Annex 3: Recommendations To Assure The Quality, Safety and Efficacy of BCG Vaccines

Annex 3
Recommendations to assure the quality, safety and

efficacy of BCG vaccines
Replacement of Annex 2 of WHO Technical Report Series, No. 745, and
Amendment to Annex 12 of WHO Technical Report Series, No. 771
Introduction 139
General considerations 139
Special considerations 140
Scope of the Recommendations 141
BCG vaccine strains 141
Potency-related tests 142
Part A. Manufacturing recommendations 143
A.1 Definitions 143
A.2 General manufacturing recommendations 145
A.3 Control of source materials 147
A.4 Control of vaccine production 149
A.5 Filling and containers 152
A.6 Control tests on final lot 152
A.7 Records 156
A.8 Retained samples 156
A.9 Labelling 156
A.10 Distribution and transport 157
A.11 Stability, storage and expiry date 157
Part B. Nonclinical evaluation of BCG vaccines 159
Part C. Clinical evaluation of BCG vaccines 160
C.1 General considerations 160
C.2 Special considerations 162
C.3 Post-marketing surveillance 164
Part D. Guidelines for NRAs 164
D.1 General 164
D.2 Release and certification 166
Authors and acknowledgements 166
References 170
137
WHO Expert Committee on Biological Standardization Sixty-second report
Appendix 1
History and genealogy of BCG substrains 174
Appendix 2
Summary protocol for manufacturing and control of BCG vaccine 175
Appendix 3
Model certificate for the release of BCG vaccine by NRAs 184
Recommendations published by the WHO are intended to be

scientific and advisory in nature. Each of the following sections
constitutes guidance for national regulatory authorities (NRAs)
and for manufacturers of biological products. If an NRA so desires,
these Recommendations may be adopted as definitive national
requirements, or modifications may be justified and made by the NRA.
It is recommended that modifications to these Recommendations
be made only on condition that such modifications ensure that the
vaccine is at least as safe and efficacious as that prepared in accordance
with the recommendations set out below. The parts of each section
printed in small type are comments for additional guidance, intended
for the benefit of manufacturers and NRAs.
WHO Technical Report Series No. 979, 2013
138
Annex 3
Introduction
The last revision of the Requirements for dried bacille Calmette–Guérin (BCG)
vaccine for human use was in 1985, and an amendment which updated the
section on the expiry date was published in 1988 (1, 2). Recent WHO consultation
meetings (3–6) have addressed issues concerning the improvement of vaccine
characterization and quality control assays of BCG vaccine to reflect current state-
of-the-art technology. In addition, a recommendation to replace the International
Reference Preparation for BCG vaccine by substrain-specific Reference Reagents
evaluated by collaborative studies has been proposed. This document provides:
recommendations for the production and control of BCG vaccines (Part A);
guidelines for nonclinical evaluation (Part B); guidelines for the content of the
clinical development programme applicable to BCG vaccines (Part C); and
recommendations for NRAs (Part D). The guidelines for nonclinical evaluation
apply to classic BCG vaccine products that are still in need of such evaluation,
including newly manufactured products requiring clinical trial studies or those
produced following changes in the manufacturing process. The clinical part of this
document aims to provide a basis for assessment of efficacy and safety of BCG
vaccines in pre-licensing clinical trials as well as in post-marketing surveillance,
monitoring consistency of production and clinical testing of new classic BCG
vaccine products. If important changes have been introduced to an authorized
production process, the need for preclinical and clinical testing should be
considered on a case-by-case basis in consultation with the NRA(s) concerned.
General considerations
Tuberculosis (TB) was declared a global emergency by WHO in 1993, and
Mycobacterium tuberculosis (M. tuberculosis) is now considered to be responsible
for more adult deaths than any other pathogen. Vaccination with BCG still
remains the standard for TB prevention in most countries because of its efficacy
in preventing life-threatening forms of TB in infants and young children. It is
inexpensive and usually requires only one administration in either newborns or
adolescents (7, 8). As there is currently no suitable alternative, BCG will remain
in use for the foreseeable future and may continue to be used as a prime vaccine in
a prime-boost immunization schedule in conjunction with new TB vaccines (4).
BCG vaccine contains a live, attenuated strain of M. bovis that was
originally isolated from cattle with tuberculosis and cultured for a period of 13
years and a total of 231 passages (7). The BCG vaccine was first used to immunize
humans in 1921. Following its introduction into the WHO Expanded Programme
on Immunization (EPI) in 1974, the vaccine soon reached global coverage rates
exceeding 80% in countries endemic for TB (9).
139
Over the years, different BCG vaccine seed strains have evolved from the
original vaccine strain for production. A number of BCG vaccine strains that
are used worldwide differ in terms of their genetic and phenotypic properties,
and their reactogenicity and immunogenicity profile when given to infants
and children. With this background of a diversity of substrains, manufacturing
processes, immunization schedules and levels of exposure to environmental
mycobacteria and virulent M. tuberculosis infection, different levels of protective
efficacy of BCG vaccines in adult populations have been reported (10). However,
the data are insufficient to make recommendations on whether one strain should
be preferred over the other (11). The United Nations agencies are the largest
supplier of BCG vaccines, distributing more than 120 million doses each year to
more than 100 countries. Worldwide, the most commonly used vaccine strains
are currently Danish 1331, Tokyo 172-1 and Russian BCG-I because they are
supplied by the United Nations Children’s Fund (UNICEF) which purchases the
vaccines through a published prequalification process which determines their
eligibility for use in national immunization programmes (12).
There has been particular concern over the safety of BCG vaccination
in subjects infected with the human immunodeficiency virus (HIV) (8). WHO
previously recommended that in countries with a high burden of TB, a single
administration of BCG vaccine should be given to all healthy infants as soon as
possible after birth, unless the child presented a symptomatic HIV infection (9).
However, recent evidence shows that children who were HIV-infected when
vaccinated with BCG at birth, and who later developed acquired immunodeficiency
syndrome (AIDS), were at increased risk of developing disseminated BCG
disease. Among these children, the benefits of potential prevention of severe TB
are outweighed by the risks associated with the use of BCG vaccine; thus the use
of BCG vaccines at birth in relation to HIV-infected infants should follow the
recommendations of the Global Advisory Committee on Vaccine Safety (GACVS)
(13, 14).
Special considerations
The formulation of international requirements for freeze-dried BCG vaccine is
complicated by the following: (a) a number of different substrains derived from
the original strain of BCG are used in vaccine manufacture; (b) a number of
different manufacturing and testing procedures are employed; (c) it is difficult to
identify a link between significant differences in vitro and in vivo between different
BCG vaccine strains and any possible differences in protective efficacy against
TB in humans; (d) vaccines are produced with different total bacterial content
and numbers of culturable particles; and (e) vaccines intended for administration
by different routes are prepared. Therefore, the following considerations should
be borne in mind regarding the scope of these recommendations, BCG vaccine
strains, and potency-related tests.
140
Annex 3
Scope of the Recommendations

These revised Recommendations refer to freeze-dried BCG vaccines prepared
from substrains derived from original BCG for use in the prevention of TB. Where
BCG vaccine is issued in liquid form, the application of these Recommendations
is entirely the responsibility of the NRA. In that case, only the relevant parts of
this document apply since the limited stability of liquid BCG limits the possibility
of completing the full recommended control test schedule. Although many of
the principles expressed in this document (e.g. manufacturing, quality control)
are expected to apply also to new recombinant BCG and other live, attenuated
mycobacterial vaccines modified by molecular biology techniques, these novel
vaccines are outside the scope of these Recommendations. The same pertains
to the use of BCG for immunotherapy (e.g. treatment of bladder cancer).
However, applicability of issues on nonclinical and clinical evaluations should
be considered on a case-by-case basis. These Recommendations have been
formulated primarily to cover vaccines intended for intradermal and percutaneous
administration. Although WHO recommends intradermal administration of the
vaccine, preferably in the deltoid region of the arm using syringe and needle,
other administration methods such as percutaneous application by the multiple
puncture technique are practised in some countries (9, 15–17).
BCG vaccine strains

The original BCG vaccine strain was formerly distributed by the Pasteur Institute
of Paris and subcultured in different countries using different culture conditions
that were not standardized. Over the years, more than 14 substrains of BCG
have evolved and have been used as BCG vaccine strains in different parts of
the world (see Appendix 1). Recently, the various substrains have been studied
by comparative genomics (18, 19). BCG vaccine strains were thus divided into
the “early” strains, in which the original characteristics of “authentic Pasteur”
were conserved with fewer deletions, insertions and mutations in the genome of
the bacilli than the “late” strains. “Early” strains are represented by BCG Russia
BCG-I, BCG Moreau-RJ, BCG Tokyo 172-1, BCG Sweden, and BCG Birkhaug;
and the “late” strains include BCG Pasteur 1173P2, BCG Danish 1331, BCG
Glaxo (Copenhagen 1077) and BCG Prague. The genomic sequences of BCG
Pasteur 1173P2 as a “late” strain, and BCG Tokyo 172-1 and BCG Moreau as
“early” strains were determined (18–20). There is insufficient direct evidence to
suggest that various BCG substrains differ significantly in their efficacy to protect
against TB in humans. However, evidence from animal and human studies
indicates differences in the immune responses induced by different BCG vaccine
strains (12, 21). Although the “early” strains may confer better protection against
141
TB in some animal studies (18, 22), commonly administered BCG vaccine

strains including both evolutionary “early” and “late” strains induce comparable
protective immunity against TB (23).
Only master seed lots that have been shown to be acceptable by
laboratory and clinical tests on batches derived from them should be used for
the production of working seed lots and/or final product. A suitable seed lot of
BCG should yield vaccines that give protection in experimental animals, produce
a relatively high level of immunological responses to M. tuberculosis antigens
including tuberculin sensitivity in humans, and have an acceptably low frequency
of adverse reactions (see section A.3.1).
Some manufacturers of freeze-dried BCG vaccine have modified their
master seed lot strain to make it more suitable for their particular production
procedure. The seed lots prepared in this way may not retain the same
immunogenic properties, and should be used only with the approval of the NRA.
In practice, a product prepared from BCG seed lots may generally be
investigated in humans only for the properties of producing tuberculin sensitivity
and vaccination lesions. The former should be measured by the distribution of
tuberculin reactions according to size in persons vaccinated with a given dose
of BCG vaccine. A low dose of tuberculin should be employed (e.g. equivalent
to 5 IU of the First WHO International Standard for purified protein derivative
(PPD) of M. tuberculosis, or 2 tuberculin units (TU) of a batch of PPD RT23 with
Tween 80).
Currently three substrain-specific Reference Reagents for BCG vaccines
are available: BCG Danish 1331, Tokyo 172-1 and Russian BCG-I.
Potency-related tests
There is some evidence that BCG seed lots that have been shown to produce
vaccines with protective potency in laboratory animals and tuberculin sensitivity
in humans will give effective protection against TB in humans. It should be noted

that tuberculin sensitivity is a marker for cell-mediated immune responses to
mycobacteria and not a direct indicator of protective immunity. A number of
alternative laboratory tests have been developed primarily for research purposes
but, to date, none have been proved reliable indicators of protective immune
conversion following administration of different vaccines.
Studies in animals should include protection tests, tests of vaccination
lesions, and tests for tuberculin conversion. Immunizing efficacy should be
measured in terms of degree of protection afforded to the test animals against a
challenge with virulent M. tuberculosis. Sensitizing efficacy should be measured
by the average dose of vaccine that will convert a negative tuberculin reaction
in guinea-pigs to a positive one, as well as by the reaction time during which
142
Annex 3
the conversion takes place. In these animal tests, the inclusion for comparative
purposes of an in-house reference BCG vaccine prepared from a seed lot known
to be effective in animals and humans is recommended.
Currently there is no biomarker which directly correlates to clinical
efficacy of BCG vaccine. These Recommendations are intended to be used for
ensuring the manufacture of consistent lots. This means that new lots should
not significantly differ from those that have already been shown to be safe and
effective in humans.
At present, for batch control purposes, much reliance is placed on tests
for the estimation of the total bacterial content and for the number of culturable
particles. It is not possible to specify single requirements for the total bacterial
content and for the number of culturable particles for all vaccines (24), since
different substrains and methods of manufacture may yield different specifications
for these parameters. For example, although the number of culturable bacteria in
a single human dose may differ for different vaccines, these vaccines may show
satisfactory properties as regards their ability to induce adequate sensitivity to
tuberculin and their safety in humans. It is therefore essential that clinical studies
for dose optimization in humans be carried out to estimate suitable total bacterial
contents and the number of culturable particles for a particular manufacturer’s
product. For a particular vaccine, the difference between the lower and upper
specification for the number of culturable particles should not be larger than
fourfold. In addition, it is necessary to perform animal experiments that give an
indication of the safety and efficacy of the vaccines to the satisfaction of the NRA.
Part A. Manufacturing recommendations

A.1 Definitions
A.1.1 International name and proper name
The international name should be “freeze-dried BCG 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

Freeze-dried BCG vaccine is a freeze-dried preparation containing live bacteria
derived from a culture of the bacillus of Calmette and Guérin, known as BCG,
intended for intradermal injection. The name of the freeze-dried vaccine
intended for percutaneous vaccination, should be “freeze-dried BCG vaccine,
percutaneous”. The preparation should satisfy all the recommendations
formulated below.
143
A.1.3 International Reference Preparations and Reference Reagents

The First WHO International Reference Preparation for BCG vaccine was
established in 1965 and the First WHO International Standard for PPD of
M. tuberculosis was established in 1951. Because of the age of these preparations,
the need for replacements has been recognized, especially for the First WHO
International Reference Preparation for BCG vaccine which is a live bacterial
preparation. WHO has initiated the development of replacements. These were
presented to the WHO Expert Committee on Biological Standardization in 2009
and 2010 as candidates for the First WHO International Reference Reagents
for BCG vaccines of substrain Danish 1331, Tokyo 172-1 and Russian BCG-I
(25, 26). These materials are available through the WHO web site (http://www.
who.int/entity/bloodproducts/catalogue/BlooFeb2013.pdf). These International
Reference Reagents cover the major proportion of BCG vaccine strains currently
used in production. The establishment of substrain Moreau-RJ as the WHO
International Reference Reagent for BCG vaccine is currently in progress and
is scheduled for submission to the Committee in 2012 for adoption. These
preparations are intended as International Reference Reagents, if required, for:
■ periodical consistency monitoring of quantitative assays such as
viability estimates (such as culturable particle count and modified
ATP assays);
■ residual virulence/local reactogenicity assays and protection assays
in animal models for nonclinical evaluation.
They are also intended:
■ as reference BCG substrains;
■ for identity tests using multiplex polymerase chain reaction (PCR)
as included in the collaborative study or in other molecular biology
techniques.
The National Institute for Biological Standards and Control (NIBSC),

England, distributes the WHO International Reference Reagents for BCG vaccines.
A.1.4 Terminology
The definitions given below apply to the terms as used in these Recommendations.
They may have different meanings in other contexts.
Final bulk: the homogeneous finished liquid vaccine present in a single
container from which the final containers are filled, either directly or through
one or more intermediate containers derived from the initial single container.
Final lot: a number of sealed, final containers that are equivalent with
respect to the risk of contamination during filling and, when it is performed,
144
Annex 3
freeze-drying. A final lot should therefore have been filled from a single container
and freeze-dried in one continuous working session.
In-house reference: a batch of vaccine prepared from the same BCG
strain as the tested vaccine and used in parallel to the vaccine tested in:
■ quantitative assays such as viability estimates (such as culturable
particle count and modified ATP assays);
■ residual virulence assays.
Master seed lot: a bacterial suspension of a single substrain originated
from the bacillus of Calmette and Guérin that has been processed as a single lot
and is of uniform composition. A seed lot should be maintained in the freeze-
dried form stored at –20 °C or below (in the liquid form it is stored at –80 °C
or below) in order to maintain viability. In each manufacturing establishment, a
master seed lot is that from which material is drawn for inoculating media for the
preparation of working seed lots or single harvests.
Single harvest: the material obtained from one batch of cultures that
have been inoculated with the working seed lot (or with the inoculum derived
from it), harvested and processed together.
Working seed lot: a quantity of bacterial organisms of a single substrain
derived from the master seed lot by growing the organisms and maintaining
them in aliquots in the freeze-dried form stored at –20 °C or below (in the liquid
form stored at –80 °C or below). The working seed lot should be prepared from
the master seed lot by as few cultural passages as possible (e.g. 3–6 passages from
the master seed lot), having the same characteristics as the master seed lot and
intended for inoculating media for the preparation of single harvests.
A.2 General manufacturing recommendations

The general manufacturing recommendations for manufacturing establishments
contained in WHO’s Good manufacturing practices: main principles for
pharmaceutical products (27) and Good manufacturing practices for biological
products (28) should apply to establishments manufacturing BCG vaccine. In
addition, the compliance with current good manufacturing practices should
apply with the addition of the following.
Details of standard operating procedures for the preparation and testing of
BCG vaccines adopted by the manufacturer, together with evidence of appropriate
validation of each production step, should be submitted for the approval of the
NRA. As required, proposals for the modification of manufacturing and control
methods should also be submitted for approval to the NRA before they are
implemented.
The NRA should satisfy itself that adequate control of the manufacturing,
shipping and storage of the BCG vaccine has been achieved. NRAs may consider
145
that a formal clinical lot-to-lot consistency study is not necessary if there are
adequate and satisfactory data provided to support consistency of manufacture.
However, several different lots of the product should be used in randomized
studies and should elicit comparable immune responses in similar populations.
The degree of consistency in producing satisfactory final lots is an
important factor in judging the efficacy and safety of a particular
manufacturer’s product.
The data that should be considered in determining the consistency of production

should include the results obtained with consecutive vaccine lots when tested as
described in Part A, section 6 (e.g. the test for viability in Part A, section 6.7, and
the thermal stability test in Part A, section 6.8).
More than two consecutive vaccine lots should have been satisfactorily
prepared before any vaccine from a given manufacturer, or resulting from a
new method of manufacture, is released. In subsequent routine production, if a
specified proportion of vaccine lots or a specified number of consecutive vaccine
lots fails to meet the requirements, the manufacture of BCG vaccine should be
discontinued and should not be resumed until a thorough investigation has been
made and the cause or causes of the failures determined to the satisfaction of
the NRA.
Conventionally, production of BCG vaccine should take place in a
dedicated area, completely separate from areas used for production of other
medicines or vaccines, and using dedicated separate equipment. Such areas
should be so situated and ventilated that the hazard of contamination is reduced
to a minimum. No animals should be permitted in the vaccine production areas.
Tests for the control of vaccine that require cultures to be made of contaminating
microorganisms should be carried out in a completely separate area. Tests in
which animals are used should also be carried out in a completely separate area.
For the purposes of these requirements, the processes of vaccine production
that should take place in dedicated facilities are all operations up to and including
the sealing of the vaccine in the final containers.
In some countries, the production of BCG vaccine – although isolated –
is carried out in a building in which other work takes place. This should
be done only after consultation with, and with the approval of, the NRA.
If production takes place in part of a building, the work carried out in
other parts of the building should be of such a nature that there is no
possibility of cross-contamination with the BCG vaccine.
No cultures of microorganisms other than the BCG vaccine strain approved

by the NRA for vaccine production should be introduced into the manufacturing
areas. In particular, no strains of other mycobacterial species, whether pathogenic
or not, should be permitted in the BCG vaccine production area.
146
Annex 3
BCG is susceptible to sunlight. Therefore, the procedures for the

preparation of the vaccine should be so designed that all cultures and vaccines are
protected from direct sunlight and ultraviolet light at all stages of manufacture,
testing and storage, until the vaccine is issued.
BCG vaccine should be produced by a staff consisting of healthy persons
who do not work with other infectious agents; in particular, they should not work
with virulent strains of M. tuberculosis, nor should they be exposed to a known
risk of tuberculosis infection. Precautions should also be taken to ensure that
no worker should be employed in the preparation of BCG vaccine unless he or
she has been shown by medical examination to be free from TB. The scope and
nature of the medical examination should be at the discretion of the NRA. It may
include a radiological examination and/or a validated immunological blood assay
that should be repeated at intervals or when there is reason to suspect illness. The
frequency of radiological examination should be at the discretion of the NRA,
taking into consideration the incidence of TB in the country.
It is advisable to keep radiation exposure to a minimum, but the
examination should be of sufficient frequency to detect the appearance of
early active TB. It is estimated that, if workers in BCG vaccine laboratories
were given one or two conventional X-ray examinations of the chest each
year, not using fluoroscopic methods, and if the best available techniques
were employed to minimize the radiation dose, the doses received
would be considerably lower than the maximum permissible doses for
workers occupationally exposed to radiation that have been set by the
International Commission on Radiological Protection (29, 30).
Should an examination reveal signs of TB or suspected TB in a worker, he or

she should no longer be allowed to work in the production areas and the rest of
the staff should be examined for possible TB infection. In addition, all cultures
should be discarded and the production areas decontaminated. If it is confirmed
that the worker has TB, all vaccine made while he or she was in the production
areas should be discarded, and all distributed batches should be recalled.
Persons not normally employed in the production areas should be
excluded from them unless, after a medical examination, including radiological
examination, they are shown to be free from TB. In particular, persons working
with mycobacteria other than the BCG seed strain should be excluded at all times.
A.3 Control of source materials

A.3.1 Seed lot system
The production of vaccine should be based on the seed lot system. A seed lot
prepared from a strain approved by the NRA (see Part D, section 1.1) should be
prepared under conditions satisfying the requirements of Part A, sections 2, 3
and 4.
147
The BCG vaccine strain used should be identified by historical records

that include information on its origin and subsequent manipulation. It would be
preferable for the master seed lot to have protection proven clinically through
clinical studies on a batch derived from it by a production process that is
representative of the commercial process. It is also recommended to use a batch
derived from such a clinically “validated” seed lot as an in-house reference in the
laboratory to help ensure consistency in production.
If a working seed lot is being used, the total number of passages for a
single production harvest should not exceed 12, including the passages necessary
for preparing the working seed lot.
A.3.2 Tests on seed lot

A.3.2.1 Antimicrobial sensitivity test
An antimicrobial sensitivity test should be carried out as part of the ongoing
characterization of BCG vaccine strains. It would be appropriate to test this
property at the level of master or working seed lot.
A.3.2.2 Delayed hypersensitivity test

When a new working seed lot is established, a suitable test for delayed
hypersensitivity in guinea-pigs is carried out; the vaccine is shown to be not
significantly different in activity from the in-house reference.
A.3.2.3 Identity test

The bacteria in the master and working seed lots are identified as M. bovis BCG
using microbiological techniques (e.g. morphological appearance of the bacilli
in stained smears and the characteristic appearance of the colonies grown on
solid media). Manufacturers are encouraged to carry out the test using molecular
biology techniques (e.g. PCR test) to identify the specific substrain of BCG. The
techniques will also provide relevant information to ensure genetic consistency

in production, from master seed through working seed and to final product (4).
A.3.2.4 Test for bacterial and fungal contamination

Each master and working seed lot should be tested for bacterial and fungal
contamination by appropriate tests, as specified in Part A, section 5.2 of General
requirements for the sterility of biological substances (31), or by the validated
methods approved by the NRA.
A.3.2.5 Test for absence of virulent mycobacteria

The test for absence of virulent mycobacteria, described in Part A, section 4.2.3,
should be made in at least 10 healthy guinea-pigs injected with a quantity of
148
Annex 3
vaccine not less than 50 single human doses and should be observed for at least
six weeks. If none of the animals shows signs of progressive TB and at least 90%
survive the observation period (i.e. should one of the 10 animals die), the seed lot
should be considered to be free from virulent mycobacteria.
If more than 10% of the guinea-pigs die during the observation period
(i.e. should two out of 10 animals die) and freedom from progressive TB disease
is verified, the test should be repeated on at least 10 more guinea-pigs. On the
second occasion, the seed lot passes the test if not more than 10% of the animals
die during the observation period (i.e. should one of the 10 animals die) and the
autopsy does not reveal any sign of TB.
A.3.2.6 Test for excessive dermal reactivity

The test for excessive dermal reactivity, described in Part A, section 6.4.2,
should be made in six healthy guinea-pigs, each weighing not less than 250 g
and having received no treatment likely to interfere with the test. Each guinea-
pig should be injected intradermally, according to a randomized plan, with
0.1 ml of the reconstituted vaccine and of vaccine dilutions 1:10 and 1:100. The
same dilutions of the appropriate international Reference Reagent or in-house
reference should be injected into the same guinea-pigs at randomly selected sites.
The guinea-pigs should be observed for at least four weeks. The vaccine complies
with the test if the reactions it produces at the sites of injection are not markedly
different from those produced by the appropriate international Reference Reagent
or in-house reference.
A.3.3 Production culture medium

The production culture medium should contain no substances known to cause
toxic or allergic reactions in humans. The use of material originating from
animals should be discouraged. However, if constituents derived from animals
are necessary, approval of the NRA should be sought and the materials should
comply with current policy on TSEs (32–37). A risk assessment for TSE would
need to be included for the materials of the culture medium. WHO’s revised
Guidelines on transmissible spongiform encephalopathies in relation to biological
and pharmaceutical products (32) provide guidance on risk assessments for
master and working seeds and should be consulted. Substances used in that
medium should meet such specifications as the NRA may prescribe.
A.4 Control of vaccine production

A.4.1 Control of single harvests
All cultures should be examined visually, and any that have grown in an
uncharacteristic manner should not be used for vaccine production.
149
A.4.2 Control of final bulk

A.4.2.1 Final bulk
The final bulk should be prepared from a single harvest or by pooling a number
of single harvests.
A.4.2.2 Test for bacterial and fungal contamination

The final bulk should be tested for bacterial and fungal contamination by
appropriate tests as specified in Part A, section 5.2 of the General requirements for
the sterility of biological substances (31), or by the validated methods approved
by the NRA. No vaccine lot should be passed for use unless the final bulk has
been shown to be free from such contamination.

The test for absence of virulent mycobacteria should be carried out on each final
bulk or final lot.
At least six healthy guinea-pigs, all of the same sex, each weighing
250–400 g should be used. They should not have received any treatment or diet,
such as antibiotics, that is likely to interfere with the test. A sample of the final
bulk intended for this test should be stored at 4 °C for not more than 72 hours
after harvest.
A dose of BCG organisms corresponding to at least 50 single human
doses of vaccine intended for intradermal injection should be injected into each
guinea-pig by the subcutaneous or intramuscular route.1 The guinea-pigs should
be observed for at least six weeks. If, during that time, they remain healthy, gain
weight, show no signs of progressive TB and not more than one dies, the final
bulk should be considered to be free from virulent mycobacteria.
At the end of the observation period, the animals should be killed and
examined postmortem for macroscopic evidence of progressive TB disease.

Similarly, any animals that die before the end of the observation period should
be subjected to a postmortem examination.
Should one third of the guinea-pigs die (i.e. should two out of six
animals die) during the observation period (and freedom from progressive TB
disease is verified), the test should be repeated on at least six more guinea-pigs.
On the second occasion, the vaccine lot passes the test if not more than one
1
When a more concentrated vaccine, intended for administration by the percutaneous route, is tested, a
dilution factor approved by the NRA should be applied so that the mass of BCG injected corresponds to at
least 50 human doses of intradermal vaccine.
150
Annex 3
animal dies during the observation period and the autopsy does not reveal any
sign of TB.
Should a vaccine lot fail to satisfy the requirements of this test because
animals die from causes other than TB, the procedure to be followed by
the manufacturer should be determined with the approval of the NRA.
If signs of TB disease are seen, the vaccine lot should be rejected, all subsequent
vaccine lots should be withheld, and all current vaccine stocks should be held
pending further investigation. The manufacture of BCG vaccine should be
discontinued and it should not be resumed until a thorough investigation has
been made and the cause or causes of the failure determined and appropriate
actions have been taken. Production should be allowed to resume only upon the
approval of the NRA.
A.4.2.4 Test for bacterial concentration

The bacterial concentration of the final bulk should be estimated by a validated
method approved by the NRA and should have a value within a range approved
by the NRA (see Part D, section 1.2).
Based on manufacturers’ experience, the opacity method is the method
of choice. The International Reference Preparation of Opacity,2 or an
equivalent Reference Preparation approved by the NRA, may be employed
in comparative tests.
Clumping issues should be considered during validation of the assay.
A.4.2.5 Test for number of culturable particles

The number of culturable particles on a solid medium of each final bulk should
be determined by an appropriate method approved by the NRA. Alternatively,
a bioluminescense or other biochemical method can be used (38, 39), provided
that the method is properly validated against the culturable particle test for the
production step in question. If properly validated, such tests can be used as
equivalent methods. Regular calibration with the reference method as agreed
with the NRA would be relevant.
The medium used in this test should be such that the number of
culturable particles may be determined at an optimal time point (usually
3–5 weeks) after the medium has been inoculated with dilutions of the
vaccine.
2
The International Reference Preparation of Opacity is in the custody of the NIBSC, Potters Bar, England,
which supplies samples on request.
151
There are various methods of determining the number of culturable

particles in BCG vaccine, and it is essential that only one culture
method be used for all the vaccine lots produced by a manufacturer (5).
It is also desirable for assay validation that the clumping issue should
be considered and that tests should be carried out in parallel with the
appropriate international Reference Reagent or in-house reference, e.g.
the same vaccine production that has been used in clinical trials and has
assured safety (including immunogenicity) and efficacy.
A.4.2.6 Substances added to the final bulk

Substances used in preparing the final bulk should meet such specifications as the
NRA may prescribe. In particular, the NRA should approve the source(s) of any
animal-derived raw materials which should comply with the WHO Guidelines
on tissue infectivity distribution in transmissible spongiform encephalopathies (34).
Substances added to improve the efficiency of the freeze-drying process
or to aid the stability of the freeze-dried product should be sterile and of high and
consistent quality, and should be used at suitable concentrations in the vaccine.
A.5 Filling and containers

The general requirements concerning filling and containers given in Good
manufacturing practices for biological products (28) should apply to vaccine
filled in the final form.
The containers should be in a form that renders the process of
reconstitution as simple as possible. Their packaging should be such that
the reconstituted vaccine is protected from direct sunlight.
A.6 Control tests on final lot

Tests on the final lot should be performed after reconstitution, except for
appearance and residual moisture tests. The diluent supplied or recommended for
reconstitution should be used, unless such diluent would interfere with any of the
tests, in which case some other suitable fluid should be used. The vaccine should
be reconstituted to the concentration at which it is to be used for injection into
humans; however, an exception may be made in the case of the test for absence
of virulent mycobacteria (Part A, section 6.4.1), when a higher concentration
of reconstituted vaccine may be necessary. It would be appropriate to monitor
periodically the antimicrobial sensitivity in final lots.
A.6.1 Inspection of final containers

Every container in each final lot should be inspected visually, and those showing
abnormalities should be discarded.
152
Annex 3
The appearance of the freeze-dried vaccine and the reconstituted vaccine

should be described with respect to form and colour. If reconstitution with the
product diluent does not allow for the detection of particulates, an alternative
diluent may be used.
A.6.2 Identity test

An identity test should be performed on samples of the vaccine from each final
lot. The identity test for final lots should be used to identify the product as BCG
as approved by the NRA. The identity of each final lot of vaccine should be
verified by the morphological appearance of the bacilli in stained smears and by
the characteristic appearance of the colonies grown on solid media. A validated
nucleic acid amplification technique (such as PCR) should preferably be used.
A.6.3 Test for bacterial and fungal contamination

Samples from each final lot should be tested for bacterial and fungal contamination
by appropriate tests as specified in Part A, section 5.2 of the General requirements
for the sterility of biological substances (31), or by the validated methods approved
by the NRA.
A.6.4 Safety tests

Provided the test for virulent mycobacteria has been carried out with satisfactory
results on the final bulk vaccine, it may be omitted on the final lot.
If the test for the absence of virulent mycobacteria, applied to the final
bulk, is unsatisfactory (and freedom from progressive TB disease is verified), it
should be repeated with a sample of a final lot (see Part A, section 4.2.3).
A.6.4.2 Test for excessive dermal reactivity

Provided the test has been carried out with satisfactory results on the working
seed lot and on at least three consecutive final lots produced from it, the test may
be omitted on the final lot.
Historically the omission of the test with satisfactory results on five
consecutive final lots has been accepted by the authorities (40).
A.6.5 Test for bacterial concentration

The total bacterial content of the reconstituted vaccine should be estimated for
each vaccine lot by a validated method approved by the NRA, and should have a
value within a range approved by the NRA (see Part D, section 1.2).
153
The estimation of tota1 bacterial content may be made either directly,

by determining the dry weight of organisms, or indirectly by an opacity
method that has been calibrated in relation to the dry weight of the
organisms.
The clumping issue should be considered during validation of the assay.
A.6.6 Test for residual moisture

The average moisture content of a freeze-dried vaccine should be determined by
a validated method accepted by the NRA. Values should be within limits of the
preparations shown to be adequately stable in the stability studies of the vaccine.
A.6.7 Tests for viability

A.6.7.1 Test for number of culturable particles
The number of culturable particles of each final lot should be determined by an
appropriate method approved by the NRA (see Part A, section 4.2.5). The viable
count should have a value within a range approved by the NRA that should not
be wider than a fourfold difference between the lower and upper levels of the
specification for numbers of culturable particles (see Part D, section 1.2). By
comparison with the results of the test for number of culturable particles carried
out on final bulk, as described in Part A, section 4.2.5, the percentage survival
on freeze-drying may be calculated and this value should be not less than one
approved by the NRA. The appropriate international Reference Reagent or
in-house reference should be used for every test in order to validate the assay.
The purpose of including the appropriate international Reference Reagent
or in-house reference is to have a check on the quality and consistency
of the culture medium and the accuracy of the technique used for the
determination of the number of culturable particles. It is not intended
to adjust the count of the vaccine by comparison with the Reference

Preparation.
The clumping issue should be considered during validation of the assay.
The survival rate after freeze-drying is usually not less than 20%.
A.6.7.2 Rapid test for viability

As an alternative to the colony counting method, a bioluminescense or other
biochemical method can be used provided that the method is properly validated
against the culturable particle test for the production step in question. If properly
validated, such tests may be considered by the NRA to replace the culturable
particle test.
154
Annex 3
The bioluminescence reaction occurring in fireflies depends on the

presence of adenosine triphosphate (ATP), luciferin luciferase, oxygen
and magnesium ions. This reaction can be reproduced in vitro by mixing
these components. If all components except ATP are present in excess,
the amount of light emitted is proportional to the amount of ATP coming
from the vaccine.
Since ATP is present in all living cells and is immediately destroyed when
the cell dies, ATP is a reliable marker for living cells.
Studies have shown that, if properly validated, measurement of ATP

using the bioluminescence reaction can be used to estimate the viable
count of freeze-dried BCG vaccine within 1–2 days as accurately as
other, more time-consuming methods, once the mean content of ATP
per culturable particle has been estimated for a given vaccine production.
A.6.8 Thermal stability test

The thermal stability test is part of the characterization and consistency
demonstration of vaccine production. The requirement for this test should be
at the discretion of the NRA and, if required, each final lot should be tested for
thermal stability by a validated method approved by the NRA. If the production
consistency is demonstrated, this test may be omitted on the final lot subject to
NRA approval (6).
If performed, the test should involve the determination of the number
of culturable particles before and after the samples have been held at appropriate
temperatures and for appropriate periods.
For example, the thermal stability test may be carried out by taking
samples of the vaccine and incubating them at 37 °C for 28 days.
The percentage decrease in the number of culturable particles is then compared

with that of samples of the same vaccine lot stored at 2–8 °C. The number of
culturable particles in the vaccine after heating should be not less than 20% of
that stored at 2–8 °C (41). The absolute value should be approved by the NRA.
The viability test should also be performed with the appropriate international
Reference Reagent or in-house reference for checking validity of the assay. One
method of determining the number of culturable particles should be adhered to,
as suggested in Part A, section 4.2.5.
The purpose of including the appropriate international Reference Reagent
or in-house reference is to check the quality and consistency of the
medium used for determination of the number of culturable particles. It
is not intended to adjust the count of the vaccine by comparison with
the Reference Preparation.
155
All manufacturers should keep their product for the approved storage period
and should determine the number of culturable particles from time to time to
demonstrate that the number is being maintained at an adequate level.
In some countries, the thermal stability test is carried out only after the
vaccine has been stored for 3–4 weeks after freeze-drying, since it is
considered that the degree of stability during the first three weeks may
not be related to the long-term stability of the product.
As a guide to stability, some manufacturers of freeze-dried BCG vaccine

determine the residual moisture content of the final vaccine, since failure
to achieve a certain degree of desiccation results in an unstable product.
However, such a test cannot be regarded as an alternative to tests involving
the determination of the number of culturable particles.
A.7 Records
The recommendations in section 8 of Good manufacturing practices for biological
products (28) should apply.
Written records should be kept of all seed lots, all cultures intended for
vaccine production, all single harvests, all final bulk vaccines, and all vaccine in
the final containers produced by the manufacturing establishments, including
all tests irrespective of their results.
The records should be of a type approved by the NRA. An example of a
suitable protocol is given in Appendix 2.
A.8 Retained samples

The recommendations in section 9.5 of Good manufacturing practices for
biological products (28) should apply.
It is desirable that samples should be retained for at least one year after the
expiry date of the final lot.
A.9 Labelling
The recommendations in section 7 of Good manufacturing practices for biological
products (28) should apply, including the following guidance.
The label, and/or the packaging insert in some countries, printed on or
affixed to each container should show the volume and nature of the diluent. Also,
this label, or the label on the carton holding several final containers, or the leaflet
accompanying the containers, should carry the following additional information:
■ the fact that the vaccine fulfils the requirements of this document;
■ instructions for use of the vaccine and information concerning
contraindications and the reactions that may follow vaccination;
156
Annex 3
■ the volume and nature of the diluent to be added to reconstitute

the vaccine, specifying that only the diluent supplied by the
manufacturer should be used;
■ the conditions recommended during storage and transport, with
information on the reduced stability of the vaccine if exposed to
temperatures higher than that stated on the label;
■ warnings that the vaccine should be protected from direct sunlight;
■ a statement that the reconstituted vaccine should be used as soon as
possible, or should be stored at 2–8 °C, protected from direct sunlight
and used within six hours (42);
■ information on antimicrobial sensitivity.
The label for the diluent should state “Reconstituting fluid for BCG
vaccine [proprietary name]”.
A.10 Distribution and transport

The recommendations given in section 8 of Good manufacturing practices for
biological products (28) should be followed, along with the guidance provided
in Safe vaccine handling, cold chain and immunizations (43). Further guidance
is provided in Model guidance for the storage and transport of time- and
temperature-sensitive pharmaceutical products (44).
It is desirable that samples should be retained for at least one year after
the expiry date of the final lot.
A.11 Stability, storage and expiry date

A.11.1 Stability testing
Adequate stability studies form an essential part of vaccine development. The
recommendations provided in WHO’s Guidelines for stability evaluation of
vaccines should be followed (46). Stability testing should be performed at different
stages of production if stored for a given time period, namely as appropriate for
single harvests or pool of single harvests, final bulk or final lot. Stability-indicating
parameters should be defined or selected appropriately according to the stage of
production. It is advisable to assign a storage period to all in-process materials
during vaccine production, particularly intermediates such as single harvests and
final bulk, and a shelf-life period to the final lots.
BCG vaccines require special precautions to ensure sufficient stability.
In this connection the most important measures are lyophilization, the use of an
effective stabilizer, and proper sealing of vaccine containers.
157
Historically the use of ampoules sealed under vacuum was the most
common practice for increasing stability. However, vacuum-sealing is
difficult compared to sealing in the presence of inert gas. There were
no significant differences between BCG vaccines sealed under vacuum
and under nitrogen or carbon dioxide at either 4 °C or 37 °C (41).
Manufacturers now prepare BCG vaccines in vials/ampoules and, under
well-validated conditions, the product is adequately stable.
A.11.2 Storage conditions

The Guideline for establishing or improving primary and intermediate vaccine
stores (47) should apply.
Storage conditions should be based on stability studies and approved by
the NRA. Before being distributed by the manufacturing establishment, or before
being issued from a depot for the storage of vaccine, all vaccines in their final
containers should be stored constantly at 2–8 °C (41, 48) and vaccine diluents
should be stored as recommended by the manufacturer. Freeze-dried BCG
vaccines, regardless of their substrain, are sensitive to ultraviolet and fluorescent
light. They should be protected from direct sunlight (41).
BCG vaccines are sensitive to light as well as to heat. Normally, these
vaccines are supplied in vials/ampoules made from dark brown glass,
which gives them some protection against light damage, but care should
still be taken to keep them covered and protected from strong light at all
times (48).
Freeze-dried BCG vaccines may be kept frozen at –15 °C to –25 °C if cold

chain space permits, but this is neither essential nor recommended (41).
Precautions should also be taken to maintain the vaccine during transport

and up to the time of use at the temperature and under the storage
conditions recommended by the manufacturer.
A.11.3 Expiry date

The expiry date should be approved by the NRA and should be based on the
stability of the final product, as well as on the results of the stability tests referred
to in section A.11.1 above. It is established for each batch by adding the shelf-life
period to the date of manufacture. Most freeze-dried BCG vaccines are stable at
temperatures of 2–8 °C for at least two years (41) from the date of manufacture.
The storage of final product at –20 °C to extend the shelf-life should be validated.
A.11.4 Expiry of reconstituted vaccine

Stability studies should be undertaken on reconstituted vaccine. Freeze-
dried BCG vaccines become much more heat-sensitive after they have been
158
Annex 3
reconstituted with diluent (41). After multidose containers of freeze-dried BCG

have been reconstituted, the vaccine should be used as soon as possible. Any
reconstituted vaccine remaining should be stored at 2–8 °C until used, and the
expiry time should be defined by stability studies (4, 42, 45).
Part B. Nonclinical evaluation of BCG vaccines

Details on the design, conduct, analysis and evaluation of nonclinical studies are
available in the WHO Guidelines on nonclinical evaluation of vaccines (49).
Nonclinical testing of a new strain (i.e. a strain derived by selection from
existing BCG strains in Appendix 1) or of a strain from a new manufacturer
of a BCG vaccine is a prerequisite for initiation of clinical studies in humans.
Nonclinical testing includes immunogenicity, protection studies (proof of
concept) and safety testing in animals. The vaccine lots used in nonclinical studies
should be adequately representative of the formulation intended for clinical
investigation and, ideally, should be the same lots, manufactured according to
current good manufacturing practice (cGMP), as those used in clinical studies.
If this is not feasible, the lots used clinically should be comparable to those used in
the nonclinical studies with respect to potency, stability and other characteristics
of quality. The technical manufacturing consistency lots may often be used for
these purposes.
New manufacturers of BCG vaccine for human use will need to refer to
the range of nonclinical safety and characterization tests that are recommended
for existing licensed BCG vaccines. Although there is currently no requirement
for additional nonclinical testing beyond that already described for licensed
BCG vaccines, the development of new variants of BCG, the potential for new
fermentation technologies and the possibility of novel live vaccines against TB
have shown that additional nonclinical studies beyond that required for licensed
BCG vaccine can be helpful in demonstrating that a new BCG product has
satisfactory nonclinical efficacy, safety and stability.
Guideline example on protective potency testing: Hartley guinea-pigs
are used for potency testing. The guinea-pigs are vaccinated with a
small amount of BCG (~10 3 colony-forming units (CFU)). Eight weeks
after the vaccination, the guinea-pigs are challenged with virulent
M. tuberculosis H37Rv (ATCC 27294) by the pulmonary route with a
low dose (10–15 CFU) per animal. Five weeks after the infection, the
guinea-pigs are killed and the spleen and the lung lobes are removed.
These organs are homogenized separately. Appropriate dilutions are
inoculated on to duplicate solid medium and incubated at 37 °C for three
weeks. The number of M. tuberculosis H37Rv colonies is counted, and is
expressed as mean log10 CFU per tissue. The CFU results are compared
between the vaccinated and non-vaccinated groups (50).
159
If there are two pharmacologically relevant species for the clinical candidate (one
rodent and one non-rodent), both species should be used for short-term (up to one
month duration) toxicology studies. If the toxicological findings from these studies
are similar in both species, longer-term studies in one species are usually considered
sufficient; the rodent species should be considered unless there is a rationale for
using non-rodents. Studies in two non-rodent species are not appropriate. Other
in vivo studies should address both potency (such as tuberculin sensitivity and
immunological tests) and safety issues (such as tests for excessive dermal reactivity
and absence of virulent mycobacteria) of the classical BCG vaccines.
It may be of benefit for new BCG vaccine developers to consider the
points raised in recent meetings establishing recommendations for new live
vaccines against TB (51, 52).
Part C. Clinical evaluation of BCG vaccines

Clinical trials should adhere to the principles described in the Guidelines for
good clinical practice (GCP) for trials on pharmaceutical products (53) and
the general principles described in WHO’s Guidelines on clinical evaluation of
vaccines: regulatory expectations (54). All clinical trials should be approved by the
relevant NRAs and local ethics committees. Continued licence of BCG vaccines
should be viewed in the light of ongoing post-marketing data on the safety,
immunogenicity and effectiveness of BCG vaccines in the target population.
Part C considers the provision of clinical data required (a) when a new
candidate “classical” BCG vaccine derived from (the same master seed of) one of
the recognized strains (see Appendix 1) is developed; (b) when there have been
major changes to the manufacturing process of an established vaccine, including
preparation of a new master seed lot of an established strain; (c) when technology
transfer of existing vaccine is planned to a new manufacturer; and (d) when
revalidation of existing vaccines used in national immunization programmes is
considered.
Vaccines manufactured using a “new strain” (i.e. a strain derived by

selection from existing BCG strains in Appendix 1) should require a full clinical
development programme that provides evidence of safety, efficacy and the
reactogenicity profile in all target age-groups.
Other vaccines against M. tuberculosis derived from M. bovis or other
mycobacterial strains cannot be considered as BCG. They would require a full
clinical development programme and are not included here.
C.1 General considerations

C.1.1 Comparative or placebo‑controlled clinical trials
It would not be considered ethical to conduct a placebo-controlled trial of
protective efficacy of a BCG vaccine in a TB-endemic area, particularly in
160
Annex 3
infants. A comparative trial with a licensed, or internationally accepted (WHO

prequalified), BCG vaccine could be accepted.
C.1.2 Value of PPD response

It is recognized that the response to tuberculin PPD is not an indicator of a
protective immune response. Nonetheless this has been used for more than
50 years to indicate a cellular immune response to an infection with M. tuberculosis
or as evidence of “successful” BCG vaccination. At best, a PPD reaction is an
indicator of exposure to antigens of TB, and the generation of a cellular immune
response. Thus, it can be used in a PPD-naive population as an indicator of an
immune response to the BCG vaccine (55). Other immunological measures may
be more closely related to M. tuberculosis infection or vaccination, but currently
none has been agreed as a correlate of protection from infection or disease.
C.1.3 BCG in HIV‑infected infants

A very important safety consideration with regard to vaccination policy is to
establish, during clinical trials, the potential for disseminated BCG disease in
immunocompromised children. In this regard, the use of BCG vaccines at birth
should follow the recommendations from the GACVS (13, 14). The GACVS
recommendations consider the policies for immunization exclusion of infants
known to be infected with HIV, infants symptomatic for HIV infection, and
those infants born to mothers known to be HIV-infected and who therefore may
be infected.
C.1.4 Post‑vaccination reactions and complications

Vaccines intended for intradermal or percutaneous injection should be given
strictly intradermally or percutaneously, and vaccinators should be trained
accordingly. Incorrect vaccination technique can result in adverse reactions,
including discharging ulcers, abscesses and keloid scars.
Current BCG vaccines have a known reactogenicity profile after
intradermal inoculation (56). Local reaction at the vaccination site is normal
after a BCG vaccination. It may take the form of a nodule that, in many cases,
will break down and suppurate. The reaction developing at the vaccination
site usually subsides within 2–5 months and in practically all children leaves a
superficial scar of 2–10 mm in diameter. The nodule may persist and ulcerate.
Swelling of regional lymph nodes may also be seen, and this may be regarded as
a normal reaction, but the size should be limited.
Keloid and lupoid reactions may occur at the site of the vaccination.
Children with such reactions should not be revaccinated. Inadvertent subcutaneous
injections produce abscess formations and may lead to ugly retracted scars.
Among the major complications, suppurative lymphadenitis has been observed.
161
In the case of certain vaccines, it has been revealed that there is a strong correlation
between the incidence of these complications in newborns and the number of
culturable particles in the vaccine.
The concentration of the vaccine should be shown to be effective and
tolerated in the age groups for which the vaccine is intended.
A reduction of the dose for newborns may be based on the evidence and
approved by the NRA (57).
The NRA should issue guidelines for the treatment of complications.
C.2 Special considerations

This section is limited to the clinical development of new “classical” BCG vaccines
manufactured following these Recommendations and using strains of BCG that
are derived from (the same master seed of) one of the strains recognized in
Appendix 1.
The use of comparative studies with a licensed BCG vaccine can provide
evidence of the similarity of safety and immune responses to a new classical BCG
vaccine product.
The target population for the vaccine would be newborns or infants
according to current recommendations on the use of BCG vaccines.
The nonclinical expectations for a new classical BCG vaccine are outlined
in Part B.
For such a new classical BCG vaccine, these nonclinical studies should be
conducted in comparison to an existing licensed BCG vaccine, preferably
derived from the same BCG substrain. It would be expected that the
results of preclinical studies would be similar for the new vaccine product
and for the comparator.
The clinical development programme should ideally be designed to show the

safety and protective efficacy of the vaccine. However, for such a new classical
BCG vaccine product, comparative studies with an existing licensed BCG vaccine,
using immunological responses as a marker for efficacy, may be acceptable to the
responsible NRA.
Comparable PPD response (proportion of PPD converters, intensity of
response) may be acceptable.
Clinical studies should provide evidence of safety in all the potential target
populations, including those with a high incidence of diseases that may affect the
safety or efficacy of the new vaccine product. In phase I and phase II studies this
should include evaluation of:
■ safety and reactogenicity in healthy adults (comparative);
162
Annex 3
■ end-points;
■ safety and reactogenicity – can include healthy HIV-infected adults;
■ immune responses – non-inferior PPD response, and may include
other immunological markers.
These studies are difficult to interpret as adults will most likely have
received BCG vaccination at birth. Dose-finding studies may be considered
unnecessary for these vaccines. The safety in HIV-infected individuals
and in infants needs to be considered.
Dose-finding and age de-escalation can be included in these studies, but review
at each step by a suitable independent safety committee should be considered.
In phase III studies evaluation should be made of:
■ safety and reactogenicity in infants (comparative)
■ end-points
■ safety and reactogenicity
■ non-inferior PPD immune response.
Post‑marketing risk management

As it may not be practically possible to evaluate protective efficacy for a new
classical BCG vaccine, the responsible NRA in the country of manufacture
should require post-marketing surveillance activities for safety and effectiveness
in a suitable environment. Sentinel surveillance sites in an endemic country may
be considered.
In the past, the responsible NRA of a country of manufacture required a
demonstration that adequate control of BCG vaccine had been achieved,
by arranging for studies of some of the final lots to be performed at
regular intervals in children.
Studies of immunological responses to M. tuberculosis antigens should be

made, such as sensitivity to tuberculin. In at least 100 tuberculin-negative
persons per year, records of tuberculin-induced reaction (distribution of
tuberculin reactions by size) by a defined dose of tuberculin3 in vaccines,
local skin lesions (nature and size of reaction at injection site), and the
occurrence of untoward vaccination reactions should be obtained. Such
tests should be performed in parallel on two or more vaccine lots in
the same population group, one of the vaccine lots being preferably a
reference vaccine.
3
An intradermal test with a dose of tuberculin equivalent to 5 IU of tuberculin PPD is suitable. A description
of an appropriate method and a design for a study to assess BCG vaccines in humans are available on
application to the World Health Organization, 1211 Geneva 27, Switzerland.
163
C.3 Post‑marketing surveillance

The NRA in the country of manufacture may require periodic updates of reports
on BCG safety and immunogenicity.
C.3.1 BCG vaccine used in a national immunization programme

As in all immunization programmes, the adverse events following immunization
with BCG vaccines should be monitored and recorded.
The following events are important after BCG vaccination (58):
■ injection site abscesses

■ BCG lymphadenitis
■ disseminated BCG diseases
■ osteitis/osteomyelitis.
Appropriate training of health-care workers is important as some medical

incidents can be related to immunization even if they have a delayed onset.
C.3.2 WHO prequalified BCG vaccines

Prequalified vaccines may be used in a wide range of countries worldwide.
Periodic safety update reports supplied to WHO should include specific analysis
of countries where the vaccine has been used.
Part D. Guidelines for NRAs

D.1 General
The general recommendations for NRAs provided in the Guidelines for national
authorities on quality assurance for biological products (59) should apply. These
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. The detailed production and control procedures, as well
as any significant change in them that may affect the quality, safety or efficacy
of BCG vaccine, should be discussed with and approved by the NRA. For
control purposes, the NRA should obtain the WHO International Reference
Reagents as comparators for potency-related testing and, where necessary,
should establish national working Reference Preparation(s) calibrated against
the international reference.
In addition, the NRA should provide a reference vaccine or approve
one used by a manufacturer, and should give directions concerning the use of
the reference vaccine in specified tests. The NRA should also give directions to
164
Annex 3
manufacturers concerning the BCG substrain to be used in vaccine production,

the total content of bacteria, the number of culturable particles, and the stability
required of the vaccine, and should specify the requirements to be fulfilled by
the manufacturer in accordance with the provisions of Part A of this document,
including those for consistency of quality in respect of the points referred to in
Part A, section 2.
D.1.1 BCG vaccine strain

The substrain of BCG (maintained in the form of a seed lot) used in the
production of vaccine should be derived from the original strain maintained by
Calmette and Guérin and should be identified by historical records that include
information on its origin and subsequent manipulation. On the basis of cultures
and biochemical and animal tests, the BCG seed lot should show characteristics
that conform to those of BCG and generally differ from those of other
mycobacteria. The identity test should be supplemented by molecular biology
techniques to identify the specific BCG substrain used. The seed lot should show
consistency in the morphological appearance of colonies and genetic stability on
serial subculture. It should also have been shown to yield vaccines that, upon
administration by intradermal injection to children and adults, induce relevant
immunological responses to M. tuberculosis antigens, including sensitivity to
tuberculin, and with a low frequency of untoward effects. In addition, the seed lot
should have been shown to give adequate protection against TB in experimental
animals in tests for protective potency.
D.1.2 Concentration of BCG vaccine

The concentration of BCG vaccine varies with different vaccine products and is
dependent on a number of factors, such as the substrain of BCG used and the
method of manufacture. It is therefore essential for each manufacturer, as well
as for each method of manufacture, for the optimum potency of vaccine to be
ascertained by trials in tuberculin-negative subjects (newborns, older children,
and adults) to determine the response to vaccination in respect of the induction
of relevant immunological responses to M. tuberculosis antigens – including
sensitivity to tuberculin, the production of acceptable local skin lesions, and the
occurrence of a low frequency of untoward reactions. As a result of such trials, the
NRA should give directions to the manufacturer concerning the total bacterial
content and the number of culturable particles required for the vaccine.
If a manufacturer changes its procedure for preparing BCG vaccine, and
if the NRA considers that the change might affect the final product, it may
be necessary to conduct further clinical trials in order to determine the
optimum content of BCG organisms in the new product.
165
D.2 Release and certification

A vaccine lot should be released only if it fulfils the national requirements
and/or Part A of these Recommendations. Before any vaccine lot is released
from a manufacturing establishment, the recommendations for consistency
of production provided in the Guidelines for national authorities on quality
assurance for biological products (59) should be met. In addition, the general
recommendations for NRAs provided in the Guidelines for independent lot
release of vaccines by regulatory authorities (60) should be followed. A protocol
based on the model given in Appendix 2, signed by the responsible official of the
manufacturing establishment, should be prepared and submitted to the NRA in
support of a request for release of a vaccine for use.
A statement signed by the appropriate official of the NRA (or authority as
appropriate) 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 date of manufacture, 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 certificate should be based on
the model given in Appendix 3. The purpose of the certificate is to facilitate the
exchange of vaccines between countries.
Authors and acknowledgements

The scientific basis for the revision of the Requirements published in WHO
Technical Report Series, Nos. 745 and 771 was developed at meetings from 2003
to 2007 attended by:
Dr L. Barker, Aeras Global Tuberculosis Vaccine Foundation, Rockville,
MD, USA; Professor M. Behr, Montreal General Hospital, Quebec, Canada; Dr T.

Bektimirov, Tarassevich State Research Institute for Standardization and Control
of Medical Biological Preparations, Moscow, Russian Federation; Dr T. Brewer,
Brigham and Women’s Hospital, Boston, MA, USA; Dr R. Brosch, Institut
Pasteur, Paris, France; Dr L.R. Castello-Branco, Fundacao Ataulpho de Paiva,
Brazilian League Against Tuberculosis, Rio de Janeiro, Brazil; Dr M. Chouchkova,
National Center of Infectious and Parasitic Diseases, Sofia, Bulgaria; Dr S. Cole,
Institut Pasteur, Paris, France; Dr M. Corbel, National Institute for Biological
Standards and Control, Potters Bar, England; Dr J. Darbyshire, MRC Clinical
Trials Unit, London, England; Dr J. Daviaud, World Health Organization,
Geneva, Switzerland; Dr H. Dockrell, London School of Hygiene and Tropical
Medicine, London, England; Dr U. Fruth, World Health Organization, Geneva,
Switzerland; Mr I. Fujita, Japan BCG Laboratory, Tokyo, Japan; Dr M. Gheorghiu,
166
Annex 3
Institut Pasteur, Paris, France; Dr E. Griffiths, Health Canada, Ottawa, Canada;

Dr C. Gutierrez Pèrez, Institut Pasteur, Paris, France; Dr K. Haslov, Statens Serum
Institut, Copenhagen, Denmark; Dr M.M. Ho, National Institute for Biological
Standards and Control, Potters Bar, England; Dr S. Jadhav, Serum Institute of
India Ltd, Pune, India; Dr S.E. Jensen, Statens Serum Institut, Copenhagen,
Denmark; Dr I. Knezevic, World Health Organization, Geneva, Switzerland;
Dr M. Lagranderie, Institut Pasteur, Paris, France; Mr F. Leguellec, Aventis Pasteur,
Val de Reuil, France; Dr Y. Lopez Vidal, Universidad Nacional Autonoma de
Mexico, Mexico City, Mexico; Dr G. Marchal, Institut Pasteur, Paris, France; Dr H.
Mayanja-Kizza, Makarere Medical School, Kampala, Uganda; Dr S. Mostowy,
McGill University Health Centre, Montreal General Hospital, Montreal, Quebec,
Canada; Dr M. Roumaintzeff, Lyons, France; Dr M. Seki, BCG Laboratory,
Tokyo, Japan; Dr V. Vincent, Institut Pasteur, Paris, France; Dr F. Weichold, Aeras
Global Tuberculosis Vaccine Foundation, Rockville, MD, USA; Dr D. Wood,
World Health Organization, Geneva, Switzerland; Dr S. Yamamoto, National
Institute of Infectious Diseases, Tokyo, Japan; Dr I. Yano, Japan BCG Laboratory,
Tokyo, Japan; Dr D. Young, Imperial College of Science, Technology & Medicine,
London, England; Dr T. Zhou, World Health Organization, Geneva, Switzerland.
The first draft was prepared by Dr H-N. Kang, World Health Organization,
Geneva, Switzerland, taking into account comments received from:
Dr M. Andre, Agence Française de Sécurité Sanitaire des Produits de Santé,
Lyons, France; Dr I.S. Budiharto, Bio Farma, Bandung, Indonesia; Dr H. Chun,
Korea Food and Drug Administration, Seoul, Republic of Korea; Dr M. Corbel,
National Institute for Biological Standards and Control, Potters Bar, England;
Dr R. Dobbelaer, Lokeren, Belgium; Dr S. Gairola, Serum Institute of India,
Hadapsar, India; Dr L. Grode, Vakzine Projekt Management GmbH, Hannover,
Germany; Dr M.M. Ho, National Institute for Biological Standards and Control,
Potters Bar, England; Dr S. Jadhav, Serum Institute of India Ltd, Pune, India;
Professor D. Levi, Tarassevich State Research Institute for Standardization and
Control of Medical Biological Preparations, Moscow, Russian Federation; Dr S.
Morris, United States Food and Drug Administration, Silver Spring, MD, USA;
Dr V. Öppling, Paul-Ehrlich-Institute, Langen, Germany; Dr M. Roumiantzeff,
Lyons, France; Dr K. Shibayama, National Institute of Infectious Diseases,
Tokyo, Japan; Dr J. Southern, Cape Town, South Africa; Mr A. Supasansatorn,
Ministry of Public Health, Nonthaburi, Thailand; Dr S. Wahyuningsih, National
Agency of Drug and Food Control, Jakarta Pusat, Indonesia; Dr K.B. Walker,
National Institute for Biological Standards and Control, Potters Bar, England;
Dr S. Yamamoto, National Institute of Infectious Diseases, Tokyo, Japan; Mrs A.
Zhao, National Institute for the Control of Pharmaceutical & Biological Products,
Beijing, China.
Following the informal consultation meeting on standardization and
evaluation of BCG vaccines in September 2009, held in Geneva, Switzerland, draft
167
Recommendations were revised by Dr H.-N. Kang, World Health Organization,

Geneva, Switzerland, taking into account information on current manufacturing
and regulatory practice provided at the meeting attended by:
Dr M. Andre, Agence Française de Sécurité Sanitaire des Produits
de Santé, Lyons, France; Dr L.F. Barker, Aeras Global Tuberculosis Vaccine
Foundation, Rockville, MD, USA; Dr A. Bhardwaj, Central Research Institute,
Kasuli, Himachal Pradesh, India; Dr M. Brennan, Aeras Global TB Vaccine
Foundation, Rockville, MD, USA; Dr I.S. Budiharto, Bio Farma, Bandung,
Indonesia; Dr L.R. Castello-Branco, Fundacao Ataulpho de Paiva, Brazilian
League Against Tuberculosis, Rio de Janeiro, Brazil; Dr M. Chouchkova, National
Center of Infectious and Parasitic Diseases, Sofia, Bulgaria; Dr H. Chun, Korea
Food and Drug Administration, Seoul, Republic of Korea; Dr M. Corbel, National
Institute for Biological Standards and Control, Potters Bar, England; Dr B. Eisele,
Vakzine Projeckt Management GmbH, Hannover, Germany; Dr S. Gairola, Serum
Institute of India, Hadapsar, India; Dr L. Grode, Vakzine Projeckt Management
GmbH, Hannover, Germany; Dr K. Haslov, Statens Serum Institut, Copenhagen,
Denmark; Dr M.M. Ho, National Institute for Biological Standards and Control,
Potters Bar, England; Dr P. Hubrechts, Statens Serum Institut, Copenhagen,
Denmark; Dr G. Hussey, South African TB Vaccine Initiative, Cape Town, South
Africa; Dr H.-N. Kang, World Health Organization, Geneva, Switzerland; Dr I.
Knezevic, World Health Organization, Geneva, Switzerland; Professor D. Levi,
Tarassevich State Research Institute for Standardization and Control of Medical
Biological Preparations, Moscow, Russian Federation; Dr Y. Lopez Vidal,
Universidad Nacional Autonoma de Mexico, Copilco-Universidad, Mexico City,
Mexico; Dr K. Markey, National Institute for Biological Standards and Control,
Potters Bar, England; Dr C. Martin, Universidad de Zaragoza, Zaragoza, Spain;
Dr S. Morris, FDA/CBER, Bethesda, MD, USA; Dr V. Öppling, Paul-Ehrlich-
Institute, Langen, Germany; Dr M. Roumiantzeff, Lyons, France; Dr. M. Seki,
Japan BCG Laboratory, Tokyo, Japan; Dr K. Shibayama, National Institute of
Infectious Diseases, Tokyo, Japan; Dr J. Southern, Advisor to Medicines Control
Council in South Africa, Cape Town, South Africa; Mr A. Supasansatorn, Ministry

of Public Health, Nonthaburi, Thailand; Dr S. Wahyuningsih, National Agency
of Drug and Food Control, Jakarta Pusat, Indonesia; Dr K.B. Walker, National
Institute for Biological Standards and Control, Potters Bar, England; Dr S.
Yamamoto, Japan BCG Laboratory, Tokyo, Japan; Dr Z. Yan, Chengdu Institute
of Biological Products, Chengdu, China; Dr L. Zhang, Chengdu Institute of
Biological Products, Chengdu, China; Mrs A. Zhao, National Institute for the
Control of Pharmaceutical & Biological Products, Beijing, China.
Several draft Recommendations were subsequently prepared by Dr M.M.
Ho, National Institute for Biological Standards and Control, Potters Bar,
England, with support from the drafting group of Dr M. Corbel, Milton Keynes,
England; Dr R. Dobbelaer, Lokeren, Belgium; Dr H.-N. Kang, World Health
168
Annex 3
Organization, Geneva, Switzerland; Dr J. Southern, Cape Town, South Africa;

and Dr K.B. Walker, National Institute for Biological Standards and Control,
Potters Bar, England.
Following the meeting of the drafting group in March 2011, held in
Potters Bar, England, draft Recommendations were updated taking into account
comments received from:
Dr M. Andre, Agence Française de Sécurité Sanitaire des Produits de
Santé, Lyons, France; Dr M. Chouchkova, National Center of Infectious and
Parasitic Diseases, Sofia, Bulgaria; Dr H. Dockrell, London School of Hygiene
and Tropical Medicine, London, England; Dr S. Gairola, Serum Institute of
India, Hadapsar, India; Dr P. Hubrechts, Statens Serum Institut, Copenhagen,
Denmark; Dr J. Joung, Korea Food and Drug Administration, Chungchengbuk-do,
Republic of Korea; Dr M. Lagranderie, Institut Pasteur, Paris, France; Professor
D. Levi, Tarassevich State Research Institute for Standardization and Control of
Medical Biological Preparations, Moscow, Russian Federation; Dr Y. Lopez Vidal,
Universidad Nacional Autonoma de Mexico, Facultad de Medicina, Mexico
City, Mexico; Dr V. Öppling, Paul-Ehrlich-Institute, Langen, Germany; Mrs G.
Trisnasari, Bio Farma, Bandung, Indonesia; Dr V. Vincent, Institut Pasteur, Paris,
France; Dr S. Wahyuningsih, National Agency of Drug and Food Control, Jakarta
Pusat, Indonesia; Dr T. Ying, China National Biotec Group, Beijing, China; Mrs A.
Zhao, National Institute for the Control of Pharmaceutical & Biological Products,
Beijing, China.
The draft Recommendations were posted on the WHO Biologicals web
site for public consultation from 24 May to 23 June 2011. The WHO/BS/2011.2157
document was then prepared by Dr M.M. Ho, National Institute for Biological
Standards and Control, Potters Bar, England; Dr H.-N. Kang, World Health
Organization, Geneva, Switzerland; Dr J. Southern, Cape Town, South Africa;
Dr K.B. Walker, National Institute for Biological Standards and Control, Potters Bar,
England, taking into account comments received from the following reviewers:
Dr C. Ahn, Korea Food and Drug Administration, Chungchengbuk-do,
Republic of Korea; Dr M. Andre, Agence Française de Sécurité Sanitaire des
Produits de Santé, Lyons, France; Dr M. Brennan, Aeras Global TB Vaccine
Foundation, Rockville, MD, USA; Dr I.S. Budiharto, Bio Farma, Bandung,
Indonesia; Ms Y. Choi, Korea Food and Drug Administration, Chungchengbuk-do,
Republic of Korea; Dr M. Chouchkova, National Center of Infectious and Parasitic
Diseases, Sofia, Bulgaria; Dr S. Gairola, Serum Institute of India, Hadapsar, India;
Dr K. Haslov, Statens Serum Institut, Copenhagen, Denmark; Dr H. Hozouri,
Pasteur Institute of Iran, Tehran, Islamic Republic of Iran; Dr P. Hubrechts,
Statens Serum Institut, Copenhagen, Denmark; Dr J. Joung, Korea Food and Drug
Administration, Chungchengbuk-do, Republic of Korea; Mr H. Kim, Korea Food
and Drug Administration, Chungchengbuk-do, Republic of Korea; Mr J.-G. Kim,
Korea Food and Drug Administration, Chungchengbuk-do, Republic of Korea;
169
Ms Y.L. Kim, Korea Food and Drug Administration, Chungchengbuk-do, Republic

of Korea; Dr K. Lee, Korea Food and Drug Administration, Chungchengbuk-do,
Republic of Korea; Dr Y. Lopez Vidal, Universidad Nacional Autonoma de
Mexico, Copilco-Universidad, Mexico City, Mexico; Dr V. Öppling, Paul-Ehrlich-
Institute, Langen, Germany; Dr M. Roumiantzeff, Lyons, France; Dr M. Seki,
Japan BCG Laboratory, Tokyo, Japan; Dr J. Shin, World Health Organization,
Geneva, Switzerland; Mr S.-C. Shin, Green Cross, Yongin, Republic of Korea;
Dr V. Vincent, Institut Pasteur, Paris, France; Dr S. Yamamoto, Japan BCG
Laboratory, Tokyo, Japan.
Further changes were made to WHO/BS/2011.2157 by the Expert
Committee on Biological Standardization, resulting in the current document.
References
1. Requirements for dried BCG vaccines. In: WHO Expert Committee on Biological Standardization.
Thirty-sixth report. Geneva, World Health Organization, 1987 (WHO Technical Report Series,
No. 745), Annex 2.
2. Requirements for dried BCG vaccines. In: WHO Expert Committee on Biological Standardization.
Thirty-eighth report. Geneva, World Health Organization, 1988 (WHO Technical Report Series,
No. 771), Annex 12.
3. Corbel MJ et al. Report on a WHO consultation on the characterization of BCG strains, Imperial
College, London 15–16 December 2003. Vaccine, 2004, 22:2675–2680.
4. Ho MM et al. Report on a WHO consultation on the characterization of BCG vaccines held at
WHO Headquarters, Geneva, 8–9 December 2004. Vaccine, 2005, 23:5700–5704.
5. Knezevic I, Corbel MJ. WHO discussion on the improvement of the quality control of BCG vaccines
held at Pasteur Institute, Paris, France, 7 June 2005. Vaccine, 2006, 24:3874–3877.
6. Ho MM et al. Report on a WHO informal consultation on standardization and evaluation on BCG
vaccines, 22–23 September 2009, WHO, Geneva, Switzerland. Vaccine, 2010, 28:6945–6950.
7. State of the art of new vaccine research and development. Geneva, World Health Organization,
2006 (WHO/IVB/06.01; http://whqlibdoc.who.int/hq/2006/WHO_IVB_06.01_eng.pdf, accessed
6 February 2013).
8. Fine PEM et al. Issues relating to the use of BCG in immunization programmes: a discussion
document. Geneva, World Health Organization, 1999 (WHO/V&B/99.23; http://whqlibdoc.who.int/
hq/1999/WHO_V&B_99.23.pdf, accessed 6 February 2013).
9. BCG vaccine: WHO position paper. Weekly Epidemiological Record, 2004, 79:27–38.
10. Colditz GA et al. Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the
published literature. Journal of the American Medical Association, 1994, 271:698–702.
11. Global Advisory Committee on Vaccine Safety, 12–13 June 2007. Weekly Epidemiological Record,
2007, 82:252–259.
12. Ritz N et al. Influence of BCG vaccine strain on the immune response and protection against
tuberculosis. FEMS Microbiology Review, 2008, 32:821–841.
13. Revised BCG vaccination guidelines for infants at risk for HIV infection. Weekly Epidemiological
Record, 2007, 82:193–196.
170
Annex 3
14. Use of BCG vaccine in HIV-infected infants. Weekly Epidemiological Record, 2010, 85:32–33.
15. Birkhaug K. Protective value of the intracutaneous and percutaneous methods of BCG vaccination.
Acta Medica Scandinavica, 1944, CXVII, fasc. III–IV:274–312.
16. Mande R, Fillastre C, Pinelli L. A propos des complications de la vaccination B.C.G. Annales de
Pédiatrie, 1976, 23:219–225.
17. Mori T, Yamauchi Y, Shiozawa K. Lymph node swelling due to bacilli Calmette-Guérin vaccination
with multipuncture method. Tubercle and Lung Disease, 1996, 77:269–273.
18. Brosch R et al. Genome plasticity of BCG and impact on vaccine efficacy. Proceedings of the
National Academy of Sciences, 2007, 104:5596–5601.
19. Seki M et al. Whole genome sequence analysis of Mycobacterium bovis bacillus Calmette-Guérin
(BCG) Tokyo 172: a comparative study of BCG vaccine substrains. Vaccine, 2009, 27:1710–1716.
20. Gomes LHF et al. Genome sequence of Mycobacterium bovis BCG Moreau, the Brazilian vaccine
strain against tuberculosis. Journal of Bacteriology, 2011, 193:5600–5601.
21. Castillo-Rodal AI et al. Mycobacterium bovis BCG substrains confer different levels of protection
against Mycobacterium tuberculosis infection in a BALB/c model of progressive pulmonary
tuberculosis. Infection and Immunity, 2006, 74:1718–1724.
22. Hayashi D et al. Comparable studies of immunostimulating activities in vitro among
Mycobacterium bovis bacillus Calmette-Guerin (BCG) substrains. FEMS Immunology and Medical
Microbiology, 2009, 56:116–123.
23. Horwitz MA et al. Commonly administered BCG strains including an evolutionarily early strain
and evolutionarily late strains of disparate genealogy induce comparable protective immunity
against tuberculosis. Vaccine, 2009, 27:441–445.
24. In vitro assays of BCG products. Geneva, World Health Organization, 1977 (WHO working
document WHO/TB/Technical guide/77.9).
25. Biological substances and reference reagents. In: WHO Expert Committee on Biological
Standardization. Sixtieth report. Geneva, World Health Organization, 2013 (WHO Technical Report
Series, No. 977), Annex 4.
26. Biological substances: International standards and reference reagents. In: WHO Expert Committee
on Biological Standardization. Sixty-first report. Geneva, World Health Organization, 2013 (WHO
Technical Report Series, No. 978), Annex 7.
27. Good manufacturing practices: main principles for pharmaceutical products. In: WHO Expert
Committee on Specifications for Pharmaceutical Preparations. Forty-fifth report. Geneva, World
Health Organization, 2011 (WHO Technical Report Series, No. 961), Annex 3.
28. Good manufacturing practices for biological products. In: WHO Expert Committee on Biological
Standardization. Forty-second report. Geneva, World Health Organization, 1992 (WHO Technical
Report Series, No. 822), Annex 1.
29. International Commission on Radiological Protection. Report of Committee II on permissible dose
for internal radiation. London, Pergamon Press, 1959 (ICRP Publication 2).
30. United Nations Scientific Committee on the Effect of Atomic Radiation. Ionizing radiation: levels
and effects. Report. New York, NY, United Nations, 1972.
31. 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 (WHO Technical Report Series, No. 530),
Annex 4.
32. Guidelines on transmissible spongiform encephalopathies in relation to biological and
pharmaceutical products. Geneva, World Health Organization, 2003 (WHO/BCT/QSD/2003.01).
171
33. Recommendations for the evaluation of animal cell cultures as substrates for the manufacture
of biological medicinal products and for the characterization of cell banks. In: WHO Expert
Committee on Biological Standardization. Sixty-first report. Geneva, World Health Organization,
2013 (WHO Technical Report Series, No. 978), Annex 3.
34. WHO Guidelines on tissue infectivity distribution in transmissible spongiform encephalopathies.
Geneva, World Health Organization, 2006 (http://www.who.int/bloodproducts/tse/WHO%20
TSE%20Guidelines%20FINAL-22%20JuneupdatedNL.pdf, accessed 11 April 2011).
35. WHO tables on guidelines on tissue infectivity distribution in transmissible spongiform
encephalopathies. Geneva, World Health Organization, 2010 (http://www.who.int/bloodproducts/
tablestissueinfectivity.pdf, accessed 11 April 2011).
36. Update of the opinion on TSE infectivity distribution in ruminant tissues. Brussels, European
Commission, 2002 (http://ec.europa.eu/food/fs/sc/ssc/out296_en.pdf, accessed 11 April 2011).
37. Note for guidance on minimizing the risk of transmitting animal spongiform encephalopathy
agents via human and veterinary medicinal products (EMEA/410/01 Rev. 2 - October 2003)
adopted by the Committee for Proprietary Medicinal Products (CPMP) and by the Committee for
Veterinary Medicinal Products (CVMP). Official Journal of the European Union, 2004 (http://www.
emea.europa.eu/docs/en_GB/document_library/Scientific_guideline/2009/09/WC500003700.
pdf, accessed 24 February 2013).
38. Ho MM et al. Report of an international collaborative study to establish the suitability of using
modified ATP assay for viable count of BCG vaccine. Vaccine, 2008, 26(36):4754–4757.
39. Jensen SE et al. Development and validation of an ATP method for rapid estimation of viable units
in lyophilised BCG Danish 1331 vaccine. Biologicals, 2008, 36(5):308–314.
40. BCG vaccine, freeze-dried. In: European Pharmacopoeia, 6th ed. Strasbourg, Directorate for the
Quality of Medicines of the Council of Europe (EDQM), 2008 (01/2008:0163).
41. Temperature sensitivity of vaccines. Geneva, World Health Organization, 2006 (WHO/IVB/06.10).
42. 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; http://www.who.int/
vaccines-documents/DocsPDF99/www9924.pdf, accessed 11 April 2011).
43. Safe vaccine handling, cold chain and immunizations. Geneva, World Health Organization, 1998
(WHO/EPI/LHIS/98.02).
44. Model guidance for the storage and transport of time- and temperature-sensitive pharmaceutical
products (jointly with the Expert Committee on Biological Standardization). In: WHO Expert
Committee on Specifications for Pharmaceutical Preparations. Forty-fifth report. Geneva, World
Health Organization, 2011 (WHO Technical Report Series, No. 961), Annex 9.
45. Proper handling and reconstitution of vaccines avoids programme errors. Vaccines and Biologicals
Update, 2000, 34 (http://www.who.int/vaccines-documents/DoxNews/updates/updat34e.pdf,
accessed 11 April 2011).
46. Guidelines for stability evaluation of vaccines. In: WHO Expert Committee on Biological
Standardization. Fifty-seventh report. Geneva, World Health Organization, 2011 (WHO Technical
47. Guideline for establishing or improving primary and intermediate vaccine stores. Geneva, World
Health Organization, 2002 (WHO/V&B/02.34; http://www.who.int/vaccines-documents/DocsPDF02/
www715.pdf, accessed 11 April 2011).
48. Immunization in practice: a practical resource guide for health staffs, 2004 update. Geneva, World
Health Organization, 2004 (WHO/IVB/04.06).
172
Annex 3
49. WHO guidelines on nonclinical evaluation of vaccines. In: WHO Expert Committee on Biological
Standardization. Fifty-fourth report. Geneva, World Health Organization, 2005 (WHO Technical
50. Yamamoto T et al. Mycobacterium bovis BCG vaccination modulates TNF-a production after
pulmonary challenge with virulent Mycobacterium tuberculosis in guinea pigs. Tuberculosis, 2007,
87:155–165.
51. Walker KB et al. The second Geneva consensus: recommendations for novel live TB vaccines.
Vaccine, 2010, 28:2259–2270.
52. Kamath AT et al. New live mycobacterial vaccines: the Geneva consensus on essential steps
towards clinical development. Vaccine, 2005, 23:3753–3761.
53. Guidelines for good clinical practice (GCP) for trials on pharmaceutical products. In: WHO Expert
Committee on the Use of Essential Drugs. Sixth report. Geneva, World Health Organization, 1995
(WHO Technical Report Series, No. 850), Annex 3.
54. Guidelines on clinical evaluation of vaccines: regulatory expectations. In: WHO Expert Committee
on Biological Standardization. Fifty-second report. Geneva, World Health Organization, 2004 (WHO
Technical Report Series, No. 924), Annex 1.
55. Roth A et al. Vaccination technique, PPD reaction and BCG scarring in a cohort of children born in
Guinea-Bissau 2000–2002. Vaccine, 2005, 23:3991–3998.
56. Hoft DF et al. Clinical reactogenicity of intradermal bacille Calmette-Guérin vaccination. Clinical
Infectious Disease, 1999, 28:785–790.
57. WHO Expert Committee on Tuberculosis. Ninth report. Geneva, World Health Organization, 1974
(WHO Technical Report Series, No. 552).
58. Surveillance of adverse events following immunization. Geneva, World Health Organization, 1997
(WHO/EPI/TRAM/93.02. Rev 1).
59. 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 (WHO Technical Report Series, No. 822), Annex 2.
60. Guidelines for independent lot release of vaccines by regulatory authorities. In: WHO Expert
Committee on Biological Standardization. Sixty-first report. Geneva, World Health Organization,
2013 (WHO Technical Report Series, No. 978), Annex 2.
173
Appendix 1
History and genealogy of BCG substrains
Note: This diagram provides only a historical overview of the use of different substrains derived from BCG vaccine
strain. It does not indicate any WHO “qualification” or “approval” of the strains or vaccines in the context of this
document.
* Yamamoto S, Yamamoto T. Historical review of BCG vaccine in Japan. Japanese Journal of Infectious Disease, 2007,
60:331–336.
174
Annex 3
Appendix 2
Summary protocol for manufacturing and control of BCG
vaccine
The following protocol is intended for guidance, and indicates the information that
should be provided as a minimum by the manufacturer to the NRA. Information
and tests may be added or omitted as required by the NRA, 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
recommendations of 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
NRA or NCL of the country in which the vaccine was produced stating that the
product meets national requirements as well as Part A of the recommendations
of this document published by WHO.
Summary information on the finished product (final lot)

International name:
Trade name:
Product licence (marketing authorization) number:
Country:
Name and address of manufacturer:
Site of manufacture of final lot:
Name and address of licence-holder
(if different):
BCG substrain:
Authority that approved the BCG substrain:
Date approved:
Final bulk number:
Volume of final bulk:
Final product:
Type of vaccine: Intradermal/Percutaneous/Other
Final lot number:
Type of container:
175
Number of doses per container:

Number of filled containers in this final lot:
Date of manufacture of final lot:
Date on which last determination of the bacterial count
was started, or date of start of period of validity:
Shelf-life approved (months):
Expiry date:
Diluent:
Storage conditions:
Volume of single human dose:
Volume of vaccine per container:
Number of doses per container:
Summary of the composition (include a summary of the
qualitative and quantitative composition of the
vaccine per human dose):
Release date:
Production information
A genealogy of the lot numbers of all vaccine components used in the formulation
of the final product will be informative.
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.
Control of source materials (section A.3)

The information requested below is to be presented on each submission. Full details
on master and working seed lots are required on first submission only and whenever
a change has been introduced.
Master seed lot

Origin of seed lot:
Master seed lot number:
Passage level:
Date of preparation of seed lot:
Date of receipt of seed lot (if applicable):
Date of reconstitution of seed lot ampoule:
Date approved by the NRA:
176
Annex 3
Working seed lot

Working seed lot number:
Passage level:
Date of reconstitution of seed lot ampoule:
Date approved by the NRA:
Tests on working seed lot production (section A.3.2)

Antimicrobial sensitivity test (section A.3.2.1)
Method used:
Date test started:
Date test completed:
Results:
Delayed hypersensitivity test (section A.3.2.2)

Vaccine Reference vaccine
Method used:
Dilutions injected:
Inoculation route:
Number of guinea-pigs given injection:
Observation period (specification):
Date test started:
Result:
Identity test (section A.3.2.3)

Method used:
Date test started:
Results:
Test for bacterial and fungal contamination (section A.3.2.4)

Method used:
Number of containers tested:
Volume of inoculum per container:
Volume of medium per container:
177
Incubation Media used Inoculum Date test Date test Results

began ended
20–25 °C
30–36 °C
Negative
control
Test for absence of virulent mycobacteria (section A.3.2.5)

Method used:
No. of human doses injected per guinea-pig:
Inoculation route:
No. of guinea-pigs given injection:
Weight range of guinea-pigs:
Date test started:
Health of animals during test:
Weight gains (losses):
Result:
Test for excessive dermal reactivity (section A.3.2.6)

Method used:
Dilutions injected:
Inoculation route:

Date test started:
Mean diameter of lesions (for
each dilution):
Result:
Production of culture medium (section A.3.3)

Any components of animal origin:
Certificate for BSE/TSE-free:
178
Annex 3
Control of vaccine production (section A.4)

Control of single harvests (section A.4.1)
Derived from master seed lot number:
Working seed lot number:
Passage level from master seed:
Culture medium:
Number and volume of containers inoculated:
Date of inoculation:
Temperature of incubation:
Date of harvest:
Results of visual inspection:
Control of final bulk (section A.4.2)

Tests for bacterial and fungal contamination (section A.4.2.2)
Method used:

began ended
20–25 °C
30–36 °C
Negative
control

(if not performed on final lot)
Method used:
Inoculation route:
Date test started:
179

Result:
Test for bacterial concentration (section A.4.2.4)

Method used:
Date test started:
Specification:
Result:
Test for number of culturable particles (section A.4.2.5)

Method used:
Date test started:
Specification:
Result:
Details of working Reference Preparation:
Substances added (section A.4.2.6)

Any components of animal origin:
Certificate for BSE/TSE-free:
Filling and containers (section A.5)

Lot number:
Date of filling:
Volume of final bulk filled:
Filling volume per container:
Number of containers filled (gross):
Date of freeze-drying:
Number of containers rejected during inspection:
Number of containers sampled:
Total number of containers (net):
Maximum period of storage approved:
Storage temperature and period:
Control tests on final lot (section A.6)

Inspection of final containers (section A.6.1)
Appearance:
Date of test:
Specification:
180
Annex 3
Result:
Recommended reconstitution fluid:
Volume of reconstitution fluid per final container:
Identity test (section A.6.2)

Method used:
Date test started:
Specification:
Result:
Tests for bacterial and fungal contamination (section A.6.3)

Method used:
Specification:

began ended
20–25 °C
30–36 °C
Negative
control
Safety tests (section A.6.4)

(if not performed on final bulk)
Method used:
Inoculation route:
Date test started:
181
Specification:
Result:
Test for excessive dermal reactivity (section A.6.4.2) if applicable

Method used:
Dilutions injected:
Inoculation route:
Date test started:
Mean diameter of lesions (for
each dilution):
Specification:
Result:
Test for bacterial concentration (section A.6.5)

Method used:
Date test started:
Specification:
Result:
Test for residual moisture (section A.6.6)

Method:
Date:
Specification:
Result:
Tests for viability (section A.6.7)

Test for number of culturable particles (section A.6.7.1)
Method used:
Medium:
Date test started:
Before lyophilization After lyophilization
No. of containers tested:
Mean count of culturable
particles per ml:
Mean survival rate (%):
182
Annex 3
Specification:
Result:
Rapid test for viability (section A.6.7.2) if applicable

Method:
Date:
Specification:
Result:
Thermal stability test (section A.6.8)

Method used:
Date test started:
Unheated containers Heated containers
No. of containers tested:
Culturable particles in each
container per ml:
Specification:
Result:
Submission addressed to NRA

Name of responsible person (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 BCG vaccine, whose number
appears on the label of the final container, meets all national requirements and/
or satisfies Part A1 of the WHO Recommendations to assure the quality, safety
and efficacy of BCG vaccines (2013)2
Signature:
Name (typed):
Date:
1
With the exception of provisions on distribution and shipping, which the NRA may not be in a position
to assess.
2
WHO Technical Report Series, No. 979, Annex 3.
183
Appendix 3
Model certificate for the release of BCG vaccine by NRAs
Lot release certificate
Certificate no.
The following lot(s) of BCG vaccine produced by 1
in , whose numbers appear on the labels of the final

2
containers, complies with the relevant specification in the marketing authorization

and provisions for the release of biological products 3 and Part A4 of the WHO
Recommendations to assure the quality, safety and efficacy of BCG vaccines
(2013)5 and comply with WHO good manufacturing practices: main principles
for pharmaceutical products; 6 Good manufacturing practices for biological
products; 7 and Guidelines for independent lot release of vaccines by regulatory
authorities.8
The release decision is based on 9
The certificate may include the following information:

■ name and address of manufacturer;
■ site(s) of manufacturing;
■ trade name and common name of product;
■ marketing authorization number;
■ lot number(s) (including sub-lot numbers, packaging lot numbers
if necessary);
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 NRA.
4
With the exception of provisions on distribution and shipping, which the NRA may not be in a position
to assess.
5
6
7
8
9
Evaluation of summary protocol, independent laboratory testing, and/or specific procedures laid down in
defined document etc., as appropriate.
184
Annex 3
■ type of container;
■ number of doses per container;
■ number of containers/lot size;
■ date of start of period of validity (e.g. manufacturing date) and/or
expiry date;
■ storage condition;
■ signature and function of the authorized person and authorized
agent to issue the certificate;
■ date of issue of certificate;
■ certificate number.
The Director of the NRA (or other authority as appropriate):
Name (typed)
Signature
Date
185

Annex 3: Recommendations To Assure The Quality, Safety and Efficacy of BCG Vaccines

Uploaded by

Copyright:

Available Formats

Annex 3: Recommendations To Assure The Quality, Safety and Efficacy of BCG Vaccines

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Annex 3: Recommendations To Assure The Quality, Safety and Efficacy of BCG Vaccines

Uploaded by

Copyright:

Available Formats

Annex 3

Recommendations to assure the quality, safety and

Recommendations published by the WHO are intended to be

Scope of the Recommendations

BCG vaccine strains

TB in some animal studies (18, 22), commonly administered BCG vaccine

in humans will give effective protection against TB in humans. It should be noted

Part A. Manufacturing recommendations

A.1.2 Descriptive definition

A.1.3 International Reference Preparations and Reference Reagents

The National Institute for Biological Standards and Control (NIBSC),

A.2 General manufacturing recommendations

The data that should be considered in determining the consistency of production

No cultures of microorganisms other than the BCG vaccine strain approved

BCG is susceptible to sunlight. Therefore, the procedures for the

Should an examination reveal signs of TB or suspected TB in a worker, he or

A.3 Control of source materials

The BCG vaccine strain used should be identified by historical records

A.3.2 Tests on seed lot

A.3.2.2 Delayed hypersensitivity test

A.3.2.3 Identity test

techniques will also provide relevant information to ensure genetic consistency

A.3.2.4 Test for bacterial and fungal contamination

A.3.2.5 Test for absence of virulent mycobacteria

A.3.2.6 Test for excessive dermal reactivity

A.3.3 Production culture medium

A.4 Control of vaccine production

A.4.2 Control of final bulk

A.4.2.2 Test for bacterial and fungal contamination

A.4.2.3 Test for absence of virulent mycobacteria

examined postmortem for macroscopic evidence of progressive TB disease.

A.4.2.4 Test for bacterial concentration

Clumping issues should be considered during validation of the assay.

A.4.2.5 Test for number of culturable particles

There are various methods of determining the number of culturable

A.4.2.6 Substances added to the final bulk

A.5 Filling and containers

A.6 Control tests on final lot

A.6.1 Inspection of final containers

The appearance of the freeze-dried vaccine and the reconstituted vaccine

A.6.2 Identity test

A.6.3 Test for bacterial and fungal contamination

A.6.4 Safety tests

A.6.4.2 Test for excessive dermal reactivity

A.6.5 Test for bacterial concentration

The estimation of tota1 bacterial content may be made either directly,

The clumping issue should be considered during validation of the assay.

A.6.6 Test for residual moisture

A.6.7 Tests for viability

to adjust the count of the vaccine by comparison with the Reference

The clumping issue should be considered during validation of the assay.

A.6.7.2 Rapid test for viability

The bioluminescence reaction occurring in fireflies depends on the

Studies have shown that, if properly validated, measurement of ATP

A.6.8 Thermal stability test

The percentage decrease in the number of culturable particles is then compared

As a guide to stability, some manufacturers of freeze-dried BCG vaccine

A.8 Retained samples

■ the volume and nature of the diluent to be added to reconstitute