor FOXL2 is mutated in blepharophimosis/ptosis/epican-
60. Paraskevas M, Scully RE. Hilus cell tumor of the ovary. A
thus inversus syndrome. Nat Genet. 2001;27(2):159–66.
clinicopathological analysis of 12 Reinke crystal-positive and 80. Roth LM, Liban E, Czernobilsky B. Ovarian endometrioid tumors
nine crystal-negative cases. Int J Gynecol Pathol. 1989;8(4): mimicking Sertoli and Sertoli-Leydig cell tumors: Sertoliform vari-
299–310. ant of endometrioid carcinoma. Cancer. 1982;50(7):1322–31.
322 M. M. Desouki
81. Tacha D, Qi W, Zhou D, Bremer R, Cheng L. PAX8 mouse 100. Conlon N, Schultheis AM, Piscuoglio S, et al. A survey of
monoclonal antibody [BC12] recognizes a restricted epitope and DICER1 hotspot mutations in ovarian and testicular sex cord-
is highly sensitive in renal cell and ovarian cancers but does not stromal tumors. Mod Pathol. 2015;28(12):1603–12.
cross-react with b cells and tumors of pancreatic origin. Appl 101. Prat J, Young RH, Scully RE. Ovarian Sertoli-Leydig cell
Immunohistochem Mol Morphol. 2013;21(1):59–63. tumors with heterologous elements. II. Cartilage and skeletal
82. Kato N, Fukase M, Ono I, Matsumoto K, Okazaki E, Motoyama muscle: a clinicopathologic analysis of twelve cases. Cancer.
T. Sertoli-stromal cell tumor of the ovary: immunohisto- 1982;50(11):2465–75.
chemical, ultrastructural, and genetic studies. Hum Pathol. 102. Robboy SJ, Scully RE, Norris HJ. Primary trabecular carcinoid of
2001;32(8):796–802. the ovary. Obstet Gynecol. 1977;49(2):202–7. Epub 1977/02/01
83. Szyfelbein WM, Young RH, Scully RE. Struma ovarii simulating 103. Young RH, Scully RE. Ovarian tumors of probable wolffian
ovarian tumors of other types. A report of 30 cases. Am J Surg origin. A report of 11 cases. Am J Surg Pathol. 1983;7(2):125–35.
Pathol. 1995;19(1):21–9. 104. Tiltman AJ, Allard U. Female adnexal tumours of probable
84. Young RH, Oliva E, Scully RE. Small cell carcinoma of the ovary, Wolffian origin: an immunohistochemical study comparing
hypercalcemic type. A clinicopathological analysis of 150 cases. tumours, mesonephric remnants and paramesonephric derivatives.
Am J Surg Pathol. 1994;18(11):1102–16. Histopathology. 2001;38(3):237–42.
85. Cannistra SA. Cancer of the ovary. N Engl J Med. 105. Scully RE. Sex cord tumor with annular tubules a distinc-
1993;329(21):1550–9. tive ovarian tumor of the Peutz-Jeghers syndrome. Cancer.
86. Mancari R, Portuesi R, Colombo N. Adult granulosa cell tumours 1970;25(5):1107–21.
of the ovary. Curr Opin Oncol. 2014;26(5):536–41. 106. Young RH, Welch WR, Dickersin GR, Scully RE. Ovarian sex
87. Jarboe EA, Hirschowitz SL, Geiersbach KB, Wallander ML, Tripp cord tumor with annular tubules: review of 74 cases including 27
SR, Layfield LJ. Juvenile granulosa cell tumors: immunoreactiv- with Peutz-Jeghers syndrome and four with adenoma malignum
ity for CD99 and Fli-1 and EWSR1 translocation status: a study of of the cervix. Cancer. 1982;50(7):1384–402.
11 cases. Int J Gynecol Pathol. 2014;33(1):11–5. 107. Gloor E. Ovarian sex cord tumor with annular tubules.
88. Kalfa N, Ecochard A, Patte C, et al. Activating mutations of Clinicopathologic report of two benign and one malignant cases
the stimulatory g protein in juvenile ovarian granulosa cell with long follow-ups. Virchows Arch A Pathol Anat Histol.
tumors: a new prognostic factor? J Clin Endocrinol Metab. 1979;384(2):185–93.
2006;91(5):1842–7. 108. Han Y, Li S, Wu L, Zhang X, Cao D. Non-Peutz-Jeghers
89. Bessiere L, Todeschini AL, Auguste A, et al. A hot-spot of in- syndrome- associated ovarian sex cord tumor with annular
frame duplications activates the oncoprotein AKT1 in juvenile tubules: report of a malignant case. J Obstet Gynaecol Res.
granulosa cell tumors. EBioMedicine. 2015;2(5):421–31. 2016;42(2):224–7.
90. Heravi-Moussavi A, Anglesio MS, Cheng SW, et al. Recurrent 109. Young RH, Scully RE. Ovarian Sertoli-Leydig cell tumors. A
somatic DICER1 mutations in nonepithelial ovarian cancers. N clinicopathological analysis of 207 cases. Am J Surg Pathol.
Engl J Med. 2012;366(3):234–42. 1985;9(8):543–69.
91. Witkowski L, Mattina J, Schonberger S, et al. DICER1 hotspot 110. Young RH, Scully RE. Well-differentiated ovarian Sertoli-Leydig
mutations in non-epithelial gonadal tumours. Br J Cancer. cell tumors: a clinicopathological analysis of 23 cases. Int J
2013;109(10):2744–50. Gynecol Pathol. 1984;3(3):277–90.
92. Young RH, Lawrence WD, Scully RE. Juvenile granulosa cell 111. Young RH, Perez-Atayde AR, Scully RE. Ovarian Sertoli-Leydig
tumor–another neoplasm associated with abnormal chromosomes cell tumor with retiform and heterologous components. Report of
and ambiguous genitalia. A report of three cases. Am J Surg a case with hepatocytic differentiation and elevated serum alpha-
Pathol. 1985;9(10):737–43. fetoprotein. Am J Surg Pathol. 1984;8(9):709–18.
93. Young RH. New and unusual aspects of ovarian germ cell tumors. 112. McClean GE, Kurian S, Walter N, Kekre A, McCluggage
Am J Surg Pathol. 1993;17(12):1210–24. WG. Cervical embryonal rhabdomyosarcoma and ovarian Sertoli-
94. Witkowski L, Carrot-Zhang J, Albrecht S, et al. Germline and Leydig cell tumour: a more than coincidental association of two
somatic SMARCA4 mutations characterize small cell carcinoma rare neoplasms? J Clin Pathol. 2007;60(3):326–8.
of the ovary, hypercalcemic type. Nat Genet. 2014;46(5):438–43. 113. Young RH. Sex cord-stromal tumors of the ovary and testis: their
95. Erdreich-Epstein A, Monforte HL, Lavey RS, Joshi S, Phillips similarities and differences with consideration of selected prob-
JD, Villablanca JG. Successful multimodality therapy of recurrent lems. Mod Pathol. 2005;18(Suppl 2):S81–98.
multifocal juvenile granulosa cell tumor of the ovary. J Pediatr 114. Foulkes WD, Priest JR, Duchaine TF. DICER1: mutations, microR-
Hematol Oncol. 2002;24(3):229–33. NAs and mechanisms. Nat Rev Cancer. 2014;14(10):662–72.
96. Frausto SD, Geisler JP, Fletcher MS, Sood AK. Late recurrence of 115. Schultz KA, Pacheco MC, Yang J, et al. Ovarian sex cord-stromal
juvenile granulosa cell tumor of the ovary. Am J Obstet Gynecol. tumors, pleuropulmonary blastoma and DICER1 mutations: a
2004;191(1):366–7. report from the International Pleuropulmonary Blastoma Registry.
97. Oliva E, Alvarez T, Young RH. Sertoli cell tumors of the ovary: Gynecol Oncol. 2011;122(2):246–50.
a clinicopathologic and immunohistochemical study of 54 cases. 116. Slade I, Bacchelli C, Davies H, et al. DICER1 syndrome: clarify-
Am J Surg Pathol. 2005;29(2):143–56. ing the diagnosis, clinical features and management implications
98. Young RH, Scully RE. Ovarian Sertoli cell tumors: a report of 10 of a pleiotropic tumour predisposition syndrome. J Med Genet.
cases. Int J Gynecol Pathol. 1984;2(4):349–63. 2011;48(4):273–8.
99. Ferry JA, Young RH, Engel G, Scully RE. Oxyphilic Sertoli cell 117. Seidman JD. Unclassified ovarian gonadal stromal tumors.
tumor of the ovary: a report of three cases, two in patients with the A clinicopathologic study of 32 cases. Am J Surg Pathol.
Peutz-Jeghers syndrome. Int J Gynecol Pathol. 1994;13(3):259–66. 1996;20(6):699–706.
Secondary Tumors of the Ovary
10
Kelley Carrick and Wenxin Zheng
© Science Press & Springer Nature Singapore Pte Ltd. 2019 323
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_10
324 K. Carrick and W. Zheng
[3–5] to approximately 21–30% in Eastern nations [6, 7]. metastasize to the ovary via blood vessels or lymphatics, as
The proportion of STOs and the relative frequency of pri- evidenced by the notable intravascular tumor identified in
mary sites of origin vary across different geographic regions many cases of STO. Direct spread is a frequent pathway for
due to geographic differences in the prevalence of various carcinomas of the fallopian tube and uterus to involve the
extraovarian primary neoplasms and their relative tendency ovary, for some colorectal and appendiceal carcinomas, and
to metastasize to the ovaries [1]. While gastric cancer, which for the rare mesothelioma [2, 12]. Tumor may also reach the
has a relatively high rate of ovarian spread, accounts for ovary through the fallopian tube lumen, a route deemed to
23.4% and 30.4% of tumors metastatic to the ovary in Japan account for spread of some carcinomas from the uterine cor-
and Korea, respectively, it is found in a reported 4.5% of pus and cervix [13, 14]. Transperitoneal spread, another
STOs in the Netherlands, which reflects its incidence in route of ovarian involvement, is typically accompanied by
those countries [7–9]. In several series examining secondary ovarian surface implantation, involvement of the superficial
ovarian malignancies in Europe and the USA, tumors of ovarian cortex, and general peritoneal spread. A good exam-
colorectal, breast, endometrial, and appendiceal origin have ple of this kind of spread is from high-grade serous carci-
been demonstrated to be the most common [4, 5, 9]. Patient noma of the fallopian tube.
age at diagnosis of a STO appears to be associated with the
origin and typical age distribution of the primary tumor [1].
For some of the most common types of metastatic tumors 10.5 G
eneral Pathologic Findings
(intestinal, gastric, and breast), the average age of patients and Clinical Correlation
presenting with ovarian involvement is lower than the aver-
age age of those without ovarian spread, which suggests that STOs may have considerable morphologic overlap with
the increased vascularization of ovaries in young women ovarian primary neoplasms, presenting diagnostic challenge
renders them more receptive to metastasis [2]. in many cases. The most challenging issues in differential
diagnosis typically involve mucinous tumors, endometrioid,
and endometrioid-like tumors. Over the years, several gross
10.3 Clinical Presentations and microscopic features have been proposed in the litera-
ture to aid in the distinction of primary versus secondary
Clinically, ovarian metastasis may be symptomatic or asymp- ovarian tumors, the most important of these being laterality
tomatic, and may present synchronously or metachronously and size of the ovarian tumor. Several studies have shown
with the extraovarian primary tumor. Ovarian metastatic that an algorithm using tumor size and laterality can accu-
tumor tends to remain asymptomatic until it reaches a certain rately distinguish a substantial majority of primary and met-
size. Presenting symptoms, found in a reported 70% of astatic tumors (bilateral tumors of any size, or unilateral
patients, are nonspecific and do not clearly differ from those tumor less than 10 cm more likely to be metastatic; unilat-
related to primary ovarian neoplasia [10]. Symptoms related eral tumor greater than or equal to 10 cm more likely an
to an ovarian mass and to an advanced stage of disease can ovarian primary neoplasm) [15, 16]. A 2008 study examin-
include abdominal pain, weight loss, and increasing abdomi- ing 194 primary and metastatic ovarian tumors demon-
nal girth [4]. As STOs, like primary tumors of the ovary, can strated that adjusting the size criterion of the algorithm to
induce ovarian stromal luteinization with consequent hor- 13 cm rather than 10 cm optimized performance of the algo-
mone production (“functioning stroma”), the patient may rithm, correctly classifying 87% of tumors overall, includ-
also experience consequences of increased estrogen, proges- ing 98% of primary tumors and 82% of metastatic tumors
terone, or androgens including abnormal uterine bleeding, [17]. Tumor size and laterality are helpful guidelines, espe-
virilization, and hirsutism [11]. Importantly, the ovarian cially in the operative setting. Exceptions are common,
metastasis may be the presenting sign of disease from a however, particularly among cases of colorectal and endo-
small, clinically occult non-ovarian primary tumor (for cervical adenocarcinomas [15, 17]; thus, laterality and
example, a small gastric carcinoma). In some cases, the pri- tumor size alone are not sufficient for determination of pri-
mary tumor may not be discovered until years after discov- mary site. In fact, there is a growing body of evidence to
ery of the ovarian metastasis or until autopsy. The primary indicate that a significant proportion of STOs are unilateral
tumor may remain unknown in up to 15–20% of cases [4, 9]. and relatively large [18]. A recent study from the Netherlands
including 2312 cases of tumor metastatic to the ovary from
various primary sites demonstrated that STOs from all sites
10.4 R
outes of Tumor Metastasis taken together were bilateral in the majority (46.3%) of
to the Ovary cases, with 63.9% of breast, 62.9% of gastric, and 58.9% of
appendiceal primaries generating bilateral metastases. In
Tumor may secondarily involve the ovary via any of several contrast, a minority (40.2%) of colorectal cancers had bilat-
routes of spread, which accounts for some of the more typi- eral metastases in this study [9]. Another study including 19
cal gross and microscopic features of STO. Tumor may tumors metastatic from the lower and upper gastrointestinal
10 Secondary Tumors of the Ovary 325
tract demonstrated that over 80% of these tumors were 7. Tubules and follicle-like spaces can be seen in metastases
greater than 10 cm [19]. from various sites and can mimic ovarian primary tumors,
Based on the above understanding and our own experi- particularly sex cord-stromal tumors [12].
ence, we summarize general gross features which are sug- 8. Ovarian stromal luteinization and resulting hormonal
gestive of metastasis to the ovary as follows: manifestations (“functioning stroma”) are nonspecific
and can appear in association with primary or metastatic
1. Bilateral ovarian involvement (notable exception: tubo- tumors.
ovarian serous carcinoma, which is often bilateral). As
metastatic tumor may be small and not enlarge the ovary, As noted above, STOs and primary ovarian neoplasms
adequate sampling is needed to rule out metastasis. may have overlapping morphologic and clinical features that
2. Tumor size less than 10–12 cm. confound diagnosis. Thus, in addition to careful gross and
3. A multinodular growth pattern. microscopic examination of the tumor, consideration of met-
4. Presence of tumor on the ovarian surface and/or in the astatic disease and clinical correlation are necessary when
superficial cortex. the diagnosis is not clear. This assessment may require intra-
operative and/or postoperative evaluation for an extraovarian
As with gross features, microscopic features of STOs may primary neoplasm if pathologic examination suggests the
bear similarity to microscopic features of primary ovarian possibility of a metastasis, or if the pattern of tumor spread is
tumors, particularly in mucinous and endometrioid/ unusual for an ovarian primary neoplasm. A directed panel
endometrioid- like tumors. Microscopic features favoring of immunohistochemical stains is useful in many cases and
metastasis include the following [2, 12, 20]: should be undertaken in any case in which the primary ver-
sus secondary nature of the ovarian tumor is uncertain. As
1. Presence of signet ring cells primary and secondary ovarian tumors often have overlap-
2. Infiltrative growth pattern with stromal desmoplasia ping immunohistochemical staining patterns (particularly in
3. Notable variation in growth pattern from one area to another the case of mucinous ovarian tumors arising in association
4. Histomorphology unusual for an ovarian primary with a teratoma, which have an intestinal immunophenotype
5. Multinodular growth pattern indistinguishable from metastatic tumor of gastrointestinal
6. Involvement of the ovarian surface and superficial cortex origin), a panel of immunostains should be employed and
7.
Hilar and/or extraovarian lymphovascular space correlated with gross, microscopic, and clinical findings. It is
involvement also important to note that metastatic tumors may have an
immunohistochemical staining pattern that deviates from the
Other pearls that can be useful in determining the primary usual, expected pattern. For typical immunohistochemical
versus secondary nature of an ovarian tumor include the staining features of primary and secondary ovarian tumors
following: with mucinous and endometrioid/endometrioid-like mor-
phology, see Table 10.1. Additional information regarding
1. A history of an extraovarian primary tumor should raise the use of immunohistochemical stains is provided in subse-
suspicion of metastasis, even if this history is remote. quent sections.
2. Metastatic tumors may be unusually large with thin-
walled cysts mimicking a primary ovarian neoplasm,
even if cysts are not a feature of the extraovarian primary 10.6 Secondary Ovarian Tumors
tumor [2, 12]. from the Gastrointestinal Tract
3. If tumor is bilateral and morphology is mucinous or endo-
metrioid/endometrioid-like, the probability of metastasis 10.6.1 Krukenberg Tumor
increases, as bilateral mucinous and endometrioid ovar-
ian primaries are relatively rare. Definition: The term “Krukenberg tumor” has not been con-
4. A carcinoma with mucinous morphology and either sistently defined over the years but has historically referred
advanced stage or association with pseudomyxoma peri- to tumors metastatic to the ovary that have a significant com-
tonei is likely to be metastatic. ponent of mucin-containing signet ring cells. A recent source
5. For endometrioid/endometrioid-like tumors, an adenofi- defines “Krukenberg tumor” as a metastatic adenocarcinoma
bromatous background or presence of endometriosis in which signet ring cells comprise at least 10% of the neo-
favors a primary ovarian origin. plasm [2]. The stomach, most often the pylorus, is the most
6. The presence of foci of histologically benign-appearing common source for these tumors, representing the primary
and low-grade proliferative mucinous epithelium (border- site in greater than 60% of cases [11]. Other possible primary
line/atypical proliferative features) that might suggest an sites include colon, rectum, breast, gallbladder, bile ducts,
ovarian origin is nonspecific and does appear in meta- appendix, and less commonly the small intestine, pancreas,
static tumors. urinary bladder, renal pelvis, and cervix [11 24]. Rare
Table 10.1 Typical immunohistochemical staining features of primary and secondary tumors with mucinous and endometrioid/endometrioid-like morphology [19, 21–23]
326
Primary site of
origin CK7 CK20 PAX8 ER PR SATB2 CDX2 DPC4/SMAD4 P16
Ovarian Endometrioid: Endometrioid: Endometrioid: Endometrioid: Endometrioid: Endometrioid: Endometrioid: Endometrioid: Endometrioid:
Positive Negative Positive Usually positive Usually positive Negative Negativef Positive Negative to
Mucinous: Mucinous: Mucinous: Mucinous: Mucinous: Mucinous: Mucinous: often Mucinous: Positive patchy positive
Positivea,b Variable positive positive/ usually usually Negativea,e positivef Mucinous:
to negativea,b negativec negatived negatived Negative to
patchy positive
Colorectal Negative Positive Negative Negative Negative Positive Positive Positive (approx. Negative to
(75–90% of (CK20>CK7) 90% of cases) patchy positive
cases negative)g
Appendiceal Negative (70% Positive Negative Negative Negative Positive Positive Positive Negative to focal
of cases)h (CK20>CK7) positive
Gastric Positive/negative Positive/negative Negative Negatived Negatived Negative or Signet ring Positive Negative/positive
positive type: variable
staining
Intestinal type:
positive
Pancreatobiliary Positive Negative/ Negative Negative Negative Negative Negative/ Negative (approx. Negative to focal
positivei positivej 50% of pancreatic positive
carcinomas
negative)
Endocervical Positive Negative Positive/ Negativek Negativek Negativek Negative/ Positive Strong diffuse
negativek positivek positivel
a
Primary ovarian epithelial tumors are typically CK7+/CK20−; thus a CK7−/CK20+ immunophenotype suggests a metastasis. Notable exception: The subset of ovarian mucinous tumors of intes-
tinal phenotype arising in association with a teratoma has an immunohistochemical staining profile indistinguishable from that of metastatic tumors of intestinal origin (CK7−/CK20+/SATB2+/
CDX2+/ER−/PR−/PAX8−)
b
Ovarian primary mucinous tumors of intestinal type are typically positive for CK7, often with diffuse strong positivity for this marker; they are often positive for CK20 as well, although with
patchy variable staining. As noted above, the subset of ovarian mucinous tumors of intestinal phenotype arising in association with teratomas have an immunohistochemical staining profile indis-
tinguishable from that of metastatic tumors of intestinal origin (CK7−/CK20+). Ovarian seromucinous tumors, separated from ovarian mucinous tumors in the 2014 WHO Classification of Tumors
of Female Reproductive Organs, typically have a CK7+/CK20−/PAX8+/CDX2− immunophenotype
c
PAX8 staining is variable in ovarian tumors of mucinous type unassociated with a teratoma, negative in ovarian mucinous tumors arising in association with a teratoma, and typically positive in
ovarian seromucinous tumors
d
ER and PR are typically negative in ovarian mucinous tumors (those with and without an associated teratoma) and positive in ovarian seromucinous tumors. Weak ER or PR staining does not rule
out stomach as a primary site
e
SATB2 is typically negative in ovarian mucinous tumors unassociated with a teratoma, positive in ovarian mucinous tumors associated with a teratoma, negative in ovarian seromucinous tumors,
and positive in metastatic tumors of colorectal or appendiceal origin
f
Ovarian mucinous tumors unassociated with teratoma show variable staining for CDX2. CDX2 is typically positive in ovarian mucinous tumors arising in association with a teratoma, and negative
in ovarian seromucinous tumors. Endometrioid adenocarcinoma may have focal positivity for CDX2 in morules and in foci of squamous differentiation
g
Adenocarcinomas of rectal origin may be positive for CK7
h
Approximately one-third of appendiceal mucinous neoplasms are positive for CK7; in contrast, appendiceal adenocarcinomas of usual intestinal type typically have negative to focal positivity for
CK7, like colonic adenocarcinomas
i
Approximately one-third of pancreatic ductal adenocarcinomas are positive for CK20, with focal positivity most common
j
Approximately 30% of pancreatic ductal adenocarcinomas are positive for CDX2
k
PAX8 is positive in the majority of endocervical adenocarcinomas. ER and PR are negative in the majority of endocervical adenocarcinomas. Endocervical adenocarcinomas may infrequently be
positive for SATB2. CDX2 is expressed in endocervical adenocarcinoma of intestinal type and in some nonintestinal types of endocervical adenocarcinoma
l
K. Carrick and W. Zheng
Endocervical adenocarcinoma expresses strong diffuse positivity for p16 in the vast majority of cases related to HPV; HPV-unrelated subtypes are negative for p16
10 Secondary Tumors of the Ovary 327
Fig. 10.3 Krukenberg tumor of gastric origin. Classic signet ring Fig. 10.5 Krukenberg tumor of gastric origin. Signet ring cells in an
cells active stroma
Fig. 10.4 Krukenberg tumor of gastric origin. Signet ring cells line Fig. 10.6 Krukenberg tumor of gastric origin. The tumor must have a
tubules significant component of signet ring cells (10%) to be classified as a
Krukenberg tumor, but cell types and architectural patterns may vary,
as in this case. Moderately well-formed small glands are present in a
sometimes with an abrupt transition from one architectural background of conspicuous edema, a feature of many Krukenberg
tumors
pattern to another. Tumor cells may be abundant and obvi-
ous, or inconspicuous and sparsely scattered within a fibrous
stroma. The proportion of signet ring cells often varies mark- may be cellular, may have a storiform architecture, and may
edly in different areas of the tumor, being inconspicuous or be prominently edematous or mucoid. Stromal reactivity/
absent in some areas. Signet ring cells have a variable amount desmoplasia may be present or absent. Stromal luteinization
of eosinophilic or basophilic intracytoplasmic mucin and is often present and may be prominent [11]. Lymphovascular
may be disposed singly or in cords, clusters, sheets, or lining invasion is commonly seen, most often in the ovarian hilum
part or all of a tubule (Figs. 10.3, 10.4, and 10.5). Signet ring and at the tumor periphery.
cells are often accompanied by variably well-formed tumor Biomarkers: Periodic acid-Schiff (PAS), Alcian blue, or
glands of various sizes, including both small and large glands mucicarmine stains may be used to highlight intracytoplas-
and cysts (Figs. 10.6 and 10.7). Hollow and solid sertoliform mic mucin. A directed panel of immunohistochemical stains
tubules may be present. Extracellular mucin may be abun- may be employed to aid in the determination of the primary
dant [11]. Stroma may be fibrous with a fibroma-like appear- site. Typical immunohistochemical staining profiles of the
ance, rendering diagnosis more challenging when tumor more common primary sites generating Krukenberg tumors
cells are few, especially in the intraoperative setting. Stroma are as follows [22, 26, 27]:
10 Secondary Tumors of the Ovary 329
Microscopic: The microscopic appearance of these tumors Clinical Features: In one 2006 series of 86 cases of
is similar to that of metastatic tumors of intestinal origin. The colorectal adenocarcinoma metastatic to the ovary, patients
tumors are formed of medium-sized to large tubular glands ranged in age from 19 to 85 years with a median age of 51;
with a pseudoendometrioid appearance, characterized by 24% of patients were younger than 40 [18]. A single case in
columnar epithelium with only focal intracytoplasmic mucin. a 12-year-old patient has been documented [32]. The major-
Other microscopic features, commonly shared by metastatic ity of patients have known intestinal carcinoma prior to diag-
tumors of intestinal origin, include segmental necrosis of epi- nosis of the ovarian tumor, while a minority (less than 10%)
thelium lining the neoplastic glands, high-grade nuclear fea- present with an ovarian mass and postoperative workup dis-
tures, variable to brisk mitotic activity, and frequent closes the intestinal primary. Some patients present with
intraluminal “dirty” necrosis. In addition to the moderate to symptoms related to their colorectal mass (rectal bleeding,
well-formed glands typical of tumors of intestinal type, these change in bowel habits, etc.), while others present with
tumors may show papillary, cribriform, trabecular, and nested symptoms related to either the ovarian mass or the stromal
patterns. Some tumors have abundant intracytoplasmic mucin luteinization generated by the tumor (abnormal uterine
with cells ranging from goblet type to foveolar type, raising bleeding, breast tenderness, etc.). Some patients present with
the differential diagnosis of a primary ovarian mucinous carci- nonspecific symptoms such as abdominal or pelvic pain [18
noma, but have the varied histologic patterns and desmoplasia 32]. Importantly, patients with ovarian metastasis from pri-
typical of mucinous neoplasms metastatic to the ovary. mary colorectal adenocarcinoma may have no symptoms
Microscopic features common to metastatic tumors in general related to the primary tumor and may have a unilateral ovar-
include areas of bland, benign-appearing mucinous epithelium ian mass greater than 10 cm [33].
and cysts lined by flattened epithelium, prominent stromal Pathologic Findings:
edema in areas, and notable morphologic variability within a Gross: Metastasis may be unilateral or bilateral. Several
small zone of tumor [30–32]. Signet ring cells, if present, by series have documented a higher percentage of unilateral
definition do not comprise more than 10% of the tumor. than of bilateral metastasis [9, 33, 34], although this has not
Biomarkers: Gastric adenocarcinoma of intestinal type been demonstrated in all series. Tumor size ranges from 2 to
shows variable positivity for CK7 and CK20, like gastric 24 cm and may be greater than 10 cm [18, 33]. Tumors tend
adenocarcinoma of diffuse type. Unlike gastric adenocarci- to have a smooth capsule without gross evidence of surface
noma of diffuse type, CDX2 is typically strongly, diffusely involvement by tumor, and lack the bosselation typical of
expressed. SATB2 may be positive or negative (limited data Krukenberg tumors. A minority are ruptured [33]. The cut
exists on SATB2 staining in gastric adenocarcinomas). There surface may be solid or solid and cystic, and may show large
may be weak staining for ER and/or PR. Tumor may be posi- thin-walled cysts suggesting an ovarian primary neoplasm.
tive for CA 19-9 and/or HepPar1 [21, 22]. Tissue is soft, friable, yellow, red, or gray, and may contain
Differential Diagnosis: The differential diagnosis mucinous or clear fluid, recent or remote hemorrhage, and
includes primary ovarian mucinous and endometrioid adeno- necrosis (Fig. 10.8) [2, 32].
carcinomas, and metastatic adenocarcinomas that may have
a mucinous, endometrioid, or pseudoendometrioid morphol-
ogy (primarily tumors of intestinal or appendiceal origin,
less commonly tumors of gallbladder or biliary tract origin).
Management and Outcomes: Limited data is available
on this infrequently reported ovarian metastasis. Of the four
patients reported in the 2006 series noted above, the three
patients with follow-up information were deceased within 1
year of discovery of the ovarian metastasis [30].
Microscopic: Microscopically, tumors often have a pseu- pseudoendometrioid morphology, tumor glands are typically
doendometrioid, mucinous, or mixed pseudoendometrioid of moderate size but may range from small and tubular to
and mucinous character [34]. The pseudoendometrioid large and cystic (Figs. 10.11 and 10.12). Broad areas of
appearance is most typical of metastatic colorectal adenocar- necrosis may be present (Fig. 10.13). Frequently encoun-
cinoma and overall resembles typical primary colorectal tered architectural patterns include a cribriform pattern and a
adenocarcinoma; this pattern features columnar cells with “garland” pattern (neoplastic epithelium draped at the
only focal intracytoplasmic mucin (scattered goblet cells periphery of necrotic material). Intraluminal “dirty” necrosis
may be present) and marked nuclear atypia with nuclear and segmental necrosis of glandular epithelium are common
stratification and brisk mitotic activity (Figs. 10.9 and 10.10). features (Figs. 10.14 and 10.15). A papillary pattern, charac-
A mucinous-type epithelium, with more extensive well- terized by well-formed papillae or micropapillae, is occa-
differentiated mucin-rich cells, is a less commonly encoun- sionally seen [17, 18, 32, 34]. Rarely, a classic pattern of
tered cell type. In tumors characterized by the more common colloid carcinoma with clusters of neoplastic cells floating in
Fig. 10.9 Metastatic colonic adenocarcinoma. Like the majority of Fig. 10.11 Metastatic colonic adenocarcinoma. Tumor glands are
secondary ovarian tumors of colorectal origin, this tumor has pseudoen- most often of moderate size but may range from small and tubular to
dometrioid cytologic features, characterized by columnar cells with large and cystic, as in this case. Large cystic glands may mimic a pri-
only focal intracytoplasmic mucin. This area has a cribriform pattern mary ovarian mucinous neoplasm
Fig. 10.14 Metastatic colonic adenocarcinoma. Intraluminal “dirty” Fig. 10.15 Metastatic colonic adenocarcinoma. Segmental necrosis of
necrosis is a common feature glandular epithelium and a “garland pattern” in which the neoplastic
epithelium is draped at the periphery of necrotic material are often seen
Fig. 10.16 Metastatic colonic adenocarcinoma with unusual morpho- mullerian clear-cell adenocarcinoma. Right: Tumor may occasionally
logic patterns. Left: A papillary pattern, with large well-formed papil- have a colloid appearance, with groups of tumor cells floating in abun-
lae. A micropapillary pattern may also be seen. Center: Adenocarcinoma dant dissecting mucin
of small or large bowel origin may have clear cytoplasm, mimicking
10 Secondary Tumors of the Ovary 333
Fig. 10.17 Metastatic colonic adenocarcinoma. In contrast to endometrioid adenocarcinoma, tumor is typically immunoreactive for CK 20 (left),
CDX2 (center), and SATB2 (right)
positive for CK20 (80–100% of cases), but may infre- gallbladder, bile ducts, pancreas, uterine cervix, endome-
quently show no or only focal positivity for CK20 trium, or urachus. Adenocarcinoma of breast and lung origin
(decreased CK20 staining is associated with microsatel- may also occasionally mimic endometrioid adenocarcinoma
lite unstable carcinomas); typically positive for SATB2 [32]. Of these, mucinous and endometrioid adenocarcinomas
(96%), CDX2 (72–100%), and DPC4/SMAD4; negative of ovarian origin are the most difficult to exclude when con-
for PAX8, ER, and PR (Fig. 10.17). sidering metastasis from a colorectal primary. Features help-
–– Mucinous tumors of intestinal phenotype, ovarian origin ful in discriminating between the top differential diagnostic
(excluding the subset of ovarian mucinous carcinomas considerations are as follows:
arising in association with a teratoma): Typically positive Features favoring metastatic colon cancer:
for CK7, although infrequently may be negative; CK20
expression variable, sometimes negative but more often • A known colorectal primary.
patchy positive; variable expression of CDX2 and PAX8; • Gross findings of bilaterality, multinodularity of tumor,
typically positive for DPC4/SMAD4; typically negative and ovarian surface involvement.
for SATB2, ER, and PR. • Prominent “dirty” necrosis, segmental necrosis of tumor
–– Mucinous tumors of intestinal phenotype, ovarian origin, epithelium, higher nuclear grade and more mitotic activ-
arising in association with a teratoma: ity than is typical of endometrioid adenocarcinoma with a
Immunohistochemical staining profile is indistinguish- similar degree of glandular differentiation, and lympho-
able from that of metastatic tumors of intestinal origin vascular space invasion.
(CK7−/CK20+/SATB2+/CDX2+/PAX8−/ER−/PR−).
–– Seromucinous tumors of ovarian origin: Typically CK7+/ Features favoring primary ovarian endometrioid
PAX8+/ER+/PR+/CK20-/CDX2−. adenocarcinoma:
–– Endometrioid tumors of ovarian or endometrial origin:
CK7+/ER+/PR+/PAX8+/DPC4/SMAD4+/CK20−; may • Squamous differentiation, an adenofibromatous compo-
focally express CDX2. nent, and/or a background of endometriosis.
• Gross findings of unilaterality of tumor, an often cystic
Immunohistochemical staining may be used in the workup cut surface, and presence of chocolatey material related to
of mucinous and endometrioid/pseudoendometrioid tumors remote hemorrhage and endometriosis.
involving the ovary, although there is overlap in staining pat- • Primary endometrioid adenocarcinoma usually has less
terns between ovarian primary and metastatic neoplasms; thus, prominent necrosis than does metastatic colorectal adeno-
circumspect interpretation and correlation with clinical find- carcinoma; however, it may show the “dirty” and/or seg-
ings are required. Note that CDX2 is not specific for intestinal mental epithelial necrosis more characteristic of colorectal
carcinoma and may be demonstrated in tumors of endometrial, adenocarcinoma [2].
pancreatobiliary, gastric, lung, bladder, and ovarian origin. For
additional information regarding immunohistochemical stain- Features favoring primary ovarian mucinous
ing features of tumors in this differential, please see Table 10.1. adenocarcinoma:
Differential Diagnosis: The differential diagnosis
includes primary ovarian tumors of mucinous and endome- • Presence of a teratoma.
trioid types, and metastatic tumors with similar morphology • Large unilateral tumor (although metastatic colorectal car-
that may arise from the appendix, small bowel, stomach, cinoma may also be unilateral and large) [9, 18, 33, 34].
334 K. Carrick and W. Zheng
• Like primary ovarian endometrioid adenocarcinoma, pri- bowel adenocarcinoma accounted for only 1.7% of ovarian
mary ovarian mucinous adenocarcinoma has a lower inci- metastases in one large series [9]. A 2017 case report and
dence of multinodularity and ovarian surface involvement review of the literature detailing 72 cases of adenocarcinoma
than does metastatic colorectal adenocarcinoma. of small bowel origin metastatic to the ovary noted a mean
• Microscopically, small areas of invasive carcinoma with patient age of 46.7 years in the 12 cases well documented
extensive benign-appearing mucinous tumor favor an enough to report, with solitary metastasis to the ovary in the
ovarian primary mucinous tumor, although metastatic majority of these cases and peritoneal dissemination in a sub-
colorectal cancers may also have deceptively bland epi- stantial minority of cases [39]. Although the duodenum is the
thelium and large cysts that mimic a benign or borderline area of the small bowel most frequently involved by adenocar-
primary ovarian tumor (Figs. 10.11, 10.12). cinoma (with most of the tumors arising around the ampulla of
• Primary ovarian mucinous carcinoma does not typically Vater), the jejunum was the most frequently involved section
show the “dirty” necrosis or lymphovascular space inva- of small bowel generating ovarian metastasis in this 2017
sion more characteristic of metastatic colorectal cancer. series, giving rise to 69% of the 12 well-documented cases.
• Ovarian primary mucinous carcinomas have goblet cells Pathologic Findings:
more frequently than do metastatic colonic adenocarcino- Gross: Literature regarding gross findings in metastatic
mas, although metastatic colonic adenocarcinomas may tumors of small bowel origin is limited. Metastatic tumor
have goblet cells as well [2, 12]. may be bilateral or unilateral, and may be cystic or solid and
cystic. In the 2017 series noted above, 51% of the tumors
Metastatic colorectal cancer is the most common primary were bilateral and 49% unilateral. The largest documented
site generating metastatic tumor to the ovary in many series tumor was 26 cm [39].
in Western nations, and has a greater tendency to be unilat- Microscopic: Literature regarding microscopic findings in
eral and greater than 10 cm relative to other primary sites. metastatic tumors of small bowel origin is likewise limited.
This evidence is contradictory to the conventional concept of Microscopically, the tumors most often appear as conventional
metastatic cancer, in which metastatic tumor is more fre- intestinal type adenocarcinoma as would be seen in metastasis
quently bilateral and less than 10 cm in size. Therefore, we from the large bowel, and thus may have a pseudoendometri-
recommend that the threshold should be relatively low for oid or mucinous character. Tumor differentiation varies from
considering the possibility of metastatic colon cancer for well differentiated to moderately to poorly differentiated [39].
tumors with microscopic features suggesting that diagnosis. Adenocarcinoma of small bowel origin may occasionally have
A directed panel of immunohistochemical stains should be an unusual, strikingly clear cytoplasm, which may mimic
employed when the primary site is not clear, although over- ovarian clear-cell carcinoma or the secretory variant of endo-
lap in immunohistochemical staining patterns mandates that metrioid adenocarcinoma. Tumor with this unusual morphol-
caution in interpretation is warranted, and that clinical find- ogy is characterized by glands and cysts lined by cells with
ings must be taken into consideration. abundant clear cytoplasm, including some areas with subnu-
Management and Outcomes: Colorectal adenocarci- clear or supranuclear vacuoles. These cases do typically show
noma metastatic to the ovary qualifies as stage IV disease the “dirty” necrosis characteristic of intestinal type adenocar-
and has had a poor outcome relative to the primary ovarian cinomas. Tumors demonstrating both conventional intestinal
neoplasms in the differential diagnosis. The treatment strat- type adenocarcinoma morphology and the more unusual clear-
egy is not currently well defined. A few studies have sup- cell morphology noted above have been described [35].
ported a higher median overall survival in patients undergoing Biomarkers:
metastasectomy [36, 37].
–– Non-ampullary small intestinal adenocarcinomas tend to
have a CK7+/CK20+ immunophenotype (66%). CK7
10.6.4 Metastasis from Small Intestine expression may be focal or diffuse. Approximately 33%
have a CK7+/CK20− immunophenotype. Both of these
Definition: Adenocarcinoma of small bowel origin meta- staining patterns contrast with that of colorectal adeno-
static to the ovary. carcinoma (usually CK7−/CK20+) [22].
Clinical Features: Most intestinal tumors metastatic to the –– Approximately 60–70% of small intestinal adenocarcino-
ovary are from the large bowel, although occasional tumors mas stain positively for CDX2 [22].
originate from the small intestine. Risk factors for adenocarci- –– A subset (46%) of small intestinal adenocarcinomas stain
noma of the small intestine include Crohn’s disease, celiac positively for SATB2, although staining tends to be less
disease, Peutz-Jeghers syndrome, familial adenomatous pol- strong and diffuse than that seen in colorectal adenocarci-
yposis (FAP), and hereditary nonpolyposis colorectal cancer nomas [40].
syndrome [38]. In accordance with its low incidence, small- –– Tumors are negative for PAX8, ER, and PR.
10 Secondary Tumors of the Ovary 335
Fig. 10.18 Pseudomyxoma peritonei involving the omentum. The Fig. 10.19 Metastatic LAMN. Abundant mucin and neoplastic epithe-
external surface shows the characteristic gelatinous appearance lium involve the ovarian surface
Fig. 10.21 Metastatic LAMN. Neoplastic epithelium is tall columnar Fig. 10.23 Adenocarcinoma of appendiceal origin metastatic to the
and bland with mucin-rich cytoplasm ovary. Tumor in this field has nonspecific medium-sized gland mor-
phology and destructive stromal invasion. The tumor showed a diversity
of morphology in other sections, with both gland formation and signet
ring cells
tumor as an adenocarcinoma with neuroendocrine differen- As primary and metastatic tumors to the ovary may have
tiation [53, 54]. Of note, the involved appendix may be firm overlapping immunohistochemical staining patterns, immu-
and thickened but may not have a grossly identifiable dis- nostaining may be of limited value in certain cases; how-
crete mass; thus, the appendiceal primary tumor may be ever, a panel of immunostains should be employed when the
overlooked at operation [53]. Even when an appendiceal pri- source of tumor is uncertain. Positive staining for PAX8,
mary tumor is not identified, a goblet cell carcinoid-like pat- CK7, ER, or PR favors a primary ovarian tumor, although
tern within an ovarian tumor should suggest an appendiceal negative staining for these markers does not exclude an
primary tumor. Appendiceal tumors meeting the criteria for ovarian primary. Rare mucinous primary ovarian tumors
“goblet cell carcinoid,” as strictly defined, rarely metastasize with an intestinal phenotype associated with a teratoma may
to the ovary. Likewise, typical carcinoids of appendiceal ori- generate pseudomyxoma peritonei, a finding almost always
gin rarely metastasize to the ovary [32]. ascribable to an appendiceal primary neoplasm [51]; these
Biomarkers: tumors have an immunophenotype indistinguishable from
metastatic tumors of intestinal origin (CK7−/CK20+/
–– LAMNs and mucinous adenocarcinomas of appendiceal CDX2+/SATB2+/PAX8−).
origin typically express CK20, CDX2, DPC4/SMAD4, Management and Outcomes: Overall, low-grade appen-
and SATB2. Approximately one-third express CK7, with diceal mucinous neoplasms (LAMNs) involving the ovary
approximately 25–75% of cells staining. exhibit more indolent behavior and have a better prognosis
–– Appendiceal adenocarcinomas of conventional intestinal than appendiceal adenocarcinomas [56]. The prognosis and
type are immunophenotypically similar to colorectal treatment of LAMN are dependent on tumor stage. LAMN
adenocarcinomas. pursues a progressive clinical course when it is widely dis-
–– Appendiceal “goblet cell carcinoids” have a mixed immu- seminated in the peritoneum [46]. Currently, complete cyto-
nophenotype that shows both neuroendocrine and glandu- reductive surgery plus hyperthermic intraperitoneal
lar differentiation; most are positive for CK20, up to 70% chemotherapy (HIPEC) is the most commonly employed
are positive for CK7, and staining with neuroendocrine treatment for pseudomyxoma peritonei (PMP).
markers synaptophysin and chromogranin A is focal Regarding appendiceal adenocarcinomas, the reported
rather than diffuse [22]. 5-year survival rates range from 18.7 to 55%. Patients with
mucinous carcinoma have a better prognosis than those with
Differential Diagnosis: non-mucinous carcinoma. Patients with peritoneal carcino-
LAMN: The differential diagnosis of secondary ovarian matosis have a poor prognosis. As with LAMN, treatment
involvement with LAMN is with primary ovarian mucinous depends on disease stage [46].
cystadenoma, mucinous borderline tumor, and mucinous
carcinoma. Metastatic LAMN is favored by the presence of
an appendiceal neoplasm, bilateral ovarian involvement, 10.7 M
etastasis from the Pancreas, Biliary
pseudomyxoma peritonei and/or ovarii, scalloped neoplastic Tract, and Liver
glands, and subepithelial clefting. The presence of a tera-
toma or PAX-8 positivity favors an ovarian primary. 10.7.1 Pancreas
Importantly, a lack of grossly identifiable appendiceal rup-
ture does not rule out that the tumor originated in the appen- Definition: Metastatic adenocarcinoma originating from the
dix, as the rupture site may be small and difficult to identify, pancreas, most often adenocarcinoma of ductal type. Ovarian
or healed over by fibrosis [55]. metastases of pancreatic mucinous cystadenocarcinoma, aci-
Appendiceal adenocarcinoma: The differential diagnosis nar cell carcinoma, neuroendocrine tumors, and solid pseu-
of metastatic appendiceal adenocarcinoma is with primary dopapillary tumor have also been infrequently to rarely
ovarian mucinous and endometrioid adenocarcinomas, and reported.
with metastatic adenocarcinomas of mucinous and endome- Clinical Features: The frequency of pancreatic adeno-
trioid/pseudoendometrioid types originating elsewhere. The carcinoma metastatic to the ovary has varied in recent
distinction of a primary ovarian neoplasm from metastatic Western series, although the pancreas typically represents
tumor of appendiceal or other origin depends to a great the source of fewer than 10% of nongenital tract primary
degree on adequate sampling of the ovarian neoplasm and tumors generating a clinically apparent ovarian metastasis
clinical correlation. General features favoring metastatic car- [4]. Ovarian spread is often part of disseminated disease but
cinoma include a known extraovarian primary tumor, bilater- has been the dominant clinical finding or presenting clinical
ality, multinodular growth pattern, involvement of the finding in some cases [57]. Patients are usually in their mid
ovarian surface, high stage of disease, and histologic features to late years of life, with reported mean ages ranging from 56
unusual for a primary ovarian neoplasm. to 63 years.
10 Secondary Tumors of the Ovary 339
Definition: Tumor of hepatocellular origin metastatic to the Neuroendocrine tumors arise in various organs including the
ovary. gastrointestinal tract, pancreas, thyroid, lung, and skin. The
Clinical Features: Hepatocellular carcinoma metasta- most common type of neuroendocrine tumor giving rise to
sizes to the ovary less commonly than pancreatobiliary ade- ovarian metastasis is the well-differentiated type of neuroen-
nocarcinoma, and only a few such cases have been reported docrine tumor historically referred to as “carcinoid tumor.”
[67, 68]. All patients have been adults. Ovarian metastases They are most often of small bowel origin, although occasion-
have been discovered prior to discovery of the liver tumor, ally of appendiceal, colonic, gastric, pancreatic, or lung origin
synchronously, and after detection of the liver tumor. (of note, the term “carcinoid tumor” is no longer commonly
Pathologic Findings: used in reference to neuroendocrine tumors of the gastrointes-
Gross: Ovarian tumors in the two reported series range tinal tract, and has been replaced by the term “well-differenti-
from 4 to 11 cm. Most tumors have a solid cut surface. A ated neuroendocrine tumor” (NET)). The following discussion
green hue may provide a clue to the diagnosis [2, 67, 68]. focuses on well-differentiated neuroendocrine tumors/carci-
Microscopic: Features typical of metastatic disease, such noid tumor. High-grade neuroendocrine carcinoma, a more
as ovarian surface involvement, are seen in some cases. clinically aggressive and poorly differentiated tumor type
342 K. Carrick and W. Zheng
exemplified by, but not restricted to, small-cell carcinoma of white to yellow and may resemble an ovarian fibroma or the-
the lung, metastasizes to the ovary from various organs less coma grossly [2, 32, 71, 72].
often than well-differentiated neuroendocrine tumors, and is Microscopic: Metastatic carcinoid tumor in the ovary
discussed briefly in relevant sections below. shows microscopic features similar to primary carcinoid tumor
Definition: Tumor arising from cells that release hor- in the ovary and other organs. Tumor cells are characteristi-
mones into the bloodstream in response to a signal from the cally uniform with round to ovoid nuclei, finely granular chro-
nervous system. matin, inconspicuous nucleoli, and moderate-to-abundant
Clinical Features: Evaluation of literature related to eosinophilic cytoplasm (Fig. 10.30). Mitoses are infrequent,
well-differentiated neuroendocrine tumor/carcinoid tumor but occasional pleomorphic cells or cells with nucleolar prom-
metastatic to the ovary is complicated by the varied criteria inence may be seen. These classic cytologic features are the
that have been employed over the years for diagnosis of such most frequently encountered, but cells may occasionally have
tumors. If strict diagnostic criteria are applied, tumors meet- clear or oncocytic cytoplasm, or a spindled morphology. Cells
ing the diagnostic criteria for well-differentiated endocrine are typically disposed in one or more architectural patterns
tumor/carcinoid tumor from any source rarely metastasize to common to neuroendocrine neoplasms in general including
the ovary [32]. Two relatively large series of carcinoid tumors insular, trabecular, ribbonlike, acinar, or solid tubular patterns
metastatic to the ovary have been reported [71, 72]. Rare (Fig. 10.31). When the acinar pattern is encountered, acini
cases of pancreatic neuroendocrine neoplasms metastatic to often contain eosinophilic secretions, which may calcify in a
the ovary have also been reported outside of these series psammomatous or nonpsammomatous fashion [2, 32, 71, 72].
[73–75]. In addition to the above-noted common architectural patterns,
Carcinoid tumors account for approximately 2% of larger cysts and follicle-like spaces are sometimes seen
metastases that form detectable ovarian masses [2]. Most (Fig. 10.32) [12]. Metastatic carcinoid tumor often has a
metastatic carcinoid tumors are of ileal origin, although less prominent paucicellular fibrous stroma, which may be exten-
frequently the primary site is the jejunum, appendix, colon, sively hyalinized (Fig. 10.33). Vascular invasion may be seen
stomach, pancreas, or lung [71–79]. In the largest series (35 occasionally [12].
cases) of carcinoid tumors metastatic to the ovary to date, Biomarkers:
patients ranged from 21 to 82 years of age, with a median age
of 57 years. Patients may present with manifestations of the –– Well-differentiated neuroendocrine tumors/carcinoid
carcinoid syndrome (flushing and diarrhea), may have symp- tumors from any primary site typically express selected
toms related to an intestinal or ovarian mass, or may be cytokeratins and a variety of neuroendocrine markers,
asymptomatic. A majority of patients with ovarian metasta- although the immunostaining profile varies somewhat
sis have extraovarian metastasis as well, which contrasts depending on the site of origin.
with the rarity of extraovarian spread of primary ovarian car- –– Well-differentiated neuroendocrine tumors/carcinoid
cinoid tumor. tumors express LMWCKs (CAM5.2) in nearly 100% of
Outside of the two large series of carcinoid tumors meta- cases, and HMWCKs (AE1/AE3) a bit less frequently.
static to the ovary noted above, rare pancreatic neuroendo-
crine neoplasms metastatic to the ovary have been reported
including a VIPoma [73], a glucagonoma [74], and an
ACTH-secreting tumor [75]. Ovarian metastasis was discov-
ered years after diagnosis of the pancreatic primary tumor in
the patients with VIPoma and glucagonoma. In the patient
with metastatic ACTH-secreting pancreatic endocrine tumor,
however, the initial presentation was due to bilateral ovarian
masses, hirsutism, and Cushing’s syndrome; subsequent
workup disclosed the pancreatic neuroendocrine tumor.
Pathologic Findings:
Gross: Carcinoid tumor metastatic to the ovary is usually
bilateral, in contrast with primary ovarian carcinoid tumor,
which is almost always unilateral. These tumors are typically
of relatively modest size, ranging from microscopic foci up
to 9.5 cm, and have a smooth or bosselated external surface.
The cut surface is typically solid, but may be solid and cystic
or predominantly cystic, with cysts containing clear, watery
Fig. 10.30 Metastatic well-differentiated neuroendocrine tumor of
fluid. Tumors typically show the discrete to confluent nodu- ileal origin. Tumor cells are uniform with round to ovoid nuclei, incon-
lar architecture often seen in metastatic lesions. The tumor is spicuous nucleoli, and subtly granular eosinophilic cytoplasm
10 Secondary Tumors of the Ovary 343
Granulosa cell tumor: The Call-Exner bodies of granu- improved the course of disease in these patients. Surgical
losa cell tumor may resemble the acinar structures of carci- resection is recommended when possible. Nonsurgical treat-
noid tumor, although the cytologic features of these tumors ment options for patients with well-differentiated gastroen-
are distinct. Granulosa cell tumor has round, ovoid, or angu- tero-pancreatic neuroendocrine tumors include somatostatin
lated, grooved nuclei and scant indistinct cytoplasm. The analogues, multi-kinase inhibitors, targeted therapy, chemo-
cells are rather haphazardly oriented relative to each other therapy, and radiolabeled somatostatin analogues [82].
and to the lumen of the Call-Exner body. In contrast, carci-
noid tumors typically have round to ovoid nuclei with finely
granular chromatin, no nuclear grooving, more distinct and 10.9 Metastatic Cervical Carcinoma
abundant cytoplasm, and a more orderly polarity around the
acinar spaces. Granulosa cell tumors of adult type typically Definition: Tumor metastatic to the ovary from the uterine
express inhibin, calretinin, CD99, and WT-1, but are nega- cervix. Ovarian metastases from cervical carcinomas of vari-
tive for neuroendocrine markers [23]. ous histologic types, including adenocarcinoma, squamous
Sertoli/Sertoli-Leydig cell tumor: The hollow and solid carcinoma, high-grade neuroendocrine carcinoma/small-cell
tubules of Sertoli/Sertoli-Leydig cell tumor may mimic the carcinoma, mixed high-grade neuroendocrine carcinoma/
acinar and ribbonlike structures of carcinoid tumor, although small-cell carcinoma and adenocarcinoma, adenosquamous
carcinoid tumor usually has longer, thicker ribbons with a carcinoma, glassy cell carcinoma, transitional cell carcinoma,
more orderly architecture. Of note, Sertoli/Sertoli-Leydig and undifferentiated carcinoma, have been reported [83, 84].
cell tumor may have a component of carcinoid tumor, typi- Of these, cervical adenocarcinoma is the most frequent histo-
cally minor in extent and associated with heterologous ele- logic type to develop ovarian metastasis [2, 32, 85].
ments. Sertoli/Sertoli-Leydig cell tumors stain for inhibin, Clinical Features: Ovarian spread of cervical carcinomas
MART-1, and calretinin, which are typically negative in car- of all types has been considered infrequent. A 2006 study of
cinoid tumors [2, 23, 32]. 3471 patients with stage Ib to IIb cervical cancer who under-
Brenner tumor: Brenner tumor, with its fibromatous went radical hysterectomy and bilateral salpingo-
stroma and nests of bland epithelium, may mimic primary or oophorectomy demonstrated ovarian metastasis in 5.31% of
metastatic insular carcinoid tumor. In Brenner tumor, epithe- those with endocervical adenocarcinoma and only 0.79% of
lial nests are of transitional type with ovoid, grooved nuclei. those with cervical squamous cell carcinoma [85]. Patients
Immunohistochemical stains for neuroendocrine markers are have ranged in age from approximately 29 to 73 years, with
negative. a mean age of 49.9 years in one large study [85, 86].
Endometrioid adenocarcinoma: These tumors may also Ovarian metastasis may present synchronously or meta-
have small tubules that resemble the acinar structures of car- chronously in reference to the cervical primary tumor. In
cinoid tumor. Consideration of nuclear and architectural fea- some cases, ovarian metastasis becomes apparent several
tures and immunohistochemical staining should resolve the years after diagnosis of the primary tumor, while in other
differential diagnosis. Squamous elements or morular meta- cases the ovarian metastasis is the presenting sign of a clini-
plasia excludes carcinoid tumor. Endometrioid adenocarci- cally unsuspected cervical tumor [13, 83, 87]. Cervical
noma is typically positive for PAX-8, ER, and PR, which are tumors giving rise to ovarian metastasis are clearly invasive
typically negative in carcinoid tumor. Endometrioid adeno- in most cases, and some are of advanced stage, with extra-
carcinoma and other carcinomas may contain neuroendo- uterine involvement or a bulky primary tumor [83, 86]. In a
crine cells that stain focally for neuroendocrine markers, substantial minority of cases, however, the cervical tumor is
though staining is less diffuse than in carcinoid tumor. Of not clinically evident prior to discovery of the ovarian metas-
note, both endometrioid adenocarcinoma and ovarian carci- tasis, and may demonstrate only small foci of superficial
noid tumor may express CDX2 [23, 81]. invasion or no unequivocal stromal invasion. Extension of
Management and Outcomes: Among the 17 patients cervical tumor into the lower uterine segment or uterine cor-
with carcinoid tumor metastatic to the ovary in the 2007 pus has been identified as a possible risk factor for ovarian
series by Strosberg and colleagues [72], two deaths occurred metastasis, postulated mechanisms including transtubal
in the follow-up interval (range of follow-up 8–146 months), spread of tumor [13, 86].
yielding a projected 5-year survival rate of 94%. All patients Pathologic Findings:
underwent surgical resection of the ovarian masses, and most Gross:
underwent surgical resection of the primary tumor, with or Metastatic endocervical adenocarcinoma: For endocer-
without more extensive debulking. Fifteen patients received vical adenocarcinoma metastatic to the ovary, metastasis
long-term depot-octreotide therapy, and two received strep- may be bilateral or unilateral. Some tumors are unilateral
tozocin-based chemotherapy. It is postulated that cytoreduc- and large (reported tumors have ranged up to 30 cm), mim-
tive surgery and treatment with octreotide substantially icking a primary ovarian mucinous neoplasm. Tumors have a
10 Secondary Tumors of the Ovary 345
Fig. 10.34 Large ovarian metastasis from an endocervical adenocarci- Fig. 10.35 Metastatic endocervical adenocarcinoma of usual type.
noma, with a smooth bosselated external surface The tumor has the hybrid endometrioid-mucinous features typical of
endocervical adenocarcinomas of usual type. The cells are columnar
with varied glandular, papillary, and cribriform architecture
smooth or nodular external surface (Fig. 10.34). The cut sur-
face may be predominantly solid or multicystic, sometimes
with solid or papillary areas [13, 86].
Metastatic cervical squamous carcinoma: Squamous
cell carcinoma metastatic to the ovary may be unilateral or
bilateral. Reported tumors have ranged up to 17 cm. Tumors
show a smooth or nodular external surface, and a solid, solid
and cystic, or predominantly cystic cut surface [83, 86, 87].
Microscopic:
Metastatic endocervical adenocarcinoma: Among the
various subtypes of endocervical adenocarcinoma, the usual
(HPV related) type is the most common to exhibit ovarian
metastasis. In the ovary, these tumors show microscopic fea-
tures similar to their primary endocervical counterparts. The
majority of these tumors have a pseudoendometrioid appear-
ance with hybrid endometrioid and mucinous features, char-
acterized by an endometrioid appearance at low magnification
Fig. 10.36 Metastatic endocervical adenocarcinoma of usual type.
but with varying amounts of apical mucin evident at higher Tumor cells are columnar and stratified with apoptosis and apical mito-
magnification. Nuclei are typically hyperchromatic and sis. Apical mucin varies in these tumors and is scant in this case, simu-
elongated, with numerous apical mitoses and frequent apop- lating the appearance of an endometrioid adenocarcinoma primary or
metastatic to the ovary
tosis (Figs. 10.35 and 10.36). In the ovary, the growth
pattern(s) may be borderline-like, confluent glandular, cribri-
form, papillary, or villoglandular, and may not be readily have distinctive clear and/or pale eosinophilic, abundant
recognized as invasive, thus simulating a primary ovarian cytoplasm with distinct cell borders. They may show single-
mucinous borderline tumor or well-differentiated primary cell infiltration and foci of signet ring and intestinal type dif-
ovarian mucinous carcinoma with a confluent invasive pat- ferentiation with goblet cells or Paneth-like neuroendocrine
tern [13]. Of note, a “carpeting” pattern of tumor spread, in cells [88].
which endocervical adenocarcinoma spreads superficially in Metastatic cervical squamous cell carcinoma: With
the endometrium without obvious myometrial invasion or involvement by cervical squamous cell carcinoma, the ova-
lymphovascular invasion, may be seen in endocervical ade- ries show nodules of metastatic tumor replacing normal
nocarcinomas with adnexal involvement [86]. ovarian parenchyma (Figs 10.37 and 10.38) [86]. Microscopic
Endocervical adenocarcinoma of the rarer gastric type features are typical, except that some tumors show striking
(HPV unassociated) has also been reported to metastasize to cyst formation within the squamous nests [83, 87]. Similar to
the ovary, although less commonly. These tumors typically the “carpeting” pattern of tumor spread noted in some cases
346 K. Carrick and W. Zheng
Fig 10.39 Metastatic
endocervical adenocarcinoma
of usual type. This HPV-
related neoplasm shows
strong, diffuse positivity for
p16 (left) and HR-HPV
infection by in situ
hybridization (right)
10 Secondary Tumors of the Ovary 347
tive stromal invasion, microscopic features uncommon for a 37.5% for stage IIa, and 18.0% for stage IIb, despite admin-
primary ovarian neoplasm, prominent lymphovascular inva- istration of adjuvant therapy (radiotherapy and/or chemo-
sion, and a known cervical primary tumor all suggest metas- therapy) to 92.3% of patients [85].
tasis. Strong, diffuse staining with p16 is likewise strongly
suggestive of a metastatic HPV-related tumor. Identification
of HPV infection by in situ hybridization or molecular meth- 10.10 Metastatic Endometrial Carcinoma
ods confirms the diagnosis. Other points helpful in the dif-
ferential diagnosis are as follows: Definition: Metastasis to the ovary originating from an
Metastatic endocervical adenocarcinoma: The differen- endometrial adenocarcinoma. The most common histologic
tial diagnosis of metastatic endocervical adenocarcinoma type of endometrial adenocarcinoma to secondarily involve
includes ovarian mucinous borderline tumor, primary ovarian the ovary is endometrioid adenocarcinoma. Metastases of
mucinous adenocarcinoma with a confluent invasive pattern, endometrial serous carcinoma, uterine carcinosarcoma, and
and other primary and metastatic carcinomas of mucinous other types of endometrial adenocarcinoma are less
and endometrioid/pseudoendometrioid types (most notably common.
primary and metastatic endometrioid adenocarcinoma, and Clinical Features: Synchronous endometrial and ovarian
adenocarcinomas of intestinal, appendiceal, gastric, and pan- carcinoma is found in approximately 10–15% of patients
creatic origin). The most challenging differential is with pri- with ovarian cancer and 5% of patients with endometrial
mary ovarian mucinous neoplasia, which metastatic cancer [89–91]. When encountered, this finding raises the
endocervical adenocarcinoma may mimic grossly and micro- important question of whether the two lesions represent
scopically; thus, an appropriate index of suspicion for possi- independent primary tumors or whether one lesion is a
ble metastasis is warranted. Clinical correlation and metastasis from the other. When histology is dissimilar the
immunohistochemical staining should be employed when the distinction is relatively easy, but when similar the distinction
diagnosis is uncertain. Clues to an ovarian primary endome- may be challenging. Making this distinction is essential for
trioid adenocarcinoma include squamous morules, endome- prognostication and treatment planning. Patients with syn-
triosis, or an endometrioid adenofibroma. Primary and chronous primary endometrial and ovarian carcinomas tend
metastatic endometrioid adenocarcinomas typically express to be younger, present with early-stage disease, and have a
vimentin, ER, and PR; show only patchy positivity for p16; more favorable overall prognosis than do patients who pres-
and are negative for CEA; in contrast, endocervical adenocar- ent with only an endometrial or ovarian carcinoma at the
cinomas are typically CEA positive, p16 block positive (when same clinical stage [89, 91–94]. Likewise, patients with syn-
HPV related), vimentin negative, and negative to focally posi- chronous, independent primary cancers have a better prog-
tive for ER and PR. Please see Table 10.1 for typical immuno- nosis than do patients with endometrial cancer with ovarian
histochemical staining features of other tumors with mucinous metastasis [95].
and endometrioid/endometrioid-like morphology. Pathologic Findings:
Metastatic cervical squamous cell carcinoma: A squa- Gross: Endometrial tumors secondarily involving the
mous tumor in the ovary may be metastatic or rather a pri- ovary are usually smaller than 5 cm and bilateral. These
mary ovarian tumor associated with a teratoma, endometriotic tumors may be solid or solid and cystic on the cut surface.
cyst, or massive squamous overgrowth in an endometrioid General gross features favoring metastatic tumor include
adenocarcinoma [32]. Adequate gross evaluation and sam- bilaterality, a multinodular growth pattern, and tumor on the
pling are essential to evaluate for associated features. Primary ovarian surface (with the caveat that tumor may develop in
ovarian squamous carcinoma is rare and, before an ovarian the ovary in the context of surface endometriosis; thus, a
primary squamous carcinoma is diagnosed, the possibility of careful search for endometriosis can be helpful).
metastasis from an occult squamous carcinoma in the cervix Microscopic: Endometrial tumors secondarily involving
or elsewhere should be entertained [32]. Ovarian transitional the ovary have microscopic features similar to their endome-
cell carcinoma should also be excluded prior to making the trial counterparts (Figs. 10.40, 10.41, and 10.42). Microscopic
diagnosis of metastatic squamous cell carcinoma, although features favoring metastasis include a multinodular growth
this tumor is rare. pattern, involvement of the ovarian surface (again with the
Management and Outcomes: A 2006 study including 52 caveat that tumor may develop in the ovary in the context of
patients with stage Ib to IIb cervical cancer and ovarian surface endometriosis), and hilar or extraovarian lymphovas-
metastasis who underwent radical hysterectomy, pelvic cular space invasion. An unusual finding occurring occasion-
lymphadenectomy, and bilateral salpingo-oophorectomy ally in the setting of an endometrial endometrioid
demonstrated a poor outcome for these patients, unrelated to adenocarcinoma with squamous differentiation is that of
FIGO stage or histologic type. Five-year survival rates for keratin deposits or degenerated mature squamous cells with
patients with ovarian metastasis were 46.6% for stage Ib, an associated foreign-body giant cell reaction on the surface
348 K. Carrick and W. Zheng
Fig. 10.40 Metastatic endometrioid adenocarcinoma of uterine corpus Fig. 10.42 Metastatic endometrioid adenocarcinoma of uterine corpus
origin. These tumors show morphology similar to their uterine counter- origin. Squamous differentiation may be seen in endometrioid adeno-
parts. Papillary, glandular, and cystic patterns may be seen carcinomas primary to the ovary or metastatic from the uterine corpus.
Squamous differentiation is not seen in adenocarcinomas of intestinal,
gastric, or appendiceal origin
ovarian endometrioid adenocarcinoma, and other primary Table 10.3 Endometrioid tumors of ovary and endometrium, features
favoring ovarian metastasis from endometrial primary
and metastatic tumors with an endometrioid-like architec-
tural pattern. Differential diagnosis includes some high- 1. Histologic similarity of the tumors
grade serous and clear-cell carcinomas, Sertoli/Sertoli-Leydig 2. Large endometrial tumor, small ovarian tumor(s)
3. Atypical endometrial hyperplasia/endometrial intraepithelial
cell tumor, endometrioid yolk sac tumor, and metastatic
neoplasia (EIN) present
tumors of endocervical, gastrointestinal, and appendiceal 4. Deep myometrial invasion with direct extension into adnexa and/
origin [12]. Helpful features that can be used to resolve the or vascular space invasion in myometrium
differential diagnosis are as follows: 5. Spread elsewhere in typical pattern of endometrial carcinoma
Metastatic endometrioid adenocarcinoma versus pri- 6. Ovarian tumors bilateral and/or multinodular
mary ovarian endometrioid adenocarcinoma: Literature 7. Hilar location, vascular space invasion, surface implants,a or
combination in ovary
over the last 20 years or so has focused on the view that syn-
8. Ovarian endometriosis absentb
chronous tumors of the endometrium and ovary more often
9. Similar molecular genetic or karyotypic abnormalities
represent independent primary tumors. The idea of synchro- in both tumors
nous endometrial and ovarian carcinomas representing two a
Infrequently, primary ovarian endometrioid adenocarcinoma may arise
separate primary tumors in the majority of cases has been from endometriosis involving the ovarian surface
supported by the fact that most synchronous tumors have a b
In addition to ovarian endometriosis, the presence of an endometrioid
favorable overall prognosis, which would be unexpected if adenofibroma in the ovary suggests an ovarian primary tumor
either tumor were metastatic. Most tumors with synchronous
involvement of the endometrium and ovary are of endometri- also be used for tumor types other than endometrioid. In
oid type. When tumors are of the serous type or any other many cases there is strong evidence regarding whether an
histologic type, there is a greater statistical likelihood that ovarian tumor represents a metastasis or an independent pri-
the ovarian tumor is metastatic [32]. mary tumor, but in some cases this distinction is difficult or
Over the years, various studies have delineated histologic impossible.
and genetic criteria to aid in the determination of whether Recently, studies from different research groups using
synchronous endometrial and ovarian tumors with similar different molecular techniques have suggested that tumors
histology represent independent primary tumors or meta- that would be categorized as independent synchronous pri-
static disease [97]. Historic criteria favoring synchronous mary tumors by historic morphologic criteria are often clon-
independent endometrioid primary tumors are presented in ally related and may thus represent dissemination/metastasis
Table 10.2, while features favoring metastasis to the ovary from one site to another [98–100]. They furthermore suggest
are listed in Table 10.3 [97]. Beyond the features listed in that the relatively good prognosis of most of these patients
Table 10.3, another feature suggesting ovarian metastasis is might be explained by isolated ovarian spread of an indolent
the presence of tumor within the fallopian tube lumen. This endometrial tumor, possibly through the fallopian tube
feature, and some of those listed in Tables 10.2 and 10.3, can lumen, without invasion of vascular structures or easy access
to the peritoneal cavity [100]. This evolving concept has not
Table 10.2 Endometrioid tumors of ovary and endometrium, features gained widespread acceptance at this time, and further work
favoring independent primary tumors is required. An alternate explanation that has been proffered
1. Histologic dissimilarity of the tumors for the similar molecular alterations identified in some syn-
2. No or only superficial myometrial invasion of the endometrial chronous endometrial and ovarian endometrioid adenocarci-
tumor nomas is that a common carcinogenic agent acting on
3. No vascular space invasion of the endometrial tumor endometrium and ectopic endometrial tissue within the ovary
4. Atypical endometrial hyperplasia/endometrial intraepithelial could induce similar genetic events in the two locations, sim-
neoplasia (EIN) present
ulating metastasis from one site to another. It is furthermore
5. Absence of other evidence of spread of endometrial tumor
6. Ovarian tumor unilateral (80–90%) of cases
noted that a metastatic lesion may exhibit a different molecu-
7. Ovarian tumor located in parenchyma lar profile from the primary tumor due to tumor heterogene-
8. No vascular space invasion, surface implants,a or predominant ity and progression. Currently, ovarian endometrioid
hilar location in ovary adenocarcinomas are evaluated for their likely primary or
9. Absence of other evidence of spread of ovarian tumor secondary nature according to the morphologic criteria pre-
10. Ovarian endometriosis presentb sented in Tables 10.2 and 10.3.
11. Dissimilar molecular genetic or karyotypic abnormalities Endometrioid adenocarcinoma versus serous carci-
in the tumor
noma: Lower nuclear grade, a more orderly glandular and
a
Infrequently, primary ovarian endometrioid adenocarcinoma may arise
from endometriosis involving the ovarian surface
villous pattern with smooth luminal borders, squamous dif-
b
In addition to ovarian endometriosis, the presence of an endometrioid ferentiation, and a lack of associated serous tubal intraepithe
adenofibroma in the ovary suggests an ovarian primary tumor lial carcinoma (STIC) favor endometrioid adenocarcinoma.
350 K. Carrick and W. Zheng
ovarian mass. In other patients, the ovarian metastasis is tic challenge [12]. The few reported cases of uterine leio-
diagnosed synchronously with the uterine primary tumor, or myosarcoma metastatic to the ovary with microscopic
years after diagnosis of the uterine primary [32, 106]. In one description available have had the spindled myoid cells with
reported case, ovarian and extraovarian metastases were variable nuclear atypia and increased mitotic activity charac-
reported 17 years after diagnosis of a uterine low-grade teristic of LMS. One tumor showed prominent myxoid
endometrial stromal sarcoma [108]. change [106].
LMS metastasizes to the ovary less frequently than does Biomarkers:
ESS but is probably more common than the rare reports in
the literature suggest, particularly in patients with wide- –– ESS typically expresses CD10, ER, PR, and WT-1, with
spread disease [2, 109]. In the aforementioned 1990 study by varied extent and intensity. Many also express SMA,
Young and Scully, three LMS metastatic to the ovary cytokeratin, or AR in a patchy fashion. Expression of des-
occurred in patients 35, 44, and 49 years of age. In the first min and CD34 is rare. Of note, this immunohistochemical
patient, a large ovarian metastasis became symptomatic staining profile is characteristic of the portion of tumor
14 months after hysterectomy. In the second patient, the resembling proliferative-type endometrium; any meta-
ovarian metastasis occurred in the setting of widespread dis- plastic elements, such as endometrioid glands, sex cord
ease, while in the third, ovarian involvement was micro- elements, or smooth muscle, tend to acquire the immuno-
scopic [106]. phenotype of the corresponding metaplastic element [23].
Pathologic Findings: –– Leiomyosarcomas typically express the smooth muscle
Gross: markers SMA, desmin, and h-caldesmon, although des-
Low-grade ESS: Low-grade ESS metastatic to the ovary min may be negative. Leiomyosarcomas may also express
is more often bilateral, and may vary in size from micro- CD10, ER, and PR. Cytokeratin is frequently expressed in
scopic to greater than 15 cm. Tumors are predominantly a patchy fashion.
solid or solid and cystic, rarely multicystic, and may have
discernible nodularity. The cut surface is white to gray white, Differential Diagnosis: The differential diagnosis of ESS
with or without foci of yellow coloration, hemorrhage, and metastatic to the ovary includes primary ovarian ESS and
necrosis [106]. A wormlike pattern of vascular plugging may primary ovarian sex cord-stromal tumors with similar histol-
rarely be seen on the cut surface [32]. ogy such as fibroma, thecoma, granulosa cell tumor, and
Uterine LMS: Uterine LMS metastatic to the ovary may Sertoli/Sertoli-Leydig cell tumor. Features helpful in the dif-
be bilateral or unilateral and ranges in size from microscopic ferential diagnosis are as follows:
up to 15 cm. Available gross descriptions detail cut surfaces Metastatic ESS versus primary ovarian ESS: Primary
with multiple firm tumor nodules or the lobulated, white-tan, ovarian ESS is rare but does occur [107, 110]. Given the rar-
fleshy appearance typical of LMS [106]. ity of primary ovarian ESS, when ESS is encountered in the
Microscopic: ovary, metastasis must be considered. A clinical history of a
Low-grade ESS: Low-grade ESS metastatic to the ovary primary uterine ESS is of course strongly suggestive of met-
can show all the varied morphologic patterns that may be astatic tumor. Bilaterality of tumor favors metastasis,
seen in primary ESS, such as fibrous or myxoid change, although bilateral ovarian primary ESS has also been docu-
smooth muscle differentiation, endometrioid gland forma- mented [110]. Associated ovarian endometriosis favors an
tion, and epithelioid cells [32, 108]. The diagnostic difficulty ovarian primary neoplasm [32, 106, 107].
presented in some cases is due to a lack of the classic archi- Metastatic ESS versus sex cord-stromal tumors: As the
tectural patterns of ESS. Metastatic tumor typically shows distinctive vascular pattern and “tonguelike” or “wormlike”
the small round to ovoid nuclei characteristic of ESS, infiltration pattern typical of primary uterine ESS may be
although the “tonguelike” pattern of uterine invasion, and the seen only focally in an ovarian metastasis, and as metastatic
characteristic vascular pattern of prominent interspersed ESS may show morphologic features overlapping with some
small arterioles, may be focal or absent. A diffuse pattern is primary ovarian sex cord-stromal tumors, a primary ovarian
the most common architectural pattern [32]. Tumor may also sex cord-stromal tumor may present a plausible differential
have a nested, nodular, clustered, corded, tubular, or single- diagnosis. A diffuse growth pattern in the metastatic ESS
cell pattern. Diagnostic challenge may arise secondary to may suggest a granulosa cell tumor. The presence of paucic-
finding fibrous bands or large pauci- to acellular fibromatous ellular or acellular fibromatous areas may resemble fibroma.
areas suggestive of fibroma, or hyaline plaques suggesting a Areas with hyaline plaques may raise a differential diagnosis
diagnosis of thecoma [32, 106]. with thecoma. Areas of a cord-like or sex cord-like growth
Uterine LMS: Uterine leiomyosarcoma metastatic to the may suggest a granulosa cell tumor or a Sertoli/Sertoli-
ovary is most often seen in the setting of more widespread Leydig cell tumor. Careful microscopic examination should
disease and typically does not present a significant diagnos- demonstrate absence of the nuclear features characteristic of
352 K. Carrick and W. Zheng
granulosa cell tumor. The pattern of small arterioles charac- limited benefit. There is some evidence that secondary cyto-
teristic of ESS is often seen at least focally in the ovary and reductive surgery may improve outcome. Targeted therapies
is a significant clue to the diagnosis. Additionally, the are under investigation [115].
“tonguelike,” often intravascular pattern of tumor infiltration
typical of ESS is not usually seen in the ovary, but is often
seen in extraovarian foci of tumor, facilitating diagnosis. 10.12 Metastatic Tumors from the Tubal
Bilaterality and extraovarian spread of tumor are typical of Fimbriated End
metastatic ESS but rare with the sex cord-stromal tumors in
the differential [32, 106]. More thorough sampling to uncover Definition: Tumors originating in the fallopian tube fimbria
areas of more revealing histology, and a directed battery of with secondary involvement of the ovary. Historically, carci-
immunohistochemical stains, may be performed in cases in nomas with predominant ovarian involvement were thought
which there is diagnostic uncertainty. to be derived from the ovarian surface epithelium. Mounting
LMS and other non-carcinomatous uterine tumors: Outside evidence from a number of related studies [116–119], how-
of ESS, other non-carcinomatous uterine tumors metastatic to ever, suggests that a majority of high-grade serous ovarian
the ovary are relatively rare. Uterine LMS metastatic to the carcinomas actually arise in the fimbriae rather than the ovar-
ovary tends to occur in the setting of more widespread disease, ian surface epithelium. Studies have demonstrated that the
and does not typically present much diagnostic difficulty when long-sought-for precursor of ovarian high-grade serous car-
encountered. As with primary ovarian ESS, LMS occurring cinoma appears to develop from an occult intraepithelial car-
primarily in the ovary or periovarian tissue is rare; thus, LMS cinoma in the fimbria designated “serous tubal intraepithelial
encountered in the ovary is more likely representative of carcinoma” (STIC), which may involve the ipsilateral and
metastasis. Ovarian metastasis of uterine choriocarcinoma is contralateral ovarian surfaces and/or peritoneal surfaces sec-
rare and must be distinguished from primary ovarian chorio- ondarily. Another possible mechanism of spread is implanta-
carcinoma of gestational or germ cell origin [12, 104, 111]. If tion of normal fimbrial epithelium on the denuded ovarian
choriocarcinoma is identified in the ovary and is not clearly surface at the site of rupture when ovulation occurs, causing
metastatic from a uterine or tubal gestational choriocarcinoma, the development of cortical inclusion cysts, which may
the ovarian tumor must be adequately sampled to rule out tera- develop neoplastic changes over time [120]. Invasive tumor
tomatous or other germ cell elements. If no teratoma or other involving the distal fallopian tube may also extend directly
germ cell elements are found, it may be impossible to differen- into the ovarian parenchyma, sometimes aided by the pres-
tiate between a primary ovarian choriocarcinoma of either ence of tubo-ovarian adhesions [2].
gestational or germ cell origin and metastasis from a uterine Clinical Features: The clinical presentation of ovarian
choriocarcinoma that has regressed [2]. Ovarian metastasis of carcinoma developing secondarily from the tubal fimbriae is
uterine adenosarcoma has also been documented [103]. Of similar to that of tumors which would historically have been
note, ovarian/adnexal adenosarcoma may represent a primary classified as primary ovarian neoplasia. Patients with symp-
tumor, particularly if it is associated with endometriosis. tomatic tumors are typically postmenopausal, with a mean
Management and Outcomes: Adnexal spread of low- age ranging from 56 to 63 years [121–123]. In symptomatic
grade ESS occurs in approximately 15% of cases [12]. The patients, the most common symptoms include an abdominal/
tumor may behave indolently and patients may have pro- pelvic mass, and abdominal pain and distention due to bulky
longed survival, with recurrence developing more than a tumor or ascites. Abnormal uterine bleeding/discharge is
decade after initial diagnosis [112]. Stage is the most impor- also a relatively common symptom [124]. Some patients will
tant prognostic factor for this tumor. At this time, the optimal have tumor identified in an endometrial biopsy/curettage,
treatment of recurrent or metastatic low-grade endometrial endocervical curettage, or Pap smear. The majority of
stromal sarcoma remains to be determined. Surgical resec- patients are white, non-Hispanic [121]. Patients with BRCA1
tion is the primary treatment and is recommended when pos- or BRCA2 germline mutations are at increased risk for the
sible. The tumors are relatively chemo- and radioresistant. development of serous tubal intraepithelial carcinoma
Hormonal treatment with progestin therapy or aromatase (STIC), with or without tubal invasion and ovarian spread.
inhibitors may be effective in achieving disease control in Among symptomatic patients with fallopian tube carcinoma,
some cases due to the frequent presence of ER and PR, women with BRCA-associated tumors present at a slightly
although further study is required [112, 113]. Targeted thera- younger age than those with sporadic tumors, although avail-
pies are also under investigation [114]. able literature suggests that both patient groups have similar
Uterine LMS is an aggressive neoplasm with a poor out- clinicopathologic features [125, 126].
come regardless of clinical stage. Surgical resection is the Pathologic Findings: The majority of malignant neo-
cornerstone of treatment for early-stage tumor. Advanced- plasms developing in the fallopian tube fimbriae are high-
stage, recurrent, or metastatic LMS is treated with cytotoxic grade serous carcinomas (Figs. 10.43, 10.44, 10.45, 10.46,
chemotherapy, although currently available regimens are of and 10.47), and a smaller percentage are endometrioid
10 Secondary Tumors of the Ovary 353
Fig. 10.44 Serous tubal intraepithelial carcinoma (STIC) involving carcinoma (STIC) shows a mutation pattern of staining with p53
tubal fimbria. Note malignant epithelium of serous type with a high (strong, diffuse staining in this case) (right). Caution is warranted in the
nuclear grade and nuclear disarray, contrasting with that of normal cili- interpretation of tubal intraepithelial carcinoma, as adenocarcinoma
ated benign tubal type epithelium (left). Serous tubal intraepithelial metastatic to the fallopian tube may closely mimic STIC
Fig. 10.45 Serous tubal intraepithelial carcinoma involving tubal fim- denced by strong, diffuse nuclear staining in >75% of the tumor cells.
bria, with underlying mucosal invasion (left). Both the STIC and the In approximately 10% of cases, a p53 mutation is indicated by virtually
invasive carcinoma show a mutation pattern of p53 staining, as is typi- no p53 staining (“null pattern”)
cal of high-grade serous carcinoma. In this case the mutation is evi-
354 K. Carrick and W. Zheng
Fig. 10.50 Mammary
carcinoma of ductal type
metastatic to the ovary.
Mammary carcinomas often
express GATA3 (left) and ER
(right). Of note, endometrioid
adenocarcinomas also
infrequently express GATA3.
Expression of PAX-8 and/or
CA125 strongly favors
endometrioid adenocarcinoma
s quamous cell carcinoma, mesothelioma, and rare endo- Serous carcinoma: Is positive for CK7 and negative for
metrial adenocarcinomas [141]. CK20, and may show positivity for ER and mammaglobin,
–– Use of a directed panel of immunostains is recommended although mutation-type staining for p53 and positivity for
when ancillary testing is needed, as no single marker is PAX8, CA125, and/or WT1 strongly favor serous carcinoma.
adequately informative. Carcinoid tumor: Is positive for synaptophysin in virtu-
ally 100% of cases, and with other neuroendocrine markers
Differential Diagnosis: The differential diagnosis (NSE, PGP9.5, and chromogranin) in many cases.
includes primary ovarian surface epithelial carcinomas (endo- Granulosa cell tumor: Is typically unilateral and shows
metrioid, serous, or undifferentiated carcinoma), other meta- ovoid or angulated small nuclei with scant indistinct cyto-
static carcinomas, granulosa cell tumor, carcinoid tumor, plasm and characteristic nuclear grooves in at least some
desmoplastic small round cell tumor, and hematopoietic neo- cells. These tumors are usually EMA negative but may show
plasms [12]. Metastatic breast tumors with a predominantly patchy or punctate marking for LMWCKs, whereas carcino-
glandular architecture may resemble endometrioid adenocar- mas tend to have strong diffuse marking for EMA and vari-
cinoma, and tumors with an insular pattern may mimic carci- ous cytokeratins. An inhibin+/calretinin+/WT1+/CD99+/
noid tumor. A tumor with a diffuse/dispersed or single file FOXL2+/GATA3−/mammaglobin−/GCDFP− immunophe-
pattern may suggest leukemia or lymphoma. Metastatic notype is consistent with granulosa cell tumor [23].
breast carcinoma may also mimic granulosa cell tumor [2]. Desmoplastic small round cell tumor: Is characterized by
Careful attention to gross and microscopic features will pro- small round uniform cells with hyperchromatic nuclei and
vide a clue to the diagnosis in many cases, but immunostain- scant cytoplasm, and may have tubular or glandular forma-
ing and clinical correlation will be required for some cases. tions or signet ring cells that suggest the possibility of a met-
Features helpful in the differential diagnosis are as follows: astatic breast carcinoma. Although these tumors may occur
Endometrioid adenocarcinoma: Shares the CK7+/ in the elderly, most tumors develop in adolescents or young
CK20−/ER+ immunoprofile typical of breast carcinoma and adults. They show a distinctive immunohistochemical stain-
may also mark with mammaglobin and GATA3 less fre- ing pattern with positivity for epithelial (keratin, EMA),
quently, although positivity for PAX8 and/or CA125 strongly muscular (desmin), and neural (NSE) markers, which con-
favors endometrioid adenocarcinoma. Endometrioid adeno- trasts with that of metastatic breast carcinoma [12, 142].
carcinoma tends to be unilateral and may arise in the setting Hematolymphoid neoplasms: If the metastatic breast car-
of endometriosis or an endometrioid adenofibroma. cinoma has a dispersed, diffuse, or cord-like pattern, the
10 Secondary Tumors of the Ovary 357
findings could suggest a hematolymphoid neoplasm. Use of cases, and melanin pigment in approximately 50% of cases.
immunohistochemical stains for keratins and hematolym- Unexpected findings include clear cells, rhabdoid cells, and
phoid antigens is confirmatory of the diagnosis. myxoid stroma. Architecturally, the tumor tends to have a
Management and Outcomes: A 2016 study of 28 patients solid growth pattern. A distinctive architectural feature found
with ovarian metastasis from previously treated breast cancer in many metastatic melanomas is a pattern of discrete
demonstrated a mean survival of 49.5 months (approximately rounded aggregates with a nevoid appearance. Other possi-
4 years) after discovery of the ovarian metastasis. Survival ble patterns include follicle-like spaces and a pseudopapil-
was 120.8 months for patients with an ovarian metastasis lary appearance [32, 145–147]. Melanoma metastatic to the
alone versus 106.9 months for patients with multiple second- ovary may grossly and microscopically mimic an ovarian
ary sites, a difference of nearly 14 months, implying that mul- primary neoplasm but, like other tumors metastatic to the
tifocal metastasis is a poor prognostic feature [132]. A 2010 ovary, metastatic melanoma tends to have a multinodular
study of 29 women with ovarian involvement with metastatic gross and microscopic appearance.
breast cancer compared outcomes between patients who The differential diagnosis for metastatic melanoma
underwent macroscopic resection of metastases versus includes primary ovarian melanoma and, depending on the
patients who did not; data indicated that survival improved cytoarchitectural features of the case, may include granulosa
significantly when optimal debulking surgery was performed, cell tumor of juvenile or adult type, small-cell carcinoma of
even when patients had advanced pelvic disease [133]. hypercalcemic type, undifferentiated carcinoma, lipid-poor
steroid cell tumor, pregnancy luteoma, and sarcoma [32]. A
panel of immunohistochemical stains is very helpful in
10.14 Metastatic Malignant Melanoma establishing the diagnosis of melanoma, especially in cases
where melanin pigment is not identified. S-100 marks at
Autopsy studies have demonstrated ovarian involvement by least 98% of melanomas, regardless of histologic type,
metastatic melanoma in approximately 20% of patients who although S-100 is not a specific marker, and thus is most
died of melanoma, although clinical detection of ovarian helpful as a screen for melanoma. HMB-45 and melan-A are
metastasis is rare [2, 12]. Data from occasional case reports both relatively specific for melanocytic type cells and have a
and three large series of melanoma metastatic to the ovary sensitivity of 60–80%. SOX-10 positivity has also been
indicate that patients may be symptomatic from the ovarian found in the majority of melanomas. Of note, S-100 and
metastasis, with abdominal swelling or pain [143–147]. A melan-A are positive in some ovarian sex cord-stromal
majority of patients with ovarian metastasis also have syn- tumors [23, 150].
chronous extraovarian metastasis. Most of the reported cases Once a diagnosis of melanoma has been established, it
have had an antecedent cutaneous melanoma, prior removal must be determined whether the melanoma is primary or
of a pigmented skin lesion, or rarely an ocular melanoma metastatic. Primary ovarian malignant melanoma is rare but
[144, 148]. The interval may be long (up to 22 years in one may occur in the setting of teratoma, including monodermal
reported case), and in some patients no prior history of ante- teratoma (struma ovarii) [151, 152]; thus, a background of
cedent melanoma is described. In these instances, the pri- teratoma is an important clue to the possible primary nature
mary melanoma may be unknown to the clinician, or may of the neoplasm. In these rare cases, junctional activity at the
have regressed [149]. base of squamous epithelium may be seen. Bilaterality and
Pathologists know that melanoma may show different multinodular growth suggest metastatic tumor, even in the
cytologic features and different architectural patterns, and absence of an identifiable primary tumor. In some cases it
may appear in unexpected locations; thus, the diagnosis may may not be possible to determine whether the melanoma is
be challenging. In the ovaries, metastatic melanoma may be primary or metastatic [32].
bilateral or unilateral, and may be solid, solid and cystic, or
even predominantly cystic. No specific gross features have
been described except for brown or black coloration, which 10.15 M
etastasis from Kidney and Lower
may be focal, in some tumors. Tumors average approxi- Urinary Tract
mately 10 cm but may be as large as 20 cm [145–147].
Microscopically, the most common cell type is a variably Metastasis from kidney: Renal cell carcinoma rarely metas-
pleomorphic, large epithelioid cell with abundant eosino- tasizes to the ovaries. Most reported cases of renal cell carci-
philic cytoplasm. Small cells with scant cytoplasm are also noma metastatic to the ovaries have been of the clear cell
common. Fewer tumors have a spindled cytomorphology. type [153–155], although other types (chromophobe, collect-
Admixed cell types may be seen. Prominent nucleoli, char- ing duct, and unclassified types) have been infrequently
acteristic of melanoma, are seen in approximately 80% of reported [153, 156]. In some cases, the ovarian tumor is dis-
cases, nuclear pseudoinclusions in approximately 25% of covered prior to diagnosis of the renal primary. In other
358 K. Carrick and W. Zheng
cases, the ovarian tumor is discovered after diagnosis of the clear-cell carcinoma and lipid-rich ovarian steroid cell
renal primary, with one reported ovarian metastasis present- tumors are as follows:
ing 14 years after discovery of the renal primary.
The differential diagnosis of metastatic clear-cell renal –– Clear-cell renal cell carcinoma: The most useful panel
cell carcinoma involving the ovary includes mullerian clear- of antibodies for distinguishing clear-cell renal cell carci-
cell carcinoma and lipid-rich ovarian steroid cell tumor. noma from mullerian clear-cell carcinoma is CK7, CD10,
Although some morphologic overlap exists between these and RCC, as clear-cell renal cell carcinoma typically
entities (most notably, clear cytoplasm), certain distinctive expresses CD10 and RCC but not CK7, while mullerian
morphologic features and variations in the immunostaining clear-cell carcinoma typically expresses CK7 but not
patterns facilitate diagnosis. Morphologic features helpful in CD10 or RCC. Clear-cell renal cell carcinoma also typi-
the differential diagnosis are as follows: cally expresses CA-IX and vimentin, while mullerian
clear-cell carcinoma often expresses Napsin-A and p504s.
• Metastatic clear-cell renal cell carcinoma: Tends to Of note, both tumors usually express PAX8, and fewer
maintain the expected morphologic pattern. Grossly, than 10% of mullerian clear-cell carcinomas express ER
tumors are solid or solid and cystic, and uniformly or and PR [23, 157].
focally yellow to orange [153–155]. Classic morphologic –– Steroid cell tumors: Are typically inhibin+/calretinin+/
features of clear-cell renal cell carcinoma include a pre- melan-A+/EMA−. Approximately 40–50% stain for
dominance of cells with clear cytoplasm disposed in sheets cytokeratin [23].
or in tubules containing intraluminal blood or eosinophilic
material. The prominent vascular pattern of renal cell car- Metastasis from lower urinary tract: Urothelial carci-
cinoma is also usually present. In contrast to the relatively noma metastatic to the ovary from the urinary bladder, ure-
uniform clear-cell cytology typically shown by metastatic ter, or renal pelvis is rare (Figs. 10.52 and 10.53). This may
clear-cell renal cell carcinoma, mullerian clear-cell carci- mimic a primary tubo-ovarian transitional cell carcinoma,
noma typically has greater cytologic and architectural malignant Brenner tumor, or undifferentiated carcinoma
variation, with a mix of clear, eosinophilic, cuboidal, hob- [32]. A single case of urothelial carcinoma of renal pelvis
nail, and flat cells. The cells may be disposed in various origin and three cases of urothelial carcinoma of bladder ori-
architectural patterns including sheets, papillae, and lining gin with glandular differentiation and signet ring morphol-
cystic spaces, usually with at least focal presence of the ogy are reported to have metastasized to the ovary and
round tubulocystic glands characteristic of mullerian generated Krukenberg tumors [158, 159]. Thorough sam-
clear-cell carcinoma. Luminal mucin, eosinophilic hyaline pling of the ovarian tumor may disclose benign or borderline
globules, and papillae with stromal hyalinization also Brenner elements, or the presence of benign mucinous ele-
favor mullerian clear-cell carcinoma [32, 155]. ments, which provide clues to the diagnosis of a primary
• Ovarian steroid cell tumors of lipid-rich type: Do not malignant Brenner tumor. Most primary ovarian transitional
show the tubular growth pattern with intraluminal blood cell carcinomas are thought to be high-grade serous carci-
or eosinophilic material typical of clear-cell renal cell car- noma with transitional like morphology; these tumors may
cinoma. Careful morphologic examination should pro- have areas of more classic serous-type morphology, or less
vide a clue to the diagnosis. commonly endometrioid morphology, which establish the
primary nature of the tumor. Undifferentiated ovarian carci-
Typical immunohistochemical staining patterns helpful in noma tends to have high-grade cytologic features and may
distinguishing clear-cell renal cell carcinoma from mullerian have pseudopapillae due to necrosis, but lacks the true
disseminated disease with associated abdominal or pelvic leukemia, and 22% of cases of chronic lymphocytic leuke-
lymph node involvement. In a minority of cases, however, mia [177]. The percentage of cases involving the ovary was
there is exclusive involvement of a single ovary, suggesting noted to decrease over the time period of the study, likely
that the ovary is the primary site [167]. reflecting increasing efficacy of treatment.
Ovarian involvement by lymphoma may be a small inci- Leukemic involvement of the ovary infrequently produces
dental finding or a large mass (masses ranging up to 25 cm a symptomatic ovarian mass. In a 1997 study focusing on 11
have been reported). The external surface is typically smooth cases of granulocytic sarcoma of the female genital tract, 7
or bosselated and intact. The cut surface is usually white to cases had ovarian involvement and produced masses ranging
tan, fleshy to firm or rubbery, and solid, with occasional from 5 to 14 cm. Three of the tumors had a green cut surface.
areas of cystic degeneration, hemorrhage, or necrosis [168– Myeloid differentiation was readily recognizable in three of
171]. Most lymphomas involving the ovary are of B-cell the cases, whereas the other cases presented more of a diag-
type, the most common subtypes including diffuse large nostic challenge due to primitive cytologic features without
B-cell type, Burkitt lymphoma, and follicular lymphomas. the classic eosinophilic myelocytes. Sclerosis was present in
Rare tumors are of T-cell type, reported cases including the majority of tumors, and prominent in one tumor [176].
T-cell anaplastic large-cell lymphoma and precursor Microscopic features of leukemia in the ovary are similar to
T-lymphoblastic lymphoma [167–169]. Primary plasmacy- those of lymphoma, although nuclei have paler, finer chro-
toma of the ovary and secondary ovarian involvement by matin than do most lymphomas, and the cytoplasm is more
multiple myeloma have also been rarely reported [172–174]. abundant, pale pink, or deeply eosinophilic. Once leukemia
Cytologically, lymphomas in the ovary are similar to their enters the differential diagnosis, the diagnosis can be con-
counterparts in lymph nodes and extranodal sites, although firmed by staining for chloroacetate esterase or a directed
sclerosis may be more prominent in the ovary, and may battery of immunohistochemical stains. The differential
induce unusual patterns including single file, trabecular, diagnosis includes lymphoma and, as noted above with lym-
insular, and storiform arrangements. Lymphomas may also phoma, other tumors consisting of singly disposed cells such
diffusely involve the ovary, and spare or obliterate physio- as granulosa cell tumor, dysgerminoma, undifferentiated car-
logic structures [12, 175]. cinoma, melanoma, and metastatic carcinomas including
Possibly due to the propensity of these tumors to have breast carcinoma.
associated sclerosis, lymphoma cells in the ovary may appear
more epithelioid or spindled than in extraovarian sites, caus-
ing diagnostic difficulty. The differential diagnosis includes 10.18 Other Rare Metastatic Tumors
other tumors characterized by relatively small cells that may
be singly disposed, including granulosa cell tumor, dysger- Virtually any tumor can metastasize to the ovary, and the list
minoma, small-cell carcinoma of hypercalcemic type, granu- of such tumors is not limited to those reported above. Thus,
locytic sarcoma, undifferentiated carcinoma, melanoma, and it is critical that when one encounters an ovarian tumor with
breast carcinoma. The differential diagnostic considerations cytologic or architectural features uncommon to primary
can be excluded with a directed panel of immunohistochemi- ovarian neoplasia, metastasis is considered.
cal stains [12, 175]. Head and neck tumors of thyroid, salivary gland,
Leukemia involving the ovary: The most substantial esophageal, and sinonasal origin: Head and neck tumors
data regarding leukemic involvement of the ovaries consists have infrequently been the source of ovarian metastasis. The
of autopsy studies and case series reported before 2000, thyroid is a rare source of ovarian metastasis, even in autopsy
which would be expected to reflect treatment efficacy at the series [2]. Few cases of thyroid follicular carcinoma and pap-
time [168, 170, 176–178]. More recent information consists illary thyroid carcinoma have been reported to metastasize to
of one series of 124 leukemic ovarian tumors which focused the ovary, up to 12 years after the initial presentation in the
on survival data [179], and infrequent case reports [180, thyroid [182–184]. In a recent study, anaplastic thyroid car-
181]. Leukemic involvement of the ovary appears to be less cinoma was found to involve the ovary in 2% of 45 cases
common than lymphomatous involvement of the ovary; in examined at autopsy [185]. As thyroid tissue may develop in
studies that have included both leukemias and lymphomas, the ovary as a component of a teratoma or as a monodermal
there were only three cases of ovarian leukemia among a teratoma, malignant thyroid tissue in the ovary raises the dif-
total of 61 cases. In a 1987 autopsy study of 1206 cases of ferential diagnosis of malignant struma ovarii. If other tera-
acute and chronic leukemias gathered between 1958 and tomatous elements are not found on thorough sampling, a
1982, the ovary was involved by leukemia in 11% of the diagnosis of malignant struma ovarii should be made only
cases of acute myelogenous leukemia, 9% of cases of chronic when the possibility of spread from a thyroid neoplasm has
myelogenous leukemia, 21% of cases of acute lymphoblastic been explored and excluded [182].
10 Secondary Tumors of the Ovary 361
19.
Strickland S, Wasserman J, Giassi A et al (2016) 37. Lee SJ, Lee J, Lim HY et al (2010) Survival benefit from ovar-
Immunohistochemistry in the diagnosis of mucinous neoplasms ian metastatectomy in colorectal cancer patients with ovarian
involving the ovary: The added value of SATB2 and biomarker dis- metastasis: a retrospective analysis. Cancer Chemother Pharmacol
covery through protein expression database mining. Int J Gynecol 66:229–35
Pathol 35:191–208 38. Aparicio T, Zaanan A, Svrcek M et al (2014) Small bowel adeno-
20. Lee K, Young R (2003) The distinction between primary and meta- carcinoma: epidemiology, risk factors, diagnosis and treatment. Dig
static mucinous carcinomas of the ovary: gross and histologic find- Liver Dis 46:97–104
ings in 50 cases. Am J Surg Pathol 27:281–292 39. Mitsushita J, Netsu S, Suzuki K et al (2017) Metastatic ovar-
21. Moh M, Krings G, Ates D et al (2016) SATB2 expression dis- ian tumors originating from a small bowel adenocarcinoma—
tinguishes ovarian metastases of colorectal and appendiceal ori- a case report and brief literature review. Int J Gynecol Pathol
gin from primary ovarian tumors of mucinous or endometrioid 36:253–260
type. Am J Surg Pathol 40:419–32 40. Kim CJ, Baruch-Oren T, Lin F et al (2016) Value of SATB2 immu-
22.
Krasinskas AM, Goldsmith JD (2014) Immunohistology nostaining in the distinction between small intestinal and colorectal
of the Gastrointestinal Tract. In: Dabbs DJ (ed) Diagnostic adenocarcinomas. J Clin Pathol 69:1046–1050
Immunohistochemistry: Theranostic and Genomic Applications. 4th 41. Chang HK, Yu E, Kim J et al (2010) Adenocarcinoma of the small
edn. Saunders Elsevier, Philadelphia, p 508–539 intestine: a multi-institutional study of 197 surgically resected
23. Rabban JT, Longacre TA (2014) Immunohistology of the female cases. Hum Pathol 41:1087–96
genital tract. In: Dabbs DJ (ed) Diagnostic Immunohistochemistry: 42. Hong SH, Koh YH, Rho SY et al (2009) Primary adenocarcinoma
Theranostic and Genomic Applications. 4th edn. Saunders Elsevier, of the small intestine: presentation, prognostic factors and clinical
Philadelphia p 653–709 outcome. Jpn J Clin Oncol 39:54–61
24. Young R (2006) From Krukenberg to today: the ever present prob- 43. Zaaimi Y, Aparicio T, Laurent-Puig P et al (2016) Advanced small
lems posed by metastatic tumors in the ovary: part I. Advances in bowel adenocarcinoma: molecular characteristics and therapeutic
Anatomic Pathology 13:205–227 perspectives. Clin Res Hepatol Gastroenterol 40:154–60
25. Grabiec J, Owen D (1985) Carcinoma of the stomach in young per- 44. Ronnett BM, Kurman RJ, Shmookler BM et al (1997) The mor-
sons. Cancer 56:388–396 phologic spectrum of ovarian metastases of appendiceal adenocar-
26. Bhargava R, Dabbs D (2014) Immunohistology of the Breast. In cinomas: a clinicopathologic and immunohistochemical analysis
Dabbs DJ (ed) Diagnostic Immunohistochemistry: Theranostic and of tumors often misinterpreted as primary ovarian tumors or meta-
Genomic Applications. 4th edn. Saunders Elsevier, Philadelphia, static tumors from other gastrointestinal sites. Am J Surg Pathol
p 710–761 21:1144–55
27. Yang C, Sun L, Zhang L et al (2017) Diagnostic utility of SATB2 in 45. Merino MJ, Edmonds P, LiVolsi V (1985) Appendiceal carcinoma
metastatic Krukenberg tumors of the ovary: an imunohistochemical metastatic to the ovaries and mimicking primary ovarian tumors. Int
study of 70 cases with comparison to CDX2, CK7, chromogranin, J Gynecol Pathol 4:110–20
and synaptophysin. Am J Surg Pathol 42:160–171 46. Misdraji J (2015) Mucinous epithelial neoplasms of the appendix
28. Yu P, Huang L, Cheng G et al (2017) Treatment strategy and and pseudomyxoma peritonei. Mod Pathol 28 Suppl 1:S67–79
prognostic factors for Krukenberg tumors of gastric origin: report 47. Young RH, Gilks CB, Scully RE (1991) Mucinous tumors of the
of a 10-year single-center experience from China. Oncotarget appendix associated with mucinous tumors of the ovary and pseu-
8:82558–82570 domyxoma peritonei. A clinicopathologic analysis of 22 cases sup-
29. Rosa F, Marrelli D, Morgagni P et al (2016) Krukenberg tumors of porting origin in the appendix. Am J Surg Pathol 15:415–429
gastric origin: the rationale for surgical resection and perioperative 48. Dietrich CS 3rd, Desimone CP, Modesitt SC et al (2007) Primary
treatments in a multicenter western experience. World J Surg 40:921–8 appendiceal cancer: gynecologic manifestations and treatment
30. Lerwill MF, Young RH (2006) Ovarian metastases of intestinal- options. Gynecol Oncol 104:602–6
type gastric carcinoma: a clinicopathologic study of 4 cases with 49. Young RH (2004) Pseudomyxoma peritonei and selected other
contrasting features to those of the Krukenberg tumor. Am J Surg aspects of the spread of appendiceal neoplasms. Semin Diagn
Pathol 30:1382–8 Pathol 21:134–50
31. Pilotti S, Rilke F, De Palo G (1982) Krukenberg tumor. Am J Surg 50. Ronnett BM, Kurman RJ, Zahn CM et al (1995) Pseudomyxoma
Pathol 6(5):486–8 peritonei in women: a clinicopathologic analysis of 30 cases with
32. Young RH (2007) From Krukenberg to today: the ever present emphasis on site of origin, prognosis, and relationship to ovar-
problems posed by metastatic tumors in the ovary. Part II. Adv Anat ian mucinous tumors of low malignant potential. Hum Pathol
Pathol 14(3):149–77 26:509–24
33. Judson K, McCormick C, Vang R, Yemelyanova AV, Wu LS,
51. Ronnett BM, Seidman JD (2003) Mucinous tumors arising in ovar-
Bristow RE, Ronnett BM (2008) Women with undiagnosed ian mature cystic teratomas: relationship to the clinical syndrome of
colorectal adenocarcinomas presenting with ovarian metastases: pseudomyxoma peritonei. Am J Surg Pathol 27:650–7
clinicopathologic features and comparison with women having 52. Stewart CJ, Ardakani NM, Doherty DA et al (2014) An evaluation
known colorectal adenocarcinomas and ovarian involvement. Int J of the morphologic features of low grade mucinous neoplasms of
Gynecol Pathol 27:182–90 the appendix metastatic in the ovary, and comparison with primary
34. Lash RH, Hart WR (1987) Intestinal adenocarcinomas metastatic to ovarian mucinous tumors. Int J Gynecol Pathol 33:1–10
the ovaries. A clinicopathologic evaluation of 22 cases. Am J Surg 53. Hristov AC, Young RH, Vang R et al (2007) Ovarian metastases of
Pathol 11:114–21 appendiceal tumors with goblet cell carcinoidlike and signet ring
35. Young RH, Hart WR (1998) Metastatic intestinal carcinomas
cell patterns: a report of 30 cases. Am J Surg Pathol 31:1502–11
simulating primary ovarian clear cell carcinoma and secretory 54. Tang LH, Shia J, Soslow RA et al (2008) Pathologic classifica-
endometrioid adenocarcinoma. A clinicopathologic and immuno- tion and clinical behavior of the spectrum of goblet cell carcinoid
histochemical study of five cases. Am J Surg Pathol 22:805–815. tumors of the appendix. Am J Surg Pathol 32:1429–1443
36. Pinto C, Pini S, Di Fabio F et al (2014) Treatment strategy for rectal 55. Moore L, Gajjar K, Jimenez-Linan M et al (2016) Prevalence of
cancer with synchronous metastases: 65 consecutive Italian cases appendiceal lesions in appendectomies performed during surgery
from the Bologna Multidisciplinary rectal cancer group. Oncology for mucinous ovarian tumors: a retrospective study. Int J Gynecol
86:135–42 Cancer 26(8):1386–9
10 Secondary Tumors of the Ovary 363
56. Ronnett BM, Yan H, Kurman RJ et al (2001) Patients with pseudo- metastases occurred, including involvement of the right ovary. Arch
myxoma peritonei associated with disseminated peritoneal adenomu- Surg 45:613–622
cinosis have a significantly more favorable prognosis than patients 78. Young RH, Scully RE (1985) Ovarian metastasis from cancer of the
with peritoneal mucinous carcinomatosis. Cancer 92(1):85–91 lung: problems in interpretation—a report of seven cases. Gynecol
57. Young RH, Hart WR (1989) Metastases from carcinomas of the Oncol 21:337–350
pancreas simulating primary mucinous tumors of the ovary: a 79. Irving JA, Young RH (2005) Lung carcinoma metastatic to the
report of seven cases. Am J Surg Pathol 13:748–756 ovary: a clinicopathologic study of 32 cases emphasizing their
58. Vakiani E, Young RH, Carcangiu ML et al (2008) Acinar cell car- morphologic spectrum and problems in differential diagnosis. Am J
cinoma of the pancreas metastatic to the ovary: a report of 4 cases. Surg Pathol 29:997–1006
Am J Surg Pathol 32:1540–1545 80. Srivastava A, Hornick JL (2009) Immunohistochemical staining for
59. Basturk O, Farris AB, Adsay V (2014) Immunohistology of the CDX-2, PDX-1, NESP-55, and TTF-1 can help distinguish gastro-
Pancreas and Hepatobiliary Tract. In: Dabbs DJ (ed) Diagnostic intestinal carcinoid tumors from pancreatic endocrine and pulmo-
Immunohistochemistry: Theranostic and Genomic Applications. 4th nary carcinoid tumors. Am J Surg Pathol 33:626–32
edn. Saunders Elsevier, Philadelphia p 540–583 81. Rabban JT, Lerwill MF, McCluggage WG et al (2009) Primary
60. Saung MT, Zheng L (2017) Current standards of chemotherapy for ovarian carcinoid tumors may express CDX-2: a potential pitfall
pancreatic cancer. Clin Ther Nov; 39:2125–2134 in distinction from metastatic intestinal carcinoid tumors involving
61. Young RH, Scully RE (1990) Ovarian metastasis from carcinoma of the ovary. Int J Gynecol Pathol 28:41–48
the gallbladder and extrahepatic bile ducts simulating primary tumors 82. Pasricha G, Padhi P, Daboul N et al (2017) Management of well-
of the ovary: a report of six cases. Int J Gynecol Pathol 9:60–72 differentiated gastroenteropancreatic neuroendocrine tumors
62. Corr BR, Mantia-Smaldone G, Cantor J et al (2013) Metastatic cholan- (GEPNETs): a review. Clin Ther 39:2146–2157
giocarcinoma to the ovary: a case series. Int J Gynecol Pathol 32:562–5 83. Young RH, Gersell DJ, Roth LM et al (1993) Ovarian metasta-
63. Khunamornpong S, Lerwill MF, Siriaunkgul S et al (2008)
sis from cervical carcinomas other than pure adenocarcinomas: a
Carcinoma of extrahepatic bile ducts and gallbladder metastatic to report of 12 cases. Cancer (phila) 15;71:407–418
the ovary: a report of 16 cases. Int J Gynecol Pathol 27:366–379 84. Ramalingam P, Malpica A, Deavers MT (2012) Mixed endocervical
64. Khunamornpong S, Siriaunkgul S, Suprasert P et al (2007)
adenocarcinoma and high-grade neuroendocrine carcinoma of the
Intrahepatic cholangiocarcinoma metastatic to the ovary: a report cervix with ovarian metastasis of the former component: a report of
of 16 cases of an underemphasized form of secondary tumor of 12 cases. Int J Gynecol Pathol 31:490–496
the ovary that may mimic primary neoplasia. Am J Surg Pathol 85. Shimada M, Kigawa J, Nishimura R et al (2006) Ovarian metastasis
31:1788–99 in carcinoma of the uterine cervix. Gynecol Oncol 101:234–237
65. Zhuang X, Xiao YP, Tan LH et al (2017) Efficacy and safety of 86. Reyes C, Murali R, Park KJ (2015) Secondary involvement of
chemotherapy with or without targeted therapy in biliary tract can- the adnexa and uterine corpus by carcinomas of the uterine cer-
cer: A meta-analysis of 7 randomized controlled trials. J Huazhong vix: a detailed morphologic description. Int J Gynecol Pathol
Univ Sci Technolog Med Sci.37:172–178 34:551–63
66. Rizvi S, Gores GJ (2017) Emerging molecular therapeutic targets 87. Nguyen L, Brewer CA, DiSaia PJ (1998) Ovarian metastasis of
for cholangiocarcinoma. J Hepatol Sep;67:632–644 stage IB1 squamous cell cancer of the cervix after radical paramet-
67. Khunamornpong S, Siriaunkgul S, Chunduan A (1999) Metastatic rectomy and oophoropexy. Gynecol Oncol 68:198–200
hepatocellular carcinoma of the ovary. Int J Gynecol Obstet 88. Karamurzin YS, Kiyokawa T, Parkash V et al (2015) Gastric type
Feb;64:189–191 endocervical adenocarcinoma: an aggressive tumor with unusual
68. Young RH, Gersell DJ, Clement PB et al (1992) Hepatocellular car- metastatic patterns and poor prognosis. Am J Surg Pathol 39:1449–57
cinoma metastatic to the ovary: a report of three cases discovered 89. Zaino R, Whitney C, Brady MF et al (2001) Simultaneously
during life with discussion of the differential diagnosis of hepatoid detected endometrial and ovarian carcinomas—a prospective clini-
tumors of the ovary. Hum Pathol 23:574–580 copathologic study of 74 cases. Gynecol Oncol 83:355–362
69. Su J-S, Chen Y-T, Wang R-C et al (2013) Clinicopathological char- 90. Kline RC, Wharton JT, Atkinson EN et al (1990) Endometrioid car-
acteristics in the differential diagnosis of hepatoid adenocarcinoma: cinoma of the ovary: retrospective review of 145 cases. Gynecol
A literature review. World J Gastroenterol. 19:321–327 Oncol 39:337–46
70. Wang EA, Stein JP, Bellavia RJ et al (2017) Treatment options 91. Soliman PT, Slomovitz BM, Broaddus RR et al (2004) Synchronous
for unresectable HCC with a focus on SIRT with Yttrium-90 resin primary cancers of the endometrium and ovary: a single institution
microspheres. Int J Clin Prac 71(11) review of 84 cases Gynecol Oncol 94:456–462
71. Robboy SJ, Scully RE, Norris HJ (1974) Carcinoid metastatic to 92. Ramus SJ, Elmasry K, Luo Z et al (2008) Predicting clinical out-
the ovary. A clinicopathologic analysis of 35 cases. Cancer (Phila) come in patients diagnosed with synchronous ovarian and endome-
33:798–811 trial cancer. Clin Cancer Res 14:5840–5848
72. Strosberg J, Nasir A, Cragun J et al (2007) Metastatic carcinoid 93. Jain V, Sekhon R, Pasricha S et al (2017) Clinicopathological char-
tumor to the ovary: a clinicopathologic analysis of seventeen acteristics and prognostic factors of synchronous endometrial and
cases. Gynecol Oncol 106(1):65–68 ovarian cancers- a single-institute review of 43 cases. Int J Gynecol
73. Mark J, Bush S, Glazer E et al (2015) Metastatic VIPoma present- Cancer 27:938–946
ing as an ovarian mass. Int J Surg Case Rep 17:167–169 94. Lim YK, Padma R, Foo L et al (2011) Survival outcome of women
74. Watt DG, Pandanaboyana S, Herrington CS et al (2013) Pancreatic with synchronous cancers of endometrium and ovary: a 10 year ret-
glucagonoma metastasizing to the right ovary five years after initial rospective cohort study. J Gynecol Oncol 22:239–243
surgery: a case report. JOP 10;14:510–514 95. Bese T, Sal V, Kahramanoglu I, Tokgozoglu N et al (2016)
75. Oberg KC, Wells K, Seraj IM et al (2002) ACTH secreting islet cell Synchronous primary cancers of the endometrium and ovary with
tumor of the pancreas presenting as bilateral ovarian tumors and the same histopathologic type versus endometrial cancer with
Cushing’s syndrome. Int J Gynecol Pathol 21(3):276–280 ovarian metastasis: a single institutional review of 72 cases. Int J
76. Brown BL, Scharifker DA, Gordon R et al (1980) Bronchial carci- Gynecol Cancer 26:394–406
noid tumor with ovarian metastasis. A light microscopic and ultra- 96. Kim KR, Scully RE (1990) Peritoneal keratin granulomas with
sound study. Cancer 46:543–546 carcinomas of endometrium and ovary and atypical polypoid
77. Hopping RA, Dockerty MB, Masson JC (1942) Carcinoid tumor of adenomyoma of endometrium. A clinicopathologic analysis of 22
the appendix. Report of a case in which extensive intra-abdominal cases. Am J Surg Pathol 14:925–932
364 K. Carrick and W. Zheng
97. Scully RE, Young RH, Clement PB (1998) Tumors of the ovary, 117. Kindelberger DW, Lee Y, Miron A et al (2007) Intraepithelial car-
maldeveloped gonads, fallopian tube, and broad ligament. Armed cinoma of the fimbria and pelvic serous carcinoma: Evidence for
Forces Institute of Pathology, Washington, D.C. a causal relationship. Am J Surg Pathol 31:161–169
98. Chao A, Wu RC, Jung SM et al (2016) Implication of genomic 118. Crum CP (2009) Intercepting pelvic cancer in the distal fallopian
characterization in synchronous endometrial and ovarian cancers tube: theories and realities. Mol Oncol 3:165–170
of endometrioid histology. Gynecol Oncol 143:60–67 119. Kurman RJ, Shih IeM (2010) The origin and pathogenesis of
99. Anglesio MS, Wang YK, Maassen M et al (2016) Synchronous epithelial ovarian cancer: a proposed unifying theory. Am J Surg
Endometrial and Ovarian Carcinomas: Evidence of Clonality. J Pathol 34:433–43
Natl Cancer Inst 108(6):djv428 120. Nik NN, Vang R, Shih IeM et al (2014) Origin and pathogenesis of
100. Schultheis AM, Ng CK, De Filippo MR et al (2016) Massively pelvic (ovarian, tubal, and primary peritoneal) serous carcinoma.
Parallel Sequencing-Based Clonality Analysis of Synchronous Annu Rev Pathol 9:27–45
Endometrioid Endometrial and Ovarian Carcinomas. J Natl 121. Vang R, Wheeler JE (2011) Diseases of the fallopian tube and
Cancer Inst 108(6):djv427 paratubal region. In: Kurman R, Hedrick Ellenson L, Ronnet
101. Fadare O, Parkash V, Gwin K et al (2013) Utility of α-methylacyl- B (eds) Blaustein’s Pathology of the Female Genital Tract,
coenzyme-A racemase (p504s) immunohistochemistry in dis- 6th ed. Springer Science+Business Media LLC, Boston, MA,
tinguishing endometrial clear cell carcinomas from serous and p 529–578
endometrioid carcinomas. Hum Pathol 44:2814–21 122. Alvarado-Cabrero I, Young RH, Vamvakas EC et al
102. Al-Maghrabi JA, Butt NS, Anfinan N et al (2017) Infrequent (1999) Carcinoma of the fallopian tube: a clinicopathologic study
Immunohistochemical Expression of Napsin A in Endometrial of 105 cases with observations on staging and prognostic fac-
Carcinomas. Appl Immunohistochem Mol Morphol 25:632–638 tors. Gynecol Oncol 72:367–379
103. Koyuncuoğlu M, Saatli B, Yıldırım N (2017) Ovarian metastasis 123. Stewart SL, Wike JM, Foster SL et al (2007) The incidence of
of Müllerian adenosarcoma of the cervix with sarcomatous over- primary fallopian tube cancer in the United States. Gynecol Oncol
growth. Turk J Obstet Gynecol 14:195–198 107:392–397
104. Acosta-Sison H (1958) The relative frequency of various anatomic 124. Obermair A, Taylor KH, Janda M et al (2001) Primary fallopian
sites as the point of first metastasis in 32 cases of chorioepithe- tube carcinoma: the Queensland experience. Int J Gynecol Cancer
lioma. Am J Obstet Gynecol 75:1149–1152 11:69–72
105. Milingos D, Doumplis D, Savage P, Seckl M, Lindsay I, Smith JR 125. Cass I, Holschneider C, Datta N et al (2005) BRCA-mutation-
(2007) Placental site trophoblastic tumor with an ovarian metasta- associated fallopian tube carcinoma: a distinct clinical phenotype?
sis. Int J Gynecol Cancer 17:925–927 Obstet Gynecol 106:1327–1334
106. Young RH, Scully RE (1990) Sarcomas metastatic to the ovary. A 126. Levine DA, Argenta PA, Yee CJ et al (2003) Fallopian tube and
report of 21 cases. Int J Gynecol Pathol 9:231–252 primary peritoneal carcinomas associated with BRCA muta-
107. Young RH, Prat J, Scully RE (1984) Endometrial stromal sarco- tions. J Clin Oncol 21:4222–4227
mas of the ovary. A clinicopathologic analysis of 23 cases. Cancer 127. Wheal A, Jenkins R, Mikami Y et al (2017) Primary Mucinous
53:1143–1155 Carcinoma of the Fallopian Tube: Case Report and Review of
108. Khalifa MA, Hansen CH, Moore JL Jr et al (1996) Endometrial Literature. Int J Gynecol Pathol 36:393–399
stromal sarcoma with focal smooth muscle differentiation: recur- 128. Movahedi-Lankarani S, Krishnamurti U, Bell DA et al (2017)
rence after 17 years: a follow-up report with discussion of the Protocol for the examination of specimens from patients
nomenclature. Int J Gynecol Pathol 15:171–176 with primary tumors of the ovary, fallopian tube, or perito-
109. Leitao MM, Sonoda Y, Brennan MF et al (2003) Incidence of neum. Version: OvaryFallopian 1.0.0.1. College of American
lymph node and ovarian metastasis in leiomyosarcoma of the Pathologists (CAP)
uterus. Gynecol Oncol 91:209–212 129. McCluggage WG, Hirschowitz L, Gilks CB et al (2017)
110. Oliva E, Egger JF, Young RH (2014) Primary endometrioid stromal The Fallopian Tube Origin and Primary Site Assignment in
sarcoma of the ovary: a clinicopathologic study of 27 cases with Extrauterine High-grade Serous Carcinoma: Findings of a Survey
morphologic and behavioral features similar to those of uterine low- of Pathologists and Clinicians. Int J Gynecol Pathol 36:230–239
grade endometrial stromal sarcoma. Am J Surg Pathol 38:305–315 130. Bolton KL, Chenevix-Trench G, Goh C et al (2012) Association
111. Savage J, Adams E, Veras E et al (2017) Choriocarcinoma in between BRCA1 and BRCA2 mutations and survival in women
Women: Analysis of a Case Series With Genotyping. Am J Surg with invasive epithelial ovarian cancer. JAMA 307:382–390
Pathol 41:1593–1606 131. De Jaeghere E, Vandecasteele K, Claes K et al (2017) Incorporating
112. Garavaglia E, Pella F, Montoli S et al (2010) Treatment of recur- PARP-inhibitors into clinical routine: A tailored treatment strat-
rent or metastatic low-grade endometrial stromal sarcoma: three egy to tackle ovarian cancer. Acta Clin Belg 72:6–11
case reports. Int J Gynecol Cancer 20:1197–1200 132. Pimentel C, Becquet M, Lavoue V et al (2016) Ovarian metas-
113. Nakamura K, Nakayama K, Ishikawa M et al (2016) Letrozole as tasis from breast cancer: a series of 28 cases. Anticancer Res
second-line hormonal treatment for recurrent low-grade endome- 36:4195–4200
trial stromal sarcoma: A case report and review of the literature. 133. Bigorie V, Morice P, Duvillard P et al (2010) Ovarian metastases
Oncol Lett 12:3856–3860 from breast cancer: a report of 29 cases. Cancer 116:799–804
114. Baek MH, Park JY, Rhim CC et al (2016) Immunohistochemical 134. Gagnon Y, Tetu B (1989) Ovarian metastases of breast carcinoma.
Characterization of Histone Deacetylase as a Potential Prognostic A clinicopathologic study of 59 cases. Cancer 64(4):892–898
Marker and Therapeutic Target in Endometrial Stromal Sarcoma. 135. Struewing JP, Hartge P, Wacholder S et al (1997) The risk of can-
Anticancer Res 36:2527–34 cer associated with specific mutations of BRCA1 and BRCA2
115. Ricci S, Stone RL, Fader AN (2017) Uterine leiomyosarcoma: among Ashkenazi Jews. N Eng J Med 336:1401–1408
Epidemiology, contemporary treatment strategies and the impact 136. Curtin JP, Barakat RR, Hoskins WJ (1994) Ovarian disease in
of uterine morcellation. Gynecol Oncol. Apr;145:208–216 women with breast cancer. Obstet Gynecol 84:449–452
116. Piek JM, Kenemans P, Verheijen RH (2004) Intraperitoneal serous 137. Yagi Y, Maeda T, Yoshimitsu Y et al (2014) A case of breast cancer
adenocarcinoma: a critical appraisal of three hypotheses on its treated with chemotherapy after resection of giant ovarian metas-
cause. Am J Obstet Gynecol 191:718–732 tasis. Gan To Kagaku Ryoho 41:633–635
10 Secondary Tumors of the Ovary 365
138. Harris M, Howell A, Chrissohou M et al (1984) A comparison of ments on ovarian tumors metastatic from the urinary tract. Int J
the metastatic pattern of infiltrating lobular carcinoma and infil- Gynecol Pathol 25:147–150
trating ductal carcinoma of the breast. Br J Cancer 50:23–30 159. Young RH, Scully RE (1998) Urothelial and ovarian carcinomas
139. Bennett JA, Young RH, Chuang AY et al (2018) Ovarian metastasis of identical cell types: problems in interpretation. A report of
of breast cancers with signet ring cells: a report of 17 cases includ- three cases and review of the literature. Int J Gynecol Pathol
ing 14 Krukenberg tumors. Int J Gynecol Pathol 37:507–515 7:197–211
140. Hagemann IS, Pfeifer JD, Cao D (2013) Mammaglobin expres- 160. Min KW, Paik SS, Han H et al (2014) Tumour-to-tumour metas-
sion in gynecologic adenocarcinomas. Hum Pathol 44:628–635 tasis of lung adenocarcinoma to ovarian serous cystadenoma. J
141. Miettinen M, McCue PA, Sarlomo-Rikala M et al (2014) GATA3: Obstet Gynaecol 34:650–651
a multispecific but potentially useful marker in surgical pathol- 161. Huysentruyt CJ, Vandevijver NM, Jan van Suylen R et al (2010)
ogy: a systematic analysis of 2500 epithelial and nonepithelial Adenocarcinoma of the fetal lung-type metastatic to the ovary. Int
tumors. Am J Surg Pathol 38:13–22 J Gynecol Pathol 29:339–40
142. Thway K, Noujaim J, Zaidi S et al (2016) Desmoplastic Small 162. Yu L, Li X, Yang W (2009) Pulmonary blastoma metastatic to the
Round Cell Tumor: Pathology, Genetics, and Potential Therapeutic ovary. Int J Gynecol Pathol 28:59–62
Strategies. Int J Surg Pathol 24:672–684 163. Martín-Hernández R, Villanueva MM, Sánchez MN et al (2015)
143. Fenzl V, Prkacin I, Skrtic A (2011) Malignant melanoma ovarian Ovarian Metastasis of a Thymoma: Report of a Case and Literature
metastasis mimicking acute tuboovarian abscess. Eur J Gynaecol Review. Int J Gynecol Pathol 34:374–8
Oncol 32:582–584 164. Young RH, Kozakewich HP, Scully RE (1993) Metastatic ovarian
144. Mandato VD, Kobal B, Di Stefano A et al (2010) Choroidal mela- tumors in children: a report of 14 cases and review of the litera-
noma metastasized to the ovary: case report and review of the lit- ture. Int J Gynecol Pathol 12:8–19
erature. Eur J Gynaecol Oncol. 31:109–113 165. Freeman C, Berg JW, Cutler SJ (1972) Occurrence and prognosis
145. Fitzgibbons PL, Martin SE, Simmons TJ (1987) Malignant mela- of extranodal lymphomas. Cancer 29:252–60
noma metastatic to the ovary. Am J Surg Pathol 11:959–964 166. Halpin TF (1975) Gynecologic implications of Burkitt's tumor.
146. Gupta D, Deavers MT, Silva EG et al (2004) Malignant melanoma Obstet Gynecol Surv 30:351–358
involving the ovary: a clinicopathologic and immunohistochemi- 167. Vang R, Medeiros LJ, Warnke RA et al (2001) Ovarian non-
cal study of 23 cases. Am J Surg Pathol 28:771–80 Hodgkin's lymphoma: a clinicopathologic study of eight primary
147. Young RH, Scully RE (1991) Malignant melanoma metastatic cases. Mod Pathol 14:1093–1099
to the ovary: a clinicopathologic analysis of 20 cases. Am J Surg 168. Osborne BM, Robboy SJ (1983) Lymphomas or leukemia pre-
Pathol 15:849–860 senting as ovarian tumors. An analysis of 42 cases. Cancer
148. Lobo J, Pinto C, Freitas M et al (2017) Ovarian metastasis from 52:1933–1943
uveal melanoma with MLH1/PMS2 protein loss in a patient with 169. Monterroso V, Jaffe ES, Merino MJ et al (1993) Malignant lym-
germline MLH1 mutated lynch syndrome: consequence or coinci- phomas involving the ovary. A clinicopathologic analysis of 39
dence? Virchows Arch 470:347–352 cases. Am J Surg Pathol 17:154–170
149. Karateke A, Tuğ N, Sahin D (2011) Metastatic ovarian malignant 170. Chorlton I, Norris HJ, King FM (1974) Malignant reticuloendo-
melanoma with no obvious primary. J Turk Ger Gynecol Assoc thelial disease involving the ovary as a primary manifestation:
12:181–182 a series of 19 lymphomas and 1 granulocytic sarcoma. Cancer
150. Prieto VG. Immunohistology of melanocytic neoplasms (2014) In 34:397–407
Dabbs DJ (ed) Diagnostic Immunohistochemistry:Theranostic and 171. Fox H, Langley FA, Govan AD et al (1988) Malignant lymphoma
Genomic Applications. 4th edn. Saunders Elsevier, Philadelphia presenting as an ovarian tumour: a clinicopathological analysis of
151. McCluggage WG, Bissonnette JP, Young RH et al (2006) Primary 34 cases. Br J Obstet Gynaecol. 5:386–390
malignant melanoma of the ovary: a report of 9 definite or prob- 172. Emery JD, Kennedy AW, Tubbs RR et al (1999) Plasmacytoma
able cases with emphasis on their morphologic diversity and of the ovary: a case report and literature review. Gynecol Oncol
mimicry of other primary and secondary ovarian neoplasms. Int J 73:151–154
Gynecol Pathol 25:321–329 173. Mondal SK, Chatterjee S, Mandal S et al (2015) Primary extra-
152. Zikry J, Korta DZ, Chapman LW et al (2017) Melanoma arising medullary plasmacytoma of ovary: Report of a rare neoplasm. J
in an ovarian cystic teratoma: a systematic review of presentation, Cancer Res Ther 11:923–4
treatment, and outcomes. Arch Gynecol Obstet 296:397–404 174. Zhong YP, Zhang JJ, Huang XN (2012) Multiple myeloma
153. Liang L, Huang H, Dadhania V et al (2016) Renal cell carcinoma with rupture of ovarian plasmacytoma. Chin Med J (Engl)
metastatic to the ovary or fallopian tube: a clinicopathological 125:2948–2950
study of 9 cases. Hum Pathol 51:96–102 175. Hirsch MS (2011) Metastatic Tumors of the Ovary. In: Crum
154. Young RH, Hart WR (1992) Renal cell carcinoma metastatic to CP, Nucci MR, Lee KR (eds) Diagnostic Gynecologic and
the ovary: a report of three cases emphasizing possible confusion Obstetric Pathology, 2nd edn. Elsevier Saunders, Philadelphia,
with ovarian clear cell carcinoma. Int J Gynecol Pathol 11:96–104 p 972–988
155. Fadare O, Desouki MM, Gwin K et al (2017) Clear Cell 176. Oliva E, Ferry JA, Young RH et al (1997) Granulocytic sarcoma of
Renal Cell Carcinoma Metastatic to the Gynecologic Tract: A the female genital tract: a clinicopathologic study of 11 cases. Am
Clinicopathologic Analysis of 17 Cases. Int J Gynecol Pathol J Surg Pathol 21:1156–1165
37:525–535 177. Barcos M, Lane W, Gomez GA et al (1987) An autopsy study
156. Kassouf W, Binsaleh S, Cohen DD et al (2004) Bellini duct carci- of 1206 acute and chronic leukemias (1958 to 1982). Cancer
noma with ovarian metastasis. Can J Urol. 11:2461–2462 60:827–837
157. Netto GJ, Epstein JI (2014) Immunohistology of the bladder, kid- 178. Pais RC, Kim TH, Zwiren GT et al (1991) Ovarian tumors in
ney and testis. In: Dabbs DJ (ed) Diagnostic Immunohistochem relapsing acute lymphoblastic leukemia: a review of 23 cases. J
istry:Theranostic and Genomic Applications. 4th edn. Saunders Pediatr Surg 26:70–74
Elsevier, Philadelphia 179. Cunningham I (2013) The clinical behavior of 124 leukemic
158. Irving JA, Vasques DR, McGuinness TB et al (2006) Krukenberg ovarian tumors: clues for improving the poor prognosis. Leuk
tumor of renal pelvic origin: report of a case with selected com- Lymphoma 54:1430–6
366 K. Carrick and W. Zheng
180. Terek MC, Ozkinay E, Zekioglu O et al (2003) Acute leukemia in 186. Longacre TA, O'Hanlan K, Hendrickson MR (1996) Adenoid cys-
pregnancy with ovarian metastasis: a case report and review of the tic carcinoma of the submandibular gland with symptomatic ovar-
literature. Int J Gynecol Cancer 13:904–8 ian metastases. Int J Gynecol Pathol 15:349–55
181. Terek MC, Ozkinay E, Zekioglu O et al (2003) Acute leukemia in 187. Riopel MA, Ronnett BM, Kurman RJ (1999) Evaluation of diag-
pregnancy with ovarian metastasis: a case report and review of the nostic criteria and behavior of ovarian intestinal-type mucinous
literature. Int J Gynecol Cancer 13:904–8 tumors: atypical proliferative (borderline) tumors and intraepi-
182. Young RH, Jackson A, Wells M (1994) Ovarian metastasis
thelial, microinvasive, invasive, and metastatic carcinomas. Am J
from thyroid carcinoma 12 years after partial thyroidectomy Surg Pathol 23:617–35
mimicking struma ovarii: report of a case. Int J Gynecol Pathol 188. Eichhorn JH, Young RH, Scully RE (1993) Nonpulmonary small
13:181–185 cell carcinomas of extragenital origin metastatic to the ovary.
183. Logani S, Baloch ZW, Snyder PJ et al (2001) Cystic ovar- Cancer 71:177–186
ian metastasis from papillary thyroid carcinoma: a case report. 189. Açıkalın A, Paydaş S, Güleç ÜK et al (2014) A unique case of merkel
Thyroid 11:1073–1075 cell carcinoma with ovarian metastasis. Balkan Med J 31:356–359
184. Brogioni S, Viacava P, Tomisti L et al (2007) A special case of 190. Zukerberg LR, Young RH (1990) Chordoma metastatic to the
bilateral ovarian metastases in a woman with papillary carcinoma ovary. Arch Pathol Lab Med 114:208–210
of the thyroid. Exp Clin Endocrinol Diabetes 115:397–400 191. Rubio A, Schuldt M, Chamorro C et al (2015) Ovarian Small
185. Besic N, Gazic B (2013) Sites of metastases of anaplastic thyroid Cell Carcinoma of Pulmonary Type Arising in Mature Cystic
carcinoma: autopsy findings in 45 cases from a single institution. Teratomas With Metastases to the Contralateral Ovary. Int J Surg
Thyroid 23:709–713 Pathol 23:388–92
Peritoneum and Broad Ligament
11
M. Ruhul Quddus, Sharon Liang, Wenxin Zheng,
and C. James Sung
© Science Press & Springer Nature Singapore Pte Ltd. 2019 367
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_11
368 M. Ruhul Quddus et al.
Fig. 11.2 Ultrastructural image of mesothelial cells showing long 11.1.2.2 Granulomatous Peritonitis
slender microvilli on the surface of the cells. This gives the typical Many infectious and noninfectious agents may inflict a gran-
“window” effect around the mesothelial cells under light microscope ulomatous response in peritoneum. Presence of diffuse
11 Peritoneum and Broad Ligament 369
g ranulomata in the peritoneum may be misinterpreted as dis- also be encountered. Sometimes histiocytic proliferation
seminated carcinomatosis. Histological examination is may occur around necrotic nodules, known as necrotic
required to confirm the diagnosis. pseudo-xanthomatous tubercles, with or without pigmenta-
Mycobacterial infection is the most common infection in tion. Pigmented histiocytic proliferation is rarely seen, and
the peritoneum (Fig. 11.5), showing an increase in incidence typically is secondary to deposition of melanin pigments, a
in recent years, especially in immunocompromised patients condition known as peritoneal melanosis, usually associated
[4]. The mode of transmission could be via systemic dissemi- with ruptured mature cystic teratoma [6]. These pigment-
nation or direct spread of intraperitoneal infection. Caseating laden histiocytes lack organelles, a differentiating point from
necrosis, epithelioid cells, and Langhans-type giant cells are malignant melanoma.
characteristics of this lesion. Acid-fast stain or immunofluo-
rescence stains may detect mycobacterium; however, they 11.1.2.4 Peritoneal Fibrosis and Adhesions
may also be negative. Molecular testing for mycobacterium, Peritoneal fibrous adhesions and reactive fibrosis often occur
typically via polymerase chain reaction (PCR), can help to after inflammation and abdominal surgery. Severe adhesions
resolve difficult cases. Rare cases of granulomatous peritoni- may result in intestinal obstruction. Patients with cirrhosis,
tis may also be caused by fungal infection. pulmonary tuberculosis, or asbestos exposure may develop
Noninfectious granulomata are caused by non-pathogens localized transparent plaques on the surface of the liver or
such as surgical glove talcum powders, suture materials, con- spleen. Diffuse fibrosis in the abdomen is known as sclerotic
trast medium, and other iatrogenic agents (Fig. 11.6). Fecal or fibrotic peritonitis (Fig. 11.7). This condition is typically
contamination, pancreatic enzymes, digestive secretions, and idiopathic, although occasionally asbestos has been impli-
bile resulting from intestinal perforation or ruptured neo- cated. Sarcoidosis, abdominal peritoneal dialysis, and use of
plasm, such as dermoid cyst or ovarian tumors, may induce beta-anti-adrenergic agents may also be contributory [7].
noninfectious granulomas. Keratin-induced foreign body Sclerosing peritonitis is a rare, often fatal condition of perito-
granulomas may be seen in endometrial or ovarian carcino- neal inflammation secondary to peritoneal dialysis. It may
mas with squamous differentiation, as well as teratomas with follow renal or liver transplantation. The patients are typically
keratin formation. It has been reported that the presence of young females of reproductive age, presenting with abdomi-
keratin granulomata without viable tumor cells does not nal pain, abdominal distention, ascites, intestinal obstruction,
affect prognosis in patients with endometrioid carcinoma [4, and a pelvic tumor. The condition may also be associated
5]. Sarcoidosis and Crohn’s disease may rarely cause granu- with anticonvulsant treatment [8, 9]. Sclerotic peritonitis is
lomatous peritonitis. thought to be a reactive process and it may be confused with
desmoplastic type mesothelioma of the peritoneum.
11.1.2.3 Non-granulomatous Histiocytic
Lesions 11.1.2.5 Rare Subtypes of Peritonitis
Endometriosis can trigger histiocytic response in the perito- Other rare types of peritonitis include eosinophilic peritoni-
neum and omentum. Decidualization, endosalpingiosis, and tis and peritonitis secondary to collagen vascular diseases,
endocervicosis (endocervical mucinous-type epithelia) may including systemic lupus erythematosus.
370 M. Ruhul Quddus et al.
a b
Fig. 11.8 Mesothelial hyperplasia, various morphologic types. (a) Mesothelial cells proliferate to become stratified or papillary. (b). Small tubu-
lar growth. (H&E 400) (c) Marked papillary growth (H&E, 100×)
11 Peritoneum and Broad Ligament 371
a b
Fig. 11.9 Reactive mesothelial cells may appear sheetlike (a) but the cells appear monotonous or pleomorphic with abundant cytoplasm (H&E,
200×) and (b) without nuclear atypia (H&E, 400×)
The morphologic appearance of reactive mesothelial cells Table 11.1 Differences between marked mesothelial hyperplasia ver-
sus diffuse malignant mesothelioma
varies depending upon the degree of proliferation and com-
plexity of architectural patterns. In mild hyperplasia, the Marked mesothelial Diffuse malignant
hyperplasia mesothelioma
nuclei are small, regular, round, or oval, with centrally placed
Clinical presentations No previous history of Previous history
nucleoli. The cytoplasm may be eosinophilic and vacuolated. asbestos exposure of asbestos
In more severe cases, mesothelial cells may be multinucle- exposure
ated with enlarged nuclei, abundant cytoplasm, and promi- Gross findings Not grossly visible Grossly visible,
nent nucleoli, mimicking adenocarcinoma; however, cells of necrotic
Cytologic morphology Monotonous Pleomorphic
mesothelial hyperplasia are typically uniform in size with
population of small or
abundant cytoplasm. Conversely, adenocarcinoma cells usu- enlarged, distinct cell
ally have pleomorphism, much larger nuclei, increased N/C border
ratios, and frequent mitotic figures. Cytoplasm Abundant eosinophilic Prominent
Florid mesothelial hyperplasia must be differentiated cytoplasm, with vacuolation
occasional vacuoles
from diffuse malignant mesothelioma (DMM). The differ-
Tissue infiltration None or superficial Deeply
ences between the two scenarios are summarized in pseudoinvasion infiltrative
Table 11.1. The presence of grossly visible nodules, necro- Immunohistochemical CK +, vimentin +, p53 +, EMA +
sis, vacuolated cytoplasm, nuclear pleomorphism, and deep staining desmin +
tissue infiltration supports the diagnosis of diffuse malignant
mesothelioma (DMM). Positive immunostaining with p53
and EMA also supports the diagnosis of diffuse malignant 11.1.3.2 Peritoneal Inclusion Cyst
mesothelioma (DMM) [10, 11]. Peritoneal inclusion cysts may have a variety of clinical presenta-
Mesothelial hyperplasia must also be differentiated from tions. It is usually seen in females of reproductive age, often with
serous borderline tumor of the peritoneum or ovary. The dif- a prior history of abdominopelvic surgery, inflammation, or
ferences of ovarian or peritoneal tumors are usually quite endometriosis, which may indicate that inflammation is causal in
apparent. The presence of columnar-type tumor cells, absent this entity [12]. In most cases, peritoneal inclusion cysts are inci-
microvilli (the so-called window around mesothelial cells), dental findings during intra-abdominal surgery. Some patients
crack artifacts around tumor cell nests, seromucinous pro- may present with lower abdominal pain or a pelvic mass. Some
teinaceous materials, and abundant psammoma bodies may evidence suggests that hormonal changes may affect the disease
support a diagnosis of serous tumor. Characteristics of serous process. Therapeutic options include observation or surgical
tumor will be discussed later in this chapter and in other resection. Most cases are benign although recurrences may occur
chapters. in 50% of the cases. Surgery is generally curative.
372 M. Ruhul Quddus et al.
a b
Fig. 11.10 Multicystic peritoneal inclusion cyst is composed of multiple cysts separated by fibrous septae lined by (a) flat mesothelial cells
(H&E, 100×), with (b) may show occasional squamous metaplasia (H&E, 200×)
Multicystic peritoneal inclusion cysts (MPIC) have also 11.1.3.3 Calcifying Fibrous Pseudotumor
been described as benign cystic mesothelioma, cystic peri- This rare benign soft-tissue lesion is defined as a paucicel-
tonitis, or postsurgical peritoneal inclusions. Use of the term lular collagenous reaction associated with inflammation and
“mesothelioma” often creates unnecessary anxiety among dystrophic calcifications. It possibly represents a reactive
the clinicians and patients, so many have elected to avoid process and often presents as an incidental finding in the vis-
this term in clinical practice. The lesion is composed of mul- ceral peritoneum of the small intestine and the stomach, typi-
tiple thin-walled cysts forming grapelike clusters attached cally in younger patients below 20 years of age. Grossly the
to peritoneal surfaces. The cysts may range from several lesion presents as a well-defined solid tumor up to 20 cm.
millimeters up to 20 cm, containing clear serous fluid.
The cut surfaces are gritty (or sandy). Microscopically dense
Microscopically the cysts are separated by fibrous stroma collagen fibers are arranged in a concentric pattern and
lined by single- or multilayered flat or cuboidal mesothelial admixed with benign-appearing fibroblasts, lymphocytes,
cells with benign-appearing nuclei or slight nuclear pleo- and plasma cells. Focal psammoma bodies or dystrophic cal-
morphism (Fig. 11.10a, b). Focal mesothelial hyperplasia, cifications are present. It can undergo torsion and present as
adenomatous changes, or squamous differentiation may be acute peritonitis [13]. It has been suggested that the lesion
encountered. may be associated with inflammatory myofibroblastic tumor
The differential diagnoses of multicystic peritoneal inclu- [14, 15] or sclerotic phase of inflammatory myofibroblastic
sion cysts include malignant mesothelioma, cystic mesothe- tumor [16, 17] which are discussed later in the chapter.
lioma, cystic lymphangioma, and adenomatoid tumor.
Malignant mesothelioma presents as papillary or tubulo- 11.1.3.4 Splenosis
papillary neoplasm, containing tumor cells with nuclear In splenosis, multiple nodules of ectopic splenic tissue are
pleomorphism and marked proliferative activity. Multicystic present in the abdominal cavity. The condition typically
lymphangioma, another differential diagnosis of mesotheli- occurs as a result of autoimplantation of splenic tissue after
oma, is a congenital condition often seen in children, which traumatic rupture of the spleen and is typically asymptom-
typically occurs in the small intestinal mesentery, omentum, atic. Splenosis is generally an incidental finding many
mesentery of the sigmoid colon, and retroperitoneum. The months or even years after splenectomy during subsequent
cysts are characterized by spongiform spaces lined by endo- intra-abdominal surgery or autopsy. Some female patients
thelial cells. Smooth muscle cells and lymphocytes are often may present with abdominal pain, having symptoms similar
present in the cyst wall of cystic lymphangioma, but not in to those of endometriosis, adnexal tumor, or diffuse perito-
MPIC. Adenomatoid tumors typically are seen in the fallo- neal carcinomatosis [18–20]. The number of the nodules
pian tubes or uterine serosa with a classic gross and micro- may vary from case to case with some cases demonstrating
scopic appearance and generally lack marked inflammatory over 100. The nodules vary in size from several millimeters
infiltrate. The differential diagnosis of multicystic peritoneal to centimeters. The nodules are typically composed of red
inclusion cysts from cystic mesothelioma is discussed later pulp surrounded by a dense collagenous pseudocapsule
in the chapter. devoid of smooth muscle or elastic fibers. Larger splenic
11 Peritoneum and Broad Ligament 373
11.1.4 Mesothelioma
Adenomatoid Tumor
Adenomatoid tumor is a benign mesothelial derived tumor
often seen near the oviducts in females and epididymis in
Fig. 11.11 Focal decidual stromal reaction with rare trophoblastic
cells in peritoneum (patient had a recent history of ectopic pregnancy) males, uterine serosa, broad ligament, and occasionally near
(H&E, 100×) the ovarian hilum and ovary.
374 M. Ruhul Quddus et al.
ally demonstrate more nuclear pleomorphism when com- exposure remains controversial. One recent study reported
pared to those of a well-differentiated papillary that 29% of 96 male patients had occupational asbestos
mesothelioma; however they may be difficult to distin- exposure while none of the 113 female patients had any
guish in some cases. Scattered ciliated cells in the lining history of asbestos exposure [37]. Some authors also noted
epithelium can help to identify a serous tumor if present. that mesothelioma in female patients may be related to
Immunohistochemical characteristics of serous border- radiation, chronic inflammation, organic chemicals, or non-
line tumor are discussed later in the chapter. asbestos fibers. Approximately 40% of female patients sur-
4. Treatment and prognosis: The treatment of choice for vived more than 4 years, much more than their male
well-differentiated papillary mesothelioma is surgery; counterparts [38].
however, this may not be curative. This tumor typically
presents with an indolent clinical course, with many 1. Clinical characteristics: The clinical presentation of dif-
patients surviving several years or even several fuse malignant mesothelioma is similar to that of ovarian
decades. or other peritoneal malignancies. Patients typically report
abdominal discomfort, abdominal distention, indigestion,
Low-Grade Cystic Mesothelioma and weight loss. Many patients present with ascites.
Most cystic mesothelial lesions appear as a reactive pro- Occasional cases may present with retroperitoneal, intes-
cess, e.g., peritoneal inclusion cysts, but rare cases may tinal, or pelvic tumors; intussusception; deep umbilical
represent a true neoplastic process such as low-grade cystic subcutaneous nodules; or even cervical or inguinal
mesothelioma (Fig. 11.16a) [34]. Unlike peritoneal inclu- lymphadenopathy. The tumor may be confused with a
sion cysts, cystic mesothelioma may be lined by atypical disseminated ovarian tumor when both ovaries are
mesothelial cells. Focally the tumor may show a classic involved. Diagnosis requires a thorough histologic exam-
pattern similar to that seen in malignant mesothelioma. ination, although some cases are diagnosed by cytologic
Under low power, this tumor may be confused with an ade- examination of the ascites fluid [39].
nomatoid tumor (Fig. 11.16b). Cystic mesothelioma typi- 2. Gross examination: Both the visceral and parietal perito-
cally behaves in an indolent fashion, but may infrequently neum will show diffuse thickening with extensive tumor
recur after resection. nodules and plaques. The tumor may encase or infiltrate
visceral organs. In contrast to the behavior of malignant
Diffuse Malignant Mesothelioma (DMM) epithelial tumors, these tumors less frequently show local
Fewer cases of peritoneal diffuse malignant mesothelioma invasion and metastasis.
(DMM) are encountered compared to those occurring in the 3. Microscopic examination: Diffuse malignant mesotheli-
pleura. DMM in peritoneum composes 10–20% of all cases oma of the abdominal and pleural cavities has identical
of malignant mesothelioma and predominantly affects histological features. The tumor may show tubular or pap-
middle-aged or elderly males [35, 36]. The causal relation- illary or tubulo-papillary morphologic features
ship between diffuse malignant mesothelioma and asbestos (Fig. 11.17a, b). Some may present as solid sheets
a b
Fig. 11.15 Well-differentiated papillary mesothelioma. (a) Low-power view shows broad fibrovascular core lined by flat to cuboidal mesothelial
cells (H&E, 200×). (b) Higher power view shows that the mesothelial cells are rather uniform with no mitosis or prominent nucleoli (H&E, 400×)
376 M. Ruhul Quddus et al.
a b
Fig. 11.16 Low-grade cystic mesothelioma. (a) Gross examination reveals numerous thin-walled cysts. (b) Low-power view may be similar to
adenomatoid tumor (H&E, 100×)
(Fig. 11.17c) and others may appear granulomatous. The Most patients were alive and free of disease on follow-
tumor may occasionally show invasion of sub-peritoneal up between 18 months and 11 years [38].
tissue such as the omentum. Intra-abdominal lymph nodal 4.
Immunohistochemical characteristics: Immuno
involvement may also be identified in some instances. histochemistry is critical to the accurate diagnosis of
The cytology of the tumor cells may vary, but vague api- mesothelioma. Mesotheliomas are frequently positive for
cal cytoplasmic protrusions are commonly seen as well as specific tumor markers, e.g., calretinin, WT-1, CK 5/6,
moderate-to-severe nuclear pleomorphism. Although and mesothelin (Table 11.2). Rarely, rather less specific
there is more nuclear pleomorphism and mitotic activity markers, e.g., D2-40 and h-caldesmon, are positive.
than in benign mesothelial tumors, the nuclear pleomor- Similar to serous tumors, malignant mesothelial cells are
phism and frequency of mitotic figures are much less pro- often positive for pancytokeratins, EMA, CK7, and
nounced than those seen in low-grade serous CA125. In contrast to serous tumors, malignant mesothe-
adenocarcinoma (Fig. 11.18). lial cells do not express CEA, LeuM1, B72.3, Ber-EP4,
Compared to pleural malignant mesothelioma, abdom- ER, and PR. However, some ovarian or peritoneal serous
inal malignant mesothelioma may have the following tumors may not express CEA and LeuM1. Therefore
unusual morphological characteristics: staining with B72.3, Ber-EP4, and other more specific
(a) Decidualization (Fig. 11.19) characterized by solid markers may be a better panel to be used in distinguishing
sheets of polygonal eosinophilic tumor cells with these two entities [41, 42].
abundant cytoplasm, well-demarcated cell borders, 5. Ultrastructure: Mesothelial cells are characterized by
and prominent nucleoli [40]. Two-thirds of this sub- long slender microvilli whereas in adenocarcinoma the
type occurs in females, even as early as in adoles- tumor cells demonstrate short, stout microvilli. Malignant
cence. This rare subtype is associated with a high mesothelial cells may contain cytoplasmic glycogen but
mortality rate. not mucin. Intermediate filaments may be present near
(b) Biphasic differentiation, which is uncommon. the nuclei.
(c) Rare tumors may be associated with marked inflam-
6. Differential diagnosis: Differentiating diffuse malignant
matory infiltrate (Fig. 11.20), often with lymphoid mesothelioma from atypical mesothelial hyperplasia, diffuse
follicles. Others may be entirely replaced by fibrous fibroblastic malignant mesothelioma, and reactive fibrosis is
tissue, resulting in diagnostic challenge. already briefly discussed previously in this chapter.
(d) Localized malignant mesothelioma: These tumors Differentiating diffuse malignant mesothelioma from
should be distinguished from diffuse malignant meso- pelvic high-grade serous adenocarcinoma is the most
thelioma because of their localized presentation and important, yet most difficult, distinction to be made. The
better prognosis. These are localized circumscribed characteristics of these two tumors are summarized in
lesions of the serosal membranes having similar micro- Table 11.2. Features supporting a diagnosis of mesotheli-
scopic appearance to that of diffuse variant. The median oma include prominent tubulo-papillary structures, follicu-
age is 63 with male preponderance (male/female 2:1). lar tumor cells, moderate amount of eosinophilic cytoplasm,
11 Peritoneum and Broad Ligament 377
a b
Fig. 11.17 Diffuse malignant mesothelioma. (a–b) Papillary and tubular structures may be present (H&E, 200×). (c) Diffuse solid sheet pattern
(H&E, 200×)
abdominal desmoplastic round cell tumors react to 4. Differential diagnoses: DSRCT should be differentiated
WT1 (C-terminal) in contrast to serous carcinomas of from several other small round blue cell neoplasms such
the ovaries/fallopian tubes and mesotheliomas which as neuroblastoma, extra-skeletal Ewing sarcoma, primi-
typically react to N-terminal of WT1 [47]. The com- tive neuroectodermal tumor, immature teratoma, lym-
bined immunohistochemical characteristics are helpful phoma, rhabdomyosarcoma, rhabdoid tumor, poorly
in differentiating this tumor from other diagnostic differentiated stromal tumors, adult-type or juvenile gran-
considerations. ulosa cell tumor, and small-cell carcinoma associated
3. Genetics: This tumor frequently contains a translocation with hypercalcemia. In general, the age of the patient,
of t(11;22)(p13;q12) resulting in fusion of EWS1 gene on lack of extra-abdominal primary tumor, tumor distribu-
chromosome 22 with Wilms tumor-suppressor gene tion, and characteristic microscopic and immunohisto-
(WT1) on chromosome 11 [48, 49]. This fused EWS/WT1 chemical staining patterns help support a diagnosis of
may be detected using RT-PCR. This translocation is DSRCT. DSRCT reacts to C-terminal of the WT1 which
unique to this tumor and therefore its detection confirms is helpful to differentiate from some of the tumors dis-
the diagnosis [50]. cussed here as the others react with the N-terminal of
WT1. The EWS/WT1 gene fusion can be detected by RT-
PCR and the presence supports the diagnosis of
DSRCT. The presence of unusual features such as tubu-
lar, glandular, cystic, papillary, anastomosing trabecular,
and single-cell cord-like patterns may contribute to diag-
nostic difficulty. Chromosome analysis usually clinches
the diagnosis.
5. Treatment of prognosis: Patients with DSRCT require
chemotherapy and radiation therapy after surgical resec-
tion. Neoadjuvant chemotherapy has also been adopted.
Some patients may benefit from bone marrow transplan-
tation. Unfortunately more than 90% of patients die of the
disease despite aggressive therapy.
a b
Fig. 11.22 Intra-abdominal desmoplastic small round cell tumor (DSRCT). (a) Low-power view shows cluster of small round blue cells separated
by abundant fibrous stroma (H&E, 200×). (b) High-power view shows rather uniform neoplastic cells with scant cytoplasm and inconspicuous cell
border (H&E, 400×)
380 M. Ruhul Quddus et al.
a b
Fig. 11.24 Solitary fibrous tumor. (a) Nonspecific solitary fibrous tumor is composed of coarse collagen fibers accompanied by elongated slit-like
spaces. The tumor cells are elongated and inconspicuous (H&E, 200×). (b) Cellular type solitary fibrous tumor is composed of highly cellular
benign spindle cells (H&E, 400×). (c) CD34 immunostain positivity of solitary fibrous tumor (IHC, 100×)
11.1.6.1 Pseudomyxoma Peritonei (PMP) in goblet cells of gastrointestinal tract and not by ovarian
Secondary mucinous involvement of the peritoneum is known mucinous tumors. Some scholars have suggested that pseudo-
as pseudomyxoma peritonei (PMP), but the terminology is not myxoma peritonei does not arise from ruptured ovarian
specific as the term is used to describe several clinical scenar- tumors; on the contrary, the associated ovarian borderline
ios where the patient presents with mucinous ascites [61–63]. tumors are in fact secondary or metastatic lesions [69].
In the past most authors considered pseudomyxoma peritonei
to originate from the ovary. This was because more frequently 1. Clinical characteristics: The clinical symptoms of pseu-
than not, both ovaries were also involved, and demonstrated domyxoma peritonei are vague and nonspecific. They
histological features of mucinous borderline tumor. In recent typically occur in patients ranging from 33 to 82 years
years it has being shown that the majority of pseudomyxoma (mean 47 years) of age. Patients may present initially
peritonei originate from the gastrointestinal tract, most com- with abdominal distention, pain, or fullness depending on
monly from the vermiform appendix, while some originate the severity of disease. With disease progression, more
from the colorectum, or rarely from the pancreas and biliary mucin accumulates, increasing abdominal distention,
tract [64, 65]. A few cases of pseudomyxoma peritonei have which can be accompanied by a mechanical or functional
been found to be associated with ovarian mucinous borderline ileus. On rare occasions patients may incidentally be
tumors associated with ovarian mature cystic teratomas [66– diagnosed during surgical procedures or hernia repair.
68]. Genetic studies have revealed that most pseudomyxoma 2. Pathologic findings: A low-grade mucinous neoplasm of
peritonei express the MUC2 protein, which is only expressed the vermiform appendix is the leading source of second-
382 M. Ruhul Quddus et al.
Fig. 11.25 Appendiceal mucinous neoplasm. Low-power view show- Fig. 11.26 Abundant acellular mucin pool dissecting peritoneum par-
ing muscular wall of the appendix infiltrated by the mucinous tumor tially surrounded by reactive mesothelial cells (H&E, 200×)
(H&E, 100×)
a b
Fig. 11.27 Low-grade mucinous neoplasm resulting in pseudomyxoma peritonei. (a) A few neoplastic cells are floating in the mucin pool (H&E,
400×); (b) occasionally intestinal type mucinous epithelium is recognizable (H&E, 400×)
ary peritoneal mucinous tumors (Fig. 11.25). The primary similar to those of a hyperplastic polyp or low-grade
appendiceal mucinous lesions may range from mucinous mucinous neoplasm. Some authors prefer the term “acel-
adenomas to low-grade appendiceal mucinous neoplasms lular mucin” or mucinous epithelium consistent with
to frankly invasive adenocarcinoma [70]. Other sources pseudomyxoma peritonei. The peritoneal lesion may be
of secondary peritoneal mucinous tumor are usually asso- accompanied by conspicuous fibroblastic reaction and
ciated with high-grade colorectal adenocarcinoma with reactive mesothelial cells. The ovaries may also show
destructive invasion. abundant acellular mucin with or without free-floating
Morphology of the epithelial component of PMP dif- tumor cells dissecting the ovarian parenchyma
fers and such morphologic differences determine the clin- (Fig. 11.27a). Intestinal type epithelium is usually evident
ical course and prognosis. Low-grade tumors usually (Fig. 11.27b). Various designations have been used in the
demonstrate acellular mucin or scant, bland, mucin- literature describing this phenomenon, e.g., well-
containing epithelial cells floating in mucinous fluid differentiated mucinous adenocarcinoma, disseminated
(Fig. 11.26). The epithelial cells, when present, may be peritoneal mucinous adenosis, and secondary low-grade
11 Peritoneum and Broad Ligament 383
a b
Fig. 11.28 High-grade mucinous neoplasm of the gastrointestinal tract secondarily involving the peritoneum. (a) Mucin and neoplastic cells
(mucicarmine, 200×). (b) Higher power view showing prominent nucleoli of the tumor cells (mucicarmine, 400×)
peritoneal mucinous neoplasm. The current authors favor 4. Immunohistochemical characteristics: Immunohis
the designation of low-grade peritoneal mucinous tochemical staining characteristics of most peritoneal sec-
neoplasm [70]. ondary mucinous neoplasms are similar to those of the
High-grade mucinous neoplasms usually show more appendix, expressing a CK7−/CK20+ phenotype, as
epithelial cells and moderate to marked nuclear pleomor- opposed to the CK7+/CK20− phenotype seen in primary
phism and prominent nucleoli (Fig. 11.28a, b). Signet- ovarian tumors. This supports the recent evidence that most
ring cells, when present, should alert the pathologists to secondary peritoneal mucinous tumors originate from the
search for gastric carcinoma or invasive lobular carci- appendix. It is important to recognize that occasional appen-
noma of the breast. dicular tumors may be positive for both CK7 and CK20 in
It is noteworthy that peritoneal secondary mucinous varying amounts. The positive staining should be diffuse
tumors often involve bilateral ovaries, and can have strik- and strong if to be supportive in differentiating one site from
ing similarities to primary ovarian mucinous cystadeno- the other; in some cases staining will be equivocal.
mas, intestinal type mucinous borderline tumors, or 5. Pathology report: Pseudomyxoma peritonei is a clinical
primary ovarian mucinous adenocarcinoma [71]. Cross term and therefore should not be used as a diagnosis in a
sections of the tumor may reveal numerous cysts of varying pathology report. The pathology report should consider
sizes containing abundant mucus. The ovarian capsule may (a) the nature of the appendiceal or ovarian tumor (benign,
be involved, and therefore must be differentiated from borderline, or malignant), and whether the primary lesion
mechanical involvement of acellular mucin associated with was ruptured; (b) the nature of the intra-abdominal dis-
a ruptured ovarian primary mucinous tumor. It is important ease, classified as acellular mucinous ascites, cellular
to note that some mucinous tumors metastatic to the ova- mucin, organizing mucin, or septated (lobulated) mucin;
ries may undergo “reverse maturation” causing their epi- and (c) the grade of the epithelium if present in the mucin
thelium to appear similar to that seen in a mucinous (benign, borderline, or malignant). Pathology reports that
cystadenoma; therefore, the determination of primary ver- explicitly characterize these components help clinicians
sus secondary ovarian mucinous lesions should not be plan appropriate management of these cases.
based on the identification of “precursor” epithelium. 6. Treatment and prognosis: The clinical course of a peri-
3. Sampling: Proper and adequate sampling of peritoneal toneal secondary mucinous tumor is determined by the
secondary mucinous neoplasms is paramount. All lesions presence or absence of neoplastic cells within the
should be extensively sampled, including grossly normal mucin, their grade, and the mucin volume [72]. Intra-
vermiform appendix. The vermiform appendix should be abdominal ascites with acellular mucin is associated
entirely submitted for microscopic examination to deter- with better prognosis. The clinical course may be indo-
mine the nature of any epithelial lesion, and to establish lent if the mucinous epithelial cells appear benign or
primary source of the tumor. Immunohistochemistry and mildly atypical. Patients whose mucinous ascites is
molecular characterization may be needed to determine associated with mucinous adenocarcinoma may be
the origin of the tumor. associated with lymph node or visceral metastasis.
384 M. Ruhul Quddus et al.
a c
Fig. 11.29 Gliomatosis peritonei. (a) Numerous small gray-white (H&E, 40×). Inset: Higher power view on the right (H&E, 100×, 200×).
tubercles of various sizes and shapes in the omentum (gross photo- (C) positive staining for GFAP (IHC, 40×)
graph). (b) The small tubercles represent typical neurofibrillary tissue
Fig. 11.31 (a) Implanted leiomyoma following laparoscopic hysterec- abdomen mimicking ruptured ectopic pregnancy. Gross and micros-
tomy with morcellation (H&E, 20×). (b) Infarcted leiomyoma. Had copy (H&E, 20×). Positive smooth muscle actin (SMA) immunostain
morcellation 9 years ago, Now pregnant, and presented with acute (IHC, 20×)
when Peik et al. reported dysplastic changes in prophylacti- larger than the involvement of the surface of either ovary, and
cally removed fallopian tubes of women predisposed to total submission and examination of both fallopian tubes do
developing ovarian carcinoma [82]. Crum et al. called atten- not reveal any in situ or invasive carcinoma [84].
tion to the distal fallopian tube and proposed that migrating Investigators have recently used advanced molecular
tumor cells from the fallopian tube are an important source of techniques to characterize the relationship between serous
serous carcinoma in the omentum, rather than arising in the tubal intraepithelial carcinoma (STIC) and high-grade serous
peritoneum [83]. As a result of this theory, it has now been carcinoma (HGSC) in detail. They have found similarities
proposed that primary peritoneal serous carcinoma should between STIC and HGSC which may support the fallopian
only be diagnosed when ovarian parenchymal involvement is tube origin theory. Perhaps more interestingly it was also dis-
less than 5 × 5 mm, involvement of extra-ovarian sites are covered that not all morphologically bona fide STICs are of
11 Peritoneum and Broad Ligament 387
tubal origin [85]. The fallopian tube mucosa can also be a like lesion (Fig. 11.32b). The clinical and pathological char-
receptacle for metastatic carcinomas arising elsewhere, and acteristics as well as the pathogenesis of endometriosis are
the authors cautioned pathologists to be extremely careful in discussed in detail in Chap. 12.
diagnosing STIC without first considering the possibility of
metastasis. Routinely used immunohistochemical stains can
be used to determine if a STIC-like lesion is tubal or non- 11.2.2 Serous (Tubal Type Epithelium) Lesions
gynecologic in origin. PAX 8, WT1, CK7, ER-negative
lesions mimicking STIC may represent metastatic tumors of Tubal type, or serous, epithelial lesions of the peritoneum
non-gynecologic origin. CDX2 positivity will point towards include endosalpingiosis and serous neoplasms. Many con-
a metastatic intestinal lesion. A word of caution here is that troversies abound regarding the pathogenesis, interaction,
positive p53 immunoreactivity should not be considered and even definition of these lesions among the gynecologic
diagnostic of STIC as metastatic breast and colon cancers pathology community. Primary serous adenocarcinoma of
may be p53 positive [86]. In the context of uterine and non- the peritoneum is defined as a serous adenocarcinoma involv-
uterine HGSC, it may be argued that STIC may represent a ing primarily the peritoneum, with relatively uninvolved or
metastasis rather than the site of origin, particularly when only superficially involved normal ovaries. A set of criteria
widespread disease is present. have recently been proposed as described earlier in this chap-
Although assigning a primary site of tumor has limited ter [84]; however, using the size and distribution of tumor as
prognostic and therapeutic relevance in the case of a serous absolute criteria to determine the origin of the disease was
carcinoma with widespread pelvic and omental/peritoneal
involvement, a recent attempt has been made to achieve that
goal based on a survey of pathologists and clinicians [87]. a
It appears, at least as of now, that the origin of peritoneal
serous carcinoma is still evolving. The validity of the sec-
ondary Müllerian system theory cannot be disregarded com-
pletely. The secondary Müllerian system, in theory, offers
explanation for the pathogenesis of all types of Müllerian
epithelia, e.g., serous, endometrioid, mucinous, and transi-
tional, whereas the proposed fallopian tube origin theory can
only explain the origin of serous-type epithelium in perito-
neum. As late Stuart C. Lauchlan once stated “Far from
being a mere receptacle for metastasis from other sites, the
omentum is an important source of primary peritoneal (sec-
ondary Müllerian system) carcinoma” [79].
As a result of the above discussion, it has been proposed
that serous Müllerian tumors be designated as “pelvic serous
carcinoma,” rather than arbitrarily assigned to a specific pel- b
vic organ such as ovary, fallopian tube, or peritoneum.
11.2.1 Endometriosis
disputed by some workers as they have suggested that most rectal pelvic peritoneum. It is typically grossly invisible,
high-grade serous adenocarcinomas may derive from early although some may appear as multiple small (1–2 mm)
serous lesions of the fimbria and become widely dissemi- white or yellow spots. Alternatively, the lesion may appear
nated in the peritoneum. Low-grade serous adenocarcinoma opaque or as semitranslucent cysts or granules.
is felt to arise possibly from implants of papillary carcinoma 2. Microscopic examination: Endosalpingiosis may appear as
of the ovaries (serous borderline tumors of the ovaries, or rounded, oval, or cystically dilated glands with no stroma.
even micropapillary serous borderline tumors). Primary The outline of the focus may be somewhat irregular with
serous adenocarcinoma of the peritoneum may be rare, but is slightly crowded glandular epithelium, occasionally with
certainly not nonexistent. intraluminal papillary protrusions. The glands are lined by
a single layer of tubal type epithelium (Fig. 11.33a), com-
11.2.2.1 Endosalpingiosis posed of all three epithelial cells of the normal fallopian
Endosalpingiosis is the presence of benign tubal type epithe- tube: pale ciliated cells, secretory cells, and slightly darker
lium in the peritoneum, sub-peritoneal tissue, or retroperito- peg cells, typically with no cytologic atypia or mitosis. The
neal lymph nodes; it is more frequently encountered in the glands are surrounded by dense connective tissue but not
peritoneum compared to the pelvis [79]. The pathogenesis of endometrial stroma. Occasional mononuclear inflamma-
endosalpingiosis remains unknown. It was thought to arise tory cells may be present. Psammoma bodies may be pres-
from implanted desquamated tubal epithelium from ent within the lumen or nearby stroma. Epithelial tufting,
chronically inflamed fallopian tubes, similar to the implants or more complex epithelial growth, may be seen and some
arising from serous borderline tumor. An alternate explanation authors have designated these lesions “proliferative endo-
was that endosalpingiosis is a metaplastic process of the sec- salpingiosis” [79, 80].
ondary Müllerian system. Given the recently proposed theory Rarely endosalpingiosis may display cytologic atypia
on the role of early tubal lesions of the fimbria in the pathogen- described as “atypical endosalpingiosis.” The lining epi-
esis of serous tumors, the metaplastic theory of secondary thelium appears stratified with varying degrees of cyto-
Müllerian system has been revisited. To avoid confusion, some logic atypia but lacks the characteristics of a typical
investigators suggested using the term “serous change” instead serous borderline tumor. Atypical endosalpingiosis may
of endosalpingiosis. Endosalpingiosis is reported to be ten coexist with peritoneal serous borderline tumor.
times more frequent than endometriosis in the peritoneum, 3. Differential diagnosis: There is much controversy in dif-
whereas in the pelvis the ratio is reversed [79]. ferentiating atypical endosalpingiosis from peritoneal
serous borderline tumors (see below). The differences
1. Gross examination: The lesion is most commonly seen on between endosalpingiosis and serous borderline tumor
the serosa of the uterus, fallopian tubes, ovaries, and utero- are shown in Table 11.3.
a b
Fig. 11.33 Endosalpingiosis. (a) The glandular epithelial cells show of Müllerian duct remnant in a transgender receiving exogenous andro-
characteristic tubal “serous”-type epithelium with readily visible cilia. gen therapy (H&E, 600×)
Note the absence of endometrial stroma (H&E, 400×). (b) Virilization
11 Peritoneum and Broad Ligament 389
Another differential diagnosis consideration is mesonephric Fig. 11.34 Serous borderline tumor of the peritoneum. Morphologically
duct remnants. Mesonephric duct remnants typically identical to ovarian serous borderline tumor (H&E, 200×)
occur next to the fallopian tubes or within the broad liga-
ments. Mesonephric remnants are typically lined by a
single layer of cuboidal or low-columnar epithelium with- invasive implants. Noninvasive implants can further be
out cilia and are surrounded by layers of smooth muscle. subclassified as desmoplastic or non-desmoplastic (epi-
Virilization of the mesonephric remnants has recently thelial) implants. It may be difficult to differentiate des-
been described in female-to-male transgender patients moplastic noninvasive implants from invasive implants.
receiving testosterone therapy (Fig. 11.33b) [88]. The following features may be helpful: Noninvasive
implants are characterized by localized nondestructive
11.2.2.2 Peritoneal Serous Borderline Tumor fibroblastic growth. Clinically they appear as fibrous
Peritoneal serous borderline tumors are morphologically plaques the surgeon can easily strip from the underlying
identical to ovarian serous borderline tumors. Currently tissue. The relative ratio of epithelial cells to desmoplas-
many unresolved issues exist pertaining to its definition, eti- tic tissue is low, especially when compared to invasive
ology, and related pathology. It is usually associated with implants. Conversely, invasive implants are character-
infertility or chronic lower abdominal pain, typically in ized by epithelium-rich, disseminated, irregular, destruc-
women under 35 years. The tumor presents in the perito- tive growth equivalent to low-grade serous carcinoma.
neum, outside of the ovaries, with localized or disseminated, The typical lobular architecture of the omental fat is dis-
miliary or granular lesions accompanied by pelvic peritoneal turbed. Clinically it is important to differentiate noninva-
and omental fibrous adhesions. Peritoneal serous borderline sive implants from invasive implants because the
tumors may arise from peritoneal endosalpingiosis or prolif- treatments and prognosis are entirely different. Invasive
erative endosalpingiosis. implants have now been considered as low-grade serous
carcinoma. Pathological characteristics of these two
1. Microscopic examination: Peritoneal serous borderline lesions are summarized in Table 11.4. It is worth men-
tumors may present with papillary structures with cellular tioned that these two lesions may be present concurrently
stratification, small clusters of epithelial hyperplasia, and in the same patient (Figs. 11.35, 11.36, and 11.37).
interspersed loose cell clusters. Nuclear pleomorphism
and mitosis may be present but there is no true invasion Occasionally the epithelial lining of noninvasive implants
(Fig. 11.34). Morphologically it may be similar to nonin- may present with atypical features such as papillary, solid,
vasive, desmoplastic implants of ovarian serous border- cribriform, sievelike, or clustering patterns [90–92].
line tumor. Up to 85% of cases are associated with These growth patterns are generally considered features
endosalpingiosis. The survival rate is reported to be 95%, of malignancy, but most experts agree that if these abnormal
which is similar to that of noninvasive peritoneal implants features are localized and confined, their clinical significance
of ovarian serous borderline tumors [89]. may be similar to that of microinvasion of ovarian serous
2. Implants: Most peritoneal implants associated with
borderline tumor, with no significant adverse effects on
serous tumors arise from ovarian serous borderline prognosis [93, 94] (see Chap. 5 for additional morphological
tumor. They can be subclassified as noninvasive and characteristics and further discussion).
390 M. Ruhul Quddus et al.
Table 11.4 Noninvasive implants versus invasive implants strated that the presence of noninvasive implants has neg-
Noninvasive implants Invasive implants ative impact on long-term prognosis [96] with recurrence
Histology Well demarcated and Irregularly infiltrate visceral in 44%. Survival in this study ranged among 10% less
distinct from the peritoneum or omental than 5 years, 19% between 5 and 10 years, 10% between
underlying tissue adipose tissue with
destructive growth pattern
10 and 15 years, and 5% for more than 15 years. The pres-
Involve interstitial Involve interstitial spaces of ence of invasive implants (i.e., low-grade serous carci-
spaces of omentum omentum resulting in noma) and grossly visible residual disease are major
with intact omental destruction of normal adverse prognostic indicators. Studies have shown that
lobules, well adipose lobules; reactive the average time to recurrence may be as early as 2 years,
demarcated fibrosis surrounds adipose
cells and after recurrence many patients progress to low-grade
Nonsolid or cribriform Solid sheets or non- serous adenocarcinoma [97]. Only a minor fraction of
structures with small cribriform tumor growth patients respond to chemotherapy.
nests or isolated tumor extending beyond
cells surrounded by desmoplastic fibrosis;
small clefts prominent clefts around
11.2.2.3 Malignant Neoplasms
tumor cells and abundant
isolated tumor cells Low-Grade Peritoneal Serous Carcinoma (LGSC)
Cytology None to moderate Prominent nuclear The morphology of low-grade peritoneal serous carcinoma
nuclear pleomorphism, pleomorphism, abundant (LGSC), including psammocarcinoma, is similar to that of
rare mitoses mitoses, and atypical
mitoses invasive implants of serous borderline tumors, characterized
Ancillary Diploid Diploid or aneuploid by the presence of papillary growth pattern (Fig. 11.38) [98];
tests as a result invasive implants are now being considered as
Surgical Easy to strip off Difficult, may result in low-grade serous carcinoma. The cells in LGSC are gener-
stripping bleeding or perforation ally of low nuclear grade, lack frequent mitosis, have absent
to rare atypical mitosis, and lack lymphovascular space inva-
Sometimes it may be difficult to determine if an implant sion. Solid growth may be observed. The average age is
is associated with invasion or not. Biopsy tissue may be too between 50 and 60 years. Atypical clinical presentations
small and overtly calcified, or may have processing artifacts. include abdominal pain or mass-forming lesions. More than
In these instances, it may be essential to designate this lesion 40% of patients are diagnosed incidentally. The primary dif-
as “implant, invasion indeterminate.” ferences between low-grade peritoneal serous carcinoma and
peritoneal serous borderline tumor are the presence of stro-
3. Differential diagnosis: Compared with ovarian serous
mal invasion and confluent growth that may be seen in the
borderline tumors, there are both similarities and differ- former. On rare occasions progression of a low-grade perito-
ences. Some diagnostic dilemmas exist: (A) How to dif- neal serous carcinoma to a high-grade serous carcinoma may
ferentiate primary peritoneal serous borderline tumor occur [99].
from peritoneal implants associated with ovarian serous Psammocarcinoma is a subtype of low-grade peritoneal
borderline tumor, and more importantly whether there is serous carcinoma characterized by having psammoma bod-
any clinical significance? (B) Is the biological behavior of ies in more than 75% of the papillary growths or cell nests
peritoneal serous borderline tumor different from ovarian (Fig. 11.39) along with the presence of stromal invasion.
serous borderline tumor with peritoneal implants? (C) There may be slight nuclear atypia. Typically, the tumor cell
What is the definition of atypical endosalpingiosis? Does nests are composed of less than 15 cells. The patients are
it exist? And if so, does it give rise to peritoneal serous typically around 40 years old, presenting with abdominal
borderline tumor? (D) When accompanied by desmopla- pain or masses. Careful sampling of the excised specimen is
sia, how to differentiate desmoplastic implants of ovarian required to rule out the presence of concomitant high-grade
serous borderline tumor from primary peritoneal serous serous adenocarcinoma.
carcinoma? (E) How to differentiate invasive implants The prognosis of low-grade peritoneal serous carcinoma
from noninvasive implants especially when the lesions and peritoneal psammocarcinoma is fairly good. No patient
elicit conspicuous fibrous stromal reaction. deaths due to psammocarcinoma are recorded in the litera-
4. Prognosis: The prognosis of peritoneal serous borderline ture [100, 101]. The long-term prognosis remains unclear
tumor is largely determined by FIGO staging, the pres- due to limited case reports available.
ence or absence of invasive implants, and the quantity of
postoperative residual disease. Recent studies have shown High-Grade Peritoneal Serous Carcinoma (HGSC)
that 16% of patients with extra-ovarian disease suffered In the past, the incidence of primary peritoneal serous ade-
recurrence or died of disease [95]. Another study demon- nocarcinoma was one-tenth that of ovarian serous adeno-
11 Peritoneum and Broad Ligament 391
a b
c d
Fig. 11.35 Noninvasive implant. (a) Solitary implant is seen on the e pithelia (IHC, 40×). (e) Higher magnification shows prominent fibrous
surface of the peritoneum (H&E, 40×). (b) Typical features of serous tissue surrounding small clusters of tumor cells or single tumor cells
borderline tumor seen in higher power view (H&E, 200×). (c) without tissue invasion (H&E, 200×) (a, c, d, and e: courtesy of
Desmoplastic noninvasive implant (H&E, 40×). (d) Immunostaining W. Dwayne Lawrence, MD)
with pancytokeratin shows preponderance of stroma over actual
392 M. Ruhul Quddus et al.
a b
Fig. 11.36 Comparing noninvasive implant and invasive implant. (a) 100×). (b) Invasive implants with abundant tumor growths destroying
Noninvasive implant with papillary proliferations between fat lobules fat lobules of the omentum (H&E, 40×) (B: courtesy of W. Dwayne
in the omentum. No invasion into surrounding adipose tissue (H&E, Lawrence, MD)
a b
c d
Fig. 11.37 Invasive implant. (a) Invasive implant in the intestinal wall tumor cells resembling low-grade serous carcinomatous cells sur-
(H&E, 20×). (b and c) Invasive implants diffusely present in the omen- rounded by clefts (H&E, 400×)
tal adipose tissue (H&E, 20×). (d) Higher magnification view shows
11 Peritoneum and Broad Ligament 393
Fig. 11.39 Psammocarcinoma,
a low-grade peritoneal serous
adenocarcinoma with more
psammoma bodies than
neoplastic cells. Gross picture
is on the left, low-power view
of the microscopic picture is on
the upper right, while
magnified view is on the low
right (gross photo: courtesy of
W. Dwayne Lawrence, MD)
carcinoma [102]. The traditional criteria for diagnosis of microscopically absent tumor in the ovary, or tumor limited
peritoneal serous adenocarcinoma included (a) size of to the surface of bilateral ovaries without parenchymal
bilateral ovaries being normal or near normal; (b) the tumor invasion; (d) tumor present on the surface as well as paren-
being present mostly in the peritoneum, with only limited chyma of the ovary, but tumor size less than 5 mm2; (e)
tumor involvement on the surface of bilateral ovaries; (c) ovarian parenchymal involvement, with tumor size less
394 M. Ruhul Quddus et al.
than 5 mm2; and (f) the histology of the tumor being similar
to that of ovarian serous adenocarcinoma of any grade. The
above criteria were based mainly on the size and distribu-
tion of the tumor, which was rather arbitrarily defined. In
recent years, it has been proposed that most ovarian and
peritoneal serous adenocarcinomas arise from the fimbria
of the fallopian tube. This new school of thought chal-
lenged the traditional concepts, terminology, and diagnos-
tic criteria of pelvic serous adenocarcinoma [103]. As a
result, many previously diagnosed primary peritoneal
serous adenocarcinomas would now be considered to arise
from the fallopian tube. Female patients with BRCA gene
mutations are at a higher risk for developing these tumors
(see chapter on fallopian tube) [104, 105]; however, this
theory has only gained acceptance in cases of serous neo-
plasia. Tumors of other histologic subtypes cannot be Fig. 11.40 High-grade peritoneal serous adenocarcinoma. Tumor
explained by this proposed theory. Recent proposed guide- cells may form solid sheet or papillary structures with highly atypical
lines in the diagnosis of primary peritoneal HGSC are dis- nuclei (H&E, 400×)
cussed earlier in this chapter.
Immunohistochemical staining is of little value in differ- The prognosis and management of high-grade peritoneal
entiating these tumors. Unlike endometrial serous adenocar- serous adenocarcinoma are similar to those of ovarian high-
cinoma which lacks WT-1 expression, peritoneal, ovarian, grade serous adenocarcinoma. The most important prognos-
and fallopian tube serous adenocarcinomas all express WT-1. tic indicators are the stage of the tumor, whether optimal
Peritoneal serous adenocarcinoma should also be differenti- tumor debulking is achieved, and tumor response to chemo-
ated from diffuse malignant mesothelioma, especially the therapeutic agents. The average 5-year survival rate is
papillary variant. Malignant mesothelioma expresses cal- approximately 40%.
retinin, h-caldesmon, and CK 5/6, whereas peritoneal serous
adenocarcinoma expresses Ber-EP4 and B72.3 (see previous
section). 11.2.3 Mucinous Diseases
The clinical presentation of patients with peritoneal
serous adenocarcinoma is similar to that of ovarian serous 11.2.3.1 Endocervicosis
adenocarcinoma. The patients are typically between 50 and Peritoneal endocervicosis is defined by the presence of
60 years old. The clinical symptoms are nonspecific, with benign endocervical type mucinous epithelium in the perito-
many presenting with abdominal distention, gastrointestinal neum. This condition is exceedingly rare and may involve
discomfort, increased abdominal girth, and constipation. the serosa of the uterus, urethrorectal cul-de-sac, vaginal
Physical examination may reveal abdominal or pelvic apex, paracervical soft tissue, and urinary bladder according
masses, with or without ascites. Serum CA 125 may be ele- to isolated case reports [108, 109]. When the urinary bladder
vated. The lifetime risk of ovarian carcinoma in BRCA1 is involved, it typically occurs in the posterior wall and
gene-mutated patients is estimated to be 20–50% [106, 107], dome. Microscopic examination shows benign endocervical
whereas the lifetime risk of ovarian carcinoma in a BRCA2 type glands present between smooth muscle layers of the
gene-mutated patient is 15–30% [106, 107]. lamina propria (Fig. 11.41). A few case reports described the
Peritoneal serous adenocarcinoma typically presents as presence of infiltrative growth pattern, slight epithelial
an omental cake. The peritoneal surfaces are covered by dis- atypia, and a periglandular stromal reaction [109]. These fea-
seminated tumor nodules or solid masses. Most nodules are tures may falsely lead to the impression of well-differentiated
located on peritoneal surfaces with invasion into the subepi- adenocarcinoma. The pathogenesis of endocervicosis
thelial mesenchyme, especially omental adipose tissue. remains controversial. Proposed theories include peritoneal
Theoretically, the ovarian tissue should remain normal, and metaplasia from the secondary Müllerian system or a lesion
only have surface involvement, or minimal involvement as arising from the urachal remnant.
described above (Fig. 11.40). Certain subtypes of peritoneal
serous carcinoma may grow on the surface of the peritoneum 11.2.3.2 Mucinous Neoplasms
without tissue invasion, and therefore may be confused with Primarily peritoneal endocervical type mucinous neoplasm
malignant mesothelioma or peritoneal adenocarcinoma of an is a tumor that occurs in extra-ovarian sites, primarily in the
unknown primary. retroperitoneum, in the absence of a primary ovarian muci-
11 Peritoneum and Broad Ligament 395
Fig. 11.41 Endocervicosis. The glandular epithelium shows obvious Fig. 11.42 Peritoneal decidual reaction (mucicarmine, 400×)
endocervical type mucinous differentiation (H&E, 200×)
Fig. 11.43 Disseminated peritoneal leiomyomatosis. The lesions are Fig. 11.44 Endosalpingiosis involving lymph node in the peritoneal
composed of nodules of smooth muscle cells (H&E, 40×) cavity (H&E, 40×)
partum state, or surgical castration. The lesional cells nodes in pelvic malignancies. The glands are typically located
often express progesterone receptor [115]. in the capsule of the lymph node or between subcortical lym-
4. Treatment and prognosis: Even if not completely excised, phoid follicles. Histologically endosalpingiosis of the lymph
disseminated peritoneal leiomyomatosis is self-limiting. node is similar to endosalpingiosis in other locations
Rare cases have been reported to recur or undergo malig- (Fig. 11.44). Endosalpingiosis in the lymph node must be dif-
nant transformation, resulting in death [114, 116]. Some ferentiated from metastatic low-grade serous neoplasia from
cases of disseminated peritoneal leiomyomatosis may be the ovary. Borderline and malignant serous neoplasms have
treated with gonadotropin-releasing hormone. been reported to originate in pelvic lymph nodes [117].
Fig. 11.45 Intranodal decidual reaction. The presence of a central gland Fig. 11.47 Ectopic adrenal rest (H&E, 200×)
suggests that this focus possibly represents endometriosis (H&E, 100×)
11.3.1.2 Cyst
Cysts of varying size, ranging from microscopic to up to
20 cm or larger, may be seen in the broad ligament. The epi-
thelial lining of some of the largest cysts is typically nonspe-
cific and cannot be further characterized. Some of these cysts
may derive from the Müllerian duct or mesothelium. A few
cases may derive from the mesonephric duct. The presence
of ciliated cells may help to differentiate Müllerian duct-
derived cyst from cysts of the mesothelium or mesonephric
duct. Complications of these cysts include torsion, infarc-
Fig. 11.46 Intranodal leiomyomatosis. The lymph node is almost
completely replaced by smooth muscle cells with small foci of residual tion, or infection.
lymphoid tissue (H&E, 200×)
11.3.1.3 Endometriosis
11.3 S
ection 3: Diseases of the Broad Endometriosis may involve the broad ligament. It is fre-
Ligament quently associated with endometriosis of the pelvis and other
pelvic organs. Detailed clinical presentations, pathological
11.3.1 Tumorlike Lesions characteristics, and pathogenesis of endometriosis are dis-
cussed in detail in Chap. 12.
11.3.1.1 Embryonal Remnants
The most commonly seen embryonal remnants in the broad
ligament are the mesonephric duct remnants, which are com- 11.3.2 Neoplasms
posed of small tubules lined by cuboidal epithelium without
cilia but with a distinct basal membrane. These tubules are The incidence of neoplasia arising in the broad ligament is
frequently surrounded by circular smooth muscle tissue. low, only 20% of that of ovarian tumors. Only about 2% of
Extensive sampling of the peritoneum may reveal ectopic broad ligament tumors are borderline or malignant, while
adrenocortical tissue near the ovarian vein in over 20% of 25% of ovarian tumors are borderline or malignant [122–
females. This ectopic adrenal cortical tissue is arranged in 124]. The clinical presentation of broad ligament tumors is
trabecular or cord-like structures, similar to those seen in similar to that of ovarian tumors.
398 M. Ruhul Quddus et al.
a b
Fig. 11.49 (a) Female adnexal tumor of probably Wolffian origin nantly sievelike structure with cystic dilatations of variable shapes and
(FATWO). Often shows lobular growth pattern with areas that may sizes (H&E, 100×)
appear as solid sheets (H&E, 100×). (b) This tumor displays predomi-
Fig. 11.50 FATWO showing small lumens surrounded by tumor cells with little nuclear pleomorphism (H&E, left 100× and right 400×)
FATWO, it is also frequently present inside the lumen of 7. Treatment and prognosis: Most FATWO has a benign
the fallopian tube. Other differential diagnoses include clinical course with rare instances of malignant behavior
Sertoli-Leydig cell tumor and other ovarian surface epi- [129]. There is no reliable morphological predictor of
thelial tumors. It is noteworthy that inhibin cannot reli- malignant potential and recurrent tumors are usually mor-
ably differentiate FATWO from Sertoli-Leydig cell tumor phologically similar to the primary tumor. Treatment of
as both tumors are positive [128]. choice is surgical resection. Chemotherapy is reserved for
400 M. Ruhul Quddus et al.
malignant tumors. Because FATWO possesses a low 8. Clement PB, Young RH, Hanna W, et al. Sclerosing peritonitis
associated with luteinized thecomas of the ovary. A clinicopatho-
potential for malignancy, these patients should be fol- logical analysis of six cases. Am J Surg Pathol. 1994;18:1–13.
lowed long term. 9. Staats PN, McCluggage WG, Clement PB, et al. Luteinized theco-
mas (thecomatosis) of the type typically associated with sclerosing
11.3.2.3 Ependymoma peritonitis: a clinical, histopathologic, and immunohistochemical
analysis of 27 cases. Am J Surg Pathol. 2008;32:1273–90.
Ependymoma of the broad ligament is extremely rare [130]. 10. Kafiri G, Thomas DM, Shepherd NA, et al. p53 expres-
The average age of presentation is 38 years (range 13–48). sion is common in malignant mesothelioma. Histopathology.
Typical clinical presentations include a mass, acute abdomi- 1992;21:331–4.
nal pain, or even intraperitoneal bleeding requiring urgent 11. Henderson DW, Shilkin KB, Whitaker D. Reactive mesothelial
hyperplasia vs mesothelioma, including mesothelioma in situ: a
surgery [131]. brief review. Am J Clin Pathol. 1998;110:397–404.
12. Vallerie AM, Lerner JP, Wright JD, et al. Peritoneal inclusion
1. Gross examination: The tumor is usually solid, cystic, or cysts: a review. Obstet Gynecol Surv. 2009;64:321–34.
nodular. The cut surface is soft with patchy hemorrhage 13. Ben-Izhak O, Itin L, Feuchtwanger Z, Lifschitz-Mercer B,
Czernobilsky B. Calcifying fibrous pseudotumor of mesen-
or necrosis. try presenting with acute peritonitis: case report with immuno-
2. Microscopic examination: The morphology of ependy- histochemical study and review of literature. Int J Surg Pathol.
moma of the broad ligament is similar to that occurring in 2001;9:249–53.
the central nervous system. The tumor may be composed 14. Hill KA, Gonzalez-Crussi F, Chou PM. Calcifying fibrous pseu-
dotumor versus inflammatory myofibroblastic tumor: a histo-
of papillary structures, tightly packed tubules, or solid logical and immunohistochemical comparison. Mod Pathol.
areas. The papillary or tubular structures are lined by flat 2001;14:784–90.
or columnar ciliated cells. The nuclei may be centrally 15. Nascimento AF, Ruiz R, Hornick JL, et al. Calcifying fibrous
placed or apical, rounded, or elongated. Perivascular ‘pseudotumor’: clinicopathologic study of 15 cases and analysis
of its relationship to inflammatory myofibroblastic tumor. Int J
rosettes, psammoma bodies, and small foci of mature car- Surg Pathol. 2002;10:189–96.
tilage may be present. 16. Van Dorpe J, Ectors N, Geboes K, et al. Is calcifying fibrous pseu-
3. Differential diagnosis: The main differential diagnosis dotumor a late sclerosing stage of inflammatory myofibroblastic
includes papillary serous adenocarcinoma as both tumors tumor? Am J Surg Pathol. 1999;23:329–35.
17. Sigel JE, Smith TA, Reith JD, et al. Immunohistochemical anal-
may contain papillary structures and psammoma bodies. ysis of anaplastic lymphoma kinase expression in deep soft tis-
The presence of perivascular rosettes and positivity for sue calcifying fibrous pseudotumor: evidence of a late sclerosing
GFAP favor a diagnosis of ependymoma. stage of inflammatory myofibroblastic tumor? Ann Diagn Pathol.
4. Treatment and prognosis: There is limited experience
2001;5:10–4.
18. Lim C, McIlroy K, Briggs G, et al. Splenosis mimicking lym-
regarding treatment options and prognosis because of the phoma. Pathology. 2007;39:183–5.
rarity of this tumor. Some tumors may present with 19. Peitsidis P, Akrivos T, Vecchini G, et al. Splenosis of the peritoneal
metastases, and some may recur after several decades. cavity resembling an adnexal tumor: case report. Clin Exp Obstet
Gynecol. 2007;34:120–2.
20. Vydianath B, Gurumurthy M, Crocker J. Solitary ovarian spleno-
sis. J Clin Pathol. 2005;58:1224–5.
21. Doss BJ, Jacques SM, Qureshi F, Ramirez NC, Lawrence
References WD. Extratubal secondary trophoblastic implants:
Clinicopathologic correlation and review of the literature. Hum
1. Egli G, Newton M. The transport of carbon particles in the human Pathol. 1998;29:1984–7.
female reproductive tract. Fertil Steril. 1961;12:151–5. 22. Ali CR, Fitzgerald C. Omental and peritoneal secondary tropho-
2. Ben-Chetrit E, Levy M. Familial Mediterranean fever. Lancet. blastic implantation – an unusual complication of IVF. Reprod
1998;351:659–64. Biomed Online. 2006;12:776–68.
3. Sohar E, Gafni J, Pras M, et al. Familial Mediterranean fever. 23. Thacher SSGD, True ID, DeCherney AH. Pelvic trophoblastic
A survey of 470 cases and review of the literature. Am J Med. implants after laparoscopic treatment of tubal ectopic pregnancy.
1967;43:227–53. Fertil Steril. 1989;52:337–9.
4. Wu TI, Chang TC, Hsueh S, et al. Ovarian endometrioid carci- 24. Tsutsumi O, Ando K, Momoeda M. Ruptured isthmal pregnancy
noma with diffuse pigmented peritoneal keratin granulomas: a following laparoscopic salpingostomy in the ipsilateral tube. Int J
case report and review of the literature. Int J Gynecol Cancer. Gynaecol Obstet. 1997;57:187–9.
2006;16:426–9. 25. Bucella D, Buxant F, Anaf V, et al. Omental trophoblastic implants
5. Chen KT, Kostich ND, Rosai J. Peritoneal foreign body granu- after surgical management of ectopic pregnancy. Arch Gynecol
loma to keratin in uterine adenoacanthoma. Arch Pathol Lab Med. Obstet. 2009;280:115–7.
1978;102:174–7. 26. Kim KR, Scully RE. Peritoneal keratin granulomas with carcino-
6. Fukushima M, Sharpe L, Okagaki T. Peritoneal melanosis second- mas of endometrium and ovary and atypical polypoid adenomy-
ary to a benign dermoid cyst of the ovary: a case report with ultra- oma of endometrium. A clinico-pathological analysis of 22 cases.
structural study. Int J Gynecol Pathol. 1984;2:403–9. Am J Surg Pathol. 1990;14:925–32.
7. Dehner LP, Coffin CM. Idiopathic fibrosclerotic disorders 27. van der Horst C, Evans AJ. Peritoneal keratin granulomas compli-
and other inflammatory pseudotumors. Semin Diagn Pathol. cating endometrial carcinoma: a report of two cases and review of
1998;15:161–73. the literature. Int J Gynecol Cancer. 2008;18:549–53.
11 Peritoneum and Broad Ligament 401
28. Kim JY, Jung K-J, Sung NK, et al. Cystic adenomatoid tumor of 48. Gerald WL, Ladanyi M, de Alava E, et al. Clinical, pathologic,
the uterus. Am J Roentgen. 2002;179:1068–70. and molecular spectrum of tumors associated with t(11;22)
29. Daya D, McCaughey WT. Well-differentiated papillary mesothe- (p13;q12): desmoplastic small round-cell tumor and its variants.
lioma of the peritoneum. A clinicopathologic study of 22 cases. J Clin Oncol. 1998;16:3028–36.
Cancer. 1990;65:292–6. 49. Kim J, Lee K, Pelletier J. The desmoplastic small round cell tumor
30. Hoekstra AV, Riben MW, Frumovitz M, et al. Well-differentiated t(11;22) translocation produces EWS/WT1 isoforms with differ-
papillary mesothelioma of the peritoneum: a pathological analysis ing oncogenic properties. Oncogene. 1998;16:1973–9.
and review of the literature. Gynecol Oncol. 2005;98:161–7. 50. Ordi J, de Alava E, Torne A, et al. Intraabdominal desmoplastic
31. Goldblum J, Hart WR. Localized and diffuse mesotheliomas of small round cell tumor with EWS/ERG fusion transcript. Am J
the genital tract and peritoneum in women. A clinic-pathological Surg Pathol. 1998;22:1026–32.
study of nineteen true mesothelial neoplasms, other than adeno- 51. Coffin CM, Watterson J, Priest JR, et al. Extrapulmonary inflam-
matoid tumors, multicystic mesotheliomas and localized fibrous matory myofibroblastic tumor (inflammatory pseudotumor). A
tumor. Am J Surg Pathol. 1995;19:1124–37. clinicopathologic and immunohistochemical study of 84 cases.
32. Baker PM, Clement PB, Young RH. Malignant peritoneal meso- Am J Surg Pathol. 1995;19:859–72.
thelioma in women: a study of 75 cases with emphasis on their 52. Coffin CM, Dehner LP, Meis-Kindblom JM. Inflammatory myofi-
morphologic spectrum and differential diagnosis. Am J Clin broblastic tumor, inflammatory fibrosarcoma, and related lesions:
Pathol. 2005;123:724–37. an historical review with differential diagnostic considerations.
33. Churg A, Allen T, Borczuk AC, et al. We-differentiated pap- Semin Diagn Pathol. 1998;15:102–10.
illary mesothelioma with invasive foci. Am J Surg Pathol. 53. Griffin CA, Hawkins AL, Dvorak C, et al. Recurrent involvement
2014;38:990–8. of 2p23 in inflammatory myofibroblastic tumors. Cancer Res.
34. Sawh RN, Malpica A, Deavers MT, et al. Benign cystic mesothe- 1999;59:2776–80.
lioma of the peritoneum: a clinicopathologic study of 17 cases 54. Treissman SP, Gillis DA, Lee CL, et al. Omental- mesenteric
and immunohistochemical analysis of estrogen and progesterone inflammatory pseudotumor. Cytogenetic demonstration of genetic
receptor status. Hum Pathol. 2003;34:369–74. changes and monoclonality in one tumor. Cancer. 1994;73:1433–7.
35. Attanoos R, Gibbs AR. Pathology of malignant mesothelioma. 55. Brunnemann RB, Ro JY, Ordonez NG, et al. Extrapleural solitary
Histopathology. 1997;30:403–18. fibrous tumor: a clinicopathologic study of 24 cases. Mod Pathol.
36. Baker PM, Clement PB, Young RH. Malignant peritoneal meso- 1999;12:1034–42.
thelioma in women: a study of 75 cases with emphasis on their 56. Young RH, Clement PB, McCaughey WT. Solitary fibrous tumors
morphologic spectrum and differential diagnosis. Am J Clin ('fibrous mesotheliomas') of the peritoneum. A report of three cases
Pathol. 1999;123:724–37. and a review of the literature. Arch Pathol Lab Med. 1990;114:493–5.
37. Hemminki K, Li X. Time trends and occupational risk factors 57. Morimitsu Y, Nakajima M, Hisaoka M, et al. Extrapleural solitary
for peritoneal mesothelioma in Sweden. J Occup Environ Med. fibrous tumor: clinicopathologic study of 17 cases and molecular
2003;45:451–5. analysis of the p53 pathway. APMIS. 2000;108:617–25.
38. Allen TC, Cagle PT, Churg AM, et al. Localized malignant meso- 58. Mosquera JM, Fletcher CD. Expanding the spectrum of malignant
thelioma. Am J Surg Pathol. 2005;29:866–73. progression in solitary fibrous tumors: a study of 8 cases with a
39. Rekhi B, Pathuthara S, Ajit D, et al. “Signet-ring” cells-a caveat discrete anaplastic component—is this dedifferentiated SFT? Am
in the diagnosis of a diffuse peritoneal mesothelioma occurring in J Surg Pathol. 2009;33:1314–21.
a lady presenting with recurrent ascites: an unusual case report. 59. Vallat-Decouuvelaere A-V, Dry SM, Fletcher CDM. Atypical and
Diagn Cytopathol. 2010;38:435–9. malignant solitary fibrous tumors in extrathoracic locations: evi-
40. Shanks JH, Harris M, Banerjee SS, et al. Mesotheliomas with dence of their comparability to intra-thoracic tumors. Am J Surg
deciduoid morphology: a morphologic spectrum and a variant not Pathol. 1998;22:1501–11.
confined to young females. Am J Surg Pathol. 2000;24:285–94. 60. Gonzalez-Crussi F, deMello DE, Sotelo-Avila C. Omental-
41. Attanoos RL, Webb R, Dojcinov SD, et al. Value of mesothe- mesenteric myxoid hamartomas. Infantile lesions simulating
lial and epithelial antibodies in distinguishing diffuse peritoneal malignant tumors. Am J Surg Pathol. 1983;7:567–78.
mesothelioma in females from serous papillary carcinoma of the 61. Sugarbaker PH, Ronnett BM, Archer A, et al. Pseudomyxoma
ovary and peritoneum. Histopathology. 2002;40:237–44. peritonei syndrome. Adv Surg. 1996;30:233–80.
42. Comin CE, Saieva C, Messerini L. H-caldesmon, calretinin, 62. Brueggen C, Baird G, Meisheid A. Pseudomyxoma peritonei
estrogen receptor, and Ber-EP4: a useful combination of immu- syndrome of appendiceal origin: an overview. Clin J Oncol Nurs.
nohistochemical markers for differentiating epithelioid peritoneal 2007;11:525–32.
mesothelioma from serous papillary carcinoma of the ovary. Am J 63. Fox H. Pseudomyxoma peritonei. Br J Obstet Gynaecol.
Surg Pathol. 2007;31:1139–48. 1996;103:197–8.
43. Fois AG, Pirina P, Arcadu A, et al. Desmoplastic small round cell 64. Szych C, Staebler A, Connolly DC, et al. Molecular genetic evidence
tumors of the pleura: a review of clinical literature. Multidiscip supporting the clonality and appendiceal o rigin of Pseudomyxoma
Respir Med. 2017;19:12–22. peritonei in women. Am J Pathol. 1999;154:1849–55.
44. Ordonez NG. Desmoplastic small round cell tumor: I: a histopath- 65. Ronnett BM, Shmookler BM, Diener-West M, et al.
ologic study of 39 cases with emphasis on unusual histological Immunohistochemical evidence supporting the appendiceal ori-
patterns. Am J Surg Pathol. 1998;22:1303–13. gin of pseudomyxoma peritonei in women. Int J Gynecol Pathol.
45. Ordonez NG. Desmoplastic small round cell tumor: II: an 1997;16:1–9.
ultrastructural and immunohistochemical study with empha- 66. Marquette S, Amant F, Vergote I, et al. Pseudomyxoma peritonei
sis on new immunohistochemical markers. Am J Surg Pathol. associated with a mucinous ovarian tumor arising from a mature
1998;22:1314–27. cystic teratoma. A case report. Int J Gynecol Pathol. 2006;25:340–3.
46. Zhang PJ, Goldblum JR, Pawel BR, et al. Immunophenotype of 67. Ronnett BM, Seidman JD. Mucinous tumors arising in ovarian
desmoplastic small round cell tumors as detected in cases with mature cystic teratomas: relationship to the clinical syndrome of
EWS-WT1 gene fusion product. Mod Pathol. 2003;16:229–35. pseudomyxoma peritonei. Am J Surg Pathol. 2003;27:650–7.
47. Barcena C, Oliva E. WT1 expression in the female genital tract. 68. Vang R, Gown AM, Zhao C, et al. Ovarian mucinous tumors asso-
Adv Anat Pathol. 2011;18:454–65. ciated with mature cystic teratomas: morphologic and immunohis-
402 M. Ruhul Quddus et al.
tochemical analysis identifies a subset of potential teratomatous uterine high-grade serous carcinoma. Findings of a survey of
origin that shares features of lower gastrointestinal tract mucinous pathologist and clinicians. Int J Gynecol Pathol. 2016;36:230–9.
tumors more commonly encountered as secondary tumors in the https://doi.org/10.1097/PGP.0000000000000336.
ovary. Am J Surg Pathol. 2007;31:854–69. 88. Singh K, Sung CJ, Lawrence WD, Quddus MR. Testosterone
69. Hart WR. Mucinous tumors of the ovary: a review. Int J Gynecol induced "virilization" of mesonephric duct remnants and cervical
Pathol. 2005;24:4–25. squamous epithelium in female-to-male transgenders: a report of
70. Panarelli NC, Yantiss RK. Mucinous neoplasms of the appendix three cases. Int J Gyncol Pathol. 2017;36:328–33.
and peritoneum. Arch Pathol Lab Med. 2011;135:1261–8. 89. Bell DA, Weinstock MA, Scully RE. Peritoneal implants of ovar-
71. Ronnett BM, Kurman RJ, Shmookler BM, et al. The morphologic ian serous borderline tumours: histologic features and prognosis.
spectrum of ovarian metastases of appendiceal adenocarcino- Cancer. 1988;62:2212–22.
mas: a clinicopathologic and immunohistochemical analysis of 90. Seidman JD, Kurman RJ. Subclassification of serous borderline
tumors often misinterpreted as primary ovarian tumors or meta- tumors of the ovary into benign and malignant types. A clinico-
static tumors from other gastrointestinal sites. Am J Surg Pathol. pathologic study of 65 advanced stage cases. Am J Surg Pathol.
1997;21:1144–55. 1996;20:1331–45.
72. Ronnett BM, Yan H, Kurman RJ, et al. Patients with pseudomyx- 91. Burks RT, Sherman ME, Kurman RJ. Micropapillary serous
oma peritonei associated with disseminated peritoneal adenomu- carcinoma of the ovary. A distinctive low-grade carcinoma
cinosis have a significantly more favorable prognosis than patients related to serous borderline tumors. Am J Surg Pathol.
with peritoneal mucinous carcinomatosis. Cancer. 2001;92:85–91. 1996;20:1319–30.
73. Baratti D, Kusamura S, Deraco M. The fifth international work- 92. Bell KA, Smith Sehdev AE, Kurman RJ. Refined diagnostic cri-
shop on peritoneal surface malignancy (Milan, Italy, December teria for implants associated with ovarian atypical proliferative
4–6, 2006): methodology of disease-specific consensus. J Surg serous tumors (borderline) and micropapillary serous carcinomas.
Oncol. 2008;98:258–62. Am J Surg Pathol. 2001;25:419–32.
74. Ferguson AW, Katabuchi H, Ronnett BM, et al. Glial implants in 93. Gilks CB, Alkushi A, Yue JJ, et al. Advanced-stage serous border-
gliomatosis peritonei arise from normal tissue, not from the asso- line tumors of the ovary: a clinicopathological study of 49 cases.
ciated teratoma. Am J Pathol. 2001;159:51–5. Int J Gynecol Pathol. 2003;22:29–36.
75. Ehdaivand S, Simon RA, Sung CJ, Steinhoff MM, Lawrence WD, 94. Longacre TA, McKenney JK, Tazelaar HD, et al. Ovarian serous
Quddus MR. Incidental gynecologic neoplasms in morcellated tumors of low malignant potential (borderline tumors): outcome-
uterine specimens: a case series with follow-up. Hum Pathol. based study of 276 patients with long- term (> or =5-year) follow-
2014;45:2311–7. up. Am J Surg Pathol. 2005;29:707–23.
76. Lauchlan SC. Conceptual unity of the Mullerian tumor group: a 95. Seidman JD, Kurman RJ. Ovarian serous borderline tumors: a
histologic study. Cancer. 1968;22:601–10. critical review of the literature with emphasis on prognostic indi-
77. Lauchlan SC. The secondary mullerian system revisited. Int J cators. Hum Pathol. 2000;31:539–57.
Gynecol Pathol. 1994;13:73–9. 96. Silva EG, Gershenson DM, Malpica A, et al. The recurrence and
78. Quddus MR, Sung CJ, Lauchlan SC. Benign and malignant serous the overall survival rates of ovarian serous borderline neoplasms
and endometrioid epithelium in the omentum. Gynecol Oncol. with noninvasive implants is time dependent. Am J Surg Pathol.
1999;75:227–32. 2006;30:1367–71.
79. Quddus MR, Sung CJ, Zhang C, Moore RG, Ou J, Steinhoff 97. Gershenson DM, Silva EG, Tortolero-Luna G, et al. Serous bor-
MM, Lawrence WD. The presence and location of epithelial derline tumors of the ovary with noninvasive peritoneal implants.
implants and implants with epithelial proliferation may pre- Cancer. 1998;83:2157–63.
dict a higher risk of recurrence in serous borderline ovarian 98. Schmeler KM, Gershenson DM. Low-grade serous ovarian can-
tumors: a Clinicopathologic study of 188 cases. Hum Pathol. cer: a unique disease. Curr Oncol Rep. 2008;10:519–23.
2012;43:747–52. 99. Quddus MR, Rashid LB, Hansen K, et al. High-grade serous car-
80. Mingels MJ, van Ham MA, de Jievit IM, et al. Mullerian precur- cinoma arising in a low-grade serous carcinoma and micropapil-
sor lesions in serous ovarian cancer patients: using SEE-Fim and lary serous borderline tumor of the ovary in a 23- year-old woman.
SEE-End protocol. Mod Pathol. 2014;27:1002–13. Histopathology. 2009;54:771–3.
81. Klymneko Y, Kim O, Stack MS. Complex determinants of epithe- 100. Poujade O, Uzan C, Gouy S, et al. Primary psammocarcinoma of
lial: mesenchymal phenotypic plasticity in ovarian cancer. Cancers the ovary or peritoneum. Int J Gynecol Cancer. 2009;19:844–6.
(Basal). 2017;9:104. https://doi.org/10.3390/cancers9080104. 101. Bodnar L, Wcislo G, Gornas M, et al. Serous peritoneal psam-
82. Piek JM, van Diest PJ, Zweemer RP, et al. Dysplastic changes in mocarcinoma with an aggressive course: a case and review of the
prophylactically removed fallopian tubes of women predisposed literature. Ginekol Pol. 2009;80:632–6.
to developing ovarian cancer. J Pathol. 2001;195:451–6. 102. Fromm GL, Gershenson DM, Silva EG. Papillary serous carci-
83. Crum CP, Drapkin R, Miton A, et al. The distal fallopian tube: noma of the peritoneum. Obstet Gynecol. 1990;75:89–95.
a new model of pelvic serous carcinogenesis. Curr Opin Obstet 103. Kindelberger DW, Lee Y, Miron A, et al. Intraepithelial carcinoma
Gynecol. 2007;19:3–9. of the fimbria and pelvic serous carcinoma: evidence for a causal
84. Singh N, Gilks CB, Wilkinson N, McCluggage WG. Assignment relationship. Am J Surg Pathol. 2007;31:161–9.
of primary site in high-grade serous tubal, ovarian and peritoneal 104. Leeper K, Garcia R, Swisher E, et al. Pathologic findings in pro-
carcinoma: a proposal. Histopathology. 2014;65:149–54. phylactic oophorectomy specimens in high-risk women. Gynecol
85. Singh R, Cho KR. Serous tubal intraepithelial carcinoma or not? Oncol. 2002;87:52–6.
Metastases to fallopian tube can masquerade as in-situ lesions. 105. Levine DA, Argenta PA, Yee CJ, et al. Fallopian tube and primary
Arch Pathol Lab Med. 2017;141:1313–5. peritoneal carcinomas associated with BRCA mutations. J Clin
86. Rabban JT, Vohra P, Zaloudek CJ. Nongynecologic metastasis Oncol. 2003;21:4222–7.
to fallopian tube mucosa. A potential mimic of tubal high-grade 106. Frank TS, Manley SA, Olopade OI, et al. Sequence analysis of
serous carcinoma and benign tubal mucinous metaplasia o muci- BRCA1 and BRCA2: correlation of mutation s with family his-
nous hyperplasia. A. J Surg Pathol. 2015;39:35–51. tory and ovarian cancer risk. J Clin Oncol. 1998;16:2417–25.
87. McCluggage WG, Hirschowtz L, Golks CB, Wolkinson N, Singh 107. Breast Cancer Linkage Consortium. Cancer risk in BRCA2 muta-
N. The fallopian tube origin and primary site assignment in extra- tion carriers. J Natl Can Inst. 1999;91:1310–6.
11 Peritoneum and Broad Ligament 403
108. Lim S, Kim JY, Park K, et al. Mullerianosis of the mesosalpinx: a 119. Kheir SM, Mann WJ, Wilkerson JA. Glandular inclusions in
case report. Int J Gynecol Pathol. 2003;22:209–12. lymph nodes. The problem of extensive involvement and relation-
109. Heretis J, Stamatiou V, Papadimitriou V, et al. Endocervicosis of ship to salpingitis. Am J Surg Pathol. 1981;5:353–9.
the bladder report of a case and review of the current literature. Int 120. Hsu YK, Rosenshein NB, Parmley TH, et al. Leiomyomatosis in
J Clin Exp Pathol. 2009;2:91–4. pelvic lymph nodes. Obstet Gynecol. 1981;57:91S–3S.
110. Banerjee R, Gough J. Cystic mucinous tumors of the mesen- 121. Horie A, Ishii N, Matsumoto M, et al. Leiomyomatosis in the pel-
tery and retroperitoneum: report of three cases. Histopathology. vic lymph node and peritoneum. Acta Pathol Jpn. 1984;34:813–9.
1988;12:527–32. 122. Aslan i M, Ahn GH, Scully RE. Serous papillary cystadenoma of
111. Pearl ML, Valea F, Chumas J, et al. Primary retroperitoneal muci- borderline malignancy of broad ligament. A report of 25 cases. Int
nous cystadenocarcinoma of low malignant potential: a case J Gynecol Pathol. 1988;7:131–8.
report and literature review. Gynecol Oncol. 1996;61:150–2. 123. Altaras MM, Jaffe R, Corduba M, et al. Primary paraovarian cyst-
112. Park U, Han KC, Chang HK, et al. A primary mucinous cys- adenocarcinoma: clinical and management aspects and literature
toadenocarcinoma of the retroperitoneum. Gynecol Oncol. review. Gynecol Oncol. 1990;38:268–72.
1991;42:64–7. 124. Altaras M, Rosen D, Shapira J, et al. Advanced primary ovarian
113. Altinok G, Usubutun A, Kucukali T, et al. Disseminated peritoneal carcinoma in pregnancy. Am J Obstet Gynecol. 1989;160:1210–1.
leiomyomatosis. A benign entity mimicking carcinomatosis. Arch 125. Cox R, Vang R, Epstein JI. Papillary cystadenoma of the epididy-
Gynecol Obstet. 2000;264:54, 55. mis and broad ligament: morphologic and immunohistochemical
114. Heinig J, Neff A, Cirkel U, et al. Recurrent leiomyomatosis peri- overlap with clear cell papillary renal cell carcinoma. Am J Surg
tonealis disseminata after hysterectomy and bilateral salpingo- Pathol. 2014;38:713–8.
oophorectomy during combined hormone replacement therapy. 126. Kariminejad MH, Scully RC. Female adnexal tumor of probable
Eur J Obstet Gynecol Reprod Biol. 2003;111:216–8. Wolffian origin. Cancer. 1973;31:671–7.
115. Due W, Pickartz H. Immunohistologic detection of estrogen and 127. Fukunaga M, Bisceglia M, Dimitri L. Endometrioid carcinoma of
progesterone receptors in disseminated peritoneal leiomyomato- the fallopian tube resembling a female adnexal tumor of probable
sis. Int J Gynecol Pathol. 1989;8:46–53. Wolffian origin. Adv Anat Pathol. 2004;11:269–72.
116. Fulcher AS, Szucs RA. Leiomyomatosis peritonealis disseminata 128. Zheng W, Senturk BZ, Parkash V. Inhibin immunohistochemical
complicated by sarcomatous transformation and ovarian torsion: staining: a practical approach for the surgical pathologist in the
presentation of two cases and review of the literature. Abdom diagnoses of ovarian sex cord-stromal tumors. Adv Anat Pathol.
Imaging. 1998;23:640–4. 2003;10:27–38.
117. Prade M, Spatz A, Bentley R, Dullivard P, Bognel C, Robboy 129. Brescia RJ, Cardoso de Almeida PC, Fuller AF Jr, et al. Female
SJ. Borderline and malignant serous tumors arising in pelvic adnexal tumor of probable Wolffian origin with multiple recur-
lymph nodes: evidence of origin in benign glandular inclusions. rences over 16 years. Cancer. 1985;56:1456–61.
Int J Gynecol Pathol. 1995;14:87–91. 130. Bell DA, Woodruff JM, Scully RE. Ependymoma of the broad
118. Horn LC, Bilek K. Frequency and histogenesis of pelvic retro- ligament. A report of two cases. Am J Surg Pathol. 1984;8:203–9.
peritoneal lymph node inclusions of the female genital tract. 131. Yamamoto Y, Takehara K, Okame S, Komatsu M, Shiroyama Y,
An immunohistochemical study of 34 cases. Pathol Res Pract. Yokoyama T. A case of ependymoma arising from the peritoneum.
1995;191:991–6. Int Can Conf J. 2016;5:121–5.
Endometriosis and Endometriosis-
Associated Tumors 12
Rosalia C. M. Simmen, Charles Matthew Quick,
Angela S. Kelley, and Wenxin Zheng
© Science Press & Springer Nature Singapore Pte Ltd. 2019 405
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_12
406 R. C. M. Simmen et al.
suspected endometriosis receive empiric medical therapy, are stratified into four stages: stage 1 (minimal), 2 (mild), 3
with hormonal suppression and/or pain control, to avoid the (moderate), and 4 (severe). Stage 1 endometriosis is charac-
inherent risks of diagnostic surgery. To assist with concep- terized by small, isolated, superficial endometriosis implants
tion in infertile women, assisted reproductive technologies on the peritoneum or within the ovaries. On the other hand,
such as in vitro fertilization may be employed [13]. As a stage 4 endometriosis is characterized by numerous endome-
chronic condition, endometriosis-related symptoms may triosis lesions, which may be superficial or deeply infiltrat-
progress, regress, or remain stable until menopause, when ing, in addition to dense peritoneal or pelvic adhesions, and
endometriosis generally becomes quiescent. There is increas- possibly a large ovarian endometriosis cyst. The majority of
ing evidence, however, that endometriosis may persist even women with endometriosis are assigned to stages 1–2; how-
in the postmenopausal state [14]. Women with endometriosis ever, there is little correlation between stage of endometrio-
have a two- to threefold increased risk for ovarian cancers sis and severity of pain symptoms [21]. A recent consensus
and other malignancies such as endometrial and breast statement by the World Endometriosis Society has incorpo-
cancers [15]. rated additional endpoints that are highly relevant to women
The definitive diagnosis of endometriosis can only be with endometriosis [22].
made surgically, with histopathologic visualization of both At present, there are no good serum markers with suffi-
endometrial glands and stroma within a surgical biopsy. cient accuracy currently to assess the severity of endometrio-
Endometriosis lesions may vary in appearance at the time of sis. Serum Cancer Antigen-125 (CA-125) levels have been
laparoscopy; the classic description is that of a blue or black reported to significantly differ in patients with pelvic versus
“powder burn lesion,” but implants may also be hemorrhagic extra-pelvic (i.e., pelvic > extra-pelvic) lesions [23]. Another
or nonpigmented [16, 17]. However, even with experienced study failed to confirm this finding and instead observed that
laparoscopic surgeons, there may be inconsistencies between extra-pelvic endometriosis was associated with higher
suspected endometriosis (by visual appearance) and con- patient serum CA-125 levels and lesions displaying increased
firmed endometriosis by histopathology [16, 18, 19], high- expression of the epithelial-mesenchymal transcription fac-
lighting the importance of histologic evaluation. tor ZEB1 than pelvic endometriosis [24]. The use of mag-
netic resonance imaging (MRI) to distinguish endometriotic
tissue from adhesions and fibrosis has been recently reported
12.2 Clinical Relevance and may be suitable for the diagnosis of endometriosis at
unusual anatomic sites in symptomatic patients [25].
The American Society for Reproductive Medicine (ASRM) Given the nonspecific nature of endometriosis symptoms,
has developed the most commonly accepted classification the diagnosis of the condition is often delayed. Indeed, while
system for endometriosis [20]. This guideline accounts for the mean age at diagnosis is ~25–29 years, most patients
lesion size, appearance, anatomic location, and depth of manifest symptoms of pelvic pain and dysmenorrhea at the
invasion of endometriotic lesions visualized during diagnos- start of menarche. Symptoms of endometriosis do not differ
tic surgery (Table 12.1). Based on these parameters, patients between women surgically diagnosed during adolescence
compared with those diagnosed as adults [26]. A summary of
the differential diagnosis of endometriosis depending on the
Table 12.1 American Society for Reproductive Medicine Classification symptom was provided in Mounsy et al. [27]. Dysmenorrhea
for Endometriosisa
in endometriosis may be primary or secondary. Generalized
Depth of endometrial pelvic pain, while characteristic of endometriosis, may also
Stage Location (size) explants
arise from malignant or benign neoplasms, pelvic adhesions,
I. Minimal Peritoneum (1–3 cm) Superficial
Ovary (<1 cm) Superficial pelvic inflammatory disease, obstructive genital anomalies,
II. Mild Peritoneum (>3 cm) Deep endometriosis or non-gynecologic causes. Women with dyspareunia may
Ovary (<1 cm) Superficial have endometriosis but the differential diagnosis also
Ovary-filmy adhesions Superficial includes pelvic infection and bowel or urinary pathology.
III. Moderate Peritoneum (>3 cm) Deep endometriosis
The diagnosis becomes more challenging if the lesions are
Ovary (1–3 cm) Deep endometriosis
Ovary Dense adhesions located in distal sites. Thus, direct visualization (via laparos-
Ovary-filmy adhesions Superficial copy) and histological confirmation of biopsied lesions must
Fallopian tubes(<1 cm) Dense adhesions be strongly considered for cases of suspected endometriosis
IV. Severe Peritoneum (>3 cm) Deep endometriosis to facilitate timely initiation of appropriate therapy.
Ovary (1–3 cm) Deep endometriosis
Ovary Dense adhesions Clinically useful serum biomarkers for diagnosis and
Fallopian tube (>2 cm) Dense adhesions staging of endometriosis and/or to differentiate endometrio-
Cul de sac Complete obliteration sis subtypes are currently lacking. Cancer Antigen (CA) 125,
Reference [20]
a
CA 19-9, and the cytokine interleukin-6 have been suggested
12 Endometriosis and Endometriosis-Associated Tumors 407
as possible noninvasive markers; however, an extensive best explained by the migration or transplantation of endo-
meta-analysis raised questions on their diagnostic specificity metrial cells into ectopic, extrauterine locations.
and accuracy [28]. One study reported that a four-marker Nevertheless, retrograde menstruation occurs frequently
panel of CA-125, macrophage chemotactic protein-1, leptin, in women with patent fallopian tubes who do not develop
and macrophage migration inhibitory factor could diagnose endometriosis [39, 40]. In addition, endometriosis has been
48% of subjects with 93% accuracy [29], but this procedure noted in previously hysterectomized women, and even in
has yet to be incorporated clinically. Tumor necrosis factor men undergoing treatment for prostate cancer [41]. Such
levels in the peritoneal fluids were found to be higher in findings suggest that menstruation is not definitively required
women with than in those without endometriosis [30]; how- for the development of endometriosis and highlight the mul-
ever, the test may have limited application since it relies tifactorial nature of the condition.
upon an invasive procedure.
Recent studies using mouse models of endometriosis [31] 12.3.1.2 Metaplasia
and women with the condition [32] have tested the feasibility An alternative theory is that of coelomic metaplasia.
of using circulating microRNAs (miRNAs) as biomarkers Coelomic epithelial cells are found within the peritoneal cav-
for endometriosis. While the results appear promising, the ity, such as on the surface of the ovary, in addition to the
accuracy and specificity of these identified miRNAs have yet pleural space. It is proposed that coelomic epithelial cells
to be rigorously vetted in clinical settings. may undergo metaplastic transformation into endometrial
cells, leading to ectopic endometrial lesions which ultimately
develop into endometriosis [2, 42]. Metaplasia may be
induced by exposure to chemical insults (such as menstrual
12.3 Pathogenesis of Endometriosis fluid) or high levels of estrogen [43].
Support for this theory includes observations that endo-
Several theories have been proposed to explain the pathogen- metriosis may be found in distant sites such as the pleural
esis of endometriosis. Below, we discuss four theories which cavity, where coelomic epithelium exists, and that endome-
may account for the development and persistence of endo- triosis can occur in prepubertal girls prior to the onset of
metriosis in affected women. It is important to recognize that menarche [33, 44]. Furthermore, case reports have demon-
no individual theory has been accepted to fully explain this strated surgical evidence of endometriosis in women with
complex, multifactorial disorder [33, 34]. Mayer-Rokitansky-Kuster-Hauser syndrome (defined as con-
genital agenesis of the uterus, cervix, and vagina due to fail-
ure of the Mullerian duct to develop), indicating that a
12.3.1 Histogenesis functional uterus is expendable for the development of endo-
metriosis [44–46].
12.3.1.1 Retrograde Menstruation
The concept of retrograde menstruation is the most cited 12.3.1.3 Stem Cell Theory
explanation for the pathogenesis of endometriosis. This In normal menstrual cycles, the human endometrium is con-
theory involves the retrograde flow of endometrial glands tinually regenerated by both local adult progenitor cells and
and stromal cells, sloughed during menses, through the fal- bone marrow-derived multipotent stem cells [41]. Stem cells
lopian tubes and into the peritoneal cavity [2, 35]. are undifferentiated cells which are characterized by their
Implantation and proliferation of these endometrial cells ability to self-renew and to later develop into a variety of
then occur within ectopic sites, where they are resistant to mature cell types [39]. Support for the role of endometrial
apoptosis [9]. stem cells in the pathogenesis of endometriosis and their
Support for the theory of retrograde menstruation has potential involvement in ectopic endometrial proliferation
been demonstrated in animal models [36, 37]. In addition, and differentiation have largely come from experimental
the prevalence of endometriosis is increased in women with mouse and nonhuman primate models [41].
obstructive outflow tract anomalies, such as cervical steno- In women with endometriosis, refluxed stem cells (via
sis, where the likelihood of retrograde menstruation is higher retrograde menstruation) deposited into the peritoneal
[1, 38]. The risk of endometriosis also increases with shorter cavity can subsequently undergo differentiation into endo-
menstrual cycles, increased menstrual frequency, and heavier metrial glands and stroma [35, 41]. The migration of extra-
menstrual flow, lending support to the role of menstruation in uterine stem cells through lymphatic or vascular channels
the pathogenesis of endometriosis [2]. Though not explained may contribute to the development of distant endometrio-
by retrograde menstruation, endometriosis lesions which sis lesions [39]. Understanding the phenomenon of “stem
develop in perineal scars following an obstetric laceration, or cell trafficking” is a growing field of endometriosis
in abdominal incisions after cesarean delivery, are likewise research [47].
408 R. C. M. Simmen et al.
12.3.1.4 Tubal Origin of Ovarian Endometriosis significant gene expression similarities between the fallopian
The fallopian tube has been recently proposed as a cellular tube and ovarian endometriosis, compared to expression pro-
contributor to the development of endometriosis [48]. About files between the endometria and ovarian endometriosis. It
a decade ago, it was noticed that there are very early morpho- was concluded that a substantial portion of ovarian endome-
logic changes in cases of ovarian endometriosis, defined as triosis may originate from the fallopian tube [51].
“initial endometriosis” [49]. Subsequently, a study of the cel- The above findings are relevant in light of growing evi-
lular origin of ovarian low-grade serous carcinoma found dence that ovarian epithelial cancers, and possibly endometri-
that ovarian epithelial inclusions (also called as endosalpin- oid cancers, may originate in the fallopian tubes [52–54].
giosis) are most commonly derived from the fallopian tube Additionally, epidemiologic evidence suggests that bilateral
[50]. Considering that initial endometriosis (Fig. 12.1) mor- salpingectomy (defined as the surgical removal of the bilat-
phologically overlaps with ovarian epithelial inclusions, it eral fallopian tubes) may decrease lifetime risk of ovarian
was proposed that ovarian endometriosis, which develops cancer in high-risk (e.g., women with BRCA1/2 gene muta-
from ovarian “initial endometriosis,” may also arise from the tions) as well as in low-risk populations [55–57]. If future
fallopian tube. In a study to evaluate this hypothesis, micro- investigations confirm the contribution of the fallopian tubes
array analysis was used to compare gene expression between to the pathogenesis of endometriosis, the findings may have
the fallopian tube and the endometrium of patients with ovar- significant therapeutic and preventative implications for
ian endometriosis and corresponding lesions. There were reproductive-age women who have completed childbearing.
a b
c d
Fig. 12.1 Initial endometriosis. Shown here are ovarian epithelial-like is present on the top right (c). The dramatic differences noted in the
inclusions in the ovarian cortex. Stromal changes including microcapil- stroma surrounding the ovarian epithelial inclusions are presented at a
lary vessels are present surrounding the epithelial inclusions (a). higher magnification (d). The non-spindle stroma enriched with micro-
Stromal and vascular changes are evident in a magnified view (b). capillary vessels represents the earliest morphologic change of
Another inclusion-like structure shows fresh bleeding adjacent to the endometriosis
glandular structure (c, left and mid-right), while typical ovarian stroma
12 Endometriosis and Endometriosis-Associated Tumors 409
significantly reduce the systemic inflammatory profiles in sis accounts for 5–10% of women with disease. Endometriosis
these women, suggesting lesions as major drivers of sys- may also be established at distant locations such as the pleu-
temic inflammation [88]. The enhanced inflammatory status ral space, diaphragm, or breast [16]. The implantation and
of women with endometriosis is likely caused by the highly proliferation of endometrial cells in ectopic sites result in
estrogenic dependence of the disease, given the demon- inflammation, fibrosis, and distortion of normal anatomy.
strated cross talk between pro-inflammatory molecules (e.g., The ovary and peritoneum are the most frequent locations
IL-6) and estradiol during early disease progression [89] and of pelvic endometriosis. Other pelvic sites include the bowel
the role of estrogens in the recruitment of pro-inflammatory [97] or bladder. Extra-pelvic lesions are less common, likely
molecules within ectopic lesions [90]. Thus, recently identi- deeply infiltrating, and found in such anatomic locations as
fied estrogen receptor antagonists that can concurrently sup- hepatobiliary and urinary systems, upper abdomen (lung,
press estrogenic and inflammatory activities [91] may show thorax), abdominal wall, and adrenal glands.
promise as preventative and treatment strategies for Diagnosis of pelvic and extra-pelvic lesions can be chal-
endometriosis. lenging. Endometriosis of the uterine cervix is generally
As mentioned earlier, the role of progesterone resistance asymptomatic and can be mistaken for cervical neoplasia
in the progression of endometriosis may be linked to the due to the presence of a cervical mass [98, 99]. Intestinal
enhanced inflammatory status of women with the disease. endometriosis may present with abdominal pain or gastroin-
Gene expression analyses have shown that loss of progester- testinal bleeding [100], and the differential diagnoses may
one receptor expression in lesions is associated with their include diverticulitis, appendicitis, Crohn disease, irritable
higher expression of inflammatory cytokines [58, 59]. bowel syndrome, carcinoma, and lymphoma. Patients subse-
Mechanistically, it has been demonstrated that inflammatory quently diagnosed with thoracic endometriosis may initially
molecules such as IL-1 and tumor necrosis factor (TNF)-α present with shoulder pain, catamenial pneumothorax, and/
can significantly reduce progesterone receptor expression or hemoptysis [101]. In patients with bowel endometriosis,
[81]. Conversely, progestin treatment of endometriotic stro- lesions display a characteristic “comet” appearance and have
mal cells can suppress TNF-α-induced inflammation [92]. been associated with obliteration of the cul-de-sac and pelvic
The significant contribution of pro-inflammatory mole- pain [102]. Abdominal wall endometriosis occurs when
cules to endometriosis raises the interesting potential for non- endometrial cells attach to the fascia or dermis at the time of
steroidal anti-inflammatory drugs such as prostaglandin obstetrical or gynecological surgery. Approximately 1% of
synthesis inhibitors (e.g., cyclooxygenase-2 inhibitors) and women who have had a caesarean delivery subsequently pre-
diets rich in anti-inflammatory components (e.g., resveratrol sented with focal pain and/or palpable mass near the surgical
in grapes) in the management of endometriosis. While there scar which was diagnosed as endometriosis [103].
is sufficient support to these possibilities in animal models
[93, 94] and in studies using endometriotic stromal cells [95], 12.4.1.1 Peritoneal Endometriosis
their potential has yet to be achieved in a clinical setting [96]. Peritoneal endometriosis, also termed superficial endome-
triosis, is typically present on the surface of the peritoneum
or the serosa of the peritoneal organs. Lesions of endometri-
12.4 Pathologic Features of Endometriosis osis can be single or in clusters. Grossly, they may present as
raised, cystic, polypoid, or nodular.
Endometriosis, pathologically, is defined by the presence of Peritoneal endometriosis may show different colors under
ectopic functional endometrial tissue, which may be accom- gross or laparoscopic examination, based on the “age” of the
panied by cyclic bleeding induced by hormonal changes and disease. Early lesions are typically colorless and 2–3 mm in
associated with adjacent tissue response and accompanying size (Fig. 12.2).The lesions then may become red (fresh or
adhesion or scar formation. recent bleeding) (Fig. 12.3), blue or black (old or remote
bleeding) (Figs. 12.4 and 12.5), and white (inflammation and
fibrosis) (Fig. 12.2), representing different stages of growth.
12.4.1 Clinicopathologic Types It takes years for an early cystic colorless lesion to develop
into a whitish scar-like lesion. Peritoneal lesions may be
Endometriosis lesions are most commonly located in the pel- multifocal (i.e., other lesions are located within a 2 cm area)
vic surface of peritoneum and ovary (peritoneal endometrio- or multicentric (i.e., lesions are located beyond 2 cm from
sis), in the ovary as cysts lined by endometrioid mucosa the main lesion); as many as 50 lesions may involve the
(ovarian endometriomas), and in pelvic structures between peritoneum.
the rectum and vagina as a solid mass comprised of endome- Presence of endometroid epithelia as well as endometrial
triotic tissue with local adipose and fibromuscular tissue stromal cells varies depending on the disease status.
(deep-infiltrating endometriosis). Rectovaginal endometrio- Typically, both can be found in more than 95% of red lesions,
12 Endometriosis and Endometriosis-Associated Tumors 411
Fig. 12.2 Peritoneal endometriosis. A laparoscopic view of the perito- Fig. 12.4 Peritoneal endometriosis. Shown here are one black,
neal cavity reveals several transparent cystic structures, ranging from 2 “powder-burn” spot in the center and another purple spot in the lower
to 4 mm in size, which represent foci of endometriosis prior to bleeding. area. Both represent relatively old hemorrhage in endometriotic lesions.
Also shown are several white raised nodules (2–3 mm in size) probably Based on the color, the purple lesion is more likely less established than
representing endometriosis with fibrotic changes the black lesion
Fig. 12.3 Peritoneal endometriosis. This laparoscopic view of the Fig. 12.5 Peritoneal endometriosis. Multiple blue and black lesions
peritoneum shows several red-tan, irregular-shaped, slightly raised representing old hemorrhage within foci of endometriosis are shown
lesions, which represent foci of endometriosis with fresh bleeding
and in only 50–60% of bluish or black lesions. The finding (Figs. 12.6, 12.7, 12.8, and 12.9). Large cysts can
clinicopathologic appearance of peritoneal endometriosis is form around the ovary and may acutely rupture, causing
summarized in Table 12.3. release and adherence of their contents to the abdominal
cavity.
12.4.1.2 Ovarian Endometriosis
Endometriosis in the ovary typically presents as a cystic 12.4.1.3 Deep-Infiltrating Endometriosis
lesion with either a single cyst or multilocular cysts. Ovarian Deep-infiltrating endometriosis (also called adenomyosis
endometriotic cysts are termed endometriomas. The cystic externa) is the most severe clinical form of endometriosis
wall is typically thickened because of fibrotic reaction. and in >95% of cases is associated with severe pain. It typi-
Blood clot or condensed blood containing chocolate-like cally presents as solid, multifocal nodules larger than 0.5 cm
material (aptly named “chocolate cyst”) is a common gross in diameter which can grow up to 5–6 cm in size [104, 105].
412 R. C. M. Simmen et al.
Fig. 12.6 Ovarian endometrioma. The ovary shows the presence of a blue
cystic lesion on the surface. A brown viscous material seen oozing onto the
surface of the ovarian mass represents an unopened ovarian endometrioma
a b
Fig. 12.13 (a) Endometriosis showing proliferative-type endometrial glands. Proliferative endometrium is present in this focus of endometriosis.
(b) Endometriosis showing secretory-type endometrial glands. Secretory endometrium is seen in this focus of endometriosis
Fig. 12.14 CD10 is positive for the stromal cells of endometriosis. Stromal cells of endometriosis stained with H&E (left panel) and immunos-
tained with anti-CD10 antibody (right panel)
Fig. 12.23 Endometriosis involving a dermal scar. Resected skin and der-
mal scar are shown on the top panel. Cross section of the specimen shows
Fig. 12.21 Endometriosis involving the endocervical mucosa.
several foci of endometriosis (black spots) and fibrosis (bottom panel)
Microscopically, the superficial endometriosis is located directly
beneath the squamous mucosa
12.5 O
ther Histological Features
of Endometriosis
Fig. 12.32 Endometriotic cyst found in a term pregnancy. High levels alization (left), and the epithelial lining is not obvious until stained for
of progesterone during pregnancy commonly cause decidual changes in cytokeratin 7 (CK7) (right)
endometriosis. The ovarian cyst shown here displays extensive decidu-
Fig. 12.33 Endometriosis with metaplasia. Metaplasia including Fig. 12.34 Endometriosis with reactive epithelial changes. Nuclear
squamous, mucinous, eosinophilic, and ciliated metaplasia may occur atypia is apparent in this focus of endometriosis; however, epithelial
in endometriosis. Mucinous metaplasia can be seen in this image proliferation and mitotic areas are not visible. The atypical nuclei rep-
resent reactive and/or degenerative changes secondary to the hostile
environment created within endometriosis
alter the presence and/or appearance of endometriotic glands.
Endometriosis, particularly cystic lesions with absent or The stromal component of lesions may also be obscured
scant glands, may be a result of the hostile surrounding envi- or effaced by conditions such as fibrosis, smooth muscle
ronment. Those lesions with glands displaying hyperplasia metaplasia, and presence of infiltrates of foamy and pig-
or atypical hyperplasia are collectively referred to as atypical mented histiocytes [112]. The inflammatory status affects
endometriosis (see below). the histologic diagnosis of endometriosis since these
Reactive “nuclear atypia” in endometriotic glands pres- changes can cause associated morbidities such as infec-
ents as enlargement of the nucleus, hyperchromasia, and tion within endometriotic cysts, florid mesothelial hyper-
multiple, tiny nucleoli, with variable degrees of nucleocyto- plasia, peritoneal inclusion cysts, and pseudoxanthomatous
megaly (Figs. 12.34 and 12.35). These atypical nuclei are not salpingitis [112]. Complication of the diagnosis also
uncommon and represent degenerative changes in the toxic arises when other conditions, such as peritoneal leiomyo-
environment. They could be confused with the hyperplastic matosis or gliomatosis, are admixed with the endometri-
process seen in true atypical endometriosis. otic foci.
420 R. C. M. Simmen et al.
Menstrual Fragments
(1)
OVARY
Inflammation
(2)
Estrogen Signaling
Progesterone Signaling
5
ENDOMETRIOMA
Genetic Mutations
Early menarche (3) (ARID1A, PIK3CA)
Late menopause
Infertility
ATYPICAL ENDOMETRIOSIS
Genetic Mutations
(4)
(ARID1A, PIK3CA, PTEN, b-catenin)
5 Others???
OVARIAN CARCINOMA
(OCC, EOC, LGSC) Proliferation
Anchorage-Independence
Invasiveness
Fig. 12.36 Model for multistep progression of benign endometriotic blown carcinoma (4). Not currently known is whether and how additional
lesions to ovarian tumors. Menstrual fragments (from retrograde men- genetic dysregulations result in specific ovarian cancer types (OCC,
struation) attach to the ovary (1), a process promoted by pro-inflammatory, EOC, LGSC). Early menarche, late menopause, and infertility are con-
pro-estrogenic, and anti-progestogenic events (2). Mutations in key genes sidered risk factors for the development of endometrioma to ovarian can-
(e.g., ARID1A, PIK3CA) are considered to promote the progression of an cers (5). OCC, ovarian clear cell carcinoma; EOC, endometrioid ovarian
endometrioma to atypical endometriosis (3) and subsequently to full- carcinoma; LGSC, low-grade serous carcinoma
(γH2AX), and enhanced activation of the apoptotic pathway, low-grade serous carcinomas and exhibits higher potential
was also demonstrated in EOC and contiguous endometrio- for malignancy and recurrence [138]. By contrast, the non-
sis at greater frequencies than in benign endometriotic serous subtype presents mostly at FIGO stage 1 and has
lesions [131]. Thus, benign endometriotic lesions carrying higher overall survival. One study reported that 30% of bor-
cancer-associated mutations in ARID1A and PIK3CA likely derline serous tumors exhibit endometriosis [139]. In a large
serve as precursors for ovarian malignancies. cohort of women with subfertility [140], an increased risk for
In a review of databases pathologically screened for sus- borderline tumors was associated with both ovarian and
pected cases of atypical endometriosis, Stamp et al. [132] extra-ovarian endometriosis. However, in studies conducted
reported that loss of ARID1A in atypical endometriosis was as part of the Ovarian Cancer Association Consortium, an
consistently associated with the development of ARID1A- association between borderline tumors and self-reported
negative endometriosis-associated ovarian cancer. In another endometriosis was not observed [141]. Further studies are
comparison of atypical endometriosis and neoplastic lesions needed to clarify these associations.
by targeted next-generation sequencing, ARID1A and Numerous analyses have shown that extra-pelvic endome-
PIK3CA genes were confirmed as frequently mutated; muta- triosis can also develop into malignant tumors. Abdominal
tions in genes associated with Notch and Wnt/β-catenin endometriosis-associated clear cell carcinoma [142, 143] has
signaling pathways [131] were also noted. Nevertheless, it been reported and additionally metastasis in the bladder and
remains unclear if endometriosis-associated EOC and OCCC the lymph nodes was also detected [140]. Malignant transfor-
evolve similarly, since the expression of other genes such as mation of deep-infiltrating endometriosis has also been
hepatocyte nuclear factor 1β, hypoxia-inducible factor 1α, reported in the pancreas [144] as well as in the appendix lead-
and NFκ-B p65 was not identical between EOC and OCCC ing to intestinal neoplasia [145]; colon/rectum leading to
and their respective coexisting endometriotic lesions [133]. endometrial stromal sarcoma [146]; vagina [147]; small intes-
Related to the latter, it was recently proposed that OCCC and tine [114]; cervix [148]; and bladder [149]. Because these con-
EOC may originate from distinct cells within endometriotic ditions are relatively rare, the mechanisms underlying their
tissue [134] . progression to neoplasia have yet to be extensively assessed.
Another potential contributory mechanism to ovarian
tumorigenesis implicates the stromal compartment [135].
When compared with more distant ovarian stroma, 12.7 Conclusions
endometriosis-associated ovarian tumors display increased
expression of steroidogenic genes and enzymes involved in The possibility that endometriosis can progress from a
steroid hormone synthesis. Early changes in endometriotic benign to a neoplastic condition, in addition to its comorbidi-
stroma involving the activation of steroid hormone produc- ties, provides a strong impetus for the early detection and
tion may thus lead to abnormal proliferation of adjacent effective management of the disease. To address this chal-
ovarian epithelia to promote neoplastic growth. lenge, it is imperative in the long term to focus resources on
(1) the identification of new targets to monitor early disease
onset, eliminate disease recurrence, and allow targeted ther-
12.6.3 Seromucinous Carcinoma and Other apy; (2) the development of new hormonal treatments with
Endometriosis-Associated Tumors minimal side effects and contraindications while preserving
fertility and ovarian function; and (3) the fine-tuning of cur-
Seromucinous tumors and borderline tumors are less com- rent technologies and development of new procedures to
mon neoplasms of the ovary. Seromucinous tumors typically improve accuracy of lesion diagnosis. In the short term, con-
display papillary architecture, an admixture of cell types, tinuous vigilance to reduce recurrence should be part of the
including endocervical-like mucinous, eosinophilic, squa- standard of care for diagnosed patients.
mous, clear, and signet-ring cells, and morphological overlap
with low-grade serous, mucinous, and endometrioid carcino-
mas [136] . These tumors are associated with endometriosis
at a relatively high frequency (23 of 92 cases) [117]. Similar
References
to EOC and OCCC, they show loss of ARID1A expression 1. Burney RO, Giudice LC. Pathogenesis and pathophysiology of
due to somatic inactivation occurring with significant fre- endometriosis. Fertil Steril. 2012;98:511–9.
quency (8 of 24 cases; 33%) [137]. Borderline tumors, first 2. Vercellini P, Vigano P, Somigliana E, Fedele L. Endometriosis:
described as “semi-malignant” ovarian tumors, display bio- pathogenesis and treatment. Nat Rev Endocrinol. 2014;10:261–75.
3. Practice Committee of American Society for Reproductive
logical characteristics intermediate between malignant and Medicine. Multiple gestation associated with infertility ther-
benign tumors and are classified to be of serous or non- apy: an American Society for Reproductive Medicine Practice
serous subtypes. The serous subtype is closely related to Committee opinion. Fertil Steril. 2012;97:825–34.
12 Endometriosis and Endometriosis-Associated Tumors 423
4. Fuldeore M, Yang H, Du EX, et al. Healthcare utilization and costs 25. Gui B, Valentini AL, Ninivaggi V, et al. Shining light in a dark
in women diagnosed with endometriosis before and after diag- landscape: MRI evaluation of unusual localization of endometrio-
nosis: a longitudinal analysis of claims databases. Fertil Steril. sis. Diagn Interv Radiol. 2017;23:272–81.
2015;103:163–71. 26. DiVasta AD, Vitonis AF, Laufer MR, Missmer SA. Spectrum
5. Soliman AM, Yang H, Du EX, Kelley C, Winkel C. The direct and of symptoms in women diagnosed with endometriosis during
indirect costs associated with endometriosis: a systematic litera- adolescence vs adulthood. Am J Obstet Gynecol. 2018;218:324
ture review. Hum Reprod. 2016;31:712–22. e1–e11.
6. Parasar P, Ozcan P, Terry KL. Endometriosis: epidemiology, 27. Mounsey AL, Wilgus A, Slawson DC. Diagnosis and management
diagnosis and clinical management. Curr Obstet Gynecol Rep. of endometriosis. Am Fam Physician. 2006;74:594–600.
2017;6:34–41. 28. Nisenblat V, Bossuyt PM, Shaikh R, et al. Blood biomarkers for
7. Bedaiwy MA, Alfaraj S, Yong P, Casper R. New develop- the non-invasive diagnosis of endometriosis. Cochrane Database
ments in the medical treatment of endometriosis. Fertil Steril. Syst Rev. 2016;5:CD012179.
2017;107:555–65. 29. Seeber B, Sammel MD, Fan X, et al. Panel of markers can accu-
8. Krishnamoorthy K, Decherney AH. Genetics of endometriosis. rately predict endometriosis in a subset of patients. Fertil Steril.
Clin Obstet Gynecol. 2017;60:531–8. 2008;89:1073–81.
9. Jiang QY, Wu RJ. Growth mechanisms of endometriotic cells in 30. Drosdzol-Cop A, Skrzypulec-Plinta V. Selected cytokines and gly-
implanted places: a review. Gynecol Endocrinol. 2012;28:562–7. codelin A levels in serum and peritoneal fluid in girls with endo-
10. Smorgick N, Marsh CA, As-Sanie S, Smith YR, Quint metriosis. J Obstet Gynaecol Res. 2012;38:1245–53.
EH. Prevalence of pain syndromes, mood conditions, and asthma 31. Seifer BJ, Su D, Taylor HS. Circulating miRNAs in murine exper-
in adolescents and young women with endometriosis. J Pediatr imental endometriosis. Reprod Sci. 2017;24:376–81.
Adolesc Gynecol. 2013;26:171–5. 32. Wang W-T, Zhao Y-N, Han B-W, Hong S-J, Chen Y-Q. Circulating
11. Sinaii N, Cleary SD, Ballweg ML, Nieman LK, Stratton P. High microRNAs identified in a genome-wide serum microRNA
rates of autoimmune and endocrine disorders, fibromyalgia, expression analysis as noninvasive biomarkers for endometriosis.
chronic fatigue syndrome and atopic diseases among women with J Clin Endocrinol Metab. 2013;98:281–9.
endometriosis: a survey analysis. Hum Reprod. 2002;17:2715–24. 33. Fritz MA, Speroff L. Clinical gynecologic endocrinology and
12. Fassbender A, Burney RO, O DF, D’Hooghe T, Giudice L. Update infertility. Philadelphia: Wolters Kluwer Health/Lippincott
on biomarkers for the detection of endometriosis. Biomed Res Int. Williams & Wilkins; 2011.
2015;2015:130854. 34. Schenken RS. Endometriosis: pathogenesis, clinical features, and
13. Somigliana E, Vigano P, Benaglia L, et al. Management of endome- diagnosis. Waltham, MA: Walters Kluwer Health; 2016.
triosis in the infertile patient. Semin Reprod Med. 2017;35:31–7. 35. Sourial S, Tempest N, Hapangama DK. Theories on the pathogen-
14. Gemmell LC, Webster KE, Kirtley S, et al. The management of esis of endometriosis. Int J Reprod Med. 2014;2014:9.
menopause in women with a history of endometriosis: a system- 36. D’Hooghe TM, Bambra CS, Suleman MA, et al. Development of
atic review. Hum Reprod Update. 2017;23:481–500. a model of retrograde menstruation in baboons (Papio anubis).
15. Wilbur MA, Shih IM, Segars JH, Fader AN. Cancer implica- Fertil Steril. 1994;62:635–8.
tions for patients with endometriosis. Semin Reprod Med. 37. Dehoux JP, Defrere S, Squifflet J, et al. Is the baboon model appro-
2017;35:110–6. priate for endometriosis studies? Fertil Steril. 2011;96:728–33 e3.
16. Marchino GL, Gennarelli G, Enria R, et al. Diagnosis of pelvic 38. Song X, Zhu L, Ding J, Xu T, Lang J. Clinical characteristics of
endometriosis with use of macroscopic versus histologic findings. congenital cervical atresia and associated endometriosis among 96
Fertil Steril. 2005;84:12–5. patients. Int J Gynaecol Obstet. 2016;134:252–5.
17. Jansen RP, Russell P. Nonpigmented endometriosis: clinical, 39. Sasson IE, Taylor HS. Stem cells and the pathogenesis of endome-
laparoscopic, and pathologic definition. Am J Obstet Gynecol. triosis. Ann N Y Acad Sci. 2008;1127:106–15.
1986;155:1154–9. 40. Halme J, Hammond MG, Hulka JF, Raj SG, Talbert LM. Retrograde
18. Albee RB Jr, Sinervo K, Fisher DT. Laparoscopic excision of menstruation in healthy women and in patients with endometrio-
lesions suggestive of endometriosis or otherwise atypical in sis. Obstet Gynecol. 1984;64:151–4.
appearance: relationship between visual findings and final histo- 41. Hufnagel D, Li F, Cosar E, et al. The role of stem cells in the etiol-
logic diagnosis. J Minim Invasive Gynecol. 2008;15:32–7. ogy and pathophysiology of endometriosis. Semin Reprod Med.
19. Walter AJ, Hentz JG, Magtibay PM, Cornella JL, Magrina 2015;33:333–40.
JF. Endometriosis: correlation between histologic and visual find- 42. Vinatier D, Orazi G, Cosson M, Dufour P. Theories of endome-
ings at laparoscopy. Am J Obstet Gynecol. 2001;184:1407–11; triosis. Eur J Obstet Gynecol Reprod Biol. 2001;96:21–34.
discussion 11–3 43. Matsumoto T, Yamazaki M, Takahashi H, et al. Distinct β-catenin
20. American Society for Reproductive Medicine. Revised American and PIK3CA mutation profiles in endometriosis-associated ovar-
Society for Reproductive Medicine classification of endometrio- ian endometrioid and clear cell carcinomas. Am J Clin Pathol.
sis: 1996. Fertil Steril. 1997;67:817–21. 2015;144:452–63.
21. Milingos S, Protopapas A, Drakakis P, et al. Laparoscopic man- 44. Troncon JK, Zani AC, Vieira AD, et al. Endometriosis in a patient
agement of patients with endometriosis and chronic pelvic pain. with Mayer-Rokitansky-Kuster-Hauser syndrome. Case Rep
Ann N Y Acad Sci. 2003;997:269–73. Obstet Gynecol. 2014;2014:376231.
22. Johnson NP, Hummelshoj L, Adamson GD, et al. World 45. Mok-Lin EY, Wolfberg A, Hollinquist H, Laufer
Endometriosis Society consensus on the classification of endome- MR. Endometriosis in a patient with Mayer-Rokitansky-Kuster-
triosis. Hum Reprod. 2017;32:315–24. Hauser syndrome and complete uterine agenesis: evidence to
23. Hirsch M, Duffy JMN, Deguara CS, Davis CJ, Khan support the theory of coelomic metaplasia. J Pediatr Adolesc
KS. Diagnostic accuracy of Cancer Antigen 125 (CA125) for Gynecol. 2010;23:e35–7.
endometriosis in symptomatic women: A multi-center study. Eur J 46. Cho MK, Kim CH, Oh ST. Endometriosis in a patient with
Obstet Gynecol Reprod Biol. 2017;210:102–7. Rokitansky-Kuster-Hauser syndrome. J Obstet Gynaecol Res.
24. Furuya M, Masuda H, Hara K, et al. ZEB1 expression is a poten- 2009;35:994–6.
tial indicator of invasive endometriosis. Acta Obstet Gynecol 47. Pluchino N, Taylor HS. Endometriosis and stem cell trafficking.
Scand. 2017;96:1128–35. Reprod Sci. 2016;23:1616–9.
424 R. C. M. Simmen et al.
48. Wang Y, Mang M, Wang Y, et al. Tubal origin of ovarian endome- 69. Zhao Y, Chen Y, Kuang Y, et al. Multiple Beneficial roles of
triosis and clear cell and endometrioid carcinoma. Am J Cancer repressor of estrogen receptor activity (REA) in suppressing the
Res. 2015;5:869–79. progression of endometriosis. Endocrinology. 2016;157:900–12.
49. Zheng W, Li N, Wang J, et al. Initial endometriosis showing direct 70. Lv J, Zhu Q, Jia X, Yu N, Li Q. In vitro and in vivo effects of tumor
morphologic evidence of metaplasia in the pathogenesis of ovar- suppressor gene PTEN on endometriosis: an experimental study.
ian endometriosis. Int J Gynecol Pathol. 2005;24:164–72. Med Sci Monit. 2016;22:3727–36.
50. Li J, Abushahin N, Pang S, et al. Tubal origin of ‘ovarian’ low- 71. Govatati S, Kodati VL, Deenadayal M, et al. Mutations in the
grade serous carcinoma. Mod Pathol. 2011;24:1488. PTEN tumor gene and risk of endometriosis: a case-control study.
51. Yuan Z, Wang L, Wang Y, et al. Tubal origin of ovarian endome- Hum Reprod. 2014;29:324–36.
triosis. Mod Pathol. 2014;27:1154–62. 72. Yoo JY, Kim TH, Fazleabas AT, et al. KRAS activation and over-
52. Corzo C, Iniesta MD, Patrono MG, Lu KH, Ramirez PT. Role expression of SIRT1/BCL6 contributes to the pathogenesis of
of fallopian tubes in the development of ovarian cancer. J Minim endometriosis and progesterone resistance. Sci Rep. 2017;7:6765.
Invasive Gynecol. 2017;24:230–4. 73. Naqvi H, Ilagan Y, Krikun G, Taylor HS. Altered genome-wide
53. Erickson BK, Conner MG, Landen CN Jr. The role of the fallo- methylation in endometriosis. Reprod Sci. 2014;21:1237–43.
pian tube in the origin of ovarian cancer. Am J Obstet Gynecol. 74. Houshdaran S, Nezhat CR, Vo KC, et al. Aberrant endometrial
2013;209:409–14. DNA methylome and associated gene expression in women with
54. Li J, Fadare O, Xiang L, Kong B, Zheng W. Ovarian serous carci- endometriosis. Biol Reprod. 2016;95:93.
noma: recent concepts on its origin and carcinogenesis. J Hematol 75. Anglesio MS, Papadopoulos N, Ayhan A, et al. Cancer-associated
Oncol. 2012;5:8. mutations in endometriosis without cancer. N Engl J Med.
55. Long Roche KC, Abu-Rustum NR, Nourmoussavi M, Zivanovic 2017;376:1835–48.
O. Risk-reducing salpingectomy: let us be opportunistic. Cancer. 76. Zondervan KT, Rahmioglu N, Morris AP, et al. Beyond endome-
2017;123:1714–20. triosis genome-wide association study: from genomics to phe-
56. Walker JL, Powell CB, Chen LM, et al. Society of Gynecologic nomics to the patient. Semin Reprod Med. 2016;34:242–54.
Oncology recommendations for the prevention of ovarian cancer. 77. Bulun SE, Monsavais D, Pavone ME, et al. Role of estro-
Cancer. 2015;121:2108–20. gen receptor- beta in endometriosis. Semin Reprod Med.
57. Perez-Lopez FR, Ceausu I, Depypere H, et al. Interventions to 2012;30:39–45.
reduce the risk of ovarian and fallopian tube cancer: A European 78. Han SJ, Jung SY, Wu SP, et al. Estrogen receptor beta modulates
Menopause and Andropause Society Position Statement. apoptosis complexes and the inflammasome to drive the pathogen-
Maturitas. 2017;100:86–91. esis of endometriosis. Cell. 2015;163:960–74.
58. Aghajanova L, Giudice LC. Molecular evidence for differences 79. Simmen RC, Kelley AS. Reversal of fortune: estrogen receptor-
in endometrium in severe versus mild endometriosis. Reprod Sci. beta in endometriosis. J Mol Endocrinol. 2016;57:F23–7.
2011;18:229–51. 80. Al-Sabbagh M, Lam EW, Brosens JJ. Mechanisms of endometrial
59. Pabona JM, Simmen FA, Nikiforov MA, et al. Kruppel-like factor progesterone resistance. Mol Cell Endocrinol. 2012;358:208–15.
9 and progesterone receptor coregulation of decidualizing endo- 81. Grandi G, Mueller MD, Papadia A, et al. Inflammation influences
metrial stromal cells: implications for the pathogenesis of endo- steroid hormone receptors targeted by progestins in endometrial
metriosis. J Clin Endocrinol Metab. 2012;97:E376–92. stromal cells from women with endometriosis. J Reprod Immunol.
60. Fassbender A, Verbeeck N, Bornigen D, et al. Combined 2016;117:30–8.
mRNA microarray and proteomic analysis of eutopic endome- 82. Tosti C, Biscione A, Morgante G, et al. Hormonal therapy for
trium of women with and without endometriosis. Hum Reprod. endometriosis: from molecular research to bedside. Eur J Obstet
2012;27:2020–9. Gynecol Reprod Biol. 2017;209:61–6.
61. Tamaresis JS, Irwin JC, Goldfien GA, et al. Molecular classifica- 83. Donnez J, Taylor RN, Taylor HS. Partial suppression of estra-
tion of endometriosis and disease stage using high-dimensional diol: a new strategy in endometriosis management? Fertil Steril.
genomic data. Endocrinology. 2014;155:4986–99. 2017;107:568–70.
62. Browne H, Taylor H. HOXA10 expression in ectopic endometrial 84. Taylor HS, Giudice LC, Lessey BA, et al. Treatment of
tissue. Fertil Steril. 2006;85:1386–90. endometriosis-associated pain with elagolix, an oral GnRH antag-
63. Fang Z, Yang S, Gurates B, et al. Genetic or enzymatic disruption onist. N Engl J Med. 2017;377:28–40.
of aromatase inhibits the growth of ectopic uterine tissue. J Clin 85. Matalliotakis I, Cakmak H, Matalliotakis M, Kappou D, Arici
Endocrinol Metab. 2002;87:3460–6. A. High rate of allergies among women with endometriosis. J
64. Bulun SE, Zeitoun KM, Takayama K, Sasano H. Estrogen biosyn- Obstet Gynaecol. 2012;32:291–3.
thesis in endometriosis: molecular basis and clinical relevance. J 86. Harris HR, Costenbader KH, Mu F, et al. Endometriosis and the
Mol Endocrinol. 2000;25:35–42. risks of systemic lupus erythematosus and rheumatoid arthritis in
65. Han SJ, Hawkins SM, Begum K, et al. A new isoform of steroid the Nurses’ Health Study II. Ann Rheum Dis. 2016;75:1279–84.
receptor coactivator-1 is crucial for pathogenic progression of 87. Ahn SH, Khalaj K, Young SL, et al. Immune-inflammation gene
endometriosis. Nat Med. 2012;18:1102–11. signatures in endometriosis patients. Fertil Steril. 2016;106:1420–
66. Lin SC, Li YH, Wu MH, et al. Suppression of COUP-TFII by pro- 31 e7.
inflammatory cytokines contributes to the pathogenesis of endo- 88. Monsanto SP, Edwards AK, Zhou J, et al. Surgical removal of
metriosis. J Clin Endocrinol Metab. 2014;99:E427–37. endometriotic lesions alters local and systemic proinflammatory
67. Yu J, Boicea A, Barrett KL, et al. Reduced connexin 43 in eutopic cytokines in endometriosis patients. Fertil Steril. 2016;105:968–
endometrium and cultured endometrial stromal cells from sub- 77 e5.
jects with endometriosis. Mol Hum Reprod. 2014;20:260–70. 89. Burns KA, Thomas SY, Hamilton KJ, et al. Early endometriosis in
68. Heard ME, Simmons CD, Simmen FA, Simmen RC. Kruppel-like females is directed by immune-mediated estrogen receptor alpha
factor 9 deficiency in uterine endometrial cells promotes ectopic and IL-6 cross-talk. Endocrinology. 2018;159:103–18.
lesion establishment associated with activated notch and hedge- 90. Greaves E, Temp J, Esnal-Zufiurre A, et al. Estradiol is a critical
hog signaling in a mouse model of endometriosis. Endocrinology. mediator of macrophage-nerve cross talk in peritoneal endome-
2014;155:1532–46. triosis. Am J Pathol. 2015;185:2286–97.
12 Endometriosis and Endometriosis-Associated Tumors 425
91. Zhao Y, Gong P, Chen Y, et al. Dual suppression of estrogenic and a common and under-recognised form of endometriosis. J Clin
inflammatory activities for targeting of endometriosis. Sci Transl Pathol. 2009;62:530–3.
Med. 2015;7:271ra9. 112. Clement PB. The pathology of endometriosis: a survey of the
92. Grandi G, Mueller M, Bersinger N, et al. Progestin suppressed many faces of a common disease emphasizing diagnostic pitfalls
inflammation and cell viability of tumor necrosis factor-alpha- and unusual and newly appreciated aspects. Adv Anat Pathol.
stimulated endometriotic stromal cells. Am J Reprod Immunol. 2007;14:241–60.
2016;76:292–8. 113. Wei JJ, William J, Bulun S. Endometriosis and ovarian cancer:
93. Ahmad SF, Akoum A, Horne AW. Selective modulation of the a review of clinical, pathologic, and molecular aspects. Int J
prostaglandin F2alpha pathway markedly impacts on endometrio- Gynecol Pathol. 2011;30:553–68.
sis progression in a xenograft mouse model. Mol Hum Reprod. 114. Ogi Y, Yamaguchi T, Kinugasa Y, et al. A case of small intestinal
2015;21:905–16. endometrioid adenocarcinoma. Surg Case Rep. 2016;2:97.
94. Rudzitis-Auth J, Menger MD, Laschke MW. Resveratrol is a 115. Scott RB. Malignant changes in endometriosis. Obstet Gynecol.
potent inhibitor of vascularization and cell proliferation in experi- 1953;2:283–9.
mental endometriosis. Hum Reprod. 2013;28:1339–47. 116. Krawczyk N, Banys-Paluchowski M, Schmidt D, Ulrich U,
95. Taguchi A, Wada-Hiraike O, Kawana K, et al. Resveratrol sup- Fehm T. Endometriosis-associated malignancy. Geburtshilfe
presses inflammatory responses in endometrial stromal cells Frauenheilkd. 2016;76:176–81.
derived from endometriosis: a possible role of the sirtuin 1 path- 117. Karpathiou G, Chauleur C, Corsini T, et al. Seromucinous ovarian
way. J Obstet Gynaecol Res. 2014;40:770–8. tumor. A comparison with the rest of ovarian epithelial tumors.
96. Mendes da Silva D, Gross LA, Neto EPG, Lessey BA, Savaris Ann Diagn Pathol. 2017;27:28–33.
RF. The use of resveratrol as an adjuvant treatment of pain 118. Somigliana E, Parazzini F, Stoppelli S, et al. Association between
in endometriosis: a randomized clinical trial. J Endocr Soc. endometriosis and cancer: a comprehensive review and a critical
2017;1:359–69. analysis of clinical and epidemiological evidence. Gynecol Oncol.
97. Nezhat C, Li A, Falik R, et al. Bowel endometriosis: diagnosis and 2006;101:331–41.
management. Am J Obstet Gynecol. 2017;218(6):549–62. 119. Schnack TH, Høgdall E, Thomsen LN, Høgdall C. Demographic,
98. Jaiman S, Gundabattula SR, Pochiraju M, Sangireddy JR. Polypoid clinical, and prognostic factors of ovarian clear cell adenocarci-
endometriosis of the cervix: a case report and review of the litera- nomas according to endometriosis status. Int J Gynecol Cancer.
ture. Arch Gynecol Obstet. 2014;289:915–20. 2017;27:1804–12.
99. Benacerraf BR, Groszmann Y, Hornstein MD, Bromley B. Deep 120. Park J-Y, Kim D-Y, Suh D-S, et al. Significance of ovarian endo-
infiltrating endometriosis of the bowel wall: the comet sign. J metriosis on the prognosis of ovarian clear cell carcinoma. Int J
Ultrasound Med. 2015;34:537–42. Gynecol Cancer. 2018;28:11–8.
100. Yantiss RK, Clement PB, Young RH. Endometriosis of the intes- 121. Kim SI, Lee JW, Lee M, et al. Genomic landscape of ovarian clear
tinal tract: a study of 44 cases of a disease that may cause diverse cell carcinoma via whole exome sequencing. Gynecol Oncol.
challenges in clinical and pathologic evaluation. Am J Surg Pathol. 2018;148:375–82.
2001;25:445–54. 122. Czernobilsky B, Morris WJ. A histologic study of ovarian endo-
101. Nezhat C, Kazerooni T, Berker B, et al. Laparoscopic manage- metriosis with emphasis on hyperplastic and atypical changes.
ment of hepatic endometriosis: report of two cases and review of Obstet Gynecol. 1979;53:318–23.
the literature. J Minim Invasive Gynecol. 2005;12:196–200. 123. Fukunaga M, Nomura K, Ishikawa E, Ushigome S. Ovarian atypi-
102. Seval MM, Cavkaytar S, Atak Z, Guresci S. Postcoital bleeding cal endometriosis: its close association with malignant epithelial
due to cervical endometriosis. BMJ Case Rep. 2013;2013 tumours. Histopathology. 1997;30:249–55.
103. Hensen JH, Van Breda Vriesman AC, Puylaert JB. Abdominal 124. Terada T. Endometrioid adenocarcinoma of the ovary arising in
wall endometriosis: clinical presentation and imaging fea- atypical endometriosis. Int J Clin Exp Pathol. 2012;5:924–7.
tures with emphasis on sonography. AJR Am J Roentgenol. 125. Moll UM, Chumas JC, Chalas E, Mann WJ. Ovarian carcinoma
2006;186:616–20. arising in atypical endometriosis. Obstet Gynecol. 1990;75:537–9.
104. Koninckx PR, Ussia A, Adamyan L, et al. Deep endometriosis: 126. Tanase Y, Furukawa N, Kobayashi H, Matsumoto T. Malignant
definition, diagnosis, and treatment. Fertil Steril. 2012;98:564–71. transformation from endometriosis to atypical endometriosis and
105. Tosti C, Pinzauti S, Santulli P, et al. Pathogenetic mechanisms of finally to endometrioid adenocarcinoma within 10 years. Case
deep infiltrating endometriosis. Reprod Sci. 2015;22:1053–9. Rep Oncol. 2013;6:480–4.
106. Redwine DB. Diaphragmatic endometriosis: diagnosis, surgical 127. Wiegand KC, Shah SP, Al-Agha OM, et al. ARID1A mutations
management, and long-term results of treatment. Fertil Steril. in endometriosis-associated ovarian carcinomas. N Engl J Med.
2002;77:288–96. 2010;363:1532–43.
107. Nezhat C, Main J, Paka C, et al. Multidisciplinary treat- 128. Ayhan A, Mao T-L, Seckin T, et al. Loss of ARID1A expression
ment for thoracic and abdominopelvic endometriosis. JSLS. is an early molecular event in tumor progression from ovarian
2014;18:e2014.00312. endometriotic cyst to clear cell and endometrioid carcinoma. Int J
108. Parker RL, Dadmanesh F, Young RH, Clement PB. Polypoid endo- Gynecol Cancer. 2012;22:1310.
metriosis: a clinicopathologic analysis of 24 cases and a review of 129. Yamamoto S, Tsuda H, Takano M, et al. Loss of ARID1A protein
the literature. Am J Surg Pathol. 2004;28:285–97. expression occurs as an early event in ovarian clear-cell carcinoma
109. Stewart CJ, Bharat C. Clinicopathological and immunohisto- development and frequently coexists with PIK3CA mutations.
logical features of polypoid endometriosis. Histopathology. Mod Pathol. 2012;25:615.
2016;68:398–404. 130. Lakshminarasimhan R, Andreu-Vieyra C, Lawrenson K, et al.
110. Syrcle S, Pelch K, Schroder A, et al. Altered gene expression Down-regulation of ARID1A is sufficient to initiate neoplastic
profile in vaginal polypoid endometriosis resembles peritoneal transformation along with epigenetic reprogramming in non-
endometriosis and is consistent with increased local estrogen pro- tumorigenic endometriotic cells. Cancer Lett. 2017;401:11–9.
duction. Gynecol Obstet Investig. 2011;71:77–86. 131. Chene G, Ouellet V, Rahimi K, et al. The ARID1A pathway in ovar-
111. Boyle DP, McCluggage WG. Peritoneal stromal endometriosis: ian clear cell and endometrioid carcinoma, contiguous endometriosis,
a detailed morphological analysis of a large series of cases of and benign endometriosis. Int J Gynaecol Obstet. 2015;130:27–30.
426 R. C. M. Simmen et al.
132. Stamp JP, Gilks CB, Wesseling M, et al. BAF250a expression in ian cancer: a pooled analysis of case–control studies. Lancet
atypical endometriosis and endometriosis-associated ovarian can- Oncol. 2012;13:385–94.
cer. Int J Gynecol Cancer. 2016;26:825–32. 142. Liu H, Leng J, Lang J, Cui Q. Clear cell carcinoma arising from
133. Er TK, Su YF, Wu CC, et al. Targeted next-generation sequenc- abdominal wall endometriosis: a unique case with bladder and
ing for molecular diagnosis of endometriosis-associated ovarian lymph node metastasis. World J Surg Oncol. 2014;12:51.
cancer. J Mol Med. 2016;94:835–47. 143. Ferrandina G, Palluzzi E, Fanfani F, et al. Endometriosis-
134. Cochrane DR, Tessier-Cloutier B, Lawrence KM, et al. Clear cell associated clear cell carcinoma arising in caesarean section scar:
and endometrioid carcinomas: are their differences attributable to a case report and review of the literature. World J Surg Oncol.
distinct cells of origin? J Pathol. 2017;243:26–36. 2016;14:300.
135. Blanco LZ Jr, Kuhn E, Morrison JC, et al. Steroid hormone 144. Mederos MA, Villafañe N, Dhingra S, et al. Pancreatic endo-
synthesis by the ovarian stroma surrounding epithelial ovarian metrial cyst mimics mucinous cystic neoplasm of the pancreas.
tumors: a potential mechanism in ovarian tumorigenesis. Mod World J Gastroenterol. 2017;23:1113.
Pathol. 2017;30:563–76. 145. Klingbeil KD, Azab B, Moller MG. Low-grade appendiceal muci-
136. Taylor J, McCluggage WG. Ovarian seromucinous carcinoma: nous neoplasm and endometriosis of the appendix. World J Surg
report of a series of a newly categorized and uncommon neo- Oncol. 2017;15:226.
plasm. Am J Surg Pathol. 2015;39:983–92. 146. Kilzieh R, Rakislova N, Torné A, et al. Endometrial stromal sar-
137. Wu CH, Mao TL, Vang R, et al. Endocervical-type mucinous coma arising in colorectal endometriosis. Int J Gynecol Pathol.
borderline tumors are related to endometrioid tumors based on 2017;36:433–7.
mutation and loss of expression of ARID1A. Int J Gynecol Pathol. 147. Cozzolino M, Nasioudis D, Sisti G, Coccia ME. Malignant
2012;31:297–303. transformation of vaginal endometriosis—a review of literature.
138. Hauptmann S, Friedrich K, Redline R, Avril S. Ovarian borderline Gynecol Obstet Investig. 2017;82:105–12.
tumors in the 2014 WHO classification: evolving concepts and 148. Hashiguchi M, Kai K, Nishiyama S, et al. Clear cell carcinoma
diagnostic criteria. Virchows Arch. 2017;470:125–42. of the uterine cervix presented as a submucosal tumor arising in
139. Akbarzadeh-Jahromi M, Shekarkhar G, Sari Aslani F, et al. a background of cervical endometriosis. Int J Gynecol Pathol.
Prevalence of endometriosis in malignant epithelial ovarian tumor. 2018;37:88–92.
Arch Iran Med. 2015;18:844–8. 149. Tarumi Y, Mori T, Kusuki I, et al. Endometrioid adenocarcinoma
140. Buis CC, van Leeuwen FE, Mooij TM, et al. Increased risk for arising from deep infiltrating endometriosis involving the bladder:
ovarian cancer and borderline ovarian tumours in subfertile a case report and review of the literature. Gynecol Oncol Rep.
women with endometriosis. Hum Reprod. 2013;28:3358–69. 2015;13:68.
141. Pearce CL, Templeman C, Rossing MA, et al. Association
between endometriosis and risk of histological subtypes of ovar-
Complications of Early Pregnancy
and Gestational Trophoblastic Diseases 13
Philip P. C. Ip, Yan Wang, and Annie N. Y. Cheung
© Science Press & Springer Nature Singapore Pte Ltd. 2019 427
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_13
428 P. P. C. Ip et al.
Trophoectoderm
Inner cell mass
Fig. 13.6 Photomicrographs showing CMV inclusions (arrows) found Fig. 13.7 Photomicrograph of a second-trimester placenta of a fetus
at the chorionic villi of a case of intrauterine death. Immunohistochemistry diagnosed to have trisomy 21 during prenatal screening
confirms CMV antigens (inset)
13.3.2 Ectopic Pregnancy that in cases when continued growth of residual incompletely
removed trophoblasts infiltrated deeply in the muscle wall of
13.3.2.1 R isk Factors and Clinical the fallopian tube or other sites persistent pain and irregular
Manifestations vaginal bleeding with persistently raised hCG in blood and
Ectopic pregnancy refers to the implantation of conceptus urine may result. This may be described as persistent ectopic
outside uterus or at unusual sites within the uterus. It occurs pregnancy [21]. Ectopic pregnancy may occur, mainly in
most commonly at the fallopian tube, followed by the ova- cases treated by non-radical surgery.
ries, uterine cornu, cervix, vagina, and other organs. Tubal
pregnancy occurs mainly at the ampulla, followed by isth- 13.3.2.3 P
athological Examination of Excised
mus and fimbria [14]. Fallopian Tube
In recent decades, the incidence of ectopic pregnancy has
increased significantly, from 0.4% to 1.6% according to some Macroscopic Examination
studies [19]. The risk factors include history of tubal surgery, Fallopian tube excised for ectopic pregnancy should be
sexually transmitted diseases, spontaneous or therapeutic examined for presence or absence of rupture and embryos.
abortion, pregnancy at advanced maternal age, use of intra- The fallopian tube may display variable degree of local or
uterine contraceptive device, as well as smoking. In addition, diffuse dilatation and serosal congestion. Blood clot and pla-
structural abnormalities of the fallopian tube including con- cental tissue may protrude from rupture. Pathologist should
genital defects, chronic salpingitis, recanalization after tubal check carefully whether there is amniotic sac or fetal tissue
ligation, as well as cilia dysfunction also increase the risk. in the blood clot. If embryo or placental tissue cannot be
Chronic salpingitis includes non- granulomatous (e.g., identified at the ruptured site, the entire fallopian tube should
Chlamydia trachomatis and mycoplasma) and granulomatous be cut open for examination. Adequate number of blocks
(e.g., Mycobacterium tuberculosis) salpingitis. should be taken.
It is estimated that 2–10% of pregnancies following
assisted reproduction implant at the fallopian tube although Microscopic Examination
the reason is unclear. It may be due to exogenous Microscopic examination is important to identify chorionic
gonadotropin-induced secondary changes in hormone levels, villi at possible ectopic sites. If villi are absent, the presence
or due to diseases underlying infertility [20]. The increased of extravillous implantation-site intermediate trophoblasts
application of assisted reproduction may thus increase the can still confirm the diagnosis of ectopic pregnancy. The
incidence of ectopic pregnancy. Tubal inflammation with implantation-site intermediate trophoblast is often found at
scarring and incomplete obstruction of the tube may also the smooth muscle bundles of the fallopian tube infiltrating
explain the increased incidence. Moreover, the increased to the serosa or replacing the blood vessel wall.
sensitivity of imaging and laboratory technologies may have The villous morphology may be normal in tubal pregnan-
diagnosed early ectopic pregnancy that cannot be diagnosed cies. In some cases, villous morphological changes may
in previous era. occur, such as villous fibrosis, edema, and hydropic degener-
Common symptoms of tubal pregnancy include abdomi- ation (Fig. 13.9). While one should bear in mind the rare pos-
nal pain and irregular vaginal bleeding, accompanied by his- sibility of ectopic hydatidiform mole and choriocarcinoma,
tory of amenorrhea. Some patients may develop shock due to avoiding overdiagnosis is equally important [22, 23]. Florid
intra-abdominal hemorrhage. However, one must bear in polar trophoblastic proliferation in association with hydropic
mind that atypical medical history and negative signs cannot villi as well as extensive aggregates of extravillous tropho-
completely rule out ectopic pregnancy. In particular, unrup- blast may be found in tubal ectopic gestation. On the other
tured ectopic pregnancy may be painless with even absence hand, the absence of circumferential trophoblastic prolifera-
of adnexal tenderness during examination. Positive preg- tion and conspicuous stromal karyorrhexis should shed doubt
nancy test supports diagnosis of pregnancy but cannot distin- on the diagnosis of ectopic hydatidiform mole. Examination
guish between abnormal intrauterine pregnancy and ectopic of all sampled tissue as well as adjunct immunohistochemical
pregnancy. and molecular tests can usually allow proper diagnosis.
13.4.2 Epidemiology
Table 13.2 WHO classification of gestational trophoblastic disease
Molar pregnancies Gestational trophoblastic disease has the highest incidence
Complete hydatidiform mole in African, Asian, and Latin American countries than in
Partial hydatidiform mole
Caucasian societies. It is estimated that gestational tropho-
Invasive hydatidiform mole
blastic disease occurs in 2–13/1000 pregnancies in Asia
Abnormal (non-molar) villous
lesions while the incidence in the United States and Europe is 0.5–
Nonneoplastic lesions 1.84/1000 pregnancies [26–36]. It is noteworthy that there
Exaggerated placental site may be variations in the diagnostic definitions of hydatidi-
Placental site nodule and form moles and persistent trophoblastic disease. Moreover,
plaque some uterine curettings or tissue passed per vagina related to
Neoplasms clinical diagnosis of missed abortion has not been sent for
Choriocarcinoma
histopathological examination leading to underdiagnosis of
Placental site trophoblastic
tumor hydatidiform mole. The incidence of choriocarcinoma is
Epithelioid trophoblastic often difficult to appraise due to lack of diagnostic tissue
tumor biopsy to differentiate postmolar choriocarcinoma from
invasive mole. It is reported to occur in around 1–9 in 40,000
pregnancies [37]. The incidences of PSTT and ETT are
blasts of the chorionic villi, implantation site, or chorion [13, lower than choriocarcinoma.
24], with close association with underlying chromosomal The main factors associated with increased risk of gesta-
composition [4, 5, 24, 25]. They may be considered as trans- tional trophoblastic disease include maternal age of <15 or
plants with pure or dominant paternal origin that invades the >40 years and a prior hydatidiform mole, although dietary and
maternal body, sometimes in uncontrolled manner. socioeconomical statuses have also been implicated [28, 30,
Hydatidiform moles or molar pregnancies are the most com- 36, 38–41]. Indeed, it is observed that the incidence of hyda-
mon type of GTD) and the majority regresses after uterine tidiform moles in Asia has apparently been declining [42, 43].
evacuation. However, a small proportion of hydatidiform Reduction in birth rates as well as improvement in diet and
moles may develop persistently elevated serum human cho- economy have been proposed to be the attributing factors.
rionic gonadotropin (hCG) in association with possibility of
developing choriocarcinoma. Treatment therefore needs to
be considered. Such cases are grouped under gestational tro- 13.5 Hydatidiform Mole
phoblastic neoplasia (GTN) cases together with the frankly
malignant members of the GTD) family (Table 13.2) [26]. It 13.5.1 Complete Hydatidiform Mole
is noteworthy that gestational trophoblastic neoplasia can
occur after non-molar pregnancy, including spontaneous 13.5.1.1 Genetic Composition
abortion, ectopic pregnancy, or even term pregnancy with In complete hydatidiform mole, there is villous hydropic
live delivery. change with little or no fetal development. It is usually
434 P. P. C. Ip et al.
Table 13.3 Similarities and differences between types of molar and non-molar diseases with hydropic chorionic villi [modified from [4, 5, 24]
Complete mole Early complete mole Partial mole Hydropic abortus
Clinical features Second-trimester vaginal First-trimester missed Late first or early Late first or early
bleeding, high hCG levels abortion second-trimester missed second-trimester missed
abortion abortion
Gross features Large amount of tissue with Usually normal Few vesicles, gestational Small amount, may have
prominent vesicles sac may present gestational sac
Fetal parts Absent Absent Present but may be Usually absent
abnormal
Villous population Wide range of large and small Mainly small hydropic Two populations of Similar size
hydropic villi villi hydropic and nonhydropic
Villous shape Round, bulbous Prominent club-shaped Irregular Round and smooth
stromal projections
Scalloped villous Rare Rare Prominent Rare
contour
Trophoblastic Common, irregular Absent Common, round Occasional, single cell
inclusions
Villous trophoblastic Marked, circumferential, marked Mild, polarized, Mild, with trophoblast Usually absent. If found,
hyperplasia cytologic atypia mild-to-moderate atypia snouts. Focal mild atypia normally polarized. No
atypia
Central cisterns Common Not prominent Not prominent if in first Absent
trimester
Stroma appearance Mucoid, bluish, no fibrosis Mucoid, bluish, cellular Nonhydropic villi are Variable
with immature stellate fibrotic, with ectatic blood
cells vessels
Stromal karyorrhexis Common Prominent Absent Absent
Nucleated red blood None None Present Present
cells
Extravillous Hyperplastic, marked cytologic Mild hyperplasia, Normal, no atypia Normal, no atypia
trophoblasts atypia, may show exaggerated mildly atypical
placental site
P57 immunostain Negative Negative Positive Positive
Ki-67 of High (>70%) High (>70%) High (>70%) Low (<25%)
cytotrophoblasts
Karyotype Diandric diploidy (46XX, 46XY) Diandric diploidy Diandric triploidy Biparental dipliody
(46XX, 46XY) (69XXY, 69XXX) (46XX, 46XY)
d iploid (46, XX or 46 XY) with a paternal only genome [24, transparent, and of various sizes (Fig. 13.11). Some may
44, 45] (Fig. 13.10). reach 1 cm or more in diameter. The vesicles are often
admixed with blood clot and decidua while normal placental
13.5.1.2 Clinical Features or fetal tissues are not found, except in cases in which there
Patients usually present with second-trimester vaginal bleed- is a concurrent twin pregnancy [52].
ing, and a large-for-dates uterus, accompanied by markedly Histologically, the chorionic villi are irregular in shape
elevated serum hCG levels [46]. There may be hyperemesis and sizes. Some may be strikingly edematous (Fig. 13.12)
gravidarum and other symptoms related to preeclampsia, while some may show club-shaped stromal projections
hyperthyroidism, and hyperreactio luteinalis [47, 48]. Rarely, (Fig. 13.13) [53, 54]. They may appear avascular, and
there is vaginal passage of molar vesicles or manifestations enlarged with cistern formation. Numerous small blood
related to metastases. In those presenting in the first trimes- vessels are usually present but they are more conspicuous
ter, there is usually an abnormal ultrasound scan with absence with CD31 immunohistochemistry [55, 56]. Karyorrhectic
of fetal heartbeat [49]. The typical “snowstorm” pattern may debris is prominent and especially in early moles
not be well developed in such early cases and the clinical (Fig. 13.14) [46, 53, 55].
diagnosis is commonly a missed abortion [50, 51]. In first-trimester complete hydatidiform moles, the cen-
tral cisterns may be small or absent. The villous stroma is
13.5.1.3 Pathological Findings pale to bluish in hematoxylin and eosin-stained sections
The classical appearance of grossly evident vesicles may not (Fig. 13.15), and fibrosis is usually not a conspicuous find-
be readily appreciable especially if presentation is in the first ing. Intravillous trophoblastic inclusions are not common but
trimester. When present, the vesicles are grapelike, semi- may be seen. There is circumferential hyperplasia of the
13 Complications of Early Pregnancy and Gestational Trophoblastic Diseases 435
Sperm, Sperm,
23X 23X
Both Both
Sperm, Sperm,
23X or 23X or
23Y 23Y
Fig. 13.10 Chromosome compositions in normal conceptus and hyda- chromosome and duplicates. In dispermic complete mole (c), two
tidiform moles. Normal fertilization involves fusion of one haploid sperms enter an empty ovum and unite. In contrast, two haploid sets of
chromosome from both the father and mother (a). Diploid chromosome paternal chromosomes fuse with an ovum with intact maternal haplo-
composition is also seen in most complete moles. In monospermic type to produce triploid genome in a partial mole (d) [47]
complete mole (b), one sperm enters an empty ovum with no maternal
13.5.1.4 Biomarkers
The p57 immunohistochemistry is useful in confirming diag-
nosis (Fig. 13.16) [57, 58]. It has been shown to correlate
with genotyping and can serve as a reliable marker for diag-
nosis of complete hydatidiform moles, as well as identifying
mosaic conceptions [59]. The p57 is the protein product of
the cyclin-dependent kinase inhibitor 1C (CDKI1C) gene
Fig. 13.14 Photomicrograph of a complete mole with cistern forma-
(p57, Kip2) on chromosome 11p15.5, which is a paternally
tion and apoptosis of stromal cells
imprinted, maternally expressed gene. Lack of maternal
chromosomes in complete mole renders the loss of expres-
sion in the villous trophoblasts and stromal cells. P57 immu-
noreactivity is usually retained in the villous intermediate
trophoblast and implantation-site extravillous intermediate
trophoblasts among decidua and may serve as an internal
positive control. Almost all complete hydatidiform moles are
p57 negative.
It is also important to be aware of aberrant p57 expres-
sion in some special scenarios. Rare cases of complete
moles may show aberrant expression due to the retention of
maternal copy of chromosome 11 such as in trisomies. In
androgenetic/biparental mosaic/chimeric conceptuses
(which may show either typical complete mole morpholo-
gies or absence of trophoblastic hyperplasia), there is dis-
cordant p57 expression with different expression profile in
the villous cytotrophoblasts and stromal cells in same villi,
Fig. 13.15 Photomicrograph of an early complete mole with club-shaped i.e., positive immunoreactivity in cytotrophoblast but nega-
villi and myxomatous stroma. Cistern formation is inconspicuous tive in villous stromal cells, or vice versa [59, 60]. Divergent
13 Complications of Early Pregnancy and Gestational Trophoblastic Diseases 437
expression may also be seen in twin gestations, in which the fertilization of an empty ovum by one (monospermic) or two
p57 is absent in the villi of the complete hydatidiform mole (dispermic) sperms (Fig. 13.10). In 80–90%, it is a result of
but retained in those from the non-molar conceptus [61]. fertilization of an empty ovum by one sperm (homozygous)
Familial biparental diploid products of gestation or com- and in 10–20% of an ovum by two sperms (heterozygous)
plete moles related to NLRP7 mutations may show variable with absence or subsequent loss of maternal chromosomes.
levels of p57 expression and evidence of fetal development Such absence of maternal alleles facilitates the diagnosis of
and mild trophoblastic proliferation resembling triploid par- complete moles using microsatellite analysis (Fig. 13.17).
tial mole. Such differential diagnosis should be kept in mind Mitochondria DNA of maternal origin, however, exists [64].
particularly in cases with a history of recurrent molar Rarely, they are tetraploid (containing four paternal haploid
pregnancies. chromosomes, with a 92 XXXX karyotype) [44, 65–67].
Stromal apoptotic index has been shown to be higher in
complete hydatidiform mole than in partial mole of normal Biparental Complete Moles
placenta [56]. Overexpression of mRNA and protein of the Rare cases of recurrent complete moles are biparental dip-
transcription factor Nanog has also been shown to increase the loidy (as opposed to androgenetic diploidy) and are thought
risk of persistent gestational trophoblastic disease [62, 63]. to be familial in origin. They have been shown to be related
to maternal mutations in NLRP7 or KHDC3L (C6orf221)
13.5.1.5 Genetic Profile genes [16, 17, 68–70]. The mutations cause multiple epigen-
Complete hydatidiform mole has a diploid androgenic only etic defects which result in the failure to establish maternal
genome (two sets of paternal chromosomes) arising from identity at imprinted loci and with abnormal expression of
Fig. 13.17 Microsatellite 120 122 124 126 128 130 132 134 136 138 140 142 144 146 148 150 152 154 156 158
polymorphisms of the decidua
(upper panel) and villi (lower 560
panel) of a case of complete
hydatidiform mole. The 490
patient is heterozygous for the
marker generating alleles of
420
136–153 bp. The
hydatidiform mole is
homozygous giving rise to 350
allele 145 bp
280
210
140
70
0
PATIENT 136 153
800
600
400
200
0
HM 145
438 P. P. C. Ip et al.
imprinted genes. It is estimated in a recent study that reces- among patients with recessive mutations and ovum donation
sive NLRP7 and KHDC3L mutations were found in 55% and was found to be helpful [71].
5% of patients with recurrent moles, respectively [69]. Indeed, in a study on products of gestations from patients
Genotyping of available molar tissues from these patients with two defective NLRP7 alleles, all the conceptuses were
confirmed the diploid biparental contribution to all molar tis- found to be biparental diploid (Figs. 13.18 and 13.19).
sues from patients with recessive mutations in the known Variable p57 (KIP2) expression was found [70]. Positive p57
genes. Such genetic predisposition can be identified by expression was found in cases with missense NLRP7 muta-
appropriate genetic tests. Suitable genetic counselling and tions and was strongly associated with the presence of
assisted reproduction may be provided in experienced cen- embryonic tissues of inner cell mass origin and mild tropho-
ters. Live births (7–15% of pregnancies) have been reported blastic proliferation, features often considered supporting
Fig. 13.18 Multiplex short tandem repeat genotyping results for a dip- For example, at marker D16S539, the complete mole received a 278 bp
loid biparental complete hydatidiform mole from a patient with bial- allele from the father and a 286 bp allele from the mother [contributed
lelic mutations in NLRP7. Genotypes at three informative markers are by Dr. Rima Slim, McGill University Health Center Research Institute
shown and demonstrate at each of the three markers the presence of one Glen Site]
allele inherited from the mother and another inherited from the father.
13 Complications of Early Pregnancy and Gestational Trophoblastic Diseases 439
13.5.2.1 Definition
Partial hydatidiform mole shows diandric triploidy (one
maternal and two paternal sets of chromosomes). They con-
tain a mixture of normal-sized and enlarged hydropic chori-
onic villi with localized and mild degree of trophoblastic
hyperplasia [75, 85, 86].
IHC: p57
Positive (villous stroma, Negative
cytotrophoblast)
Genotyping
Partial Complete
Non-
Hydatidiform Hydatidiform
Molar
Mole Mole
Fig. 13.24 Flowchart of diagnosis in cases suspected of hydatidiform mole [modified from Banet N et al. Mod Pathol 2014 [59]]
Fig. 13.26 Hydropic chorionic villi (V) with florid trophoblast prolif-
eration impinging at the myometrium (M) of the uterus in this case of
invasive mole
13.5.3.5 Prognosis and Outcome cases in which the p57 expression profile is equivocal,
Invasive mole is often a clinical diagnosis. Patients usually molecular genotyping can be used to make a definitive diag-
have persistent elevated hCG levels but without residual nosis [57, 58, 66, 73, 93]. Hydropic abortus typically shows
molar tissue identified in the uterus on repeated curettage. biparental diploidy, while complete and partial hydatidiform
These cases are considered persistent gestational trophoblas- moles show androgenetic diploidy and diandric triploidy,
tic disease or gestational trophoblastic neoplasia as it is usu- respectively (see under complete and partial hydatidiform
ally not possible to distinguish invasive/metastatic mole moles for rare exceptions).
from choriocarcinoma. Nonetheless, most patients are cured In a problematic hydropic abortus, the spectrum of villous
if treated with chemotherapy [24]. abnormality may range from small, fibrotic villi to larger,
hydropic villi, which can potentially mimic partial hydatidi-
form mole [72, 73, 75]. The chorionic villi in a hydropic
13.6 S
pecific Issues of Diagnosis abortus usually do not exceed two to three times the size of
and Management in Relation those of the background small villi and do not have typical
to Hydropic Villi central cisterns. It is important not to overinterpret dilated
stem villi as central cisterns (Fig. 13.27). Another clue to
13.6.1 Hydropic Abortus and Abnormal support a hydropic abortus includes attenuated (rather than
(Non-molar) Villous Lesions snouting) surface trophoblasts. Villous trophoblastic hyper-
plasia is commonly observed in non-molar abortuses with
In routine surgical pathology practice, hydropic change of cho- abnormal karyotype, such as trisomies, digynic triploid preg-
rionic villi is a common finding in uterine evacuation samples. nancies, and placental mesenchymal dysplasia [72, 74, 75].
The challenge is to distinguish hydropic abortus from hydatidi- If molecular genotyping is not readily available, it is accept-
form moles. There is significant interobserver and intraob- able to sign out such cases as “abnormal villous morphology,
server variability in morphologic diagnoses even among features indeterminate for partial hydatidiform mole”
experienced pathologists who subspecialize in gynecologic (Fig. 13.28) [24]. These patients may be followed up for a
pathology [61, 73, 100]. Table 13.3 lists the similarities and short duration until the serum hCG is normalized.
differences between entities with problematic hydropic change.
From a practical standpoint, when hydropic change is
identified in the initial sections selected for histology, all 13.7 Choriocarcinoma
remaining tissue of the same specimen should be examined
to exclude hydatidiform moles. All recent curettage speci- 13.7.1 Definition
mens should also be reviewed. Application of p57 immunos-
tain should help to exclude most cases of complete Choriocarcinoma is a malignant gestational trophoblastic
hydatidiform moles, with the exception of androgenetic/ tumor in which there is simultaneous proliferation of inter-
biparental mosaic/chimeric conceptuses and twin gestations mediate, cytotrophoblasts and syncytiotrophoblasts of the
(see under complete hydatidiform moles) [60]. In difficult chorionic villi.
444 P. P. C. Ip et al.
13.7.2 Clinical Features Fig. 13.29 Metastatic choriocarcinoma to the lung as hemorrhagic
nodules
Choriocarcinomas are seen in women of reproductive age
group, and rarely in teens and postmenopausal women [101].
They are two times more common in non-Caucasians than
Caucasians with highest incidence reported in Asians,
Africans, and Latin Americans [29, 102–109]. The usual pre-
sentation is abnormal vaginal bleeding or extrauterine hem-
orrhage and a highly elevated serum hCG level [101, 110].
The majority of choriocarcinomas (50%) are developed after
a complete hydatidiform mole, an abortion (25%), normal
pregnancy (22.5%), and ectopic pregnancy (2.5%) [50, 80,
97, 109, 111, 112]. The tumors usually develop after a
latency of a few months to more than 14 years (with a mean
of 13 months after a complete hydatidiform mole and a mean
of 1–3 months after a normal pregnancy) [50, 80, 97, 101,
111]. The remainder of cases develop after a partial hydatidi-
form mole. The risk of post-molar choriocarcinoma is 2–3%
and <0.5% for complete and partial moles, respectively.
Fig. 13.30 Frozen section of a lung nodule reveals metastatic
Rarely, the presentation is the incidental identification of a
choriocarcinoma
choriocarcinoma in a term placenta during microscopic
examination [113, 114]. It is important to identify intrapla-
cental choriocarcinoma for further investigation and follow- clear intermediate trophoblasts and cytotrophoblasts are
up due to the significant risk of metastasis in both the mother usually surrounded by multinucleated syncytiotropho-
and baby. blasts in the periphery forming a biphasic and plexiform
pattern (Figs. 13.30 and 13.31) [24, 110]. All the tumor
cells show significant cytologic atypia and nuclear hyper-
13.7.3 Pathological Findings chromasia, appreciable on low magnification. They have
prominent nucleoli and are mitotically active. Some cells,
Grossly, the tumor is typically hemorrhagic and alternates especially the intermediate trophoblasts, may contain
with fleshy and necrotic areas (Fig. 13.29). It may be large striking cytoplasmic clearing. Tumor necrosis and hemor-
and involves both endometrium and myometrium with rhage are often a prominent feature. In tumors treated with
extension into cervix with extensive tissue destruction preoperative chemotherapy, the number of syncytiotropho-
[115]. Ectopic sites include fallopian tubes and/or ovaries blasts may be proportionally less than intermediate tro-
are uncommon [116, 117]. Histologically, the tumor grows phoblasts and cytotrophoblasts but can be highlighted by
in diffuse and infiltrating sheets and invades the myome- hCG immunohistochemistry (Fig. 13.32) [118]. Rare cases
trium or surrounding structures. The sheets of mononu- of choriocarcinoma may contain a minor component of
13 Complications of Early Pregnancy and Gestational Trophoblastic Diseases 445
13.7.4 Biomarkers
Fig. 13.37 Solid aggregates of PSTT tumor cells with nuclear pleo- Fig. 13.39 The tumor cells of PSTT are often negative for p63 and
morphism and brisk mitotic figures may be helpful to distinguish from poorly differentiated non-
keratinizing carcinoma in a cervical biopsy
Fig. 13.38 The cytoplasm of PSTT tumor cells is positive for hPL Fig. 13.40 Focal immunoreactivity for hCG can be demonstrated in
PSTT
13.8.4 Biomarkers
13.8.6 Differential Diagnoses
The tumor cells are positive for cytokeratins, hPL
(Fig. 13.38), CD10, inhibin, Mel-CAM (CD146), and HLA- The differential diagnoses include choriocarcinoma, epithe-
G. It is often negative for p63 (Fig. 13.39). The hCG staining lioid leiomyosarcoma, poorly differentiated carcinomas,
is usually limited to the multinucleated cells (Fig. 13.40). melanomas, and exaggerated placental site reaction [25]
The MIB1 proliferative index (ki-67) is usually >10% [10, (Tables 13.4 and 13.6).
120, 139]. Unlike placental site trophoblastic tumor, choriocarci-
noma usually has a combination of features including very
high serum hCG, a hemorrhagic mass, and a plexiform
13.8.5 Genetic Profile growth but lacking the angiocentric and angioinvasive pat-
tern of placental site trophoblastic tumor (see under differen-
There is usually a paternal X chromosome. There are occa- tial diagnosis of choriocarcinoma). Poorly differentiated
sional cases which show genetic imbalances [124, placental site trophoblastic tumor may be indistinguishable
140–143]. from some choriocarcinomas and may rarely coexist.
13 Complications of Early Pregnancy and Gestational Trophoblastic Diseases 449
Fig. 13.41 Spindle-shaped PSTT cells with eosinophilc degeneration Fig. 13.42 Spindle-shaped PSTT cells resembling leiomyosarcoma
resemble keratinizing squamous cell carcinoma of cervix
Distinction from epithelioid leiomyosarcoma, poorly differ- other pelvic sites, and metastases to lymph nodes, lungs,
entiated carcinomas, and melanomas in the uterine corpus or and liver. Half of these patients may die from tumor.
cervix may be difficult particularly during interpretation of fro- Metastases and recurrent tumors respond poorly to che-
zen section or small biopsies (Figs. 13.41 and 13.42). motherapy and often result in fatality. Pathologic features
Identification of more typical histopathological features and associated with poor outcome include extensive necrosis,
immunohistochemical profiling are helpful in the differential cells with clear cytoplasm, deep myometrial invasion, and
diagnoses. PSTT usually shows permeation of blood vessel wall, >5 mitotic figures per 10 high-power fields. In multivari-
splitting apart of well-preserved myometrial cells, and conspicu- ate analysis, FIGO stage III/IV, a latency of ≥2 years
ous fibrinoid deposit. Epithelioid leiomyosarcomas may be con- since last pregnancy, and presence of clear cells are inde-
firmed by absence of the distinctive vascular pattern of placental pendently associated with a poor prognosis [133,
site trophoblastic tumor; immunoreactivity for h-caldesmon, 144–147].
desmin, and actin; and negative for hPL and inhibin. Poorly dif-
ferentiated carcinomas usually show some histologic evidence of
a better differentiated component (squamous or glandular) and
an immunoprofile of hPL/hCG/inhibin nonreactivity. 13.9 Epithelioid Trophoblastic Tumor
13.9.1 Definition
13.8.7 Prognosis and Outcome
Epithelioid trophoblastic tumor is a malignant trophoblastic
The majority of patients present at FIGO stage I [132]. tumor of intermediate trophoblasts. The cell of origin is
Approximately 30% are high stage with involvement of believed to be chorionic-type trophoblasts [148].
450 P. P. C. Ip et al.
Exaggerated placental site is the unusual prominence of Placental site trophoblastic tumor is favored in the presence
implantation-site intermediate trophoblasts found at the of a mass (clinically or radiologically), high serum hCG lev-
implantation site of a placenta (synonym: exaggerated pla- els, destructive myoinvasion, typical angioinvasive pattern,
cental site reaction) [139, 163]. tumor necrosis, and a MIB1 proliferative index >10%. A low
MIB1 index and presence of decidua and villi are in favor of
exaggerated placental site.
13.10.2 Pathological Findings
The intermediate trophoblasts are immunoreactive for cyto- Nodules or plaques are usually discovered in hysterectomy
keratins, hPL, inhibin, Mel-CAM (CD146), and MIB1 pro- or uterine curettage specimens when the patient is undergo-
liferative index (ki-67) <1% [122, 123, 165]. ing investigations for other gynecological conditions, such as
452 P. P. C. Ip et al.
abnormal menstrual bleeding. Patients are usually in their pleomorphic nuclei with smudged nuclear chromatin.
reproductive years and had an antecedent pregnancy dating Mitotic figures are typically absent. Some cells may contain
back to 8 years (with a mean of 3 years). Rarely, the presen- cytoplasmic eosinophilic hyaline material. Fibrinoid necro-
tation is of someone being investigated for postmenopausal sis and dystrophic calcification are common. Rarely, necrotic
bleeding. The usual sites of involvement are lower segment or degenerated chorionic villi with ghost outline are also
endometrium, and endocervix [167]. Only 25% of lesions seen [168].
are grossly visible as tan to brown solid nodules.
Microscopically, the lesion is hypocellular and has abundant
eosinophilic hyalinized stroma in which individual and clus- 13.11.3 Biomarkers
ters of cells are seen (Fig. 13.46). The cells have variable
amount of eosinophilic to clear cytoplasm, and irregular and The cells are immunoreactive for cytokeratins, EMA, CD10,
p63, and inhibin (Fig. 13.47). Implantation-site trophoblast
markers such as hPL and Mel-CAM are less often positive.
The MIB1 proliferative index (ki-67) is <8%. Unlike epithe-
lioid trophoblastic tumors, nodules are negative for cyclin E
[11, 13, 123, 155, 170].
a b
Fig. 13.47 The chorionic intermediate trophoblasts at this PSN are negative for inhibin (a) but positive for p63 (b)
13 Complications of Early Pregnancy and Gestational Trophoblastic Diseases 453
trophoblastic tumor is often a large infiltrative mass with 7. Mittal K, Soslow R, McCluggage WG. Application of
immunohistochemistry to gynecologic pathology. Arch
high MIB1 index. Pathol Lab Med. 2008;132:402–23. https://doi.org/10.1043/
Distinction of a placental site nodule from epithelioid tro- 1543-2165(2008)132[402:aoitgp]2.0.co;2.
phoblastic tumor, which bears the same type of intermediate 8. Xue WC, Khoo US, Ngan HY, Chan KY, Ip PP, Tsao SW, Cheung
trophoblasts, may be difficult especially in curettage sam- AN. c-mos immunoreactivity aids in the diagnosis of gestational
trophoblastic lesions. Int J Gynecol Pathol. 2004;23:145–50.
ples. A nodule usually has a lower cellularity, more diffusely 9. Zhang HJ, Xue WC, Siu MK, Liao XY, Ngan HY, Cheung
hyalinized, and the cells are nonimmunoreactive for cyclin AN. P63 expression in gestational trophoblastic disease: cor-
E, and have a MIB1 index of <8%. In cases in which the relation with proliferation and apoptotic dynamics. Int J
distinction between a nodule and a tumor cannot be made Gynecol Pathol. 2009;28(2):172–8. https://doi.org/10.1097/
PGP.0b013e318189555b.
with certainty (in which cases the nodules are larger than 10. Singer G, Kurman RJ, McMaster MT, Shih Ie M. HLA-G immu-
usual, focally infiltrative border, increased cellularity, mod- noreactivity is specific for intermediate trophoblast in gestational
erate nuclear atypia, presence of mitotic figures, and MIB1 trophoblastic disease and can serve as a useful marker in differen-
index of 8–10%), it has been suggested that the term “atypi- tial diagnosis. Am J Surg Pathol. 2002;26:914–20.
11. Shih IM, Nesbit M, Herlyn M, Kurman RJ. A new Mel-CAM
cal placental site nodule” may be used [155, 158]. (CD146)-specific monoclonal antibody, MN-4, on paraffin-
Other differential diagnoses of placental site nodule embedded tissue. Mod Pathol. 1998;11:1098–106.
include squamous cell carcinoma and epithelioid smooth 12. Mao TL, Kurman RJ, Jeng YM, Huang W, Shih Ie M. HSD3B1
muscle tumors and are discussed under epithelioid tropho- as a novel trophoblast-associated marker that assists in the dif-
ferential diagnosis of trophoblastic tumors and tumorlike lesions.
blastic tumor. Am J Surg Pathol. 2008;32:236–42. https://doi.org/10.1097/
PAS.0b013e31812e0046.
13. Shih Ie M. Trophogram, an immunohistochemistry-based algo-
13.11.5 Prognosis and Outcome rithmic approach, in the differential diagnosis of trophoblastic
tumors and tumorlike lesions. Ann Diagn Pathol. 2007;11:228–
34. https://doi.org/10.1016/j.anndiagpath.2007.04.001.
These are benign lesions and follow-up on these patients 14. Xue WC, Cheung AN. Complications of early pregnancy, includ-
has been uneventful. Lesions considered atypical placen- ing trophoblastic neoplasia. In: Zheng W, editor. Pathology of
tal site nodules are suggested to be premalignant to epi- obstetrics and gynaecology; 2010.
15. Royal College of Obstetricians and Gynecologists (2011) The
thelioid trophoblastic tumors [158, 171]. Patients with investigation and treatment of couples with recurrent first trimes-
atypical placental site nodule should be evaluated by ter and second-trimester miscarriage. Green-top Guideline No. 17.
imaging studies to exclude the presence of a mass lesion, 16. Messaed C, Chebaro W, Di Roberto RB, Rittore C, Cheung A,
and should be under close surveillance with serial hCG Arseneau J, Schneider A, Chen MF, Bernishke K, Surti U, Hoffner
L, Sauthier P, Buckett W, Qian J, Lau NM, Bagga R, Engert JC,
measurements. Coullin P, Touitou I, Slim R. NLRP7 in the spectrum of repro-
ductive wastage: rare non-synonymous variants confer genetic
susceptibility to recurrent reproductive wastage. J Med Genet.
2011;48:540–8. https://doi.org/10.1136/jmg.2011.089144.
17. Slim R, Ao A, Surti U, Zhang L, Hoffner L, Arseneau J, Cheung
References A, Chebaro W, Wischmeijer A. Recurrent triploid and disper-
mic conceptions in patients with NLRP7 mutations. Placenta.
1. Moore KL, Persaud TVN, Torchia MG. Before we are born : 2011;32:409–12. https://doi.org/10.1016/j.placenta.2011.02.001.
essentials of embryology and birth defects. Philadelphia, PA: 18. Murdoch S, Djuric U, Mazhar B, Seoud M, Khan R, Kuick R,
Elsevier/Saunders; 2016. Bagga R, Kircheisen R, Ao A, Ratti B, Hanash S, Rouleau GA,
2. Paradinas FJ, Elston CW. Gestational trophoblastic diseases. In: Slim R. Mutations in NALP7 cause recurrent hydatidiform moles
Fox H, Wells M, editors. Haines & Taylor obstetrical and gyn- and reproductive wastage in humans. Nat Genet. 2006;38:300–2.
aecological pathology. Edinburgh: Churchill Livingstone; 2003. https://doi.org/10.1038/ng1740.
p. 1359–430. 19. Bouyer J, Coste J, Shojaei T, Pouly JL, Fernandez H, Gerbaud L,
3. Shih IM, Mazur MT, Kurman RJ. Gestational trophoblastic dis- Job-Spira N. Risk factors for ectopic pregnancy: a comprehensive
ease and related lesions. In: Kurman R, editor. Blaustein’s pathol- analysis based on a large case-control, population-based study in
ogy of the female genital tract. New York: Springer Verlag; 2002. France. Am J Epidemiol. 2003;157:185–94.
4. Cheung A. Gestational trophoblastic disease. In: Robboy S, 20. Fernandez H, Gervaise A. Ectopic pregnancies after infertil-
Mutter G, Prat J, Bentley R, Russell P, Anderson M, editors. ity treatment: modern diagnosis and therapeutic strategy. Hum
Robboy’s pathology of the female reproductive tract. China: Reprod Update. 2004;10:503–13. https://doi.org/10.1093/
Elsevier Churchill Livingstone; 2009. p. 881–907. humupd/dmh043.
5. Lage JM. Gestational trophoblastic diseases. In: Robbey S, Russel 21. Zhang Y, Chen J, Lu W, Li B, Du G, Wan X. Clinical characteristics
R, editors. Pathology of the female reproductive tract. London: of persistent ectopic pregnancy after salpingostomy and influence
Elsevier; 2001. on ongoing pregnancy. J Obstet Gynaecol Res. 2017;43:564–70.
6. Lee CQ, Gardner L, Turco M, Zhao N, Murray MJ, Coleman N, https://doi.org/10.1111/jog.13251.
Rossant J, Hemberger M, Moffett A. What is trophoblast? A combi- 22. Burton JL, Lidbury EA, Gillespie AM, Tidy JA, Smith O, Lawry
nation of criteria define human first-trimester trophoblast. Stem Cell J, Hancock BW, Wells M. Over-diagnosis of hydatidiform mole in
Rep. 2016;6:257–72. https://doi.org/10.1016/j.stemcr.2016.01.006. early tubal ectopic pregnancy. Histopathology. 2001;38:409–17.
454 P. P. C. Ip et al.
23. Sebire NJ, Lindsay I, Fisher RA, Savage P, Seckl MJ. Overdiagnosis 43. Martin BH, Kim JH. Changes in gestational trophoblastic
of complete and partial hydatidiform mole in tubal ectopic preg- tumors over four decades. A Korean experience. J Reprod Med.
nancies. Int J Gynecol Pathol. 2005;24:260–4. 1998;43:60–8.
24. Hui P, Baergen RN, Cheung AN, Fukunaga M, Gersell DJ, Lage 44. Kajii T, Ohama K. Androgenetic origin of hydatidiform mole.
JM, Ronnett BM, Sebire NJ, Wells M. Gestational trophoblastic Nature. 1977;268:633–4.
disease. In: Kurman RJ, Young RH, Carcangiu ML, Herrington S, 45. Lawler SD, Povey S, Fisher RA, Pickthall VJ. Genetic studies on
editors. In: WHO classification of tumours of female reproductive hydatidiform moles. II. The origin of complete moles. Ann Hum
organs. 4th ed. Lyon: IARC; 2014. p. 155. Genet. 1982;46:209–22.
25. Clement P, Young R. Trophoblastic lesions, miscellaneous pri- 46. Paradinas FJ, Browne P, Fisher RA, Foskett M, Bagshawe KD,
mary uterine neoplasms, hematopoietic neoplasms, and metastatic Newlands E. A clinical, histopathological and flow cytometric
neoplasms to the uterus. In: Clement P, Young R, editors. Atlas of study of 149 complete moles, 146 partial moles and 107 non-
gynecologic surgical pathology. 3rd ed. Philadelphia: Saunders; molar hydropic abortions. Histopathology. 1996;28:101–10.
2013. 47. Szulman AE. Syndromes of hydatidiform moles. Partial vs. com-
26. Ngan HY, Seckl MJ, Berkowitz RS, Xiang Y, Golfier F, Sekharan plete. J Reprod Med. 1984;29:788–91.
PK, Lurain JR. Update on the diagnosis and management of gesta- 48. Garner EI, Goldstein DP, Feltmate CM, Berkowitz RS. Gestational
tional trophoblastic disease. Int J Gynaecol Obstet. 2015;131(Suppl trophoblastic disease. Clin Obstet Gynecol. 2007;50:112–22.
2):S123–6. https://doi.org/10.1016/j.ijgo.2015.06.008. https://doi.org/10.1097/GRF.0b013e31802f17fc.
27. Atrash HK, Hogue CJ, Grimes DA. Epidemiology of hyda- 49. Fowler DJ, Lindsay I, Seckl MJ, Sebire NJ. Histomorphometric
tidiform mole during early gestation. Am J Obstet Gynecol. features of hydatidiform moles in early pregnancy: relationship
1986;154:906–9. to detectability by ultrasound examination. Ultrasound Obstet
28. Bagshawe KD, Dent J, Webb J. Hydatidiform mole in England Gynecol. 2007;29:76–80. https://doi.org/10.1002/uog.3880.
and Wales 1973–83. Lancet. 1986;2:673–7. 50. Berkowitz RS, Goldstein DP. Chorionic tumors. N Engl J Med.
29. Buckley JD. The epidemiology of molar pregnancy and choriocar- 1996;335:1740–8. https://doi.org/10.1056/NEJM199612053352306.
cinoma. Clin Obstet Gynecol. 1984;27:153–9. 51. Berkowitz RS, Goldstein DP. Clinical practice. Molar pregnancy.
30. Hayashi K, Bracken MB, Freeman DH Jr, Hellenbrand N Engl J Med. 2009;360:1639–45. https://doi.org/10.1056/
K. Hydatidiform mole in the United States (1970–1977): a statis- NEJMcp0900696.
tical and theoretical analysis. Am J Epidemiol. 1982;115:67–77. 52. Baergen RN, Kelly T, McGinniss MJ, Jones OW, Benirschke
31. Ngan HY, Kohorn EI, Cole LA, Kurman RJ, Kim SJ, Lurain K. Complete hydatidiform mole with a coexistent embryo. Hum
JR, Seckl MJ, Sasaki S, Soper JT. Trophoblastic disease. Int Pathol. 1996;27:731–4.
J Gynaecol Obstet. 2012;119(Suppl 2):S130–6. https://doi. 53. Keep D, Zaragoza MV, Hassold T, Redline RW. Very early com-
org/10.1016/S0020-7292(12)60026-5. plete hydatidiform mole. Hum Pathol. 1996;27:708–13.
32. Ngan HY, Tam KF, Lam KW, Chan KK. Relapsed gestational 54. Sebire NJ, Fisher RA, Rees HC. Histopathological diagnosis
trophoblastic neoplasia: a 20-year experience. J Reprod Med. of partial and complete hydatidiform mole in the first trimes-
2006;51:829–34. ter of pregnancy. Pediatr Dev Pathol. 2003;6:69–77. https://doi.
33. Hando T, Ohno M, Kurose T. Recent aspects of gestational tro- org/10.1007/s10024-002-0079-9.
phoblastic disease in Japan. Int J Gynaecol Obstet. 1998;60(Suppl 55. Kim KR, Park BH, Hong YO, Kwon HC, Robboy SJ. The vil-
1):S71–6. lous stromal constituents of complete hydatidiform mole differ
34. Kim JH, Park DC, Bae SN, Namkoong SE, Kim SJ. Subsequent histologically in very early pregnancy from the normally devel-
reproductive experience after treatment for gestational tropho- oping placenta. Am J Surg Pathol. 2009;33:176–85. https://doi.
blastic disease. Gynecol Oncol. 1998;71:108–12. https://doi. org/10.1097/PAS.0b013e31817fada1.
org/10.1006/gyno.1998.5167. 56. Kim MJ, Kim KR, Ro JY, Lage JM, Lee HI. Diagnostic and patho-
35. Altieri A, Franceschi S, Ferlay J, Smith J, La Vecchia genetic significance of increased stromal apoptosis and incom-
C. Epidemiology and aetiology of gestational trophoblastic dis- plete vasculogenesis in complete hydatidiform moles in very early
eases. Lancet Oncol. 2003;4:670–8. pregnancy periods. Am J Surg Pathol. 2006;30:362–9. https://doi.
36. Palmer JR. Advances in the epidemiology of gestational tropho- org/10.1097/01.pas.0000194299.27463.21.
blastic disease. J Reprod Med. 1994;39:155–62. 57. LeGallo RD, Stelow EB, Ramirez NC, Atkins KA. Diagnosis of
37. Lurain JR. Gestational trophoblastic disease I: epidemiology, hydatidiform moles using p57 immunohistochemistry and HER2
pathology, clinical presentation and diagnosis of gestational tro- fluorescent in situ hybridization. Am J Clin Pathol. 2008;129:749–
phoblastic disease, and management of hydatidiform mole. Am 55. https://doi.org/10.1309/7XRL378C22W7APBT.
J Obstet Gynecol. 2010;203:531–9. https://doi.org/10.1016/j. 58. McConnell TG, Murphy KM, Hafez M, Vang R, Ronnett
ajog.2010.06.073. BM. Diagnosis and subclassification of hydatidiform moles
38. La Vecchia C, Parazzini F, Decarli A, Franceschi S, Fasoli M, using p57 immunohistochemistry and molecular genotyping:
Favalli G, Negri E, Pampallona S. Age of parents and risk of ges- validation and prospective analysis in routine and consultation
tational trophoblastic disease. J Natl Cancer Inst. 1984;73:639–42. practice settings with development of an algorithmic approach.
39. Sebire NJ, Foskett M, Fisher RA, Rees H, Seckl M, Newlands Am J Surg Pathol. 2009;33:805–17. https://doi.org/10.1097/
E. Risk of partial and complete hydatidiform molar pregnancy in PAS.0b013e318191f309.
relation to maternal age. BJOG. 2002;109:99–102. 59. Banet N, DeScipio C, Murphy KM, Beierl K, Adams E, Vang
40. Matsuura J, Chiu D, Jacobs PA, Szulman AE. Complete hydatidi- R, Ronnett BM. Characteristics of hydatidiform moles: analy-
form mole in Hawaii: an epidemiological study. Genet Epidemiol. sis of a prospective series with p57 immunohistochemistry and
1984;1:271–84. https://doi.org/10.1002/gepi.1370010306. molecular genotyping. Mod Pathol. 2014;27:238–54. https://doi.
41. Martin PM. High frequency of hydatidiform mole in native org/10.1038/modpathol.2013.143.
Alaskans. Int J Gynaecol Obstet. 1978;15:395–6. 60. Lewis GH, DeScipio C, Murphy KM, Haley L, Beierl K, Mosier
42. Cheung A. Gestational trophoblastic diseases. In: Ho F, Wu P, edi- S, Tandy S, Cohen DS, Lytwyn A, Elit L, Vang R, Ronnett
tors. Topics in pathology for Hong Kong. Hong Kong: Hong Kong BM. Characterization of androgenetic/biparental mosaic/chi-
University Press; 1995. p. 147–63. meric conceptions, including those with a molar component:
13 Complications of Early Pregnancy and Gestational Trophoblastic Diseases 455
morphology, p57 immnohistochemistry, molecular genotyp- 75. Genest DR. Partial hydatidiform mole: clinicopathological fea-
ing, and risk of persistent gestational trophoblastic disease. Int tures, differential diagnosis, ploidy and molecular studies, and gold
J Gynecol Pathol. 2013;32:199–214. https://doi.org/10.1097/ standards for diagnosis. Int J Gynecol Pathol. 2001;20:315–22.
PGP.0b013e3182630d8c. 76. Paradinas FJ, Sebire NJ, Fisher RA, Rees HC, Foskett M, Seckl
61. Ronnett BM, DeScipio C, Murphy KM. Hydatidiform MJ, Newlands ES. Pseudo-partial moles: placental stem vessel
moles: ancillary techniques to refine diagnosis. Int J hydrops and the association with Beckwith-Wiedemann syndrome
Gynecol Pathol. 2011;30:101–16. https://doi.org/10.1097/ and complete moles. Histopathology. 2001;39:447–54.
PGP.0b013e3181f4de77. 77. Niemann I, Petersen LK, Hansen ES, Sunde L. Predictors of low
62. Siu MK, Wong ES, Chan HY, Ngan HY, Chan KY, Cheung risk of persistent trophoblastic disease in molar pregnancies.
AN. Overexpression of NANOG in gestational trophoblastic Obstet Gynecol. 2006;107:1006–11. https://doi.org/10.1097/01.
diseases: effect on apoptosis, cell invasion, and clinical out- AOG.0000210635.24543.3b.
come. Am J Pathol. 2008;173:1165–72. https://doi.org/10.2353/ 78. Yeung C-W, Cheung ANY. Negative pregnancy test in patients with
ajpath.2008.080288. trophoblastic diseases. Curr Obstet Gynecol Rep. 2014;3:102–6.
63. Shih Ie M, Kuo KT. Power of the eternal youth: Nanog expression https://doi.org/10.1007/s13669-013-0067-2.
in the gestational choriocarcinoma. Am J Pathol. 2008;173:911–4. 79. Baasanjav B, Usui H, Kihara M, Kaku H, Nakada E, Tate S,
https://doi.org/10.2353/ajpath.2008.080624. Mitsuhashi A, Matsui H, Shozu M. The risk of post-molar ges-
64. Chiu PM, Liu VW, Ngan HY, Khoo US, Cheung AN. Detection of tational trophoblastic neoplasia is higher in heterozygous than
mitochondrial DNA mutations in gestational trophoblastic disease. in homozygous complete hydatidiform moles. Hum Reprod.
Hum Mutat. 2003;22:177. https://doi.org/10.1002/humu.9160. 2010;25:1183–91. https://doi.org/10.1093/humrep/deq052.
65. Murphy KM, Descipio C, Wagenfuehr J, Tandy S, Mabray J, 80. Matsui H, Iizuka Y, Sekiya S. Incidence of invasive mole and cho-
Beierl K, Micetich K, Libby AL, Ronnett BM. Tetraploid par- riocarcinoma following partial hydatidiform mole. Int J Gynaecol
tial hydatidiform mole: a case report and review of the literature. Obstet. 1996;53:63–4.
Int J Gynecol Pathol. 2012;31:73–9. https://doi.org/10.1097/ 81. Agarwal R, Teoh S, Short D, Harvey R, Savage PM, Seckl
PGP.0b013e31822555b3. MJ. Chemotherapy and human chorionic gonadotropin concen-
66. Lipata F, Parkash V, Talmor M, Bell S, Chen S, Maric V, Hui trations 6 months after uterine evacuation of molar pregnancy: a
P. Precise DNA genotyping diagnosis of hydatidiform mole. retrospective cohort study. Lancet. 2012;379:130–5. https://doi.
Obstet Gynecol. 2010;115:784–94. https://doi.org/10.1097/ org/10.1016/s0140-6736(11)61265-8.
AOG.0b013e3181d489ec. 82. Cheung AN, Chan KK. Perplexing hCG profile after evacuation
67. Fukunaga M, Endo Y, Ushigome S. Clinicopathologic study of hydatidiform mole. Lancet. 2012;379:98–100. https://doi.
of tetraploid hydropic villous tissues. Arch Pathol Lab Med. org/10.1016/s0140-6736(11)61518-3.
1996;120:569–72. 83. Garrett LA, Garner EI, Feltmate CM, Goldstein DP, Berkowitz
68. Brown L, Mount S, Reddy R, Slim R, Wong C, Jobanputra RS. Subsequent pregnancy outcomes in patients with molar preg-
V, Clifford P, Merrill L, Brown S. Recurrent pregnancy loss in nancy and persistent gestational trophoblastic neoplasia. J Reprod
a woman with NLRP7 mutation: not all molar pregnancies can Med. 2008;53:481–6.
be easily classified as either “partial” or “complete” hydatidi- 84. Sebire NJ, Fisher RA, Foskett M, Rees H, Seckl MJ, Newlands
form moles. Int J Gynecol Pathol. 2013;32:399–405. https://doi. ES. Risk of recurrent hydatidiform mole and subsequent preg-
org/10.1097/PGP.0b013e31826cbf6a. nancy outcome following complete or partial hydatidiform molar
69. Nguyen NM, Slim R. Genetics and epigenetics of recurrent pregnancy. BJOG. 2003;110:22–6.
hydatidiform moles: basic science and genetic counselling. Curr 85. Hui P, Martel M, Parkash V. Gestational trophoblastic diseases:
Obstet Gynecol Rep. 2014;3:55–64. https://doi.org/10.1007/ recent advances in histopathologic diagnosis and related genetic
s13669-013-0076-1. aspects. Adv Anat Pathol. 2005;12:116–25.
70. Nguyen NM, Zhang L, Reddy R, Dery C, Arseneau J, Cheung 86. Lawler SD, Fisher RA, Pickthall VJ, Povey S, Evans MW. Genetic
A, Surti U, Hoffner L, Seoud M, Zaatari G, Bagga R, Srinivasan studies on hydatidiform moles. I. The origin of partial moles.
R, Coullin P, Ao A, Slim R. Comprehensive genotype-phenotype Cancer Genet Cytogenet. 1982;5:309–20.
correlations between NLRP7 mutations and the balance between 87. Jauniaux E. Partial moles: from postnatal to prenatal diag-
embryonic tissue differentiation and trophoblastic prolifera- nosis. Placenta. 1999;20:379–88. https://doi.org/10.1053/
tion. J Med Genet. 2014;51:623–34. https://doi.org/10.1136/ plac.1999.0390.
jmedgenet-2014-102546. 88. Fine C, Bundy AL, Berkowitz RS, Boswell SB, Berezin AF,
71. Nguyen N, Khawajkie Y, Mechtouf N, Sauthier P, Arseneau J, Doubilet PM. Sonographic diagnosis of partial hydatidiform
Rahimi K, Breguet M and Slim R (2017) The genetics of recurrent mole. Obstet Gynecol. 1989;73:414–8.
hydatidiform moles: new findings and lessons from the analysis of 89. Jacobs PA, Szulman AE, Funkhouser J, Matsuura JS, Wilson
111 patients. XIX Biannual World Congress, International Society CC. Human triploidy: relationship between parental origin of the
of Study of Trophoblastic Diseases, Amsterdam. additional haploid complement and development of partial hyda-
72. Buza N, Hui P. Partial hydatidiform mole: histologic param- tidiform mole. Ann Hum Genet. 1982;46:223–31.
eters in correlation with DNA genotyping. Int J Gynecol Pathol. 90. Sundvall L, Lund H, Niemann I, Jensen UB, Bolund L,
2013;32:307–15. https://doi.org/10.1097/PGP.0b013e3182626011. Sunde L. Tetraploidy in hydatidiform moles. Hum Reprod.
73. Chew SH, Perlman EJ, Williams R, Kurman RJ, Ronnett 2013;28:2010–20. https://doi.org/10.1093/humrep/det132.
BM. Morphology and DNA content analysis in the evaluation of 91. Golubovsky MD. Postzygotic diploidization of triploids as a
first trimester placentas for partial hydatidiform mole (PHM). Hum source of unusual cases of mosaicism, chimerism and twinning.
Pathol. 2000;31:914–24. https://doi.org/10.1053/hupa.2000.9085. Hum Reprod. 2003;18:236–42.
74. Sebire NJ, May PC, Kaur B, Seckl MJ, Fisher RA. Abnormal 92. Schammel DP, Bocklage T. p53 PCNA, and Ki-67 in hydropic
villous morphology mimicking a hydatidiform mole associated molar and nonmolar placentas: an immunohistochemical study.
with paternal trisomy of chromosomes 3,7,8 and unipaternal Int J Gynecol Pathol. 1996;15:158–66.
disomy of chromosome 11. Diagn Pathol. 2016;11:20. https://doi. 93. Bifulco C, Johnson C, Hao L, Kermalli H, Bell S, Hui P. Genotypic
org/10.1186/s13000-016-0471-9. analysis of hydatidiform mole: an accurate and practical method
456 P. P. C. Ip et al.
of diagnosis. Am J Surg Pathol. 2008;32:445–51. https://doi. tionship with infantile choriocarcinoma. Fetal Pediatr Pathol.
org/10.1097/PAS.0b013e3181520034. 2005;24:21–9.
94. DeScipio C, Haley L, Beierl K, Pandit AP, Murphy KM, Ronnett 115. Fu Y, Lu W, Zhou C, Xie X. Primary cervical chorio-
BM. Diandric triploid hydatidiform mole with loss of maternal carcinoma: report of four cases and literature review.
chromosome 11. Am J Surg Pathol. 2011;35:1586–91. https://doi. Int J Gynecol Cancer. 2007;17:715–9. https://doi.
org/10.1097/PAS.0b013e31822d5cff. org/10.1111/j.1525-1438.2007.00819.x.
95. Rice LW, Berkowitz RS, Lage JM, Goldstein DP, Bernstein 116. Chan DP, Wong WP. Extrauterine gestational choriocarcinoma.
MR. Persistent gestational trophoblastic tumor after partial hyda- Report of two cases. Obstet Gynecol. 1970;35:730–3.
tidiform mole. Gynecol Oncol. 1990;36:358–62. 117. Ober WB, Maier RC. Gestational choriocarcinoma of the fallo-
96. Hancock BW, Nazir K, Everard JE. Persistent gestational tropho- pian tube. Diagn Gynecol Obstet. 1981;3:213–31.
blastic neoplasia after partial hydatidiform mole incidence and 118. Mazur MT. Metastatic gestational choriocarcinoma. Unusual
outcome. J Reprod Med. 2006;51:764–6. pathologic variant following therapy. Cancer. 1989;63:1370–7.
97. Seckl MJ, Fisher RA, Salerno G, Rees H, Paradinas FJ, 119. Shen DH, Khoo US, Ngan HY, Ng TY, Chau MT, Xue WC,
Foskett M, Newlands ES. Choriocarcinoma and partial hyda- Cheung AN. Coexisting epithelioid trophoblastic tumor and cho-
tidiform moles. Lancet. 2000;356:36–9. https://doi.org/10.1016/ riocarcinoma of the uterus following a chemoresistant hydatidi-
S0140-6736(00)02432-6. form mole. Arch Pathol Lab Med. 2003;127:e291–3. https://doi.
98. Bynum J, Murphy KM, DeScipio C, Beierl K, Adams E, org/10.1043/1543-2165(2003)127<e291:CETTAC>2.0.CO;2.
Anderson D, Vang R, Ronnett BM. Invasive complete hyda- 120. Kalhor N, Ramirez PT, Deavers MT, Malpica A, Silva
tidiform moles: analysis of a case series with genotyping. Int EG. Immunohistochemical studies of trophoblastic tumors.
J Gynecol Pathol. 2016;35:134–41. https://doi.org/10.1097/ Am J Surg Pathol. 2009;33:633–8. https://doi.org/10.1097/
PGP.0000000000000232. PAS.0b013e318191f2eb.
99. Gaber LW, Redline RW, Mostoufi-Zadeh M, Driscoll SG. Invasive 121. Mao TL, Kurman RJ, Huang CC, Lin MC, Shih Ie
partial mole. Am J Clin Pathol. 1986;85:722–4. M. Immunohistochemistry of choriocarcinoma: an aid in dif-
100. Fukunaga M, Katabuchi H, Nagasaka T, Mikami Y, Minamiguchi ferential diagnosis and in elucidating pathogenesis. Am J
S, Lage JM. Interobserver and intraobserver variability in the diag- Surg Pathol. 2007;31:1726–32. https://doi.org/10.1097/
nosis of hydatidiform mole. Am J Surg Pathol. 2005;29:942–7. PAS.0b013e318058a529.
101. Ober WB, Edgcomb JH, Price EB Jr. The pathology of choriocar- 122. Shih IM, Kurman RJ. Ki-67 labeling index in the differential
cinoma. Ann N Y Acad Sci. 1971;172:299–426. diagnosis of exaggerated placental site, placental site trophoblas-
102. Llewellyn-Jones D. Trophoblastic tumors; geographical variations tic tumor, and choriocarcinoma: a double immunohistochemical
in incidence and possible aetiological factors. J Obstet Gynaecol staining technique using Ki-67 and Mel-CAM antibodies. Hum
Br Commonw. 1965;72:242–8. Pathol. 1998;29:27–33.
103. Shanmugaratnam K, Muir CS, Tow SH, Cheng WC, Christine B, 123. Shih IM, Kurman RJ. Immunohistochemical localization of
Pedersen E. Rates per 100,000 births and incidence of choriocar- inhibin-alpha in the placenta and gestational trophoblastic lesions.
cinoma and malignant mole in Singapore Chinese and Malays. Int J Gynecol Pathol. 1999;18:144–50.
Comparison with Connecticut, Norway and Sweden. Int J Cancer. 124. Yap KL, Hafez MJ, Mao TL, Kurman RJ, Murphy KM, Shih Ie
1971;8:165–75. M. Lack of a y-chromosomal complement in the majority of ges-
104. Nakano R, Sasaki K, Yamoto M, Hata H. Trophoblastic disease: tational trophoblastic neoplasms. J Oncol. 2010;2010:364508.
analysis of 342 patients. Gynecol Obstet Investig. 1980;11:237–42. https://doi.org/10.1155/2010/364508.
105. Poen HT, Djojopranoto M. The possible etiologic factors of hyda- 125. Rodriguez E, Melamed J, Reuter V, Chaganti RS. Chromosomal
tidiform mole and choriocarcinoma: preliminary report. Am J abnormalities in choriocarcinomas of the female. Cancer Genet
Obstet Gynecol. 1965;92:510–3. Cytogenet. 1995;80:9–12.
106. Brinton LA, Bracken MB, Connelly RR. Choriocarcinoma inci- 126. Sheppard DM, Fisher RA, Lawler SD. Karyotypic analysis and
dence in the United States. Am J Epidemiol. 1986;123:1094–100. chromosome polymorphisms in four choriocarcinoma cell lines.
107. Ringertz N. Hydatidiform mole, invasive mole and choriocar- Cancer Genet Cytogenet. 1985;16:251–8.
cinoma in Sweden 1958–1965. Acta Obstet Gynecol Scand. 127. Savage J, Adams E, Veras E, Murphy KM, Ronnett
1970;49:195–203. BM. Choriocarcinoma in women: analysis of a case series with
108. Smith HO. Gestational trophoblastic disease epidemiology and genotyping. Am J Surg Pathol. 2017;41:1593–606. https://doi.
trends. Clin Obstet Gynecol. 2003;46:541–56. org/10.1097/pas.0000000000000937.
109. Soper JT. Gestational trophoblastic disease. Obstet 128. Cheung AN, Zhang HJ, Xue WC, Siu MK. Pathogenesis of cho-
Gynecol. 2006;108:176–87. https://doi.org/10.1097/01. riocarcinoma: clinical, genetic and stem cell perspectives. Future
AOG.0000224697.31138.a1. Oncol. 2009;5:217–31. https://doi.org/10.2217/14796694.5.2.217.
110. Smith HO, Kohorn E, Cole LA. Choriocarcinoma and gestational 129. Morgan JM, Lurain JR. Gestational trophoblastic neoplasia: an
trophoblastic disease. Obstet Gynecol Clin N Am. 2005;32:661– update. Curr Oncol Rep. 2008;10:497–504.
84. https://doi.org/10.1016/j.ogc.2005.08.001. 130. Rodabaugh KJ, Bernstein MR, Goldstein DP, Berkowitz
111. Lurain JR, Brewer JI, Torok EE, Halpern B. Natural history RS. Natural history of postterm choriocarcinoma. J Reprod Med.
of hydatidiform mole after primary evacuation. Am J Obstet 1998;43:75–80.
Gynecol. 1983;145:591–5. 131. Kohorn EI. The new FIGO 2000 staging and risk factor scoring
112. Bagshawe KD, Golding PR, Orr AH. Choriocarcinoma after hyda- system for gestational trophoblastic disease: description and criti-
tidiform mole. Studies related to effectiveness of follow-up prac- cal assessment. Int J Gynecol Cancer. 2001;11:73–7.
tice after hydatidiform mole. Br Med J. 1969;3:733–7. 132. Baergen RN, Rutgers JL, Young RH, Osann K, Scully
113. Jacques SM, Qureshi F, Doss BJ, Munkarah A. Intraplacental RE. Placental site trophoblastic tumor: a study of 55 cases
choriocarcinoma associated with viable pregnancy: pathologic and review of the literature emphasizing factors of prognos-
features and implications for the mother and infant. Pediatr Dev tic significance. Gynecol Oncol. 2006;100:511–20. https://doi.
Pathol. 1998;1:380–7. org/10.1016/j.ygyno.2005.08.058.
114. Sebire NJ, Lindsay I, Fisher RA, Seckl MJ. Intraplacental cho- 133. Hassadia A, Gillespie A, Tidy J, Everard RGNJ, Wells M,
riocarcinoma: experience from a tertiary referral center and rela- Coleman R, Hancock B. Placental site trophoblastic tumour:
13 Complications of Early Pregnancy and Gestational Trophoblastic Diseases 457
c linical features and management. Gynecol Oncol. 2005;99:603– 153. Meydanli MM, Kucukali T, Usubutun A, Ataoglu O, Kafkasli
7. https://doi.org/10.1016/j.ygyno.2005.06.054. A. Epithelioid trophoblastic tumor of the endocervix: a case
134. Chang YL, Chang TC, Hsueh S, Huang KG, Wang PN, Liu report. Gynecol Oncol. 2002;87:219–24.
HP, Soong YK. Prognostic factors and treatment for placen- 154. Cheung AN. Pathology of gestational trophoblastic diseases. Best
tal site trophoblastic tumor-report of 3 cases and analysis of 88 Pract Res Clin Obstet Gynaecol. 2003;17:849–68.
cases. Gynecol Oncol. 1999;73:216–22. https://doi.org/10.1006/ 155. Mao TL, Seidman JD, Kurman RJ, Shih Ie M. Cyclin E and p16
gyno.1999.5344. immunoreactivity in epithelioid trophoblastic tumor—an aid
135. Scully RE, Young RH. Trophoblastic pseudotumor: a reappraisal. in differential diagnosis. Am J Surg Pathol. 2006;30:1105–10.
Am J Surg Pathol. 1981;5:75–6. https://doi.org/10.1097/01.pas.0000209854.28282.87.
136. Young RH, Scully RE. Placental-site trophoblastic tumor: current 156. Shih IM, Kurman RJ. p63 expression is useful in the distinction
status. Clin Obstet Gynecol. 1984;27:248–58. of epithelioid trophoblastic and placental site trophoblastic tumors
137. Young RH, Scully RE, McCluskey RT. A distinctive glomerular by profiling trophoblastic subpopulations. Am J Surg Pathol.
lesion complicating placental site trophoblastic tumor: report of 2004;28:1177–83.
two cases. Hum Pathol. 1985;16:35–42. 157. Xu ML, Yang B, Carcangiu ML, Hui P. Epithelioid trophoblastic
138. Mazzucco G, Colla L, Monga G. Kidney involvement in a patient tumor: comparative genomic hybridization and diagnostic DNA
affected by placental site trophoblastic tumor. Am J Kidney Dis. genotyping. Mod Pathol. 2009;22:232–8. https://doi.org/10.1038/
2011;57:516–20. https://doi.org/10.1053/j.ajkd.2010.11.019. modpathol.2008.165.
139. Shih IM, Kurman RJ. The pathology of intermediate trophoblastic 158. Tsai HW, Lin CP, Chou CY, Li CF, Chow NH, Shih IM, Ho
tumors and tumor-like lesions. Int J Gynecol Pathol. 2001;20:31–47. CL. Placental site nodule transformed into a malignant epi-
140. Hui P, Wang HL, Chu P, Yang B, Huang J, Baergen RN, Sklar J, thelioid trophoblastic tumour with pelvic lymph node and
Yang XJ, Soslow RA. Absence of Y chromosome in human pla- lung metastasis. Histopathology. 2008;53:601–4. https://doi.
cental site trophoblastic tumor. Mod Pathol. 2007;20:1055–60. org/10.1111/j.1365-2559.2008.03145.x.
https://doi.org/10.1038/modpathol.3800941. 159. Lewin SN, Aghajanian C, Moreira AL, Soslow RA. Extrauterine
141. Hui P, Riba A, Pejovic T, Johnson T, Baergen RN, Ward epithelioid trophoblastic tumors presenting as primary lung car-
D. Comparative genomic hybridization study of placental cinomas: morphologic and immunohistochemical features to
site trophoblastic tumour: a report of four cases. Mod Pathol. resolve a diagnostic dilemma. Am J Surg Pathol. 2009;33:1809–
2004;17:248–51. https://doi.org/10.1038/modpathol.3800025. 14. https://doi.org/10.1097/PAS.0b013e3181b9cd67.
142. Hui P, Parkash V, Perkins AS, Carcangiu ML. Pathogenesis of 160. Narita F, Takeuchi K, Hamana S, Ohbayashi C, Ayata M, Maruo
placental site trophoblastic tumor may require the presence of a T. Epithelioid trophoblastic tumor (ETT) initially interpreted as
paternally derived X chromosome. Lab Investig. 2000;80:965–72. cervical cancer. Int J Gynecol Cancer. 2003;13:551–4.
143. Xue WC, Guan XY, Ngan HY, Shen DH, Khoo US, Cheung 161. Deavers MT, Kalhor N, Silva EG. Diagnostic problems with
AN. Malignant placental site trophoblastic tumor: a cytogenetic trophoblastic lesions. Arch Pathol Lab Med. 2008;132:168–74.
study using comparative genomic hybridization and chromo- https://doi.org/10.1043/1543-2165(2008)132[168:DPWTL]2.0
some in situ hybridization. Cancer. 2002;94:2288–94. https://doi. .CO;2.
org/10.1002/cncr.10424. 162. Lurain JR. Gestational trophoblastic disease II: classification
144. Feltmate CM, Genest DR, Wise L, Bernstein MR, Goldstein DP, and management of gestational trophoblastic neoplasia. Am
Berkowitz RS. Placental site trophoblastic tumor: a 17-year expe- J Obstet Gynecol. 2011;204:11–8. https://doi.org/10.1016/j.
rience at the New England Trophoblastic Disease Center. Gynecol ajog.2010.06.072.
Oncol. 2001;82:415–9. https://doi.org/10.1006/gyno.2001.6265. 163. Young RH, Kurman RJ, Scully RE. Proliferations and tumors of
145. Papadopoulos AJ, Foskett M, Seckl MJ, McNeish I, Paradinas FJ, intermediate trophoblast of the placental site. Semin Diagn Pathol.
Rees H, Newlands ES. Twenty-five years’ clinical experience with 1988;5:223–37.
placental site trophoblastic tumors. J Reprod Med. 2002;47:460–4. 164. Wan SK, Lam PW, Pau MY, Chan JK. Multiclefted nuclei. A help-
146. Finkler NJ. Placental site trophoblastic tumor. Diagnosis, clinical ful feature for identification of intermediate trophoblastic cells in
behavior and treatment. J Reprod Med. 1991;36:27–30. uterine curetting specimens. Am J Surg Pathol. 1992;16:1226–32.
147. Schmid P, Nagai Y, Agarwal R, Hancock B, Savage PM, Sebire 165. Yeh IT, O'Connor DM, Kurman RJ. Intermediate trophoblast:
NJ, Lindsay I, Wells M, Fisher RA, Short D, Newlands ES, further immunocytochemical characterization. Mod Pathol.
Wischnewsky MB, Seckl MJ. Prognostic markers and long-term 1990;3:282–7.
outcome of placental-site trophoblastic tumours: a retrospec- 166. Dotto J, Hui P. Lack of genetic association between exaggerated
tive observational study. Lancet. 2009;374:48–55. https://doi. placental site reaction and placental site trophoblastic tumor.
org/10.1016/S0140-6736(09)60618-8. Int J Gynecol Pathol. 2008;27:562–7. https://doi.org/10.1097/
148. Fadare O, Parkash V, Carcangiu ML, Hui P. Epithelioid tropho- PGP.0b013e31816d1d00.
blastic tumor: clinicopathological features with an emphasis on 167. Young RH, Kurman RJ, Scully RE. Placental site nodules and
uterine cervical involvement. Mod Pathol. 2006;19:75–82. https:// plaques. A clinicopathologic analysis of 20 cases. Am J Surg
doi.org/10.1038/modpathol.3800485. Pathol. 1990;14:1001–9.
149. Hamazaki S, Nakamoto S, Okino T, Tsukayama C, Mori M, 168. Huettner PC, Gersell DJ. Placental site nodule: a clinicopatho-
Taguchi K, Okada S. Epithelioid trophoblastic tumor: morpholog- logic study of 38 cases. Int J Gynecol Pathol. 1994;13:191–8.
ical and immunohistochemical study of three lung lesions. Hum 169. Shih IM, Seidman JD, Kurman RJ. Placental site nodule and char-
Pathol. 1999;30:1321–7. acterization of distinctive types of intermediate trophoblast. Hum
150. Kuo KT, Chen MJ, Lin MC. Epithelioid trophoblastic tumor of Pathol. 1999;30:687–94.
the broad ligament: a case report and review of the literature. Am 170. Shitabata PK, Rutgers JL. The placental site nodule: an immuno-
J Surg Pathol. 2004;28:405–9. histochemical study. Hum Pathol. 1994;25:1295–301.
151. Shih IM, Kurman RJ. Epithelioid trophoblastic tumor: a neo- 171. Chen BJ, Cheng CJ, Chen WY. Transformation of a post-
plasm distinct from choriocarcinoma and placental site tro- cesarean section placental site nodule into a coexisting epi-
phoblastic tumor simulating carcinoma. Am J Surg Pathol. thelioid trophoblastic tumor and placental site trophoblastic
1998;22:1393–403. tumor: a case report. Diagn Pathol. 2013;8:85. https://doi.
152. Palmer JE, Macdonald M, Wells M, Hancock BW, Tidy org/10.1186/1746-1596-8-85.
JA. Epithelioid trophoblastic tumor: a review of the literature. J
Reprod Med. 2008;53:465–75.
Overview of Placenta Pathology
14
John Paul B. Govindavari and Anna R. Laury
J. P. B. Govindavari · A. R. Laury (*) As the placenta develops, the entire blastocyst surface is ini-
Department of Pathology and Laboratory Medicine, Cedars-Sinai tially covered by villi. To form the membranes, most of the
Medical Center, Los Angeles, CA, USA
newly formed villi regress and the intervillous space obliter-
e-mail: [email protected]
© Science Press & Springer Nature Singapore Pte Ltd. 2019 459
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_14
460 J. P. B. Govindavari and A. R. Laury
ates, a process which spreads over the blastocyst surface. trophoblastic cover and poorly developed fetal capillar-
The chorion, obliterated intervillous space, villous remnants, ies, and make up very little of the term placenta
and trophoblastic shell fuse, forming the smooth chorion. (Table 14.1).
The small cells that line the inner surface of the trophoblast
eventually form the amnionic epithelium. Early in gestation
the amnionic mesoderm and the chorionic mesoderm fuse, 14.1.5 General Rules of Pathologic Evaluation
but the fusion of the amnion and chorion is never complete,
and the two membranes can always easily slide against each When approaching a placenta, it is helpful to use a system-
other [1, 2]. atic approach to inspect the three main components: the disc,
the membranes, and the umbilical cord. The order is less
important than a standard routine, although in practice it may
14.1.3 Twin/Multiple Pregnancy Development be easiest to examine the cord and membranes prior to the
disc as these will be removed before obtaining the placental
This will be covered later in the chapter. weight. Additionally, it is important not to overlook any
extraplacental tissue (such as a fetus papyraceus or large
amounts of blood clot) that may be present [2, 4–7].
14.1.4 Chorionic Villi
a b
Fig. 14.1 Tertiary chorionic villi. (a) Three types of villi are readily contain scant connective tissue, with more than 50% of the volume con-
identified in this photomicrograph. The large stem villus (SV) is notable sisting of capillaries. (b) Immature intermediate villus (IV) with reticu-
for its dense fibrotic stroma and large vessels. The mature intermediate lar stroma and fluid-filled channels. (c) Most of the villi in this picture
villus (MV) has a dense cellular stroma, and over 50% of the volume is represent terminal villi (TV). There is scant connective tissue and a thin
taken up by capillaries. Terminal villi (TV) are the smallest villi, and trophoblastic layer, which directly contacts the dilated capillaries
refers to the presence or absence of disrupted cotyledons. that the cotyledons may be disrupted, but still comprise a
In an intact placenta, the entire maternal surface will appear complete placenta when pieced together, though this can be
smooth and shiny due to the (thin) layer of decidua that is very difficult to determine with certainty (Fig. 14.3).
normally present. Areas of disruption appear roughened Missing placental tissue affects accurate weight measure-
and reddish, due to exposed villi (Fig. 14.2a, b). ment and more importantly can indicate retained placental
Completeness refers to receipt of the entire placenta. Note tissue within the uterus.
462 J. P. B. Govindavari and A. R. Laury
a b
Fig. 14.2 Integrity of the placental disc. (a) Maternal surface with exposed villi, near where the umbilical cord emerges. (b) The focus of
mostly intact cotyledons, which are identified by their smooth, slightly disruption closer up. The roughened surface is clearly in contrast with
shiny appearance. There is a focus of roughening, corresponding to the smooth, shiny intact cotyledons
14.2.1.2 Shape
While usually round to ovoid, placental shape is highly vari-
able, and of not much importance as long as the overall
growth/weight is appropriate. Uterine cavity abnormalities
(i.e., septate uteri, scars, and leiomyomas) can result in
unusual shapes. For example, should a placenta implant in
the lateral uterine sulcus, a bilobate placenta can result, with
a velamentously inserted cord between the lobes (Fig. 14.5).
In the early stages of placental development, the entire sac
is covered by chorionic villi, most of which regress and flat-
Fig. 14.3 Completeness of the placental disc. This maternal surface is ten out to form the smooth fetal membranes. A succenturiate
composed predominantly of intact cotyledons; however, one fragment
is received separately (bottom of image). In this instance, the placenta lobe (Fig. 14.6) results when an area away from the main
is likely disrupted but complete placental disc fails to regress [7, 9].
14 Overview of Placenta Pathology 463
14.2.1.3 Color
Color of the villous tissue or placental parenchyma is mostly
determined by the hemoglobin content, with immature placen-
tal parenchyma appearing significantly paler than discs at
term. Color is not particularly specific, but may correlate with
a clinical setting or suggest a diagnosis. Hydropic placentas,
which are usually seen with hydropic infants and fetuses, will
have a very pale color with coarse texture on cut section.
Massive fetomaternal hemorrhage results in a remarkably pale
placenta, often with an otherwise unremarkable exam, though
intervillous thrombi are not unusual (Fig. 14.7). Very deep red
placentas suggest increased blood content or congestion, such
as in twin-twin transfusion syndrome [2, 5–7, 10] (Fig. 14.8).
14.2.1.4 Weight
Fig. 14.4 Disruption of the placental disc. This placenta required man-
ual removal for presumed placenta accreta. The maternal surface is The placenta should be weighed with the membranes
markedly disrupted; completeness cannot be assessed trimmed and the cord removed. As stated previously, fixation
Fig. 14.7 Placental pallor in a term placenta. This cross section of the
parenchyma reveals notable placental pallor, as well as a small intervil-
lous thrombus (top middle), but is otherwise normal. This is an example
of massive fetomaternal hemorrhage (image courtesy of Stephanie
Yagi, DO)
Fig. 14.5 Placental shape. This is a bilobed singleton placenta with a mem-
branously inserted umbilical cord located between the two discrete lobes
Fig. 14.6 Succenturiate lobe. There is a separate island of well-developed Fig. 14.8 Twin-twin transfusion syndrome. The contrast between the
placental parenchyma located away from the main disc. Note the large- markedly pale donor twin side and the very dark, congested, recipient
sized vessels traversing the membranes to reach the accessory lobe twin territory is readily apparent
464 J. P. B. Govindavari and A. R. Laury
can increase the weight by up to 10%. There are established the infant and more may be removed when retrieving cord
normal weight ranges for placentas, and small placentas blood, etc. [12, 13]. Long cords are associated with an
(<10th percentile) and large placentas (>90th percentile) are increased incidence of cord entanglement, cord prolapse, true
associated with varying pathologies and maternal conditions. knots, excessive coiling, and constriction. Short cords are cor-
Ideally, weight norms should be established for the popula- related with a range of neonatal problems (from fetal distress
tion in which the pathologist practices, though this is not and low APGAR scores to depressed IQ and developmental
always simple. Large-for-gestational-age placentas can be anomalies). However, as cord length is thought to arise from
seen in diabetic pregnancies, maternal anemia, or retropla- fetal movement, it is unclear whether the short cord is a con-
cental hematoma. Fetal factors leading to a large for gesta- sequence or cause of these abnormalities [12, 13].
tion age can include states of fetal stress such as hypoxia or
respiratory distress. Small-for-gestational-age placentas can 14.2.2.3 Cord Insertion
be seen in cases of reduced uteroplacental blood flow, such The umbilical cord may insert into the chorionic plate at any
as gestational hypertension or preeclampsia, as well as a location; the insertion site is described qualitatively as cen-
variety of fetal conditions [2, 6, 7, 10, 11]. tral, eccentric (sometimes called paracentral), marginal, and
velamentous (membranous). Central and eccentric insertions
account for more than 90% of placentas. Velamentous is the
14.2.2 Umbilical Cord least frequent (and most worrisome); the rest are marginal
(Fig. 14.9).
The umbilical cord connects the developing fetus to the pla- Velamentous insertion is considered more susceptible to
centa. The cord contains three vessels: two umbilical arteries vessel rupture, and is associated with intrauterine growth
which carried deoxygenated blood from the fetus to the pla- restriction, stillbirth, and neonatal death. Interestingly, the
centa, and a single, larger, umbilical vein, which carries oxy- incidence of velamentous insertion increases with maternal
genated blood from the placenta to the fetus. The protective smoking, advanced maternal age, and diabetes mellitus, as
gelatinous connective tissue in which the vessels course is well as in multiple gestations, congenital malformations, and
called Wharton’s jelly; it is composed mainly of mucopoly- pregnancies achieved with in vitro fertilization. The vessels
saccharides with scattered fibroblasts and macrophages. At of these cords are more vulnerable to injury because they
term, the umbilical cord is pearly white in color, about 60 cm lack the protective layer of Wharton’s jelly as they course
in length, averages 1.0–1.5 cm in diameter, and is helical in through the membranes to the disc (Fig. 14.10). Marginally
shape, with an average of 1–3 coils for every 10 cm [12, 13]. inserted cords (those less than 1 cm from the disc edge) are
The cord inserts into the fetal surface of the placenta disc similarly associated with adverse outcomes, though not as
[14]. When examining the umbilical cord, it is important to
note color, length, diameter, coiling, insertion, and configu-
ration, and to assess the vasculature both for number of ves-
sels and for thrombi or hematomas.
14.2.2.1 Color
The normal cord is pearly white. If discolored, there will
usually be a similar discoloration of the fetal surface and/or
membranes. Green or brown discoloration is seen with
meconium discharge before delivery. Yellowed cords are
associated with maternal hyperbilirubinemia, or acute funisi-
tis accompanying an ascending infection. Focal plaque-like
discolorations of the cord are often noted in candidal infec-
tions of the amniotic fluid. Red discoloration can be iatro-
genic but may also be due to necrosis, thrombosis, or
hemolysis in the setting of fetal demise [12, 13, 15].
14.2.2.2 Length
The average length of an umbilical cord at term is 60 cm. A
long cord is defined as one measuring over 70 cm, and short
Fig. 14.9 Marginal insertion of the umbilical cord. The cord insertion
cords as 40 cm or less. It is important to note that calling a is directly at the disc edge. There is a vessel running unprotected
cord a “short cord” at the time of gross examination may not through the membranes (left side); however, the cord itself does not
be accurate due to the fact that some length is left attached to insert into the membranes
14 Overview of Placenta Pathology 465
14.2.2.4 Coiling Fig. 14.11 Cord coiling. (a) A normally coiled umbilical cord. (b)
The usual umbilical cord has a counterclockwise/left-turn Hypercoiled umbilical cord (>3 coils per 10 cm) in a preterm gestation.
(c) Hypocoiled umbilical cord (<1 coil per 10 cm). There is essentially
coil. The average umbilical coiling index (UCI) is approxi- no cord twist in this term cord
mately 0.3 coils/cm (or about 1 coil every 5 cm) [16]. Both
hyper- (>90th percentile) and hypocoiled (<10th percentile)
cords are associated with adverse outcomes in the preg- 14.2.2.6 Thrombi
nancy, with hypercoiled cords associated with fetal growth Venous thromboses are more common than arterial thrombo-
restriction and hypocoiled cords associated with meconium sis (with the latter being more often lethal). The cause is usu-
staining, low Apgar scores, and NICU admission [17] ally physical compression of umbilical vessels and/or
(Fig. 14.11). damage to the vessel walls, but there may be an underlying
cause (e.g., maternal diabetes, hypercoagulable state).
14.2.2.5 Single Umbilical Artery Grossly, areas of thrombosis can show swelling and discolor-
As stated previously, there are normally two arteries and ation, and at times are visible on cut section; old thrombi
one vein in the human umbilical cord. Initially, a second may calcify [12, 13].
umbilical vein is present, but it atrophies during the second
month of pregnancy [18–20]. Single umbilical artery
(SUA) is the commonest true congenital anomaly of 14.2.2.7 Hematoma
humans, with a prevalence ranging from 0.2% to 11%, While rare, true hematomas have serious consequences when
depending on the population studied [18–20]. The loss of they occur, and are associated with a 50% fetal mortality
an umbilical artery is secondary to the thrombotic atrophy rate. They occur due to rupture of one or more umbilical ves-
of a normal artery or, less commonly, due to primary agen- sels with varying underlying causes including short cords,
esis of an artery. The vast majority of children with an iso- trauma, inflammation, aneurysms, hemangiomas, and cord
lated SUA grow and develop normally. However, the rate entanglement. Grossly, there will be elongated, fusiform
of chromosomal abnormalities and/or associated congeni- swelling of the cord and marked dark red discoloration. It is
tal malformations is increased in fetuses with important not to confuse this with iatrogenic cord hemor-
SUA. Congenital malformations associated with single rhage associated with cord clamping. Cut sectioning will
umbilical artery include neural tube and cardiac defects, as reveal hemorrhage throughout the cord. The cord surface
well as respiratory, gastrointestinal, musculoskeletal, and may range from unstained (acute rupture) to red discolor-
genitourinary anomalies [19]. ation (in subacute to remote ruptures) [12, 13].
466 J. P. B. Govindavari and A. R. Laury
a b
Fig. 14.12 (a) False knot. Dilated umbilical vessels are present in the midportion of the cord. (b) True knot. This knot was easily reducible; note
the absence of congestion or apparent constriction on either side of the knot
Fig. 14.14 Amniotic band. The presence of free strips of amnion, seen
here adjacent to the umbilical cord insertion site, indicates disruption of
the amniotic sac. This example likely occurred at the time of delivery
tions with demise of one or more fetuses, the tissue may not
be completely resorbed. In this situation, a firm mummified
fetus papyraceous can be identified; depending on the gesta-
tional age of demise and time of delivery the tissue can range
c from very firm and sclerotic to semi-firm and necrotic.
Placenta membranacea, also called placenta diffusa, is a rare
developmental abnormality of the human placenta in which
fetal membranes are covered diffusely by chorionic villi of
varying thickness [2] (Fig. 14.15).
a b
c d
Fig. 14.15 Plaques on the fetal membranes. (a) Squamous metaplasia. associated with prolonged oligohydramnios. They are removable. (c)
These small plaques are commonly seen in term placentas, especially Yolk sac remnant. Present as a single, firm nodule in the free membranes
near the umbilical cord insertion site. They are not removable. (b) or fetal surface of the disc, this is a normal, if unusual, finding. (d) Fetus
Amnion nodosum. The plaques (upper left, inset) are pathologic and papyraceous. The dark pigment of the retina often stands out (arrow)
abruption”). If bleeding is confined behind the placenta for a prenatal ultrasound. The majority of these are not otherwise
significant amount of time, the overlying hematoma com- clinically significant and represent chorionic or subchorionic
presses the villous tissue and can lead to infarct, with notable hemorrhage; the appearance can be relatively dramatic
thinning of the disc in that area (Fig. 14.16). Chronic mar- (Fig. 14.18).
ginal retroplacental hemorrhage can result in chorioamniotic
hemosiderosis, which reflects hemorrhage into the amniotic 14.2.4.3 Parenchyma
fluid [2]. Cut sectioning of the normal placental disc reveals a dark
red, spongy surface. Serial sectioning of the disc at regular
14.2.4.2 Fetal Surface intervals (approximately 1–2 cm) is important to unveil any
The fetal surface is assessed for subchorionic fibrin and infarcts or thrombi.
thrombotic material, both of which can increase through Obstruction of the uteroplacental (maternal) circulation
pregnancy. Patchy fibrin deposition is usually of little signifi- leads to placental infarcts, defined in part by villous necrosis.
cance. A dull, opaque appearance to the fetal surface can be Minor infarcts are seen in about a quarter of placentas, but
a sign of ascending infection, with more severe cases appear- they are not of much clinical significance unless at least 30%
ing more dramatic due to the numerous neutrophils and of the placenta is involved. Single infarcts, especially at the
tissue necrosis (Fig. 14.17). Thrombosed fetal vessels are not placental margin, are not unusual in term placentas, but
rare, with chalky deposits representing calcified old throm- remember that any infarct in a preterm placenta is considered
boses [2, 5, 10]. In some cases, a cyst may be identified on abnormal [2, 4, 5]. Compared to adjacent tissue, placental
14 Overview of Placenta Pathology 469
a b
Fig. 14.16 Retroplacental hemorrhage. (a) Maternal surface showing adherent blood clot, consistent with hemorrhage occurring prior to delivery.
(b) Cross section of remote retroplacental hemorrhage; the underlying placental parenchyma is thinned due to the presence of the confined blood
a b
Fig. 14.17 Discolored membranes. (a) Green-tinged membranes are suggestive of meconium staining. (b) Dull, opaque membranes are often
associated with a significant acute fetal inflammatory response, such as to amniotic fluid infection
infarcts are firmer and granular; the appearance progresses pallor that retains the usual consistency of villous parenchyma
from red and hemorrhagic to yellow to tan-white over time. (as contrasted with the denser firmness of an infarct). The base
Palpation can aid in detecting earlier infarcts, as can formalin of the triangle is usually located at the basal plate and can
fixation. Infarcts are more common at the periphery, with sometimes be visualized more easily in a fixed specimen [23].
central infarcts more indicative of true pathology, especially Intervillous thrombi are found within the parenchyma,
if multiple [2, 5] (Fig. 14.19). expanding the intervillous space. Early thrombi are fresh red
Clusters of avascular villi are caused by prolonged or recur- clots, which evolve to laminated thrombi, and eventually
rent occlusions within the fetal vasculature, which can eventu- white lesions. They can be differentiated from infarcts by
ally lead to a whole branch of the villous tree becoming their shiny quality and lack of granularity [2, 5] (Fig. 14.20).
avascular and atrophic. If there is a large enough focus of avas- An entity that can be confused with infarcts and thrombi
cular villi, it may be visible grossly as an area of triangular is the chorangioma, a firm fleshy lesion usually found under
470 J. P. B. Govindavari and A. R. Laury
a b
Fig. 14.18 Placental surface cysts. (a) A relatively large subamniotic/subchorionic cyst, or hematoma. As here, this finding is most often seen near
the cord insertion. (b) Multiple small amniotic and subamniotic/chorionic cysts, some with hemorrhage. These are of no clinical consequence
cm 1 2 3 4 5
14.2.6.1 Chorionicity
One of the first things that should be assessed grossly is the
chorionicity (how many discs are present) and amnionicity
(how many amniotic sacs are present) of the placenta. Zygosity
cannot be determined by placental examination, with the excep-
Fig. 14.21 Chorangioma. (a) This mass lesion is well defined, multi- tion of monochorionic placentas, which occur only with mono-
lobulated, and somewhat fleshy. (b) The lesion is clearly demarcated zygosity (taking care to exclude fused discs). Useful algorithms
from the surrounding parenchyma. (c) On high power, the mass is com- have been described previously regarding the assessment of
posed of small, bland capillary-type vascular structures set within in a twin placentation (histologic depictions of various membrane
hyalinized stroma
configurations are presented later in this chapter):
Perinatal mortality of twins is roughly five times that of (a) Separate or fused discs: Separate discs (connected by only
singletons, largely due to increased prematurity rates in the membranes) are diagnostic of diamniotic-dichorionic pla-
twin population. Monochorionic twins, in particular, have centas. Fused discs are dichorionic and usually diamniotic
472 J. P. B. Govindavari and A. R. Laury
14.2.6.2 Anastomoses
Regarding inter-twin anastomoses, it is useful to note that
Fig. 14.23 Laser ablation of anastomoses for twin-twin transfusion
chorionic arteries are identified by their tendency to run syndrome. The irregular to oval areas in the center (arrow) are the sites
superficial to their accompanying veins (arteries travel over of in utero laser ablation
veins). AA and VV anastomoses are superficial and do not
penetrate the chorionic plate, and are usually of minimal sig-
nificance. Arteriovenous (AV) anastomoses, however, are dye is injected through these, allowing the examiner to visu-
deeper, occur at the villous capillary level, and cannot be alize the anastomoses [11, 25, 26]. In practice, this is very
seen when examining the chorionic plate. One can instead rarely performed.
describe/sample the sites of an unpaired artery and vein in
close proximity to one another [11, 25, 26] (Fig. 14.22). 14.2.6.3 Laser Treatment
Injection studies are not routinely needed, but may be In cases of severe chronic TTTS, possible prenatal treat-
pursued for academic purposes, at clinician/parent request, ments include serial amnioreduction, and fetoscopic laser
or in cases of laser ablation for twin-twin transfusion syn- coagulation (ablation) of the communicating vessels.
drome (TTTS). To perform an injection study, the umbilical Depending on the timing of the photocoagulation, gross find-
cords are sectioned, with a stump left for cannulation of the ings range from hemorrhagic clotted vessels with interrup-
vessels. Manual “milking” of the superficial vessels removes tion of dye filling (if examined within 1 month of
remaining intravascular blood. Umbilical catheters are photocoagulation) to absence of inter-twin anastomoses with
inserted into the umbilical vein and arteries of each cord and subchorionic fibrin deposition [25] (Fig. 14.23).
14 Overview of Placenta Pathology 473
a b
Fig. 14.24 Embryologic remnants in the umbilical cord. (a) Allantoic remnant. As seen here, these are most often lined by mucin-rich epithe-
duct remnant. These structures are lined by a cuboidal to flat epithe- lial cells, but rarely other tissue types are noted. A muscular layer or
lium, resembling transitional type epithelium or urothelium. A muscu- associated small vessels are occasionally seen
lar wall is always lacking. (b) Vitelline (omphalomesenteric) duct
474 J. P. B. Govindavari and A. R. Laury
a b
c d
e f
Fig. 14.25 Acute inflammation in the umbilical cord. (a) Funisitis. umbilical vessel. (d) A robust neutrophilic infiltrate extending beyond
Low power of the umbilical cord with acute inflammation marginating the vessel and into Wharton’s jelly (funisitis). (e) Gross appearance of a
out of the vessel and spilling into Wharton’s jelly (arrow). (b) Higher thrombosed umbilical artery (arrow). (f) Thrombosed and calcified
power of an umbilical vessel with neutrophils present between the mus- umbilical artery
cle fibers. (c) Acute inflammation confined to the muscular wall of the
14 Overview of Placenta Pathology 475
tion below), these findings can also be seen with in utero (c) Chorion, including the sclerotic villi of early gestation
meconium exposure. When associated with meconium, the (chorion laeve) and trophoblasts
inflammation tends to be more pronounced in the umbilical (d) Maternal decidua
cord than the chorionic plate, and meconium-laden macro-
phages should be identifiable histologically. 14.3.2.2 Meconium
Meconium, the bile-stained intestinal content of the fetus, is
14.3.1.3 Thrombi occasionally expressed at delivery or prior to delivery. If
Thrombi will show usual organization with alternating expelled into the amniotic fluid well prior to delivery, it can
fibrin and clot within the lumen, as in any other vessel [12, stain and irritate the fetal membranes, will be preserved
13, 27]. within macrophages, and can therefore be identified histo-
logically. Meconium expressed at the time of delivery will
rinse off easily and won’t be found within macrophages on
14.3.2 Membranes histologic examination. Meconium-laden macrophages are
large, are round to ovoid, have slightly granular yellow-
When examining the membranes histologically (Fig. 14.26), brown material filling their cytoplasm, and are usually iden-
the pathology to be evaluated ranges from pigment in the tified in the subamniotic spongy layer. If meconium has been
membranes to evidence of an inflammatory response and to present in the amniotic cavity for many hours, the amnionic
searching for maternal arteriopathy within the decidua. epithelium may show degenerative changes, including vacu-
olization of the cytoplasm, heaping up of cells, hobnailing,
14.3.2.1 Basic Structure dissociation, loss of cells, and necrosis. Meconium staining
Histology of the extraplacental membranes consists of: affects the amnion before the chorion, as the macrophages
travel through the layers of the membranes (Fig. 14.27).
(a) Cuboidal amniotic epithelium, with its underlying base-
ment membrane 14.3.2.3 Squamous Metaplasia
(b) A hypocellular “spongy” layer with scattered fibroblasts, This is a finding that is present in more than half of term
which is really a potential space placentas. It is not a true pathologic finding, but a sign of
placental maturity. The amniotic epithelium becomes strati-
fied with focal keratinization of the epithelium, resembling
squamous epithelium. It is most often identified near the
umbilical cord insertion (Fig. 14.28a).
7
1
2 3
a b
c
d
Fig. 14.28 Fetal membranes. (a) Squamous metaplasia. Keratinizing sac remnant. A well-defined, circumscribed, and calcified plaque, just
epithelium is present, in contrast to the normal cuboidal amniocytes under the amniotic epithelium. (d) Fetus papyraceous, with ghost out-
(upper right of image). (b) Amnion nodosum. Grossly identified plaques lines of devitalized and calcified fetal tissues
are composed of degenerated squamous cells, hair, and debris. (c) Yolk
14.3.2.4 Amnion Nodosum absent, and there are usually few signs of degeneration [21].
This condition is associated with prolonged oligohydram- Amniotic bands need to be assessed grossly.
nios of any cause, though the mechanism is not understood.
Microscopically, the nodules are composed of sloughed fetal 14.3.2.6 Acute Chorioamnionitis
squamous cells and hair, mixed with hyaline and protein- The pattern of pathologic findings commonly referred to as
aceous debris. There is no associated inflammation. The nod- “chorioamnionitis” is most often indicative of ascending
ules are usually situated superficially on the amniotic surface, amniotic fluid infection. Its definition is the presence of neu-
but sometimes extend through to the spongy layer trophils (PMNs) within the fetal membranes. Eosinophils
(Fig. 14.28b). Other lesions such as yolk sac remnant and may be present, especially in protracted infections, though
fetus papyraceous may be seen in Fig. 14.28c, d chronic inflammatory cells are generally not a significant
respectively. component of the response. PMNs may be present within the
decidua and/or chorion, even in the absence of an infectious
14.3.2.5 Amniotic Bands process. The term “acute chorioamnionitis” should be
Histologic sections show normal-appearing amnionic epithe- reserved for instances of inflammation extending into the
lium and underlying connective tissue. Inflammation is subepithelial hypocellular space to involve all layers of the
14 Overview of Placenta Pathology 477
a b
Fig. 14.29 Acute chorioamnionitis. (a) Acute inflammation extends infiltrates and necrosis of the amniotic epithelium (arrow). (c)
all the way up from the decidua (bottom of image) through the chorion Chorionitis. The neutrophils are dense within the chorion, but have not
and spongy layer into the amnion and is present just under the epithe- yet crossed into the reticular or amniotic layers; chorioamnionitis
lium. (b) Necrotizing chorioamnionitis with confluent neutrophilic should not be reported here
membranes [15, 23, 28] (Fig. 14.29). See also the section on chorionic villi. These free villi are covered by a thin syn-
amniotic fluid infection sequence. cytiotrophoblastic layer, with a discontinuous cytotro-
phoblastic layer, which is in contact with sinusoidal
capillaries. The dense collection of capillaries and sinu-
14.3.3 Placental Disc soids in terminal villi comprises at least 50% of the stro-
mal volume in these small structures. Syncytial knots,
Findings within the parenchyma of the disc (and its compart- which are aggregates of syncytiotrophoblast nuclei pres-
ments) are often related to lesions initially discovered upon ent along one surface of terminal villi, are also a sign of
gross examination, and to histological findings of the umbili- villous maturity (Fig. 14.30).
cal cord and membranes. (a) Hypermaturity: In instances of underperfusion, or
A brief discussion of villous maturation is merited before placental hypoxia, the tertiary villi respond to
additional findings of the disc are elaborated. Various types decreased blood flow and poor oxygenation by
of villi have been discussed previously; what follows is a exhibiting accelerated maturation. There is an
summary of what to look for in terminal villi when assessing increase in mature tertiary free villi (relative to ges-
maturity. tational age), as well as an increase in syncytial
knots (so-called Tenney-Parker change). It is gener-
1. Maturity mature terminal villi are found in the third tri- ally agreed that knots in >20% villi preterm or >30%
mester of pregnancy, and are the smallest versions of at term are considered increased, though this is a
478 J. P. B. Govindavari and A. R. Laury
a b
c d
Fig. 14.31 Maternal decidual arteriopathy. (a) Transformed vessels muscular wall. (c) Fibrinoid necrosis with acute atherosis. The brightly
within the decidua; there is no discernable smooth muscle layer and the eosinophilic vessels stand out from the pale decidual cells in this exam-
lumens are widely patent. (b) Decidual vessels with retention of the ple. (d) Fibrinoid necrosis with atherosis, higher power. At this power,
the lipid-laden macrophages within the fibrinoid material are apparent
a b
c d
e f
Fig. 14.32 Villitis of unknown etiology (VUE). (a) Focal VUE. A one full-thickness section). (d) High-grade VUE, also a diffuse pattern.
cluster of hypercellular villi are present in the middle of the image. This example is relatively subtle; the top half of the image is composed
Though the finding is focal, this would be classified as high grade as of affected villi while the lower half is relatively spared. (e) Higher
>10 contiguous villi are involved. (b) On high power, the infiltrate is power appearance of the diffusely involved, sclerotic villi. (f) An immu-
identified as lymphocytic. Also note the lack of villous capillaries in the nohistochemical stain for CD3 (nuclear) confirms that the infiltrate is
affected villi. (c) High-grade VUE, diffuse pattern (involves >30% of composed predominantly of lymphocytes
14 Overview of Placenta Pathology 481
absent (Fig. 14.33). Foci of more recent infarction show increased, or more prominent, syncytial knots, which
more subtle changes, including villous crowding, con- retain their nuclear basophilia. In foci of avascular villi,
gestion, a mild inflammatory response, and stromal the villous stroma is eosinophilic and hyalinized, and
karyorrhexis [4–6, 10]. It is important to distinguish lacks stromal capillaries. Finally, the intervillous space
infarcts from avascular villi. is predominantly open and contains maternal blood; it is
6. Avascular Villi not obliterated as with an infarct [6, 10, 23] (Fig. 14.34).
Clusters of avascular villi tend to be closely apposed to 7. Intervillous Thrombi
normal villi, and thus appear sharply demarcated from If enough time has passed for the thrombus to be prop-
the surrounding parenchyma. This finding is differenti- erly organized, it will have the usual appearance of an
ated from an infarct in several ways. First, the tropho- organized clot, with the alternating red and pink lines of
blastic layer is still viable due to perfusion by maternal Zahn. The red is composed of entrapped red blood cells
blood in the intervillous space. There may also be and the pink is composed of fibrin. Grossly, it can be
a b
Fig. 14.33 (a) Edge of a placental infarct. The infarcted villi (top) are bordered by a rim of ischemic parenchyma, including prominent syncytial
knots. (b) The center of the infarct shows ghostlike villous structures with near-complete loss of nuclear basophilia
a b
*
*
Fig. 14.34 (a) The cluster of avascular villi (*) stands out from the contrasted with unaffected tissue on the right. Note the open intervil-
background of unaffected villi due to its brightly eosinophilic hyalin- lous space occupied by maternal blood, and the lack of significantly
ized stroma. Note the lack of fetal capillaries, and retained nuclear increased intervillous fibrin deposition
basophilia of the trophoblasts. (b) A larger focus of avascular villi (left)
482 J. P. B. Govindavari and A. R. Laury
a b
Fig. 14.35 (a) Intervillous thrombus with prominent laminations ischemia, including increased fibrin deposition, prominent syncytial
(lines of Zahn). (b) The edge of a large intervillous thrombus. As shown knots, and a mild chronic inflammatory response. Over time, the villi
here, the parenchyma just next to a thrombus can show evidence of can become completely infarcted
difficult at times to distinguish these from infarcts but (b) Partial hydatidiform mole (PHM): In a well-developed
microscopic examination usually makes it clear which partial hydatidiform mole, two relatively discrete vil-
pathology is present (Fig. 14.35). lous populations can be seen. One population is
8. Meconium composed of large hydropic structures, and the other of
Meconium-laden macrophages can also be seen in the smaller, more normal-appearing villi. The hydropic
chorionic plate of the disc, under the amniotic epithe- villi have occasional central cisterns, exhibit areas of
lium, just as in the membranes. If there is a history of lacy trophoblastic hyperplasia (without atypia), and are
meconium-stained fluids or if the fetal surface has a notable for their unusual peripheral architecture. The
green tinge, it may be prudent to carefully examine the hydropic villi of partial moles often have prominent
specimen microscopically. invaginations of the surface, creating a scalloped
9. Hydropic Change appearance (“fingers and toes,” or “coast of Norway”).
This finding is usually noted in very early gestations, There should also be evidence of fetal tissue, often
such as in therapeutic or missed abortion specimens. noted in the form of nucleated red blood cells.
Swollen or dilated villi, with lightly basophilic hypocel- Immunohistochemical staining for p57 will be positive
lular stroma and few to no residual vessels, are the hall- within the villous stromal cell nuclei (Fig. 14.36c, d).
marks of hydropic change. These changes can be (c) Complete hydatidiform mole (CHM): Well-developed
associated with degeneration due to fetal/embryonic examples of complete hydatidiform moles will have
demise, aneuploidy, partial hydatidiform mole (PHM), diffuse, rounded enlargement of the villi. Central cis-
or complete hydatidiform mole (PHM). A brief over- terns and inclusions are easily identified, and lacy cir-
view of molar pregnancy is present below; more infor- cumferential trophoblastic hyperplasia will be found.
mation on this topic is presented in Chap. 13. Prominent implantation site with notable trophoblas-
(a) Hydropic degeneration due to demise: In hydropic tic atypia is often associated with CHM, but this can
degeneration, a sign of lost pregnancy, there will be be a very subjective feature. No fetal tissues are pres-
enlarged round to oval villi with edematous stroma, a ent. CHM is due to uniparental (paternal) diploidy,
few residual vessels, atrophy of the syncytiotropho- and as such immunohistochemical staining for the
blasts, and evidence of embryonic or fetal tissue. maternally expressed protein p57 will be absent in the
While pseudocystic spaces may be seen, so-called villous stromal nuclei. Staining can be seen in the
inclusions (spaces lined by trophoblasts) within the extravillous trophoblasts, and does not exclude an
villi will be absent, and trophoblastic hyperplasia will interpretation as CHM (Fig. 14.36e–g).
not be present. In instances of demise, there is often Histologic diagnosis of suspected molar pregnancies,
an admixture of variably sclerotic and edematous villi particularly early in gestation, can be frustrating; it is not
[5] (Fig. 14.36a, b). entirely sensitive or specific, and suffers from high inter-
14 Overview of Placenta Pathology 483
a b
c d
e f
Fig. 14.36 (a) Hydropic and sclerotic changes in a trisomy 16 concep- atypia is lacking and circumferential proliferation is absent. Villous
tus. Central cisterns are present, but inclusions and trophoblast hyper- contours tend to be complex and irregular. (e) Complete hydatidiform
plasia are absent. The edematous and sclerotic changes are present mole (CHM) showing edematous change of most villi, with prominent
along a spectrum; a discrete dual population is not present. (b) Hydropic central cisterns. Circumferential trophoblastic hyperplasia is present
change in a trisomy 16 conceptus; note also the irregular villous con- (*). (f) Trophoblast inclusions (*) are a feature of CHM. (g)
tours without trophoblast hyperplasia. (c) Partial hydatidiform mole Trophoblastic hyperplasia with atypia in a CHM. (h)
(PHM). Large edematous villi with central cisterns (right) are admixed CHM. Immunohistochemistry for p57 shows complete loss of nuclear
with a second population of smaller, more typical-appearing (or fibrous) staining in the villous trophoblasts. Extravillous trophoblasts often
villi (left). Note the irregularly contoured villous in the lower left. (d) show multifocal positive nuclear staining
PHMs exhibit some degree of trophoblastic hyperplasia (*), but notable
484 J. P. B. Govindavari and A. R. Laury
g h
and intra-observer variability. Apart from immunohisto- more fibrous appearance of normal stem villi), and
chemical staining for p57 and clinical impression, there prominent fibromuscular vasculature with fibroblastic
are few useful tools to assist in separating molar gesta- overgrowth. Trophoblastic hyperplasia is absent
tions from non-molar pregnancies. In particular, it can be (Fig. 14.37b–d). As the pregnancy progresses, the abnor-
challenging to distinguish a PHM from other types of mal vascular component may become more pronounced,
aneuploid or hydropic gestations. In the clinical setting of and evidence of degenerative changes may appear [30].
a patient actively desiring pregnancy, the distinction can 11. Chorangiosis/Chorangiomatosis/Chorangioma
be immediately relevant. All three of the following conditions can be seen in
Conventional karyotyping and ploidy analysis cannot cases of chronic placental underperfusion, and are
distinguish between diandric triploidy and other triploid thought to be an adaptive response:
gestations, though a diploid result can exclude a partial (a) Chorangiosis: To diagnose chorangiosis, one needs
mole (and karyotype can identify a trisomy, for example). >/= 10 fields of placental parenchyma with >/= 10
More advanced genetic techniques may be of help, though terminal villi, each with 10 or more capillaries per
they are not yet routinely in use at most facilities [5, 29]. cross section. This is seen more frequently in diabetic
10. Placental Mesenchymal Dysplasia and preeclamptic pregnancies, and the placenta is
Placental mesenchymal dysplasia (PMD) is a rare, often large for gestational age. This is likely a normal
benign condition that is characterized by placentomeg- finding at high elevations.
aly. It is associated with Beckwith-Weidman syndrome (b) Chorangiomatosis: This extremely unusual lesion can
(in about a quarter of the cases), fetal growth restriction, be either localized or multifocal, and consists of
and gestational hypertension. Clinically, PMD is easily chorangioma-like lesions that, rather than forming a
mistaken for a molar pregnancy due to the striking cys- single nodule, extend into adjacent stem villi. The
tic villous changes; however, unlike molar gestations, it lesional vessels are surrounded by pericytes and there
can be associated with a viable, and even normal, fetus/ is increased stromal collagenization.
infant. (c) Chorangioma: This is an intraparenchymal nodule
On gross examination there will be an abundance of composed of capillaries, stromal cells, and surround-
tissue, or a large-for-gestational-age placenta ing trophoblasts. A villous structure is expanded by
(Fig. 14.37). Grossly visible cystic structures and pale, the abnormal vascular proliferation, and there may be
friable zones are typically interspersed within an other- associated trophoblastic hyperplasia with pleomor-
wise relatively normal-appearing parenchyma phism and atypia. These lesions are usually located in
(Fig. 14.37a). In third-trimester or term placentas, the the subchorionic or marginal zones of the disc, and
chorionic plate vasculature may be prominent, com- there can be either a single nodule or (less often) mul-
posed of dilated and tortuous vessels. Histologically, the tiple nodules. The majority of chorangiomas are inci-
changes of PMD vary somewhat with gestational age. dental, but if large (>5 cm) they can be associated
Even in early gestation, PMD is most notable for the with complications including polyhydramnios, fetal
large, hydropic stem villi with central cisterns (not inclu- thrombocytopenia, heart failure, and hydrops [24]
sions), myxoid change (in marked contrast with the (Table 14.2).
14 Overview of Placenta Pathology 485
a b
c d
Fig. 14.37 Placental mesenchymal dysplasia (PMD). (a) Gross with cisterns are one of the hallmark findings in PMD. (c) Myxoid
appearance. This 13-week conceptus was clinically compatible with a change with prominent vasculature and fibroblastic stromal overgrowth.
complete mole; the specimen contained >70 g of tissue. There are many (d) Higher power showing the unusual vascular changes
easily appreciated cystic structures (circled). (b) Edematous stem villi
a b
Fig. 14.38 (a) Diamniotic dichorionic twin placenta. Chorion (*) is present between the two layers of amnion. (b) Diamniotic monochorionic
twin placenta. The two layers of amnion are directly apposed
12. Microscopic Examination of the Twin Placenta The pathologic findings of MVM include both gross and
Microscopically, twin placentas are examined in much microscopic changes. Grossly, there is evidence of poor pla-
the same fashion as singletons, with a major exception cental growth (a small [<10th percentile]-for-gestational-age
being confirming chorionicity when analyzing the layers disc), and evidence of maternal vascular disruption/occlu-
of the dividing membranes. Dividing membranes from a sion such as infarcts and retroplacental hemorrhage
diamniotic, dichorionic placenta (Fig. 14.38a) will have (Fig. 14.39a).
a trophoblast layer in the middle; monochorionic divid- The microscopic changes include confirmation of infarc-
ing membranes will have amnion directly apposed to tion, abnormal spiral arteries (maternal decidual arteriopa-
amnion (Fig. 14.38b). thy), and features of placental hypoxia. The histologic
13. Lasers findings of infarction and decidual arteriopathy are described
Laser-treated vessels show necrosis with focal hemor- above (Fig. 14.39b–d). Features of placental hypoxia include
rhage, clusters of avascular villi, and fibrin deposition in accelerated villous maturation (villous hypermaturity; see
the underlying parenchyma [25]. above) and distal villous hypoplasia. Distal villous hypopla-
sia is characterized by a notable reduction in distal/terminal
villi in relation to the intermediate and stem villi. The remain-
14.4 Constellation Syndromes ing villi tend to be elongated, with prominent syncytial knots,
and evidence of accelerated maturation. This finding is best
Constellation syndromes are a group of disorders that encom- appreciated on low power, and should involve at least 30% of
pass findings in multiple parts of the placenta, both gross and one full-thickness section [5, 23].
microscopic, and are due to a single underlying etiology.
a b
c d
Fig. 14.39 Maternal vascular malperfusion in a placenta from an upper portion of the image. This could be seen immediately adjacent to
intrauterine fetal demise at 31 weeks. (a) Gross appearance of multiple an infarct, but this change was present diffusely throughout the pla-
central infarcts. (b) Maternal decidual arteriopathy with fibrinoid centa. (d) A smaller “microinfarct” which was not appreciated grossly.
necrosis. (c) Low-power appearance of a large infarct (bottom of The agglutinated focus of villi in the top middle of the image shows
image), contrasted with the hypermature villi (for 31 weeks) in the nuclear smudging and loss of fetal capillaries
The established findings of FVM are predominantly tified in the most distal portions of the villous tree, resulting
microscopic, but relevant gross findings include the presence in randomly scattered small foci of avascular villi
of an umbilical cord thrombus, thrombi within the chorionic (Fig. 14.34).
plate vasculature, or a tight umbilical cord knot. The princi- The other pattern, which is due to segmental/complete
pal finding microscopically is clusters of avascular villi, as occlusion of a large fetoplacental vessel (often by thrombus),
described previously. To be identified as a “cluster” there leads to larger, more discrete foci of degenerating down-
should be at least 3–4 villous structures affected, with large stream villi. These villi initially show degenerative changes
foci including 10 or more villous cross sections; at least 3 (villous stromal karyorrhexis, vascular remodeling) and
separate foci should be identified to qualify as a diagnosable eventually lose all of their vessels, which results in a large
lesion (Fig. 14.40d–e). focus of avascular villi (Fig. 14.40f). When extensive, this
Fetal vascular malperfusion can have variable distribution pattern has been termed fetal thrombotic vasculopathy and
patterns. If associated with obstructive umbilical cord lesions has been associated with increased risk of fetal CNS injury,
such as hypercoiling or stricture, abnormal placental inser- as well as other adverse outcomes.
tion site, or long-standing fetal entanglements, the overall The associated vascular changes, other than a well-
histological features are suggestive of globally increased developed unequivocal thrombus, are more subtle and con-
venous pressure and poor circulation. The changes are iden- troversial. Additional vascular lesions that may be important
488 J. P. B. Govindavari and A. R. Laury
a b
c d
e f
Fig. 14.40 Global fetal vascular malperfusion in the setting of a tight rhexis. Thought to be an early sign of fetal vascular malperfusion,
umbilical cord knot. (a) A large, organizing thrombus is present in a karyorrhectic debris with extravasated/fragmented red blood cells are
chorionic plate vessel (*). (b) Downstream from the large thrombus, present in the villous stroma. (e) A cluster of avascular villi (right side
foci of intramural fibrin deposition are present (*). (c) Partially obstruc- of image) are present adjacent to uninvolved villi. (f) A larger zone of
tive vascular changes in a stem villous. (d) Villous stromal karyor- avascular villi; uninvolved villi are present at the upper right
14 Overview of Placenta Pathology 489
Table 14.3 Grade and stage of the inflammatory response to ascending amniotic fluid infection (Amsterdam Placental Workshop Group
Consensus Statement [23])
Stage Grade
Maternal inflammatory response
(Evaluated in the membranes)
1. Acute chorionitis or subchorionitis: Neutrophils within subchorionic fibrin, subchorial 1. Not severe
intervillous space, or at the choriodecidual interface 2. Severe: Confluent neutrophils or
2. Acute chorioamnionitis: Neutrophils extend into fibrous chorion and/or amnion subchorionic microabscesses
3. Necrotizing chorioamnionitis: Karyorrhexhis of neutrophils, amniocyte necrosis, and/or
amnion basement membrane hypereosinophilia
Fetal inflammatory response
(Evaluated in the large fetal vessels of the umbilical cord and placental surface)
1. Chorionic vasculitis or umbilical phlebitis: Neutrophils are present within the vessel wall 1. Not severe
2. Involvement of umbilical vein, and at least one artery: Neutrophils are present in the walls 2. Severe: Confluent or near-confluent
of at least two umbilical vessels neutrophils within a vessel wall,
3. Necrotizing funisitis: Neutrophils extend through a vessel wall into Wharton jelly, associated with vascular smooth muscle
with karyorrhextic/necrotic debris attenuation
a
14.5 Specific Infections
14.5.1 Candida
14.5.2 Listeria Fig. 14.42 Candida infection. (a) Section of a membrane roll reveals
numerous yeast forms, some with signs of budding. (b) GMS stain of
the same placenta highlights the organisms
Listeria monocytogenes infection occurs in approximately
9/100,000 pregnancies, and about 20% of these infections white nodules can be identified scattered throughout the pla-
result in fetal or perinatal demise. Grossly, the amniotic fluid cental parenchyma. Histologically, the hallmark of listeriosis
appears meconium stained (even in premature deliveries is the presence of prominent acute intervillositis with micro-
unlikely to have passed meconium), and on cut section small abscess formation, which is what forms the white nodules
14 Overview of Placenta Pathology 491
seen grossly. The villi themselves are relatively spared from growth restriction. Grossly, the placenta affected by syphilis is
inflammation, as this is a hematogenously spread infection thick, friable, and frequently pale. In about half of the cases, the
that is acquired via the maternal blood. Organisms are usu- umbilical cord has a “barber pole” appearance with chalky
ally appreciable with a gram stain [6, 28]. white necrotic debris surrounding the vessels. Histologically,
the “barber pole” appearance is due to severe necrotizing
funisitis, with a prominent (mostly degenerated) acute inflam-
14.5.3 Toxoplasmosis matory cell infiltrate in Wharton’s jelly; spirochetes may be
identified with silver stains if antibiotic therapy had not yet
Infection with toxoplasmosis can have severe consequences, been initiated. The villi may show chronic lymphoplasmacytic
with 5–15% of infections resulting in stillbirth. Among sur- villitis, increased nucleated red blood cells, villous edema, and
viving infants, though most will appear asymptomatic, at increased numbers of Hofbauer cells [6, 15, 38].
least 85% will develop chorioretinitis if they are not treated.
Grossly, there are no specific features of this protozoal infec-
tion in the placenta. Histologically, the intracellular organ- 14.5.7 Zika
isms are found in amniotic epithelial cells, stromal cells
between amnion and chorion, and villous stromal Hofbauer Infection with the Zika virus has recently become a wide-
cells. Pseudocysts and true cysts may occasionally be seen, spread topic of interest as a significant congenital infec-
which harbor many organisms. Rupture of these cysts results tion. While not all mothers who have been exposed to the
in a lymphohistiocytic or granulomatous villitis with multi- virus will give birth to affected infants, the recent surge
nucleated giant cells and necrosis [6, 28]. in incidence and awareness has certainly made it a con-
sideration for obstetricians. Serologic testing of the
mother and infant is the mainstay of diagnosis, but there
are reports of placental pathology associated with con-
14.5.4 Cytomegalovirus genital infections. The histologic changes, when present,
tend to be mild, and are nonspecific. The villi are
Cytomegalovirus (CMV) is the most common congenital viral enlarged, partly due to villous edema, but also due to
infection, occurring in 1–2% of newborns. Congenital CMV increased numbers of Hofbauer cells in the villous
acquired early in gestation can lead to prematurity, hydrops, stroma. Interestingly, no significant acute or chronic vil-
growth restriction, and fetal demise. Grossly, the placenta may litis has been noted. RNA probes for the virus will high-
appear hydropic. Histologic findings are most often present light the villous stromal/Hofbauer cells, indicating the
only in cases with severe clinical findings. The histologic fea- presence of Zika within them [39].
tures include lymphoplasmacytic villitis, hyalinized villi, or
avascular villi with hemosiderin-laden macrophages. Viral
inclusions, when present, can be found in villous stromal References
Hofbauer cells and endothelium. Immunohistochemical stains
for CMV may be helpful [34, 35]. 1. Norwitz ER, Schust DJ, Fished SJ. Implantation and the survival of
early pregnancy. N Engl J Med. 2001;345:1400–8.
2. Kaplan CG. Gross pathology of the placenta: weight, shape, size,
colour. J Clin Pathol. 2008;61:1285–95.
14.5.5 Herpes Simplex 3. Schneider H. Oxygenation of the placental-fetal unit in humans.
Respir Physiol Neurobiol. 2011;178:51–8.
4. Longtine MS, Nelson DM. Placental dysfunction and fetal pro-
The rare placenta infected with herpes simplex virus (HSV) gramming: the importance of placental size, shape, histopathology,
is almost always both grossly and microscopically and molecular composition. Semin Reprod Med. 2011;29:187–96.
unremarkable. Rarely, features including necrotizing lym- 5. Nagi AN. A monograph placental examination and pathology.
phoplasmacytic villitis, villous fibrosis and calcification, Biomedica. 2017;27:81–99.
6. Redline RW. Placental pathology: a systematic approach with clini-
chronic inflammation surrounding fetal vessels, or HSV cal correlations. Placenta. 2008;29:86–91.
inclusions (in villi, the cord, and membranes) are noted. 7. Salafia CM, Yampolsky M, Misra DP, Slakhter O, Haas D, Eucker
Necrotizing funisitis has also been described [36, 37]. B, Thorp J. Placental surface shape, function, and effects of mater-
nal and fetal vascular pathology. Placenta. 2010;31:958–62.
8. Garrod A, Batra G, Ptacek I, Heazell AEP. Duration and methods of
tissue storage alters placental morphology-implications for clinical
14.5.6 Syphilis and research practice. Placenta. 2013;34:1116–9.
9. Salafia CM, Yampolsky M, Shlakhter A, Mandel DH, Schwartz
While about two-thirds of infants infected with Treponema pal- N. Variety in placental shape: when does it originate? Placenta.
2012;33:164–70.
lidum are asymptomatic at birth, about one-third are stillborn. 10. Redline RW. Classification of placental lesions. Am J Obstet
There is also a high incidence of prematurity and intrauterine Gynecol. 2015;213:S21–S8.
492 J. P. B. Govindavari and A. R. Laury
11. Almog B, Sehata F, Aljabri S, Levin I, SHalom-Paz E, Shrim 26. Hubinont C, Lewi L, Bernard P, Marbaix E, Debieve F, Jauniaux
A. Placenta weight percentile curves for singleton and twins deliv- E. Anomalies of the placenta and umbilical cord in twin gestations.
eries. Placenta. 2011;32:58–62. Am J Obstet Gynecol. 2015;213:S91–S102.
12. Baergen RN. Umbilical cord pathology. Surg Pathol Clin.
27. Avagilano L, Marconi AM, Candiani M, Barbera A, Bulfamante
2013;6:61–85. G. Thrombosis of the umbilical vessels revisited. An observational
13. Benirschke K, Burton GJ, Baergen RN. Anatomy and pathology of study of 317 consecutive autopsies at a single institution. Hum
the umbilical cord. In: Pathology of the human placenta. Berlin, Pathol. 2010;41:971–9.
Heidelberg: Springer; 2012. p. 309–75. 28. Kim CJ, Romero R, Chaemsaithong P, Chaiyasit N, Yoon BH,
14. Pathak S, Hook E, Hackett G, Murdoch E, Sebire NJ, Jessop F, Kim YM. Acute chorioamnionitis and funisitis: definition, patho-
Lees C. Cord coiling, umbilical cord insertion and placental shape logic features, and clinical significance. Am J Obstet Gynecol.
in an unselected cohort delivering at term: relationship with com- 2015;213:S29–52.
mon obstetric outcomes. Placenta. 2010;31:963–8. 29. Seckl MJ, Sebire NJ, Berkowitz RS. Gestational trophoblastic dis-
15. Redline RW. Inflammatory responses in the placenta and umbilical ease. Lancet. 2010;376:717–29.
cord. Semin Fetal Neonatal Med. 2006;11:296–301. 30. Parveen Z, Tongson-Ignacio JE, Fraser CR, Killeen JL, Thompson
16. Ernst LM, Minturn L, Huang MH, Curry E, Su EJ. Gross patterns KS. Placental mesenchymal dysplasia. Arch Pathol Lab Med.
of umbilical cord coiling: correlations with placental histology and 2007;131:131–7.
stillbirth. Placenta. 2013;34:583–8. 31. Heider A. Fetal vascular malperfusion. Arch Pathol Lab Med.
17. Patil NS, Kulkarni SR, Lohitashwa R. Umbilical cord coiling index 2017;14:1484–9.
and perinatal outcome. J Clin Diagn Res. 2013;7:1675–7. 32. Hood IC, Desa DJ, Whyte RK. The inflammatory response in can-
18. Bombrys AE, Neiger R, Hawkins S, Sonek J, Croom C, McKenna didal chorioamnionitis. Hum Pathol. 1983;14:984–90.
D, Ventolini G, Habli HH, Sibai B. Pregnancy outcome in isolated 33. Schwartz DA, Reef S. Candida albicans placentitis and funisitis:
single umbilical artery. Am J Perinatol. 2008;25:239–42. early diagnosis of congenital candidemia by histopathologic exami-
19. Hua M, Odibo AO, Macones GA, Roehl KA, Crane JP, Cahill nation of umbilical cord vessels. Pediatr Infect Dis J. 1990;9:661–5.
AG. Single umbilical artery and its associated findings. Obstet 34. Chou D, Ma Y, Zhang J, McGrath C, Parry S. Cytomegalovirus
Gynecol. 2010;115:930–4. infection of trophoblast cells elicits an inflammatory response:
20. Murphy-Kaulbeck L, Dodds L, Joseph KS, Van den Hof M. Single a possible mechanism of placental dysfunction. Am J Obstet
umbilical artery risk factors and pregnancy outcomes. Obstet Gynecol. 2006;194:535–41.
Gynecol. 2010;116:843–50. 35. McDonagh S, Maidi E, Chang HT, Pereira L. Patterns of human
21. Goldfarb CA, Sathienkijkanchai A, Robin NH. Amniotic constric- cytomegalovirus infection in term placentas: a preliminary analy-
tion band: a multidisciplinary assessment of etiology and clinical sis. J Clin Virol. 2006;35:210–5.
presentation. JBJS. 2009;91:68–75. 36. Finger-Jardim F, Teixeira LO, de Oliveira GR, Barral MF, da Hora
22. Herva R, Karkinen-Jaaskelainen M. Amniotic adhesion mal-
VP, Gonalves CV, Soares MA, de Martinez AM. Herpes simplex
formation syndrome: fetal and placental pathology. Clin Ter. virus: prevalence in placental tissue and incidence in neonatal cord
1984;29:11–9. blood. J Med Virol. 2014;86:519–24.
23. Khong TY, Mooney EE, Ariel I, Balmus NC, Boyd TK, Brundler 37. Heifetz SA, Bauman M. Necrotizing funisitis and herpes simplex
MA, Derricott H, Evans MJ, Faye-Petersen OM, Gillan JE, Heazell infection of placental and decidual tissues: study of four cases.
AE. Sampling and definitions of placental lesions: Amsterdam pla- Hum Pathol. 1994;25:715–22.
cental workshop group consensus statement. Arch Pathol Lab Med. 38. Seffield JS, Sanchez PJ, Wendel GD, Fong DW, Margraf LR, Zeray
2016;140:698–713. F, McIntire DD, Rogers BB. Placental histopathology of congenital
24. Ogino S, Redline RW. Villous capillary lesions of the placenta: dis- syphilis. Obstet Gynecol. 2002;100:126–33.
tinctions between chorangioma, chorangiomatosis, and chorangio- 39. Rosenberg AZ, Yu W, Hill DA, Reyes CA, Schwartz DA. Placental
sis. Hum Pathol. 2000;31:945–54. pathology of Zika virus: viral infection of the placenta induces
25. De Paepe ME. Examination of the Twin Placenta. Semin Perinatol. villous stromal macrophage (Hofbauer cell) Proliferation and
2015;39:27–35. Hyperplasia. Arch Pathol Lab Med. 2017;141:43–8.
Placenta and Pregnancy-Related
Diseases 15
Erica Schollenberg, Anna F. Lee, Jefferson Terry,
and Mary Kinloch
© Science Press & Springer Nature Singapore Pte Ltd. 2019 493
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_15
494 E. Schollenberg et al.
Table 15.1 Indications for placental examination human tissues must be conducted by the laboratory regard-
Maternal history less of whether medical examination is indicated.
Pre-existing diabetes mellitus, vascular disease, or other The placental requisition for examination should include,
gestationally significant medical condition at a minimum, the gestational age and the indication for
Substance abuse
examination. Some hospitals use purpose-designed placental
Previous obstetric history
requisitions with fields to record relevant obstetric details
Prior recurrent pregnancy loss or late fetal/neonatal loss
Prior significant obstetric complication
and indications for examination, which facilitates provision
Maternal conditions arising in pregnancy of relevant clinical history and appropriate requests for
Peripartum fever or suspected infection examination. Placentas are most commonly accessioned as
Gestational diabetes surgical specimens under the maternal record; a special pro-
Preeclampsia or pregnancy-induced hypertension vision in the laboratory information system or reporting sys-
Maternal trauma tem ideally also makes the report available in the neonate’s
Current pregnancy chart.
Preterm labor or delivery (<37 completed weeks gestation)
Post-term delivery (≥42 completed weeks gestation)
Prolonged rupture of membranes (>24 h prior to delivery)
Excessive bleeding prepartum
15.1.2 Gross Examination and Sampling
Suspected placental abruption
Oligohydramnios or polyhydramnios Placentas may be examined in the laboratory in the fresh- or
Placenta accreta/increta/percreta or previa formalin-fixed state. The advantages of formalin fixation
Meconium-tinged amniotic fluid include better characterization of some types of discrete
Termination for anomalies or suspected gestational trophoblastic lesions, decreased infection potential, less mess from unfixed
disease blood, and better fixation of sampled blocks if processing is
Fetal hydrops done on the same day as block sampling. On the other hand,
Fetal demise in utero
even with large volumes of fixative, formalin permeates and
Neonatal
Perinatal or neonatal demise
fixes bloody tissues very slowly, mitigating some of the
Need for resuscitation or ventilation, poor Apgar (<7), low cord abovementioned advantages. Examination in the fresh state
blood pH allows for sampling for specialized molecular genetic tech-
Neonatal anemia niques that are incompatible with formalin, faster turnaround
Low birth weight (<10th percentile) times, and use of considerably less fixative. The chosen
Macrosomia (birth weight >95th percentile) examination protocol depends on the workflow and resources
Seizure of the specific lab.
Sepsis or suspected infection
A placental gross examination should include descriptors
Admission to neonatal intensive care unit
of the umbilical cord, membranes, and disc including fetal
Major congenital anomalies, including confirmed or suspected
aneuploidy surface, maternal surface, and parenchymal cut surface
Multiple gestation with discordant weight, premature delivery, and/ (Table 15.2). Use of standardized grossing templates facili-
or clinically significant question regarding chorionicity tates thorough and reproducible examinations [2]. Explicit
Placental comparison of observed disc weight with gestational age-
Small or large placenta (either absolutely or relative to neonatal adjusted expected weight is recommended [3]. It should be
size)
noted that formalin fixation affects placental measurements,
Abnormal umbilical cord insertion (marginal or velamentous) or
length (short or excessively long) including increasing disc weight by up to nearly 8% [4]. It is
Abnormal umbilical cord vessels (single umbilical artery, suspected often more useful to document macroscopic lesions photo-
thrombosis) graphically than it is to submit numerous sections of similar-
Umbilical cord true knot appearing areas.
Discrete lesions of the disc (mass, hematoma, infarction) According to recently published consensus guidelines,
Abnormal color (abnormal pallor, green-stained membranes) minimum sections recommended for microscopic examina-
Table adapted from [1] tion are roll of extraplacental membrane from the margin to
the rupture site, two cross sections of the umbilical cord from
s ubmitted for pathologic examination. If resources allow, it different areas of the cord, three full-thickness sections of the
is advisable to store unsubmitted placentas from all remain- placental disc, and representative samples of discrete lesions
ing deliveries for a period of time (1 or 2 weeks) to allow for [5]. Increasing the number of grossly normal sections sub-
placental examination if the need becomes evident postna- mitted for histologic examination will increase sensitivity;
tally. Refrigerated storage without formalin fixation pre- however, most grading schemes for common and clinically
serves tissues sufficiently. In some jurisdictions, disposal of relevant pathological entities are based on examination of a
15 Placenta and Pregnancy-Related Diseases 495
Table 15.2 Elements of a placental gross examination and recognized placental reporting protocols do not presently
description exist; however, development of local reporting protocols
Unique patient identifiers may improve clinician understanding and facilitate research.
Placenta received fresh or in formalin
Singleton or multiple gestation (see Sect. 15.12)
Umbilical cord 15.2 Placental Infection
Attached or separate; number of segments
Number of vessels
Placental inflammation and infection may sometimes be
Length and average diameter
used interchangeably, but they imply different disease states
Insertion (central, eccentric, distance to margin, marginal,
velamentous, furcate, etc.) in the placenta. Placental infections can have inflammatory
Coiling (number of total coils or coils per 10 cm) cells, but not all placentas with inflammatory cells have
General descriptors (edematous, discolored, surface lesions, etc.) infections. Most inflammatory patterns in the placenta can be
Discrete lesions (varices, true knots, vessel thrombosis, etc.) distinguished by location of the inflammatory infiltrate and
Membranes cellular composition. This section deals with common and
Insertion into disc (normal, circummarginate, circumvallate) typical patterns of placental infections, and the section fol-
Rupture site distance from margin lowing covers inflammatory patterns not associated with
General descriptors (translucent, opaque, green, brown,
infectious etiologies.
hemorrhagic)
Velamentous vessels
Disc
Weight (after removal of membranes and umbilical cord) 15.2.1 Ascending Intrauterine Infection
Dimensions
Shape (round, oval, bilobed, accessory lobe, etc.) Ascending intrauterine infection is also referred to as amni-
Fetal surface otic fluid infection sequence or syndrome. The histopatho-
General descriptors (color, etc.) logic descriptor chorioamnionitis should not be used
Discrete lesions
interchangeably with these clinicopathologic terms.
Chorionic plate vessels (distribution, thrombosis, etc.)
Identifiable microbes are typically bacterial, and this type of
Maternal surface
Completeness
placental infection has its sentinel declaration in the placental
Presence, size, and quality of hematomas membranes, which are the first point of contact with the outside
Discrete lesions world. The closed cervical os and mucous plug act as natural
Cut surface of parenchyma barriers against outside contamination of the amniotic sac
General descriptors (congested, pale, firm, mottled, etc.)
Discrete lesions (number, size, color, consistency, location, etc.) 15.2.1.1 Clinical Features
The incidence of ascending uterine infection is inversely
associated with gestational age. There is an over 90% asso-
set number of tissue sections, and increasing the sections ciation of acute chorioamnionitis with spontaneous preterm
submitted may lead to overestimation of grade. In general, delivery at 20 weeks; this association steadily decreases with
submission of extra sections of grossly normal placenta increasing gestational age to 5.1% at term [7, 8].
should be restricted to specific indications requiring increased The most common organisms are cervicovaginal bacteria
sensitivity (e.g., searching for maternal metastases). from the lower genital tract. Common organisms include
Important gross findings and abnormalities are discussed Ureaplasma urealyticum, Escherichia coli, Gardnerella vag-
in the sections below. The reader is also referred to other com- inalis, and group B Streptococcus (GBS).
prehensive resources on gross pathology of the placenta [6]. A requisition stating “query acute chorioamnionitis” is a
common indication for placental examination when there is
maternal fever or tachycardia. Other possible clinical find-
15.1.3 Standardized Reporting ings include fetal tachycardia, uterine tenderness, foul-
smelling vaginal discharge, and maternal white cell count
The value of standardized reporting for improving clinical increase. Common clinical scenarios are preterm rupture of
interpretation and research in cancer has been well estab- membranes and prolonged premature rupture of membranes.
lished, and generally accepted cancer reporting protocols are Risk factors and contributors include many etiologies such
now in widespread use. Considerable progress has been as incompetent cervix, previous infection (such as bacterial
made in defining, refining, and standardizing many placental vaginosis), cervical instrumentation such as cerclage, and
pathological processes [5], and placental pathology report- vaginal hemorrhage. The clinical severity and histologic
ing could similarly benefit from this approach. Universally findings do not always correlate.
496 E. Schollenberg et al.
15.2.1.2 Gross Features Table 15.3 Staging and grading of maternal and fetal inflammatory
Color and opacity are both important descriptors of the pla- responses in ascending intrauterine infection
cental membranes. Normal placental membranes are shiny Maternal inflammatory response
and translucent, almost transparent. The neutrophilic exu- Stage Acute subchorionitis or Grade Not severe as defined
1 chorionitis 1
date from an ascending intrauterine infection causes opacifi-
Stage Acute chorioamnionitis; Grade Severe: confluent
cation of the membranes, making them thick and dull-white 2 polymorphonuclear 2 polymorphonuclear
(Fig. 15.1). The neutrophilic exudate can also cause a green- leukocytes extend into leukocytes or
brown discoloration; however, this is not a specific finding fibrous chorion and/or subchorionic
amnion microabscesses
because meconium staining and biliverdin breakdown can
Stage Necrotizing
also cause similar discoloration. Even in cases of severe 3 chorioamnionitis:
inflammation, the gross findings may be relatively karyorrhexis of
unimpressive. polymorphonuclear
leukocytes, amniocyte
necrosis, and/or amnion
15.2.1.3 Microscopic Features basement membrane
The Amsterdam Placental Workshop Group Consensus hypereosinophilia
Statement from July 2016 provides the latest standard in the Fetal inflammatory response
diagnostic criteria for acute chorioamnionitis [5] (Table 15.3). Stage Chorionic vasculitis or Grade Not severe as defined
There is consensus that location (staging) and composition 1 umbilical phlebitis 1
Stage Involvement of the Grade Severe: near-confluent
of the inflammatory response should be documented. In
2 umbilical vein and one 2 intramural
addition, classification of the severity (grading), and separa- or more umbilical polymorphonuclear
tion of fetal from maternal inflammatory responses, is arteries leukocytes with
useful. attenuation of vascular
smooth muscle
Stage Necrotizing funisitis
Maternal Inflammatory Response 3
See Table 15.3 for the recommended staging and grading
Reproduced with permission from [5]
system.
Stage 1: Acute subchorionitis or chorionitis includes
acute inflammation in the form of neutrophils beneath the (Fig. 15.2). Stage 2: Acute chorioamnionitis refers to a neu-
chorion, with minimal extension into the cellular chorion, trophilic infiltrate within the chorion (above the cellular cho-
without the presence of neutrophils elsewhere. This stage is rion), with or without early involvement of the amnion
considered an early response to an ascending infection. (Fig. 15.3). Stage 3: Necrotizing chorioamnionitis includes
Strictly speaking, this is not a true chorioamnionitis any cases with neutrophil karyorrhexis and/or overt necrotic
damage to the amnion (Fig. 15.4).
Note that even severe/advanced chorioamnionitis can be
surprisingly patchy, with relative or complete sparing of the
chorionic plate or membranes in some fields.
a b
Fig. 15.2 Acute subchorionitis or chorionitis (maternal inflammatory response stage 1) denotes neutrophilic inflammation within subchorionic
fibrin, beneath the chorion, with (a) no or (b) minimal extension into the cellular chorion (original magnifications 200×)
a b
Fig. 15.3 Acute chorioamnionitis (maternal inflammatory response can be graded as either (a) mild or moderate (grade 1) or (b) severe
stage 2) refers to a neutrophilic infiltrate within the chorion, above the (grade 2) (original magnification 100×)
cellular chorion, with or without early involvement of the amnion, and
a b
Fig. 15.4 Low- (a) and higher-power (b) views of necrotizing chorioamnionitis (maternal inflammatory response stage 3), with neutrophil karyor-
rhexis, amniocyte necrosis, and hypereosinophilia of the amnion basement membrane (original magnifications 40× and 100×)
Like the maternal inflammatory response, the inflamma- least two umbilical cord cross sections. In general, the fetal
tion can be variable and patchy in different parts of the fetal inflammatory response lags stagewise behind the maternal
circulation. This is part of the reason for recommending at inflammatory response. The fetal inflammatory response
498 E. Schollenberg et al.
a b
Fig. 15.5 (a) Umbilical phlebitis (fetal inflammatory response stage 1) with fetal neutrophils migrating out through a chorionic plate vessel
with fetal neutrophils migrating from the fetal circulation via the umbil- (original magnifications approximately 100×)
ical vein. (b) Chorionic vasculitis (fetal inflammatory response stage 1)
a b
Fig. 15.8 (a) Neutrophil collection on the surface of an umbilical cord in a case of (b) Candida infection (original magnification 200×)
a b
Fig. 15.9 (a) Acute necrotizing villitis and acute intervillositis, caused genes (gram-positive rods) on its surface (gram stain; original magnifi-
by Listeria monocytogenes infection (original magnification 100×). (b) cation 400×)
Umbilical cord cross section showing colonies of Listeria monocyto-
Microscopic Features
See Table 15.4 for characteristic features of syphilis and
CMV infection of the placenta.
Distinction between infectious villitis of this type and the
much more commonly encountered chronic villitis of unknown
etiology (VUE) is discussed below in the section on VUE.
Fig. 15.10 Intervillous abscess in placental Listeria infection (original
magnification 20×) 15.2.2.3 Placental Parasitic Infections
Parasitic infections of the placenta are individually and col-
space. The clinical history is usually sufficient for lectively uncommon in North American practice. Two are
differentiation. worth mentioning for their specific clinical contexts and
characteristic microscopic appearances.
Prognosis Placental malaria is a significant public health concern in
Pregnancy outcome depends on the gestational age at time of many parts of the world; in North America it is encountered
infection. In the second trimester, fetal loss is common. The mainly in immigrants and recent travelers. Malarial infection
neonate remains vulnerable to Listeria sepsis and meningitis of the placenta occurs mainly within the maternal blood
in the weeks following delivery. (intervillous) space and correlates with maternal parasitemia.
The most common causative species is Plasmodium falci-
15.2.2.2 Chronic Infectious Villitis parum. The microscopic appearance is typically one of
chronic histiocytic intervillositis with hemozoin pigment as
Clinical Features well as organisms within maternal erythrocytes, although the
The incidence is related to the two etiologic infectious agents histologic appearance and parasite burden depend on the
associated with most of these cases: Treponema pallidum timing and chronicity of infection (Fig. 15.12) [10, 11].
15 Placenta and Pregnancy-Related Diseases 501
a b
Fig. 15.11 Villitis caused by cytomegalovirus (a, original magnification 200×; b, immunohistochemical stain for cytomegalovirus showing
nuclear positivity, original magnification 100×)
a b
Fig. 15.14 Chronic histiocytic intervillositis in (a) early spontaneous abortion and (b) fetal demise at term (original magnifications 100× and 200×)
15 Placenta and Pregnancy-Related Diseases 503
etiology. Evaluation of vessels is crucial to identify the pres- vascular space. The similarity of MFI and MPFD suggests
ence of associated vascular damage (similar in appearance to they represent parts of a common pathogenetic spectrum.
that seen in fetal vascular malperfusion). The underlying etiology of MFI/MPFD is unclear but is
likely multifactorial and may involve aberrant maternal
15.3.2.3 Differential Diagnosis immune responses and/or coagulation abnormalities [21].
The most important consideration in the differential diagno-
sis of VUE is not to miss a significant placental infection. 15.3.3.1 Clinical Features
See Table 15.7 for distinguishing features. MFI/MPFD typically presents with intrauterine growth
restriction (in up to 100% of cases) and premature delivery
15.3.2.4 Prognosis (up to 60%); intrauterine demise may occur in up to 50% of
High-grade chronic villitis of unknown etiology and villitis affected pregnancies [21]. Oligohydramnios and elevated
with associated vascular damage have been shown to be maternal serum alpha-fetoprotein may be associated prenatal
associated with poor outcome, in particular adverse neuro- findings. MFI/MPFD may be suspected by the combination
logical outcomes [19, 20]. of growth restriction, oligohydramnios, and echogenic pla-
centa seen on prenatal ultrasound. There may be a history of
infertility.
15.3.3 Maternal Floor Infarction and Massive
Perivillous Fibrinoid Deposition
Table 15.7 Chronic villitis of unknown etiology versus villitis of
Maternal floor infarction (MFI) and massive perivillous infectious etiology
fibrinoid deposition (MPFD) are rare entities with overlap-
Feature VUE Infectious villitis
ping phenotypes characterized by excessive deposition of
Prevalence Common Rare
perivillous fibrin and fibrinoid that obscures the maternal Gestational Most common in Any trimester
age third trimester
Table 15.6 Grading chronic villitis of unknown etiology Inflammatory Lymphocytes, Histiocytes (including
cells especially T cells, granuloma formation or
Low grade Fewer than ten contiguous villi involved per focus and histiocytes giant cells) or plasma cells
Focal Only present on one slide may predominate
Multifocal Present on multiple slides Distribution Patchy, often with Diffuse
High At least one focus of inflammation with more than ten basal predominance
grade involved villi Specific None Viral inclusions or
Patchy Multiple foci on one or more slides findings microorganisms
Diffuse Involving more than 30% of sampled distal villi Special stains N/A Viral and histochemical
Table adapted from [5] stains may be informative
a b
Fig. 15.15 In villitis of unknown etiology (VUE), lymphohistiocytic inflammation of the villi is typically patchy (a) and associated with agglu-
tination of distal villi (b) (original magnification 200×)
504 E. Schollenberg et al.
a b
Fig. 15.16 (a) In massive perivillous fibrinoid deposition, excess fibrinoid appears and makes the placenta firm with a pattern of lacelike tan
fibrinoid. (b) Low-power view showing excessive deposition of fibrinoid around villi (original magnification 10×)
15.3.3.2 Gross Features Table 15.8 Diagnostic criteria for maternal floor infarction/massive
perivillous fibrin deposition
Placentas affected by MFI/MPFD are usually small for ges-
tational age but may be normal or even large by weight Pattern Diagnostic criteria
Classic Basal fibrinoid involving entire maternal floor
depending on the extent of fibrinoid deposition. Unfixed pla-
At least one slide with fibrinoid thickness ≥3 mm
centas are unusually firm and stiff, and the maternal surface Transmural Encasement of ≥50% of villi in at least one slide
appears pale. Sectioning reveals dense, firm, tan-white Non-basal (focal) fibrinoid distribution pattern
parenchyma in a distribution varying from conspicuous ori- Borderline Encasement of ≥25% to <50% of villi in at least
entation along most or all of the maternal surface (the MFI one slide
pattern) to more discrete foci extending from the maternal Non-basal (focal) fibrinoid distribution pattern
surface to the chorionic plate and involving 50% or more of Table adapted from [21]
the placental disc volume (the MPFD pattern) (Fig. 15.16a).
but fibrinoid is not abundant, and intervillous histiocytic
15.3.3.3 Microscopic Features infiltrates are present. The gross and microscopic appearance
Despite differences in distribution, MFI/MPFD both involve of MFI/MPFD can resemble primary infarction; however,
extensive perivillous deposition of fibrin and fibrinoid mate- the villi in MFI/MPFD will be separated by abundant fibrin/
rial that fills the intervillous spaces (Fig. 15.16b). Extravillous fibrinoid. VUE and fetal vascular malperfusion may produce
trophoblast within the fibrinoid material may be prominent. similar regressive changes in affected villi, but complete vil-
Encasement of villi produces regressive changes including lus encasement by fibrinoid will not be present.
loss of villous trophoblast, vessels, and stromal cells.
Inflammatory cells, including lymphocytes and neutrophils, 15.3.3.5 Prognosis
may be seen in the vicinity of degenerating villi. Mass effect Although rare, MFI/MPFD is an important entity to recog-
disrupts local maternal blood flow leading to the features of nize because it is associated with poor outcomes and has a
maternal vascular malperfusion and development of intervil- significant risk of recurrence in future pregnancies. In addi-
lous thrombi in adjacent parenchyma. Semiquantitative diag- tion to those described above, MFI/MPFD is also a risk fac-
nostic criteria for MFI/MPFD based on gross and microscopic tor for neonatal death and neurological impairment [22].
findings have been proposed (Table 15.8).
a b
Fig. 15.18 (a) Meconium appears as tan, poorly refractile pigment tive changes (original magnification 400×). (b) Meconium extending to
within the chorioamniotic tissues and histiocytes. Amniocyte cytoplas- the umbilical vessels may induce smooth muscle vasospasm and necro-
mic vacuolization and edema, seen here, are commonly associated reac- sis (original magnification 100×)
506 E. Schollenberg et al.
The caustic nature of meconium may also induce vaso- 15.5.1 Umbilical Cord Length and Diameter
spasm and vascular medial necrosis in umbilical and chori-
onic plate vessels, which may have serious consequences for Umbilical cord length increases with gestational age with a
the fetus (Fig. 15.18b). In this situation, meconium and vacu- normal range at 38 weeks gestation of approximately
olated meconium-laden macrophages are seen in and around 35–70 cm [23, 24]. Unusually long umbilical cords pose an
the media of umbilical and chorionic plate vessels. increased risk of fetal cord entanglement, true knot forma-
tion, and cord prolapse, all of which may obstruct umbilical
blood flow with potentially disastrous consequences
15.4.4 Differential Diagnosis (Fig. 15.19). Short umbilical cords, especially when less
than 10 cm, are associated with limb-body wall complex and
Meconium can be differentiated from hemosiderin by the fetal developmental abnormalities. Accurate assessment of
presence of the above-described reactive changes and nega- umbilical cord length is frequently limited by submission of
tive iron special staining. Lipofuscin, which is not commonly incomplete cords for assessment.
encountered in the placenta, resembles meconium by light Umbilical cord diameter also increases with gestational
microscopy and cannot be definitively differentiated by spe- age with a normal range of approximately 0.8–1.2 cm at
cial staining. 38 weeks gestation [25]. Ground substance (Wharton’s jelly)
Meconium may be artifactually displaced in the potential surrounding the umbilical vessels provides a protective cush-
space between the amnion and chorion after delivery. This ion that prevents vascular occlusion (Fig. 15.20a). Narrow
can be differentiated from true meconium staining by its dis- umbilical cords have reduced ground substance and are at
tribution, which may involve but is not restricted to the increased risk of vascular compression and fetal vascular
amniotic-chorionic cleft. malperfusion (FVM) (Fig. 15.20b). Increased umbilical cord
diameter is usually the result of edema (Fig. 15.21). When
umbilical cord edema is marked, it may appear as cystic
15.4.5 Prognosis spaces on prenatal imaging; however, these are almost
always pseudocystic spaces. True umbilical cord cysts,
Most infants born with pre- or perinatal meconium exposure which most commonly arise from the omphalomesenteric
have no clinical sequelae. Although meconium can be asso- duct remnant, are rare.
ciated with poor neonatal outcomes, it is not an independent
predictor, and prognostically relevant placental abnormali-
ties associated with meconium release should be sought.
Prolonged intrauterine exposure to thick meconium can
rarely result in meconium aspiration syndrome, a form of
neonatal respiratory distress caused by massive meconium
aspiration. Meconium aspiration syndrome cannot be reli-
ably predicted by placental examination. Meconium-
associated vascular necrosis is a significant risk factor for
central nervous system damage and neonatal death [20].
a b
Fig. 15.20 (a) Connective tissue and ground substance surrounds and stents umbilical vessels. (b) Narrow umbilical cords (0.4 cm at 38 weeks,
in this case) lack protective ground substance (original magnifications 100×)
15.5.4 Umbilical Cord Insertion Fetal vascular malperfusion (FVM) is a diagnostic term that
has been recommended to replace fetal thrombotic vascu-
The site of umbilical cord insertion into the placental disc is lopathy and fetal vascular flow restriction to describe abnor-
generally thought to initiate centrally and then “migrate” as malities in fetal perfusion of the placenta [5, 33]. FVM is
15 Placenta and Pregnancy-Related Diseases 509
a b
Fig. 15.24 (a) Marginal, (b) velamentous, and (c) furcate umbilical cord insertions are risk factors for mechanical obstruction of umbilical blood
flow
defined by pathologic features of thrombosis or ischemia in ties (see previous section) and/or presence of oligohydram-
the fetal vasculature of the placenta. nios [34]. Compromised fetal cardiac output may be primary,
as in structural cardiac defects, or secondary, such as in high-
output failure secondary to fetal anemia.
15.6.1 Clinical Features
Most causes of FVM fall within two groups: obstruction of 15.6.2 Gross Features
umbilical cord blood flow and compromised fetal cardiac
output. Of these, umbilical cord flow restriction is more Thrombosis of fetal vessels may be partially or completely
common and is usually related to umbilical cord abnormali- occlusive and in larger vessels may be seen grossly. Clusters
510 E. Schollenberg et al.
a b
Fig. 15.25 Redundancies, outpouchings, or varices of umbilical vessels are sometimes referred to as false knots. They have no apparent func-
tional significance, whether simple (a) or complex (b)
a b
Fig. 15.26 Placentas from intrauterine fetal demises at (a) 36 weeks and (b) 20 weeks of gestation. Note the true knots in the umbilical cords with
differential congestion of vessels on either side
of avascular distal villi may also be seen on gross examina- wall in the chorionic plate or stem villus. Histologic features
tion as an area of pale parenchyma. Placentas with extensive of early true thrombosis include mural fibrin deposition
FVM may be small for gestational age. (Fig. 15.27a) with loss of appreciable endothelium between
the thrombus and vessel wall (Fig. 15.27b). Over time, nonoc-
clusive thrombosis results in restoration of the vascular lumen
15.6.3 Microscopic Features and fibrin incorporated into the vessel wall (Fig. 15.28a),
while occlusive thrombosis causes luminal obliteration with or
Fetal vascular thrombosis is more likely to be recognized on without recanalization (Fig. 15.28b). Mural fibrin eventually
microscopic examination as fibrin layered against a vessel calcifies, which is a sign of remote thrombosis (Fig. 15.28c).
15 Placenta and Pregnancy-Related Diseases 511
a b
Fig. 15.27 Fetal vascular thrombosis and post-delivery or post-demise shows loss of intervening endothelium and extension of fibrin into the
intravascular coagulation can both present as mural fibrin collections vascular intima (b, original magnification 200×)
(a, original magnification 20×), but true fetal vascular thrombosis
Downstream stream effects of FVM present initially as setting of fetal demise if temporally incongruous features of
villous stromal hemorrhage and karyorrhexis (Fig. 15.29a), FVM are present (e.g., focal intramural fibrin, calcification),
which represent degenerative changes in villous capillaries. but making such a differentiation should be approached con-
With time the villi become avascular (Fig. 15.29b). servatively. See the section on late fetal demise, below, for an
Hemosiderin resulting from villous stromal hemorrhage may approach to this differential diagnosis.
be present in avascular villi. Foci of villous stromal hemor-
rhage, karyorrhexis, and avascular villi are usually sharply
demarcated from adjacent unaffected villi (Fig. 15.30). 15.6.5 Prognosis
Intervillous thrombi can be seen in the setting of FVM
where they arise from mixing of maternal and fetal blood The clinical significance of FVM is increased risk of still-
related to stromal villous hemorrhage extending into the birth, fetal growth restriction, and development of neurologi-
maternal blood space, but are not specific for FVM. A rare cal abnormalities [34–36]. Grading schemes have been
finding in FVM is erythropoietic islands arising in chorionic developed in an attempt to quantify FVM for prognosis. The
plate and stem villous stromal tissue adjacent to areas of fetal most recent grading scheme is summarized in Table 15.9.
vascular thrombosis. Aside from low and high grades, the findings can be described
as either segmental (occurring in the distribution of a chori-
onic or stem villous vessel) or global (diffuse changes
15.6.4 Differential Diagnosis attributable to obstruction of flow to or from an umbilical
vessel).
Mimics of FVM include amniotic fluid infection
inflammation- related mural thrombi, but these are easily
recognized by the intermixed acute inflammatory infiltrate 15.7 Maternal Vascular Malperfusion
(Fig. 15.31). Avascular distal villi may result from chronic
villitis, which can be indistinguishable from avascular villi In the normally developing placenta, uterine artery remodel-
resulting from FVM. Differentiation of chronic villitis from ing begins in the late first trimester, and placental perfusion
FVM as the cause of avascular distal villi is based on the lack commences around 12 weeks gestation. Arteries lose their
of an apparent vascular distribution and the presence of muscular media and become thin-walled. The uterine spiral
inflamed adjacent villi, if present. arteries therefore become a low-pressure arterial supply to
Intrauterine fetal demise leads to fetal vascular changes the placental disc that is resistant to systemic vasoconstric-
similar to FVM. Pre-demise FVM can be recognized in the tive signals. The term maternal vascular malperfusion refers
512 E. Schollenberg et al.
a b
Fig. 15.28 (a) The intravascular component of nonocclusive fetal vas- thrombi lead to obliteration of the vascular lumen (original magnifica-
cular thrombosis disappears with time, but fibrin incorporated into the tion 100×). (c) Calcification of intramural fibrin is a marker of remote
vascular wall remains (original magnification 200×). (b) Occlusive fetal vascular thrombosis (c, original magnification 100×)
to both pathologies of the uterine (decidual) vessels and the with pathology of the decidual vessels. Hypertensive
downstream effects on placental development [5]. disorders affect approximately 5–10% of pregnancies
and are relatively more common in young and older
mothers.
15.7.1 Clinical Features The clinical presentation may reflect acute or chronic fail-
ure of adequate provision of oxygen and nutrients to the
Pre-existing hypertension, pregnancy-induced hyperten- fetus, such as intrauterine growth restriction, fetal distress, or
sion, acute hypertension/preeclampsia, and related disor- even fetal demise in utero. Placental abruption is associated
ders are the maternal clinical features typically associated with hypertensive disorders of pregnancy.
15 Placenta and Pregnancy-Related Diseases 513
a b
Fig. 15.29 (a) Upstream restriction of fetal blood flow results in mal karyorrhexis) (original magnification 200×). (b) Over time these
degeneration of downstream distal villi vasculature. Initially this degenerative debris are cleared leaving avascular villi (original magni-
appears as villous hemorrhage and endothelial apoptosis (villous stro- fication 200×)
15.7.3.2 Decidual Vasculopathy The presence of residual smooth muscle in basal decidual
Decidual vasculopathy or arteriopathy includes a variety of vessels near term, with or without persistence of invading
abnormalities ranging from incomplete adaptation of vessels trophoblast, indicates incomplete adaptation for pregnancy.
to more severe pathologies. Pathology of the decidual arter- Decidual mural hypertrophy describes hypertrophic muscu-
ies correlates imperfectly with the presence of hypertensive lar layers in either membranous or basal decidual arteries and
disorders of pregnancy and fetal or placental growth is pathologic. A muscular layer with thickness greater than one-
restriction. third of the vessel diameter is a useful criterion (Fig. 15.33).
Acute atherosis refers to subendothelial foamy macrophages
of an arterial wall in any part of the decidua and may be accom-
panied by acute fibrinoid necrosis (Fig. 15.34). Unremodeled
Table 15.9 Grading of fetal vascular malperfusion (FVM)
arteries provide the substrate for the development of this acute
Low-grade One chorionic plate or stem villus thrombus lesion. Luminal thrombosis may be present.
FVM and/or
<45 avascular distal villi Decidual vasculopathy is often accompanied by perivas-
High-grade Two or more chorionic plate or stem villus cular inflammation with lymphocytes, although this finding
FVM thrombi on its own is not diagnostic.
and/or
≥45 avascular distal villi, in at least two foci
15.7.3.3 Histologic Sequelae
Adapted from [5] In addition to the vascular changes described above, the fol-
lowing accompanying histologic changes may be seen:
accelerated villous maturation or distal villous hypoplasia,
villous infarcts, and basal or retroplacental hematomas.
15.7.4 Prognosis
a b
Fig. 15.34 (a) Acute atherosis of decidual vessels, with fibrinoid necrosis of the walls and (b) foamy macrophages (original magnifications 200×)
There is a risk of recurrence in subsequent pregnancies, become paler, firmer, and more homogeneous (Fig. 15.35).
especially if preeclampsia was severe or had preterm onset. This tinctorial and textural difference is more evident in the
fixed than in the fresh state. They usually involve the basal
portion of the disc with variable extension toward the fetal
15.8 Placental Ischemia and Infarction surface.
The size and/or quantity of infarcted parenchyma should be
Villous ischemia results from inadequacy or interruption of estimated in the gross description, either in absolute terms or
maternal blood supply via the uterine spiral arteries. This is as an approximate percentage of placental disc volume. The
often due to the decidual vascular pathology described in the location should also be noted, as small marginal infarcts are
section above on maternal vascular malperfusion, but can not uncommon at term. It is not necessary to exhaustively
also be caused by maternal conditions not necessarily affect- sample infarcted parenchyma if the macroscopic appearance
ing the arteries themselves. Premature separation of the pla- is typical; if multiple lesions appear grossly similar, one repre-
cental disc from the uterine arterial supply (abruption) also sentative block often suffices. Sampling of surrounding appar-
causes ischemia of the supplied portion of the disc. ently unaffected parenchyma is often more informative.
In general, central infarcts are abnormal since this should
be an area of optimized maternal blood supply. Additionally,
15.8.1 Clinical Features large infarcts (>1–2 cm), multiple infarcts, and any infarcts
in the preterm or undersized placenta indicate inadequacy of
The underlying cause may be a primary condition affecting utero-placental perfusion.
maternal vascular health, such as pre-existing hypertension, The placenta may have other gross features of maternal
diabetic vasculopathy, smoking, cocaine use, thrombophilia, vascular malperfusion, most notably hypoplasia and/or gross
or hemoglobinopathy. Maternal hypotension or shock can evidence of abruption.
also cause ischemia of the placenta. The cause may be pri-
marily a gestational one, as in decidual vasculopathy, or ana-
tomically abnormal placental implantation (e.g., in the lower 15.8.3 Microscopic Features
uterine segment or overlying a submucosal leiomyoma).
Because the normal term placenta has significant excess Histologically, early villous infarcts appear as crowding or
capacity, small infarcts near or at term, especially near the collapse of the villous architecture with loss of the interven-
disc margin, rarely cause problems. ing intervillous maternal blood space (Fig. 15.36). Infiltration
by neutrophils may be a prominent, though transient, feature
(Fig. 15.37). As the infarct matures, there is progressive loss
15.8.2 Gross Features of nuclear staining in the villous stromal cells. The syncytio-
trophoblastic nuclei usually first appear smudged and even-
The macroscopic appearance of an infarct, as in other organs, tually also lose their nuclear staining (Fig. 15.38). A remote
depends on the age of the lesion. Acute infarcts may appear infarct is a good example of coagulative or white infarction
darker than surrounding parenchyma, whereas older infarcts (despite the dual blood supply of the placenta), in which
516 E. Schollenberg et al.
a b
Fig. 15.35 Placental disc cross sections showing multiple infarcts at varying stages, from (a) recent (dark red and hemorrhagic) to (b) remote
(pale and firm)
a b
Fig. 15.36 (a) At low power, early villous infarcts appear as localized increased syncytial knots and nuclear smudginess are apparent (original
areas of villous condensation and congestion with collapse of the inter- magnification 200×)
villous space (original magnification 10×). (b) At higher power,
a b
Fig. 15.37 (a, b) Infarcts may be transiently surrounded by a rim of infiltrating neutrophils which quickly undergo karyorrhexis (original magni-
fications 10× and 200×)
15 Placenta and Pregnancy-Related Diseases 517
a b
Fig. 15.38 (a) Remote infarcts appear as well-defined areas of agglutinated pale villi (original magnification 20×). (b) At high power, villous
outlines are maintained but nuclei are faded (original magnification 200×)
a b
Fig. 15.39 Intervillous thrombi are common discrete parenchymal lesions that can mimic infarcts. On gross (a) and microscopic (b) examination,
however, they can be seen to not involve villi (except peripherally) and have characteristic lines of Zahn (b, original magnification 20×)
nuclear detail is lost but architectural outlines of villi remain 15.8.4 Differential Diagnosis
(so-called ghost villi).
Aside from frank infarction, the villi in an affected placenta Another relatively common discrete gross parenchymal
may elsewhere show changes of acute and/or chronic ischemia lesion is the intervillous thrombus (IVT). These are local-
on the basis of inadequate perfusion of the placenta. In severe ized areas of intervillous hemorrhage, composed most
cases of chronic underperfusion, distal villous hypoplasia may often of both maternal and fetal blood. They can be distin-
be evident (described in next section). Other changes attrib- guished macroscopically from infarcts in that they are often
uted to ischemia include increased syncytial knots and/or polygonal instead of round, are not typically basally
prominent protrusion of syncytial knots on villous surfaces located, and are composed of laminated layers. On micro-
(Tenney-Parker change). Villous agglutination and perivillous scopic examination they can be seen to be composed of
fibrin deposition are non-specific features of villous injury that alternating layers of red cells and fibrin with platelets (lines
may precede histologic evidence of infarction. of Zahn). Ischemic or infarcted villi may be present at the
Examination of sampled decidual vessels may or may not periphery, but there are no ghost villi within the center of
disclose evidence of decidual vasculopathy. the lesion (Fig. 15.39).
518 E. Schollenberg et al.
a b
Fig. 15.40 Chorionic villi in the late first trimester are relatively edematous-appearing stroma. Developing fetal vessels contain nucle-
homogeneous in size, shape, and appearance. There are distinct layers ated red blood cells (a, original magnification 100×, b, original magni-
of synctiotrophoblast and cytotrophoblast lining the villi, which have fication 400×)
A reasonably accurate clinical estimate of gestational age 15.9.1.2 Villi in the Second Trimester
must be provided to the pathologist evaluating a specimen. In Over the course of the second trimester, the villous stroma
order to objectively assess disorders of villous maturation, becomes less edematous and more collagenized, with pro-
one must be familiar with the range of villous shapes, sizes, gressively more vascularization (Fig. 15.41). As the villous
and vascularity in the normally developing placenta. tree branches, there is more variability in villous size.
a b
Fig. 15.41 Villi in the second trimester undergo further branching, resulting in greater variability of size. Villi become more collagenized and
more vascularized (a, original magnification 100×, b, original magnification 400×)
a b
Fig. 15.42 These term villi are mostly small distal villi with abundant capillaries and vasculosyncytial membranes. Larger villi are stem villi
containing conducting vessels (a, original magnification 100×, b, original magnification 400×)
the thinned syncytiotrophoblast directly (the site of maternal- be assessed immediately adjacent to a discrete pathologic
fetal gas and nutrient exchange). The larger villi are stem villi lesion.
containing conducting venous and arterial branches with no It is important to remember that the changing appearance
vasculosyncytial membranes; these often become lined by a of various villous types exists on a spectrum; even occasional
layer of fibrin rather than syncytiotrophoblast. immature villi may be encountered at term. One should be
familiar with the expected villous populations seen across a
15.9.1.4 Assessment of Villous Maturation range of gestational ages, in order to recognize aberrations of
Villous maturity cannot be assessed in a single field, but development or maturation.
instead is a gestalt based on multiple complete disc cross The most helpful way to assess villous development is by
sections. Appearance of the villi at the margin and immedi- comparison with whole-slide normal age-matched controls.
ately under the chorionic plate may give a false impression Single textbook figures usually do not suffice to show the full
of accelerated maturation. Likewise, maturation should not range of villous morphologies and sizes at any gestational age.
520 E. Schollenberg et al.
usually be more informative than placental examination in usually characterized grossly by large placental size, large-
isolation. caliber thick-walled and tortuous chorionic plate vessels, and
hydropic villi (Fig. 15.49). Some of the microscopic features
15.9.5.2 Gross Features (edema and cisterns) resemble those of molar pregnancy, but
Placentas may be either very small for gestational age (which rather than trophoblastic hyperplasia, abnormal vascular
is typical for trisomy 18 and digynic triploidy), small-to- development predominates, including thick-walled stem
appropriate for gestational age (trisomy 21), or large for ges- vessels as well as chorangiomatous capillary proliferations
tational age (monosomy X with hydrops, molar pregnancy). (Fig. 15.50). This type of placental maldevelopment is clas-
sically associated with Beckwith-Wiedemann syndrome,
15.9.5.3 Microscopic Features although other genetic, imprinting, and chromosomal etiolo-
Villi may appear completely normal. Not uncommonly, vil- gies have been described [48].
lous development is diffusely but non-specifically altered. In
general, clues to aneuploidy include accelerated or delayed
villous maturation, variable villous morphology in different
areas (“villous dysmaturity”), villous edema, villous stromal
hypercellularity, complex villous outlines, predominance of
intermediate villi, trophoblast (pseudo)inclusions, and stip-
pling of trophoblast basement membranes (Figs. 15.47 and
15.48). None of these features have individually been shown
to predict aneuploidy (and all are occasionally seen in dip-
loid gestations), but in the correct clinical setting, the gestalt
may be useful in suggesting chromosomal testing to explain
a pregnancy loss or anomalous fetus.
Placental mesenchymal dysplasia is a specific placental phe- Fig. 15.47 Aneuploid placenta (trisomy 18) at mid-gestation, with
notype sometimes referred to as “pseudo-mole” [47]. It is immature convoluted villi (original magnification 20×)
a b
Fig. 15.48 Non-specific features that can be seen in aneuploid placentas are (a) trophoblast pseudo-inclusions and (b) stippling and edema
(manifested as clefting beneath the trophoblast basement membrane) (original magnifications 100×)
15 Placenta and Pregnancy-Related Diseases 523
a b
Fig. 15.49 Two cases of placental mesenchymal dysplasia with abnormal vessels and cyst-like masses, at (a) 32 weeks and (b) 35 weeks. Both
neonates had congenital anomalies, but neither had genetically confirmed Beckwith-Wiedemann syndrome
15.9.7.4 Prognosis
Fetuses of diabetic mothers are at increased risk of macroso-
Fig. 15.50 Placental mesenchymal dysplasia. The cystic lesions seen mia, congenital anomalies, neonatal respiratory distress,
grossly correspond to edematous stem villi (original magnification 20×) neonatal hypoglycemia, early spontaneous abortion, as well
as late (term or post-dates) stillbirth. A long-term implication
for the maternal patient diagnosed with gestational diabetes
15.9.7 Effects of Maternal Diabetes is an estimated 50% risk of developing type 2 diabetes within
on Placental Development 20 years [52].
a b
Fig. 15.51 Acute placental abruption is associated with retroplacental hematoma formation that deforms the maternal surface (a, formalin-fixed,
b, fresh)
Fig. 15.53 Extensive chronic abruption leads to grossly visible heme Fig. 15.54 Early ischemic changes associated with subacute abruption
breakdown products that stain the placental membranes green-brown include loss of nuclear chromatin detail and cytoplasmic eosinophilia in
(diffuse chorioamniotic hemosiderosis). Marginal thrombus is a typical affected syncytiotrophoblast (original magnification 200×)
concurrent finding and is present in this example
a b
Fig. 15.55 Characteristic microscopic findings in chronic abruption include hemosiderin in the (a) membranous tissues and (b) chorionic plate
(Perls’ Prussian blue stain; both original magnifications 100×)
retromembranous degenerating blood, retromembranous bleeding into this potential space (Fig. 15.55b). Small areas
thrombus, and hemosiderin-laden macrophages and hemo- of remote infarction related to areas of chronic abruption
siderin deposits (Fig. 15.55a). Hemosiderin deposits in the may be present, particularly at the disc margin (Fig. 15.56).
chorionic plate are also seen in chronic abruption and may Chronic abruption-oligohydramnios sequence (CAOS)
conform to the juncture of the amnion and chorion, implying describes chronic abruption associated with oligohydram-
526 E. Schollenberg et al.
Fig. 15.57 Placenta increta in the lower uterine segment and cervix in
a hysterectomy specimen
Fig. 15.58 Placenta accreta with chorionic villi implanted immedi-
ately above a layer of extravillous trophoblast and myometrium, with-
mal implantation. Rarely grossly evident adherent myome- out intervening decidua (original magnification 40×)
trium is evident.
In more severe cases, hysterectomy may have been referred to this finding as occult placenta accreta or basal
required either after placental delivery due to persistent hem- plate myometrial fibers (Fig. 15.59); there is a plausible
orrhage, or the hysterectomy specimen may be received with connection and case-control evidence that this finding is
the placenta still attached (Fig. 15.57). In this case, it is help- associated with placenta accreta in subsequent pregnancies
ful to approach the examination and sampling similarly to [61, 62].
that used for invasive endometrial cancers; that is, the In placenta increta, gross myometrial thinning will also
following should be documented: anatomic location of
be evident microscopically, and the absence of decidua can
implantation (fundal, anterior/posterior, lower uterine seg- be confirmed. Note that the myometrium of the lower uterine
ment, cervix), relationship to parametrial and paracervical segment is physiologically very thin in late pregnancy (com-
resection margins, depth of invasion (on cross section), parison with a section of uterine wall not underlying the pla-
thickness of uninvolved myometrium, and involvement of cental disc provides a control). The microscopic picture of
serosal surface. Photographs are helpful. A standardized pro- placenta percreta may be less clear as frank serosal perfora-
tocol has been published [60]. Principles of standard placen- tion will often be accompanied by hemorrhage and/or dis-
tal examination also apply. ruption due to surgical dissection off an adjacent structure
(Fig. 15.60).
15.11.1.3 Microscopic Features
In the case of a delivered placenta not accompanied by a
uterus, microscopic examination is probably insensitive for 15.11.1.4 Prognosis
confirming a clinical suspicion of placenta accreta. This is in Abnormal placental implantation is a risk factor for preterm
part due to the fact that the diagnostic features may be hidden delivery, perinatal mortality [58], indicated Cesarean sec-
in retained products. Furthermore, considering the very lim- tion, postpartum hemorrhage, delayed placental delivery,
ited sampling typical for placental pathology (three blocks hysterectomy, and even maternal mortality. The recurrence
from a 400 to 500 g organ), it is not unexpected that diagnos- risk in subsequent pregnancies for women who do not
tic microscopic fields may be missed simply due to extremely undergo hysterectomy is high.
limited sampling of the maternal surface.
When myometrium is present on the basal surface in a
microscopic section, there is most often an intervening layer 15.11.2 Placenta Previa and Vasa Previa
of basal fibrin and extravillous trophoblast underlying villi,
as opposed to villi lying immediately upon myometrial cells Placenta and vasa previa describe placental implantation
(Fig. 15.58). Immunohistochemical stains can be used to dif- over the inner cervical os. In placenta previa, the placental
ferentiate decidua and trophoblast if necessary: decidual disc covers the os, while in vasa previa, the disc regresses
cells are CD10 positive (cytoplasmic/membranous) and leaving placental membrane with velamentous umbilical
keratin negative (cytoplasmic), whereas extravillous tropho- vessels covering the os (Fig. 15.61). Placenta and vasa previa
blast stains oppositely. predispose to abruption; features of chronic abruption are
The significance of incidentally seeing the histologic typically found. Occasionally, large hematomas with evi-
picture of placenta accreta in a patient without clinical evi- dence of recurrent bleeding may be present. Associated disc
dence of abnormal implantation is debated. Some have infarction and atrophic regression may also be seen. Since
528 E. Schollenberg et al.
a b
Fig. 15.59 Myometrial cells can be seen adherent to the basal surface and magnification with desmin immunohistochemical stain, c, original
of the placenta, with only rare decidual cells intervening between them magnification 400×)
and the basal trophoblast (a, original magnification 100×, b, same field
Fig. 15.61 A case of delivered placenta previa with vasa previa. The
Fig. 15.60 The right side of the picture demonstrates increta, with section of velamentous vessels with deficient parenchyma overlaid the
thinning of the normal uterine wall thickness, and the left side of the cervical os
picture shows percreta, with chorionic villi invading into extrauterine
space (original magnification 20×)
and higher-order multiples are around 300 times less fre- Table 15.10 Specific details to include in gross description of a mul-
quent than twins [63, 64]. tiple gestation placenta
Risk of multiple gestation is increased in some ethnic Whether cords have been matched to specific neonates (e.g., “cord
clamp on twin B”)
groups, with advancing maternal age, with higher parity, and
Chorionicity (monochorionic, dichorionic, trichorionic, etc.)
with a personal or family history of prior twins or multiples.
Amnionicity: within each chorionic sac, how many amniotic sacs
An increasingly important contributor to multifetal preg- are present (monoamniotic, diamniotic)?
nancy rates is the use of assisted reproductive technologies Description of dividing membrane (opaque versus translucent)
[65]. Many pregnancies that begin as multiple gestations Number of discrete placental discs? If multiple, separate, or fused?
have asymptomatic occult spontaneous abortion of one twin Separate discs: weigh each separately
prior to the second trimester (so-called vanished twin). Single or fused disc: weigh together
(To be compared against multiple gestation-specific normal values)
Intercordal distance (distance between cord insertions)
For monochorionic placentas, placental share (percentage of the
15.12.2 Gross Features fetal surface covered by each twin’s chorionic plate vessels)
The integrity of intact anastomoses can be tested by twin. This allows for identification of anastomoses involving
injecting air using a syringe into the fetal surface vessel from vessels too thin to inject with air. For detailed information on
one twin (near the equator), then gently massaging the air this technique, the reader is directed to the technical paper
bubble into the parenchyma, and observing the bubble [74]. Air and dye injection maneuvers require a relatively
appearing in an adjacent fetal surface vessel from the other intact placenta. Fragmented placentas cannot be successfully
twin. A highly sensitive technique for identifying vascular injected. A case in which one or more fetuses died in utero
anastomoses during gross pathology examination has been may be challenging to inject.
developed by a group in the Netherlands [72–74]. This tech- Identifying inter-twin vascular anastomoses can provide
nique involves injection of different-colored pathology dyes an explanation for in utero hemodynamic variations/abnor-
into one artery and one vein of the umbilical stumps of each malities in the twins. Another purpose of identifying anasto-
moses is to evaluate the quality and success of laser ablation.
The gross appearance of ablation sites depends on the time
elapsed between ablation and delivery. Initially they may not
be visible, but as the ablated parenchyma ages, the sites
become firm, fibrotic, and yellow, similar to subchorionic
infarction or fibrin deposition (Fig. 15.65a).
Symptomatic TTTS may be associated with congested
parenchyma on the recipient twin’s side and pale paren-
chyma on the donor twin’s side (Fig. 15.65b). The twin pla-
centa associated with TTTS could include demise-related
changes if one or both twins undergo demise (Fig. 15.66).
a b
Fig. 15.65 (a) Monochorionic, diamniotic twin placenta delivered at cular equator by visual inspection or by air injection technique. Some
36 weeks of gestation. Selective fetoscopic laser ablation was per- firm yellow patches are visible where the laser was applied. These
formed at 23 weeks of gestation. Both twins were live-born. Gross resemble infarcts microscopically. (b) Same placenta, demonstrating
examination shows no residual intact vascular anastomoses at the vas- maternal surface with differential congestion
532 E. Schollenberg et al.
15.12.5.5 Prognosis
Outcome depends on severity (including Quintero stage) of
the hemodynamic imbalance. In general, the earlier in preg-
nancy the condition is detected, the more severe its potential
Fig. 15.66 Monochorionic, diamniotic twin placenta from a preg- consequences. There is high mortality particularly for TTTS
nancy complicated by twin-to-twin transfusion syndrome. Selective presenting in the second trimester, if left untreated, with
fetoscopic laser ablation was attempted at 16 weeks, but pregnancy was
excess mortality for the donor rather than the recipient twin.
complicated by extreme preterm labor and delivery at 22 weeks result-
ing in demise of both twins. Ablation sites at the vascular equator are
visible as subchorionic hemorrhagic patches. No residual vascular
anastomoses could be demonstrated by air injection technique or by 15.13 Fetal Hydrops
visual inspection
fetal normoblastemia due to chronic hypoxia and/or fetal 15.13.3 Microscopic Features
anemia. The histology of ablation sites is non-specific and is
related to tissue damage and repair [75]. The histology In a hydropic placenta, there is diffuse or focal chorionic vil-
depends on the interval between laser ablation and delivery. lous edema. The affected chorionic villi are expanded by fluid
There may initially be a thrombohematoma that becomes a and enlarged. Edematous villi are especially prominent in
remote organized thrombus. Infarcted villi, avascular villi, third-trimester placentas, when villi are normally expected to
and subchorionic fibrin deposition are also seen (Fig. 15.67). differentiate into small terminal villi with reduced stroma and
Coexisting findings may include chorioamnionitis if there is numerous close-packed profiles of capillary vessels.
premature rupture of membranes secondary to the ablation Edematous villi show capillaries that appear small in caliber
15 Placenta and Pregnancy-Related Diseases 533
a b
Fig. 15.69 Changes following intrauterine fetal demise include involution of large fetal vessels, (a) stem vessels, and (b) distal capillaries (origi-
nal magnifications 100×)
a b
Fig. 15.70 Chorangiomas typically appear as (a) round or lobulated red-tan masses, (b) microscopically comprised of lobules of capillaries
within chorionic tissue histologically similar to adjacent villi (original magnification 100×)
Fig. 15.71 Primary placental choriocarcinoma with cytotrophoblastic Intrahepatic cholestasis of pregnancy (ICP) typically pres-
and syncytiotrophoblastic components (original magnification 40×) ents in the third trimester with maternal pruritus, liver dys-
function, and increased serum bile acids [97]. The
15.15.3 Placental Metastasis pathogenesis of ICP is unclear but appears to be complex
with pregnancy hormones and genetic susceptibility playing
Metastases of maternal malignancies to the placenta are central roles. ICP is characterized by an increased risk of
exceedingly rare. Melanoma is most common; carcinomas perinatal complications, particularly preterm birth and still-
(e.g., breast) occur less frequently. Placental tumor deposits birth, which is related to the level of maternal serum bile
may metastasize further to the fetus with disastrous conse- acids. The etiology of fetal distress has not been defined but
quences. Fetal metastases to the placenta are also rare and com- likely includes the direct toxic effects of increased bile acids
prised of tumors that may present congenitally such as on the fetal heart [98]. Meconium staining of the fetus and
neuroblastoma and leukemia. These typically appear histologi- placenta is a common finding in ICP. Specific placental fea-
cally as limited to circulating tumor cells in the fetal vascula- tures of ICP have not been identified, although ICP may be
ture. Metastasis from fetus to mother has not been reported. associated with MVM-like changes [99, 100]. Interestingly,
15 Placenta and Pregnancy-Related Diseases 537
treatment of ICP with ursodeoxycholic acid has been treatment in a multicenter prospective study. Autoimmunity.
2015;48(1):40–5. https://doi.org/10.3109/08916934.2014.939267
reported to protect the placenta including a decreased inci- .
dence of VUE [99, 100]. 15. Boyd TK, Redline RW. Chronic histiocytic intervillositis: a pla-
cental lesion associated with recurrent reproductive loss. Hum
Pathol. 2000;31(11):1389–96.
16. Kim CJ, Romero R, Chaemsaithong P, et al. Chronic inflam-
15.16.2 Acute Fatty Liver of Pregnancy mation of the placenta: definition, classification, pathogenesis,
and clinical significance. Am J Obstet Gynecol. 2015;213(4
Acute fatty liver of pregnancy presents as rapidly progressive Suppl):S53–69. https://doi.org/10.1016/j.ajog.2015.08.041.
maternal acute liver failure, which may include severe mani- 17. Becroft DM, Thompson JM, Mitchell EA. Placental villitis
of unknown origin: epidemiologic associations. Am J Obstet
festations such as hepatic encephalopathy, coagulopathy or Gynecol. 2005;192(1):264–71.
disseminated intravascular coagulation, or hypoglycemia 18. Redline RW. Villitis of unknown etiology: noninfectious chronic
[101]. It is fortunately rare, occurring in fewer than 1 in 5000 villitis in the placenta. Hum Pathol. 2007;38(10):1439–46.
pregnancies. Maternal and fetal mortality are high. The pla- 19. Tamblyn JA, Lissauer DM, Powell R, et al. The immu-
nological basis of villitis of unknown etiology - review.
centa may appear normal or may show effects of preeclamp- Placenta. 2013;34(10):846–55. https://doi.org/10.1016/j.
sia, a commonly associated condition [39]. placenta.2013.07.002.
20. Redline RW. Severe fetal placental vascular lesions in term
infants with neurologic impairment. Am J Obstet Gynecol.
2005;192(2):452–7.
References 21. Faye-Petersen OM, Ernst LM. Maternal floor infarction and mas-
sive perivillous fibrin deposition. Surg Pathol Clin. 2013;6(1):101–
1. Langston C, Kaplan C, Macpherson T, et al. Practice guideline 14. https://doi.org/10.1016/j.path.2012.10.002.
for examination of the placenta: developed by the Placental 22. Adams-Chapman I, Vaucher YE, Bejar RF, et al. Maternal floor
Pathology Practice Guideline Development Task Force of infarction of the placenta: association with central nervous sys-
the College of American Pathologists. Arch Pathol Lab Med. tem injury and adverse neurodevelopmental outcome. J Perinatol.
1997;121(5):449–76. 2002;22(3):236–41. https://doi.org/10.1038/sj.jp.7210685.
2. Heerema-McKenney A, editor. Appendix: sample templates for 23. Baergen RN. Umbilical cord pathology. Surg Pathol Clin.
placental examination. In: Diagnostic pathology: placenta. 1st ed. 2013;6:61–85. https://doi.org/10.1016/j.path.2012.11.003.
Philadelphia: Elsevier; 2015. 24. Georgiadis L, Keski-Nisula L, Harju M, et al. Umbilical cord
3. Kraus FT, Redline RW, Gersell DJ, Nelson DM, Dicke length in singleton gestations: a Finnish population-based retro-
JM. Placental pathology: atlas of nontumor pathology. Silver spective register study. Placenta. 2014;35(4):275–80. https://doi.
Spring, MD: American Registry of Pathology; 2004. org/10.1016/j.placenta.2014.02.001.
4. Fox GE, Van Wesep R, Resau JH, et al. The effect of immersion 25. Proctor LK, Fitzgerald B, Whittle WL, et al. Umbilical cord
formaldehyde fixation on human placental weight. Arch Pathol diameter percentile curves and their correlation to birth weight
Lab Med. 1991;115(7):726–8. and placental pathology. Placenta. 2013;34(1):62–6. https://doi.
5. Khong TY, Mooney EE, Ariel I, et al. Sampling and definitions org/10.1016/j.placenta.2012.10.015.
of placental lesions: Amsterdam placental workshop group con- 26. Machin GA, Ackerman J, Gilbert-Barness E. Abnormal umbili-
sensus statement. Arch Pathol Lab Med. 2016;140(7):698–713. cal cord coiling is associated with adverse perinatal outcomes.
https://doi.org/10.5858/arpa.2015-0225-CC. Pediatr Dev Pathol. 2000;3(5):462–71. https://doi.org/10.1007/
6. Kaplan CG. Color atlas of gross placental pathology. 2nd ed. s100240010103.
New York, NY: Springer Verlag; 2007. 27. de Laat MW, Franx A, van Alderen ED, et al. The umbilical coil-
7. Russel P. Inflammatory lesions of the human placenta. I. Clinical ing index, a review of the literature. J Matern Fetal Neonatal Med.
significance of acute chorioamnionitis. Am J Diagn Gynecol 2005;17(2):93–100.
Obstet. 1979;1:127–37. 28. Khong TY. Evidence-based pathology: umbilical cord coiling.
8. Tita AT, Andrews WW. Diagnosis and management of clinical Pathology. 2010;42(7):618–22. https://doi.org/10.3109/00313025
chorioamnionitis. Clin Perinatol. 2010;37(2):339–54. https://doi. .2010.520309.
org/10.1016/j.clp.2010.02.003. 29. de Laat MW, van Alderen ED, Franx A, et al. The umbilical coil-
9. Janakiraman V. Listeriosis in pregnancy: diagnosis, treatment, and ing index in complicated pregnancy. Eur J Obstet Gynecol Reprod
prevention. Rev Obstet Gynecol. 2008;1(4):179–85. Biol. 2007;130(1):66–72.
10. Kalilani-Phiri L, Thesing PC, Nyirenda OM, et al. Timing of 30. Ernst LM, Minturn L, Huang MH, et al. Gross patterns of umbili-
malaria infection during pregnancy has characteristic maternal, cal cord coiling: correlations with placental histology and still-
infant and placental outcomes. PLoS One. 2013;8(9):e74643. birth. Placenta. 2013;34(7):583–8. https://doi.org/10.1016/j.
https://doi.org/10.1371/journal.pone.0074643. placenta.2013.04.002.
11. Brabin BJ, Romagosa C, Abdelgalil S, et al. The sick placenta- 31. Yampolsky M, Salafia CM, Shlakhter O. Probability distributions
the role of malaria. Placenta. 2004;25(5):359–78. https://doi. of placental morphological measurements and origins of variabil-
org/10.1016/j.placenta.2003.10.019. ity of placental shapes. Placenta. 2013;34(6):493–6. https://doi.
12. Kravetz J. Congenital toxoplasmosis. BMJ Clin Evid. 2013, pii: 0906. org/10.1016/j.placenta.2013.03.003.
13. Revaux A, Mekinian A, Nicaise P, et al. Antiphospholipid syn- 32. De Paepe ME, Luks FI. What-and why-the pathologist should know
drome and other autoimmune diseases associated with chronic about twin-to-twin transfusion syndrome. Pediatr Dev Pathol.
intervillositis. Arch Gynecol Obstet. 2015;291(6):1229–36. 2013;16(4):237–51. https://doi.org/10.2350/13-03-1315-MISC.1.
https://doi.org/10.1007/s00404-014-3536-6. 33. Kraus FT. Fetal thrombotic vasculopathy: perinatal stroke, growth
14. Mekinian A, Costedoat-Chalumeau N, Masseau A, et al. Chronic restriction, and other sequelae. Surg Pathol Clin. 2013;6(1):87–
histiocytic intervillositis: outcome, associated diseases and 100. https://doi.org/10.1016/j.path.2012.10.001.
538 E. Schollenberg et al.
34. Saleemuddin A, Tantbirojn P, Sirois K, et al. Obstetric and peri- 53. Shinde GR, Vaswani BP, Patange RP, et al. Diagnostic performance of
natal complications in placentas with fetal thrombotic vascu- ultrasonography for detection of abruption and its clinical correlation
lopathy. Pediatr Dev Pathol. 2010;13(6):459–64. https://doi. and maternal and foetal outcome. J Clin Diagn Res. 2016;10(8):QC04–
org/10.2350/10-01-0774-OA.1. 7. https://doi.org/10.7860/JCDR/2016/19247.8288.
35. Redline RW, Ariel I, Baergen RN, et al. Fetal vascular obstruc- 54. Redline RW, Wilson-Costello D. Chronic peripheral separation of
tive lesions: nosology and reproducibility of placental reac- placenta. The significance of diffuse chorioamnionic hemosidero-
tion patterns. Pediatr Dev Pathol. 2004;7(5):443–52. https://doi. sis. Am J Clin Pathol. 1999;111(6):804–10.
org/10.1007/s10024-004-2020-x. 55. Redline RW, O’Riordan MA. Placental lesions associated with
36. Chisholm KM, Heerema-McKenney A. Fetal thrombotic cerebral palsy and neurologic impairment following term birth.
vasculopathy: significance in liveborn children using pro- Arch Pathol Lab Med. 2000;124(12):1785–91. https://doi.
posed society for pediatric pathology diagnostic criteria. Am org/10.1043/0003-9985(2000)1242.0.CO;2.
J Surg Pathol. 2015;39(2):274–80. https://doi.org/10.1097/ 56. Kobayashi A, Minami S, Tanizaki Y, et al. Adverse peri-
PAS.0000000000000334. natal and neonatal outcomes in patients with chronic
37. Ananth CV, Friedman AM. Ischemic placental disease and risks abruption- oligohydramnios sequence. J Obstet Gynaecol Res.
of perinatal mortality and morbidity and neurodevelopmen- 2014;40(6):1618–24. https://doi.org/10.1111/jog.12395.
tal outcomes. Semin Perinatol. 2014;38(3):151–8. https://doi. 57. Wu S, Kocherginsky M, Hibbard JU. Abnormal placentation:
org/10.1053/j.semperi.2014.03.007. twenty-year analysis. Am J Obstet Gynecol. 2005;192(5):1458–61.
38. Lo JO, Mission JF, Caughey AB. Hypertensive disease 58. Vinograd A, Wainstock T, Mazor M, et al. Placenta accreta is an
of pregnancy and maternal mortality. Curr Opin Obstet independent risk factor for late pre-term birth and perinatal mor-
Gynecol. 2013;25(2):124–32. https://doi.org/10.1097/ tality. J Matern Fetal Neonatal Med. 2015;28(12):1381–7. https://
GCO.0b013e32835e0ef5. doi.org/10.3109/14767058.2014.955004.
39. Benirschke K, Baergen RN, Burton G. Pathology of the human 59. Wortman AC, Alexander JM. Placenta accreta, increta, and
placenta. 6th ed. Berlin: Springer; 2012. percreta. Obstet Gynecol Clin North Am. 2013. https://doi.
40. Fox H, Sebire NJ. Pathology of the placenta. 3rd ed. Amsterdam: org/10.1016/j.ogc.2012.12.002.
Saunders Elsevier; 2007. 60. Dannheim K, Shainker SA, Hecht JL. Hysterectomy for placenta
41. Higgins M, McAuliffe FM, Mooney EE. Clinical associations accreta; methods for gross and microscopic pathology exami-
with a placental diagnosis of delayed villous maturation: a retro- nation. Arch Gynecol Obstet. 2016;293(5):951–8. https://doi.
spective study. Pediatr Dev Pathol. 2011;14(4):273–9. https://doi. org/10.1007/s00404-015-4006-5.
org/10.2350/10-07-0872-OA.1. 61. Stanek J, Drummond Z. Occult placenta accreta: the missing link
42. Al-Adnani M, Marnerides A, George S, et al. “Delayed vil- in the diagnosis of abnormal placentation. Pediatr Dev Pathol.
lous maturation” in placental reporting: concordance 2007;10(4):266–73.
among consultant pediatric pathologists at a single special- 62. Linn RL, Miller ES, Lim G, et al. Adherent basal plate myometrial
ist center. Pediatr Dev Pathol. 2015;18(5):375–9. https://doi. fibers in the delivered placenta as a risk factor for development
org/10.2350/12-02-1604-OA.1. of subsequent placenta accreta. Placenta. 2015;36(12):1419–24.
43. Kingdom J, Huppertz B, Seaward G, et al. Development of the https://doi.org/10.1016/j.placenta.2015.10.004.
placental villous tree and its consequences for fetal growth. Eur J 63. Martin JA, Hamilton BE, Osterman MJ, et al. Births: final data for
Obstet Gynecol Reprod Biol. 2000;92(1):35–43. 2015. Natl Vital Stat Rep. 2017;66(1):1.
44. Genest DR, Roberts D, Boyd T, et al. Fetoplacental histology as 64. Statistics Canada. Table 102-4515—Live births and fetal deaths
a predictor of karyotype: a controlled study of spontaneous first (stillbirths), by type (single or multiple), Canada, provinces and
trimester abortions. Hum Pathol. 1995;26(2):201–9. territories, annual (number). http://www5.statcan.gc.ca/cansim/
45. Van Lijnschoten G, Arends JW, Thunnissen FB, et al. A morpho- a47. Accessed 21 Dec 2017.
metric approach to the relation of karyotype, gestational age and 65. Cunningham FG, Williams JW, editors. Multifetal pregnancy. In:
histological features in early spontaneous abortions. Placenta. Williams obstetrics. 24th ed. New York: McGraw-Hill Med; 2014.
1994;15(2):189–200. 66. Murata M, Ishii K, Kamitomo M, et al. Perinatal outcome and
46. van Lijnschoten G, Arends JW, De La Fuente AA, et al. Intra- clinical features of monochorionic monoamniotic twin gestation. J
and inter-observer variation in the interpretation of histological Obstet Gynaecol Res. 2013;39(5):922–5. https://doi.org/10.1111/
features suggesting chromosomal abnormality in early abortion jog.12014.
specimens. Histopathology. 1993;22(1):25–9. 67. Behrendt N, Galan HL. Twin-twin transfusion and laser ther-
47. Faye-Petersen OM and Kapur RP. Placental mesenchymal apy. Curr Opin Obstet Gynecol. 2016;28(2):79–85. https://doi.
dysplasia. Surg Pathol Clin. 2013. https://doi.org/10.1016/j. org/10.1097/GCO.0000000000000247.
path.2012.11.007. 68. Quintero RA, Morales WJ, Allen MH, et al. Staging of twin-twin
48. Cohen MC, Roper EC, Sebire NJ, et al. Placental mesenchy- transfusion syndrome. J Perinatol. 1999;19(8 Pt 1):550–5.
mal dysplasia associated with fetal aneuploidy. Prenat Diagn. 69. Zhao D, Lipa M, Wielgos M, et al. Comparison between mono-
2005;25(3):187–92. https://doi.org/10.1002/pd.1103. chorionic and dichorionic placentas with special attention to vas-
49. Huynh J, Dawson D, Roberts D, et al. A systematic cular anastomoses and placental share. Twin Res Hum Genet.
review of placental pathology in maternal diabetes melli- 2016;19(3):191–6. https://doi.org/10.1017/thg.2016.19.
tus. Placenta. 2015;36(2):101–14. https://doi.org/10.1016/j. 70. De Paepe ME, Shapiro S, Greco D, et al. Placental markers of
placenta.2014.11.021. twin-to-twin transfusion syndrome in diamniotic-monochorionic
50. Clarson C, Tevaarwerk GJ, Harding PG, et al. Placental weight in twins: a morphometric analysis of deep artery-to-vein anasto-
diabetic pregnancies. Placenta. 1989;10(3):275–81. moses. Placenta. 2010;31(4):269–76. https://doi.org/10.1016/j.
51. Lilja M. Infants with single umbilical artery studied in a national placenta.2009.12.024.
registry. 3: a case control study of risk factors. Paediatr Perinat 71. Denbow ML, Cox P, Taylor M, et al. Placental angioarchitecture
Epidemiol. 1994;8(3):325–33. in monochorionic twin pregnancies: relationship to fetal growth,
52. Cunningham FG, Williams JW, editor. Diabetes. In: Williams fetofetal transfusion syndrome, and pregnancy outcome. Am J
obstetrics. 24th ed. New York: McGraw-Hill Med; 2014. Obstet Gynecol. 2000;182(2):417–26.
15 Placenta and Pregnancy-Related Diseases 539
72. Slaghekke F, Lewi L, Middeldorp JM, et al. Residual anastomo- 86. Barros A, Freitas AC, Cabral AJ, et al. Giant placental chorioan-
ses in twin-twin transfusion syndrome after laser: the Solomon gioma: a rare cause of fetal hydrops. BMJ Case Rep. 2011; https://
randomized trial. Am J Obstet Gynecol. 2014;211(3):285.e1–7. doi.org/10.1136/bcr.02.2011.3880.
https://doi.org/10.1016/j.ajog.2014.05.012. 87. Shafqat G, Iqbal F, Rizvi F. Chorangioma of the placenta with
73. Slaghekke F, Lopriore E, Lewi L, et al. Fetoscopic laser coagu- hydrops foetalis. J Pak Med Assoc. 2009;59(6):411–2.
lation of the vascular equator versus selective coagulation for 88. Suri V, Aggarwal N, Deo ND, et al. Placental chorioangioma
twin-to-twin transfusion syndrome: an open-label randomised with hydrops foetalis: a case report. Acta Obstet Gynecol Scand.
controlled trial. Lancet. 2014;383(9935):2144–51. https://doi. 2005;84(6):603–4.
org/10.1016/S0140-6736(13)62419-8. 89. Santamaria M, Benirschke K, Carpenter PM, et al. Transplacental
74. Lopriore E, Slaghekke F, Middeldorp JM, et al. Accurate and sim- hemorrhage associated with placental neoplasms. Pediatr Pathol.
ple evaluation of vascular anastomoses in monochorionic placenta 1987;7(5–6):601–15.
using colored dye. J Vis Exp. 2011. https://doi.org/10.3791/3208. 90. Isaacs H. Fetal hydrops associated with tumors. Am J Perinatol.
75. Emery SP, Nguyen L, Parks WT, et al. Histological appearance of 2008;25(1):43–68. https://doi.org/10.1055/s-2007-1004826.
placental solomonization in the treatment of twin-twin transfusion 91. Bianchi DW, editor. Nonimmune hydrops fetalis. In: Fetology.
syndrome. AJP Rep. 2016;6(2):e165–9. 2nd rev. ed. New York: McGraw-Hill; 2010.
76. Desilets V, Audibert F, Society of Obstetrician and Gynaecologists 92. Cunningham FG, Williams JW, editors. Abortion. In: Williams
of Canada. Investigation and management of non-immune fetal obstetrics. 24th ed. New York: McGraw-Hill Education/Medical;
hydrops. J Obstet Gynaecol Can. 2013;35(10):923–38. 2014.
77. Society for Maternal-Fetal Medicine (SMFM), Norton ME, 93. Association of Directors of Anatomic and Surgical Pathology,
Chauhan SP, et al. Society for maternal-fetal medicine (SMFM) Silverman JF, Fletcher CD, et al. Critical diagnoses (critical val-
clinical guideline #7: nonimmune hydrops fetalis. Am J Obstet ues) in anatomic pathology. Hum Pathol. 2006;37(8):982–4.
Gynecol. 2015;212(2):127–39. https://doi.org/10.1016/j. 94. Genest DR. Estimating the time of death in stillborn fetuses:
ajog.2014.12.018. II. Histologic evaluation of the placenta; a study of 71 stillborns.
78. Bukowski R, Hansen NI, Pinar H, et al. Altered fetal growth, placen- Obstet Gynecol. 1992;80(4):585–92.
tal abnormalities, and stillbirth. PLoS One. 2017;12(8):e0182874. 95. Parast MM, Crum CP, Boyd TK. Placental histologic criteria
https://doi.org/10.1371/journal.pone.0182874. for umbilical blood flow restriction in unexplained stillbirth.
79. Machin GA. Differential diagnosis of hydrops fetalis. Am Hum Pathol. 2008;39(6):948–53. https://doi.org/10.1016/j.
J Med Genet. 1981;9(4):341–50. https://doi.org/10.1002/ humpath.2007.10.032.
ajmg.1320090410. 96. Sebire NJ, Lindsay I, Fisher RA, et al. Intraplacental chorio-
80. Kraus FT. Clinical syndromes with variable pathologic features. carcinoma: experience from a tertiary referral center and rela-
Semin Diagn Pathol. 2007;24(1):43–7. tionship with infantile choriocarcinoma. Fetal Pediatr Pathol.
81. Knisely AS. The pathologist and the hydropic placenta, fetus, or 2005;24(1):21–9.
infant. Semin Perinatol. 1995;19(6):525–31. 97. Williamson C and Geenes V. Intrahepatic cholestasis of
82. Sarno M, Sacramento GA, Khouri R, et al. Zika virus infection pregnancy. Obstet Gynecol. 2014. https://doi.org/10.1097/
and stillbirths: a case of hydrops fetalis, hydranencephaly and AOG.0000000000000346.
fetal demise. PLoS Negl Trop Dis. 2016;10(2):e0004517. https:// 98. Williamson C, Miragoli M, Sheikh Abdul Kadir S, et al. Bile
doi.org/10.1371/journal.pntd.0004517. acid signaling in fetal tissues: implications for intrahepatic cho-
83. de Mooij YM, van den Akker NM, Bekker MN, et al. Aberrant lestasis of pregnancy. Dig Dis. 2011;29(1):58–61. https://doi.
lymphatic development in euploid fetuses with increased org/10.1159/000324130.
nuchal translucency including Noonan syndrome. Prenat Diagn. 99. Patel S, Pinheiro M, Felix JC, et al. A case-control review of pla-
2011;31(2):159–66. https://doi.org/10.1002/pd.2666. centas from patients with intrahepatic cholestasis of pregnancy.
84. Rohanizadegan M, Abdo SM, O’Donnell-Luria A, et al. Utility Fetal Pediatr Pathol. 2014;33(4):210–5. https://doi.org/10.3109/1
of rapid whole-exome sequencing in the diagnosis of Niemann- 5513815.2014.899413.
Pick disease type C presenting with fetal hydrops and acute liver 100. Geenes VL, Lim YH, Bowman N, et al. A placental phenotype for
failure. Cold Spring Harb Mol Case Stud. 2017;3(6):pii: a002147. intrahepatic cholestasis of pregnancy. Placenta. 2011;32(12):1026–
https://doi.org/10.1101/mcs.a002147. 32. https://doi.org/10.1016/j.placenta.2011.09.006.
85. Delbecque K, Gaillez S, Schaaps JP. Histopathological diag- 101. Liu J, Ghaziani TT, Wolf JL. Acute fatty liver disease of preg-
nosis of a type VII mucopolysaccharidosis after pregnancy ter- nancy: updates in pathogenesis, diagnosis, and management. Am
mination. Fetal Pediatr Pathol. 2009;28(1):1–8. https://doi. J Gastroenterol. 2017;112(6):838–46. https://doi.org/10.1038/
org/10.1080/15513810802547943. ajg.2017.54.
Principles and Practical Guidelines
of Intraoperative Consultation 16
Hannah Goyne, Emily Paull Acheson,
and Charles Matthew Quick
Abstract carcinoma are just some examples. The most important first
Frozen section analysis of the gynecologic tract is a com- step in evaluating any specimen is to ask oneself, what does
plex task due to the number of organs, tumors, and entities the surgeon want to know, followed by how do I process this
involved. Awareness of the importance of a thorough specimen to answer this question?
gross evaluation, careful selection of samples, and the Gross evaluation of gynecologic specimen is often com-
numerous histologic features associated with this plex. In general, all specimens should be properly oriented,
endeavor is a must. This chapter acts to serve as a quick measured, and weighed before sectioning. The serosa
reference to some of the more common epidemiologic, should be carefully inspected to look for tumor deposits,
gross, and histologic features of some of the more com- which should be differentially inked if present. Many spec-
mon entities encountered in intraoperative consultation of imens, such as uteri and ovaries, require complex section-
the female reproductive system. ing to assure enough tissue is exposed and that the extent of
disease can be delineated. Selection of the optimal tissue
Keywords for freezing is predicated on a proper gross exam, identifi-
Intraoperative consultation · Frozen section cation of sometimes subtle features, and the avoidance of
Endometrium · Myometrium · Products of conception pitfalls such as tumor necrosis. Random sections may be
Ovary · Fallopian tube · Peritoneum · Lymph node prescribed in the case of a grossly unapparent lesion.
Additionally, if loss of diagnostic material is a concern
(such as in grossly unremarkable fallopian tubes), frozen
sectioning may be contraindicated. A description of these
16.1 Introduction intricacies for each type of specimen that may be encoun-
tered is detailed below.
The performance of intraoperative consultation centered on Histologic evaluation of frozen section slides requires a
gynecologic pathology is a daunting task. In general, this working knowledge of the vast number of tumors that may
practice can help to inform the surgeon of the appropriate affect a patient. Obviously, this is complicated by the fact
course of action in ongoing cases. This may take the form of that many tumors are capable of displaying subtle histologic
a staging procedure or the cessation of surgery. The different changes that may be overlooked. Furthermore, evaluation of
organs that comprise the gynecologic tract often require dif- “frozens” can be made more difficult by the relatively
fering types of analysis to help complete this goal: diagnosis decreased histologic resolution that is part and parcel of fro-
of a mass in ovarian carcinoma, intraoperative staging of an zen sectioning. When evaluating a case, never hesitate to ask
endometrial carcinoma, and margin evaluation in a vulvar for additional sections or deeper levels if the possibility of
either will help. The purpose of this chapter is to serve as a
primer on how to approach various specimens and scenarios
H. Goyne · E. P. Acheson · C. M. Quick (*) that could occur during intraoperative consultation of the
Department of Pathology, College of Medicine, gynecologic tract. Brief statements on some of the more
University of Arkansas for Medical Sciences,
Little Rock, AR, USA
common entities have been provided, yet more detailed
e-mail: [email protected]; [email protected]; descriptions of the entities herein are presented elsewhere in
[email protected] this book.
© Science Press & Springer Nature Singapore Pte Ltd. 2019 541
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_16
542 H. Goyne et al.
16.2 F
rozen Section Analysis of Vaginal used to specify margins. Mass lesions and areas of discolor-
and Vulvar Samples ation should be identified and described, and the distance to
each margin is measured. In general, a perpendicular section
16.2.1 Purpose of Frozen Section Evaluation of the closest margin should be chosen for freezing.
a b
Fig. 16.1 (a) Squamous cell carcinoma of the vulva presenting as a tan, circular, plaque-like mass. Note the adjacent areas of high-grade dysplasia
identified by their exophytic, white configuration. (b) Patchy high-grade vulvar dysplasia presenting as an ill-defined, white discoloration
16 Principles and Practical Guidelines of Intraoperative Consultation 543
these mimics is suspected, then a descriptive diagnosis, 16.3 Frozen Section Analysis of the Cervix
warning of atypical cells, is appropriate.
16.3.1 Purpose of Frozen Section Evaluation
16.2.3.3 Vulvar Intraepithelial Neoplasia (VIN)
Two variants of VIN exist, classic and differentiated type. In general, frozen section evaluation of the cervix outside of
The classic type is associated with HPV, particularly type 16, the setting of possible extension of endometrial adenocarci-
whereas differentiated VIN is associated with chronic inflam- noma is rare. On rare occasion, a surgeon may send a biopsy,
matory states as opposed to HPV infection. Histologically, or small conization specimen, to determine if invasive dis-
classic VIN has atypia found throughout the layers of epithe- ease is present and, thus, which type of hysterectomy should
lium. Enlarged nuclei, hyperchromasia, multinucleation in be performed. Another possible scenario involves margin
the lower epithelial layers, and abnormal mitotic figures are evaluation of trachelectomy specimen, although this is not
characteristic of the lesions. Apoptosis may be present. The always performed at certain institutions. In cases with pos-
surface may show atypical parakeratosis with or without sible lymph node involvement, nodal tissue may be sent to
koilocytes. Low-grade dysplasia demonstrates increased rule out metastasis. Frozen sectioning of therapeutic LEEP
thickness of the basal layer, occasional binucleation, and specimen is generally contraindicated, as it may destroy
enlarged nuclei. Koilocytic atypia is often present, but under- diagnostic tissue.
stated when compared to cervical koilocytes. High-grade
dysplasia displays full-thickness loss of maturation. The
cells have a high nuclear to cytoplasmic ratio giving the epi- 16.3.2 Gross Specimen Processing
thelium a basaloid appearance.
Differentiated VIN is often associated with an inflamma- Small cone specimen for diagnosis should be measured
tory dermatosis. They are typically seen in older women with and evaluated for gross evidence of disease. In the case of
clinical presentations of lichen sclerosis and lichen simplex a grossly apparent lesion, one representative section of
chronicus. Histologically, they have severe basal atypia, said lesion should be submitted for sectioning. If no lesion
elongation of the rete ridges, and abrupt keratinization [2]. is grossly apparent, the two random, perpendicular sec-
The diagnosis of differentiated-type VIN may be difficult; tions of the cervix should be submitted. Trachelectomy
the presence of dermal hyalinization (indicative of lichen specimen may, on occasion, be sent for evaluation of the
sclerosis), severe basal atypia, and increased, or atypical, margins. The ectocervical and endocervical aspects of the
mitotic figures in the basal layers can be helpful. A more specimen should be identified, and the outer surface of the
detailed explanation of the features and diagnosis of VIN can cervix and paracervical tissue should be carefully inked.
be found in Chap. 6 (Volume I). Application of a drying agent, such as alcohol, should be
applied to the inked surface, which should be allowed a
short time to sufficiently dry before sectioning. Once the
16.2.4 Diagnosis and Reporting ink is dried, the specimen should be opened by a single
excision, allowing visualization of the endocervical canal.
When reporting the frozen section, mention the presence of Depending on the location and configuration of the lesion,
squamous cell carcinoma and/or dysplasia and if the lesion is a shave (if the lesion is distant from the edge) or perpen-
at the margin. If the margins are negative, report the distance dicular (if the lesion is close to the edge) should be taken.
to that margin whether obtained via gross or microscopic In the case of deeply invasive or ulcerative lesions, a sec-
means. If dysplasia is present, report it and state if it is low or tion demonstrating the lesion and underlying radial margin
high grade. The diagnosis may be challenging when there is should be taken as well.
tangential sectioning, obscuring inflammation, involvement
of adnexal structures by in situ neoplasia, melanocytic prolif-
erations, herpetic infection, squamous eddies in irritated seb-
orrheic keratosis, and pseudoepitheliomatous hyperplasia 16.3.3 Common Entities
associated with a granular cell tumor [1]. When situations
such as this arise, requesting of additional levels as well as a 16.3.3.1 Squamous Cell Carcinoma
brief chart review can be helpful in providing guidance. In Squamous cell carcinoma may be exophytic or ulcerative.
the case of difficult squamous lesions, a diagnosis of inva- Histologically, the lesion is similar to squamous cell
sion should only be rendered when unequivocal evidence of carcinoma seen elsewhere and is composed of irregularly
invasion is present; if lacking, a diagnosis of “suspicious for shaped jagged nests of squamous epithelium in the stroma
invasion” will allow the surgeon to make a therapeutic deci- below the epithelial basement membrane. A desmoplastic
sion based on their observations. stromal response is typically present.
544 H. Goyne et al.
16.3.3.2 Endocervical Adenocarcinoma uterine segment. If the mass approximates the cervix, make
Similar to their squamous counterparts, endocervical adeno- a longitudinal section (parallel to and along the endocervical
carcinomas may present as a polypoid, as an exophytic mass, canal) to evaluate for cervical invasion. Any suspicious areas
or as an ulcerative lesion. In most cases, irregular, infiltrative should be submitted for frozen section. If no polyps or
glands with prominent desmoplasia can be seen extending masses are present, inspect the endometrium lining for any
into the endocervical stroma. On occasion, these tumors may firm, white areas, as this may represent carcinoma. Avoid
lack desmoplasia as well as obvious invasion. In this latter rubbing the endometrial lining. Thinly serially section the
case, it is appropriate to report the margin status alone and endomyometrium (bread loafing). Inspect the myometrium
defer the final diagnosis to examination of permanent for the area of deepest invasion, and submit a full-thickness
sections. section for frozen evaluation. Grossly, invasion of the myo-
metrium presents as a tan to white mass, often with a firm
consistency, replacing the normal myometrial texture
16.3.4 Diagnosis and Reporting (Figs. 16.3 and 16.4). The best time to evaluate the depth of
myometrial invasion is to examine the sections by removing
In cases sent for diagnosis, such as small cones, the reporting the oozing blood by a cutting blade or knife immediately
of the presence of invasive disease is the primary objective. after sectioning the endomyometrium. Adenomyosis may
In cases where there is no unequivocal evidence of invasion, mimic invasion and is characterized grossly by thick myo-
a descriptive diagnosis is appropriate. Trachelectomy speci- metrial trabeculae with pinpoint dark or red hemorrhages.
mens are usually sent for evaluation of the margins, as the
histotype of the tumor is known in most cases.
16.4.3 Common Entities
16.4 F
rozen Section Analysis 16.4.3.1 Endometrial Polyp
of Endometrial Lesions One of the most common causes of abnormal uterine bleed-
ing is an endometrial polyp. Characteristic histologic fea-
16.4.1 Purpose of Frozen Section Evaluation tures include fibrous or collagenous stroma, thick-walled
vessels, and/or benign glands with an irregular architecture,
When evaluating endometrial lesions, it is important to often displaying cystic change.
determine if carcinoma is present and, if so, the type. In addi-
tion, the grade, depth of invasion into the myometrium, pres- 16.4.3.2 Adenomyosis
ence of cervical stromal invasion, and involvement of serosal, Ectopic endometrial glands and stroma in the myometrium
ovarian, or fallopian tube tissue should be detailed if define adenomyosis. These foci may become involved by
possible. endometrial carcinoma and should not be confused with
deep invasion (Fig. 16.5a, b). The presence of endometrial
stroma and normal glands surrounding the malignant glands
16.4.2 Gross Specimen Processing is a helpful clue. Surrounding desmoplasia should be absent.
First, measure and weigh the uterus. Carefully inspect the 16.4.3.3 Endometrioid Intraepithelial
serosal surface and cervix (if present) for any lesions or Neoplasia (EIN)
irregularities. If an area suspicious for serosal implants is This lesion is the precancerous counterpart to endometrioid
present, it should be inked and sampled. Orient the uterus adenocarcinoma. Five criteria are necessary to make this
using the peritoneal reflection, which is lower posteriorly diagnosis and include the following: (1) EIN stands out/dif-
than anteriorly. Alternatively, if adnexa are present, the fal- fers from background endometrial glands, (2) there is gland
lopian tubes typically lie anterior to the ovaries. Inspect the crowding (gland to stroma ratio >50%), (3) size exceeds
surface of the ovaries and diameter of the fallopian tubes. 1 mm, (4) benign diagnostic mimics are ruled out, and (5)
Orient, remove, and serially section the adnexa for evidence carcinoma is excluded. This lesion is discussed in detail in
of any mass or lesion. Ink the anterior and posterior radial Chap. 17 (Volume I).
surgical margins, and bivalve the uterus, along the lateral
edges, with a pair of sharp scissors or by holding the uterus 16.4.3.4 Endometrioid Adenocarcinoma
upside down with a long handle forceps and then cutting Most of these lesions are characterized histologically by
with a sharp knife (Fig. 16.2a, b). Inspect the endometrial well-defined, tubular glands that are lined by malignant,
cavity for polyps and/or masses, and measure them if pres- stratified or pseudostratified, columnar epithelium.
ent. Take note how far away the mass is from the cervix/lower Architecturally the glands are complex and appear fused or
16 Principles and Practical Guidelines of Intraoperative Consultation 545
a b
Fig. 16.2 (a) A properly oriented uterus and bilateral adnexa. (b) After aspects with sharp scissors. Note the carcinoma present in the lower
removal of the adnexa and inking of the lower uterine segment and uterine segment in both the anterior and posterior halves
radial cervical margin, the uterus should be opened along the lateral
16.4.3.6 Endometrial Serous Carcinoma pleomorphism. Nucleoli are prominent and mitotic figures
Pure serous carcinomas account for about 1–10% of endo- are frequent.
metrial carcinomas. The tumors can have a solid, papillary,
micropapillary, glandular, or slit-like glandular pattern 16.4.3.7 Carcinosarcoma
(Fig. 16.8a, b). The nuclei are high grade and show marked Uterine carcinosarcomas are also known as malignant mixed
Mullerian tumors (MMMT). They tend to occur in elderly
women, are highly aggressive, and have a poor prognosis.
Grossly they are often polypoid/exophytic (Fig. 16.9).
Histologically, any subtype of endometrial carcinoma com-
bined with a high-grade malignant mesenchymal component
should be present. The neoplasms may contain homologous
(tissues normally found in the uterus) or heterologous (tissues
not normally found in the uterus, e.g., cartilage) elements [4].
a b
Fig. 16.5 (a) Adenomyosis may be recognized grossly as thick myo- adenomyosis. Note the presence of smooth outer contours of the
metrial trabeculae with areas of hemorrhage (photo courtesy of involved areas as well as uninvolved adenomyosis
Katelynn Campbell, M.D.). (b) Endometrial adenocarcinoma involving
16 Principles and Practical Guidelines of Intraoperative Consultation 547
a b
Fig. 16.6 (a) Well-differentiated endometrioid adenocarcinoma with fused, cribriforming glands and immature morular metaplasia. (b)
Labyrinthine glands with intraluminal papillary growth and necrotic luminal debris
a b
Fig. 16.7 (a) Clear cell carcinoma of the endometrium growing as tubules and cysts. Note the prominent cellular hobnailing. (b) Clear cell carci-
noma with small, hyalinized papillae
a b
Fig. 16.8 (a) High-grade serous carcinoma. Note the presence of numerous exfoliated tumor cells and “cracks and crevices” within the epithelial
surface. Nuclear pleomorphism and prominent nucleoli are easily seen. (b) High-grade serous carcinoma with solid growth
548 H. Goyne et al.
16.6 F
rozen Section Analysis
16.5.3 Diagnosis and Reporting of Myometrial Lesions
If viable chorionic villi (Fig. 16.11), with or without implan- 16.6.1 Purpose of Frozen Section Evaluation
tation site (Fig. 16.12), are identified, “intrauterine preg-
nancy” may be reported. Scant scattered villi may represent The purpose of the majority of intraoperative consultation
contamination from a tubal pregnancy, and this should be centered on the myometrium is to rule out malignancy in a
discussed with the consulting clinician. Implantation site presumed leiomyoma. Occasionally, other neoplasms such
alone without villi is highly suggestive of intrauterine preg- as endometrial stromal sarcoma or uterine tumor resembling
nancy; however, implantation site contamination from a ovarian sex cord tumor may be a diagnostic consideration.
tubal pregnancy cannot be completely ruled out in this sce-
nario. The presence of decidualized endometrium may be
seen in the setting of any pregnancy, including ectopic, and 16.6.2 Gross Specimen Processing
therefore is not indicative of an intrauterine pregnancy in
isolation. The specimen usually consists of a hysterectomy or resection
of specific lesions from the uterus. In the case of a leiomy-
oma, the tumor is typically composed of a firm, well-
circumscribed mass (Fig. 16.13a–d), which is measured and
weighed. It is important to document the weight of the speci-
men as it may impact patient charges in some practice set-
tings. Serially section the mass at 1 cm intervals, and inspect
the cut surface for necrosis and/or hemorrhage. Any soft
areas, tissue adjacent to necrosis, and areas of hemorrhage
should be sampled. Leiomyomata may have hydropic change
and degeneration making gross evaluation for malignancy
challenging (Fig. 16.14a–d). If the specimen is not well-
circumscribed, this may suggest malignancy.
16.6.3.1 Leiomyoma
Fig. 16.11 Scattered, immature chorionic villi present within a large Leiomyomata are the most common gynecologic tumor
blood clot and are commonly seen in hysterectomies regardless of the
550 H. Goyne et al.
a b
c d
Fig. 16.13 (a) Multiple leiomyomata distorting the myometrium and uterine specimen with a large, pedunculated serosal leiomyoma and a
endometrial cavity. (b) Numerous submucosal leiomyomata projecting smaller intramural leiomyoma
into the endometrial cavity. (c) A single submucosal leiomyoma. (d) A
surgical indication. Grossly, they are firm, well-circum- Finally, cotyledonoid dissecting leiomyomas may grossly
scribed, tan-white masses that have a whorled, bulging appear invasive and similar to low-grade endometrial stro-
appearance on cut surface (Fig. 16.13a–d). They should mal tumors. Histologically, the tumors are composed of
lack hemorrhage and necrosis; however, degenerative fascicles of bland spindle cells with cigar-shaped nuclei.
changes may be present (Fig. 16.14a–d) and are typically The fascicles intersect at 90° angles. Epithelioid and myx-
centrally located and well demarcated. Long-standing oid variants can be seen as well. Mitotic activity should be
leiomyomata may undergo calcification as well. The rare to absent. The proliferation of these tumors is stimu-
lesions may bulge into the endometrial cavity or form sub- lated by progesterone; therefore, mitotic activity may
serosal nodules. When large and/or numerous, they can increase in the secretory phase of the menstrual cycle.
severely distort the uterus and endometrial cavity. Several Mitotic activity may increase after surgery, during post-
clinical settings and variants of leiomyoma may grossly be menopausal hormone or tamoxifen therapy. If mitotic fig-
concerning for malignancy. In the setting of previous ures are seen, they should not be atypical. Tumor necrosis
embolization, hormonal therapy, and pregnancy, the leio- should be absent; however, infarct-type necrosis is allowed.
myoma may undergo infarction or apoplexy, leading to a Nuclear atypia should not be present. Occasionally long-
hemorrhagic cut surface. Hydropic leiomyomas and highly standing leiomyomata undergo ancient or symplastic
cellular leiomyomas may lack the distinct, fascicular, changes and can have bizarre nuclei, but mitotic activity
bulging cut surface of otherwise more typical leiomyomas. should not be increased.
16 Principles and Practical Guidelines of Intraoperative Consultation 551
a b
c d
Fig. 16.14 (a) The bosselated cut surface of a leiomyoma with degen- degeneration of a leiomyoma. (d) Patchy cystic and red degeneration in
erative changes. (b) Degenerative change in a leiomyoma often consists a large leiomyoma
of centrally located cystic change or discoloration. (c) So-called “red”
a b
Fig. 16.15 (a) The cut surface of a leiomyosarcoma demonstrating irregular borders and extensive hemorrhage and necrosis. (b) Leiomyosarcoma
with a fleshy cut surface, degenerative change (lower left) and necrosis (far right)
Fig. 16.16 Nuclear pleomorphism in leiomyosarcoma is often pro- Fig. 16.17 An epithelioid leiomyosarcoma composed of round cells
nounced; as in this example, albeit, atypia in isolation is not diagnostic with ample clear cytoplasm. In some cases, these tumors may be diffi-
of malignancy cult to classify as mesenchymal
pinpoint hemorrhages in the myometrium. Occasionally, arterioles scattered throughout the lesion. These tumors are
areas of adenomyosis may for a mass lesion, termed distinguished from endometrial stromal nodules by the pres-
adenomyoma. ence of myometrial infiltration (>3 mm) and/or lymphovas-
cular invasion.
16.6.3.5 Endometrial Stromal Lesions
Endometrial stromal lesions are typically divided into two
types: stromal nodule and stromal sarcoma. Endometrial 16.6.4 Diagnosis and Reporting
stromal sarcomas are rare, accounting for 0.2% of all uterine
malignancies and 15% of uterine sarcomas [6]. Myometrial If malignancy is present, the surgeon may elect to perform a
invasion appears as fingerlike projections, and, grossly, they hysterectomy instead of myomectomy alone. If a benign
have been described as a “bag of worms” within the myome- diagnosis is made, such as a leiomyoma, conservative man-
trium (Fig. 16.19a, b). Histologically, they resemble agement may be implemented. This is particularly important
proliferative endometrial stroma with numerous thin-walled for women who want to maintain their fertility.
16 Principles and Practical Guidelines of Intraoperative Consultation 553
a b
Fig. 16.19 (a) Endometrial stromal sarcoma presenting as multiple wormlike tan-yellow protrusions (lower left aspect). (b) Innumerable polyp-
oid projections in an extensive endometrial stromal sarcoma
554 H. Goyne et al.
implantation site, and even embryonic tissue in advanced red, nucleoli. Numerous mitotic figures, pleomorphism, and
cases is diagnostic (Fig. 16.20). apoptotic bodies are easily appreciated. High-grade serous
carcinoma may be diagnosed in the presence of invasion,
16.7.2.2 Serous Carcinoma which may be bilateral in up to 20% of cases.
The most common primary carcinoma of the fallopian tube is
serous carcinoma, which accounts for 90% of all tubal carci- 16.7.2.3 Endometriosis
nomas. Serous tubal intraepithelial carcinoma (STIC) is the in Endometriosis is very common and is found in about 10% of
situ form of high-grade serous carcinoma and may be identi- tubal specimens. Histologically, it is characterized by endo-
fied on frozen section; however, frozen sectioning of grossly metrial glands and stroma in the muscularis, subserosa, or
normal tubes to identify STIC is contraindicated, as previously mesosalpinx (Fig. 16.23). Older lesions may be surrounded
stated. If a mass lesion is present, a portion may be sampled by fibrosis and hemosiderin-laden macrophages.
for frozen sectioning (Fig. 16.21). The epithelium may show
thickening, loss of polarity, and high nuclear-to-cytoplasmic 16.7.2.4 Transitional Cell Lesions
ratios with loss of cilia (Fig. 16.22). The nuclei are hyperchro- Walthard cell rests are benign and consist of transitional
matic and enlarged and have prominent, sometimes cherry metaplasia of mesothelial cells or subcolumnar cells
Fig. 16.20 Cross section of an advanced tubal ectopic pregnancy dem- Fig. 16.22 Serous tubal intraepithelial carcinoma (STIC) and a small
onstrating from left to right: tubal wall, implantation site, chorionic focus of invasive serous carcinoma. Note the increased thickness of the
villi, and embryonic tissue tubal epithelium and clusters of shed cells present within the lumen
Fig. 16.21 Section of serous carcinoma involving the wall of the fal- Fig. 16.23 Endometriosis with pre-decidual change within the wall of
lopian tube forming a grossly appreciable mass lesion the fallopian tube
16 Principles and Practical Guidelines of Intraoperative Consultation 555
Fig. 16.24 Walthard cell rests on the serosal surface of the fallopian Fig. 16.25 Adenomatoid tumor of the fallopian tube. These tumors
tube. Grossly these may appear as small paratubal cysts appear to be composed of small glands or vascular spaces at low power
(Fig. 16.24). They are commonly seen adjacent to the fimbria sequela of the disease results in fusion of tubal plicae and is
and serosa. Traditionally, transitional cell carcinoma has termed follicular salpingitis. Mycobacterium tuberculosis
been described as a carcinoma that resembles malignant uro- causes granulomatous salpingitis and can mimic a mass.
thelial epithelium, but is now believed to be variants of high-
grade serous carcinomas by many investigators [7, 8].
16.7.3 Diagnosis and Reporting
16.7.2.5 Salpingitis Isthmica Nodosa
The prevalence of salpingitis isthmica nodosa varies, but is If the fallopian tube is grossly normal, the pathologist should
relatively common. It is a condition that is analogous to ade-render a negative gross description to the submitting surgeon
nomyosis (i.e., tubal epithelium is located in the muscular and recommend fixation and permanent sections. When an
wall of the fallopian tube). The gross appearance is typicallyinvasive high-grade carcinoma is present, this can be reported
a yellow to white nodule near the tubal isthmus. as such, with an additionally comment favoring serous carci-
noma if appropriate. Serous tubal intraepithelial carcinoma
16.7.2.6 Adenomatoid Tumor is difficult to diagnose on frozen section. If atypia is present,
The most common benign tumor of the fallopian tube is the a diagnosis of “severe tubal atypia” with deference to perma-
adenomatoid tumor. Grossly, the tumor is small, well- nent sections is warranted. If products of conception are
circumscribed, and usually located on the serosal surface. identified either grossly or on frozen section, a diagnosis of
The cut surface is typically white. The lesion is composed of tubal ectopic pregnancy should be rendered.
small irregular gland-like spaces lined by a single layer of
benign mesothelial cells (Fig. 16.25). Intervening smooth
muscle or hyalinized stroma may be present. 16.8 Frozen Section Analysis
of Ovarian Masses
16.7.2.7 Infectious Disorders
Pelvic inflammatory disease (PID) is the clinical term for 16.8.1 Purpose of Frozen Section Evaluation
infectious salpingitis. Women with this condition are at a
greatly increased risk of developing an ectopic pregnancy. The primary purpose of an intraoperative consultation for an
Most cases of PID are caused by sexually transmitted disease ovarian/pelvic mass is to determine whether the process is
(e.g., Neisseria gonorrhoeae and Chlamydia trachomatis), benign or malignant and potentially to provide a precise diag-
but two other causes include instrumentation and use of nosis or tumor grade. If benign, the surgeon can proceed con-
intrauterine devices. Grossly, early cases will present as a servatively, which is especially important in women who
purulent exudate (pyosalpinx) in the fallopian tube, followed wish to maintain fertility. If malignant, the pathologist should
by tubo-ovarian adhesions, and finally by hydrosalpinx. try to determine whether it is a primary or metastatic tumor as
Histologically, early infection will result in acute inflamma- this will alter treatment. This can be particularly challenging
tion followed by chronic inflammation and adhesions. Late in mucinous tumors, in which there are many overlapping
556 H. Goyne et al.
adenoma may be present. Borderline tumors can be they tend to more solid than borderline tumors, and histo-
distinguished from low-grade serous carcinomas by the logically stromal invasion or markedly complex epithelial
absence of significant micropapillary or cribriform growth growth (extensive micro-papillae or cribriforming) should be
and lack of invasion into the stroma. This may be challeng- present. Clear spaces between individual nests of tumor cells
ing on frozen section due to the nature of complexity of the and the stroma are a helpful diagnostic feature. Psammoma
lesions, limited sampling, and oblique sectioning. The bodies are often numerous. The nuclei are round to oval and
pathologist should look for papillary structures and epithe- may have nucleoli, but should lack significant atypia. The
lium with a small cleft separating them from the surrounding mitotic index is higher than that of borderline serous tumors
stroma as well as a stromal response to aid in the diagnosis but should be less than 10 per 10 high-power fields.
of invasion (Fig. 16.29). Grossly, high-grade serous carcinoma demonstrates areas
Low-grade serous carcinomas often merge morphologi- of solid growth with necrosis. There may be papillary or cys-
cally with (i.e., arise from) serous borderline tumors. Grossly, tic growth as well (Fig. 16.30). High-grade serous carcino-
mas demonstrate a wide array of morphologic patterns.
Papillae (Fig. 16.31), pseudo-endometrioid cribriforming
glands, solid growth, and transitional-type epithelium have
been demonstrated in these tumors. The most reliable histo-
a b
Fig. 16.28 (a) Stratified, tufted epithelium in a borderline tumor. (b) The epithelial proliferation in a borderline tumor will often lead to free float-
ing islands of relatively bland epithelium
558 H. Goyne et al.
16.8.3.4 Clear Cell Neoplasms *More in depth discussion of these entities can be found in
Clear cell carcinomas comprise about 5% to 12% of ovarian Chap. 9.
epithelial-stromal tumors [2]. Benign and borderline tumors
are exceptionally rare in this category. Approximately 40% 16.8.4.1 Fibroma/Thecoma
of clear cell carcinomas are confined to the ovary [2]. Fibromas, along with fibrothecomas, are the most common
Carcinomas often display areas of solid growth, but invari- sex cord stromal tumor, constituting 80% of this group [2].
ably demonstrate cystic areas as well. Hemorrhage and Fibromas are usually unilateral; however, they are also the
necrosis are often present. The histologic features of clear most common bilateral sex cord stromal tumor. Grossly, the
16 Principles and Practical Guidelines of Intraoperative Consultation 561
a b
Fig. 16.40 (a) Brenner tumors are firm, typically solid masses with a tan-white cut surface. (b) Occasionally, cystic areas or mucinous cystic
tumors may be present
a b
c d
Fig. 16.43 (a) Micro- and macrocystic growth in an adult-type granu- in an adult-type tumor. (d) High-power examination can often reveal
losa cell tumor. Note the presence of small follicles within the solid coffee bean-shaped nuclei with occasional nuclear grooves. While help-
components. (b) Trabecular, corded, and diffusely solid (right side) ful in some cases, this is not a diagnostic feature of adult-type granulosa
growth in an adult-type granulosa cell tumor. (c) Diffuse or solid growth cell tumors
16 Principles and Practical Guidelines of Intraoperative Consultation 563
a b
Fig. 16.44 (a) A Sertoli-Leydig cell tumor with a prominently cystic growth pattern. (b) Solid growth in a Sertoli-Leydig cell tumor
Fig. 16.45 A mature cystic teratoma with solid and cystic growth and hair
16.8.5.2 Mature Cystic Teratoma Fig. 16.46 Immature teratomas may appear as frankly malignant,
solid tumors, as seen in this example, or as a solid component in an
The mature cystic teratoma is the most common germ cell otherwise mature teratoma. Areas of fleshy, solid growth should always
tumor of the ovary. The classic appearance is a unilocular be preferentially sampled when evaluating these lesions
cyst filled with keratinaceous debris, oily brown to tan seba-
ceous material, and hair (Fig. 16.45). A raised protuberance,
known as Rokitansky’s nodule, may be present in the cyst 16.8.5.3 Dysgerminoma
composed of adipose tissue with overlying skin. Dysgerminoma is a rare, malignant germ cell tumor account-
Histologically, a wide variety of tissues may be present. The ing for approximately 1% of all ovarian germ cell tumors [2].
most common tissues include skin, skin appendages, adipose They are most common in the second and third decades of
tissue, respiratory epithelium, cerebral cortex, bone, carti- life. Of note, it is one of the most common neoplasms seen
lage, cerebellum, meninges, and peripheral nerve. Areas of during pregnancy. The tumors grow rapidly and may be
grossly apparent solid growth with necrosis may be seen in bilateral in a small number of cases. Grossly, the tumors are
immature teratoma (Fig. 16.46); because of this, when evalu- solid and firm with a uniform, tan-pink, cerebriform cut sur-
ating a mature teratoma, any and all areas that are fleshy, face (Fig. 16.47). Histologically, the tumor is composed of
solid, or necrotic should be sampled. cords and nests of uniform, polygonal cells, with ample
564 H. Goyne et al.
Infiltrative glands and single cells that encompass normal lesions, as to not miss small areas of solid growth. Sections
ovarian structures are suggestive of metastasis as well [2, 9]. should be taken from solid, soft, and/or fleshy areas. In
It is important to remember that the residual ovarian stroma lesions grossly suspicious for borderline tumor or carcinoma,
surrounding the metastasis may become luteinized, mimick- but microscopically benign, additional sections (up to 4–5)
ing a sex cord stromal neoplasm. Awareness of this phenom- should be taken. If a mucinous carcinoma is encountered, the
enon, especially in younger patients, can help to avoid next step should be to rule out metastasis. This is particularly
misinterpretation. true when cancer involves both ovaries, is smaller than 10 cm
(particularly <5 cm), and/or is associated with psuedomyx-
oma peritonei. It may be helpful to remember that primary
16.8.7 Diagnosis and Reporting ovarian mucinous carcinoma is a rare finding. In addition, it
is also important to remember that “benign”-appearing muci-
Age is an important factor to consider when encountering an nous epithelium does not rule out metastasis, as some metas-
ovarian neoplasm. The most common ovarian neoplasms tasis (such as pancreatic carcinoma) may show remarkably
overall, and among women younger than 50 years of age, are bland lining epithelium.
of germ cell origin. Epithelial neoplasms are most common in Benign mesenchymal processes, such as spindle cell
women older than 50 years of age. Overall, stromal neoplasms lesions, may be reported as “benign or low-grade spindle cell
are uncommon [10]. When considering the benign neoplasms, neoplasms.” If marked cellularity, necrosis, pleomorphism,
benign cystic teratoma is the most common overall and in or atypia is present, the pathologist may report “atypical or
women younger than 50. The second most common subset is malignant spindle cell lesion,” depending on the degree of
serous and the third most common mucinous [10]. concern.
The most common tumors of low malignant potential are
serous borderline tumors, mucinous borderline tumors, and
endometrioid borderline tumors, in that order. Of note, endo- 16.9 F
rozen Section Analysis of Peritoneal
metrioid borderline tumors are rare [10]. The most common Lesions
malignant tumors of the ovaries are serous carcinomas and
tend to increase in incidence with age. 16.9.1 Purpose of Frozen Section Evaluation
In cases where a diagnosis of carcinoma can be made, it is
best practice to report a subtype (e.g., serous, mucinous, The frozen section analysis of peritoneal masses is not
endometrioid, etc.) if possible. The grade of the tumor (i.e., uncommon in the setting of a gynecologic tumor. This is
low vs. high) should be reported as well, as it may prompt a attributable to serous carcinoma’s tendency to seed the peri-
more extensive staging procedure. In some cases there will toneal cavity causing widespread disease. In many surgeries
be ambiguity regarding the primary source of the tumor, in this setting, it is technically easy (and commonplace) to
especially in cases of mucinous carcinoma. In these cases, biopsy the omentum or a peritoneal implant in order to estab-
the possibility of metastasis should be conveyed to the sur- lish a diagnosis or extent of spread. That being said, there are
geon, triggering a search for other possible sources of tumor a number of processes that may involve the peritoneal cavity
in the abdomen. In general the number of sections of tumor and mimic serous carcinoma or other malignancies.
examined intraoperatively should be reported to convey the Knowledge of these entities is critical in constructing a
extent of evaluation to the operating room. If carcinoma is robust differential diagnosis.
uncertain, a descriptive diagnosis is preferable, and the
pathologist should defer to permanent sections.
In the case of borderline tumors, adding the phrase “at 16.9.2 Gross Specimen Processing
least borderline tumor” is appropriate given the heteroge-
neous nature of these lesions. Again the pathologist should Grossly, peritoneal specimens sent for intraoperative consul-
report how many sections were examined. About a fourth of tation are noncomplex. They are usually removed entirely or
these lesions will show areas of carcinoma on additional sent as small biopsies. Inking is generally not indicated as
sampling. Subtyping (i.e., serous, mucinous, endometrioid) margin status is inconsequential. Occasionally sectioning of
these lesions is appropriate when possible [2]. larger specimen, such as omentum, should be performed to
Ovarian mucinous lesions are well known for low diag- identify focal lesions. Representative sections, avoiding
nostic accuracy at the time of frozen section evaluation. A necrosis, of larger lesions should be submitted, while freez-
good rule of thumb is to open every cyst for large cystic solid ing the entirety of small samples is routine practice.
566 H. Goyne et al.
Fig. 16.53 Occasionally single cells and irregularly shaped glands Fig. 16.55 Metastasis from a signet ring cell tumor. A central pool of
may be present in desmoplastic implants; these are not indicative of mucin contains occasional, small histiocytoid cells, while larger aggre-
invasive disease gates of signet ring cells can be seen in the upper left and bottom right
corners of the image
Fig. 16.57 Reactive mesothelial cells seen growing in tufts and small Fig. 16.58 Mesothelioma composed of bland, eosinophilic cells grow-
papillary structures. Note that the lining of the structures is low cuboi- ing in papillae. This tumor is remarkably similar to low-grade serous
dal and that even at low magnification it is apparent that atypia is absent carcinoma, and discrimination between the two may not be possible on
frozen section
Reactive (papillary) mesothelial hyperplasia is a common 16.9.3.4 I nclusions, Cysts, and Other Reactive
histologic finding, which is rarely appreciated as a gross or Ectopic tissues
entity. It is usually composed of relatively bland, polygonal During the course of any abdominal surgery, a myriad of
cells lining the omental or peritoneal surface forming tufts or masses and cysts may be identified and sampled. These
papillae (Fig. 16.57). These lesions may occasionally dem- lesions are usually readily identifiable as benign, and in most
onstrate necrosis; however severe atypia should be absent. cases statement of such with a descriptive diagnosis will suf-
Most cases display mild atypia at most. Spindled growth fice. The following lesions and tissues are the most com-
may occur in some tumors, but, again, should lack nuclear monly encountered:
atypia. In most cases, an inciting event such as previous sur-
gery, inflammation, or endometriosis is the cause of these • Endometriosis—Endometriosis presents as a blue, brown, or
proliferations, although it may not be apparent in the sam- black discoloration or cyst. Histologic examination will
pled material. Low-grade, well-differentiated mesotheliomas reveal varying proportions of endometrial glands, stroma,
are typically small and incidentally identified at the time of and hemorrhage, which may be in the form of pigment-laden
surgery. These tumors share many histologic features with macrophages (Fig. 16.59). Occasionally, endometriosis may
their reactive counterparts, and their discrimination should become mass forming or polypoid, mimicking malignancy.
not be attempted during frozen section analysis. If a well- • Endosalpingiosis—deposits of tubal-type epithelium,
differentiated mesothelioma is suspected, the surgeon should which may be seen throughout the body. Endosalpingiosis
be altered to the possibility, and the tumor should be heavily ranges from small cysts to larger, partially cystic masses,
sectioned for histologic evaluation. which may contain papillary structures. The identification
High-grade mesothelioma is relatively uncommon, yet of otherwise unremarkable fallopian tube epithelium,
should be in the differential diagnosis of most low-grade composed of ciliated and secretory cells, is the major his-
(and some high-grade) serous carcinomas (Fig. 16.58). tologic hallmark (Fig. 16.60).
These tumors are often grossly large, bulky, and widespread, • Peritoneal inclusion cysts—small to large simple cysts
which may mimic high-grade serous carcinoma. In addition, lined by a bland mesothelial lining (Fig. 16.61).
they frequently display nuclear atypia, mitotic activity, peri- • Decidua—Any peritoneal surface may display decidual
toneal invasion, and necrosis. To further complicate evalua- changes similar to those seen in the endometrium
tion between these lesions, high-grade mesothelioma may (Fig. 16.62). These may present as tan/white plaques or
grow in papillary, solid, tubule-papillary, and sarcomatoid nodules, are usually associated with pregnancy, and, thus,
patterns. Suffice to say that in the vast majority of cases, the seen at the time of cesarean section.
differential diagnosis includes low- and high-grade serous • Granulomatous peritonitis—Keratin, infections, sarcoid,
carcinoma, including the primary peritoneal variant, and the gallstones, and foreign material may all lead to granu-
diagnosis should be based on permanent sections and use of loma formation in the peritoneum. Keratin granulomas
immunohistochemistry. are relatively common and are usually seen in the setting
16 Principles and Practical Guidelines of Intraoperative Consultation 569
Fig. 16.59 A serosal deposit of endometriosis with decidual change. Fig. 16.61 Low magnification view of a peritoneal inclusion cyst. The
When identified during cesarean section, the endometrium may show lining epithelium is low cuboidal and lacks atypia
this effect
Fig. 16.60 A small, cystic, peritoneal deposit of endosalpingiosis. Fig. 16.62 A carpet-like plaque of peritoneal deciduosis lining the
Note the presence of psammoma bodies, which is a common outside of the fallopian tube
occurrence
of mature cystic teratomas or severe inflammation with Grossly, it appears as red or brown nodules that may be
squamous metaplasia. Gallstone peritonitis may be seen mistaken for lymph nodes. This finding may be associated
following cholecystectomy, and foreign bodies may be with prior abdominal trauma.
identified in patients with extensive surgical histories and • Sclerosing mesenteritis—a reactive thickening of the
in cases of bowel perforation associated with diverticular mesenteric fat, which can mimic a malignant process.
disease or Crohn’s disease. Histologically, this lesion is composed of lobules of
• Peritoneal leiomyomatosis—diffuse involvement of the fat, separated by fibrous bands (Fig. 16.63). Fat necro-
peritoneal surfaces by innumerable, “benign” smooth sis, calcification, and inflammation are usually present
muscle tumors. This disease may primary, which is rare, and are a useful hint to the reactive nature of this
or secondary to prior morcellation of uterine tumors. disease.
• Gliomatosis—is the presence of mature glial tissue with • Fat necrosis/infarcted epiploica—Nodules of necrotic fat
small, round nuclei and fibrillary, eosinophilic cytoplasm may be identified by the surgeon and may mimic a malig-
growing on the peritoneal surfaces. This phenomenon nant process due to the firmness of the masses.
may be seen sporadically or associated with a teratoma. Histologically the presence of variably sized, necrotic fat
• Splenosis—the presence of histologically unremarkable, cells, chronic inflammation, and calcification can help to
splenic-type tissue located anywhere in the peritoneum. make the diagnosis.
570 H. Goyne et al.
17.1.1 Epidemiology
U. Krishnamurti · M. Mosunjac · K. Z. Hanley (*)
Emory University Hospital, Atlanta, GA, USA
As recently as the 1940s, cervical cancer was a major cause
e-mail: [email protected]
of death among women of childbearing age in the United
G. Deftereos
States. The American Cancer Society’s estimates for cervical
University of Utah School of Medicine, Salt Lake City, UT, USA
© Science Press & Springer Nature Singapore Pte Ltd. 2019 571
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3_17
572 U. Krishnamurti et al.
cancer in the United States for 2017 are that about 12,820 Cameron, the director of the American Cancer Society,
new cases of invasive cervical cancer will be diagnosed and sponsored the First National Cytology Conference in
that 4210 women will die of cervical cancer [1]. However, Boston, in 1948, to showcase the unmistakable success of
worldwide, cervical cancer incidence is over 500,000 cases the Papanicolaou method in detecting early cervical disease
annually with 275,000 deaths/year [2]. and funded the training of pathologists by George
About 79 million Americans are infected with HPV. Of Papanicolaou. It was Cameron who christened the test as the
19 million people who newly acquire sexually transmitted “Pap test” [8]. In 1954, Dr. Papanicolaou produced the Atlas
diseases each year, the majority (about 14 million people) of Exfoliative Cytology, which was truly a masterpiece [9].
become newly infected by HPV each year. Of these, 49% of The generalization of exfoliative cytology was also sup-
cases are in young people (15–24 years of age) [3]. Clinical ported by the National Cancer Institute and the US Public
manifestations of HPV infection include genital warts (con- Health Service. With a simple Pap test, cervical cancer inci-
dylomas), intraepithelial neoplasia, and carcinomas of the dence and death rates declined by more than 60% over the
lower anogenital tract, particularly uterine cervical squa- last 40 years [1, 10, 11]. Even with newer advances such as
mous cell carcinoma (SCC) [3]. HPV and other biomarker testing, cervical cytology is still
the most successful cancer prevention program! Even in
women who have been HPV vaccinated, it is a highly spe-
17.1.2 History of Pap Test cific screening regimen. Cervical cytology is the first-line
screening test in many situations, especially where resources
The Pap test was invented by and named after George are limited.
Papanicolaou, who started his research in 1923. He first
announced his method of examining vaginal smears, now
commonly known as the “Pap test,” at the 1928 Battle Creek 17.2 C
ollection Of Cervical Cytology
Conference in Michigan [4]. There is controversy whether Samples [12]
credit for the “Pap test” should be given to Papanicolaou or
to Aurel Babes. A year earlier in 1927, Aurel Babes and It is crucial to properly collect and submit cervical cytology
Constantin Daniel presented their findings from examination samples. One half to two thirds of false negatives are due to
of cellular material obtained from the uterine cervix on slides patient conditions present at the time of sample collection
that were air dried and stained with Giemsa stain to diagnose and operator-related issues [13, 14].
cervical cancer [5]. The method by Babes was radically dif- Samples for Pap test should be collected within approxi-
ferent from that by Papanicolaou. Dr. Papanicolaou should mately 2 weeks (10–18 days) after the first day of a woman’s
receive the credit for the use of exfoliative cytology, the wet last menstrual period. Within 48 h prior to the test, the woman
fixation, the staining technique, the systematic classification should not douche and use tampons, birth control foams, jel-
of cells with intermediate stages between the normal and the lies, vaginal creams, or vaginal medications. Also, she should
cancerous cell, and the envisioning that the method could be refrain from intercourse 48 h prior to the test [15].
applied to large numbers of women in the cancer-bearing The laboratory requisition must be legibly and accurately
period of life to detect cervical cancer in its early stages, i.e., filled out before obtaining the cellular sample.
the “Pap test” as is commonly employed [6]. Information in the laboratory requisition as required by
The Pap test was finally on the map only after a leading CLIA’88 [16] are:
article in the American Journal of Obstetrics and Gynecology
in 1941 by Dr. G. Papanicolaou and Dr. Herbert Traut, a • Name (any name change in the past 5 years should be
gynecologist [7]. A monograph entitled Diagnosis of Uterine noted)
Cancer by the Vaginal Smear that they published contained • Age and/or date of birth
drawings of the various cells seen in patients with no disease, • Menstrual status
inflammatory conditions, and preclinical and clinical carci- • Previous abnormal cervical cytology result
noma. The outstanding illustrations in the monograph were • Previous treatment, biopsy, or surgical procedure
by Hashima Murayama, who was later a staff illustrator with • Risk status for developing cervical cancer, e.g., “high risk”
the National Geographic Society. Both Papanicolaou and his • Source of specimen, e.g., cervical, vaginal
wife dedicated the rest of their lives to teaching the technique • Hormone/contraceptive use
to other physicians and laboratory personnel [8]. • Relevant clinical findings (abnormal bleeding, grossly
The purpose of the Pap test is to detect precancerous visible lesion, etc.)
abnormalities, which can be destroyed when detected (using
loop electrocautery excision procedure, cryotherapy, or sur- Laboratories should have a written procedure specifying
gical excision), thus preventing cancer. Dr. Charles the requirements for proper specimen identification. The
17 Gynecologic Cytology 573
specimen vial or glass slide must be labeled with a unique Table 17.1 Comparison of LBC systems
identifier which is usually the patient’s first and last names. SurePath ThinPrep
A second identifier such as date of birth, medical record Cell separation method from the Centrifugation Filtration
number, social security number, or collection date may be suspension
Cell deposition method on slide Sedimentation Controlled
used based on the laboratory policy.
pressure
A cervical cytology specimen is usually collected with Circular diameter for microscopy 13 mm 20 mm
the patient in the dorsolithotomy position. A sterile or single-
use bivalve speculum of appropriate size is inserted into the
vagina without lubrication. The position of the speculum then the collection device is discarded. The cap is tightened
should allow for complete visualization of the cervical os so that the torque line on the cap passes the torque line on
and ectocervix. The endocervix, transformation zone, and the vial.
ectocervix should all be sampled [17]. The currently avail-
able collection devices include endocervical brushes, 17.2.1.2 Collection for SurePath [22]
wooden and plastic spatulas, and plastic “broom-type” For SurePath cytology, the sample is collected using either a
samplers. broom-like device or combination brush/spatula with detach-
able heads. The detachable head is dropped into the BD
SurePath™ vial and then the cap tightened and transported
17.2.1 Liquid-Based Cytology to the laboratory.
While both SurePath and ThinPrep systems are automated
Liquid-based cytology (LBC) is an alternative to conven- and result in a well-preserved approximate monolayer of
tional Pap smears. The US FDA approved the ThinPrep cells, with a background devoid of blood and mucus, there
(Hologic, Marlborough, MA) in 1996 and SurePath (previ- are some differences between the two liquid-based systems
ously known as Auto-Cyte Prep) in 1999 (BD TriPath, (Table 17.1).
Burlington, NC). Both liquid-based cytology and conven-
tional smears are acceptable for cytologic screening [18, 19]. 17.2.1.3 Advantages of LBC
In the United States, in 2006–2007, liquid-based cytology • Immediate fixation with enhanced nuclear and cytoplas-
was used in approximately 75% of Pap tests, and by 2014 it mic details.
was estimated to be about 93% of Pap tests. • Less screening time than a conventional smear.
Liquid-based methods require the use of collection • Additional samples can be prepared as needed.
devices that have been approved by the FDA for use with the • Cleaner background.
particular specimen preparation instrument, and manufactur- • Reduced unsatisfactory rate.
ers’ directions must be followed [20]. • Suitability for other tests, e.g., HPV testing,
chlamydia/gonorrhea, and trichomonas testing.
17.2.1.1 Collection for ThinPrep [21] • Suitability for automated analysis.
For ThinPrep cytology, the broom or brush/spatula can be
used for collection. For collection using the brooms, the cen- 17.2.1.4 Disadvantages of LBC
tral bristles of the broom are inserted into the endocervical • Abnormal cells are dispersed.
canal deep enough to allow the shorter bristles to fully con- • Epithelial cells especially endocervical cells and imma-
tact the ectocervix. The broom is gently pushed and rotated ture metaplastic cells appear smaller than in conventional
in a clockwise direction five times. With the “broom-like” smears.
device, the ectocervical and endocervical specimens are col- • Scanty LBC preparations can be difficult to screen and
lected simultaneously. interpret.
For collection using the brush/spatula, adequate sampling • More expensive than conventional test.
from the ectocervix is obtained by rotating the plastic spatula • Background material such as blood, mucus, or tumor dia-
360 degrees around the entire exocervix while maintaining thesis when present appears different.
tight contact with exocervical surface. After that an endocer-
vical brush device is inserted into the cervix until only the
bottommost fibers are exposed. This is slowly rotated 1/4 or 17.2.2 Conventional Cervical Cytology
1/2 turn in one direction.
For both collection methods, the collection device (broom Air-drying greatly limits specimen evaluation by obscuring
or spatula and brush) is rinsed as quickly as possible into the cellular details. Therefore, for conventional cervical cytology
PreservCyt solution by rotating it in the vial ten times and specimens, to prevent air-drying, it is important to immedi-
then swirled vigorously to further release the material, and ately fix the sample. This can be achieved by either immersing
574 U. Krishnamurti et al.
the slide in alcohol or by using spray cytology fixatives. While ogy terminology (LAST), and the increasing use of HPV
applying, spray fixatives should be 6–10 in. (15–25 cm) from vaccinations. All these were justifications for the 2014
the glass slide. Specimens fixed in alcohol can be transported Bethesda System. There was no consensus workshop held
to the laboratory in the alcohol container itself or transported for the 2014 Bethesda System update. Following literature
in an air-dried manner after 20–30 min of fixation. review, the proposed content for the atlas was conceived by a
select group of cytopathologists, clinicians, and epidemiolo-
gists. Based on the feedback received from individuals in 59
17.3 C
ervical Cytology Reporting System countries worldwide on the proposed updates, the positions
and Its Progresses and contents were fine-tuned and then included into the 2014
Bethesda System and the most recent atlas. The 3rd edition
In the first few decades of cervical cytology, there was poor of the atlas published in 2015 has a new chapter on a risk
reproducibility and lack of standardization of laboratory assessment-based management [25].
reports of Papanicolaou smears [23]. The Bethesda System In addition to standardizing the reporting of cervical
(TBS) is a standardized system widely used for reporting cytology worldwide, TBS has advanced our understanding
Pap smear results. The goal of the first Bethesda workshop, of HPV biology and provided the framework necessary for
which was convened in December 1988 and chaired by Dr. the development of systematic and evidence-based cervical
Robert Kurman, was to find a better way of reporting cervi- cancer screening and management guidelines [25].
cal cytology. It is named The Bethesda System (TBS) TBS has clear criteria for specimen adequacy, and a
because the first conference where the system was devised “statement of adequacy” is an essential part of the report
was held in Bethesda, Maryland. serving as an important quality assurance element. The ter-
TBS terminology has the following fundamental guiding minology is based on the underlying pathobiology of the
principles: morphologic changes of cervical epithelial abnormalities.
TBS has standard reporting terms and criteria for each inter-
• Terminology used by the laboratory must communicate pretive category. Squamous intraepithelial lesion (SIL) is
appropriate and relevant clinical relevant information to divided into low- and high-grade lesions. The two-tier termi-
the clinician. nology reflects the different biologic states between low-
• Terminology should be consistent across different labora- grade lesion (representing productive HPV infections and
tories but also flexible so that it can be modified to suit the self-limiting state) and high-grade lesion (representing a pre-
needs of various laboratories and diverse geographic cancer state).
settings.
• Terminology should be continuously updated to repro-
duce the most up-to-date understanding of the tumor biol- 17.4 E
valuation and Reporting of Cervical
ogy of cervical neoplasia and incorporate advances in Cytology Specimens
laboratory practice. Additional workshops were convened
in 1991 and 2001. The Pap smear/cervical cytology report should have the
following main headings (Table 17.2):
There have been three editions of the Cervical Cytology
Bethesda System atlas that encompass the definitions and Table 17.2 The 2014 Bethesda System for reporting cervical
cytologic criteria of cervical cytology interpretations along cytology
with explanatory notes and images. It was in 1994 and 2004 Specimen type
that the first two editions of TBS atlas were published [23, Indicate conventional smear (Pap smear) vs. liquid-based
24]. The workshop in 2001 utilized the Internet, providing preparation vs. other
everyone an opportunity for input. This resulted in more than Specimen adequacy
2000 comments and more than 400 participants from over • Satisfactory for evaluation (describe presence or absence of
endocervical/transformation zone component and any other
two dozen countries in the actual meeting [24]. With the quality indicators, e.g., partially obscuring blood, inflammation,
increasing use of liquid-based cytology by many laborato- etc.)
ries, images and criteria specific to liquid-based cytology • Unsatisfactory for evaluation (specify reason)
were included in the 2004 atlas. Specimen rejected/not processed (specify reason)
Specimen processed and examined, but unsatisfactory for
Between 2001 Bethesda System and 2014 Bethesda evaluation of epithelial abnormality because of (specify reason)
System, there had been many changes in the practice of cer- General categorization (optional)
vical cytology and pathology, including the primary HPV • Negative for intraepithelial lesion or malignancy
screening with Pap as triage, the new screening and manage- • Other: See Interpretation/Result (e.g., endometrial cells in a
ment guideline (ASCCP 2012), the changes in histopathol- woman ≥45 years of age)
17 Gynecologic Cytology 575
a b
Fig. 17.2 Pap smear unsatisfactory for evaluation. (a) Unsatisfactory due to scant cellularity (SurePath, Papanicolaou stain, 100×). (b)
Unsatisfactory for evaluation due to scant cellularity with predominantly acute inflammation SurePath, Papanicolaou stain, 200×)
Table 17.2 presents the recommended Bethesda System This section should provide a brief description of the test
terminology for reporting the findings. method(s) and report the result in a manner that can be easily
The main interpretation categories for adequate speci- understood by the clinician.
mens are:
• Negative for intraepithelial lesion or malignancy (NILM) 17.4.6 Automated Review [32]
• Epithelial cell abnormality, which can be further specified
as squamous cell or glandular cell If a case is examined by an automated device, such as by
• Other (endometrial cells in a woman ≥45 years of age computer assistance, the device used should be specified.
with specification if the specimen is “negative for squa- Irrespective of the result, the report should state if the spec-
mous intraepithelial lesion”) imen was processed successfully or if there was review of
• Other malignant neoplasms (specify) only the device identified fields of view or a full manual
screening. If only device identified fields of view were
The negative for intraepithelial lesion or malignancy cat- reviewed, the interpretation as well as adequacy informa-
egory reports nonneoplastic findings and organisms that may tion obtained from the computer assessment must be
be present. reported. Even if manual screening/review is not per-
The squamous epithelial cell abnormalities include atypi- formed, a laboratorian with appropriate training and autho-
cal squamous cells undetermined significance (ASC-US), rization must review and verify the results generated by the
low-grade squamous intraepithelial (LSIL), atypical instrument.
578 U. Krishnamurti et al.
17.4.7 Educational Notes and Comments The nucleus is eccentrically placed with a finely granular
Appended to Cytology Reports chromatin. Normal endocervical cells have inconspicuous
(Optional) nucleoli, but reactive endocervical cells can have prominent
nucleoli (Fig. 17.5a, b).
The suggestions should be concise and in keeping with clini-
cal follow-up guidelines published by professional organiza- 17.5.2.2 Endometrial Cells
tions. References to relevant publications may be included. These can be seen in younger women, and benign endome-
The 2012 consensus guidelines of the American Society trial cells are not reportable in women less than 45 years of
for Colposcopy and Cervical Pathology (ASCCP) are fol- age. They are usually identified in the first 12 days of the
lowed for managing patients with abnormal cervical cancer menstrual cycle. They appear as small dark hyperchromatic
screening tests and cancer precursors [26]. cells arranged in balls (Fig. 17.6a). Sometimes, endome-
trial cells show a dual cell population of larger glandular
cells in the periphery and small, dark, stromal cells in the
17.5 Normal Cellular Elements center (Fig. 17.6b). Endometrial cells may be seen out of
a b
Fig. 17.4 Parabasal cells. (a) Parabasal cell in a normal Pap smear (SurePath, Papanicolaou stain, 100×). (b) Predominance of parabasal cells in
atrophy (SurePath, Papanicolaou stain, 600×)
a b
Fig. 17.5 Endocervical cells. (a) Normal endocervical cells in a honeycomb pattern; (SurePath, Papanicolaou stain, 100×). (b) Singly dispersed
endocervical cells with a basally placed nucleus and intracytoplasmic mucin (SurePath, Papanicolaou stain, 100×)
a b
Fig. 17.6 Endometrial cells. (a) Endometrial cells as small dark hyperchromatic cells arranged in balls. (b) Endometrial cells as a dual cell popu-
lation of larger glandular cells in the periphery and small, dark, stromal cells in the center (exodus) (SurePath, Papanicolaou stain, 600×)
580 U. Krishnamurti et al.
phase, in women with endometritis, intrauterine devices, or and fragments of squamous cytoplasm. Excessive cytolysis
endometrial polyps [33]. Endometrial cells are in the dif- may render a Pap smear difficult to evaluate.
ferentials of hyperchromatic crowded groups that also
include lower uterine segment cells, HSIL, squamous cell 17.5.3.3 Trophoblastic Cells and Decidual Cells
carcinoma, AIS, small-cell carcinoma, and endometrial Trophoblastic cells are seen very rarely, in about 0.1% of Pap
carcinoma. smears from pregnant women. Syncytiotrophoblastic cells are
large with abundant cytoplasm, multiple nuclei with slightly
irregular nuclear contours, and a granular chromatin pattern.
17.5.2.3 Lower Uterine Segment Syncytiotrophoblasts may sometimes show a tail that represents
Sometimes lower uterine segment (LUS) or endometrium where the cells separate from the placenta (Fig. 17.9). The pres-
may be sampled, particularly if there is a shortened endocer- ence of these cells does not reliably indicate an imminent abor-
vical canal as in cases of women post-conization. Lower tion [34]. Decidual cells are isolated cells with degenerative
uterine segment is seen as large fragments of glands and changes. Decidual cells have abundant granular cytoplasm, a
stroma. Compared to endocervical cells, LUS cells are more large vesicular nucleus, and a prominent nucleolus.
hyperchromatic, show a higher nuclear to cytoplasmic ratio,
and more likely show mitoses.
17.5.3.2 Lactobacilli
These are gram-positive rod-shaped bacilli (Fig. 17.8). They
Fig. 17.8 Lactobacilli. Note the numerous rod-shaped bacilli causing
produce lactic acid and reduce the pH and are believed to cytolysis with fragments of squamous cytoplasm. Also present are
possibly offer protection from infection by pathogens such numerous neutrophils (SurePath, Papanicolaou stain, 600×)
as Candida. Lactobacilli can cause cytolysis, resulting in a
smear with numerous bacilli, bare intermediate cell nuclei,
Table 17.5 Three subtypes of protective response of cells to local or systemic injury
Squamous metaplasia Keratotic change Tubal metaplasia
Definition Metaplastic process in which glandular Protective mechanism usually in Metaplastic process in which normal
endocervical epithelium is replaced by association with HPV endocervical epithelium is replaced by
squamous type of cells Includes previously used terms epithelium that resembles cells of the
Immature metaplasia is a transitional • Keratosis normal fallopian tube (Fig. 17.12)
developmental stage in which cells keep • Parakeratosis
characteristics of both endocervical and • Dyskeratosis
squamous epithelium (Fig. 17.10a, b) • Hyperkeratosis
Mature metaplastic cells are
morphologically indistinguishable from
squamous cells
Criteria Immature metaplasia Miniature superficial squamous Ciliated columnar cells in groups or singly
Sheets and single cells cells with orangeophilic High N/C ratio nuclei might be enlarged
Higher N/C ratio cytoplasm and small round dark pleomorphic and hyperchromatic
Oval nuclei with smooth borders nucleus (Fig. 17.11) Cytoplasm may be vacuolated
Finely granular chromatin Hyperkeratosis
No nucleoli Anucleated mature squamous
Dense cyanophilic cytoplasm with sharp cells
borders “Ghost cells without nuclei”
“Spider cells” (Fig. 17.10b, c)
Differential LSIL ASC-US AGUS
diagnosis HSIL Adenocarcinoma
ASC-US
582 U. Krishnamurti et al.
a b
Fig. 17.10 (SurePath, Papanicolaou stain). (a) Mature metaplastic with extended cytoplasmic processes called “spider cells” resulted from
cells with dense cytoplasm and well-defined cell borders (400×). (b) mechanical force used to extract cells. N/C ratio increased, nuclear
Cluster of immature metaplastic cells with both features of columnar membranes smooth, chromatin powdery with no nucleoli (600×)
endocervical cells and squamous cells (400×). (c) Metaplastic cells
Fig. 17.11 (SurePath, Papanicolaou stain). Cluster of anucleated squa- Fig. 17.12 Tubal metaplasia (SurePath, Papanicolaou stain). Tubal
mous cells (hyperkeratosis) admixed with some cells that still contain metaplasia. Endocervical cells with prominent terminal bar and cilia. N/C
small pyknotic and hyperchromatic nuclei (parakeratosis) (400×) ratio slightly increased, nuclei round to oval with fine chromatin (600×)
17 Gynecologic Cytology 583
Table 17.6 Definition, morphological features, and differential diagnosis of atrophy and pregnancy-related changes in the cervical smear
Atrophy Pregnancy-associated change
Definition Thinned squamous epithelium is a result of decreased Alterations in squamous epithelium due to
hormonal stimulation in postmenopausal women pregnancy-induced hormonal changes. Relative low
level of estrogen and high level of progesterone lead
to incomplete maturation of the squamous cells
resulting in an intermediate cell—dominant pattern
Criteria Flat monolayered sheets with parabasal-like cells with no Intermediate-like cells with boat-shaped pattern
nuclear overlap (Fig. 17.13a) (navicular cells) are frequently seen (Fig. 17.14a)
Slight increase in N/C ratio (Fig. 17.13b) Basophilic to clear cytoplasm rich in glycogen
Chromatin evenly dispersed Vesicular nuclei with delicate chromatin pattern
Nuclear membranes regular
Sight anisonucleosis (Fig. 17.13b, d)
“Blue blobs” refer to globular collections of basophilic
amorphous material and crushed overlapping parabasal
cells (Fig. 17.13c)
Degenerated small parakeratotic-like cells and histiocytes
may be present
Background Inflammatory exudate
Clear
Differential diagnosis HSIL HPV changes (Fig. 17.14b)
ASC-US
LSIL
a b
c d
Fig. 17.13 Atrophy (SurePath, Papanicolaou stain). (a) Low power of absence of nucleoli, and smooth nuclear borders and no overlap (400×).
atrophic smear showing smaller single and parabasal cells with cyano- (c) Tighter cluster of atrophic cells, periphery of the cluster, shows typi-
philic cytoplasm and slightly enlarged nuclei (100×). (b) High power cal features as seen in picture b (400×). (d) Slight anisonucleosis often
showing parabasal cells with increased N/C ratio, fine chromatin, seen in atrophic smear and should not be interpreted as ASC-US (400×)
584 U. Krishnamurti et al.
a b
Fig. 17.14 Pregnancy changes (SurePath, Papanicolaou stain). (a) change in which there is perinuclear clearing with condensation and
Smear composed mostly of intermediate cells showing peripheral fold- folding-like features of the cytoplasm at the periphery. However,
ing of cytoplasm giving cells boat-like features that are known as navic- nuclear features are typical of LSIL (600×)
ular cells (400×). (b) Contrast those to HPV changes with koilocytic
Table 17.7 The most common subtypes of benign reactive cellular changes
Inflammation and repair (Fig. 17.15a–e) Radiation IUD
Cytoplasmic Cytoplasmic vacuolization Cytoplasmic Vacuolization
changes Polychromasia vacuolization Glandular cells present singly or in small
Perinuclear halos Polychromasia clusters
Increased N/C ratio Large bizarre cells Small dark cells with scant cytoplasm
“Spiderlike” metaplastic reactive cells Normal N/C ratio Increased N/C ratio
Cohesive sheets in the “school of fish” pattern Fig. 17.15f
Cytoplasmic boundaries well defined
Nuclear changes Binucleation Multinucleation Wrinkled, degenerated chromatin
Prominent single or multiple nucleoli Prominent nucleoli Prominent nucleoli
Enlargement
Non-overlapping
Smooth chromatin with mild hyperchromasia
Smooth nuclear borders
Background Inflammation may be seen Calcifications
Actinomyces
Differential Nonkeratinizing SCC Herpes cytopathic Adenocarcinoma
diagnosis Endocervical adenocarcinoma effect HSIL
LSIL
Recurrent carcinoma
smears [37]. When e xuberant, such changes pose a diag- 17.6.1.4 G landular Cell Status Post
nostic challenge and can mimic premalignant and malig- Hysterectomy
nant conditions (Table 17.7). Benign-appearing glandular cells can be seen in approximately
Lymphocytic (follicular) cervicitis is a subtype of chronic 2% of vaginal Pap smears in women after hysterectomy. The
cervicitis characterized by the presence of mature lymphoid findings are more common in women with the history of radia-
follicles. In addition to the above inflammatory reactive tion therapy compared to women with hysterectomy alone. It is
changes, numerous t lymphoid cells with tangible body mac- postulated that those glandular cells r epresent therapy-induced
rophages are seen in those Pap smears [38]. metaplasia of squamous epithelium [39].
17 Gynecologic Cytology 585
a b
c d
e f
Fig. 17.15 Reactive changes/repair (SurePath, Papanicolaou stain). (600×). (d) Compare with atypical repair showing a cluster with some
(a) Cluster of intermediate cells mixed with neutrophils showing anisonucleosis, nucleoli, and coarser chromatin (400×). (e) Tight
enlarged nuclei with nucleoli. However, nuclear boarders are smooth, inflammatory perinuclear white halos present in this reactive cluster of
and cells show no significant overlap or anisonucleosis (400×). (b) intermediate cells. Multinucleated endocervical cell. Cluster of colum-
Reactive metaplastic cells showing increased N/C ratio and “school of nar cells, one with two nuclei, enlarged nuclei with coarser chromatin,
fish” appearance (600×). (c) Changes consistent with repair. Cells with smooth nuclear borders, and non-overlapping cells (600×). (f) Reactive
enlarged nuclei but very prominent nucleoli and smooth nuclear board- glandular cells associated with IUD. Cells show nuclear enlargement,
ers. Neutrophils are admixed with cells and cells are non-overlapping nucleoli and cytoplasmic vacuolization (400×)
586 U. Krishnamurti et al.
a b
c d
Fig. 17.16 Specific organisms (SurePath, Papanicolaou stain). (a) Bacterial vaginosis: coccobacilli adhered to the surface of superficial
Candida: clustered superficial and intermediate cells that appear to be cell in relatively clean background. Features are suggestive of bacterial
speared by pseudo-hyphal elements of Candida giving it appearance of vaginosis and shift in vaginal flora (400×). (d) Actinomyces: cluster of
kebab (600×). (b) Trichomonas: pear-shaped organism presents bacteria with thin filaments at periphery (600×). (e) Herpes; multinucle-
between superficial cells showing small nucleus and eosinophilic gran- ated cell with intranuclear eosinophilic inclusions consistent with
ules. Some organisms are adhered to the cytoplasm of the cells and are “Cowdry B-type” inclusions (400×)
recognized by presence of cytoplasmic red granules (600×). (c)
588 U. Krishnamurti et al.
17.7 A
typical Squamous Cells tive of HSIL. Within ASC, greater than 90% of cases should
and Squamous Intraepithelial Lesions be ASC-US and less than 10% should be ASC-H. Compared
to ASC-US, ASC-H has a higher predictive value for the
17.7.1 Atypical Squamous Cells diagnosis of HSIL (47, 48). Positivity for HR-HPV in ASC-H
is as high as that in HSIL cytology [45].
When we talk about atypical cells in cytology in general, we It should be emphasized that cytology interpretation is
usually refer to a set of morphological changes that do not fit based on the entire specimen instead of single cells.
into clearly defined categories. It is important to define the
“atypical” category as precisely as we can to avoid the over- 17.7.1.1 A typical Squamous Cell
use of the term to avoid potential overtreatment or treatment of Undetermined Significance
delay. (ASC-US)
Historically the term “atypia in squamous epithelium” Squamous cell changes that are suspicious for but quantita-
encompassed a wide spectrum of changes including “inflam- tively or qualitatively do not fulfill the criteria for LSIL are
matory atypia, “reactive atypia,” “suspicious findings,” or classified as ASC-US. The ASC-US interpretation is influ-
findings of uncertain nature. In the 1988 and 1991 Bethesda enced by the quantity of the atypical squamous cells, the
Systems, the term atypical cells of undetermined signifi- degree of abnormalities, specimen preservation artifacts, and
cance (ASC-US) was used to define cellular changes that are clinical settings [46]. Cellular changes close to LSIL and
more marked than can be attributable to reactive changes, but atypical parakeratosis represent the most common situations
quantitatively or qualitatively fall short of a definitive diag- when an ASC-US interpretation is made (Table 17.9).
nosis of squamous intraepithelial lesion (SIL) [23, 41]. In the In reality, ASC-US lesion does not exist. The interpreta-
2001 Bethesda System, ASC was created to describe squa- tion of ASC-US is the result of the light microscopy capabil-
mous cell changes that suggest SIL but is qualitatively or ity and the pathologist’s interpretation ability. The ASC-US
quantitatively insufficient for a definitive interpretation of category represents the need for interpretation comfort zone
such [42]. for pathologists. When pathologists are afraid of missing
Although ASC is defined as cytologic changes suggestive SIL, the interpretation of ASC-US will increase. The fre-
of SIL, the ASC category usually can be associated with a quency of ASC-US interpretation can be reduced when the
wide spectrum of changes, including artifact, atrophy, pathologists’ experience increases.
inflammation, LSIL, HSIL, and even cancers. HSIL is identi- Morphological nuclear features of ASC-US include varia-
fied in cervical biopsies in about 10–20% of patients with an tion in nuclear size with nuclear enlargement no more than
ASC-US interpretation. While ASC diagnoses should not be three times the intermediate cell nucleus, mild hyperchroma-
overused, there is no ideal rate of ASC. However, it has been sia, and limited nuclear membrane irregularities. Most com-
suggested by expert panels that the rate should be kept to less mon cytoplasmic changes are dense orangeophilic cytoplasm
than 5% of all Pap cases. In a high-risk population, the ASC/ and koilocyte-like changes. Not all of those changes have to
SIL ratio should not exceed 3:1 [43, 44]. be present in a single Pap smear. One or more of those mor-
Due to its equivocal nature, some authors suggest remove phological features is sufficient for diagnosis of ASC-US
the ASC category. However, elimination of ASC category (Table 17.9).
would decrease the sensitivity of Pap test. Approximately
40–50% patients with an interpretation of ASC are infected
with the high-risk HPV. As the most frequent Pap test inter- Table 17.9 Definition and cytomorphologic criteria for ASC-US
pretation category, ASC is also the most common interpreta- ASC-US (Fig. 17.17a–e)
tion prior to a biopsy diagnosis of HSIL. There are several Definition Changes suggestive of but do not
fulfill the criteria of LSIL
reasons to keep the ASC category: (1) we need an interpreta-
Criteria 2.5–3× nuclear enlargement Fig. 17.17a
tion category to covey the equivocal cellular changes, (2) compared to intermediate cells
there is no clear-cut of cellular changes between interpreta- Variation in nuclear size Fig. 17.17a, b
tion categories, (3) ASC is not a new concept and has been Mild hyperchromasia without Fig. 17.17b
widely used, and (4) gynecologists are familiar with the granularity
interpretation of ASC-US. Limited irregularities of nuclear Fig. 17.17c
membranes
ASC is further separated into atypical squamous cells of
Vacuoles resembling koilocytes
undetermined significance (ASC-US) and atypical squamous but not well defined
cells-cannot exclude high-grade squamous intraepithelial Dense orangeophilic cytoplasm, Fig. 17.17e
lesion (ASC-H) [25]. This separation is necessary as most parakeratotic cells
ASC cases show changes more suggestive of LSIL, and a Presence of some but not all Fig. 17.17d
small proportion of ASC cases display features more sugges- HPV-related changes
17 Gynecologic Cytology 589
a b
c d
Fig. 17.17 ASC-US (SurePath, Papanicolaou stain). (a) Single atypi- (d) Three atypical cells. Top one with enlarged, hyperchromatic, and
cal metaplastic cell with nucleus 2–3× of adjacent intermediate cell irregular nucleus. Middle one with binucleation and bottom one with
showing mild hyperchromasia, nuclear irregularity, and perinuclear increased N/C ratio but lacking other features such as hyperchromasia
clearing suggestive of HPV changes (600×). (b) Single atypical inter- and nuclear irregularities. Taken together features are suggestive but not
mediate cell with enlarged wrinkled and hyperchromatic nucleus quantitatively or qualitatively sufficient for the diagnosis of LSIL
(600×). (c) A group of superficial and intermediate cells showing mild (600×). (e) Atypical parakeratotic cells showing slight enlargement and
nuclear enlargement and nuclear membrane irregularities with coarser hyperchromasia as well as some anisonucleosis. Such findings are best
chromatin. Features are suggestive but not diagnostic of LSIL (400×). categorized as ASC-US as they do not fulfill all criteria for LSIL (400×)
590 U. Krishnamurti et al.
a b
c d
Fig. 17.18 ASC-H (SurePath, Papanicolaou stain). (a) Isolated single with increased N/C ratio and scant cytoplasm. Features of HSIL are
cell that appears significantly smaller than surrounding cells, with high present; however quantitatively only those two cells were identified as
N/C ratio and small rim of dense cytoplasm, nuclear coarse chromatin, atypical, and therefore diagnosis of ASC-H is best suited (400×). (d)
and nuclear membrane irregular. Findings are suggestive for HSIL Crowded sheet pattern of ASC-H showing cohesive overlapping cells
however quantitatively insufficient for definitive diagnosis (400×). (b) with high N/C ratio, slightly hyperchromatic, and some with irregular
Single cluster of cells with high N/C ratio, wispy scant cytoplasm, and boarders. Such clusters can be seen in atrophy or reactive changes.
anisonucleosis with one prominent large nucleus with coarser chroma- Follow-up biopsy on this 47-year-old women showed HSIL (600×)
tin but relatively smooth nuclear boarders (600×). (c) Two small cells
philic cytoplasm, atypical parakeratotic cells are arranged in rent understanding of HPV infection biology. While LSIL
three-dimensional clusters with larger hyperchromatic and represents HPV production stage and is usually due to a tran-
irregular nuclei with increased N/C ratio [52]. However, the sient infection, HSIL represents a precancerous lesion. The
atypia and the N/C ratio fall short of LSIL due to lack of major goal of cervical cancer screening is to detect and treat
definitive features of HPV-associated changes. In such cases, HSIL. This two-tiered system has been widely adopted. In
it is appropriate to interpret the changes as ASC-US if the 2012, the Lower Anogenital Squamous Terminology (LAST)
changes are suggestive of LSIL and ASC-H and if the Standardization Consensus Conference also recommended a
changes suggest HSIL [53] (Fig. 17.17e). two-tiered histopathologic nomenclature with a single set of
diagnostic term for all HPV-associated preinvasive squa-
mous lesions of the lower anogenital tract regardless of
17.7.2 Intraepithelial Lesion anatomic site and regardless of gender. This histopathology
nomenclature mirrors the Bethesda cytology reporting for
The Bethesda System uses the two-tiered system for report- SIL [54]. This two-tiered histopathologic nomenclature in
ing cervical cytology: LSIL and HISL. This reflects our cur- the anogenital tract may be further qualified with appropriate
592 U. Krishnamurti et al.
Table 17.11 Morphological differences of typical and atypical repair Table 17.12 Morphological features of LSIL
Morphological features of typical and atypical repair Morphological features of LSIL
Atypical repair Usually in superficial Large size Fig. 17.19a, b
(Fig. 17.15d) Typical repair (Fig. 17.15c) or intermediate Mature with well-defined
Cell Sheets with crowding Cohesive flat sheets with squamous cells cytoplasm
arrangement and overlapping no overlapping and (Fig. 17.20a–e) Nuclei > than 3× that of Fig. 17.19a, b
crowding normal intermediate cells
Discohesion with No single cells Slightly increased N/C
single cells ratio
Steaming Loss of nuclear Nuclear streaming Hyperchromatic or Fig. 17.19a, b
streaming and “School of fish” normochromatic nuclei
polarity arrangement Keratotic cells with Fig. 17.19c
Pleomorphism Marked Mild to moderate enlarged nuclei
pleomorphism of pleomorphism of nuclei Anisonucleosis
nuclei No nucleoli
Nuclei Large nucleoli Large nucleoli HPV-related changes Sharply delineated large Fig. 17.19d
Coarse chromatin Pale evenly distributed irregular perinuclear
chromatin clearing (koilocytic
Mitosis Mitoses Mitosis change) associated with
the enlarged and/or
binucleated nucleus
–IN terminology. –IN refers to the generic intraepithelial
neoplasia terminology, without specifying the location. For a
specific location, the appropriate term should be used. For a From time to time, pathologists may encounter cases
–IN3 lesion, for example, CIN3 is used for cervical lesion, showing features of LSIL and also a few cells suggestive of
AIN3 for anal lesion, VIN3 for vulvar lesion, PeIN3 for HSIL. Since 2001, there have been literatures suggesting the
penile lesion, VaIN3 for vaginal lesion, and PAIN3 for peri- use of an intermediate category (LSIL-cannot exclude HSIL,
anal lesion. LSIL-H) to handle cases with such changes. The risk of
HSIL found in these cases was between typical LSIL and
17.7.2.1 L ow-Grade Squamous Intraepithelial typical HSIL. Subsequently, the issue was submitted to
Lesion (LSIL) online open discussions, and a decision was made to limit
LSIL includes koilocytes and CIN1. LSIL is most commonly terminology to LSIL and HSIL. There are several reasons for
seen in intermediate or superficial cells. The classical fea- such decisions: (1) LSIL-H is an intermediate category, (2)
tures of LSIL were described by Reagan and Hamonic in the management guideline applies to LSIL and HSIL, (3)
1956 and are included in Table 17.12 [55]. disease biology does not support the intermediate category,
Although perinuclear halo is frequently seen in LSIL, the (4) LSIL-H has poor reproducibility, (5) LSIL-H may be
diagnosis of LSIL does not require the presence of a perinu- overused, and (6) the use of LSIL-H can cause confusion to
clear halo. Also, an interpretation of LSIL cannot be made in clinicians. The 2014 Bethesda System recommends the
cells showing only perinuclear halo without the typical interpretation of both LSIL and ASC-H when a smear shows
nuclear features of LSIL. Nuclear changes are the basis of definitive LSIL and a few cells suggestive of HSIL. There are
LSIL. also other options: (1) report LSIL and comment the possi-
bility of HSIL, and (2) report ungradable SIL and specify
17.7.2.2 H igh-Grade Squamous Intraepithelial reason for such in the comment. These should represent a
Lesion (HSIL) small proportion of cases as most cases should be separated
HSIL is a precancerous lesion and includes CIN2 and CIN3. into either LSIL or HSIL. In general, the management of
In high-grade lesions, the overall size of the cells is reduced LSIL and ASC-H is the same. In younger women, however,
compared to LSIL, but nuclear to cytoplasmic ratio is greatly the addition of ASC-H in the pathology report will result in
increased due to minimal cell differentiation. Two major pat- colposcopy.
terns of HSIL can be seen: individual small cells with high
N/C ratio (Fig. 17.20a, b) and syncytial clusters of dysplastic Differential Diagnosis and Problematic
cells (Fig. 17.20c). Syncytial hyperchromatic groups are Patterns of HSIL
more prominent and are easier to spot, while single dysplas- There are mimickers of HSIL. Different patterns in HSIL
tic cells can be overlooked during routine screening due to may generate a different set of differential diagnosis. For
their small size. In both patterns, nuclei are hyperchromatic example, HSIL in small hyperchromatic crowded groups
with coarse chromatin and irregular nuclear membranes. should be differentiated from immature squamous metapla-
Mitoses can be observed in syncytial clusters (Table 17.13). sia, atrophy, endometrial cells, and tubal metaplasia. While
17 Gynecologic Cytology 593
a b
c d
Fig. 17.19 LSIL (SurePath, Papanicolaou stain). (a) Binucleated and can be seen in HPV-associated lesions. Keratinizing cells showing
enlarged irregular nuclei, more than three times of intermediate cell nuclear abnormalities and increased N/C ratio should be categorized
nuclei and coarse chromatin. Less prominent koilocytic change with according to degree of abnormality. Increase in nuclear size and nuclear
some cytoplasmic clearing and condensation of cytoplasm at periphery irregularities in addition to presence of a classic koilocyte supports the
(600×). (b) Disproportionally large cells with multiple nuclei. Nuclear diagnosis of LSIL. Compare with Fig. 17.17e (400×). (d) HPV cyto-
chromatin does not always have to be coarse and in this case is “washed pathic effect consisting of clear perinuclear halo,“koilocyte” with
out” and pale; however multinucleation, irregular shapes of nuclei, and nuclear enlargement, coarse chromatin, and nuclear membrane irregu-
large size of this cell are consistent with HPV changes (600×). (c) larities (600×)
Parakeratotic cells with slightly increased size of hyperchromatic nuclei
increased nuclear to cytoplasmic ratio can be seen in any of coarseness of chromatin, absence of nucleoli, and quality
the HSIL mimickers, marked nuclear membrane irregulari- and amount of cytoplasm should be helpful clues
ties, coarse chromatin with significant anisonucleosis, over- (Table 17.15).
lap and nuclear crowding, and atypical mitotic figures are the
major hallmarks of HSIL (Table 17.14). HSIL with Endocervical Gland Involvement
Single HSIL cells should be differentiated from mimick- Colonization of endocervical glands by HSIL poses a diag-
ers with a similar pattern. These may include immature squa- nostic problem in cervical smears. Such architectural
mous metaplasia, isolated atrophic cells, inflammatory cells, arrangement of HSIL within endocervical glands can mimic
decidualized stromal cells, and isolated endocervical cells. atypical and malignant glandular lesion such as AIS and
The morphological features of those atypical single cells adenocarcinoma. Clues to recognizing the squamous nature
mimickers with high nuclear to cytoplasmic ratio are some- of the lesion include spindling of the cells in the central
times even more difficult to discern from features of HSIL areas of cellular clusters and flattening of the nuclei at the
than when atypical cells are in syncytial group; however, periphery.
594 U. Krishnamurti et al.
a b
Fig. 17.20 HSIL (SurePath, Papanicolaou stain). (a) Low power view in Figure b. showing cells with large, hyperchromatic and irregular
of cluster of seven cells that appears smaller than the rest of the smear nuclei with coarse chromatin consistent with HSIL (400×). (c) Crowded,
but with extremely high N/C ratio (100×). (b) High power of the cluster hyperchromatic pattern of HSIL arranged in syncytial group (400×)
Table 17.13 High-grade squamous intraepithelial lesion (HSIL) syncytial aggregates of malignant cells with large nuclei,
Morphological features of HSIL coarse chromatin with chromatin clearing, and prominent
Immature Usually smaller parabasal-sized Fig. nucleoli. Tumor cells are usually surrounded by a back-
squamous or cells 17.20a ground of blood and necrotic debris often referred to “dirty
parabasal cells Single discrete cells or Fig.
background” or tumor diathesis (Fig. 17.23a–c). Cervical
syncytium-like groups 17.20b
Nuclear changes Enlarged nuclei Fig.
smears with keratinizing squamous cell carcinomas contain
High N/C ratio 17.20a, b less prominent “dirty background” and usually show single
Marked nuclear contour with malignant cells of different shapes and sizes including
irregularities of nuclear “tadpole-shape” (Fig. 17.23d). Tumor cells in keratinizing
membranes
Marked hyperchromasia
squamous cell carcinoma often show orangeophilic cyto-
Marked chromatin coarseness plasm and large irregular markedly hyperchromatic nuclei
Mitoses Present usually with inconspicuous nucleoli (Table 17.16).
a b
Fig. 17.21 HSIL vs. atrophy (SurePath, Papanicolaou stain). (a) shows slightly overlapping cluster of smaller cells with hyperchromatic
Cluster shows overlapping cells with increased N/C ratio and anisonu- nuclei with smooth boarders, slight anisonucleosis. N/C ratio is not as
cleosis, coarse chromatin, irregular nuclear boarders, and mitotic high; changes are consistent with atrophy (400×)
figures consistent with syncytial pattern of HSIL (600×). (b) Cluster
obtain endocervical cells, increased awareness, and increased reactive cellular changes of endocervical cells share some
incidence of adenocarcinoma. morphological features of AGC and can be difficult to dif-
Atypical glandular cells (AGC) refer to cellular changes ferentiate. Glandular extension of HSIL also can mimic AGC
that fall between definitive benign reactive process and and accounts for about 30% of women referred to colpos-
unequivocal AIS or adenocarcinoma [56–58]. Glandular cel- copy for AGUS diagnosis [44].
lular abnormalities are considerably less common than squa- Cytological features of AGC are related to underlying his-
mous type, and the AGC diagnosis should represent less than tological abnormality and the origin of atypical glandular
1% of all cervicovaginal smears. Even though AGC should cells. The most common underlying lesion associated with
be reserved for changes in which the diagnosis of AIS or the interpretation of AGS is squamous intraepithelial lesions
adenocarcinoma was seriously considered, it is still a chal- (including low- and high-grade lesions), followed by benign
lenging diagnosis for many cytopathologists [59]. Marked lesions (such as polyps and microglandular hyperplasia) and
596 U. Krishnamurti et al.
a b
Fig. 17.22 HSIL vs. squamous metaplasia (SurePath, Papanicolaou ers (600×). (b) Compared to similar cluster of cells, however, nuclei are
stain). (a) Cluster of cells with slightly increased N/C ratio, dense cyto- larger, nuclear boarded irregular, and coarser chromatin (600×)
plasm and fine, smudged chromatin, and smooth regular nuclear board-
then adenocarcinoma in situ (AIS) and invasive adenocarci- Category of AGC-favor neoplastic includes the diagno-
noma [60]. sis in which some but not all the features of AIS is present.
AGC can be further subclassified as “not otherwise speci- Nuclear feathering is one of the crucial features of AIS, and
fied” or “favor neoplastic” depending on the severity of mor- when not present AIS cannot be confidently diagnosed in
phological changes present. AGC can also be classified as cervical smears.
AGC-NOS, AGC-endocervical, and AGC-endometrial based
on the particular changes present. In cases interpreted as
AGC-favor neoplastic, the cytological abnormalities are 17.10 Adenocarcinoma
more pronounced quantitatively and qualitatively and par-
tially overlap with features of adenocarcinoma in situ Invasive adenocarcinoma should be subclassified as endo-
(Fig. 17.24a–cb, Table 17.17). AIS typically shows groups of cervical or endometrial origin whenever it is possible.
cells in rosette formations with crowding and feathering of However, such distinction may not be possible because of
nuclei. Nuclei are elongated with coarser chromatin but the overlapping morphological features. Adenocarcinoma
without conspicuous nucleoli (Fig. 17.24dc). There is cells from both endocervical and endometrial origin contain
increased N/C ratio. If some but not all these characteristics enlarged nuclei with prominent nucleoli and coarsely
are present, the changes can be interpreted as AGC-favor clumped chromatin (Fig. 17.25a–d). Morphological features
neoplastic (Table 17.17). Pap test detects only 38–50% of of invasive adenocarcinoma, nonkeratinizing SCC, and
AIS, and only 1.1–4.7% of women diagnosed with AGC HSIL colonizing the glands can overlap and are delineated
have biopsy-confirmed AIS [61, 62]. in Table 17.18.
17 Gynecologic Cytology 597
a b
c d
Fig. 17.23 Invasive squamous carcinoma (SurePath, Papanicolaou nonkeratinizing squamous cell carcinoma (600×). (c) Keratinized squa-
stain). (a) Cluster of cells with features of HSIL and adjacent tumor mous cell carcinoma showing a large keratinized cluster of cells
diathesis. Findings are suggestive of nonkeratinized squamous cell car- admixed with necrotic debris (600×). (d) High power shows atypical
cinoma (400×). (b) Syncytial cluster of cells with highly pleomorphic keratinized and nonkeratinized cells including “tadpole” cells with
cells, coarse chromatin, irregular nuclear boarders, and marked overlap. hyperchromatic irregular nuclei and elongated cytoplasmic “tail”
Marked anisonucleosis with other features of HSIL raises a concern for (600×)
Table 17.16 Morphological features of keratinized SCC, nonkeratinized SCC, and HSIL
Keratinized SCC Nonkeratinized SCC HSIL
Cell Cells are predominantly isolated, Single or in aggregates with poorly defined Single or in syncytial aggregates
arrangement less often in aggregates cytoplasmic boarders
Cell size Marked variation Mostly smaller cells compared to keratinizing Small cells
squamous cell carcinoma
Cell shape Caudate and spindle-shaped cells Round to elongated cells Usually round cells
are seen
Cytoplasm Dense orangeophilic Wispy scant cytoplasm with indistinct boarders Delicate bluish cytoplasm
Nuclei Irregular, multinucleated Markedly irregular nuclear borders Markedly irregular nuclear
borders
Coarse irregular chromatin with Coarse irregular chromatin with chromatin clearing Coarse irregular chromatin
chromatin clearing
Nucleoli Less common than in Often present and may be prominent Inconspicuous
nonkeratinized ca
Tumor Less common than in Often present Not present
diathesis nonkeratinized carcinoma
598 U. Krishnamurti et al.
a b
Fig. 17.24 AGUS (SurePath, Papanicolaou stain). (a) Atypical endo- of endocervical cells showing enlarged nuclei and prominent nucleoli
cervical cells show nuclear enlargement, some prominent nucleoli and (600×). (c) Endocervical AIS - Pseudostratified cluster with enlarged
a sheetlike arrangement (400×). (b) Crowding and overlapping pattern elongated and hyperchromatic nuclei suggestive of “feathering” (600×)
Table 17.17 Morphological features of reactive glandular cells, AGC, and AIS
Morphological features of benign reactive glandular changes, AGC, and AIS
Benign reactive AGC “not otherwise specified” AIS (Fig. 17.19d)
Glandular changes (Fig. 17.19a–cb)
Cellular Sheets and strips Small to large tissue fragment Small to large tissue fragment
arrangement Honeycomb if in sheets No honeycomb arrangement Syncytial arrangement
Crowding Usually no nuclear crowding or nuclear Variable nuclear crowding and Obvious nuclear crowding and
overlap overlapping overlapping
Vertical orientation of cells at
periphery
Feathering No No Yes
Nuclei Rounded to oval Rounded to oval Coarse granular Oval to elongated nuclei,
Mild variable hyperchromasia chromatin “cigar-shaped”
Nucleoli may be present Nucleoli may be present Marked hyperchromasia
No nucleoli
N/C ratio Slightly increased N/C ratio Variable increase in cell size and N/C Markedly increased N/C ratio
ratio
Cytoplasm Adequate volume of cytoplasm Cytoplasm scant to moderate Cytoplasm scant
Mitoses Absent Mitoses may be present Frequent mitoses
17 Gynecologic Cytology 599
a b
c d
Fig. 17.25 Adenocarcinoma (SurePath, Papanicolaou stain). (a) Tight cells with prominent nucleoli are seen at the periphery of the cluster,
tridimensional cluster of glandular cells with enlarged nuclei, promi- endometrial adenocarcinoma (600×). (d) Irregular cluster of cells with
nent nucleoli, and coarse chromatin. Patient with endometrial serous high N/C ratio, with coarse chromatin, and some with prominent nucle-
carcinoma (600×). (b) Cluster of glandular cells with wispy cytoplasm oli. Cells at the periphery retained columnar shape, endocervical adeno-
and large nuclei with prominent nucleoli, endocervical adenocarcinoma carcinoma (600×)
(400×). (c) Three-dimensional cluster with a grape-like structure. Large
Small-cell carcinomas of the cervix are rare, and they are 17.11.2 Malignant Melanoma
aggressive malignancies that occur in a wide age range and are
usually asymptomatic in the early stage. Small-cell carcinoma Up to 5% of malignant melanomas in women occur in the
is commonly associated with HPV type 18. Morphologically, gynecologic tract. Although the most common site is vulva,
600 U. Krishnamurti et al.
melanomas can arise in the vagina and cervix. Melanomas be present but usually as a minor component [67]
are characterized by isolated or loosely cohesive cells that (Fig. 17.26c).
can be epithelioid or spindled with large nucleoli and cyto-
plasmic pigment [65] (Fig. 17.26a). 17.11.4.1 Metastatic Tumors
The most common metastatic tumors detected in Pap smears
are ovarian and fallopian tube serous adenocarcinomas. They
17.11.3 Non-Hodgkin Lymphomas have the usual cytological features of high-grade adenocarci-
noma including single cells or three-dimensional clusters of
Most cases of non-Hodgkin lymphomas in the cervix and large cells with irregular nuclear borders and prominent nucle-
vagina are seen in advanced stage of disease. Rarely, lym- oli. Psammoma bodies can be detected and are seen as lami-
phoma may arise as a primary tumor in the cervix or the nated dark blue calcifications on Papanicolaou stain. Tumors
vagina. Cytological samples may be negative if mucosa is not from the rectum, colon, bladder, and urethra can spread
involved. If mucosa is involved, the smears are usually cellular directly to the cervix or vagina [68]. Metastatic tumor can also
and are similar in morphology to lymphomas originating in be from as the breast, kidney, lung, and pancreas. When poorly
the lymph nodes. The major morphologic characteristics of a differentiated, such tumors exhibit large hyperchromatic
non-Hodgkin lymphoma include discohesive malignant cells malignant cells without any distinguishing features. It is
with high N/C ratio, prominent nucleoli, and mitotic activity important to recognize those as malignant, but specifying the
[66] (Fig. 17.26b). Lymphoglandular bodies may be seen. primary site may not be possible without clinical history,
Differential diagnosis includes follicular cervicitis, small- biopsy, and additional immunohistochemical stains.
cell carcinoma, and invasive nonkeratinized squamous cell
carcinoma.
17.12 Newly Developed Techniques
17.11.4 M
alignant Mixed Mesodermal 17.12.1 Automated Screening
Tumors
17.12.1.1 ThinPrep Imaging System (TIS) [69]
Malignant mixed mesodermal tumors usually occur in the The ThinPrep Imaging System (TIS) uses location-guided
uterine corpus. When detected by Pap smear, it mostly repre- screening to assist the cytotechnologist in reviewing
sents a tumor that has originated in the uterine body and has ThinPrep Pap slides.
extended to the cervix. Much of the tumor that sheds into the
cervix originates from the high-grade epithelial part of the Major components of the TIS are:
tumor and will have features of high-grade adenocarcinoma • Image processor for computer image analysis
or undifferentiated carcinoma. Malignant spindle cells may • Review Scope providing automated microscope location
17 Gynecologic Cytology 601
a b
Fig. 17.26 Metastatic malignancies (SurePath, Papanicolaou stain). macrophages and lack of spectrum of maturation forms of lymphocytes
(a) Single large cells with prominent nucleoli and spiderlike cytoplas- indicate hematopoietic malignancy, large cell lymphoma, and primary
mic projections filled with brown pigment. Patient was diagnosed with uterine (600×). (c) Large multinucleated spindled cell with prominent
vaginal/cervical malignant melanoma (600×). (b) Numerous lymphoid nucleoli, primary rhabdomyosarcoma of the uterus (400×)
cells with scant cytoplasm in loose groups. Absence of tangible body
The image processor has three principal components: –– It holds 10 cassettes and each cassette can hold 25
• Image processor controller with internal server slides.
–– Comprised of computer and software to capture and –– It scans slides and selects 22 fields of view (FOV),
analyze the slide images as well as store the results of most likely to contain abnormal cells. FOV is based on
the analysis. optical density measurements and other features. The x
–– The server is the central data manager that stores, and y coordinates of the 22 FOV are stored in a data-
retrieves, and transmits information based on the slide ID. base that is retrieved later.
• Imaging processor • Computer user interface equipment
–– Is a tabletop unit with hardware used to image the cassettes –– Consists of the monitor, keyboard, and mouse that are
of previously stained and coverslipped ThinPrep slides. like those in a personal computer.
602 U. Krishnamurti et al.
The Review Scope has three components: review.” Of slides requiring further manual review, it selects
• Microscope with automated stage. at least 15% for a second manual review. Slides classified as
• Twenty-two fields of view (FOV) are presented to the unsatisfactory for analysis must also undergo manual review.
reviewer by automatic slide movement. However, the BD FocalPoint Slide Profiler is not approved
• Manual review of the slide may also be performed. for screening women at high risk for cervical cancer. On
• An automated marking system allows the reviewer to evaluation in a clinical trial, the BD FocalPoint Slide Profiler
mark sites either physically or electronically for further detected significantly more abnormal slides (ASC-US or
review. greater) than regular practice (86% vs. 79%) [77]. BD
• The display unit is used to communicate with the reviewer. FocalPoint Slide Profiler archives about 16–17% of Paps as
• The pod is used to control the microscope. not requiring further review resulting in a modest gain in
• It is used by the cytotechnologist to navigate to the 22 productivity [77, 78].
FOV. It was in 2008 that the FDA approved the BD FocalPoint
• To navigate from one FOV to the next is done using the GS Imaging System which consists of the BD FocalPoint
shortest distance. Slide Profiler and a BD FocalPoint GS Review Station,
• Cytotechnologist can change objectives and advance or which uses location-guided screening to assist the cytotech-
return to FOV. nologist. The SurePath slide is first evaluated by the BD
FocalPoint Slide Profiler, which then uses algorithms to
If no abnormal cells are detected in any of the 22 FOV, the select ten fields of view (FOV) most likely to harbor abnor-
case can be reported as negative. If any abnormal cells are mal cells. These ten FOV are presented to the cytotechnolo-
found, the entire slide must be reviewed. gist at the Review Station. If no abnormality is detected in
Implementation of TIS has been found to increase pro- the FOV, the slide can be reported as negative with no further
ductivity [70, 71], since the 22 FOV represent about 25% of review. If any abnormality is found in the FOV, the slide
the screening area [72]. In addition, studies have shown that must undergo full manual review.
TIS detected more abnormal cases (including ASC-US or Evaluation of the BD FocalPoint GS Imaging System
more severe lesions) than manual review (82% vs. 76%) showed that detection of HSIL+ increased by 19.6% and of
[70]. Other studies have reported that TIS resulted in a higher LSIL+ by 9.6% in the computer-assisted arm, but there were
detection rate of LSIL and HSIL [73, 74]. significant decreases in specificity. The two study arms of
regular practice vs. computer-assisted screening were not
17.12.1.2 B
D (Becton Dickinson) FocalPoint GS statistically significant for sensitivity and specificity of
Imaging System [75] ASC-US+ [79]. The BD FocalPoint GS Imaging System also
Components of the BD FocalPoint GS Imaging System are: increases productivity [80].
testing has a very high negative predictive value (NPV). If HPV detections platforms for histological specimens have
HR-HPV is negative, risk of HSIL in 3 years is very low. been made [85, 86].
Almost all HSIL cases are seen in women with persistent Other biomarkers for HPV have been proposed, which are
HR-HPV infection. discussed in the relative section, but which are not currently
In the search for suitable platforms for clinical HPV used in the clinical setting.
detection, primarily in screening algorithms for cervical car-
cinoma, but also for prognostic and differential diagnostic
purposes in other squamous carcinomas, it has become evi- 17.12.3 ain Platforms for Molecular HPV
M
dent that DNA-based methods are currently the most suitable Testing
way of approaching this issue and has led to the development
of different platforms. In contrast to other viral infections, 17.12.3.1 Platforms for HPV DNA Detection
the use of other virological methods, such as culture in vitro Cobas® 4800 HPV Genotyping. The cobas® HPV test
culture and immunological methods, such as serology-based (Roche Molecular Diagnostics, Branchburg, NJ) is a qualita-
assays, is not suitable for HPV detection. This limitation is tive test, on a real-time PCR platform, that is approved by the
mainly associated with HPV infection natural history, but FDA to detect HPV in liquid cytology specimens. This test
also with characteristics of the HPV replicative circle: HPV specifically identifies 14 HR-HPV subtypes, including 16
infection and proliferation are tightly linked to the presence and 18. Results are reported as negative or positive for HPV
of well-differentiated squamous cell, with infection of lower 16, HPV 18, and for pooled “other” HR-HPV subtypes,
strata of the squamous epithelium and with capsid produc- which include types 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66,
tion occurring in terminally differentiated squamous cells and 68.
[81]. This renders in vitro culturing of HPV difficult and not Hybrid Capture II. The Hybrid Capture II (HCII), avail-
suitable for clinical testing. While there are HPV infections able from Qiagen (Hilden, Germany; previously available
recognized in many mammalian species, there is a species- through Digene, Gaithersburg, MD), was the first FDA-
specific differentiation for all the different HPV types, mak- approved assay for HPV detection and employs a sandwich
ing the introduction of animal models problematic. Further, enzyme-linked immunosorbent assay (ELISA) platform.
the introduction of “viruslike particles” (VPLs) has not only The DNA of HPV, if present in the sample, hybridizes to a
aided in the production of HPV vaccinations but has also led series of HPV-specific RNA probes. The produced hybrid
to the development of serologic assays. Nonetheless, the pro- DNA-RNA molecules are then captured by proprietary anti-
duction of HPV antibodies happens in a subset of patients body. This is detected with the use of a second, peroxidase-
that have prolonged HPV infections and only in interval of conjugated antibody and a chemiluminescence substrate.
12–15 months after initial HPV infection [82]. Therefore, The emitted light is captured and measured by a luminome-
this kind of detection is incompatible with the purposes of a ter. This test can detect either low-risk HPV types (LR-HPV;
screening test and is moreover not suited for detecting HPV 6, 11, 42, 43, 44) or HR-HPV (16, 18, 31, 33, 35, 39, 45, 51,
infections that are acute. 52, 56, 58, 59, 68).
The importance of HPV genotyping is also reflected in Roche Linear Array (LA). This is a reverse line blot
the American Society for Colposcopy and Cervical Pathology assay for HPV genotyping, which is available commercially.
(ASCCP) for cervical cancer screening guidelines, as spe- Roche Linear Array is based on the aforementioned
cific HR-HPV type (HPV 16/18 vs. other) determine man- PGMY09/11 PCR system and detects 37 HPV types (6, 11,
agement of women who are HR-HPV positive [83]. 16, 18, 26, 31, 33, 35, 39, 40, 42, 45, 51, 52, 53, 54, 55, 56,
Testing for HPV status in formalin-fixed, paraffin- 58, 59, 61, 62, 64, 66, 67, 68, 69, 70, 71, 72, 73, 81, 82, 83,
embedded (FFPE) histological samples, mainly for prognos- 83, IS39, CP6108) (Roche Diagnostics, Pleasanton, CA).
tic reasons (but also as an aid in differential diagnosis), More in detail, the products of HPV PCR are created using
especially in head and neck squamous cell carcinomas, is biotin-labeled primers, which are hybridized to a type-
currently mainly done using either DNA in situ hybridization specific HPV probe-containing strip, and are then detected
(ISH) or p16 immunohistochemistry (IHC). Both these tests with the aid of streptavidin-conjugated horseradish peroxi-
have limitations, such as the fact that they are highly depen- dase, along with a chromogenic substrate.
dent on the experience of the pathologist, but are also labor Amplification (PCR)-Based Methods. One of the
intensive. In addition, HPV ISH has a relatively low sensitiv- advantages of PCR-based methods is the high analytical sen-
ity, and p16 IHC is not 100% specific for demonstrating sitivity; however, a limitation they face is that assays that are
HPV tumorigenesis as it is a surrogate marker of HPV infec- type-specific can only detect a specific HPV type or may be
tion and does not directly identify the presence of HPV [84, also other, closely related types, which make these less than
85]. More recently, efforts to validate the use of existing optimal for clinical testing. Despite the use of PCR-based
604 U. Krishnamurti et al.
platforms mainly in Europe, there are no available PCR- quently hybridized to a set of types of specific beads [93] and
based HPV detection assays in the US market. Two well- attached to HPV type-specific probes. The beads that have
described examples can employ nondegenerate consensus HPV DNA hybridized to them are then labeled with
primers, used under low stringent amplification conditions, phycoerythrin-conjugated streptavidin which is detected by
such as the GP5/GP6 and GP5+/GP6+ primer set [87, 88], the Luminex 100 instrument. This assay has been validated
which detects a total of 27 HPV types, and the MY09/MY11 for clinical use for detection and genotyping of 37 HPV
primer set [88]. The MY09/MY11 degenerate primers target types (6, 11, 16, 18, 26, 31, 33, 35, 39, 40, 42, 45, 51, 52, 53,
specific, highly conserved sequences of HPV 6, 11, 16, 18, 54, 55, 56, 58, 59, 61, 62, 64, 66, 67, 68, 69, 70, 71, 72, 73,
and 33, located at a 450-bp region of the L1 ORF of the HPV 81, 82, 83, 83, IS39, CP6108).
[89] and uses nested primers for HPV types 16, 18, and 31. Cervista. The Cervista HR-HPV test detects DNA
Subsequently, mixed primers were developed to address genomic sequences of L1, E6, and E7 HPV genes, but offers
sensitivity-reproducibility issues associated with the use of no genotyping. The Cervista HPV 16/18 assay offers identi-
degenerate primers, such as the PGMY09/11 and SPF primer fication and differentiation of the two most common
sets and systems. HR-HPV types. Clinical comparison between the Cervista
PGMY09/11 primers were created in order to improve on platform and Aptima, and mRNA-based HPV detection
MY09/MY11, and, by comparison to the original MY09/ assay offered by the same company (described below),
MY11/HMB01 primer set, the PGMY09/11 are found to be reported a low specificity for Cervista [94].
associated with higher sensitivity and specificity, but also in
the reproducibility of detecting HPV types that were previ- 17.12.3.2 Platforms for HPV mRNA Detection
ously not efficiently primed (a more significant issue in The main platform, Aptima HPV, is commercially available
lower viral loads). for HPV detection and genotyping in the United States through
SPF1/SPF2 (Microgen Bioproducts, Camberley, UK) is a Hologic (Marlborough, MA). The Aptima detects mRNA
short consensus fragment system for HPV PCR, consisting from E6 and E7 HPV genes and is designed to detect 14
of six primers (four forward and two reverse) which amplify HR-HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59,
a short (65-bp) fragment of HPV DNA, while the updated 66, and 68). The FDA approved the complementary Aptima
version of this platform (SPF10) consists of the addition of genotype assay detection, and genotyping of HPV 16 and
four additional primers (ten total), thus increasing analytical HPV 18/45 (it does not distinguish among the two types) is
sensitivity [90]. The advantage of this PCR test is that, as the offered for women with positive Aptima HPV detection test.
amplicon size is small, it is more suitable for testing on FFPE Given the central role that the E6 and E7 genes and their
blocks or specimens of lower quality/degraded DNA. product in HPV pathogenesis, as well as their documented
Roche Diagnostics (Pleasanton, CA) produced a HPV overexpression, which correlates with integration of the
PCR detection system, utilizing a 96-well microplate, genomes of HPV in cervical carcinoma, testing for HR-HPV
launched in Europe for clinical use of the Amplicor HPV E6/E7 mRNA in cervical samples could have higher speci-
test, able to detect a total of 13 HR-HPV types [91]. The kit ficity, compared with DNA testing, for the high-grade cervi-
uses 12 different biotin-labeled primers, targeted to amplify cal lesion detection. There are studies available that have
a 165-bp region in the L1 gene of the genome of these HPV suggested that HPV mRNA testing ameliorates specificity
viruses, subsequently captured by a fixed probe set. However, for CIN2-or-higher lesions (CIN2+), in comparison to DNA
this system is no longer commercially available. assays [95–97], but also higher positive predictive value
The INNO-LiPA HPV Genotyping II Extra. This is a [98]. However, these results have been contradicted by dif-
genotyping assay (Fujirebio, Tokyo, Japan) which is based ferent publications, showing no significant differences
on the SPF PCR system and employs type-specific probes between DNA and RNA testing methods and specifically in
for 32 HPV genotypes (6, 11, 16, 18, 26, 31, 33, 35, 39, 40, their performance for the detecting of CIN2+ [99, 100].
42, 43, 44, 45, 51, 52, 53, 54, 56, 58, 59, 61, 62, 66, 67, 68, There is a long series of HPV-related testing platforms
70, 73, 81, 82, 83, 89). In a comparison study between this available in literature [101] that are beyond the scope of this
platform and the Roche LA, the authors found that, while chapter and are thus not included in this list.
results were comparable (for shared genotypes), the sensitiv-
ity for HPV 16 was higher with the LA and was higher for
HPV 31, 53, and 66 with the INNO-LiPA [92]. This platform 17.12.4 Biomarkers and Molecular
is not commercially available in the United States. Application
The Liquid Bead Microarray (LBMA) Assay. The
LBMA assay utilizes the Luminex-based platform (Luminex Our knowledge of the biology of HPV and its role in cervical
Corporation, Austin, TX). In this process DNA of HPV is cancer is changing how we screen for this disease. Direct detec-
amplified using biotin-labeled consensus primers, subse- tion of HPV, e.g., by testing for HPV DNA or HPV-produced
17 Gynecologic Cytology 605
mRNA of known oncogenic genes (E6, E7), has become part Methylation is a known epigenetic factor important in
of the screening strategies for cervical cancer. Further, as HPV cancer, and there are several studies that have looked into
types 16 and 18 account for the majority of cervical cancers but methylation of both HPV and human genomes, in associa-
also the types with the highest virulence, there is the need for tion with cervical cancer. It has been suggested that HPV 16
HPV genotyping in cervical cancer guidelines. shows differential methylation patterns, especially with
It has become clear that the majority of women with pre- regard to L1, L2, and E2 genes in association with cervical
cancerous lesions will regress, and only a minority will prog- cancer [107–109]. A series of studies have suggested that
ress to developing invasive cervical carcinoma. Therefore, measurement of individual CpG DNA methylation sites on
there is the need to find biomarkers that can separate women HR-HPV genomes can be utilized for epidemiological pur-
with similar clinical characteristics into those who will prog- poses, but also to follow up women with documented
ress and others who will regress. HR-HPV infections. The same studies have shown a correla-
tion between methylation levels and persistence of infection
for HPV 16, with yearly increments of 0.5–0.7% in cases of
17.12.5 Cytological and Histological persistent infection [107, 110]. There is a potential utility for
Correlation of Disease tracking HR-HPV infections in order to predict risk of sub-
sequent to high-grade dysplasia and/or cancer.
The Bethesda classification of squamous intraepithelial lesions Further, a long list of human genes investigated as possi-
can be correlated with the histologic diseases. In this aspect, ble methylation biomarkers of cervical cancer [111–123],
cytologic LSIL correlates histologic LSIL (including koilo- with numbers exceeding 100 genes proposed as biomarkers
cytic changes and CIN1). Cytologic HSIL corresponds to his- of this kind; however, no single gene has been proven suffi-
tologic HSIL (including CIN2 and CIN3). In contrast to the ciently sensitive and specific for identifying lesions in the
progression from tubular adenoma to high-grade dysplasias to CIN3+ category. The more feasible approach seems to be
invasive tumor in colon cancer, HPV infection and cervical that of gene panels, and two genes with evidence of good
carcinogenesis, does not show a linear progression from low- sensitivity/specificity performance are MAL and CADM1
grade to high-grade dysplasia. Instead, LSIL is associated with [112, 113, 118, 124, 125].
HPV infections that are transient, while HSIL represents a pre- Other epigenetic markers are microRNAs, with a multi-
cancerous lesion. However, an LSIL diagnosis in patients por- tude of available studies attempting a microRNA expression
tends a higher risk for a subsequent HSIL diagnosis, probably assessment in cervical carcinomas and in serum of women
in association with the likelihood of repeated HPV infections with and without cervical cancer, which have reported lists
in the same patients. of microRNAs upregulated and downregulated [126–134]
and may play a role in cervical carcinoma (Table 17.19).
17.12.5.1 HR-HPV Detection and Genotyping
Another important observation is that the specific HPV type Table 17.19 List of microRNAs reported to be upregulated or down-
detected in LSIL is indicative of the risk of a subsequent regulated in cervical cancer
HSIL or carcinoma diagnosis. Multiple studies, both cross- microRNAs miR-9, miR-10a, miR-15b, References:
sectional [102–104] and prospective [105, 106], have shown upregulated in miR-20a, miR-20b, miR-21, [30–32,
cervical cancer miR-27a, miR-34a, miR-127, 34–38]
the different virulence and carcinogenicity associated with
miR-133a, miR-133b, miR-141,
different types of HR-HPV. In fact, the current ACCP guide- miR-145, miR-155, miR-196a,
lines recommend that women be managed differently if HPV miR-199a*, miR-199a,
types 16 and 18 are found upon genotyping [83]. miR-199b, miR-199-s,
miR-200a, miR-203, miR-214,
miR-221, miR-224, miR-500,
17.12.5.2 Other Proposed Biomarkers miR-505, miR-711, miR-888,
Other possible biomarkers have been proposed, which miR-892b, miR-1246, miR-
mainly relate to genetic, epigenetic proteomic/metabolic 1290, miR-2392, miR-3147,
miR-3162 miR-4484
changes that are seen in association with HPV infections and
microRNAs Let-7a, Let-7b, Let-7c, References:
cervical cancer. Many of these, including those listed, are not downregulated miR-26a, miR-27b, miR-30d, [30, 31, 37]
clinically used, but could shed further light into the biology in cervical miR-34a, miR-100, miR-103b,
of HPV and on the interactions between HR-HPV viruses cancer miR-125a, miR-125b, miR-126,
and different mechanisms of host response, which may result miR-141, miR-143, miR-145,
miR-149, miR-200b, miR-200c,
in cervical cancer genesis and progression or in spontaneous miR-203, miR-204, miR-375,
regression. As the presence of HPV infection does not equate miR-451, miR-3185, miR-3196,
with carcinogenesis, there are several proposed markers of miR-3960, miR-4324, miR-
cervical cancer, as well as markers of clinical outcome. 4467, miR-4488, miR-4525
606 U. Krishnamurti et al.
Other proposed biomarkers include serum protein mark- (Roche Molecular Diagnostics, Branchburg, NJ) as a pri-
ers of disease [135, 136], histone acetylation status [137], mary screening method for cervical cancer 25 years old or
metabolomics, and transcriptomics [138], and their clinical older women [154].
validity remains to be ascertained. There are definite advantages of HPV DNA testing, as
evinced by several studies, such as higher sensitivity for pre-
cancerous lesions, higher precision (reproducibility), but it
17.12.6 The Relationship Between Cytology has actually lower specificity, when compared to cytomor-
and HPV Testing phological primary screening [155–158]. What is now con-
sidered a landmark population-based study on the viability
Since its inception and implementation, cervical cancer of HPV DNA testing for cervical cancer screening, exam-
screening by Pap test has resulted in a continuous decline in ined more than 4000 Planned Parenthood patients, showed
cervical cancer incidence and cervical cancer-related mortal- that 30% of women under the age of 30 had positive results
ity in the United States in recent decades [139]. Pap test has for HR-HPV and would thus require colposcopy follow-up,
known limitations for its low sensitivity (30–60% for a sin- thus rendering primary HPV testing non-feasible for this age
gle Pap smear) [140–143] and poor intra- and interobserver group [159]. These results have contributed to the current
concordance [144, 145]. Despite its shortcomings, Pap test guidelines and FDA approval for HPV testing as a method
has proven so effective that by the mid-1990s, the epidemio- for primary screening in women over age 30 or older.
logically most significant factor of risk for cervical carci- In Europe, where primary cervical carcinoma screening by
noma in the United States was actually not being screened HPV testing was first implemented, randomized trials com-
for it with the use of Pap smears [146, 147]. paring HPV versus cytological primary screening in female
Understanding the importance of HR-HPVs in cervical population age 30 or older concluded that testing for HPV is
cancer pathogenesis, along with the subsequent development a better screening method for cervical carcinoma, in compari-
of platforms for molecular detection of these HPVs, has per- son to cytology [160–162]. Moreover, after concluding that
mitted the introduction of molecular testing in the cervical negative results in HPV testing are associated with low risk
cancer screening guidelines [83, 148] and has contributed in for either CIN3/carcinoma in situ/invasive cervical carcinoma
improving this screening process. In fact, subsequent meta- in the following 5 years, studies have suggested prolonging
analysis studies showed how triage of patients with ASC-US the follow-up screening interval to 5 years [163, 164].
diagnoses, utilizing HPV-specific testing, improves accuracy Despite all the advantages that HPV testing clearly offers
(meaning more sensitive with comparably specific or compa- for cervical cancer screening, primary HPV screening has
rably sensitive and more specific) compared to repeat cytol- some limitations, mainly connected to the disease course of
ogy [149–153]; as such, HPV testing has become the HPV infections and cervical carcinomas. Indeed, HPV test-
preferred follow-up method, although follow-up cytology is ing offers a snapshot of HPV infection status at any given
still acceptable [83]. Further studies have permitted to iden- moment, and thus a single test cannot offer information on
tify the benefit for specific subgroups within the female pop- the transient or persistent nature of an HPV infection, which
ulation, and we thus now know that, because of the very high becomes a bigger issue in the younger female population, in
prevalence of and transient nature of the vast majority of which HPV infections are much more prevalent.
HPV infections in young women, testing for HPV is not indi- In fact, most cervical HPV infections are not persistent;
cated for women 20 years old or younger [83]. Moreover, however, it is this minority of persistent HPV infections
conservative follow-up is indicated for this group, based on (usually defined persistent at 6 months, if due to the same
current guidelines [83, 148]. Based on the same guidelines, HPV type and with no intervening negative HPV tests) that
concurrent Pap test and HPV testing (referred to as co- are shown to be associated with the development of CIN3 or
testing) are preferred or acceptable strategy women aged worse [15, 165–172]. The findings in normal controls in the
30–64. Finally, the importance of HPV genotyping is also PATRICIA HPV vaccine clinical trial showed that the longer
reflected in the ASCCP for cervical cancer screening as it is HPV infections were associated with increased risks of sub-
an acceptable follow-up for women 30 years old or more, sequent CIN lesions, with a 6-month length of infection car-
with negative Pap smears but positive HPV co-testing. Given rying the highest risk for CIN detected on follow-up,
our knowledge on the increased virulence and oncogenicity followed by infections that are persistent, but last less than
of HPV 16 and 18, women may be sent for colposcopy if 6 months. Expectedly, transient HPV infections were associ-
they have HPV 16 or 18 detected, while they may undergo ated with lower risks for developing CIN. More specifically,
co-testing again after 1 year if positive for other HR-HPV women with 6 months or longer infection due to HPV 16 or
types [83]. HPV 33 had a 25 times higher risk for CIN3/CIS/cancer, in
Following implementation in several European countries, comparison to other, non-oncogenic, persistent HPV infec-
in 2014, the FDA has approved the Roche cobas test for HPV tions. The added risk for CIN3/CIS/cancer was also seen
17 Gynecologic Cytology 607
with infection by HPV types 18, 31, and 45, when persistent Table 17.20 Initial management guidelines of Pap and HPV testings
[173]. With this in mind, it is not yet fully understood Pap test Combined tests (Pap and
whether it is the persistence of HPV DNA positivity that Management alone HPV)
truly is important in development of precancerous lesions, or Immediate colposcopy SqCC
it represents a marker for the presence of subclinical HSIL HSIL, HPV+
HSIL AGC, HPV+
(CIN2/CIN3).
HSIL, HPV−
A large Finnish (n = 54,218) general population study
ASC-H, HPV+
among females ages between 25 and 65 compared cytomor- ASC-H
phological and HR-HPV molecular (DNA targeting, fol- AGC ASC-H, HPV−
lowed by cytology for positive DNA results) screening AGC, HPV−
methods. For both study cohorts, the cytology diagnosis of LSIL, HPV+
LSIL and worse warranted colposcopy and biopsy. The ASC-US, HPV+
study findings suggested that, although HPV testing alone LSIL
had high sensitivity, the specificity of the HPV test alone is Repeat testing in ASC-US NILM, HPV+
6–12 months LSIL, HPV−
significantly inferior, compared to Pap smear. In fact, intro-
Repeat testing in 3 years ASC-US, HPV−
ducing cytology follow-up for positive molecular results
NILM
increased specificity to nearly identical values of those of
Repeat in 5 years NILM, HPV−
Pap smear alone, for the entire age population tested (99.2%
vs. 99.1% for CIN2+, P = 0.13). In conclusion, these find-
ings suggested that cytology triage following a positive recommendation is against HPV testing, in order to avoid
HPV testing result may reduce the referral rate to colpos- harm by overtreatment. In females aged 30–65, combined
copy to levels comparable to those of cytology screening. testing (Pap and HPV tests) is recommended in 5-year inter-
More specifically, for females 35 years old or more, molecu- vals. An alternative method is Pap test every 3 years in this
lar HPV testing with subsequent Pap smear on positive age group. It should be emphasized that HPV testing is lim-
results actually improved specificity, when compared to ited to HR-HPV and the HPV test mentioned in the guide-
cytology-based screening. Contrarily (and expectedly), lines refers to only HR-HPV.
women younger than 35 had referrals more frequently when Management algorithm is very complex and includes
screened by HPV testing vs. screened by Pap smears detailed follow-up guidelines for women with abnormal test
(RR = 1.27, 95% CI = 1.01–1.60), and this could mean more results. The principle is to apply equal management for equal
unnecessary follow- up with more lesions identified and risks. Readers can refer to the original publication if inter-
treated that would not progress to cancer [174, 175]. ested (ref 26). The major management guideline is summa-
Additional studies evinced that CIN3+ lesions (CIN3, CIS, rized in Table 17.20.
invasive carcinoma, adenocarcinoma in situ) showed that It should be noted that guidelines cannot replace clinical
the combination of HPV testing with Pap smear follow-up judgment. Currently, there is no test or treatment that is
was a more sensitive screening strategy compared to cytol- 100% effective. Women who undergo regular Pap and HPV
ogy alone in detecting these CIN3+ lesions [176] and testing can still develop cervical cancers.
showed that this is a viable approach and could be the way
screening is done in the future.
17.13 C
ytology of Ascites and Pelvic
Washings
17.12.7 P
ap Test and HPV Testing
Management Guidelines 17.13.1 Formation of Ascites
The recently published consensus guidelines by the American The pelvic and peritoneal cavities are lined by serous mem-
Society of Colposcopy and Cervical Pathology (ASCCP), branes, which consist of connective tissue rich in capillaries
the American Society of Cancer (ASC), as well as the and lymphatics, covered by a layer of mesothelial cells. The
College of American Pathologists (CAP) emphasize the bal- pelvic and peritoneal cavities normally contain a small
ance between benefits and harms associated with Pap test, amount of clear fluid ranging from 7 to 16 mL [177]. This
HPV testing, and related treatment (ref 26). Based on this fluid is continuously produced by ultrafiltration of plasma
guideline, all women should start cervical precancer and through semipermeable endothelial cells. The fluid is
cancer screen at the age of 21. In women aged 21–29, Pap absorbed through lymphatics and capillaries. The balance
test is recommended to be repeated in 3-year intervals. As between formation and reabsorption determines the amount
transient HPV infection is common in this population, the of serous fluid, influenced by hydrostatic pressure in capil-
608 U. Krishnamurti et al.
laries, plasma oncotic pressure, and permeability of capillar- [186]. As low-stage non-myoinvasive papillary serous carci-
ies. Accumulation of fluid between the parietal and visceral noma of the endometrium can present with extrauterine dis-
serous lining of body cavities is termed effusion, which is ease at the time of initial diagnosis [187–189] and
always considered pathologic. The effusion in the peritoneal high-grade, high-stage endometrioid carcinomas are often
cavity is known as ascites. Effusions can be transudate or associated with positive PW suggestive of extrauterine dis-
exudate. Transudates are usually benign and caused by ease, patients with these diseases are treated with adjuvant
increased hydrostatic pressure, e.g., congestive heart failure chemotherapy [190].
and cirrhosis. Exudates can be malignant and are usually due In recent years, laparoscopic hysterectomy, laparoscopy-
to serous membrane damage as in carcinomatosis or infec- assisted vaginal hysterectomy, and robotic hysterectomies
tion. The peritoneal cavity can hold up to 20 L of fluid. have gained popularity in gynecologic oncology. The
Draining (paracentesis) and microscopic examination of the intrauterine balloon used during these minimally invasive
fluid can provide valuable clinical information on the cause surgical procedures creates positive pressure inside the
of ascites. For this reason, microscopic examination of pel- uterine cavity, and as a result, tissue fragments from the
vic and peritoneal fluids is important to rule out an underly- uterine cavity may be “displaced” into vascular spaces,
ing malignancy. into the fallopian tubes, and onto peritoneal surfaces. It is
important to recognize these artifacts, since misinterpreta-
tion may lead to overtreatment with unnecessary adjuvant
17.13.2 Clinical Significance of Pelvic therapy (Fig. 17.27a, b).
Washings It is controversial whether minimally invasive surgical
procedures are associated with a higher incidence in posi-
Pelvic lavage or washing (PW) cytology was first described tive PW. Some studies concluded that endometrial carci-
in a study by Keettel and Elkins in 1956 as a potential way noma cells exfoliate from the tumor surface because of
to detect early spread of ovarian carcinoma cells [178]. positive intrauterine pressure and gain access to the perito-
Since then many studies evaluated the clinical value of PW neal cavity via the fallopian tubes [191]. These studies have
and found correlation with prognosis in patients with gyne- shown that PWs from minimally invasive hysterectomies
cologic malignancies [179–181]. PW is now routinely per- were more likely to have malignant cells, as compared to
formed for staging of gynecologic malignancies, and the those p erformed without the uterine manipulator. A study
results are significant component of the final pathologic and by Hu et al. [192] on patients treated for low-stage and low-
clinical stage of ovarian cancers. The role and clinical sig- grade endometrial carcinomas showed that PWs were more
nificance of PW in endometrial tumors has been evaluated likely to have malignant tumor cells if minimally invasive
by several studies [182–185]. Due to the lack of correlation surgical methods were used (20.6% versus 0%). Another
between positive PW and clinical outcome, PW results are study by Krizova et al. [193] concluded similar findings,
no longer part of the staging process of uterine malignancies specifically 5/40 (12.5%) in minimally invasive versus
a b
Fig. 17.27 (a) Cluster of cytologically low-grade endometrioid adeno- metrioid adenocarcinoma in pelvic washings. Robotic hysterectomy for
carcinoma in pelvic washings. Robotic hysterectomy for non- non-myoinvasive FIGO grade1 endometrioid adenocarcinoma
myoinvasive FIGO grade1 endometrioid adenocarcinoma (ThinPrep, (ThinPrep, Papanicolaou stain, 400×)
Papanicolaou stain, 200×). (b) Cluster of cytologically low-grade endo-
17 Gynecologic Cytology 609
Fig. 17.31 Traumatically exfoliated folded sheets of benign mesothe- Fig. 17.33 Ascitic fluid with numerous well-preserved and degener-
lial cells are characteristic findings in pelvic washings (ThinPrep, ated benign mesothelial cells, inflammatory cell, and histiocytes (Cell
Papanicolaou stain, 100×) block, H&E stain, 100×)
a b
Fig. 17.35 (a) High power view of benign mesothelial cells. Eccentric 400×). (b) Benign mesothelial cells occasionally show some degree of
nucleus, smooth nuclear membrane, washed out chromatin, and small nuclear size and shape variation within a sheet or group of mesothelial
nucleolus are characteristic findings (ThinPrep, Papanicolaou stain, cells (ThinPrep, Papanicolaou stain, 200×)
a b
Fig. 17.37 (a) Reactive mesothelial cells often exhibit nuclear size ing nuclear enlargement and prominent nucleoli (ThinPrep,
and shape variation, binucleation, and hyperchromasia (ThinPrep, Papanicolaou stain, 200×). (c) Mitotic figures are not uncommon find-
Papanicolaou stain, 200×). (b) Sheet of reactive mesothelial cells show- ings in reactive mesothelial cells (cell block, H&E, 100×)
tight three-dimensional hyperchromatic clusters, especially specimen are essential to distinguish these benign Mullerian
when stromal cells are also present (Fig. 17.39a, b) [200, cells from metastatic carcinoma. IHC have limited or essen-
211]. Their resemblance to endometrial cells on cervical tially no utility in distinguishing endometriosis or endosal-
smears can help in identifying them. Nevertheless, mild pingiosis from low-grade serous or endometrioid neoplasia,
cytologic atypia, degenerative changes, and even mitotic fig- as their staining profile is identical.
ures can be misleading findings for malignancy. Characteristic Psammoma bodies are rarely seen in PW and usually best
histologic features of endometriosis are often appreciated on appreciated on cell block sections (Fig. 17.40). If seen on
cell block preparations. Correlation with clinical history, smear preparations, they can be confused with collagen
intraoperative findings, and corresponding surgical resection balls. The red or purple staining characteristic on
614 U. Krishnamurti et al.
Table 17.21 Cytologic features of benign and malignant mesothelial Papanicolaou stain and cracking artifact can help in their rec-
cells [197, 200, 206]
ognition (Fig. 17.30). The presence of psammoma bodies
Benign mesothelial cells Malignant mesothelial cells without associated epithelial cells does not indicate malig-
Uniform cells Morphologic variation nancy, as they can be seen in several benign/low-grade con-
Monolayer sheets Cell balls and papillary
ditions, such as endosalpingiosis, mesothelial hyperplasia,
fragments
Low N/C ratio High N/C ratio
ovarian serous cystadenomas, and serous borderline tumors
Round to oval uniform nuclei Nuclear pleomorphism [200]. They are also common findings in serous carcinomas
Evenly distributed fine chromatin Hyperchromasia and may even be the predominant findings, as seen in psam-
Inconspicuous to prominent Prominent macronucleolus mocarcinoma [212].
nucleolus Epithelial cells from benign serous ovarian tumors (cyst-
Absent or rare mitotic figures Mitotic figures common adenofibroma, cystadenoma) may be seen in PW if the tumor
Absent psammoma bodies Psammoma bodies common ruptures or involves the ovarian surface. As both tumors
N/C nuclear to cytoplasm characteristically show minimal epithelial proliferation and
small non-branching papillae, they have identical appear-
ance in PW, including simple papillae of cohesive orderly
and uniform columnar cells with cilia and minimal cytologic
atypia (Fig. 17.41). Psammoma bodies may be present.
Correlation with corresponding surgical specimen is essen-
tial to avoid interpretation of these cells as malignant.
a b
Fig. 17.39 (a) Cluster of benign endometrial glands on pelvic washing from a patient with ovarian surface endometriosis (ThinPrep, Papanicolaou
stain, 100×). (b) Corresponding cell block shows endometrial gland fragments and stroma, characteristic of endometriosis (cell bock, H&E, 200×)
17 Gynecologic Cytology 615
Fig. 17.40 Psammoma bodies are often seen on pelvic washings from Fig. 17.42 A “second population” of high-grade carcinoma cells often
patients with ovarian surface of peritoneal involvement by serous bor- “stand out” on low power from the background mesothelial cells. Large
derline tumors or serous carcinoma (cell block, H&E, 200×) three-dimensional groups of tumor cells are easy to appreciate
(ThinPrep, Papanicolaou stain, 20×)
Table 17.22 Cytologic features of low-grade serous tumors, high-grade serous tumors, and their mimickers [195, 197, 201, 211, 213]
Cytologic
features Endosalpingiosis Serous cystadenoma Serous borderline tumor High-grade serous carcinoma
Cellularity Low Low to moderate Moderate, variable Hypercellular
Cell Tightly cohesive small Cohesive, orderly Crowded and rare single Crowded and numerous single
arrangement groups
Papillae Simple, small, Simple, Branching, large, 3D Irregular, large, branching 3D
non-branching non-branching
Pleomorphism None to minimal None to minimal Uncommon, degenerative atypia Marked and obvious
Nuclear Low N/C ratio, fine Low N/C ratio, fine Variable N/C ratio, moderate Variation in size, shape, high
features chromatin, inconspicuous chromatin, 1–2 small atypia, coarse chromatin, rare N/C ratio, coarse chromatin,
nucleoli nucleoli small prominent nucleoli prominent nucleoli
Cytoplasm Scant basophilic Moderate basophilic Minimal cytoplasmic vacuoles Prominent cytoplasmic vacuoles
Mitosis Absent Absent Rare Common
Psammoma Common Sometimes Common Common
bodies
N/C nuclear to cytoplasm, 3D three-dimensional
a b
Fig. 17.43 (a) High-grade serous carcinoma of the ovary. Marked of markedly atypical carcinoma cells. Mitotic figure is noted (ThinPrep,
cytologic atypia, nuclear hyperchromasia, nucleomegaly, and promi- Papanicolaou stain, 200×). (c) High-grade serous carcinoma of the
nent nucleoli are readily seen (ThinPrep, Papanicolaou stain, 200×). (b) ovary. Marked cytologic atypia, nuclear hyperchromasia, nucleomeg-
High-grade serous carcinoma of the ovary. Three-dimensional cluster aly, and prominent nucleoli are readily seen (cell block, H&E, 200×)
17 Gynecologic Cytology 617
Table 17.23 Cytomorphologic characteristics and immunohistochemical features of the common metastatic gynecologic malignancies in pelvic
fluid cytology [199, 217, 222, 236–251]
Tumor type Architecture Nucleus Cytoplasm IHC Pitfalls and pearls
Ovarian surface epithelial tumors
High-grade Branching Irregular, pleomorphic, Vacuolated, “soap P53+ (80%) or 0 (null), WT-1+ in
serous carcinoma papillary, single hyperchromatic, bubble” p16+, WT-1+ mesothelial cells,
cells prominent nucleolus some ER+/−,
PR−/+, p53 wild
type in 20%
Mucinous Strips of tall Pleomorphism, Mucin displacing CK7+>>CK20+, Rule out
carcinoma mucinous cells, hyperchromasia, nucleus, signet ring CDX2+/−, mCEA+ PAX8, metastasis
3D clusters prominent nucleolus cells DPC4+
Endometrioid Cohesive Elongated, enlarged, Scant, fine vacuoles or ER+, PR+, PTEN−, High grade may
carcinoma clusters, acini less pleomorphic than granules, squamous or WT-1−, vimentin+, nuclear be p53+
HGSC, inconspicuous mucinous β catenin+
nucleoli differentiation
Clear cell Hyalinized Round nuclei, less Clear to eosinophilic, Napsin A+, HNF-1β+
carcinoma papillae, acini, pleomorphism than fine vacuoles, distinct
often single cells HGSC, prominent cell borders, hyaline
nucleolus globules
Germ cell tumors
Dysgerminoma Cellular, large Irregular, coarse Scant, delicate OCT-4, PLAP+, CD117+ D2-40+, AE1/
tumor cells, clumped chromatin, one vacuolated cytoplasm AE3+
single and or multiple nucleoli perinuclear dot
clusters like
Yolks sac tumor Papillary and Vesicular nuclei, 1–2 Scant vacuolated AFP+, SALL-4+,
cohesive groups prominent nucleoli cytoplasm, intracellular glypican-3+, PLAP+, AE1/
hyaline globules AE3+
Embryonal Tight clusters and Large pleomorphic, Scant delicate OCT-4+, CD30+, AE1/
carcinoma acini coarse chromatin, cytoplasm, indistinct AE3+, CD117−
prominent nucleoli cell borders
Immature Clusters of small Small round uniform, Delicate fibrillary AE1/AE3+, neuroendocrine
teratoma cohesive cells, neuroendocrine cytoplasm markers+, GFAP+
rare rosettes chromatin pattern
Sex cord tumors
Granulosa cell Small and large Eccentric Scant delicate Inhibin A+, calretinin+, Luteinized
tumor clusters hyperchromatic, EMA−, CD99+ stromal cells
(especially if granular chromatin, membranous, SF-1+ stain similarly,
ruptured) grooves, more mesothelial cells
pleomorphism in calretinin+
juvenile type
Sertoli-Leydig Tight clusters, Grade dependent: Clear abundant to scant Inhibin A+, calretinin+, Luteinized
cell tumor balls or papillae hyperchromatic with EMA−, CD99+ stromal cells
inconspicuous nucleoli (membranous), Melan A+, stain similar,
or sarcomatoid and SF-1+ mesothelial cells
pleomorphic calretinin+
Biphasic tumors
Carcinosarcoma Epithelial cells in Pleomorphic, Vacuolated, dependent AE1/AE3+, PAX8+ in
(MMMT) clusters, papillae, hyperchromatic, on epithelial epithelial component, AE1/
spindled cell prominent nucleolus, component. May have AE3+ in mesenchymal
component may rarely malignant heterologous features component, MyoD1,
be present spindled cells (rhabdoid, chondroid) myogenin for rhabdoid diff
3D three-dimensional, HGSC high-grade serous carcinoma, + positive, − negative, +/− can be positive, −/+ rarely positive, diff differentiation
ered positive in a low-stage, low-grade endometrioid carci- 17.14.6 Cytologic Diagnosis of Primary
noma. Iatrogenic contamination of the peritoneal cavity, Peritoneal Tumors
associated with robotic and laparoscopic hysterectomies,
and transit of tumor cells via fallopian tubes are other possi- Primary peritoneal lesions of the pelvic peritoneum encoun-
bilities to consider. tered in gynecologic pathology include various benign,
Carcinosarcomas often exfoliate the carcinomatous com- low malignant potential and malignant tumors of both
ponent, and sarcomatoid cell clusters with elongated bizarre mesothelial and Mullerian origin. Most primary Mullerian
nuclei are rarely seen [226]. tumors of the peritoneum are serous neoplasms, and occa-
17 Gynecologic Cytology 619
a b
Fig. 17.48 (a) Primary peritoneal malignant mesothelioma. Low (ThinPrep, Papanicolaou stain, 200×). (c) Primary peritoneal malignant
power shows large clusters and papillary fragments of mesothelial cells mesothelioma. Low power shows large clusters and papillary fragments
(ThinPrep, Papanicolaou stain, 100×). (b) Primary peritoneal malignant of mesothelial cells (cell block, H&E, 100×)
mesothelioma. Large papillary fragment of atypical mesothelial cells
significant number of tumor cells, and (2) the characteristic ings. Histologically, MM shows a cystic papillary or a diffuse
histologic features of adenomatoid tumor are not appreciated growth pattern. The diagnosis of MM on PW is challenging
on PW. The cytologic features of adenomatoid tumor have because invasion into adjacent normal structures is required
only been described in fine-needle aspiration specimens that for a definitive diagnosis and invasion cannot be evaluated in
show sheets and clusters of round to oval tumor cells with cytology [230]. In cellular specimens, MM usually presents
eccentric vesicular nuclei, fine chromatin, and vacuolated as large clusters, balls, or papillary fragments of atypical
cytoplasm [228, 229]. mesothelial cells (Fig. 17.48a–c). Cytologic atypia of meso-
Primary malignant mesothelioma (MM) of the perito- thelial cells is variable in MM. Compared to benign meso-
neum is rare. It is more often seen in males, with only around thelial cells, however, MM cells are larger, with higher
8% of cases encountered in women. History of asbestos nuclear to cytoplasmic ratio and prominent nucleoli. The
exposure is a known risk factor. Ascites and diffuse perito- mesothelial origin can be recognized based on the presence
neal involvement by tumor are the usual intraoperative find- of dense central and lacy peripheral cytoplasm, distinct cell
17 Gynecologic Cytology 621
borders, and intercellular windows [195, 231]. Binucleation, original tumor. On PW, residual carcinoma cells are often
multinucleation, and mitotic figures (including atypical arranged as single cells, but small papillary clusters may
forms) are readily identified in MM. The absence of a “for- be seen. The presence of psammoma bodies is variable,
eign” cell population and identification of spectrum of cyto- and they trigger additional workup (such as to perform
logic atypia ranging from benign/reactive to atypical to additional levels and IHC stains on cell block).
malignant are helpful features to distinguish MM from carci- Cytologically, carcinoma cells after chemotherapy show
nomas. The cytologic features of reactive mesothelial cells, bizarre degenerative changes, multinucleation, abundant
malignant mesothelioma, and adenocarcinoma are summa- vacuolated eosinophilic cytoplasm with an overall low
rized in Table 17.24. Rare variants of MM are challenging to nuclear to cytoplasmic ratio, smudgy clumped chromatin,
diagnose on a biopsy sample and even more challenging on and prominent nucleoli [234]. Foamy histiocytes, necrotic
fluid cytology. These include the sarcomatoid, lymphohistio- debris, cholesterol crystal, foreign body giant cells, and
cytoid, rhabdoid, and deciduoid variants. Their diagnosis acute inflammatory cells are also common findings. As
heavily relies on IHC and the awareness of these rare sub- residual malignant cells may be sparse and poorly pre-
types of MM [232]. served and appear bizarre and more atypical than the orig-
inal tumor, an undifferentiated malignancy or a second
primary tumor may even enter the differential diagnosis.
17.15 Cytologic Changes After Treatment Therefore, IHC stains may be of a great value to identify
residual microscopic disease in PW and to confirm their
Neoadjuvant chemotherapy followed by interval debulk- origin. A study by Wang et al. showed the IHC staining
ing surgery is widely used for advanced-stage high-grade profile of treated ovarian carcinoma cells is identical to
ovarian and primary peritoneal carcinomas. Clinical those prior to treatment [235]. It is important to note that
assessment of treatment response and need for additional tumor typing and grading are not possible on PW obtained
chemotherapy are based on histologic evaluation for during interval debulking surgery and should be deferred
residual disease on surgical specimen. Histologic features to the surgical specimen.
often seen in high-grade ovarian carcinomas following
neoadjuvant chemotherapy include degenerative changes
in tumor cells (such as with smudgy chromatin, bizarre 17.16 Ancillary Studies
atypia, and cytoplasmic vacuoles), numerous psammoma
bodies without associated viable tumor cells, fibrosis, Metastatic high-grade carcinoma is usually easy to diagnose
foamy histiocytes, necrosis, and acute inflammation on PW, as tumor cells “stand out” as a second (foreign) pop-
[233]. It is important to be familiar with these cytologic ulation of three-dimensional cellular aggregates with readily
changes, as residual tumor cells often do not resemble the recognizable cytologic features of malignancy. On the other
Table 17.24 Cytomorphologic characteristics of reactive mesothelial cells, mesothelioma, and adenocarcinoma on pelvic fluid specimen [194,
195, 197, 200, 201, 231]
Cytologic features Reactive mesothelial cells Mesothelioma Adenocarcinoma
Cellularity Moderately cellular Hypercellular Hypercellular
Cell arrangement Monolayer or small groups Large 3D cell groups with knobby Single cells, 3D cell groups with smooth
with knobby outline outline, papillae outline (community borders), true glands
Pleomorphism Minimal or absent Present, giant cells, multinucleated Variation in cell size and shape, bizarre
cells atypical cells
Intercellular Present Present Absent
windows
Cell borders Microvilli, cytoplasmic blebs Prominent microvilli, cytoplasmic Sharp cell borders
blebs
Cytoplasm Two-tone, degenerative Two-tone, fine vacuoles (lipid, Vacuolated (mucin)
vacuoles glycogen)
Nuclear features Reactive, uniform nuclei, Hyperchromatic, central, Eccentric, irregular, abnormal chromatin
prominent nucleoli may be pleomorphism, macronucleoli
seen
Distinct second Absent Spectrum of changes without obvious Distinct second population
population second population
Nuclear features of Absent Nuclear enlargement, prominent Nuclear enlargement, prominent nucleoli,
malignancy nucleoli, atypical mitosis may be seen atypical mitosis present
3D three-dimensional
622 U. Krishnamurti et al.
Table 17.25 Commonly used immunohistochemical markers and their staining pattern in pelvic fluid specimens [201, 232, 246, 252, 253]
Carcinoma Mesothelial
IHC marker cells cells Potential pitfalls
MOC-31 M
BerEP-4 M
B72.3 M or C
CD15 C or M Only 30–60% of ovarian carcinomas+
(Leu-M1)
CK7 C C Most primary ovarian carcinomas+
CK20 C Only rare ovarian carcinomas (mucinous) are CK20+
AE1/AE3 C C Not useful
EMA C MI
mCEA C Only + in mucinous carcinomas
CK5/6 C Some carcinomas are positive
Calretinin C and N Sex cord tumors+
WT-1 N N Stains surface epithelial cells from adnexa, small-cell carcinoma of hypercalcemic type,
some sex cord tumors
D2-40 M Lymphatic endothelial cells+
Thrombomodulin M Stain urothelial cells, some carcinomas+
ER, PR N 5% of mesothelial cells are ER+, <5% of mesothelial cells are PR+
PAX8 N N* Renal and thyroid tumors+, mesothelial cells from pelvic surface epithelium+
M membranous, C cytoplasmic, N nuclear, MI microvillous
N* PAX8 expression, especially with polyclonal antibody, has been reported in benign peritoneal mesothelial cells and mesothelioma [253]
2. Fackler MJ, McVeigh M, Mehrotra J, Blum MA, Lange J, Lapides specimen adequacy. New York, NY: Springer; 1994. https://doi.
A, et al. Quantitative multiplex methylation-specific PCR assay for org/10.1007/978-1-4684-0201-8.
the detection of promoter hypermethylation in multiple genes in 24. Solomon D, Nayar R. The Bethesda system for reporting cervical
breast cancer. Cancer Res. 2004;64(13):4442–52. cytology: definitions, criteria, and explanatory notes, vol. xxiii. 2nd
3. Gential HPV infection-CDC fact sheet 2017 [cited 2017 June ed. New York: Springer; 2004. p. 191.
29]. Available from: https://www.cdc.gov/std/hpv/hpv-factsheet- 25. Nayar R, Wilbur DC, SpringerLink (Online Service). The Bethesda
march-2017.pdf system for reporting cervical cytology definitions, criteria, and
4. Naylor B. The century for cytopathology. Acta Cytol. explanatory notes. https://doi.org/10.1007/978-3-319-11074-5.
2000;44(5):709–25. 26. Massad LS, Einstein MH, Huh WK, Katki HA, Kinney WK,
5. Babes A. Diagnostic du cancer du col utérin par les frottis (diag- Schiffman M, et al. 2012 updated consensus guidelines for the
nosis of cancer of the uterine cervix by smears). Presse Med. management of abnormal cervical cancer screening tests and can-
1928;29:451–4. cer precursors. J Low Genit Tract Dis. 2013;17(5 Suppl 1):S1–S27.
6. Diamantis A, Magiorkinis E, Androutsos G. What's in a name? 27. Lin SN, Taylor J, Alperstein S, Hoda R, Holcomb K. Does specu-
Evidence that Papanicolaou, not babes, deserves credit for the pap lum lubricant affect liquid-based Papanicolaou test adequacy?
test. Diagn Cytopathol. 2010;38(7):473–6. Cancer Cytopathol. 2014;122(3):221–6.
7. Papanicolaou G, Traut H. The diagnostic value of vaginal smears in 28. Feit TD, Mowry DA. Interference potential of personal lubricants
carcinoma of the uterus. Am J Obstet Gynecol. 1941;42:193–206. and vaginal medications on ThinPrep pap tests. J Am Board Fam
8. Chantziantoniou N. The wars against cervical cancer. J Am Soc Med. 2011;24(2):181–6.
Cytopathol. 2014;2:275–9. 29. Sweeney BJ, Haq Z, Happel JF, Weinstein B, Schneider
9. Papanicolaou G. Atlas of Exfoliative cytology. Cambridge: Harvard D. Comparison of the effectiveness of two liquid-based
University Press; 1954. Papanicolaou systems in the handling of adverse limiting factors,
10. Anderson GH, Boyes DA, Benedet JL, Le Riche JC, Matisic JP, such as excessive blood. Cancer. 2006;108(1):27–31.
Suen KC, et al. Organisation and results of the cervical cytology 30. Kenyon S, Sweeney BJ, Happel J, Marchilli GE, Weinstein B,
screening programme in British Columbia, 1955-85. Br Med J Schneider D. Comparison of BD Surepath and ThinPrep pap
(Clin Res Ed). 1988;296(6627):975–8. systems in the processing of mucus-rich specimens. Cancer
11. Christopherson WM, Scott MA. Trends in mortality from uterine Cytopathol. 2010;118(5):244–9.
cancer in relation to mass screening. Acta Cytol. 1977;21(1):5–9. 31. Eversole GM, Moriarty AT, Schwartz MR, Clayton AC, Souers R,
12. Aisner S, Austin M, Bishop J, et al. Cervical cytology practice Fatheree LA, et al. Practices of participants in the college of ameri-
guidelines. Acta Cytol. 2001;45(02):201–26. can pathologists interlaboratory comparison program in cervico-
13. Vooijs GP, Elias A, van der Graaf Y, Poelen-van de Berg M. The vaginal cytology, 2006. Arch Pathol Lab Med. 2010;134(3):331–5.
influence of sample takers on the cellular composition of cervical 32. Wilbur DC, Marianne PU, Nayar R. Computer-assisted interpreta-
smears. Acta Cytol. 1986;30(3):251–7. tion of cervical cytology. In: The Bethesda system for reporting cer-
14. McGoogan E, Colgan TJ, Ramzy I, Cochand-Priollet B, Davey DD, vical cytology definitions, criteria, and explanatory notes [Internet].
Grohs HK, et al. Cell preparation methods and criteria for sample SpringerLink (Online Service). vol. XXIV, 3rd ed. New York:
adequacy. International academy of cytology task force summary. Springer. p. 321. https://doi.org/10.1007/978-3-319-11074-5. 08
Diagnostic cytology towards the 21st century: an international illus. in color.
expert conference and tutorial. Acta Cytol. 1998;42(1):25–32. 33. Yancey M, Magelssen D, Demaurez A, Lee RB. Classification
15. Ho GY, Burk RD, Klein S, Kadish AS, Chang CJ, Palan P,
of endometrial cells on cervical cytology. Obstet Gynecol.
et al. Persistent genital human papillomavirus infection as a 1990;76(6):1000–5.
risk factor for persistent cervical dysplasia. J Natl Cancer Inst. 34. Fiorella RM, Cheng J, Kragel PJ. Papanicolaou smears in preg-
1995;87(18):1365–71. nancy. Positivity of exfoliated cells for human chorionic gonado-
16. Hingorani SR, Tuveson DA. In search of an early warning system tropin and human placental lactogen. Acta Cytol. 1993;37(4):
for pancreatic cancer. Cancer Biol Ther. 2003;2(1):84–6. 451–6.
17. NCCLS. Papanicolaou Technique; Approved Guideline. NCCLS 35. Crothers BA, Booth CN, Darragh TM, Means MM, Souers RJ,
Document 15-A (ISBN 1-56238-238-1) 1994;14(8). Thomas N, et al. Atrophic vaginitis: concordance and interpretation
18. Moyer VA. Force USPST. Screening for cervical cancer: U.S. pre- of slides in the College of American Pathologists Cervicovaginal
ventive services task force recommendation statement. Ann Intern Interlaboratory Comparison Program in Gynecologic
Med. 2012;156(12):880–91, W312 Cytopathology. Arch Pathol Lab Med. 2012;136(11):1332–8.
19. Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam 36. Shield PW, Daunter B, Wright RG. Post-irradiation cytology of cer-
SL, Cain J, et al. American Cancer Society, American Society for vical cancer patients. Cytopathology. 1992;3(3):167–82.
Colposcopy and Cervical Pathology, and American Society for 37. Colgan TJ, Woodhouse SL, Styer PE, Kennedy M, Davey
Clinical Pathology screening guidelines for the prevention and DD. Reparative changes and the false-positive/false-negative
early detection of cervical cancer. CA Cancer J Clin. 2012;62(3): Papanicolaou test: a study from the College of American Pathologists
147–72. Interlaboratory Comparison Program in Cervicovaginal Cytology.
20. Boon ME, de Graaff Guilloud JC, Rietveld WJ. Analysis of five Arch Pathol Lab Med. 2001;125(1):134–40.
sampling methods for the preparation of cervical smears. Acta 38. Halford JA. Cytological features of chronic follicular cervi-
Cytol. 1989;33(6):843–8. citis in liquid-based specimens: a potential diagnostic pitfall.
21. Hong SM, Kelly D, Griffith M, Omura N, Li A, Li CP, et al. Multiple Cytopathology. 2002;13(6):364–70.
genes are hypermethylated in intraductal papillary mucinous neo- 39. Videlefsky A, Grossl N, Denniston M, Sehgal R, Lane JM,
plasms of the pancreas. Mod Pathol. 2008;21(12):1499–507. Goodenough G. Routine vaginal cuff smear testing in post-
22. Hu Y, Xie H, Liu Y, Liu W, Liu M, Tang H. miR-484 suppresses pro- hysterectomy patients with benign uterine conditions: when is it
liferation and epithelial-mesenchymal transition by targeting ZEB1 indicated? J Am Board Fam Pract. 2000;13(4):233–8.
and SMAD2 in cervical cancer cells. Cancer Cell Int. 2017;17:36. 40. Nayar R, Wilbur DC. The pap test and Bethesda 2014. Cancer
23. Kurman RJ, Solomon D, SpringerLink (Online Service). The
Cytopathol. 2015;123(5):271–81.
Bethesda system for reporting cervical/vaginal cytologic diagno- 41. Barlow JF. The 1988 Bethesda system for cervical vaginal cytol-
ses definitions, criteria, and explanatory notes for terminology and ogy: pros and cons. S D J Med. 1991;44(6):157–8.
17 Gynecologic Cytology 625
42. Solomon D, Davey D, Kurman R, Moriarty A, O’Connor D, Prey 61. Lee KR, Minter LJ, Granter SR. Papanicolaou smear sensitivity for
M, et al. The 2001 Bethesda system: terminology for reporting adenocarcinoma in situ of the cervix. A study of 34 cases. Am J
results of cervical cytology. JAMA. 2002;287(16):2114–9. Clin Pathol. 1997;107(1):30–5.
43. Gupta N, Srinivasan R, Nijhawan R, Rajwanshi A, Dey P, Suri V, 62. Roberts JM, Thurloe JK. Comparative sensitivities of ThinPrep
et al. Atypical squamous cells and low-grade squamous intraepi- and Papanicolaou smear for adenocarcinoma in situ (AIS) and
thelial lesion in cervical cytology: cytohistological correla- combined AIS/high-grade squamous intraepithelial lesion (HSIL):
tion and implication for management in a low-resource setting. comparison with HSIL. Cancer. 2007;111(6):482–6.
Cytopathology. 2011;22(3):189–94. 63. Attipoe MF, Sturgis CD. Small cell carcinoma of the uterine cervix
44. Jones BA, Davey DD. Quality management in gynecologic cytol- in a pregnant patient diagnosed with liquid based cytology and cell
ogy using interlaboratory comparison. Arch Pathol Lab Med. block immunocytochemistry. Case Rep Pathol. 2014;2014:971464.
2000;124(5):672–81. 64. Park HJ, Choi YM, Chung CK, Lee SH, Yim GW, Kim SW, et al.
45. Wright TC Jr, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ, Pap smear screening for small cell carcinoma of the uterine cer-
Conference AS-sC. 2001 consensus guidelines for the management vix: a case series and review of the literature. J Gynecol Oncol.
of women with cervical cytological abnormalities. J Low Genit 2011;22(1):39–43.
Tract Dis. 2002;6(2):127–43. 65. Setia N, Goulart RA, Leiman G, Otis CN, Modem R, Pantanowtiz
46. Nayar R, Wilbur DC. The pap test and Bethesda 2014. “the reports L. Cytomorphology of cervicovaginal melanoma: ThinPrep versus
of my demise have been greatly exaggerated.” (After a quotation conventional Papanicolaou tests. Cytojournal. 2010;7:25.
from mark twain). Acta Cytol. 2015;59(2):121–32. 66. Hanley KZ, Tadros TS, Briones AJ, Birdsong GG, Mosunjac
47. Levine PH, Elgert PA, Sun P, Simsir A. Atypical repair on
MB. Hematologic malignancies of the female genital tract diagnosed
pap smears: clinicopathologic correlates in 647 cases. Diagn on liquid-based pap test: Cytomorphologic features and review of
Cytopathol. 2005;33(3):214–7. differential diagnoses. Diagn Cytopathol. 2009;37(1):61–7.
48. Yelverton CL, Bentley RC, Olenick S, Krigman HR, Johnston WW, 67. Hanley KZ, Oprea-Ilies G, Ormenisan C, Seydafkan S, Mosunjac
Robboy SJ. Epithelial repair of the uterine cervix: assessment of MB. Atypical findings on cervicovaginal smears correlate with
morphologic features and correlations with cytologic diagnosis. Int cervical involvement by malignant mixed Mullerian tumors of the
J Gynecol Pathol. 1996;15(4):338–44. uterus. Acta Cytol. 2015;59(4):319–24.
49. Abati A, Jaffurs W, Wilder AM. Squamous atypia in the atrophic 68. Giordano G, Gnetti L, Pilato FP, Viviano L, Silini EM. The role
cervical vaginal smear: a new look at an old problem. Cancer. of cervical smear in the diagnosis and management of extrauter-
1998;84(4):218–25. ine malignancies metastatic to the cervix: three case reports. Diagn
50. Keating JT, Wang HH. Significance of a diagnosis of atypical Cytopathol. 2010;38(1):41–6.
squamous cells of undetermined significance for Papanicolaou 69. House MG, Guo M, Iacobuzio-Donahue C, Herman JG. Molecular
smears in perimenopausal and postmenopausal women. Cancer. progression of promoter methylation in intraductal papillary
2001;93(2):100–5. mucinous neoplasms (IPMN) of the pancreas. Carcinogenesis.
51. Acs G, Gupta PK, Baloch ZW. Glandular and squamous atypia and 2003;24(2):193–8.
intraepithelial lesions in atrophic cervicovaginal smears. One insti- 70. Biscotti CV, Dawson AE, Dziura B, Galup L, Darragh T, Rahemtulla
tution’s experience. Acta Cytol. 2000;44(4):611–7. A, et al. Assisted primary screening using the automated ThinPrep
52. Kubba LA, Patel K, Du H, Hahn EA, Sturgis CD. Atypical para- imaging system. Am J Clin Pathol. 2005;123(2):281–7.
keratotic spires and HCII HPV results: correlation in liquid-based 71. Schledermann D, Hyldebrandt T, Ejersbo D, Hoelund B. Automated
cervicovaginal cytology specimens interpreted as ASC-US. Diagn screening versus manual screening: a comparison of the ThinPrep
Cytopathol. 2007;35(8):476–81. imaging system and manual screening in a time study. Diagn
53.
Abramovich CM, Wasman JK, Siekkinen P, Abdul-Karim Cytopathol. 2007;35(6):348–52.
FW. Histopathologic correlation of atypical parakeratosis diag- 72. Dawson AE. Can we change the way we screen? The ThinPrep
nosed on cervicovaginal cytology. Acta Cytol. 2003;47(3):405–9. imaging system. Cancer. 2004;102(6):340–4.
54. Darragh TM, Colgan TJ, Thomas Cox J, Heller DS, Henry MR, Luff 73. Chivukula M, Saad RS, Elishaev E, White S, Mauser N, Dabbs
RD, et al. The lower anogenital squamous terminology standardiza- DJ. Introduction of the thin prep imaging system (TIS): experience
tion project for HPV-associated lesions: background and consensus in a high volume academic practice. Cytojournal. 2007;4:6.
recommendations from the College of American Pathologists and 74. Lozano R. Comparison of computer-assisted and manual screening
the American Society for Colposcopy and Cervical Pathology. Int J of cervical cytology. Gynecol Oncol. 2007;104(1):134–8.
Gynecol Pathol. 2013;32(1):76–115. 75. Huh WK, Ault KA, Chelmow D, Davey DD, Goulart RA, Garcia
55. Reagan JW, Hamonic MJ. Dysplasia of the uterine cervix. Ann N Y FA, et al. Use of primary high-risk human papillomavirus testing
Acad Sci. 1956;63(6):1236–44. for cervical cancer screening: interim clinical guidance. J Low
56. Miller RA, Mody DR, Tams KC, Thrall MJ. Glandular lesions of Genit Tract Dis. 2015;19(2):91–6.
the cervix in clinical practice: a cytology, histology, and human 76. Lee JS, Kuan L, Oh S, Patten FW, Wilbur DC. A feasibility study
papillomavirus correlation study from 2 institutions. Arch Pathol of the AutoPap system location-guided screening. Acta Cytol.
Lab Med. 2015;139(11):1431–6. 1998;42(1):221–6.
57. Selvaggi SM. Glandular epithelial abnormalities on thinprep(R) 77. Wilbur DC, Prey MU, Miller WM, Pawlick GF, Colgan TJ. The
pap tests: clinical and cytohistologic correlation. Diagn Cytopathol. AutoPap system for primary screening in cervical cytology.
2016;44(5):389–93. Comparing the results of a prospective, intended-use study with
58. Simsir A, Hwang S, Cangiarella J, Elgert P, Levine P, Sheffield MV, routine manual practice. Acta Cytol. 1998;42(1):214–20.
et al. Glandular cell atypia on Papanicolaou smears: interobserver 78. Troni GM, Cariaggi MP, Bulgaresi P, Houssami N, Ciatto
variability in the diagnosis and prediction of cell of origin. Cancer. S. Reliability of sparing Papanicolaou test conventional reading in
2003;99(6):323–30. cases reported as no further review at AutoPap-assisted cytologi-
59. Ajit D, Gavas S, Joseph S, Rekhi B, Deodhar K, Kane
cal screening: survey of 30,658 cases with follow-up cytological
S. Identification of atypical glandular cells in pap smears: is it a hit screening. Cancer. 2007;111(2):93–8.
and miss scenario. Acta Cytol. 2013;57(1):45–53. 79. Wilbur DC, Black-Schaffer WS, Luff RD, Abraham KP, Kemper
60. Kaferle JE, Malouin JM. Evaluation and management of the AGUS C, Molina JT, et al. The Becton Dickinson FocalPoint GS imaging
Papanicolaou smear. Am Fam Physician. 2001;63(11):2239–44. system: clinical trials demonstrate significantly improved sensitiv-
626 U. Krishnamurti et al.
ity for the detection of important cervical lesions. Am J Clin Pathol. 96. Castle PE, Eaton B, Reid J, Getman D, Dockter J. Comparison of
2009;132(5):767–75. human papillomavirus detection by Aptima HPV and cobas HPV
80. Sweeney BJ, Wilbur DC. Effects on cervical cytology screening tests in a population of women referred for colposcopy following
productivity associated with implementation of the BD FocalPoint detection of atypical squamous cells of undetermined significance
guided screener imaging system. Acta Cytol. 2013;57(2): by pap cytology. J Clin Microbiol. 2015;53(4):1277–81.
147–52. 97. Iftner T, Becker S, Neis KJ, Castanon A, Iftner A, Holz B, et al.
81. Deftereos G, Kiviat NB. Human papillomaviruses: cervical cancer Head-to-head comparison of the RNA-based Aptima human pap-
and warts. In: Kaslow RASL, Le Duc JW, editors. Viral Infections illomavirus (HPV) assay and the DNA-based hybrid capture 2
in humans. 5th ed. New York, NY: Springer; 2014. p. 1063–104. HPV test in a routine screening population of women aged 30 to
82. Deftereos G, Kiviat NB. Detection and clinical management of 60 years in Germany. J Clin Microbiol. 2015;53(8):2509–16.
cervical pathology in the era of HPV. Curr Obstet Gynecol Rep. 98. Nolte FS, Ribeiro-Nesbitt DG. Comparison of the Aptima
2014;3(2):107–15. and Cervista tests for detection of high-risk human papillo-
83. Massad LS, Einstein MH, Huh WK, Katki HA, Kinney WK,
mavirus in cervical cytology specimens. Am J Clin Pathol.
Schiffman M, et al. 2012 updated consensus guidelines for the 2014;142(4):561–6.
management of abnormal cervical cancer screening tests and can- 99. Muangto T, Chanthasenanont A, Lertvutivivat S, Nanthakomon T,
cer precursors. Obstet Gynecol. 2013;121(4):829–46. Pongrojpaw D, Bhamarapravatana K, et al. Experience of com-
84. Mahajan A. Practical issues in the application of p16 immunohisto- bined liquid based cervical cytology and high-risk HPV mRNA
chemistry in diagnostic pathology. Hum Pathol. 2016;51:64–74. for cervical cancer screening in Thammasat University Hospital.
85. Kerr DA, Sweeney B, Arpin RN 3rd, Ring M, Pitman MB, Wilbur Asian Pac J Cancer Prev. 2016;17(9):4409–13.
DC, et al. Automated extraction of formalin-fixed, paraffin- 100. Virtanen E, Kalliala I, Dyba T, Nieminen P, Auvinen
embedded tissue for high-risk human papillomavirus testing of E. Performance of mRNA- and DNA-based high-risk human
head and neck squamous cell carcinomas using the Roche Cobas papillomavirus assays in detection of high-grade cervical lesions.
4800 system. Arch Pathol Lab Med. 2016;140(8):844–8. Acta Obstet Gynecol Scand. 2017;96(1):61–8.
86. Pettus JR, Wilson TL, Steinmetz HB, Lefferts JA, Tafe LJ. Utility 101. Poljak M, Kocjan BJ, Ostrbenk A, Seme K. Commercially avail-
of the Roche Cobas 4800 for detection of high-risk human papillo- able molecular tests for human papillomaviruses (HPV): 2015
mavirus in formalin-fixed paraffin-embedded oropharyngeal squa- update. J Clin Virol. 2016;76 Suppl 1:S3–S13.
mous cell carcinoma. Exp Mol Pathol. 2017;102(1):47–9. 102. Kadish AS, Hagan RJ, Ritter DB, Goldberg GL, Romney SL,
87. Clifford GM, Vaccarella S, Franceschi S, Tenet V, Umulisa MC, Kanetsky PA, et al. Biologic characteristics of specific human
Tshomo U, et al. Comparison of two widely used human papilloma- papillomavirus types predicted from morphology of cervical
virus detection and genotyping methods, GP5+/6+-based PCR fol- lesions. Hum Pathol. 1992;23(11):1262–9.
lowed by reverse line blot hybridization and multiplex type-specific 103. Lorincz AT, Reid R, Jenson AB, Greenberg MD, Lancaster W,
E7-based PCR. J Clin Microbiol. 2016;54(8):2031–8. Kurman RJ. Human papillomavirus infection of the cervix: rela-
88. Remmerbach TW, Brinckmann UG, Hemprich A, Chekol M,
tive risk associations of 15 common anogenital types. Obstet
Kuhndel K, Liebert UG. PCR detection of human papillomavirus Gynecol. 1992;79(3):328–37.
of the mucosa: comparison between MY09/11 and GP5+/6+ primer 104. Bergeron C, Barrasso R, Beaudenon S, Flamant P, Croissant O,
sets. J Clin Virol. 2004;30(4):302–8. Orth G. Human papillomaviruses associated with cervical intraep-
89. Manos MM, Ting Y, Wright DK, Lewis AJ, Broker TR, Wolinsky ithelial neoplasia. Great diversity and distinct distribution in low-
SM. The use of polymerase chain reaction amplification for the and high-grade lesions. Am J Surg Pathol. 1992;16(7):641–9.
detection of genital human papillomavirus. In: Furth M, Greaves 105. Campion MJ, McCance DJ, Cuzick J, Singer A. Progressive
M, editors. Molecular diagnostics of human cancer. Cold Spring potential of mild cervical atypia: prospective cytological, col-
Harbor, NY: Cold Spring Harbor Laboratory Press; 1989. p. 209–14. poscopic, and virological study. Lancet. 1986;2(8501):237–40.
90. Kleter B, van Doorn LJ, Schrauwen L, Molijn A, Sastrowijoto S, ter 106. Kataja V, Syrjanen K, Syrjanen S, Mantyjarvi R, Yliskoski M,
Schegget J, et al. Development and clinical evaluation of a highly Saarikoski S, et al. Prospective follow-up of genital HPV infec-
sensitive PCR-reverse hybridization line probe assay for detec- tions: survival analysis of the HPV typing data. Eur J Epidemiol.
tion and identification of anogenital human papillomavirus. J Clin 1990;6(1):9–14.
Microbiol. 1999;37(8):2508–17. 107. Mirabello L, Schiffman M, Ghosh A, Rodriguez AC, Vasiljevic
91. Phillips S, Cornall AM, Machalek DA, Garland SM, Bateson D, N, Wentzensen N, et al. Elevated methylation of HPV16 DNA is
Garefalakis M, et al. Comparison of the Roche Cobas(R) 4800 HPV associated with the development of high grade cervical intraepi-
assay to Roche Amplicor for detection of high-risk human papillo- thelial neoplasia. Int J Cancer. 2013;132(6):1412–22.
mavirus. Eur J Clin Microbiol Infect Dis. 2016;35(8):1305–7. 108. Vasiljevic N, Scibior-Bentkowska D, Brentnall A, Cuzick J,
92. Barbieri D, Nocera M, Gallinella G, Gentilomi GA, Plazzi M, Costa Lorincz A. A comparison of methylation levels in HPV18, HPV31
S, et al. Comparison of HPV sign genotyping test with INNO-LiPA and HPV33 genomes reveals similar associations with cervical
HPV genotyping extra assay on histologic and cytologic cervical precancers. J Clin Virol. 2014;59(3):161–6.
specimens. Diagn Microbiol Infect Dis. 2012;74(1):43–8. 109. Wentzensen N, Sun C, Ghosh A, Kinney W, Mirabello L,
93. Feng Q, Cherne S, Winer RL, Balasubramanian A, Lee SK, Hawes Wacholder S, et al. Methylation of HPV18, HPV31, and HPV45
SE, et al. Development and evaluation of a liquid bead microar- genomes and cervical intraepithelial neoplasia grade 3. J Natl
ray assay for genotyping genital human papillomaviruses. J Clin Cancer Inst. 2012;104(22):1738–49.
Microbiol. 2009;47(3):547–53. 110. Lorincz AT, Brentnall AR, Vasiljevic N, Scibior-Bentkowska D,
94. Ge Y, Christensen P, Luna E, Armylagos D, Schwartz MR, Mody Castanon A, Fiander A, et al. HPV16 L1 and L2 DNA meth-
DR. Performance of Aptima and Cobas HPV testing platforms ylation predicts high-grade cervical intraepithelial neoplasia in
in detecting high-grade cervical dysplasia and cancer. Cancer women with mildly abnormal cervical cytology. Int J Cancer.
Cytopathol. 2017;125(8):652–7. 2013;133(3):637–44.
95. Szarewski A, Mesher D, Cadman L, Austin J, Ashdown-Barr L, Ho 111. Lorincz AT. Cancer diagnostic classifiers based on quantitative
L, et al. Comparison of seven tests for high-grade cervical intraepi- DNA methylation. Expert Rev Mol Diagn. 2014;14(3):293–305.
thelial neoplasia in women with abnormal smears: the predictors 2 112. Hesselink AT, Heideman DA, Steenbergen RD, Gok M, van
study. J Clin Microbiol. 2012;50(6):1867–73. Kemenade FJ, Wilting SM, et al. Methylation marker analy-
17 Gynecologic Cytology 627
sis of self-sampled cervico-vaginal lavage specimens to triage 127. Lee JW, Choi CH, Choi JJ, Park YA, Kim SJ, Hwang SY, et al.
high-risk HPV-positive women for colposcopy. Int J Cancer. Altered MicroRNA expression in cervical carcinomas. Clin
2014;135(4):880–6. Cancer Res. 2008;14(9):2535–42.
113. Verhoef VM, Heideman DA, van Kemenade FJ, Rozendaal L, 128. Gocze K, Gombos K, Juhasz K, Kovacs K, Kajtar B, Benczik M,
Bosgraaf RP, Hesselink AT, et al. Methylation marker analysis and et al. Unique microRNA expression profiles in cervical cancer.
HPV16/18 genotyping in high-risk HPV positive self-sampled Anticancer Res. 2013;33(6):2561–7.
specimens to identify women with high grade CIN or cervical 129. Wilting SM, Snijders PJ, Verlaat W, Jaspers A, van de Wiel MA,
cancer. Gynecol Oncol. 2014;135(1):58–63. van Wieringen WN, et al. Altered microRNA expression associ-
114. Kalantari M, Osann K, Calleja-Macias IE, Kim S, Yan B, Jordan ated with chromosomal changes contributes to cervical carcino-
S, et al. Methylation of human papillomavirus 16, 18, 31, and genesis. Oncogene. 2013;32(1):106–16.
45 L2 and L1 genes and the cellular DAPK gene: considerations 130. Nagamitsu Y, Nishi H, Sasaki T, Takaesu Y, Terauchi F, Isaka
for use as biomarkers of the progression of cervical neoplasia. K. Profiling analysis of circulating microRNA expression in cer-
Virology. 2014;448:314–21. vical cancer. Mol Clin Oncol. 2016;5(1):189–94.
115. Lorincz AT, Brentnall AR, Scibior-Bentkowska D, Reuter C, 131. Sharma S, Hussain S, Soni K, Singhal P, Tripathi R,
Banwait R, Cadman L, et al. Validation of a DNA methyla- Ramachandran VG, et al. Novel MicroRNA signatures in HPV-
tion HPV triage classifier in a screening sample. Int J Cancer. mediated cervical carcinogenesis in Indian women. Tumour Biol.
2016;138(11):2745–51. 2016;37(4):4585–95.
116. Hansel A, Steinbach D, Greinke C, Schmitz M, Eiselt J, 132. Jia W, Wu Y, Zhang Q, Gao GE, Zhang C, Xiang Y. Expression
Scheungraber C, et al. A promising DNA methylation signature profile of circulating microRNAs as a promising fingerprint
for the triage of high-risk human papillomavirus DNA-positive for cervical cancer diagnosis and monitoring. Mol Clin Oncol.
women. PLoS One. 2014;9(3):e91905. 2015;3(4):851–8.
117. Vasiljevic N, Scibior-Bentkowska D, Brentnall AR, Cuzick 133. Li MY, Hu XX. Meta-analysis of microRNA expression profiling
J, Lorincz AT. Credentialing of DNA methylation assays for studies in human cervical cancer. Med Oncol. 2015;32(6):510.
human genes as diagnostic biomarkers of cervical intraepithe- 134. Deftereos G, Corrie SR, Feng Q, Morihara J, Stern J, Hawes SE,
lial neoplasia in high-risk HPV positive women. Gynecol Oncol. et al. Expression of mir-21 and mir-143 in cervical specimens
2014;132(3):709–14. ranging from histologically normal through to invasive cervical
118. Hesselink AT, Heideman DA, Steenbergen RD, Coupe VM, cancer. PLoS One. 2011;6(12):e28423.
Overmeer RM, Rijkaart D, et al. Combined promoter methylation 135. Dasari S, Wudayagiri R, Valluru L. Cervical cancer: biomarkers
analysis of CADM1 and MAL: an objective triage tool for high- for diagnosis and treatment. Clin Chim Acta. 2015;445:7–11.
risk human papillomavirus DNA-positive women. Clin Cancer 136. Yang P, Chen N, Yang D, Crane J, Yang S, Wang H, et al. The
Res. 2011;17(8):2459–65. ratio of serum Angiopoietin-1 to Angiopoietin-2 in patients with
119. Verhoef VM, Bosgraaf RP, van Kemenade FJ, Rozendaal L, cervical cancer is a valuable diagnostic and prognostic biomarker.
Heideman DA, Hesselink AT, et al. Triage by methylation- PeerJ. 2017;5:e3387.
marker testing versus cytology in women who test HPV-positive 137. Beyer S, Zhu J, Mayr D, Kuhn C, Schulze S, Hofmann S, et al.
on self-collected cervicovaginal specimens (PROHTECT-3): Histone H3 acetyl K9 and histone H3 tri methyl K4 as prog-
a randomised controlled non-inferiority trial. Lancet Oncol. nostic markers for patients with cervical cancer. Int J Mol Sci.
2014;15(3):315–22. 2017;18(3) https://doi.org/10.3390/ijms18030477.
120. Eijsink JJ, Lendvai A, Deregowski V, Klip HG, Verpooten G, 138. Yang K, Xia B, Wang W, Cheng J, Yin M, Xie H, et al. A compre-
Dehaspe L, et al. A four-gene methylation marker panel as tri- hensive analysis of metabolomics and transcriptomics in cervical
age test in high-risk human papillomavirus positive patients. Int J cancer. Sci Rep. 2017;7:43353.
Cancer. 2012;130(8):1861–9. 139. Institute NC. Surveillance, epidemiology, and end results pro-
121. Lai HC, Ou YC, Chen TC, Huang HJ, Cheng YM, Chen CH, et al. gram Bethesda, MD 2017 [cited 2017 7/24/2017]. Available from:
PAX1/SOX1 DNA methylation and cervical neoplasia detection: a https://seer.cancer.gov/statfacts/html/cervix.html.
Taiwanese Gynecologic Oncology Group (TGOG) study. Cancer 140. Fahey MT, Irwig L, Macaskill P. Meta-analysis of pap test accu-
Med. 2014;3(4):1062–74. racy. Am J Epidemiol. 1995;141(7):680–9.
122. Wilhelm T. Phenotype prediction based on genome-wide DNA 141. Cuzick J, Szarewski A, Terry G, Ho L, Hanby A, Maddox P, et al.
methylation data. BMC Bioinformatics. 2014;15:193. Human papillomavirus testing in primary cervical screening.
123. Su PH, Hsu YW, Huang RL, Weng YC, Wang HC, Chen YC, Lancet. 1995;345(8964):1533–6.
et al. Methylomics of nitroxidative stress on precancerous 142. Sedlacek TV. Cost-effectiveness of methods to enhance sensitivity
cells reveals DNA methylation alteration at the transition from of Papanicolaou testing. JAMA. 1999;282:1419–20.
in situ to invasive cervical cancer. Oncotarget. 2017;8(39): 143. Joste NE, Rushing L, Granados R, Zitz JC, Genest DR, Crum CP,
65281–91. et al. Bethesda classification of cervicovaginal smears: reproduc-
124. De Vuyst H, Franceschi S, Plummer M, Mugo NR, Sakr SR, ibility and viral correlates. Hum Pathol. 1996;27(6):581–5.
Meijer CJ, et al. Methylation levels of CADM1, MAL, and 144. Ismail SM, Colclough AB, Dinnen JS, Eakins D, Evans DM,
MIR124-2 in cervical scrapes for triage of HIV-infected, high-risk Gradwell E, et al. Observer variation in histopathological diag-
HPV-positive women in Kenya. J Acquir Immune Defic Syndr. nosis and grading of cervical intraepithelial neoplasia. BMJ.
2015;70(3):311–8. 1989;298(6675):707–10.
125. De Strooper LM, Hesselink AT, Berkhof J, Meijer CJ, Snijders 145. Dalla Palma P, Giorgi Rossi P, Collina G, Buccoliero AM,
PJ, Steenbergen RD, et al. Combined CADM1/MAL methyla- Ghiringhello B, Gilioli E, et al. The reproducibility of CIN
tion and cytology testing for colposcopy triage of high-risk HPV- diagnoses among different pathologists: data from histology
positive women. Cancer Epidemiol Biomark Prev. 2014;23(9): reviews from a multicenter randomized study. Am J Clin Pathol.
1933–7. 2009;132(1):125–32.
126. Chen J, Yao D, Li Y, Chen H, He C, Ding N, et al. Serum 146. Miller AB. An epidemiological perspective on cancer screening.
microRNA expression levels can predict lymph node metastasis in Clin Biochem. 1995;28(1):41–8.
patients with early-stage cervical squamous cell carcinoma. Int J 147. NIH releases consensus statement on cervical cancer. Am Fam
Mol Med. 2013;32(3):557–67. Physician. 1996;54(7):2310, 2315–6.
628 U. Krishnamurti et al.
148. Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam 163. Carozzi F, Visioli CB, Confortini M, Iossa A, Mantellini P,
SL, Cain J, et al. American Cancer Society, American Society Burroni E, et al. hr-HPV testing in the follow-up of women with
for Colposcopy and Cervical Pathology, and American Society cytological abnormalities and negative colposcopy. Br J Cancer.
for Clinical Pathology screening guidelines for the preven- 2013;109(7):1766–74.
tion and early detection of cervical cancer. Am J Clin Pathol. 164. Dillner J, Rebolj M, Birembaut P, Petry KU, Szarewski A, Munk
2012;137(4):516–42. C, et al. Long term predictive values of cytology and human pap-
149. Kulasingam SL, Kim JJ, Lawrence WF, Mandelblatt JS, Myers illomavirus testing in cervical cancer screening: joint European
ER, Schiffman M, et al. Cost-effectiveness analysis based on cohort study. BMJ. 2008;337:a1754.
the atypical squamous cells of undetermined significance/low- 165. Liaw KL, Hildesheim A, Burk RD, Gravitt P, Wacholder S, Manos
grade squamous intraepithelial lesion triage study (ALTS). J Natl MM, et al. A prospective study of human papillomavirus (HPV)
Cancer Inst. 2006;98(2):92–100. type 16 DNA detection by polymerase chain reaction and its asso-
150. Arbyn M, Sasieni P, Meijer CJ, Clavel C, Koliopoulos G, Dillner ciation with acquisition and persistence of other HPV types. J
J. Chapter 9: clinical applications of HPV testing: a summary of Infect Dis. 2001;183(1):8–15.
meta-analyses. Vaccine. 2006;24(Suppl 3):S3/78–89. 166. Liaw KL, Glass AG, Manos MM, Greer CE, Scott DR, Sherman
151. Arbyn M, Roelens J, Simoens C, Buntinx F, Paraskevaidis E, M, et al. Detection of human papillomavirus DNA in cytologically
Martin-Hirsch PP, et al. Human papillomavirus testing versus normal women and subsequent cervical squamous intraepithelial
repeat cytology for triage of minor cytological cervical lesions. lesions. J Natl Cancer Inst. 1999;91(11):954–60.
Cochrane Database Syst Rev. 2013;(3):Cd008054. https://doi. 167. Rodriguez AC, Schiffman M, Herrero R, Hildesheim A, Bratti
org/10.1002/14651858.CD008054.pub2. C, Sherman ME, et al. Longitudinal study of human papilloma-
152. Carozzi FM, Confortini M, Cecchini S, Bisanzi S, Cariaggi MP, virus persistence and cervical intraepithelial neoplasia grade
Pontenani G, et al. Triage with human papillomavirus testing of 2/3: critical role of duration of infection. J Natl Cancer Inst.
women with cytologic abnormalities prompting referral for col- 2010;102(5):315–24.
poscopy assessment. Cancer. 2005;105(1):2–7. 168. Nobbenhuis MA, Walboomers JM, Helmerhorst TJ, Rozendaal
153. Arbyn M, Roelens J, Cuschieri K, Cuzick J, Szarewski A, Ratnam L, Remmink AJ, Risse EK, et al. Relation of human papilloma-
S, et al. The APTIMA HPV assay versus the hybrid capture 2 virus status to cervical lesions and consequences for cervical-
test in triage of women with ASC-US or LSIL cervical cytol- cancer screening: a prospective study. Lancet. 1999;354(9172):
ogy: a meta-analysis of the diagnostic accuracy. Int J Cancer. 20–5.
2013;132(1):101–8. 169. Kjaer SK, Frederiksen K, Munk C, Iftner T. Long-term absolute
154. Aministration USFaD. FDA approves first human papillomavirus risk of cervical intraepithelial neoplasia grade 3 or worse follow-
test for primary cervical cancer screening Silver Spring, MD2014 ing human papillomavirus infection: role of persistence. J Natl
[cited 2017 7/24/2017]. Available from: https://wayback.archive- Cancer Inst. 2010;102(19):1478–88.
it.org/7993/20170112222903/http://www.fda.gov/NewsEvents/ 170. Koutsky LA, Holmes KK, Critchlow CW, Stevens CE, Paavonen
Newsroom/PressAnnouncements/ucm394773.htm. J, Beckmann AM, et al. A cohort study of the risk of cervical
155. Bulkmans NW, Berkhof J, Rozendaal L, van Kemenade FJ, Boeke intraepithelial neoplasia grade 2 or 3 in relation to papillomavirus
AJ, Bulk S, et al. Human papillomavirus DNA testing for the infection. N Engl J Med. 1992;327(18):1272–8.
detection of cervical intraepithelial neoplasia grade 3 and cancer: 171. Trottier H, Mahmud SM, Lindsay L, Jenkins D, Quint W, Wieting
5-year follow-up of a randomised controlled implementation trial. SL, et al. Persistence of an incident human papillomavirus infec-
Lancet. 2007;370(9601):1764–72. tion and timing of cervical lesions in previously unexposed
156. Mayrand MH, Duarte-Franco E, Rodrigues I, Walter SD, young women. Cancer Epidemiol Biomark Prev. 2009;18(3):
Hanley J, Ferenczy A, et al. Human papillomavirus DNA versus 854–62.
Papanicolaou screening tests for cervical cancer. N Engl J Med. 172. Schlecht NF, Kulaga S, Robitaille J, Ferreira S, Santos M,
2007;357(16):1579–88. Miyamura RA, et al. Persistent human papillomavirus infec-
157. Naucler P, Ryd W, Tornberg S, Strand A, Wadell G, Elfgren K, tion as a predictor of cervical intraepithelial neoplasia. JAMA.
et al. Human papillomavirus and Papanicolaou tests to screen for 2001;286(24):3106–14.
cervical cancer. N Engl J Med. 2007;357(16):1589–97. 173. Jaisamrarn U, Castellsague X, Garland SM, Naud P, Palmroth J,
158. Ronco G, Giorgi-Rossi P, Carozzi F, Dalla Palma P, Del Mistro A, Del Rosario-Raymundo MR, et al. Natural history of progression
De Marco L, et al. Human papillomavirus testing and liquid-based of HPV infection to cervical lesion or clearance: analysis of the
cytology in primary screening of women younger than 35 years: control arm of the large, randomised PATRICIA study. PLoS One.
results at recruitment for a randomised controlled trial. Lancet 2013;8(11):e79260.
Oncol. 2006;7(7):547–55. 174. Kotaniemi-Talonen L, Nieminen P, Anttila A, Hakama M. Routine
159. Kulasingam SL, Hughes JP, Kiviat NB, Mao C, Weiss NS, cervical screening with primary HPV testing and cytology triage
Kuypers JM, et al. Evaluation of human papillomavirus test- protocol in a randomised setting. Br J Cancer. 2005;93(8):862–7.
ing in primary screening for cervical abnormalities: compari- 175. Leinonen M, Nieminen P, Kotaniemi-Talonen L, Malila N,
son of sensitivity, specificity, and frequency of referral. JAMA. Tarkkanen J, Laurila P, et al. Age-specific evaluation of primary
2002;288(14):1749–57. human papillomavirus screening vs conventional cytology in a
160. Arbyn M, Ronco G, Anttila A, Meijer CJ, Poljak M, Ogilvie G, randomized setting. J Natl Cancer Inst. 2009;101(23):1612–23.
et al. Evidence regarding human papillomavirus testing in sec- 176. Anttila A, Kotaniemi-Talonen L, Leinonen M, Hakama M, Laurila
ondary prevention of cervical cancer. Vaccine. 2012;30(Suppl P, Tarkkanen J, et al. Rate of cervical cancer, severe intraepithe-
5):F88–99. lial neoplasia, and adenocarcinoma in situ in primary HPV DNA
161. Meijer CJ, Berkhof H, Heideman DA, Hesselink AT, Snijders screening with cytology triage: randomised study within organ-
PJ. Validation of high-risk HPV tests for primary cervical screen- ised screening programme. BMJ. 2010;340:c1804.
ing. J Clin Virol. 2009;46(Suppl 3):S1–4. 177. Froudarakis ME. Diagnostic work-up of pleural effusions.
162. Ronco G, Dillner J, Elfstrom KM, Tunesi S, Snijders PJ, Arbyn Respiration. 2008;75(1):4–13.
M, et al. Efficacy of HPV-based screening for prevention of inva- 178. Keettel WC, Elkins HG. Experience with radioactive colloidal
sive cervical cancer: follow-up of four European randomised con- gold in the treatment of ovarian carcinoma. Am J Obstet Gynecol.
trolled trials. Lancet. 2014;383(9916):524–32. 1956;71(3):553–68.
17 Gynecologic Cytology 629
179. Creasman WT, Rutledge F. The prognostic value of peritoneal neal mesothelioma in females from serous papillary carcinoma of
cytology in gynecologic malignant disease. Am J Obstet Gynecol. the ovary and peritoneum. Histopathology. 2002;40(3):237–44.
1971;110(6):773–81. 200. Ali SZ, Cibas ES. In: Rosenthal D, editor. Serous cavity fluids and
180. Day TG Jr, Smith JP. Diagnosis and staging of ovarian carcinoma. cerebrospinal fluid cytolopathology. New York: Springer; 2012.
Semin Oncol. 1975;2(3):217–22. 201. Zuna RE, Rao RN, Shidham VB. Diagnostic cytopathology of
181. Davidson W, Madan R, O’Neil M, Tawfik OW, Fan F. Utility of peritoneal washing, mesothelioma, metastatic carcinoma in effu-
peritoneal washing cytology in staging and prognosis of ovarian sions. In: Shidham VB, Atkinson BF, editors. Cytopathologic
and fallopian tube neoplasms: a 10-year retrospective analysis. diagnosis of serous fluids. Philadelphia: Saunders Elsevier; 2007.
Ann Diagn Pathol. 2016;22:54–7. p. 91–115.
182. Fadare O, Mariappan MR, Hileeto D, Wang S, McAlpine JN, 202. Tyagi R, Gupta N, Bhagat P, Gainder S, Rai B, Dhaliwal LK, et al.
Rimm DL. Upstaging based solely on positive peritoneal wash- Impact of SurePath((R)) liquid-based preparation in cytological
ing does not affect outcome in endometrial cancer. Mod Pathol. analysis of peritoneal washing in practice of gynecologic oncol-
2005;18(5):673–80. ogy. J Cytol. 2017;34(2):95–100.
183. Fadare O, Mariappan MR, Wang S, Hileeto D, McAlpine J, Rimm 203. Moriarty AT, Schwartz MR, Ducatman BS, Booth CN, Haja
DL. The histologic subtype of ovarian tumors affects the detection J, Chakraborty S, et al. A liquid concept--do classic prepara-
rate by pelvic washings. Cancer. 2004;102(3):150–6. tions of body cavity fluid perform differently than ThinPrep
184. Milosevic MF, Dembo AJ, Thomas GM. The clinical significance cases? Observations from the College of American Pathologists
of malignant peritoneal cytology in stage I endometrial carci- Interlaboratory Comparison Program in Nongynecologic
noma. Int J Gynecol Cancer. 1992;2(5):225–35. Cytology. Arch Pathol Lab Med. 2008;132(11):1716–8.
185. Tebeu PM, Popowski Y, Verkooijen HM, Bouchardy C, Ludicke 204. Selvaggi SM. Diagnostic pitfalls of peritoneal washing cytology
F, Usel M, et al. Positive peritoneal cytology in early-stage and the role of cell blocks in their diagnosis. Diagn Cytopathol.
endometrial cancer does not influence prognosis. Br J Cancer. 2003;28(6):335–41.
2004;91(4):720–4. 205. Balassanian R, Wool GD, Ono JC, Olejnik-Nave J, Mah MM,
186. American Joint Committee on Cancer. AJCC cancer staging man- Sweeney BJ, et al. A superior method for cell block prepara-
ual. 7th ed. Philadelphia, PA: Lippincott Raven Publishers; 2010. tion for fine-needle aspiration biopsies. Cancer Cytopathol.
187. Hanley KZ, Fadare O, Fisher KE, Atkins KA, Mosunjac 2016;124(7):508–18.
MB. Clinical significance of positive pelvic washings in uterine 206. Shidham VB. The panorama of different faces of mesothelial cells.
papillary serous carcinoma confined to an endometrial polyp. Int J In: Atkinson BF, Shidham VB, editors. Cytopathologic diagnosis
Gynecol Pathol. 2016;35(3):249–55. of serous fluids. 1st ed. Philadelphia: Saunders Elsevier; 2007.
188. Snyder MJ, Bentley R, Robboy SJ. Transtubal spread of serous p. 19–31.
adenocarcinoma of the endometrium: an underrecognized 207. Pisharodi LR, Bedrossian CW. Cytopathology of serous neoplasia
mechanism of metastasis. Int J Gynecol Pathol. 2006;25(2): of the ovary and the peritoneum: differential diagnosis from meso-
155–60. thelial proliferations. Diagn Cytopathol. 1996;15(4):292–5.
189. Stewart CJ, Doherty DA, Havlat M, Koay MH, Leung YC, Naran 208. Assaly M, Bongiovanni M, Kumar N, Egger JF, Pelte MF, Genevay
A, et al. Transtubal spread of endometrial carcinoma: correla- M, et al. Cytology of benign multicystic peritoneal mesothelioma
tion of intra-luminal tumour cells with tumour grade, perito- in peritoneal washings. Cytopathology. 2008;19(4):224–8.
neal fluid cytology, and extra-uterine metastasis. Pathology. 209. Malpica A, Sant’Ambrogio S, Deavers MT, Silva EG. Well-
2013;45(4):382–7. differentiated papillary mesothelioma of the female perito-
190. Zaino RJ. FIGO staging of endometrial adenocarcinoma: a critical neum: a clinicopathologic study of 26 cases. Am J Surg Pathol.
review and proposal. Int J Gynecol Pathol. 2009;28(1):1–9. 2012;36(1):117–27.
191. Delair D, Soslow RA, Gardner GJ, Barakat RR, Leitao MM Jr. 210. Haba T, Wakasa K, Sasaki M. Well-differentiated papillary
Tumoral displacement into fallopian tubes in patients undergoing mesothelioma in the pelvic cavity. A case report. Acta Cytol.
robotically assisted hysterectomy for newly diagnosed endome- 2003;47(1):88–92.
trial cancer. Int J Gynecol Pathol. 2013;32(2):188–92. 211. Zuna RE, Mitchell ML. Cytologic findings in peritoneal wash-
192. Hu S, Sun Y, Brown A. Incidence of positive peritoneal washings ings associated with benign gynecologic disease. Acta Cytol.
obtained during hysterectomy for endometrial adenocarcinoma: a 1988;32(2):139–47.
comparison of robotic-assisted hysterectomy versus total abdomi- 212. Pantanowitz L, Otis CN, Goulart RA. Cytologic findings
nal hysterectomy. Lab Investig. 2010;90(247A): 23–24. of psammocarcinoma in peritoneal washings. Acta Cytol.
193. Krizova A, Clarke BA, Bernardini MQ, James S, Kalloger SE, 2009;53(3):263–7.
Boerner SL, et al. Histologic artifacts in abdominal, vaginal, lapa- 213. Zuna RE, Mitchell ML, Mulick KA, Weijchert WM. Cytohistologic
roscopic, and robotic hysterectomy specimens: a blinded, retro- correlation of peritoneal washing cytology in gynecologic disease.
spective review. Am J Surg Pathol. 2011;35(1):115–26. Acta Cytol. 1989;33(3):327–36.
194. Shield P. Peritoneal washing cytology. Cytopathology. 214. Zuna RE, Behrens A. Peritoneal washing cytology in gynecologic
2004;15(3):131–41. cancers: long-term follow-up of 355 patients. J Natl Cancer Inst.
195. Rodriguez EF, Monaco SE, Khalbuss W, Austin RM, Pantanowitz 1996;88(14):980–7.
L. Abdominopelvic washings: a comprehensive review. 215. Tauchi PS, Caraway N, Truong LD, Kaplan AL, Ramzy I. Serous
Cytojournal. 2013;10:7. surface carcinoma of the peritoneum: useful role of cytology in dif-
196. Wojcik EM, Naylor B. “Collagen balls” in peritoneal washings. ferential diagnosis and follow-up. Acta Cytol. 1996;40(3):429–36.
Prevalence, morphology, origin and significance. Acta Cytol. 216. Hart WR. Diagnostic challenge of secondary (metastatic) ovar-
1992;36(4):466–70. ian tumors simulating primary endometrioid and mucinous neo-
197. RM D. The art and science of cytopathology. Exfoliative cytology. plasms. Pathol Int. 2005;55(5):231–43.
2nd ed. Chicago, IL: ASCP Press; 2012. p. 269–338. 217. Khunamornpong S, Settakorn J, Sukpan K, Suprasert P,
198. McGowan L. Peritoneal fluid washings. Acta Cytol. Siriaunkgul S. Mucinous tumor of low malignant potential (“bor-
1989;33(3):414–5. derline” or “atypical proliferative” tumor) of the ovary: a study
199. Attanoos RL, Webb R, Dojcinov SD, Gibbs AR. Value of meso- of 171 cases with the assessment of intraepithelial carcinoma and
thelial and epithelial antibodies in distinguishing diffuse perito- microinvasion. Int J Gynecol Pathol. 2011;30(3):218–30.
630 U. Krishnamurti et al.
218. Pisharodi LR, Bedrossian CW. Cytologic diagnosis of pseu- 236. Attanoos RL, Dojcinov SD, Webb R, Gibbs AR. Anti-mesothelial
domyxoma peritonei: common and uncommon causes. Diagn markers in sarcomatoid mesothelioma and other spindle cell neo-
Cytopathol. 1996;14(1):10–3. plasms. Histopathology. 2000;37(3):224–31.
219. Badyal RK, Khairwa A, Rajwanshi A, Nijhawan R, Radhika S, 237. Attanoos RL, Webb R, Dojcinov SD, Gibbs AR. Malignant epi-
Gupta N, et al. Significance of epithelial cell clusters in pseudo- thelioid mesothelioma: anti-mesothelial marker expression corre-
myxoma peritonei. Cytopathology. 2016;27(6):418–26. lates with histological pattern. Histopathology. 2001;39(6):584–8.
220. Ronnett BM, Kurman RJ, Zahn CM, Shmookler BM, Jablonski 238. Chu AY, Litzky LA, Pasha TL, Acs G, Zhang PJ. Utility of D2-40,
KA, Kass ME, et al. Pseudomyxoma peritonei in women: a clini- a novel mesothelial marker, in the diagnosis of malignant meso-
copathologic analysis of 30 cases with emphasis on site of origin, thelioma. Mod Pathol. 2005;18(1):105–10.
prognosis, and relationship to ovarian mucinous tumors of low 239. Comin CE, Saieva C, Messerini L. h-caldesmon, calretinin, estro-
malignant potential. Hum Pathol. 1995;26(5):509–24. gen receptor, and Ber-EP4: a useful combination of immuno-
221. Ronnett BM, Shmookler BM, Sugarbaker PH, Kurman histochemical markers for differentiating epithelioid peritoneal
RJ. Pseudomyxoma peritonei: new concepts in diagnosis, ori- mesothelioma from serous papillary carcinoma of the ovary. Am J
gin, nomenclature, and relationship to mucinous borderline (low Surg Pathol. 2007;31(8):1139–48.
malignant potential) tumors of the ovary. Anat Pathol. 1997;2: 240. Hecht JL, Lee BH, Pinkus JL, Pinkus GS. The value of Wilms
197–226. tumor susceptibility gene 1 in cytologic preparations as a marker
222. Gupta R, Mathur SR, Arora VK, Sharma SG. Cytologic features for malignant mesothelioma. Cancer. 2002;96(2):105–9.
of extragonadal germ cell tumors: a study of 88 cases with aspira- 241. Lal A, Bourtsos EP, Nayar R, DeFrias DV. Cytologic features of
tion cytology. Cancer. 2008;114(6):504–11. granulosa cell tumors in fluids and fine needle aspiration speci-
223. Park JY, Kim DY, Suh DS, Kim JH, Kim YM, Kim YT, et al. mens. Acta Cytol. 2004;48(3):315–20.
Outcomes of surgery alone and surveillance strategy in young 242. McCluggage WG. Immunohistochemistry in the distinction
women with stage I malignant ovarian germ cell tumors. Int J between primary and metastatic ovarian mucinous neoplasms. J
Gynecol Cancer. 2016;26(5):859–64. Clin Pathol. 2012;65(7):596–600.
224. Lurain JR, Rice BL, Rademaker AW, Poggensee LE, Schink JC, 243. Mhawech-Fauceglia P, Wang D, Menesses T, Chandavarkar U,
Miller DS. Prognostic factors associated with recurrence in clini- Ough F, Lin Y, et al. Pax-8 is a reliable marker in making the diag-
cal stage I adenocarcinoma of the endometrium. Obstet Gynecol. nosis in advanced stage epithelial ovarian carcinoma and primary
1991;78(1):63–9. peritoneal carcinoma for neoadjuvant chemotherapy on cell block
225. Zaino RJ, Kurman RJ, Diana KL, Morrow CP. Pathologic mod- and biopsy specimens. Histopathology. 2012;60(6):1019–20.
els to predict outcome for women with endometrial adenocarci- 244. Mittal K, Soslow R, McCluggage WG. Application of immuno-
noma: the importance of the distinction between surgical stage histochemistry to gynecologic pathology. Arch Pathol Lab Med.
and clinical stage – a gynecologic oncology group study. Cancer. 2008;132(3):402–23.
1996;77(6):1115–21. 245. Selvaggi SM, Guidos BJ. Immature teratoma of the ovary on fluid
226. Kanbour AI, Buchsbaum HJ, Hall A, Kanbour AI. Peritoneal cytology. Diagn Cytopathol. 2001;25(6):411–4.
cytology in malignant mixed mullerian tumors of the uterus. 246. Sheibani K, Shin SS, Kezirian J, Weiss LM. Ber-EP4 anti-
Gynecol Oncol. 1989;33(1):91–5. body as a discriminant in the differential diagnosis of malig-
227. Wheeler YY, Burroughs F, Li QK. Fine-needle aspiration of a nant mesothelioma versus adenocarcinoma. Am J Surg Pathol.
well-differentiated papillary mesothelioma in the inguinal her- 1991;15(8):779–84.
nia sac: a case report and review of literature. Diagn Cytopathol. 247. Vang R, Gown AM, Barry TS, Wheeler DT, Ronnett
2009;37(10):748–54. BM. Immunohistochemistry for estrogen and progesterone
228. Kalyani R, Das S. Adenomatatoid tumor: cytological diagnosis of receptors in the distinction of primary and metastatic muci-
two cases. J Cytol. 2009;26(1):30–2. nous tumors in the ovary: an analysis of 124 cases. Mod Pathol.
229. Makkar M, Dayal P, Gupta C, Mahajan N. Adenomatoid tumor of 2006;19(1):97–105.
testis: a rare cytological diagnosis. J Cytol. 2013;30(1):65–7. 248. Vang R, Gown AM, Barry TS, Wheeler DT, Yemelyanova A,
230. Husain AN, Colby TV, Ordonez NG, Allen TC, Attanoos RL, Seidman JD, et al. Cytokeratins 7 and 20 in primary and second-
Beasley MB, et al. Guidelines for pathologic diagnosis of malig- ary mucinous tumors of the ovary: analysis of coordinate immu-
nant mesothelioma: 2017 update of the consensus statement from nohistochemical expression profiles and staining distribution in
the international mesothelioma interest group. Arch Pathol Lab 179 cases. Am J Surg Pathol. 2006;30(9):1130–9.
Med. 2018;142(1):89–108. 249. Vang R, Gown AM, Wu LS, Barry TS, Wheeler DT, Yemelyanova
231. Patel NP, Taylor CA, Levine EA, Trupiano JK, Geisinger A, et al. Immunohistochemical expression of CDX2 in primary
KR. Cytomorphologic features of primary peritoneal mesothe- ovarian mucinous tumors and metastatic mucinous carcinomas
lioma in effusion, washing, and fine-needle aspiration biopsy involving the ovary: comparison with CK20 and correlation with
specimens: examination of 49 cases at one institution, including coordinate expression of CK7. Mod Pathol. 2006;19(11):1421–8.
post-intraperitoneal hyperthermic chemotherapy findings. Am J 250. Vrdoljak-Mozetic D, Stankovic T, Krasevic M, Versa-Ostojic D,
Clin Pathol. 2007;128(3):414–22. Stemberger-Papic S, Rupcic S. Intraoperative cytology of clear
232. Suster S, Moran CA. Applications and limitations of immuno- cell carcinoma of the ovary. Cytopathology. 2006;17(6):390–5.
histochemistry in the diagnosis of malignant mesothelioma. Adv 251. Xiang L, Zheng W, Kong B. Detection of PAX8 and p53 is ben-
Anat Pathol. 2006;13(6):316–29. eficial in recognizing metastatic carcinomas in pelvic washings,
233. Tiwana KK, Nibhoria S, Kaur M, Monga T, Gupta especially in cases with suspicious cytology. Gynecol Oncol.
R. Postchemotherapy histopathological evaluation of ovarian 2012;127(3):595–600.
carcinoma: a 40-case study. Chemother Res Pract. 2015;2015: 252. Dabbs DJ. Diagnostic immunohistochemistry theranostic and
197871. genomic applications. 3rd ed. Philadelphia: Saunders Elsevier;
234. McCluggage WG, Lyness RW, Atkinson RJ, Dobbs SP, Harley I, 2010.
McClelland HR, et al. Morphological effects of chemotherapy on 253. Chapel DB, Husain AN, Krausz T, McGregor SM. PAX8 expres-
ovarian carcinoma. J Clin Pathol. 2002;55(1):27–31. sion in a subset of malignant peritoneal mesotheliomas and
235. Wang Y, Wang Y, Zheng W. Cytologic changes of ovarian epithe- benign mesothelium has diagnostic implications in the differen-
lial cancer induced by neoadjuvant chemotherapy. Int J Clin Exp tial diagnosis of ovarian serous carcinoma. Am J Surg Pathol.
Pathol. 2013;6(10):2121–8. 2017;41(12):1675–82.
Index
© Science Press & Springer Nature Singapore Pte Ltd. 2019 631
W. Zheng et al. (eds.), Gynecologic and Obstetric Pathology, Volume 2, https://doi.org/10.1007/978-981-13-3019-3
632 Index
Brenner tumors, 132, 133, 203–227, 298, 299, 343, 344, 561 Colorectal cancers, 324, 329
Broad ligament, 21, 54, 79, 80, 226, 310, 367–384, 386–390, Colposcopy, 70, 592, 595, 606, 607
393–400, 413, 443 Columnar epithelial cells, 226
Compaction/morula, 124
Complete hydatidiform mole (CHM), 433, 435, 437, 439–441, 443,
C 444, 446, 482, 483
Calcification, 17, 60, 71, 86, 87, 109, 152, 158, 175, 215, 216, 219, Complex papillary, 165
249, 253, 266–268, 277, 278, 280, 282, 286, 287, Confidence interval, 553
293, 310, 312, 313, 373, 398, 489, 511, 512, 534, Confined placental mosaicism, 460
550, 569, 584 Confluent growth, 390
Calcifying fibrous pseudotumor, 372 Congenital adrenal hyperplasia, 95
Call-Exner body, 344 Corpus albicans cysts, 104
Calretinin, 7, 14, 41, 62, 79, 87, 98, 103, 281, 289, 293, 298, 312, 316, Corpus luteum cysts, 108, 109, 414
344, 368, 394, 623 Corticosteroids, 274
Cambium layer, 41, 42 Cotyledonoid dissecting leiomyoma, 19, 21, 550
Candida albicans, 490 Crystals of Reinke, 292, 293, 295, 296, 299, 300, 302
Carcinoids with mixed patterns, 246 Cystadenofibroma, 103, 142–144, 205, 614
Carcinoid tumor, 244, 310, 339, 341–344 Cystadenoma, 67, 103, 125, 126, 141, 142, 144, 145, 204–208, 214,
Carcinoma in situ (CIS), 606, 607 614, 622
Carcinomatosis, 167, 369, 608 Cysteine, 303
Carcinosarcoma, 12, 13, 25, 31, 33, 34, 41, 43, 58–61, 123, 187–189, Cystic corpora lutea, 92, 108
237, 279, 318, 546, 548, 618 Cystic granulosa cell tumor, 89
Cell adhesion molecule (CAM), 430 Cystic nephroma, 317
Cell block, 611, 613, 615, 617, 621, 622 Cytokeratin (CK), 14, 16, 41, 79, 81, 99, 115, 211, 216, 221, 223, 262,
Cell-renal cell carcinoma, 358 279, 291, 297, 298, 351, 368, 377, 380, 395, 398, 445, 446,
Cellular fibroma, 275, 277–279, 303 448–452
Cellular leiomyoma, 19, 29 Cytomegalovirus (CMV), 102, 430, 431, 491, 500, 501, 533, 575
Cervical adenocarcinoma, 344 Cytospin, 611, 612
Cervical cancer, 344, 347, 571, 572, 578, 591, 602–606 Cytotrophoblasts (CT), 253, 262, 427–429, 431, 435, 436, 441,
Cervical intraepithelial neoplasia (CIN), 606 443–445, 460, 518
Cervix, 14, 34, 41, 61, 67, 122, 213, 214, 312, 325, 329, 333, 335,
373, 384, 407, 410, 416, 420, 422, 432, 439, 527, 543, 544,
546, 572, 573, 594, 599, 600 D
Chondrosarcoma, 60, 61, 240 Decidua, 429, 431, 434, 437, 451, 461, 475–477, 513, 514, 526, 527,
Chorangiocarcinoma, 535 534, 568
Chorangioma, 469–471, 484, 535, 536 Decidual changes, 419, 447, 568, 569
Chorangiosis, 484, 520, 523 Decidual vasculopathy, 513–515, 517
Choriocarcinoma, 88, 253, 261, 262, 352, 430, 432, 433, 436, 439, De-differentiated carcinoma, 10
441, 443–448, 450, 535 Delayed villous maturation, 478, 520, 522, 523
Chorion, 428, 429, 433, 460, 466, 471, 475–477, 486, 489, 491, 496, Depth of invasion, 406, 527, 544, 546, 548
497, 505, 506, 525, 529 Dermoid cysts, 234, 248, 308, 369, 617
Chorionic gonadotropin, 293 Desmin, 3, 5–7, 10, 13, 14, 16, 20, 21, 24–28, 34, 36–40, 42, 80, 115,
Chorionic intermediate trophoblasts, 428, 452 187, 279, 281, 284, 287, 289, 292, 293, 351, 356, 378, 380,
Chromogranin A, 216, 222, 293, 338, 343 449
Chromosome abnormalities, 431 Desmoplastic non-invasive implants, 149–151, 389, 391
Chronic deciduitis, 489 Desmoplastic small round-cell tumor (DSRCT), 225, 356, 378, 379
Chronic intervillositis, 501–502 Diandric diploidy, 434
Chronic salpingitis, 64, 69, 71, 432 Diandric triploidy, 440, 443, 484
Chronic villitis, 479, 491, 502, 503, 511 DICER1 syndrome, 317
Ciliated metaplasia (tubal metaplasia), 419 Diffuse decidual leukocytoclastic necrosis, 466
Circumvallate placentas, 466 Diffuse leiomyomatosis, 21
Cistern, 434, 436, 440, 485, 522 Diffuse malignant mesothelioma (DMM), 371, 375, 377–379
Classic vulvar intraepithelial neoplasia, 543 Diffusion-weighted imaging, 134
Clear cell adenocarcinoma (CCA), 398 Dissecting gonadoblastoma, 266
Clear cell carcinoma, 23, 58, 61, 129, 131, 135, 146, 161, 191, 195, Dissecting leiomyoma, 21
253, 258, 259, 295, 355, 358, 420, 545, 560, 617, 618 Disseminated peritoneal leiomyomatosis (DPL), 22, 384, 395–396
Clear cell cystadenoma, 189 Distal villous fetal thrombotic vasculopathy, 508
Clear cell renal cell carcinoma, 358 Distal villous hypoplasia (DVH), 486, 514, 517, 520, 521
Clear cell tumors, 164 Distal villous immaturity, 520, 523
Clitoromegaly, 313 Divergent differentiation, 5, 266
Clue cells, 586 Driver gene, 131, 409
Coagulative necrosis, 17, 18, 249 Driving factor, 95
Coagulopathy, 537 Dualistic model of ovarian carcinogenesis, 121–123, 126
Coelomic epithelium, 407 Dysfunctional uterine bleeding, 104
Collagen balls, 609 Dysgerminoma, 221, 231, 249, 251, 253, 254, 258, 260–262, 264,
Colloid carcinoma, 331 266–268, 311, 360, 563, 564, 618
Colonize, 70, 354 Dyskeratosis, 581
Index 633
Dysmenorrhea, 104, 406 Enlarged villi, 434, 439, 440, 482, 532
Dyspareunia, 405 Eosinophilic clear cell carcinoma, 35
Dysplasia, 542, 543 Eosinophilic metaplasia, 64
Eosinophilic secretion, 226, 244, 342
Ependymoma, 247, 400
E Epidermoid cyst, 562
Early pregnancy, 84, 87, 427–434, 436–441, 443–453, 534 Epithelial membrane antigen (EMA), 9, 10, 13, 14, 24, 81, 115, 211,
E-cadherin, 122 216, 217, 221, 222, 225, 251, 258, 262, 275, 279, 280, 284,
Eclampsia, 486, 514, 523 287, 289, 292–294, 297, 298, 303, 304, 306, 310, 312, 313,
Ectopic 316, 317, 319, 356, 368, 371, 376, 378, 398, 450,
decidua, 69, 86, 89–90, 414, 417, 418 452, 622
decidual reaction, 377 Epithelial-mesenchymal transition (EMT), 122, 385
pregnancy, 53, 61, 65–69, 90, 99, 107, 373, 429, 432, 433, 444, Epithelial metaplasia, 32
548, 553, 555 Epithelial non-invasive implants, 149, 389, 391, 566
Edema of umbilical cord, 506, 508 Epithelioid angiomatosis, 27
Embryo, 67–69, 86, 90, 91, 427, 431, 432, 459, 460, 534 Epithelioid leiomyoma, 19, 20
Epithelial membrane antigen (EMA), 9, 10, 13, 14, 24, 81, 192, 197, Epithelioid leiomyosarcoma, 26, 27, 448–450, 551, 552
211, 216, 218, 225, 251, 253, 260, 275, 284, 287, 292, 293, Epithelioid trophoblastic tumor (ETT), 27, 433, 445, 446, 449, 450,
310, 313, 317, 350, 356, 376, 378, 398, 452, 622 452, 453
Embryonal carcinoma, 236, 253, 258–262, 264, 318, 618 Epoophoron, 81
Embryonal rhabdomyosarcoma (ERMS), 41, 42 Estradiol, 80, 92, 104, 410
Endocervical adenocarcinoma, 324, 344, 345, 544, 584 Estrogen, 5, 10, 19, 31, 59, 79, 84, 86, 97, 130, 148, 157, 159, 174,
Endocervical adenocarcinoma, usual type, 214, 346, 350 195, 214, 225, 274, 282, 286, 298, 300, 310, 319, 324, 407,
Endocervical curettage, 352 409, 410, 421, 581
Endocervicosis, 122, 369, 394, 395 Estrogen receptor (ER), 59, 156, 159, 160, 196, 214, 284, 333, 338,
Endodermal sinus, 192, 196, 253, 255, 564 340, 376
Endometrial adenocarcinoma, 282, 347, 373, 543, 545, 546 Estrogen receptor alpha (ERα), 131
Endometrial carcinoma, 34, 62, 72, 129, 131, 182, 183, 300, 347–350, European Medicines Agency, 135
541, 544–546, 608, 609 Ewing's sarcoma, 225, 226, 247, 248, 379
Endometrial clear cell carcinoma, 617 Exaggerated placental site (EPS), 430, 433, 434, 446, 448, 451, 452
Endometrial curettage, 13, 526 Exogenous hormones, 94, 395
Endometrial endometrioid carcinoma, 182 Expansile growth, 207, 208
Endometrial glands in background, 544 Extramammary Paget disease (EMPD), 542
Endometrial hyperplasia, 183, 218, 275, 282 Extravillous trophoblast, 432–436, 446, 482, 483, 504, 527
Endometrial intraepithelial carcinoma (EIC), 384
Endometrial serous carcinoma (ESC), 59, 60, 62, 348, 546, 548
Endometrial stromal nodule (ESN), 1–8, 16, 552 F
Endometrial stromal sarcoma (ESS), 2, 3, 7, 8, 12, 13, 22, 25, 40, 43, Fasciitis, 38
187, 222, 351, 352, 422, 552, 554 Female adnexal tumor of probable Wolffian origin, 224, 226, 275, 310,
Endometrial stromal tumor(s), 2–3 398–400
with sex cord-like elements, 6, 7, 14, 281 Fetal growth restriction, 465, 484, 511, 512, 518, 521
with smooth muscle differentiation, 5–7, 25 Fetal inflammatory response, 469, 473, 489, 490, 496–498, 506
Endometrioid adenocarcinoma, 34, 131, 157, 161, 166, 304, 335, 338, Fetal thrombotic vasculopathy, 487, 508
344, 349, 350, 354–356, 398, 544, 548, 560 Fetiform teratoma, 234, 238–239
Endometrioid adenofibroma, 175 Fetus papyraceous, 467, 468, 476
Endometrioid carcinoma, 131, 183, 185 Fibroblast, 17, 123–125, 253, 273, 277, 280, 282, 283, 372, 380, 460,
with ciliated cell differentiation, 131 464, 475
with corded and hyalinized component, 317 Fibrosis, 24, 54, 65, 66, 72, 87, 94, 95, 112, 278, 286, 310, 336, 369,
with mucinous differentiation, 166 370, 376, 378, 384, 390, 410, 412, 413, 416, 417, 419, 431,
with sertoli-like or sex cord-like elements, 310 432, 434, 479, 491, 554, 621
with squamous differentiation, 131, 179, 347, 369 Fibroma, 111, 112, 114, 128, 215, 274, 275, 277–279, 281, 284, 289,
Endometrioid cystadenoma, 174–175 300, 301, 303, 327, 339, 351, 352, 360
Endometrioid intraepithelial neoplasia (EIN), 34, 544 Fibromatosis, 38, 110–113, 279
Endometrioid stromal sarcoma, 187, 279 Fibrosarcoma, 187, 211, 240, 274, 275, 279–282, 286
Endometrioid tumor, 59, 174–176, 182, 183, 185, 333, 335 Fibrothecoma, 111, 274, 277, 279, 280, 282, 284, 285, 307,
Endometrioma, 104, 173, 174, 177, 411, 414, 421, 556 319, 560
Endometriosis, 4, 70, 96, 103, 115, 129, 130, 133, 183, 185, 203, 259, Follicular cervicitis, 575, 580, 581, 584, 600
325, 387, 397, 405, 406, 408, 413, 415, 416, 419, 422, 559, Follicular phase, 85, 104
612, 619 Foreign body, 66, 116, 208, 347, 369, 373, 569
Endometriosis-associated tumors, 420, 422 Food and Drug Administration (FDA), 573, 602, 603, 606
Endometriotic cyst, 104, 105, 109, 115, 419, 559 Fumarate hydratase, 29, 30
Endometritis, 580
Endomitosis, 124
Endoreplication, 124, 125 G
Endosalpingiosis, 58, 64, 70, 96, 102–104, 142, 151–153, 167, 369, Gartner’s ducts, 226
384, 387–389, 395, 396, 408, 414, 417, 568–570, 610, Gastric cancer, 324, 329, 566
612–614, 617, 622 Gastric type, adenocarcinoma, 346
634 Index
Gastrointestinal stromal tumor (GIST), 279, 280, 380, 415 Human androgen receptor gene assay, 132
Germ cell tumor, 196, 197 Human chorionic gonadotrophin, 84, 87–89, 105, 428, 433, 548
Germline mutation, 29, 55, 57, 72, 129, 131, 135, 141, 221, 303, 317, Human immunodeficiency virus, 65
352, 354 Human papillomavirus, 214, 346, 347, 350, 543, 572, 574, 582, 603, 607
Gestational diabetes mellitus, 494, 520, 523 Human placental lactogen (hPL), 428
Gestational hypertension, 464, 484 Hyaline degeneration (hyalinization), 250
Gestational trophoblastic diseases (GTD), 429, 432, 433 Hydatid cyst, 58, 70
Ghost villi, 479, 517 Hydatidiform mole, 231, 429–433, 436–439, 441, 443, 445
Gland-poor endometriosis, 43 Hydatids of Morgagni, 70
Glandular cell abnormality, 577 Hydropic abortion, 431, 439, 441
Glandular epithelium, 240, 331, 350, 388, 395, 567 Hydropic degeneration, 18, 432, 482
Glassy cell carcinoma, 344 Hydropic leiomyoma, 18, 19, 28, 550
Gliomatosis peritonei, 234, 235, 238, 384, 385 Hydrops fetalis, 532
Goblet cell carcinoids, 244–246, 291, 335, 337, 338 Hypercalcemic type, small-cell carcinoma, 220, 221, 226
Gonadal ridge, 274 Hyperchromatic crowded groups, 580, 592, 595
Gonadoblastoma, 250, 253, 254, 264–268 Hyperkeratosis, 582
Gonadoblastoma with malignant germ cell tumor, 250 Hyperplasia with atypia, 483
Gonadotropin-releasing hormone (GnRH), 19, 22, 396 Hyperreactio luteinalis, 88–89, 95, 434
Gonorrhea, 68, 99, 573 Hyperthecosis, 84, 95–97, 279, 286, 293, 300
Gorlin’s syndrome, 274, 277, 279 Hypertrophy of myometrium, 43
Granular cell tumor, 543
Granulomatous peritonitis, 369, 568
Granulosa cell, 6, 14, 16, 82, 85, 86, 90, 92, 104–106, 274–277, 282, I
285, 299–308, 310, 312 Immature intermediate villi, 460, 462, 518, 520
Granulosa cell tumors (GCT), 14, 85, 88, 105, 221, 243, 258, 277, Immature teratoma, 114, 234–238, 262, 379, 384, 563, 618
279, 282, 284, 286, 299, 303, 306–309, 317, 319, 344, 351, Immune hydrops fetalis, 533
356, 561, 562, 617, 618 Immunne checkpoint, 135
Growing teratoma syndrome, 238 Immunocytochemistry (immunohistochemistry), 3, 10–12, 14, 16, 19,
Growth factors, 86 21, 25, 26, 30, 32–37, 39, 40, 43, 55, 62, 79, 80, 87, 103,
Gynandroblastoma, 275, 303, 317 133, 195, 213, 214, 220–222, 250, 257, 260, 277, 299, 430,
Gynecologic Cytology, 571–623 434, 484
Implantation, 68, 86, 90, 91, 410, 427, 431, 432, 521, 549
Implantation of fertilized egg, 68, 459
H Implantation site, 68, 91, 239, 428, 432, 433, 436, 446, 451, 452, 482,
Hair-an syndrome, 233, 275 548, 549, 554
Hamartoma, 379, 380 Implantation site intermediate trophoblast, 432
Haploid, 435, 437 Implants, 128, 149–151, 153, 167, 168, 234, 235, 240, 339, 389, 390,
Heave menstrual bleeding (menorrhagia), 38, 300 392, 406, 526, 544, 564, 566, 567
Hemangioma, 61, 224, 465 Imprinted genes, 438
Hemangiosarcoma (angiosarcoma), 27, 43, 240 Inclusion cysts, 57, 64, 102, 103, 122, 142, 144, 619
Hemangiopericytoma, 42, 286, 287, 289 Incomplete abortion, 433, 440
Hepatoid carcinoma, 223 Infantile uterus, 231, 232
Hereditary leiomyomatosis and renal cell carcinoma syndrome, 29 Infarcted appendix epiploica, 373
Herpes, 491, 500, 586, 587 Infarct type necrosis, 447, 550
Herpes simplex virus (HSV), 491, 500 Infiltrative pattern of invasion, 212
Heterogeneous tumors, 12, 126, 173, 243 Inflammatory myofibroblastic tumor, 1, 21, 26, 28, 38–40, 372, 379,
High grade adenocarcinoma of the appendix, 600 380
High-grade endometrioid stromal sarcoma, 59 Inhibin, 7, 14, 16, 80, 85, 87, 92, 98, 225, 307, 398, 399, 429, 445,
High-grade neuroendocrine carcinoma, 341, 344 448, 449, 451, 452, 622
High-grade peritoneal serous carcinoma, 58, 127 Initial endometriosis, 130, 408
High-grade serous carcinoma (HGSC), 54–60, 62, 65, 123, 125–130, Insular carcinoid, 244, 303, 344
133–136, 141, 154–161, 303, 305, 324, 354, 380, 390, 554, Insulin-like growth factor II mRNA-binding protein 3 (IMP-3),
555, 564 24, 26
High-grade squamous intraepithelial lesion (HSIL), 450, 575, 577, Intercellular bridges, 368
588, 591–593, 595, 596 Intermediate trophoblast, 69, 260, 428, 429, 436, 444–446, 449, 451, 453
High-risk HPV, 588 Intermediate villi, 518, 522
Hilar cell nodular hyperplasia, 94 Intervillous spaces, 427, 428, 459, 469, 479, 481, 489, 490, 499, 500,
Hilar cell tumor/hilus cell tumor, 99, 292 502, 504, 516
Hilus cell, 71, 80, 97–99, 144 Intestinal hamartoma, 317
Hilus cell hyperplasia, 71, 97–99 Intestinal type mucinous carcinoma, 132, 559, 617
Hobnail cell, 166, 177, 197, 350, 545 Intra-abdominal desmoplastic small round cell tumor, 378–379
Homeobox, 122 Intraepithelial carcinoma, 132, 148, 165, 208, 209, 352, 559
Homologous adenosarcoma, 31 Intraepithelial neoplasia, 542, 572
Homosexual precocity, 263 Intrahepatic cholestasis of pregnancy (ICP), 536
Homunculus, 234, 238 Intranodal decidua, 396
Hormone replacement therapy, 92 Intraplacental choriocarcinoma, 444, 445
Index 635
Intrauterine device, 100, 101, 555, 580 Lymphoma, 17, 43, 62, 221, 359, 360, 410, 564, 600
Intravenous leiomyomatosis (IVL), 7, 16, 21–22 Lymphovascular invasion, 4, 10, 36, 37, 241, 307, 328, 339, 347, 355,
Invasive implants, 147, 149–151, 153, 389, 390, 392, 614 359, 552
Invasive mole, 433, 442, 443 Lymphovascular space invasion, 305, 334, 339, 343, 347, 390
Lynch syndrome, 181, 193, 195
Lynch syndrome related endometrial cancer, 185
J
Juvenile granulosa cell tumor, 307, 379, 561
M
Macrofollicular, 300, 301, 307, 561
K Maffucci’s syndrome, 274, 281
Keratinizing squamous cell carcinoma, 449, 594, 597 Malignant Brenner tumor, 123, 131–133, 215, 218, 219, 358, 359, 560
Koilocytes, 543, 592 Malignant melanoma, 239, 249, 304, 357, 369, 599
Krukenberg tumors, 246, 291, 317, 325, 327–330, 336, 337, 339, Malignant mesothelioma (MM), 162, 372, 374–378, 394, 612, 619,
340, 355 620
Malignant mixed mesodermal tumors, 600
Malignant mixed Müllerian tumor (MMMT), 59, 123, 133–134, 237
L Massive ovarian edema, 110, 111, 279, 286
Large cell neuroendocrine carcinoma, 222 Massive perivillous fibrin deposition, 504
Large solitary luteinized follicle cyst, 89, 105 Maternal inflammatory response, 489, 496, 497
Large vessel fetal thrombotic vasculopathy, 487 Mature cystic teratomas, 132, 133, 210, 232–234, 239, 381, 569
Leiomyoma, 7 Mature intermediate villi, 460, 462
with bizarre nuclei, 19, 25, 29, 30, 279 Mature solid teratomas, 234
Leiomyomatosis, 7, 21, 384, 395–397, 419, 569 Mature teratomas, 232
Leiomyosarcoma (LMS), 7, 10, 12, 13, 16, 18, 21–29, 37, 40, 61, 281, Mayer-Rokitansky-Kuster-Syndrome, 407
350, 351, 380, 396, 551–553 Medulloepithelioma, 247, 317
Leiomyosarcoma with osteoclast-like giant cells, 5 Meigs’ syndrome, 274, 279
Leukemia, 43, 359, 360, 536 Melanocytic lesions, 542
Leydig cell, 14, 71, 80, 97, 266, 273, 274, 297, 315, 562 Melanocytic tumor, 249
Leydig cell proliferations, 71 Melanoma cell adhesion molecule, 429
Leydig cell tumor, 274, 275, 277, 292, 293, 295, 296, 299, 303, 310, Melanosis, 54, 369
314–318 Menarche, 135, 154, 254, 274, 405–407, 421
Lichen sclerosus (sclerosis), 543 Menopause, 41, 57, 80, 84, 85, 96, 154, 274, 406, 421
Lichen simplex chronicus, 543 Menstrual cycle, 19, 54, 104, 108, 550, 578
Lineage infidelity, 122, 131 Menstruation, 407, 421
Lipid-rich Sertoli cell tumor, 310 Merkel cell carcinoma, 361
Lipofuscin, 54, 71, 98, 506 Mesenchymal neoplasms, 551
Lipoleiomyoma, 19, 21, 551 Mesenchyme, 122, 234, 394, 427
Lipoma, 21 Mesonephric ducts, 226, 389, 397, 398
Liposarcoma, 21, 61, 380 Mesonephric remnants, 226, 389
Liquid-based cytology, 573, 574 Mesothelial hyperplasia, 64, 113, 370–372, 376, 419, 568, 614
Listeria, 431, 490, 500, 502 Mesothelioma, 41, 62, 90, 113, 157, 162, 168, 223–225, 279, 324,
Lobular endocervical glandular hyperplasia, 62 372, 376, 379, 395, 567, 568, 621, 622
Lobulated ovary, 292 Mesovarium, 98
Loop electrosurgical excision procedure of cervix (LEEP), 543 Metaplastic papillary tumor, 58, 64
Loss of heterozygosity (LOH), 25, 55, 133, 153, 181, 190, 210, 221, Metastasizing leiomyoma, 22, 396
279 Metastatic carcinoma, 62, 88, 152, 213, 613
Lower uterine segment (LUS), 31, 34, 344, 450, 515, 526–528, 544, Metastatic renal cell carcinoma, 295
545, 580 Microcystic, elongated, and fragmented (MELF), 548
Low-grade endometrial stromal sarcoma (LGESS), 1, 2, 20, 279, 281, Microcystic stromal tumor (MST), 275, 283, 289, 296–298
350, 352 Microfollicular, 90, 300, 301, 307, 561
Low-grade endometrial stromal sarcoma with limited infiltration, 3 Microglandular hyperplasia, 595
Low-grade mucinous neoplasm of the appendix, 381 Microinvasion, 132, 148, 149, 152, 165, 190, 208, 209
Low-grade peritoneal serous carcinoma (LGSC), 390 Microinvasive carcinoma, 209
Low-grade serous carcinoma, 58, 59, 123, 126, 127, 130, 133, 141, Microphthalmia-associated transcription factor (MiTF), 36
157–164, 389, 422, 557 MicroRNA/miRNA, 407, 409, 605
Low grade squamous intraepithelial lesion (LSIL), 592, 602, 605 Mitogen-activated protein kinase (MAPK), 153, 160, 409
Low-risk HPV, 602, 603 Mitotic activity, 8, 15, 19, 21, 28, 29, 31, 33, 40, 56, 59, 64, 65, 87, 90,
Luminal epithelium (LE), 428 111, 156, 158, 160, 162, 209, 221, 222, 224, 225, 227, 236,
Luteal phase, 17 330, 331, 333, 351, 376, 550, 551, 553, 568, 590, 600, 619
Luteinization, 105, 232, 301 Mitotically active cellular fibroma, 278
Luteinized follicular cyst, 89, 105, 106, 109 Mitotically active leiomyoma, 19
Luteinized thecoma, 87, 275, 279, 282–286, 293 Mitotic index, 24, 25, 29, 197, 283, 293, 295, 307, 557
Luteinized thecoma associated with sclerosing peritonitis, 282–287 Mixed epithelial-mesenchymal tumors, 58
Luteinizing hormone (LH), 84 Mixed germ cell-sex cord-stromal tumors, 264
Luteoma of pregnancy, 87, 89, 308 Mixed germ cell tumors, 249, 264, 265, 268, 317
636 Index
Monodermal teratomas, 298 Nuclear grooves, 15, 90, 241, 299, 301–304, 307, 356, 562, 617
Morula, 427, 459 Nuclear molding, 599
Morular metaplasia, 34, 547
Mucinous adenocarcinoma, 316, 339, 382, 383, 395
Mucinous adenofibromas, 205 O
Mucinous borderline tumor, 148, 164, 207, 209–212, 338, 340, 345, Omental mesenteric myxoid hamartoma, 380
347, 381, 559, 560 Opportunistic salpingectomy, 135
with intraepithelial carcinoma, 209 Ovarian atrophy, 96
with microinvasion, 209, 211 Ovarian cortex, 80–82, 90, 92, 97, 102, 103, 126, 128, 273, 283, 313,
Mucinous carcinoid, 244, 292, 329 324, 398, 408
Mucinous carcinoma, 63 Ovarian cortical fibromatosis, 112–113
gastric type, 345 Ovarian epithelial carcinomas, 129–134, 220, 356
signet-ring cell type, 225 Ovarian epithelial inclusion, 122, 346, 408, 417
Mucinous cell carcinoid, 245, 291 Ovarian hilus, 71, 80, 296
Mucinous cyst, 561 Ovarian inclusion cysts, 142
Mucinous cystadenocarcinoma, 338 Ovarian infection, 99, 102
Mucinous cystadenoma, 103, 132, 143, 204–206, 338–340, 383, 395, Ovarian pregnancy, 91, 239, 262
559, 560 Ovarian remnant syndrome, 114–115
Mucinous metaplasia, 58, 62 Ovarian stromal hyperplasia, 286
Mucinous tumor, 132, 206, 208, 310, 334, 343, 382, 383 Ovarian stromal hyperthecosis, 286, 293
Müllerian adenosarcoma, 31, 33, 34 Ovarian surface epithelium, 57, 79, 80, 102, 113, 121, 141, 142, 352,
Müllerian duct, 397 353
Multispectral fluorescence imaging, 134 Ovarian torsion and infarction, 111, 112
Mural nodules, 63, 204, 210, 211 Ovulatory cycle, 82, 83
Mutation, 61, 126, 128, 190, 195, 196
Myofibroblastoma, 380
Myometrial invasion, 1, 9–11, 27, 34, 345, 447, 449, 544, 546, 548, 552 P
Myxoid Paget’s disease, 542
degeneration, 17, 431 Pap smears, 167, 352, 573–575, 580–584, 588, 599, 600, 606, 607
hamartoma, 379 Papillary cystadenoma (with von hipple-lindau disease), 398
leiomyoma, 19–21, 28, 40 Papillary proliferation of the endometrium, 143, 207, 392, 612
leiomyosarcoma, 12, 18, 21, 27–28, 40, 551 Papillary tubal hyperplasia, 58, 64
material, 20, 255 Papilloma, 58, 64, 143, 418
Parabasal cells, 578, 590
Parakeratosis, 543, 588, 590
N Parasitic leiomyoma, 384
Necrotic pseudoxanthomatous nodules, 369 Parietal, 256, 258, 259, 367, 375, 608
Necrotizing chorioamnionitis, 477, 496, 497 Partial hydatidiform mole (PHM), 433, 439–441, 443, 444, 482, 483
Necrotizing funisitis, 490, 491, 496, 498 Pelvic inflammatory disease (PID), 64, 66–68, 83, 91, 99, 100, 114,
Negative for intraepithelial lesion, 574, 577, 586, 590 406, 555
Neisseria gonorrhoeae, 65, 555 Pelvic washings, 153, 607–609
Neuroblastoma, 222, 226, 359, 379, 536 Perimenopause, 97, 98
Neuroblastoma, 222, 226, 247, 349, 361, 379, 536 Peripheral neuroectodermal tumor, 222
Neuroectodermal-type tumors, 247 Peritoneal fibrosis, 286, 369
Neuroendocrine carcinoma, 222 Peritoneal inclusion cysts, 224, 371, 372, 375, 419, 568, 569, 612
non-small-cell type, 222 Peritoneal serous borderline tumors, 167–168, 388–390
Neuroendocrine differentiation, 335, 337 Perivascular epithelioid cell tumor (PEComa), 1, 35–38
Neuroendocrine neoplasms, 243, 342 Peutz-Jeghers syndrome (PJS), 62, 274, 310, 312, 334
Neurofibromatosis type 1 (NF1), 129 Phoshatase and tensin homolog (PTEN), 166, 409
Neuron-specific enolase (NSE), 225, 293 Picket fence, 578
Nevoid basal cell carcinoma syndrome, 281 Pituitary adenoma, 249
Nodular goiter, 317 accreta, 462, 463, 494, 526–528
Non-gestational choriocarcinoma, 261, 262, 445, 446, 556 increta, 527
Non granulomatous histiocytic lesions, 369 membranacea, 467
Non-hilar cell, 296 percreta, 526, 527
Non immune hydrops fetalis (NIHF), 533 Placental alkaline phosphatase, 197
Non-invasive implants, 389 Placental infections, 499–503
Non-invasive low-grade serous carcinoma, 127 Placental inflammation, 495, 496, 498–501
Non-keratinizing carcinoma, 448, 594 Placental mesenchymal dysplasia (PMD), 439, 441, 443, 484, 485,
Non-keratinizing squamous cell carcinoma, 448 522, 523
Non-luteinized follicle cyst, 104, 105 Placental site nodules and plaques, 430, 433
Non-small cell neuroendocrine carcinoma (NSCNEC), 222 Placental site trophoblastic tumors, 69, 350, 430, 433, 445, 446,
Nuclear atypia, 21, 24–27, 29, 37, 42, 64, 156, 158, 175, 189, 190, 448–453
194, 207, 215, 236, 245, 256, 278–282, 286–288, 291, 293, Placentomegaly, 484
295, 307, 310, 312, 316, 317, 319, 336, 350, 371, 377, 390, Pleuropulmonary blastoma, 317
419, 436, 446, 450, 548, 551, 553, 558, 566, 568 P53 nonsense mutation(null-pattern), 156
Index 637
Polycystic ovarian syndrome (PCOS), 94–95, 110 Secretory cell outgrowth, 55, 127, 128
Polyembryoma, 258, 261–263 Sectioning and Extensively Examining the Fimbria (SEE-FIM), 53,
Polypoid adenomyoma, 418 72, 134
Polypoid endometriosis, 70, 415, 418 Selective estrogen receptor modulators, 19
Polyvesicular, 192, 196, 256, 564 Seminoma, 249, 250
Postmenopausal bleeding, 3, 12, 14, 23, 35, 300, 452 Sentinel lymph node, 570
Precancer, 607 Seromucinous, 422
Pre-eclampsia, 88, 440, 464, 473, 485, 514, 523 Seromucinous adenofibromas, 164
Premature ovarian failure (POF), 91–92, 94 Seromucinous borderline tumors, 164–167, 214
Preovulatory follicle, 82, 83, 104 Seromucinous carcinoma, 123, 129–131, 133, 135, 157, 164, 166
Preterm birth, 526, 536 Seromucinous cystadenoma, 164
Primary carcinoid tumor, 342 Seromucinous tumors, 131, 164, 203, 205, 207, 333, 422
Primary carcinoma, 329, 354, 554 Serous, 58
Primary follicles, 81, 82, 92, 93, 104 Serous adenocarcinoma, 376, 378, 387, 388, 393, 394, 400
Primary tumor, 34, 43, 132, 238, 246, 262, 323–325, 327, 330, 335, Serous adenofibroma, 58, 61, 143
338, 340, 342–344, 347, 349–354, 357, 379, 385, 399, 535, Serous borderline tumor (SBT), 58, 72, 126, 141–142, 144–153, 159,
600, 617, 621 163, 164, 167, 209, 370, 371, 375, 384, 388–391, 398, 556,
Primitive neuroectodermal tumor (pnet), 226, 235, 237, 247, 379 557, 560, 565, 566, 570, 614, 615
Primordial follicle, 81, 82, 94, 104 Serous borderline tumor−micropapillary variant, 126, 127
Primordial germ cells, 231, 232 Serous carcinoma, 53, 55, 57–60, 64, 102, 122, 123, 125–129, 134,
Progesterone, 19, 54, 70, 79, 80, 84, 86, 108, 116, 159, 180, 195, 274, 135, 146, 153, 154, 156, 158, 164, 166, 168, 180, 181, 189,
324, 409, 410, 419, 550, 581 196, 203, 215, 218–220, 225, 260, 305, 318, 352, 358, 359,
Progesterone receptor (PR), 5, 10, 26, 34, 80, 116, 148, 157, 159, 214, 386, 387, 389, 390, 392, 394, 420, 421, 546, 547, 554–558,
225, 284, 396, 409, 410 564–568, 608, 614, 617–619
Progestin, 59, 68, 70, 89, 410, 415 Serous cystadenoma, 67, 103, 104, 126, 127, 142–144, 158, 164, 398,
Progestin therapy, 13, 352, 395 556–557
Proliferative phase, 4, 12, 281, 578 Serous endometrial intraepithelial carcinoma, 56
Prophylactic bilateral salpingectomy, 127 Serous ovarian cancer, 219
Psammoma body, 293 Serous psammocarcinoma, 220, 390
Pseudocarcinomatous hyperplasia, 64, 65 Serous serous carcinoma, 129
Pseudoepitheliomatous hyperplasia, 543 Serous tubal epithelial carcinoma (STIC), 133
Pseudomyxoma peritoneii, 567 Serous tubal intraepithelial carcinoma (STIC), 53–58, 62, 72, 123,
Pseudopapillary, nodular, 307 127–130, 134, 141, 154, 168, 349, 352, 354, 384, 386, 387,
p53 signatures, 54, 55, 127–130, 141 554, 555
Pushing border, 40 Serous tubal intraepithelial lesion, 58
Serous tumor, 126, 144–146, 316, 371, 375
Sertoli cell, 6, 14, 227, 299, 310
R Sertoli cell tumor (SCT), 90, 275, 311
Red degeneration, 551 Sertoli-Leydig cell tumor (SLCT), 16, 243, 247, 274, 310, 314–316,
Resistant ovary syndrome, 94 343, 344, 349–351, 399, 561, 563, 618
Rete ovarii, 81, 132, 144, 223, 414, 417, 418 Sertoli-Leydig cell tumor with heterologous Elements, 316
Rete testis, 81, 223 Severe preeclampsia, 521
Retinal anlage tumors, 249 Sex-cord element, 299–300
Retrodifferentiation, 254 Sex cord-stromal tumor (SCSTs), 16, 180, 227, 247, 265, 266, 268,
Retroplacental hemorrhage, 467–469, 486 273–275, 277, 279, 282, 292, 300, 303, 310, 312, 313, 317,
Rhabdomyoma, 533 319, 325, 351
Rhabdomyosarcomas (RMS), 26, 32, 33, 41, 42, 61, 222, 226, 237, Sex cord-stromal tumor with annular tubules (SCTAT), 267, 268, 274,
314, 317, 379 275, 312, 313
Sex steroid hormone, 409
Signet-ring-like cells, 193
S Signet ring stromal tumors (SRSTs), 279, 289–292, 317
Salpingitis isthmica nodosa, 58, 62, 65, 68, 69, 555 Simple cysts, 70, 102, 103, 105, 398, 568
Sarcomas, 1, 8–13, 19, 26, 31, 33, 61, 239, 280, 318, 380, 552, 615 Simple papillae, 614
Sarcomatoid, 133, 279, 281, 282, 300, 301, 303, 304, 307, 315, 316, Single exon gene, 303
319, 568, 618, 621 Single umbilical artery (SUA), 465, 494, 507, 508, 523
Sarcomatous overgrowth, 32–34 Single-exon gene, 303
Schiller-duval body, 192, 258–260, 263, 308, 564 Small cell carcinoma of the hypercalcemic type (SCCH), 222, 251,
Schistosomiasis, 65, 102 253, 304, 309, 329, 357, 360
Sclerosing stromal tumor (SST), 274, 275, 279, 284, 286–289, 291, Small cell carcinoma, pulmonary type, 222, 361
298, 303, 329 Small placentas, 464
Sebaceous tumors, 248 Smooth muscle tumor, 1, 5–7, 16, 19–26, 28, 35, 37, 39, 40, 279, 395,
Seborrheic keratosis, 543 429, 450, 453, 569
Secondary follicles, 81, 82, 104 Smooth muscle tumors of uncertain malignant potential, 16, 24, 26, 28
Secondary Müllerian system, 121, 122, 384 Solid, (pseudo)endometrioid, transitional cell carcinoma-like (set),
Secondary tumors, 127, 323–361 128
Secretory cell expansion (SCE), 55, 127 Solitary fibrous tumor, 35, 42, 380, 381
638 Index