WHO

WHO
WHO
Technical
Technical
Technical
Report
Report
Report
Series
Series
Series
941941
941

WHO
WHO
WHOEXPERT
EXPERT
EXPERTCOMMITTEE
COMMITTEE
COMMITTEE

BIOLOGICAL STANDARDIZATION
BIOLOGICAL STANDARDIZATION
BIOLOGICAL STANDARDIZATION
This
This
This
report
report
report
presents
presents
presentsthe
the
recommendations
the
recommendations
recommendations of of
a of
WHO
a WHO
a WHOExpert
Expert
Expert
Committee
Committee
toto
Committee
the
to
the
commissioned
adoption
the
commissioned
adoption
commissioned
adoption
of of
international
of
toto
international
coordinate
international
to
coordinate
coordinate
activities
activities
activities
recommendations
recommendations
recommendations
leading
leading
forfor
leading
the
for
the
the ON
ON
ONBIOLOGICAL
BIOLOGICAL
BIOLOGICAL
production
production
and
and
production
and
thethe
and
and
and
control
control
establishment
the
control
establishment
establishment
of of
of of
vaccines
of
vaccines
vaccines
international
of
and
international
and
and
other
international
other
other
biologicals
biological
biological
biologicals
biological
biologicals
reference
reference
reference STANDARDIZATION
STANDARDIZATION
STANDARDIZATION
materials.
materials.
materials.
TheThe The
report
report
report
starts
starts
starts
with
withwith
a discussion
a discussion
a discussionof of
general
of
general
generalissues
issues
issues
brought
brought
brought
toto the
to
theattention
theattention
attention
of of
the
of
theCommittee
theCommittee
Committee andandand
provides
provides
provides information
information
information
onon the
onthestatus
thestatus
status
andand and
development
development
development of of
reference
of
reference
referencematerials
materials
materialsforforfor
various
various
various
antibodies,
antibodies,
antibodies,antigens,
antigens,
antigens,
blood
blood
blood
products
products
products andand and
related
related
related
substances,
substances,
substances, cytokines,
cytokines,
cytokines, growth
growth
growth
factors,
factors,
factors,
andandand
endocrinological
endocrinological
endocrinological
substances.
substances.
substances. TheTheThe
second
second
second
part
part
part
of of
the
of
the
report,
the
report,
report,
of ofparticular
ofparticular
particular
relevance
relevance
relevancetotomanufacturers
to
manufacturers
manufacturers andandand
national
national
national
regulatory
regulatory
regulatory authorities,
authorities,
authorities, Fifty-sixth
Fifty-sixth
Fifty-sixth
Report
Report
Report
contains
contains
containsguidelines
guidelines
guidelines onon quality,
onquality,
quality,
safety
safety
safety
andandand
effieffi
cacy
effi
cacy
cacy
of of
live
of
live
live
attenuated
attenuated
attenuated rotavirus
rotavirus
rotavirus
vaccines;
vaccines;
vaccines;
DNADNADNA
vaccines;
vaccines;
vaccines; a abiosafety
abiosafety
biosafety
risk
riskrisk
assessment
assessment
assessment forfor
production
for
production
productionandandand
quality
quality
quality
control
control
control
of of
human
of
human
human
inflinfl
uenza
infl
uenza
uenza
pandemic
pandemic
pandemic vaccines;
vaccines;
vaccines;
recommendations
recommendations
recommendations forforfor
inactivated
inactivated
inactivatedrabies
rabies
rabies
vaccines
vaccines
vaccines
produced
produced
produced in in
cell
in
cell
substrates
cell
substrates
substrates andandand
embryonated
embryonated
embryonated eggs;
eggs;
eggs;
forforwhole
for
whole
whole
cell
cell
pertussis
cell
pertussis
pertussis
vaccine;
vaccine;
vaccine;
and andand
forforproduction,
for
production,
production, control
control
control
and
andand
regulation
regulation
regulation
of of
human
of
human
human plasma
plasma
plasma
forforfractionation.
for
fractionation.
fractionation.
Also
Also
Also
included
included
included
areare
aare
list
a list
aof
list
of
recommendations,
of
recommendations,
recommendations,guidelines
guidelines
guidelines
and
and
and
other
other
other
documents
documents
documents forfor
biological
for
biological
biological
substances
substances
substances
used
used
used
in in in
WHO Technical Report Series — 941
WHO Technical Report Series — 941
WHO Technical Report Series — 941
medicine,
medicine,
medicine,
and
andand
of of
international
of
international
international
standards
standards
standards
and
and
and
reference
reference
reference
reagent
reagent
reagent
forfor
biological
for
biological
biologicalsubstances.
substances.
substances.

ISBN
ISBN
ISBN
978-92-4-120941-0
978-92-4-120941-0
978-92-4-120941-0

Geneva
Geneva
Geneva

TRS941cover.indd
TRS941cover.indd
TRS941cover.indd
1 1 1 4.10.2007
4.10.2007
4.10.2007
15:49:13
15:49:13
15:49:13
The World Health Organization was established in 1948 as a specialized
agency of the United Nations serving as the directing and coordinating authority SELECTED WHO PUBLICATIONS OF RELATED INTEREST
for international health matters and public health. One of WHO’s constitutional
functions is to provide objective and reliable information and advice in the field of
human health, a responsibility that it fulfils in part through its extensive programme WHO Expert Committee on Biological Standardization
of publications. Fifty-fifth report.
The Organization seeks through its publications to support national health WHO Technical Report Series, No. 932, 2006 (137 pages)
strategies and address the most pressing public health concerns of populations web site www.who.int/biologicals
around the world. To respond to the needs of Member States at all levels of
development, WHO publishes practical manuals, handbooks and training mate- WHO Expert Committee on Biological Standardization
rial for specific categories of health workers; internationally applicable guide- Fifty-fourth report.
lines and standards; reviews and analyses of health policies, programmes and WHO Technical Report Series, No. 927, 2005 (154 pages)
research; and state-of-the-art consensus reports that offer technical advice and
recommendations for decision-makers. These books are closely tied to the WHO Expert Committee on Biological Standardization
Organization’s priority activities, encompassing disease prevention and control, Fifty-third report.
the development of equitable health systems based on primary health care,
WHO Technical Report Series, No. 926, 2004 (109 pages)
and health promotion for individuals and communities. Progress towards better
health for all also demands the global dissemination and exchange of informa-tion
that draws on the knowledge and experience of all WHO’s Member coun-tries and WHO Expert Committee on Biological Standardization
the collaboration of world leaders in public health and the biomedical sciences. Fifty-second report.
WHO Technical Report Series, No. 924, 2004 (234 pages)
To ensure the widest possible availability of authoritative information and guid-
ance on health matters, WHO secures the broad international distribution of its
publications and encourages their translation and adaptation. By helping to
promote and protect health and prevent and control disease throughout the world,
WHO’s books contribute to achieving the Organization’s principal objec-tive — the
attainment by all people of the highest possible level of health

The WHO Technical Report Series makes available the findings of various international
groups of experts that provide WHO with the latest scientific and technical advice on
a broad range of medical and public health subjects. Members of such expert groups
serve without remuneration in their personal capacities rather than as representatives
of governments or other bodies; their views do not necessarily reflect the decisions or
the stated policy of WHO. An annual subscription to this series, comprising about six
such reports, costs Sw. fr. 132.– or US$ 106.– (Sw. fr. 92.40 in developing countries).
For further information, please contact Marketing and Dissemination,
World Health Organization, 20 avenue Appia, 1211 Geneva 27, Switzerland (tel.:
+41 22 791 2476; fax: +41 22 791 4857; e-mail: [email protected]).

Further information on these or other WHO publications can be obtained from

Marketing and Dissemination, World Health Organization, 1211 Geneva 27, Switzerland
 This report contains the collective views of an international group of experts and
does not necessarily represent the decisions or the stated policy of the World Health Organization

WHO Technical Report Series

941

WHO EXPERT COMMITTEE

ON BIOLOGICAL
STANDARDIZATION

Fifty-sixth Report

Geneva 2007
WHO Library Cataloguing-in-Publication DataPublications of the World Health Organization enjoy copyright
protection in accordance with the
WHO Library Cataloguing-in-Publication Data

WHO Expert Committee on Biological Standardization. Meeting (2007: Geneva, Switzerland)

Fifty-sixth report/WHO Expert Committee on Biological Standardization.

(WHO technical report series ; no. 941)

1. Biological products - standards. 2. Vaccines - standards. 3. Reference standards.

4. Guidelines. I. World Health Organization. II. Series.

ISBN 92 4 120941 0 (NLM classification: QW 800)

ISBN 978 92 4 120941 0
ISSN 0512-3054

© World Health Organization 2007

All rights reserved. Publications of the World Health Organization can be obtained
from WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27,
Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail: [email protected]).
Requests for permission to reproduce or translate WHO publications — whether for sale
or for noncommercial distribution — should be addressed to WHO Press, at the above
address (fax: +41 22 791 4806; e-mail: [email protected]).
The designations employed and the presentation of the material in this publication do not
imply the expression of any opinion whatsoever on the part of the World Health Organization
concerning the legal status of any country, territory, city or area or of its authorities, or
concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent
approximate border lines for which there may not yet be full agreement.
The mention of specific companies or of certain manufacturers’ products does not imply
that they are endorsed or recommended by the World Health Organization in preference to
others of a similar nature that are not mentioned. Errors and omissions excepted, the names
of proprietary products are distinguished by initial capital letters.
All reasonable precautions have been taken by the World Health Organization to verify
the information contained in this publication. However, the published material is being
distributed without warranty of any kind, either expressed or implied. The responsibility for
the interpretation and use of the material lies with the reader. In no event shall the World
Health Organization be liable for damages arising from its use.
This publication contains the collective views of an international group of experts and does
not necessarily represent the decisions or the stated policy of the World Health Organization.

Typeset in Switzerland.
Printed in Singapore.
Contents

Introduction 1
Opening remarks by Secretary of the Expert Committee 2

General 2
Developments in biological standardization 2
WHO programmatic issues 2
Vaccines and biological therapeutic products 4
Maintenance of capacity for production of MCR-5 human diploid
fibroblast cells 6
Blood products and related in vitro diagnostic devices (IVDs) 8
Advancement of technical expertise of regulatory authorities in the area
of blood products and IVDs 10
Quality, safety and efficacy of animal plasma-derived antisera 10
Global needs in standardization of products derived by biotechnology 12
International Non-proprietary Names for gene therapy products 13
Reports from the WHO International Laboratories and
WHO Collaborating Centres 14
Feedback from users of WHO biological standardization products 19

International guidelines, recommendations and other matters related

to the manufacture and quality control of biologicals 20
Guidelines for assuring the quality and nonclinical safety evaluation
of DNA vaccines 20
Recommendations for rabies vaccines 21
Guidelines to assure the quality, safety and efficacy of live attenuated
rotavirus vaccines (oral) 22
WHO Biosafety guidelines for the production and quality control of
human influenza pandemic vaccine strains 23
Recommendations for whole cell pertussis vaccines 24
Recommendations for the production, control and regulation of human
plasma for fractionation 25
WHO guidelines on transmissible spongiform encephalopathies
in relation to biological and pharmaceutical products 25
Guidelines to assure the quality, safety and efficacy of human cells
and tissues for transplantation 27
Guidelines for good manufacturing practice for biological products 29
Guidelines for good manufacturing practice for blood
establishments 30
Stability of vaccines 31
Flavivirus vaccines — regulatory expectations 32
Guidelines for acellular pertussis vaccines 33
Recommendations for bacille Calmette-Guérin vaccines 34

International reference materials 35

Proposals for discontinuation of reference preparations 35

iii
Antigens and related substances 36
Yellow fever vaccine — minimum specifications in International Units
per 0.5 ml dose 36
Smallpox vaccine — stability studies 37
Poliovirus, Sabin type 3 — neurovirulence test reference 38
Haemophilus influenzae type b capsular polysaccharide 39

Antisera 40
Dengue virus antibody, human serum 40
Japanese encephalitis virus, human serum 41
Anti-human platelet antigen-1a 42

Blood products and related substances 43

World Health Organization/International Society of Thrombosis
and Haemostasis Liaison Committee report 43
Anti-A and anti-B blood grouping reagents 44
Prothrombin mutation 46
Vitamin B12 and folate in human serum 47
Blood coagulation factor V (plasma) human 47
Blood coagulation factor XI (plasma) human 48
Thromboplastin, rabbit, plain 49

Cytokines, growth factors and endocrinological substances 49

Vascular endothelial growth factor 49
Keratinocyte growth factor 50
Measurement of relative potencies of thermal degradation samples
of the WHO international standard of interferon alpha 2b 51

Diagnostic reagents 53
Reference materials for in vitro diagnostic devices 53
HIV-1 RNA nucleic acid amplification test 53
Anti-HIV tests 54

Annex 1
Guidelines for assuring the quality and non-clinical safety evaluation
of DNA vaccines 57

Annex 2
Recommendations for inactivated rabies vaccine for human use
produced in cell substrates and embryonated eggs 83

Annex 3
Guidelines to assure the quality, safety and efficacy of live attenuated
rotavirus vaccines (oral) 133

Annex 4
Recommendations for the production. control and regulation of human
plasma for fractionation 189

iv
Annex 5
WHO biosafety risk assessment and guidelines for the production
and quality control of human influenza pandemic vaccines 265

Annex 6
Recommendations for whole cell pertussis vaccine 301

Annex 7
Biological Substances: international standards and reference reagents 335

Annex 8
Recommendations, guidelines and other documents for biological substances
used in medicine 337

v
vi
Members
Dr W.G. van Aken, Amstelveen, the Netherlands
Dr R. Dobbelaer, Head, Biological Standardization, Scientific Institute of Public
Health — Louis Pasteur, Brussels, Belgium (Chairman)
Dr F. Fuchs, Director — Lyon Site, French Agency for Safety of Health Products,
Lyon, France
Dr V. Grachev, Deputy Director, MP Chumakov Institute of Poliomyelitis and Viral
Encephalitides, Moscow, Russian Federation
Dr A. Homma, Director, Bio-Manguinhos, Oswaldo Cruz Foundation, Rio de
Janeiro, Brazil
Dr T. Kurata, Director General, National Institute of Infectious Diseases, Tokyo,
Japan
Dr J. Löwer, Director, Paul Ehrlich Institut, Langen, Germany
Dr P. Minor, Head, Division of Virology, National Institute for Biological Standards
and Control, Potters Bar, Herts., England
Dr F. Reigel, Head, Swissmedic, Biological Medicines and Laboratories, Agency
for Therapeutic Products, Berne, Switzerland (Rapporteur)
Professor G.N. Vyas, Department of Laboratory Medicine, University of California,
San Francisco, California, USA (Vice-Chairman)

Representatives of other organizations

Council of Europe, European Directorate for the Quality of Medicines
Mr J.-M. Spieser, European Directorate for the Quality of Medicines, European
Pharmacopoeia Commission, Strasbourg, France
Dr K. H. Buchheit, European Directorate for the Quality of Medicines, European
Pharmacopoeia Commission, Strasbourg, France
Developing Country Vaccine Manufacturer’s Network
Dr S. Jadhav, Executive Director, Serum Institute of India Ltd, Pune, India
European Diagnostic Manufacturers Association
Dr H. Bayer, Roche Diagnostics, Mannheim, Germany
Eye Bank Association of America
P. Dahl, Executive Director/CEO, The Eye-Bank for Sight Restoration, Inc
New York, USA
International Association of Biologicals
Dr A. Eshkol, Vice-President, Scientific Affairs, Serono International SA,
Plan-Les-Ouates, Switzerland
International Federation of Clinical Chemistry and Laboratory Medicine
Professor J.-C. Forest, Centre Hospitalier Universitaire de Québec, Québec,
Canada

vii
 International Federation of Pharmaceutical Manufacturers and Associations
Dr A. Sabouraud, Responsible Pharmacist, Quality Control of Development
Products, Sanofi Pasteur SA, Marcy l’Etoile, France
Dr M. Dûchene, Head of Product Life Cycle Management, GlaxoSmithKline
Biologicals, Rixensart, Belgium
International Society of Blood Transfusion/European Plasma Fractionation Association
Dr P. Strengers, Amsterdam, the Netherlands
International Society on Thrombosis and Haemostasis
Professor I. Peake, Deputy Director, Division of Genomic Medicine, University of
Sheffield, Royal Hallamshire Hospital, Sheffield, England
Plasma Protein Therapeutics Association
Dr R. Büchel, Director PPTA Source/European Plasma Collection Committee,
Brussels, Belgium
United States Pharmacopeia
Dr R. Dabbah, Division Department of Standards Development US Pharmacopeia,
Rockville, MD, USA

viii
Secretariat
Dr Y. Arakawa, Director, Department of Bacterial Pathogenesis and Infection Control,
National Institute of Infectious Disease, Tokyo, Japan (Temporary Adviser)
Dr T. Barrowcliffe, Head, Haematology, National Institute for Biological Standards
and Control, Potters Bar, Herts., England (Temporary Adviser)
Dr A. Bristow, Head, Technology Development and Infrastructure, National Institute
for Biological Standards and Control, Potters Bar, Herts., England (Temporary
Adviser)
Dr T. Burnouf, Human Plasma Product Services, Lille, France (Temporary Adviser)
Dr D. Calam, Pewsey, Wilts., England (Temporary Adviser)
Dr F. Cardoso de Melo, Head of Hemotherapic and Biological Products Unit,
Agencia Nacional de Vigilancia Sanitaria, Ministerio da Saude, Brasilia DF,
Brazil (Temporary Adviser)
Dr M. Corbel, Division of Bacteriology, National Institute for Biological Standards
and Control, Potters Bar, Herts. England (Temporary Adviser)
Dr P. Dubord, University of British Columbia, Vancouver BC, Canada (Temporary
Adviser)
Dr M. Ferguson, National Institute for Biological Standards and Control, Potters Bar,
Herts. England (Temporary Adviser)
Mrs D. Fehily, Como, Italy (Temporary Adviser)
Dr E. Griffiths, Associate Director General, Biologics and Genetic Therapies, Health
Canada, Ottawa, Ontario, Canada (Temporary Adviser)
Dr S. Inglis, Director, National Institute for Biological Standards and Control, Potters
Bar, Herts., England (Temporary Adviser)
Mrs T. Jivapaisarnpong, Director, Division of Biological Products, Department of
Medical Sciences, Ministry of Public Health, Nonthaburi, Thailand (Temporary
Adviser)
Dr V.K. Kashyap, Director, National Institute of Biologicals, Ministry of Health and
Family Welfare, Noida, India (Temporary Adviser)
Dr H. Klein, National Institutes of Health, Warren G. Magnuson Clinical Center,
Department of Transfusion Medicine, Bethesda, MD, USA (Temporary Adviser)
Dr R.G. Kreeftenberg, Product Manager, Netherlands Vaccine Institute, Bilthoven,
Netherlands (Temporary Adviser)
Dr M. Luz Pombo, Chief, Department Control of Vaccines and Recombinant
Products, Instituto Nacional de Higiene “Rafael Rangel”, Caracas, Venezuela
(Temporary Adviser)
Dr M. Nubling, Paul Ehrlich Institut, Langen, Germany (Temporary Adviser)
Dr S.-N. Park, Director, Division of Viral Vaccines, Korea Food and Drug
Administration, Seoul, Republic of Korea (Temporary Adviser)
Dr M. Powell, Medicines and Healthcare Products Regulatory Agency, London,
England (Temporary Adviser)

ix
 Dr C. Schärer, Swissmedic, Swiss Agency for Therapeutic Products, Inspectorates/
Biologicals, Berne, Switzerland (Temporary Adviser)
Dr J.A. Southern, Advisor to Medicines Control Council in South Africa, Simonstown,
South Africa (Temporary Adviser)
Dr S. Thorpe, National Institute for Biological Standards and Control, Potters Bar,
Herts., England (Temporary Adviser)
Dr Y. Vajaradul, The Bangkok Biomaterial Center, Siriraj Hospital, Mahidol University,
Bangkok, Thailand (Temporary Adviser)
Dr J. Weir, Acting Deputy Director, Office of Vaccines Research and Review,
Center for Biologics Evaluation and Research, Food and Drug Administration,
Rockville, MD, USA (Temporary Adviser)
Dr D. Wood, Coordinator, Quality Assurance and Safety of Biologicals, World Health
Organization, Geneva, Switzerland (Secretary)
Dr J. Wood, National Institute for Biological Standards and Control, Potters Bar,
Herts. England (Temporary Adviser)
Professor Hongzhang Yin, Division of Biological Products, State Food and Drug
Administration, Beijing, People’s Republic of China (Temporary Adviser)

x
Introduction
The WHO Expert Committee on Biological Standardization met in Geneva
from 24 to 28 October 2005. The meeting was opened by Dr J.M. Okwo-
bele, Director, Department of Immunization, Vaccines and Biologicals on
behalf of the Director-General.
Dr Okwo-bele emphasized the importance of the work of the Committee,
which as one of the longest standing WHO Committees started its work
as early as 1947. He stressed the importance of its work in developing,
establishing and promoting written technical standards to assure the quality,
safety and efficacy of vaccines, biological therapeutic products, blood
products and selected in vitro diagnostic devices (IVDs). He also noted that
WHO biological reference preparations are important tools that allow the
comparability of data worldwide in diverse fields of medical practice.
Dr Okwo-bele summarized the key policy developments in biological
standardization since the Committee had last met. He highlighted high level
discussions between WHO and the European Union (EU) Commission
that led to the acceptance by the EU of the principle that WHO biological
reference materials could be regarded as “higher order” reference materials
in the context of EU IVD legislation. This decision was based on the example
provided by the 2nd International Standard for Hepatitis B surface antigen,
established by the Committee in 2004.
He also reported that, in response to the need of the Secretariat and the advice
from the Committee, the meetings of the Expert Committee on Biological
Standardization would be run in a new way. The first and last days would
be plenary sessions, whereas the middle three days would consist of two
parallel tracks, one for vaccines and biological therapeutic products and the
other for blood products and related IVDs.
Dr Okwo-bele informed the Committee that new experts had been appointed
to the Expert Advisory Panel on Biological Standardization. This would
strengthen support to WHO by increasing the diversity of subject experts,
and the geographic and gender balance of the Committee. He also highlighted
key public health issues that would be considered by the Committee during
its meeting. This included establishment of technical documents for the
safe production of pandemic influenza vaccines; guidance on regulatory
expectations for live attenuated rotavirus vaccines; concerns about the
transmission of variant Creutzfeldt–Jakob disease (vCJD) through blood
transfusion; and, the safety of products from human plasma. Dr Okwo-bele
reminded the participants at the meeting that only 60% of blood products
worldwide were produced under stringent regulatory oversight. The new
document on plasma for fractionation would be expected to contribute
significantly to improving the safety of patients.

1
 Dr Okwo-bele reminded the members of the Committee that they do not
represent organizations or governments, but had been invited by WHO
because of their individual expertise. He reminded the Committee that
all decisions made should be based on sound scientific principles. Finally
he invited all participants in the meeting to actively contribute in their
respective capacities.
Dr Roland Dobbelaer was elected as Chairman of the meeting, Dr Girash
Vyas, Vice-Chairman and Dr Franz Reigel, as Rapporteur. After all the
participants had been introduced, the Committee adopted the agenda
(WHO/BS.04.2005) and the timetable proposed.

Opening remarks by Secretary of the Expert Committee

Dr David Wood welcomed the participants. In order to manage the two
parallel sessions for the middle days of the meeting, Dr Johannes Löwer
was elected as Chairman, Professor G.N. Vyas as Vice-chairman and Dr
Franz Reigel as Rapporteur for the track for blood products and Dr Roland
Dobbelaer as Chairman, Dr Takeshi Kurata as Vice-Chairman and Dr Morag
Ferguson as Rapporteur for the track of vaccines and biological therapeutic
products.

General
Developments in biological standardization
WHO programmatic issues
Dr David Wood recalled the WHO mandate of its 192 Member States
to develop, establish and promote international standards for biological
products. In operational terms this mandate covers vaccines, biological
therapeutic products, blood products and selected IVDs.
He discussed recent organizational changes that are intended to provide
clear strategic directions, an appropriate organizational structure, and
appropriate financial and human resources to fulfil the ambitions of WHO.
The new (11th) WHO General Programme of Work proposes a global health
agenda for the years 2006–2015, and identifies among the key challenges
to health the need to close gaps in implementation of knowledge. The
Organization is also emphasizing a results-based management system
using performance indicators. The Committee stressed that WHO should
ensure that biological standardization activities remain highly integrated
whatever the organizational location of the unit(s) that perform the
activities.
Dr Wood presented a process for establishing priorities for written standards
for public health issues of global importance and also the establishment of

2
a portfolio of proposed new and replacement reference preparations. He
described a review process for the draft priorities by stakeholders including
validation by the Expert Advisory Panel for Biological Standardization and
the Committee. The Committee agreed that it is essential to continue to
support the research base to develop standards so that the WHO biological
standardization process produces relevant standards that contribute to
global harmonization. Moreover WHO aims to develop strategies to assess
the impact of WHO standards, to strengthen the global capacity to regulate
biologicals and, in the long term, to ensure that only biological products of
assured quality are used in health systems.
WHO aims to foster a broader geographical representation for its collaborating
centres, and to strengthen the relationships between these centres. With
regard to the biological reference materials programme, Dr Wood reported
some recent developments such as transfer of custodianship of antibiotic
standards from the National Institute for Biological Standards and Control,
England to the European Directorate for the Quality of Medicines (EDQM)
in France; progress with development of regional working reference
materials (an initiative being implemented by the WHO Regional Office for
South-East Asia); a newly established working group on reference material
stability; progress in capacity building in biological standardization (such
as training in biological standardization being provided by the collaborating
centres); and new categories of reference materials (such as genetic testing
standards, and a proposal to consider standards for cell therapy reference
materials). The Committee strongly supports the concept of networking
between the laboratories that operationalize the WHO biological standards
programme.
Promotion of WHO biological standards will be improved through a variety
of means such as publications and web sites, e-learning materials, interactions
with regulators, for example via the recently established Developing
Countries Vaccines Regulators Network and the proposed network of
blood products regulators. New members are being actively recruited to
strengthen the expertise and the geographical and gender balance of the
Expert Advisory Panel (EAP). A web-based collaborative workspace has
been established which provides the possibility for interactive consultations,
and had been used, for example, to consult the EAP on the agenda for the
present meeting. The web space will be increasingly used to consult the
EAP on policy and strategy issues as well as scientific matters. It will help
to develop a collective sense of community and improve knowledge sharing.
Contacts with other users will be improved to promote WHO standards and
to receive input on needs and concerns.
In summary, WHO’s biological standardization programme will provide
global standards for biological medicines, promote and implement global

3
 norms and standards for biological medicines, and continue to strive to
provide the most relevant products to improve global health.

Vaccines and biological therapeutics

Dr David Wood outlined recent developments in immunization, changes at
WHO, recent activities to support the science base for norms and standards,
and short-term priorities.
The new Global Immunization Vision and Strategy 2006–2015 provides
an ambitious framework for immunization for the next decade and will
contribute to improved global health through expanding the scope of
immunization activities in many countries. This will include introduction
of new vaccines for children, reaching older age groups with vaccines,
using vaccines in epidemic preparedness and control and also using
vaccines to prevent some types of cancer. Many new vaccines currently
under development have the potential to prevent childhood deaths (for
example vaccines against rotavirus, measles, pertussis, tetanus, yellow
fever, malaria, human immunodeficiency virus (HIV), tuberculosis (TB),
meningococcal A/C diseases, Japanese encephalitis and streptococcal
pneumonia). Dr Wood emphasized that new regulatory paradigms were
emerging. For example, some vaccines were being registered first in the
countries with the higher disease burdens rather than in the developed
world. This placed large responsibilities on the regulators in these countries
and emphasized the need for continually updated, or new, WHO guidelines
and recommendations.
Dr Wood gave an overview of the background and processes of the new
strategic direction and competency review at WHO. The core competences
at WHO headquarters in the IVB Department were in the Innovation Unit
(IVR Unit), Quality, Safety and Standards (QSS Team), and Access (EPI+
Team). Biological standardization activities were to be managed by the QSS
team. The impact of this change would be to strengthen norms and standards
activities in the vaccine field, and represented a strong commitment of the
Organization.
Dr Wood summarized some important recent scientific activities, namely
a meeting held in April 2005 on the application of molecular methods in
the control of vaccines, that included a review of safety issues associated
with residual cellular DNA in vaccines, and an informal consultation on
the scientific bases for regulatory evaluation of candidate human vaccines
manufactured in plants, that occurred in January 2005. The reports of both
meetings had been submitted for publication in peer reviewed journals. He
also highlighted a WHO/International Association for Biologicals/EDQM
workshop on neurovirulence tests for live virus vaccines held in January 2005,
the report of which was submitted for publication in a peer-reviewed journal.

4
 Dr Wood presented the short-term priorities in developing norms and
standards in the vaccines area and invited comments from the Committee.
The agreed written standards and Working Group activities are given in
Table 1. He also informed the Committee of the establishment of indicators
for the performance of the biological standards programme. For the period
2006–2007 these are listed in Table 2.

Table 1
Short-term priorities for vaccines and biological therapeutics
1a. Written standards (to specify production, QC, nonclinical and clinical trial
standards) to be established by the ECBS in 2006–2007

Document Justification for assignment as a priority

Proposed revised standards
OPV (TRS 904, 2002) 1. Inclusion of biosafety specifications for OPV
stockpile production in post-OPV era
2. Elimination of quality control need for wild
polio type 1
IPV (TRS 910, 2002) 1. Update of biosafety specificatons for
production in post-OPV era
2. Inclusion of Sabin-IPV specifications
BCG (TRS 745, 1985) Recommended by ECBS
Hepatitis B (TRS 786, 1989) Recommended by ECBS
Acellular pertussis (TRS 878, 1998) Recommended by WHO informal consultation
(April 05)
Proposed new standards
Meningococcal A conjugate Clinical trial applications for countries in
African meningitis belt expected before end
2007
Stability evaluation of vaccines Recommended by ECBS
Human papilloma vaccines Anticipating strong interest from countries in
vaccine introduction projects and need for
prequalification
Combination vaccines Increasing evidence of complex quality
interactions in combination formulations
coupled with continued trend for new
combination applications to national
regulatory authorities
Projects to be started in 2006–2007
(will become priorities for establishment
by ECBS in 2008–2009
Live influenza vaccine (TRS 638, 1978) New vaccine licensed in one country; not
covered by the current document
New adjuvants Broad range of new adjuvants in development
Inactivated Japanese encephalitis vaccine New cell culture vaccines in development
(TRS 771, 1988)

5
1b. Guidelines to be established by the ECBS, or Working Groups active
in 2006–2007 to improve safety of biologicals
Document Justification for assignment as priority
Proposed revised standards

Cell substrate guidelines (TRS 878, 1998) Recommended by ECBS, Global Advisory
Committee on Vaccine Safety
GMP for biologicals (TRS 822, 1992) Recommended by ECBS
Proposed working group activities

Safety of gene therapies Serious adverse events (leukaemias)

documented
Safety of virus vectors WHO Informal Consultation (2003)
recommended topic be addressed in 2–3
years
Safety of biotherapeutic medicines Serious adverse events (aplastic anaemia;
unwanted immunogenicity) documented

1c. Global strategies to be developed in 2006–2007

Activity Justification for activity
To promote and monitor implementation
of norms and standards
Develop tools to assess the impact of Monitoring implementation currently done
WHO quality standards in an ad hoc fashion
Develop tools to promote WHO quality WHO manuals dealing with testing
standards procedures for vaccine potency require
producing or updating
To facilitate access to and appropriate
use of international biological reference
materials for vaccines
Develop training materials Current guidance (TRS 932) does not extend
to use of reference materials
To globally align priorities with the WHO
collaborating centres for biological
reference materials
Develop agreed policies and priorities Recommended by ECBS
ECBS, WHO Expert Committee on Biological Standardization; TRS, Technical Report Series.

Maintenance of capacity for production of MRC-5 human

diploid fibroblast cells
Human diploid fibroblasts (HDFs) are cell lines with a maximum lifespan of
approximately 70 population doublings. These cells are used for production
of a number of licensed human vaccines. A cell seed of one HDF cell line,
MRC-5, was established at population doubling level (PDL) 7 before the
emergence of bovine spongiform encephalopathy (BSE) and associated

6
Table 2
Indicators for the performance of the biological standards programme
for vaccines and biological therapeutic products
Indicator Status Target for Target for
end 2007 end 2009
Number of new or revised standards and 0 15 30
reference materials established by the
WHO Expert Committee on Biological
Standardization
Number of new guidelines established, or 0 5 8
globally coordinated research projects under
way, that contributes to improved safety of
biological medicines
Number of global strategies developed to 0 3 6
promote and monitor implementation of norms
and standards and to facilitate access to
WHO collaborating centres.

transmissible spongiform encephalopathy (TSE) regulatory issues. In

practice, MRC-5 cells are often used at around population doubling level
(PDL) 35 although it has been demonstrated that they will grow to PDL 50
without loss of viability.
Despite this lifespan restriction, early calculations of vaccine production
capacity indicated that the total cell production capacity from the ampoules
of MRC-5 cells stored in the master bank would easily provide for their
application as production cell substrates for viral vaccines into the distant
future. However, in reality much of this capacity is lost when low passage
cells are provided for new manufacturing processes and a considerable
proportion of the total replicative capacity of the cells lines is not utilized.
Although there are a number of alternative cell substrates that could be
investigated (eg new alternative HDF cells, HDF cells immortalized with
telomerase and tumour cell lines) it would take a long time to evaluate,
to address regulatory issues and to implement use of alternative cell
substrates in new vaccine manufacturing processes. The Committee
therefore considered a proposal from the National Institute for Biological
Standards and Control (NIBSC) to anticipate the eventual depletion of
MRC-5 early passage stocks by two means, firstly by establishing new
seed stocks of the existing HDFs and secondly by revisiting the current
constraints on maximum population doubling limits for production (WHO/
BS/05.2012).
NIBSC proposes the establishment of a new seed stock of MRC-5 cells
utilizing six ampoules at PDL 7 to generate 384 ampoules at PDL 12. This,
on the basis of replies received from manufacturers, is considered sufficient
to meet the current demands of manufacturers of viral vaccines and their

7
 anticipated needs in the near future. It is anticipated that the complete
history of any serum used in this renewal process would be available.
NIBSC envisage that after consensus from manufacturers on this approach
and the preparation of the seed stock in late 2005, the testing and qualification
regime for the new stock of cells facilitated by WHO would be undertaken
in 2006. A complete testing programme and test data would be presented to
the committee in 2006.
The committee agreed with the proposal to prepare a new bank of MRC-5
cells under GMP at PDL 12 but advised NIBSC to seek scientific advice
from key regional regulatory authorities and to investigate the regulatory
implications of this proposal before proceeding.

Blood products and related in vitro diagnostics

The Committee was informed about the strategy of WHO for strengthening
national regulatory authorities, building technical capacity to address specific
risks and promoting the formation of regional networks of national regulatory
authorities. The activities of WHO in 2004–2005 were summarized, and
attention was drawn to the proposals for establishing Recommendations for
the production, control and regulation of plasma for fractionation at this
meeting, which included guidance on plasma from apheresis and whole
blood donations. The document was also intended to provide guidance for
the increasing number of Plasma Contract Fractionation Programmes and
on the need for the implementation of good manufacturing practice (GMP)
in blood establishments.
A high level WHO–European Commission (EC) meeting on IVDs was
held on 21 April 2005 in Brussels to advocate the adoption of the WHO
Biological Reference Materials for IVDs within the EC legal framework.
As an example of coordination and harmonization of biological reference
materials, the EC has agreed to adopt the second WHO International
Standard (a global reference preparation) for hepatitis B surface antigen
(HBsAg) in the category of higher order reference material. This action
establishes the principle that WHO reference materials may be regarded as
higher order reference materials. While it welcomed this development, the
Committee nevertheless stressed that urgent action is needed to ensure that
other WHO International Standards for the control of in vitro diagnostic
procedures will be taken as higher order reference materials within the
Common Technical Specifications of lists A and B supporting the European
Directive on in vitro diagnostics.
A consultation meeting was held at WHO headquarters in September 2005
on tissue infectivity distribution in TSEs, and the risk of transmission of
vCJD through blood and blood products. An update of the TSE tissue

8
infectivity categories, first published by WHO in 2003, was agreed in order
to enable an assessment of the risk of transmission of vCJD via blood and
blood products and the selection of measures to avoid the as yet unknown
ways in which the disease might spread. The updated document provides
evidence-based information to Member States, especially those countries
where surveillance systems for BSE and vCJD are not yet in place.
The Committee received a report on a WHO Regional Workshop on Quality
Assurance and Safety of Plasma and Plasma Derivatives held in Cairo,
Egypt from 31 January–3 February 2005. In feedback from the course WHO
was urged to promote consistency in the Region in respect of regulations,
licensing and inspection of blood establishments and plasma fractionation
facilities. The need for a network of national regulatory authorities
within the Region in order to promote harmonization of regulatory
policies was agreed upon. A Steering Committee was formed, comprising
representatives from the national regulatory authorities of Bahrain, Egypt,
the Islamic Republic of Iran and Tunisia that should facilitate developments
in this context. The need for adherence to GMP throughout all steps from
collection of starting material to manufacture and release of product to
assure the safety, quality and efficacy of blood components and plasma
derivatives was stressed. Member States should adopt and apply common
regional guidelines on GMP standards for the provision of safe and effective
blood, blood components and plasma for fractionation. Independent quality
control of blood screening tests should be considered by national regulatory
authorities. The appropriate selection of tests is a key intervention to assure
safety of blood products.
The shortage of production of animal-derived sera (antisera) for therapeutic
use was highlighted. Antisera are largely unavailable, consequently, high
rates of mortality and morbidity are still a reality for some otherwise treatable
conditions. Improvements are urgently needed in some production systems
in developing countries and overall there is a need to provide appropriate
regulatory controls.
The Committee was informed of a meeting held in May 2005 in Bethesda,
MD, USA of key regulators to further develop the concept agreed at the
last meeting of the Expert Committee on Biological Standardization for the
establishment of the forum of regulators with regulatory oversight of blood,
labile blood products and plasma products. The outcome of the meeting and
the further development of this group are described in detail later in this
report.
The Committee noted the priorities identified by WHO to strengthen the
regulation and regulatory oversight of the quality and safety of blood
products, haematological products and IVDs worldwide. However, the
Committee strongly reiterated its opinion stated in 2004 that human and

9
 financial resources available at WHO for work in this important field of
global health remain inadequate and urgently need to be addressed.

Advancement of technical expertise of regulatory authorities

in the area of blood products and in vitro diagnostics
The Committee was reminded of the recommendation made at its 2004
meeting, for WHO to promote worldwide cooperation of national regulatory
authorities to facilitate improvement of their technical expertise regarding
blood, blood products and IVDs, and to establish a global network of
regulatory authorities. At this meeting, the Committee encouraged
WHO to take appropriate steps for realization of the global network by
developing, as a first step, the terms of reference. It also advised that the
group, when formed, should report to the Expert Committee on Biological
Standardization. The Committee was informed of the progress that had
been made. A first meeting was held in Bethesda, MD, USA, in May 2005
to discuss the objectives of the group. These were agreed to be:
— scientific assessment of current and emerging threats to the safety and
availability of blood products;
— scientific assessment of the impact, i.e. potential benefits and drawbacks,
of new technologies in the field of blood products;
— exploration of opportunities among regulatory authorities to cooperatively
address emerging public health challenges;
— exploration of opportunities for regulatory harmonization particularly
in response to emerging public health challenges (eg actions to prevent
transmission of vCJD, tools for removing TSE agents, and virus
inactivation in blood components).
The Committee affirmed its decision of 2004 that WHO would establish a
group for regulators, with the objectives outlined above, and agreed that it
should be called the WHO Blood Regulators Network. The Committee also
recommended that the network should report to it, and that WHO should
provide some support for the activities of the network. The Committee
further recommended that WHO should invite countries that have expressed
an interest in joining the network and that the group would benefit from
some representation from developing countries (eg chairs of regional blood
regulators networks).

Quality, safety and efficacy of animal plasma-derived antisera

The Committee aw informed of the important role of specific therapeutic
treatments (eg snake antivenoms) and passive immunizations for some
infectious diseases (eg rabies immunoglobulins and diphtheria antitoxins)
that are based on antisera derived from the plasma of immunized animals.
Although such products have been used for many years, there are a number

10
of current problems. There has been a sharp reduction in the number of
manufacturers of animal antisera and some production systems in the
developing world are fragile. In some places there is poor regulatory control
over the manufacture of antivenoms and over the import of such products.
Furthermore, access to antisera was in some cases hampered by high
costs; logistical problems in distributing animal sera; and a general lack of
knowledge about prevention and correct medical management of diseases
that could effectively be treated by antisera.
A review of the potential risk of transmission of animal viruses to humans
from the use of animal sera concluded that no cases have been attributed
to these products. It is very likely that, by analogy to the manufacturing
process for products from human plasma, that current manufacturing steps
for animal sera have the potential for virus inactivation and/or virus removal,
but may not be recognized as such by the manufacturers. Improvements
to methods of manufacturing animal antisera can be identified that may
reduce the clinical side-effects linked to animal proteins. The Committee
cautioned though that the impact of any changes to a manufacturing process
on product efficacy should be carefully considered.
The Committee was presented with a series of proposals for possible action
by WHO, namely, to support production of antisera in the developing
world according to GMP; to support viral validation studies of existing
manufacturing steps when needed; to reinforce the capacity of local NRAs
and to establish a pre-qualification system for antisera following the
existing WHO experience in prequalification of other therapeutic products.
The proposed activities should be targeted to the local NRA and local
manufacturers, and they may be organized at the regional level to facilitate
development of regional NRA networks. The Committee advised focusing
on those activities that would have most impact, i.e., GMP for antisera,
good animal husbandry practices, collection practices for animal plasma,
and implementation of existing manufacturing steps possibly contributing
to viral safety (eg caprylic acid precipitation, low pH treatment).
The Committee was also presented with a proposal to develop guidance
and training materials on the prevention, diagnosis and management of
diseases treatable by antisera (especially envenomations) and that WHO
recommendations on the production, control and regulation of animal
plasma-derived antisera should be developed. These could be elaborated
in parallel with, and as a result of, regional workshops which would also
contribute to the goal of education of local inspectors. These activities will
enhance the visibility and awareness of the therapeutic importance of antisera
and support the production and availability of safe antisera products.
The Committee agreed with the planned activities and urged WHO to
mobilize resources to support this area of work.

11
 Global needs in standardization of products derived
by biotechnology
The Committee was informed that control of chronic diseases was an
increasingly important global public health problem that needed to be acted
upon urgently. The means for preventing and controlling most chronic
diseases are well established, and include biotechnological interventions.
The number of approved innovative biotherapeutic products is expected
to increase substantially over the coming years (eg some 500 monoclonal
antibody-based products are currently in the pipeline in the European
Union (EU) and USA alone). In addition, the imminent expiration of the
patents on many biotechnology products will result in a substantial increase
in “follow-on” or “biosimiliar” products.
WHO is receiving requests from countries for advice on appropriate
regulatory oversight for biological therapeutic products, because the
potential for the success of therapeutic biological products used in the
treatment of a wide variety of chronic diseases is being tempered by
concerns over the quality, safety and availability of such products, leading
to uncertainty for national regulatory authorities. The Committee heard
that inappropriate assay methods or poor potency determinations (eg for
recombinant streptokinase) can lead to life or death clinical problems;
and adverse drug reactions, for example unwanted antibody development
in some individuals, could occur following administration of a variety of
products. Increased risks of infections have been seen with blockers of
tumour necrosis factor α and the potential for substandard and counterfeit
biotech products is an important cause for concern.
The Committee received a report on the existing WHO written standards
and reference materials that are, or may be, relevant to biological therapeutic
products, and on the support for the science base in this area provided
through activities such as the series of WHO informal consultations on
standards for cytokines, growth factors and endocrinological substances
and a WHO working group on gene therapies.
The need for further WHO initiatives was discussed. These included
updating of existing WHO written standards, the establishment of new WHO
written standards to promote and develop appropriate understanding of the
risks and benefits of biotechnological products, continued development of
new reference materials and strengthening of technical capacity in national
regulatory authorities in this area.
The Committee took note of the proposed activities and agreed that WHO
should organize a meeting of interested parties to review the issues in
depth and help WHO develop consensus on the global needs, priorities
and potential role for global standardization in the area of biotherapeutics

12
for the major chronic diseases. The Committee also recommended that
WHO should facilitate the strengthening of technical capacity of national
regulatory authorities for regulating biological therapeutic products and
also collate information on substandard or counterfeit biological medicines
for the treatment of chronic diseases.

International Non-proprietary Names for gene therapy

products
An International Nonproprietary Name (INN) is a unique global name for
a single pharmaceutical substance. The use if INNs was initiated in 1953
and they are perceived as an essential component in the regulatory process
of many countries (eg the EU, Japan and the USA). An INN is required
for licensing of products and although assigned to certain biologicals,
particularly biotechnology products (such as rDNA products and monoclonal
antibodies), by convention they are not assigned to vaccines or natural blood
products.
In January 2005, the WHO Executive Board emphasized the importance
of INNs for global public health programmes. However the Board decided
that INNs should only be considered for products at the phase III of clinical
development and only for products already in commercial development.
The Committee was informed that WHO had held a consultation in
January 2005 on INNs for gene therapy products. The participants had
agreed that there was a need for INNs for gene therapy products and had
proposed a possible systematic nomenclature scheme for such products.
However the experts advised that the proposed system should not apply, at
this stage, to other gene transfer products such as DNA vaccines or vaccines
based on live viral vectors.
The scope of gene therapy products is very wide and many trials are being
undertaken globally. These involve a wide range of types of transferred genes
such as antigen, tumour suppressor, cytokine, drug resistance, deficiency
correction and others. A range of different types of gene therapy vectors
are also currently used e.g. adenoviruses, adeno-associated viruses, herpes
simplex virus, certain poxviruses, retroviruses, lentiviruses and plasmid
DNA.
It was agreed that gene therapy has become a clinical reality and that the INN
nomenclature system would contribute to patient safety, provided that both
the vector and the gene transferred could be rapidly identified. Both United
States Adopted Names (USAN)/US Food and Drug Administration (FDA)
and the European Agency for the Evaluation of Medical Products (EMEA)
had given much thought to possible nomenclature systems, and it was agreed
to build on the experience obtained with other complex biologicals (such

13
 as fusion proteins). Therefore the consultation recommended that an INN
system for gene therapy products should be based on a two word scheme.
The first word would define the expression gene and the second the vector
component.
The first word to describe the expression gene would have a prefix which
would be a distinctive name e.g. al-, bel- or val-; an infix which would
identify the gene using existing infixes as for the protein for which the gene
codes; and a suffix — gen or gene. The second word for the vector component
would have a prefix which again contributes to a distinctive name; an infix
e.g. lenti (lentivirus) or retro (other retroviruses) or adeno (adenovirus); and
a suffix to indicate the viral vector “vec”. The consultation agreed that for
medicinal products administered by transfecting a patient’s cells ex vivo,
the cells themselves were considered as the route of administration only,
and not included in the INN.
The Committee noted this proposal and that further work may be required
to refine the recommended INN policy as experience is gained with
applications for gene therapy products.

Reports from the WHO International Laboratories

and WHO Collaborating Centres
The Committee was informed of recent developments at the various
WHO International Laboratories and Collaborating Centres for biological
standardization.

National Institute for Biological Standards and Control,

Potters Bar, England
The Committee was provided with an updated overview from the National
Institute for Biological Standards and Control (NIBSC) regarding the
contribution of the Institute to the WHO biological standardization
programme. The Director, Dr Inglis, outlined significant changes to the way
in which NIBSC would be managed by the UK Department of Health. He
stressed the need for, and was assured of, strong support from the Committee
to ensure that under the new management structure the NIBSC “brand”
was protected and also that maintaining a full spectrum of activities across
all biological medicines, and not just work on vaccines, was of critical
importance for global public health reasons.
The Director gave an overview of the standards programme currently in
progress. In 2004–2005 approximately 5000 shipments of biological
reference preparations were made, representing increase on previous
years of about 10%. The overall inventory of such standards is now over
2 million vials and ampoules. Ongoing work includes 83 active standards
development projects, 61 projects already endorsed by the Committee

14
and 22 new projects identified since 2004. A complete list of all projects
designed to capture all the relevant information and providing a resource
for retaining knowledge over time is now established and was provided to
the Committee. The database provides a tool for the management of projects
through regular reporting of progress, ensuring that key steps are followed,
and resource planning for example in statistics and filling suite capacity.
The list will be reviewed annually by project leaders together with the WHO
secretariat in preparation for the meeting of the Committee. New projects
will be brought to the Committee for discussion and endorsement. The
22 new projects identified since 2004 were agreed by the Committee.
The alignment between the NIBSC programme and potential priorities
for new or replacement reference preparations identified by WHO was
reviewed. In most cases there was a good match, but there were some
exceptions. The Committee therefore recommended further discussion on
the need for WHO international standards for rotavirus, meningococcal A
antisera; meningococcal B antisera; PCR-enhanced reverse transcriptase
(PERT) testing, and a standard for residual cellular DNA levels in biological
medicines.
The Committee received a report on progress made at NIBSC in developing
training in biological standardization. A training course was currently
being piloted. The intention was to construct a lecture based course in the
principles of biological standardization to be supplemented by subsequent
practical training through placements. In the long term a manual on
biological standardization was envisaged.
The risks to global biological standardization if NIBSC were to suffer a
catastrophic loss of stocks were also discussed. Offsite storage of a limited
number of important reference materials has already been undertaken.
The Committee was informed about the ongoing and projected work on
standards in haemostasis and thrombosis from 2005 to 2009 at NIBSC.
Highest priority is given to replacement standards. For new standards,
highest priority is generally given to new therapeutic materials that are, or
are about to be licensed, and to important diagnostic tests.
Finally it was noted, and agreed by the Committee, that the implications of
ISO 34 for the ECBS standards setting process need to be considered by
WHO and the Collaborating Laboratories.

Central Laboratory of the Netherlands Red Cross Blood Transfusion

Service (Sanquin), Amsterdam, the Netherlands
The Committee heard a presentation on behalf of the WHO International
Laboratory at Sanquin Diagnostic Services. It reported that Sanquin had
decided recently to discontinue custodianship for the WHO Biological

15
 Standardization Programme by the end of 2006. The Committee was
informed that the organization had transferred their own activities on
biological standards to another organization. Infrastructure for the
warehousing and shipment of biological standards will be reduced in the near
future. Furthermore employees with expertise in biological standardization
had left the organization and, strategically, Sanquin Diagnostic Services
had decided to focus on their core activities, namely diagnostic services
for the blood transfusion service and national health care needs. Various
attempts had failed to convince the Dutch Government to continue funding
of the important custodianship activity. The Committee recommended that
WHO express appreciation to Sanquin for their work over the years as an
International Laboratory, but also regret that they have decided to stop
performing this function. Furthermore WHO should urgently review the
broader implications for its biological standards programme of the loss of
this laboratory as a collaborating centre.
Transfer of custodianship of the WHO reference preparations currently held
by Sanquin to another party selected by WHO is required. Although Sanquin
Diagnostic Services will cooperate with WHO in the proper transfer of the
activities to this party, the transfer should be completed by the end of 2006
before the shutdown of the facilities.
The status of ongoing standards projects that are the currently the
responsibility of Sanquin was reviewed. Sanquin gave an undertaking to
complete a study of mono-specific anti-HCV antibodies. Sanquin has also
started the preparation of the replacement standard for W1042 anti-hepatitis
B immunoglobulin and will prepare a candidate replacement standard. The
laboratory will also design a protocol for evaluation of the candidate standard
together with another collaborating centre. However Sanquin requested that
completion of the study be taken over by the “new” custodian.
The Committee noted that sustainable government funding of biological
standardization activities had became increasingly difficult to secure in
recent years, a development which may have significant impact for the
biological standardization programme of WHO. A comprehensive advocacy
programme to explain the benefits of biological standardization to policy-
makers was needed.

Center for Biologics Evaluation and Research, Food and Drug

Administraton, Bethesda, MD, United States
The Committee was informed of CBER’s mission as a regulatory authority,
its activities as WHO Collaborative Centre and of recent organizational
changes including the transfer of certain products to the Center for
Drugs Evaluation and Research (CDER). The latter included therapeutic
biologicals, antibiotics, insulin and human growth hormone. It was also

16
noted that medical devices, with the exception of those used in blood banks,
were the responsibility of the Center for Devices and Radiological Health.
CBER’s activities include collaborative studies for reference materials,
research and testing for improving standardization and control of biological
products, training of laboratory personnel, providing expertise to many WHO
working groups and informal consultations, serving on evaluation teams
assessing NRAs according to WHO criteria, and improving communications
about biological medicines. CBER also acted as adviser on quality control
and laboratory related regulatory issues to a PAHO network of quality
control laboratories and national regulatory authorities.
The presentation highlighted the development by CBER of a West Nile virus
panel for the evaluation of nucleic acid amplification tests, and an RNA
subgroup panel for HIV-1 for use by manufacturers to assess compliance
with sensitivity and specificity requirements. CBER had hosted scientists
from Latin America for training in testing diphtheria and pertussis vaccines.
The Committee recommended WHO to request CBER to consider the
applicability of extending the training it offers to PAHO scientists in vaccine
quality control to scientists from other parts of the world.
The current priorities for blood products included updated product
specifications for leukocyte depleted platelets and HBsAg test sensitivity.

WHO Collaborating Center for Research and Reference Services

for Immunological and Biological Products, Japan
The Committee was informed of the activities and different functions
of the WHO Collaborating Center for Research and Reference Services
for Immunological and Biological Products, Japan (NIID). As a WHO
collaborative centre it focuses on four major items:
— standardization and recommendations on the calibration of test methods
for the quality control of biological products and other related activities
(e.g. laboratory tests replacing animals models);
— standardization and calibration of reference preparations for the quality
control of biological products;
— training in methods for the quality control of biological products; and
— collaboration and consultation with other national and international
standard-setting organizations.
Some of the recent activities of the NIID were outlined for consideration by
the Committee. For example, attention was drawn to the fact that various
viruses and certain vaccines, such as whole virus influenza vaccine, may be
associated with leukopenia. A mouse model that is used in Japan to assess
the leukopenic activity of vaccines and monitor consistency of production
was described. This is an example of research efforts towards developing

17
 clinically relevant laboratory test models which, together with postmarketing
surveillance of adverse effects, may contribute towards the improvement of
quality assurance systems for biological products. The NIID representative
proposed that the Expert Committee on Biological Standardization, as well
as each NRA, should make concrete plans on a budget for promoting quality
control activities for biological products.
The NIID representative expressed support for the WHO/Expert Committee
on Biological Standardization initiative to develop Regional Working
Reference Standards in each Region. It was suggested that this might
contribute to reducing the heavy workload of NIBSC involved in providing
numerous International Reference Standards. He presented an example of
an existing regional cooperation between China, Japan and the Republic of
Korea, in quality control of biological products and infection control and
surveillance leading to the establishment of a regional reference standard for
mamushi (Gloydius blomhoffii) antivenom. The Committee recommended
that WHO request NIID to provide further details on the methods used to
standardize this reference material, to see if the principles used might be
generally applicable.

WHO Collaborative Centre for Quality Assurance of Blood Products

and in vitro Diagnostic Devices, Paul Ehrlich Institute, Germany
The Committee was informed of the approval of the institute as a WHO
Collaborative Centre on 6 June 2005 by WHO. A reported was received
on the activities of the Paul Ehrlich Institute in quality assurance of blood
products and of IVDs and the contribution of the Institute in the development
of WHO’s strategy for safe blood/plasma-derived medicinal products
through GMP and educational tools for blood collection establishments.
The Institute trains inspectors in GMP at the regional level, and participates
in workshops on aspects of blood transfusion, licensing and batch release of
biological medicinal products.
The Paul Ehrlich Institute offers training courses for assessors working in
national regulatory authorities on the quality assurance of blood products
and IVDs. For blood products, training topics include bacterial and viral
safety, pyrogen testing, immunoglobulins, antisera, monoclonal antibodies,
coagulation factors, albumin and other blood products. For IVDs,
training topics include control of serological tests, control of nucleic acid
amplification tests and regulatory oversight.
The Paul Ehrlich Institute contributes to the development of guidelines and
recommendations and to the quality assurance of IVDs. It also contributes
to the human and financial resources available to WHO through, for
example, secondment of a staff member to WHO, waiving fees for courses,
or covering its own expenses for the majority of WHO activities.

18
Finally it was noted that various biological reference materials had been
developed at the Paul Ehrlich Institute. The Committee recommended that
WHO should review the suitability of these reference materials as potential
WHO International Standards.
Taking into account all of the above reports from collaborating centres, the
Committee recommended that WHO should:
1. collectively negotiate with the collaborating centres to ensure there is
agreement on which collaborating centre will take forward which new
reference material project, and
2. collate and review the training activities under way in each collaborating
centre to ensure maximum efficiency and synergy in training activities.

Feedback from users of WHO biological standardization products

The Committee was informed of the activities by the WHO South-East
Asia Region in the area of biological standardization through a project to
develop Regional Working Reference Standards for vaccines. This was
being developed by a Regional National Control Laboratory Network
involving the three vaccine producing countries of the Region (India,
Indonesia and Thailand) and was discussed at a WHO meeting held in
Bangkok, Thailand in November 2004. Based on regional public health
needs and regional priorities, the laboratories agreed to initiate work to
develop Regional Working Reference Standards for whole cell pertussis
vaccine and inactivated Japanese encephalitis vaccine.
The meeting in Bangkok had recommended that WHO should develop
advocacy material for biological standardization and also conduct an
assessment of the feasibility of establishing regional biological standards
in South-East Asia. The Committee was informed of the outcome of the
feasibility assessment. It was concluded that the initiative was indeed
feasible and would contribute significantly to strengthening regional
collaboration in the regulation of vaccines. A recommendation was made
that consideration should be given to expanding the planned activities to
include additional countries. Parallels were drawn with an already established
regional chemical standards programme that included not only countries of
the WHO South-East Asia Region, but also countries from the adjacent
WHO Western Pacific Region. Establishing cross-regional collaborations
for biological standards would, in time, become very beneficial. Moreover
during the course of the feasibility assessment it had become apparent that
an inventory of critical reagents for batch release activities for vaccines
would be useful, and WHO should explore the possibility of helping to
source these materials. It was concluded that the priorities for use of
resources should be training activities rather than laboratory facilities such
as filling capacities for standards.

19
 A need for increased geographical representation in WHO collaborative
studies from laboratories was raised in discussion. Consequently the
Committee requested WHO to consider publishing the details of forthcoming
collaborative studies and requesting expressions of interest in participating
in the studies. The Committee also acknowledged the important role that
proficiency studies may play in increasing capacity and so recommended
that WHO should investigate the inclusion of proficiency testing for vaccines
within the South-East Asia Region national control laboratory network.

International guidelines, recommendations

and other matters related to the manufacture
and quality control of biologicals
Guidelines for assuring the quality and nonclinical safety
evaluation of DNA vaccines
A new approach to vaccination involving the direct introduction of
plasmid DNA containing the gene encoding the antigen against which
an immune response is sought into appropriate host tissues with the in
situ production of the target antigen(s) has been developed in the past
10 years. This approach offers a combination of potential advantages over
the more traditional approaches, including the stimulation of both B and
T cell responses, improved thermal stability of the vaccine, absence of
adventitious agents and the relative ease of large scale manufacture. Many
scientific publications have addressed the potential of DNA vaccination
and immune responses in animal models have been obtained using genes
from a variety of infectious agents including influenza virus, hepatitis
B virus, human immunodeficiency virus, rabies virus, lymphocytic
choriomeningitis virus, malaria and mycoplasma. However, the immune
responses observed in animal models have generally not been reproduced in
humans when phase I clinical trials were undertaken and many approaches
have been and are being followed in order to enhance the human immune
response.
DNA vaccines have been licensed for veterinary use and efficacy in
animal target species has been observed. Although the quality and safety
considerations for vaccines for veterinary use differ from those for vaccines
for human use, experience with veterinary DNA vaccines can provide
valuable information for the control and use of human DNA vaccines.
A revision of the WHO guidelines for DNA vaccines was initiated in 2002
to reflect the rapidly accumulating knowledge on these vaccines and an
updated draft was considered in detail in July 2005 at a WHO Consultation
attended by regulators, industry representatives and academic experts. On the
basis of this meeting, a proposed revised version of the guidelines had been

20
prepared (WHO/BS/05.2013) and was considered by the Committee. This
revision includes information on plasmid DNA vaccines for prophylactic
and therapeutic use but does not cover plasmid DNA vaccines for use in
gene therapy, DNA vaccines derived in eukaryotic cells, bacterial vectored
plasmid DNA vaccines or synthetic oligonucleotides.
These guidelines are intended to provide information and guidance to
national regulatory authorities and vaccine manufacturers concerning the
characteristics, production, quality control and nonclinical development of
DNA vaccines. The section on nonclinical development had been drawn up
in response to the use of different experimental approaches to enhancing the
efficacy of DNA vaccines, which may raise specific safety concerns.
After making suitable amendments, the Committee recommended adoption
of the guidelines and agreed that the text should be appended to its report
(Annex 1).

Recommendations for rabies vaccines

The last revision of the WHO requirements for rabies vaccines for human
use was in 1980 although an additional document, WHO Requirements for
rabies vaccine (inactivated) for human use produced in continuous cell lines
was published in 1987 to take into consideration advances in the development
of cell culture derived vaccines. An amendment which updated the section
on the International Standard for Rabies Vaccine was published in 1994.
Since these two sets of requirements for rabies vaccines were published
there have been many developments in the production and quality control
of vaccines as well as in their overall regulation. In particular, safety issues
have been carefully reconsidered over the last 10 years.
The availability and use rabies vaccines produced in cell culture or purified
from embryonated duck egg has dramatically decreased the number of
human deaths from rabies worldwide, but most notably in countries where
canine rabies is endemic. The revised scope of the WHO recommendations
(WHO/BS/05.2015) encompasses vaccines produced in cell substrates,
ranging from primary cells (hamster kidney and chick embryo fibroblasts),
human and monkey diploid cells (MRC-5 and FRhL-2 cells), to continuous
cell lines such as Vero cells. Vaccines purified from duck embryos are also
within the scope of the document. However, neural tissues are no longer
considered as suitable substrates for the production of rabies vaccine.
Relevant guidance documents published since the last revision of the
requirement for rabies vaccines have also been considered in this revision,
including the WHO guidelines on animal cell substrates which include the
revised requirement for not more than 10 ng of residual host cell DNA per
single human dose.

21
 In addition to the substrates for vaccine production, the major issues
addressed in this revision are biosafety levels used for production and quality
control areas; testing for adventitious agents; the genetic characterization of
virus strains used in vaccine production; the inactivation process; the test
for effective inactivation; the use of in vitro assays for determination of the
antigen content as a measure of consistency of production; and potency
tests. The stability of the final product and intermediates, guidance on the
quality control of vaccines to be administered by the intradermal route and
clinical evaluation of vaccines are also addressed in separate sections.
The Committee, after making suitable amendments, advised that the
recommendations are adopted and appended to its report (Annex 2). In
addition, in view of the extensive text on testing for adventitious agents in
this document, the Committee recommended that the Secretariat consider
development of a separate document on testing for adventitious agents. This
would facilitate consistency between all WHO guidance documents which
include testing for adventitious agents.

Guidelines to assure the quality, safety and efficacy of live

attenuated rotavirus vaccines (oral)
The high incidence of rotavirus disease around the world has prompted
international agencies and partnerships including the WHO, the Global
Alliance for Vaccines and Immunization (GAVI), and the Children’s Vaccine
Program at the Program for Appropriate Technology in Health (PATH) to
identify rotavirus vaccine development as one of their highest priorities.
Examples of vaccines approved or in development include multivalent
human–bovine and human–rhesus reassortant vaccines and monovalent
vaccines containing a single attenuated human rotavirus strain, natural
bovine–human reassortants and animal rotavirus strains.
A WHO Consultative group met in February 2005 to review the scientific
basis for quality, safety and efficacy of live attenuated rotavirus vaccines.
Issues considered included the control of vaccine source materials and
vaccine production, nonclinical and clinical evaluation of vaccines based on
existing WHO guidelines in general and issues of quality, safety and efficacy
specific to rotavirus. Draft guidelines (WHO/BS/05.2014) were subsequently
developed and reviewed at an WHO informal consultation in August 2005.
Issues relevant to the production and control of rotavirus vaccines are
the sensitivity of the virus at acid pH; testing for adventitious agents and
qualification of the vaccine source materials; the need for a full passage
history of the seed materials to identify all substrates through which they
have been passed; the levels of impurities such as residual cellular DNA;
assays to assess the concentration of virus throughout the production
process; and the applicability of suitable reference materials which are likely

22
to be product-specific; and stability of the final product in the presence or
absence of the vaccine diluent which is added prior to vaccination.
Claims of attenuation of the strains used in the production of rotavirus
vaccines are based on clinical experience in human subjects rather than
on animal models which are not readily available or laboratory markers
of attenuation which are not well-defined. Potential laboratory markers
of product consistency include genetic sequence information which may
demonstrate that a new virus seed is similar to the previous seed and that
each could be distinguished from the parent virus.
Clinical issues are considered in detail in the draft guidelines and, although
many of the points of possible concern are generic, the guidelines emphasize
that each candidate needs to be examined individually as there may be
significant product specific issues. These will include the dose required to
result in immunization, the possibility of transmission to contacts of vaccinees
and the genetic stability of the virus on replication in the gut of recipients.
The Committee agreed that the use of primary cells for the production of
rotavirus vaccine should be discouraged, as primary cell cultures prepared
from wild animals often show a high frequency of viral contamination.
After making suitable amendments, the Committee agreed that the
Guidelines should be adopted and appended to its report (Annex 3).

WHO Biosafety guidelines for the production and quality

control of human influenza pandemic vaccine strains
WHO guidance on the production of pilot lots of inactivated influenza vaccines
from reassortants derived from avian influenza viruses was prepared in 2003
in response to the threat of a pandemic posed by highly pathogenic H5N1
avian influenza viruses and the need to begin development of experimental
vaccines. This threat still persists and several countries are planning large-
scale production of H5N1 vaccine. The earlier guidance has therefore been
reassessed in light of the larger scale of intended production and also in light
of the experience gained from developing and testing vaccine reference viruses
derived by reverse genetics from highly pathogenic avian influenza viruses.
Revised guidelines had thus been prepared (WHO/BS/05.2026) which detailed
international biosafety expectations for large-scale vaccine production and
control and specified steps to minimize the risk of introducing influenza virus
strains with pandemic potential from a vaccine manufacturing facility into
the community. The risk assessment approach used previously for production
of pilot lots of vaccine was followed. However, it is anticipated that the risks
associated with large scale production are likely to be different from those
associated with pilot lots, e.g. the “open” aspect of some production processes
and the quantity of virus-containing waste. Moreover the revised document

23
 had been expanded to encompass current vaccine development pathways. The
guidelines consider a range of hazards associated with manufacturing and
laboratory testing of vaccines for use during a pandemic which are dependent
on the type of virus strain used for production (reassortant versus wild type),
method of production (egg-based versus cell-based) and intended viability of
the virus in the product (inactivated versus live attenuated).
Although the guidance was influenced considerably by the experience gained
with the H5 strains, it will also be applicable to future threats of pandemics
from other potential pandemic strains, such as H2 or highly pathogenic H7.
In addition it covers the possibility that live attenuated virus vaccines, as well
as inactivated vaccines, may be developed as potential pandemic vaccines.
The Committee, after making suitable amendments, advised that the
recommendations should be adopted and appended to its report (Annex 5).
The Committee also recommended that some of the matters raised in the
document be discussed in an international regulatory meeting.

Recommendations for whole cell pertussis vaccines

The last revision of requirements for diphtheria, tetanus, pertussis (DTP)
and combined DTP vaccines was in 1989. Since that time, there have been
a number of developments in the methods of production and quality control
of whole cell pertussis vaccines. Progress was reviewed by a WHO working
group on several occasions (Bethesda, USA, November 2000; Geneva,
Switzerland, July 2003) and key issues presented to the Committee in 2004.
A consensus on a scientific basis for the revision of the requirements was
reached at the Informal Consultation on the Standardization and Control
of Pertussis Vaccines, held in Geneva in March 2005. The main changes
that were incorporated into a draft revised standard (WHO/BS/05.2016)
include an update on reference reagents; the use of the mouse weight gain
test to assess specific toxicity; a clarification of the estimated potency; and
new sections on stability, nonclinical and clinical evaluation of whole cell
pertussis vaccines. In addition, a number of clarifications on technical details
of various tests as well as on the interpretation of the results were provided.
This revision of the recommendations for whole cell pertussis vaccine is
part of the revision of recommendations for DTP and should be read in
conjunction with the revisions already made for diphtheria and tetanus
vaccines (WHO Technical Report Series, No. 927, 2005, Annex 5).
The Committee, after making suitable amendments, advised that the
recommendations should be adopted and appended its report (Annex 6).
In addition, the Committee encouraged surveillance of pertussis strains to
further evaluate the effect of vaccination on circulation of wild-type strains,
and carriage of strains in populations.

24
Recommendations for the production, control and regulation
of human plasma for fractionation
Human plasma is a source of important medicinal products which are obtained
by a combination of large-scale processing steps called “fractionation”. It
is important that these products have an optimal quality and safety profile.
The current WHO requirements for the collection, processing, and quality
control of blood, blood components, and plasma derivatives were published
in 1992 (WHO Technical Report Series, No. 840, 1994). Subsequently,
supplementary guidance was developed on specific topics, such as the
recently published WHO guidelines on viral inactivation and removal
procedures (WHO Technical Report Series, No. 924, 2004), to address the
measures necessary to eliminate blood-borne viruses during processing of
plasma into plasma derivatives.
A drafting group reviewed and revised those parts of the 1992 Requirements
dealing with plasma for fractionation, and prepared draft Recommendations
for consultation (WHO/BS/05.2019). The intended purpose of the new
document was to provide guidance, to both developed and developing
countries, on the production, control and regulation of human plasma for
fractionation as a source material for plasma derived medicinal products.
The Committee considered that the document, by summarizing experience
and information, would serve as a guide to blood establishments to enable
them to understand and facilitate the implementation of appropriate
procedures for the production and control of the starting plasma material,
and to facilitate the provision of safe fractionated plasma products at country
level. It should also be helpful in establishing the supervision by the national
regulatory authorities in their role of assessing the quality and safety of
plasma, prepared either locally or imported, for fractionation, therefore
contributing to the improved quality and safety of human plasma products
worldwide. Manufacturers of plasma derivatives would also benefit from
the document which would provide a common worldwide expectation for
the quality criteria to be applied to plasma for fractionation.
The Committee discussed the proposed Recommendations in detail and,
after making suitable amendments, advised that the Recommendations
should be adopted and appended to its report (Annex 4).

WHO guidelines on transmissible spongiform encephalopathies

in relation to biological and pharmaceutical products
The Committee was informed of the work being done to revise the existing
WHO guidelines on transmissible TSEs to reflect new scientific knowledge.
A variant form of Creutzfeldt–Jakob disease (vCJD), a fatal brain disease
of humans, first appeared in the mid-1990s, as a result of the epidemic

25
 of bovine spongiform encephalopathy (BSE or “mad-cow” disease) in the
United Kingdom. Since first reports in 1996, 157 cases of vCJD have been
reported in the United Kingdom, 14 in France, three in Ireland, and single
cases in Canada, Italy, Japan, the Netherlands, Portugal, Saudi Arabia,
Spain and the USA. Cases of BSE and vCJD have been decreasing in the
United Kingdom in recent years, but both diseases have appeared in other
countries.
Until recently, all vCJD cases were attributed to consumption of beef
products contaminated with the infectious agent of BSE. Since December
2003, two individuals have been identified with vCJD infections probably
acquired from blood transfusions—one with typical vCJD symptoms and
the other with preclinical vCJD involving the spleen and lymph nodes but
not the brain. The fact that the two vCJD infections followed transfusions
of blood from clinically healthy persons who became ill more than a year
after donating blood implies that other blood donors currently incubating
the disease might also be potential sources of infection for recipients. The
possible extent of future blood-borne spread of vCJD infections is still
unknown. The identification of these cases has intensified the concern about
previously unconsidered ways in which the disease might spread. Except
for the two transfusion-transmitted infections, no cases of vCJD have been
linked to any medicinal product to date, and guidelines from WHO and
other authorities have been in place for many years to minimize risk from
such products.
A meeting held at WHO, Geneva, in September 2005 brought together experts
and regulators from around the world to revise the existing WHO Guidelines
on Transmissible Spongiform Encephalopathies (TSEs) in relation to
Biological and Pharmaceutical Products, which provide guidance to both
regulators and manufacturers on ways to prevent potential transmission
of vCJD through human blood and blood products, as well as through
medicinal products prepared with bovine derived materials. The primary
aim of the meeting was to provide updated evidence-based information to
the national regulatory authorities of WHO Member States, especially to
those where BSE has not yet been reported and where surveillance systems
for BSE and vCJD may not be in place. The information is intended to assist
authorities in conducting risk assessments and selecting measures to reduce
the risk of transmitting vCJD by human blood and blood products and other
medicinal products of biological origin (biologicals).
In 2003, WHO Guidelines encouraged authorities to consider introduction
of precautionary measures to keep human blood and blood products safe
from the then-theoretical risk of transmitting vCJD while continuing to
ensure an adequate supply of these important medicines. The experts at the
September 2005 meeting acknowledged that transfusion-transmitted vCJD

26
is no longer just a theoretical possibility and again advised authorities to
assess the vCJD risk in the context of their own national situations, weighing
the potential benefits of adopting precautionary policies to reduce that risk
against the impact that policies might have on the supply of blood.
Earlier WHO meetings had repeatedly stressed that, when feasible, the use
of tissues or body fluids of ruminant origin in the preparation of biologicals
should be avoided. When bovine materials must be used, they should be
obtained from sources assessed to have negligible risk from the infectious
agent of BSE. Most bovine tissues, including bovine muscle, used to
manufacture biologicals, if carefully selected by taking into account the
geographical distribution of BSE and collected according to guidelines,
have little risk of contamination with BSE agent. The recent findings of
disease-associated proteins in the muscles of sheep with scrapie (a disease
similar to BSE but not known to infect humans) and the recognition of
BSE itself in a goat, reinforce the need for manufacturers of biologicals
to maintain the precautionary safety measures laid down in the previous
WHO guidelines. Ruminant blood and blood derivatives, such as fetal calf
serum in cell cultures and bovine serum albumin stabilizers, are also used
to prepare biologicals. Bovine blood has not been identified as a source
of infection, and if properly collected, fetal bovine serum has a negligible
risk. The blood of sheep with experimental BSE or natural scrapie can be
infectious and, because scrapie and BSE agents behave similarly in sheep
and goats, the use of the blood of small ruminants in preparing biologicals
should either be avoided or animals selected very carefully from sources
known to be free of TSEs.
There is a continuing need to ensure that all national regulatory authorities
with limited resources have ready access to reliable and up-to-date
information when assessing TSE risks and evaluating product safety. That
information includes guidance to help minimize or eliminate the risk for
transmitting BSE and vCJD to humans via biologicals and other medicinal
products. The Committee recommended that the revised 2005 guidelines on
minimizing TSE risk should be adopted as the current WHO position and
be published on the WHO web site (www.who.int/biologicals).

Guidelines to assure the quality, safety and efficacy of human

cells and tissues for transplantation
The Committee was given an introduction to the World Health Assembly
Resolution 57.18 regarding allogeneic transplantation accepted in May
2005. The aims of the resolution are:
— to implement effective national oversight of procurement, processing
and transplantation of human cells tissues and organs;
— to cooperate in establishing guidelines to harmonize global practices;

27
 — to consider setting up ethics commissions to ensure the ethical use of
cell, tissue and organ transplantation;
— to extend the use of living kidney donations when possible, in addition
to donations from deceased donors;
— to take measures to protect the poorest and most vulnerable groups from
“transplant tourism” and from the sale of tissues and organs, including
giving attention to the wider problem of international trafficking in
human tissues and organs.
The Committee was reminded that, during its previous meeting it had
agreed with a proposal to develop global guidelines to assure the quality
and safety of cells and tissues used in transplantation. The Committee was
provided with an outline of the principles used to draft the guideline (WHO/
BS/05.2010) namely to:
— build on existing international guidance;
— to distil out key fundamental requirements;
— to focus on tissue or cell products that are commonly traded across
international frontiers; and
— to produce clear, user-friendly documents.
Accordingly, existing international guidance and regulations were reviewed
to identifiy cross-cutting core requirements and to identify core requirements
for specific tissues and cells. The resulting draft document aims to assist
regulators, tissue bank operators and tissue and cell transplant surgeons
globally by providing guidance on safety and quality requirements for the
most used and distributed human tissues and cells. The specifications are
product-based. They are designed to define international expectations for
essential testing, donor selection and processing requirements for selected
tissues or cells. They also provide guidance on the information to be provided
with the product either on the label or in the accompanying documentation.
The draft document did not include specific quality criteria because it was
intended that these would be added in the form of an aide-memoire in the
near future.
A list of tissue and cell products that are transplanted in the human body
and that are used widely, and commonly cross international boundaries, has
been established. This list was agreed at a global consultation with tissue
and cell banking experts and regulators held in Ottawa from 29 November
to 2 December 2004 and selection criteria and specifications were defined
for the following products: frozen bone and tendon; freeze-dried bone;
fresh, glycerolized or cryopreserved skin; fresh or glycerolized amniotic
membrane; cryopreserved cardiac valve; fresh or culture-medium preserved
cornea; fresh haematopoietic stem cells; and cryopreserved cord blood stem
cells.

28
The Committee was informed that the draft guidelines represents a first
step towards providing guidance from WHO in the field of human cells
and tissue transplantation, and that the document will evolve with the
expected rapid developments in the field. The Committee was concerned
that no description of quality criteria was included in the document and
that GMP aspects also needed to be addressed. Accordingly, the Committee
recommended that the document should not be appended to the report of
the Committee, but that the current draft (WHO/BS/05.2010) would be
made available through other channels in the form of an aide memoire. The
Committee also recommended that the current draft be further developed to
include detailed guidance on quality requirements for cells and tissues and
the role of national regulatory authorities in the inspection of facilities, and
that a revised document should be resubmitted to the Committee.

Guidelines for good manufacturing practice for biological

products
The Committee was reminded that GMP is that part of quality assurance
which ensures that products are consistently produced and controlled to the
quality standards appropriate to their intended use and as required by the
marketing authorization. The principles of GMP apply to both drugs and
biologicals. The main principles that apply to both classes of product are
set out in WHO Good Manufacturing Practices (WHO Technical Report
Series) and were last revised in 2003. These technical standards are used in
many Member States for product and establishment licensure and are also
used as the basis for assessing compliance for prequalification of vaccines
for supply to United Nations agencies.
The WHO GMP includes an annex specifically for biological medicines
which was last updated in 1992. Taking into account the developments
in the field since then, the Committee was asked to decide if there is a
continued need for specific WHO GMP guidance for biological products
and if so, if all areas of biological medicinal products are adequately
covered. Topics that might be addressed included blood and plasma-derived
products; production of microorganisms requiring biological containment;
gene and tissue therapies; biological products containing biotechnology-
derived proteins (GMOs); computerized systems (which would not be
unique to biologicals but would need to be developed in collaboration with
colleagues responsible for GMP of chemical medicines); animal testing for
batch release purposes; and a classification system for GMP deficiencies
for biological medicines.
The Committee expressed its firm opinion that the application of GMP in the
production of all pharmaceuticals including biologicals is critical and that
the regulations need to be kept up to date. The Committee noted that revision

29
 of the existing GMP guidelines is urgently needed and recommended that
work on the revision should be started without delay.

Guidelines for good manufacturing practice for blood

establishments
The importance of the implementation of GMP in blood establishments to
improve the quality, safety and availability of plasma for fractionation as
starting material was noted by the Committee. Quality assurance should
cover all stages leading to the finished product, from collection (including
donor selection) to storage, transport, processing, quality control and
delivery of the finished product.
Blood or plasma used as a source material for the manufacture of blood
products should be collected by establishments and be tested in laboratories
which are subject to inspection and approved by a competent national
regulatory authority.
In blood establishments, a systematic approach to ensure compliance at
all steps is needed as well as assessment of compliance by a competent
regulatory authority. Many countries have urged WHO to help build up the
necessary technical expertise in the competent authorities.
The Committee was informed about the experiences from several workshops
held in different regions of the world. It was recognized that the need for
implementing quality assurance in blood establishments was generally
understood. However the workshops showed that technical assistance to build
up a GMP culture was urgently needed. The development and harmonization of
GMP standards at a regional level and the need for training of representatives
of national regulatory authorities, were also recognized as important.
The Pharmaceutical Inspection Convention/Pharmaceutical Inspection Co-
operation Scheme (PIC/S) GMP guide which is already used in 28 countries
is widely accepted as a basis for the establishment of GMP guidelines
in different regions. This guide had been used as the working document
in the regional training workshops organized by WHO for authorities,
blood establishments, fractionators and inspectors. In the context of blood
establishments, the PIC/S GMP guideline introduces the application
of quality assurance principles in all steps involved in the collection,
preparation and testing activities; supports the systematic application of
donor selection criteria for each donation; reduces errors and technical
problems in collection, preparation, testing and distribution; guarantees the
release of products which comply with safety and quality requirements;
ensures adequate documentation and full traceability for each collection and
product; and provides an enabling environment for continuous improvement
of the collection, preparation and testing of starting materials.

30
Examples of building up of regional networks among regulatory authorities
and training of inspectors were presented to the Committee. At the WHO
Regional Office for the Americas/Pan American Health Organization
(AMRO–PAHO) Workshop on GMP, held in Buenos Aires from 28 June–
2 July 2004, a three step procedure was followed:
— building up a common understanding between national regulatory
authority, blood establishments and fractionators;
— building up a common language regarding the GMP standards and their
interpretation; and
— specific training (e.g. for inspectors).
As an essential part of the workshop so-called mock inspections were
performed in different blood establishments where essential aspects of
GMP were discussed with the participants on site, thus enhancing the
participants’ understanding of compliance with GMP in blood collection
and processing.
The Committee agreed with a proposal for a workplan spanning several
years to continue building up technical expertise in GMP for blood
establishments that would include training workshops at a regional level.
The Committee recommended that in the short term, workshops should
be continued using the PIC/S guidelines and that WHO Guidelines be
developed in the mid-term taking into account the experience from the
above workshops.

Stability of vaccines
The Committee was informed that the Secretariat is developing guidance on
regulatory expectations for stability evaluations of vaccines. This guidance
is intended to assist manufacturers and national regulatory authorities in
the design, conduct and evaluation of stability studies for vaccines, to
complement current recommendations for particular vaccines. The need for
this guidance had been identified by the Committee in 2000 and a drafting
group met in 2002 to develop an outline and identify key issues. A further
meeting was held in 2004 and further drafts had been developed in 2005.
The proposed scope of the guidance will cover prophylactic and therapeutic
products for infectious diseases: viral and bacterial, live and non-live,
combinations, rDNA products, DNA vaccines, and vectored vaccines. The
guidance will cover study design issues such as testing frequency; expression
of data; statistical analysis (e.g. trending); and provide links with work on
the stability of reference materials.
The proposed document will include sections on accelerated stability testing
and stability of intermediates, together with regulatory considerations such
as stability assessment required for clinical trial approval, for licensing,

31
 and for lot release. The document will also comment on differences and
similarities in the stability testing of pharmaceuticals and biologicals and
propose the use of the term accelerated stability instead of accelerated
degradation. This is because for pharmaceutical products the tests look
for degradation products whereas for vaccines the tests look for remaining
active product. The selection of assays and other “stability indicating
parameters” to detect changes in the stability profile of a vaccine will be
addressed, as will an assessment of the appropriateness of thermal stability
tests for lot release of live attenuated and other vaccines. Annexes addressing
specific issues for some vaccines e.g. oral polio vaccine, diphtheria, tetanus
(OPV), measles, D, T, whole cell pertussis (wP), meningococcal, BCG,
Haemophilus influenzae type b (Hib), inactivated poliovirus (IPV), rabies,
HepB etc may be included.
The Committee was informed that a further draft together with a questionnaire
would be circulated to manufacturers and national regulatory authorities
worldwide for comment and that a further consultation is planned for 2006
prior to the submission of the guidelines to the Committee in 2006. The
Committee was asked to review the balance between general principles
and specific recommendations for a vaccine or a type of vaccine (e.g., live
attenuated) and the implications of having general guidelines co-existing
with recommendations for specific vaccines. The Committee was also asked
for advice as to whether a shelf-life (storage period) should be assigned
to all stages of production (e.g., source materials seed lots/cell seeds;
intermediates; harvest; bulk; final bulk).
The Committee commented that the proposed guidance was comprehensive
but considered that the document should recommend definition of stability
indicating parameters during vaccine development including physicochemical
characteristics. The thermostability of reconstituted vaccine and storage
conditions for opened vials should also be addressed. It also requested that
the need for testing samples at the frequency cited in other guidelines and
also at one year after the shelf-life should be considered.
In conclusion, the Committee agreed to the plan of action for these new
guidelines.

Flavivirus vaccines — regulatory expectations

The requirements for Japanese encephalitis (JE) vaccine (inactivated) for
human use (WHO Technical Report Series, No. 771, Annex 6) focus on
control of production of neural tissue derived vaccine. A second guidance
document (WHO Technical Report Series No. 910, 2002 Annex 3) makes
specific reference to live, attenuated JE vaccine SA 14-14-2, produced
on primary hamster kidney cells which is widely used in some countries.
Inactivated vaccines containing Beijing 1 or SA 14-14-2 virus produced in

32
cell culture and a live, recombinant, Vero cell derived vaccine containing a
yellow fever/ SA 14-14-2 chimeric virus are now in development.
The Committee was informed that consideration is being given to how the
new JE vaccines should be evaluated, given that efficacy trials with clinical
end-points were considered unsuitable for ethical and practical reasons. A
WHO Consultation had therefore been held in September 2004 to discuss
and achieve consensus on a scientific rationale for correlates of protection,
primary end-points that may be used during clinical evaluation and additional
criteria that may be used in support of registration of new JE vaccines.
The consultation had concluded that neutralizing antibodies provide the
best evidence that protective immunity has been established. The available
evidence included passive transfer studies, linear titre protection relationship,
correlation with neutralization capacity of monoclonal antibodies, protective
immunity against heterologous strains and supportive data from vaccine
efficacy trials. It was therefore agreed that a primary end-point of trials
should be a quantitative analysis of the neutralizing antibody response against
a homologous target virus, that head-to-head comparison with a registered
vaccine should be undertaken and that non-inferiority to the comparator
vaccine according to predefined margins based on percentage seroconversion
demonstrated. A threshold antibody titre of 1 in 10 in a plaque reduction
neutralization (PRNT)50 assay would be considered indicative of immunity.
Additional information in support of registration would include evidence
of immunological memory, qualitative analysis of neutralizing antibody
response, evaluation of T cell immune response and demonstration that
vaccine-induced immunity protects animals in passive protection studies.
The Committee noted the issues discussed including the strains of virus
which should be used in neutralization tests and in the vaccines used in
head-to-head comparisons of old and new vaccines and recommended
that the Secretariat develop guidance on environmental assessment of live,
recombinant (chimeric) flavivirus vaccines. It also recommended that the
recommendations on inactivated JE vaccines be revised and a section on
nonclinical and clinical guidance be added.

Guidelines for acellular pertussis vaccines

Guidelines on production and quality control of acellular pertussis (aP)
vaccines were adopted by WHO in 1996. These guidelines were based
on the concept of “equivalence to lots of known clinical efficacy” and for
equivalence of a new vaccine to be assessed against a homologous approved
aP vaccine. However, this is no longer feasible because such lots are no
longer available and, moreover, there are no correlates of protection, no
consensus on antigens that are important for protection and no generally
accepted animal model to predict clinical efficacy.

33
 The 1996 guidelines also recognized a need for research to standardize
immunogenicity assays in mice by using homologous reference vaccine and
mouse reference sera, on reliable methods for monitoring individual antigens
in final bulk, on immunological markers of protection against pertussis and
the identification of assays that better predict protective efficacy in humans.
The guidelines also required that rigorous post-licensing monitoring for
safety and effectiveness be undertaken. Agreement on studies to be conducted
was reached at a WHO consultation in 2000 and the results of collaborative
studies which were reported at a second consultation in 2003 indicated
progress on potency and toxicity issues. The need for licensing criteria for
new products and new formulations was also identified at that time.
A WHO consultation held in March 2005 recommended revision of the 1996
guidelines for acellular pertussis vaccines by means of a two-step approach.
The first step is the development of licensing criteria for clinical evaluation
of new vaccines and the second, the development of the scientific basis for
recommendations on production and quality control, focusing on potency
and toxicity tests. The issues to be considered in the clinical evaluation of
vaccines include the design of the comparability studies (e.g. should they
be conducted with whole-cell pertussis, or a licensed acellular pertussis
vaccine; should they require tests of superiority/equivalence or non-
inferiority; and what should be the duration of protection as determined in
postmarketing surveillance). Additional issues relate to consideration of the
acellular pertussis component as part of new combination vaccines and the
interactions between components; the postmarketing surveillance required
for combination vaccines; the impact of the immunization schedule and the
concomitant use of vaccines.
The Committee considered that revision of the guidelines on acellular
pertussis vaccine is not urgent as such vaccines are not a priority for WHO
prequalification. However, as a first step, and to anticipate future trends, the
development of guidance for clinical evaluation of new acellular pertussis
vaccines should be initiated.

Recommendations for bacille Calmette-Guèrin vaccines

The Committee was reminded that, despite its long history of use, several
questions regarding BCG vaccines remain unresolved. These include
the parameters to assess the efficacy of BCG vaccines, the significance
of evolution of BCG vaccine strains, the clinical implications of genetic
differences that can be detected between BCG strains, treatment for
adverse events following immunization, and whether some BCG strains
are antibiotic resistant. Another question is how to test BCG vaccines used
as part of a novel TB vaccine prime–boost strategy. Requirements for the
production and quality control of BCG vaccines were established by the

34
Committee (WHO Technical Report Series, No. 745, 1987) and a revision
process has been initiated by the Secretariat through a Working Group. An
interim report was provided to the Committee for feedback and guidance.
The proposed revision of the Recommendations would address a new
method of establishing identity; relevant parameters to assess consistency
of production; improved specifications for potency (currently not defined
in terms of protective efficacy in an animal model); and tests for cultivable
bacteria.
The Working Group also proposed that a repository of currently used strains
of BCG vaccine should be established for the purpose of defining their
genetic characteristics and evaluating the relevance of genetic testing for
monitoring consistency of production. Further the repository may be used
for the evaluation of new assays for quality control of BCG vaccines (e.g.
rapid tests for viability as opposed to assaying colony forming units and in
vivo protection tests). The Committee was informed that such a repository
of BCG strains has been established at NIBSC.
The Committee was also reminded that it had considered a replacement
for the 1st International Reference Preparation, established in 1965 on
several occasions. The Working Group which met in June 2005, reviewed
the intended purpose of such a replacement. They agreed that a replacement
standard should be established primarily for calibration of secondary
(internal or working) standards which in turn should be used for controlling
the variability of viable count assays (used for estimating the numbers of
live bacteria in a preparation). Other potential applications include serving
as a reference for residual virulence in local reactogenicity assays, as a
reference for protection assays for new BCG vaccines and as a calibrant
for new tests for quality control of BCG vaccines. The Working Group
noted that it was not clear if separate standards for the three sub-strains
that now account for most BCG production would be required. Additional
information was needed that could be obtained from a collaborative study
on rapid methods for assaying viability.
The Committee agreed that scientific studies on the genetic analysis of BCG
strains and standardization of potency tests should be pursued, but advised
that a revision of the current Recommendations should be timed to take
advantage of new data likely to be generated from the work in progress.

International reference materials

Proposals for discontinuation of reference preparations
The Committee considered proposals that the following preparations should
be discontinued:

35
 Newcastle (Hitchner B1 strain) disease vaccine, live; 1st International
Reference Preparation, 1967 coded NV. A recent exercise to reconcile
the WHO Catalogue with the NIBSC catalogue of international reference
materials had revealed that this preparation was not transferred from the
Central Veterinary Laboratories (CVL), (now the Veterinary Laboratory
Agencies (VLA)), Weybridge, England to NIBSC during the exercise to
transfer the custodianship of International Reference materials previously
held by CVL. This was possibly because of safety considerations. The
representative from the European Directorate for the Quality of Medicines
(EDQM) offered to host this material if it can be determined that some
of the standard still exists because some live attenuated vaccines are still
manufactured and there may be a need for this material in the future. The
Secretariat was therefore requested to make further efforts to find out
whether the residual stock of this material is still at VLA, Weybridge and if
so to transfer it to EDQM.
Scarlet fever streptococcus antitoxin, equine; 1st International Standard,
1952 coded SC. This standard was proposed for disestablishment as less
than 50 ampoules are now available. Nevertheless, the Committee agreed
with a suggestion that this material may of use in the future in comparisons
of, or development of, assays. It was therefore proposed that the Secretariat
publish the intention to discontinue this standard on their web site and if
no objection is received, the proposal to discontinue this material would
be reconsidered by the Committee in 2006. If there is a need to retain this
material, EDQM would be prepared to hold it as a joint standard.

Antigens and related substances

Yellow fever vaccine—minimum specification in International
Units per 0.5-ml dose
The First International Standard for yellow fever vaccine, NIBSC Code
99/616, was established in November 2003 with an assigned potency of
104.5 International Units (IU) per ampoule. The WHO requirements for
potency of yellow fever vaccine require that “The titre of the vaccine shall
not be less than 1000 mouse median lethal dose (LD50) or its equivalent
in plaque-forming units (PFU), in the dose recommended by the
manufacturer for use in humans”. However, most laboratories established
the relationship between LD50 and PFU potency tests many years ago and
the relationship is not necessarily valid today. It is therefore desirable to
update the minimum specification so that a minimum dose is expressed
in IU.
Collaborative study data indicated that 103 LD50/0.5-ml dose may be
equivalent to 104 IU/0.5 ml dose. However, this was found to be inappropriate

36
when data from assays of routine production batches were reviewed,
particularly after the stability test. An alternative approach currently being
investigated is to base the minimum potency on the minimum PFU or IU
which results in seroconversion in clinical trials. Relevant reports identified
so far indicate that a vaccine virus dose of 100 LD50 from one manufacturer
resulted in 100% seroconversion, but 32 LD50 gave neutralizing antibody
in only 19 out of 23 (87%) subjects. From the LD50/PFU/IU conversion
factor of this manufacturer this suggests that 76 IU would result in 100%
seroconversion. Data from a second manufacturer indicated that 1000 PFU
(or 250 IU) resulted in 100% seroconversion
As it appears unlikely that there will be sufficient data from recent clinical
trials to support the establishment of a minimum dose in IU/0.5 ml, the
Committee endorsed a proposal that WHO convene a meeting of yellow
fever vaccine manufacturers to discuss this issue further.

Smallpox vaccine—stability studies

The Committee recalled that at its previous meeting a report of a collaborative
study to assess the suitability of two candidate smallpox vaccines for the
establishment of a replacement International Reference Preparation had
been presented. The study also compared chorio-allontoic membrane
(CAM) and tissue culture assays. The two candidate replacement standards
had been prepared in different ways, one was a cell-culture-grown New
York City Board of Health strain and the second was prepared from the
Lister strain grown on calf flank. The other study samples represented a
range of vaccine strains produced by a variety of methods and included the
existing International Reference Preparation (IRP).
The Committee was asked to note the results of additional stability studies
(WHO/BS/05.2023) on the candidate replacement preparations undertaken
at the recommendation of the Committee in 2004. These studies included
regular assessments of real-time stability (at –20 ºC) of the candidate
preparations and additional accelerated thermostability data on the two
candidates generated in tests done concurrently with the current IRP at
+4 ºC, +20 °C and other elevated temperatures. The results of these studies
indicated that at higher temperatures, 20 ºC, 37 ºC and 60 ºC, there was no
statistically significant difference between the samples, and the Arrhenius
plot was uninformative. There appears to be little difference in the stability
of the candidates and the current IRP
The Committee was informed that the fill data on the two candidate materials
which were both commercial vaccines were not yet available, but that
demand was currently low and a delay in recommending the establishment
of a replacement would not compromise availability of a reference
preparation. The Committee therefore agreed that the final report including

37
 the fill data and the recommendation on the replacement for the current IRP
be submitted for the 2006 meeting of the Committee. The Committee also
noted that there may be concerns regarding the import and use of the calf
derived vaccine in some countries

Poliovirus, Sabin, type 3—Neurovirulence test reference

Neurovirulence testing is an important part of the safety testing of oral polio
vaccines. The current test in primates is designed so that each monovalent
bulk intended for use in OPV production is tested for neurovirulence against
an appropriate reference preparation. In addition to being used in testing in
primates the reference materials are also used in the transgenic mouse test.
Only those bulks that pass one or other of these tests are used for vaccine
production. The reference material is also included in the validation and
implementation phases of the transgenic mouse test, which laboratories
need to complete before using the method for quality control purposes.
The above activities lead to a continuous requirement for the reference
preparations.
The current WHO reference materials are +2 passages from the Sabin
original seed viruses and were produced by Behringwerke, Germany, for
WHO in the mid-1970s. Although there are still reasonable quantities of
the WHO(SO+2)/I and WHO(SO+2)/II (the references for Types 1 and 2,
respectively), the stocks of the reference for Type 3 (WHO(SO+2)/III) are
very low. The Committee was informed that the properties of this reference
are crucial and it is important to establish a replacement material which has
similar properties to the previous reference. This is because if a replacement
is too virulent, it’s use could result in batches passing that previously would
have failed or, if the replacement is too attenuated, it may result in too many
batches failing that previously would have passed and this could result in
supply problems.
Four candidate replacements for the current WHO neurovirulence reference
preparation (WHO(SO+2)/III) were produced under normal manufacturing
conditions by four commercial suppliers of oral polio vaccine. Two studies
have been undertaken to evaluate the properties of these materials (WHO/
BS/05.2020). In the first study, four vaccines were evaluated by mutation
analysis by PCR and restriction enzyme cleavage (MAPREC) analysis
and in transgenic mouse tests using both a local TgMPVR21/Bx strain
of mice and also in the WHO-approved TgMPVR21 mice. Two vaccines,
vaccine #1 and vaccine #2, were taken forward into the second phase of a
collaborative study and examined with MAPREC tests, transgenic mouse
tests and monkey neurovirulence tests. As a result of this study, vaccine #1
is the most likely replacement. Consequently, 3 litres of the preparation will
be distributed in ampoules, titred and the stability determined in a small

38
collaborative study and its behaviour relative to its intended use will be
explored.
The Committee agreed that, once established, the replacement standard
should be routinely used, the old one should be discontinued and
remaining stocks should not be used. At the time that the replacement
is established, all laboratories undertaking testing will have to establish
revised C values. In addition, the Committee noted that the change in
reference material will have implications for licences and changes
to licences, which will have to be approved by the national regulatory
authority, will be required.
The Committee noted the plans for the replacement of this reference material
and requested that a full report should be submitted in 2006.

Haemophilus influenzae type b capsular polysaccharide—

1st International Standard
A wide variety of physicochemical methods are used for the quantification of
the content of the capsular polysaccharide PRP (polyribosyl ribitol phosphate;
5-d-ribitol-(1→1)-β-d-ribose-3-phosphate) in purified polysaccharides, bulk
conjugates and final lots of Haemophilus influenza type b (Hib) vaccines.
These include phosphorus assay, ribose determination, high-performance
anion exchange chromatography-pulsed amperometric detection (HPAEC-
PAD) and immunochemical assays. Biological testing is carried out only to
ensure safety.
A meeting to discuss WHO recommendations for the production and quality
control of Hib vaccines in 1996 and a WHO/NIBSC workshop on the use
of physicochemical methods for the characterization of Hib conjugate
vaccines in 1998 recommended that a PRP reference preparation should be
made available under the auspices of WHO. The need identified was for a
reference preparation containing a known quantity of polysaccharide that
can be used to cross-calibrate various methods to quantify the PRP content
of the bulk saccharide, bulk conjugate and final formulations to facilitate
the calibration of assays.
The candidate standard, NIBSC code 02/208, was therefore prepared. This
is a freeze dried preparation of PRP unconjugated polysaccharide diluted to
5 mg/ml using sterile distilled water (DW). Ten laboratories from eight
different countries participated in a collaborative study to assess its suitability;
to determine its PRP content in SI units (per weight (w) basis), and to evaluate
its suitability for use as a standard for PRP quantification assays (including
ribose, phosphorus and HPAEC-PAD assays) of final fills and bulks of
Hib vaccines (WHO/BS/05.2018). The mean PRP content as determined
by the ribose assays carried out by seven of the participating laboratories

39
 was 4.933 ± 0.267 mg/ampoule (expanded uncertainty calculated using a
coverage factor of 2.45 which corresponds approximately to a 95% level of
confidence).
On the basis of the results from this study, the Committee recommended that the
preparation coded 02/208 should be established as the 1st International Standard
for Haemophilus influenzae type b capsular polysaccharide polyribosyl ribitol
phosphate (PRP) with an assigned unitage of 4.933 ± 0.267 mg/ampoule.
The Committee requested that additional information be provided in the
instructions for use concerning the use of the standard for calibration of
different assays. They also recommended follow-up studies, especially to
assess the value of the International Standard for assays on final fills.

Antisera
Dengue virus antibody, human serum, 1st Reference Reagent
The WHO Steering Committee on Flaviviruses identified a need for reference
materials for antibodies to dengue virus as it was envisaged that the use of
standards calibrated using a common unitage would facilitate the comparison
of data between laboratories and of the responses to different vaccines. The
availability of such materials, along with Vero cells and four serotypes of
dengue virus for use in dengue neutralization tests, would facilitate the
standardization of antibody assays for the detection of specific antibodies
to dengue virus. These materials should be of use in the assessment of
antibodies induced in vaccine trials as well as in naturally infected individuals
in diagnostic laboratories.
A collaborative study divided into two phases were undertaken to assess
the suitability of a freeze-dried polyvalent plasma (NIBSC code 02/186) to
serve as an international reference preparation for dengue virus antibodies
and to assess the variation in neutralization assays for antibodies to dengue
virus (BS/WHO/05.2009).
Thirteen laboratories in six countries took part in the studies. In the first phase
participants were sent coded samples which included a duplicate sample of
the candidate standard and a negative control. They also received Vero cells
and four dengue virus serotypes. In the second phase, participants assayed
uncoded samples using the method, cells and viruses in routine use in their
laboratory. Data from 32 assays on each of the four dengue virus serotypes
were analysed. The 50% plaque reduction neutralization titre (PRNT50) for
each sample was calculated when assayed against each virus and potencies
relative to the candidate standard were calculated.
The polyvalent anti-dengue virus 1+2+3+4 serum (NIBSC code 02/186) had
measurable titres against all four dengue virus serotypes tested. However,

40
the expression of neutralizing antibody titre of the monovalent anti-dengue
sera, and a coded duplicate of 02/186, relative to the candidate standard
only marginally improved the percentage geometric coefficient of variation,
a measure of the variation between laboratories and assays, compared to
those obtained with the expression of antibody content as a 50% plaque
reduction neutralization titre.
The Committee was informed that the number of available ampoules of the
polyvalent anti-dengue 1+2+3+4 preparation 02/186 is smaller than often
is the case for an International Standard, as a limited amount of serum was
available for processing at the time this project was initiated. Nevertheless,
the Committee agreed that the preparation will be of use in facilitating the
standardization of neutralization assays for antibodies to dengue virus. The
Committee thus endorsed the establishment of the polyvalent anti-dengue
1+2+3+4 preparation, 02/186, as a WHO reference reagent with a unitage
of 100 units for each virus so that laboratories assaying sera from recipients
of vaccines currently in development can compare results. The Committee
also agreed that the monovalent anti-dengue antisera preparations be issued
as a reference panel but with no unitage assigned, for use in assay validation
studies.
They also noted that efforts are currently under way to identify alternative
sources of plasma containing antibodies to dengue serotypes 1+2+3+4.

Japanese encephalitis virus, human serum

At present there are no internationally available reference materials for
antibodies to JE virus. The WHO Steering Committee on Flaviviruses
identified the need for suitable reference materials as it was envisaged that
the use of standards calibrated in International Units would facilitate the
comparison of data between laboratories in the assessment of antibodies
induced in vaccine trials as well as in naturally infected individuals in
diagnostic laboratories.
A candidate standard designated NIBSC code 02/182 was prepared from
the serum from six recipients of an internationally-available, licensed
inactivated vaccine (Nakayama-NIH strain). These individuals were bled
before administration of the first dose of vaccine and a control serum for
use in neutralization tests is therefore also available (NIBSC code 02/184).
A collaborative study was undertaken to assess the suitability of the
materials to serve as international reference materials (WHO/BS/05.2008).
Vero cells and three strains of JE virus were supplied to participants for
use in this study. Five participants in five countries submitted data from
assays on the candidate reference material and seven coded samples. The
coded samples included the negative control serum and a duplicate of the
candidate standard. Data from 11 assays were analysed. The expression of

41
 neutralizing antibody titre relative to the candidate standard only marginally
improved the percentage geometric coefficient of variation, a measure of
the variation between laboratories and assays.
Although data are available from only a few assays, the results of the study
indicate that the PRNT50 results obtained for serum from recipients of killed
vaccines, including the candidate standard, vary depending on the virus
used in the neutralization tests. Thus higher PRNT50 results are obtained
when the challenge virus is homologous to the vaccine strain than when
a heterologous virus is used. Potencies expressed relative to the candidate
standard are therefore affected by the strain of virus used in assays. The titres
obtained with the candidate material and Nakayama virus as challenge virus
in the neutralizing antibody assay differ considerably from those obtained
when Beijing or SA14-14-2 are used as challenge virus.
The Committee agreed that the candidate standard appears to be a strain-
specific serum and is therefore unsuitable to serve as the International
Standard for antibodies to JE virus. They nevertheless endorsed the view
that it may be helpful for laboratories validating assays and should be
made available by NIBSC for this purpose. They also noted that sera from
naturally immune donors or recipients of the SA14-14-2 live vaccine may
be suitable to serve as a reference material for antibodies to JE virus and
that the availability of such materials is being investigated.

Anti-human platelet antigen-1a

Alloantibodies against human platelet antigens (HPA) are involved in
neonatal alloimmune thrombocytopaenia (NAIT), platelet refractoriness
(PR) and post-transfusion purpura (PTP). Detection of the relevant HPA
antibody is essential to the diagnosis and treatment of patients and a variety of
methods have been used to detect alloantibodies against platelets. However,
there is considerable variation between laboratories in the sensitivity of
antibody detection tests and there is a clear need for standardization in HPA
antibody detection. To date, two reference materials have been prepared by
the NIBSC (anti-HPA-1a; NIBSC code 93/710 and anti-HPA-5b; NIBSC
code 99/666) and both have been established as International Reference
Reagents by WHO. Both are minimum potency preparations containing low
titre antibodies and they are used to determine the sensitivity of tests for
the respective antibodies. However, these preparations are not useful for
the quantitation of antibodies, where higher titre antibodies are required
for constructing a standard curve over a dilution range which reflects the
variable reaction strengths found in patient samples.
Quantitation of anti-D antibodies has become routine in the monitoring of
pregnancies with RhD immunization and for measuring the quantity of anti-D
in anti-D preparations for prophylaxis, and international standards allow the

42
comparison of results between laboratories. For the former this has led to better
treatment with the reliable identification of severe fetal anaemia requiring
antenatal treatment with intrauterine transfusion of donor red cells. Antenatal
treatment for NAIT using intra-uterine platelet transfusion is available but as
it carries considerable risk it is desirable to restrict its use to the most severe
cases. Currently there is no reliable predictor of severity, but the introduction
of a standard for quantitation of HPA-1a antibodies will allow laboratories
to measure anti-HPA-1a uniformly. This will facilitate multicentre trials and
allow the relevance of anti-HPA-1a quantitation to be determined. Therefore,
plasma samples containing potent anti-HPA-1a were pooled and freeze
dried. In addition, three individual plasma samples were selected which had
varying levels of anti-HPA-1a activity which was determined by a variety of
quantitative assays using the proposed standard as a reference.
An international collaborative study which was conducted in 39 laboratories
in 24 countries and showed that the anti-HPA-1a activity in three test samples
could be reliably determined using the proposed standard. The assays used to
determine anti-HPA-1a activity were monoclonal antibody immobilization
of platelet antigens (MAIPA) assay (30 participants); commercial or in-house
monoclonal antibody capture enzyme-linked immunosorbent assay (MACE
ELISA) (5 participants); platelet immunofluorescence test read by flow
cytometry (platelet immunofluorescence test (PIFT flow)), (3 participants).
The suitability of the candidate standard 03/152 was confirmed by the
International Society for Blood Transfusion Platelet Immunology Working
Party and by the International Society of Haemostasis and Thrombosis/
Scientific Standardisation Committee (ISTH/SSC) Subcommittee on Platelet
Immunology.
Based on the results of the collaborative study the Committee recommended
that the material, coded 03/152, should be established as the 1st International
Standard for quantitation of anti-HPA-1a, with an assigned unitage of 100
International Units, and that its intended use is in quantitative assays to
determine the anti-HPA-1a activity in clinical samples.
The Committee also recommended that this material does not replace the
earlier International Reference Reagent (93/710) which is not discontinued.

Blood products and related substances

World Health Organization/International Society of
Haemostasis and Thrombosis Liaison Committee report
The activities of a new WHO/ISTH (International Society of Haemostasis
and Thrombosis) liaison committee were presented. The aims of this
committee are to promote the timely flow of biological standards related

43
 to haemostasis and thrombosis through a sound peer review process in
and between organizations through to final establishment by the Expert
Committee on Biological Standardization; to advise the Expert Committee
of priorities of various standards to be developed, replaced or discontinued
as well as on new methodologies, taking into account the needs in the
various WHO Regions; to hold regular discussions about information on the
development and assessment of such standards; to discuss and recommend
as appropriate to the ISTH/SSC and WHO/Expert Committee on Biological
Standardization issues related to the preparation, assessment and publication
of biological standards in the field of haemostasis and thrombosis.
The Expert Committee received a report on the first meeting of the liaison
committee held on 10th August 2005 in Sydney, Australia. The following
priorities were identified at the meeting: revision of WHO guidelines for
thromboplastin and development and establishment of WHO reference
preparations standards (e.g. a number of global measurement standards
for in vitro diagnosis; platelet antibodies; and particularly replacement
of the current bovine thromboplastin reference material). The liaison
committee also commented on priorities proposed by NIBSC and the other
collaborating centres for new and replacement reference preparations;
provided expert review of standards for submission to the 2005 meeting
of the Expert Committee (specifically the proposed standards for factor
V plasma, factor XI plasma, prothrombin G20210A mutation, and anti-
human platelet antigen 1a); and discussed ISTH/SSC publication of reports
and manuscripts.
The Committee heard that it was anticipated that the liaison committee
would facilitate communications between ISTH/SSC and the WHO/Expert
Committee on Biological Standardization, respond to requests from WHO
Regions for standards and guidelines, contribute to a continuity of approach
in standards setting, guarantee and maintain records of expert peer review
of all submissions to the Expert Committee, and provide a point of contact
for all WHO–ISTH collaborations.
The Committee agreed with the proposed priorities for reference
materials, that the continued need for the WHO bovine thromboplastin
reference preparation should be reviewed and that the WHO guidelines on
thromboplastins should be revised.

Anti-A and Anti-B blood grouping reagents

The quality of blood grouping reagents is clearly an important factor for
safe blood transfusion, yet there is currently no appropriate international
standardization of monoclonal anti-A or anti-B blood grouping reagents.
Although WHO has International Standards for anti-A and anti-B, they
were prepared many years ago and represent grouping reagents available

44
at that time i.e. sera from immunized donors, whereas current anti-A and
anti-B grouping reagents consist of monoclonal antibody preparations.
Consequently there is currently little or no demand for the existing WHO
standards. WHO thus initiated development of new candidate reagents.
The Committee also recalled that it had considered preliminary reports
of collaborative studies to characterize two candidate standards at its
55th meeting (WHO Technical Report Series, No. 932). At that time the
Committee noted that agreement on the value assignment had to be reached
before the reagents could be endorsed as WHO standard reagents and the
study was referred to a working group for this purpose. Advice on impact
of the methodology used in the study would also be requested from the
working group.
The results of a collaborative study, WHO/BS/05.2024, performed in
2005 in response to the recommendations from the Committee at its 55th
meeting were presented. In this study, scientific advisors from four WHO
Collaborating Centres had performed parallel titrations of the new candidate
standards for anti-A and anti-B against existing reference preparations (the
2nd WHO International Standard for anti-A (code number W1001) and
the 3rd International Standard for anti-B (code W1002); CBER US/FDA
preparations; and British Minimum Potency Reference Preparations) to
help determine minimum specifications to recommend. The specifications
agreed by the scientific advisers were then communicated to the original
study participants for comment. As a result of the new study the participants
agreed on minimum potency specifications that were intended to ensure
safety of testing, but agreed not to define an optimum specification for
sensitivity of anti-A or anti-B blood grouping tests.
The Committee, taking note of the outcome of these further studies,
recommended that the preparation coded 03/188 should be established
as the 1st International Standard for Minimum Potency of Anti-A Blood
Grouping Reagents and that a 1 in 8 dilution of 03/188 should define the
recommended minimum potency specification for anti-A blood grouping
reagents. The Committee recommended that a decision on the status of the
2nd International Standard for anti-A blood typing serum, code W1001,
should be deferred until an evaluation is made of the demand and its current
applicability. The Committee also recommended that the preparation coded
03/164 should be established as the 1st International Standard for Minimum
Potency of Anti-B Blood Grouping Reagents and that a 1 in 4 dilution of
03/164 should define the recommended minimum potency specification for
anti-B blood grouping reagents. The Committee likewise recommended
that a decision on the status of the 3rd International Standard for anti-B
blood typing serum, code W1002, should be deferred until an evaluation is
made of the demand and its current applicability.

45
 Prothrombin mutation
A non-coding mutation in the 3’ untranslated region (UTR) of the prothrombin
gene G20210A (also known as 20210G>A variant of the prothrombin gene)
is associated with an increased risk of arterial and venous thrombosis.
Carriers of the mutation have higher plasma prothrombin concentrations
than non-carriers. The mutation confers a 2- to 4-fold increased risk for
venous thrombosis in the heterozygous form and a 10-fold risk in the
homozygous form. Consequently, the determination of the G20210A
genotype is recommended as a high priority test in the investigation of genetic
thrombophilia and it has become one of the most frequent genotyping tests
performed in clinical laboratories. The frequency of testing is also likely
to increase as research continues on the association of pre-existing risk
factors, such as a previous history of thrombosis, use of oral contraceptives
or hormone replacement therapy and genetic susceptibility with venous
thrombosis and air travel.
External quality assurance schemes have shown that errors in genotyping
for the prothrombin mutation do occur. In 2003, results for tests of the
molecular genetics of thrombophilia showed that 4.5% of laboratories
had made an error when testing a sample which was heterozygous for the
prothrombin mutation G20210A and that 3.1% of laboratories made an
error when testing a sample containing the wild-type prothrombin gene
sequence. These errors can have a significant and long-lasting impact on
the patient, particularly because genotyping tests are usually carried out
only once on any one patient. This is in contrast to many other pathology
tests which are often carried out on several occasions, allowing errors to be
identified.
Forty-five laboratories participated in an international collaborative study
to assess the suitability of a panel of three gDNA samples as the 1st
International Genetic Reference Panel for Prothrombin Mutation G20210A,
Human gDNA. The participants employing a total of 22 methods evaluated
the panel against their in-house controls which were samples from known
patient and commercial controls.
The results have shown that the panel of three gDNA samples evaluated in
this international multicentre study are suitable as reference materials for
laboratories carrying out genotyping for prothrombin mutation G20210A.
Therefore the Committee endorsed the three human gDNA preparations
04/194 (Prothrombin wild type), 04/174 (Prothrombin Mutation 20210
Heterozygote), 04/196 (Prothrombin Mutation 20210 Homozygote) as
the 1st International Genetic Reference Panel for Prothrombin Mutation
G20210A, Human gDNA. A panel of materials, comprising one of each of
the above preparations, will be made available from NIBSC with the code
number 05/130.

46
Vitamin B12 and folate in human serum
The assay of the vitamins B12 and folate in blood is the current routine
procedure for determining deficiency of these vitamins. Deficiency can
result in a number of clinical conditions including megaloblastic and
pernicious anaemia.
The 1st International Reference Reagent for human serum vitamin B12
(81/563) was established by WHO in 1992. However, the preparation has
since been found to be positive for anti-hepatitis C virus (HCV) and HCV
RNA. Also, it was calibrated using a microbiological assay (Euglena assay)
which has since been replaced by automated assay systems. For these
reasons, it was decided to replace the 1st International Reference Reagents
and assign a value to the replacement preparation, independently of the
value assigned to the previous standard, using current methodology.
The aim of the study (WHO/BS/05.2025) was therefore to evaluate a batch
of lyophilized serum for folate and B12 content using current methods,
including candidate reference methods based on mass spectrometry for the
specific determination of 5-methyltetrahydrofolate (5MeTHF) and other
folate forms in serum or plasma. The candidate International Standard for
B12 and serum folate, 03/178, was assayed using a wide range of methods
in 24 laboratories in seven countries.
The Committee recommended that the preparation, coded 03/178, should be
established as the 1st International Standard for serum folate, with a unitage
of 12.1 nmol/l (5.33 ng/ml) of folate and the 2nd International Standard for
vitamin B12 with an assigned unitage of 480 pg/ml, when the freeze-dried
contents of the ampoule are reconstituted in 1 ml of distilled water. The
Committee also recommended that a corrected report be submitted to WHO
showing the vitamin B12 value in pg/ml, not ng/ml as stated in BS.05/2025.
Finally the Committee recommended that the assigned vitamin B12 value
of 480 pg/ml should be re-evaluated when a reference method has been
established.

Blood coagulation factor V (plasma) human

Estimation of factor V clotting activity (FV:C) in human blood plasma is
performed for several reasons. These include the diagnosis of very rare
congenital bleeding disorders such as FV deficiency (“parahaemophilia”)
and combined FV/FVIII deficiency, as well as investigations into coagulation
disorders linked to liver disease. The need to review the management of
rare bleeding disorders, including improved standardization of diagnostic
coagulation tests, has been highlighted. Factor V:C estimation is also
a requirement for the quality control of the therapeutic product, virus-
inactivated fresh frozen human plasma, as described in the European

47
 Pharmacopoeia. Despite the existence of several commercial reference
plasmas with assigned values there is currently no internationally accepted
unit (IU).
An international collaborative study involving 22 laboratories in 11 different
countries was undertaken to calibrate a proposed 1st International Standard for
Factor V clotting activity (WHO/BS/05.2007). Most estimates (21/23) were
obtained using thromboplastin-based methods rather than activated partial-
thromboplastin time (APTT)-based methods. There was very good agreement
between estimates with inter-laboratory variability of 3.55% and a range of
estimates from 110–124% of the proposed assigned value of the proposed
standard. There was also good agreement between results from APTT- and
thromboplastin-based methods indicating that the proposed standard will be
suitable for use with both methodologies. Factor V is recognized as a labile
coagulation factor and can suffer loss of activity on storage in the liquid state
or through cycles of freezing and thawing. Accelerated degradation studies
in three laboratories indicated that the proposed standard is extremely stable
and suitable for prolonged use.
On the basis of the results of the study the Committee endorsed the WHO
1st International Standard Factor V Plasma (03/116) with an assigned value
of 0.74 IU Factor V:C per ampoule.

Blood coagulation factor XI (plasma) human

Factor XI deficiency is inherited as an autosomal recessive bleeding disorder
known as haemophilia C. Factor XI deficiency is most frequently observed in
Ashkenazi Jews and Iraqi Jews. Patients with factor XI deficiency have also
been reported in African-American, Arab, Chinese, English, German, Indian,
Iranian, Italian, Japanese, Korean, Portuguese and Swedish populations.
There are two plasma derived concentrates available for the replacement
therapy and these concentrates are calibrated against different plasma pools.
Currently, no WHO International Standard is available for calibration of
FXI activity and laboratories employ locally collected normal plasma pools
or commercially available frozen or freeze-dried plasma as references. A
common standard is required to improve inter-laboratory agreement for
measurement of FXI activity.
Twenty-seven laboratories from 11 countries took part in a collaborative study
to assign a potency value to a proposed WHO 1st International Standard for
Blood Coagulation Factor XI, Plasma, Human (04/102) and also to calibrate
the Scientific Standardization Committee (SSC) secondary plasma standard
Lot#3 for factor XI activity (WHO/BS/05.2017). The main aim of this study
was to assign an internationally agreed potency value in IU to a batch of
pooled normal plasma relative to locally collected normal plasma pools.

48
On the basis of the results, the Committee endorsed Plasma, Human (04/102)
as the WHO 1st International Standard for FXI with an assigned value
estimated against pools of both fresh and frozen normal plasma of 0.86 IU/
ampoule. This standard is intended for use in the in vitro diagnostics area
and will be available for the calibration of commercial reference plasmas
and of secondary and in-house standards.

Thromboplastin, rabbit, plain

Thromboplastins used in the prothrombin time (PT) test for the laboratory
control of oral anticoagulant treatment must be calibrated against International
Standards to determine the International Sensitivity Index necessary to convert
PT results into an International Normalized Ratio (INR). Observations from
users show that the calibration of a given thromboplastin is generally more
precise when it is performed against an International Standard of similar
composition, and from the same species, and is one of the reasons that WHO
maintains thromboplastin International Standards from different species.
Since stocks of the existing International Standard coded RBT/90 (rabbit,
plain) are exhausted, a preparation of rabbit brain thromboplastin coded 04/162
was calibrated in an international collaborative study organized and carried
out under the auspices of the Subcommittee on Control of Anticoagulation
of the Scientific and Standardization Committee (SSC) of the International
Society on Thrombosis and Haemostasis (ISTH). Since RBT/90 was no
longer available, it could not be used for the calibration of the replacement
candidate. Therefore the study (WHO/BS/05.2029) included five secondary
reference materials for rabbit thromboplastins previously calibrated against
RBT/90 instead. Twenty-two laboratories in 13 countries participated in the
study. Most used the manual (tilt tube) technique for PT testing.
Based on the results of the study the Committee endorsed preparation
04/162 as the 3rd WHO International Standard for Thromboplastins, rabbit,
(plain) with an assigned International Sensitivity Index value, calibrated
against the five existing International Standards, of 1.15.

Cytokines, growth factors and endocrinological

substances
Vascular endothelial growth factor, VEGF165- 1st Reference
Reagent
Vascular endothelial growth factor (VEGF, also known as VEGF-A) belongs
to a gene family that includes VEGF-B, VEGF-C, VEGF-D and placental
growth factor (PlGF). Through its actions on endothelial cells, VEGF induces
vasculogenesis (differentiation of endothelial cell precursors), angiogenesis
(growth of blood vessels), survival of immature blood vessels (vascular

49
 maintenance) and lymphangiogenesis (growth of lymphatic vessels). It
is a potent vascular permeability factor in vivo and is a chemo-attractant
for monocytes and endothelial cells. VEGFs are under investigation, by
administration of the protein or as a gene therapy product, for stimulation
of angiogenesis in ischaemic disease and lymphangiogenesis in treatment
of lymphoedema. VEGF expression is elevated in some types of cancer,
and measurement of VEGF levels for diagnosis and prognosis is under
investigation. Inhibition of VEGF activity has wide potential therapeutic
application in the treatment of various cancers and other conditions involving
pathological development of blood vessels.
The availability of a reference standard for VEGF could facilitate measurement
of the potency and stability of therapeutic preparations of VEGF and
inhibitors, and measurement of VEGF levels for diagnosis.
Preparations of human sequence recombinant vascular endothelial growth
factor-165 (VEGF165) synthesized in Escherichia coli were formulated and
lyophilized at NIBSC. The first preparation, 01/424, was non-homogeneous in
appearance, and following evaluation at NIBSC, was distributed as a NIBSC
research reagent. A second preparation, 02/286, was lyophilized in a different
formulation and evaluated in a collaborative study for its suitability to serve
as a reference standard, and compared with preparation 01/424, by five
laboratories using in vitro bioassays or immunoassays (WHO/BS/05.2028).
The candidate standard was assessed for stability using thermally accelerated
degradation samples. None of these samples showed any evidence of loss
of activity at any of the elevated temperatures, including the relatively high
temperatures of 37 °C and 45 °C.
The Committee endorsed the proposal that the preparation 02/286 be
established as the WHO reference reagent for VEGF165 with an assigned
unitage of 13 000 units per ampoule. To assist laboratories currently using
mass units for VEGF165, preparation 02/286 could be considered to contain
approximately 13 µg VEGF165 per ampoule, based on the predicted content
from the manufacturer’s stated concentration and the results of this study,
and would be noted in the memorandum with instructions for use. The
Committee was also informed that preparation 01/424 will continue to be
made available as an NIBSC research reagent for at least 2 years after the
establishment of 02/286 as a WHO reference reagent, to provide continuity
for laboratories currently using this preparation and to permit them to make
a direct comparison of the two preparations in their own assay systems.

Keratinocyte growth factor—1st Reference Reagent

Keratinocyte growth factor (KGF), also known as fibroblast growth factor-
7 (FGF-7) is a single chain secreted glycoprotein with target specificity

50
restricted to epithelial cells in which it stimulates proliferation, differentiation
and migration. KGF is under investigation for a number of therapeutic
applications involving stimulation of proliferation of epithelial cells.
Measurement of endogenous KGF levels may have applications in diagnosis
and prognosis. A modified form of recombinant KGF synthesized in E. coli,
is licensed for the reduction of incidence and severity of oral mucositis in
cancer patients receiving a high dose of chemotherapy and radiation therapy.
The availability of a reference preparation would facilitate inter-laboratory
comparison of potency and immunoassay measurements of KGF and stability
studies of therapeutic formulations.
Two preparations of recombinant KGF synthesized in E. coli, one consisting
of the normal mature 163 amino acid sequence of human KGF, and the
other of a modified sequence lacking the 23 N-terminal amino acids were
tested by five laboratories using in vitro bioassays or immunoassays (WHO/
BS/05.2027). A preparation which contained no carrier protein was included
in some studies. The different preparations were clearly distinguished by
the different assay systems, demonstrating the need for seperate reference
preparations in terms of which bioactivity can be expressed, and which
could be used to assess assay specificity. Each of the candidate preparations
was judged sufficiently stable to serve as a reference standard, although the
Committee noted that some additional studies on reconstituted material will
be undertaken so that appropriate advice can be included in the information
for users.
The Committee endorsed the proposal that the preparation coded 03/150 be
established as the WHO reference reagent for human KGF, with an assigned
unitage of 4000 units of KGF per ampoule, and that the preparation coded
03/148 be established as the WHO reference reagent for human KGF (24-
163), with an assigned unitage of 9000 units of KGF(24-163) per ampoule.
In addition, it was agreed that in order to assist laboratories currently using
mass units for KGF, 03/150 could be considered to contain approximately 4 µg
KGF per ampoule and 03/148 could be considered to contain approximately
9 µg KGF (24-163) per ampoule based on the content predicted from the
manufacturer’s stated concentration and the results of this study, as would
be noted in the memorandum with instructions for use. The Committee
noted that freezing and thawing of reconstituted solutions can alter their
properties and that freezing and reuse is therefore not recommended.

Measurement of relative potencies of thermal degradation

samples of the WHO International Standard of interferon
alpha 2b (ifn-α2b), 95/566
Several International Standards for different human interferon (IFN) types
have been established which serve as primary calibrators of bioassays for

51
 secondary working reference preparations of the same IFN. An important
aspect of their suitability to serve as primary calibrators is the requirement
for long-term thermal stability of IFN activity at the repository storage
temperature of ampoules. It is often not possible to study thermal stability
in real time as normally an International Standard would be established 2–
4 years after its preparation and evaluation by collaborative study, a period of
time usually too short to allow the evaluation of potential losses of activity in
ampouled material held at the recommended storage temperature of –20 °C.
Thermal stability is therefore assessed in accelerated thermal degradation
studies. Nevertheless, since most IFNs are very stable, even accelerated
degradation studies may require protracted storage of ampoules—longer
than the limited time-frame before establishment of an International
Standard—for the procurement of data that accurately predicts rate of
loss of activity at the recommended storage temperature. For example,
accelerated thermal degradation studies on the current WHO International
Standard of IFN-α were carried out 0.9 to 1.7 years after manufacture
and, because there was no indication of thermally instability at –20 °C,
the materials were considered to be suitable for establishment (WHO/
BS/99.1911).
The continued availability of accelerated thermal degradation samples
after longer storage times provided an opportunity to re-assess whether
the thermal stability of activity confirmed the expected robust stability
of IFN-α at –20 °C. The activity of the 2nd WHO International Standard
for IFN-α2b (NIBSC code 95/566) stored for approximately 9 years has
therefore been examined (WHO/BS/05.2006). Two assay systems were
used—an antiviral assay and a reporter gene assay. The accuracy and
precision of these assays were compared to determine which method
provided the most reliable data for assessing the thermal stability of activity.
Two sets of ampoules of the standard coded 95/566 were retrieved from
storage at temperatures of –150 °C, –70 °C, –20 °C, 4 °C, 20 °C, 37 °C,
45 °C and 56 °C on 12 July 2004, approximately 9 years from the fill-date of
29 May 1995. One set was retained at NIBSC for testing in antiviral assays,
the other set was sent to The National Institute of Chemistry, Ljubljana,
Slovenia, for testing in reporter gene assays.
The results of this study provided strong evidence that the activity of the
WHO International Standard for IFN-α2b contained in ampoules of 95/566,
is extremely stable at the customary storage temperature of –20 °C. They
also confirm the expected long-term thermal stability of this preparation
and its suitability to serve as the International Standard for the duration of
its usage (to near exhaustion of the stock of ampoules), i.e. approximately
25 years. Comparison of the results derived from (a) antiviral assays and
(b) reporter gene assays showed they were very similar for this IFN-α2b
preparation. The lower intra-assay variation of estimates from reporter

52
gene assays suggests they could provide a valuable alternative method for
evaluating the thermal degradation studies of other WHO International
Standards for type I IFNs, e.g. IFN-β.
The Committee noted the contents of the report and also encouraged
publication of this, and other, examples of long-term studies of the stability
of biological reference materials.

Diagnostic reagents
Reference materials for in vitro diagnostic devices
The Committee was provided with an overview of the current activities
of the International Federation of Clinical Chemists (IFCC) related to
standardization in laboratory medicine. It was noted that the Scientific
Division of the IFCC through its Committees and Working Groups is
involved in more than 15 projects on development of reference materials
and/or reference measurement procedures following the criteria defined in
International Standards Organization (ISO) Standards 15193 and 15194.
The IFCC contributes to the development of the concept of reference
measurement laboratories and is involved in developing a process for review
of reference measurement services and criteria of recognition of reference
laboratories in laboratory medicine. Through its Committee on Traceability in
Laboratory Medicine, the IFCC establishes and implements external quality
assessment schemes (EQAS) for reference laboratories for monitoring
competence through comparative measurement campaigns. Further, the
IFCC is a major partner and founding member of the Joint Committee
on Traceability in Laboratory Medicine (JCTLM) which has established
two major Working Groups. The first is involved in evaluating nominated
reference measurement procedures and reference materials by various
national and international organizations, and publishes a list of reference
materials and reference measurement procedures that meet ISO criteria.
These lists are available for consultation through the IFCC web site. The
second Working Group on reference measurement laboratories is currently
working on defining a process for review of reference measurement services
from laboratories following ISO 17025/15195 Standards. Finally a list of
current IFCC standardization activities was provided to the Committee.

HIV-1 RNA nucleic acid amplification test assays

The 1st International Standard for HIV-1 RNA for HIV-1 NAT assays, code
97/656, was established by the Expert Committee in 1999 and has since
been extensively used for assay validation and calibration and to calibrate
secondary standards and working reagents. The 1st International Standard
was prepared from a subtype B (V3) HIV-1 polymerase chain reaction (PCR)-

53
 positive, antibody-negative plasmapheresis donation from the USA that was
diluted in screened defibrinated plasma and freeze-dried. Approximately
2300 vials were prepared, each containing, when reconstituted, a volume
of 1 ml. This material was evaluated in an international collaborative study
together with a different batch of freeze-dried diluted HIV and a liquid HIV
preparation. All three preparations gave results that were tightly grouped,
with little difference between the results from different laboratories or
those obtained by the use of different assays. The Committee designated
the preparation 97/656 as the 1st International Standard and assigned it a
potency of 5.0 log10 (100 000) IU per vial.
In 2003, it was decided to evaluate a replacement International Standard
as the number of vials remaining had decreased to ~1000 (with a mean
usage of 240 vials per annum). At current usage rates supplies would be
exhausted in 3–4 years and it was also discovered that the current standard
was contaminated with hepatitis B virus (HBV) DNA which, although not
a problem for single marker NAT assays, did present a problem for the
increasing number of multiplex NAT assays being developed and coming
into use. This was discussed at the 16th International Scientific Working
Group on the Standardisation of Genome Amplification Techniques (SoGAT)
in July 2003 where it was agreed that a replacement standard was needed
and that the second freeze-dried candidate from the first collaborative study
should be re-evaluated in a new collaborative study.
The objectives of the present collaborative study (WHO/BS/05.xxxx)
were to establish a replacement WHO International Standard for HIV-1
RNA NAT assays, to determine the real time stability of the current and
candidate preparations over the period since they had been freeze-dried
and to establish the relative potency of the candidate replacement standard
(97/650) against the WHO 1st International Standard (97/656) to ensure
continuity of the IU.
The Committee, taking note of the report, endorsed the establishment of
candidate material 97/650 as the 2nd International Standard for HIV-1 RNA
NAT assays and assigned a unitage of 5.56 log10 IU/vial. This is equivalent
to 5.56 log10 IU/ml when a freeze-dried vial is reconstituted with 1 ml
purified water.

Anti-HIV tests
The human immunodeficiency virus (HIV), a retrovirus that is the cause of
AIDS, exhibits considerable sequence diversity which forms the basis of its
genetic subtyping. A number of groups and subtypes are recognized including
HIV-1 groups M (subtypes A–K), N and O and HIV-2. Despite good progress
having been made on genotyping HIV through sequence diversity, little
progress has been made on the serotyping of HIV. Infected individuals mount

54
a strong humoral immune response, starting about 3 weeks after infection,
and antibodies that react in enzyme immunoassays (EIAs) are produced to
all the main structural proteins. Functional immune responses, such as the
development of virus-neutralizing antibodies, also occur, although they may
take a considerable time to develop and titres are usually low. Attempts to
show a relationship between virus-neutralization serotypes and sequence-
based genotypes have to date been largely unsuccessful.
During 1994, it was discovered that certain screening tests, particularly
those based on synthetic peptides, were unable to detect infection with some
of the more diverse outlier group O viruses. This resulted in the withdrawal
of a number of assays and, over the subsequent years, most assays were
modified to broaden their range and enable them to better detect antibodies
to outlier viruses and to enhance their sensitivity.
In 1998, a meeting of the WHO Working Group on Reference Preparations
for Testing HBsAg, anti-HCV and anti-HIV Diagnostic Kits recommended
that an anti-HIV reference panel consisting of a panel of plasma samples
representing the major groups and subtypes of HIV to reflect HIV diversity
around the world should be established. Such a panel would:
• assist users of serological tests in different regions of the world in
determining that the test they use is able to detect antibodies to the
subtype(s) prevalent in their region;
• help kit manufacturers and others to demonstrate the capability of their
assays to detect antibodies to all the main HIV subtypes and groups; and
• be of value in assessing the analytical sensitivity of assays with regard to
their ability to detect diluted samples.
A pilot study was performed the results of which were used to select samples
for inclusion in a definitive anti-HIV reference panel (code 02/210). The
panel consists of solvent-detergent treated anti-HIV-positive human plasma
samples that have been diluted 1 in 40 in anti-HIV-negative human serum
and freeze-dried. The anti-HIV-positive plasma samples were derived from
individuals infected with HIV-1 subtypes A, B, C and E, HIV-1 group O and
HIV-2 and were selected using the criteria of type, subtype and group, titre,
Western blot profile and availability.
Thirteen laboratories from around the world took part in the collaborative
study. Participants were requested to test the reference panel for anti-HIV in
as wide a range of assays as possible. Where appropriate, it was recommended
that serial dilutions were performed to facilitate the comparison of analytical
sensitivity between assays. For assays such as Western blots, assessment
of their performance undiluted was considered sufficient. The intention
was that a dossier detailing the performance of reference panel samples
in a range of tests would be provided with the panel to enable the recipient

55
 to compare the performance of the panel with kits of their choice in their
own laboratory with the results from elsewhere. Assigning a unitage to the
samples was not considered appropriate.
The Committee was informed that 24 datasets hads been received.
Preliminary results show that the anti-HIV-negative human serum sample
was negative in all assays and that all HIV-positive samples were detected
in all assays, with the exceptions that an anti-HIV-2 EIA failed to detect
some HIV-1 samples and a particle agglutination test failed to detect the
group O sample. Once all data have been included in the analysis and the
results discussed with the participants, a full report will be submitted to the
Committee.
The Committee noted the progress that was being made and requested that
a full report be submitted at its 2006 meeting.

56
© World Health Organization
WHO Technical Report Series No 941, 2007

Annex 1
Guidelines for assuring the quality and nonclinical
safety evaluation of DNA vaccines

This document provides information and guidance to national regulatory

authorities and vaccine manufacturers concerning the characteristics,
production, control and nonclinical development of DNA vaccines. The text is
written in the form of Guidelines instead of Recommendations because further
work is still needed to develop and standardize appropriate methods and criteria
that will assure the consistent quality, safety and stability of these vaccines.
Guidelines allow greater flexibility than Recommendations with respect to
expected future developments in the field and indicate present deficiencies.

1. Introduction
2. Scope of the guidelines
3. Definitions
4. General manufacturing considerations
5. Manufacture and control of bulk purified plasmid (drug substance)
5.1 General information
5.2 Manufacture
5.3 Characterization
5.4 Control of bulk purified plasmid
5.5 Reference standards or materials
5.6 Stability
6. Manufacture and control of final formulated vaccine (drug product)
6.1 Composition
6.2 Manufacture
6.3 Control of materials
6.4 Control of final formulated vaccine
6.5 Reference materials
6.6 Stability
7. Nonclinical safety evaluation
7.1 Introduction
7.2 General considerations on safety
7.3 Considerations on the nonclinical safety programme
7.4 Toxicity studies
7.5 Immunogenicity studies
7.6 Biodistribution, persistence and integration
7.7 Genotoxicity
7.8 Developmental and reproductive toxicity
Authors
Acknowledgements
References

57
1. Introduction
Vaccination involves priming the immune system of a host with an
infectious agent or components of an infectious agent modified in a manner
to ensure that the vaccine does not cause any harm or disease to the host,
but ensures that when the host is confronted with that infectious agent,
its immune system can respond adequately control the invading organism
before it causes any ill effect. For over a hundred years immunizaton has
been achieved by one of two basic approaches:
• introducing into the host, specific antigens against which the immune
system will react directly; or
• introducing attenuated living organisms which replicate within the host
without causing disease and synthesize the appropriate antigens which
subsequently prime the immune system.
Since the early 1990s a radically new approach to vaccination has been
actively and vigorously developed. This involves the direct introduction
of plasmid DNA containing the gene encoding the antigen against which
an immune response is sought into appropriate host tissues and the in situ
production of the target antigen(s). This approach offers a combination
of potential advantages over the more traditional approaches, including
the stimulation of both B and T cell responses, improved stability of the
vaccine, absence of any infectious agents and the relative ease of large scale
manufacture. Many scientific publications address the potential of DNA
vaccination and immune responses in animal models have been obtained
using genes from a variety of infectious agents including influenza virus,
hepatitis B virus, human immunodeficiency virus, rabies virus, lymphocytic
choriomeningitis virus, West Nile virus, malaria and mycoplasma. In many
cases protection from disease in animal models has also been demonstrated
and many aspects of the immune response generated by the injection of
plasmid DNA vaccines have been revealed although much remains to be
understood. The value and advantages of DNA vaccines will be assessed on
a case-by-case basis and their applicability will depend on the nature of the
organism being immunized against, the nature of the antigen and the type
of immune response required for protection.
DNA vaccines progressed rapidly into phase I clinical trials. However, the
immune responses observed in animal models have generally not been
reproduced in humans and many approaches have been and are being
followed to enhance the human immune response. These approaches
function in different ways such as in enhanced uptake, stability of expression,
modulation of the immune response, or in adjuvanting, and include:
• complexing the DNA with polymers (enhances uptake, improves
stability);

58
• encapsulating the DNA on or within microparticles (assists uptake,
presentation and stability);
• optimization of the codon usage of the gene encoding the antigen of
interest (enhances expression);
• encoding a variety of T cell epitopes either instead of or in addition to a
full size protein antigen (modulates the immune response by targeting T
cell stimulation);
• optimizing administration, e.g. by particle-mediated delivery (gene gun)
or electroporation (enhances uptake, modulates immune response);
• route of administration, e.g. mucosal versus parenteral (modulates the
immune response);
• boosting with viral vectors or protein antigen following an initial priming
with plasmid DNA (improves immune response); and
• co-administration of DNA encoding an immune stimulatory molecule
(molecular adjuvant), e.g. a cytokine (improves immune response, modulates
the immune response).
Other approaches may also be under development now or in the future.
The above approaches to enhancing the efficacy of a DNA vaccine may
raise specific safety concerns and these should be addressed in appropriate
nonclinical safety testing.
DNA vaccines are also being developed for veterinary use and efficacy in
animal target species is being observed in some trials. Potentially protective
immune responses are being observed against many infectious agents in
several target species including fish, and companion and farm animals. A
DNA vaccine against West Nile virus for use in horses was first licensed
in the USA in 2005. Although the quality and safety considerations for
vaccines for veterinary use differ from those for human use, experience with
veterinary DNA vaccines can provide valuable information for the control
and use of human DNA vaccines.
These guidelines concentrate on the quality control and on nonclinical
testing of vaccines based on bacterial plasmid DNA intended for use in
humans. The purpose of this document is to provide guidance on:
• appropriate methods for the control of the manufacture and characterization
of plasmid DNA vaccines;
• appropriate approaches for the nonclinical testing of plasmid DNA
vaccines; and
• information specific to plasmid DNA vaccines that should be included in
submissions by manufacturers to national regulatory authorities in support
of applications for the authorization of clinical trials and for marketing.
The development and application of DNA vaccines continues to evolve.
Since these guidelines were first adopted in 1996 (1), many clinical trials of

59
 DNA vaccines have taken place and much experience in their manufacture
and control has accrued. This revision reflects the experience gained,
especially in relation to the data derived from nonclinical safety testing
and the concerns expressed in the first version of these guidelines. The
control of these vaccines should continue to be approached in a flexible
manner to enable further modifications as more experience is gained in
their production and use. The intention is to provide a scientifically sound
basis for the manufacture and control of these vaccines for use in humans
so as to ensure their consistent safety and efficacy. Individual countries may
wish to use this document to develop their own national guidelines for DNA
vaccines.

2. Scope of the guidelines

This document provides guidance on quality and nonclinical aspects of
DNA vaccines intended for use in humans.
The active constituent of a DNA vaccine is a plasmid molecule that contains
the gene for a component of a pathogenic organism under the control of a
mammalian expression system, and possesses DNA sequences necessary
for replication and selection in bacteria. Although a vaccine is generally
defined as a biological medicinal product for the prophylaxis of infectious
disease, “DNA vaccines” are also being developed for therapeutic use, either
against infectious disease or for other diseases such as cancer. In the case of
cancer, the relevant gene often has a human origin (e.g. a cytokine) rather
than a microbiological origin. DNA vaccines against infectious disease
may also contain plasmids expressing genes of human origin which act as
molecular adjuvants. It is clear that the manufacture and quality control of
plasmid DNA for any of the above indications will be essentially identical
and consequently, these guidelines are applicable to DNA vaccines for
therapeutic as well as prophylactic use. The detailed design of relevant
nonclinical safety testing should take into account the proposed use of the
DNA vaccine and the risk–benefit situation.
The guidelines cover DNA vaccines regardless of their method of delivery.
The quality section of these guidelines will be applicable to DNA plasmids
that contain mammalian viral replicons; however, different requirements
may apply to nonclinical testing of such products and the present guidelines
do not address these. Similarly, many aspects of the guidelines may be
applicable to vaccines based on RNA, although again, different requirements
are likely to apply especially for nonclinical safety testing for these types
of vaccine. Plasmid DNA vaccines for use in gene therapy, DNA vaccines
derived in eukaryotic cells, vaccines in which a bacterial cell acts as a carrier
for a plasmid DNA encoding a relevant antigen and nucleic acid vaccines

60
 made entirely by chemical means such as synthetic oligonucleotides are all
outside the scope of these guidelines.
The quality section addresses the control of the bulk purified plasmid (drug
substance) including control of the manufacturing process and the starting
materials, characterization of the purified plasmid and control of the final
formulated vaccine (drug product) including formulation, the control of
materials used in formulation and characterization of the final vaccine.
The nonclinical safety section addresses the approaches to be followed during
testing of the vaccine prior to clinical use. The background information
provided in the quality section should be considered when designing
appropriate nonclinical safety studies.
In general, recommendations in these guidelines are relevant to the product
at the time of application for marketing approval. Some relevant information
is provided with respect to products in development in these guidelines;
otherwise, the respective national regulatory authority should be consulted
prior to clinical development on a case-by-case basis (2, 3).
The control and nonclinical testing of each vaccine should be considered
individually and any special features should be taken into account.
Furthermore, the application of these guidelines to a particular vaccine should
reflect its intended clinical use. Thus, different criteria will apply to a vaccine
that is to be used prophylactically in healthy children universally, than to one
that is to be used therapeutically for treating a life-threatening condition.

3. Definitions
The definitions given below apply to the terms as used in these guidelines
only. They may have different meanings in other contexts.
Bulk purified plasmid (drug substance). The purified plasmid before final
formulation. It is obtained from one or more bulk harvests, and is kept in
one or more containers designated as a single homogeneous production lot
and used in the preparation of the final dosage form (drug product).
Final formulated vaccine (drug product). The finished formulated vaccine
product. It may be freeze-dried and/or contain excipients and/or adjuvants.
Master cell bank (MCB). A homogeneous suspension of bacterial cells,
already transformed by the plasmid containing the desired gene, dispensed
in aliquots into individual containers for storage. All containers are treated
identically during storage and, once removed from it, are not returned to the
cell bank.
Plasmid. A circular, extrachromosomal bacterial DNA element which
undergoes autonomous replication in bacterial cells. It usually carries a

61
 few genes, some of which confer resistance to various antibiotics; such
resistance is often used to discriminate between organisms that contain the
plasmid and those that do not.
Working cell bank (WCB). A homogeneous suspension of bacterial cells
derived from a single vial of the master cell bank dispensed in aliquots into
individual containers for storage. All containers are treated identically and,
once removed from storage, are not returned to the cell bank. Typically, a
single or a defined number of aliquots is used to manufacture a batch of
vaccine. In some cases, a working cell bank may not be established and
vaccine manufacture may begin from an aliquot of the master cell bank.

4. General manufacturing considerations

Plasmid DNA vaccines are considered to be similar to bacterial and viral
vaccines produced by traditional methods, where adequate control of the
starting materials and manufacturing process is just as important as that of
the product. The guidelines therefore place considerable emphasis on “in-
process” controls for assuring the safety and effectiveness of the vaccine as
well as on comprehensive characterization of the vaccine itself.
The general manufacturing requirements contained in good manufacturing
practices (GMP) for pharmaceutical (4) and biological (5) products will
apply to plasmid DNA vaccines. DNA vaccines for use in clinical trials
should also be prepared under GMP conditions. Appropriate attention
therefore needs to be given to the quality of all reagents used in production,
including the components of fermentation media. Many of the general
requirements for the quality control of biological products, such as tests for
potency, endotoxin, stability and sterility, also apply to DNA vaccines.
It is recognized that the level of detail required by a regulatory agency
increases as product development proceeds. During the initial phases of
clinical development, the information contained in a clinical trial application
should be adequate to allow an assessment of the safety risks from the
manufacturing process. This would include, for example, testing of the cell
banks for identity, identification and specifications for all materials used in
the process, assessment of risks from biologically-sourced materials, a brief
description of the process and tests, results of testing of the clinical trial
material and preliminary stability of the drug product.
For late-stage clinical trials the level of detail would increase to include
preliminary evidence of consistency of manufacture and validation.
Changes made to the product composition (e.g. addition of adjuvant or
preservatives) or manufacture (process, site or scale) during the development
of clinical and postapproval manufacturing lots may have a significant

62
 impact on safety and/or efficacy. Any change in the production of a DNA
vaccine places a responsibility on the manufacturer to show that the product
is equivalent to that used in preclinical studies or earlier stage clinical trials.
Such changes should be evaluated on a case-by-case basis to determine what
supporting data should be provided to show comparability of the modified
version with the previous one.

5. Manufacture and control of bulk purified plasmid

(drug substance)
5.1 General information
A brief overview of the development and manufacture of the product
including a justification for the selection of the gene(s) of interest, the
rationale for the development of the product and the proposed route of
administration should be provided.

5.2 Manufacture
5.2.1 Description of manufacturing process and process controls
Information should be provided on the manufacturing process, which
typically starts with a vial of the cell bank, and includes fermentation,
harvest, purification, filling into bulk containers and storage.
A flow chart should be provided illustrating the manufacturing steps from
the cell bank up to the drug substance. The chart should include all steps
(i.e. unit operations), identification of materials, major equipment and in-
process controls.
A description of each process step in the flow chart should be provided
and each step justified. Information should be included on, for example,
scale; culture media, buffers and other additives; major equipment; and
process controls, including in-process tests and operational parameters with
acceptance criteria. Information on procedures used to transfer material
between steps, equipment, areas, and buildings, as appropriate, and shipping
and storage conditions should be provided by the time of application for
marketing authorization.
Data on the consistency of fermentation conditions, culture growth and cell
and plasmid yield should be presented. Criteria for the rejection of culture
lots should be established. The maximum level of cell growth and scale to
be permitted during production should be specified, based on information
on the stability of the host-cell/plasmid system up to and beyond the level
of fermentation used in production by the time of application for marketing
authorization.

63
 The characteristics of bacterial cells/plasmids at the end of fermentation
should be investigated. This investigation should include, as a minimum,
plasmid copy number and restriction enzyme mapping.
For materials such as filter membranes and chromatography resins, information
on conditions of use should be provided. Filter membranes and chromatography
resins should be dedicated to a single product. In the event that a column or
filter is re-used for a single product, conditions of re-use should be provided.
The methods used for harvesting, extraction and purification should be
described in detail and justified. The process should be designed to remove
process- and host-related contaminants, such as endotoxin, host RNA,
chromosomal DNA and any other materials considered undesirable in the
final product.

5.2.2 Control of materials

The materials used in the manufacture of the drug substance (e.g. raw
materials, starting materials, solvents, reagents and catalysts) should be
listed and information given on where each material is used in the process.
Information on the quality and control of these materials should be provided.
Reference to internationally accepted pharmacopoeias or details on the
specifications should be provided.

5.2.2.1 Control of source and starting materials of biological or animal origin

Information regarding the source, manufacture and characterization of all
biologically-sourced materials or materials that may have used biological
materials during manufacture should be provided. Summaries of the viral
safety information should be provided including appropriate certification
where applicable.

5.2.2.2 Source, history and generation of the host cell

Information should be provided on the bacterial host cell including its
source, phenotype and genotype.
The complete nucleotide sequence of the plasmid DNA should be provided.
In addition, the identity, source, isolation and sequence of the gene encoding
the antigen(s); a description of the steps involved in the construction of
the entire plasmid; a detailed functional map of the plasmid; information
on the source and function of component parts of the plasmid known to
have biological activities, such as origins of replication, viral/eukaryotic
promoters and other expression signals and genes encoding selection
markers, should be provided. A clear rationale should be provided for the
use of specific regions of DNA, such as the promoter or a gene encoding
a selection marker and special attention should be given to the nature of a
selection marker.

64
 Any modifications to the original native sequence(s) of the antigen should be
described and explained. The location of mammalian promoters in relation
to antibiotic resistance genes and the use of novel promoters or inducers
should be carefully considered. Certain sequences with properties of mobile
elements, such as insertion sequences or retroviral-like long terminal
repeats (LTRs), should be avoided. Oncogenes are not recommended unless
justified. It is also recommended that genes encoding enzymatic activity
or a biological function be either inactivated by genetic manipulation to
remove any undesirable activity, or justified. Further, although the relevance
at this stage may not be understood, as part of characterization, a DNA
sequence homology check of the plasmid with the international databases
(e.g. the National Center for Biotechnology Information, National Institute
for Health, USA, and/or other international nucleotide databases) should be
performed to investigate the presence of unintended sequences of biological
significance such as those encoding cellular growth functions or alternative
and unanticipated reading frames.
The identity of the plasmid after transfection into the bacterial cell to be used
for production should be confirmed in addition to the phenotype of the cell.
Representative restriction enzyme maps may be useful. Rearrangements of
the plasmid within the host bacterial cell are not acceptable.

5.2.2.3 Cell banking system, characterization and testing

The production of a plasmid DNA vaccine should be based on a cell bank
system involving an MCB and preferably a WCB. For early stage clinical
trials it may be appropriate to use the MCB although sponsors are encouraged
to prepare a WCB for later clinical studies.
A well-characterized bacterial cell containing the plasmid should be cloned
and used to establish the MCB. The preparation of the MCB and WCB
should be conducted according to GMP with appropriate precautions taken
to prevent contamination. Information should be provided on the origin,
form and storage conditions. Evidence for the viability of the MCB and
WCB under storage and recovery conditions should also be provided by the
time of application for marketing authorization. New WCBs should be fully
characterized and meet established acceptance criteria. Specific phenotypic
features that can form a basis for identification of the transformed cell
should be described.
The DNA sequence of the entire plasmid should normally be confirmed
at the stage of the MCB. Evidence that the MCB and WCB are free from
extraneous microbial agents should be provided.
The genetic stability of the plasmid should be confirmed by characterization
of the plasmid (copy number, size and sequence) after extended cell growth
(end of production) at some stage during development.

65
5.2.3 Process development and in-process control
The developmental history of the manufacturing process should be provided.
Tests and acceptance criteria for critical steps of the manufacturing process
should be developed to ensure, and provide feedback on, the control of the
process.
Validation of the manufacturing process should demonstrate reproducible
and consistent clearance of process and host-related contaminants to levels
acceptable for intended use in humans. Data from validation studies on
the purification procedures may be required to demonstrate clearance of
undesirable contaminants at each purification step and overall.
Process validation is not generally required for a product used in early-
stage clinical trials although critical steps such as aseptic processing,
sterilization of final product and cleaning validation—particularly when
multi-product facilities or contract manufacturing organizations are used
for the manufacturing—should be validated or carefully controlled prior to
initiation of clinical development.

5.3 Characterization
5.3.1 Characterization of bulk purified plasmid
A summary of the characterization of the drug substance should be provided
including its identity, strength, biological activity and purity. Rigorous
characterization by chemical, physical and biological methods will be
essential paying particular attention to the use of a wide range of analytical
techniques which are based on different principles.
During development, the sequence of the entire plasmid should be determined.
Attention should be paid to possible modification of the DNA because of
the possibility that such modifications may influence the biological and
immunological properties of the plasmid vaccine.
Potential impurities in the purified product should be described and investigated.
These impurities include host cell residues, residual RNA and chromosomal
DNA, materials used in the manufacturing process and media components.
Data should be provided on the contaminants present in the purified plasmid,
with estimates of their maximum acceptable or achievable levels. Denatured
plasmid DNA and partial degradation by nucleases are typically observed as
part of analytical procedures such as polyacrylamide gel electrophoresis, high
performance liquid chromatography and capillary electrophoresis.

5.3.2 Consistency of manufacturing

A number of batches should be characterized as fully as possible to determine
consistency. Any differences between one batch and another outside the
accepted range for the parameters tested should be noted. The data obtained

66
 from such studies should be used as the basis for the specification. During
early clinical development, demonstration of consistency may be limited
and occur as manufacturing experience is gained during the clinical
development phases. Characterization of consistency of lots is generally
done either during phase 3 or, if the phase 3 manufacturing process has
not been scaled up for commercial manufacture, after phase 3 and prior to
submission of a licence or marketing application.

5.4 Control of bulk purified plasmid

A specification for the drug substance should be established and justified.
Descriptions of analytical methods and acceptance limits for the drug
substance, including assay validation information should be provided.
A summary of the results of testing of all batches produced should be
provided.
It is recommended that the specification includes an assessment of the identity,
nature and quantity of the plasmid, purity, biological activity, endotoxin
content and sterility or bioburden. A justification of the specification should
be provided. Early in development the specification may be limited and
have wide acceptance criteria.
Not all the tests conducted during product characterization need to be carried
out on each batch of vaccine. Some tests are required only to establish the
validity or acceptability of a procedure, whereas others might be performed
on a limited series of batches to establish consistency of production. Thus,
a comprehensive analysis of the initial production batches should be
undertaken to establish consistency with regard to identity, purity, potency
and stability but thereafter a limited series of tests may be appropriate.

5.4.1 Identity
Each batch should undergo an appropriate selection of the tests used to
characterize the purified plasmid in order to confirm its identity. However,
the specific tests that adequately characterize any particular plasmid on
a lot-to-lot basis, may depend on both the nature of the plasmid and its
method of production and purification. Typically, restriction analysis will be
the primary approach used to confirm identity; however, in vitro or in vivo
expression of the plasmid accompanied by confirmation of the identity of
the expressed antigen should also be considered.

5.4.2 Nature and quantification of plasmid, and biological activity

Quantification of the plasmid is usually by absorbance at 260 nm. The
proportion of supercoiled plasmid should be determined and specifications
set.

67
 The biological activity of each batch of the drug substance should be
determined using a suitable well characterized assay together with an
appropriate in-house reference preparation. For biological activity,
whenever possible, the antigen should be expressed in vitro by transfection
of a suitable cell line and the expressed protein characterized, for example,
by immunofluorescence or by Western blot. Where possible, the in vitro
assay should be shown to correlate with immunogenic activity or efficacy in
an animal model. Alternatively, the plasmid should be shown to possess the
relevant immunogenic or biological activity in an animal model.

5.4.3 Purity
Limits based on process capability and regulatory guidance should be
established for all impurities detected and these should be identified and
characterized as appropriate. The degree of contamination with chromosomal
DNA, RNA and proteins should be assessed and limits established, and the
criteria for rejection should be established and specified. It is important that
the techniques used to demonstrate purity be based on as wide a range of
physicochemical properties as possible. Where multiproduct facilities or
contract manufacturing organizations are used for the manufacturing process,
freedom from contamination with other products should be demonstrated.

5.5 Reference standards or materials

An in-house reference preparation should be established for use in assay
standardization. Information on the reference standards or reference
materials used for testing of the drug substance should be provided by the
time of application for marketing authorization.
A suitable batch, preferably one that has been clinically evaluated, should
be fully characterized in terms of its chemical composition, purity and
biological activity, including, where possible, full sequencing, and retained
for use as a chemical and biological reference material.

5.6 Stability
The stability assessment should be in compliance with the International
Conference on Harmonisation (ICH) guideline Q5C “Stability testing of
biotechnological/biological products”. The types of studies conducted, protocols
used and the results of the studies should be summarized in an appropriate
format such as tables, graphs or a narrative document. The summary should
include results as well as drawing conclusions with respect to appropriate
storage conditions and retest date or shelf-life. Limited stability information
would be expected during initial clinical development. Further data on stability
to support the expiry date of the drug substance for licence should be based
on long-term, real-time stability studies under actual conditions of use.

68
6. Manufacture and control of final formulated
vaccine (drug product)
6.1 Composition
The final composition of the vaccine should be documented.

6.2 Manufacture
A flow chart should be provided that illustrates the manufacturing steps
from the bulk purified plasmid to the final formulated vaccine. The chart
should include all steps (i.e. unit operations), identification of materials,
major items of equipment and in-process controls. In some cases, this may
involve simple dilution of the purified bulk; in other cases, a more complex
formulation may be envisaged.
A description of each process step depicted in the flow chart should be
provided. Information should be included on, for example, scale, buffers
and other additives, major equipment, and process controls, including in-
process tests and operational parameters with acceptance criteria.

6.3 Control of materials

Details of excipients, adjuvants or any other component of the vaccine in
addition to the plasmid constituting the active substance should be provided,
including their source, specification, method of conjugation, if appropriate,
and final concentration in the vaccine.

6.4 Control of final formulated vaccine

A specification for the drug product should be established and justified.
Descriptions of analytical methods and acceptance limits for the drug
product, including information on assay validation should be provided. It is
recommended that the specification includes an assessment of the identity,
nature and quantity of the plasmid, purity, potency, endotoxin content and
sterility. A justification of the specification should be provided. Early in
development, the specification may be limited with wide acceptance
criteria.
A summary of the results of the testing on all batches produced should be
provided.
The appropriateness of performing tests on the bulk purified plasmid versus
the formulated vaccine should be considered on a case-by-case basis and
justified.
Several batches of vaccine, including the final dosage form, should be
characterized as fully as possible to determine consistency. Any differences

69
 between one batch and another should be noted. The data obtained from
such studies should be used as the basis for the drug product specification.
Not all the tests conducted during product characterization need to be carried
out on every batch of vaccine. Some tests are required only to establish the
validity or acceptability of a procedure, whereas others might be performed
on a limited series of batches to establish consistency of production. Thus, a
comprehensive analysis of the initial production batches should be undertaken
to establish consistency with regard to identity, purity, potency and stability
but thereafter a more limited series of tests may be appropriate.

6.4.1 Identity
Each batch of vaccine should be subjected to an appropriate selection of
the tests used to characterize the purified plasmid to confirm its identity.
The specific tests that adequately characterize any particular plasmid on
a lot-to-lot basis, however, may depend on both the nature of the plasmid
and the method of its production and purification. Depending on the scope
of identification tests, confirmation of the sequence or restriction enzyme
mapping and verification of expression following transient transfection,
will be necessary.

6.4.2 Purity
The purity of each batch of vaccine should be determined and be shown
to be within specified limits. The analysis should include sensitive and
reliable assays for contaminants of bacterial-cell origin and strict upper
limits should be specified for their content in the bulk purified plasmid.
Where multiproduct facilities or contract manufacturing organizations are
used for the manufacturing process, freedom from contamination with other
products should be demonstrated.

6.4.3 Potency
The potency of each batch of the vaccine should be determined using a
suitable well-characterized assay together with an appropriate in-house
reference preparation. A potency assay should be established that can be
correlated to functional activity. This may take the form of a quantitative
in vitro expression system or may require titration in a defined animal
model to determine the minimum quantity of vaccine required to induce an
appropriate immune response. The potency of the final vaccine formulation
should be established unless otherwise justified and should be correlated
with vaccine efficacy.

6.4.4 Sterility
Each batch of final product should be tested for sterility.

70
6.4.5 Other tests
The final product specification should include tests for endotoxin, visual
appearance and pH. Other tests, such as the residual moisture, if the product
is lyophilized, may be required to confirm the physical characteristics of the
product as well as the formulation.

6.4.6 Multi-component vaccines

Additional factors must be considered when more than one plasmid forms
the final formulated vaccine form. Plasmids in multi-component vaccines
may encode additional antigens or cytokines or other biologically active
molecules which enhance the efficacy or affect the safety of the vaccine.
For each plasmid, the development overview, the control of production
and the characterization of the bulk purified plasmid must be described
as above. Careful consideration has to be given to the control of the final
formulated vaccine. For example, potency may depend on the combination
of plasmids and their interaction and not on any single plasmid component
of a multicomponent vaccine.

6.5 Reference materials

A suitable batch of the final formulated vaccine, or bulk purified plasmid,
preferably one that has been clinically evaluated, should be fully characterized
in terms of its chemical composition, purity and biological activity, including,
where possible, full sequencing, and retained for use as a chemical and
biological reference material. This material should be used as the basis for
evaluation of product quality for production batches.

6.6 Stability
Adequate stability studies form an essential part of vaccine development. The
stability of the final product in the container proposed for use should, therefore,
be determined and the results used to set a shelf life under appropriate
storage conditions. Real time stability studies should be undertaken for
this purpose, but accelerated stability studies at elevated temperatures may
provide complementary supporting evidence for the stability of the product
and confirm the stability indicating nature of the assays used to determine
stability. The stability assessment should comply with ICH guideline Q5C.

7. Nonclinical safety evaluation

Although the recommendations set out below should be considered generally
applicable, individual products may present particular safety concerns. The
nonclinical safety evaluation should be considered on a product-specific
basis taking into account the intended clinical use of the product. It is

71
 important to note that when addressing some of the nonclinical and clinical
issues a clear understanding of the product characteristics would be required
to enable appropriate studies relating to the toxicology and pharmacology
of the product to be designed.

7.1 Introduction
With the advent of DNA vaccines in the 1990s, several potential safety issues
associated with administering plasmid DNA to humans were discussed: e.g.
integration into the host’s chromosomes, immunopathological reactions,
risks related to expression of cytokines or co-stimulatory molecules,
antibodies against the injected DNA and potential biological activity of the
expressed antigen. This revision of the WHO guidelines takes into account
the accumulated experience of the use of DNA vaccines in both nonclinical
and clinical development. Many phase 1 clinical trials in humans have been
conducted with prophylactic DNA vaccines and at least one phase 2 clinical
trial with a therapeutic vaccine is in progress. Studies using different doses
of up to multi-milligrams, different schedules, routes of administration and
delivery devices have been performed. In general, plasmid DNA appears
to be safe and well-tolerated, i.e. no serious adverse reactions or major
concerns related to the monitored parameters have been noted. At present,
prime–boost strategies with other types of vectors, e.g. plasmid DNA/
adenovirus and multiple immunization strategies are being investigated to
improve immunogenicity. Many animal studies have been performed both
with laboratory animals and with veterinary target animal species. It may
be useful for sponsors to consult the published literature on experience with
relevant animal models in the veterinary field.

7.2 General considerations on safety

This guidance is intended to address the issues specifically related to DNA
vaccines. It should be read in conjunction with the guidance provided in
the WHO guidelines on nonclinical evaluation of vaccines (2) and the
nonclinical section (part A) of the WHO guidelines on clinical evaluation
of vaccines: regulatory expectations (3).
The general aim of nonclinical safety evaluation is to determine whether new
medicinal products have the potential to cause unexpected and/or undesirable
effects. The study design should be based on the intended clinical use and
focus on tests that are relevant for a plasmid DNA product. Classical safety or
toxicological testing, as recommended for chemical drugs, will have limited
application to a plasmid DNA product. As for other biotechnology products,
studies with animal models for safety assessment consider two issues:
1) selection of the animal species and physiological state; and 2) the manner
of delivery, including the dose, the route of administration and the treatment

72
regimen (frequency and duration). The nonclinical safety evaluation should
provide information about a safe starting dose for the initial human trials.
Using animal models to gain knowledge about the response to be expected
in humans, always carries the risk that the animal species used in the model
does not predict the human response or shows a non-relevant species-specific
response. Relevant animal models should be used where possible, i.e.
species/models whose immunogenic or biological response to the delivery
system and expressed gene product would be expected to be similar to the
response in humans.
The risk–benefit evaluation of a product is related to the actual product
and its intended use. For example, a prophylactic DNA vaccine for use in
healthy children will have a risk–benefit ratio different to that of a therapeutic
vaccine against cancer for which there is no other available treatment. Thus,
for these and other reasons, it is likely that a flexible approach will be
necessary for the nonclinical safety evaluation of DNA vaccines.
If modifications to the manufacturing process or the DNA product are made
during the development programme, the potential impact on the product
should be considered. Modifications of the genetic sequence, the use of
alternative promoter or enhancer sequences, or other changes to the product,
may require additional nonclinical safety evaluation. The scientific rationale
for the approach taken should be provided.
Although safety testing will undoubtedly be required, the range of tests that
need to be carried out should be decided on a case-by-case basis, preferably in
consultation with the national control authority. A suitable range of biological,
molecular, biochemical, immunological, toxicological and histopathological
investigative techniques should be used in the assessment of a DNA vaccine’s
effect, over an appropriate range of doses during both acute and repeated
exposure. Where there is experience with nonclinical and clinical use of a
particular delivery vector by a particular route of administration, it may be
possible to use information from previous studies or literature in place of
further experimental work.
Several potential safety issues associated with administering plasmid DNA
into humans are discussed below:
• The injected DNA taken up by cells of the host may integrate into the
host's chromosomes and cause an insertional mutagenic event.
• The long-term expression of a foreign antigen may result in an undesired
immunopathological reaction.
• The use of genes encoding cytokines or co-stimulatory molecules may
pose additional risks.
• Antibodies may be formed against the injected DNA itself and these may
contribute towards undesired autoimmune reactions.

73
 • The expressed antigen may itself have biological activity.
• Expression of other gene sequences in mammalian or bacterial cells may
pose a risk.

DNA insertion
It is known that DNA taken up by mammalian cells in culture can integrate into
the cellular genetic material and be faithfully maintained during replication.
This is the basis of the production of some recombinant therapeutic proteins.
Theoretically, the introduction of extraneous DNA into a susceptible cell
type in vivo could cause a transformation event leading to the formation of
tumour cells by the insertion of an active oncogene, by insertional activation
of a host-cell proto-oncogene or by insertional deactivation of a suppressor
gene. DNA insertion can occur in one of three ways: by random integration
of the plasmid into the genome, by homologous recombination between
homologous sequences carried by the plasmid and the genome or by a direct
integrating mechanism such as is employed by retroviruses. The most likely
means in the present context would be random integration.
After injection of naked DNA into an animal, only a small proportion of the
DNA molecules enter cells and of those only a fraction are likely to enter the
nucleus. The probability of any DNA molecule integrating into a chromosome
is very low. When consideration is given to the probability of insertional
mutation occurring at a growth regulatory gene and to the multi-step process
of oncogenesis, the risk of a tumorigenic event becomes exceedingly low.
Several investigators have assessed the ability of particular plasmids to
integrate in vivo based on the association of plasmid DNA with genomic
DNA after gel purification (6, 7). In most cases negative results have been
obtained at a sensitivity in the region of one potential plasmid integration
event per microgram of host-cell DNA (corresponding to approximately
150 000 diploid cells). However, alternative formulations or administration
devices such as co-inoculation of a plasmid encoding a growth promoting
factor, or electrostimulation, can lead to an increased potential for integration
of plasmid DNA, and an investigation of the potential integration of the
plasmid DNA in vivo into the host’s chromosomes should form part of the
nonclinical safety testing of a DNA vaccine. Integration studies may not
be necessary for a plasmid DNA vaccine if prior information on a similar
plasmid, with the same mode of administration already exists. There would
be a need to reassess integration if there was a significant change in the
method of delivery, especially any change potentially involving an increase
in the capacity of plasmid DNA to enter the nucleus.

Immunopathological reaction
Clinical studies to date have shown that plasmid DNA is well tolerated.
Immunopathological reactions such as general immunosuppression and

74
inflammation have not been reported. Advances in understanding the
mechanism of the immune response to an antigen which is expressed from
injected DNA have been made although much remains unclear. Knowledge
of the duration of expression of an antigen from injected DNA is limited
although some reports suggest that expression could continue for many
months, which means that the possibility of tolerance may remain a concern.
In nonclinical investigations to date, tolerance has not been observed in
adult animals and humans and the initial concern may have been overstated.
Tolerance can be induced in neonatal mice; this may be because the mouse
immune system at birth is immature and if development of tolerance is a
concern for a specific product, a more relevant animal model is desirable.

Risks of genes encoding cytokines or co-stimulatory molecules

The co-administration of genes encoding regulatory cytokines or other
immunostimulatory molecules has been used to improve the immune
response to a DNA vaccine. However, such molecular adjuvants may have
additional unintended consequences such as the possibility of stimulating
one arm of the immune response at the expense of the other or, in theory,
leading to immunopathological reactions, e.g. immunosuppression, chronic
inflammation or autoimmunity. Considerable data on the safety (and
usefulness) of this approach has accrued especially on their use in clinical
trials of gene therapy and therapeutic studies with recombinant protein
in humans; however, this approach should be followed carefully. Studies
should continue to address the possibility that some cytokines may produce
local or systemic toxic effects. The persistence of a cytokine expressing
plasmid should be monitored. It cannot be ruled out that studies in animals
responsive to the encoded human cytokine or models using the analogous
animal genes may be useful to clarify safety issues.

Autoimmune reactions
Bacterial DNA can promote the production of IgG anti-DNA auto-
antibodies generally associated with the development of auto-immune
glomerulonephritis in diseases such as systemic lupus erythematosus and
this initially raised the theoretical concerns that DNA vaccination using
bacterial plasmid DNA might induce such disease. Consistent with such a
possibility, sensitive enzyme-linked immunosorbent assay analysis of serum
samples from humans and animal models have shown that repeated DNA
vaccination can stimulate a ≤ 5-fold increase in anti-DNA auto-antibody
levels. Such an increase may not be detected by less sensitive clinical
antinuclear antibody screening and the levels observed are well below
that associated with the development of autoimmune disease. Although
the possibility of anti-DNA antibody production should be considered in
the development of plasmid DNA vaccines, analysis in the nonclinical
programme is not generally warranted. Improvements in the efficiency of

75
 DNA delivery and/or increases in vaccine dose and frequency may change
the need for analysis of anti-DNA antibodies.

Unwanted biological activity

Consideration must be given to the possibility that the in vivo synthesized
antigen may exhibit unwanted biological activity. If necessary, appropriate
steps must be taken, e.g. by deletion mutagenesis, to eliminate this activity
while retaining the desired immune response.

Expression of other gene sequences

If other gene constructs are included in the plasmid, such as antibiotic
resistance genes for manufacturing reasons, then the possibility of expression
of such gene sequences in mammalian cells or in microorganisms which
are potentially pathogenic, and the possible clinical consequences of such
expression, should be considered.
It is encouraging that data acquired to date demonstrate the safety of plasmid
DNA. The data set is expanding, but the above issues must continue to be
addressed, especially as measures are being sought to increase the efficacy
of DNA vaccines.

7.3 Considerations on the nonclinical safety programme

In designing the nonclinical safety programme for a DNA vaccine product, the
WHO guidelines on nonclinical evaluation of vaccines should be consulted in
addition to the guidance provided here. Sponsors are also advised to perform
a literature search for relevant published literature which should be used as
part of safety evaluation and for justification of their safety programme.
Every product should be evaluated on a case-by-case basis. As a general
rule, nonclinical safety assessment should be performed on every novel
vaccine or vaccine/adjuvant formulation.
The following parameters should be considered and incorporated into the
design of the nonclinical study: the nature of the vaccine, administration
route (e.g. intranasal, intravenous, intramuscular or oral), formulation (e.g,
liposome encapsulation, poloxamers) and any technique used to improve the
uptake of the plasmid (e.g. electroporation). Data from a similar construct
may be used to support the nonclinical safety assessment, but will require
careful consideration.
When the drug product consists of more than one individual plasmid,
testing of the drug product (formulated vaccine) is preferred unless there is
reason to suspect one particular plasmid may present a significantly higher
risk. The data from a drug product with several plasmids may be applicable
for a drug product that contains a subset of the plasmids, unless there is
interference between plasmids.

76
7.4 Toxicity studies
Studies should follow good laboratory practice (GLP) regulations and
should test the final formulation intended for clinical use. However, certain
assays might be conducted following the principles of GLP when newly
developed methods have not been fully established.
The toxicity study may be combined with an assessment of local tolerance,
immunogenicity and biodistribution evaluations. When possible, the dose,
route and schedule should follow that intended for clinical use and the
number of doses should be equal to or exceed that intended in the clinic.
The dose interval can be shortened to 2–4 weeks or to another appropriate
interval related to the intended dose regimen, although the kinetics of the
immune response and potential toxic effect of the gene product should be
considered. Relevant biodistribution parameters may also be investigated to
allow any findings to be correlated with presence or expression of the gene
product.
The animal model should be relevant and the product should be immunogenic
in the chosen species. It is recommended that the study include an
appropriate number of different doses and a vehicle or other appropriate
control group. Toxicity assessments should be performed after both an
acute and a recovery (follow-up) period, e.g. at 2–3 days and 14–21 days
after the last administration. In general, studies in non-human primates are
not required before proceeding to human trials. However, if the expected
toxicity is species-specific, primates or transgene mouse models may be
more predictive of clinical toxicity. Where possible the same lot of material
should be used in the nonclinical safety evaluation and the initial clinical
study.
Assessments should include daily clinical observations and injection site
reactions, food consumption and body weight. Laboratory assessments
should include clinical chemistry and haematology. Postmortem
investigations should include macroscopic and microscopic assessments in
an appropriate range of tissues, e.g. the injection site, spleen, liver, kidney,
intestines, brain, bone marrow, ovaries/testes, lungs, lymph nodes, heart and
adrenals. Findings should be assessed in relation to the pattern and severity
of the effects and the relevance in relation to the intended product.
The studies should address product-specific concerns including the need
for auto-antibody testing. In general, testing for anti-DNA-antibodies is no
longer required. Local inflammatory response (e.g. myositis), organ-specific
autoimmunity, immunopathology and other relevant parameters may need
to be included. In particular where the encoded antigen is a self-antigen, or
may show self-antigen mimicry, a wider range of studies, including auto-
antibodies, may be necessary.

77
7.5 Immunogenicity studies
The purposes of the immunogenicity analysis are to define an appropriate
dose, route, schedule and formulation of the vaccine for clinical trials and
to provide justification (i.e. the benefit side of the risk–benefit equation) and
rationale for the clinical trial. The results of assays to show immunological
activity in an animal model should be presented. These could comprise
antigen-specific antibody titres, serum neutralization titres, seroconversion
rates, activation of cytokine-secreting cells and/or measures of cell-
mediated immune responses. The studies should optimally be designed
to give information about the duration of the immune response and may
be combined with challenge or protection studies. If the DNA vaccine
is intended to express, for example, a human cytokine, considerations
should be given to the species-specificity, i.e. the animal species should be
responsive to the cytokine, if possible, or a model using the corresponding
animal cytokine may be used.

7.6 Biodistribution, persistence and integration

After inoculation of plasmid DNA into an experimental animal system,
assays to assess the distribution, duration and potential integration of the
plasmid should be performed, together with an assessment of the germ line,
unless otherwise justified.
Biodistribution and persistence studies are required, unless substantial
experience has already been gained with an almost identical or similar
product. Biodistribution and persistence should be investigated using, for
example, sensitive nucleic acid detection techniques, and the justification
for the chosen assays should be stated. The assay limit of sensitivity, its
specificity, and the potential for tissue- or preparation-specific inhibitors
should be established during the assay validation. Among the nucleic acid
detection techniques, quantitative polymerase chain reaction (PCR)-based
assays have been the most commonly used and most reliable means of
assessing plasmid levels in biodistribution and integration studies. The
amount of plasmid in the relevant tissues should be quantified and the
persistence of plasmid at each site over time should be monitored at an
appropriate number of time-points, both early (e.g. 1–7 days) and late
(e.g. 2-3 months). The duration and sites of expression of the encoded
proteins over time should be investigated. If the encoded protein product
is expected to persist for a considerable length of time, the impact of this
should be addressed.
Size fractionation/gel-purification assays, where plasmid co-migrates
with genomic DNA, may be used as a parameter for potentially integrated
plasmid DNA (6, 7). The sensitivity and specificity of the chosen assay
should be well-documented.

78
 Published studies on biodistribution of DNA vaccines indicate that
intramuscular, subcutaneous, intradermal or particle-mediated delivery
does not result in long-term persistence of plasmid at ectopic sites. Some
studies have reported ≤ 30 copies of plasmid per 100 000 host cells persist
at the site of injection, after 60 days (8–10), whereas other studies have
shown that the plasmid can persist at the injection site at levels greater
than 500 copies per 100 000 cells 6 months post-administration (7). If
≤ 30 copies of plasmid DNA per 100 000 host cells persist after 60 days, then
further integration assessment may not be necessary. As more experience
is gained, this issue will need to be addressed again.
Depending on the potential for integration and the proposed clinical
indication, further studies may be required to investigate directly any actual
integration, e.g. by specially designed PCR-primers (11), or the potential
for tumour formation or disruption of normal gene expression.
It may also be necessary to investigate the distribution and clearance of
the material used for delivery of the DNA, e.g. complexing material, if
sufficient information is not already available in the published literature.
The studies should identify sites of uptake after in vivo delivery.

7.7 Genotoxicity
The standard battery of genotoxicity and conventional carcinogenicity
studies is not applicable to DNA vaccines. However, genotoxicity studies
may be required to address a concern about a specific impurity or novel
chemical component, e.g. a complexing material that has not been tested
previously.

7.8 Developmental and reproductive toxicity

Integration into reproductive tissue may result in germline alteration. The
possibility of distribution to, integration or expression in germline cells
must be investigated unless otherwise justified, e.g. the clinical indication or
patient population indicates that such studies are not warranted. Confirmation
of absence of germline alterations may be gained from investigating the
presence and persistence of the plasmid DNA in gonadal tissue from
both male and female animals, using e.g. PCR technology. Persistence
of plasmids in gonadal tissue over time require further investigation, e.g.
of ova and sperm cells and considerations of potential effects on fertility
and general reproductive function. In addition, embryo-fetal and perinatal
toxicity studies may be required if women of childbearing potential are to be
exposed to the product, depending on intended clinical use and population.
Such studies may not be required prior to clinical studies in populations
with life-threatening diseases, provided appropriate measures are taken to
minimize risks. Before a DNA vaccine is used in children or newborns, it

79
 should be tested for safety and immunogenicity in adults, and appropriate
nonclinical models, e.g. with juvenile animals, should be considered for the
study of toxicity and induction of immunological tolerance.

Authors
The draft of these revised Guidelines was prepared by Dr J. Robertson, National
Institute for Biological Standards and Control, Potters Bar, England; Mr J. Ackland,
The Biologics Consulting Group, San Mateo, California, USA; Dr A. Holm, Danish
Medicines Agency, Copenhagen, Denmark.

Acknowledgements
Acknowledgements are due to the following experts for their comments, advice
and information given at an Informal WHO Consultation on the WHO guidelines for
assuring the quality and preclinical safety evaluation of DNA vaccines held at WHO
Headquarters, Geneva, from 4–5 July 2005:

Dr T. Bektimirov, L.A. Tarasevich State Research Institute for Standardization and

Control of Medical Biological Preparations, Moscow, Russian Federation; Dr K.
Cichutek, Paul-Ehrlich-Institute, Langen, Germany; Mr D. Daout, Serum Institute of
India, Ltd, Nyon, Switzerland; Dr F. Ferre, Althea Technologies Inc., San Diego,
California, USA; Dr D. Gannaway, PowderMed Ltd, Oxford, England; Dr H. Golding,
Center for Biologics Evaluation and Research, Food and Drug Administration,
Bethesda, Maryland, USA; Dr J. Joung, Korea Food and Drug Administration, Seoul,
Republic of Korea; Dr D. Kaslow, Vical Incorporated, San Diego, California, USA;
Dr A. Kojima, National Institute of Infectious Diseases, Tokyo, Japan; Dr R. Krause,
International Federation of Pharmaceutical Manufacturers Associations, Geneva,
Switzerland; Dr B. Ledwith, Merck Research Laboratories, West Point, Pennsylvania,
USA; Dr P. Manyike, South African AIDS Vaccine Initiative, Cape Town, South Africa;
Dr S. Phumiamorn, Department of Medical Sciences, Ministry of Public Health,
Thailand; Dr T. Richie, Naval Medical Research Center, Silver Spring, Maryland,
USA; Dr G. Sharpe, Cobra Bio-Manufacturing plc., Keele, Staffordshire, England;
Dr G. Thiry, International AIDS Vaccine Initiative, New York, USA; Dr J. Ulmer, Chiron
Corporation, Emeryville, California, USA; Dr J. van der Laan, National Institute for
Public Health and the Environment, BA Bilthoven, The Netherlands; Dr Y. Wang,
National Institute for the Control of Pharmaceutical and Biological Products, Temple
of Heaven, Beijing, People’s Republic of China; Dr D.Weiner, University of
Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA.

Thanks are also due to the following for their written comments and advice:

Dr D. Gannaway, PowderMed Ltd., Oxford, England; Dr J. Joung, Korea Food and

Drug Administration, Seoul, Republic of Korea; Dr D. Kaslow, Vical Incorporated,
San Diego, California, USA; Dr J. Lebron, Merck Research Laboratories, West Point,
Pennsylvania, USA; Dr S. Phumiamorn, Department of Medical Sciences, Ministry of
Public Health, Thailand; Dr F. Reigel, Biological Medicines and Laboratories,
Swissmedic Agency for Therapeutic Products, Berne, Switzerland; Dr R. Sheets,

80
 Vaccine Research Center, National Institute of Allergy and Infectious Diseases,
National Institutes of Health, Bethesda, MD, USA; Dr G. Thiry, International AIDS
Vaccine Initiative, New York, USA; Dr G. Vicari, Via Principessa Clotilde, Rome, Italy.

WHO Secretariat
Dr D. Wood, Coordinator, Quality Assurance and Safety: Biologicals, World Health
Organization, Geneva, Switzerland; Dr J. Shin, Scientist, Quality Assurance and
Safety: Biologicals, World Health Organization, Geneva, Switzerland.

References
1. Guidelines for assuring the quality of DNA vaccines. In: WHO Expert
Committee on Biological Standardization. Forty-seventh report. Geneva,
World Health Organization, 1998, Annex 3 (WHO Technical Report Series,
No. 878).
2. Guidelines on nonclinical evaluation of vaccines. In: WHO Expert Committee
on Biological Standardization. Fifty-fourth Report. Geneva, World Health
Organization, 2005, Annex 1 (WHO Technical Report Series, No. 927).
3. Guidelines on clinical evaluation of vaccines: regulatory expectations. In:
WHO Expert Committee on Biological Standardization. Fifty-second report.
Geneva, World Health Organization, 2004, Annex 1 (WHO Technical Report
Series, No. 924).
4. Good manufacturing practices for pharmaceutical products. In: WHO Expert
Committee on Specifications for Pharmaceutical Preparations. Thirty-seventh
report. Geneva, World Health Organization, 2003, Annex 4 (WHO Technical
Report Series, No. 908).
5. Good manufacturing practices for biological products. In: WHO Expert
Committee on Biological Standardization. Forty-second report. Geneva,
World Health Organization, 1992, Annex 1 (WHO Technical Report Series,
No. 822).
6. Ledwith BJ et al. Plasmid DNA vaccines: assay for integration into host
genomic DNA. Developments in Biologicals, 2000, 104:33–43.
7. Ledwith BJ et al. Plasmid DNA vaccines: investigation of integration into
host cellular DNA following intramuscular injection in mice. Intervirology,
2000, 43:258–272.
8. Kim BM et al. In vivo kinetics and biodistribution of a HIV-1 DNA vaccine
after administration in mice. Archives of Pharmacal Research, 2003,
26:493–498.
9. Pilling AM et al. The assessment of local tolerance, acute toxicity, and DNA
biodistribution following particle-mediated delivery of a DNA vaccine to
minipigs. Toxicolologic Pathology, 2002, 30:298–305.
10. Bureau MF et al. Intramuscular plasmid DNA electrotransfer: biodistribution
and degradation. Biochimica Biophysica Acta, 2004, 1676:138–148.
11. Wang Z et al. Detection of integration of plasmid DNA into host genomic
DNA following intramuscular injection and electroporation. Gene Therapy,
2004, 11:711–721.

81
© World Health Organization
WHO Technical Report Series No 941, 2007

Annex 2
Recommendations for inactivated rabies vaccine
for human use produced in cell substrates and
embryonated eggs

These Recommendations provide information and guidance to national

regulatory authorities and vaccine manufacturers concerning the
characterization, production and control of rabies vaccines in order to
facilitate their international licensure and use. Each of the following sections
constitutes a recommendation. The parts of each section that are printed
in normal type have been written in the form of requirements so that if
a national regulatory authority so desires these parts may be adopted
as definitive national requirements. The parts of each section printed in
small type are comments and additional guidance. It is recommended that
modifications to these Recommendations be made only on condition that
the modifications ensure that the vaccine is at least as safe and efficacious
as that prepared in accordance with the Recommendations set out below.
In order to facilitate the international distribution of vaccine made in
accordance with these recommendations, a summary protocol is given
in the Appendix.

Introduction
General considerations
Part A. Manufacturing recommendations
A.1 Definitions
A.2 General manufacturing recommendations
A.3 Control of source materials
A.4 Control of vaccine production
A.5 Filling and containers
A.6 Control tests on final product
A.7 Records
A.8 Samples
A.9 Labelling
A.10 Distribution and shipping
A.11 Stability, storage and expiry date
A.12 Intradermal route of administration
Part B. Nonclinical evaluation of rabies vaccines
Part C. Clinical evaluation of rabies vaccines
Part D. National control requirements
D.1 General
D.2 Release and certification
Authors
Acknowledgements
References
Appendix
Summary protocol

83
 Introduction
The last revision of the requirements for rabies vaccines for human use
was in 1980 (1). However, an additional document, WHO Requirements
for rabies vaccine (inactivated) for human use produced in continuous
cell lines was published in 1987 to take into consideration advances in the
development of cell culture derived vaccines (2). An amendment which
updated the section on the International Standard for Rabies Vaccine was
published in 1994 (3, 4).
The following Recommendations are for inactivated rabies vaccine for human
use produced only in cell substrates and embryonated eggs. They replace all
previous requirements (1–4). The scope of the present recommendations
encompasses vaccines produced in cell substrates, ranging from primary
cells (hamster kidney and chick embryo fibroblasts), diploid cells, to
continuous cell lines such as Vero cells. Purified vaccines produced using
duck embryos are also within the scope of the document.
However, vaccines produced in mammalian neural tissues are not considered
in this or any other document because their use is no longer recommended.

General considerations
Rabies is an under-reported, neglected and deadly disease estimated to
cause more than 50 000 human deaths annually, most of which occur in the
poorest regions of the world (5). Once clinical symptoms are evident, the
prognosis for survival is poor and death is almost inevitable. The population
at risk includes 2.5 billion people currently living in regions in which
rabies is endemic. Half of the victims of dog bites and rabies subsequent
deaths from occur in children younger than 15 years of age, as they are
the population most at risk. (6). Clearly, greater efforts should be made
to improve the control of rabies, a zoonotic disease with the highest case
fatality rate known to humans.
One of the most important elements in the effective control of human rabies
is the use of efficacious vaccines. Vaccines produced in mammalian neural
tissues have been in use for more than 100 years. However, it is the availability
and use of rabies vaccines produced in cell culture and embryonated eggs
that has dramatically decreased the number of human deaths throughout
the world, most notably in countries where canine rabies is endemic. For
example, in Thailand, the introduction of cell culture vaccines together with
reduced dosage intradermal regimens decreased the incidence of human
rabies by 80% in 15 years (5). Similar progress has been documented in
other countries where nerve tissue vaccines have been replaced by rabies
vaccines produced in cell culture and embryonated eggs.

84
This document focuses on the recommendations for production, control
and evaluation of rabies vaccines, which as stated above, are one of the
most important elements in the effective control of human rabies. However,
vaccine should always be considered as part of the complete treatment and
additional information on recommendations for the treatment of the disease
is available in the Report of the Expert Consultation on Rabies (7).
WHO requirements for rabies vaccines were published in 1981 and 1987.
The former requirements encompassed vaccines derived from mammalian
neural tissue as well as vaccines produced using embryonated eggs and
variety of cell substrates, whereas the latter covered only vaccines produced
in continuous cell lines. Since that time, there have been many developments
in the production and quality control of vaccines as well as in their overall
regulation. In particular, considerable attention has been given to safety
issues.
The scientific basis for the present revision of the requirements for rabies
vaccines was developed at the meetings of a working group held at WHO,
Geneva, in May 2003 and May 2004. The issues identified for revision were:
the scope of the document; the substrates for vaccine production that the
revised document would cover; the inactivation process; the test for effective
inactivation; potency test; the use of in vitro assays for determination of the
antigen content as a measure of consistency of production; stability test
and the value of the accelerated degradation test; and national regulatory
authority requirements. Further details of these discussions and the rationale
for the proposed revisions are available in the meeting reports (8, 9).
Rabies vaccines produced in mammalian neural tissues (brain of adult
animals such as sheep and goats; brain of suckling animals such as mouse,
rat and rabbit) have been in use worldwide for many years. It is well known
that their use has led to adverse reactions following immunization, such
as encephalomyelitis and polyneuritis (10, 11). Although the risk of such
adverse reactions is reduced when the virus is grown in the brains of newborn
animals, such as rats and mice, before the development of myelin in the
brain, the safety profile of these vaccines is still unacceptable. Moreover,
there is evidence for a lack of potency of these neural tissue vaccines,
leading to inadequate protection in humans, making a strong argument for
the discontinuation of their production and use (12–14). The present revised
recommendations are intended to improve control of rabies disease by
promoting vaccines of assured quality as part of pre-exposure vaccination
and post-exposure prophylaxis. To facilitate the international distribution of
vaccine produced in accordance with these recommendations, a summary
protocol is given in the Appendix.
Recently developed methods for genetic sequencing of rabies virus have
been considered in this document. Given that the genetic characteristics are

85
 part of the identity of the vaccine strains, it would be beneficial to include
this information in the licensing of new vaccines and to use sequencing in
monitoring for subtle genetic changes of vaccine strains over time.
The approach to potency testing remains the same as previously recommended.
The National Institute of Health (NIH) potency test, based on a mouse
protection assay, is recognized as a reliable assay. A review of the data on a
single-dilution NIH test has led to the development of criteria for the validation
of a modified NIH assay. The latter has been extensively used for lot release of
rabies vaccine for veterinary use while the experience in testing vaccines for
human use is still at the experimental stage. More data are needed to support
this approach and to provide a basis for a standardized testing procedure.
Several studies conducted over the last 10 years have provided useful
information on the value and potential use of the in vitro assays for
measurement of the antigen content in vaccines (National Institute for
Biological Standards and Control (NIBSC) and Agence Française de Sécurité
Sanitaire des Produits de Santé (AFSSAPS) studies). Such assays have been
successfully used by several manufacturers to control antigen concentration
during production and in the final formulation of a product. However, in
vitro data concerning antigen concentration in the final vaccine have not
generally been reported and direct correlations between such determinations
and evidence of protection in humans need to be evaluated. Correlations
between the in vitro assays and the NIH test have proven challenging owing
primarily to the variability of the NIH test. Further characterization of the
reagents used for in vitro assays may clarify the potential of these methods
to assess the quality as well as the quantity of antigen. Both manufacturers
and national regulatory authority/national control laboratory staff are
encouraged to further develop and use these assays, and to accumulate more
data on their application to the control of rabies vaccines.
Because the intradermal route of administration has been used for some
rabies vaccines initially developed for intramuscular administration, some
additional considerations are discussed in a separate section of this document
(see Intradermal route of administration). Furthermore, guidance for
nonclinical and clinical testing of vaccines intended to be administered by
the intradermal route are provided in sections B and C of this document.
Given that new rabies vaccines might be developed in the near future, a section
on clinical evaluation of vaccines describes specific considerations for the
evaluation of data generated in clinical trials and provides some guidance on
how to assess the immunogenicity and safety of rabies vaccines.
The stability evaluation of vaccines is addressed in a separate section of
these recommendations. The importance of real-time studies under intended
storage conditions is emphasized.

86
 Relevant guidance documents published since the last revision of the
requirements for rabies vaccines have been considered in the present revision.
The following documents are mentioned as relevant and should be consulted
for further information. Updated requirements for the characterization of
continuous cell lines used for the preparation of biologicals, adopted in 1998
(15), provide current recommendations for vaccines produced using cell
substrates. It is important to note that the WHO recommendation for 10 ng
of residual host cell DNA per single human dose for products manufactured
using continuous cell lines remains the same as recommended in 1996. In
addition, guidance to reduce possible risk of contamination of vaccines by
transmissible spongiform encephalopathies (16) and updated requirements
for blood products (17) including human albumin, used in some vaccines as
a stabilizer, are now available.
In recent years concerns have been raised over the safety of thiomersal in
vaccines, especially those given to infants. These concerns have been based
primarily on data regarding the toxicity of a related substance, methyl mercury,
and from data on chronic exposure to mercury via the food chain. Such safety
concerns have led to initiatives in some countries to eliminate, reduce or
replace thiomersal in vaccines, both in monodose and multidose preparations.
It is important to note that concerns about the toxicity of thiomersal are
theoretical and there is currently no compelling scientific evidence of a safety
problem with its use in vaccines, although public perception of risk remains
in some countries. WHO policy is clear on this issue, and the Organization
continues to recommend the use of vaccines containing thiomersal for global
immunization programmes because the benefits of using such products far
outweigh any theoretical risk of toxicity (18).

Part A. Manufacturing recommendations

A.1. Definitions
A.1.1 International name and proper name
The international name should be rabies vaccine for human use. The proper
name should be the equivalent name in the language of the country of
origin.
The use of the international name should be limited to vaccines that
satisfy the requirements formulated below.

A.1.2 Descriptive definition

Rabies vaccine for human use is a freeze-dried or liquid preparation of well
characterized, laboratory adapted and attenuated virus with stable biological
characteristics, grown in cell substrates or embryonated eggs, and inactivated

87
 by a suitable method. The preparations for human use should satisfy all the
recommendations formulated below.

A.1.3 International standards

The fifth International Standard for Rabies Vaccines was established by
the WHO Expert Committee on Biological Standardization in 1991, with a
potency of 16 IU of Rabies Vaccine per ampoule. Recent research has indicated
that the glycoprotein and ribonucleoprotein components of inactivated rabies
vaccines play an important role in conferring protection. For this reason,
the Committee also assigned 10 IU of Rabies Virus PM-Glycoprotein and
135 IU of Rabies PM-Ribonucleoprotein to the contents of each ampoule of
the International Standard. It is recognized, however, that these components
might differ antigenically in the different virus strains used for vaccine
production; the International Standard may therefore be inappropriate for the
estimation of glycoprotein and ribonucleoprotein components of vaccines
not derived from the Pitman-Moore (PM) strain.
The Second International Standard for Rabies Immunoglobulin was
established by WHO in 1993. It is a preparation of human immunoglobulin
and each ampoule contains 30 IU.
The fifth International Standard for Rabies Vaccine and the first International
Standard for Rabies Immunoglobulin were initially held at the Statens
Serum Institute in Copenhagen, Denmark. Since 1997 this standard has
been in the custody of the National Institute for Biological Standards
and Control, Potters Bar, Hertfordshire, EN6 3QG, England (web site:
http://www.nibsc.ac.uk).
The international reference materials mentioned above are intended for the
calibration of national reference materials for use in the quality control of
rabies vaccines. They are distributed free of charge, on request, to national
control laboratories. The WHO catalogue of international biological
standards should be consulted for the latest list of appropriate international
standards and reference materials (http://www.who.int/biologicals/IBRP/
Catalogue.htm).

A.1.4 Terminology
The following definitions are given for the purpose of these Recommendations
only.
Adventitious agents: contaminating microorganisms including bacteria,
fungi, mycoplasmas, and endogenous and exogenous viruses that have been
unintentionally introduced.
Cell seed: a quantity of well-characterized cells of human or animal origin
stored frozen at –100 ºC or below in aliquots of uniform composition

88
derived from a single tissue or cell, one or more of which would be used for
the production of a master cell bank.
Final bulk: The material after completion of preparations for filling,
homogenous with respect to mixing of all components, and present in the
container from which the final containers are filled. The final bulk may be
prepared from one or more purified bulk materials.
Final lot: a collection of sealed final containers of freeze-dried or liquid
vaccine that is homogeneous with respect to the risk of contamination
during the filling process or the preparation of the finished vaccine. A final
lot must therefore have been filled or prepared in one working session from
a single final bulk.
Master cell bank: a quantity of fully characterized cells of human or animal
origin stored frozen in liquid nitrogen in aliquots of uniform composition
derived from the cell seed, one or more of which may be used for the
production of a manufacturer’s working cell bank.
Production cell culture: a collection of cell cultures that have been prepared
together from one or more containers from the working cell bank or in the
case of primary cell cultures, from the tissues of one or more animals.
Purified bulk material: a pool of inactivated and processed single harvests
before preparation of the final bulk. The pool may be prepared from one or
more single harvests and may yield one or more final bulks.
Single virus harvest: a virus suspension of the same virus working seed
lot inoculated, incubated and harvested together from either a group of
embryonated eggs or a cell culture in one production run. Multiple harvests
from the same production cell culture may be pooled and considered a
single virus harvest.
Virus master seed lot: a quantity of virus, physically homogeneous, that has
been prepared as a single lot. It is used for the preparation of working seed
lots.
Virus working seed lot: a seed lot prepared from the master seed lot with
no more than five passages removed from the master seed lot. Both passage
level and the method of passaging should be approved by the national
regulatory authority.
Working cell bank (WCB): a quantity of cells of uniform composition
derived from one or more ampoules of the master cell bank, which may
be used for the production cell culture. In normal practice, a cell bank
is expanded by serial subculture up to passage number (or population
doubling, as appropriate) selected by the manufacturer, at which point the
cells are combined to give a single pool and preserved cryogenically to

89
 form the WCB. One or more of the cryotubes from such a pool may be used
for the production of cell culture.

A.2 General manufacturing recommendations

The general requirements for manufacturing establishments contained
in Good manufacturing practices for pharmaceuticals (19) and Good
manufacturing practices for biological products (20) should apply to
establishments manufacturing inactivated rabies vaccine, with the addition
of the following:
Rabies has the highest case-fatality rate of any currently recognized
infectious disease; therefore the assignement of an appropriate biosafety
level for specific work with the virus at the production as well as at the
control facilities is an essential precautionary measure.
The assignment of a virus to a biosafety level for production and quality
control facilities must be based on a risk assessment. Such an assessment will
take the risk group, as well as other factors, into consideration in establishing
the appropriate biosafety level. For example, a virus assigned to risk group
2 may generally require Biosafety Level 2 facilities, equipment, practices
and procedures for safe conduct of work. However, if particular phases of
production require work with live virus and/or exposure to large quantities
and/or high titre of virus as well as exposure to aerosol, then Biosafety Level 3
may be more appropriate to provide the necessary degree of safety, because it
ensures superior containment in the production and quality control facilities.
The biosafety level assigned for the specific work is therefore a result of
professional judgement based on a risk assessment rather than an automatic
assignment of a laboratory biosafety level according to the particular risk
group designation of the pathogenic agent to be used. Further guidance on the
risk assessment and assignment of appropriate biosafety level are available in
the WHO laboratory biosafety manual (21). However, countries should draw
up their own national classification of microorganisms, by risk group.
Personnel employed in the production and control facilities should be healthy
and should receive regular medical examinations. They should be adequately
trained and protected against accidental infection with rabies virus. Steps
should be taken to ensure that all the personnel in the production and control
areas have been immunized against rabies and maintain an antibody titre of
at least 0.5 IU per ml of serum as measured by the rabies fluorescent focus
inhibition test (RFFIT) and fluorescent anti-virus neutralization (FAVN).
Periodic titre control and if required, booster injections are recommended
for people who are at continuous risk of rabies exposure. Further guidance
on the need for boosters is available in the Report of the Expert Consultation
on Rabies (7).

90
 The production of rabies vaccine should be conducted by dedicated staffs
who do not handle other infectious microorganisms, animals, or tissue
cultures in the same working day.
Steps should also be taken to minimize the risks of transmission of rabies
virus from the production facility to the outside environment.

A.3 Control of source materials

A.3.1 Substrates for virus production
Rabies vaccines may be produced in human diploid cells, in continuous cell
lines, in primary hamster kidney cells or in primary chick embryo fibroblast
cells. For human diploid and continuous cell lines section 3.1.1 should apply;
for primary hamster kidney cells section 3.1.2 should apply; for primary
chick embryo fibroblasts section 3.1.3 should apply; for embryonated duck
eggs section 3.1.4 should apply. Section 3.1.5 applies to all types of cell
substrates.

A.3.1.1 Diploid cells and continuous cell lines

The use of a diploid cell or continuous cell line for the manufacture of rabies
vaccines should be based on the cell seed system. The cell seed should
be approved by the national regulatory authority. The maximum number
of passages (or population doublings) by which the working cell bank is
derived from the cell seed should be approved by the national regulatory
authority.
WHO has established a cell bank of Vero cells characterized in
accordance with the requirements in the report of the WHO Expert
Committee on Biological Standardization (15), which is available as a well
characterized starting material to manufacturers for preparation of their
own master and working cell seeds on request to the Coordinator, Quality
Assurance and Safety of Biologicals, WHO, Geneva, Switzerland.

A.3.1.1.1 Identity test

Cell seed should be characterized according to the Requirements for
animal cell lines used as substrates for production of biologicals (15), as
appropriate to continuous cells or human diploid cells.
The testing performed on a replacement master cell bank (derived from
the same clone or from an existing master or working cell bank) should
be the same as for the establishment of the initial master cell bank,
unless a justified exception is made. The WCB should be identified
by means of, for example, biochemical (e.g. isoenzyme analysis),
immunological, and cytogenetic marker tests, approved by the national
regulatory authority.

91
A.3.1.2 Primary hamster kidney cells for production of rabies vaccines
A.3.1.2.1 Animals
Syrian hamsters 10–14 days old may be used as the source of kidneys for cell
culture. Only hamster stock approved by the national regulatory authority
should be used as the source of tissue and should be derived from a closed,
healthy colony. A closed colony is a group of animals sharing a common
environment and having their own caretakers who have no contact with other
animal colonies. The animals are tested according to a defined programme to
ensure freedom from specified pathogens, including the absence of antibodies
to these pathogens. When new animals are introduced into the colony, they
should be maintained in quarantine in vermin-proof quarters for a minimum
of two months and shown to be free from these specified pathogens. The
parents of animals to be used as a source of tissue should be maintained
in vermin-proof quarters. Neither parent hamsters nor their progeny should
previously have been used for experimental purposes, especially those
involving infectious agents. The colony should be monitored for zoonotic
viruses and markers for contamination at regular intervals.
At the time the colony is established, all founder animals should be tested to
determine freedom from antibodies to the following pathogens: microorganisms
pathogenic for hamsters (e.g. Mycobacterium tuberculosis, lymphoma virus,
papilloma virus, polyomavirus, adenoviruses and retroviruses), lymphocytic
choriomeningitis virus, pneumonia virus of mice, reovirus type-3, minute
virus of mice, Sendai, Hantaan virus, SV-5, Toolans H-a, mouse polio, mouse
hepatitis virus, and Kilham rat virus. Hamster antibody production (HAP),
mouse antibody production (MAP) and rat antibody production (RAP) tests
should also be performed. A test for retroviruses using a sensitive polymerase
chain reaction (PCR) based reverse transcriptase (Rtase) assay also should
be included. The results of such assays need to be interpreted with caution
because Rtase activity is not unique to retroviruses and may derive from
other sources, such as retrovirus-like elements that do not encode a complete
genome (20). Nucleic acid amplification tests for retrovirus may also be used.
A PCR test for hamster polyoma virus should be used on a selected number
of hamster tissues, especially kidneys, to qualify the colony, and at intervals
thereafter.
After the colony is established, it should be monitored by testing a
representative group of animals. The choice of tests and testing procedures
for monitoring as well as the appropriate number of animals should be
approved by the national regulatory authority. In addition, the colony should
be screened for the presence of pathogenic bacteria, including mycobacteria;
fungi and mycoplasma. This should be performed in all of the animals over
a defined period of time. The screening programme should be approved by
the national regulatory authority.

92
 Any animal that dies should be investigated to determine the cause of death.
If the presence of an infectious agent is demonstrated in the colony, the
national regulatory authority should be informed and the manufacture of
vaccine should be discontinued. In this case, manufacture should not be
resumed until a thorough investigation has been completed and precautions
have been taken against the infectious agent being present in the product,
and only then with the approval of the national regulatory authority.
At the time of kidney harvest, the animals should be examined for the presence of
any abnormalities and if kidney abnormalities or other evidence of pathology is
found, the affected animals are not to be used for production of rabies vaccine.
Each group of control cultures derived from a single group of animals used
to produce a single virus harvest should remain identifiable as such until all
testing, especially for adventitious agents, is completed.

A.3.1.2.2 Cell cultures for virus propagation

Kidneys derived from animals which comply with the guidelines set out in section
A.3.1.2.1. should be dissected and minced under conditions approved by the
national regulatory authority. A primary cell suspension is obtained after trypsin
digestion and this is distributed into cell culture vessels with growth medium.
Penicillin and other Beta-lactam antibiotics should not be used at any stage of
manufacture. Minimal concentrations of suitable antibiotics such as kanamycin
and neomycin may be used if approved by the national regulatory authority.

A.3.1.3 Chicken eggs used for primary chick embryo fibroblast preparation
If the vaccine is to be produced in primary chick embryo fibroblasts, the
eggs to be used should be from a closed, specific-pathogen-free flock. This
flock should be monitored at regular intervals for agents pathogenic to
birds. These include Mycobacterium avium, fowl pox virus, avian leukosis
virus (ALV) and other avian retroviruses, Newcastle disease virus and
other avian parainfluenza viruses, avian encephalomyelitis virus, infectious
laryngotracheitis virus, avian reticulo-endotheliosis virus, Marek’s disease
virus, infectious bursal disease virus, avian adenoviruses — group 1, avian
infectious bronchitis virus, avian nephritis virus, avian orthoreoviruses,
chicken anaemia virus, egg drop syndrome virus, influenza A virus, turkey
rhinotracheitis virus, Haemophilus paragallinarum, Salmonella enterica
Gallinarum and relevant Mycoplasma spp.
In some countries, all birds are bled when a colony is established, and
thereafter 5% of the birds are bled each month. The resulting serum samples
are screened for antibodies to the relevant pathogens. Any bird that dies
should be investigated to determine the cause of death.
The flock must not have been vaccinated.

93
A.3.1.4 Embryonated duck eggs
If the vaccine is to be produced in the embryonated eggs of ducks, the
eggs to be used should be from a closed, specific-pathogenic-free flock. The
flock should be regularly monitored for agents pathogenic to ducks. The
day old ducklings should come from an establishment or a hatchery where
duck virus enteritis, duck viral hepatitis, Salmonella enteritidis, Salmonella
anatum, Salmonella aertrycke, avian tuberculosis, psittacosis-orinthosis,
fowl cholera (pasteurellosis), egg drop syndrome, avian influenza (type
A) adenovirus group III (EDS), avian rotavirus, avian encephalomyelitis,
avian J virus, infectious serositus (new duck disease), coliform septicaemia,
spirochaetosis (duck tick fever), reticuloendotheliosis virus and Newcastle
disease have not been reported during the past 12 months.
In some countries, all ducks are bled when a colony is established,
and thereafter 5% of the ducks are bled each month. The resulting
serum samples are screened for antibodies to the relevant pathogens.
Any ducks that dies should be investigated to determine the cause of
death.
Live vaccine against avian influenza should never have been used in the
supply flocks. If any other vaccine is used the name and nature of the
vaccine, source of vaccine and date of vaccination should be reported.

A.3.1.5 Cell culture medium

Serum used for the propagation of cells should be tested to demonstrate
freedom from bacteria, fungi and mycoplasmas, according to the requirements
given in Part A, sections 5.2 and 5.3 of the revised Requirements for Biological
Substances No. 6 (23), and from infectious viruses. Suitable tests for detecting
viruses in bovine serum are given in Appendix 1 of the Recommendations for
Production and Control of Poliomyelitis Vaccine (Oral) (24).
Validated molecular tests for bovine viruses may replace the cell culture
tests of bovine sera. As an additional monitor of quality, sera may
be examined for freedom from phage and for an acceptable limit of
bacterial endotoxin.
Irradiation may be used to inactivate potential contaminant viruses.
The acceptability of the source(s) of any components of bovine, sheep
or goat origin used in culture media should be approved by the national
regulatory authority. These components should comply with current WHO
guidelines in relation to animal transmissible spongiform encephalopathies
(16).
Human serum should not be used. If human albumin is used it should meet
the revised Requirements for the collection, processing and quality control
of blood, blood components and plasma derivatives (Requirements for

94
 Biological Substances No. 27) (17), as well as current WHO guidelines in
relation to human transmissible encephalopathies (16).
The use of human albumin as a component of a cell culture medium
requires careful consideration due to potential problems with the validity
period of albumin in relation to the intended long-term storage of rabies
vaccines.
Penicillin and other beta-lactam antibiotics should not be used at any stage
of manufacture. Minimal concentrations of suitable antibiotics such as
kanamycin and neomycin may be used if approved by the national regulatory
authority.
Nontoxic pH indicators may be added, e.g. phenol red in a concentration
of 0.002%. Only substances that have been approved by the national
regulatory authority may be added.
If trypsin of animal origin is used for preparing cell cultures it should
be tested and found free of cultivable bacteria, fungi, mycoplasmas
and infectious viruses, especially bovine or porcine parvoviruses, as
appropriate. The methods used to ensure this should be approved by the
national regulatory authority.
The source(s) of trypsin of bovine origin, if used, should be approved by the
national regulatory authority. Bovine trypsin, if used, should comply with
current WHO guidelines in relation to animal transmissible spongiform
encephalopathies (16).

A.3.2 Virus seed

A.3.2.1 Strains of virus
The strains of rabies virus used in the production of all seed lots should
be well characterized, laboratory adapted and attenuated1 with stable
biological characteristics. The strains should be identified by historical
records including the information on its origin. They should have been
shown, to the satisfaction of the national regulatory authority, to yield safe
and immunogenic vaccines when inactivated.
Vaccine strains used for production of vaccines derived from cell substrates
and embryonated eggs known to induce protection in humans against rabies
include, but are not restricted to, the Pitman Moore virus, Pasteur Virus,
the Vnukovo –32, the Flury LEP, and the CTN. These are examples from

1
Previously used term “fixed” is based on the defined time in which clinical symptoms of the
disease appear in animals when inoculated intracerebrally (e.g. rabbits, mice and sheep). This
was obtained by serial passaging of the virus in rabbits. Although this characteristic reflects
attenuation of virus in an appropriate animal model it is not a guarantee of the suitability of a rabies
virus for production of vaccines for human use. The latter should be based on the immunogenicity
and safety of an inactivated virus in humans.

95
 some of currently licensed vaccines and should not be interpreted as a
recommendation.
The choice of virus, its full characterization and adaptation to the production
substrate should be justified in the overall evaluation of rabies vaccines for
licensing.
Vaccine strains should be characterized by molecular and serological
methods, including the use of monoclonal antibodies for the characterization
of rabies virus. This should also include animal inoculation. In addition,
sequencing of at least the glycoprotein and nucleoprotein genes of master
or working seed should be considered.
All master and working seed lots should comply with the current guidelines
to minimize the risks of transmission of animal transmissible spongiform
encephalopathies (16).

A.3.2.1.1 Virus seed lot system

Vaccine production should be based on the virus seed lot system. The
working virus seed lot should be not more than five passages removed from
the master virus seed lot, which should have been thoroughly characterized.
Vaccines should be made from the working seed lot without further
intervening passage. Virus seed lots should be maintained either in the dried
or in the frozen form and each lot should be stored separately. If frozen, the
seed lots should be kept continuously at a temperature below –60 °C.

A.3.2.1.2 Tests on virus seed lots

Seed lots should have been shown, to the satisfaction of the national
regulatory authority, to be capable of yielding vaccine that meets all the
manufacturing requirements listed here.
The virus master and working seed lots should be identified as rabies virus
by methods approved by the national regulatory authority.
Monoclonal antibodies which react specifically with rabies virus
nucleocapsid and glycoprotein may be used to identify the virus as
rabies.
All master and working seed lots should comply with the current guidelines
to minimize the risks of transmission of animal transmissible spongiform
encephalopathies (16).

A.3.2.1.2.1 Tests for bacteria, fungi and mycoplasmas

Each virus seed lot should be tested for bacterial, fungal, and mycoplasmal
contamination by appropriate tests according to Part A, section 5.2 of the
Requirements for biological substances no.6. General requirements for the
sterility of biological substances (23).

96
A.3.2.1.2.2. Tests for adventitious agents
The virus master or working seed lots used for the production of vaccine in cell
substrates or embryonated eggs should be free from detectable adventitious
agents. Seed lots produced in cell substrates should comply with the
recommendations in Part A. Section 3.1.1 applies to seed production in human
diploid or continuous cell lines; section 3.1.2 applies to primary hamster
kidney cells; and section 3.1.3 applies to seed lots produced using primary
chick embryo fibroblasts; section 3.1.4 applies to embryonated duck eggs.
For these tests the virus should first be neutralized by a specific anti-rabies
serum.
The individual tests on the seed virus should be designed to satisfy the
requirements of the national regulatory authority. The anti-rabies serum
should be free of known adventitious viruses.
Tests in suckling mice. A sample of the virus suspension should be tested
for the presence of adventitious agents pathogenic to mice by intracerebral
inoculation of 0.01 ml and intraperitoneal inoculation of at least 0.1 ml
into at least 10 suckling mice. The mice should be less than 24 hours old
and originate from more than one litter. They should be observed daily
for at least 14 days. All mice that die within the first 24 hours following
inoculation or that show signs of illness should be examined for evidence of
viral infection. This should be done macroscopically and by subinoculation
of appropriate tissue suspensions by the intracerebral and intraperitoneal
routes into at least five additional suckling mice, which should be observed
daily for 14 days.
In some countries, in addition, a blind passage is made of a suspension
of the pooled emulsified tissue (minus skin and viscera) of all mice
surviving the original 14-day test.
The virus seed passes the test if at least 80% of the mice originally inoculated
remain healthy and survive the observation period, and if none of the mice
shows evidence of infection with any adventitious agent attributable to the
virus seed.
Tests in adult mice. A sample of the virus suspension should be tested for
the presence of adventitious agents pathogenic to mice by intracerebral
inoculation of 0.03 ml, intraperitoneal inoculation of at least 0.25 ml,
and inoculation of 0.01 ml into the footpad in at least 20 adult mice, each
weighing 15–20 g. The mice should be observed for at least 4 weeks. All mice
that die within the first 24 hours of inoculation or that show signs of illness
should be examined for evidence of viral infection. This should be done
macroscopically by direct observation and by subinoculation of appropriate
tissue suspensions by the intracerebral and intraperitoneal routes into at least
five additional mice, which should be observed for 3 weeks.

97
 The virus seed passes the test if at least 80% of the inoculated animals
remain healthy and survive the observation period, and if none of the mice
shows evidence of infection with any adventitious agent attributable to the
virus seed.
Tests in cell cultures. The neutralized seed virus should be tested for freedom
from adventitious viruses in three sensitive cell culture systems:
— the cell line used for production;
— a different cell line from a different species; and
— human diploid cells.
Ten millilitres of the neutralized seed virus should be inoculated into each
cell system and the cells incubated at 35–37 °C for 14 days.
For virus seeds produced in human diploid cells, cell cultures should be
held for 28 days for the detection of cytomegalovirus.
The cells should be observed microscopically for cytopathic changes. At
the end of the observation period, the cells or fluids should be tested for
haemadsorbing viruses and other adventitious agents as specified in Part
A, sections 4.1.1.1. and 4.1.1.2. For the tests to be valid, at least 80% of the
culture vessels should be available and suitable for evaluation at the end of
the observation period. For the seed virus to be satisfactory, no cytopathic
changes or adventitious agents should be detected. Control cell cultures
should be included in the tests.

A.3.2.1.2.3. Additional tests if chick cell cultures are used for production of virus seed
If chicken cell cultures are used, a sample of fluids pooled from the control
cultures should be tested for avian retroviruses such as avian leukosis virus,
by a method approved by the national regulatory authority.
A test for retroviruses using a sensitive polymerase chain reaction
(PCR)-based reverse transcriptase (Rtase) assay may be used. The
results of such assays need to be interpreted with caution because
Rtase activity is not unique to retroviruses and may derive from other
sources, such as retrovirus-like elements that do not encode a complete
genome (22). Nucleic acid amplification tests for retrovirus may also be
used.

A.3.2.1.3 Virus content

A titration of the virus content of each seed lot should be made. Such
titrations may be done either in cell culture or by the inoculation of mice.
If mice are used, they should be inoculated by the intracerebral route with
0.03 ml quantities of suitable dilutions of the virus seed lot. Although the
previously recommended end-point for this in vivo titration was death of
the mice, it is reasonable instead to use clinical signs of paralysis as the

98
 end-point and to kill the animals when they reach this stage. Mice that show
no signs of rabies infection such as ruffled fur, slow and shaky movements
or paralysis should be observed for 14 days. The minimum titre of the seed
should be specified by the manufacturer, according to the product, cells and
virus strain.

A.3.2.1.2.4. Additional tests if duck embryos are used for production of virus seed
If duck embryos are used for the production of virus seed, tests for
Mycobacteria and avian viruses should be performed.

A.4. Control of vaccine production

A.4.1 Control (of) cell cultures
Penicillin and other beta-lactams should not be used at any stage of
manufacture.
Minimal concentrations of other suitable antibiotics, such as kanamycin
and neomycin, may be used where approved by the national regulatory
authority.
At least 5% or 500 ml, whichever is greater, of the cell suspension at the
concentration employed for seeding vaccine production cultures, should be
used to prepare control cultures.
If bioreactor technology is used, the national regulatory authority should
determine the size and treatment of the cell sample to be examined.

A.4.1.1 Tests of control cell cultures

The control cell cultures should be treated in a similar way to the production
cell cultures, but they should remain uninoculated so that they can be used
for the detection of extraneous viruses.
The control cell cultures should be incubated under the same conditions
as the inoculated cultures for two weeks or until the last harvest of virus
from the production cultures—whichever is the later—and should be
examined during this period for evidence of cytopathic changes. For
the test to be valid, not more than 20% of the control cell cultures
should have had to be discarded because of accidental contamination or
damage.
At the end of the observation period, the control cell cultures should be
examined for degeneration caused by infectious agents. If this examination,
or any of the tests specified in this section, shows evidence of the presence
in a control culture of any adventitious agent, the rabies virus grown in
the corresponding inoculated cultures shall not be used for vaccine
production.

99
A.4.1.1.1 Tests for haemadsorbing viruses
At the end of the observation periods, 25% of the control cells should
be tested for the presence of haemadsorbing viruses by using guinea-pig
erythrocytes. If the guinea-pig erythrocytes have been stored, the duration
of storage should not have exceeded 7 days and the temperature of storage
should have been in the range of 2–8 °C.
In tests for haemadsorbing viruses, calcium and magnesium ions should be
absent from the buffer medium.
In some countries the national regulatory authority requires that tests
for haemadsorbing viruses should also be done with erythrocytes from
other species, including human beings (blood group 0), monkeys, and
chickens (or other avian species).
The results of all tests should be noted after incubation of the
erythrocytes with the cultured cells for 30 minutes at 0–4 °C and again
after a further incubation for 30 minutes at 20–25 °C. For the test with
monkey erythrocytes, the results should be noted a third time, after a
final incubation for 30 minutes at 34–37 °C.

A.4.1.1.2 Tests for other adventitious agents in control cell fluids

At the end of the observation period a sample of the pooled fluids from each
group of control cultures should be tested for adventitious agents. Ten millilitres
of each pool should be tested in the same cells, but not the same batch of cells,
as those used for virus production, and additional 10 ml samples of each pool
should be tested in human cells and at least one other sensitive cell system.
The inoculated cultures should be incubated at 35–37 °C and should be
observed for at least 14 days.
For the tests to be valid, at least 80% of the culture vessels should be available
and suitable for evaluation at the end of the test period.
If any cytopathic changes due to adventitious agents occur in any of the
cultures, the virus harvest produced from the batches of cells from which
the control cells were taken should be discarded.

A.4.1.1.3 Identity test (cell line)

At the production level, and for vaccines produced in human diploid cells
or continuous cell lines only, the cells should be identified by using one of
the methods specified in current requirements for the use of animal cells as
in vitro substrates for production of biologicals (15). The method(s) should
be approved by the national regulatory authority.
Methods for identity testing include, but are not limited to, biochemical
(e.g. isoenzyme analysis), immunological (e.g. HLA assays), cytogenetic

100
 tests (e.g. for chromosomal markers), and tests for genetic markers
(DNA finger-printing).

A.4.1.1.4 Additional tests on control cells if avian embryo cells are used for production
A sample of the control fluid taken at the end of the observation period of
the control cell cultures should be tested for avian retroviruses such as avian
leukosis virus, by a method approved by the national regulatory authority.
In some countries the complement fixing test (COFAL) is used for
detecting avian leukosis viruses, and liver or kidney cell cultures of
embryos are used for detecting adenoviruses. A test for retroviruses
using a sensitive PCR-based RTase assay may be used. The results of
such assays need to be interpreted with caution because RTase activity
is not unique to retroviruses and may derive from other sources, such as
retrovirus-like elements that do not encode a complete genome (nucleic
acid amplification tests for retrovirus may also be used).
Only those cells shown to be free from contamination should be used.

A.4.1.1.5 Additional tests on control cells if other cell cultures are used
When other cell substrates are used for the growth of rabies virus, additional
appropriate tests should be considered.

A.4.2 Control of production in embryonated duck eggs

A.4.2.1 Control of (uninoculated) embryonated duck eggs
A sample of 2% of, but in any case not less than 20 and not more than 50,
uninoculated embryonated eggs from the batch used for vaccine production
should be incubated under the same conditions as the inoculated eggs. At
the time of virus harvest, the uninoculated eggs should be processed in
the same manner as the inoculated eggs, and the extract from the control
embryos should be shown to be free from haemagglutinating agents and
from adenoviruses, avian retroviruses such as avian leukosis virus, and other
extraneous agents by tests approved by the national regulatory authority.
A test for retroviruses using a sensitive PCR-based RTase assay may be used.
The results of such assays need to be interpreted with caution because RTase
activity is not unique to retroviruses and may derive from other sources,
such as retrovirus-like elements that do not encode a complete genome (22).
Nucleic acid amplification tests for retrovirus may also be used.

A.4.2.2 Single harvests from embryonated duck eggs

After the eggs have been incubated for a suitable period they should be
inoculated with seed virus. After further incubation for a suitable period, only
living, typical duck embryos should be harvested with aseptic precaution.

101
 The time of harvest should be defined starting from the inoculation of the
virus and should not be more than 14 days. Embryos inoculated at the
same time and harvested together may be pooled and the viral harvest kept
separate until completion of the sterility test (Part A, section 4.2.1.).

A.4.3 Control of single virus harvests and purified bulk material

After inoculation of the production cells with the virus working seed lot,
neither inoculated nor control cell cultures should at any time be at a
temperature outside the range approved by the national regulatory authority
for the defined incubation periods. The optimal range for pH, multiplicity of
infection, cell density and time of incubation should be established for each
manufacturer, and be approved by the national regulatory authority.
The appropriate time for harvest should be defined as number of days after virus
inoculation and should be approved by the national regulatory authority.
It is advisable that the inoculated cell cultures are processed in such a
manner that each virus suspension harvested remains identifiable as a single
harvest and is kept separate from other harvests until the results of all the
tests described in part A sections 4.1 and 4.2 have been obtained.
Only virus harvests satisfying the recommendations below should be pooled
and used in the preparation of the inactivated virus harvest.

A.4.3.1. Sterility tests on single virus harvests

A sample removed from each virus harvest should be tested for bacterial
and fungal contamination by appropriate methods, according to Part A,
section 5.2 of the Requirements for biological substances no.6. General
requirements for the sterility of biological substances (23). In addition, test
on mycoplasma contamination should be performed.
Any virus harvest in which contamination is detected must be discarded.

A.4.3.2 Identity
The single harvest contains virus that should be identified as rabies virus
using specific antibodies.

A.4.3.3 Virus titration for infectivity

Each virus harvest or pool of harvests should be tested for infectivity in a
sensitive assay. Both mice and cell culture of defined sensitivity are suitable
for testing infectivity. Manufacturers should set an in-house specification
for the titre of each harvest or pool of harvests.

A.4.3.4 Determination of antigen content

Assays for determination of glycoprotein antigen for determining the
antigen content of the final bulk have been shown to be suitable for

102
 monitoring consistency of production. Such assays include the single
radial immunodiffusion and enzyme immunoassay (EIA) test. Selection of
antibodies and other reagents is of critical importance.

A.4.3.5 Monitoring consistency of production

Virus titre, as well as determination of the antigen content mentioned above,
are appropriate parameters for monitoring consistency of production.
Therefore, internal specifications should be set.

A.4.3.6 Purification and/or concentration and of virus harvests

One or more single harvests may be purified and/or concentrated by
methods demonstrated to yield safe, potent and immunogenic vaccine. The
virus harvest should be inactivated by a validated method at a defined stage
of the process which may be before, during or after any concentration and
purification.
The process should be approved by the national regulatory authority and
should be shown to give consistent results.

A.4.3.7 Test for residual cellular DNA

For viruses grown in continuous cell lines, purified bulk should be tested
for residual cellular DNA. The purification process should be shown to
consistently reduce the amount of host cell DNA. As recommended in
the Requirements for the use of animal cells as in vitro substrates for the
production of biologicals (15), the amount of residual cell DNA should be
less than 10 ng per purified human dose. The assay for determination of
residual cell DNA with defined sensitivity for detection of specified levels
should be approved by the national regulatory authority. The specification set
for the level of residual DNA should comply with current WHO guidelines
for cell substrates.

A.4.3.8 Test for residual animal serum

If animal serum is used during production, the concentration of bovine
serum albumin (BSA) may be used as an indicator of animal serum in the
purified bulk which should not be greater than 50 ng per human dose or its
equivalent.
In some countries, tests are carried out to estimate the amount of
residual animal serum in the final vaccine.

A.4.3.9 Tests for residual materials

Each manufacturer should demonstrate, by testing each virus purified bulk
or by validation of the manufacturing process, that any residual materials
used in manufacture are consistently reduced to a level acceptable to the
national regulatory authority.

103
A.4.4 Inactivation procedure
A.4.4.1 Methods and agents
The methods and agents used for inactivation should be validated and
approved by the national regulatory authority. Given that presence of virus
aggregates may result in ineffective inactivation, great care should be taken
to avoid this. If it is not possible, virus aggregates should be removed before
starting the inactivation procedure.
Chemical or physical means, such as filtration, may be used to remove
aggregates.
If clarification is performed on a crude virus suspension, it is advisable
to start inactivation within 24 hours.
In the case of vaccine produced in embryonated eggs the method should
also be shown to inactivate avian leukosis viruses, as demonstrated by tests
in tissue culture, or, in the case of vaccine produced in the primary hamster
kidney cells, any adventitious agent that may be present, as demonstrated
by tests using tissue culture or by animal inoculation.
The method used should be shown to be consistently effective in the hands
of the manufacturer. The kinetics of inactivation should be demonstrated by
the manufacturer to be consistently effective using at least five consecutive
batches. The total inactivation time to complete inactivation should be
determined. The inactivation time used routinely should be at least double
the period required to inactivate the virus completely.
As a part of validation of the inactivation process, virus samples taken
at appropriate times, should be inoculated immediately into the sensitive
substrate (e.g. mice, cell cultures), to determine the inactivation curve. This
provides information on the reproducibility of the inactivation process.
Various methods for inactivating rabies virus have been used with
success. The concentration of the inactivating chemical, the temperature,
and the length of time necessary for inactivation must be defined by each
manufacturer for a particular production process. Satisfactory vaccines
may be prepared by treating virus suspensions from tissue culture at
2–8 °C with beta-propiolactone at a dilution of 1:3500 to 1:5000 for
24 h, or until inactivation is complete, as demonstrated by the results of
the test for effective inactivation specified below.
The conditions for storage of concentrated bulk intermediates should to
be validated and approved by the national regulatory authority.
Formaldehyde may also be used as an inactivating agent in the
production of rabies vaccines. Tests for free formaldehyde should be
performed at appropriate intervals and the concentration maintained

104
 at specified levels. Ultraviolet irradiation may be used to facilitate the
inactivation process.

A.4.4.2 Tests for effective inactivation

Each purified bulk material should be tested in an appropriate test system
for effective inactivation of the virus before the addition of preservatives
and other substances. The sensitivity of the assay should be determined
according to the rabies virus used for production and the most sensitive
assay should be used. Tests should be performed immediately after
inactivation.
If samples are not tested immediately after inactivation they should
be stored frozen at < –60°C. The conditions of storage should be
validated to confirm that there is no loss of virus titre. If test is performed
at a later stage of production, appropriate biosafety levels should be
maintained.
The rabies virus amplification test, for testing for the presence of live virus,
should be performed in the cell culture used for vaccine production or a type
of cell line demonstrated to be of greater sensitivity. The national regulatory
authority should approve the cell line and the method used. Manufacturers
are encouraged to use cells such as Vero cells, BHK-21 and neuroblastoma
cells which are known to be highly sensitive to rabies virus.
The rabies virus amplification test should be done as follows. At least
25 ml of bulk vaccine corresponding to at least 25 human doses
should be inoculated on five cell cultures of the type used for vaccine
production, or a type of at least equal sensitivity. At least 3 cm2 of cell
sheet should be used per millilitre of vaccine. After adsorption of the
inoculum for an appropriate time, medium should be added such that the
ratio of medium to vaccine is not more than 1:3. The cultures should be
observed for at least 21 days. The cell cultures may be stained directly
for the presence of rabies virus by immunofluorescence. Otherwise,
5 millilitres of each culture fluid should be pooled on days 14 and 21
and 0.03 ml of this pool should be inoculated intracerebrally into each
of 20 weanling mice of 12–15 g. These mice should be observed for
14 days. Any symptoms caused by rabies virus should be confirmed by
the immunofluorescence assay. At the end of the observation period, no
cytopathic effects should be detected.
A test involving the use of immunofluorescence for the detection of cells
infected with rabies virus at day 21, which is shown to be as sensitive
as mice inoculation and approved by the national regulatory authority
may be used. A specification for the proportion of cells checked by
immunofluorescence should be set and approved by the national
regulatory authority. The inclusion of trypsinisation of the cells on day

105
 7 should be considered as this may increase the sensitivity of the
amplification assay.
For certain products such as purified duck embryo vaccine (PDEV) the
virus may not be adapted to growth in cell culture. Virus amplification
test may, therefore, be performed in the same substrate as the one used
for production. The virus is inoculated in yolk sac in pre-incubated eggs.
Absence of virus can be confirmed by the fluorescent antibody test.
The bulk material passes the test if the product is shown, to the satisfaction
of the national regulatory authority, to be free from residual live virus.

A.4.5 Preparation and control of the final bulk

A.4.5.1 Preservatives and other substances added
In preparing the final bulk, only adjuvant, preservatives and other substances
such as human albumin approved by the national regulatory authority should
be added. Such substances should have been shown by appropriate tests not to
impair the safety or effectiveness of the product at the concentration used.
If beta-propiolactone has been used for inactivation, the procedure should
be such that this chemical is not detectable in the final bulk. The test used
for determination of beta — propiolactone should be of defined sensitivity,
performed at the time intervals appropriate for the kinetics of inactivation
for the vaccine in question. The test should be approved by the national
regulatory authority.
No antibiotics should be added to rabies vaccine for human use after the
virus has been harvested.

A.4.5.2 Antigen content of the final bulk

Assays for determination of glycoprotein antigen of the final bulk have
been shown to be suitable for monitoring consistency of production. Such
assays include the single radial immunodiffusion and EIA test. Selection of
antibodies and other reagents is of critical importance.
Since the presence of adjuvant may affect results, it is advisable to
perform the assay before the adjuvant is added. Alternatively, antigen
may be eluted from the adjuvant prior to assay.
Some manufacturers test glycoprotein content of the purified, concentrated
bulk to determine the dilution of the bulk to be used in the preparation
of the final bulk.

A.4.5.3 Sterility tests

Each final bulk should be tested for sterility according to Part A, section
5.2 of the revised Requirements for biological substances no. 6. General
requirements for the sterility of biological substances (23).

106
A.5. Filling and containers
The requirements concerning filling and containers given in Good
Manufacturing Practices for Biological Products (20) should apply to
vaccine filled in the final form.
Care should be taken to ensure that the material of which the container
is made does not adversely affect the virus content of the vaccine under
the recommended storage conditions.

A.6. Control tests on final product

Samples should be taken from each filling lot for the tests described in the
following sections.

A.6.1 Identity test

An identity test should be performed on at least one labelled container from
each filling lot by an appropriate method.
The test for potency described in Part A, section 6.5 may serve as an
identity test.

A.6.2 Sterility test

Each filling lot should be tested for bacterial and fungal sterility according
to Part A, section 5.2 of the revised Requirements for biological substances
no. 6. General requirements for the sterility of biological substances (23).

A.6.3 General safety (innocuity) test

Each final lot should be tested for unexpected toxicity (sometimes called
abnormal toxicity) using a general safety (innocuity) test approved by the
national regulatory authority.
This test may be omitted for routine lot release once consistency of
production has been well established to the satisfaction of the national
regulatory authority and when good manufacturing practices are in
place. Each lot, if tested, should pass a test for general safety.

A.6.4 Antigen content

If not done on the final bulk (4.3.2), antigen content should be determined
and be within limits approved by the national regulatory authority.

A.6.5 Potency test of vaccine in final containers

The potency of each final lot should be determined. Before being tested,
dried vaccine should be reconstituted to the form in which it is to be used
in humans.

107
 The NIH test as described in Laboratory techniques in rabies (10) should be
used to evaluate consistency of production of the vaccine in question. This
should also be used to test product stability for the purpose of establishing
shelf life as well as to calibrate reference preparations.
In this test mice are immunized and subsequently challenged with rabies
virus. The test is conducted by vaccinating groups of mice, on two occasions,
7 days apart, with dilutions of an appropriate reference material calibrated
against the International Standard for Rabies Vaccine the and vaccine being
tested. Seven days after the last vaccination, the immunized animals and
a control group of mice are challenged intracerebrally with the > 5 LD50
of challenge virus standard (CVS). The titre of the challenge virus should
be confirmed by inoculation of at least three tenfold dilutions into further
groups of mice. The mice are observed for 14 days and the 50% effective
dose (ED50) of the reference and test vaccine is determined on the basis of
survival rate of mice. Humane endpoints may also be used if validated.
The potency of the test vaccine in IU should be determined by comparing
the ED50 of the test vaccine with that of the reference vaccine calibrated in
IU by comparison with the International Standard for Rabies Vaccine using
appropriate statistical methods.
The assay with defined criteria for validity and test procedure, the method
for statistical calculation together with the minimum number of assays to
be performed for adequate interpretation of the results and the confidence
limits of the assay should be approved by the national regulatory authority.
In particular, calibration of reference vaccine against the International
Standard as well as use, storage and handling of CVS should be well defined
in the approved test procedure.
The confidence limits of the assay should be in the range of 25–400%.
The number of tests to be undertaken on each batch is dependent on the
consistency of assays in an individual laboratory. If consistency of testing
in a laboratory is well demonstrated, one assay may be sufficient. However,
additional assays may increase the precision of the potency estimate.
The potency should be at least 2.5 IU per single human dose.
In some countries, more than one assay is performed. In this case, the
estimated geometric mean potency is based on two valid tests and
should be at least 2.5 IU per human dose.
In some countries, a modified NIH test is in use. Following licensing,
and once consistency in production and quality control of the vaccine
has been further confirmed on a continuous basis over at least two
years, the determination of potency in routine lot release may, with the
approval of the national regulatory authority, be based on the modified

108
NIH assay, based on a single dilution. This assay will provide qualitative
(or semi-quantitative) results.
Several prerequisites and conditions should be considered before
designing a single dilution assay:
• The one dilution assay is advantageous when vaccine lots
consistently give a lower limit for the estimated potency well in
excess of 2.5 IU per single human dose. This is more likely to
be consistently achieved where the antigen content of the final
container vaccine is based on the assay of rabies glycoprotein
content.
• This assay is suitable for testing a large number of batches each
year, in particular for testing several batches at the same time.
• Consistency of testing results is essential.
• Several factors such as virus strain, the homogeneity of the CVS
challenge preparation and the strain and quality of mice may affect
reproducibility of the results of tests.
The following criteria for validity of a single dilution assay should be
taken into account:
• The full dilution assay should be well established with high
percentage of valid results for defined period of time.
• Reference vaccine and CVS should have a good record of values
within the specified range for the laboratory in question.
• Sub-potent vaccines should be included in the validation.
• Acceptance and rejection criteria must be defined.
To further confirm consistency on a continuous basis, the potency of
about 10 recent batches of vaccine should be tested using the full
dilution assay. If potency expressed in IU is within the specified range of
values and if the expectations of linearity and parallelism are consistently
satisfied, then fewer doses may be used and the assumptions of linearity
and parallelism need not be tested in each assay.
A one-dilution assay is based on the same principles for evaluating the
response as the three-dilution assays. The assay involves the selection
of a dose of the reference vaccine, expressed as a fraction of 2.5 IU (i.e.,
of the minimum potency of a single human dose), that elicits a minimum
protective effect in mice, and comparing its effect with the response elicited
by the same fraction of a human dose of the test vaccine. If the response to
the test vaccine is significantly greater than the response to the reference
vaccine (P ≤ 0.05), the potency of the test vaccine is satisfactory.

109
 One dilution assays provide assurance that the lower limit of the estimated
potency is in excess of the minimum requirement. A disadvantage
of such an approach is that strictly quantitative estimates of vaccine
potency will not be possible.
Lot release based on a simplified approach will require periodic review
to ensure that the validity of all procedures is maintained. The timing of
the review should be decided on a case by case basis depending on
the number of batches of vaccine produced annually and/or performed
regularly (at least every 2 years), as agreed by the national regulatory
authority.
If a batch of vaccine fails to meet the specification set for the modified
test, a full NIH mouse protection test should be performed.
Manufacturers are encouraged to support data generated by NIH potency
assay by the determination of antigen content using an in vitro assay in
order to ensure overall consistency of production.
The design of the test as well as statistical analysis of the data should be
approved by the national regulatory authority.
A suitable challenge strain, CVS-11, is available upon request from the
World Health Organization, Geneva, Switzerland. Such a request should be
approved by the relevant national regulatory authority.

A.6.6 Ovalbumin content

For vaccines produced in embryonated eggs only, the ovalbumin content of
each filling lot, if not done on the final bulk, should be determined and be
within limits approved by the national regulatory authority.

A.6.7 Residual moisture test on freeze-dried vaccine

The residual moisture in a representative sample of each freeze-dried
lot may be determined by a method approved by the national regulatory
authority. The upper limit of moisture content should be specified by the
national regulatory authority. Generally, moisture levels of less than 3% are
considered satisfactory.

A.6.8 Test for pyrogenic substances

Each final lot should be tested for pyrogenic substances. The test should be
approved by the national regulatory authority.

A.6.9 Test for residual animal serum protein

A sample of the final lot should be tested to verify that the level of bovine
serum albumin in the final reconstituted vaccine is less than 50 ng per
human dose.

110
A.6.10 Adjuvant
If an adjuvant has been added to the vaccine, its content should be determined
by a method approved by the national regulatory authority.
Where aluminium compounds are used the concentration of aluminium
should not be greater than 1.25 mg per single human dose. If other substances
are used as adjuvant or those with adjuvanted effect, a specification should
be set.
When aluminium hydroxide is used as the adjuvant, the degree of adsorption
should be determined in the final bulk. This should not be less than 95%.

A.6.11 Preservative
If a preservative has been added to the vaccine, the content of preservative
should be determined by a method approved by the national regulatory
authority.
The amount of preservative in the vaccine dose should be shown neither
to have any deleterious effect on the antigen nor to impair the safety of
the product in humans. The preservative, its use at different stages of the
manufacturing process as well as the residual amount present in the product
should be approved by the national regulatory authority.
If any modification of preservative content in an already licensed vaccine
is made, general principles for vaccine evaluation described in the WHO
Guidelines on regulatory expectations related to the elimination, reduction
or replacement of thiomersal in vaccines, should be followed (25).

A.6.12 Inspection of final containers

Each container in each filling lot should be inspected, and those showing
abnormalities should be discarded.

A.6.13 Test for residual cellular DNA

For viruses grown in continuous cell lines, the final product should be tested
for residual cellular DNA if this test has not been carried out at final bulk
stage (section A.4.3.7).

A.7. Records
The recommendations given in Good manufacturing practices for biological
products (20) should apply.

A.8. Samples
Vaccine samples should be retained as recommended in Good Manufacturing
Practices for biological products (20, Annex 1).

111
A.9. Labelling
The recommendations given in Good manufacturing practices for biological
products (20) should apply, with the addition of the following.
The label on the container or package should include the following information:
— the designation of the strain of rabies virus contained in the vaccine;
— the minimum potency of vaccine determined by NIH test and expressed
in IU per human dose;
— the substrate used for the preparation of the vaccine;
— the method used for inactivating the virus;
— the nature and amount of stabilizer, preservative or additive present in
the vaccine;
— the volume and nature of diluent;
— the use of vaccines after reconstitution if the vaccine is in the dried form;
— the expiry date should be indicated on both the primary and secondary
packaging.
It is desirable that the label carry the names both of producer and of the
source of the bulk material, if the producer of the final vaccine did not
prepare it. The nature and residual amount of the antibiotics present in
the vaccine, if any, may be included.

A.10. Distribution and shipping

The recommendations given in Good manufacturing practices for biological
products (20) should apply.

A.11. Stability, storage and expiry date

A.11.1 Stability
Stability evaluation is an important part of the quality assessment. The
purpose of stability studies is to ensure that the vaccine at the end of its shelf
life, storage period or period of use, still has the required characteristics
supporting quality, safety and efficacy.

A.11.1.1 Stability for licensing

Studies that support stability of a vaccine for the purpose of licensing have to be
performed as real time real condition studies. Stability-indicating parameters
should be carefully selected. They should always include, but should not
be limited to, the potency test as part of real time studies under conditions
recommended for storage. Tests should be conducted to determine the loss of
potency at appropriate time intervals during storage. Final containers from at
least three batches of vaccine derived from different bulks should be tested
on the expiry date to demonstrate stability during storage.

112
 Data from accelerated stability testing for a product stored for limited
periods at temperatures that may affect stability could support preliminary
data from ongoing real time stability studies but should not replace them.
However, further data on stability to support shelf life of the product
should be based on long-term stability studies under real conditions and
should be submitted to the national regulatory authorities for approval.
Any modification of the shelf life approved as part of licensing requires
additional stability data to support the proposed modification and should be
approved by the relevant national regulatory authority. Following licensure,
stability should be monitored throughout the proposed shelf-life.

A.11.1.2 Stability for lot release

There is no additional value in performing an accelerated stability test for
the purpose of lot release.

A.11.1.3 Stability at different stages of manufacturing process

Stability testing should be performed at different stages of production, namely
on single harvests, final bulk and final lot. Stability indicating parameters
should be selected according to the stage of production. It is advisable to
assign a shelf-life to all materials during vaccine production, in particular
intermediates such as single harvests, purified bulk and final bulk.

A.11.1.4 Stability for clinical trial approval

For vaccines under development, stability data, such as those described under
11.1.1, are expected for the purpose of clinical trial approval. However,
stability data generated for a more limited period are acceptable at this stage.
Appropriate documentation to support the stability profile of a vaccine
should be submitted to the competent national regulatory authority at all
stages mentioned above.

A.11.2 Storage conditions

Recommended storage conditions and the defined maximum duration of storage
should be based on stability studies as described above and approved by the
national regulatory authority. For rabies vaccines, both liquid and freeze-dried,
a temperature of 2–8 °C has been found satisfactory. This should ensure that
the minimum potency specified on the label of the container or package will be
maintained after release until the end of the shelf-life, if the conditions under
which the vaccine is stored are in accordance with those stated on the label.

A.11.3 Expiry date

The expiry date should be defined on the basis of shelf life supported by
the stability studies as described above (section A.11) and approved by the
national regulatory authority.

113
A.12. Intradermal route of administration
Vaccines to be administered by the intradermal route should meet the
same quality, safety and efficacy specifications as defined in the WHO
recommendations for production and control for vaccines for intramuscular
use. This means that the potency of such vaccines, if reconstituted in the
volume intended for intramuscular use, should be at least 2.5 IU per single
dose. In addition, manufacturers should provide clinical evidence that the
vaccine is immunogenic and safe when administered intradermally.
In some countries a volume of 0.1ml per intradermal site has been found
appropriate due to practical aspects of vaccine administration.
Ideally, vaccines intended to be administered by intradermal route should
be developed for this purpose. This includes appropriate studies in which
the immunogenicity and safety of vaccines are demonstrated by testing
vaccine, the potency of which is assigned for an intradermal dose.
For vaccines originally developed for intramuscular administration,
intradermal use should be supported by nonclinical and clinical data
(see sections B and C). In addition, potency should be assigned for the
intradermal dose.
Rabies vaccines formulated with an adjuvant should not be administered
intradermally.
Intradermal injections must be administered by staff trained in this technique.
Further details on immunization regimens and practices to be followed
when a vaccine is to be administered by the intradermal route are available
in the report of the WHO Expert Consultation on Rabies (7).

Part B. Nonclinical evaluation of new rabies

vaccines
Preclinical testing is a prerequisite for the initiation of clinical trials in
humans and includes immunogenicity studies (proof of concept) and safety
testing in animals. The vaccine lots used in preclinical and nonclinical
studies should be adequately representative of the formulation intended for
use in the clinical investigation and, ideally, preclinical testing should be
done on the same lots as proposed for the clinical trials. If this is not feasible,
then these lots should be comparable with respect to potency, stability, and
other characteristics of quality. Details of the design, conduct, analysis
and evaluation of nonclinical data are available in the WHO Guidelines for
nonclinical evaluation of vaccines (26).
If a new rabies vaccine is intended to be used intradermally, the issue of
appropriate formulation for this purpose should be addressed early in its

114
 development. Dose–response and minimum potency for induction of a protective
immune response in a relevant animal model should be demonstrated.

Part C. Clinical evaluation of rabies vaccines

C.1 Evaluation of new rabies vaccines for licensing
C.1.1 General considerations for the clinical assessment of rabies vaccines
The clinical development programme for rabies vaccines should evaluate
their use for pre-exposure and post-exposure prophylaxis including different
vaccination schedules and routes of administration, the onset, extent and
duration of protection, and the need for and timing of booster vaccination.
Clinical trials should adhere to the principles described in the Guidelines for
good clinical practice (27) as well as to those formulated for design, conduct
and analysis of vaccine clinical trials described in the WHO Guidelines for
clinical evaluation of vaccines (28). All clinical trials should be approved by
the relevant national regulatory authority.
For ethical reasons, it is impossible to conduct placebo controlled clinical
efficacy studies involving an unvaccinated group that might be or has been
exposed to rabies infection. Efficacy has been demonstrated previously in
well designed but uncontrolled studies for tissue culture- and avian-derived
vaccines in individuals exposed to rabies infection. Long-term evidence has
shown that vaccines that met the minimum WHO potency requirement of
2.5 IU per dose induce adequate immunogenicity and protection. An antibody
concentration of at least 0.5 IU per ml on days 14 and 28 or 30, after initial
vaccination, is generally considered to be adequate. Therefore a satisfactory
concentration of neutralizing antibodies could be used as a predictor of clinical
efficacy. Nevertheless every effort should be made to obtain information on
the protective capacity of vaccines during their actual use.
The clinical development programme must be tailored to the type of
vaccine, taking into consideration the particular formulation, including any
adjuvant content, the production process and intended use of a vaccine. For
example, vaccines containing antigens, adjuvants or other components with
which there is little or no previous experience in humans will require more
extensive clinical study than those that closely resemble licensed vaccines.

C.1.2 Characterization of vaccine lots for clinical trials

By the beginning of the later stages of clinical development, a vaccine
should have been fully characterized in terms of its physicochemical and
biological properties and the final manufacturing process. Final release and
end of shelf-life specifications and batch release testing procedures may not
have been established at this stage since they may partly depend on the total
clinical data.

115
 Consistency of manufacturing for the vaccine lots used in clinical trials
should be demonstrated and well documented. These lots should be
adequately representative of the formulation intended for marketing. At a
minimum, candidate vaccines for clinical trials should be prepared under
conditions of good manufacturing practice for clinical trial material (29).
Full GMP will be required at the later stages of clinical development (19,
20). All the lots to be used for clinical trials should be released by national
control laboratories.
Any change in the manufacturing process during vaccine development
should be assessed carefully regarding any possible impact on the safety,
immunogenicity and likely efficacy of the vaccine and the need for additional
nonclinical and clinical investigations (see section 1.7 below).
Similarly if the vaccine has been the subject of a transfer of production
clinical data may be required to assess its safety, immunogenicity and likely
protective efficacy

C.1.3 Immunogenicity studies

Dose-finding studies to identify appropriate regimens for induction of
protective immune responses should be performed.
Initial immunogenicity studies should be performed in healthy adult
volunteers who have not been exposed to rabies and have not been previously
vaccinated. After the vaccine has been proven to be immunogenic in
rabies naïve healthy adults, further studies should be conducted to
demonstrate immunogenicity in target populations according to the intended
use:
Pre-exposure prophylaxis—persons resident in endemic areas should be
enrolled in the trial. The population should include the elderly and persons
with different vaccine histories to establish the suitability of the vaccine for
naïve and previously immunized persons. If the vaccine is to be licensed for
pre-exposure prophylaxis of children then adequate data should be obtained
in various age groups.
Post-exposure prophylaxis—these studies can only be done in high risk
areas and populations and should follow production of immunogenicity
data as for pre-exposure prophylaxis. Subjects known or thought likely to
have been exposed to rabies infection should be vaccinated and followed
up to test for immunogenicity and efficacy. For the initial evaluation of
new vaccines, schedules recommended by WHO should be used (7). For
post-exposure prophylaxis regimens, the following schedules for antibody
testing are recommended as a minimum: days 0, 14, 28 or 30, 90, 180, 360.
It is imperative to include a blood sample taken on day 0 and 7 in order to
identify and exclude previously vaccinated subjects.

116
 For new rabies vaccines intended to be administered intradermally, the
suitability of the formulation for this purpose should be tested in the target
population (see section 1.8).

C.1.4 Assessment of the immune response

When assessing pre-exposure use, immunogenicity should be determined in
terms of the time to onset, antibody titres and duration of likely protection.
Variability of the immune response between subjects is an important element
and should be reported wherever possible.
The ability of the vaccine to induce an immunological memory and
consequently an anamnestic response after booster doses should be tested.
Data generated in one study using a specific priming regimen should not be
extrapolated to different regimens or routes of administration.
It would be beneficial to define the highest potency, determined by the
NIH test and expressed in IU per dose, that induces further increases in
antibody levels (the so called “saturation point”).
The appropriate time intervals for taking the samples should be defined
taking into account study objectives. Serum samples should be divided into
aliquots and stored securely so that they can be made available in the event
that re-evaluation is required.
Immunogenicity should be assessed using one of the two serological assays:
rapid fluorescent focus inhibition test (RFFIT) or fluorescent antibody virus
neutralization (FAVN). The assay used to determine levels of neutralizing
antibodies should be approved by the national regulatory authority. It has been
demonstrated that the degree of homology between the strain of challenge virus
used in the RFFIT to measure the immune response after vaccination and the
strain of seed virus used for vaccine production profoundly affects reported rabies
virus neutralizing antibodies (RVNA) values (30). The use of a heterologous
challenge virus strain (CVS) may result in lower levels of neutralizing
antibodies than those obtained with homologous CVS in the same assay.

C.1.5 Quality assurance of immunogenicity testing

It is of critical importance to perform immunogenicity testing in a laboratory
with well established RFFIT/FAVN testing protocols and that implements
quality assurance of testing procedures. For this purpose, the following
should be in place:
• written (standard operating procedure (SOP)
• temperature monitoring and control on relevant equipment
• all equipment should be calibrated and evaluated annually. This should
include all pipettes, CO2 incubators, water baths, refrigerators/freezers,
hoods and plate washers

117
 • cells should be monitored to determine that they are free of mycoplasma
contamination
• passage of cells should be documented
• pre-batch acceptance of all reagents to verify fitness for purpose, e.g.
fetal calf serum and fluorescein isothiocyanate (FITC)-labelled anti-
rabies antibody.
• virus
— virus strain and passage history should be documented
— virus should be passaged to avoid production of defective interfering
particles
— titre established for stock virus
— establish working dilution
• cells
— passage history
— free from contamination
— maintain same growth curve when utilized
• serum
— store refrigerated or frozen
— clearly labelled and destroyed adequately after usage
• assay
— validity criteria of assay:
• all test criteria should be documented for every test
• back titration to ensure working virus titre is within specification
• good cell sheet is required, reasonable confluence
• a negative serum and a reference serum calibrated in IU must be
included in all assays
• end-point dilution of reference needs to be monitored
• analysis of serum titres should be conducted according to
documented procedures
• training and proficiency
— all technicians need to be appropriately trained and competence
demonstrated
— participation in proficiency testing schemes (PTS) is recommended.

C.1.6 Analysis and interpretation of the data

Collection, recording, analysis and interpretation of data should be conducted
according to GCP guidelines. Methodological and statistical considerations
described in WHO guidelines should be taken into account (28).
Data generated in clinical trials should be submitted to the national
regulatory authority as described in the Summary protocol for vaccine
evaluation (28). Data should be presented stating the batch number, vaccine
presentation, potency of the vaccine used, nature and volume of a diluent
where appropriate, and other relevant characteristics of tested vaccine. In

118
 addition to general statements (e.g. regarding study sites, investigators,
objectives, inclusion and exclusion criteria), a number of issues specific to
rabies should be described including:
— nature of exposure (WHO category 1-3);
— status and confirmation of rabid animal;
— nature of wound care;
— immunization schedule;
— details of postexposure prophylaxis (PEP) treatment including time and
sites of administration of rabies immune globulin, nature, volume and
other details of rabies immune globulin as a product;
— care and treatment of adverse side-effects;
— other medications given;
— control group using well-established rabies vaccine.
Immunogenicity data should include the total number and percentage
of subjects in whom titres are above and less than 0.5 IU/ml, GMT
with confidence interval, range of antibody titers. Safety data should be
presented as total number, percentage and type of adverse events. It would
be beneficial for the whole scientific community to publish clinical trial
data in a peer-reviewed journal.

C.1.7 Studies to support change in manufacturing processes

Changes in production methods or scale-up before or following licensing will
necessitate further product characterization to demonstrate comparability
with the lots used in earlier studies of safety and immunogenicity. Changes
that do not require clinical data should be defined in the national regulations
and the national regulatory authority consulted regarding all changes prior
to their implementation. However, some changes could affect the safety and
immunogenicity of the vaccine and therefore it is likely efficacy and such
changes should be supported by additional clinical data. The extent of the
clinical data needed depends on the nature and extent of the changes made.
The design of such studies rests on the demonstration of non-inferiority in
terms of eliciting protective immune responses and safety (28).

C.1.8 Studies to support a new route of administration

Clinical data cannot be extrapolated between routes of administration and
clinical trials of administration by the proposed new route should be performed.
Dose finding studies should be conducted to determine the optimal intradermal
dose and volume of administrations. Immunization schedules recommended
by WHO are described elsewhere (7).
For vaccines already licensed for the intramuscular route of administration,
the following issues should be carefully considered before proposing such
a formulation for intradermal use:

119
 • Presence of preservative in a vaccine: consideration should be given to
the type of preservative and the residual amount in the final vaccine.
• Potential impact of the use of opened multi-dose vials in the field: the
level of compliance with good immunization practice in a particular area/
country and the actual risk of contamination should be taken into account
by the national regulatory authority when this route of administration is
submitted for approval.
Every effort should be made to reduce the potential risk of contamination
of the vaccine in the multi-dose vials (31).
The principles of the study design are as for section 1.7 above.

C.2 Clinical evaluation as part of postmarketing surveillance

C.2.1 Monitoring vaccine efficacy, effectiveness and safety
Every effort should be made to improve scientific understanding of the
protection and safety of rabies vaccines in humans by conducting active
postmarketing surveillance.
It is particularly important that data are collected on any vaccine failures
including detailed data on the post-exposure prophylaxis administered.
Details on the proper procedure for investigation of the treatment failures
are provided elsewhere (7).
Given that limited safety data are obtained in pre-licensure studies, it is very
important that safety should also be monitored as part of postmarketing
surveillance.
Data generated in post-marketing surveillance should be submitted to the
national regulatory authority.

Part D. Recommendations for national

regulatory authorities
D.1 General
The general recommendations for national regulatory authorities provided
in the Guidelines for national authorities on quality assurance for biological
products should be followed (32). These specify that no new biological substance
should be licensed until consistency of production has been established.
The detailed production and control procedures as well as any change in
them that may affect the quality, safety or efficacy of rabies vaccine should
be discussed with and approved by the national regulatory authority.
The national regulatory authority should obtain the International Standard for
potency testing and, where necessary, establish national working reference

120
 preparation(s) calibrated against the International Standard. In addition,
challenge virus standard should be obtained from a reliable source, stored
and used as appropriate. The national regulatory authority should be able to
provide the standard for potency testing as well as challenge virus standard
on request (Part A, section 6.5).

D.2 Release and certification

A vaccine lot should be released only if it fulfils Part A of these
Recommendations. Before any vaccine lot is released from a manufacturing
establishment, the recommendations for consistency of production provided
in Guidelines for national authorities on quality assurance for biological
products (32) should be met.
A statement signed by the appropriate official of the national control
laboratory should be provided if requested by a manufacturing establishment
and should certify whether or not the lot of vaccine in question meets all
national requirements, as well as Part A of these recommendations. The
certificate should also state the lot number, the number under which the lot
was released, and the number appearing on the labels of the containers. In
addition, the date of the last satisfactory potency test as well as the expiry
date assigned on the basis of shelf life should be stated. A copy of the official
national release document should be attached.
The purpose of the certificate is to facilitate the exchange of rabies
vaccine between countries. A model of a suitable certificate is given in
the appendix.

Authors
The scientific basis for the revision of the Requirements published in 1981, 1987
and 1994 was developed at two meetings of the working group held at the World
Health Organization, Geneva, in May 2003 and in May 2004 attended by the
following people:

Dr H. Bourhy, Rabies Laboratory, National Reference Centre for Rabies, WHO

Collaborating Centre for Reference and Research on Rabies, Pasteur Institute,
Paris, France; Dr L. Bruckner, Institute of Virology and Immunoprophylaxis,
Mittelhäusern, Switzerland; Dr W. Correa de Moura, INCQS/FIOCRUZ, Rio de
Janeiro, Brazil; Dr M. Ferguson, National Institute for Biological Standards and
Control, Potters Bar, England; Dr V. Grachev, Deputy Director, Institute of Poliomyelitis
and Viral Encephalitis, Moscow, Russian Federation; Dr R. Gibert, Laboratories and
Controls Department, French Health Products Safety Agency, Lyon, France;
Dr A. Kumar, Center for Biologics Evaluation and Research, Food and Drug
Administration, Rockville, MD, USA; Dr R. Levis, Center for Biologics Evaluation and
Research, Food and Drug Administration, Rockville, MD, USA; Dr L. Markoff, Center
for Biologics Evaluation and Research, Food and Drug Administration, Rockville,

121
 MD, USA; Dr H. Meyer, Department of Virology, Paul Ehrlich Institute, Langen,
Germany; Dr S. Morgeaux, Laboratories and Controls Department, French Health
Products Safety Agency, Lyon, France; Dr A. Tahlan, Central Drugs Laboratory,
Central Research Institute, Kasauli, India; Dr G. Reiner, Bulk Manufacturing
Germany, Chiron Vaccines, Marburg, Germany; Dr A. Sabouraud, Quality Control of
Development Products, Sanofi Pasteur, Marcy l’Etoile, France; Dr A. Costa, Access
to Technologies, WHO, Geneva, Switzerland; Dr J. Daviaud, Access to Technologies,
WHO, Geneva, Switzerland; Dr N. Dellepiane, Access to Technologies, WHO,
Geneva, Switzerland; Dr C. Rodriguez-Hernandez, Access to Technologies, WHO,
Geneva, Switzerland; Dr David Wood, Quality Assurance and Safety of Biologicals,
Geneva, Switzerland; Dr Ivana Knezevic, Quality Assurance and Safety of
Biologicals, Geneva, Switzerland.

The first draft of these revised Recommendations was prepared by Dr Ivana

Knezevic, Quality and Safety of Biologicals, WHO and Dr Morag Ferguson, National
Institute for Biological Standards and Control, England, following discussions at a
meeting of the working group held in May 2004.

The second draft of these Recommendations was prepared by Dr Ivana Knezevic,

Quality and Safety of Biologicals, WHO, taking into account information on the
current manufacturing and regulatory practice obtained from a survey undertaken
in 2005, outcomes of the Expert Consultation on Rabies held in 2004 as well as
comments from the experts consulted.

The third draft was prepared by Dr Ivana Knezevic, Quality Assurance and Safety
of Biologicals, WHO, Dr Morag Ferguson, National Institute for Biological Standards
and Control, and Dr David Wood, Quality Assurance and Safety of Biologicals,
WHO after an informal WHO Consultation held in July 2005, with the following
participants:

Dr H. Bourhy, Rabies Laboratory, National Reference Centre for Rabies, WHO

Collaborative Centre for Reference and Research on Rabies, Institut Pasteur, Paris,
France; Dr D. Briggs, Department of Diagnostic, Medicine/Pathobiology College of
Veterinary Medicine, Kansas State University, Manhattan, USA; Dr Cardoso de
Melo, Head of Hemotherapic and Biological Products Unit, Ministry of Health,
Brasilia, Brazil; Dr H. Chader, National Control Laboratory of Pharmaceutical
Products, Ministry of Health and Population, Clinic Ahmed AROUA, Yahia, Hydra,
Algeria; Dr F. Cliquet, WHO Collaborating Centre for Research and Management
on Zoonoses Control, Research Laboratory on Rabies and Wild Animal Diseases,
Malzeville, France; Dr G. Dong, National Institute for the Control of Pharmaceutical
and Biological Products, People’s Republic of China; Dr M. Ferguson, National
Institute of Biological Standards and Control, Potters Bar, Herts., England;
Dr R. Gibert, Laboratories and Controls Department, French Health Products Safety
Agency, Lyon, France; Dr V. Grachev, Institute of Poliomyelitis and Viral
Encephalitides, Academy of Medical Sciences of the Russian Federation, Moscow,
Russian Federation; Dr E. Griffiths, Biologics and Genetic Therapies, Ottawa,
Ontario, Canada; Dr R.L. Ichhpujani, Ministry of Health and Family Welfare, New
Delhi, India; Dr T. Jivapaisarnpong, Division of Biological Products, Department of
Medical Sciences, Ministry of Public Health, Nonthaburi, Thailand; Dr R. Levis,
Division of Viral Products, Food and Drug Administration, Center for Biologics

122
Evaluation and Research, Rockville Pike, Rockville, MD, USA; Mme L. Mahjoubi,
National Control Laboratory of Medicine, Ministry of Public Health, Tunis, Tunisia;
Dr H. Meyer, Paul Ehrlich Institute, Langen, Germany; Mrs D.P. Mora, National
Regulatory Authority of Cuba, State Control Center for the Quality of Drugs,
CECMED, Havana, Cuba; Dr Hoang Quang Huy, National Center for Quality Control
of Biological Products, Hanoi, Viet Nam; Dr N. Miranda, Laguna, Manila, Philippines;
Dr H. Rahman, Directorate of Drug Administration, Ministry of Health and Family
Welfare, Government of the People’s Republic of Bangladesh, Dhaka, Bangladesh;
Dr Lucky S. Slamet, Deputy for Therapeutic Products, Narcotic, Psychotropic and
Addictive Substance Control, National Agency of Drug and Food Control of the
Republic of Indonesia, Jakarta Pusat, Indonesia; Dr L. Teferi, Head, Drug
Administration and Control Authority, Human and Veterinary Drug Evaluation and
Registration Division, Drug Administration and Control Authority, Addis Ababa,
Ethiopia; Dr H. Wilde, Queen Saovabha Memorial Institute, Thai Red Cross Society
and Department of Medicine, Chulalongkor University, Bangkok, Thailand; Dr E.I.P.
Arisetianingsih, National Laboratory of Drug and Food Control, NADFC, Jakarta
Pusat, Indonesia; Dr S. Morgeaux, Laboratories and Controls Department, French
Health Products Safety Agency, Lyon, France; Dr D. Smith, Biologics and Genetic
Therapies, Ottawa, Ontario, Canada; Dr M. Frazatti-Gallina, Instituto Butantan/
Fundaçao Butantan, São Paulo, Brazil; Dr S. Jadhav, Serum Institute of India Ltd,
Pune, India; Mrs Catherine Chamberlin, European Directorate for the Quality of
Medicines, Strasbourg, France; Dr C. Malerczyk, Clinical Research and Medical
Affairs, Chiron Vaccines, Marburg, Germany; Dr G. Reiner, Bulk Manufacturing
Germany, Chiron Vaccines, Marburg, Germany; Dr A. Sabouraud, Quality Control
of Development Products, Sanofi Pasteur, Marcy l’Etoile, France; Dr J. Sokhey,
WHO Regional Office for South-East Asia, World Health House, New Delhi India;
Dr D. Wood, Coordinator, Quality Assurance and Safety of Biologicals, WHO,
Geneva Switzerland; Dr M.P. Kieny, Director, Initiative for Vaccine Research, WHO,
Geneva, Switzerland; Dr F. Meslin, Coordinator for Strategy Development and
Monitoring of Zoonoses, Foodbone Diseases and Kinetoplastidae (ZFK), CPE/CDS,
WHO, Geneva, Switzerland; Dr N. Dellepiane, Access to Technologies, WHO,
Geneva, Switzerland; Dr A. Costa, Access to Technologies, WHO, Geneva,
Switzerland; Dr C. Hernandez Rodriguez, Access to Technologies, WHO, Geneva,
Switzerland; Dr I. Knezevic, Quality Assurance and Safety of Biologicals, WHO,
Geneva Switzerland.

The first draft of the section on clinical evaluation of vaccines was prepared
by Dr D. Briggs, Kansas State University, Manhattan, USA, Dr I. Knezevic,
Quality Assurance and Safety of Biologicals, WHO, Geneva Switzerland and
Dr M. Ferguson, National institute for Biological Standards and Control, Potters Bar,
England, following a discussion on 13 July 2005 attended by following people:
Dr H. Wilde, Dr D. Briggs, Dr V. Gratchev, Dr M. Ferguson, Dr. C. Malerczyk,
Dr T. Jivapaisarnpong, Dr R.L. Ichhpujani, Dr R. Levis, Dr J. Sokhey, Dr J. Lang and
Dr I. Knezevic. Taking into account comments and suggestions made on the first
draft by all participants of the consultation, in particular by the participants of the
discussion held on 13 July 2005, the draft was finalized.

123
Acknowledgements
Acknowledgements are due to the following experts for their comments and advice
on these recommendations: Dr V. Grachev, Deputy Director, Institute of Poliomyelitis
and Viral Encephalitis, Moscow, Russian Federation; Dr A. Tahlan, Central Drugs
Laboratory, Central Research Institute, Kasauli, India; Dr H. Bourhy, Rabies
Laboratory, National Reference Centre for Rabies, WHO Collaborative Centre for
Reference and Research on Rabies, Institut Pasteur, Paris, France; Dr Powell,
Medicines and Healthcare Regulatory Agency, London, UK; Dr D. Briggs, Department
of Diagnostic, Medicine/Pathobiology College of Veterinary Medicine, Kansas State
University, Manhattan, USA; Dr S. Jadhav, Serum Institute of India Ltd, Pune, India;
Mrs Catherine Chamberlin, European Directorate for the Quality of Medicines,
Strasbourg, France; Dr H. Wilde, Queen Saovabha Memorial Institute, Thai Red
Cross Society and Department of Medicine, Chulalongkor University, Bangkok,
Thailand; Dr A. Sabouraud, Quality Control of Development Products, Sanofi Pasteur,
Marcy l’Etoile, France; Dr G. Reiner, Bulk Manufacturing Germany, Chiron Vaccines,
Marburg, Germany; Dr N. Miranda, Laguna, Manila, Philippines; Dr Jivapaisanpong,
Division of Biological Products, Ministry of Public Health, Northabmi, Thailand;
Dr F. Meslin, Coordinator for Strategy Development and Monitoring of Zoonoses,
Foodbone Diseases and Kinetoplastidae (ZFK), CPE/CDS, WHO, Geneva,
Switzerland; Dr N. Dellepiane, Access to Technology, WHO, Geneva, Switzerland.

References
1. Requirements for rabies vaccine for human use. In: WHO Expert
Committee on Biological Standardization. Thirty-first report. Geneva,
World Health Organization, 1981, Annex 2 (WHO Technical Report Series,
No. 658).
2. Requirements for rabies vaccine (inactivated) for human use produced
in continuous cell lines. In: WHO Expert Committee on Biological
Standardization. Thirty-ninth report. Geneva, World Health Organization,
1987 (WHO Technical Report Series, No. 760).
3. Requirements for rabies vaccine for human use. In: WHO Expert Committee
on Biological Standardization. Forty-third report. Geneva, World Health
Organization, 1994, Annex 4 (WHO Technical Report Series, No. 840).
4. Requirements for rabies vaccine (inactivated) for human use produced
in continuous cell lines. In: WHO Expert Committee on Biological
Standardization. Forty-third report. Geneva, World Health Organization,
1994, Annex 5 (WHO Technical Report Series, No. 840).
5. WHO position paper on rabies vaccines. Weekly Epidemiological Records,
No. 14, 2002, 77:109–119.
6. Rupprecht CE, Hanlon CA, Hemachudha T. Rabies re-examined. The
Lancet Infectious Diseases, 2002, 2:327–343.
7. WHO Expert Consultation on Rabies. First report. Geneva, World Health
Organization, 2005 (WHO Technical Report Series, No. 931).
8. Meeting report: discussion on WHO requirements for rabies vaccine
for human use: potency assay, 20 May 2003. Geneva, World Health

124
 Organization, 2003 (http://www.who.int/biologicals/publications/
meetings/areas/vaccines/rabies/en/Rabies%20vaccine%20meeting%20Final
%20May%202003.pdf).
9. Meeting Report: Discussion on WHO requirements for rabies vaccine for
human use 4–5 May 2004. Geneva, World Health Organization, 2004
(http://www.who.int/biologicals/publications/meetings/areas/vaccines/rabies/
en/Rabies%20vaccine%20meeting%20Final%20May%202004.pdf).
10. Laboratory techniques in rabies. 4th edition. Geneva, World Health
Organization, 1996.
11. Hemachuda T et al. Myelin basic protein as an encephalitogen in
encephalomyelitis and polyneuropathy folowing rabies vaccination. New
England Journal of Medicine, 1987, 316, 3:9–74.
12. Parviz S, Luby S, Wilde H. Postexposure treatment of rabies in Pakistan.
Clinical Infectious Diseases, 1998, 27:751–755.
13. Parviz et al. Rabies deaths in Pakistan : results of ineffective post-exposure
treatment. International Journal of Infectious Diseases, 2004, 8:346–352.
14. Chanthavanich P et al. Antibody response to suckling mouse brain rabies
vaccines for post exposure treatment. Transactions of the Royal Society of
Tropical Medicine and Hygiene 1987, 81:260-263.
15. Requirements for the use of animal cells as in vitro substrates for the
production of biologicals. In: WHO Expert Committee on Biological
Standardization. Forty-seventh report. Geneva, World Health Organization,
1998, Annex 1 (WHO Technical Report Series, No. 878).
16. Guidelines on transmissible spongiform encephalopathies in relation to
biological and pharmaceutical products. Geneva, World Health Organization,
2003 (WHO/BCT/QSD/03.01).
17. Requirements for the collection, processing and quality control of blood,
blood components and plasma derivatives. (Requirements for Biological
Substances No. 27, revised 1992). In: WHO Expert Committee on Biological
Standardization. Forty-third report. Geneva, World Health Organization,
1994 (WHO Technical Report Series, No. 840).
18. Vaccines and biologicals: Recommendations from the Strategic Advisory
Group of Experts. Weekly Epidemiological Record, 2002, 37:306.
19. Good manufacturing practices for pharmaceutical products. In: WHO Expert
Committee on Specifications for Pharmaceutical Preparations. Thirty-seventh
Report, Geneva, World Health Organization, 2003, Annex 4 (WHO Technical
Report Series, No. 908).
20. Good manufacturing practices for biological products. Geneva, World Health
Organizaton, 1992 (WHO Technical Report Series, No. 822):20–30.
21. Laboratory biosafety manual, 3rd edition. Geneva, World Health
Organization, 2004.
22. Robertson JS et al. Assessing the significance of reverse transcriptase
activity in chick cell-derived vaccines. Biologicals 1997, 25:403–414.
23. General requirements for the sterility of biological substances (Requirements
For Biological Substances No.6, revised 1973). In: WHO Expert Committee
on Biological Standardization. Twenty-fifth report, Geneva, World Health

125
 Organization, 1973 (Technical Report Series, No. 530) and amended In:
WHO Expert Committee on Biological Standardization. Forty-sixth report,
Geneva, World Health Organization, 1998 (WHO Technical Report Series,
No. 872).
24. Recommendations for the production and control of poliomyelitis vaccine
(oral). In: WHO Expert Committee on Biological Standardization. Fiftieth
report. Geneva, World Health Organization, 2002 (WHO Technical Report
Series, No. 904).
25. Guidelines on regulatory expectations related to the elimination, reduction
or replacement of thiomersal in vaccines. In: WHO Expert Committee
on Biological Standardization. Fifty-third report. Geneva, World Health
Organization, 2004, Annex 4 (WHO Technical Report Series, No. 926).
26. Guidelines for nonclinical evaluation of vaccines. In: WHO Expert Committee
on Biological Standardization. Fifty-fourth report. Geneva, World Health
Organization, 2005 (WHO Technical Report Series, No.927) annex 1.
27. Guidelines for good clinical practices (GCP) for trials on pharmaceutical
products. In: WHO Expert Committee on the Use of Essential Drugs,
Sixth report. Geneva, World Health Organization, 1995, Annex 3 (WHO
Technical Report series, No. 850).
28. Guidelines for clinical evaluation of vaccines: regulatory expectations. In:
WHO Expert Committee on Biological Standardization. Fifty-second report.
Geneva, World Health Organization, 2004, Annex 1 (WHO Technical Report
Series, No. 924).
29. Good Manufacturing Practices: supplementary guidelines for the
manufacture of the investigational pharmaceutical products for clinical trials
in humans. In: WHO Expert Committee on Specifications for Pharmaceutical
Preparations: Thirty-Fourth Report, 1996. Annex 7 (WHO Technical Report
Series, No. 863).
30. Moore SM et al. The influence of homologous vs. heterologous challenge
virus strains on the serological test results of rabies virus neutralization
assays. Biologicals, 2005, Dec 33(4):269–76.
31. WHO policy statement. The use of opened multi-dose vials of vaccine in
subsequent immunization sessions. Geneva, World Health Organization,
2000 (WHO/V&B/00.09).
32. Guidelines for national authorities on quality assurance for biological
products. Geneva, World Health Organization, 1992 (WHO Technical
Report Series, No. 822):31–46.

126
 Appendix
Summary protocol for production and testing
of inactivated rabies vaccine for human use
produced in cell substrates and embryonated eggs
Identification of final lot
Name and address of manufacturer ______________________________
Lot number of vaccine ________________________________________
Date of manufacture of final lot _________________________________
Date of start of the potency test _________________________________
Expiry date _________________________________________________
Total volume of final lot _______________________________________
Temperature of storage _______________________________________

3. Control of source materials1

3.1 Substrate for virus production
Name and identification of cell substrate _________________________

Cell seed and WCB

Origin and short history of master and working cell bank
(date of approval by national regulatory authority) __________________
Authority that approved cell seed _______________________________
Date the MWCB was established and approved by national
regulatory authority
Quantity of cell stored ________________________________________
The passage level of the WCB __________________________________
Storage conditions ___________________________________________
Percentage of all WCB ampoules tested __________________________

Identity test (WCB)

Method used _______________________________________________
Results ____________________________________________________

Serum used in cell culture medium

Origin of serum used _________________________________________
Tests performed on serum _____________________________________
Results ____________________________________________________

Trypsin used for preparation of cell cultures

Origin of trypsin used _________________________________________

1
Section numbers correspond to those used in the main text of the annex.

127
 Tests performed on trypsin ____________________________________
Results ____________________________________________________

3.2 Virus seed

Strain of virus
Name and short description of history, origin, process of attenuation,
and adaptation ____________________________________________
Date of preparation of master virus seed lot _______________________
Number of passages between isolation and primary seed _____________
Date of preparation of working virus seed lot ______________________
Number of passages between master and working seed ______________
Virus seed lot system
Number of subcultures between master
virus seed lot and production _________________________________
Method for identification of the virus seed lot _____________________
Results ____________________________________________________
Tests for bacteria, fungi, and mycoplasmas
Method used _______________________________________________
Results ____________________________________________________
Tests for adventitious agents
Tests in suckling mice
No. of animals tested _________________________________________
Quantity injected ________________________________________
Observation period _______________________________________
Results (survival numbers, etc) _____________________________
Tests in adult mice
No. of animals tested _____________________________________
Quantity injected ________________________________________
Observation period _______________________________________
Results (survival numbers, etc.,) ____________________________
Tests in guinea-pigs
No. of animals tested _____________________________________
Quantity injected ________________________________________
Observation period _______________________________________
Results (survival numbers, etc.) _____________________________
Tests in cell cultures
Methods _______________________________________________
Results ________________________________________________
Virus content
Method of titration ___________________________________________
Results ____________________________________________________

128
4. Control of vaccine production
4.1 Control of cell cultures
Tests for haemadsorbing viruses
Method ____________________________________________________
Results ____________________________________________________

Tests for other adventitious agents

Method ____________________________________________________
Results ____________________________________________________

Identity test (cell line)

Method ____________________________________________________
Results ____________________________________________________

4.2 Control of production in embryonated duck eggs

Control of (uninoculated) embryonated duck eggs
Method ____________________________________________________
Results ____________________________________________________

4.3 Control of single virus harvests and purified bulk material

Sterility tests of single virus harvests
Have all the harvests included been tested for sterility? ______________
Results of these tests _________________________________________

Pooling of single virus harvests

No. of viral harvests included __________________________________
Date of pooling _____________________________________________

Purification of virus harvests

Method ____________________________________________________
Degree of purity achieved _____________________________________

Animal serum in purified bulk

Method ____________________________________________________
Results (concentration) _______________________________________

4.4 Inactivation procedure

Method ____________________________________________________
Date ______________________________________________________
Temperature ________________________________________________
Tests for effective inactivation
Volume and concentration of bulk material injected _____________

129
 No. of mice injected ______________________________________
Weight of mice __________________________________________
Duration of observation ___________________________________
Other animals (if used) ____________________________________
Results of tests ______________________________________________
Rabies virus amplification test
Amount of vaccine tested (ml) ______________________________
Results ________________________________________________

4.5 Preparation and control of final bulk

Preservatives and other substances added
Concentration of phenol (if used) _______________________________
Other preservatives (type and concentration) ______________________
Other substances added _______________________________________

Antigen content of the final bulk

Method ____________________________________________________
Results ____________________________________________________

Sterility tests
Date of test _________________________________________________
Result _____________________________________________________

Other tests (chemical, biochemical)

Type of test _________________________________________________
Results ____________________________________________________

5. Filling and containers

Date of filling _______________________________________________
Quantity of containers ________________________________________
Volume of vaccine per container ________________________________

Control for defective vials

Methods ___________________________________________________
Results ____________________________________________________

6. Control tests on final product

6.1 Identity test
Method ____________________________________________________
Result _____________________________________________________

6.2 Sterility test

No. of containers examined ____________________________________

130
 Method ____________________________________________________
Date at start of test ___________________________________________
Date at end of test ___________________________________________
Result _____________________________________________________

6.3 Innocuity tests

Mice Guinea-pigs
No. of animals _____________________ ____________________
Route of injection _____________________ ____________________
Volume of injection _____________________ ____________________
Date of injection _____________________ ____________________
Date of end of test _____________________ ____________________
Result _____________________ ____________________

6.4 Antigen content

Type of test _________________________________________________
Result _____________________________________________________

6.5 Potency test of vaccine in final containers

Type of test _________________________________________________
Date of immunization of mice __________________________________
Reference vaccine (potency) ___________________________________
Challenge strain _____________________________________________
Date of challenge ____________________________________________
ED50 test vaccine1 ____________________________________________
ED50 reference vaccine1 _______________________________________
Calculated IU/single human dose _______________________________
Confidence limits ____________________________________________
Results of other potency tests __________________________________

6.6 Stability test

Duration and temperature of incubation __________________________
Result _____________________________________________________

6.7 Residual moisture test on freeze-dried vaccine

Method ____________________________________________________
Result _____________________________________________________

1
ED50, quantity of vaccine that protects 50% of animals against infection with the challenge
strain.

131
6.8 Inspection of final containers
Result _____________________________________________________

6.9 Test for pyrogenic substances

Method ____________________________________________________
Results ____________________________________________________

6.10 Test for adjuvant

Date of test _________________________________________________
Nature and concentration of adjuvant per single human dose __________
Degree of adsorption _________________________________________

Internal certification
Certification by person taking overall responsibility for production of the vaccine
I certify that lot no. _________________ of rabies vaccine satisfies Part A
of the WHO Recommendations for inactivated rabies vaccine for human
use produced in cell substrates and embryonated eggs.
Signature __________________________________________________
Name (typed) _______________________________________________
Date ______________________________________________________
The protocol must be accompanied by a sample of the label and a copy of
the leaflet.

Release certification by the national regulatory authority

Whenever rabies vaccines produced in continuous cell lines are to be
exported, they should be accompanied by a release certificate from the
national regulatory authority.

Sample release certificate

I hereby certify that batch no.____________ of rabies vaccine produced by
(name of producer) in continuous cell lines meets all national requirements
as well as Part A of the WHO Recommendations for inactivated rabies
vaccine produced in cell substrates and embryonated eggs for human use.
The date of the last satisfactory potency test carried out by the national
regulatory authority is ________________________________________
The final lot has been released by us under no. _____________________
The number appearing on the label of the containers is ______________
Signature __________________________________________________
Name (typed) _______________________________________________
Date ______________________________________________________

132
© World Health Organization
WHO Technical Report Series No 941, 2007

Annex 3
Guidelines to assure the quality, safety and efficacy
of live attenuated rotavirus vaccines (oral)

Introduction
General considerations
Special considerations
Use of primary cells for the production of rotavirus vaccine
Part A. Guidelines on manufacturing
A.1 Definitions
A.2 General manufacturing recommendations
A.3 Control of source materials
A.4 Control of vaccine production
A.5 Filling and containers
A.6 Control tests on final product
A.7 Records
A.8 Samples
A.9 Labelling
A.10 Distribution and shipping
A.11 Storage and expiry date
Part B. Nonclinical evaluation of live attenuated rotavirus vaccines
Part C. Clinical evaluation of live attenuated rotavirus vaccines
Part D. Guidelines for national regulatory authorities
D.1 General
D.2 Release and certification
Authors
Acknowledgements
References
Appendix 1
Summary protocol of manufacturing and control of live attenuated
rotavirus vaccines (oral)
Appendix 2
Genealogy of vaccine production process
Appendix 3
Model certificate for the release of live attenuated rotavirus vaccines

133
 This document provides guidance to national regulatory authorities and
vaccine manufacturers on the production, quality control and evaluation of
safety and efficacy of live attenuated rotavirus vaccines (oral) by outlining,
in detail, international regulatory expectations for product characterization.
It should be read in conjunction with the WHO guidelines on nonclinical
evaluation of vaccines (1), and the WHO guidelines on clinical evaluation
of vaccines: regulatory expectations (2), in order to understand the whole
process of vaccine evaluation. Advice that is specific to the nonclinical
or clinical evaluation of live attenuated rotavirus vaccines is provided to
supplement these two generic documents.
The following text is written in the form of guidelines instead of
recommendations. Guidelines allow greater flexibility than recommendations
with respect to expected future developments in the field.

Introduction
The high incidence of rotavirus disease around the world has prompted
international agencies including the World Health Organization (WHO),
the Global Alliance for Vaccines and Immunization (GAVI), Children’s
Vaccine Program and the Rotavirus Vaccine Programme at the Program
for Appropriate Technology in Health (PATH) to identify rotavirus vaccine
development as one of their highest priorities. In response to this interest
in the development of candidate rotavirus vaccines, WHO held an informal
consultation on the Quality, Safety and Efficacy Specifications for Live
Attenuated Rotavirus Vaccines in Mexico City, from 8–9 February 2005 (3).
The experts present considered it timely for WHO to develop production and
quality control guidelines and technical specifications for these vaccines.
Draft guidelines on production and quality control specifications for live
attenuated rotavirus vaccine (oral) were developed by a small drafting
group established by WHO and were comprehensively reviewed by a WHO
Informal Consultation on the proposed WHO Guidelines to Assure the
Quality, Safety and Efficacy of Live Attenuated Rotavirus Vaccines held in
Geneva, from 29–30 August 2005. After taking into account the comments
made during the consultation, the revised document was submitted to and
approved by the WHO Expert Committee on Biological Standardization at
its 56th meeting.
The scope of this document covers live attenuated rotavirus vaccines (oral),
and vaccine candidates, that are intended for international use. The aim
of vaccination against rotavirus infection is to induce immunity against
relevant rotavirus serotypes in one series of inoculations for the prevention of
rotavirus gastroenteritis. The prevalence of multiple and different rotavirus
serotypes throughout the world complicates vaccine development strategies.
According to the data available in 2005, the majority of rotavirus disease

134
in humans is caused by five distinct serotypes of rotavirus, designated G1,
G2, G3, G4 and G9 on the basis of the molecular composition of one outer
coat viral protein (VP7). In the last decade, intense surveillance of rotavirus
and improved laboratory assay methods have resulted in the detection of
additional G serotypes that play a role in human disease. Although serotypes
G1–G4 still predominate worldwide, serotypes G5, G8 and G9 have been
isolated in Latin America, Africa and India respectively, and are beginning
to cause a higher percentage of cases of the disease. The relative benefits of
monovalent, multivalent or regional specific vaccines will remain unclear
until efficacy data demonstrating heterotypic protection against relevant
rotavirus serotypes become available.
The scope of this document is restricted to live vaccines, but the properties
of the possible candidates are varied, particularly with respect to their
degree of attenuation. This affects the extent to which they grow in culture
and in the human gut and applies to strains currently being investigated
as well as those studied previously. There will therefore be major quality
issues specific to a particular vaccine, such as the assay of vaccine potency,
the stability of the virus in production, the yield of virus and the extent
of contamination by cell-derived materials. Clinical issues which will
vary between candidates include the dose required to obtain immunity in
recipients, the possibility of transmission to contacts of vaccinees and the
genetic stability of the virus on replication in the gut of recipients. Although
many of the points of possible concern considered in this document are
generally applicable to rotavirus vaccines, it must be remembered that each
candidate must be examined individually and that this may raise significant
product-specific issues.
Several oral rotavirus vaccines are currently approved or in various stages
of vaccine development. Each vaccine is the result of a unique approach
in developing an attenuated phenotype and to the prevention of rotavirus
disease. As more vaccines become readily available around the world,
data will be provided to help in making decisions on the compositions of
future vaccines. Examples of vaccines approved or in development include
multivalent human–bovine and human–rhesus reassortant vaccines and
monovalent vaccines containing a single attenuated human rotavirus strain;
natural human–bovine reassortants or animal rotavirus strains. Vaccines for
which controlled efficacy studies have been completed have demonstrated
high levels of efficacy (> 80%) against severe rotavirus gastroenteritis and
have also shown a range of heterotypic protection against multiple serotypes.
The impact of introduction of the vaccine on the epidemiology of natural
rotavirus infection is as yet unknown.
All of the vaccine candidates are claimed to be attenuated; however, the
basis for the attenuation phenotype of the current vaccine candidates is

135
 unknown. Typically the method of attenuation is to exploit the host range
restriction properties of animal rotaviruses (Jennerian) by laboratory
or natural (neonatal rotavirus strains) genetic reassortment with human
rotavirus with the desired serotypes or insertion of multigenic mutations
by serial passage in cell culture. Because laboratory markers of attenuation
are not well defined and animal models demonstrating attenuation are not
readily available, the claim of attenuation is based on clinical experience
in human subjects. How the product will actually behave in clinical use is
therefore based on clinical trials and consistency of production rather than
on any specific laboratory tests. Potential laboratory markers of product
consistency include information on the genetic sequence of the virus seed
to show that a new seed material is similar to the previous seed and that
each can be distinguished from the parent virus. Multiple passages of virus
seed materials under defined cell culture conditions as well as examination
of vaccine virus isolated from stool samples of vaccinees may be helpful
to generate data on the conditions favouring genetic stability of the virus.
Studies on consistency of production would need to take into account the
variability inherent in replication of RNA virus and assess the presence of
minority populations, as revealed for example by the occurrence of mixed
base signals in sequencing studies. If minority populations are detected,
it will be necessary to assess their biological importance, for example, by
careful comparison of the level of heterogeneity between the master or
working seed and higher passage levels e.g. clinical trial material.
The development of any new rotavirus vaccines must take into account the
events which led to the withdrawal of one vaccine (RotaShield) from the
marketplace. RotaShield was introduced in the United States in August 1998
and was withdrawn less than 1 year later when an epidemiological relationship
was established between RotaShield vaccination and intussusception (IS).
Early estimates suggested a risk of one case per 2500 children immunized.
Re-analysis of the case–control study that examined intussusception and
RotaShield revealed that the majority of the cases of IS associated with the
first dose occurred in children 4 months of age or older. This did not comply
with the manufacturer’s recommendation that the first dose should be given
at 2 months of age and thus changed the early estimates of attributable risk
of IS in the target population (4). The detailed pathogenic mechanisms for
IS are unclear, but are very likely to be complex.

General considerations
Considerable experience in the manufacturing, testing and clinical
evaluation of rotavirus vaccines has been gained from over 20 years of
their development. Candidate vaccines have been and are being studied in
a number of countries with diverse economic conditions and geographical

136
boundaries. The regulatory approvals of two rotavirus vaccines and the
submission for approval of another provide a solid foundation for the
preparation of these guidelines and for the continuing development of new
rotavirus vaccines; they also support the global introduction of rotavirus
vaccines. These experiences have enabled standards to be set for vaccine
efficacy, identified safety concerns, highlighted areas of scientific weakness,
and led to the introduction of new manufacturing technologies and testing
strategies. Issues relevant to the development of rotavirus vaccines are
discussed below.
Rotavirus is an acid-labile virus which is rapidly inactivated at an acid pH
and has a half-life of less than 12 minutes at pH 2.0. Because rotavirus
vaccines are intended to be administered to infants by the oral route, the
virus would be inactivated by stomach gastric acid prior to reaching the site
of infection in the upper gastrointestinal tract. To prevent inactivation of the
virus by gastric acid, antacids or buffers are usually administered before or
in combination with the oral rotavirus vaccination. The composition of the
antacid and the mode of administration (in combination with vaccine or
administered separately) will depend upon the biological characteristics of
the vaccine virus.
The processing of live rotavirus vaccines, like that of many other live viral
vaccines, involves cell disruption and, if any, incomplete purification of
the virus. In-process steps for the inactivation of adventitious agents are
not included for live viral vaccines, as these steps may compromise the
live nature of the vaccine itself. As a result, validation of clearance of any
adventitious agents may not be possible. For these reasons a comprehensive
set of tests for adventitious agents and qualification of the vaccine source
materials is essential as part of the vaccine safety control. As with any
viral vaccines, production of rotavirus vaccines also involve cells, virus
seeds and biological reagents (such as growth supplements, serum, trypsin
and any virus stabilizers used in the final product). Hence, each of these
vaccine source components must be shown to be free from adventitious
agents. The full passage history of the seed materials used for vaccine
development should identify all substrates through which they have been
passed to aid the development of appropriate programmes of testing for
adventitious agents.Viral seed lots should be assessed for absence of
adventitious agents from all species that they may have been exposed to
from isolation, through passage, and during production, including those
that may be present in the raw materials used at each of these stages. The
testing required will depend on available documentation on cell substrate
donor, virus seed passage and derivation history and the seed virus may
require purification (e.g. molecular cloning and plaque purification), if
passage and derivation history is uncertain, or if it is contaminated with a
known agent.

137
 A risk assessment for transmissible spongiform encephalopathies would
need to be included for the seed materials. The revised WHO Guidelines
on transmissible spongiform encephalopathies in relation to biological
and pharmaceutical products (5) provide guidance on risk assessments for
master and working seeds and should be consulted.
Many candidate vaccines are being developed in Vero cells. WHO has set
a limit of 10 ng of DNA per human dose for parenterally administered
vaccines based upon a WHO risk assessment (6) that established a 2 × 108
safety factor. For orally delivered vaccines, the acceptable limit of residual
cellular DNA has not been previously established. However, recent studies
in rats, which compared the uptake efficiency of Vero cell DNA following
administration by the oral and intramuscular routes indicated that Vero
cell DNA given orally could be found in rat tissues at a concentration
of 10 000-fold less than when equivalent amounts of Vero DNA were
given parenterally (7). Similarly, experiments in mice using polyoma
DNA also showed a difference in uptake between the oral and parenteral
routes (8). These preliminary data and the lack of evidence that cellular
DNA is tumorigenic suggest that a DNA level of 100 µg/dose in an orally
administered vaccine is equivalent to 10 ng in parenterally administered
vaccines. In addition to the quantity of DNA, the size distribution of DNA
should be evaluated during assessments of manufacturing consistency and
process validation. It may be desirable for manufacturers to include steps
in the process to reduce the level of high-molecular-weight cellular DNA.
The national regulatory authority may require the further purification of
virus harvests derived from continuous cell lines to remove cellular DNA,
and/or the use of DNase treatment to reduce the size of DNA fragments. If
the virus harvests are derived from diploid cell cultures, further purification
is not required.
The concentration of virus can be determined by infectivity titrations, e.g.
plaque-forming or focus-forming assays. However, confirming the quantity
of each rotavirus serotype in the final mixture of a multivalent, as opposed
to a single component, vaccine is challenging. Conventional infectivity
titrations require an antibody specific to each of the vaccine components.
Molecular methods of measuring virus concentration that do not require
the use of specific antibody, such as quantitative real-time polymerase
chain reaction (PCR), have been developed and are being evaluated. One
method uses Vero cell monolayers which are seeded in 96-well plates and
inoculated with dilutions of a multivalent reference standard and samples.
Viral replication is allowed to proceed for 23–24 hours and the cells are
then lysed by addition of Triton and one freeze–thaw cycle. Cell lysates
are assayed by real-time quantitative reverse transcriptase-PCR (QPCR)
and rotavirus nucleic acid is quantified (9). Individual QPCR reactions are
performed employing primer/probe sets specific to each of the reassortant

138
strains. Virus concentration is determined by parallel line analysis between
the reference standard and each sample. International titration standards
do not yet exist and WHO is considering developing such reagents and
their subsequent characterization by international collaborative study. It
should be borne in mind however that the differences in the properties of
the candidate vaccines in terms of their attenuation may mean that such
reference material will be product-specific. Manufacturers should set aside
a preparation for use as an in-house standard.
The real-time stability of each serotype in the final vaccine formulation
should be measured to verify that the infectivity titre is at or above that
level of virus required for vaccine efficacy at the end of the product shelf-
life. The thermal stability of each individual serotype should be determined
in an appropriate stability study programme. It is advisable to assign a
shelf-life to the intermediates that are intended to be stored on the basis of
stability studies. WHO is developing further guidance on this issue. Based
on the results of the stability testing programme, an accelerated stability test
should be done on each new batch of vaccine.
If the final product is mixed with vaccine diluent prior to use, the stability
of this combination should also be determined.
The consistency of a manufacturing process for live virus vaccine can best
be determined by monitoring the content and quality of virus produced
throughout the process, as well as the level of impurities following
processing and the stability of the final product. Manufacturers should
select relevant parameters based upon the specific production process used
and develop methods to establish a programme to assure the consistency
of production process. For example, the amount of virus produced by a
standardized production process and the amount lost during processing
should be consistent; the quality of the virus produced can be assessed by
its thermal stability profile; the stability of the genomic sequence through
multiple cell culture passages may be evaluated; process impurities and
residuals such as residual host cell protein, residual cellular DNA, endotoxin,
bovine serum, trypsin and antibiotics can be measured and their reduction
during processing can be monitored. If the consistency of production has
been validated, the monitoring programme could be conducted based on
a reduced frequency of testing, but not on a batch-to-batch basis, with the
agreement of the national regulatory authority.

Special considerations
Use of primary cells for the production of rotavirus vaccine
The majority of live attenuated rotavirus vaccines that have been developed
or are currently under development are produced in continuous cell lines

139
 or diploid cells which are based on the cell bank system, so the main
focus of these guidelines is on vaccines produced under such conditions.
But the vaccine may also be produced in primary cells, in which case, the
manufacturers should consult additional relevant guidelines or documents
to ensure the safety of the final product. However, as suitable alternative
cell substrates become available, primary cell cultures are less likely to be
used in the future for many reasons (6).The following text provides brief
guidance on issues that need to be considered if primary cells are used for
rotavirus vaccine production, but there are some specific issues relating to
the production of rotavirus vaccines in primary cell substrates which are not
covered in these guidelines.
If vaccine is produced in primary cells obtained directly from trypsinized
tissue of normal healthy animals, such animals should be of a species
approved by the national regulatory authority and not have been previously
employed for experimental purposes. However, primary cell cultures
prepared from wild animals show a high frequency of viral contamination.
The number of viruses isolated and the frequency of isolation depend on
many factors, including methods of isolation, test cell systems used, number
of passages, duration of incubation and co-cultivation, and are directly
proportional to the duration of the incubation period of the cultures. The
frequency of cell culture contamination can be reduced by careful screening
of the source animals to be used in production for the absence of antibodies
to the relevant viruses. The use of animals bred in a carefully controlled
colony, especially those which are specific-pathogen free, is strongly
recommended. In addition, the use of secondary or tertiary cells on which
testing for adventitious agents could be performed will reduce the frequency
of contamination of the production cell culture. Production of uninfected
control cultures and the extensive testing required for relevant adventitious
agents becomes more challenging when producing live rotavirus vaccines
in primary cells.
A live attenuated rotavirus vaccine prepared from primary calf kidney
cells has been approved for use in one country. There are no existing WHO
guidelines for the production of vaccines on primary calf kidney cells. The
WHO Requirements for the use of animal cells as in vitro substrates for
the production of biologicals (6) provide general recommendations. The
WHO Recommendations for the production and control of poliomyelitis
vaccine (Oral) (10), Guidelines for the production and control of Japanese
encephalitis vaccine (live) for human use (11) and Guidelines for the
production of candidate live attenuated dengue virus vaccines (12) provide
guidance on the use of primary cell substrates of monkey, hamster and
dog kidney, respectively, and illustrate the principles that should be
followed for vaccines produced in primary cell substrates from other
species.

140
Part A. Guidelines on manufacturing
A.1 Definitions
A.1.1 International name and proper name
The international name should be Live attenuated rotavirus vaccine (LARV)
(oral) with additions to indicate the virus serotype(s) of the vaccine. The
proper name should be equivalent to the international name in the language
of the country of origin.
The use of the international name should be limited to vaccines that satisfy
the specifications formulated below.

A.1.2 Descriptive definition

Live attenuated rotavirus vaccine (oral) is a sterile vaccine preparation
containing one or more virus serotypes, which have been grown through a
seed lot system, prepared in a suitable approved cell substrate, formulated in a
form suitable for oral administration and satisfying all the recommendations
formulated in this document.
At least five types of live attenuated rotavirus vaccine have been developed
and/or are under development:
— a candidate rhesus rotavirus strain live attenuated vaccine;
— a candidate human-rhesus rotavirus strain live attenuated vaccine;
— a human rotavirus strain live attenuated vaccine;
— a candidate human–bovine reassortant rotavirus strain live attenuated vaccine;
— a lamb rotavirus strain live attenuated vaccine.
The preparation should satisfy all of the specifications formulated in these
guidelines.
Live attenuated rotavirus vaccine with an appropriate stabilizer may be
freeze-dried or liquid.

A.1.3 International reference preparations

No international reference preparations are available at the time of preparing
this document.

A.1.4 Terminology
The definitions given below apply to this document only.
Candidate vaccine. A vaccine under development which is used in human
clinical trials to assess its safety and efficacy.
Virus master seed lot. A quantity of virus of uniform composition derived
from an original isolate, processed at one time and passaged for a documented
number of times.

141
 Virus working seed lot. A quantity of virus of uniform composition derived
from the master seed lot by a limited number of passages by a method
approved by the national regulatory authority and fully characterized. The
virus working seed lot is used for production of vaccine.
Cell seed. A quantity of well-characterized cells, derived from a single tissue
or cell of human, animal or other origin, stored frozen in liquid nitrogen in
aliquots of uniform composition, one or more of which may be used for the
production of a master cell bank.
Cell bank. A collection of ampoules containing material of uniform
composition stored under defined conditions, each ampoule containing an
aliquot of a single pool of cells.
Master cell bank (MCB). A quantity of fully characterized cells of human,
animal or other origin stored frozen in liquid nitrogen in aliquots of uniform
composition, derived from the cell seed. The master cell bank is itself an
aliquot of a single pool of cells generally prepared from a selected cell clone
under defined conditions, dispensed into multiple containers and stored under
defined conditions. The master cell bank is used to derive all working cell
banks. The testing performed on a replacement master cell bank (derived from
the same cell clone, or from an existing master or working cell bank) is the
same as for the initial master cell bank, unless a justified exception is made.
Working cell bank (WCB). A quantity of cells of uniform composition
derived from one or more ampoules of the master cell bank at a finite
passage level, dispensed in aliquots into individual containers appropriately
stored, usually stored frozen in liquid nitrogen, one or more of which would
be used for production of cell culture. All containers are treated identically
and once removed from storage, are not returned to the stock.
Production cell culture. A cell culture derived from one or more ampoules
of the WCB or primary tissue used for the production of vaccines.
Adventitious agents. Contaminating microorganisms of the virus seed or
cell substrate or materials used in its culture, which may include bacteria,
fungi, mycoplasmas, and endogenous and exogenous viruses.
Single harvest. A virus suspension of one virus type harvested from cell
cultures prepared from a single production run.
Monovalent virus pool. A homogenous pool of a number of single harvests of
the same virus serotype, collected into a single vessel before clarification.
Final bulk. The finished biological preparation present in the container from
which the final containers are filled. The final bulk may be prepared from
one or more clarified monovalent virus pools and may contain one or more
virus serotypes.

142
 Filling lot (final lot). A collection of sealed final containers of vaccine that
are homogeneous with respect to the risk of contamination during the filling
process or freeze-drying. A filling lot must therefore have been filled or
prepared in one working session.
Focus forming unit (FFU). Viral infectivity identified on cell monolayers
using rotavirus-specific antiserum.
Plaque forming unit (PFU). The smallest quantity of a virus suspension that
will produce a plaque in monolayer cell cultures.
Cell culture infective dose 50% (CCID50 ). The quantity of a virus suspension
that will infect 50% of cell cultures.
Unit of infectivity (UI). Relative viral infectivity of a sample inoculated
in 96-well Vero cell monolayers measured by QPCR against a defined
reference standard preparation.

A.2 General manufacturing recommendations

The general manufacturing requirements contained in Good manufacturing
practices for biological products (13) should apply to establishments
manufacturing oral rotavirus vaccine, with the addition of the following:
Production steps and quality control operations involving manipulations of
live rotavirus should be conducted under Biosafety Level 2 (14).
Separate areas or a campaigned programme for the manufacturing of
different virus serotypes are required.
In production areas used for bulk formulation and filling, multiple
serotypes may be present at the same time and these production areas may
be campaigned with other vaccines provided sufficient cleaning validation
and product changeover data is provided.

A.3 Control of source materials

A.3.1 Cell cultures for virus production
A.3.1.1 Master cell bank and working cell bank
The use of a continuous cell line such as Vero cells (low passage, non-
tumorigenic phenotype) or diploid cells such as fetal rhesus lung cells,
FrHL-2 (well-defined non-tumorigenic phenotype) for the manufacture
of rotavirus vaccines should be based on the cell bank system. The cell
substrates and cell banks should conform with the WHO Requirements for
use of animal cells as in vitro substrates for the production of biologicals (6),
as appropriate to continuous cells and diploid cells, and should be approved
by the national regulatory authority. The maximum number of passages (or
population doublings) allowable between the cell seed and the WCB and the

143
 production cells should be approved by the national regulatory authority.
Additionally, for Vero cells, the MCB or WCB cells should be propagated to
or beyond the maximum production level and be examined for the presence
of retroviruses and tumorigenicity in an animal test system (6).
WHO has established a cell bank of Vero cells characterized in accordance
with the requirements in the forty-seventh report of the WHO Expert
Committee on Biological Standardization (6), which is available as a well-
characterized source material (15) to manufacturers for preparation of their
own MCB and WCB on application to the Coordinator, Quality Assurance
and Safety of Biologicals, WHO, Geneva, Switzerland.
The master cell bank, which is made in sufficient quantities and stored in a
secure environment is used as the source material to make manufacturer’s
working cell banks. In normal practice a master cell bank is expanded by serial
subculture up to a passage number (or population doubling, as appropriate)
selected by the manufacturer and approved by the national regulatory authority,
at which point the cells are combined to give a single pool distributed into
ampoules and preserved cryogenically to form the WCB.
The manufacturer’s WCB is used for the preparation of production cell
culture, and thus for production of vaccine batches.

A.3.1.1.1 Cell culture medium

Serum used for the propagation of cells should be tested to demonstrate
freedom from bacteria, fungi and mycoplasmas, according to the
requirements given in Part A, sections 5.2 and 5.3 of the General requirements
for the sterility of biological substances (16), as amended in 1995 and from
infectious viruses. Suitable tests for detecting viruses in bovine serum are
given in Appendix 1 of the WHO Recommendations for production and
control of poliomyelitis vaccine (Oral) (10).
Validated molecular tests for bovine viruses may replace the cell culture
tests of bovine sera.
As an additional indicator of quality, sera may be examined for freedom
from phage and endotoxin. Irradiation may be used to inactivate potential
contaminant viruses.
The acceptability of the source(s) of any components of bovine, porcine,
ovine or cervine origin used should be approved by the national regulatory
authority. These components should comply with current WHO guidelines
in relation to animal transmissible spongiform encephalopathies (5).
If trypsin is used for preparing cell cultures and aiding in virus infection, it
should be tested and found free of cultivable bacteria, fungi, mycoplasmas and
infectious viruses, especially bovine or porcine parvoviruses, as appropriate.

144
 The methods used to ensure this should be approved by the national regulatory
authority. The trypsin should be gamma irradiated if possible.
Human serum should not be used. If human albumin is used, it should meet
the revised Requirements for the collection, processing and quality control
of blood, blood components and plasma derivatives (Requirements for
Biological Substances No. 27) (17), as well as current guidelines in relation
to human transmissible encephalopathies (5).
Penicillin and other beta-lactams should not be used at any stage of the
manufacture. Other antibiotics may be used at any stage in the manufacture
provided that the quantity present in the final product is acceptable to the
national regulatory authority.
Nontoxic pH indicators may be added, e.g. phenol red at a concentration of
0.002%. Only substances that have been approved by the national regulatory
authority may be added.

A.3.1.1.2 Tests on master and working cell banks

Tests on the MCB and WCB are performed in accordance with WHO
Requirements for use of animal cells as in vitro substrates for the production
of biologicals (6, 15). It is important to show that the cell banks are free of
adventitious agents relevant to the species used in its derivation. Cell banks
should be assessed for absence of adventitious agents that may have been
present during production, including those that may be present in the source
materials used at each of these stages.
Full characterization may be performed on either the MCB or on the WCB
(6, 15).

A.3.2 Virus seeds

A.3.2.1 Virus strains
Virus strains of rotavirus used for master and working seed lots to produce
vaccines have in some cases been derived by genetic reassortment of animal
rotavirus with human rotavirus with the desired serotypes or in other cases
by multiple passages of human rotavirus in cell culture. The seed lot viruses
should comply with the specifications of this section. Viruses may be passed
in continuous, diploid, and/or primary cell lines. The candidate vaccine
strains should be approved by the national regulatory authority.
• The strains of rotavirus used in the production of candidate rotavirus
vaccines should be identified by historical records, which will include
information on the origin of each strain, method of attenuation, whether
the strains have been biologically cloned prior to generation of the master
seed lots, genetic sequence information and the passage level at which
attenuation for humans was demonstrated by clinical trials.

145
 • The vaccine production strain(s) should have been shown by appropriate
laboratory tests (see A.3.2.3) and/or clinical studies to yield vaccines that
are safe and efficacious in humans. Only strains approved by the national
regulatory authority should be used.
• The immunogenicity of the virus strains, based upon the quantity of
infectious virus of each serotype present in the vaccine that induces
seroconversion when susceptible individuals are immunized with the
vaccine, should be established in a dose–response study. Any potential
interference or potentiation between the serotypes in an infectivity assay
should be evaluated prior to establishing this value. The immunizing dose
established in this way serves as a basis for establishing parameters for
potency at the time of release, stability and expiry date.

A.3.2.2 Virus seed lot system

The production of vaccine should be based on the virus master seed lot
and virus working seed lot system. Seed lots should be prepared in the
same type of cells and by the same production process as those used for
production of final vaccine. Virus seed lots should be stored in a dedicated
temperature-monitored refrigerator at a temperature that ensures stability
and the storage arrangement should ensure appropriate security of the virus
seed lots.
Seed lots of rotavirus used in the production of live attenuated rotavirus
vaccine (oral) should be identified by historical records, which should include
information on their origin. Only virus seed lots that are approved by the
national regulatory authority should be used (see General considerations).
The virus master seed lot is generally produced from a biological clone of the
attenuated parent virus (animal/human reassortant or cell-culture passaged
human virus) and is made in sufficient quantities and stored in a secure
environment and is used as the source material to make the manufacturer’s
virus working seed lots. Either the virus master seed lots or the virus
working seed lots should be fully characterized and be tested extensively for
adventitious agents, and approved by the national regulatory authority.
The virus master seed lot also serves as a benchmark against which to
compare virus produced by subsequent passage in cell culture or shed virus
following vaccination.
The manufacturer’s virus working seed lot is used for the production of
vaccine batches and is prepared from the master virus seed lot. It is
recommended that a large lot of virus working seed be set aside as the
basic material that the manufacturer should use for the preparation of each
batch of vaccine. The virus working seed lot should be prepared by defined
number of passages from the virus master seed lot by a method and a passage
level from the original virus seed that is established through clinical and

146
 vaccine development studies. Once the passage level of the working seed
lot is established, it may not be changed without approval from the national
regulatory authority.

A.3.2.3 Tests on virus master and working seed lots

A.3.2.3.1 Identity
Each seed lot should be identified by virus type by an immunological assay
or by sequencing. A molecular identity test, electropherotyping, is one
recognized method for identifying rotaviruses. Comparing human and animal
rotaviruses and differentiating between subgroups can be achieved through
electropherotyping. Electropherotyping can be used as a means for identifying
gross alterations in genomic differences between the parent wild type viruses
and/or animal donor virus (if employed) and the vaccine strains. The identity
of the vaccine virus is confirmed by comparing the electrophoretic profile
of the vaccine virus to the RNA electrophoretic profile of a known human
rotavirus serotype. Other molecular identity tests may include RNA/RNA
hybridization and enzyme restriction maps or genetic sequences of PCR-
amplified VP7 gene segments. The tests should be approved by the national
regulatory authority.

A.3.2.3.2 Genotype/phenotype characterization

Genotype and phenotype stability of the seed lots upon passage should be
measured using relevant assays to ensure uniformity of vaccine lots. Genetic
characterization of the viruses has played a major role in the development
of the vaccines already licensed or currently nearing licensure. It helps to
assess the genetic and phenotypic stability of the seed lots on passage, and
molecular methods have been used intensively. Other approaches including
in vitro phenotypic markers might be considered.
Genetic characterization of the seed lot viruses has included determination
of the nucleotide sequence of the complete viral genome, analysis of the
molecular basis of the attenuated phenotype and determination of the
genetic stability of the virus seed lots by comparison of the nucleotide
sequence of the viral genome at different passage levels. It should be
noted that full-length sequencing may not identify minority populations of
variants that may be present in vaccines. In some studies the entire genomic
sequence of all 11 gene segments of the virus master seed lot has been
sequenced and compared with the sequence of the virus working seed lot,
vaccine production lots and vaccine virus shed in stools of vaccinees. If the
attenuation loci are known, sequencing around these sites can be performed
instead of on the entire genome. Given the frequency of errors of the RNA
replicating polymerase, base changes are likely to be found. The potential
impact of any changes observed (silent or in protein-coding regions) should
be evaluated.

147
 Specific assays for markers of attenuation are still in development and the
manufacturer should make an effort to investigate appropriate assays that
are relevant to their vaccine.

A.3.2.3.3 Tests for bacteria, fungi and mycoplasmas

Each virus seed lot should be tested for bacterial, fungal and mycoplasmal
contamination by appropriate tests as specified in Part A, sections 5.2 and
5.3, of the General requirements for the sterility of biological substances,
as amended in 1995 (16).

A.3.2.3.4 Tests for adventitious viruses

Each virus seed lot should be tested in cell cultures for adventitious viruses
appropriate to the origin and the passage history of the seed virus (18).
Neutralization of rotavirus is necessary for many tests because the virus
is cytopathogenic. Antisera used for this purpose should be shown to be
free from antibodies that may neutralize specific adventitious viruses being
tested for. If neutralization of rotavirus is not possible, the test sample may
be passaged in trypsin-free media prior to initiating the assay, to reduce the
ability of rotavirus to infect the indicator cell substrates. The cells inoculated
should be observed microscopically for cytopathic changes. At the end of the
observation period, the cells should be tested for haemadsorbing viruses.
Each master or working seed lot should also be tested in animals that may
include guinea-pigs, mice and embryonated chicken eggs, as appropriate.
For test details refer to the WHO Requirements for measles vaccines (Live)
(19), and the European Pharmacopoeia, 2002 (18)
Additional testing for specific adventitious viruses may be performed, for
example, using PCR amplification techniques.

A.3.2.3.5 Virus concentration

Each seed lot should be assayed for infectivity in a sensitive assay in a cell
culture system. An immunofocus or plaque forming assay may be used in
MA-104, Vero or other sensitive cells to determine virus concentration. The
assay is based on the visualization of infected areas (plaques or focus of
infection) of a cell monolayer directly or by probing with rotavirus-specific
antibodies. Results should be recorded as focus-forming units (FFU/ml) or
plaque-forming units (PFU/ml).
A cell culture infectious dose assay may also be used to determine virus
concentration. Results should be recorded as CCID50/ml.
Alternatively, quantitative PCR detection of virus replication in a cell
culture system may be used to provide an appropriate measure of infectivity.
Results should be recorded as units of infectivity (UI/ml).

148
 The detailed procedures for carrying out the tests and for interpreting the
results should be those approved by the national regulatory authority.

A.4 Control of vaccine production

A.4.1 Control of cell cultures
From the cell suspension used to prepare cell cultures for growing attenuated
rotavirus, an amount of processed cell suspension equivalent to at least 5%
or 500 ml of cell suspension, whichever is greater, should be used to prepare
control cultures of uninfected cells. If fermenter technology is used, the size
and treatment of the cell sample to be examined should be approved by the
national regulatory authority.
These control cultures should be observed microscopically for changes
attributable to the presence of adventitious agents for at least 14 days after
the day of inoculation of the production cultures or at the time of final virus
harvest if this is later. The control cultures should be incubated under essentially
similar conditions to those used for the production cultures with agreement of
the national regulatory authority. At the end of the observation period, fluids
collected from the control culture from each single virus harvest should be
tested for the presence of adventitious agents as described below (A.4.1.2).
Samples that are not tested immediately should be stored at –60 ºC or below.
If any test shows evidence of the presence of adventitious agents in control
cultures, the harvest of virus from these cultures should not be used for
vaccine production.
For the test to be valid, not more than 20% of the control culture vessels
should have been discarded for nonspecific accidental reasons by the end
of the test period.

A.4.1.1 Tests for haemadsorbing viruses

At the end of the observation period, cells comprising no less than 25% of
the control cells should be tested for the presence of haemadsorbing viruses,
using guinea-pig red blood cells. If the red blood cells have been stored, the
duration of storage should not have exceeded 7 days, and the temperature
of storage should have been in the range of 2–8 ºC.
In some countries, the national regulatory authority requires that additional
tests for haemadsorbing viruses be performed using other species of red
blood cells including those from humans (blood group O), monkeys and
chickens (or other avian species). In all tests readings should be taken after
incubation for 30 minutes at 0–4 ºC, and again after a further incubation
for 30 minutes at 20–25 ºC. A further reading for the test with monkey red
blood cells should be taken after an additional incubation for 30 minutes at
34–37 ºC.

149
 For the tests to be valid, not more than 20% of the culture vessels should
have been discarded for nonspecific accidental reasons by the end of the
test period.

A.4.1.2 Tests for other adventitious agents

At the end of the observation period, a sample of the pooled fluid from each
group of control cultures should be tested for adventitious agents. For this
purpose, 10 ml of each pool should be tested in the same cells, but not the
same batch of cells, as those used for the production of virus, and additional
10-ml samples of each pool should be tested in human cells and at least one
other sensitive cell system.
Each sample should be inoculated into bottles of these cell cultures in such
a way that the dilution of the pooled fluid in the nutrient medium does
not exceed 1 in 4. The area of the cells should be at least 3 cm2 per ml of
pooled fluid. At least one bottle of each kind of cell cultures should remain
uninoculated to serve as a control.
The inoculated cultures and control cultures should be incubated at a
temperature of 35–37 ºC and should be observed for cytopathic effects for
a period of at least 14 days.
For the tests to be valid, not more than 20% of the culture vessels should have
been discarded for nonspecific accidental reasons by the end of the test period.
Some national regulatory authorities require that, at the end of this
observation period, a subculture is made by inoculating the fluid onto
fresh cells in the same culture system and observed for at least 7 days.
Furthermore, some national regulatory authorities require that these cells
should be tested for the presence of haemadsorbing viruses.

A.4.1.3 Identity test

At the production level, the cells should be identified by means of tests
approved by the national regulatory authority. Suitable methods are, but are
not limited to, biochemical tests (e.g. isoenzyme analyses), immunological
tests (e.g. HLA assays), cytogenetic tests (e.g. for chromosomal markers),
and tests for genetic markers (e.g. DNA fingerprinting).

A.4.2 Cell cultures for vaccine production

A.4.2.1 Tests for adventitious agents
On the day of inoculation with the virus working seed lot, each cell culture
and/or cell culture control should be examined for degeneration caused by
infectious agents. If such examination shows evidence of the presence in
a cell culture of any adventitious agents, the culture should not be used
for vaccine production. Prior to infection, samples of each cell culture are
removed for sterility and mycoplasma testing.

150
 If animal serum is used for cell cultures before the inoculation of virus,
it should be removed and replaced with serum-free maintenance medium,
after the cells have been washed with serum-free medium, if appropriate.

A.4.2.2 Tests for bacteria, fungi and mycoplasmas

A volume of at least 20 ml of the pooled supernatant fluids from the production
cell culture should be tested for bacterial, fungal and mycoplasmal sterility
by appropriate tests as specified in Part A, sections 5.2 and 5.3, of the
General requirements for the sterility of biological substances, as amended
in 1995 (16).

A.4.3 Control of single harvests and monovalent virus pools

A.4.3.1 Virus inoculation
Cell cultures are inoculated with rotavirus working seed at a defined
multiplicity of infection. After viral adsorption, cell cultures are fed with
maintenance medium and incubated within a defined temperature range and
for a defined period of time, usually established based upon the degree of
cytopathic effect.
The range of multiplicity of infection, temperature, pH and time period
of incubation will depend on the vaccine strain and production, a defined
range should be established by the manufacturer and be approved in the
marketing authorization by the national regulatory authority.

A.4.3.2 Monovalent virus pools

A virus single harvest is harvested within a defined time period post-
inoculation. A monovalent virus pool may be the result of one or more single
harvests (from multiple tissue culture flasks, cell factories or bioreactors) in
which all harvests were derived from one or a small number of ampoules of
the WCB and the same virus working seed lot recovered at the same time.
If multiple single harvests are taken, each single harvest should be sampled
for testing, stabilized and stored under suitable conditions until pooling. No
antibiotics should be added at the time of harvesting or at any later stage
of manufacture. Samples of monovalent virus pools should be taken for
testing and stored at a temperature of –60 ºC or below.

A.4.3.3 Tests on single harvest or monovalent virus pools

Tests may be done on single harvest or on virus pool. If the tests are done on
the virus pool, the protocol should be approved by the national regulatory
authority.

A.4.3.3.1 Sampling
Samples required for the testing of virus harvests should be taken immediately
on harvesting prior to further processing. If the tests for adventitious agents

151
 as described in Part A, section A.4.3.3.4, are not performed immediately,
the samples taken for these tests should be kept at a temperature of –60 ºC
or below and subjected to no more than one freeze–thaw cycle.

A.4.3.3.2 Identity
Each single harvest or virus pool should be identified as the appropriate
rotavirus serotype by immunological assay or by a molecular-based assay,
e.g. electropherotyping, RNA/RNA hybridization or PCR. The tests should
be approved by the national regulatory authority.

A.4.3.3.3 Tests for bacteria, fungi and mycoplasmas

Each single harvest or virus pool should be shown to be free from bacterial,
fungal and mycoplasmal contamination by appropriate tests as specified in
Part A, sections 5.2 and 5.3, of the General requirements for the sterility of
biological substances, as amended in 1995 (16).

A.4.3.3.4 Tests for adventitious agents

For the purposes of the recommendations set out in this section of Part A,
the volume of each single harvest or virus pool taken for neutralization and
testing should be at least 10 ml and should be such that a total of at least
50 ml or the equivalent of 500 doses of final vaccine, whichever is the
greater, has been withheld from the corresponding final bulk.
Each virus pool should be tested in cell cultures for adventitious viruses
appropriate to the passage history of the seed virus (18). Neutralization of
rotavirus is necessary for many tests because the virus is cytopathogenic.
Antisera used for this purpose should be shown to be free from antibodies
that may neutralize the adventitious viruses being tested for. If neutralization
of rotavirus is not possible, the test sample may be passaged in trypsin-
free media prior to initiating the assay to reduce the ability of rotavirus to
infect the indicator cell substrates. The cells inoculated should be observed
microscopically for cytopathic changes. At the end of the observation
period, the cells should be tested for haemadsorbing viruses.
Additional testing for specific adventitious viruses may be performed, for
example using PCR amplification techniques.

A.4.3.3.5 Virus concentration

Each virus pool should be assayed for infectivity using a sensitive assay in
cell cultures to monitor the consistency of production.
An immunofocus or plaque assay may be used in MA-104, Vero or other
sensitive cells to determine virus concentration. The assay is based on the
visualization of infected areas of a cell monolayer directly or by probing

152
 with rotavirus serotype-specific monoclonal antibodies. Results should be
recorded as FFU/ml or PFU/ml.
Alternatively, quantitative PCR detection of virus replication in a cell
culture system may be used to provide an appropriate measure of infectivity.
Results should be recorded as units of infectivity (UI/ml).
A cell culture infectious dose assay may also be used to determine virus
concentration. Results should be recorded as CCID50/ml.
The detailed procedures for carrying out the tests and for interpreting the
results should be those approved by the national regulatory authority.

A.4.3.3.6 Tests for consistency of virus characteristics

Tests for consistency of virus characteristics are performed during vaccine
development, and process validation and not intended for batch release.
Examples of studies that might be considered to characterize rotavirus are
given here. Tests should be sought to compare the rotavirus in the harvest
pool with the master seed virus, or suitable comparator, to ensure that the
vaccine virus has not undergone critical changes during its multiplication
in the production culture system. Phenotypic or genotypic characteristics
(genomic sequence analysis) may be suitable. Examples of evidence to
support the consistent quality of the virus produced may include in vitro
growth characteristics, thermal stability profile, the ratio of infectious
(triple-shelled) to non-infectious (double-shelled) particles produced and the
stability of the genomic sequence through multiple cell culture passages.
Other aspects of process consistency may also be monitored and validated,
such as process impurities and residuals as residual host cell protein,
residual cellular DNA, endotoxin, bovine serum, trypsin and antibiotics.
Their reduction during processing can be monitored to assess consistency
of the manufacturing process. The reduction level should be accepted by the
national regulatory authority.

A.4.3.3.7 Storage
Virus pools should be stored at a temperature that will ensure stability until
formulation.

A.4.3.4 Control of clarified monovalent virus pool (bulk)

The monovalent virus pool may be clarified or filtered to remove cell debris
and stored at a temperature that ensures stability before being used to
prepare the final bulk.

A.4.3.4.1 Sampling
Samples of the clarified virus pool should be taken immediately after
clarification and prior to further processing to ensure that no cells or cell

153
 debris is left. Samples should also be tested as described in this section. If
not tested immediately, the samples should be kept at a temperature below
–50 °C until testing is done.

A.4.3.4.2 Sterility
The clarified virus pool should be tested for bacterial and fungal sterility
according to the requirements in Part A, sections 5.2 of the General requirements
for the sterility of biological substances (16), by acceptable methods approved
by the national regulatory authority.

A.4.3.4.3 Virus concentration

Each clarified virus pool should be assayed for infectivity in a sensitive
assay in cell cultures to monitor the consistency of production.
An immunofocus or plaque assay may be used in MA-104, Vero or other
sensitive cells to determine virus concentration. The assay is based on the
visualization of infected areas of a cell monolayer directly or by probing
with rotavirus serotype-specific monoclonal antibodies. Results should be
recorded as FFU/ml or PFU/ml.
Alternatively, quantitative PCR detection of virus replication in a cell
culture system may be used to provide an appropriate measure of infectivity.
Results should be recorded as units of infectivity (UI/ml).
A cell culture infectious dose assay may also be used to determine virus
concentration. Results should be recorded as CCID50/ml.
The detailed procedures for carrying out the tests and for interpreting the
results should be those approved by the national regulatory authority.

A.4.3.4.4 Tests for residual cellular DNA

For viruses grown in continuous cells, the virus pool should be tested for
residual cellular DNA. The removal process, at production scale, should be
shown to consistently reduce the level of cellular DNA. The limit should
be established and approved by the national regulatory authority. Based
on relevant studies in animals, WHO established the acceptable limit of
not more than 100 µg of cellular DNA per human dose, which is likely
to provide an adequate margin of safety for orally-administered vaccines.
Consideration should also be given to determining the size of residual
cellular DNA as part of the validation process.
These tests may be conducted on the final product, in which case, the product
should also conform to the specifications described here.
These tests may be omitted from routine testing, with the agreement of the
national regulatory authority, if the manufacturing process is validated as
consistently achieving the specification.

154
A.4.4 Final bulk
The final bulk should be aseptically prepared from one or more serotypes
each derived from one or more virus pools obtained from substrates of
which control cultures pass the tests specified in Section A.4.1. The virus
concentration in the final formulation should be sufficient to ensure the dose
which is consistent with that shown to be safe and effective in human clinical
trials. The virus pools and final bulk should pass the tests specified in Sections
A.4.3.3 and A.4.4.1.
The operations necessary for preparing the final bulk lot should be conducted
in such a manner as to avoid contamination of the product.
In preparing the final bulk, any substances such as diluents or stabilizers
that are added to the product should have been shown to the satisfaction of
the national regulatory authority not to impair the safety and efficacy of the
vaccine at the concentration used.

A.4.4.1 Tests on the final bulk

A.4.4.1.1 Sterility
Each final bulk suspension should be tested for bacterial and fungal sterility
according to Part A, sections 5.2 of the General requirements for the sterility
of biological substances (16), or by a method approved by the national
regulatory authority.

A.4.4.2 Storage
Until the bulk is filled into containers, the final bulk suspension should be
stored under conditions shown by the manufacturer to allow it retain the
desired biological activity.

A.5 Filling and containers

The requirements concerning filling and containers given in Good
manufacturing practices for biological products (13) should apply to
vaccine filled in the final form.
Care should be taken to ensure that the materials of which the container and,
if applicable, transference devices and closure are made do not adversely
affect the quality of vaccine and its diluent.
When a freeze-drying process is used for vaccine production, its validation
should be submitted to the national regulatory authority for approval.
The manufacturers should provide the national regulatory authority with
adequate data to prove the stability of the product under appropriate
conditions of storage and shipping.

155
A.6 Control tests on final lot
Samples should be taken from each final lot for the following tests. Both
freeze-dried vaccine and its diluent, if applicable, should be tested and
should fulfil the requirements discussed in this section.

A.6.1 Vaccine
A.6.1.1 Inspection of final containers
Each container in each final lot should be inspected visually and those
showing abnormalities should be discarded.

A.6.1.1.1 Appearance
The appearance of the freeze-dried or liquid vaccine should be described with
respect to its form and colour. In the case of freeze-dried vaccines, a visual
inspection should be performed of the freeze-dried vaccine, its diluent and
the reconstituted vaccine. If reconstitution with the product diluent does not
allow for the detection of particulates, an alternative diluent may be used.

A.6.1.2 Identity
The virus in one or more individually labelled final containers should
be identified as rotavirus and, for multivalent vaccine formulations each
serotype should be identified, by appropriate methods approved by the
national regulatory authority. Methods such as plaque neutralization, and
immunofocus assays in cell culture are suitable to identify the presence of
rotavirus using rotavirus-specific polyclonal antiserum. PCR may also be
appropriate, in this case the virus titration by quantitative PCR may serve
as the identity test.

A.6.1.3 Sterility
Liquid or reconstituted vaccine should be tested for bacterial and fungal
sterility according to the requirements in Part A, section 5.2 of the General
requirements for the sterility of biological substances (16), or by the methods
approved by the national regulatory authority.

A.6.1.4 pH
The pH of the final lot should be tested in a pool of final containers and
an appropriate limit set to guarantee virus stability. In case of freeze-dried
vaccines, pH should be measured after reconstitution of the vaccine with
the diluent.

A.6.1.5 Residual moisture

The residual moisture in a representative sample of each freeze-dried lot
should be determined by a method approved by the national regulatory
authority and an appropriate limit set to ensure vaccine stability.

156
A.6.1.6 Virus concentration
The virus concentration in each of at least three final containers of the
rotavirus vaccine final lot should be assayed individually for infectivity in
a sensitive assay system in which interference or potentiation between the
serotypes present in the vaccine does not occur.
An immunofocus or plaque assay may be used in MA-104, Vero or other
sensitive cells to determine virus concentration. The assay is based on the
visualization of infected areas of a cell monolayer directly or by probing
with rotavirus serotype-specific monoclonal antibodies. Results should be
recorded as FFU/ml or PFU/ml.
If immunological based assays are used, the specificity and lack of cross-
reactivity of the antiserum must be verified. Alternatively, quantitative PCR
detection of virus replication in a cell culture system may be used to provide
an appropriate measure of infectivity. Results should be recorded as units of
infectivity (UI/ml).
A cell culture infectious dose assay may also be used to determine virus
concentration. Results should be recorded as CCID50/ml.
The titre of each individual serotype should be determined and should fall
within the specifications for potency. The assay method should include
suitable qualified reference reagents for each serotype in the vaccine. The
detailed procedures for carrying out the tests and for interpreting the results
should be those approved by the national regulatory authority.
The national regulatory authority should approve a reference preparation
of live attenuated rotavirus vaccine for use in tests to determine virus
concentration.
Freeze-dried vaccine should be reconstituted with its diluent to determine
virus concentration. A validated alternative diluent may be needed if the
approved diluent is toxic to the cell cultures used in the assay. If a different
diluent is used for this test, data to allow a comparison between the results
with both diluents should be submitted for the approval of the national
regulatory authority.
Internal consistency limits should be established by the manufacturer taking
into account the vaccine dose shown to be safe and effective in human
clinical trials. Specifications for virus concentration should essentially
specify the minimum titre guaranteed to be contained in one human dose
and this should be agreed with the national regulatory authority.

A.6.1.7 Accelerated stability tests

A representative number of the final containers should be exposed to an
elevated temperature for a defined time period, using conditions based on

157
 the manufacturer’s experience. The geometric mean of infectious virus titre
of the containers that have been exposed should not have been decreased by
more than a specific amount during the period of exposure. Estimation of
the virus titre of non-exposed and exposed vials should be made in parallel
and results expressed in terms of PFU, FFU, CCID50 or UI per human dose.
The maximum allowable loss of titre during the accelerated stability test
should be confirmed on the basis of the manufacturer’s experience and
approved by the national regulatory authority. For a multivalent vaccine,
if there is no significant difference in the virus loss between serotypes, the
loss may be based upon total virus concentration.

A.6.2 Diluents
The requirements given in Good manufacturing practices for pharmaceutical
products (20) should apply for the manufacturing and control of diluents
used to reconstitute live attenuated rotavirus vaccines and, if required, the
antacid buffer used. An expiry date should be established for the diluent
based upon stability data. If an antacid is to be used, the stability of the
rotavirus in the presence of the antacid should be confirmed. For lot release
of the diluent, tests for identity, appearance, pH, volume, sterility, and the
content of key components should be done.

A.7 Records
The requirements given in Good manufacturing practices for biological
products (13) should apply.

A.8 Samples
The requirements given in Good manufacturing practices for biological
products (13) should apply.

A.9 Labelling
The requirements given in Good manufacturing practices for biological
products (13) should apply, with the addition of the following.
The label on the carton enclosing one or more final containers, or the leaflet
accompanying the container, should include the following information:
— a statement that the candidate vaccine fulfils Part A of these
Guidelines;
— a statement of the nature of the preparation, specifying the designation of
the strain(s) of rotavirus contained in the vaccine, the minimum amount
of virus contained per human dose, the origin of the substrates used
in the preparation of the vaccine and whether the vaccine strains were
derived by molecular methods;

158
 — the fact that the vaccine is not to be injected;
— a statement of the nature and quantity, or upper limit, of any antibiotic
present in the vaccine;
— an indication that contact with disinfectants is to be avoided;
— a statement regarding the concomitant administration of rotavirus
vaccine with other oral vaccines such as oral poliovirus vaccine and with
other non-orally administered vaccines;
— a statement concerning administration to HIV-positive or other
immunocompromised individuals;
— a statement indicating the volume and nature of the diluent to be added
to reconstitute the vaccine, and specifying that the diluent to be used is
that supplied by the manufacturer;
— a statement that after the vaccine is reconstituted, it should be used
without delay, or if not used immediately, stored under defined conditions
and in the dark for a maximum period defined by stability studies;
— a statement concerning storage conditions (temperature), expiry date,
volume and instructions for reconstitution. Only one storage temperature
and expiry date should be stated on the label and leaflet; and
— a statement indicating whether an antacid is to be given prior to or in
combination with the vaccine at the time of vaccination.
It is desirable for the label or the leaflet to carry the names of both the
producer and the source of the bulk material if the producer of the final
vaccine did not prepare it.

A.10 Distribution and shipping

The requirements given in Good manufacturing practices for biological
products (13) should apply.
Vaccine shipments should be maintained at the approved temperatures and
parcels should contain cold-chain monitors.
For some products, freezing of the diluent should be avoided.

A.11 Storage and expiry date

The statements concerning storage temperature and expiry date of the
vaccine and its diluent, if applicable, that appear on the label and in the
leaflet, as recommended in Good manufacturing practices for biological
products (13), should be based on experimental evidence and should be
submitted for approval to the national regulatory authority.

A.11.1 Storage conditions

Before being distributed by the manufacturing establishment or before being
issued from a storage site, the vaccine should be stored at a temperature

159
 shown by the manufacturer to be compatible with a minimal titre loss.
The maximum duration of storage should be fixed with the approval of the
national regulatory authority and should be such as to ensure that all quality
specifications for final product including the minimum titre specified on the
label of the container (or package) will still be maintained until the end of
the shelf-life.

A.11.2 Expiry date

The expiry date should be defined on the basis of shelf-life and supported
by the stability studies with the approval of the national regulatory authority
and should relate to the date of the last satisfactory determination of virus
concentration, performed in accordance with Part A, section A.4.3.3.5, i.e.
the date on which the test system was inoculated.
The expiry dates for the vaccine and the diluent may be different.

Part B. Nonclinical evaluation of live

attenuated rotavirus vaccines
Nonclinical evaluation of rotavirus vaccines should be based on existing
WHO guidelines (1); however, the following rotavirus-specific issues
should be considered.
In animal studies of rotavirus vaccines, oral dosing with antacid which
corresponds to that intended for use in clinical trials is necessary; for
example, in mice or in rats (using a strain of rat susceptible to human
rotavirus infection) with the full human dose of vaccine. As rotavirus is not
neurotropic, and neural tissue passage has not been used in the derivation
of any of the vaccine strains, a neurovirulence test for each batch is not
justified, nor is there a need to test the master or working seed lots. As with
all live attenuated vaccines, special attention should be paid to evaluating
the stability of attenuation phenotype by appropriate in vitro and in vivo
assays. Nonclinical experimental studies to predict risks from excreted
vaccine virus are not expected to be informative.
On the basis of a review of the literature, manufacturers should evaluate
any risks to humans or animals if the vaccine virus is subsequently shown
to be shed from vaccinees, taking into account the likelihood of excreted
vaccine virus reassorting with wild-type viruses. Finally, the pathogenic
mechanisms of intussusception associated with oral rotavirus vaccination
are currently unknown and no suitable animal model is readily available
to evaluate the risk of intussusception. Nevertheless, manufacturers should
keep abreast of the evolving scientific knowledge and plan their strategy for
preclinical vaccine evaluation accordingly.

160
Part C. Clinical evaluation of live attenuated
rotavirus vaccines
The WHO guidelines on regulatory expectations for clinical evaluation are
appropriate for development of rotavirus vaccines (2). Consideration should
be given to the following rotavirus-specific issues.

C.1 Immune responses to rotavirus

Live attenuated rotavirus vaccines have been developed on the basis of
evidence that natural rotavirus infections elicit protective immune responses,
particularly against future severe rotavirus disease. For a period after rotavirus
vaccination, much of the serum IgA is rotavirus-specific, so that rotavirus
serum IgA levels act as a measure of seroconversion. Rotavirus serum IgA
antibody responses have been used as measures of vaccine immunogenicity
of all the candidate live attenuated rotavirus vaccines evaluated so far. Thus
rotavirus serum IgA antibody responses should also be recorded for other
vaccine candidates.
Nevertheless it should be noted that rotavirus serum IgA is not an immune
correlate of protection for rotavirus vaccines. A role for neutralizing antibody
in protective efficacy has been suggested, but the correlation between
neutralizing antibody titres and clinical efficacy has not been demonstrated.
These data have been recently reviewed for the different live attenuated
rotavirus vaccine candidates (21).

C.2 Special considerations

There are additional specific issues that may affect the safety and efficacy
of oral rotavirus vaccines. Factors which are considered important for
evaluation pre-licensure include the following:

C.2.1 Vaccine virus shedding and transmission

Manufacturers should undertake studies to determine the amounts of the
vaccine virus (if applicable, by serotype) shed by vaccinees and the duration
of shedding. They should assess the transmissibility of vaccine strains to
unvaccinated people during clinical studies of safety and efficacy. If vaccine
virus is shed in sufficient quantities to make transmission feasible and/or if
transmission from vaccines is demonstrated, then studies of the likelihood
of reversion to wild type and the likelihood of the vaccine virus reassorting
with wild-type rotaviruses are indicated.

C.2.2 Dose regimen

Clinical trials should be designed to determine the number of doses to be
administered to elicit a measurable immune response and clinical efficacy.

161
 Based on experience to date, at least two or three doses of live attenuated
oral rotavirus vaccines are necessary to overcome interfering factors which
may include maternal antibodies or concomitant administration of oral
poliovirus vaccine (OPV).

C.2.3 Concomitant administration with routine childhood vaccines

As it is intended that live attenuated rotavirus vaccines would be incorporated
into routine childhood immunization programmes, it is important to generate
information on immune responses to co-administration with the routine
childhood vaccines at the target ages of administration. In particular, co-
administration with OPV should be studied, as both are live attenuated oral
vaccines.

C.2.4 Vaccine safety

After licensure of a rhesus reassortant rotavirus vaccine, postmarketing
surveillance in one country revealed an association with intussusception, a rare
serious adverse event following administration of the vaccine to infants. This
occurred in infants given the vaccine at an older age than that recommended
by the manufacturer. In light of this experience, rotavirus vaccines must be
assessed for any vaccine-attributable intussusception within 30 days after
each dose of vaccine. However, even very large pre-licensure studies (for
example in 70 000 infants) cannot rule out an association between vaccine
and intussusception. They can only provide an estimate of the relative and
absolute risk compared to placebo together with 95% confidence intervals
that give an idea of the degree of risk that cannot be excluded. Such data are
generated in clinical trials in selected populations and in which the doses of the
regimen under study are controlled, so that the data may not predict risk in the
post-licensure period. If post-licensure studies, currently in progress, reveal
an age-related risk of vaccine-attributable intussusception, specifications may
need to be considered for the age at first dose of vaccine (22).
In planning the size of future clinical safety trials for a new rotavirus vaccine,
assumptions need to be made about the likely background incidence of
intussusception. However, it may not be feasible to obtain accurate data on
naturally occurring intussusception in all countries in which a study is to
be performed. The calculation should state the degree of risk of vaccine-
associated intussusception that the study should be able to assess. With
appropriate justifications, sample sizes of < 70 000 for future pre-licensure
clinical trials of new rotavirus vaccines may be acceptable.

C.2.5 Definition of clinical end-point in trials of protective efficacy

The results from clinical trials to date, show that rotavirus vaccines are more
efficacious against severe illness than against mild or moderate disease.
Standard definitions of a diarrhoea episode and the severity of illness

162
 (recommended as the primary end-point of efficacy studies) are critical. Several
methods for assessing the “severity” of rotavirus infection are available, for
example, based on the scale described by Ruuska et al (23) and Clark et al
(24). Hospitalization may not be appropriate as the only end-point because
this is a context-sensitive situation which may not always reflect the severity
of illness. However, hospitalization combined with grading of severity on an
appropriate scale should give comparable results between study sites.
Usually the primary analysis is focused on prevention of acute rotavirus
gastroenteritis due to the serotype(s) in the vaccine. It is acceptable that such
primary analysis be based on those cases of acute rotavirus gastroenteritis
that occur from 14 days onwards after administration of the last dose of the
tested regimen. Consideration should also be given to sensitivity analysis to
investigate efficacy against all cases of acute rotavirus gastroenteritis that
occur from the time of the first dose onwards.

C.2.6 Other factors

Several other factors need to be considered when assessing new rotavirus
vaccines, although some of these data may be generated post-licensure. Such
data include those on use in special populations and in diverse geographical
regions (developed or developing country settings), and interference by
other health factors.

C.2.6.1 Vaccine safety in immunocompromised infants

No data on immunocompromised children are currently available. As with
other live vaccines, live attenuated rotavirus vaccines should be shown to
be safe in HIV-infected infants prior to general implementation in countries
with high prevalence of HIV. Natural rotavirus infection has not been shown
to be more serious or to be associated with greater safety risks in HIV-
infected infants than in non-HIV-infected infants, although a slightly longer
period of shedding of rotavirus was observed in one study.

C.2.6.2 Seasonality of rotavirus infection

Depending on where the pre-licensure efficacy data were generated (i.e. in
areas with seasonal or year-round rotavirus infection), it may be appropriate
to generate further data in alternative geographical areas. For example, in
those settings with a marked seasonal pattern of rotavirus infection, the
efficacy of rotavirus vaccine may be higher when it is administered shortly
before the peak period for rotavirus circulation, possibly due to a booster
effect of natural infection during the season. For this reason, it may be useful
to assess efficacy in sequential seasons.

C.2.6.3 Diverse geographical and population settings

The efficacy and/or effectiveness of rotavirus vaccine needs to be studied
in diverse geographical regions with different populations and different

163
 socioeconomic conditions. Generally, data from clinical trials conducted in
one part of the world would not necessarily be predictive of vaccine efficacy
in other parts of the world (3). For example, clinical trials have shown variable
levels of rotavirus-specific serum IgA with the same vaccine in populations
in developed and less developed or developing countries. Furthermore, the
assessment of protection provided by each serotype in the vaccine and/or
against serotypes not in the vaccine (i.e. cross-protection) is inherently
limited by the serotypes circulating in the countries where the study/studies
was/were done during the observation period. Studies may lack the power
to demonstrate efficacy by specific serotype. In these circumstances, data
must be assessed on a case-by-case basis when considering any claims for
vaccine serotype-specific and heterotypic protection.
Many factors are likely to be responsible for observed differences in
immunogenicity and protective efficacy of the rotavirus vaccines in diverse
geographical settings. In designing rotavirus vaccine trials (pre- or post-
licensure), consideration should be given to controlling for the possible effects
of factors such as the predominant circulating rotavirus serotypes, other
vaccinations, malnutrition, vitamin and mineral deficiencies, concurrent enteric
infections, malaria, parasitic infections, HIV, hepatitis, immunodeficiences,
maternally transmitted antibodies and breastfeeding.

C.2.6.4 Interchangeability
At present there are no data on safety, immunogenicity or efficacy when
more than one vaccine type is used in one infant in the priming series.
Although there may be difficulties in interpreting immunogenicity data
in terms of predicting efficacy, and of complexities that would arise in
studies of safety and efficacy when more than one vaccine type is given
(e.g. a monovalent human rotavirus strain live attenuated vaccine followed
by a multivalent human-bovine reassortant rotavirus strain live attenuated
vaccine), such data would be welcome.

C.3 Postmarketing studies and surveillance

Large-scale postmarketing surveillance studies of safety and effectiveness are
indicated once vaccines have been approved. In the case of rotavirus, knowledge
of distribution of strains, establishment of country-specific laboratory
expertise and education of physicians and policy-makers will be required
to conduct these studies effectively. The aims of postmarketing surveillance
of rotavirus vaccines may include evaluation of vaccine effectiveness in
different populations and in settings in which more varied rotavirus serotypes
predominate; the impact of emerging strains of rotavirus; monitoring the risk
of vaccine-associated intussusception and of other potential adverse events;
and evaluation of compliance with the recommended vaccination schedule. A
standardized protocol for postmarketing surveillance is under development.

164
Part D. Guidelines for national regulatory
authorities
D.1 General
The general recommendations for control laboratories given in the Guidelines
for national authorities on quality assurance for biological products (25),
which specify that no new biological substance should be released until
consistency of manufacturing and quality as demonstrated by a consistent
release of batches has been established, should apply.
The detailed production and control procedures and any significant changes
made to them should be discussed with and approved by the national regulatory
authority. The national regulatory authority should obtain the working reference
from the manufacturers to establish a national working reference preparation
for use until an international reference reagent becomes available.

D.2 Release and certification

A vaccine lot should be released only if it fulfils the national requirements
and/or Part A of the present Guidelines. A protocol based on the model
given in Appendix 1, signed by the responsible official of the manufacturing
establishment, should be prepared and submitted to the national regulatory
authority in support of a request for release of vaccine for use.
A statement signed by the appropriate official of the national control
laboratory should be provided if requested by a manufacturing establishment
and should certify whether or not the lot of vaccine in question meets all
national requirements, as well as Part A of these Guidelines. The certificate
should also state the lot number, the number under which the lot was released,
and the number appearing on the labels of the containers. In addition, the
date of the last satisfactory determination of virus concentration as well as
the expiry date assigned on the basis of shelf-life should be stated. A copy
of the official national release document should be attached. The certificate
should be based on the model given in Appendix 2.
The purpose of the certificate is to facilitate the exchange of live attenuated
rotavirus vaccines between countries.

Authors
The first draft of these Guidelines (WHO/BS/05.2014) was prepared by Dr P.D.
Minor, National Institute for Biological Standards and Control, Potters Bar, Herts,
England; Dr M.L. Pombo, Instituto Nacional de Higiene Rafael Rangel, Control of
Vaccines and Recombinant Department, Caracas, Venezuela; Dr W. Wainwright,
Rotavirus Vaccine Program, PATH, Seattle, USA; with support from the WHO
Secretariat (Dr D.J. Wood and Dr T.Q. Zhou, Quality Assurance and Safety of

165
 Biologicals/Immunizaton, Vaccines and Biologicals/Family and Community Health,
WHO, Geneva, Switzerland).

A revised draft was prepared by Dr R. Dobbelaer, Biological Standardization,

Scientific Institute of Public Health — Louis Pasteur, Belgium; Dr P.D. Minor, National
Institute for Biological Standards and Control, Potters Bar, Herts, England; Dr M.L.
Pombo, Instituto Nacional de Higiene Rafael Rangel, Control of Vaccines and
Recombinant Department, Caracas, Venezuela; Dr W. Wainwright, Rotavirus
Vaccine Program, PATH, Seattle, USA; Dr M. Powell, Medicines and Healthcare
Products Regulatory Agency, London, England, with support from the WHO
Secretariat (Dr D.J. Wood and Dr T.Q. Zhou, Quality Assurance and Safety of
Biologicals/Immunizaton Vaccines and Biologicals/ Family and Community Health,
WHO, Geneva, Switzerland; Dr M.P. Kieny and Dr D. Steele, Initiative for Vaccine
Research/Immunizaton Vaccines and Biologicals/ Family and Community Health,
WHO, Geneva, Switzerland; Dr A. Bentsi-Enchill, Vaccine Assessment and
Monitoring/Immunizaton Vaccines and Biologicals/Family and Community Health,
WHO, Geneva, Switzerland and Dr L. Chocarro, Access to Technologies/
Immunizaton Vaccines and Biologicals/Family and Community Health WHO,
Geneva, Switzerland).

Acknowledgements
For the first draft of these guidelines, acknowledgements are due to the following
participants at a WHO Informal Consultation on Quality, Safety and Efficacy
Specifications for Live Attenuated Rotavirus Vaccines held in Mexico, DF, Mexico
in February 2005:

Dr C.D. Atreya, Center for Biologics Evaluation and Research, Food and Drug
Administration, Rockville Pike, Bethesda, MD, USA; Dr L. Chocarro, Access to
Technologies, World Health Organization, Geneva, Switzerland; Dr M. de los
Angeles Cortes, Regional Advisor on Vaccines and Biologicals, Regional Office
for the Americas, Pan American Health Organization, Washington, DC, USA;
Dr R. Dobbelaer, Head, Biological Standardization, Scientific Institute of Public
Health Louis Pasteur, Brussels, Belgium; Dr M. Franco, Instituto de Genetica
Humana, Pontificia Universidad Javeriana, Bogotá, Colombia; Dr R. Glass, Chief,
Viral Gastroenteritis Section, Centers for Disease Control and Prevention, Atlanta,
GA, USA; Dr H. Greenberg, Senior Associate Dean for Research, Stanford University
School of Medicine and the VA Palo Alto Health Care System, Stanford, USA;
Dr C. Lecomte, Regulatory Affairs, Glaxo SmithKline Biologicals, Rixensart,
Belgium; Dr A. MacLean, Director, Q Gen, Cooperative Research Centre for Vaccine
Technology, Queensland Institute of Medical Research, Brisbane, Queensland,
Australia; Dr P. Minor, Head, Division of Virology, National Institute for Biological
Standards and Control, Potters Bar, Herts., England; Dr I. Perez-Schael, Chief of
Section, Instituto de Biomedicina, Caracas, Venezuela; Dr M. Luz Pombo, Instituto
Nacional de Higiene Rafael Rangel, Control of Vaccines and Recombinant
Department, Caracas, Venezuela; Mr S. Prasad, Bharat Biotech International Ltd.,
Hyderabad, India; Dr L.P. Ruiz Jr., BIOVIRx Inc, Shoreview, MN, USA; Dr L. Saif,
The Ohio State University, Ohio Agricultural Research and Development Center,

166
Food Animal Health Research Program, Wooster, OH, USA; Dr A. Shaw, Executive
Director, Merck Vaccine Division, Merck and Co. Inc., West Point, USA; Dr L.S.
Slamet, Deputy for Therapeutic Products, Narcotic, Psychotropic and Addictive
Substance Control, Directorate General of Food and Drug Control, Ministry of
Health, National Agency of Drug and Food Control, Jakarta Pusat, Indonesia; Dr D.
Steele, Initiative for Vaccine Research, World Health Organization, Geneva,
Switzerland; Dr G. Thiry, Director, Project Management for Research and
Development, International AIDS Vaccine Initiative (IAVI), New York, NY, USA; Dr
Timo Vesikari, University of Tampere Medical School, Tampere, Finland;
Dr W. Wainwright, Rotavirus Vaccine Program, PATH (Seattle), Seattle, USA;
Dr R.L.Ward, Division of Infectious Diseases, Children’s Hospital Medical Center,
Cincinnati, USA; and Dr D. Wood, Coordinator, Quality Assurance and Safety of
Biologicals, World Health Organization, Geneva, Switzerland

Ma Cecilia Copello. National Regulatory Authority Argentina; Flavia Cardoso.

National Regulatory Authority, Brazil; Patricia León Treviño. National Regulatory
Authority, Colombia; Olga Jacobo Casanueva. National Regulatory Authority, Cuba;
Ana Maria Jorquera. National Regulatory Authority, Chile; Ana Agaton, National
Regulatory Authority, Venezuela; and Sonia Zamudio. National Regulatory Authority,
México.

For the second draft, acknowledgements are due to the following participants at a
WHO Informal Consultation on proposed WHO Guidelines to Assure the Quality,
Safety and Efficacy of Live Attenuated Rotavirus Vaccines in Geneva, 29–30 August
2005:

Dr C.D. Atreya, Division of Viral products, Center for Biologics Evaluation and
Research, Food and Drug Administration, Rockville, MD, USA; Dr R. Biellik,
Rotavirus Vaccine Program (RVP), PATH-Europe, Ferney-Voltaire, France; Dr J.
Bresee, Lead, Epidemiology Team, Respiratory and Enteric Virus Branch, Centers
for Disease Control and Prevention, Rockville, MD, USA; Dr F. Cardoso de Melo,
Head of Hemotherapic and Biological Products Unit, Agencia Nacional de
Vigilancia Sanitaria, Ministerio da Saude, Brasilia, Brazil; Dr R.M. Dhere,
Representative from Developing Country Vaccine Manufacturer’s Network
(DCVMN), Director, Vaccine Production, Serum Institute of India Ltd, Maharashtra,
India; Dr R. Dobbelaer, Head, Biological Standardization, Scientific Institute of
Public Health — Louis Pasteur, Brussels, Belgium; Dr Y. Lawanprasert, Food and
Drug Administration, Ministry of Public Health, Thailand; Dr C. Lecomte,
Representative from International Federation of Pharmaceutical Manufacturer’s
Association (IFPMA), Technical Regulatory Affairs, GSK Biologicals, Belgium; Dr P.
Minor, Head, Division of Virology, National Institute for Biological Standards and
Control, Potters Bar, Herts, England; Dr S. Morgeaux, Agence Française de Sécurité
Sanitaire des Produits de Santé, Direction des Laboratoires et des Contrôles, Unité
de Contrôle des Médicaments Immunologiques, Lyon, France; Dr M. Luz Pombo,
Instituto Nacional de Higiene Rafael Rangel, Control of Vaccines and Recombinant
Department, Caracas, Venezuela; Dr Lucky S. Slamet, Deputy for Therapeutic
Products, Narcotic, Psychotropic and Addictive Substance Control, National
Agency of Drug and Food Control the Republic of Indonesia, Jakarta, Indonesia;
Dr A.R. Shaw, Representative from International Federation of Pharmaceutical

167
 Manufacturer’s Association (IFPMA), Executive Director, Merck Vaccine Division,
West Point, Pennsylvania, USA; Dr J. Southern, Advisor to Medicines Control
Council in South Africa, Ministry of Health, Cape Town, South Africa; Dr A. Tahlan,
Joint Director and Government Analyst, Central Drugs Laboratory, Central Research
Institute, Kasauli, India; Dr W. Wainwright, President, Biopharm Consulting Services,
Director of Operations, BIOVIRx, Inc., Conestoga, PA, USA; Dr H. Yin, Director,
Division of Biological Products, State Food and Drug Administration (SFDA),
Beijing, People’s Republic of China; Dr Ma. Angeles Cortes, Regional Advisor on
Vaccines and Biologicals, WHO Regional Office for the Americas/Pan American
Sanitary Bureau, Washington, DC, USA; Dr J. Sokhey, WHO Regional Office for
South-East Asia, World Health House, Indraprastha Estate, New Delhi, India; Dr D.
Wood, Coordinator, Quality Assurance and Safety of Biologicals, World Health
Organization, Geneva, Switzerland; Dr Tiequn Zhou, Scientist, Quality Assurance
and Safety of Biologicals, World Health Organization, Geneva, Switzerland; Dr D.
Steele, Scientist, Initiative for Vaccine Research, World Health Organization,
Geneva, Switzerland; Dr L. Chocarro, Scientist, Access to Technologies, World
Health Organization, Geneva, Switzerland; Dr A.D. Bentsi-Enchill, Medical Officer,
Vaccine Assessment and Monitoring, World Health Organization, Geneva,
Switzerland.

References
1. WHO Guidelines on nonclinical evaluation of vaccines. In: WHO Expert
Committee on Biological Standardization. Fifty-fourth report. Geneva, World
Health Organization, 2005, Annex 1 (WHO Technical Report Series, No. 927).
2. Guidelines on clinical evaluation of vaccines: regulatory expectations.
In: WHO Expert Committee on Biological Standardization. Fifty-second
report. Geneva, World Health Organization, 2002, Annex 1 (WHO Technical
Report Series, No. 924).
3. WHO Consultative Group. Report of a meeting on quality, safety and efficacy
specifications for live attenuated rotavirus vaccines. Mexico City, Mexico,
8–9 February, 2005.Vaccine 23:5478–5487.
4. Ruuska T, Vesikari T. Rotavirus disease in Finnish children: Use of numerical
scores for clinical severity of diarrhoeal episodes. Scandinavian Journal
of Infectious Diseases, 1990, 22:259–267.
5. Guidelines on transmissible spongiform encephalopathies in relation to
biological and pharmaceutical products. Geneva, World Health Organization,
2003 (http://www.who.int.biologicals).
6. Requirements for the use of animal cells as in vitro substrates for the
production of biologicals. In: WHO Expert Committee on Biological
Standardization. Forty-seventh report. Geneva, World Health Organization,
1998, Annex 1 (WHO Technical Report Series, No. 878).
7. Lebron JA et al. Adaptation of the WHO guideline for residual DNA in
parenteral vaccines on continuous cells lines to a limit for oral vaccines.
Developments in Biologicals, 2006, 123:35–44.
8. Israel MA et al. Biological activity of polyoma viral DNA in mice and
hamsters. Journal of Virology 1979, 29:990–996.

168
9. Ranheim T et al. Development and application of a quantitative RT-PCR
potency assay for a pentavalent rotavirus vaccine (RotaTeq®). Journal
of Virological Methods, 2006, 131:193–201.
10. Recommendations for the production and control of poliomyelitis vaccine
(oral). In: WHO Expert Committee on Biological Standardization. Fiftieth
report, Geneva, World Health Organization, 2002 (WHO Technical Report
Series, No. 904).
11. Guidelines for the production and control of Japanese encephalitis vaccine
(live) for human use. WHO Expert Committee on Biological Standardization.
Fifty-first report. Geneva, World Health Organization, 2002, Annex 3 (WHO
Technical Report Series, No. 910).
12. Guidelines for the production and quality control of candidate tetravalent
dengue virus vaccines (live). WHO Expert Committee on Biological
Standardization. Fifty-fifth report. Geneva, World Health Organization, 2005,
Annex 1 (WHO Technical Report Series, No. 932).
13. Good manufacturing practices for biological products. In: WHO Expert
Committee on Biological Standardization. Forty-second report. Geneva,
World Health Organization, 1992, Annex 1 (WHO Technical Report Series,
No. 822).
14. Laboratory biosafety manual, 3rd ed. Geneva, World Health Organization,
2004.
15. Requirements for the use of animal cells as in vitro substrates for the
production of biologicals (Addendum 2003). In: WHO Expert Committee
on Biological Standardization. Fifty-fourth report. Geneva, World
Health Organization, 2005, Annex 4 (WHO Technical Report Series,
No. 927).
16. General requirements for the sterility of biological substances (Requirements
for Biological Substances No. 6, revised 1973, amendment 1995). In: WHO
Expert Committee on Biological Standardization. Forty-sixth report. Geneva,
World Health Organization, 1995, Annex 3 (WHO Technical Report Series
No. 872).
17. Requirements for the collection, processing and quality control of blood,
blood components and plasma derivatives (Requirements for Biological
Substances No. 27, revised 1992). In: WHO Expert Committee on Biological
Standardization. Forty-third report. Geneva, World Health Organization, 1994
(WHO Technical Report Series, No. 840).
18. Tests for extraneous agents in viral vaccines for human use (General method
No. 2.6.16). European Pharmacopoeia 4th ed, 148–149.
19. Requirements for measles vaccines (Live). WHO Expert Committee on
Biological Standardizaton. Thirty-eighth Report. World Health Organization,
1988, Annex 5 (WHO Technical Report Series, No. 771).
20. Good manufacturing practices for pharmaceutical products. In: WHO Expert
Committee on Specifications for Pharmaceutical Preparations. Thirty-second
report. Geneva, World Health Organization, 1992, Annex 1 (WHO Technical
Report Series, No. 823).
21. Ward R.L et al. Live vaccine strategies to prevent rotavirus disease. New
generation vaccines, 3rd ed. New York, Marcel Dekker, 2004:607–620.

169
 22. Simonsen L et al. More on RotaShield and intussusception: the role of age at
the time of vaccination. Journal of Infectious Disease, 2005, 192:S36–43.
23. Ruuska T, Vesikari T. Rotavirus disease in Finnish children: Use of numerical
scores for clinical severity of diarrhoeal episodes. Scandinavian Journal of
Infectious Diseases, 1990, 22:259–267.
24. Clark et al. Protective effect of WC3 vaccine against rotavirus diarrhea
in infants during a predominantly serotype 1 rotavirus season. Journal of
Infectious Diseases, 1988, 158:570–587.
25. Guidelines for national authorities on quality assurance for biological
products. In: WHO Expert Committee on Biological Standardization. Forty-
second report. Geneva, World Health Organization, 1992, Annex 2 (WHO
Technical Report Series, No. 822).

170
 Appendix 1
Summary protocol of manufacturing and control
of live attenuated rotavirus vaccines (oral)
The following protocol is intended for guidance, and indicates the
information that should be provided as a minimum by the manufacturer to
the national regulatory authority. The protocol must be accompanied by a
lot release certificate from the licensing authority which may or may not be
the country of manufacturing origin. Information and tests may be added
or deleted as required by the national regulatory authority of the importing
country, if applicable.
It is thus possible that a protocol for a specific product may differ in
detail from the model provided. The essential point is that all relevant
details demonstrating compliance with the licence and with the relevant
WHO guidelines for a particular product should be given in the protocol
submitted.
The section concerning the final product must be accompanied by a
sample of the label and a copy of the leaflet that accompanies the vaccine
container. If the protocol is being submitted in support of a request to permit
importation, it must also be accompanied by a lot release certificate from
the national regulatory authority of the country in which the vaccine was
produced stating that the product meets national requirements as well as the
guidelines in Part A of this document.

Summary information on the finished product (final lot)

International name: Live attenuated rotavirus
vaccine (oral)
Commercial name: ___________________________________________
Product licence (marketing authorization) number __________________
Country: ___________________________________________________
Name and address of manufacturer: ______________________________
Name and address of licence holder if different: ____________________
__________________________________________________________
Final packing lot number: _____________________________________
Type of container: ___________________________________________
Number of containers in this packing lot: _________________________
Final container lot number: ____________________________________
Number of filled containers in this final lot: _______________________
Date of manufacture (filling or lyophilizing, if applicable): ___________
Date on which last determination of virus concentration
was started: ______________________________________________

171
 Shelf-life approved (months): __________________________________
Expiry date: ________________________________________________
Storage conditions: __________________________________________
Volume of single dose: ________________________________________
Volume of vaccine per container: _______________________________
Number of doses per container: _________________________________
Virus concentration per human dose:
Serotype: __________________________________________________
Serotype: __________________________________________________
Serotype: __________________________________________________
Serotype: __________________________________________________
Nature of any antibiotics present in vaccine and amount
per human dose: ___________________________________________
Production cell substrate: ______________________________________
Bulk No. of monovalent virus pools blended in multivalent
vaccine (if applicable): ______________________________________
Diluent or antacid (if applicable): ________________________________
Lot number: ____________________________________________
Date of manufacture: _____________________________________
Expiry date: ____________________________________________
A genealogy of the lot numbers of all vaccine components used in the
formulation of the final product, diluent and antacid will be informative.
An example of a genealogy is given in Appendix 2.
The following sections are intended for the reporting of the results of the tests
performed during the production of the vaccine, so that the complete document
will provide evidence of consistency of production; thus if any test has to be
repeated, this must be indicated. Any abnormal results should be recorded on
a separate sheet. If any cell lot, virus harvest or other intermediates intended
for production of the current lot was rejected, this should also be recorded
either in the following sections or on a separate sheet.
A.3 Control of source materials
The results of tests on the same master or manufacturing working cell
bank and the same virus master and manufacturing working seed lots are
submitted to and approved by the national regulatory authority during the
procedure for granting the licence or its variations and need not be re-
submitted at the time of lot release.
A.3.1 Cell cultures for virus production
Name and identification of cell substrate: _________________________
Origin and short history: ______________________________________
Authority that approved cell bank: ______________________________

172
Master cell bank (MCB)
Lot number: ________________________________________________
Date MCB was established: ____________________________________
Date of approval by the national regulatory authority: _______________
Total number of ampoules stored: _______________________________
Passage/population doubling level of MCB: _______________________
Maximum passage/population doubling level approved
for MCB: ________________________________________________
Storage conditions: __________________________________________
Percentage of total MCB ampoules tested: ________________________
Identity test: ________________________________________________
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________
Results of tests for adventitious agents: ___________________________
Results of tests for tumorigenicity (if applicable): __________________
Tests for retroviruses (if applicable)
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________

Manufacturer’s working cell bank (MWCB)

Lot number: ________________________________________________
Date MWCB was established: __________________________________
Total number of ampoules stored: _______________________________
Passage/population doubling level of MWCB:______________________
Maximum passage/population doubling level approved for
MWCB: _________________________________________________
Storage conditions ___________________________________________
Percentage of total MWCB ampoules tested: ______________________
Identity test:
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________
Results of tests for adventitious agents: ___________________________
Results of tests for tumorigenicity (if applicable): __________________
Tests for retroviruses (if applicable):
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________
Cell culture medium
Serum used in cell culture medium
Animal origin of serum: ______________________________________

173
 Batch number: ______________________________________________
Vendor: ____________________________________________________
Country of origin: ___________________________________________
Certificate of freedom from TSE (yes/no): ________________________
Tests performed on serum: _____________________________________
Methods used ___________________________________________
Results ________________________________________________

Trypsin used for preparation of cell cultures

Animal origin of trypsin: ______________________________________
Batch number _______________________________________________
Vendor: ____________________________________________________
Country of origin: ___________________________________________
Certificate of freedom from TSE (yes/no): ________________________
Tests performed on trypsin: ____________________________________
Date of tests ____________________________________________
Methods used ___________________________________________
Results ________________________________________________

A.3.2 Virus seeds

Virus strain(s) and serotype(s): _________________________________
Substrate used for preparing seed lots: ___________________________
Origin and short history: ______________________________________
Authority that approved virus strain(s): ___________________________
Date approved: ______________________________________________

Virus master seed lot (VMS)

Lot number: ________________________________________________
Date VMS was established: ___________________________________
Date approved by the National Regulatory Authority: _______________
Total quantity of VMS stored: __________________________________
Storage conditions: __________________________________________
Passage level of VMS: ________________________________________
Maximum passage level approved for VMS: _______________________

Manufacturer’s virus working seed lot (MVWS)

Lot number: ________________________________________________
Date MVWS was established: __________________________________
Total quantity of MVWS stored: ________________________________
Storage conditions: __________________________________________
Passage level of MVWS: ______________________________________
Maximum passage level approved for MVWS: _____________________

174
Tests on virus master and working seed lots
Identity test:
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________
Genotype/phenotype characterization:
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________
Tests for bacteria, fungi and mycoplasmas
Tests for bacteria and fungi
Incubation Media used Inoculum Date test on Date test off Results
20–25 °C _________ ________ _____________ ____________ _____
30–36 °C _________ ________ _____________ ____________ _____
Negative control _________ ________ _____________ ____________ _____

Tests for mycoplasmas

Standard culture method
Inoculum Date test began Date test ended Incubation conditions Results
Solid media _______ ___________ ___________ ________________ ______
Liquid media _______ ___________ ___________ ________________ ______
Negative control _______ ___________ ___________ ________________ ______
Positive control
cultures _______ ___________ ___________ ________________ ______

Indicator cell-culture method

Cell substrate used ___________________________________________
Inoculum __________________________________________________
Date of test _____________________________________________
Passage number _________________________________________
Negative control _________________________________________
Positive controls _________________________________________
Date of staining _________________________________________
Results ________________________________________________
Tests for adventitious viruses
Volume of virus seed samples for neutralization and testing _______
Batch number(s) of antisera/antiserum used for neutralization
of virus seeds _________________________________________
Tests in tissue cultures
Type of simian cells _______________________________________
Quantity of neutralized sample inoculated _____________________
Incubation conditions _____________________________________

175
 Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Results ________________________________________________
Type of human cells __________________________________________
Quantity of neutralized sample inoculated _____________________
Incubation conditions _____________________________________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Results ________________________________________________
Other cell types ______________________________________________
Quantity of neutralized sample inoculated _____________________
Incubation conditions _____________________________________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Results ________________________________________________
Tests in animals
Test in adult mice
Weight and number of animals ______________________________
Routes and quantity of neutralized sample inoculated ____________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of animals surviving the observation period _______________
Results ________________________________________________

Test in suckling mice

Age and number of animals ________________________________
Routes and quantity of neutralized sample inoculated ____________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of animals surviving the observation period _______________
Results ________________________________________________

Test in guinea-pigs
Weight and number of animals ______________________________
Routes and quantity of neutralized sample inoculated ____________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of animals surviving the observation period _______________
Results ________________________________________________

176
 Additional tests __________________________________________
Date of tests ____________________________________________
Methods used ___________________________________________
Results ________________________________________________

Virus concentration
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________

A.4 Control of vaccine production

Production cells
Lot number: ________________________________________________
Date of thawing ampoule of MWCB: ____________________________
Passage/population doubling level at virus inoculation: ______________
Maximum passage/population doubling level approved for
vaccine production: ________________________________________
Nature and concentration of antibiotics used in production
cell culture maintenance medium: _____________________________
Identification and source of starting materials used in preparing
production cells including excipients and preservative
(particularly any materials of human or animal origin): ____________

A.4.1 Control of Cell Cultures

(Note: If more than one virus single harvest is used to produce a monovalent
virus pool, then data on each lot of control cells should be provided.)
Amount or ratio of control cultures to production cell cultures:
__________________________________________________________
Incubation conditions: ________________________________________
Period of observation of cultures: _______________________________
Date started _____________________________________________
Date ended _____________________________________________
Ratio of cultures discarded and reason: ___________________________
Results of observation: _______________________________________
Date fluids collected: _________________________________________
Date fluids pooled (if applicable): _______________________________
Tests for haemadsorbing viruses:
Quantity of cells tested ____________________________________
Type of RBC used ________________________________________
Storage time and temperature of RBC ________________________
Incubation time and temperature of RBC ______________________

177
 Date test began __________________________________________
Date test ended __________________________________________
Results ________________________________________________
Additional tests if performed ___________________________________
Tests for other adventitious agents:
Test in production cells
Date of sampling ________________________________________
Quantity of sample inoculated ______________________________
Incubation conditions _____________________________________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Uninoculated cell control __________________________________
Results ________________________________________________
Test in human cells
Type of human cells ______________________________________
Quantity of sample inoculated ______________________________
Incubation conditions _____________________________________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Uninoculated cell control __________________________________
Results ________________________________________________
Test in other cell system
Type of cells ____________________________________________
Quantity of sample inoculated ______________________________
Incubation conditions _____________________________________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Uninoculated cell control __________________________________
Results ________________________________________________
Identity test:
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________

A.4.2 Cell cultures for vaccine production

Tests for adventitious agents
Date of examination (inoculation) _______________________________
Results ____________________________________________________

178
 Tests for bacteria, fungi and mycoplasmas
Date and volume of sampling _______________________________
Volume of samples tested __________________________________
Tests for bacteria and fungi
Incubation Media used Inoculum Date test on Date test off Results
20–25 °C _________ ________ _____________ ____________ _____
30–36 °C _________ ________ _____________ ____________ _____
Negative control _________ ________ _____________ ____________ _____

Tests for mycoplasmas

Standard culture method
Inoculum Date test on Date test off Incubation conditions Results
Solid media _______ ___________ ___________ ________________ ______
Liquid media _______ ___________ ___________ ________________ ______
Negative control _______ ___________ ___________ ________________ ______
Positive control
cultures _______ ___________ ___________ ________________ ______

Indicator cell-culture method

Cell substrate used ___________________________________________
Inoculum __________________________________________________
Date of test _________________________________________________
Passage number _____________________________________________
Negative control _____________________________________________
Positive controls _____________________________________________
Date of staining _____________________________________________
Results ____________________________________________________

A.4.3 Control of single harvests

For multivalent vaccine, the following information for each virus serotype
should be submitted.
If more than one single harvest is used to prepare a monovalent virus pool,
the following information for each single harvest should be submitted.
Virus serotype _______________________________________________
Lot number of single harvest ___________________________________
Date of virus inoculation ______________________________________
Multiplicity of infection _______________________________________
Incubation conditions ________________________________________
Date of harvesting ___________________________________________
Volume harvested ____________________________________________
Date of sampling ____________________________________________
Volume of sampling __________________________________________
Storage conditions and period __________________________________

179
 Monovalent virus pool (pre-clarification)
Lot number of virus pool ______________________________________
Date of pooling _____________________________________________
Virus single harvests pooled
Lot number _________________ Volume pooled ________________
Volume of virus pool after pooling ______________________________
Date of sampling ____________________________________________
Volume of sampling __________________________________________
Storage of samples (if applicable) _______________________________

Tests on single harvest or monovalent virus pools

(Tests may be done on individual single harvest or on the virus pools as
approved by the national regulatory authority.)
Identity
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________
Tests for bacteria, fungi and mycoplasmas
Tests for bacteria and fungi
Incubation Media used Inoculum Date test began Date test ended Results
20–25 °C _________ ________ _____________ ____________ _____
30–36 °C _________ ________ _____________ ____________ _____
Negative control _________ ________ _____________ ____________ _____

Tests for mycoplasmas

Standard culture method
Inoculum Date test began Date test ended Incubation conditions Results
Solid media _______ ___________ ___________ ________________ ______
Liquid media _______ ___________ ___________ ________________ ______
Negative control _______ ___________ ___________ ________________ ______
Positive control
cultures _______ ___________ ___________ ________________ ______

Indicator cell-culture method

Cell substrate used _______________________________________
Inoculum _______________________________________________
Date of test _____________________________________________
Passage number _________________________________________
Negative control _________________________________________
Positive controls _________________________________________
Date of staining _________________________________________
Results ________________________________________________

180
 Tests for adventitious agents
Volume of samples for neutralization and testing ___________________
Batch number(s) of antisera/antiserum used for neutralization _________

Tests in tissue cultures

Type of simian cells ______________________________________
Quantity of neutralized sample inoculated _____________________
Incubation conditions _____________________________________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Results ________________________________________________
Type of human cells ______________________________________
Quantity of neutralized sample inoculated _____________________
Incubation conditions _____________________________________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Results ________________________________________________
Type of other cells _______________________________________
Quantity of neutralized sample inoculated _____________________
Incubation conditions _____________________________________
Date test began __________________________________________
Date test ended __________________________________________
Ratio of cultures viable at end of test _________________________
Results ________________________________________________

Primary passage Subculture passage

Cell Specification Test No. of Results Test No. of Results
substrate initiation flasks initiation flasks
date tested date tested

Cytopathic effect

Haemadsorption

Positive control
virus

Negative control

Virus concentration
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________

181
 Tests for consistency of virus characteristics
(Tests are performed during vaccine development and process validation,
may not be required for batch release.)
Item tested _________________________________________________
Date of test _________________________________________________
Methods used _______________________________________________
Results ____________________________________________________
Storage conditions and period ________________________________
Control of clarified monovalent virus pool
Lot number of monovalent virus pool ____________________________
Date of clarification __________________________________________
Methods used for clarification __________________________________
Volume of virus pool before clarification _________________________
Volume of virus pool after clarification ___________________________
Date of sampling ____________________________________________
Volume of sampling __________________________________________
Storage conditions of samples __________________________________
Date test
Specification initiated Method Results
Sterility ___________ __________ _________ _________
Virus concentration ___________ __________ _________ _________
Tests for residual ___________ __________ _________ _________
cellular DNA ___________ __________ _________ _________
A.4.4 Final bulk
Lot number ________________________________________________
Date of formulation __________________________________________
Total volume of final bulk formulated ____________________________
Monovalent virus pools used for formulation
Serotype Lot number Volume added Virus concentration
________ _____________ ______________ ________________
________ _____________ ______________ ________________
________ _____________ ______________ ________________
Name Lot number Volume added
Stabilizer if used ______________ ____________ ______________
Diluent used ______________ ____________ ______________
Date test
Specification initiated Method Results
Sterility ___________ __________ _________ _________
Storage conditions
and period _______________________________________________

182
 Approved storage
period ___________________________________________________

A.5 Filling and containers

Lot number ________________________________________________
Date of filling _______________________________________________
Volume of final bulk filled _____________________________________
Filling volume per container ___________________________________
Number of containers filled (gross) ______________________________
Date of lyophilization (if applicable) _____________________________
Number of containers rejected during inspection ___________________
Number of containers sampled _________________________________
Total number of containers (net) ________________________________
Maximum period of storage approved ____________________________
Storage temperature and period _________________________________

A.6 Control tests on final lot

A.6.1 Vaccine
Inspection of final containers
Appearance _____________________________________________
Date of test _____________________________________________
Results ________________________________________________
Before reconstitution _____________________________________
After reconstitution _______________________________________
Diluent used ____________________________________________
Lot number of diluent used ________________________________

Identity
Date test began and ended _________________________________
Method used ____________________________________________
Results ________________________________________________
Lot number of reference reagents ____________________________
Date test
Specification initiated Method Results
Sterility ___________ __________ _________ _________
Diluent used ____________________________________________
Lot number of diluent used ________________________________

pH
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________

183
 Diluent used ____________________________________________
Lot number of diluent used ________________________________
Residual moisture (if applicable)
Date of test _____________________________________________
Method used ____________________________________________
Results ________________________________________________

Virus concentration
Date titration began and ended ______________________________
Method used for titration __________________________________
Results
Serotype Virus titre
___________________________ ___________________________
___________________________ ___________________________
___________________________ ___________________________
___________________________ ___________________________
Lot number of reference virus ______________________________
Lot number of other reference reagents if used _________________
Diluent used ____________________________________________
Lot number of diluent used ________________________________

Accelerated stability tests

Duration of exposure _____________________________________
Temperature of exposure __________________________________
Date titration began and ended ______________________________
Method used for titration __________________________________
Results ________________________________________________
Total virus titre
Exposed sample _________________________________________
Non-exposed sample _____________________________________
Titre reduction __________________________________________
Lot number of reference virus ______________________________
Lot number of other reference reagents if used _________________
Diluent used ____________________________________________
Lot number of diluent used ________________________________

A.6.2 Diluents
Nature and volume ___________________________________________
Lot number ________________________________________________
Date of manufacture _________________________________________
Storage conditions and period __________________________________
Expiry date _________________________________________________

184
 Date test
Specification initiated Method Results
Sterility ___________ __________ _________ _________
Identity ___________ __________ _________ _________
pH ___________ __________ _________ _________
Physical inspection ___________________________________________
Content of key components:
__________________________________________________________
__________________________________________________________
__________________________________________________________
__________________________________________________________
Certification by the manufacturer
Name of head of production (typed) _____________________________
Certification by the person from the control laboratory of the manufacturing
company taking over responsibility for the production and control of the
vaccine:
I certify that lot no. ______________ of live attenuated rotavirus vaccine
(oral), whose number appears on the label of the final container, meets all
national requirements and satisfies Part A of the Requirements for Biological
Substances No. __________ of live attenuated rotavirus vaccines (oral).
Signature: __________________________________________________
Name (typed): ______________________________________________
Date: ______________________________________________________

185
 Appendix 2
Genealogy of vaccine production process
Virus Master Seed
Type 1
Lot no.

Virus Working Seed

Type 1
Lot no.

Single Harvest Single Harvest Single Harvest

Type 1 Type 1 Type 1
Lot no. 1 Lot no. 2 Lot no. …

Monovalent Virus Pool

Type 1
Lot no.

Clarified Monovalent Virus Pool

Type 1
Lot no.
Virus Master Seed
Type 2
Lot no.

Virus Working Seed

Type 2
Lot no.

Single Harvest Single Harvest Single Harvest

Type 2 Type 2 Type 2
Lot no. 1 Lot no. 2 Lot no. …

Monovalent Virus Pool

Type 2
Lot no.

Clarified Monovalent Virus Pool

Type 2
Lot no.
Virus Master Seed
Type …
Lot no.

Virus Working Seed

Type …
Lot no.

Single Harvest Single Harvest Single Harvest

Type … Type … Type …
Lot no. 1 Lot no. 2 Lot no. …

Monovalent Virus Pool

Type …
Lot no.

Clarified Monovalent Virus Pool

Type …
Lot no.

186
 Clarified Monovalent Virus Pool Genealogy of final bulk lot and final lot Clarified Virus Harvest Pool
Type 1 Monovalent Type …
Lot no. 1 Lot no. …

Clarified Monovalent Virus Pool Clarified Virus Harvest Pool

Type 1 Monovalent Type …
Lot no. 2 Lot no. 2

Clarified Monovalent Virus Pool Clarified Monovalent Virus Pool

Type 1 Type …
Lot no. … Lot no. 1

Clarified Monovalent Virus Pool Clarified Monovalent Virus Pool Clarified Monovalent Virus Pool
Type 2 Type 2 Type 2
Lot no. 1 Lot no. 2 Lot no. …

Final Bulk Final Bulk

Lot no. 1 Lot no. …

Final Lot Final Lot Final Lot Final Lot Final Lot Final Lot
Lot no. 1 Lot no. 2 Lot no. … Lot # 1 Lot no. 2 Lot no. …

Packaging Lot Packaging Lot Packaging Lot

Lot no. 1 Lot no. 2 Lot no. …

Diluent Diluent Diluent

Lot no. … Lot no. … Lot no. …

187
 Appendix 3
Model certificate for the release of live attenuated
rotavirus vaccines
This certificate is to be provided by the national regulatory authority of
the country where the vaccines have been manufactured, on request by the
manufacturer
Certificate no. ________________

Lot release certificate

The following lot(s) of live attenuated rotavirus vaccine (oral) produced by
____________________________1 in _______________2 whose numbers
appear on the labels of the final containers, meet all national requirements3
and Parts A4, B and C of Guidelines for Live Attenuated Rotavirus Vaccine
(oral), (2007)5 and comply with Good Manufacturing Practices for
pharmaceutical products6 and Good manufacturing practices for biological
products.7
As a minimum, this certificate is based on examination of the summary
protocol of manufacturing and control.

Final lot no. Number of released Expiry date

human doses
in this final lot

The director of the national regulatory authority

(or authority as appropriate):
Name (typed) _______________________________________________
Signature __________________________________________________
Date ______________________________________________________

1
Name of manufacturer.
2
Country of origin.
3
If any national requirements are not met, specify which one(s) and indicate why release of the
lot(s) has nevertheless been authorized by the national regulatory authority.
4
With the exception of provisions on distribution and shipping, which the national regulatory
authority may not be in a position to assess.
5
WHO Technical Report Series, No. 941, 2007, Annex 3.
6
WHO Technical Report Series, No. 823, 1992, Annex 1.
7
WHO Technical Report Series, No. 822, 1992, Annex 1.

188
© World Health Organization
WHO Technical Report Series No xxxx, 200x

Annex 4
Recommendations for the production, control
and regulation of human plasma for fractionation

1 Introduction
2 International Biological Reference Preparations
3 Glossary
4 General considerations
4.1 Range of products made from human blood and plasma
4.2 Composition of human plasma
4.3 Pathogens present in blood and plasma
5 Measures to exclude infectious donations
5.1 Appropriate selection of blood/plasma donors
5.2 Screening of blood/plasma donations for infectious markers
5.3 Epidemiological surveillance of donor population
5.4 Strict adherence to Good Manufacturing Practices
5.5 Post-donation events
6 Production of plasma for fractionation
6.1 Methods used to obtain plasma for fractionation
6.2 Characteristics of plasma for fractionation
6.3 Premises and devices for collection of plasma for fractionation
6.4 Blood/plasma collection process
6.5 Separation of plasma
6.6 Freezing of plasma
6.7 Storage of plasma
6.8 Compliance with plasma fractionator requirements
6.9 Release of plasma for fractionation
6.10 Packaging of plasma
6.11 Transportation of plasma
6.12 Recall system
7 Quality assurance system and Good Manufacturing Practices
7.1 Organisation and personnel
7.2 Documentation system
7.3 Premises and equipment
7.4 Materials
7.5 Validation programme

189
 7.6 Quality monitoring data
7.7 Virology safety testing
7.8 Electronic information system
7.9 Storage and transport
7.10 Change control system
7.11 Quality assurance auditing
7.12 Defect reporting system
7.13 Quality agreement between blood establishment and fractionator
7.14 Blood/plasma establishment audit and inspection
8 Regulatory control of plasma for fractionation
8.1 Role of national regulatory authority
8.2 Establishment licence and inspections
8.3 Impact of Good Manufacturing Practices
8.4 Inspections
Authors
References
Appendix 1
Plasma products and clinical applications
Appendix 2
Donor selection
Appendix 3
Donor immunization and plasmapheresis for specific immunoglobulins
Appendix 4
Contract plasma fractionation programme
Appendix 5
Technical points to consider in establishing plasma specifications
criteria and obligations between blood establishment and plasma fractionator

190
1. Introduction
Human plasma is a source of important medicinal products which are
obtained by a combination of large-scale processing steps known as
“fractionation”. It is important that these products have an appropriate
quality and safety profile.
Recognizing the importance of the provision of safe blood, blood components
and plasma derivatives, the 58th World Health Assembly in 2005 (WHA
Resolution 58.13) (1) expressed its support for “the full implementation of
well-organized, nationally coordinated and sustainable blood programmes
with appropriate regulatory systems” and stressed the role of “voluntary,
non-remunerated blood donors from low-risk populations”. The provision
of blood, blood components and plasma derivatives from voluntary, non
remunerated donors should be the aim of all countries.
The WHO requirements for the collection, processing, and quality control
of blood, blood components, and plasma derivatives were published in 1994
(2). Numerous developments have taken place since that time, requiring
updates of both technical and regulatory guidelines to be made available
at the global level. The recently published WHO guidelines on viral
inactivation and removal procedures (3) address the measures necessary to
eliminate or reduce the risk from blood-borne viruses during the processing
of plasma into plasma derivatives.
The present Recommendations are intended to provide guidance on the
production, control and regulation of human plasma for fractionation as a
source material for plasma derived medicinal products. Such information is
necessary for the manufacture of safe plasma derivatives in both developed
and developing countries worldwide.
The current document, by bringing together experience and information,
will serve as a guide to blood establishments in their implementation of
appropriate procedures for the production and control of the starting plasma
material, and will facilitate the provision of safe fractionated plasma products
at the national level. It is intended to assist national (medicines) regulatory
authorities in establishing the supervision necessary for assessment of the
quality and safety of plasma for fractionation, either prepared locally or
imported, and will therefore contribute to improved quality and safety of
human plasma products worldwide. Manufacturers of plasma derivatives
(fractionators) may use these guidelines when discussing the quality criteria
of plasma for fractionation with representatives of blood establishments and
the national regulatory authority.
This guidance document addresses only human plasma sourced for the
manufacture of plasma derivatives. Plasma for clinical use is not discussed,
nor is there any consideration of plasma from other species.

191
2. International Biological Reference Preparations
The full list of WHO Biological Reference Preparations relevant to
blood products and related substances is available at: http://www.who.int/
bloodproducts/ref_materials/
The biological activity of blood products should be measured by comparison
with the relevant International Standard. Activity is usually expressed in
International Units (IU), but may in some cases be expressed in SI units.

3. Glossary
The definitions given below apply to the terms used in these
Recommendations. They may have different meanings in other contexts.

Apheresis
Procedure whereby blood is removed from the donor, separated by physical
means into components and one or more of them returned to the donor.

Blood collection
Procedure whereby a single donation of blood is collected in an anticoagulant
and/or stabilizing solution, under conditions designed to minimize
microbiological contamination of the resulting donation.

Blood component
A constituent of blood (erythrocytes, leukocytes, platelets or plasma) that
can be prepared under such conditions that it can be used either directly or
after further processing for therapeutic applications.

Blood establishment
Any structure or body that is responsible for any aspect of the collection
and testing of human blood or blood components, whatever their intended
purpose, and for their processing, storage, and distribution when intended
for transfusion.1

Donor
A person who gives blood or plasma used for fractionation.

Factor VIII
Blood coagulation factor VIII, deficient in patients with haemophilia A.
Also called antihaemophilic factor.

1
A blood centre is a blood establishment.

192
Factor IX
Blood coagulation factor IX, deficient in patients with haemophilia B.

First-time tested donor

A person whose blood or plasma is tested for the first time for infectious
disease markers in a blood establishment.

Fractionation
A (large-scale) process by which plasma is separated into individual
protein fractions, that are further purified for medicinal use (variously
referred to as plasma derivatives, fractionated plasma products or plasma-
derived medicinal products). The term fractionation is used to describe a
sequence of processes, including: plasma protein separation steps (typically
precipitation and/or chromatography), purification steps (typically ion-
exchange or affinity chromatography) and one or more steps for the
inactivation or removal of blood-borne infectious agents (most specifically
viruses and, possibly, prions).

Fractionator
A company or an organization performing plasma fractionation to manufacture
plasma-derived medicinal products.

Genome equivalents (GE)

The amount of nucleic acid of a particular virus assessed using nucleic acid
testing.

Good Manufacturing Practice (GMP)

That part of quality assurance which ensures that products are consistently
produced and controlled to the quality standards appropriate to their intended
use and as required by the marketing authorization or product specification.
It is concerned with both production and quality control.

Hepatitis A virus (HAV)

A non-enveloped, single-stranded RNA virus, causative agent of hepatitis A.

Hepatitis B surface antigen (HBsAg)

The antigen on the periphery of hepatitis B virus.

Hepatitis B virus (HBV)

An enveloped, double-stranded DNA virus, causative agent of hepatitis B.

Hepatitis C virus (HCV)

An enveloped, single-stranded, RNA virus, causative agent of hepatitis C.

Hepatitis E virus (HEV)

A non-enveloped, single-stranded RNA virus, causative agent of hepatitis E.

193
 Hepatitis G virus (HGV) (or GB virus C (GBV-C))
An enveloped single-stranded RNA virus, causative agent of hepatitis G.

Human immunodeficiency virus (HIV)

An enveloped, single-stranded RNA virus, causative agent of acquired
immunodeficiency syndrome (AIDS).

Incidence
The rate of newly-acquired infection identified over a specified time period
in a defined population.

Inventory hold period

Period during which the plasma for fractionation is on hold pending
identification and elimination of possible window-phase donations.

Intravenous immunoglobulin (IVIG)

Also known as immune globulin, intravenous.

Look-back
Procedure to be followed if it is found retrospectively that a donation from
a high-risk donor should have been excluded from processing.

Manufacture
All operations of procurement of materials (including collection of
plasma for fractionation) and products; production; quality control; release;
storage; distribution; and quality assurance of plasma-derived medicinal
products.

Nucleic acid testing (NAT)

A method to detect viral genome that uses amplification techniques such as
polymerase chain reaction.

National regulatory authority

WHO terminology for referring to national medicines regulatory authorities.
Such authorities promulgate medicines regulations and enforce them.

Plasma
The liquid portion remaining after separation of the cellular elements from
blood collected in a receptacle containing an anticoagulant, or separated
by continuous filtration or centrifugation of anticoagulated blood in an
apheresis procedure.

Plasmapheresis
Procedure in which whole blood is removed from the donor, the plasma
is separated from the cellular elements and at least the red blood cells are
returned to the donor.

194
Plasma products
A range of medicinal products (as listed in Appendix 1) obtained by the
process of fractionation of human plasma. Also called plasma derivatives,
fractionated plasma products, or plasma-derived medicinal products.

Plasma for fractionation

Recovered plasma or source plasma used for the production of plasma
products.

Plasma master file

A document which provides all relevant detailed information on the
characteristics of the entire human plasma used by a fractionator as starting
material and/or raw material for the manufacture of subintermediate or
intermediate plasma fractions, constituents of the excipient and active
substance(s), which are part of a medicinal product.
Prevalence
The rate of infection identified, including both past and present infections,
at a specified point in time or over a specified time period in a defined
population.

Prion
The infectious particle associated with transmissible spongiform
encephalopathies. It is believed to consist only of protein and to contain
no nucleic acid.

Production
All operations involved in the preparation of plasma-derived medicinal
products, from collection of blood or plasma, through processing and
packaging, to its completion as a finished product.

Recovered plasma
Plasma recovered from a whole blood donation and used for fractionation.

Repeat-tested donor
A person whose blood or plasma has been tested previously in the blood
establishment for infectious disease markers.
Replacement donor
Person who gives blood upon request of a specific patient or patient’s family
or acquaintance, which in principle is intended to be used specifically for
the treatment of that patient.

SD-plasma
Solvent/detergent-treated pooled plasma intended as a substitute for fresh
frozen plasma (FFP).

195
 Serious adverse event
Any untoward occurrence associated with the collection, testing,
processing, storage and distribution of blood and blood components
that might lead to death or life-threatening, disabling, or incapacitating
conditions for patients or which results in, or prolongs, hospitalization or
morbidity.

Serious adverse reaction

An unintended response in a donor associated with immunization that is
fatal, life-threatening, disabling, incapacitating, or which results in, or
prolongs, hospitalization or morbidity.

Source plasma
Plasma obtained by plasmapheresis for further fractionation into plasma
products.

Traceability
Ability to trace each individual unit of blood or blood component derived
thereof from the donor to its final destination, whether this is a recipient,
one or more batches of medicinal product or disposal. The term is used to
describe both forward tracing (donation to disposition) and reverse tracing
(disposition to donation).

TT virus (TTV)
A non-enveloped, single-stranded DNA virus, causing post-transfusion
hepatitis of unknown etiology.

Viral inactivation
A process of enhancing viral safety in which the virus is intentionally
“killed”.

Viral removal
A process of enhancing viral safety by removing or separating the virus
from the protein(s) of interest.

West Nile virus (WNV)

An enveloped single-stranded RNA virus, causative agent of West Nile
fever.

4 General considerations
4.1 Range of products made from human blood and plasma
Human blood is the source of a range of medicinal products. Blood products
obtained from the processing of single donations of blood or plasma,

196
 generally known as blood components, include red cell concentrates,
platelet concentrates, leukocyte concentrates and plasma for transfusion.
Small pools, usually of less than 10 donations, mainly for the production of
platelet concentrates, can also be prepared by blood establishments. Small
pool cryoprecipitate is produced in some countries. The safety of these
blood components depends largely on the criteria used for selection of the
donors and the screening of donations.
Other blood products are obtained by the industrial processing of plasma
of a large number of donations (up to tens of thousands) that are pooled
together. These products include pooled virally-inactivated plasma for
transfusion that is not fractionated, and the purified plasma products, also
known as plasma derivatives, that are obtained by a fractionation process
that combines protein purification and viral inactivation and removal
steps.
Table 1 summarizes the range of products made from human blood and
plasma, illustrating the diversity of source material and manufacturing
methods involved, and, consequently, the complex regulation needed to
ensure their quality and safety, in particular with regard to the control of
risks of infection.
Plasma-derived products are regarded as medicinal products worldwide
and their marketing authorization, which involves the official approval of
the production process and quality assurance (QA) system used as well as
of product efficacy, should be the responsibility of the national regulatory
authority in all Member States. The national regulatory authority has the
duty to enforce regulations, to evaluate the quality and safety of products,
and to conduct regular assessment and inspection of the manufacturing
sites.
An important part of the evaluation of the marketing authorization for
plasma products relates to the production and control of the starting plasma
used for fractionation, and is the focus of these Guidelines.

4.2 Composition of human plasma

Human plasma is a complex biological material composed of hundreds of
biochemical entities, some of which have not yet been fully characterized.
Among these are albumin, various classes of immunoglobulins, coagulation
factors, anticoagulants, protease inhibitors, and growth factors. The
complexity of plasma is illustrated in the Table 2.
The concentrations of the various protein components vary from about
40 g/litre (albumin) down to a few nanograms/ml for some coagulation
factors. Plasma protein molecular mass varies from several million daltons

197
 (the von Willebrand multimer complex) to tens of thousands Daltons (for
example, albumin).

Table 1
Range of blood/plasma products derived from single donor or pooled donations

Single-donor blood components

■ Whole blood
■ Red cell concentrate
■ Platelet concentrate (obtained by apheresis)
■ Leukocyte concentrate
■ Plasma for transfusion
■ Cryoprecipitate
■ Cryo-poor plasma

Small-pool blood components

■ Platelet concentrates (obtained from whole blood)a
■ Cryoprecipitateb

Large-pool, unfractionated virally inactivated plasma product

■ Plasma for transfusion, solvent-detergent (SD) treated (4)

Large-pool products purified by fractionation of plasma

■ See the list of products in Appendix 1

a
Usually 4–10 platelet concentrates derived either from platelet-rich-plasma or from buffy coats.
b
Rarely produced. Pooled cryoprecipitate should ideally be subjected to a viral inactivation treatment. Also used
as a fibrinogen source for fibrin sealant (fibrin glue).

Human plasma for fractionation is the starting material for the manufacture
of a range of medicinal products used for the treatment of a variety of life-
threatening injuries and diseases. A list which includes the most established
clinical use of these products is provided in Appendix 1.

4.3 Pathogens present in blood and plasma

A number of infectious agents can be present in human blood but not all
blood-borne pathogens can be transmitted by plasma for transfusion or by
plasma derivatives (7). Some pathogens are exclusively associated with
blood cells, or are at least partially sensitive to the freeze–thaw process
that takes place during the manufacture of plasma and plasma products. In
addition, the multiple sterilizing filtration steps systematically included in
the manufacture of plasma products, as for any other parenteral preparation,
eliminate micro-organisms larger than 0.2 µm. Table 3 summarizes the
major infectious risks linked to blood-borne pathogens and presents the
current evidence on risks of infection from cellular components, plasma
and fractionated plasma products.

198
Table 2
Selected proteins of human plasma

Major proteins
Daltons mg/litre
• Albumin 68 000 40 000
• IgG 150 000 12 500
Protease inhibitors
• Alpha-2-macroglobulin 815 000 2 600
• Alpha-1-antitrypsin 52 000 1 500
• C1-esterase inhibitor 104 000 170
• Antithrombin 58 000 100
• Heparin cofactor II 65 000 100
• Alpha-2-antiplasmin 69 000 770
Protease
• ADAMTS13 190 771
Fibrinolytic proteins
• Plasminogen 92 000 200
• Histidine-rich glycoprotein 75 000 100
Coagulation factors and anti-coagulant proteins
• Fibrinogen 340 000 3 000
• Fibronectin 250 000 300
• Prothrombin 72 000 150
• Factor XIII 320 000 730
• Protein S 69 000 729
• Von Willebrand Factor (monomer) 220 000 710
• Factor IIa 72 000 150
• Factor X 59 000 710
• Factor V 286 000 777
• Factor XI 80 000 775
• Factor IX 57 000 775
• Factor XII 76 000 740
• Protein C 57 000 774
• Factor VII 50 000 7 0.5
• Factor VIII 330 000 7 0.3
Cytokinesb
• Interleukin-2 15 000 Traces
• Granulocyte colony-stimulating factor (G-CSF) 20 000 < 30 pg/ml
• Erythropoietin 34 000 0.3 µg/litre
Source: Adapted from references 5 and 6.
a
Factor II is the zymogen plasma protein which upon activation generates thrombin, one of the components of
fibrin sealant (fibrin glue).
b
There are several cytokines present in traces in plasma. G-CSF and erythropoietin for therapeutic use are
obtained by recombinant technology.

Some of the viruses listed in Table 3 are highly pathogenic (e.g. HIV, HCV
and HBV), others are pathogenic only in certain recipient populations (e.g.
cytomegalovirus (CMV) and B19) and a few are currently considered to be
non-pathogenic (HGV and TTV).
Historically, clinical use of single-donor blood components and pooled
plasma products (plasma derivatives) has been associated with transmission
of blood-borne viruses (HBV, HCV, HIV, HAV and B19) (3). The

199
 implementation of validated virus inactivation and removal steps into the
manufacturing process of plasma derivatives has now virtually eliminated
the risks of infection from HIV, HBV, and HCV (3) and has also avoided the
transmission of some emerging infectious agents, such as WNV (8, 9).
The infective agents for the bacterial and parasitic infections most commonly
associated with transfusions of cellular blood components are reliably
removed during the processing and aseptic filtration of plasma products, as
are residual blood cells.
Table 3
Evidence of transmission of infectious agents by human blooda

Cellular blood Plasma

Infectious agents Components Plasma products
Viruses
HIV I and II + + +
HBV + + +
HCV + + +
Hepatitis Delta virus + + +
HAV + + +
HEV + + +
HGV + + +
TT virus + + +
Parvovirus B19 + + +
Human T-cell leukaemia virus I and II + – –
Cytomegalovirus + – –
Epstein–Barr virus + – –
West Nile virus + ? –
Dengue virus + ? –
Human herpes virus-8 ? – –
Simian foamy virus ?b ? –
Severe acute respiratory syndrome (SARS) virus ?c ? –
Bacteria
Spirochaete (syphilis) + – –
Parasites
Babesia microti (babesiosis) + – –
Plasmodium falciparum (malaria) + – –
Leishmania (Leishmaniasis) + – –
Trypanosoma cruzi (Chagas Disease) + – –
Unconventional agents /TSE
Creutzfeldt Jakob disease agent – – –
Variant Creutzfeld Jakob disease agent + ? –d
HIV, human immunodeficiency virus; HBV, hepatitis B virus; HCV, hepatitis C virus; HAV, hepatitis A virus; HEV,
hepatitis E virus; HGV, hepatitis G virus; TSE, transmissible spongiform encephalopathies.
+, evidence of transmission; –, no evidence of transmission; ?, questionable or unknown.
a
Most viral transmissions associated with plasma products took place prior to the introduction of efficient viral
inactivation or removal procedures.
b
Transmitted by contact with animal blood but not reported to follow transfusion.
c
Limited epidemiological surveys have not revealed transmission of SARS coronavirus by transfusion but further
confirmation may be needed.
d
Investigational studies performed by plasma fractionators using spiked TSE agents indicate that several
purification steps used in the manufacture of some plasma products are likely to remove prion agents. These data
may not necessarily be extrapolated to clearance of the endogenous form of the TSE agent in human blood.

200
4.4 Strategies to ensure safety of plasma products
A combination of measures to exclude infectious donations, together
with steps to inactivate or remove potential contaminating viruses during
processing, has significantly reduced the risk of disease transmission by
plasma products.
There are four distinct complementary approaches to virus risk reduction
for plasma products:
• minimizing the virus content of the plasma pool by:
— implementing a quality system to select donors;
— screening blood/plasma donations; and
— performing plasma manufacturing pool testing.
• inactivating and removing residual viruses during plasma fractionation
and processing (3);
• adherence to GMP at all steps of the production; and
• recognizing and responding appropriately to post-donation events
affecting plasma donations that have already been processed.
In-process and finished product virus inactivation and/or removal procedures
have been shown to play a powerful role in ensuring the viral safety of
plasma products, in particular from HIV, HBV, and HCV risks (3). Those
procedures have also recently been shown to provide a sufficient margin of
safety against emerging lipid-enveloped viruses, such as WNV (8, 9).
Although procedures for the inactivation and removal of viruses may
therefore seem to offer the fractionator an ideal means for counterbalancing
occasional lapses in the identification of risk donations, such an assumption
would be incorrect. As powerful as the contribution of properly validated and
implemented steps for virus inactivation and removal has been shown to be,
it remains essential to limit the virus load at the stage of the plasma pool by
avoiding, through donor selection and donation screenings, the inclusion of
a high-titre infectious donation. The synergistic effects of reduced viral load
in the plasma pool and validated viral inactivation and removal procedures
are well illustrated for resistant non-enveloped viruses, such as parvovirus
B19, for which viral reduction procedures used during fractionation alone
may not be sufficient to ensure safety (10, 11).
Exclusion of infectious donations, and retrospective identification of any
infectious donation that would have passed undetected through the screening
and testing nets, require the highest priority at the blood establishment.
The blood establishment should establish a reliable mechanism to ensure
consistent identification of such donations.
Neither of the sets of measures described above can, in isolation, provide
sufficient assurance of safety against all potential blood-borne pathogens.

201
 For this reason, the manufacture of plasma for fractionation according to
Good Manufacturing Practices (GMP) is necessary to ensure the optimal
quality and margin of safety of this raw material for the manufacture of
medicinal plasma products.

5. Measures to exclude infectious donations

The safety and quality of plasma for fractionation results from the
combination of several cumulative prevention measures
— appropriate selection of blood/plasma donors;
— testing of blood/plasma donations;
— epidemiological surveillance of the donor population;
— strict adherence to GMP; and
— post-donation information system.
Such information on collection and testing of plasma is requested by
some regulatory authorities as part of a plasma master file (12) used in
the evaluation of the marketing authorization of plasma-derived medicinal
products. However, the plasma master file is not a universally used regulatory
document.

5.1 Appropriate selection of blood/plasma donors

Plasma for fractionation should be obtained from carefully selected, healthy
donors who, after review of their medical history (the donor questionnaire),
medical examination and laboratory blood tests, would be considered not to
present an increased risk for transmission of infectious agents by plasma-
derived products (see Appendix 2). Local national regulatory authorities
are pivotal in setting up at the national level a harmonized donor selection
criteria framework appropriate to the country in which plasma is collected,
taking into account the type of products to be manufactured, the relevant
risks of infection, and the epidemiological situation. The local national
regulatory authority should also be part of any decision making process
intended to modify the donor selection and donation testing procedures.
Specific selection criteria may be added by the plasma fractionator as part
of the contractual agreement with the provider of plasma.
Regulatory agencies and a number of organizations have published
regulations and recommendations concerning the criteria for the selection
of donors of whole blood and of plasma obtained by apheresis (see for
instance Guide to the preparation, use and quality assurance of blood
components of the Council of Europe (13). In general these regulations and
recommendations can be used as reference documents for the collection
of plasma for fractionation, although some specifications may differ from
those of plasma for transfusion. Examples of criteria for the selection

202
 of donors for the collection of plasma for fractionation are presented in
Appendix 2. These are not intended to constitute an absolute reference or an
exhaustive list of requirements, but rather to provide examples and explain
critical points for consideration.
A regular donor is someone who routinely donates blood or plasma in the
same centre in accordance with the minimum time intervals. The period
taken into account may vary from country to country. A repeat donor is
someone who has donated before in the same establishment, but not within
the period of time required to be considered as regular donation. Plasma
fractionators may implement their own criteria for donors’ eligibility to
improve safety margins. Whenever possible, plasma for fractionation should
be collected through a donation system that relies on regular and repeat
donors. Obtaining plasma from regular and repeat donors makes a major
contribution to ensuring optimal historical medical information about the
donors, and therefore to detecting potential risk factors.
In some countries family or replacement donors may constitute a significant
proportion of the population of blood plasma donors, and — depending on
the situations — have been found (14) or not (15) to be at a higher risk than
regular/repeat donors of having markers of viral infections. The decision
to use this plasma for fractionation is to be made jointly by the plasma
fractionator and the national regulatory authorities and should be based on
both a careful epidemiological assessment and the evaluation of other safety
measures in place for screening of donations for viruses.
Plasma may be collected by plasmapheresis from donors who have acquired
immunity through natural infection or through active immunization. Specific
information on this issue can be found in Appendix 3.

5.2 Screening of blood/plasma donations for infectious markers

5.2.1 Screening tests
The following tests, considered mandatory by all regulatory agencies,
are relevant to the preparation of plasma for fractionation and should be
performed on each blood or plasma donation:
— an approved test for HBsAg;
— an approved test for anti-HIV; and
— an approved test for anti-HCV.
The results of all three tests should be negative. Testing for HIV p24
antigen and HCV core antigens may increase the sensitivity. Initially
reactive donations should be retested in duplicate by the same assay. A
repeatedly reactive donation should not be used for therapeutic applications
and should usually be destroyed unless useful for non-therapeutic use

203
 or investigations. A sample of the donation should be evaluated by a
confirmatory test and if confirmation is positive a system should exist to
notify and counsel the donor. It is recommended that national algorithms
should be developed and used to enable consistent resolution of discordant
or unconfirmed results.

5.2.2 Other tests

The screening of plasma for fractionation for anti-human T-lymphotropic
virus (HTLV) is not required as the virus is cell-associated and susceptible
to inactivation by the freeze–thaw process.
In some countries, testing for anti-HBc is performed on whole blood
donations as a means to reduce the risks of exposure to donations of blood
components that are hepatitis B-positive (16). However, donations of plasma
for fractionation obtained from whole blood that are both anti-HBc positive
and HBsAg negative, and which contain a sufficient titre of antibodies against
hepatitis B surface antigen (anti-HBs) are usually used for fractionation: the
scientific rationale is to maintain a sufficient titre of anti-HBs antibody in
the plasma pool to neutralize any HBV that may be present. The minimum
titre of anti-HBs for an anti-HBc positive/HBsAg negative plasma donation
to be accepted for fractionation may be specified by the plasma fractionator
and/or the national regulatory authority. Currently, the minimum titre of
anti-HBs antibodies required by some plasma fractionators ranges from 50
to 100 IU/l. Alternatively, the plasma donation may be identified by the
plasma collector as being anti-HBc positive and the plasma fractionator
may conduct additional tests. The setting of a minimum limit, if any, for
the titre of anti-HBs antibody usually involves a risk assessment that takes
into consideration the sensitivity of the HBsAg screening test, the testing
or not of HBV by nucleic acid testing (NAT), and the efficiency of the viral
reduction techniques (3, 17).
Additional testing for other agents or markers may be required by the
national regulatory authority, taking into consideration the epidemiological
situation in any given area or country, or the frequency of donating blood or
plasma, and at the specific request of the plasma fractionator.

5.2.3 Nucleic acid testing

NAT of plasma for fractionation may be performed for the following
viruses: HCV, HBV, HIV, HAV, and/or B19. If NAT is performed by the
fractionator, following current practice using mini-pool samples, a specific
logistics system may have to be developed at the blood establishments
to collect and provide labelled samples in a form suitable for the test. In
addition to performing mini-pool testing, fractionators re-test the plasma
manufacturing pool for the absence of various viral markers.

204
5.2.4 Test kits
A system should exist in the country or region for approval of tests kits,
such as an official approval system by the national regulatory agency or a
delegated laboratory. The required sensitivity of the tests for the different
antigens or antibodies should be determined by the national regulatory
authority. In addition, the test kits to be used should be agreed by the
fractionator that will receive the plasma for fractionation.

5.2.5 Quality control of screening

The quality of the screening of blood/plasma donations relies on a number
of measures, such as:
— validation of new techniques before implementation;
— internal control of reagents and techniques on a daily basis;
— confirmation of positive tests by an appropriate laboratory; and
— external proficiency testing which involves the testing of a panel of sera
circulated to laboratories by an approved reference institution.
Details on sampling, test equipment, validation of performance of assays,
test interpretation and downloading and follow-up of reactives can be found
in section 7 on QA and GMP in these guidelines.

5.2.6 Look-back
A system should be in place to perform a look-back procedure,
preferably using a computer database. A look-back is a procedure to
be followed if it is found retrospectively that a donation should have
been excluded from processing, e.g. because that unit was collected
from a donor who was subsequently rejected because of reactive viral
marker, risk behaviour, exposure to CJD/vCJD or other risks related to
infectious diseases. The blood establishment should then transmit this
information to the fractionator according to the agreements in place, and
to the national regulatory authority. Donor notification and counselling
are recommended both for purposes of donor health and for the safety of
the blood supply.

5.3 Epidemiological surveillance of donor population

To ensure optimal long-term safety of plasma for fractionation, it is highly
recommended to establish continuous epidemiological surveillance of the
donor population This is not a requirement in all regions of the world. The
objective of this survey is to know, as precisely as possible, the prevalence
and incidence, and their trends, of infectious markers that are relevant to the
safety of medicinal plasma products so that counter-measures can be taken
in a timely fashion.

205
 The system should not only be able to gather epidemiological data at
the national and regional level but also among the donor populations
that are providing blood or plasma for fractionation at individual blood
establishments within a country or a region.
The information from the epidemiological surveillance can furthermore be
used:
• to detect differences among donor populations of various collection
centres which may be associated with objective differences in viral
markers within donor populations or may reflect differences in the process
of donor selection and screening among collection centres;
• to detect trends in infectious markers which may reflect either a
change in the rate of viral markers in the population or a possible
deviation in the donor selection or screening process at specific collection
sites;
• to assess the relevance of any prevention measures such as a strengthened
donor selection process, additional exclusion criteria, or implementation
of additional screening tests to avoid contamination of plasma products.
When donations from first-time donors are used to prepare plasma for
fractionation, epidemiological data on this specific group of donors
should be included in the estimation of the risk for infectious diseases
transmitted by blood. Indeed, it has been shown that first-time donors,
who may occasionally include test-seeking individuals, constitute a group
which in some situations is more likely to have blood-borne viral markers
than regular donors group who have already gone through a selection or
deferral process (18–21). Some plasma fractionators do not fractionate
plasma from first-time donors as prevalence of infectious diseases may be
higher in this donor group. Currently, it is advisable to collect and analyse
epidemiological data at the collection sites for anti-HIV-1 and anti-
HIV-2, anti-HCV and HBsAg, since, historically, they represent the major
pathogenic risks associated with plasma products. It is the responsibility
of the local national regulatory authority to determine whether the
list should be modified or should include additional criteria, such as
emerging infectious agents, based on local or regional epidemiology. For the
three currently recommended markers, only confirmed positive tests (i.e.
tests that are repeatedly reactive in a screening test and positive in at least
one confirmatory test) should be recorded. When the plasma fractionator
performs additional tests (such as NAT tests) on donations which gave
negative results in serological tests, the results should be reported.
Recent guidelines published by the European Medicines Agency (EMEA)
entitled Guideline on epidemiological data on blood transmissible infections
(22) describe how to conduct epidemiological surveillance of the donor
population.

206
5.4 Strict adherence to good manufacturing practices
Because the pooling of thousands of plasma donations is required for the
manufacture of plasma derived medicinal products, it is necessary to ensure
full traceability between individual blood/plasma units collected and the
final plasma products manufactured. This is important to enable any quality
and safety problems, in particular problems related to infectious risks, to
be traced back to individual blood/plasma donations and to allow relevant
measures to be taken to protect the donors as well as the patients who
received the plasma-derived medicinal products.
The donor selection process, the collection of blood or plasma and the
processing of the donation, in order to obtain plasma for fractionation,
represent the first steps in the manufacturing of plasma-derived medicinal
products, and therefore should be performed in compliance with GMP.
Strict adherence to the principles of GMP and the implementation of a QA
system to address and comply with the requirements of GMP is crucial at
all stages of the production of plasma for fractionation (See section 7 on
QA and GMP in these Guidelines).

5.5 Post-donation events

There should be a system to ensure effective communication between the
blood establishment and the fractionator so that information on significant
post-donation events may be immediately transmitted to the fractionator
and the national regulatory authority. In particular, this procedure should
allow early and effective communication of any evidence for the presence
of blood-transmissible infection in a donor whose plasma has been sent for
fractionation.

6. Production of plasma for fractionation

6.1 Methods used to obtain plasma for fractionation
Human plasma for fractionation may be obtained by separation of plasma
from whole blood, or by apheresis.

6.1.1 Recovered plasma

Recovered plasma is plasma recovered by centrifugal separation from the
cells and cellular debris of whole blood under the conditions described
below.

6.1.2 Apheresis plasma (source plasma)

Apheresis plasma is obtained by a procedure in which anticoagulant-treated
blood is removed from the donor, the plasma is separated from the formed

207
 elements, and at a minimum the red cells are returned to the donor. The
separation of cellular elements and plasma may be achieved either by
centrifugation or filtration. The equipment used for the collection of plasma
by automated methods is designed for separating cellular elements and
plasma by centrifugation or filtration. The manufacturers of the equipment
provide operating manuals that include instructions for installation
validation, routine preventive maintenance, periodic performance checks
(e.g. weight scale checks), alert mechanisms (e.g. haemoglobin detector)
and troubleshooting. Annual preventive maintenance should be performed
by a qualified field service engineer. It includes e.g. visual inspection, initial
operational integrity, inspection of equipment integrity, inspection of filter
and/or centrifuge, calibration testing and safety testing. In addition, the
manufacturers of the equipment usually provide support for the installation
and train on-site technicians to maintain the equipment. Apheresis collection
potentially increases the availability of plasma for fractionation, enabling
higher donation frequency and a larger volume per donation, and is the
preferred approach for the regular collection of plasma from hyperimmune
donors who have high antibody titres against specific disorders.
In principle, the method of preparation should remove cells and cell debris
as completely as possible and should be designed to prevent the introduction
of microorganisms. No antibacterial or antifungal agent is added to the
plasma. The residual blood cell content of the plasma, in the absence of
dedicated leukoreduction filtration, may vary with the collection method.

6.2 Characteristics of plasma for fractionation

6.2.1 Plasma frozen within 24 hours of collection
Subject to appropriate handling (storage and transport), plasma frozen, at
–20 °C or –30 °C, within 24 hours of blood collection or apheresis (see
section 6.6.2.1) will normally be suitable for optimal recovery of both labile
factors (factor VIII and other coagulation factors and inhibitors) and stable
plasma proteins (usually albumin and immunoglobulins). Plasma meeting
these quality specifications is also used for direct clinical applications; it
is then referred to as fresh frozen plasma (FFP), clinical plasma or plasma
for transfusion. Table 4 sets out the main characteristics of plasma prepared
either from whole blood (recovered plasma) or by apheresis.
Both sources of plasma have been found by experience to be appropriate
for the manufacture of the whole range of plasma products. That said, the
method of collection and preparation has some impact on the characteristics
and/or yield of the proteins fractionated from the plasma. Apheresis plasma
collected from donors undergoing frequent plasmapheresis contains lower
levels of IgG than plasma units produced by moderate serial plasmapheresis
or from whole blood (23, 24). The content of various coagulation factors is

208
 usually higher in apheresis plasma than in recovered plasma (24, 25), for
various reasons that include rapid separation of blood cells and plasma,
differing ratios of anticoagulant added, and the possibility of freezing the
apheresis plasma soon after completion of collection.

Table 4
Characteristics of plasma for fractionation used in the manufacture
of labile plasma products

Characteristic Recovered plasma Apheresis

plasma
Volume, ml 100–260a 450–880b
Protein content, g/l ≥ 50 (13) ≥ 50
(each donation) (but typically greater than in apheresis plasma)
Factor VIII, IU/ml ≥ 0.7 (26) ≥ 0.7
(average) (but typically less than in apheresis plasma)
Concentration of Variable, according to donation size (volume Constant
anticoagulant of anticoagulant is fixed for a given pack type; (metered into
the acceptable blood volume range should be donation)
specified)
Acceptable donation Determined nationally, usually a maximum of Determined
frequency one donation every 2 months nationally
a
Based on a standard donation size of 450 ml, with blood:anticoagulant ratio of 7:1. The maximum volume of
blood to be collected during one donation procedure is determined by national authorities.
b
With anticoagulant. The maximum volume of plasma to be removed during one plasmapheresis procedure is
determined by national authorities.

Preservation of factor VIII and other labile factors depends on the collection
procedure and on the subsequent handling of the blood and plasma. With
good practice, an average of 0.7 IU/ml factor VIII can usually be achieved
both with apheresis and recovered plasma. Units of plasma for fractionation
with a lower activity may still be suitable for use in the production of
coagulation factor concentrates, although the final product yield may be
reduced.
The implementation of GMP in the preparation of plasma for fractionation
should ensure that the plasma bioburden is controlled, labile proteins are
conserved as far as possible, and minimal proteolytic activity is generated.

6.2.2 Plasma frozen after 24 hours of collection

Plasma may be available that does not fulfil the above-defined criteria but
still has value as a source of some plasma proteins. This would include:
• plasma separated from whole blood and frozen more than 24 h but usually
less than 72 h after collection;
• plasma, separated from whole blood stored at 4 °C, and frozen within
72 h of separation but within the assigned shelf-life of the blood);

209
 • Plasma frozen within 24 h but stored under conditions that preclude its
use for the manufacture of coagulation factors.
Provided the circumstances of manufacture and storage of such plasma
does not result in increased bioburden, the plasma may be considered
suitable for the manufacture of stable plasma proteins, but not coagulation
factors.
Plasma which is not frozen within 72 h of collection or separation from
whole blood should not be used for fractionation.

6.2.3 Plasma not meeting the requirement for fractionation

Plasma obtained by therapeutic plasma exchange does not meet the criteria
for fractionation into plasma products. Indeed, plasma from individuals
subjected to therapeutic plasma exchange for the treatment of a disease
state may present an enhanced risk of transmitting blood-borne diseases
(due to infectious risks associated with plasma) and a high risk of
irregular antibodies, and should not be offered for fractionation. In addition,
such plasma cannot be classified as being obtained from a voluntary
donor.
Plasma from autologous blood donations is excluded from use as plasma
for fractionation and may have higher prevalence of viral markers (27).

6.2.4 Hyper-immune (antibody-specific) plasma

Detailed information regarding immunization of donors for the preparation
of hyperimmune plasma is provided in Appendix 3. The following are the
three approaches for the preparation of plasma for the manufacture of
specific immunoglobulins (antibody-specific immunoglobulins):
• Individuals selected from the normal population by screening of plasma
units for antibody titres (Screening may be random, or may be informed
by knowledge of history of recovery from an infectious disease — for
example varicella).
• Individuals with a high titre of a specific antibody resulting from
prophylactic immunization.
• Volunteers recruited to a panel for a targeted immunization programme.
The clinical and ethical requirements for such a programme are considered
in Appendix 3.
Clinically relevant specific immunoglobulins include anti-D (anti-Rho), and
HAV, HBs, tetanus, varicella/herpes zoster and rabies immunoglobulins.
Hyperimmune globulins are prepared for intramuscular administration, but
products for intravenous use are also available. The typical derivation of
hyperimmune plasma of each specificity is summarized in Table 5.

210
Table 5
Types of hyperimmune plasma

Specificity Natural Prophylactic Targeted

immunity immunization immunization

Anti-D (anti-Rho) Yes No Yes

Anti-hepatitis A (anti-HAV) Yes Yes Yes
Anti-hepatitis B (anti-HBs) Yes Yes Yes
Anti-tetanus No Yes Yes
Anti-varicella/herpes zoster Yes No Possibly
Anti-cytomegalovirus (anti-CMV) Yes No No
Anti-rabies No Yes Yes

Acceptable minimum antibody potencies in individual plasma donations

for fractionation should be agreed to by the fractionator. Those will usually
depend upon:
• the size and composition of the fractionation pool (which may include
high-titre donations to increase the mean titre of the fractionation
pool);
• the characteristics of the immunoglobulin fractionation process; and
• the minimum approved potency of the final IgG product.
The following general guidance may be useful for each specificity.

6.2.4.1 Anti-D (anti-Rho)

• Antibody potency should be estimated in international units, using an
appropriate quantitative assay (e.g. auto analyser-based assay or flow
cytometry method) agreed by the fractionator.

6.2.4.2 Anti-hepatitis A
• Antibody potency should be estimated in international units, using a
quantitative assay agreed by the fractionator.
• The minimum acceptable potency in an individual donation is unlikely to
be less than 50 IU/ml.

6.2.4.3 Anti-hepatitis B
• Antibody potency should be estimated in international units, using a
quantitative assay that detects antibody to hepatitis B surface antigen
(typically radioimmunoassay (RIA) or enzyme-linked immunoassay
(ELISA)) agreed to by the fractionator.
• The minimum acceptable potency in an individual donation is unlikely to
be less than 10 IU/ml.

211
6.2.4.4 Anti-tetanus
• Antibody potency should be estimated using either a neutralization assay
or a quantitative assay with established correlation to the neutralization
assay, agreed by the fractionator.

6.2.4.5 Anti-varicella/zoster
• Antibody potency should be estimated using a quantitative assay
(typically ELISA, immunofluorescence or complement fixation) agreed
by the fractionator.
• The minimum potency should be shown to be equal to or greater than that
of a control sample provided by the fractionator.

6.2.4.6 Anti-cytomegalovirus
• Antibody potency should be estimated using a quantitative assay
(typically ELISA, immunofluorescence or complement fixation) agreed
by the fractionator.
• The minimum potency should be shown to be equal to or greater than that
of a control sample provided by the fractionator.

6.2.4.7 Anti-rabies
• Assessing plasma for rabies antibody is rarely done. A donor may be
considered to have acceptable antibody titres between 1 and 3 months
after a second (or booster) dose of vaccine. Plasma should not be collected
from persons immunized after exposure to infection with rabies virus.

6.3 Premises and devices for collection of plasma for fractionation

6.3.1 Premises
The collection of blood or plasma for fractionation should be performed
in licensed, or regulated, permanent premises or mobile sites which are
compliant with the intended activity and comply with the GMP standards
approved by the national regulatory authority. The area for blood donors
should be separated from all processing and storage areas. The area for
donor selection should allow confidential personal interviews with due
regard for the safety of donors and personnel. Before premises are accepted
for mobile donor sessions, their suitability should be assessed against the
following criteria:
— the size (to allow proper operation and ensure donor privacy);
— safety for staff and donors; and
— adequate ventilation, electrical supply, lighting, hand washing facilities,
blood storage and transport equipment, and reliable communication
capabilities.

212
6.3.2 Containers
Because plasma is a complex and variable mix of proteins in aqueous
solution, the way in which it is handled will have consequences for its
safety, quality and quantity. Furthermore, the effects of mishandling will
not always be as simple (or as obvious) as a reduction in the content of
recoverable factor VIII — they are just as likely to affect the behaviour of
the plasma when it is thawed (this is very important to the fractionator, who
requires consistency from this particularly important process step).
The containers used for the collection and storage of plasma for fractionation
should comply with the appropriate regulatory provisions and should be
under the control of the regulatory authority. Containers should also comply
with the regulatory and technical requirements of the plasma fractionator.
Containers should be labelled with batch numbers traceable to individual
donations. The quality of containers has a direct impact on the quality of the
plasma produced and it is therefore part of GMP to control the suitability of
this starting material before use.
Containers for whole blood collections are the same as for donations of whole
blood from which plasma is used for fractionation. They should be plastic,
and should have been manufactured in such a way as to assure internal
sterility; they should be hermetically sealed to exclude contamination. If
the container is not manufactured as an integral part of a blood collection
set, there should be a mechanism for docking with the collection set that
minimizes the risk of adventitious microbial contamination.
Validation studies will be required to confirm the suitability of the container
material (and the material of any tubing or harness through which plasma should
pass) during storage in contact with the plasma. Specifically, it will be necessary
to establish that the plastic is physically compatible with the proposed methods
for freezing and opening (or thawing) the packs and to establish the quantities
of extractable materials (for example, plasticizers) during the claimed periods
of storage in the liquid and frozen forms. These studies are carried out by
the manufacturer of the containers. When using collection sets and containers
previously established by a manufacturer as being suitable, a cross-reference to
such a study may be sufficient. Validated collection and storage containers for
blood/plasma are available from several manufacturers worldwide.
The choice of the containers (e.g. type of plastic bags for recovered plasma
or plastic bags or bottles for apheresis collection) has a direct impact on
the design of the container opening machine that is used at the plasma
fractionation plant at the plasma pooling stage.

6.3.3 Anticoagulants
Most anticoagulant solutions developed and introduced for the collection of
blood cellular components and plasma for transfusion are compatible with

213
 the preparation of plasma for fractionation and with the manufacture of
plasma products (although some influence on factor VIII content in plasma
has been described (28–32)). One exception is when heparin is added to
the anticoagulant solution. The main anticoagulant solutions currently in
use for collection of either whole blood or apheresis plasma are listed in
Table 6.
Anticoagulant solutions should comply with the appropriate regulatory
provisions. They can be already present in the collection container (e.g.
plastic containers used for whole blood collection) or added to the blood
flow during apheresis procedures. In both cases, information on the device
and the anticoagulant should be provided to the regulatory authorities. The
fractionator will need to know what anticoagulant has been used, and its
concentration as these may have an impact on the fractionation process.

6.4 Blood/plasma collection process

6.4.1 Procedure
A standardized and validated procedure for the preparation of the phlebotomy
site should be followed using a suitable antiseptic solution, and should be
allowed to dry (depending on the type of disinfectant). The prepared area
should not be touched before insertion of the needle. Prior to venipuncture
the containers should be inspected for defects. Any abnormal moisture or
discoloration suggests a defect. A careful check of the identity of the donor
should be performed immediately before venipuncture.
The collection of a whole blood unit used to prepare plasma for fractionation
should be performed following already established recommendations (for
instance as described in the Council of Europe Guide (13)). In particular, good
mixing of the blood with the anticoagulant solution should be ensured as soon
as the collection process starts to avoid risks of activation of the coagulation
cascade. The mixing can be done manually, every 30 to 45 seconds, and at least
every 90 seconds. Collection of one standard unit of blood should be achieved
within 15 minutes, as longer collection periods may result in activation of the
coagulation factors and cellular components.
In automated apheresis procedures, whole blood is collected from the donor,
mixed with anticoagulant, and passed through an automated cell separator.
The plasma for fractionation is separated from the cellular components
of the blood, which are returned to the donor in a series of collection/
separation and return cycles. The plasma is separated from the red blood
cells by centrifugation or filtration, or a combination of both (33, 34). The
operational parameters of the plasmapheresis equipment are defined by
the manufacturers of the equipment and by the requirements of national
regulatory authorities. In general, the anticoagulant (often 4% sodium citrate)

214
 is delivered at a rate to yield a specified ratio of anticoagulant to blood. The
volume of plasma collected from the donor during one procedure and over a
period of time is regulated. The number of collection/separation and return
cycles for each donor depends on the total volume of plasma that is to be
collected. For determining the number of cycles employed, the equipment
requires programming by input of data. These data elements may include
such parameters as donor weight and haematocrit values. The amount of
time required for the donation procedure depends on the number of cycles
(and hence the volume of plasma collected) but is generally between 35 and
70 minutes.
Table 6
Examples of anticoagulant solutions commonly used in the preparation
of plasma for fractionation
Composition Recovered Ratio per Apheresis
plasma 100ml blood plasma
ACD-A Sodium citrate dihydrate 22.0 g/l
Citric acid hydrous 8.0 g/l
Dextrose monohydrate 25.38 g/l × 15 (×)
pH (25 °C) 4.7–5.3
ACD-B Sodium citrate dihydrate 13.2 g/l
Citric acid hydrous 8.0 g/l × 25
Dextrose monohydrate 15.18 g/l
pH (25 °C) 4.7–5.3
CPD Sodium citrate dihydrate 26.3 g/l
Citric acid hydrous 3.7 g/l
Dextrose monohydrate 25.5 g/l × 14 (×)
Sodium biphosphate 2.22 g/l
Sodium hydroxide 1 N (pH adjustment)
pH (25 °C) 5.3–5.9
CPD-A Sodium citrate dihydrate 26.3 g/l
Citric acid hydrous 2.99 g/l
Dextrose monohydrate 29 g/l × 14
Sodium biphosphate 2.22 g/l
Adenine 0.27 g/l
Sodium hydroxide 1 N (pH adjustment)
pH (25 °C) 5.3–5.9
CP2D Sodium citrate dihydrate 26.3 g/l
Citric acid hydrous 3.7 g/l
Dextrose monohydrate 50.95 g/l × 14
Sodium biphosphate 2.22 g/l
Sodium hydroxide 1 N (pH adjustment)
pH (25 °C) 5.3–5.9
4% Citrate Sodium citrate dihydrate 40 g/l
Citric acid hydrous: 6.25 ×
as required for pH adjustment
pH (25 °C) 6.4–7.5
(×), seldom used; ×, commonly used.

215
6.4.2 Labelling of collection bags
There should be a secure system for procurement, printing and storing of
the bar code labels used to identify the main collection bags and the satellite
bags, associated samples and documentation to ensure full traceability at
each stage of plasma production. There should be a defined procedure for
labelling collection bags and samples — in particular to ensure that the
labels correctly identify the association between samples and donations.
Labelling should be performed in a secure manner, e.g. at the donor couch,
prior to collection, or immediately after the start of collection, to avoid
mislabelling. Duplicate number sets of bar code donation numbers should
not be used. Information on the label of the donation should include: official
name of the product; volume or weight; unique donor identification; name
of the blood establishment; shelf-life or shelf term; shelf temperature; and
name, content and volume of anticoagulant.

6.4.3 Equipment
Equipment used for the collection and further separation of blood should be
maintained and calibrated regularly, and the collection and separation process
needs to be validated. When validating the quality of the recovered plasma, a
set of quality control tests, including measurement of total proteins, residual
blood cells, haemoglobin, and relevant coagulation factors, such as Factor VIII,
should be included. In addition, markers of activation of the coagulation and
fibrinolytic systems may, if necessary, be performed with the support of the
plasma fractionator based on the specifications of the plasma for fractionation
set out by the fractionator and/or the national regulatory authority.
Likewise, apheresis equipment and apheresis procedures should be validated,
maintained and serviced. Validation criteria for assessing the quality of plasma
for fractionation also include protein recovery, residual content of blood cell
and haemoglobin, and relevant coagulation factors. Validation studies of
new apheresis procedures should also evaluate possible risks of activation
of the coagulation, fibrinolysis, and complement systems potentially induced
by the material in contact with blood (25, 35, 36); such studies are usually
performed by the manufacturer of the apheresis machines.

6.4.4 Laboratory samples

Laboratory samples should be taken at the time of blood/plasma collection.
Procedures should be designed to avoid any mix-up of samples and samples
awaiting testing should be stored at an appropriate temperature, as specified
in the operating instructions of the test kits.

6.4.5 Volume of plasma per unit

The volume of recovered plasma per container varies depending upon the
volume of whole blood collected, the respective haematocrit of the donor, and

216
 the volume of the anticoagulant solution. The volume of apheresis plasma per
container depends directly upon the volume collected during the apheresis
session and the volume of anticoagulant. The range of volume of blood and
plasma collected per donor is usually defined in national regulations taking
into consideration criteria such as the weight of the donor.
Although in most countries the volume of whole blood collected is close
to 400–450 ml per donor, in some it may be as low as 200 ml (under those
circumstances, the volume of anticoagulant solution is reduced so that the
plasma:anticoagulant ratio is constant). As a result the volume of recovered
plasma per unit (including anticoagulant) may vary from about 100 to
260 ml per container. In the case of plasmapheresis plasma, the volume
may range from about 450 to 880 ml per container, depending upon the
regulations in the country of collection.
The volume of plasma per container has direct practical impact on the fractionation
process and manufacture of plasma products. Small-volume donations (e.g.
100 ml) will require more handling by the plasma fractionation operators at the
stage of plasma preparation, at the container opening step, and during plasma
thawing. The overall container opening process will take longer, requiring
additional care to control bacterial contamination. Another consequence is
that the number of donations contributing to a plasma pool will be higher (for
instance, 20 000 plasma donations for a pool size of 2000 litres).

6.4.6 Secure holding and reconciliation

When the collection process is finished, it should be ensured that blood/
plasma donations are held at the donation site using a secure system to
avoid mishandling.
Prior to dispatching the collected donations to the blood/plasma processing
site, reconciliation of the collected donations should be performed according
to a standardized procedure. The procedure should also specify the actions
to be taken if there are found to be missing numbers or leaking containers.
Documentation should accompany the donations to the plasma processing
site, to account for all donations in the consignment.

6.4.7 Donor call-back system

A system should be in place in the blood establishment which allows recall
of a donor if further analysis or investigation is necessary.

6.5 Separation of plasma

6.5.1 Premises
Blood processing should be carried out in adequate facilities suitable for the
needs of the intended activity. The donor area and plasma processing areas

217
 should be separated whenever possible. Each area used for processing and
storage should be secured against the entry or intervention of unauthorized
persons and should be used only for the intended purpose. Laboratory areas
and plasma storage areas should be separate from the donor and processing
areas.

6.5.2 Intermediate storage and transport

Transport of the donations and samples to the processing site should be done
according to procedures that ensure both constant approved temperature
and secure confinement. This is especially important when blood/plasma is
transported from distant blood drive sessions.
Temperature monitoring is important to ensure optimal compliance and
quality. One way is to use packaging methods that can keep the blood/
plasma within the required temperature limits. Portable temperature loggers
can be used to monitor and record temperatures during the transportation of
blood/plasma to the processing site.

6.5.3 Impact of whole-blood holding period

It has been shown that whole blood anticoagulated with CPD, transported
and stored at 22 °C for up to 8 h prior to separation of plasma is suitable
for the production of plasma for fractionation, but factor VIII activity is
reduced by an additional 15–20% if blood is stored for 24 h (37). Rapid
cooling of whole blood to 22 °C +/– 2 °C immediately after collection
(e.g. using cooling units with butane-1,4-diol) (38) protects factor VIII and
may allow storage of blood for 24 h (39). A temperature of 4 °C during
transportation or storage of blood collected with either ACD, ACD-adenine,
or CPD anticoagulants consistently appears to reduce the factor VIII content,
but not necessarily that of other proteins, especially after 8 hrs of holding
time (40–43). Holding blood at 4 °C for longer than 8 h is therefore not
recommended when plasma is used for fractionation in the manufacture of
factor VIII products.

6.5.4 Centrifugation of whole blood

Documentation on blood and plasma collection should be checked at the
processing laboratory on receipt of the donations; reconciliation between
consignment and documentation received should be performed. Blood
separation procedures should be performed using a closed system and should
be validated, documented and proven to ensure that containers are correctly
identified.
Reproducible production characteristics of the plasma for fractionation,
following a validated procedure, should ensure consistency in the residual
blood cell count and protein content and quality to meet the specifications

218
 set out by the blood establishment or the national regulatory authority and
the plasma fractionator.
Comparisons have shown that CPD whole blood units that were centrifuged
under conditions of low g force for a long time and those subjected to high g
force for a short time yielded blood components of similar quality (44). Blood
separation classically starts with the isolation of the platelet-rich plasma
(PRP) fraction from whole blood by low-speed centrifugation. Subsequent
high-speed centrifugation of PRP in turn yields the corresponding platelet
concentrate and the plasma.
Fully automated systems for blood processing including removal of the
buffy-coat layer have replaced manual extraction procedures. This allows
standardized extraction and contributes to compliance with GMP in the
preparation of blood components including plasma for fractionation (45).
Blood component separation systems may be based on buffy coat extraction
by the “top and bottom” technique (46). Its efficacy in terms of yield, purity,
and standardization of blood components has been well established.
Several technical approaches have been developed to separate blood
components. The process may involve normal centrifugation to separate the
blood components, which are subsequently squeezed out from the top and
bottom simultaneously under control of a photocell. This primary separation
step results in three components: a leukocyte-poor red-cell suspension,
plasma, and a buffy-coat preparation (46). A multiple-bag system with
top and bottom drainage of the primary bag allows automatic separation
of blood components; plasma containing 14.6 ± 5.6 × 103 platelets/µl and
0.04 ± 0.035.6 × 103 leukocytes/µl is obtained (47). Blood components may be
separated by initial high-speed centrifugation (4158 g, 14 min, 22 ºC) of whole
blood in sealed triple or quadruple bag systems, followed by simultaneous
extraction of fresh plasma at the top, and the red blood cell concentrate at the
bottom, of the respective satellite bags that constitute the blood extraction
bag system — keeping the leukocyte-platelet buffy coat layer stable
throughout the process within the original extraction bag. The buffy coat
component yields the platelet concentrate after low-speed centrifugation and
removal of the plasma from the PRP. Automatic separators that subsequently
express the various components into their respective satellite bags in top
and bottom systems yield plasma containing 3 ± 3 × 106 leukocytes and
4 ± 3 × 109 platelets per unit (48). The “top and bottom” approach allows
a marked reduction in leukocyte contamination of the different blood
components (38, 49), and may yield optimal plasma volume (38).

6.5.5 Impact of leukoreduction

Recently, several countries have implemented universal leukoreduction of the
blood supply (50, 51) to avoid cell-mediated adverse events or improve viral

219
 safety of blood components. It has also been considered as a precautionary
measure against the risk of transmission of variant Creutzfeldt-Jakob disease
(vCJD). A recent study in an endogenous animal infectivity model reports
that leukoreduction of whole blood removes 42% of the vCJD infectivity
associated with plasma (52), whereas further investigation by the same group
found a ~70% removal of infectivity (R. Rohwer, unpublished data). The
impact of leukoreduction on plasma protein recovery and activation markers
appears to depend upon the chemical nature of the filters (53, 54). Some
loss of coagulation factors and sometimes an increase in the markers of
coagulation and complement activation has been found, although the impact
on the quality of fractionated plasma derivatives is not known (54, 55).
Therefore, until more scientific data are gathered, the benefits of
leukoreduction for the quality and safety of plasma products remains
debated. The decision to leukoreduce plasma for fractionation should be
assessed with the plasma fractionator and the national regulatory authority.

6.6 Freezing of plasma

Freezing is an important processing step that has an impact on some aspects
of the quality of plasma for fractionation, in particular with regard to the
content of factor VIII.
Several aspects of the freezing conditions of plasma for fractionation have
been evaluated.

6.6.1 Holding time of plasma

Holding plasma, freshly harvested from CPD-whole blood, at ~ 4 °C for up to
24 h before freezing at –20 °C for 4 months was shown to induce almost 25%
loss of factor VIII activity compared to that in plasma frozen immediately,
whereas other coagulation factors were not affected (56). Storing plasma at
22 °C for 2–4 h does not seem to induce a significant loss of factor VIII
activity; however, after 4 h, some loss of activity takes place (41, 57).
Therefore, placing recovered plasma in a freezer as soon as possible, or at
least within 4 h, after separation from cellular elements, would be favourable
to the recovery of factor VIII. Similarly apheresis plasma should be frozen
as soon as possible after completion of the collection procedure.

6.6.2 Freezing rate and freezing temperature

6.6.2.1 Freezing conditions
The regulatory requirements for the temperature at which plasma should be
frozen vary (58), and depend upon the type of proteins fractionated.
The fractionator may also wish to specify freezing conditions depending on
the intended use of the plasma.

220
 The European pharmacopoeia currently states that recovered or apheresis
plasma for fractionation to be used for the manufacturing of labile proteins
(e.g. production of factor VIII concentrate) should be frozen rapidly, within
24 hours of collection, at –30 °C or colder (26), as this temperature has
long been claimed to ensure complete solidification (59), and to be needed
for optimal freezing (60). However, freezing conditions are currently under
debate and the wording used in the European pharmacopoeia monograph
may be revised. Recovered plasma used to manufacture only stable plasma
proteins (e.g. albumin and immunoglobulins) should be frozen within 72 h
of collection at –20 °C or colder.
The US Code of Federal Regulations specify that plasma collected by
apheresis and intended as source material for further manufacturing should
be stored at –20 °C or colder immediately after collection.
The rate at which freezing proceeds is considered to be an important quality
factor, at least when coagulation factors are intended to be produced (61, 62).
Rapid freezing of plasma prevents or reduces loss of factor VIII in frozen
plasma either recovered or obtained by apheresis (23, 63, 64), whereas slow
freezing of plasma has been shown to influence the purity and recovery of
factor VIII in cryoprecipitate (61, 64–66). An ice front velocity of 26 mm/h
during freezing was recently shown to preserve factor VIII:C in plasma
better than 9 mm/h or less (57).
Therefore, freezing plasma rapidly (typically in less than 2 h, so as to ensure
a high ice front velocity) down to a core temperature of at least –20 °C, and
preferably colder, appears to be the best approach for the preservation of
labile proteins.

6.6.2.2 Impact of containers and equipment

To ensure optimal and consistent freezing and storage conditions, it
is important to use standardized plasma containers as freezing time is
influenced by container shape, volume and thickness (57, 64, 65).
Optimum conditions used by some plasma collectors to ensure reproducible
freezing are achieved by freezing “well separated” plasma packs in a stream
of moving cold air at the lowest temperature tolerable to the plastic of the
pack (a so-called “blast freezer”), and then to store the frozen packs “close-
packed” in a storage freezer at the agreed upon storage temperature. The
worst case would be to place a large number of unfrozen plasma bags, close
together, in a domestic (–18 °C to –22 °C) freezer, adding more plasma bags
for freezing each day, and storing the plasma under these conditions for
several months. With good practice at the time of loading (i.e. not putting
too many packs in at the same time and keeping them separated), a walk-in
freezer at a suitable temperature offers a workable compromise.

221
 The plasma fractionator should specify precisely to the plasma collector,
with the approval of the national regulatory authority, which precise freezing
parameters to use.

6.6.2.3 Validation of the freezing process

Recovered plasma and apheresis plasma should be shown to be frozen
in a consistent manner at the required temperature. A system should be in
place for ensuring that plasma is frozen to the correct core temperature
within the time limit agreed upon with the plasma fractionator, keeping
in mind that the speed of freezing will be influenced by the type of
plasma container as well as by the volume of plasma (64). Validation of
the freezing process by recording the temperature of plasma donations
during a freezing process allows evaluating the freezing capacity of the
equipment to be evaluated. Validation studies should be available, and
should demonstrate that the temperature of a frozen pack reaches the
proposed storage temperature following the specifications agreed upon
with the manufacturer.
As indicated above, the aim should be to achieve rapid freezing, and
thereafter to minimize temperature changes to the frozen plasma.

6.7 Storage of plasma

6.7.1 Storage conditions and validation
Plasma for fractionation should be stored at –20 °C or colder.
A multicentre study showed no detectable storage-related changes in three
pools of plasma (2 recovered CPD plasma and 1 apheresis plasma) that
had been quick-frozen at –30 °C, or colder, and stored over a period of
36 months at –20 °C, -25 °C, -30 °C, or –40 °C. An 11% reduction in
factor IX was found in one of the recovered plasma pools during storage at
–20 °C for 2 years (67). The authors concluded that plasma may be stored at
–20 °C for 2 years, or at –25 °C, –30 °C, or –40 °C for 3 years.
By keeping the average storage temperature of the frozen plasma as
constant as possible, at or below –20 °C, the original quality of the plasma
is maintained, without having any impact on the fractionation process, in
particular the cryoprecipitation step (60, 61, 66).
The European pharmacopoeia has a provision stating that if the temperature
of the plasma is between –20 °C and –15 °C for a maximum of 72 h, or if it
is above –15 °C (but colder than –5 °C) in no more than one occurrence, the
plasma can still be used for fractionation. Therefore, maintaining a constant
storage temperature of –20 °C or colder is a recommended approach to
ensure a consistent and optimal plasma quality.

222
6.7.2 Premises and equipment
Storage conditions should be controlled, monitored and checked. Temperature
records should be available to prove that the full plasma containment is stored
at the temperature agreed upon with the plasma fractionator throughout the
storage area. Appropriate alarms should be present and regularly checked; the
checks should be recorded. Appropriate actions on alarms should be defined.
Areas for storage should be secured against the entry of unauthorised persons
and should be used only for the intended purpose. Storage areas should provide
effective segregation of quarantined and released materials or components.
There should be a separate area for rejected components and material.
If a temporary breakdown of the freezing machine or failure of the electricity
supply occurs (e.g. electricity used for the stored plasma), examination of the
temperature records should be made together with the plasma fractionator
to evaluate the impact on plasma quality.

6.7.3 Segregation procedures

The following should be taken into account in the storage and boxing of
plasma for fractionation.
• Untested plasma and released plasma should be stored in separate
freezers, or if both types of plasma are stored in a single freezer a secure
segregation system should be used.
• Initially reactive plasma donations should be stored in a separate quarantine
freezer or a secure system (e.g. validated computer hold system) should
be used to prevent boxing of non-released plasma.
• Donations that are found to be unacceptable for fractionation should be
retrieved, disinfected and discarded using a secure system.
• Plasma donations for shipment to the plasma fractionator should be boxed
in a secure manner and an effective procedure (such as a computerized
system) should exist to make sure that only fully tested and released
plasma donations are boxed.
• Prior to shipment, plasma boxes should be reconciled appropriately.
• Prior to release of the plasma shipment to the fractionator, there should be
a formal review of the documentation to ensure that the plasma shipped
complies fully with the specifications agreed upon with the plasma
fractionator.
The goal of the above-mentioned measures is to make sure that donations that
do not comply with the specifications agreed upon with the fractionator will
not be released and shipped, and that traceability of donations is ensured.

6.8 Compliance with plasma fractionator requirements

Any plasma collected and prepared for fractionation should meet the
plasma product manufacturer requirements as the specifications of plasma

223
 for fractionation are part of the marketing authorization granted by the
national regulatory authority for a specific plasma derivative. In addition,
to the regulatory criteria related to donor selection and screening of
donations, the quality specifications agreed upon with the fractionator may
encompass:
— compliance with GMP during production and control;
— residual level of blood cells (platelets, leukocytes) that should be below a
certain level that may vary depending upon the requirements of different
countries or fractionators;
— protein content possibly including a minimal mean level of Factor VIII
coagulation activity if this product is manufactured;
— guarantee of an appropriate ratio of plasma:anticoagulant solution (see
Table 6) and evidence of appropriate mixing with the anticoagulant
during the collection process (for instance, clots should be absent);
— acceptable maximum titre of ABO blood group antibodies (risks
of haemolytic reactions due to the presence of ABO antibodies, or
antibodies to other blood group systems, in intravenous IgG and low-
purity factor VIII preparations have been described (68)). The European
pharmacopoeia requires an ABO titre of less than 1:64 for the release of
plasma products for intravenous use.
— maximum haemoglobin content;
— absence of haemolysis;
— colour;
— absence of opalescence (due to lipids);
— citrate (anticoagulant) range content (usually between 15 and 25 mM);
and
— minimum titre of a specific antibody when the donation is used for the
production of hyperimmune IgG such as anti-Rho, anti-HBs, anti-tetanus
or anti-rabies.

6.9 Release of plasma for fractionation

Each blood establishment should be able to demonstrate that each unit of
plasma has been formally approved for release by an authorized person
preferably assisted by validated information technology (IT)-systems.
The specifications for release of plasma for fractionation should be
defined, validated, documented and approved by quality assurance and the
fractionator.
There should be a system of administrative and physical quarantine for
plasma units to ensure that they cannot be released until all mandatory
requirements have been satisfied. In the absence of a computerized system
for control of product status, the label of the plasma unit should identify the
product status and should clearly distinguish released from non-released

224
 (quarantined) plasma. Records should demonstrate that before a plasma
unit is released, all current declaration forms, relevant medical records and
test results have been verified by an authorized person.
Before final product release, if plasma has been prepared from a donor who
has donated on previous occasions, a comparison with previous records
should be made to ensure that current records accurately reflect the donor
history.
In the event that the final product is not released due to potential impact on
plasma quality or safety, all other implicated components from the same
donation should be identified. A check should be made to ensure that (if
relevant) other components from the same donation(s) and plasma units
or other components prepared from previous donations from the same
donor(s) are identified. There should be an immediate update of the donor
record(s) to ensure that the donor(s) cannot make a further donation, if
appropriate.

6.9.1 Plasma release using electronic information systems

Special documented evidence is needed if release of plasma is subject to use
of electronic information systems (EIS) to ensure that the system correctly
releases plasma units only if all requirements are met. The following points
should be checked:
• The EIS should be validated to be fully secure against the possibility of
plasma which does not fulfil all test or donor selection criteria, being
released.
• The manual entry of critical data, such as laboratory test results, should
require independent verification by a second authorized person.
• There should be a hierarchy of personnel permitted access to enter, amend,
read or print data. Methods of preventing unauthorized entry should be
in place, such as personal identity codes or passwords which are changed
regularly.
• The EIS should block the release of plasma or other blood components
considered not to be acceptable for release. There should also be a means
to block the release of any future donation from a donor.

6.10 Packaging of plasma

The packaging requirements should be specified by the fractionator. The
specification should include the following information:
— how the plasma containers are to be packed to prevent damage during
shipment;
— that plasma of different types should be kept discrete and packaged into
separate cartons; and

225
 — that each carton should have a unique identification number or a bar
code which should be clearly displayed on the carton and recorded in the
shipping documentation.

6.11 Transportation of plasma

Although it is possible to think of transport as an extension of storage,
some additional qualification is appropriate. This need arises because of the
additional requirements for risk management during transport. Plasma is at
increased risk when:
• Responsibilities for storage and transportation conditions change
(especially when handling is the responsibility of individuals with little
understanding of the consequences of temperature elevation, as will often
be the case with contract shippers).
• Plasma is moved from one freezer or container to another (especially
if this involves even temporary exposure to ambient temperatures,
as on the loading dock of a blood establishment or a fractionation
facility).
• The usual provisions for backup in the event of failure of the refrigeration
system are not available (as during sea-transportation of several weeks
duration).
The recommendations for cold chain maintenance, as mentioned for
plasma storage, should also apply during transportation of plasma. The
arrangements for temperature control and monitoring during shipping
should be clearly defined and documented. The requirements for number
and location of temperature logging devices during shipping should be
based on a documented assessment of risk throughout the process. The
temperature to be maintained during transportation should be defined by
the fractionator in accordance with relevant regulations.
The responsibilities of organizations and individuals during shipping
should be identified; in particular any requirements for documented hand-
over checks should be specified. The final responsibility for acceptance of
quality as compliant with specification lies with the quality department of
the fractionation facility.
Table 7 summarizes some recommendations on the handling of blood and
plasma to optimize the recovery of labile proteins such as factor VIII in
plasma. These recommendations should be examined keeping in mind that
the relationship between the content of factor VIII in the starting plasma
and its recovery in factor VIII concentrates is unclear (40, 69), possibly
in part due to the loss of factor VIII that takes place during industrial
cryoprecipitation (70) as well as during purification and virus reduction
procedures.

226
Table 7
Processing of plasma for fractionation to optimize factor VIII stability

Steps Recommendations

Whole blood storage before • Up to 18 to 20 h at 22 °C ± 1 °C

plasma separation • Not more than 8 h at 4 °C

Freezing • As soon as possible, within 24 hrs of blood

collection or apheresis procedurea

Freezing rate and temperature • As specified by plasma fractionator, following

relevant regulations pertaining to the countries
where plasma will be fractionated and products
will be marketed
• < –20 °C or colder

Storage temperature • –20 °C or colder, constant

Transportation temperature • –20 °C or colder, constant

a
Collection of plasma by apheresis makes it possible to freeze plasma immediately after the end of the collection
procedure by contrast to whole blood processing.

6.12 Recall system

In the case of known or suspected quality defects of a plasma unit that
has already been shipped, a person within the blood establishment should
be nominated to assess the need for product recall and to initiate and
coordinate the necessary actions. An effective recall procedure should
be in place, including a description of the responsibilities and actions to
be taken. Actions should be taken within predefined periods of time and
should include tracing all relevant components of the donation and, where
applicable, should include look-back procedures.

7. Quality assurance system and

Good Manufacturing Practices
Human plasma for fractionation is the single most critical raw material in
the manufacture of plasma derivatives. Fractionators should only use plasma
for fractionation from blood establishments that are subject to inspection
and approved by a national regulatory authority. When the mandatory safety
testing is outsourced, the laboratories need to be inspected and approved.
The safety and quality of plasma for fractionation should be assured by
implementation of standards at the blood establishment where plasma is
prepared. These standards should be assured by implementation, at the
blood establishment, of an effective QA system based on the principles of
GMP.

227
 The QA system should ensure that all critical processes such as the purchase
of raw materials, starting materials, selection of donors, collection of blood
and plasma, production of plasma, storage, laboratory testing, dispatch and
associated quality control measures, are specified in appropriate instructions
and are performed in accordance with the principles of GMP and comply
with the appropriate regulations. The management should review the system
regularly to verify the effectiveness and introduce corrective measures if
deemed necessary.
Because the quality standards implemented at the blood establishment have
such a profound impact on the quality of plasma, it is a requirement that
their implementation be agreed between the blood establishment and the
fractionator, under the terms of the contract for plasma supply (Appendix 5).
Medicines regulatory authorities will verify that such a contract is in place
and that it complies with the regulations in force.
A blood establishment should establish and maintain an active and
operational quality assurance system covering all activities, and taking into
account the principles of GMP. The following items are of special relevance
as part of a QA system for the production of plasma for fractionation (71).

7.1 Organization and personnel

There should be an organization chart showing the hierarchical structure
of the blood establishment and clear delineation of lines of responsibilities.
All personnel should have appropriate qualifications and experience to
enable them to perform their tasks and should be provided with initial and
continued training. Only persons authorized by defined procedures and
documented as such should be involved in the production and control of
plasma. The tasks and responsibilities should be clearly documented and
understood. All personnel should have clear, documented and up to date
job descriptions.
Training programmes appropriate to the specific tasks of staff members
should be in place, and should include at least:
— relevant principles of plasma production and plasma characteristics;
— quality assurance and GMP; and
— relevant knowledge of microbiology and hygiene.
Training should be documented and training records should be maintained.
The contents of training programmes should be periodically assessed.
If certain tasks, such as separation of blood or viral safety testing, are
performed externally, these should be subject to a specific written contract.
The contract should ensure that the contract acceptor meets good practice
requirements in all disciplines relevant to the contract giver’s activity.

228
7.2 Documentation system
Every activity that may affect the quality of the blood and/or blood
component should be documented and recorded. The documentation should
be designed to ensure that the work performed is standardized and that there
is traceability of all steps in the process. The documentation should allow all
steps and all data to be checked. All documentation should be traceable and
reliable. A document control procedure should be established for review,
revision history and archiving of documents. It should include a distribution
list. All changes to documents should be acted upon promptly and should
be reviewed, dated and signed by an authorized person. Documentation
procedures should be designed, developed, validated and personnel trained
in a consistent manner.

7.3 Premises and equipment

Premises should be located, constructed, adapted and maintained to suit
the operations to be carried out. Premises should be designed to permit
effective cleaning and maintenance to minimize risk of contamination. The
tasks in each area should take place in a logical sequence to minimize the
risk of errors.
All critical equipment should be designed, validated and maintained to suit its
intended purpose and should not present any hazard to donors or operators.
Maintenance, cleaning and calibration should be performed regularly and
recorded. Instructions for use, maintenance, service, cleaning and sanitation
should be available. There should be procedures for each type of equipment,
detailing the action to be taken when malfunctions or failures occur. New and
repaired equipment should meet qualification requirements when installed
and authorized before use. Qualification results should be documented.

7.4 Materials
Only reagents and materials from approved suppliers that meet the
documented requirements and specifications should be used. Where relevant,
materials, reagents and equipment should meet the requirements of other
local legislation for medical devices. Appropriate checks on goods received
should be performed to confirm that they meet specifications. Inventory
records should be kept for traceability. Critical materials should be released
under the responsibility of quality assurance function before use.

7.5 Validation programme

All processes and equipment involved in the production and control of
plasma for fractionation should be validated. Data should be available to
ensure that the final product will be able to meet specifications.

229
7.6 Quality monitoring data
Quality control of plasma should be carried out according to a defined
sampling plan taking into account different collection and production sites,
modes of transport, methods of preparation and equipment used. Acceptance
criteria should be based on a defined specification for each type of plasma
for fractionation. These data may include monitoring of factor VIII or any
other protein quality criteria determined by the plasma fractionator, and
monitoring of residual cell counts (platelets, leukocytes, erythrocytes) when
requested by the plasma fractionator. All quality control procedures should
be validated before use.
The viral safety testing should be performed in accordance with the
recommendations of the manufacturers of the reagents and test kits. The
work record should identify the test(s) employed to ensure that entries, such
as the calculation of results, are available for review. The results of quality
control testing should be subject to periodic review.
Test results that do not satisfy the specified acceptance criteria should
be clearly identified to ensure that plasma from that donation remains in
quarantine and that the relevant samples are kept for further testing. Where
possible the performance of the testing procedures should be regularly
assessed by participation in a formal system of proficiency testing.

7.7 Virology safety testing

7.7.1 Sampling
The following are practical points to consider in ensuring that sampling is
performed appropriately:
• Sampling machine:
— Automatic sampling: Test samples should be taken automatically
and the donation number should be read from the barcode. In case
of failure of the automatic system, an appropriate system for manual
entry of records of donations should exist, and should require double
entry with digit checks;
— Sampler validation: The sampling machine should be validated and a
validation report should be available; and
— Calibration: The sampling machine should be calibrated on schedule
and records available.
• Reconciliation: There should be a reconciliation of the samples received
at the virology laboratory versus expected.

7.7.2 Test equipment

The following are practical points to consider in ensuring that the equipment
used for the virology testing performs appropriately:

230
 • Sample addition. The process of addition of samples to the test plates should
be automatic and should include identification of the barcode of the plates.
• Test processing. Ideally, the test processing should be automated. If
addition of reagents is done manually, full documentation should be
available
• Equipment. Pipettes, incubators and other items of equipment should be
fully validated and routinely calibrated and appropriate records maintained.

7.7.3 Assay performance validation

The objective of validation of assay performance is to make sure that the
performance of the virology assays, as carried out by the entity responsible
for plasma collection, is satisfactory. Points to consider include:
• Each test run should include an independent control.
• Analysis of positive controls.
• Analysis of data on non-repeatable reactives (see 7.7.5 below).
• Evidence of satisfactory participation in external proficiency schemes.

7.7.4 Test interpretation and downloading

The data from virology safety testing should be examined by the supervisor
before being officially accepted. Accepted data should be downloaded
directly to the mainframe computer, or there should be a secure system
for manual download which ensures positive release of the samples. No
transcription of results should be done as mistakes may be introduced.

7.7.5 Follow-up of reactives

The following should be given special attention:
• Identification of initial reactives. They should be identified using a secure
system.
• Repeat reactives. An acceptable system should be in place to confirm
repeat reactives, including sampling, labelling, testing, and entry of
results.
• Editing of repeat reactive. A computer algorithm should edit reactive
status to repeat reactive, or the editing should be performed by two staff
members.
• Deferral system. An appropriate deferral system should exist for repeat
reactives.
• Re-entry of deferred donors. Appropriate documentation should be in
place.

7.8 Electronic information system

Importance should be given to the introduction of an EIS for blood
establishments involved in the preparation of plasma for fractionation and

231
 when possible linked to other establishments to facilitate and speed tracing
of individual plasma donations. This will allow timely identification of the
location of donations in the chain of production of plasma products.
All software, hardware and backup procedures should be validated before
use and checked at least once a year to ensure reliability. The system should
prevent the use of duplicate donation numbers, or else the system should be
able to deal with duplication without data corruption.
Hardware and software should be protected against unauthorized use or
changes (e.g. by password protection of key functions). There should be
procedures for each type of software and hardware, detailing the action to
be taken when malfunctions or failures occur.
A backup procedure should be in place to prevent loss of records during
expected and unexpected downtime or function failures. Changes in
computerized systems should be validated, applicable documentation
revised and personnel trained, before the change is introduced into routine
use. The EIS should be maintained in a validated state.

7.9 Storage and transport

Storage and distribution routines should take place in a safe and controlled
way to assure product quality throughout storage and transport and to
exclude identification errors of plasma units. Intermediate storage and
transport should be carried out under defined conditions to ensure that set
requirements are met.

7.10 Change control system

A formal change control system should be in place for planning, evaluating
and documenting all changes that may affect the quality, traceability,
availability or effect of components or safety of components, donors or
patients. The potential impact of the proposed change should be evaluated.
The need for additional testing and validation should be determined.

7.11 Quality assurance auditing

In order to monitor the implementation and compliance with the blood
establishment quality management system, regular internal audits are
needed. These should be conducted independently by trained and competent
persons from within the organization, according to approved protocols.
Inter-institutional audits should be actively promoted.
All audit results should be documented and reported to management.
Appropriate corrective actions should be taken. Preventive and corrective
actions should be documented and assessed for effectiveness after

232
 implementation. In general the blood establishment should have procedures
for systematic improvement. Input for this process can come from complaints,
errors, inspections, audits and suggestions.

7.12 Defect reporting system

There should be systems in place to ensure that complaints, all types
of quality defects (e.g. in blood bags or test kits), and adverse events or
reactions are documented, carefully investigated for causative factors and,
where necessary, followed by the implementation of corrective actions to
prevent recurrence. This also applies to “near miss events”. The corrective
and preventive action system should ensure that nonconformity of the
product or quality problems are corrected, that recurrence of the problem is
prevented, and that the plasma fractionator is notified according to the agreed
procedure. The blood establishment should have methods and procedures
in place to channel product or quality problems into the corrective and
preventive action system.

7.13 Quality agreement between blood establishment

and fractionator
The important elements of a blood establishment quality system with critical
implications for plasma quality, will normally be addressed in a quality
agreement — an addendum to the contract for plasma supply. The quality
agreement should address at least the following areas of concern:
• agreement on specific donor selection criteria (with approval of the
national regulatory authority);
• schedule of requirements for exclusion or acceptance of donors, including
the arrangements for establishing donor identity and the provision for
possibility of self-exclusion;
• arrangements for monitoring and reporting the epidemiology of the donor
population;
• location of blood establishments (and of any facility to which a quality-
critical function, for example donation testing, has been outsourced);
• frequency of donation and the system for ensuring that donations are not
taken more frequently than allowed;
• requirements for donor screening and for donation testing, including any
provision for the preparation and testing of mini-pools;
• procedure for validation and approval of relevant test reagents and kits;
• record-keeping, including the arrangements for traceability of donors and
donations;
• specifications of plasma to be supplied, including any arrangements
for verifying compliance with specifications and documentation of
compliance;

233
 • specifications of containers to be used for blood/plasma collection and
supply;
• detailed requirements for labelling of individual plasma units (the
adhesive used for the labels should not compromise the quality of the
plasma products);
• arrangements for freezing, storage and shipment of plasma;
• notifiable events, including the arrangements for post-donation
notification;
• procedure for review and approval of any proposal for procedural
change;
• procedure and agreed frequency for audit of blood establishments by the
fractionator; and
• arrangements for notifying the fractionator of a proposed regulatory
inspection, its periodicity, and of the outcome of such an inspection.

7.14 Blood/plasma establishment audit and inspection

It is a requirement of GMP that the regulatory authorities and the plasma
fractionator should establish the basis of confidence in the quality of
critical raw materials. In the case of plasma, this is achieved by four basic
provisions:
• maintenance of a list of blood establishments approved (by the fractionator
and the regulatory authorities) for supply of plasma;
• agreement in a contract, or in the technical agreement to a contract of
supply, of the quality arrangements made at each blood establishment
approved for supply of plasma;
• regular audit of blood establishments to confirm satisfactory
implementation of the quality arrangements (these audits should be
reported in writing to the blood/plasma establishment and any remedial
actions confirmed); and
• monitoring of the quality of plasma supplied, with trending of quality-
critical parameters.
There will normally be a requirement for independent inspection and
approval of each blood establishment by the relevant regulatory authority
(see below). Such inspections should be provided for in any contract between
the plasma supplier and the fractionator, and will normally be undertaken
by the responsible authority in the country where plasma preparation is
performed. Written reports of such inspections should be made available
to the blood establishment and a remediation plan agreed upon. Reports of
regulatory inspections and associated remediation plans should be made
available to the fractionator under the terms of the contract for plasma
supply.

234
8. Regulatory control of plasma for fractionation
8.1 Role of national regulatory authority
According to the WHO Guidelines for national regulatory authorities
on quality assurance of biological products (72, 73), national regulatory
authorities have the duty to ensure that available biological products, whether
imported or manufactured locally, are of good quality, safe and efficacious,
and should thus ensure that manufacturers adhere to approved standards of
quality assurance and GMP. The responsibilities of the national regulatory
authority should also include the enforcement and implementation of
effective national regulations, and the setting of appropriate standards and
control measures. The evaluation and control of the quality, safety and
consistency of production of blood products involve the evaluation of the
starting material, production processes and the test methods to characterize
batches of the product. This requires specialist expertise by the national
regulatory authority.

8.2 Establishment licence and inspections

In many countries, national regulatory authorities have implemented a
control system based on licensing the establishments, inspecting them
regularly, and enforcing the implementation of the legal requirements and
applicable standards.
According to international GMP standards for the manufacturing of blood
products, the following two main principles are important for the control of
plasma as starting material:
• Quality assurance should cover all stages leading to the finished product,
from collection (including donor selection) to storage, transport,
processing, quality control and delivery of the finished product.
• Blood or plasma used as a source material for the manufacture of
medicinal products should be collected by establishments and be tested
in laboratories which are subject to inspection and approved by a national
regulatory authority.
These two points in the GMP requirements summarize an important basic
principle which is relevant for the manufacture of plasma derivatives
and the control of plasma as starting material. Most national regulations
therefore require that the establishments involved in the collection and
storage of plasma as a source material (e.g. plasmapheresis centres and
blood establishments) need to have an establishment licence and need to
be inspected by the competent national regulatory authority. To obtain the
licence the establishments have to fulfil a defined set of requirements to
guarantee that the plasma collected is safe and of good quality. Since each unit
collected represents a single batch, a marketing authorization for the plasma

235
 as a “product“ is not required in all countries. Under the latter condition, a
“system control”, instead of a “product control”, may be more appropriate.
In addition to the establishment licensing system some countries have also
introduced a product-specific approval system for blood components.

8.3 Impact of good manufacturing practices

The approach of implementing the principles of GMP in the production
of medicinal products is not new, and it is widely acknowledged that it is
essential in assuring the quality and safety of medicinal products. For blood
products, GMP becomes even more important and more complex due to the
biological nature of the products. Therefore, taking into account the principles
of GMP and the existence of an appropriate system of quality assurance to
address and implement these requirements in the manufacturing steps of
blood products should be a pivotal element of the preparation of plasma for
fractionation. As outlined in the previous sections, implementation of GMP
in the manufacture of blood products is essential, and quality assurance
and GMP should cover all stages, including the collection of plasma as
starting material. The implementation and enforcement of GMP in blood
establishments therefore has the following impact:
— introduces the application of quality assurance principles in all steps
involved in the collection, preparation and testing of blood components;
— supports systematic application of donor selection criteria for each
donation;
— reduces errors and technical problems in collection, preparation, testing,
and distribution;
— contributes to the release of products which comply with safety and
quality requirements;
— ensures adequate documentation and full traceability for each donation
and product;
— enables continuous improvement in collection, preparation and testing
of starting material;
— supports regional cooperation networks that may result in the formation
of centres of competence by centralizing activities in order to reach
compliance at the required level (cost-benefit for implementing quality
assurance measures);
— provides suitable tools for the national regulatory authority to assess the
compliance of a plasma collection centre.
An establishment licensing system for blood establishments by the
competent national regulatory authorities should therefore exist. The main
requirements for obtaining an establishment licence may include:
• Application of quality assurance system and GMP to all steps from
donation, to preparation, storage, testing and distribution of plasma.

236
 • Personnel directly involved in the collection, testing, processing, storage
and distribution of plasma need to be appropriately qualified and provided
with timely and relevant training.
• Adequate premises and equipment should be available.
• An adequate system to ensure traceability of plasma should be established;
traceability should be enforced through accurate donor, donation, product
and laboratory sample identification procedures, through record maintenance
and use of an appropriate labelling system.
• Requirements for selection of donors, including exclusion criteria for
donors with risk behaviours; provision of information to donors on risk
situations and the donation in general; and the use of a questionnaire to
obtain information on donor’s health.
• Requirements for testing of each donation.
• Requirements regarding traceability and documentation.
• Post-donation information system.

8.4 Inspections
In conducting regular inspections as part of the licensing procedure,
enforcement of the implementation of GMP is required aiming to ensure
the compliance of the blood establishments with the existing provisions. It
is the responsibility of the inspector from the national regulatory authority
to ensure that the manufacturers and the blood and plasma establishments
adhere to the approved standards of GMP and quality assurance, including
at sites where plasma is collected as starting material.
The inspections should be carried out by officials representing the competent
national regulatory authority. These officials should be specialized inspectors,
trained in GMP inspections, and they should be familiar with blood bank
technologies and the special features of quality assurance in the collection of
plasma. Inspections may follow common inspection procedures, including:
• an opening meeting;
• a blood establishment tour;
• inspection of main areas and activities;
— donor acceptance and identification
— donor suitability
— collection process
— processing and sampling
— plasma freezing
— testing and availability of test results
— release of plasma units
— storage, transportation and shipment
— quality assurance (including self inspection and change control)
— documentation (standard operating procedures, records, donor record
files and log books)

237
 — personnel and organization
— qualification and process validations
— error and corrective action system
— look-back information, recalls and complaints
— product quality controls
• a final meeting summarizing the inspection outcome.
A thorough inspection includes the observation of staff during performance
operation and comparison with defined written procedures. In a “system
control”, the inspection can be considered not only as a GMP inspection,
but also as an indirect product quality assessment by checking product-
specific validation and quality control data.
A written report should summarize the main aspects of the inspection
including its scope, a description of the company, the deficiencies listed,
specified and classified (e.g. as critical, major or minor), and a conclusion.
The written report will be sent to the company. The companies are requested
to notify the national regulatory authority about the specific steps which are
taken or planned to correct the failures and to prevent their recurrence. If
necessary, follow-up inspections should be performed e.g. to check that
specific corrective actions have been implementd.
The national regulatory authority should have the authority to withdraw
an establishment licence in a case where inspection results showed critical
non-compliance with the requirements or product specifications.
Information on the collection and control of the starting material, human
blood or plasma, and on the procedures conducted during the preparation
of the final blood derived medicinal product have to be documented as part
of the dossier in the marketing authorization.
In summary, the implementation of licensing and inspection systems for
blood establishments has become an important tool through which the
national regulatory authorities confirm the assurance of quality of plasma as
starting material for fractionation. The use of international standards not only
further promotes harmonization, but also facilitates regional collaboration
and information exchange between the national regulatory authorities.

Authors
The drafts of these guidelines were prepared by:

Dr T. Burnouf, Human Plasma Product Services, Lille, France; Dr A. Padilla, World

Health Organization, Geneva, Switzerland; Dr C. Schärer, Swissmedic, Swiss
Agency for Therapeutic Products, Bern, Switzerland; Dr T. Snape, Consultant,
Pickering, North Yorkshire, UK; Dr P. Strengers, International Society of Blood
Transfusion, Amsterdam, the Netherlands; Professor S. Urbaniak, Regional

238
Transfusion Centre, Aberdeen, UK; Professor W.G. van Aken, Professor of Medicine,
Amsterdam, the Netherlands.

The Drafts prepared were circulated for Consultation to Representatives of National

Regulatory Authorities, National Blood Programs and to the respective WHO
Regional Advisors in all the WHO Regional Offices. Plasma fractionators were
consulted through their respective plasma fractionation associations or through the
regulatory agencies in their countries. Both the Plasma Protein Therapeutic
Association and the International Plasma Fractionation Association presented
consolidated comments of their Members.

Acknowledgements are due to the following experts for their comments, advice
and information given during the preparation and consultation process of these
Guidelines:

Lic. M. P. Alvarez, Departamento Biológicos, CECMED, Havana, Cuba; Dr. R. S.

Ajmani, Intas Pharmaceuticals Ltd, Chinubhai Centre, Ahmedabad, India; Dr D.
Armstrong, Natal Bioproducts, South Africa; Dr. T. Barrowcliffe, National Institute for
Biological Standards, Potters Bar, Herts, UK; Dr C. Bianco, America's Blood
Centers, Washington DC, USA; Dr R. Büchel, Plasma Protein Therapeutics
Association (PPTA) Source, Brussels, Belgium; Dr E. A. Burgstaler, Transfusion
Medicine, Mayo Clinic, Rochester, Minnesota, USA; Mr A. Cadiz, Empresa
Productora de Sueros y Hemoderivados, La Habana, Cuba; Dr F. Cardoso de Melo,
Agencia Nacional de Vigilancia Sanitaria, Ministerio da Saude, Brasilia, Brazil;
Dr B. Cuthbertson, Scottish National Blood Transfusion Service, Edinburgh, UK;
Dr A. M. Cheraghali, Iran Blood Transfusion Organization, Tehran, Iran; Dr N.
Choudhury, Prathama Blood Center, Vasna, Ahmedabad, India; Dr J. R. Cruz,
Regional Advisor Laboratory and Blood Services, AMRO/PAHO, Washington, USA;
Dr. F Décary, Héma-Québec, Canada; Dr N. Dhingra, World Health Organization,
Geneva, Switzerland; Dr R. Dodd, American Red Cross, USA; Dr J. Epstein, Office
of Blood Research and Review, FDA Center for Biologics Evaluation and Research,
Bethesda, Maryland, USA; Mr T. Evers, International Plasma Fractionation
Association (IPFA), Amsterdam, the Netherlands; Ms M. Farag, Egyptian Regulatory
Authority, Cairo, Egypt; Professor A. Farrugia, Office of Devices, Blood and Tissues,
Therapeutic Goods Administration, Woden, Australia; Dr B. Flan, Laboratoire
Français du Fractionnement et des Biotechnologies, les Ulis, France; Dr J. C.
Goldsmith, Office of Blood Research and Review, FDA Center for Biologics
Evaluation and Research, Bethesda, Maryland, USA; Ms K. Gregory, AABB,
Bethesda, MD, USA; Ms M. Gustafson, Plasma Protein Therapeutics Association,
Washington, USA; Mrs T. Jivapaisarnpong, Department of Medical Sciences,
Ministry of Public Health, Nonthaburi, Thailand; Dr H. Klein, National Institutes of
Health, Clinical Research Center, Transfusion Medicine, Bethesda, MD, USA; Dr. J.
Kurz, Federal Ministry of Health and Women, Medicines & Medical Devices
Inspectorate, Vienna, Austria; Professor J. Löwer, Paul Ehrlich Institute, Langen,
Germany; Mrs B. Mac Dowell Soares, Agencia Nacional de Vigilancia Sanitaria,
Brasilia, Brazil; Dr E. Al Mansoori, Drug Control Department, Ministry of Health,
United Arab Emirates; Dr M. Maschio, Plan Nacional de Sangre, Buenos Aires,
Argentina; Dr A. Miller, Blood National Program, Montevideo, Uruguay; Dr S. Park,
Korea Food and Drug Administration, Seoul, South Korea; Professor I. Peake,

239
 International Society on Thrombosis and Haemostasis, University of Sheffield,
Sheffield, UK; Dr F. Reigel, Swissmedic, Swiss Agency for Therapeutic Products,
Bern, Switzerland; Dr A. Robinson, NHS Blood and Transplant, National Health
Service, UK; Mr D. Sato, Ministry of Health and Welfare, Japan; Professor E.
Seifried, German Red Cross, Institute of Transfusion Medicine and
Immunohaematology, Frankfurt/Main, Germany; Professor R. Seitz, Paul Ehrlich
Institute, Langen, Germany; Dr G. Silvester; European Medicines Evaluation
Agency, London, UK; Dr L. S. Slamet, National Agency of Drug and Food Control,
Indonesia; Dr T Simon, Tricore, USA; Professor J.-H. Trouvin, Afssaps, Paris, France;
Dr F. Vericat, Grifols, Barcelona, Spain; Dr E. Voets, Biological Standardization,
Scientific Institute of Public Health, Federal Public Service Health, Brussels,
Belgium; Professor G. N. Vyas, University of California, San Francisco, California,
USA; Dr M. Weinstein, Office of Blood Research and Review, FDA Center for
Biologics Evaluation and Research, Rockville, Maryland, USA; Mrs M. Wortley,
Haemonetics, Braintree USA; Professor H. Yin, Biological Products, State of Food
and Drug Administration, Beijing, People’s Republic of China; Dr Mei-Ying Yu,
Office of Blood Research and Review, FDA Center for Biologics Evaluation and
Research, Rockville, Maryland, USA.

Special thanks are also due to Dr T. Burnouf for the compilation and professional
follow up of the significant number of contributions received during the Consultation
process. Dr T. Burnouf and Dr A. Padilla, WHO Project Leader prepared the final
manuscript of these Guidelines.

References
1. Resolution WHA 58.13. Blood safety: proposal to establish World Blood
Donor Day. In: Fifty-eighth World Health Assembly. Geneva, World Health
Organization, 2005.
2. Requirements for the collection, processing and quality control of blood,
blood components and plasma derivatives. In: WHO Expert Committee
on Biological Standardization, Forty-third Report. Geneva, World Health
Organization, 1994; Annex 2 (WHO Technical Report Series, No. 840).
3. WHO Guidelines on viral inactivation and removal procedures intended to
assure the viral safety of human blood plasma products. In: WHO Expert
Committee on Biological Standardization. Fifty-second Report. Geneva,
World Health Organization, 2004, Annex 4 (WHO Technical Report Series,
No. 924).
4. Piet MP et al. The use of trin-(butyl)phosphate detergent mixtures to
inactivate hepatitis viruses and human immunodeficiency virus in plasma
and plasma’s subsequent fractionation. Transfusion, 1990, 30:591–598.
5. Ala F, Burnouf T, El-Nageh M. Plasma fractionation programmes for
developing economies. Technical aspects and organizational requirements.
Cairo, WHO Regional Publications, 1999 (Eastern Mediterranean Series).
6. Prowse C. Plasma and Recombinant Blood Products in Medical Therapy
— Appendix 1. Chichester, John Wiley & Sons, 1992.
7. Burnouf T, Radosevich M. Reducing the risk of infection from plasma
products: specific preventative strategies. Blood Reviews, 2000, 14:94–110.

240
 8. Kreil TR. West Nile virus: recent experience with the model virus approach.
Developments in Biologicals, 2004, 118:101–105.
9. Remington K et al. Inactivation of West Nile virus, vaccinia virus and viral
surrogates for relevant and emergent viral pathogens in plasma-derived
products. Vox Sanguinis, 2004, 87:10–18.
10. Schmidt I et al. Parvovirus B19 DNA in plasma pools and plasma
derivatives. Vox Sanguinisuinis, 2001, 81:228.
11. Blumel J et al. Parvovirus B19 transmission by heat-treated clotting factor
concentrates. Transfusion, 2002, 42:1473–1481.
12. Committee for Medicinal Products for Human Use. Guideline on the scientific
data requirements for a plasma master file (PMF). London, European Medicine
Evaluation Agency, 2004 (EMEA/CPMP/BWP/3794/03) (http://www.emea.eu.int).
13. Anonymous. Guide to the preparation, use and quality assurance of blood
components. 13th ed. Strasbourg, Council of Europe Publishing, 2007.
14. Sarkodie F et al. Screening for viral markers in volunteer and replacement
blood donors in West Africa. Vox Sanguinis, 2001, 80:142–147.
15. Pereira A, Sanz C, Tassies D, Ramirez B. Do patient-related blood donors
represent a threat to the safety of the blood supply? Haematologica, 2002,
87:427–433.
16. Roth WK et al. NAT for HBV and anti-HBc testing increase blood safety.
Transfusion, 2002, 42:869–875.
17. Tabor E. The epidemiology of virus transmission by plasma derivatives:
clinical studies verifying the lack of transmission of hepatitis B and C viruses
and HIV type 1. Transfusion, 1999, 39:1160–1168.
18. Wang B et al. Prevalence of transfusion-transmissible viral infections in first-
time US blood donors by donation site. Transfusion, 2003, 43:705–712.
19. Dodd RY, Notari EP, Stramer SL. Current prevalence and incidence of
infectious disease markers and estimated window-period risk in the American
Red Cross blood donor population. Transfusion, 2002, 42:975–979.
20. Watanabe KK, Williams AE, Schreiber GB, Ownby HE. Infectious disease
markers in young blood donors. Retrovirus Epidemiology Donor Study.
Transfusion, 2000, 40:954–960.
21. Muller-Breitkreutz K, Evers T, Perry R. Viral marker rates among unpaid
blood donors in Europe decreased from 1990 to 1996. Euro Surveillance,
1998, 3:71–76.
22. Committee for Medicinal Products for Human Use. Guideline on
epidemiological data on blood transmissible infections. For inclusion
in the guideline on the scientific data requirements for a plasma master
file. London, European Medicine Agency, 2005 (EMEA/CPMP/BWP/3794/03:
EMEA/CPMP/BWP/125/04) (http://www.emea.eu.int).
23. Hellstern P et al. The impact of the intensity of serial automated
plasmapheresis and the speed of deep-freezing on the quality of plasma.
Transfusion, 2001, 41:1601–1605.
24. Runkel S, Haubelt H, Hitzler W, Hellstern P. The quality of plasma collected
by automated apheresis and of recovered plasma from leukodepleted whole
blood. Transfusion, 2005, 45:427–432.

241
 25. Burnouf T, Kappelsberger C, Frank, K, Burkhardt T. Protein composition
and activation markers in plasma collected by three apheresis procedures.
Transfusion, 2003, 43:1223–1230.
26. Anonymous. Monograph of human plasma for fractionation 01/2005:0853
corrected. European Pharmacopoeia, Strasbourg, 2005.
27. Pink J, Thomson A, Wylie B. Infectious disease markers in autologous and
directed donations. Transfusion Medicine, 1994, 4:135–138.
28. de Wit HJ, Scheer G, Muradin J, van der Does J. A. Influence of the primary
anticoagulant on the recovery of factor VIII in cryoprecipitate. Vox Sanguinis,
1986, 51:172–175.
29. Griffin B, Bell K, Prowse C. Studies on the procurement of blood
coagulation factor VIII. In vitro studies on blood components prepared in
half-strength citrate anticoagulant 18 hours after phlebotomy. Vox Sanguinis,
1988, 55:9–13.
30. Prowse C, Waterston YG, Dawes J, Farrugia A. Studies on the procurement
of blood coagulation factor VIII in vitro studies on blood components
prepared in half-strength citrate anticoagulant. Vox Sanguinis, 1987,
52:257–264.
31. Rock G, Tittley P, Fuller V. Effect of citrate anticoagulants on factor VIII levels
in plasma. Transfusion, 1988, 28:248–253.
32. Beeck H et al. The influence of citrate concentration on the quality of plasma
obtained by automated plasmapheresis: a prospective study. Transfusion,
1999, 39:1266–1270.
33. Burgstaler EA. Blood component collection by apheresis. Journal of Clinical
Apheresis, 2006, 21:142–151.
34. Burgstaler EA. In: McLeod BC, Price TH, Weinstein R, eds. Apheresis:
Principles and Practice. 2nd ed. AABB Press, 2003:95.
35. Burnouf T, Kappelsberger C, Frank K, Burkhardt T. Residual cell content
in plasma from 3 centrifugal apheresis procedures. Transfusion, 2003,
11:1522–1526.
36. Smith JK. Quality of plasma for fractionation–does it matter? Transfusion
Science, 1994, 15:343 –350.
37. O’Neill EM. Effect of 24-hour whole-blood storage on plasma clotting factors.
Transfusion, 1999, 39:488 –491.
38. Hurtado C et al. Quality analysis of blood components obtained by
automated buffy-coat layer removal with a top & bottom system (Optipress
(R)II). Haematologica, 2000, 85:390–395.
39. Pietersz RN et al. Storage of whole blood for up to 24 hours at ambient
temperature prior to component preparation. Vox Sanguinis, 1989, 56:145–150.
40. Hughes C et al. Effect of delayed blood processing on the yield of factor VIII
in cryoprecipitate and factor VIII concentrate. Transfusion, 1988, 28:566–570.
41. Carlebjork G, Blomback M, Akerblom O. Improvement of plasma quality
as raw material for factor VIII:C concentrates. Storage of whole blood and
plasma and interindividual plasma levels of fibrinopeptide A. Vox Sanguinis,
1983, 45:233–242.

242
42. Nilsson L, Hedner U, Nilsson IM, Robertson B. Shelf-life of bank blood and
stored plasma with special reference to coagulation factors. Transfusion,
1983, 23:377–381.
43. Cardigan R, Lawrie AS, Mackie IJ, Williamson LM. The quality of fresh-
frozen plasma produced from whole blood stored at 4 degrees C overnight.
Transfusion, 2005, 45:1342 –1348.
44. Hogman CF, Johansson A, Bergius B. A simple method for the
standardization of centrifugation procedures in blood component
preparation. Vox Sanguinis, 1982, 43:266–269.
45. Hogman CF, Eriksson L, Ring M. Automated blood component preparation
with the Opti system: three years’ experience. Beitr Infusionstherapie, 1992,
30:100–107.
46. Hogman CF, Eriksson L, Hedlund K, Wallvik J. The bottom and top
system: a new technique for blood component preparation and storage.
Vox Sanguinis, 1988, 55:211–217.
47. Kretschmer V et al. Improvement of blood component quality--automatic
separation of blood components in a new bag system. Infusionstherapie,
1988, 15:232–239.
48. van der Meer P et al. Automated separation of whole blood in top and
bottom bags into components using the Compomat G4. Vox Sanguinis, 1999,
76:90–99.
49. Pietersz RN, Dekker WJ, Reesink HW. Comparison of a conventional
quadruple-bag system with a ‘top-and-bottom’ system for blood processing.
Vox Sanguinis, 1990, 59:205–208.
50. Masse M. Universal leukoreduction of cellular and plasma components:
process control and performance of the leukoreduction process. Transfusion
clinique et biologique, 2001, 8:297–302.
51. Seghatchian J. Universal leucodepletion: an overview of some unresolved
issues and the highlights of lessons learned. Transfusion and Apheresis
Science, 2003, 29:105–117.
52. Gregori L et al. Effectiveness of leucoreduction for removal of infectivity
of transmissible spongiform encephalopathies from blood. Lancet, 2004,
364:529–531.
53. Chabanel A et al. Quality assessment of seven types of fresh-frozen plasma
leucoreduced by specific plasma filtration. Vox Sanguinis, 2003, 85:250.
54. Cardigan R. The effect of leucocyte depletion on the quality of fresh-frozen
plasma. Br J Haematol, 2001, 114:233–240.
55. Runkel S et al. The impact of two whole blood inline filters on markers of
coagulation, complement and cell activation. Vox Sanguinis, 2005, 88:17–21.
56. Smith JF, Ness PM, Moroff G, Luban NL. Retention of coagulation factors
in plasma frozen after extended holding at 1-6 degrees C. Vox Sanguinis,
2000, 78:28–30.
57. Swärd-Nilsson A-M, Persson P-O, Johnson U, Lethagen S. Factors
influencing Factor VIII activity in frozen plasma. Vox Sanguinis, 2006, 90:33–39.
58. Farrugia A. Plasma for fractionation: safety and quality issues. Haemophilia,
2004, 10: 334–340.

243
 59. Fekete M, Kovacs M, Tollas G. The circumstances of freezing in the freeze-
drying process of haemoderivatives. Annales immunologiae Hungaricas,
1975, 18:229–236.
60. Myllyla G. Factors determining quality of plasma. Vox Sanguinis, 1998,
74:507–511.
61. Farrugia A, Prowse C. Studies on the procurement of blood coagulation
factor VIII: effects of plasma freezing rate and storage conditions on
cryoprecipitate quality. Journal of Clinical Pathology, 1985, 38:433–437.
62. Farrugia A. Factor VIII/von Willebrand factor levels in plasma frozen
to -30 degrees C in air or halogenated hydrocarbons. Thrombosis Research,
1992, 68:97–102.
63. Akerblom O. Freezing technique and quality of fresh-frozen plasma.
Infusionstherapie und Transfusionsmedizin, 1992, 19:283–287.
64. Carlebjork G, Blomback M, Pihlstedt P. Freezing of plasma and recovery
of factor VIII. Transfusion, 1986, 26:159–162.
65. International Forum — What are the critical factors in the production
and quality control of frozen plasma intended for direct transfusion or for
fractionation to provide medically needed labile coagulation factors? Vox
Sanguinis, 1983, 44:246–259.
66. Rock GA, Tittley P. The effects of temperature variations on cryoprecipitate.
Transfusion, 1979, 19:86–89.
67. Kotitschke R et al. Stability of fresh frozen plasma: results of 36-
month storage at -20°C, -25°C, -30°C and -40°C. Infusionstherapie und
Transfusionsmedizin, 2000, 27:174–180.
68. Buchta C, Macher M, Hocker P. Potential approaches to prevent uncommon
hemolytic side effects of AB0 antibodies in plasma derivatives. Biologicals,
2005, 33:41–48.
69. Pepper MD, Learoyd PA, Rajah S.M. Plasma factor VIII, variables affecting
stability under standard blood bank conditions and correlation with recovery
in concentrates. Transfusion, 1978, 18:756–760.
70. Foster PR. Control of large-scale plasma thawing for recovery of
cryoprecipitate factor VIII. Vox Sanguinis, 1982, 42:180–189.
71. PIC/S. PIC/S GMP guide for blood establishments. PE005-2, Pharmaceutical
Inspection Convention 2004.
72. Guidelines for national authorities on quality assurance for biological
products. In: WHO Expert Committee on Biological Standardization,
Forty-second Report. Geneva, World Health Organization, 1992; Annex 2
(WHO Technical Report Series, No. 822).
73. Regulation and licensing of biological products in countries with newly
developing regulatory authorities. In: WHO Expert Committee on Biological
Standardization, Forty-fifth Report. Geneva, World Health Organization, 1995;
Annex 1 (WHO Technical Report Series, No. 858).

244
 Appendix 1
Plasma products and clinical applications1
Products Main indications

Albumin

Human serum albumin Volume replacement

Blood coagulation factors

Factor VIIIa Haemophilia A

Prothrombin complex Complex liver diseases; warfarin or coumarin

derivatives reversalc

Factor IX Haemophilia B

Factor VII Factor VII deficiency

Von Willebrand factor Von Willebrand factor deficiency (type 3 and severe
forms of type 2)

Factor XI Haemophilia C (factor XI deficiency)

Fibrinogen Fibrinogen deficiency

Factor XIII Factor XIII deficiency

Activated PCC Haemophilia with anti-factor VIII (or factor IX) inhibitors

Protease inhibitors

Antithrombin Antithrombin deficiency

Alpha 1 antitrypsin Congenital deficiency of alpha 1 antitrypsin with

clinically demonstrable panacinar emphysema

C1-inhibitor Hereditary angioedema

Anticoagulants

Protein C Protein C deficiency

Fibrin sealant (fibrin glue)d Topical haemostatic/healing/sealing agent (surgical

adjunct)
a
Some factor VIII concentrates containing von Willebrand factor are effective for the treatment of von Willebrand
disease
b
Prothrombin complex contains factor II, factor VII, factor IX, and factor X. The content of factor VII may vary
depending upon products.
c
May be used, in the absence of purified plasma products, for substitutive therapy in factor VII, factor X, or
protein C deficiency. Whenever available, purified factor IX should be used to treat haemophilia B.
d
Product obtained by mixing a concentrate rich in fibrinogen and a concentrate rich in thrombin.

1
Adapted from: Ala, F, Burnouf T, El-Nageh M. Plasma fractionation programmes for developing
economies. Technical aspects and organizational requirements. Cairo, WHO Regional Publications,
1999 (Eastern Mediterranean Series).

245
 Products Main indications

Intramuscular
immunoglobulins (IMIG)

Normal (polyvalent) Prevention of hepatitis A (also rubella, and other

specific infections)

Hepatitis B Prevention of hepatitis B

Tetanus Treatment or prevention of tetanus infection

Anti-Rho (D) Prevention of haemolytic disease of the newborn

Rabies Prevention of rabies infection

Varicella/zoster Prevention of chickenpox infection

Intravenous immunoglobulins
(IVIG)

Normal (polyvalent) Replacement therapy in immune deficiency states;

Cytomegalovirus (CMV) Prevention of CMV infection (e.g. after bone marrow

transplantation)

Hepatitis B Prevention of HBV infection (e.g. liver transplant)

Rho (D) Prevention of haemolytic disease of the newborn

246
 Appendix 2
Donor selection
1 Preamble
2 Information to donors
3 Compliance with donor selection criteria
3.1 Identification of donors
3.2 Confidentiality
3.3 Questionnaire and interview
3.4 Physical examination, acceptance and deferral criteria

1. Preamble
Recognizing the importance of the provision of safe blood, blood components
and plasma derivatives, the 58th World Health Assembly in 2005 (WHA
Resolution 58.13) (1) expressed its support for “the full implementation of
well-organized, nationally coordinated and sustainable blood programmes
with appropriate regulatory systems” and stressed the role of “voluntary,
non-remunerated blood donors from low-risk populations”. The provision
of blood, blood components and plasma derivatives from voluntary, non
remunerated donors should be the aim of all countries.

2. Information to donors
Candidate donors should receive an explanation, ideally both verbally and
in writing, or by any other appropriate means such as a self-administered
questionnaire, that answers to questions about their medical history and personal
behaviour are necessary to determine whether they are eligible to donate blood
or plasma. Written information can be in the form of a leaflet explaining the
risks of infection associated with blood and plasma products; impact of social
behaviour on risks of infection and risk factors for infection. This information
is generally given by a licensed physician, or by a person under the direct
supervision of a licensed physician, who should explain the exclusion criteria
for donating blood and plasma. A convenient communication system should
ensure that risk factors are well understood by the candidate donor.
Additionally, the donor should be asked to inform the blood centre if he
or she feels unwell after the donation or if he or she forgot to mention a
possible risk factor. This is of special importance for a donation used to
prepare plasma for fractionation as it is important to be able to remove
at-risk donations prior to the industrial pooling stage to avoid the potential
need to destroy the plasma pool or the intermediates or products derived
from it.

247
3. Compliance with donor selection criteria
3.1 Positive identification of donors
Upon presentation at the blood/plasma collection site, donors should be
asked to identify themselves by stating their name, address and date of
birth, and to supply proof of a permanent place of residence to establish a
reliable means of contact, including, for example, a telephone number where
they can be contacted after donation, if needed. Proof of identity (such as
identity card, passport or driving licence) should be provided. Identification
of donors should also take place immediately before venipuncture.

3.2 Confidentiality
The premises and setting of the blood/plasma collection centre (or the
mobile collection unit) should allow for adequate confidentiality during the
donor’s interview and the selection process so that the candidate donor will
not avoid answering questions on his or her personal or private behaviour,
which otherwise would compromise the safety of the plasma donation used
for the fractionation process.

3.3 Questionnaire and interview

The assessment of each donor should be carried out by a suitably qualified
person, trained in use of donor selection criteria and will involve an interview, a
questionnaire and further direct questions if necessary. In order to obtain relevant
and consistent information about the donor’s medical history and general health,
it is recommended that the donor can review, complete and sign a pre-printed
questionnaire (computer-assisted self-administered interviewing (CASI) is
being developed in some regions), adapted to the type of donor (for instance,
first-time donor versus regular donor). The questionnaire should be drafted in
such a way that donors may easily identify whether they are in good health.
Candidate donors who are at risk of carrying a disease transmissible by
blood/plasma derived products should be able to exclude themselves
voluntarily after reading and responding to the donor information and/or
the questionnaire. Such confidential self-exclusion should also be possible
after the donation (e.g. by telephone).
The candidate donor should be asked to sign an informed consent to give
blood/plasma in which he or she acknowledges an understanding of the
moral responsibility behind the donation of blood/plasma.

3.4 Physical examination, acceptance and deferral criteria

3.4.1 Physical examination
Prior to the first donation and before subsequent blood donations and in
case of plasmapheresis at regular intervals, a physical examination should

248
 be carried out by a licensed physician or a physician substitute following an
established procedure. Local national regulatory authorities, usually after
consultation with the blood establishment, should determine the health
criteria and the respective acceptable limits to be taken into consideration
during physical examination, such as measurement of weight, blood
pressure, pulse rate and temperature, or any other criteria considered to
affect the safety of plasma-derived products or the donor.

3.4.2 Records and traceability

An appropriate computerized or, if not available, manual system should exist
to keep records of the donors, of their medical history and health status, and
to ensure efficient traceability of their donations. Such information provides
historical perspective on the health status of the donors, including previous
temporary deferrals (should they exist), and contributes to reinforcing the
judgement as to whether the donation would present a risk to the quality and
safety of plasma for fractionation.

3.4.3 Selection and exclusion criteria

The following elements have been recognized as playing a role in selecting
the safest donors:
Establishment of exclusion criteria: Relevant acceptance, deferral and
exclusion criteria for the donation of blood/plasma used for fractionation
should be formulated by the national regulatory authority and be applicable
nationwide, as national requirements. Within the scope of their role to establish
and implement effective national regulations, local national regulatory
authorities should enforce such criteria. Based on the characteristics of the
production process used to manufacture plasma-derived products, the plasma
fractionator may suggest additional or alternative exclusion criteria. For
instance, in some countries, the plasma from first-time donors is not used.
Deferral: A defined list of permanent or temporary deferral criteria used for
candidate donors from which the plasma would be used for fractionation,
should be clearly stated, made public, and incorporated in the donor
educational material. The physician performing the physical examination
should be able to identify whether the donor has been previously deferred
and, if so, for what reason. Examples of the major permanent deferral
criteria found in international guidelines include:
• clinical or laboratory evidence of blood-borne infectious diseases, e.g.
infection with HIV, HBV or HCV;
• past or present intravenous drug use.
Other exclusion criteria, either permanent or temporary, may include:
• sexual relationship between men;
• men or women who are engaged in prostitution;

249
 • subjects with haemophilia or other clotting-factor defects, in particular if
treated with clotting factors;
• sexual partners of any of the above or of someone the donor suspects may
carry the above risk factors;
• jaundice within the 12 months prior to donation, as it may be a clinical
sign of hepatitis A, B or C;
• transfusion with blood, blood components, or plasma products in the
12 months previous to donation, as blood transfusion is a risk factor for
all blood-borne infections;
• tattooing, scarification, ear piercing, acupuncture in the 12 months prior
to donation. These practices may be a vehicle for the transmission of viral
diseases unless clear evidence is provided that the procedure was carried
out under sterile conditions;
• a particular policy may be required with regard to the exclusion criteria for a
risk factor relevant to the safety of cellular blood components although it does
not create safety issues for the preparation of plasma for fractionation and
plasma derived products. For instance, risk factors for HTLV infection (e.g.
due to travel in countries where the prevalence is high) may be an exclusion
criterion for the donation of blood components, but this virus cannot be
transmitted by plasma products. It is however not advisable to introduce two
screening and quality standards for products separated from a whole blood
unit (e.g. red cell concentrates and plasma for fractionation) as this may in
itself create a risk of mishandling and error at the blood collection centre.

3.4.4 Reinstatement
When temporary deferral criteria are applied, a specific procedure conducted
by trained personnel should be in place for reinstatement of donors. Some
exclusion criteria are temporary (e.g. as long as a risk factor has been
identified) and can be waived once additional checks on the donor have
been made, or the time period for exclusion has passed.

3.4.5 Procedures
Based on such criteria, a written procedure should be in place at the
blood/plasma collection centre to control donor acceptance and deferral
criteria. The procedure should comply with the requirements of the national
regulatory authority and fractionator. Abnormal conditions should be
referred to the physician who has the responsibility of making the final
decision on the donor suitability. If the physician has any doubt about the
donor’s suitability, donation should be deferred.

Reference
1. Resolution WHA58.13. Blood safety: proposal to establish World Blood Donor Day.
In: Fifty-eighth World Health Assembly. Geneva, World Health Organization, 2005.

250
 Appendix 3
Donor immunization and plasmapheresis
for the manufacture of specific immunoglobulins
There is a need for hyperimmune plasma for the manufacture of specific
immunoglobulins that are clinically valid for therapeutic and prophylactic uses.

Donors with acquired antibodies

Plasma may be collected by plasmapheresis from donors who have acquired
immunity through natural infection or through active immunization with
approved vaccines for their own protection. Donors with medically useful
plasma may be identified by screening whole blood donations or by testing
the plasma of convalescent patients or vaccinated individuals who have
produced high-titre antibodies with the desired specificity, for example,
patients recovering from varicella-zoster infection or donors who have
been immunized with rabies vaccine. Unnecessary primary immunizations
can be avoided by this approach. Donation of plasma following natural
infection should be deferred until the potential donor is asymptomatic, and
non-viraemic.

Donors who require immunization

To ensure a sufficient supply of life-saving immunoglobulins to treat patients,
deliberate immunization of healthy volunteers may be necessary in addition
to collection of plasma from convalescent patients and donors selected by
screening for high levels of specific antibodies. The immunization of donors
requires informed consent in writing and should take into consideration all
the requirements of this Annex.
Donors should be immunized with antigens only when sufficient supplies
of material of suitable quality cannot be obtained from other appropriate
donors, or from donations selected by screening. Donors should be fully
informed of the risk of any proposed immunization procedure, and pressure
should not be brought to bear on a donor to agree to immunization. Women
capable of child-bearing should not be immunized with erythrocytes or
other antigens that may produce antibodies harmful to the fetus. Donors
with known allergies should preferably not be recruited.
Every effort should be made to use the minimum dose of antigen and number
of injections. In any immunization programme, the following should be
taken into consideration as a minimum:
— the antibody assay;
— the minimum level of antibody required;

251
 — data showing that the dose, the intervals between injections and the total
dosage proposed for each antigen are appropriate; and
— the criteria for considering a prospective donor a non-responder for a
given antigen.
A donor could be hyperimmunized with more than one immunizing
preparation as long as the safety of the procedure of multiple immunizations
is demonstrated.
Potential donors should be:
• informed by a licensed physician of the procedures, risks and possible
sequelae and how to report any adverse effects, and encouraged to take
part in a free discussion (which, in some countries, takes place with small
groups of potential donors);
• informed that they are free to withdraw their consent at any time.
In addition, donors may also be:
• encouraged to seek advice from their family doctor, or from an independent
competent counsellor, before agreeing to immunization; and
• informed that any licensed physician of their choice will be sent all the
information about the proposed immunization procedure.
All vaccines used for immunizing donors should be approved by the national
regulatory authority. Special care should be taken to ensure the safety of the
donor when a vaccine is administered at doses or according to schedules
that differ from those recommended for routine prophylactic immunization.
Erythrocyte and other cellular antigens should be obtained from an
establishment approved by the national regulatory authority. Donors should
be observed for approximately 30 min following any immunization in order
to determine whether an adverse reaction takes place. Because reactions
often occur 2–3 h after immunization, donors should be advised of this
possibility and instructed to contact the facility’s physician if a reaction is
suspected in the first 12 h after immunization. Reactions may be local or
systemic. Local reactions, which may be immediate or delayed, take the
form of redness, swelling or pain at the injection site. Systemic reactions
may include fever, chills, malaise, arthralgia, anorexia, shortness of breath
and wheezing. An insurance system should be in place to compensate for
side-effects to the donor.

Immunization with human erythrocytes

Erythrocyte donors
A donor of erythrocytes for the purposes of immunization should meet all
the general health criteria for donors (see Appendix 2). Relevant measures
should be taken to limit the risk of infectious diseases; these may vary from

252
country to country taking into consideration the relevant risks. For instance,
in some countries, the donor should never have had a blood transfusion in
order to reduce risks of vCJD. Prior to the first donation, the donor should be
found to be negative for relevant markers, which may include the following:
syphilis, HBsAg, anti-HIV, antibody to hepatitis B core antigen (anti-HBc),
anti-HCV and antibodies to human T-cell lymphotropic viruses (anti-HTLV),
and the serum level of aminotransferases should be within normal limits as
established by the national control authority. Erythrocyte phenotyping should
be done for ABO as well as for C, D, E, c and e. It is advantageous to select
red cells expressing high amounts of RhD antigen, e.g. homozygous D or
Rho, for immunization. Phenotyping for other clinically relevant specificities
is also required , especially for Kell, Fya/Fyb, Jka/Jkb and S/s. The volume
of erythrocytes drawn from a donor should not exceed 450–500 ml of whole
blood in any 12 week period. Shorter intervals may induce iron deficiency
and, possibly, anaemia. Erythrocytes obtained for immunization purposes
should be frozen (at least for 6–12 months depending upon the sensitivity
and range of the tests performed, e.g. the use of NAT) before use and the
donor should be retested and shown to be negative for the above markers of
infection before the stored cells are released and used for immunization. Pre-
storage leukoreduction of donations is considered desirable, and NAT testing
for HBV, HCV and HIV would give an additional level of safety.

Collection and storage of erythrocytes

Erythrocytes should be collected under aseptic conditions into sterile
pyrogen-free containers in an appropriate proportion of an approved
anticoagulant. They may then be dispensed in aliquots under aseptic
conditions into single-dose sterile, pyrogen-free containers for storage. The
microbiological safety of the dispensing environment should be validated.
The method selected should have been shown to provide acceptable cell
recovery in vitro (80%) or in vivo (70%). Erythrocytes should be washed
after storage to remove the cryoprotective agent (e.g. glycerol). Adequate
sterility data to support the shelf-life for stored erythrocytes should be kept
on file. A test for bacterial and fungal contamination should be done on
all blood dispensed in aliquots in an open system. The test should also be
performed on at least one single-dose vial from each lot of whole blood that
has been stored unfrozen for more than seven days. The test should be done
on the eighth day after collection and again on the expiry date. Sterility tests
should be performed following an approved procedure.

Erythrocyte recipients
The following additional testing of erythrocyte recipients is necessary:
• The recipient should be phenotyped for ABO, Rh, Kell Fya/Fyb, Jka/Jkb
and S/s antigens before immunization. The red cell donor and the recipient

253
 should be matched as far as possible for major blood group antigens
other than RhD. Only ABO-compatible erythrocytes may be transfused.
Whereas mismatching within the Rh system for C and or E is acceptable,
mismatching in the Kell, Fy, Jk and S/s systems is unacceptable.
• Screening for unexpected antibodies by methods that demonstrate
coating and haemolvtic antibodies should use the antiglobulin method or
a procedure of equivalent sensitivity.
Prospective erythrocyte recipients in whom antibody screening tests
demonstrate the presence of erythrocyte antibodies (other than those
deliberately stimulated through immunization by the plasmapheresis
centre) should be asked whether they have ever been pregnant or had a
blood transfusion, a tissue graft or an injection of erythrocytes for any
reason. This history should form part of the permanent record and should
identify the cause of immunization as clearly as possible. Recipients should
be notified in writing of any specific antibodies they have developed after
injection of erythrocytes. The plasma centre should maintain records, which
should be reviewed during inspection. The immunized donor should carry
a card or medicalert bracelet specifying the antibodies. These measures
allow optimal care of immunized donors who may require an emergency
transfusion, (e.g. following a road traffic accident) at some future time,
and for whom knowledge of the antibody status, especially mixtures of
antibodies, is important.

Immunization schedules
Erythrocytes used for immunization purposes should not be administered
as part of any plasmapheresis procedure. Such immunization may be
performed on the same day as plasmapheresis, but only after it and as a
separate procedure.
To minimize the risk of infection to the donor, the immunization schedule
should involve as few doses of erythrocytes as possible. Wherever possible,
the same red cell donor should be used throughout the immunization
programme of an individual plasma donor.
For primary immunization two injections of erythrocytes, each of a volume
of about 2–5 ml and given 3 months apart, elicit antibody formation within
three months of the second injection. Different schedules may be used for
de novo immunization. It is advantageous to choose as donors of anti-D
(anti-Rho) volunteers who are already immunized, because useful levels of
anti-D are then usually attained within a few weeks of reimmunization with
2–5 ml of erythrocytes. About 70% of immunized volunteers eventually
produce antibody levels well above 100 IU/ml. The baseline antibody titre
of every recipient of erythrocytes should be established, and the antibody
response, including both type and titre, should be monitored monthly to

254
 establish the peak level of anti-D and duration of the response. The response
of each recipient is individual, and additional injections of erythrocytes
may be required at intervals of 2–9 months to maintain anti-D levels (1).
If injections of erythrocytes are discontinued, antibody levels usually fall
appreciably within 6-12 months. Erythrocytes to be used for immunization
purposes should be selected, for each recipient, by a licensed physician or a
suitably trained and qualified person.
Donors undergoing primary immunization who have not responded to a
total of up to 150 ml erythrocytes are likely to be ‘non-responders’ and
should be removed from the panel.

Plasmapheresis schedules
Donors should comply with the requirements for health screening and
maximum plasma donation allowed by their national authorities.

Risks to recipients
Recipients of erythrocytes for immunization purposes may be at risk of:
— viral hepatitis (B and C) and HIV infection;
— other infectious diseases;
— HLA immunization;
— the production of unwanted erythrocyte antibodies that may complicate
any future blood transfusion;
— a febrile haemolytic reaction if the antigen dose is too high;
— vCJD in countries where this is endemic.

Record-keeping
Records of erythrocyte donors and of the recipients of their erythrocytes
should be maintained and cross-referenced and stored at least for the
minimum time required for blood transfusion recipients by the national
authorities.

Reference
1. Cook I et al. Frozen red cells in Rhesus immunization. British Journal of
Haematology, 1980, 44:627.

255
 Appendix 4
Contract plasma fractionation programme
The fractionation of plasma requires specialized facilities, with provision
for large-scale protein separation, purification, virus inactivation and
formulation, as well as for aseptic finishing and freeze-drying. The
preparation of plasma-derived products should be governed by the same
regulatory considerations that are applied to medicines. Manufacturers
are required to obtain manufacturing licences which should cover the
method of preparation and product characteristics. To obtain a licence, it is
necessary to demonstrate adherence to GMP. Considerable technological,
pharmaceutical and scientific expertise is required to meet these demands.
Since key utilities (such as heating, ventilation and air-conditioning (HVAC),
refrigeration and water for injection) should be maintained operational
even when the facility is not fractionating plasma, the investment in and
running costs of fractionation are substantial. The economic viability of a
fractionation facility will be determined by:
• the cost of the plasma for fractionation (in particular cost-allocation of
the whole blood collection system on plasma versus labile components);
• the operating capacity of the facility; and
• plasma availability and product demand to allow the facility to operate
continuously at near to maximum capacity.
The break-even point for minimum annual plasma throughput for economic
viability may vary greatly according to a set of parameters, these including
plasma cost, product portfolio, adequacy of the various plasma products
versus the plasma needed to cover those needs, and product yield. Therefore
such projects require a careful feasibility study.
Countries which cannot justify building and operating a fractionation
facility, may opt to have plasma collected locally and shipped for processing
in an independent facility—so-called contract or toll fractionation. Plasma-
derived products are then returned to the originating country on payment of
a fee (toll). Such arrangements can work well, subject to specific provisions
being made and adhered to. These include:
• commercial and quality agreements defining the responsibilities of both
parties (the contract giver and the contract acceptor);
• clearly defined requirements for plasma quality (including the
arrangements for donor selection, testing and traceability);
• provision for audit of the plasma collection centre (by the fractionator)
and inspection by an appropriate regulatory body;
• formal approval of the contract plasma fractionation activities by the
regulatory authority of the fractionator;

256
 • a contractual commitment to supply agreed quantities of plasma. The
annual minimal volume is dependent upon the fractionator’s overall free
capacity and specific aspects of production such as plasma pool and
product batch size;
• agreement on the arrangements for storage and shipment of plasma, with
defined provisions for monitoring and control (typically transport by sea,
at –20 °C or below);
• agreement on the range of products to be manufactured; and
• agreement on specific aspects of plasma processing (including batch
size, possible requirements for segregation of processing, agreed use or
destruction of excess intermediates, expected yield and toll fees).
Plasma products made from local plasma need to receive a specific
registration, even if the same products made from foreign plasma are
already licensed in the country of origin.
The regulatory authorities of the country where the plasma is collected
may require inspection of the fractionation centre. Table 1 summarizes the
responsibilities and roles of each party.

Table 1
Responsibilities and roles of blood establishment, plasma fractionator,
and regulatory authorities

Task Blood establishment Plasma fractionator Regulatory authority

Epidemiology Collects and Reviews the data Reviews the data

surveillance of analyses the data
donor population based on results of
screening tests

Donor selection Develops and Verifies that criteria Sets the criteria and
and interview implements the set by national inspects the blood
criteria in selection regulatory authority establishment
and interview are met; may provide
of donors additional selection
criteria

Serological testing Performs validated Agrees on the Approves test kits

of donation tests (or the tests tests kits used and and inspects the
may be sub- audits the virology blood establishment
contracted) laboratory

Post-donation Informs plasma Takes appropriate Evaluates

follow-up and fractionator (and measures if haemovigilance/post-
haemovigilance when appropriate plasma pool or donation reports with
the regulatory product quality is regards to product
authority) when compromised quality and safety
relevant information
is obtained

257
Task Blood establishment Plasma fractionator Regulatory authority

Preparation Collects blood Sets the Approves and

of plasma plasma, prepares, specifications inspects the blood
freezes, and and audits establishment
stores the plasma,
according to good
manufacturing
practice (GMP)

Nucleic acid Prepares the NAT Provides the Approves the

testing (NAT) samples following standard operating procedure and
(mini-pool) fractionator’s procedure for inspects the
specifications NAT samples plasma fractionator
and performs
(or sub-contracts)
the validated testing

Fractionation Applies the Evaluates the data

methods including fractionation methods presented in the
viral reduction following GMPs and dossiers prepared
processes described by the fractionator,
in marketing and inspects
authorization fractionation facility

Preparation of Prepares the files Reviews and

plasma product evaluates
regulatory files

GMPa Implements GMP Audits the blood Inspects blood

establishment establishment and
enforces GMP

Granting Grants the marketing

of marketing authorization
authorization

Plasma product Does Evaluates

pharmacovigilance pharmacovigilance pharmacovigilance
studies and reports with regards
informs regulatory to product quality
authorities and blood and safety
establishment when
relevant side-effects
are found

a
See sections 7 and 8 of this annex.

258
 Appendix 5
Technical points to consider in establishing plasma
specifications criteria and obligations between
blood establishment and plasma fractionator
The purpose of the contract is to have a “legally binding” document between
the plasma supplier and the fractionator.
The following is an example of the quality control and documentation
required by a plasma fractionator to acquire plasma for fractionation from
a blood establishment. It is not meant to represent the only possible way
to define plasma specifications criteria and obligations between a blood
establishment and a plasma fractionator. Depending upon the prevalence of
blood-borne diseases in a country, additional safety requirements on donor
selection and testing should be considered.

General specifications
Donors
Reference should be made to local regulations pertaining to the selection,
eligibility, and exclusion criteria for donors of blood or plasma used for
the manufacture of blood components and plasma derivatives. Newly
introduced criteria may be spelled out (such as travel restrictions related
to vCJD).

Blood establishments
Reference should be made to the official legislation of blood establishments
in the country of origin and to relevant legislation relating to plasma
fractionation.

Donation process and plasma unit specifications

The contract should cover the following aspects of the donation process and
plasma unit specifications.
• Collection process of the blood/plasma units:
— containers, collection sets and anticoagulants with relevant
registration;
— duration of the whole blood collection (e.g. less than 15 minutes (1)
for recovered plasma);
— guarantee that blood will be mixed with the anticoagulant as soon as
the collection starts, by regular manual shaking or using a validated
automated method (1);

259
 — prior to freezing, plasma is clear (light opalescence may be allowed),
yellow to — green in colour, with no sign of haemolysis or presence
of red cells (2); and
— acceptable citrate concentration range.
• Infectious markers:
— test kits used should be of acceptable sensitivity and be agreed with
manufacturer;
— anti-HIV 1 and 2, anti-HCV and HBsAg should be absent, and there
should be no laboratory evidence of syphilis;
— when applicable: specific handling of anti-HBc positive donations
(e.g. accepted only if anti-HBs antibody titre > 0.050 IU/ml and
HBsAg negative); and
— HCV NAT and HIV tests must be negative (i.e. when a blood
establishment organization performs NAT for HCV and HIV for
blood components).
• Immunohaematological markers
— anti-A and anti-B titre < 1/64 using a validated assay;
— special requirements relative to the absence of irregular antibodies.
• Cellular content and haemoglobin
— statistical records of blood cell contamination showing that the relevant
specifications are met. Some countries/fractionators have set specific
limits on the residual leukocyte content of plasma for fractionation;
— statistical records of haemoglobin contamination showing that the
relevant specifications are met.
• Protein quality control
— protein content ≥ 50 g/l after mixture with the anticoagulant;
— when plasma is used for production of factor VIII concentrate;
minimum factor VIII content to be specified for a pool sample of a
defined number of donations
• Other criteria
— minimal acceptable volume of plasma per container;
— plasma freezing conditions: core temperature, time taken to freeze,
and absence of folding to avoid a thin plasma layer that would be
more susceptible to thawing during subsequent handling;
— maximum acceptable thickness of plasma containers;
— positioning of the donation identification label (number and bar
code);
— plasma storage temperature;
— plasma density (used to determine the volume of plasma shipped to/
received by fractionator);
— maximum time elapsed between donation and shipment to the
fractionator.

260
Standard plasma
Plasma types
Plasma categories vary depending upon fractionator and local regulations.
For instance, some fractionators may classify as plasma, either from whole
blood or from apheresis, based on the time interval between the collection
procedure and freezing.
Examples of plasma categories include:
• Category A: apheresis plasma frozen within 6 hours, with a factor VIII
content ≥ 0.7 IU/ml;
• Category B: Recovered plasma with a factor VIII content ≥ 0.7 IU/ml,
obtained from whole blood kept at 20–22 °C and frozen within 6 hours (in
the absence of devices to maintain blood temperature), or frozen within
20 hours (if devices to maintain blood temperature are used);
• Category C: Plasma frozen within 24 hours after collection, or plasma
initially categorized as A or B but containing ≤ 0.7 IU factor VIII/ml.
This plasma is used to produce immunoglobulins and albumin only.

Hyperimmune plasma
Quality criteria
Acceptable criteria include:
• protein content, factor VIII, haemoglobin: usually the same as for standard
plasma;
• a minimum potency level will be set for each antibody type. Where
possible, the required potency will be specified in IU per ml when assayed
using an agreed method which includes an agreed reference control
calibrated in IU/ml. Examples of limits are as follows:
— anti-tetanus: 10 IU/ml;
— anti-varicella/zoster: 10 IU/ml;
— anti-HBs: 25 IU/ml;
• Indication of the assay procedure, procurement of standards, test
laboratory and communication procedure of the data.

Documentation
Each blood establishment delivering plasma should have an approved
organizational chart, and changes should be communicated to the plasma
fractionation centre according to an agreed procedure.
Shipping documentation should include:
• dated shipping document signed by responsible person;

261
 • certificate of origin and control of the plasma, stating for each donation
the:
— collection date;
— carton number;
— results of virology and immunohaematology screening;
— test kits used and their batch number;
— signature of the director or an authorized person;
• password-protected electronic file of the plasma donations and samples
sent, stating for each donation collection date (this needs to be agreed
with the fractionator):
— carton number;
— results of virology and immunohaematology screening;
— test kits used and their batch number;
• upon request, additional information on viral screening tests and
confirmatory assays can be provided to the fractionator;
• epidemiology data should be made available as appropriate, e.g.
annually.

Shipment
Specifications relating to shipment include the following:

Plasma donations
• Broken plasma containers are not acceptable.
• When applicable, specifications of “pig tail” used for additional screening
tests by the fractionator (e.g. length of 10–20 cm, attached to the plasma
donation, and ideally, identified with the donation number).
• Specification of the plasma container identification (labels and barcode).
• Specification on potential additional samples sent with the shipment for
additional screening tests such as NAT or for the look-back procedure.
• Statement on minimal number of plasma containers per shipping box or
carton, and positioning.

Containers for shipment

Auditing programme
The contract should cover the following aspects of the auditing programme:
• obligation of the blood establishment to be subjected to auditing by the
fractionator;
• routine auditing performed by the fractionator should follow an internally
approved and regularly revised procedure with an established list of
questions and check-points;
• special auditing performed annually/biannually based on a programme
previously communicated to the director of the blood establishment;

262
• audit reports are communicated to the director of the blood
establishment;
• list of reference documents (such as internal acceptance criteria for the
preparation of plasma for fractionation).

Notification obligations
Notification obligations cover the following:
• obligation to notify the fractionator each time the safety of a previous
donation may be questionable;
• obligation to notify the fractionator when:
— a unit positive for viral markers such as HBsAg, HIV-1 and HIV-2
antibodies, HCV antibody or syphilis has been sent by mistake;
— a deviation is subsequently discovered in any of the screening tests
performed on the plasma units supplied. In this situation, the blood
establishment should attempt to retest the implicated units if suitable
library samples are available;
— a regular donor is found to be positive for a marker although the
previous donation was found to be negative;
— the blood establishment is informed that a donor, previously
contributing to plasma for fractionator, has developed an infectious
disease potentially transmissible by plasma;
— a donation is found to have transmitted an infectious disease, or there
is strong evidence implicating a donation in disease transmission;
— the blood establishment is informed that a donor previously
contributing to plasma for fractionation: (a) has developed CJD or
vCJD (in such a case the report with the pathological findings should
be provided if available); (b) has risk factors for vCJD; or (c) is
identified as exhibiting risk behaviour or other factors that affect the
safety of the plasma;
— the blood establishment is informed that a patient has developed post-
transfusion infection following transfusion of blood component(s)
obtained from a donor who has also donated one or more units of
plasma for fractionation.
Notifications should provide the list of all donations made within a 6-month
period prior to the last donation found to be negative. The period of time
depends on local regulations and the type of disease. The fractionator may
request additional data on previous donations when thought necessary.
A communication procedure must be in place indicating information that
must be provided. This should include:
• name of qualified person at the fractionator to be contacted;
• reasons and description of the problem (under confidentiality clauses);

263
 • the time period between information being known and communication to
the fractionator;
• if the problem is related to an infectious disease, a list of all plasma
for fractionation donations made in the defined period prior to the last
donation found negative;
• name of the blood establishment, director, donation number, carton
number as indicated on the electronic file sent with the shipment, date of
shipment, date of notification and signature of the responsible person or
his or her delegate.

References
1. Anonymous. Guide to the preparation, use and quality assurance of blood
components. 13th ed. Strasbourg, Council of Europe Publishing, 2007.
2. Anonymous. Monograph of human plasma for fractionation 01/2005:0853
corrected. European Pharmacopoeia, Strasbourg, 2005.

264
© World Health Organization
WHO Technical Report Series No 941, 2007

Annex 5
WHO biosafety risk assessment and guidelines
for the production and quality control
of human influenza pandemic vaccines

This document provides guidance to national regulatory authorities and

vaccine manufacturers on the safe production and quality control of human
influenza vaccines produced in response to a threatened pandemic. The
document details international biosafety expectations for both pilot-scale
and large-scale vaccine production and control and is thus relevant to both
development and production activities. It should be read in conjunction
with the WHO Laboratory Biosafety Manual (1) and replaces the earlier
WHO guidance Production of pilot lots of inactivated influenza vaccines
from reassortants derived from avian influenza viruses: Interim biosafety
risk assessment (2). Tests required to evaluate the safety of candidate
influenza vaccine reference viruses by WHO Reference Laboratories prior
to release to vaccine manufacturers are also specified in this document.
The following text is written in the form of guidelines rather than
recommendations. Guidelines allow greater flexibility than recommendations
with respect to expected future developments in the field. These guidelines
specify steps to minimize the risk of introducing influenza virus strains
with pandemic potential from a vaccine manufacturing facility into the
community. If a national regulatory authority so desires, these guidelines
may be adopted as definitive national requirements, or modifications may
be justified and made by a national regulatory authority. It is recommended
that modifications to the principles and technical specifications of these
guidelines be made only on condition that the modifications ensure that
the risks of introducing influenza virus to the community are no greater
than as outlined in the guidelines set out below.

Summary
Introduction
Glossary
1. Scope of the risk assessment
2. Hazard identification
2.1 Hazards associated with the type of pandemic vaccine viruses
2.1.1 Hazards associated with the recipient virus
in a reassortant strain
2.1.2 Hazards arising from the inserted gene product
in a reassortant vaccine strain

265
 2.1.3 Hazards arising from reassortant viruses
2.1.3.1 Direct hazards
2.1.3.2 Indirect hazards
2.1.4 Hazards arising from the use of wild type viruses
for pandemic strain vaccine production
2.2 Hazards arising from the type of production
2.2.1 Production in eggs
2.2.2 Production in cell cultures
2.3 Factors affecting pathogenicity for humans
2.3.1 HA receptor specificity
2.3.2 HA cleavability
2.3.3 Other factors affecting pathogenicity
2.4 Hazards arising from the vaccine
2.5 Prior large scale experience with reassortants
2.6 Testing of viruses being considered for vaccine production
2.6.1 In vivo tests to evaluate pathogenicity of H5
and H7 vaccine candidates
2.6.2 Genetic stability of H5 and H7 vaccine candidates
2.6.3 Evaluation of wild type non-pathogenic H5 or
H7 viruses or reassortants derived from them.
2.6.4 In vivo tests of non H5, non H7 vaccine candidates
or reassortants derived from them.
3. Risk assessment
3.1 Health protection
3.1.1 Likelihood of harm to human health
3.2 Environmental protection
3.2.1 Nature of the work
3.2.2 Environmental considerations
3.3 Assignment of containment level
3.4 Environmental control measures
3.4.1 Specifications for “BSL2 Enhanced (pandemic
influenza vaccine)”
3.4.1.1 Facility
3.4.1.2 Personal protection
3.4.1.3 Virus monitoring
3.4.2 Specifications for “BSL3 Enhanced (pandemic
influenza vaccine)”
3.4.2.1 Facility
3.4.2.2 Personal protection
3.5 Biosafety management and implementation within a vaccine
production facility
3.5.1 Management structures
3.5.2 Medical surveillance
3.5.3 Implementation
Authors
References

266
Summary
International biosafety expectations for both the pilot-scale and large-scale
production of human vaccines for a response to a pandemic influenza strain,
and the quality control of these vaccines, are described in detail in these
WHO Guidelines. Tests required to evaluate the safety of candidate influenza
vaccine reference viruses prior to release to vaccine manufacturers are also
specified in this document which is thus relevant to both development and
production activities, and also to vaccine and biosafety regulators. A detailed
risk assessment is presented that concludes that the likelihood of direct harm
to human health would be high if non-reassortant H5 or H7 viruses with
multiple basic amino acids at the haemagglutinin (HA) cleavage site and high
in vivo pathogenicity are used for vaccine production. Such viruses could
also pose a significant risk to animal health. Stringent vaccine biosafety
control measures, defined as Biosafety Level (BSL)3 enhanced (pandemic
influenza vaccine) are defined to manage the risk from vaccine production
and quality control using such viruses in the pre-pandemic period. For
all other vaccine strains, for example reassortants derived from H5 or H7
strains in which the multiple basic amino acid HA0 cleavage site has been
removed, the direct risk to human health is very remote. Nevertheless, there
is an indirect risk to human health due to a theoretical risk of secondary
reassortment with normal human influenza viruses, resulting in a virus
with avian-like coat proteins capable of replicating in humans. Although
very unlikely, the secondary reassortant could become adapted to human
infection and transmission which, if vaccine production was taking place in
the pre-pandemic period, would have serious public health consequences.
The biosafety control measures that are proposed, defined as BSL2
enhanced (pandemic influenza vaccine), take this and also potential risks to
animal health into account. Facility and personal protection specifications
are provided for both BSL2 enhanced and BSL3 enhanced bioafety levels
and guidance is provided on biosafety management and implementation
within a vaccine production facility. Tests to be performed on candidate
vaccine reference strains prior to release to vaccine developers depend on
the type of virus but include, at a minimum, in vivo tests on ferrets or other
susceptible mammals, and, where appropriate, chickens and egg embryos,
plaque assays and sequencing.

Glossary
The definitions given below apply to the terms used in these guidelines.
They may have different meanings in other contexts.
Aerosol
A dispersion of solid or liquid particles of microscopic size in a gaseous
medium.

267
 Air balance
The necessity to keep air supply and exhaust systems in balance by means
of measurements of static pressure, fan and motor performance, and air
volumes.
Airlock: Areas found at entrances or exits of rooms that prevent air in one
space from entering another space. These generally have two doors and
a separate exhaust ventilation system. In some cases a multiple-chamber
airlock consisting of two or more airlocks joined together is used for
additional control.

Biosafety committee
An institutional committee of individuals versed in the subject of containment
and handling of infectious materials.

Biosafety level 2 (or 3) (enhanced pandemic influenza)

A specification for the containment of pandemic influenza during vaccine
manufacture and quality control testing with specialized air handling systems,
waste effluent treatment, immunization of staff, specialized training, and
validation and documentation of physical and operational requirements.

Biosafety manual
A comprehensive document describing the physical and operational practices
of the laboratory facility with particular reference to infectious materials.

Biosafety officer
A staff member of an institution who has expertise in microbiology and
infectious materials, and has the responsibility for ensuring the physical
and operational practices of various biosafety levels are carried out in
accordance with the standard procedures of the institution.

Biological indicators
The use of organisms to test the efficacy of sterilization processes.

Biological safety cabinet

Primary and partial containment work enclosure used for manipulation of
materials that may cause infections or sensitization to workers. They are
equipped with high-efficiency particulate air (HEPA) filters and may or may
not be open-fronted.

Certification
Documentation that a system qualification, calibration, validation, or
revalidation has been performed appropriately and the results are acceptable.

Decontamination
A process by which an object or material is freed of contaminating agents.

268
Floor dams
Purpose-built elevations to enclose liquid spills.

Fumigation
The process whereby gaseous chemical is applied to an enclosed space for
the purpose of sterilizing the area.

Good manufacturing practices

That part of quality assurance which ensures that products are consistently
produced as controlled to the quality standards appropriate to their intended
use and as required by the marketing authorization.

HEPA filter
A filter capable of removing at least 99.97% of all particles with a mean
aerodynamic diameter of 0.3 micrometres.

Inactivation
To render an organism incapable of replication by application of heat, or
other means.

Seed lot
A culture of microorganism distributed from a single bulk container in a
single operation, in such a manner as to ensure uniformity and stability and
to prevent contamination.
Positive pressure laminar flow hood
An enclosure with unidirectional outflowing air, generally used for product
protection.

Primary containment
A system of containment, usually a biological safety cabinet or closed
container, which prevents the escape of a biological agent into the immediate
working environment.

Respirator
A respiratory protective device with an integral perimeter seal, valves
and specialized filtration, used to protect the wearer from toxic fumes or
particulates.
Risk analysis
A formalized documented process for analysing risks.

Secondary containment
A system of containment, usually involving specialized air-handling,
airlocks and secure operating procedures, which prevents the escape of a
biological agent into the external environment or into other working areas.

269
 Sterilization
Sterility is the absence of viable microorganisms. In general, an item is
assumed to be sterile if the validation of the sterilization process applied to
it indicates that only one item in one million items subjected to the process
will contain a viable microorganism.

Validation
The documented act of proving that any procedure, process, equipment,
material, activity, or system actually leads to the expected results.

Introduction
The earlier WHO guidance Production of pilot lots of inactivated influenza
vaccines from reassortants derived from avian influenza viruses. An interim
biosafety risk assessment (2) was prepared in response to the threat of a
pandemic posed by the highly pathogenic H5N1 avian influenza viruses
and the need to begin development of experimental vaccines. This threat
persists and several countries are now planning large-scale production
of H5N1 vaccine. The risk assessment that informed the WHO biosafety
guidance for pilot-lot vaccine production (2) has therefore been reassessed
in light of the intended greater scale of vaccine production and because
production facilities are likely to be different from those used in developing
small pilot lots, and also taking into account the experience gained from
developing and testing vaccine reference viruses derived by reverse genetics
from highly pathogenic avian influenza viruses.
This document follows the risk-assessment scheme used in the WHO
biosafety guidance for pilot-lot vaccine production, but is extended to include
considerations relating to the greater production-scale needed to supply
large quantities of vaccines. The risks associated with large-scale production
are likely to be different from pilot lots, e.g. the “open” aspect of some
production processes and quantity of virus-containing waste. It also takes
into account the considerable experience gained from highly pathogenic
avian influenza viruses, and the hazards associated with such strains.
Furthermore, the range of options for vaccine development is broader than
originally considered in the WHO risk assessment for pilot lot production
and the present document has been expanded to encompass current vaccine
development pathways.

1. Scope of the risk assessment

Much effort has recently gone into the development of H5N1 vaccine
and manufacture and the guidance presented is strongly influenced by
the experience gained with this strain and our greater knowledge of H5

270
strains in general. It is, nevertheless, intended that the guidance will also be
applicable to future threats from other potential pandemic strains, such as
H2 or highly pathogenic H7.
There is a range of possible pathogenicities in the viruses used in candidate
vaccine production not only for humans but also for other mammals and
avian species. On the one hand, H5 viruses that can be highly pathogenic
to both humans and chickens have been used to produce reassortant viruses
genetically modified to be of low pathogenicity for chickens and mammals.
On the other hand, for strains inherently less pathogenic for humans, wild-
type virus might be used directly for vaccine production. Thus reassortants
derived by reverse genetics, empirically-derived reassortants, which may
or may not be genetically modified, and native wild strains are within the
scope of these guidelines.
Eggs have traditionally been used for the production of influenza vaccines,
but cell culture techniques have been recently introduced and international
expectations for production and quality control specifications defined (3).
For the development of pandemic vaccine, either method may be used; thus
both egg and cell culture production methodologies are within the scope of
this document.
Most effort to date with candidate pandemic vaccine development has
been targeted towards inactivated vaccines. In one country however two
live attenuated virus vaccines for potential pandemic strains are under
development. This may raise important issues beyond the risks to humans,
namely the potential for excreted viruses or their derivatives to infect and
replicate in non-human species particularly in those raised for commercial
purposes. As the detection of H5 and H7 influenza strains are notifiable strains
to the Office International des Epizooties (OIE), widespread dissemination
of such vaccine strains could have a significant economic impact as well as
ramifications for international trade. Developers and regulators will need
to assess both the human and the agricultural risk of live pandemic strain
vaccines under development should shedding and replication be possible.
Both vaccine types (inactivated and live) are therefore covered in the scope
of these guidelines.
Furthermore it is intended that the risk assessment and the guidelines on
containment measures should apply to all facilities and laboratories that
have a need to handle live vaccine virus. This includes not only the vaccine
manufacturing facility but also to the quality control laboratories of the
manufacturer and, if appropriate, to National Control Laboratories. The
transport of live virus materials within and between sites should comply
with international specifications (4).
Finally it should be noted that the risk assessment for vaccine manufacture
will vary according to whether production is occurring in an interpandemic

271
 period, in a pandemic alert period (as for example early in 2004 when H5N1
was threatening to circulate extensively in South East Asia) or in a pandemic
period. These guidelines are intended to describe steps to minimize the risks
associated with the production and testing of vaccines with emphasis on
the interpandemic period, while indicating modifications that may be found
appropriate during other periods.

2. Hazard identification
Hazards associated with pandemic vaccine manufacturing and laboratory
testing are dependent on the type of pandemic vaccine strain (reassortant or
wild type), method of production (egg-based or cell-based) and whether it is
an inactivated or live attenuated virus vaccine. The type of vaccine strain, the
proposed testing schedule and containment level are illustrated in Table 1.

2.1 Hazards associated with the type of pandemic vaccine virus

2.1.1 Hazards associated with the recipient virus in a reassortant strain
Pandemic vaccine reassortants have been produced on the human strain A/
PR/8/34 (PR8) as recipient virus. PR8 has had over 100 passages in each of
mice, ferrets and embryonated chicken eggs. The result of such a passage
history is complete attenuation of the virus and its inability to replicate in
humans (5).
PR8 reaches a high titre in embryonated chicken eggs and since the late 1960s,
it has been used to produce “high growth reassortants” in combination with
the prevailing influenza A vaccine strain. The use of such reassortants as
vaccine strains has increased vaccine yield many-fold. The reassortants are
produced by a mixed infection of eggs with PR8 and the nominated vaccine
strain, combined with a selection system based on anti-PR8 antibody and
growth at high dilution.
Live attenuated influenza vaccines are licensed in some countries. The
parental strains used in such vaccines, e.g. A/Ann Arbor/6/60, are also
potential recipient strains for the development of pandemic reassortant
vaccines. These parental strains possess phenotypic markers of vaccine
safety, such as temperature sensitivity, cold-adaptation and attenuation in
ferrets or rodents and moreover have a demonstrated attenuated phenotype
in humans.

2.1.2 Hazards arising from the inserted gene product in a reassortant

vaccine strain
The products of the inserted genes will be, at a minimum, the haemagglutinin
(HA) and neuraminidase (NA) of the pandemic strain virus. For reassortants
derived from highly pathogenic H5 or H7 strains by reverse genetics, the

272
 HA will have been modified so that the multiple basic amino acids at the
HA cleavage site, which are associated with high pathogenicity, will be
reduced to a single basic amino acid. Any protein derived from the wild-
type strain on its own will be neither inherently infectious nor harmful.

2.1.3 Hazards arising from reassortant viruses

2.1.3.1 Direct hazards
Without treatment, reassortant viruses may be expected to survive for at
least a short time (hours) on surfaces or in a laboratory environment and
thus provide a potential means of infection for laboratory workers. Although
the surface antigens of reassortants, particularly the HA, can contribute
to pathogenicity (5, 6) published information indicates that a reassortant
between PR8 and a wild-type human influenza virus is likely to be avirulent in
humans (5, 7–9). Although such information is difficult to interpret because
the genetic composition of the reassortants was not clear, it is known that the
degree of attenuation increases as reassortants include more PR8 genes (10,
M Tashiro, unpublished data). The reassortants created by reverse genetics
as H5N1 pandemic reference strains contain six out of eight viral genes from
PR8 and the NA and modified HA genes of the H5N1 virus. Furthermore,
the H5 HA retains a preference for α2,3 linked residues (see below), so
the ability of the H5N1 reassortants to bind to and replicate in human cells
should be minimal. It is therefore envisaged that an H5N1 reassortant derived
by reverse genetics according to WHO guidance (11) would be attenuated
for humans compared to the H5 wild type. Furthermore, it is clear that
such reassortants are expected to be of low pathogenicity in chickens and
other animals compared to the highly pathogenic parental wild strains, and
this expectation has been borne out by experience to date. Nevertheless,
as the factors affecting pathogenicity are not fully understood (see below),
genetic manipulation to remove the polybasic sites could theoretically have
unpredicted effects on both transmissibility and pathogenicity.
For reassortants derived by traditional co-cultivation methods, the gene
constellation is less predictable. There is a theoretical possibility of
developing reassortants with more than two wild-type parental genes
or even of selection of a mutant (non-reassortant) wild-type virus with
improved growth characteristics. If vaccine production takes place in
the interpandemic phase there would be a need to determine the gene
constellation of reassortants derived by traditional co-cultivation methods
in order to conduct a full risk assessment.
Reassortants with a 6:2 gene constellation based on live attenuated
recipient strains such as A/Ann Arbor/6/60, or other strains used as live
attenuated vaccines, may also be used for the production of pandemic
influenza vaccine. The attenuated A/Ann Arbor/6/60 strain has been used

273
 as a backbone in 6:2 reassortant live attenuated vaccines in clinical studies
for more than 30 years using approximately 30 different vaccine strains, and
the data demonstrate that the Ann Arbor/6/60 virus produces reassortant
vaccine strains that are attenuated for humans (12). Live vaccines derived
from the Ann Arbor strain have been licensed in one country. An adequate
level of attenuation should be expected for modified H5 reassortant strains.
For each candidate pandemic strain, this should be verified by testing as
described below (section 3.6.1).
Reassortants may be also be derived from non-H5 or non-H7 viruses (e.g.
H9N2, H2N2) and may use either PR8 or an attenuated vaccine strain. The
hazards associated with such reassortants depend on HA receptor specificity.
If a reassortant has a preference for avian cell receptors (α2,3 linked sialic
acid e.g. avian H2N2 viruses), the hazards are considered to be no different
from those associated with the above-mentioned 6:2 reassortants derived
from attenuated H5 or H7 viruses (see section 3.3). However, if a reassortant
has a preference for mammalian cell receptors (α2,6 linkages, e.g. human
H2N2 pandemic virus from 1957), or possesses both avian and mammalian
receptor specificities (e.g. H9N2), there is a greater risk of human infection
(see Table 1).

2.1.3.2 Indirect hazards

Although it is considered that, for example, an H5N1/PR8 reassortant will
be either attenuated or possibly non-infectious to humans, an indirect hazard
may exist through secondary reassortment with a human or animal influenza
virus as influenza viruses are known to exchange genes by the process of
reassortment. For secondary reassortants to be generated, several events
need to occur; firstly infection of the production staff with the reassortant
strain; secondly, for the infected worker to have a mixed infection with a
wild type influenza virus, and thirdly for a reassortment event to take place.
In practice, manufacturers have 30 years of experience with large scale
production of vaccines based on PR8 reassortants and there have been no
reported cases of human illness. However, it should be noted that this does
not rule out the possibility of infections having occurred. Additionally, at the
point when seasonal influenza vaccines become available, production staff
can be vaccinated to reduce the chances of an infection with a circulating
wild-type virus.
In practice, the lack of success in producing H5N1 reassortant vaccine
strains in 1997 (UK: avian and swine viruses; Australia and USA: avian
and PR8 viruses) suggests that the probability of producing H5 reassortants
between mammalian and avian viruses in human cells is slight. It should
also be considered that poultry and pig farmers are continually exposed
to animal influenza viruses and there have been few documented cases of
human infection in this population with a reassortant between an avian or

274
 porcine and a human influenza virus. Based on these considerations the
probability that a PR8 reassortant strain will replicate and combine with
another influenza virus(es) in human cells is considered to be minimal. The
risk of such secondary reassortments for animal species will be considered
in the environmental risk assessment section (see section 3.2).

2.1.4 Hazards arising from the use of wild type viruses for pandemic strain
vaccine production
Wild-type strains may be considered for production purposes and different
potential vaccine candidates could be:
— an avian strain with no record of human infection (surrogate virus);
— an avian strain with documented human infection (potential pandemic
virus);
— an actual human pandemic virus, or a past H2N2 pandemic virus.
The hazards from wild type vaccine strains will differ according to the
category of wild-type virus used but in all cases are compounded during
vaccine manufacturing and associated vaccine product testing, due to the
high volumes and high titres encountered. With the exception of surrogate
viruses, the use of wild type pandemic-like influenza viruses to develop
pandemic vaccine strains presents considerable biosafety risks to personnel
in vaccine manufacturing facilities and testing laboratories, and also to the
general community if manufacture is taking place for clinical studies or
stockpiling of vaccines during the interpandemic period.

2.2 Hazards arising from the production process

Vaccine manufacture follows Good manufacturing practices for biologicals
(13). Good manufacturing practices (GMP) require protection of the
product from the operator and the environment and thus amelioration of
certain hazards associated with production will require the establishment of
a suitable balance between GMP and biosafety requirements.

2.2.1 Production in eggs

Influenza vaccine has been produced in embryonated hens’ eggs on a
large scale since the early 1950s. Much experience has been gained and
some facilities are capable of handling large numbers of eggs on a daily
basis with the aid of mechaniyed egg handling, inoculation and harvesting
machines.
Hazards occur only during the production stages and quality control
laboratory activities prior to virus inactivation. The most hazardous
production stage is egg harvesting when the eggs have to be opened to
harvest the allantoic fluid. The volume and titre of virus is higher at this

275
 stage than at any other. The open nature of the operations leads to a greater
exposure to aerosols and spills. In contrast during egg inoculation, the
virus used is dilute and of a relatively small volume. The allantoic fluid that
is harvested from the eggs is invariably manipulated thereafter in closed
vessels and hazards arising from live virus during downstream processing
and during the virus inactivation process, if used, are therefore less than
during virus harvest. Collection and disposal of egg waste is potentially a
major environmental hazard. Safe disposal of the waste from egg-grown
vaccines, both within the plant and outside, is therefore critical.

2.2.2 Production in cell cultures

For pandemic influenza vaccines produced on cell cultures, the biosafety
risks associated with manufacturing will depend primarily on the nature
of the cell culture system employed. Closed systems, such as bioreactors,
normally present little to no opportunity for exposure to live virus during
normal operation, but additional safety measures must be taken during
procedures where samples are introduced into or removed from the
bioreactor, and during procedures to deal with accidental spills. Roller bottles
and cell culture flasks used for virus production may allow exposure to live
virus through aerosols, spills, and other operations during virus production
and, thereafter, additional risks are associated with the inactivation and
disposal of the large quantities of contaminated solid waste generated by
this method.
The possibility exists that genetic mutations may be selected in pandemic
vaccine viruses during passage in mammalian cells that render them more
adapted to humans. Sequence analysis of the region of the HA gene encoding
the receptor binding site may be useful. However, it should be noted that
little is known about the relation between cell substrate and virus reversion
or adaption. Beare et al. (5) tried to de-attenuate PR8 by multiple passage
in organ cultures of human tissue, but failed, whereas studies with MDCK
cells (14) demonstrated that human viruses that retained their α2,6 receptor
specificity (human-like) were likely to mutate to an α2,3 specificity (avian-
like) as this provided a replicative advantage on MDCK cells, rather than the
reverse. Overall, hazards arising from the inherent properties of a reassortant
or wild type virus are likely to be far greater than the probability of adaptation
of the virus to a more human-like phenotype.

2.3 Factors affecting pathogenicity for humans

2.3.1 HA receptor specificity
The influenza HA is responsible for attachment of virus to the target cell
and has specificity for sialic acid receptors on cell surface molecules. The
HAs present on human influenza A viruses preferentially bind to receptors

276
 containing α2,6-linked sialic acid residues, whereas avian influenza viruses
preferentially bind to α2,3-linked sialic acid (15). Human tracheal cells
have mainly α2,6 linked residues (16), so the acquisition of an avian HA by
PR8 virus is expected to minimize potential binding to human respiratory
epithelial cells. Although the α2,3 receptor specificity of avian viruses
will reduce the efficacy of such binding, it may not completely prevent
infection in humans. Moreover, the presence of avian-like receptors has
been demonstrated in human respiratory tract epithelium (17). Beare and
Webster (18) found that over 100 fold higher quantities of avian viruses
(between 106.8 and 109.2 egg infectious doses) were needed for replication
in humans and, because replication was poor, that it was not possible to
induce person-to-person transmission.
There have been many reports of human infections with avian H5N1
viruses since 1997 in south-east Asia. It is possible that exposure to high-
titre H5N1 virus in contaminated chicken or duck carcasses or animal
products may have overcome the avian specificity of HA receptor binding.
Virus replication in such human cases was much better than in the earlier
experimental studies of avian influenza viruses in humans (18); however,
the extensive replication of H5N1 viruses in these people is inexplicable on
the basis of current knowledge of receptor specificity because the viruses
isolated from them retained the α2,3 avian specificity.

2.3.2 HA cleavability
The HA of influenza virus must be cleaved into HA1 and HA2 by host
cell proteases as a prerequisite for infectivity, and this cleavage has been
correlated with virulence. The pathogenicity of H5 and H7 influenza A
viruses in chickens is largely determined by the nature of the amino acids
at the HA cleavage site. H5 and H7 viruses with multiple basic amino acid
sequences are highly pathogenic and their HA can be effectively cleaved
by the ubiquitous furin-like proteases, which are expressed in most organs
of birds and humans. In contrast, the HA of H5 and H7 viruses of low
pathogenicity for birds and certain laboratory animals contain a single
basic residue at the cleavage site, a feature common to all other subtypes
of influenza HA, and which can only be cleaved by trypsin-like proteases,
which are restricted to certain cell types, e.g. epithelial cells lining the
respiratory tract of humans and the gut of birds. Thus, HA cleavability
influences tissue specificity and is a major determinant of pathogenicity
for H5 and H7 viruses in chickens and certain laboratory animals. Multiple
basic amino acids at the cleavage site have not been observed for any other
HA subtype.
Direct evidence has been obtained that both HA cleavage and HA receptor
specificity have an effect on tissue tropism of an avian H7N1 virus, A/

277
 Fowl Plague/Rostock/34 in chicken embryos (19). Similarly, the available
evidence from the H5N1 infections in 1997 demonstrates that the high
degree of pathogenicity in chickens, mice and ferrets is directly influenced
by the presence of the multiple basic amino acids. Webby et al. (20)
demonstrated that removal of the basic amino acids changed H5N1 infections
from a fatal systemic infection to a localized non-pathogenic infection in
chickens (i.e low pathogenicity for chickens), mice and ferrets. Hatta et
al. (21) and Lipatov et al. (22) have also shown by reverse genetics that
high cleavability of H5N1 HA due to the presence of multiple basic amino
acids was an essential requirement for a lethal mouse infection. It is not
ethical to examine the pathogenicity of influenza virus infection in humans,
but an examination of H5N1 viruses by Gao et al (23) provided evidence
that pathogenicity in mice can resemble that in humans. The occurrence
of multiple organ failure after human H5N1 infections is suggestive of an
unusual tissue tropism. Although evidence for viral replication outside the
lung has been described for at least one human case (24), such evidence
remains difficult to document (25).
The available evidence suggests that virulence of the 1997 and later H5N1
viruses for humans is related to the presence of the HA multiple basic amino
acids. It is therefore considered imperative to remove them, if present in the
HA of any H5N1 virus being developed as a vaccine strain, to reduce the
potential for harm to humans. This procedure will also increase the safety
of the reassortants for avian species (see below under environmental risk
assessment) as cleavage site modifications have resulted in a reduction of
their pathogenicity in avian embryos (26). It should be noted that during
production of reassortants by reverse genetics, base substitutions are
introduced to stabilize the removal of multiple basic amino acids during
passage of reassortants.

2.3.3 Other factors affecting pathogenicity

Although it is clear from experience in south-east Asia from 1997 to the
present that H5N1 influenza viruses that display α2,3 sialic acid specificity
could replicate in humans, it must be noted that influenza virus pathogenicity
does not depend solely on HA, but is a polygenic trait. The 1997 H5N1 virus
had unusual PB2 and NS1 genes that influenced pathogenicity whereas the
2004 H5N1 viruses possess complex combinations of changes in different
gene segments that affect pathogenicity in ferrets (27). Changes in the PB2
gene of the 1997 H5N1 viruses were sufficient to attenuate them for mice
(21) and changes in the NS1 protein rendered these viruses resistant to the
effects of interferons and other cytokines produced as part of the innate
immune response (28). The changes to NS1 conferred a highly virulent
phenotype which allowed replication to proceed unchecked in vivo. In this
case even a virus with a poor affinity for its receptor was able to replicate

278
 (although not to transmit). In contrast, viruses with a gene constellation
producing PR8 internal proteins were clearly sensitive to the innate
immune mechanisms which prevent the establishment of infection by an
avian virus in humans. This may well explain why in the outbreaks of H5
avian influenza before 1997, no evidence of transmission from birds to
humans was noted. Further, prior to the 2003 outbreak in the Netherlands,
only two cases of transmission of H7 viruses from birds to humans were
documented (29, 30). Also during the many years of laboratory handling of
high-titre avian viruses (of which one H7 strain (A/FPV/Dobson) is known
to contain a gene which adapts it for replication in mammalian cells (31)),
there has only been one report in the literature of a worker being affected
by these viruses. This was a laboratory worker in Australia who developed
conjunctivitis after accidentally being exposed to a H7N7 virus directly in
the eye (32). The PR8/H5N1 6:2 reassortants and the A/Ann Arbor/6/60 live
attenuated 6:2 reassortants created by reverse genetics for the production of
H5N1 vaccine do not contain the gene constellation considered necessary
for pathogenicity in chickens, mice and ferrets and in contrast have internal
genes that confer sensitivity to the innate immune response.

2.4 Hazards arising from the vaccine

Inactivated pandemic influenza vaccines present no biosafety risks provided
that the results of the inactivation steps show complete virus inactivation, as
the viral vaccine is rendered incapable of replication.
In an interpandemic or pandemic alert period, pilot-scale live attenuated
pandemic influenza vaccines may be developed for clinical evaluation. As
there is some uncertainty concerning the biosafety risks associated with
shedding or other unintentional release into the environment following
vaccination, subjects participating in clinical trials in the interpandemic or
pandemic alert phase should be kept under appropriate clinical isolation
conditions. If this is not done, indirect hazards for humans could arise as
considered in section 3.1. Furthermore, for pandemic human influenza
vaccine strains that express H5 or H7 avian influenza genes, there will be
potential consequences for agricultural systems (section 3.2.2). If viruses
of the H5 or H7 subtype become transmissible in livestock, this would be
notifiable to OIE and could result in sanctions with serious economic and
trade implications to prevent the spread of disease.
If a human pandemic has already started, the hazards from live attenuated
vaccines elaborated above will not be relevant.

2.5 Previous large-scale experience with reassortants

Reassortants derived from PR8 have been used routinely for the production
of inactivated influenza vaccines for the past 30 years. This work involves

279
 the production of many thousands of litres of infectious egg allantoic fluids,
which create substantial aerosols of reassortant virus within manufacturing
plants. Most of the reassortants were made from wild type human strains that
had not yet been in widespread circulation. Thus, although the manufacturing
staff would have some susceptibility to infection with the wild type virus,
there have been no anecdotal or documented cases of work-related human
illness resulting from occupational exposure to the reassortants.
Similarly, reassortants derived from the A/Ann Arbor/6/60 strain have been
used for the production of live attenuated vaccine for at least 3 years and
no anecdotal or documented cases of work-related human illness have
been reported. While to date no conclusive study has been conducted to
detect silent infections for either the PR8 or live attenuated strains, and
thus infectivity in humans cannot be fully assessed, the attenuation status
of these vaccine strains continues to be supported by their excellent safety
record to date
However, unlike the situation with the human influenza strains selected
for the annual vaccine formulation, staff manufacturing an H5N1 vaccine
would have no previous immunological experience of the avian virus, and
would therefore be expected to be susceptible, although the risk of work-
related human illness and of transmission outside the facility is expected to
be slight and lower than for non-reassortant strains.

2.6 Testing of reference viruses being considered

for vaccine production
Vaccine reference viruses will be developed by a WHO laboratory or by
a laboratory approved by a national regulatory authority (hereafter, for
ease of reference, referred to as a WHO laboratory). The following tests
and specifications have been developed based on experience gained in the
evaluation of 6:2 reassortant H5N1 viruses produced on the PR8 and A/Ann
Arbor/6/60 backbones. The principles outlined should be applicable during
the interpandemic period to other reassortant strains, but exceptions may be
made if appropriately justified. Tests on wild-type viruses being considered
for vaccine production will need to be selected on a case-by-case basis.
The tests described below are usually conducted by the WHO laboratory
developing the reference strain.
In a pandemic alert period or a pandemic period, the requirement for
the conduct or the completion of some or all of these tests prior to the
distribution of a candidate reference strain may be relaxed based on the risk
assessment. For example, in a pandemic alert period, a candidate reference
strain which on the basis of molecular analyses, is expected to have a low
risk of human infection and transmission could be distributed to vaccine
manufacturers to enable them to begin preparation of their seed stocks prior

280
 to the completion of time-consuming tests such as the chicken and the ferret
pathogenicity tests. If a pandemic has already begun, and the pandemic virus
has become adapted to human infection, there may be no need to perform
all the pathogenicity tests indicated below. A risk assessment should be
performed for each candidate reference strain and the outcome will depend
on the nature of the strain and the pandemic period declared by WHO.

2.6.1 In vivo tests to evaluate pathogenicity of H5 and H7 viruses

For optimal interpretation of tests, the pathogenic properties of the candidate
reference virus, should be compared with those of the parental backbone
strain and the wild-type strain.
These tests should be performed under appropriate high laboratory
containment conditions (see section 4.3). Tests to be performed on the
candidate vaccine reference strain (see Table 1) by the WHO reference
laboratory that develops the reassortant strain include:
• The ability to plaque in the presence or absence of added trypsin. Viruses with
high pathogenicity can replicate in mammalian cell culture in the absence of
added trypsin, whereas those with low pathogenicity generally do not.
• The ability to cause chicken embryo death. Highly pathogenic viruses
cause rapid chicken embryo death upon inoculation into eggs whereas
removal of the multiple basic amino acids from a highly pathogenic strain
results in embryo survival (26).
• Pathogenicity in chickens. The chicken intravenous pathogenicity (IVP)
test is an important statutory test required by veterinary authorities, and
a reassortant virus must have an index of 1.2 or less before it can be
removed from high-level containment (33). Development of specifications
to indicate that the test articles have been correctly administered in the
IVP test would be beneficial.
• Attenuation in ferrets. The viruses should be shown to be attenuated
in ferrets or in other suitable animal models, provided they have virus
sensitivity equivalent to that of ferrets and a similar ability to discriminate
between highly pathogenic and non-pathogenic influenza viruses. These
tests compare the candidate reference virus with the wild type virus.
Detailed test procedures are described in Appendix 1. In the case of
H5N1 reassortants, the criteria used to evaluate this test are that virus
replication and clinical symptoms should be comparable to those induced
by the attenuated PR8 parent virus and should be milder than the wild-
type human H5N1 virus infection.
Ferrets were chosen because they have been used extensively as a good
indicator of influenza virus virulence for humans (reviewed by Smith
and Sweet, 34). Typically, human influenza viruses cause lethargy, nasal
discharge and occasionally fever in ferrets, and virus replication is usually

281
 limited to the respiratory system. PR8 virus has been assessed in ferrets and
found to cause few or no clinical signs, and virus replication is limited to
the upper respiratory tract. However, the 1997 and 2004 wild-type human
H5N1 viruses replicated in ferrets throughout the body, caused fever, weight
loss and occasionally death (27, 35). Thus, in the absence of human data,
the ferret is the best model to predict whether a virus will be pathogenic or
attenuated in humans.
It would be useful to be able to measure transmissibility as well as
pathogenicity of virus strains, but currently a well-characterized methodology
to do so is lacking. Intranasal administration of virus to chickens may be
one such method, and has been shown to be possible, but to date the test
is not standardized. Uninoculated birds in close contact with infected
birds in the intravenous pathogenicity test may provide some information
on transmissibility. Transmission studies in ferrets after oral and ocular
inoculation are also potentially useful, but need to be standardized.
Tests for safety in mice may provide useful information if the parent strain
is virulent in mice. Detailed test procedures are described in Appendix 1.
A reassortant virus should be used for vaccine manufacture only after
appropriate results have been obtained in the above tests. For H5 and H7
strains, the nucleotide sequence corresponding to the HA cleavage site
should be determined by the WHO laboratory to demonstrate the absence
of multiple basic amino acids in the vaccine candidate. After WHO has
declared a pandemic manufacturers may receive candidate reference strains
that have not been assessed fully for pathogenicity. In this case they should
handle the viruses appropriately depending on the nature of the virus and
the pandemic situation.

2.6.2 Genetic stability of H5 and H7 viruses

Genetic stability is an important issue as it is known that in poultry, wild-
type low-pathogenicity H5 and H7 avian viruses can become highly
pathogenic by mutation (insertion of basic amino acids at the HA cleavage
site) and this is the origin of the highly pathogenic H5 and H7 strains.
Although the derivation of low-pathogenicity candidate reference viruses
by reverse genetics involves the introduction of silent mutations in the
region of the HA cleavage site that should minimize the re-insertion of
multiple basic amino acids, during vaccine production, such viruses may
be passaged several times and it is therefore important to evaluate their
genetic stability at the cleavage site. Several attenuated reassortants have
now been produced between PR8 virus and highly pathogenic H5N1,
H5N3 and H7N1 viruses by reverse genetics (20, 26, 36, 37, FLUPAN
(http://www.nibsc.ac.ukflupan/)) and following extended passage in eggs
(up to 10), they have each retained their attenuated phenotype.

282
 Nevertheless, manufacturers should assess any H5 and H7 seed viruses
and vaccine virus harvests by sequence analysis of the HA cleavage site.
The need for studies of genetic stability for seed viruses prepared from
candidate reference strains derived by other methods should be assessed
on a case-by-case basis. At least one in vivo test (section 3.6.1) should be
applied, for example the egg embryo lethality test.

2.6.3 Evaluation of wild-type non-pathogenic H5 or H7 viruses or

reassortants derived from them
In view of the propensity for non-pathogenic H5 and H7 viruses to acquire
mutations leading to increased pathogenicity, it is advisable to conduct
the full spectrum of pathogenicity tests (in ferrets, chickens and chicken
embryos), as indicated in section 3.6.1.

2.6.4 In vivo evaluation of non-H5, H7 viruses or reassortants

derived from them
Ferret tests are required for non-H5, non-H7 candidate vaccine strains prior
to manufacture. The tests should be conducted under biocontainment levels
equivalent to that required for the production of the reference strain. The
other tests (specified in sections 3.6.1 and 3.6.2) are not required because
they are specific for reassortants derived from highly pathogenic H5 and
H7 viruses.

3. Risk assessment
3.1 Health protection
3.1.1 Likelihood of harm to human health
By virtue of PR8 attenuation, avian receptor specificity, loss of multiple
basic amino acids at the HA cleavage site and the absence of other H5N1
genes associated with pathogenicity in humans (i.e. NS1 or PB2 genes), it is
envisaged that an PR8 x H5N1 6:2 reassortant, although possibly infectious
to humans and ferrets, will have only a low probability of causing harm to
human health. On the basis of these arguments, reassortants derived from
H5 or H7 strains in which the multiple basic amino acid HA0 cleavage
site has been removed, using either PR8 or strains attenuated for humans
e.g. the A/Ann Arbor/6/60 as the recipient virus, would be likely to be
similarly attenuated. Reassortants derived from all other subtypes or from
low pathogenicity H5 and H7 subtypes, in which the multiple basic amino
acids were not present, should also be attenuated by virtue of the receptor
specificity of the avian HA and the attenuating effect of the 6 PR8 genome
segments (absence of any other avian genes). The same arguments are also
valid for reassortants prepared from live attenuated virus strains such as
A/Ann Arbor/6/60.

283
 If staff at a vaccine production plant are exposed to aerosols containing high-
titre reassortant virus, sub-clinical infections could result. If this happened,
it is very unlikely that a reassortant virus would transmit to human contacts
as it is likely that replication will be attenuated and virus shedding, if it
occurs, it would be well below the titres considered to be needed for human
infection.
However, although there is no precedent, as described above there is a
theoretical possibility of secondary reassortment with normal human
influenza viruses and that such reassortant viruses may be replication-
competent in humans, while having avian-virus like coat proteins. Although
it is very unlikely that the secondary reassortant could become adapted to
human infection and transmission, were this to happen the public health
consequences would be serious. The likelihood of such occurences can be
reduced through biosafety measures designed to limit exposure of personnel
to high-titre materials during vaccine production and testing.
If non-reassortant wild-type viruses with multiple basic amino acids at
the HA cleavage site and high in vivo pathogenicity are used for vaccine
production they would potentially be highly pathogenic and transmissible in
humans. Stringent vaccine biosafety control measures are required to manage
the risk from vaccine production using such viruses. Non-reassortant wild-
type viruses, without multiple basic amino acids at the HA cleavage site with
low in vivo pathogenicity and avian receptor specificity are likely to be less
pathogenic and less transmissible in humans (18) than the wild type viruses
described above. However the risks of secondary reassortment with normal
human viruses remain and the risk that such reassortant viruses may be able
to replicate in humans. Appropriate vaccine biosafety control measures are
required to manage the risk from vaccine production using such viruses.
Non-reassortant wild-type viruses, without multiple basic amino acids at the
HA cleavage site, with low in vivo pathogenicity and mammalian receptor
specificity (e.g. human H2N2 and H9N2) are also likely to be less pathogenic
than the wild-type viruses described above, but their ability to transmit to
humans is unknown. Consequently, because of the risks of secondary
reassortments, appropriate biosafety control measures should be considered.

3.2 Environmental protection

3.2.1 Nature of the work
Egg-based vaccine production represents a relatively open system with
several operations likely to generate virus aerosols: namely, seed virus
preparation, egg inoculation, harvest of infected egg fluids, use of laminar
outward air flows, segregation of contaminated eggs, cleaning (that may
include high powered spraying) and decontamination of contaminated egg
trays, and disposal of waste products.

284
 Prior to the virus inactivation step, cell culture production requires handling
large volumes of high-titre preparations of live influenza virus. As mentioned
above, even in closed systems such as bioreactors leaks can occur, and
spillage or other operator contact with high-titre viral solutions during the
introduction of materials into the bioreactor, taking of samples, or clean-up
procedures is possible. If roller bottles or cell culture flasks are used in place
of bioreactors there will be a higher risk of generating aerosols and spills
due to the increased manipulations required, and the volume of materials to
be properly decontaminated for disposal will be proportionally greater.

3.2.2 Environmental considerations

Influenza A viruses are endemic throughout the world in some farm animals
(pigs and horses) and some populations of wild birds, specifically birds of
the families Anseriformes (ducks, geese and swans) and Charadriiformes
(shorebirds) (38). Of the influenza A viruses, a number can cause disease in
domestic poultry, such as H5, H7 and H9. H5 and H7 are thought to be highly
pathogenic in poultry, whereas H9 is typically less so. In addition, sporadic
infections by influenza A viruses have been reported in farmed mink, wild
whales and seals, dogs and captive populations of big cats (tigers and leopards)
(38, 39). In dogs, the influenza A infections were caused by H3N8 viruses
closely related to endemic equine viruses, and in the big cats, the infections
followed consumption of dead chickens infected with H5N1 viruses.
In the case of an H5N1 reassortant, the virus will have avian receptor
specificity, and thus birds would theoretically be the species most susceptible.
The contribution of the six PR8 internal genes to replication and virulence
in birds is unknown.
However, Hatta et al. (40) have recently demonstrated, by the use of reverse
genetics, that acquisition of only one PR8 gene by an avian influenza
virus can abolish virus replication in ducks. Experimental evidence has
demonstrated that PR8 virus is attenuated not only in humans (see above),
but also chickens (37). Furthermore, a reassortant between PR8 (internal
protein genes) and the 1997 Hong Kong H5N1 virus (NA and HA with a
single basic amino acid) replicated poorly in chickens and was not lethal.
Similar studies have been performed with the 2003 Hong Kong H5N1 virus
at the WHO Collaborating Centre, Memphis, USA (R Webster, unpublished
data), where the 6:2 PR8 reassortant did not replicate or cause signs of
disease in chickens. The removal of the multiple basic amino acids from
the H5 x PR8 reassortants in both studies probably played a major role in
reducing the risk for chickens.
Although replication occurs in chicken embryos, for reasons that are
unknown, the risk of environmental transmission via such replication in
nature is remote.

285
 Pigs are uniquely susceptible to infection by all strains of influenza A virus
because they have both alpha 2,3 and alpha 2,6 receptors in abundance.
Although pigs are not susceptible to infection with PR8, a reassortant
containing a single gene (HA) from an A/New/Jersey/76 (H1N1) isolate,
infected pigs and the animals excreted virus (6). It is thus conceivable that
pigs are susceptible to infection by an H5N1 reassortant, as viruses with
avian receptor specificity are known to replicate in this species. It is also
possible that these species would be susceptible to secondary reassortments
between the H5N1 reassortant and a pig virus. There is in fact evidence that
triple reassortants between avian, pig and human influenza viruses have
circulated in pigs (41).

3.3 Assignment of containment level

The production of influenza vaccine reassortant reference viruses, by WHO
Collaborating Centres, from highly pathogenic H5 or H7 wild type viruses
should take place at a high level of biocontainment (BSL-3 enhanced or BSL-
4, as advised by WHO and national authorities) (11). The collaborating centres
provide characterized reassortant reference viruses to vaccine manufacturers
who may develop vaccine seeds and vaccines from these materials.
In consideration of the hazards associated with egg and cell culture H5
and H7 vaccine production and quality control with reassortant viruses of
demonstrated low pathogenicity in chickens and/or in ferrets (and mice if
applicable), as specified in sections 3.6.1 and 3.6.2, the assigned containment
level is BSL-2 enhanced (pandemic influenza vaccine), as defined below
(see Table 1). This applies to both pilot-scale and large-scale production
during the interpandemic phase and pandemic alert period (42) when the
site of vaccine production is geographically remote from the site of the
emerging pandemic. Any subsequent relaxation of the levels of containment
during the developing pandemic, should be decided on a case-by-case basis
after careful evaluation of the risks.
In consideration of hazards associated with egg and cell culture vaccine
production and quality control with wild-type viruses (non-H5 and non-H7)
of demonstrated low pathogenicity in ferrets, as specified in section 3.6.3,
the assigned containment level is BSL-2 enhanced (pandemic influenza
vaccine), as defined below. This applies to both pilot-scale and large-scale
production during the interpandemic phase and pandemic alert period (42)
when the site of vaccine production is geographically remote from the
site of the emerging pandemic. Any subsequent relaxation of the levels of
containment during the developing pandemic, should be decided on a case-
by-case basis after careful evaluation of the risks.
In consideration of hazards associated with cell culture vaccine production
and quality control with highly pathogenic H5 or H7 wild-type viruses, the

286
 assigned containment level is BSL-3 enhanced (pandemic influenza vaccine),
as defined below. This applies to both pilot-scale and large-scale production
during the interpandemic phase and pandemic alert period (42) when the site
of vaccine production is geographically remote from the site of the emerging
pandemic. Any subsequent relaxation of the levels of containment during
the developing pandemic, should be decided on a case-by-case basis after
careful evaluation of the risks. In addition, the parts of the facility where
such work is done (both production and quality control) should meet the OIE
requirements for containment, which include not only biosafety, but also
requirements for biosecurity. (33). In view of the open nature of large scale
egg-based vaccine production, it is not possible to operate at BSL-3 enhanced
(pandemic influenza vaccine). Therefore egg-based vaccine production from
high pathogenicity H5 or H7 wild-type strains is not recommended.
For vaccine production and quality control using other types of vaccine
virus (e.g. reassortants derived from non-H5 or H7 viruses; wild-type low-
pathogenic H5 or H7 viruses), the assigned containment level is BSL2
enhanced (pandemic influenza vaccine), as defined below. This applies
to both pilot-scale and large-scale production during the interpandemic
phase and pandemic alert period (42) when the site of vaccine production
is geographically remote from the site of the emerging pandemic. Any
subsequent relaxation of the levels of containment during the developing
pandemic, should be decided on a case-by-case basis after careful evaluation
of the risks.
It should be noted that implementation of the containment conditions
described in this section within a production and quality control testing
facility must take into account the large quantities and high titres of live
virus that are produced, the industrial scale of facilities, as well as the
rules and regulations governing the manufacture and testing of medicinal
products known as good manufacturing practices (GMP) (13). The facility
requirements for a specific biosafety level within a manufacturing plant will
differ from the facility requirements within a laboratory handling smaller
quantities of infectious material such as a laboratory producing reassortant
reference viruses or in a pilot-scale facility. It should also be noted that
these biosafety requirements apply to the production and quality control
operations involving live viruses; virus lots shown to be inactivated by a
validated process need not be handled under these conditions.

3.4 Environmental control measures

Each vaccine manufacturer must review their own control measures in light
of the intended work, the nature of laboratory and production facilities and
the need to maintain GMP. Influenza specific enhanced containment measures
(defined in 4.4.1 and 4.4.2) should be in place for open manipulations with live

287
 virus, especially virus harvesting in egg production facilities. Quality control
facilities need to meet production containment requirements, and in some
regions, a second approval will be needed to meet other requirements such
as those regulating products containing materials derived from a genetically
modified organism (GMO).
Local safety regulations provide guidance on the disposal of potentially
infectious waste. Contaminated waste from current production facilities
may reach high virus titres. Decontamination methods should be validated.
If possible, decontamination of waste should take place on site. Where this
is not possible, there should be procedures in place to ensure that material is
safely contained and transported prior to decontamination off site. Guidance
on regulations for the transport of infectious substances is available from
WHO (4). In all cases the procedures should be validated to ensure that they
function at the scale of manufacturing.
In view of the possible exposure to high titre pandemic strain virus and the
need to reduce the chance of simultaneous infection with human influenza
viruses, staff should be prophylactically vaccinated with seasonal influenza
vaccines. It is anticipated that before large scale vaccine production is
attempted, pilot lots of pandemic strain vaccine will have already been
produced. Experimental vaccines inducing protective antibody levels
are recommended for use by staff before large scale vaccine production
commences if possible. Antiviral treatment must be available in case the
situation warrants it.
Each manufacturer should also assess the risk of contamination of birds or
pigs based on the likelihood of their being in the vicinity of the manufacturing
plant, and the manufacturing controls in use. Staff or other personnel entering
the area potentially exposed to live virus should avoid visiting pig, horse or
bird facilities (e.g. farms, equestrian events, bird sanctuaries) for at least
14 days following occupational exposure. This period should be extended
to 14 days after the symptoms resolve if conjunctivitis or respiratory signs
indicating the potential development of influenza infection or disease
develop during this 14 day period.
It is also known that mice can be experimentally infected with some
influenza viruses and the PR8 strain is known to be lethal for mice. It is not
known whether a reassortant based on PR8 will be able to replicate in mice,
but steps should be taken to prevent exposure of wild mice and the escape
of laboratory mice, and rodent control measures should be in place.
3.4.1 Specifications for “BSL2 enhanced (pandemic influenza vaccine)”
Specifications for BSL2 enhanced (pandemic influenza vaccine) facilities
include the following in addition to the principles for BLS2 facilities as
specified in the WHO Laboratory biosafety manual (1).

288
3.4.1.1 Facility
The facility should be designed and operated according to the stage of the
manufacturing process to meet the demands of protection of the recipient
of the vaccine, the staff producing and testing the vaccine and of the
environment. It is noted that different solutions may be needed depending
on the risks inherent in the operation(s) conducted in an area. Specialized
engineering solutions will be required that may include:
— use of relative negative pressure biosafety cabinets when possible;
— use of high-efficiency particulate air (HEPA) filtration of air prior to
exhaust into public areas or the environment; and
— use of positive pressure with negative pressure in-line sinks prior to
exhausting to the non-viral zone.
In addition the following decontamination procedures should take place:
— decontamination of all waste from BSL-2 enhanced (pandemic influenza
vaccine) areas; and
— decontamination of manufacturing and quality control areas at the end of
a production campaign through cleaning and validated decontamination
for example gaseous fumigation.

3.4.1.2 Personal protection

• Full-body protective laboratory clothing (for example Tyvek® disposable
overalls) is to be worn in the controlled BSL-2 enhanced (pandemic
influenza vaccine) area.
• If activities cannot be contained by primary containment and open activities
are being conducted, the use of respiratory protective equipment, such
as N95, FFP3 (43) or equivalent respirators is strongly recommended.
Minimal specifications for the filtering/absorbing capacity of such
equipment should be met, and masks, if used, must be fitted properly and
the correctness of fit tested.
• Personnel should be instructed, in a written document to which they sign
their agreement, not to have any contact with birds or pigs, in particular
farm animals for 14 days after departure from the facility where vaccine
has been produced. Currently the risks involved in contact with household
dogs and cats are not considered to be significant, but the available
scientific evidence is sparse.
• Staff should be prophylactically vaccinated with seasonal inactivated
influenza vaccines.
• It is anticipated that before large scale vaccine production is attempted, pilot
lots of pandemic strain vaccine will have already been produced. Experimental
vaccines inducing protective antibody levels are recommended for use by
staff before large scale vaccine production commences if possible.
• Antiviral treatment must be available in case the situation warrants it.

289
3.4.1.3 Monitoring of decontamination
• Cleaning and decontamination methods need to be validated periodically
as part of a master validation plan to demonstrate that the protocols,
reagents and equipment used are effective in the inactivation of pandemic
influenza virus on facility and equipment surfaces, garments of personnel
and waste materials, and within cell growth and storage containers.
Once decontamination procedures for influenza virus have been fully
described and validated, there is no need to repeat them for each new
strain. Validation studies using influenza viruses may be supplemented
by studies with biological (for example bacterial) markers selected to be
more difficult to inactivate than influenza.

3.4.2 Specifications for “BSL3 enhanced (pandemic influenza vaccine)”

Specifications for BSL3 enhanced (pandemic influenza vaccine) facilities
include the following requirements in addition to the principles for BLS3
facilities as specified in the WHO Laboratory biosafety manual (1), and are
additional to the specifications given above in section 3.4.1.

3.4.2.1 Facility
The facility should be designed and operated to meet the demands of
protection of the recipient of the vaccine, the staff producing and testing the
vaccine and of the environment. This will require specialized engineering
solutions that may include:
— negative pressure secondary containment areas
— HEPA filtration on supply and exhaust air
— on-site decontamination of liquid effluent
— floor dams should be erected around bioreactors or other large scale
equipment including storage tanks to contain spillage of virus from
large virus-containing vessels

3.4.2.2 Personal protection

• All clothing worn outside the facility should be replaced by manufacturing
facility garments upon entry into the facility.
• Upon entry into the containment zone personnel are to gown in full body
protective single-use laboratory clothing (for example Tyvek® disposable
overalls).
• When open activities are being conducted, eye protection and the use of
respiratory protective equipment, such as N95, FFP3 (43) or equivalent
respirators such as positive pressure air purifying respirators is required.
Minimal specifications for the filtering/absorbing capacity of such
equipment should be met, and masks, if used, must be fitted properly and
the correctness of fit tested.

290
 • Taking a full body shower upon exit from the BSL-3 enhanced (pandemic
influenza vaccine) containment facility is recommended. It is mandatory
following situations when staff may have been exposed to vaccine virus.
• Personnel should be instructed, in a written document to which they sign
their agreement, not to have contact with animals, in particular farm
animals 14 days following departure from the facility where vaccine has
been produced. Currently the risks involved in contact with household
dogs and cats are not considered to be significant, but the available
scientific evidence is sparse.
• Staff should be prophylactically vaccinated with seasonal influenza vaccines.
• It is anticipated that before large-scale vaccine production is attempted,
pilot lots of pandemic strain vaccine will have already been produced.
Experimental vaccines inducing protective antibody levels are recommended
for use by staff before large-scale vaccine production commences, if
possible.
• Antiviral treatment must be available as necessary.

3.5 Biosafety management and implementation within a vaccine

production facility
3.5.1 Management structure
The implementation of the biosafety levels described in these guidelines
requires that the institution employ a biosafety officer who is knowledgeable in
large-scale viral production and containment, but is independent of production
in his or her reporting structure. The biosafety officer is responsible for the
independent oversight of the implementation of the biosafety practices, policies
and emergency procedures in place within the company or organization and
should report directly to the highest management levels within the company.
A biosafety officer is needed in addition to a qualified person who, in some
countries, has overall responsibility for a medicinal product.
There should also be a Biosafety Committee comprising representatives of
viral production and quality control that is responsible for reviewing the
biosafety status within the company and for coordinating preventive and
corrective measures. The institutional biosafety officer must be a member of
the Committee. The chairperson should be independent of both the production
and quality control functions. The management and governing board of the
manufacturing company should ensure that adequate priority and resources
are made available to the Committee to implement the required measures.

3.5.2 Medical surveillance

Occupational health departments at vaccine manufacturers of pandemic
strain influenza vaccines should provide training in recognizing the clinical
signs of influenza infection to company physicians, nurses and vaccine

291
 manufacturing supervisors, who must make decisions on the health of
personnel associated with the manufacture and testing of pandemic strain
influenza vaccine. Local medical practitioners caring for personnel from
the manufacturing site should receive special training in the diagnosis and
management of pandemic influenza infection. Any manufacturer embarking
on large-scale production should have documented procedures for dealing
with influenza-like illness in the staff involved, or their family members,
including diagnostic procedures and prescribed treatment protocols.
Manufacturers should ensure that staff understand that they have an
obligation to seek medical attention and to report any influenza-like illness
to the occupational health department or equivalent. Manufacturers should
hold supplies of one or more effective antiviral agent(s) and have defined
means of quarantining staff if necessary.

3.5.3 Implementation
A detailed and comprehensive risk analysis should be conducted to define
possible sources of contamination of personnel or the environment that
may arise from the production or testing of live influenza virus within the
establishment. For each procedure or system, this analysis should take
into account the concentration and stability of the virus at the site, the
potential for inhalation or injection that could result from accidents, and the
potential consequences of a major or minor system failure. The procedural
and technical measures to be taken to reduce the risk to workers and the
environment should be considered as part of this analysis. The results of
this risk analysis should be documented.
A comprehensive Biosafety Manual must be created and implemented
that fully describes the biosafety aspects of the production process and
of the quality control activities. It should define such items as emergency
procedures, waste disposal, and the requirements for safety practices and
procedures as identified in the risk analysis. The manual should be made
available to all staff of the production and quality control units, with at
least one copy present in the containment area(s). The manual should be
reviewed and updated when changes occur and at least annually.
Comprehensive guidelines outlining the response to biosafety emergencies,
spills and accidents should be prepared and made available to key personnel
for information and for coordination with emergency response units.
Rehersals of emergency response procedures are helpful. These guidelines
should be reviewed and updated annually.
The implementation of the appropriate biosafety level status in the
production and testing facilities should be verified through an independent
assessment. National requirements concerning verification mechanisms
should be in place and complied with.

292
Table 1
Comparison of properties and proposed containment for pandemic vaccine
production using different vaccine reference viruses

Vaccine virus Haemagglutinin Tests needed on Proposed containment

receptor specificity reference virusa for vaccine production

H5, H7 reassortants, α2,3 residues Ferret, chicken, BSL-2 Enhanced

from HP virusesb sequence, plaquing, (pandemic influenza
egg embryo vaccine)

H5, H7 reassortants, α2,3 residues Ferret, chicken, BSL-2 Enhanced

from NP virusesb sequence, plaquing, (pandemic influenza
egg embryo vaccine)

Non-H5, H7 α2,3 residues Ferret BSL-2 Enhanced

reassortant (pandemic influenza
vaccine)

Non-H5, H7 α2,6 residues Ferret BSL-2 Enhanced

reassortant (pandemic influenza
vaccine)

H5, H7 HP viruses α2,3 residues Not applicable BSL-3 Enhanced

(pandemic influenza
vaccine)

H5, H7 NP viruses α2,3 residues Ferret, chicken, BSL-2 Enhanced

sequence, plaquing, (pandemic influenza
egg embryo vaccine)

Non-H5, H7 viruses α2,3 residues Ferret BSL-2 Enhanced

(pandemic influenza
vaccine)

Non-H5, H7 viruses α2,6 residues Ferret BSL-2 Enhanced

(pandemic influenza
vaccine)
a
Test performed by WHO reference laboratory.
b
Highly pathogenic and nonpathogenic viruses.

293
Authors
Four background documents1 were discussed in a teleconference on 27 July 2005
convened by the World Health Organization, Geneva, Switzerland (Dr D. Wood,
S. Lambert, A. Mohammadi and B. Kay) attended by the following persons:
Dr P. Celis, European Medicines Agency, London, England; Mr T. Colegate, Chiron
Vaccines, Liverpool, England; Dr J. Katz, Centers for Disease Control, Atlanta,
USA; Dr C. Gerdil, Sanofi Pasteur, Marcy l’Etoile, France; Dr G. Grohmann,
Therapeutic Goods Administration, Woden ACT, Australia; Dr A. Hampson,
WHO Collaborative Centre for Influenza, Parkville, Victoria, Australia; Dr A. Hay,
WHO Collaborative Centre for Influenza, National Institute for Medical Research,
London, England; Dr R. Levandowski, Food and Drug Administration, Bethesda,
Maryland, USA; Mr P. Logan, Health and Safety Executive, Merseyside, England;
Dr J. Robertson, National Institute for Biological Standards and Control, Potters
Bar, Herts, England; Dr D. Swayne, Department of Agriculture, USA; Mr J. Richmond,
Atlanta, Georgia, USA.

A first draft document was prepared by the WHO Secretariat (Dr D. Wood) based
on the outcome of the teleconference and the commissioned papers. Comments
on this first draft were received from Dr Alexander, Dr A. Hampson, Dr A. Hay,
Dr P. Logan, Dr J. Robertson, Dr D. Swayne and the International Federation of
Pharmaceutical Manufacturers (IFPMA) Influenza Vaccine Supply International
Task Force. A version of the document for public comment (WHO/BS/05.2026) was
prepared by the WHO Secretariat (Dr D. Wood) taking into account the comments
received and further review by Dr J. Robertson and Dr J. Wood.

The final draft version of the document (WHO/BS/05.2026,12 October 2005) was
prepared by the Secretariat (Dr D. Wood and Dr S. Lambert) taking into account
comments from participants at a WHO informal consultation on WHO/BS/05.2026,
held in Geneva from 19–20 September 2005 attended by the following persons:
Mr T. Colegate, Chiron Vaccines, Liverpool, England; Dr G. Grohmann, Therapeutic
Goods Administration, Woden ACT, Australia; Dr I. Kallings, Swedish Institute for
Infectious Disease Control, Solna, Sweden; Dr T. Kurata, National Institute of
Infectious Diseases, Tokyo, Japan; Dr Y. Lawanprasert, Food and Drug
Administration, Nonthaburi, Thailand; Dr P. Logan, Merseyside, England; Dr J.
Lubroth; Food and Agriculture Organization of the United Nations (FAO), Rome,
Italy; Dr P. Payette, Public Health Agency of Canada, Ottawa, Canada; Mr S.
Phoshoko, National Department of Health, Pretoria, South Africa; Dr I. Raw, Instituto
Hutantan, São Paolo, Brazil; Dr J. Richmond, Southport, North Carolina, USA; Dr J.
Robertson, National Institute for Biological Standards and Control, Potters Bar,
Herts., England; Dr J-F Saluzzo, Sanofi Pasteur, Marcy l’Etoile, France; Dr N.T. Van,

1
The following series of background papers, commissioned by the WHO Secretariat, were
prepared in the period April–July 2005.
a
A review of WHO biosafety guidelines for Manufacturing Avian Influenza Vaccines (Frey,
Richmond, Robinson).
b
A risk assessment for large scale manufacture of inactivated influenza vaccines from reassortants
derived from avian influenza viruses (Wood, Robertson, Logan).
c
Industry pandemic biosafety position paper (IFPMA influenza vaccine supply international task force).
d
Conceptual risks of reassortants for the environment (Swayne).

294
 Vabiotech, National Institute of Hygiene and Epidemiology, Hanoi, Viet Nam;
Dr T.G. Webster, St. Jude’s Children’s Research Hospital, Memphis, Tennessee,
USA; Dr J. Wood, National Institute for Biological Standards and Control, Potters
Bar, Herts, England. The WHO Secretariat included Dr L. Chocarro, Access to
Technologies; Dr B. Kay, Communicable Disease Surveillance and Response;
Dr S. Lambert, Quality and Safety of Biologicals; Dr A. Mohammadi, Communicable
Disease Surveillance and Response; Dr N. Previsani, Communicable Disease
Surveillance and Response; Dr Y. Pervikov, Initiative for Vaccine Research;
Dr J. Sokhey, WHO Regional Office for South-East Asia, New Delhi, India;
Dr K. Stohr, Global Influenza Programme; Dr D. Wood, Quality and Safety of
Biologicals and Dr W. Zhang, Global Influenza Programme.

Gratitude is also due to the following individuals for their written comments:
Dr A. Hampson, WHO Collaborative Centre for Influenza, Parkville, Victoria,
Australia; Dr I. Kallings, Swedish Institute for Infectious Disease Control, Solna,
Sweden; Dr P. Payette, Public Health Agency of Canada, Ottawa, Canada;
Dr J. Robertson, National Institute for Biological Standards and Control, Potters Bar,
Herts., England; Dr J. Wood, National Institute for Biological Standards and Control,
Potters Bar, Herts., England: and the Influenza Vaccine Supply Task Force.

References
1. Laboratory biosafety manual, 3rd ed. Geneva, World Health Organization,
2004 (WHO/CDS/CSR/LYO/2004.11).
2. Production of pilot lots of inactivated influenza vaccines from reassortants
derived from avian influenza viruses. An interim biosafety risk assessment.
Geneva, World Health Organization, 2003 (http://www.who.int/csr/resources/
publications/influenza/WHO_CDS_CSR_RMD_2003_5/en/).
3. Recommendations for production and quality control of inactivated influenza
vaccines. In: WHO Expert Committee on Biological Standardization. Fifty-
fourth report. Geneva, World Health Organization, 2005, Annex 3 (WHO
Technical Report Series, No. 927).
4. Guidance on regulations for the transport of infectious substances. Geneva,
World Health Organization, 2005 (WHO/CDS/CSR/LYO/2005.22).
5. Beare AS, Schild GC, Craig JW. Trials in man with live recombinants made from
A/PR/8/34 (H0N1) and wild H3N2 influenza viruses. Lancet, 1975, ii:729–732.
6. Kilbourne ED. Influenza: viral determinants of the pathogenicity and
epidemicity of an invariant disease of variable occurrence. Philosophical
Transactions of the Royal Society of London. Series B, Biological Sciences,
1980, 288:291–297.
7. Beare AS, Hall TS. (1971) Recombinant influenza A viruses as live vaccines
for man. Lancet, ii:1271–1273.
8. Florent G. Gene constellation of live influenza A vaccines. Archives of
Virology, 1980, 64:171–173.
9. Oxford JS, McGeoch DJ, Schild GC, Beare AS. Analysis of virion RNA
segments and polypeptides of influenza A virus recombinants of defined
virulence. Nature, 1978, 273:778–779.

295
 10. Florent G et al. RNAs of influenza virus recombinants derived from parents
of known virulence for man. Archives of Virology, 1977, 54:19–28.
11. WHO guidance on development of influenza vaccine reference viruses by
reverse genetics. Geneva, World Health Organization, 2005 (WHO/CDS/CSR/
GIP/2005.6).
12. CDC (2005) Vaccine information statement 2005–06
13. Good Manufacturing Practices for Biologicals. In: WHO Expert Committee
for Biological Standardization. Forty-second report. Geneva, World
Health Organization, 1992, Annex 1 (WHO Technical Report Series,
No. 878).
14. Robertson JS, Cook P, Attwell A-M, Williams SP. Replicative advantage
in tissue culture of egg-adapted influenza virus over tissue-culture derived
virus: implications for vaccine manufacture. Vaccine, 1995, 13:1583–1588.
15. Rogers GN, D’Souza BL. Receptor binding properties of human and animal
HI influenza virus isolates. Virology, 1989, 173:317–322.
16. Nelson J et al. Influenza virus strains selectively recognize sialyooligo-
saccharides on human respiratory epithelium: the role of the host cell
in selection of haemagglutinin receptor specificity. Virus Research, 1993,
29:155–165.
17. Matrosovich MN et al. Human and avian influenza viruses target different
cell types in cultures of human airway epithelium. Proceedings of the National
Academy of Sciences, 2004, 101:4620–4624.
18. Beare AS, Webster RG. Replication of avian influenza viruses in humans.
Archives of Virology, 1991, 119:37–42.
19. Feldmann A, et al. Targeted infection of endothelial cells by avian influenza
virus A/FPV/Rostock/34 (H7N1) in chicken embryos. Journal of Virology,
2000, 74:8018–8027.
20. Webby RJ et al. Responsiveness to a pandemic alert: use of reverse
genetics for rapid development of influenza vaccines. Lancet, 2004,
363:1099–1103.
21. Hatta M, Gao P, Halfmann P, Kawaoka Y. Molecular basis for high virulence
of Hong Kong H5N1 influenza A viruses. Science, 2001, 293:1840–1842.
22. Lipatov AS et al. Efficacy of H5 influenza vaccines produced by reverse
genetics in a lethal mouse model. Journal of Infectious Diseases, 2005,
191:1216–1220.
23. Gao P et al. (1999) Biological heterogenicity, including systemic replication in
mice, of H5N1 influenza A virus isolates from humans in Hong Kong. J. Virol.
73: 3184–3189.
24. Uiprasertkul M et al. Influenza A H5N1 replication sites in humans. Emerging
Infectious Diseases, 11, 1036–1041.
25. To K et al. Pathology of fatal human infection associated with avian influenza
A H5N1 virus. Journal of Medical Virology, 2001, 63:242–246.
26. Nicolson C, Major D, Wood JM, Robertson JS. Generation of influenza
vaccine viruses on Vero cells by reverse genetics: an H5N1 candidate
vaccine virus produced under a quality system. Vaccine, 2005,
23:2943–2952.

296
27. Govorkova EA et al. Lethality to ferrets of H5N1 influenza viruses
isolated from humans and poultry in 2004. Journal of Virology, 2005,
79:2191–2005.
28. Seo SH, Hoffman E, Webster RG. Lethal H5N1 influenza viruses escape
host anti-viral cytokine responses. Nature Medicine, 2002, 9:950–954.
29. Campbell et al. Fowl plague virus from man. Journal of Infectious Disease,
1970, 122:513.
30. Kurz J, Manvell RJ, Banks J. Avian influenza virus isolated from a woman
with conjunctivitis. Lancet, 1996, 348:901–902.
31. Almond JW. A single gene determines the host range of influenza virus.
Nature, 1977, 270:617–618.
32. Taylor, H.R., Turner, A.J. A case report of fowl plague keraconjunctivitis.
British Journal of Ophthalmology, 1977; 61: 86–88.
33. Manual for standards for diagnostic tests and vaccines for terrestrial animals.
5th ed. Paris, Office International des Epizooties (OIE), 2004.
34. Smith H, Sweet C. Lessons for human influenza from pathogenicity studies
in ferrets. Reviews of Infectious Diseases, 1988, 10:56–72.
35. Zitzow LA et al. Pathogenesis of avian influenza A (H5N1) viruses in ferrets.
Journal of Virology, 2002, 76:4420–4429.
36. Liu M et al. Preparation of standardized efficacious agricultural H5N3
vaccine by reverse genetics. Virology, 2003, 314:580–590.
37. Subbarao K et al. Evaluation of a genetically modified reassortant H5N1
influenza A virus vaccine candidate generated by plasmid-based reverse
genetics. Virology (2003) 305: 192–200.
38. Swayne DE, Halvorson, DA. Influenza. In: Saif YM et al. eds. Diseases of
Poultry. 11th ed. Ames, Iowa, Iowa State University Press, 2003:135–160.
39. Keawcharoen J et al. Avian influenza H5N1 in tigers and leopards. Emerging
Infectious Diseases, 2004, 10:2189–2191.
40. Hatta M, Halfmann P, Wells K, Kawaoka Y. Human influenza A viral genes
responsible for the restriction of its replication in duck intestine. Virology,
2002, 295:250–255.
41. Webby RJ et al. Evolution of swine H3N2 influenza viruses in the United
States. Journal of Virology, 2000, 74:8243–8251.
42. WHO global influenza preparedness plan. The role of WHO and
recommendations for national measures before and during a pandemic.
Geneva, World Health Organization, 2005 (WHO/CDS/CSR/GIP/2005.5).
43. European Union. EN 149:2001. Respiratory protection devices — Filtering
half-masks to protect against particles — Requirements, testing, marketing.

297
 Appendix 1
Testing for attenuation of influenza vaccine
strains in mammals

Titration of test virus

The dose of vaccine virus or parental strain virus that produces infection in
50% of cases should be determined by titration in eggs (EID50) or cell culture
(TCID50), as appropriate. Titration of vaccine virus stock and parental virus
stocks should be determined within the same laboratory and titres should be
sufficiently high that these viruses can be compared using equivalent high
doses in mice or ferrets (107 to 106 EID50 or TCID50).

Ferrets
Experimental procedure
Outbred ferrets 4–8 months of age are sedated either by intramuscular
inoculation of a mixture of anaesthetics (e.g. ketamine (25 mg/kg), xyalazine
(2 mg/kg) and atropine (0.05 mg/kg)) or by a suitable inhalant. A standard
dose of 107 EID50/TCID50 (as appropriate) (106, if the higher dose is not
possible) in 1 ml phosphate-buffered saline is slowly administered into the
nares of the sedated animal, making sure that the virus is inhaled and not
swallowed or expelled. A group of 4–6 ferrets should be infected. One group
of ferrets (2–3 animals) should be killed on day 3 or 4 post-infection and
the following tissues should be collected for estimation of virus replication:
nasal turbinates and/or swabs, lung (tissue samples from each of four lobes
and pooled), brain (tissues from anterior and posterior sections sampled and
pooled), spleen and intestine. Additional lung tissue may be collected and
processed for haematoxylin and eosin staining for microscopic evaluation
of histopathology. The remaining animals are observed for 14 days for signs
of weight loss, lethargy (based on a previously published index (1)), and
respiratory and neurological symptoms. Neurological involvement may
be confirmed by collection of brain tissue on day 14 post-infection at the
termination of the experiment and processing as above for histopathology.

Expected outcome
Viral titres of the vaccine strain in respiratory tissues should be no greater
than in either parental strain; a substantial decrease in lung virus replication
is anticipated. Replication of the vaccine candidate should also be restricted
to the respiratory tract and replication in the spleen or intestine is not
expected. Although isolation of the vaccine strain from the brain is not
desirable, if high viral titres are found in the nasal turbinates, there may
be some detection of virus in the brain based on previous results with non-

298
 virulent human H3N2 viruses (2). The significance of such a finding may be
confirmed by performing a histopathological analysis of brain tissue on day
14 post-infection. Neurological lesions detected in heamatoxylin and eosin-
stained tissue sections confirm virus replication in the brain. Neurological
symptoms and histopathology would indicate a lack of suitable attenuation
of the vaccine candidate. Likewise clinical signs of disease such as weight
loss and lethargy would indicate lack of attenuation in the vaccine strain,
assuming that the wild-type avian virus also causes these symptoms.

Mice
Experimental procedure
The 50% lethal dose (LD50) of the vaccine strain and parental virus strains
is determined in 6–8 week old female BALB/c mice. Mice are lightly
anaesthetized with an inhalant and groups of mice (4–8 per group) are
infected intranasally with 0.05 ml of serial 10-fold dilutions of virus
(expected dose range 107 to 101 EID50). Mice are observed daily for disease
signs and the numbers of deaths at each virus dilution are recorded. The
LD50 values are calculated by the method of Reed and Muench (3). An
additional three mice infected with a high dose of virus (e.g. 106) are killed
on day 3 or 4 post-infection and organs, including the lungs and brain, are
harvested for estimation of virus replication.

Expected outcome
If the wild-type avian strain replicates in the brain and is highly lethal for
mice, the vaccine candidate should exhibit at least a 1000-fold reduction
in LD50 values. Titres of the vaccine strain in lung and brain should be
lower than those of either parental strain, consistent with an attenuation of
replication in mouse tissues.

References
1. Reuman PD, Keely S, Schiff GM (1989). Assessment of signs of influenza
illness in the ferret model. Laboratory Animal Science, 42:222–232.
2. Zitzow LA et al. Pathogenesis of avian influenza A (H5N1) viruses in ferrets.
Journal of Virology, 2002, 76:4420–4429.
3. Reed LJ, Muench H (1938). A simple method for estimating fifty percent
endpoints. American Journal of Hygiene, 27:493–497.

299
© World Health Organization
WHO Technical Report Series No 941, 2007

Annex 6
Recommendations for whole-cell pertussis vaccine

These Recommendations provide information and guidance to national

regulatory authorities and vaccine manufacturers concerning the
characterization, production and control of whole-cell pertussis vaccines
to facilitate their international licensure and use. Each of the following
sections constitutes a recommendation. The parts of each section
printed in large type have been written in the form of requirements so
that if a national regulatory authority so desires these parts may be
adopted as definitive national requirements. The parts of each section
printed in small type are comments and recommendations for guidance.
It is recommended that modifications to these Recommendations be
made only on condition that the modifications ensure that the vaccine is
at least as safe and efficacious as that prepared in accordance with the
Recommendations set out below. It is desirable that the World Health
Organization should be kept informed of any such differences.
To facilitate the international distribution of vaccine made in accordance with
these recommendations, a summary protocol is given in the Appendix.

Introduction
General considerations
Part A. Manufacturing recommendations
A.1 Definitions
A.2 General manufacturing recommendations
A.3 Control of vaccine production
A.4 Filling and containers
A.5 Control of final product
A.6 Records
A.7 Samples
A.8 Labelling
A.9 Distribution and shipping
A.10 Stability, storage and expiry date
Part B. Nonclinical evaluation of whole-cell pertussis vaccines
Part C. Clinical evaluation of whole-cell pertussis vaccines
Part D. National Control Recommendations
D.1 General
D.2 Official release and certification
Authors
Acknowledgements
References
Appendix
Summary protocol

301
 Introduction
The WHO Requirements for whole-cell pertussis vaccine were first formulated
in 1963 and the requirements for diphtheria and tetanus vaccines were
prepared in 1964. Since diphtheria, tetanus and whole cell pertussis (DTwP)
vaccines have been commonly used in a combined form, the requirements
revised in 1978 contained separate sections on all three components as a
followed by a final section that applied to a combination. The last revision
of the requirements for diphtheria, tetanus, pertussis and combined vaccines
was made in 1989 and published in 1990 (1).
Since that time a number of developments have taken place in the production,
standardization and quality control of DTwP vaccines, as well as in the
understanding of Bordetella pertussis, and it was considered that the existing
requirements should be reviewed and where appropriate revised and updated.
An amendment to the diphtheria and tetanus sections concerning single
dilution and in vitro potency assays was adopted by the Expert Committee
on Biological Standardization in 2004 (2). The present revision of the
requirements for whole-cell pertussis vaccine should therefore be considered
as part of the revision of the overall requirements for DTP. In 1998 the title
WHO Requirements was changed to WHO Recommendations to better
reflect the nature of these documents. These recommendations for whole-cell
pertussis vaccine supersede those published in 1990 (1) and should be read
in conjunction with the recommendations for diphtheria and tetanus vaccines
when whole-cell pertussis vaccine is part of DTwP combined vaccine (1, 2).
Once both the diphtheria and tetanus sections have also been fully revised, it
is the intention to combine all three sections into one document.
Since 1989 a variety of combination vaccines based on DTP and involving
a number of additional antigens have been developed and licensed. Many
countries have already included tetravalent and pentavalent vaccines containing
hepatitis B, Haemophilus influenzae type b conjugate (Hib) and inactivated
polio vaccines (IPV), in addition to DTwP, in their immunization programmes.
A need for further guidance on the evaluation of combination vaccines based on
DTwP has been recognized and will be considered as a separate document.
In addition, a number of acellular pertussis vaccines have been licensed
and used in combination with other vaccines for more than 20 years.
Separate Guidelines for the production and control of the acellular pertussis
component of monovalent or combined vaccines were developed in 1996
(3) (and their revision is also being undertaken separately).

General considerations
Pertussis is an important cause of infant death worldwide and continues to
be a public health concern even in countries with high vaccination coverage.

302
Recent estimates from WHO suggest that in 2002 about 18 351 000 cases
of pertussis occurred worldwide, the vast majority in developing countries,
and that about 294 000 of those infected died. It is further estimated that
in 2002, global vaccination against pertussis averted more than 37 million
cases and 587 000 deaths (WHO/IVB database at http://www.who.int/
immunization_monitoring/burden/estimates_burden/en/index.html).
It is clear that immunization programmes with high coverage have
significantly reduced mortality and morbidity from the disease in many
countries. However, despite its efficient prevention of clinical disease, the
vaccine appears to have had limited impact on the circulation of B. pertussis
even in countries with high vaccination coverage. The impact of vaccines on
the circulating strains is not fully understood. In addition, during the 1990s,
a significant epidemiological shift towards higher incidences of pertussis
among schoolchildren previously vaccinated, adolescents and adults has been
observed in many industrialized countries (4). This led to the consideration
of a potential need for immunization of adolescents and adults to improve
current control of whooping cough.
The optimal immunization schedule and the appropriate time for booster
dose of DTwP vaccine should be assessed in individual national programmes
taking into account the current epidemiological situation (4). Careful
epidemiological surveillance of pertussis is encouraged worldwide to monitor
disease burden and the impact of vaccination and particularly to compare
different products and vaccination schedules.
Also, a shift in the antigenic properties of B. pertussis strains in circulation
has been reported (5–7) and the continued monitoring of its potential
impact on the overall immunity of a population is crucial in controlling
the disease. Therefore, monitoring of genetic and antigenic characteristics
of the pathogen in the context of the appropriateness of the strains of B.
pertussis used in the production of both whole cell and acellular pertussis
vaccines is encouraged.
Whole-cell pertussis vaccines have been used worldwide as part of
combined DTP vaccine in national childhood immunization programmes for
decades. Although concerns about possible adverse events following their
administration have led to the adoption of acellular pertussis vaccines in some
countries, whole-cell pertussis vaccines are still widely produced and used
globally in both developed and developing countries. Whole-cell pertussis
vaccines that comply with WHO requirements, administered according to
an optimal schedule have a long and successful record in the control of
whooping cough. Furthermore, the excellent efficacy of some currently
available whole-cell pertussis vaccine has also been shown, not only in
recent clinical trials, but also on the basis of the resurgence of disease where
vaccination has been interrupted or when coverage has markedly decreased

303
 (8). Therefore, WHO continues to recommend whole-cell pertussis vaccines
for use in national immunization programmes. Further details are available
in a WHO position paper on pertussis vaccines (4).
In terms of severe adverse events, acellular pertussis and whole-cell pertussis
vaccines appear to have acceptable safety, whereas mild to moderate adverse
reactions are more commonly associated with the whole-cell pertussis
vaccine. The latter is not recommended for use in adolescents and adults. So
far, no clinically significant immunological interference has been documented
between whole-cell pertussis vaccines and other antigens when they are
offered in a combination formulation, or with other vaccines simultaneously
administered at different injection sites. This is in contrast to the reduced
antibody levels to Hib vaccine that have been observed when given in
combination with some acellular pertussis vaccines (9).
Recent developments in the production, standardization and quality control
of pertussis vaccines were reviewed by a Center for Biologics Evaluation
and Research (CBER)/WHO working group on pertussis vaccines in
November 2000 and at a WHO consultation in July 2003. However, the
scientific basis for the present revision of the requirements for whole-
cell pertussis vaccines was developed at a WHO consultation of national
regulatory authorities, vaccine manufacturers and other experts, in March
2005. Key areas covered included vaccine composition, potency evaluation
and toxicity testing.
Considerable progress has been made in understanding the nature of some
of the agglutinogens of B. pertussis (10) . These are surface proteins which,
on infection, elicit the production of antibodies that cause the agglutination
of the organism in vitro. Some have been identified as fimbriae. The
presence of fimbriae 2 and 3, formerly identified as agglutinogens 2 and 3,
in whole-cell pertussis vaccines is believed to contribute to their protective
efficacy, and a test has been included in these revised Recommendations
for the purpose of determining whether fimbriae 2 and 3 are present, before
adjuvant is added.
The evidence that vaccines shown to protect mice against intracerebral
challenge also protected immunized children against whooping cough
when such children were exposed to the disease in the home by infection
from a sibling was published in the 1950s. This correlation was the basis
for the establishment of the current potency test (11). Although the potency
test has a long record of use, it has often been criticized, especially on its
reproducibility. However, a recent WHO proficiency study involving 13
laboratories in 12 countries confirmed that the intracerebral challenge assay
was effective in distinguishing potent and sub-potent batches of vaccine
and gave consistent results both between repeat tests and between different
laboratories (12).

304
Nevertheless, the mouse protection test is technically demanding and efforts
have been made to develop alternative in vitro potency assays, such as
serological assays. However, the lack of understanding of the mechanisms
of protection in humans afforded by whole-cell pertussis vaccines, in
particular, of the value of neutralizing antibodies, the nature of the critical
antigens and the role of cell-mediated immunity, it is difficult to design an
acceptable alternative. The serological approach was extensively discussed
at an European Directorate for the Quality of Medicines/European Centre
for the Validation of Alternative Methods (EDQM/ECVAM) consultation in
2005 (13) where the issue of the relevance of simple antibody measurements
to human clinical protection was considered. It was concluded that such tests
cannot yet be considered as validated alternatives to the mouse protection
potency test for whole-cell pertussis vaccines. However, correlation between
production of agglutinins in mice and protection in children demonstrated
as early as the Medical Research Council (MRC) trials in the 1950s
should be further explored as a potential alternative or a complementary
test to the currently recommended potency test. There was also a strong
recommendation from the EDQM /ECVAM consultation to use validated
humane end-points in the mouse protection test.
The use of the WHO Opacity Standard has also been much discussed.
Comments from many manufacturers and discussion at the WHO Consultation
in 2005 indicated that the estimation of the number of bacteria using the opacity
of the bacterial suspension prior to inactivation is still a valuable parameter
in the in-process control of whole-cell pertussis vaccines. Manufacturers
are encouraged to continue to express opacity in International Units and to
specify the range of values for their own vaccine product.
The role of different toxins, such as pertussis toxin, heat labile
(dermonecrotic) toxin, tracheal cytotoxin, adenylate cyclase toxin and
endotoxin in immunity to the natural infection or in immunization is not
fully understood. A potential link between the presence of some of these
toxins and reactogenicity in humans has been reported, but the mechanisms
of their action and the contribution of individual toxins to overall toxicity
remains unclear. Nevertheless, the determination of residual toxic activity
remains an important aspect of the safety assessment. Residual levels of
active pertussis toxin and endotoxin are likely to be a major contributor to
the reactogenicity of whole-cell pertussis vaccines in humans and limits have
been established for active pertussis toxin in acellular pertussis vaccines.
The First International Standard for pertussis toxin has been established
and various methods for the determination of residual levels of this toxin in
vaccine preparations have been developed. At present, there is no scientific
basis for setting specifications for pertussis toxin and endotoxin in whole-cell
pertussis vaccine preparations, but monitoring their levels for consistency
during production is encouraged.

305
 In recent years, safety concerns have been raised over the use of thiomersal
in vaccines, especially those given to infants. These concerns have been
based primarily on data regarding the toxicity of a related substance,
methyl mercury, and from data on chronic exposure to mercury via the
food chain. Such safety concerns have led to initiatives in some countries
to eliminate, reduce or replace thiomersal in vaccines, both in single dose
and multidose presentations. It is important to note that the concerns
about the toxicity of thiomersal are theoretical and there is no compelling
scientific evidence of a safety problem with its use in vaccines, although
a public perception of risk remains in some countries. WHO policy is
clear on this issue, and the Organization continues to recommend the use
of vaccines containing thiomersal for global immunization programmes
because the benefits of using such products far outweigh any theoretical
risk of toxicity (14). In the case of whole-cell pertussis vaccines,
thiomersal has been used in the production process as an inactivating
agent as well as a preservative. Potential changes in its content, following
licensing, may affect quality, safety and efficacy of the vaccine. In the
event of any change, WHO Guidelines on regulatory expectations related
to the elimination, reduction or replacement of thiomersal in vaccines
(15) should be followed.

Changes made and issues addressed

The main changes made, and issues addressed, in the present revision are
as follows:
• Final vaccine bulk should be examined to ensure it contains predominantly
phase I bacteria that display fimbriae 2 and 3. Strains of B. pertussis used in
production should be well characterized emphasizing phase I organisms.
Markers for phase I strains (e.g. haemolytic activity) are suggested in
small print.
• The reference reagents currently available are listed and include reagents
for the determination of fimbriae 2 and 3.
• Determination of bacterial concentration is considered to be an important
in-process control test and the International Reference Preparation of
Opacity is still considered to be a valuable tool.
• The recommendation to use the mouse weight gain test to assess specific
toxicity of vaccine preparations has been upgraded to large print whereas
the details of the methodology and refined methodology are displayed in
small print.
• During monitoring of detoxification processes, as well as when validating
methods used for detoxification and establishing consistency of production,
manufacturers are encouraged to monitor levels of pertussis toxin and
endotoxin. International Standards for pertussis toxin and endotoxin are
available and results should be expressed in IU.

306
 • The estimation of potency has been upgraded from small to large print
clarifying that it should not be less than 4.0 IU per single human dose with
a lower fiducial limit of the estimated potency being not less that 2.0 IU.
• Manufacturers and control laboratories are encouraged to use validated
humane end-points in recording results of potency testing.
• A new section on the stability evaluation of vaccines has been included
which emphasizes the importance of real-time studies under intended
storage conditions and discusses the extent of stability studies needed for
different purposes and at different stages of manufacturing.
• Specific issues for nonclinical and clinical evaluation of new pertussis
vaccines as well as a need for the improvements in postmarketing
surveillance are also discussed in separate sections of this document.

Part A. Manufacturing recommendations

A.1 Definitions
A.1.1 International name and proper name
The international name should be whole-cell pertussis vaccine. The proper
name should be the equivalent of the international name in the language of
the country of origin.
The use of the international name should be limited to vaccines that satisfy
the recommendations formulated below.

A.1.2 Descriptive definition

Whole-cell pertussis vaccine is a suspension of the whole cells of one or
more strains of killed Bordetella pertussis which have been appropriately
treated to minimize toxicity and retain potency. The preparations for human
use should satisfy all the recommendations formulated below.

A.1.3 International reference materials

The WHO catalogue of international biological standards should be
consulted for the latest list of appropriate international standards and
reference materials (http://www.who.int/biologicals/IBRP/Catalogue.htm).
The third International Standard for Pertussis Vaccine was established in
1998 with a potency of 46 IU of pertussis vaccine per ampoule.
The fifth International Reference Preparation of Opacity was established in
1975 with an opacity of 10 International Units. It consists of plastic rods
simulating the optical properties of a bacterial suspension.
The WHO reference reagents of monoclonal antibodies for B. pertussis
anti- fimbriae serotype 2 and 3 were established in 2004. They are intended
for the determination of serotype of B. pertussis strains.

307
 The First International Standard for Pertussis Toxin was established in 2003
with an activity of 10 000 IU per ampoule. It is intended for the determination
of residual pertussis toxin in pertussis vaccine.
The above-mentioned International Standards/reference materials and
other reagents from the WHO Pertussis reagent Bank are in the custody
of the National Institute for Biological Standards and Control, Potters Bar,
Hertfordshire, EN6 3QG, England (web site: http://www.nibsc.ac.uk).
These reference preparations are available for calibration and establishment
of regional, national or in-house reference materials. Samples are distributed
free of charge, on request, to national control laboratories.

A.1.4 Terminology
The following definitions are given for the purpose of these recommendations
only.
Seed lot. A quantity of bacterial suspension that is derived from one strain,
has been processed as a single lot and has a uniform composition. It is used
for preparing the inoculum for the production medium.
Single harvest. A suspension of bacteria prepared from cultures of one
strain of B. pertussis inoculated, harvested and processed together.
Final bulk. The homogeneous finished vaccine from which the final
containers are filled either directly or through one or more intermediate
containers.
Final lot. A collection of sealed final containers that are homogeneous
with respect to the risk of contamination during filling. A final lot must
therefore have been filled from a single container in one continuous working
session.

A.2 General manufacturing recommendations

The general manufacturing recommendations contained in good
manufacturing practices for pharmaceuticals (16) and biological products
(17) should apply to establishments manufacturing whole-cell pertussis
vaccine.

A.3 Production control

A.3.1 Control of source materials
A.3.1.1 Strains of Bordetella pertussis
Strains of B. pertussis used in preparing vaccines should be identified by a
full record of their history, including their origin, characteristics on isolation,
and particulars of all tests made periodically to verify strain characteristics.

308
 The strains should be well characterized and chosen in such a way that the
final vaccine contains predominantly phase I cells that display fimbriae 2 and
3. They should have been shown to the satisfaction of the national regulatory
authority, to yield safe and immunogenic vaccines when inactivated.
The reference preparations of antibodies for detection of fimbriae 2 and 3
should be used.
Since haemolytic activity has been suggested as a marker for phase I
cells, colonies of B. pertussis can be examined for this characteristic on
a suitable solid medium containing blood. Alternatively, some culture
media (e.g. meat extract agar plates) support the growth of phase III/IV
isolates of B. pertussis, but not that of phase I bacteria, and these media
can be used to confirm phase I status of cultures. When culture methods
for phase I assessment are used, the media composition, blood type
and concentration, and incubation time need to be properly defined.
The strains should be maintained by a method that will preserve their ability
to yield potent vaccine.
Freeze-drying or storage in liquid nitrogen is a satisfactory method of
maintaining strains.

A.3.1.2 Seed lot system

The production of pertussis vaccine should be based on a seed lot system.
Cultures of the working seed should have the same characteristics as those
of the strain from which the parent seed lot was derived.

A.3.1.3 Culture media for production of bacteria

The media chosen for growing B. pertussis should be carefully selected and
enable the organism to grow well and to retain phase I characteristics. Given
that different media have an impact on the quality of the vaccine, every
effort should be made to use media proved as a substrate for manufacturing
a vaccine that consistently meets the potency requirements. Once the media
have been demonstrated as appropriate they should be consistently used.
Every change of media should be validated and the national regulatory
authority notified.
The acceptability of the source(s) of any components of bovine, sheep
or goat origin used in culture media should be approved by the national
regulatory authority. B. pertussis should be grown in media free from
substances likely to cause toxic or allergic reactions in humans. If any
materials of animal origin are used in seed preparation or preservation, or
in production, they should comply with the guidelines on medicinal and
other products in relation to human and animal transmissible spongiform
encephalopathies (18). When animal blood or blood products are used, they
should be removed by washing the harvested bacteria.

309
 In some countries, the use in the medium of blood from any source
is not permitted. Manufacturers are encouraged to explore the use of
media derived from non-animal sources.
Human blood or blood products should not be used in culture media for
propagating bacteria, either for seed or for vaccine.

A.3.2 Control of single harvests

A.3.2.1 Monitoring consistency of production
Consistency of production should be demonstrated. Parameters to be measured
include, but are not limited to, bacterial growth rate and some characteristics
of phase I organisms in the culture, such as haemolytic activity and presence
of fimbriae 2 and 3.
Criteria for acceptance or rejection of harvests should be defined.

A.3.2.2 Control of bacterial purity

Samples of single harvests taken before killing should be tested for purity
by microscopic examination of stained smears or by inoculation into
appropriate culture media. Single harvests should not be used for the final
bulk if contamination has occurred at any stage in their production.

A.3.2.3 Control of opacity

The opacity of each single harvest should be measured not later than 2 weeks
after harvesting and before the bacterial suspension has been subjected to any
process capable of altering its opacity. It should be measured by comparison
with the International Reference Preparation of Opacity or an equivalent
reference preparation approved by the national regulatory authority. The
opacity of bacterial suspensions should be expressed in International Units
and specifications set for each vaccine.
A bacterial suspension having the same opacity as the International Reference
Preparation of Opacity has a bacterial concentration providing 10 IU of
opacity. The relationship between such units and actual numbers of bacterial
cells may vary from vaccine to vaccine.
A spectrophotometric method validated against the opacity reference may
also be used for this purpose.

A.3.2.4 Killing and detoxification

After samples of single harvests have been taken for purposes of purity
control and opacity measurement, the bacteria shall be killed and detoxified
by a method approved by the national regulatory authority. To ensure that
the organisms have been killed, a sample should be tested in an appropriate
culture medium.

310
 B. pertussis can be killed by a number of methods whose effectiveness
depends on the concentration of the chemicals used and the temperature,
time and pH at which killing is carried out. The aim is twofold: to kill
all bacterial cells and to achieve an appropriate level of detoxification
without adversely affecting the potency or the physical characteristics
of the vaccine. The methods used and kinetics of inactivation should be
validated to the satisfaction of, and approved by, the national regulatory
authority.
After killing and detoxification, the opacity of the suspension will be
different from what it was originally. Each single harvest should, however,
still be regarded as containing the same number of bacteria.
No biologically active heat-labile toxin (dermonecrotic toxin) should
be detectable in a vaccine. The method of manufacture should be
validated to ensure that no active dermonecrotic toxin is present in
the final product. The method of detoxification should ensure vaccine
safety. At present, it is not possible to recommend limits for levels of
pertussis toxin, endotoxin, tracheal cytotoxin and adenylate cyclase
in whole-cell pertussis vaccines. Manufacturers are encouraged to
appropriately validate tests for these factors, and to ensure consistency
of production.

A.3.3 Control of final bulk

A.3.3.1 Preparation
The final bulk may consist of a single harvest or a pool of single harvests
from one or more strains. If a vaccine is prepared from two or more strains,
the proportion of each strain in the pool, as calculated in opacity units,
should remain consistent for each batch of the final bulk. The single harvest
or pool should be diluted such that the number of bacteria in a single human
dose of the final bulk is equivalent to the number of bacteria in the same
volume of a suspension showing an opacity of no more than 20 IU. The
opacity measured on the single harvests (before killing, see part A, section
A.3.2.3) should be used to calculate the bacterial concentration in the final
bulk.

A.3.3.2 Fimbriae
Each bulk should be examined, before adjuvant is added, for the presence of
fimbriae 2 and 3 to ensure that appropriate expression has occurred during
bacterial growth.

A.3.3.3 Preservative
If the vaccine is to be dispensed into multidose containers, a suitable
antimicrobial preservative should be added. Consideration should be given
to the effect of the preservative on stability of the vaccine formulation and

311
 possible interactions between the vaccine components and the preservative.
If a preservative has been added to the vaccine, the content of preservative
should be determined by a method approved by the national regulatory
authority. The amount of preservative in the vaccine dose should be shown
not to have any deleterious effect on the antigen nor impair the safety of
the product in humans. The preservative, its use at different stages of the
manufacturing process as well as its residual amount should be approved by
the national regulatory authority.
If any modification of the preservative content in an already licensed vaccine
is made, general principles for vaccine evaluation described in the WHO
Guidelines on regulatory expectations related to the elimination, reduction
or replacement of thiomersal in vaccines, should be followed (15).
Phenol should not be used as a preservative.

A.3.3.4 Adjuvants
If an adjuvant has been added to the vaccine, its nature, purity and
concentration should be determined by a method approved by the national
regulatory authority.
Either aluminium or calcium compounds may be used as mineral
carriers.
Where aluminium compounds are used as adjuvants the concentration of
aluminium should not exceed 1.25 mg. When calcium adjuvants are used,
the concentration of calcium should not exceed 1.3 mg per single human
dose.
In some countries, an upper limit of 1.25 mg of aluminium is considered
to be excessive for products containing a pertussis component and
such vaccines therefore contain only 0.1–0.3 mg of aluminium per
single human dose.
If other substances have been used as adjuvants or those with adjuvanted
effect, specifications should be set and agreed by the national regulatory
authority.
The formulation should be such that the homogeneous suspension appear
after shaking and remains as such for a specified time (e.g. time needed for
vaccine administration).

A.3.3.5 Sterility
Each final bulk shall be tested for bacterial and fungal sterility in
accordance with the requirements given in Part A, section 5, of the revised
Requirements for Biological Substances No. 6 (General Requirements for
the Sterility of Biological Substances) (19) or by a method approved by
the national regulatory authority. If a preservative has been added to the

312
 vaccine, appropriate measures should be taken to prevent it from interfering
with the test.

A.3.3.6 Specific toxicity

Each final bulk should be tested for toxicity using the mouse weight gain
test. The final bulk is considered satisfactory if the following conditions are
met:
(a) at the end of 72 hours the average weight of the group of vaccinated
mice is not less than that preceding the injection,
(b) at the end of 7 days the average weight gain per mouse is not less than
60% of that per control mouse, and
(c) no deaths occur when 10 mice are used and no more than one death
occurs when 20 mice are used.
A satisfactory method of carrying out the assay is as follows: at least
10 healthy mice each weighing 14–16 g are used for each vaccine group
and for the saline control group. Mice should be of the same sex or
segregated males and females should be distributed equally between
all groups. Mice should have access to food and water for at least
2 h before injection, and continuously after injection for the duration of
the test. The total weight of each group of mice should be measured
immediately before injection. Each mouse is given an intraperitoneal
injection of 0.5 ml of 0.85% NaCl aqueous solutions containing half of
the recommended single human dose. The mice in the control group
are inoculated with 0.5 ml of physiological saline, preferably containing
the same amount of preservative as the inoculum injected into the test
mice. The total weight of each group of mice is measured or calculated
at 72 h and again at 7 days after injection.
If vaccine fails to meet the requirements in a first test, it can be retested
once, and the results of the two valid tests should be combined.
If the average weight gain per mouse in the vaccine group is greater
than 150% of that per control mouse, ascites production should be
suspected and the test should be considered invalid.
Manufacturers are encouraged to develop refinements and alternatives
to the mouse weight gain test.
In some countries a refinement of the mouse weight gain test is used.
Mice are weighed individually immediately before injection, and
16–24 h, 72 h and 7 days after injection. On day 7 blood samples are
taken from the tail vein and leukocytes are counted. Weight change at
16–24 h is considered to reflect the presence of lipo-oligosaccharide
and an increase in the leukocyte count is considered to reflect the
presence of pertussis toxin in the vaccine.

313
 Other tests:
Cell harvests of B. pertussis to be used in the manufacture of pertussis
vaccine contain a number of biologically active molecules which may
contribute to the toxicity of the final product. Assays for some of these
substances can be used to monitor and validate the methods used for
detoxification and may also be useful in assessing final products. In the
process of validating the manufacturing procedures, manufacturers are
encouraged to monitor the following:
Pertussis toxin. A Chinese hamster ovary cell (CHO-cell) assay, based
on the clustering of cells after treatment with pertussis toxin is used
in some countries to measure pertussis toxin in vaccine. A pertussis
toxin standard is included in the assay, and a vaccine reference is used
as a positive control. All samples are serially diluted to obtain an end-
point and the concentration of the pertussis toxin in the test sample
is calculated in relation to the toxin reference. Tests for histamine
sensitizing activity in mice may also be used.
Endotoxin. B. pertussis is a Gram negative organism, thus whole-
cell pertussis vaccines contain lipo-oligosaccharide endotoxin.
The endotoxin content of vaccines can be estimated by the limulus
amoebocyte lysate assay or the rabbit pyrogen test. The limulus
amoebocyte lysate assay is preferred. Although there is no agreement
as to what constitutes an acceptable level of endotoxin in whole-cell
pertussis vaccines, monitoring of endotoxin level on a lot-to-lot basis is
encouraged as a monitor of consistency of production.

A.3.3.7 Potency
The potency of each final bulk (or of each final lot) should be determined
by comparison with that of a reference vaccine calibrated against the
International Standard for Pertussis Vaccine or an equivalent standard
vaccine approved by the national regulatory authority. The assay should be
performed by the intracerebral mouse protection test. The assay method
and the method of calculating the results should be approved by the
national regulatory authority. The potency is estimated in terms of IU in
the volume recommended for a single human dose. The vaccine passes
the recommendations for potency if the result of a statistically valid test
shows that the estimated potency of the vaccine is not less than 4.0 IU in the
volume recommended for a single human dose and if the lower fiducial limit
(P = 0.95) of the estimated potency is not less than 2.0 IU. Additional tests
may be done, but in this case the results of all valid tests must be combined
in the weighted geometric mean estimate and its lower fiducial limit.
In some countries, an upper limit of potency is also specified.
A satisfactory method of carrying out the assay is as follows:

314
Mice. Healthy mice from a strain and colony capable of giving an
adequate immune response are used. They should preferably be of the
same sex but, if this is not possible, both sexes should be distributed
equally throughout the test and the sexes segregated. Mice should be
consistent for age and weight. An example of a criterion for consistency
which has been used is that mice should weigh at least 10 g and not
more than 18 g and in a single test the weight of the mice should not
differ by more than 4 g.
The mice are randomly allocated to the different groups, and the shelf
position of the cages, the order of immunization, and the order of challenge
are also randomized. Groups of at least 16 mice should be used for each
dilution of the standard vaccine and of the vaccines under test, and at least
10 mice should be used for each dilution of the culture in the estimation of
the number of median lethal doses (LD50) in the challenge dose.
Immunization of mice. At least three dilutions of the reference vaccine
and of each lot of vaccine should be tested. Serial dilutions, not greater
than fivefold, of the vaccine to be tested and of the standard vaccine
should be made in a suitable diluent. The median effective dose (ED50)
for each preparation should be tested by the dilutions used. Each
mouse in each immunization group should be injected intraperitoneally
with 0.5 ml of the appropriate dilution.
The interval between immunization and challenge should be 14–17 days.
At least 94% of the mice immunized by each dilution of both the reference
vaccine and the test vaccines should survive until challenged, and each
mouse challenged should appear healthy prior to challenge.
The challenge. The strain used for challenge (generally B. pertussis
18323) should be approved by the national regulatory authority. To ensure
consistency of virulence from test to test, a large working challenge
lot prepared from the master culture is dispensed into ampoules and
freeze-dried or stored in liquid nitrogen.
The bacterial suspension used for challenge is prepared from a
20–24 h culture grown on Bordet–Gengou medium, or other suitable
medium that has been seeded from a rapidly growing culture not more
than 30 h old. Alternatively, aliquots of the challenge suspension may
be frozen and kept in liquid nitrogen; after thawing and dilution, they
can be used directly as the challenge culture. The suspension is diluted
with a diluent in which the organisms will remain viable, e.g. an aqueous
solution containing 10 g/l casein peptone and 6 g/l sodium chloride
adjusted to a pH of 7.1 ± 0.1. The suspension, free from particles of agar
or clumps of bacteria, is adjusted in such a way that each challenge
dose of not more than 0.03 ml contains 100–1000 times the LD50.

315
 Mice immunized with the reference vaccine and the test vaccines are
challenged at random under mild narcosis by intracerebral injection of
the challenge dose. To obtain an estimate of the LD50, dilutions of the
challenge dose are then injected into control mice by the intracerebral
route and an appropriate dilution of the challenge dose is cultured on
Bordet–Gengou medium to determine the number of colony-forming
units contained therein.
Recording of results. The mice are observed for 14 days. Mice that die
within 72 hours should be excluded from the test. To determine the ED50
of the vaccines, records should be kept of the number of mice that die
after 72 hours. Animal welfare regulations should be followed.
The use of validated humane end-points is encouraged.
Calculation of results. The ED50 values for each preparation are
determined by a statistical method that includes the transformation of
the mouse survival data into a form capable of consistently producing
a linear regression. Probits, logits and angle transformation have been
shown to be suitable. Similar methods should be used to determine the
LD50 of the challenge suspension.
Validity of the test. The test is valid if the ED50 of each vaccine is
intermediate between the largest and the smallest immunizing
doses, and the regressions do not show significant deviation from
linearity and parallelism (P < 0.05). The challenge dose should contain
100–1000 LD50 and the LD50 should contain no more than 300 colony-
forming units.
Estimate of potency. The ED50 of the vaccine under test and the
standard vaccine are calculated by a method that provides an
estimate of the limits of the 95% confidence intervals. The potency
is estimated in terms of IU in the volume recommended for a single
human dose.

A.3.3.8 pH
The pH of each final bulk should be measured and specifications set.
In some countries this test is applied to the final filled vaccine (A 5.7).

A.4 Filling and containers

The requirements concerning filling and containers given in Good
manufacturing practices for biological products (17) should apply to
vaccine filled in the final form.
Single-dose or multiple-dose containers may be used. Vaccine in multidose
containers should contain a suitable antimicrobial preservative.

316
A.5 Control of pertussis component in final lot
A.5.1 Identity
An identity test should be performed on at least one container from each
final lot.
The identity test may be based on an immunological reaction (for example,
agglutination of the organisms) with a specific antipertussis serum.
Alternatively, vaccines may also be inoculated into animals to show that
pertussis-specific antibodies (e.g. agglutinins) are present in their serum.

A.5.2 Sterility
Final containers should be tested for sterility by a method approved by the
national regulatory authority.
Many countries have regulations governing the sterility testing of the
final product. Where these do not exist, the requirements published by
WHO should be met (19). If a preservative has been added to the vaccine,
appropriate measures should be taken to prevent it from interfering with the
sterility test.

A.5.3 Potency
A potency test should be carried out as provided in Part A, section A.3.3.7,
on each final lot, if such a test has not been done on the final bulk.

A.5.4 General safety (innocuity) test

Each final lot should be tested for unexpected toxicity (sometimes called
abnormal toxicity) using a general safety (innocuity) test approved by the
national regulatory authority.
This test may be omitted for routine lot release once consistency of
production has been well established to the satisfaction of the national
regulatory authority and when good manufacturing practices are in
place. Each lot, if tested, should pass a test for general safety.

A.5.5 Adjuvant content

If an adjuvant has been added to the vaccine, its content should be determined
by a method approved by the national regulatory authority.
When aluminium compounds are used as adjuvants, the concentration of
aluminium should not exceed 1.25 mg per single human dose. If a calcium
adjuvant is used, the concentration of calcium should not exceed 1.3 mg per
single human dose.
If other substances were used as adjuvants, appropriate specifications should
be set for the substance with adjuvant effect.

317
A.5.6 Preservative content
If a preservative has been added to the vaccine, the content of preservative
should be determined by a method approved by the national regulatory
authority.
The amount of preservative in the vaccine dose should be shown not to have
any deleterious effect on the antigen or to impair the safety of the product
in humans. The preservative, its use at different stages of the manufacturing
process as well as its residual amount should be approved by the national
regulatory authority.
If any modification of thiomersal content in an already licensed vaccine
is made, general principles for vaccine evaluation described in the WHO
Guidelines on regulatory expectations related to the elimination, reduction
or replacement of thiomersal in vaccines, should be followed (15).

A.5.7 pH
The pH of each final lot should be measured and specifications set.
In some countries this test is applied to the final bulk only (A 3.3.8).

A.5.8 Inspection of final containers

Each container in each final lot should be inspected visually, and those
showing abnormalities — such as improper sealing, lack of integrity,
clumping or the presence of particles — should be discarded.

A.6 Records
The recommendations given in Good Manufacturing Practices for biological
products (17) should apply.
A model of a suitable summary protocol to be used for pertussis
vaccines is given in the Appendix.

A.7 Retained samples

The recommendations given in Good Manufacturing Practices for biological
products (17) (Annex 1) should apply.

A. 8 Labelling
The recommendations given in Good Manufacturing Practices for biological
products (17) should apply, with the addition of the following.
— the words whole-cell pertussis vaccine;
— the word “adsorbed”, if applicable;
— the name and address of the manufacturer;

318
 — the recommended storage temperature and the expiry date if kept at that
temperature; and
— the recommended single human dose and route of administration.
In addition, the label printed on or affixed to the container, or the label
on the carton, or the leaflet accompanying the container shall contain the
following:
— a statement that the vaccine satisfies the requirements of this document;
— the nature and amount of any preservative present in the vaccine
(if there is no preservative in single-dose containers, this should be
stated);
— the nature and amount of the adsorbing agent, if applicable;
— the nature and amount of any substances added to the vaccine;
— the recommended conditions for storage and transport;
— a warning that the vaccine should not be frozen;
— a warning that the vaccine should be shaken before use; and
— instructions for the use of the vaccine and information on contraindications
and the reactions that may follow vaccination.

A.9 Distribution and transport

The recommendations given in Good Manufacturing Practices for biological
products (17) should apply.

A.10 Stability, storage and expiry date

A.10.1 Stability
Stability evaluation is an important part of the quality assessment. The
purpose of stability studies is to ensure that the vaccine at the end of its shelf
life, storage period or period of use, still has the required characteristics
supporting quality, safety and efficacy.

For licensing
Studies that support stability of a vaccine for the purpose of licensing have
to be performed as real-time studies under the intended storage conditions.
Stability-indicating parameters should be carefully selected. They should
always include, but should not be limited to, the potency test. Tests should
be conducted to determine the loss of potency at appropriate time intervals
during storage. Final containers from at least three batches of vaccine derived
from different bulks should be tested on the expiry date to demonstrate
stability during storage.
Accelerated stability data for product stored for limited periods at
temperatures that may affect stability could support preliminary data

319
 from ongoing real time stability studies but should not replace them. Any
modification of the shelf life approved as part of licensing requires additional
stability data to support the proposed modification and should be approved
by the national regulatory authority. Following licensure, stability should be
monitored throughout the proposed shelf-life.

At different stages of manufacturing process

Stability testing should be performed at different stages of production,
namely single harvests, final bulk and final lot. Stability indicating parameters
should be selected according to the stage of production. Manufacturers are
encouraged to assign a shelf-life to all materials during vaccine production,
in particular to intermediates such as single harvests, purified bulk and final
bulk.

For clinical trial approval

For vaccines under development, stability data, such as those described
above, are expected for the purpose of clinical trial approval. However,
the stability data for such vaccines are generally available for a limited
period.
Appropriate documentation to support the stability profile of a vaccine
should be submitted to the competent national regulatory authority at all
stages mentioned above.

A.10.2 Storage conditions

Recommended storage conditions and defined maximum duration of storage
should be based on stability studies as described in section 10.1 above and
approved by the national regulatory authority. For pertussis vaccines, a
temperature of 2–8 °C has been found satisfactory. This should ensure that
the minimum potency specified on the label of the container or package
will still be maintained after release until the end of the shelf-life, if the
conditions under which the vaccine is stored are in accordance with what is
stated on the label.
The manufacturer should recommend conditions of storage and transport
that will ensure that the vaccine satisfies the potency requirements until the
expiry date stated on the label.
The vaccine must not be frozen.

A.10.3 Expiry date

The expiry date should be defined on the basis of the shelf-life supported by
the stability studies as described above (section 10.1) and approved by the
national regulatory authority.

320
Part B. Nonclinical evaluation of whole-cell
pertussis vaccines
Nonclinical evaluation of new pertussis vaccines
For a new whole-cell pertussis vaccine, a new formulation, or for a vaccine
produced by a manufacturer with no previous experience of such vaccines and
which has not been previously tested in humans, proof of concept in a relevant
animal model in terms of both potency and safety should be demonstrated.
In addition, a safety assessment should be undertaken before initiation of
the clinical evaluation of a new vaccine. General principles for the design,
conduct, analysis and evaluation of nonclinical data are available in the WHO
guidelines for nonclinical evaluation of vaccines (20). In particular, studies
on safety pharmacology intended to investigate the effects of a vaccine on
vital functions should be undertaken.

Part C. Clinical evaluation of whole-cell

pertussis vaccines
New whole-cell pertussis vaccines, vaccines with a new formulation, or
those intended to use a new route of administration and /or produced by
a manufacturer with no previous experience with such vaccines should
undergo clinical evaluation. This section is intended to indicate some of
the specific issues which need to be considered in the clinical testing of
such vaccines, as monocomponent vaccines or as a part of a combination.
Issues to be considered in designing clinical studies for licensing as well
as those for monitoring clinical effectiveness and safety in postmarketing
surveillance studies are discussed.
In general, clinical trials should adhere to the principles described in good
clinical practice (21) as well as to those formulated for the design, conduct
and analysis of vaccine clinical trials described in the WHO guidelines
for clinical evaluation of vaccines (22). Data generated in clinical trials
should be submitted to the national regulatory authority as described in the
Summary protocol for vaccine evaluation (22). All clinical trials should be
approved by the relevant national regulatory authority.
However, there are issues which apply specifically to pertussis clinical trials
and these should be considered in addition to the general principles mentioned
above. First, prospective randomized controlled studies of protective efficacy
(i.e. testing against a placebo) cannot be performed for ethical reasons. Second,
trials designed to measure efficacy relative to that of a licensed whole-cell
pertussis vaccine with proven efficacy, would need to be very large in order
to provide adequate precision in the efficacy estimates.

321
 An additional complexity is that many different antigens are expressed
by B. pertussis and there are many different assays that might be used for
the assessment of the immunogenicity. However, without any established
immunological correlate(s) of protection the data cannot be used to predict
efficacy.

C.1 Clinical evaluation of new whole-cell pertussis vaccines

for licensing
C.1.1 Compliance with the recommendations for production and control
Candidate vaccine should comply with the recommendations for production
and control described in part A of this document.

C.1.2 Immunogenicity and safety assessment in humans

C.1.2.1 A comparability study using a “new” wP preparation which meets
these requirements for potency and safety, and a wP-containing vaccine that
has been licensed for some years and used extensively in countries with
reliable postmarketing safety surveillance schemes may be an appropriate
approach.
C.1.2.2 The immune response in clinical trials should be assessed by using
a small range of validated assays. Selection of the assays for evaluation
of the immune response to the vaccine should be justified by the vaccine
developer; when feasible, assays that measure functional immune responses
should be employed. The assays used are unlikely to be commercially
available, and thus validation issues must be addressed.
C.1.2.3 For each assay used, the immunogenicity data obtained should
be compared both in terms of percentage of vaccinees who demonstrate a
response (e.g. the percentage who reach a specified threshold or achieve
a significant increase in antibody concentration) and in geometric mean
concentrations (GMCs).
C.1.2.4 Every effort should be made to determine antibody response to
individual, specific antigens rather than relying solely on the measurement
of antibodies against whole cells or whole-cell extracts. Because of the
historical link to clinical efficacy, the measurement of whole-cell agglutinins
is recommended. Additionally, at least one assay used should determine
antibodies against pertussis toxin.
C.1.2.5 The size of such a study and the end-points for evaluation require
justification. The immunogenicity end-points need to be set according to
the sensitivity and specificity of the assays and in the light of experience
regarding natural variation between individuals.
C.1.2.6. In the case of combination with other antigens, potential
interactions between the whole cell pertussis component and the others

322
 should be investigated as described in the WHO guidelines for clinical
evaluation of vaccines (22).
C.1.2.7 Safety assessment should be part of the comparability study
mentioned above with defined objectives of the study. The study should
have sufficient power to provide reliable rates of frequent or very common
adverse events (22).
C.1.2.8 The rates of specific adverse events should be formally compared:
the non-inferiority margin should be based on anticipated rates from the
trials conducted in the past.

C.2 Monitoring vaccine effectiveness and safety in

the population
Every effort should be made to improve current scientific understanding
of the protection in humans by providing data from active postmarketing
surveillance.
Vaccine effectiveness in the population should be reported wherever
possible.
Given that limited safety data are obtained in pre-licensure studies, all relevant
safety indicating parameters should be monitored as part of postmarketing
surveillance.
Data generated in postmarketing surveillance should be submitted to the
national regulatory authority.

Part D. Recommendations for National

Regulatory Authorities
D.1 General
The general recommendations for National Regulatory Authorities
contained in the Guidelines for National Authorities on Quality Assurance
for Biological Products (23) should apply.
The detailed production and control procedures and any change in them
that may affect the quality, safety or efficacy of whole-cell pertussis
vaccine should be discussed with and approved by the National Regulatory
Authority. The National Regulatory Authority should establish a national
working reference preparation calibrated against the International
Standard for Pertussis Vaccine.
Consistency of the production has been recognized as an essential component
in the quality assurance of whole-cell pertussis vaccines. In particular,

323
 National Regulatory Authority should carefully monitor results of tests
performed on a series of consecutive batches of the final bulk.

D.2 Official release and certification by the national regulatory

authority
A vaccine lot should be released only if it fulfils national requirements and/
or satisfies Part A of these Recommendations.
A statement signed by the appropriate official of the national regulatory
authority should be provided at the request of the manufacturing
establishment and should certify that the lot of vaccine in question satisfies
all national requirements as well as Part A of the present Requirements. The
certificate should state the number under which the lot was released by the
national regulatory authority, and the number appearing on the labels of the
containers. The official national release document should be provided to
importers of pertussis vaccines.
The purpose of the certificate is to facilitate exchange of pertussis vaccines
between countries. A model of a suitable certificate is given in the appendix.

Authors
The scientific basis for the revision of the Requirements published in 1990 was
discussed at the meeting of the working group held at the World Health Organization,
Geneva, in July 2003 attended by the following people: Dr J. Arciniega, Center for
Biologics Evaluation and Research, Rockville, MD, USA; Dr M. Corbel, National
Institute for Biological Standards and Control, Potters Bar, England; Dr R. Gaines-
Das, National Institute for Biological Standards and Control, Potters Bar, England;
Dr E. Griffiths, Centre for Biologics Research, Biologics and Genetic Therapies
Directorate Health Canada, Ottawa, Canada; Dr J.G. Kreeftenberg, Netherlands
Vaccine Institute (NVI), Bilthoven, The Netherlands; Dr S.S. Jadhav, Quality Assurance
and Regulatory Affairs, Serum Institute of India Ltd, Pune, India; and Dr D. Xing,
National Institute for Biological Standards and Control, Potters Bar, England.

The first draft of these Recommendations was prepared by Dr Ivana Knezevic,

Quality Assurance and Safety of Biologicals, World Health Organization, Geneva,
Switzerland and Dr Dorothy Xing, National Institute for Biological Standards and
Control, Potters Bar, England, following discussion at the National Institute for
Biological Standards and Control, in April 2004, attended by following participants:
Dr Elwyn Griffiths, Health Canada, Canada; Dr Michael Corbel; Dr Dorothy Xing
and Dr Rose Gaines-Das, National Institute for Biological Standards and Control,
Potters Bar, England.

Taking into account comments on the first draft, a second draft was prepared by
Dr Ivana Knezevic, Quality Assurance and Safety of Biologicals, World Health
Organization, Geneva, Switzerland and Dr Elwyn Griffiths, Health Canada, Canada,
in January 2005.

324
The third draft was prepared by Dr Rose Das; Dr Dorothy Xing and Dr Michael Corbel,
National Institute for Biological Standards and Control, Potters Bar, England;
Dr Yoshinobu Horiuchi, National Institute for Infectious Diseases, Japan; Dr Elwyn
Griffiths, Health Canada, Canada, and Dr Ivana Knezevic, Quality Assurance and
Safety of Biologicals, World Health Organization, Geneva, Switzerland, after an
informal WHO Consultation held in March 2005, with the following participants:
Dr J. Arciniega, Center for Biologics Evaluation and Research, Rockville, MD, USA;
Dr C.M. Ausiello, Instituto Superiore di Sanita, Rome, Italy; Dr T.A. Bektimirov, L.A.
Tarasevich State Research Institute for Standardization and Control of Medical
Biological Preparations, Moscow, Russian Federation; Dr P. Chagnaud, French
Health Products Safety Agency, Lyon, France; Dr M. Corbel, National Institute for
Biological Standards and Control, Potters Bar, England; Dr A. Dias, Oswaldo Cruz
Foundation/FIOCRUZ, Rio de Janeiro, Brazil; Dr R. Dobbelaer, Scientific Institute of
Public Health-Louis Pasteur, Brussels, Belgium; Dr R. Gaines-Das, National Institute
for Biological Standards and Control, Potters Bar, UK; Dr G Gallegos Flores, Comisión
de Control Analitico y Amplicción de Cobertura, Gerencia de Analisis y Desarrollo de
Pruebas Biológicas, Mexico; Dr M. Girard, Centre for Biologics Research, Biologics
and Genetic Therapies Directorate Health Canada, Ottawa, Canada; Dr E. Griffiths,
Centre for Biologics Research, Biologics and Genetic Therapies Directorate Health
Canada, Ottawa, Canada; Dr N. Guiso, National Centre of Reference of Pertussis,
Pasteur Institute, Paris, France; Dr S.R. Gupta, Directorate General of Health Services,
Ministry of Health and Family Welfare, Government of India, New Delhi, India; Dr Y.
Horiuchi, Dept. of Bacterial Pathogenesis and Infection Control (NIID), Tokyo, Japan,
Mrs T. Jivapaisarnpong, Ministry of Public Health, Thailand; Dr J.G. Kreeftenberg,
Netherlands Vaccine Institute (NVI), Bilthoven, The Netherlands; Dr B. Meade, Office
of Vaccines Review and Research, Center for Biologics Evaluation and Research,
Rockville, MD, USA; Dr A. M’henni, Institut Pasteur de Tunis, Ministry of Public Health,
Tunis, Tunisia; Dr P. Olin, Swedish Institute for Infectious Disease Control, Solna,
Sweden; Dr M. Powell, Medicines and Healthcare products Regulating Agency,
London, England; Dr A. Tahlan, Joint Director and Government Central Research
Institute, Kasauli, India; Dr C. von Hunolstein, Instituto Superiore di Sanità, Rome,
Italy; Dr D. Xing, National Institute for Biological Standards and Control, Potters Bar,
England; Dr S. Zhang, National Institute for the Control of Pharmaceutical and
Biological Products, State Food and Drug Administration, Beijing, People’s Republic
of China; Ms E. Molari, UNICEF Supply Division, Copenhagen, Denmark; Dr Ma
Verónica Ortega Adame, Comisión de Control Analitico y Amplicción de Cobertura,
Gerencia de Analisis y Desarrollo de Pruebas Biológicas, Mexico; Dr S.S. Jadhav,
Quality Assurance and Regulatory Affairs, Serum Institute of India Ltd, Pune, India;
Dr Y. Lingjiang, International Business and Co-operation, Chengdu Institute of
Biological Products, Chengdu, People’s Republic of China; Dr E. Ma Fajardo, Vaccine
Adviser and International Affairs, Finlay Institute, Cuba; Dr M. Qin, Bacterial Vaccine
Department, Chengdu Institute of Biological Products, Chengdu, People’s Republic
of China; Dr N. Harjee, CSL Consultant, Ontario, Canada; Dr B. t’Serstevens, GSK,
Rixensart, Belgium; Dr E. Vidor, Sanofi Pasteur, Lyon, France; Dr M-E. Behr Gross,
Strasbourg, France; Dr D. Wood, Coordinator, Quality Assurance and Safety of
Biologicals, World Health Organization, Geneva, Switzerland; Dr Carmen Rodriguez
Hernandez, Access to Technology, World Health Organization, Geneva, Switzerland;
Dr Ivana Knezevic, Quality Assurance and Safety of Biologicals, World Health

325
 Organization, Geneva, Switzerland; Dr Philippe Duclos, Vaccine Assessment and
Monitoring, World Health Organization, Geneva, Switzerland.

The first draft of the section on clinical evaluation of whole-cell pertussis vaccines
was prepared by Dr Mair Powel, Medicines and Healthcare Products Regulating
Agency, London, England, and Dr Ivana Knezevic, Quality Assurance and Safety
of Biologicals, World Health Organization, Geneva, Switzerland, following
discussion on 17 March 2005 attended by the following participants: Dr N. Guiso,
National Centre of Reference of Pertussis, Pasteur Institute, Paris, France; Dr B.
Meade, Office of Vaccines Review and Research, Center for Biologics Evaluation
and Research, Rockville, MD, USA; Dr P. Olin, Swedish Institute for Infectious
Disease Control, Solna, Sweden; Dr C.M. Ausiello, Instituto Superiore di Sanita,
Rome, Italy; Dr M. Powell, Medicines and Healthcare Products Regulating
Agency, London, England; Dr E. Vidor, Sanofi Pasteur, Lyon, France; Dr Philippe
Duclos, Vaccine Assessment and Monitoring, World Health Organization, Geneva,
Switzerland and I. Knezevic, Quality Assurance and Safety of Biologicals, World
Health Organization, Geneva, Switzerland. Taking into account comments and
suggestions received from the participants at this consultation, the clinical section
was finalized.

Acknowledgements
Thanks are due to the following experts for their comments and advice on these
Guidelines: Dr S.S. Jadhav, Quality Assurance and Regulatory Affairs, Serum
Institute of India Ltd, Pune, India; Dr E. Ma Fajardo, Vaccine Adviser and International
Affairs, Finlay Institute, Cuba; Dr A. Tahlan, Joint Director and Government Central
Research Institute, Kasauli, India; Dr J. Arciniega, Center for Biologics Evaluation
and Research, Rockville, MD, USA; Dr B. Meade, Office of Vaccines Review and
Research, Center for Biologics Evaluation and Research, Rockville, MD, USA; Dr
M. Powell, Medicines and Healthcare Products Regulating Agency, London, England
and Dr M-E. Behr Gross, Strasbourg, France.

References
1. Requirements for Diphtheria, tetanus, pertussis and combined vaccines.
In: WHO Expert Committee on Biological Standardization. Fortieth report.
Geneva, World Health Organization, 1990, Annex 2 (WHO Technical
Report Series, No. 800).
2. Recommendations for diphtheria, tetanus, pertussis and combined vaccines.
Amendments 2003. In WHO Expert Committee on Biological Standardization.
Fifty-fourth report. Geneva, World Health Organization, 2005, Annex 5 (WHO
Technical Report Series, 927).
3. Guidelines for the production and control of the acellular pertussis
component of monovalent or combined vaccines. In: WHO Expert Committee
on Biological Standardization. Forty-seventh Report. Geneva, World Health
Organization, 1998, Annex 2 (WHO Technical Report Series, No. 878).
4. WHO position paper on pertussis vaccines. Weekly Epidemiological Record,
2005, 80:29–40.

326
5. Halander H et al. Shifts of Bordetella pertussis Variants in Sweden from 1970
to 2003, during three periods marked by different vaccination programs.
Journal of Clinical Microbiology, 2005, 43:2856–2865.
6. Van Amersfooth et al. Analysis of Bordetella pertussis populations in
European countries with different vaccination policies. Journal of Clinical
Microbiology, 2005, 43:2837–2843.
7. Caro et al. Pulsed-field gel electrophoresis analysis of Bordetella pertussis
populations in various European countries with different vaccine policies.
Microbe and Infection, 2005, 7:976–982.
8. Miller E. Overview of recent clinical trials of acellular pertussis. Biologicals,
1999, 27:79–86.
9. Heath PT, Ramsay ME. Haemophilus influenzae type b vaccine—booster
campaign. British Medical Journal, 2003. 326:1158–1159.
10. Matoo S and Cherry JD. Molecular pathogenesis, epidemiology and clinical
manifestations of respiratory infections due to Bordetella pertussis and other
Bordetella species. Clinical Microbiology Reviews, 2005, 18:326–382.
11. Medical Research Council. Vaccination against whooping cough: Relation
between protection in children and results of laboratory tests. British
Medical Journal, 1956, 2:454–462.
12. Xing D et al. Laboratory testing of whole-cell pertussis vaccine: a WHO
proficiency study using the Kendrick test. Vaccine, 2002, 20:342–351.
13. Alternatives to whole-cell pertussis vaccine potency assay. In: EDQM
Proceedings. Geneva, European Directorate for the Quality of Medicines,
Council of Europe, 2005.
14. Vaccines and biologicals: Recommendations from the Strategic Advisory
Group of Experts. Weekly Epidemiological Record, 2002, 37:306.
15. Guidelines on regulatory expectations related to the elimination,
reduction or replacement of thiomersal in vaccines. In: WHO Expert
Committee on Biological Standardization. Fifty-third report. Geneva, World
Health Organization, 2004, Annex 4 (WHO Technical Report Series,
No. 926).
16. Good manufacturing practices for pharmaceutical products. In: WHO
Expert Committee on Specifications for Pharmaceutical Preparations.
Thirty-seventh Report. Geneva, World Health Organization, 2003. Annex 4
(WHO Technical Report Series, No. 908).
17. Good manufacturing practices for biological products. Geneva, World Health
Organization, 1992 (WHO Technical Report Series, No. 822):20–30.
18. Guidelines on transmissible spongiform encephalopathies in relation to
biological and pharmaceutical products. Geneva, World Health Organization,
2003 (WHO/BCT/QSD/03.01).
19. General requirements for the sterility of biological substances (Requirements
for Biological substances no. 6, revised 1973). In: WHO Expert Committee
on Biological Standardization. Twenty-fifth report. Geneva, World Health
Organization, 1973 (Technical Report Series, No. 530) and amended In:
WHO Expert Committee on Biological Standardization. Forty-sixth report.
Geneva, World Health Organization, 1998 (WHO Technical Report Series,
No. 872).

327
 20. Guidelines for nonclinical evaluation of vaccines. In: WHO Expert Committee
on Biological Standardization. Fifty-fourth report. Geneva, World Health
Organization, 2005, annex 1 (WHO Technical Report Series, No. 927).
21. Guidelines for good clinical practices (GCP) for trials on pharmaceutical
products. In: WHO Expert Committee on the Use of Essential Drugs, Sixth
report. Geneva, World Health Organization, 1995, Annex 3 (WHO Technical
Report Series, No. 850).
22. Guidelines for clinical evaluation of vaccines: regulatory expectations. In:
WHO Expert Committee on Biological Standardization. Fifty-second report.
Geneva, World Health Organization, 2004, Annex 1 (WHO Technical Report
Series,
No. 924).
23. Guidelines for national authorities on quality assurance for biological
products. Geneva, World Health Organization, 1992 (WHO Technical Report
Series, No. 822):31–46.

328
 Appendix
Summary protocol for whole-cell pertussis vaccine
production and testing
Summary information on final lot
Name and address of manufacturer ______________________________
__________________________________________________________
Lot no. ____________________________________________________
Date of filling _______________________________________________
Date of manufacturing ________________________________________
Nature of final product (absorbed) _______________________________
Volume of each recommended single
human dose ______________________________________________
No. of doses per final container _________________________________
No. of final containers ________________________________________
Container/closure system for the final lot _________________________
Expiry date _________________________________________________

Detailed information on manufacture and control

Strain
Identity of B. pertussis strains used in vaccine _____________________
Serological types of strains ____________________________________
Reference no. of seed lot ______________________________________
Date(s) of reconstitution of ampoule(s)
for manufacture ___________________________________________

Single harvests used for preparing final bulk

List the single harvests and indicate the medium, dates of inoculation,
temperature of incubation, dates of harvests, volumes, results of tests for
bacterial purity, methods and dates of inactivation, opacity and fimbriae
present.

Final bulk
Identification _______________________________________________
Volume ____________________________________________________
No. of opacity units (calculated from opacities of
single harvests) ___________________________________________

329
 Test for fimbriae 2 and 3
Date and results (before addition of adjuvant) _____________________
Sterility test
Sample tested and volume _____________________________________
Media, volume and temperatures of incubation
Date(s) of inoculation ________________________________________
Date(s) of end of observation __________________________________
Result of each test ___________________________________________

Specific toxicity test (mouse weight-gain test)

Strain of mice _______________________________________________
No. of mice ________________________________________________
Volume and route of injection __________________________________
Date of end of observation _____________________________________
Result of test: on a separate sheet of paper, give all relevant details on mice
in the control and test groups (survival, mean weight on day of injection and
three and seven days after injection) and indicate percentage weight of test
group as compared with control group).

Other specific toxicity tests

Mention here date and results of any other specific
toxicity test(s) which may have been performed
(e.g. tests for heat-labile toxin, lymphocytosis promoting
factor and endotoxin) ___________________________________

Potency test
Strain, weight and sex of mice __________________________________
Date of immunization ________________________________________
LD50 in challenge dose ________________________________________
No. of colony-forming units in challenge
dose ____________________________________________________
Date of challenge ____________________________________________
Date of end of observation _____________________________________
Results ____________________________________________________
Calculation method __________________________________________

No. of survivors/ Median effective

Dilution No. of animals inoculated dose (ED50 )
Reference ___________ _____________________
vaccine ___________ _____________________ __________ ml
(..IU/ml) ___________ _____________________

330
 No. of survivors/ Median effective
Dilution No. of animals inoculated dose (ED50 )
___________ _____________________
Test vaccine ___________ _____________________ __________ ml
___________ _____________________
Potency of test vaccine is ... IU per single human dose. Limits of 95%
confidence interval (in %) are ...

pH
Date of measurement _________________________________________
Result _____________________________________________________
Specification _______________________________________________

Final product
Identity test ________________________________________________
Date of test _________________________________________________
Type of test and result ________________________________________

Sterility test
No. of times the test had to be performed _________________________
No. of containers tested in each test and volume ____________________
Media, volume and temperatures of incubation ____________________
Date(s) of inoculation ________________________________________
Date(s) of end of observation __________________________________
Result of each test ___________________________________________

Potency test
If the test was not performed on the final bulk, indicate this and report the
data obtained on the final product in the space provided for potency tests in
the “final bulk” section.

Innocuity test Mice Guinea-pigs

No. of animals ___________________ __________________
Route of injection ___________________ __________________
Volume of injection ___________________ __________________
Date of start of test ___________________ __________________
Date of end of test ___________________ __________________
Results ___________________ __________________

Test for adjuvant

Date of test _________________________________________________
Nature and concentration of adjuvant per single
human dose ______________________________________________
Method of testing ____________________________________________

331
 Specification _______________________________________________
Result _____________________________________________________

Test for preservative

Date of test _________________________________________________
Nature and concentration of preservative _________________________
Method of testing ____________________________________________
Specification _______________________________________________
Result _____________________________________________________

pH
Date of measurement _________________________________________
Method of testing ____________________________________________
Specification _______________________________________________
Result _____________________________________________________

Inspection of final containers

Date of inspection ___________________________________________
Organoleptic characteristics ____________________________________
Number of containers inspected ________________________________
% of rejected containers ______________________________________

Stability test1
Indicate separately all relevant details and (as a percentage) the calculated
losses of potency per year at different temperatures, as determined by
accelerated degradation tests, and actual titres2 (with limits of 95%
confidence intervals) after storage for the maximum period claimed for the
product at the recommended temperature.

Certification by the manufacturer

Name of head and production (typed) ____________________________
Certification by person from the control laboratory of the manufacturing
company taking overall responsibility for the production and control of the
vaccine
I certify that lot No. ... of pertussis vaccine, whose number appears on the
label of the final containers, meets all national requirements3 and satisfies
Part A of the pertussis vaccine section of Requirements for Biological
Substances Nos. 8 and 10, revised 1989 and (if applicable) addenda 19...

1
Not required in summary protocols of every batch.
2
Needed only for three batches to validate the production method.
3
If any national requirement(s) is (are) not met, specify which one(s) and indicate why release of
the lot has nevertheless been authorized.

332
Signature __________________________________________________
Name (typed) _______________________________________________
Date ______________________________________________________

Certification by the national regulatory authority

If the vaccine is to be exported, attach a certificate from the national
regulatory authority as shown in Appendix 2, a label from a final container,
and an instruction leaflet for users.

333
© World Health Organization
WHO Technical Report Series No 941, 2007

Annex 7
Biological substances: international standards
and reference reagents
A list of International Standards and Reference Reagents for biological
substances was issued in WHO Technical Report Series, No. 897, 2000
(Annex 4) and an updated version is available on the Internet at http://www.
who.int/biologicals. Copies of the list may be obtained from appointed sales
agents for WHO publications or from: Marketing and Dissemination, World
Health Organization, 1211 Geneva 27, Switzerland.
At its meeting in October 2005, the WHO Expert Committee on Biological
Standardization made the following changes to the previously published
list.
These substances are held and distributed by the International Laboratory
for Biological Standards, National Institute for Biological Standards and
Control, Potters Bar, Herts., EN6 3QG, England.

Additions
Preparation Activity Status

Antigens and related substances

Haemophilus influenza type b 4.933 ± 0.267 mg/ampoule First International Standard

capsular polysaccharide of polyribosyl ribitol (2005)
phosphate (PRP)

Antisera

Anti-dengue virus types 100 units per serotype First WHO reference reagent
1, 2, 3 and 4 serum per ampoule (2005)

Anti-human platelet 100 IU per ampoule First International Standard

antigen-1a (2005)

Anti-A blood grouping No assigned activity; First International Standard

minimum potency however a 1 in 8 dilution (2005)
Reagent should define the
recommended minimum
potency specification for
anti-A blood grouping
reagents

335
Preparation Activity Status

Anti-B blood grouping No assigned activity; First International Standard

minimum potency reagent however a 1 in 4 dilution (2005)
should define the
recommended minimum
potency specification for
anti-B blood grouping
reagents

Blood products and related substances

Prothrombin mutation No assigned activity First International Genetic

G20210A Reference Panel (2005)

Folate in human serum 12.1 nmol/l of folate per First International Standard
amouple (2005)

Vitamin B12 in human serum 480 pg/ml of vitamin B12 Second International
per ampoule Standard (2005)

Coagulation factor V, plasma, 0.74 IU of Factor V:C per First International Standard
human ampoule (2005)

Coagulation factor XI, plasma, 0.86 IU per ampoule First International Standard
human (2005)

Thromboplastin, rabbit, plain International Sensitivity Third International Standard

Index (ISI) value of 1.15 (2005)

Cytokines, growth factors and endocrinological substances

Vascular endothelial growth 13 000 units per ampoule First WHO Reference
factor, human Reagent (2005)

Keratinocyte growth factor, 4000 units per ampoule First WHO Reference
human Reagent (2005)

Keratinocyte growth factor, 9000 units per ampoule First WHO Reference
(24-163), human Reagent (2005)

Diagnostic reagents

HIV-1 RNA 5.56 log10 International Second International

Units per vial Standard (2005)

336
© World Health Organization
WHO Technical Report Series No 941, 2007

Annex 8
Recommendations, guidelines and other documents
for biological substances used in medicine
The recommendations (previously called requirements) and guidelines
published by the World Health Organization are scientific and advisory in
nature but may be adopted by a national regulatory authority as national
requirements or used as the basis of such requirements.
These international recommendations are intended to provide guidance to
those responsible for the production of biologicals as well as to others who
may have to decide upon appropriate methods of assay and control to ensure
that these products are safe, reliable and potent.
Recommendations concerned with biological substances used in medicine
are formulated by international groups of experts and are published in the
Technical Report Series of the World Health Organization1 as listed here.
A historical list of requirements and other sets of recommendations is
available on request from the World Health Organization, 1211 Geneva 27,
Switzerland.
Reports of the Expert Committee on Biological Standardization published
in the WHO Technical Report Series can be purchased from:
Marketing and Dissemination
World Health Organization
1211 Geneva 27
Switzerland
Telephone: + 41 22 79 12 476
Fax: +41 22 79 14 857
e-mail: [email protected]
Individual recommendations and guidelines may be obtained free of charge
as offprints by writing to:
Quality, Safety and Standards
Department of Immunization, Vaccines and Biologicals
World Health Organization
1211 Geneva 27
Switzerland

1
Abbreviated in the following pages as TRS.

337
 Recommendations, Guidelines
and other documents
Recommendations, Guidelines Reference
and other documents

Acellular pertussis component of monovalent or Adopted 1996, TRS 878 (1998)

combined vaccines

Animal cells, use of, as in vitro substrates for the Revised 1996, TRS 878 (1998);
production of biologicals Addendum 2003, TRS 927 (2005)

Biological standardization and control: a scientific Unpublished document WHO/

review commissioned by the UK National Biological BLG/97.1
Standards Board (1997)

Biological substances: international standards and Revised 2004, TRS 932 (2006)
reference reagents, guidelines for preparation,
characterization and establishment

BCG vaccine, dried Revised 1985, TRS 745 (1987);

Amendment 1987, TRS 771 (1988)

Biological products prepared by recombinant DNA Adopted 1990, TRS 814 (1991)
technology

Blood, blood components and plasma derivatives: Revised 1992, TRS 840 (1994)
collection, processing and quality control

Blood plasma for fractionation (human) Adopted 2005, TRS 941 (2007)

Blood plasma products, human: viral inactivation Adopted 2001, TRS 924 (2004)
and removal procedures

Cholera vaccine (inactivated, oral) Adopted 2001, TRS 924 (2004)

Dengue virus vaccine (tetravalent, live) Adopted 2004, TRS 932 (2006)

Diphtheria, tetanus, pertussis (whole cell) and Revised 1989, TRS 800 (1990);
combined (DTwP) vaccines Addendum 2003, TRS 927 (2005);
Addendum 2005, TRS 941 (2007)

DNA vaccines; quality and nonclinical safety Revised 2005, TRS 941 (2007)

Good manufacturing practices for biological products Adopted 1991, TRS 822 (1992)

Haemophilus influenzae type b conjugate vaccines Revised 1998, TRS 897 (2000)

Haemorrhagic fever with renal syndrome (HFRS) Adopted 1993, TRS 848 (1994)
vaccine (inactivated)

Hepatitis A vaccine (inactivated) Adopted 1994, TRS 858 (1995)

Hepatitis B vaccine prepared from plasma Revised 1987, TRS 771 (1988)

Hepatitis B vaccines made by recombinant DNA Adopted 1988, TRS 786 (1989);
techniques Amendment 1997, TRS 889 (1999)

338
Recommendations, Guidelines Reference
and other documents

Human interferons prepared from lymphoblastoid cells Adopted 1988, TRS 786 (1989)

Influenza vaccine (inactivated) Revised 2003, TRS 927 (2005)

Influenza vaccine (live) Adopted 1978, TRS 638 (1979)

Influenza vaccines, biosafety risk assessment and Adopted 2005, TRS 941 (2007)
safe production and control for human pandemic
vaccines

Japanese encephalitis vaccine (inactivated) for Adopted 1987, TRS 771 (1988)
human use

Japanese encephalitis vaccine (live) for human use Adopted 2000, TRS 910 (2002)

Louse-borne human typhus vaccine (live) Adopted 1982, TRS 687 (1983)

Measles, mumps and rubella vaccines and Adopted 1992, TRS 848 (1994);
combined vaccine (live) Note TRS 848 (1994)

Meningococcal polysaccharide vaccine Adopted 1975, TRS 594 (1976);

Addendum 1980, TRS 658 (1981);
Amendment 1999, TRS 904 (2002)

Meningococcal C conjugate vaccines Adopted 2001, TRS 924 (2004);

Addendum 2003, TRS 926 (2004)

Monoclonal antibodies Adopted 1991, TRS 822 (1992)

Pneumococcal conjugate vaccines Adopted 2003, TRS 927 (2005)

Poliomyelitis vaccine (inactivated) Revised 2000, TRS 910 (2002);

Amendment 2003, TRS 926 (2004)

Poliomyelitis vaccine (inactivated): guidelines for Adopted 2003, TRS 926 (2004)
the safe production and quality control of inactivated
poliovirus manufactured from wild polioviruses

Poliomyelitis vaccine, oral Revised 1999, TRS 904 (2002);

Addendum 2000, TRS 910 (2002)

Quality assurance for biological products, guidelines Adopted 1991, TRS 822 (1992)
for national authorities

Rabies vaccine for human use (inactivated) Revised 2005, TRS 941 (2007)
produced in cell substrates or embryonated eggs

Regulation and licensing of biological products in Adopted 1994, TRS 858 (1995)
countries with newly developing regulatory authorities

Rotavirus vaccine (live attenuated), oral Adopted 2005, TRS 941 (2007)

Smallpox vaccine Revised 2003, TRS 926 (2004)

Sterility of biological substances Revised 1973, TRS 530 (1973);

Amendment 1995, TRS 872 (1998)

339
Recommendations, Guidelines Reference
and other documents

Synthetic peptide vaccines Adopted 1997, TRS 889 (1999)

Thiomersal for vaccines: regulatory expectations Adopted 2003, TRS 926 (2004)
for elimination, reduction or removal

Thromboplastins and plasma used to control oral Revised 1997, TRS 889 (1999)
anticoagulant therapy

Tick-borne encephalitis vaccine (inactivated) Adopted 1997, TRS 889 (1999)

Transmissible spongiform encephalopathies tissue Revised 2005, WHO (2006) http://

infectivity distribution, guidelines www.who.int/bloodproducts/TSE/en

Tuberculins Revised 1985, TRS 745 (1987)

Typhoid vaccine Adopted 1966, TRS 361 (1967)

Typhoid vaccine, Vi polysaccharide Adopted 1992, TRS 840 (1994)

Vaccines, clinical evaluation: regulatory expectations Adopted 2001, TRS 924 (2004)

Vaccines, nonclinical evaluation Adopted 2003, TRS 926 (2004)

Varicella vaccine (live) Revised 1993, TRS 848 (1994)

Yellow fever vaccine Revised 1995, TRS 872 (1998)

Yellow fever vaccine, laboratories approved by WHO Revised 1995, TRS 872 (1998)
for the production of

Yellow fever virus, production and testing of WHO TRS 745 (1987)
primary seed lot 213-77 and reference batch 168-73

340