Radiotherapy and Oncology 124 (2017) 277–284

Contents lists available at ScienceDirect

Radiotherapy and Oncology

journal homepage: www.thegreenjournal.com

Perioperative management of sarcoma

Overall survival advantage of chemotherapy and radiotherapy in the

perioperative management of large extremity and trunk soft tissue
sarcoma; a large database analysis
Omar Mahmoud a,b,⇑, Ahmet Tunceroglu a, Ravi Chokshi c, Joseph Benevenia d, Kathleen Beebe d,
Francis Patterson d, Thomas F. DeLaney e
a
Department of Radiation Oncology, Rutgers the State University of New Jersey, Robert Wood Johnson Medical School, New Brunswick; b Department of Radiation Oncology, Rutgers
the State University of New Jersey, New Jersey Medical School; c Department of Surgical Oncology; d Department of Orthopedic Surgery, Rutgers-New Jersey Medical School, Newark;
and e Department of Radiation Oncology, Harvard Medical School, Boston, USA

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: Intergroup 9514 reported promising outcomes with neoadjuvant chemoradiotherapy for large
Received 1 November 2016 extremity/trunk soft tissue sarcoma (ESTS). One decade later, optimum integration of chemotherapy
Received in revised form 10 July 2017 and radiotherapy into the perioperative management of ESTS remains to be defined.
Accepted 16 July 2017
Methods: The National Cancer Data Base was used to identify 3422 patients who underwent resection for
Available online 1 August 2017
large (>8 cm) high-grade STS between 2004 and 2013. Chi-square analysis was used to evaluate distribu-
tion of patient and tumor related factors within treatment groups while multivariate analyses were used
Keywords:
to determine the impact of these factors on patient outcome. The Kaplan Meier method and Cox propor-
Extremity and trunk soft tissue sarcoma
Adjunctive
tional hazards model were utilized to evaluate overall survival according to treatment regimen, with a
Combined modalities secondary analysis based on propensity score matching to control for prescription bias and potential con-
Radiotherapy founders imbalance.
Chemotherapy Results: Hazard ratio for death was reduced by 35% with radiotherapy and 24% with chemotherapy, com-
Survival outcomes pared to surgery alone. Combination therapy incorporating both modalities improved 5-yr survival
(62.1%) compared to either treatment alone (51.4%). The sequencing of chemotherapy and radiotherapy
or whether they were delivered as adjuvant vs. as neoadjuvant therapy did not affect their efficacy.
Age > 50 years, tumor size > 11 cm, and tumor location on the trunk/pelvis were poor prognostic factors.
Conclusion: Our analysis suggests that adjunctive modalities are both critical in the treatment of large
high-grade ESTS, improving survival when used individually and demonstrating synergy in combination,
regardless of sequencing relative to each other or relative to surgery; thus providing a framework for
future randomized trials.
Ó 2017 Elsevier B.V. All rights reserved. Radiotherapy and Oncology 124 (2017) 277–284

The rarity of extremity and trunk soft tissue sarcoma (ESTS) – a modalities has resulted in improved patient outcomes with radio-
heterogeneous group of mesenchymal tumors affecting 12,000 therapy lowering local failure to less than 10% [5,6] and systemic
cases each year [1] – remains a challenge to the completion of ran- chemotherapy reducing distant failure hazard ratios (HR) by
domized trials and has likely slowed therapeutic progress in this 21–29% [7] when compared with surgery alone. Importantly, in a
disease. large meta-analysis, the reduction in distant failure provided by
Compared with amputation, limb sparing surgery has emerged adjuvant chemotherapy translated in improved survival [7]. How-
as the treatment of choice by allowing limb preservation while ever, other studies failed to confirm this advantage with adjuvant
achieving equivalent survival [2]. In large high-grade tumors, the chemotherapy [8], and such benefit is even less clear in the neoad-
local and distant recurrence rates after surgery alone can reach juvant setting [9]. To further confound treatment decisions, the
as high as 30% and 50%, respectively [3,4]. Incorporation of adjunct timing of radiotherapy with respect to surgery is still controversial
[10,11] with no apparent local control advantage for adjuvant vs.
⇑ Corresponding author at: Department of Radiation Oncology, Rutgers The State neoadjuvant approach. As such, national guidelines provide little
University of New Jersey, New Jersey Medical School, Cancer Institute of New Jersey, guidance on the optimum timing of these modalities [12].
Robert Wood Johnson Medical School, One Robert Wood Johnson Place, Floor G2, The concept of combining chemotherapy and radiotherapy is
New Brunswick, NJ 08901-1914, USA. not novel, and, in other solid tumors, led to improvement of all
E-mail address: [email protected] (O. Mahmoud).

http://dx.doi.org/10.1016/j.radonc.2017.07.021
0167-8140/Ó 2017 Elsevier B.V. All rights reserved.
278 Adjunctive therapies in ESTS

treatment end points including survival [13,14]. RTOG 9514 tested The Chi-square test was employed to assess the distribution of
this paradigm in the setting of ESTS and demonstrated not only the these prognostic variables according to type of adjunct treatment
feasibility of neoadjuvant chemotherapy interdigitating with delivered. OS was estimated using the Kaplan–Meier method and
radiotherapy for high-grade ESTS  8 cm, but also it was associated log-rank test. The survival and hazards ratio are all reported with
with an impressive 5-year local control of 78% and distant- 95% confidence intervals (CI). The association between survival
metastasis-free-survival of 79% [15]. Similar results were reported outcomes and treatment strategy was assessed with Cox propor-
by Mullen et al. where a neoadjuvant combined approach yielded a tional hazards models. Factors significant on univariate analysis
7-year DFS of 85% vs. 50% (p = 0.004) in a matched group of histor- (UVA) were included in the multivariate analysis (MVA) model.
ical control patients [16]. As the proportional hazards model may not address the profound
Despite these results, the integration of chemotherapy (CT) and difference of covariates between the treatment subgroups,
radiotherapy (RT) in the management of large high-grade ESTS as propensity-score matching generated a well-balanced patient pop-
well as the optimal timing of these adjunctive modalities with ulation; thus eliminating covariates imbalance that may affect
respect to surgery are still debatable. Using the National Cancer treatment choice and/or treatment outcomes. The propensity score
Data Base (NCDB), we evaluated the pattern of utilization of matching model included selected variables based on the presence
adjunct therapies in this select patient subgroup, assessing the sur- of the profound imbalance and/or based on their expected impact
vival benefit of chemotherapy and radiation, alone and in combina- on the treatment selection or overall outcome. The nearest neigh-
tion, as well as the effect of timing of such therapies on overall bor technique with a caliper distance of 0.25 matching 1:1 was
survival. used. All analyses in this study were conducted in R statistical
environment (R Development core team (2015), version 3.2.2)
[19]. The analysis was conducted in a stepwise fashion starting
Methods by investigating the usage pattern and survival benefit of adjuvant
chemotherapy and/or radiotherapy, each compared independently
The American College of Surgeons and the American Cancer with surgery alone. Subsequently, the single and combined adjunc-
Society sponsor the NCDB that includes abstracted hospital registry tive approaches were compared for survival advantage for one
data from 1500 accredited facilities representing 70% of newly paradigm over the other. Finally, the survival outcomes of different
diagnosed cancer cases nationwide [17]. In 2016, the study was combined modality strategies with respect to their timing in rela-
approved by the institutional review board to acquire ESTS dataset. tion to surgery were compared.
Emulating the RTOG 9514 eligibility criteria, the NCDB coding
key [18] was employed to select patients 18 years or older, without
comorbidities, who were diagnosed with large (>8 cm) non- Results
metastatic high-grade ESTS and who underwent surgical resection
from 2004 through 2013. Among the 30,519 ESTS patients who underwent surgery for
Patients with missing treatment/follow-up information or non-metastatic disease between 2004 and 2013, 13,265 patients
whose tumors were not resected were excluded. Our analysis had high-grade disease and almost half of them had tumor larger
incorporated demographic data including socioeconomic and clin- than 8 cm. Only 3422 patients had complete follow-up information
ical covariates such as age, race, insurance status, median income, and they constituted the entire study cohort (Fig. 1). The median
ZIP code education level, treatment facility type, distance to hospi- age was 60 years (range: 18–90) and 55% were males. The median
tal, tumor size, site, grade, histology, and margin status. tumor size was 12.3 cm (IQR: 10-17cm) involving the lower limb in

Fig. 1. Number of Observations After Each Exclusion Criterion.

 O. Mahmoud et al. / Radiotherapy and Oncology 124 (2017) 277–284 279

64% of the cases with fibromatous-sarcoma as the most frequent (eTable 2). To control for possible confounding factors, a propensity
differentiated histologic type (23%). Baseline patient characteris- score matched population was created according to age, tumor site,
tics are summarized in Table 1. Although the study population con- tumor size, tumor histology, and income level (eTable 3). The sig-
stituted a select group of ultra-high-risk (UHR) patients, nificant survival advantage of combined modality was still evident
perioperative radiotherapy and chemotherapy were delivered in in this population (Fig. 2D).
only 65% and 30% of the patients, respectively, while 25% did not In the 714 patients who received combined modality therapy,
receive any adjunct therapy. Insured patients presenting with lar- the small population size prevented the detection of a significant
ger lower limb tumors and/or living in zip codes with higher edu- survival advantage of either approach with respect to timing in
cation and within 25 miles from the treatment facility were more relation to surgery (neoadjuvant vs. adjuvant) or the sequencing
likely to receive radiotherapy. The technique used was conven- of chemotherapy and radiotherapy relative to each other (MCMT
tional/conformal, intensity modulated radiotherapy and vs. ACMT and MCMT vs. NCMT). The 5-year survival was 65% (CI:
brachytherapy techniques in 80%, 18% and 2%, respectively to med- 59–73), 62% (CI: 55–69) and 57% (CI: 50–66) for NCMT, ACMT
ian doses of 50 Gy and 63 Gy in the preoperative and postoperative and MCMT, respectively (Fig. 3). A significant survival advantage
settings, respectively. Comparable patterns were seen in with any particular combined modality approach was not detected
chemotherapy that was frequently delivered in higher income either on univariate analysis or after propensity score matching to
and higher educational levels younger patients presenting with adjust for covariate imbalances (Done but not shown).
larger lower limb tumors, and/or in those diagnosed after 2009
(Table 1).
At 49 months median follow-up (IQR: 28–72 months), the 5- Discussion
year survival of grade 3–4 ESTS was significantly inferior compared
with grade 1–2 disease (65% vs. 87%; p < 0.001). Within the high- For patients undergoing surgery for high-grade large extremity
grade group, lower 5-year survival was observed in larger tumors: soft tissue sarcoma, we hypothesized that adjunctive modalities-
75%, 64% and 48% (p < 0.001) in less than 5cm, 5–8 cm and larger in combination- may increase overall survival. This analysis con-
than 8 cm tumors, respectively. The previous two comparisons firms our hypothesis and, to our knowledge, is the largest study
illustrated the poor prognosis of our select UHR patients present- to examine the survival benefit of perioperative adjunct treatment
ing with both adverse factors (high-grade and large size) constitut- in this ultra-high-risk ESTS patient population treated in hospital
ing almost one quarter of ESTS patients and their 5-year survival based setting.
was significantly inferior to those presenting with a single factor; High-grade disease has been consistently associated with worse
69% (CI: 68–70) vs. 50% (CI:48–52), respectively. Regarding the local and especially distant control rates with consequent inferior
therapeutic benefit of adjunctive modality, 5-year survival survival [20–22]. Conversely, large tumor size mostly predicts
improved from 38% (CI: 34–41) to 54% (CI:51–56) with radiother- higher rates of distant metastasis and/or death with indeterminate
apy and from 45% (CI:43–47) to 57% (CI:53–60) with chemother- impact on local control rates and/or surgical margins status [20–
apy (Fig. 2A and B). On multivariate analysis, mortality HR was 22]. Adjunctive modalities are typically indicated in high-risk ESTS
significantly reduced by 37% with radiotherapy and by 24% with patients presenting with either factor alone (high-grade or large-
chemotherapy. Poor prognostic factors included older age, larger size tumor) [4,9] to reduce recurrence rates, which approach 50%
tumor, positive surgical margins, and tumors located in the pel- [12]. When both risk factors present together, the recurrence rates
vis/trunk (Table 2). are even higher [20]. Moreover, larger tumor size remains an inde-
Different treatment subgroups were then created to analyze pendent poor prognostic factor for survival even when restricting
survival according to the number of adjunct therapy received: the analysis to high-grade disease [23]. Therefore, an adjunctive
modality is expected to significantly improve survival if it is effec-
(1) In the Single modality group, neoadjuvant radiotherapy tive in reducing the high recurrence rate predicted in our UHR
alone (NRT), adjuvant radiotherapy (ART), neoadjuvant patient cohort comprised exclusively of large and high-grade
chemotherapy (NCT) and adjuvant chemotherapy (ACT) disease.
were delivered in 495, 1027, 133 and 184 patients, In uncommon heterogeneous tumors like ESTS, the beneficial
respectively. effect of adjunct treatment is best demonstrated in the setting of:
(2) Combined modality group was delivered in only 21% of this
UHR cohort. This group was further divided into 3 subgroups (1) A large population size powered to detect a statistically sig-
according to the timing of therapy in relation to surgery: nificant survival difference;
(A) Neoadjuvant combined modality (NCMT) treatment (250 (2) A homogenous high-risk patient population with uniform
patients) consisting of NCT and NRT that were both poor outcome without treatment.
delivered before surgery.
(B) Adjuvant combined modality (ACMT) treatment (261 Due to rarity of ESTS, neither condition has been realized in
patients) consisting of ACT and ART that were both deliv- multiple sample-limited randomized studies or small single insti-
ered after surgery. tution retrospective analyses, resulting in inconsistent outcomes
(C) Mixed combined modality (MCMT) treatment (203 of randomized studies, which have led to some debate regarding
patients) consisting of NRT followed by ACT or NCT fol- the role of adjuvant chemotherapy in the management of ESTS
lowed by ART. [4]. However, the survival benefit of chemotherapy was confirmed
in a recent meta-analysis illustrating that regimens consisting of
Combined approach was more frequently used in younger doxorubicin and ifosfamide conferred an 11% absolute reduction
patients presenting with larger tumors as well as in those with in death [7]. In our analysis, chemotherapy yielded a comparable
greater income, higher education, and those receiving their treat- 5 year-survival benefit (12%, from 45% to 57%; p < 0.001). The mag-
ment after 2008 (eTable 1). As shown in Fig. 2C, the 5-year survival nitude of this survival benefit underscores the exceptional compo-
was 35.7%, 49.4% and 62.1% (p < 0.001) in the surgery alone, single sition of our UHR patient population that, unlike the patients
adjunct modality and combined adjunct modality groups, respec- included in the above meta-analysis, was restricted to high-grade
tively. On MVA, combined modality approach reduced the HR for and large tumors. Failure to differentiate between high-risk and
death (0.69; CI: 0.6–0.81) compared to single modality treatment UHR ESTS patients may dilute a small but real survival benefit
280 Adjunctive therapies in ESTS

Table 1
Baseline Characteristics Stratified by Type of Treatment.

Variable Overall Cohort N = 3422 RT N = 2237 No-RT N = 1185 p > X2 Chemo N = 1028 No Chemo N = 2394 p > X2
Facility type 0.184 0.107
Community Cancer Program 166(5) 111(5) 55(5) 33(3) 133(5)
Comprehensive Cancer Program 804(23) 558(25) 246(21) 204(20) 600(25)
Academic/Research Program 1738(51) 1120(50) 618(52) 481(47) 1257(53)
Integrated Cancer Program/Other 714(21) 448(20) 266(22) 310(30) 404(17)
Insurance
Not Insured 245(7) 131(6) 114(10) <0.001 67(7) 178(7) 0.378
Insured 3177(93) 2106(94) 1071(90) 961(93) 2216(93)
Median Income of zip code, $ 0.105 0.009
<30,000 653(19) 408(18) 245(20) 172(17) 481(20)
30,000–35,999 829(24) 536(24) 293(25) 230(22) 599(25)
36,000–45,999 900(26) 582(26) 318(27) 282(27) 618(26)
>46.000 1040(31) 711(32) 329(28) 344(34) 696(29)
Education 0.006 <0.001
<7% 762(23) 529(24) 233(20) 273(27) 489(21)
7–12.9% 1078(31) 719(32) 359(30) 326(32) 752(31)
13–20.9% 881(26) 565(25) 316(27) 239(23) 642(27)
>21% 701(20) 424(19) 277(23) 190(18) 511(21)
Distance, miles <0.001 0.116
<25 1949(57) 1346(61) 603(51) 560(54) 1389(58)
25–100 958(28) 590(26) 368(31) 307(30) 651(27)
>100 515(15) 301(13) 214(18) 161(16) 354(15)
Year of diagnosis 0.8
2004–2008 1747(51) 1138(51) 609(51) 444(43) 1303(54) <0.001
2009–2013 1675(49) 1099(49) 576(49) 584(57) 1091(46)
Age, y 0.891 <0.001
50 1049(31) 688(30) 361(30) 496(48) 553(23)
>50 2373(69) 1549(70) 824(70) 532(52) 1841(77)
Race 0.172 0.312
Black 408(12) 255(11) 153(13) 114(11) 294(12)
White 2845(83) 1879(84) 966(81) 856(83) 1989(83)
Other 169(5) 103(5) 66(6) 58(6) 111(5)
Gender 0.043 0.054
Male 1870(55) 1251(56) 619(52) 588(57) 1282(54)
Female 1552(45) 986(44) 566(48) 440(43) 1112(46)
Histology Type** <0.001 <0.001
Clear Cell 9(0.2) 3(0.1) 6(0.5) 3(0.2) 6(0.2)
Dermatofibrosarcoma 15(0.4) 8(0.3) 7(0.5) 4(0.4) 11(0.2)
Epithelioid 33(1) 20 (0.6) 13(1) 12(1) 21(0.7)
Fibromatous 781(23) 535(24) 246(21) 159(15.4) 622(26)
Leiomyosarcoma 361(10) 209(9) 152(13) 110(11) 251(10)
Lipomatous 583(17) 428(19) 155(13) 150(15) 433(18)
Neurofribromatous 172(5) 106(5) 66(6) 63(6) 109(5)
Unidentified 947(30) 631(28) 316(27) 293(28) 654(27)
Rhabdomyoscaroma 67(2) 33(1) 34(3) 32(3) 35(1)
Synovial 153(4) 97(4) 56(5) 87(8) 66(3)
Vascular 42(1) 21(1) 21(2) 11(1) 31(0.9)
Rare Subtypes 259(5) 146(4) 113(7) 104(10) 155(6)
Size 0.003 0.001
8–11 cm 1262(37) 844(37) 418(36) 335(33) 927(39)
11–14 cm 771(23) 530(24) 241(20) 262(25) 509(21)
14 cm 1389(40) 863(39) 526(44) 431(42) 958(40)
Grade 0.978 0.194
3 2016(59) 1317(59) 699(59) 588(57) 1428(60)
4 1406(41) 920(41) 486(41) 440(43) 966(40)
Surgical Margin 0.377 0.506
Positive 373(11) 252 (11) 121 (10) 106(10) 267(11)
Negative 3049(89) 1985 (89) 1064 (90) 922(90) 2127(89)
Site <0.001 0.025
Upper Limb 459(13) 301(14) 158(13) 115(11) 348(14)
Lower Limb 2182(64) 1483(66) 699(59) 672(65) 1520(63)
Pelvis/Trunk 781(23) 453(20) 328(28) 249(24) 536(22)
Chemotherapy 0.003 –
Yes 1028(30) 714(32) 314(27) – –
No 2394(70) 1523(68) 871(73) – –
Radiotherapy – 0.003
Yes 2237(65) – – 714(69) 1523(64)
No 1185(35) – – 314(31) 871(36)

**Unidentified Sarcoma: Sarcomatosis not otherwise specified (NOS), Spindle cell, Giant cell, small cell, Undifferentiated sarcoma.
Rare Sarcoma subtypes: Desmoplastic small round cell tumor, fascial, infantile, angiomatoid, fibrous histiocytoma, rhabdoid, giant cell tumor of soft part, alveolar soft part.
Fibromatous Sarcoma: Fibrosarcoma NOS, Fibromyxoma, fibrous histiocytoma.
Lipomatous Sarcoma: Myxosarcoma, Angiomyxoma, Atypical Lipoma, Liposarcoma NOS, Fibromyxolipoma, Myxoid, round cell, Pleomorphic, mixed, dedifferentiated
Liposarcoma and spindle cell lipoma.
Vascular Sarcoma: Hemangiosarcoma, malignant hemangioendothelioma, epithelioid hemangioendothelioma, malignant hemangiopericytoma, lymphangiosarcoma.
Neurofibrosarcoma: Neurosarcoma, Malignant peripheral nerve sheath tumor, malignant perineuroma.
 O. Mahmoud et al. / Radiotherapy and Oncology 124 (2017) 277–284 281

Fig. 2. (A) Overall survival with (red line) and without (blue line) radiotherapy. (B) Overall survival with (red line) and without (blue line) chemotherapy. (C) Overall survival
according to adjunct therapy received: red line – no adjunct therapy (surgery alone), green line – single modality adjunct therapy, blue line – combined modality adjunct
therapy. (D) Overall survival of the propensity score matched population stratified by single (blue line) or combined (red line) modality adjunct treatment.

for chemotherapy. For example, neoadjuvant chemotherapy did the eradication of distant microscopic disease by efficient systemic
not improve outcomes in a study where high-grade disease therapy may potentially raise the relative importance of local con-
>8 cm constituted only 43% of patients [9]; yet, a 0.45 reduction trol as a major determinant of treatment outcome. By improving
in the disease-specific mortality was observed in another neoadju- local control, adjunctive radiotherapy may, to some extent, impart
vant chemotherapy trial where the benefit was limited to high- additive survival benefit secondary to that conferred by
grade tumors larger than 10 cm [24]. chemotherapy, which was delivered in one third of our adjunctive
In a disease whose main cause of death is systemic failure, radiotherapy receiving patient cohort.
radiotherapy, traditionally associated with improved local control, Our study demonstrates that, in addition to increased survival
is expected to have little impact on survival. However, our NCDB with the use of either chemotherapy or radiotherapy, incorpora-
analysis uncovered a 37% reduction in the mortality hazard ratio tion of both adjunct treatments, regardless of their sequence, con-
with radiotherapy administration. Three potential theories can fers an even greater benefit, with a 12% increase in 5-year survival
explain this finding: First, studies in other malignancies, such as when compared with a single adjunctive modality. These findings
breast, and lung cancers, have demonstrated that, in a sufficiently are in line with studies employing combined adjunctive modalities
large population of high-risk patients, radiation therapy can confer that report 5-year local control rates ranging from 78% to 100% and
a locoregional control benefit that eventually translates into a sur- 5-year distant control rates as high as 85% [4,6,15,16,30].
vival advantage [25,26]. The increase in survival observed in our Combined modalities were employed in only 21% of this UHR
study was also demonstrated in other analyses of large datasets patient population. Concern regarding the potential toxicity of
of ESTS [27,28]. While the first study showed only 9% improvement the RTOG 9514 regimen with 5% rate of treatment-related death
in 5-year survival with the use of radiotherapy (compared to 16% [15] remains the main barrier against the generalizability of this
reported in our cohort), all ESTS patients with variable recurrence approach. Indeed, a recent report confirmed that only 22% of Stage
risk were included irrespective of size and grade of tumors [27]. III patients received trimodality therapy [31] highlighting the lack
Conversely, the SEER analysis reported a 33% reduction in the mor- of quality data to guide the incorporation of adjunct treatments in
tality hazard ratio in high-grade ESTS patients treated with radio- this setting. However, optimization of radiotherapy technique [32]
therapy (very comparable to our reported 37% mortality reduction) sequencing, dose/fractionation [33] with refinement of chemother-
[28]. Second, local recurrence was found to be among the most apeutic regimens [34] may improve tolerability of this approach.
important prognostic factors for survival and its occurrence pre- As in other malignancies, clinical factors are not the sole deter-
dicts a 2–4-fold higher risk of distant failure and death [20,29]. minants of the received treatment and socioeconomic variables
Thus, by reducing the rate of local recurrence [6], radiotherapy may be important factor [35,36]. For instance, our analysis
can indirectly reduce distant metastasis rates – expectedly high revealed that both chemotherapy and radiation therapy were more
in UHR patient – which may partially explain our findings. Third, likely to be delivered in patients with higher income and higher
282 Adjunctive therapies in ESTS

Table 2
Prognostic factors for survival on Univariate and Multivariate analysis for the entire cohort.

Variable Univariate Multivariate

HR p-Val HR p-Val
Age, y
50(REF) 1 – 1 –
>50 1.51 <0.001 1.45 <0.001
Median Income of zip code, $
<30,000(REF) 1 – 1 –
30,000–35,999 0.96 0.68 0.95 0.58
36,000–45,999 0.84 0.025 0.87 0.13
>46.000 0.78 0.001 0.89 0.27
Education
>21%(REF) 1 – 1 –
13–20.9% 1.009 0.9 1.06 0.46
7–12.9% 0.89 0.15 1.004 0.96
<7% 0.78 0.003 0.91 0.39
Distance, miles
<25(REF) 1 – – –
25–100 1.13 0.02 1.04 0.45
>100 1.006 0.93 0.88 0.18
Histology Type***
Fibromatous(REF) 1 – 1 –
Clear Cell 1.83 0.14 1.85 0.13
Dermatofibrosarcoma 0.66 0.36 0.54 0.18
Epithelioid 0.79 0.41 0.95 0.87
Leiomyosarcoma 1.1 0.29 1.04 0.65
Lipomatous 0.67 <0.001 0.66 <0.001
Neurofribromatous 1.14 0.23 1.29 0.04
Unidentified 1.04 0.5 1.05 0.42
Rhabdomyoscaroma 1.38 0.055 1.39 0.067
Synovial 0.73 0.026 0.95 0.766
Vascular 1.47 0.056 1.2 0.39
Rare Sarcoma Subtypes 0.88 0.34 0.8 0.12
Surgical Margins
Negative(REF) 1 – 1 –
Positive 1.17 0.036 1.18 0.03
Size
8–11 cm(REF) 1 – 1 –
11–14 cm 1.26 <0.001 1.28 <0.001
14 cm 1.46 <0.001 1.53 <0.001
Site
Lower Limb(REF) 1 – 1 –
Upper Limb 1.02 0.739 0.96 0.68
Pelvis/Trunk 1.23 <0.001 1.21 0.002
Chemotherapy
NO(REF) 1 – 1 –
YES 0.71 <0.001 0.76 <0.001
Radiotherapy
NO(REF) 1 – 1 –
YES 0.59 <0.001 0.63 <0.001

*** Please refer to legend in Table 1.

education levels. Further, radiation therapy was utilized to a and disease related covariates were poor prognostic markers for
greater extent in patients with insurance, raising questions regard- survival, including age older than 50 years [22,37], tumor location
ing the equitable availability of crucial therapeutic modalities. on the trunk or pelvis [22], tumor size >11 cm [22,24,30,37], and
Although the effect of these demographic variables may not be positive surgical margins [2,22,37].
identical in other international health systems, the analysis hints This study is unique not only in analyzing the role and timing of
to the gap between evidence base medicine and its pragmatic both chemotherapy and radiotherapy in a sufficiently large patient
applicability in the community. Acknowledging that demographic population, but it also surveys practice patterns in the community
factors were not the sole determinants of treatment prescription, while focusing on an extremely high-risk patient cohort. In spite of
treatment selection bias were potentially reduced via restricting a propensity score analysis performed to account for confounding
the analysis to UHR ESTS, application of the propensity score factors, the retrospective nature of the study, the heterogeneously
matching and focusing on those who got any form of adjunctive delivered treatments, and lack of specific information regarding
therapy. By reducing this selection bias, the findings in our analysis treatment details (chemotherapy regimens, number of cycles,
maybe applicable outside USA where treatment selection is not duration of chemotherapy delivery, the radiotherapy specifics as
influenced by equivalent demographic variables. target coverage and radiation dose homogeneity), toxicity, disease
In the cohort that received combined modality, our examination free survival or sites of treatment failure (local or distant) remain
could not yield firm conclusions regarding the optimum timing of notable limitations. Nevertheless, it addresses an important and
adjunctive modalities due to the small patient number in each as yet unanswered clinical question regarding the perioperative
treatment subgroup. Congruent with other studies, several patient management of these patients. The results presented here
 O. Mahmoud et al. / Radiotherapy and Oncology 124 (2017) 277–284 283

Fig. 3. Overall survival according to the timing of combined modality therapy with respect to surgery. NCMT: Neoadjuvant combined modality treatment, ACMT: Adjuvant
combined modality treatment, MCMT: Mixed combined modality treatment.

illustrate the need for multi-institutional randomized studies eval- [8] Woll PJ, Reichardt P, Le Cesne A, Bonvalot S, Azzarelli A, Hoekstra HJ, et al.
Adjuvant chemotherapy with doxorubicin, ifosfamide, and lenograstim for
uating the benefit of combined approach while depicting the opti-
resected soft-tissue sarcoma (EORTC 62931): a multicentre randomised
mal sequencing of these 2 adjunctive modalities. controlled trial. Lancet Oncol 2012;13:1045–54.
[9] Gortzak E, Azzarelli A, Buesa J, Bramwell VH, van Coevorden F, van Geel AN,
et al. A randomised phase II study on neo-adjuvant chemotherapy for ’high-
Acknowledgment risk’ adult soft-tissue sarcoma. Eur J Cancer 2001;37:1096–103.
[10] O’Sullivan B, Davis AM, Turcotte R, Bell R, Catton C, Chabot P, et al.
The NCDB is a joint project of the Commission on Cancer of the Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the
limbs: a randomised trial. Lancet 2002;359:2235–41.
American College of Surgeons and the American Cancer Society and
[11] Sampath S, Schultheiss TE, Hitchcock YJ, Randall RL, Shrieve DC, Wong JY.
neither organization have verified or are responsible for the ana- Preoperative versus postoperative radiotherapy in soft-tissue sarcoma: multi-
lytic or statistical methodology employed, or the conclusions institutional analysis of 821 patients. Int J Radiat Oncol Biol Phys
drawn from these data by the investigator. 2011;81:498–505.
[12] National Comprehensive Cancer Network. NCCN clinical practice guidelines in
oncology: Soft Tissue Sarcoma v2.2016. Available at: www.nccn.org/
Appendix A. Supplementary data professionals/physician_gls/pdf/sarcoma.pdf. Accessed September 5.
[13] Rose PG, Lappas PT. Analysis of the cost effectiveness of concurrent cisplatin-
based chemoradiation in cervical cancer: implications from five randomized
Supplementary data associated with this article can be found, in trials. Gynecol Oncol 2000;78:3–6.
the online version, at http://dx.doi.org/10.1016/j.radonc.2017.07. [14] Auperin A, Le Pechoux C, Rolland E, Curran WJ, Furuse K, Fournel P, et al. Meta-
analysis of concomitant versus sequential radiochemotherapy in locally
021. advanced non-small-cell lung cancer. J Clin Oncol 2010;28:2181–90.
[15] Kraybill WG, Harris J, Spiro IJ, Ettinger DS, DeLaney TF, Blum RH, et al. Long-
References term results of a phase 2 study of neoadjuvant chemotherapy and
radiotherapy in the management of high-risk, high-grade, soft tissue
sarcomas of the extremities and body wall: Radiation Therapy Oncology
[1] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin
Group Trial 9514. Cancer 2010;116:4613–21.
2016;66:7–30.
[16] Mullen JT, Kobayashi W, Wang JJ, Harmon DC, Choy E, Hornicek FJ, et al. Long-
[2] Rosenberg SA, Tepper J, Glatstein E, Costa J, Baker A, Brennan M, et al. The
term follow-up of patients treated with neoadjuvant chemotherapy and
treatment of soft-tissue sarcomas of the extremities: prospective randomized
radiotherapy for large, extremity soft tissue sarcomas. Cancer2012. p. 3758-
evaluations of (1) limb-sparing surgery plus radiation therapy compared with
65.
amputation and (2) the role of adjuvant chemotherapy. Ann Surg
[17] American College of Surgeons. National Cancer Data Base. Available at: http://
1982;196:305–15.
www.facs.org/quality-programs/cancer/ncdb. Accessed September 3. 2016.
[3] Rosenberg SA, Glatstein EJ. Perspectives on the role of surgery and radiation
[18] American College of Surgeons. National Cancer Data Base: Patient user file data
therapy in the treatment of soft tissue sarcomas of the extremities. Semin
dictionary items. Available at: http://ncdbpuf.facs.org/node/259. Accessed
Oncol 1981;8:190–200.
September 6. 2016.
[4] Bramwell V, Rouesse J, Steward W, Santoro A, Schraffordt-Koops H, Buesa J,
[19] R Development Core Team. R: A language and environment for Statistical
et al. Adjuvant CYVADIC chemotherapy for adult soft tissue sarcoma–reduced
Computing. 3.2.2 ed. Vienna, Australia: R Foundation for Statistical
local recurrence but no improvement in survival: a study of the European
Computing; 2015.
Organization for Research and Treatment of Cancer Soft Tissue and Bone
[20] Trovik CS, Bauer HC, Alvegard TA, Anderson H, Blomqvist C, Berlin O, et al.
Sarcoma Group. J Clin Oncol 1994;12:1137–49.
Surgical margins, local recurrence and metastasis in soft tissue sarcomas: 559
[5] Alektiar KM, Brennan MF, Healey JH, Singer S. Impact of Intensity-Modulated
surgically-treated patients from the Scandinavian Sarcoma Group Register. Eur
Radiation Therapy on Local Control in Primary Soft-Tissue Sarcoma of the
J Cancer 2000;36:710–6.
Extremity. Journal of Clinical Oncology2008. p. 3440-4.
[21] Coindre JM, Terrier P, Bui NB, Bonichon F, Collin F, Le Doussal V, et al.
[6] Yang JC, Chang AE, Baker AR, Sindelar WF, Danforth DN, Topalian SL, et al.
Prognostic factors in adult patients with locally controlled soft tissue sarcoma.
Randomized prospective study of the benefit of adjuvant radiation therapy in
A study of 546 patients from the French Federation of Cancer Centers Sarcoma
the treatment of soft tissue sarcomas of the extremity. J Clin Oncol
Group. J Clin Oncol 1996;14:869–77.
1998;16:197–203.
[22] Zagars GK, Ballo MT, Pisters PW, Pollock RE, Patel SR, Benjamin RS, et al.
[7] Pervaiz N, Colterjohn N, Farrokhyar F, Tozer R, Figueredo A, Ghert M. A
Prognostic factors for patients with localized soft-tissue sarcoma treated with
systematic meta-analysis of randomized controlled trials of adjuvant
conservation surgery and radiation therapy: an analysis of 1225 patients.
chemotherapy for localized resectable soft-tissue sarcoma. Cancer
Cancer 2003;97:2530–43.
2008;113:573–81.
284 Adjunctive therapies in ESTS

[23] Suit HD, Mankin HJ, Wood WC, Gebhardt MC, Harmon DC, Rosenberg A, et al. [31] Sherman KL, Wayne JD, Chung J, Agulnik M, Attar S, Hayes JP, et al. Assessment
Treatment of the patient with stage M0 soft tissue sarcoma. J Clin Oncol of multimodality therapy use for extremity sarcoma in the United States. J
1988;6:854–62. Surg Oncol 2014;109:395–404.
[24] Grobmyer SR. Neo-adjuvant chemotherapy for primary high-grade extremity [32] Wang D, Zhang Q, Eisenberg BL, Kane JM, Li XA, Lucas D, et al. Significant
soft tissue sarcoma. Annals of Oncology2004. p. 1667-72. Reduction of Late Toxicities in Patients With Extremity Sarcoma Treated With
[25] Douillard JY, Rosell R, De Lena M, Riggi M, Hurteloup P, Mahe MA, et al. Impact Image-Guided Radiation Therapy to a Reduced Target Volume: Results of
of postoperative radiation therapy on survival in patients with complete Radiation Therapy Oncology Group RTOG-0630 Trial. J Clin Oncol
resection and stage I, II, or IIIA non-small-cell lung cancer treated with 2015;33:2231–8.
adjuvant chemotherapy: the adjuvant Navelbine International Trialist [33] Haas RL, Miah AB, LePechoux C, DeLaney TF, Baldini EH, Alektiar K, et al.
Association (ANITA) Randomized Trial. Int J Radiat Oncol Biol Phys Preoperative radiotherapy for extremity soft tissue sarcoma; past, present and
2008;72:695–701. future perspectives on dose fractionation regimens and combined modality
[26] Clarke M, Collins R, Darby S, Davies C, Elphinstone P, Evans V, et al. Effects of strategies. Radiother Oncol 2016;119:14–21.
radiotherapy and of differences in the extent of surgery for early breast cancer [34] Palassini E, Ferrari S, Verderio P, De Paoli A, Martin Broto J, Quagliuolo V, et al.
on local recurrence and 15-year survival: an overview of the randomised trials. Feasibility of Preoperative Chemotherapy With or Without Radiation Therapy
Lancet 2005;366:2087–106. in Localized Soft Tissue Sarcomas of Limbs and Superficial Trunk in the Italian
[27] Hou CH, Lazarides AL, Speicher PJ, Nussbaum DP, Blazer 3rd DG, Kirsch DG, Sarcoma Group/Grupo Espanol de Investigacion en Sarcomas Randomized
et al. The use of radiation therapy in localized high-grade soft tissue sarcoma Clinical Trial: Three Versus Five Cycles of Full-Dose Epirubicin Plus Ifosfamide.
and potential impact on survival. Ann Surg Oncol 2015;22:2831–8. J Clin Oncol 2015;33:3628–34.
[28] Koshy M, Rich SE, Mohiuddin MM. Improved survival with radiation therapy in [35] Deb S, Pendharkar AV, Altekruse S, Ratliff JK, Desai A. 105 The effect of
high-grade soft tissue sarcomas of the extremities: a SEER analysis. Int J Radiat socioeconomic status on gross total resection, radiation therapy, and overall
Oncol Biol Phys 2010;77:203–9. survival in patients with gliomas. Neurosurgery 2016;63(Suppl 1):144–5.
[29] Eilber FC, Rosen G, Nelson SD, Selch M, Dorey F, Eckardt J, et al. High-grade [36] Lin CC, Bruinooge SS, Kirkwood MK, Hershman DL, Jemal A, Guadagnolo BA,
extremity soft tissue sarcomas: factors predictive of local recurrence and its et al. Association between geographic access to cancer care and receipt of
effect on morbidity and mortality. Ann Surg 2003;237:218–26. radiation therapy for rectal cancer. Int J Radiat Oncol Biol Phys
[30] Gronchi A, Verderio P, De Paoli A, Ferraro A, Tendero O, Majo J, et al. Quality of 2016;94:719–28.
surgery and neoadjuvant combined therapy in the ISG-GEIS trial on soft tissue [37] Maretty-Nielsen K, Aggerholm-Pedersen N, Safwat A, Jorgensen PH, Hansen
sarcomas of limbs and trunk wall. Ann Oncol 2013;24:817–23. BH, Baerentzen S, et al. Prognostic factors for local recurrence and mortality in
adult soft tissue sarcoma of the extremities and trunk wall: a cohort study of
922 consecutive patients. Acta Orthop 2014;85:323–32.