A study of protective efficacy is not feasible if the disease to be prevented does not occur at
present (e.g. smallpox) or occurs at too low a rate for a study to be performed in a
reasonable timeframe (e.g. anthrax, brucellosis, Q fever).
A study of protective efficacy is not feasible if the disease to be prevented occurs in
unpredictable short-lived outbreaks that, even if large numbers are affected, do not allow
enough time to accrue sufficient cases for an assessment of vaccine efficacy (e.g. some viral
haemorrhagic fevers).
If a study of protective efficacy is not necessary and/or not feasible, the applicant should provide
scientific justification in the Clinical Overview. In such cases, the applicant should also provide a
detailed description of the post-authorisation studies that are planned to evaluate vaccine effectiveness
(see section 4.2.2).
If a study of protective efficacy is considered necessary and feasible, the applicant should provide a
detailed description of the study design in the Clinical Overview. The design of the study will be
influenced by the incidence and characteristics of the infectious disease that is to be prevented. The
study should be designed to provide a reliable estimate of vaccine efficacy with sufficient precision
and should be conducted in a population that is representative of the target population for the vaccine.
The study should be designed to evaluate the protective efficacy of the vaccine against the disease(s)
to be prevented. The primary efficacy variable should be based on a pre-defined case definition (see
below). The study should also evaluate the protective efficacy of the vaccine against other clinically
relevant endpoints (e.g. infection, severe disease, hospitalisation and death). The study should
evaluate the duration of protection and the need for booster doses (see section 4.1.1).
The study should be designed to evaluate the protective efficacy of the vaccine in various subgroups of
the study population (e.g. age groups, ethnic groups, previous immunisation histories). The study
should also evaluate the protective efficacy of the vaccine against various strains/serotypes of the
infectious agent.
The study should be designed to evaluate the safety of the vaccine in the study population. The study
should also evaluate the immunogenicity of the vaccine in a subset of the study population (see section
4.1.1).
- Randomised controlled trials
The most reliable method for evaluating the protective efficacy of a vaccine is a randomised
controlled trial (RCT). The study should be designed to minimise bias and confounding. The study
should be conducted in a population that is representative of the target population for the vaccine.
The study should be designed to evaluate the protective efficacy of the vaccine against the disease(s)
to be prevented. The primary efficacy variable should be based on a pre-defined case definition (see
below). The study should also evaluate the protective efficacy of the vaccine against other clinically
relevant endpoints (e.g. infection, severe disease, hospitalisation and death). The study should
evaluate the duration of protection and the need for booster doses (see section 4.1.1).
The study should be designed to evaluate the protective efficacy of the vaccine in various subgroups of
the study population (e.g. age groups, ethnic groups, previous immunisation histories). The study
should also evaluate the protective efficacy of the vaccine against various strains/serotypes of the
infectious agent.
The study should be designed to evaluate the safety of the vaccine in the study population. The study
should also evaluate the immunogenicity of the vaccine in a subset of the study population (see section
4.1.1).
- Secondary attack rate studies
Secondary attack rate (SAR) studies are sometimes used when the infection to be prevented is known
or expected to be associated with a relatively high incidence of secondary cases. In these studies, an
assumption is made that vaccinees and non-vaccinees have an equal chance of acquiring the infection
from the index case. The preferred design would be to randomise the direct contacts, and sometimes
secondary contacts, of a case on an individual basis to receive or not receive the candidate vaccine.
Alternatives could include randomising individuals to immediate or delayed vaccination or randomising
all the members of each ring to the same arm, i.e. a cluster-randomised approach.
The primary analysis should be based on the intent-to-treat (ITT) population, defined as all randomised
subjects who receive at least one dose of study vaccine. The analysis should also be performed in the
per-protocol (PP) population, defined as all randomised subjects who complete the study according to
the protocol. The analysis should also be performed in various subgroups of the study population (e.g.
age groups, ethnic groups, previous immunisation histories).
The primary efficacy variable should be based on a pre-defined case definition (see below). The study
should also evaluate the protective efficacy of the vaccine against other clinically relevant endpoints
(e.g. infection, severe disease, hospitalisation and death). The study should evaluate the duration of
protection and the need for booster doses (see section 4.1.1).
The case definition should be based on clinical signs and symptoms typical of the infectious disease
together with laboratory confirmation of the aetiology. The laboratory methods used to confirm the
diagnosis should be pre-defined and justified. If there are commercially available tests, the choice of
laboratory method(s) should be based on the reported performance characteristics (i.e. the sensitivity
and specificity of the assay and whether it is deemed suitable for the trial population). In some cases,
there may be interest in selecting an assay that can detect additional pathogens that may co-infect with
the target pathogen and possibly affect the severity or course of the disease. It may also be necessary
to apply additional assays to detect such organisms if this is considered important for interpretation of
the trial results.
On occasion, such as when there are no commercially available tests available with satisfactory
performance characteristics, it may be appropriate to use experimental laboratory methods for
establishing the presence of infection. In such cases, every effort should be made during the clinical
development programme to evaluate the sensitivity, specificity and reproducibility of the methods
used. If the case definition is based on histological findings, the criteria for staging and progression
should be pre-defined in the protocol and it is recommended that there is a quality control system in
place and/or secondary readings conducted at an expert central laboratory facility.
If an organism causes disease of variable severity or a range of clinical presentations (e.g. life-
threatening invasive infections as well as localised infections) the clinical features of the case definition
should be selected in accordance with the proposed indication(s). In these instances, separate efficacy
trials using different case definitions may be necessary to support specific indications (e.g. prevention
of invasive pneumococcal disease vs. prevention of pneumococcal otitis media). In addition, for some
vaccines it may be important to compare the severity of vaccine breakthrough cases with cases that
occur in the control group to determine whether prior vaccination ameliorates or possibly enhances
the severity of the disease.
It is usual that there is active case ascertainment at least up to the time of conduct of the primary
analysis. If there is to be further follow-up after the primary analysis the decision to switch to passive
case ascertainment should consider the importance of obtaining reliable estimates of vaccine efficacy
in the longer term and information on the characteristics of cases that occur in previously vaccinated
and unvaccinated subjects over time.
When the primary endpoint is laboratory-confirmed clinical disease, the protocol should list the clinical
signs and/or symptoms that trigger contact between trial subjects and trial site staff or designated
healthcare facilities participating in the trial so that appropriate laboratory testing can be conducted to
confirm the case. Regular personal or non-personal contact with trial staff may also be used to
determine whether there have been any recent clinical signs or symptoms of potential relevance and to
determine whether cases may have been missed. If any cases bypass the designated trial healthcare
facilities and present elsewhere, attempts should be made to retrieve available data that could be used
to establish whether the case definition was met.
If the primary endpoint is not a clinically manifest infection, trial visits should be sufficiently frequent
to obtain the laboratory data of importance. Every effort should be made to minimize numbers that are
lost to follow-up and to conduct trial visits within protocol-defined windows.
4.2.2. Vaccine effectiveness
Estimates of vaccine effectiveness reflect direct (vaccine induced) and indirect (population related)
protection during routine use. Vaccine effectiveness may be estimated from studies that describe the
occurrence of the disease to be prevented in the vaccinated target population over time. For example,
these may be observational cohort studies, case-control or case-cohort studies. Alternatively,
effectiveness may be estimated from data collected during phased (e.g. in sequential age or risk
groups) introduction of the vaccine into the target population and on occasion, using other study
designs, disease surveillance networks or disease registries.
Vaccine effectiveness studies are not always necessary but may be particularly useful in some
situations and/or to address certain issues, including but not limited to the following:
Authorisation was based on nonclinical efficacy data and a comparison of immune responses
between protected animals and vaccinated humans and/or on a human challenge trial;
It is not known how long protection will last after the primary series and/or after post-primary
dose(s);
It is proposed to use the data collected to address long-term protection to support
identification of an ICP;
There are unanswered questions regarding the efficacy of a vaccine against a wide range of
pathogen subtypes;
There are scientific reasons to suspect that an estimate of vaccine efficacy documented in a
pre-authorisation trial may not be widely applicable to other populations (e.g. to subjects who
are resident in different endemic or non-endemic regions);
Different vaccine implementation strategies are in use in different countries or regions that
may impact on the estimate of vaccine effectiveness (e.g. when introduction of routine use in
infants is accompanied by a catch-up programme in older subjects and the upper age of the
catch-up). In these instances, estimates of vaccine effectiveness obtained using different
strategies can inform the optimal strategy to achieve rapid and efficient control of the disease;
There is reason to suspect that widespread use of a vaccine could result in a change in the
subtypes of a pathogen causing disease compared to the pre-vaccination era.
Vaccine effectiveness studies require a suitable infrastructure to be in place for case ascertainment and
confirmation of cases in accordance with clinical and laboratory criteria and it may not be possible to
obtain reliable data in all countries or regions. In addition, for some infectious diseases an estimate of
vaccine effectiveness is possible only in case of a naturally occurring epidemic or a deliberate release
of a pathogen in the context of bioterrorism. Furthermore, the conduct of a vaccine effectiveness study
requires that a policy decision has been made to vaccinate a sufficiently large population to support the
analysis.
Whenever it is perceived that valuable information could be gained from conducting a vaccine
effectiveness study it is important that plans are in place to enable its initiation whenever a suitable
opportunity arises in the post-authorisation period.
The role of the licence holder in designing vaccine effectiveness studies and specifying the target of,
and the population for analysis, generating protocols, and collecting and analysing the data requires
consideration on a case by case basis. In most cases, unless the incidence of the infectious disease is
very high in some regions so that a relatively small and short study is possible, a study sponsored by
the licence holder is not a practical undertaking. The only feasible way to evaluate vaccine
effectiveness is often from studies put in place by public health authorities when initiating large
vaccination programmes. Nevertheless, licence holders have a responsibility to ensure that relevant
data made available to them and/or reported in the literature from non-sponsored studies are reported
to EU Competent Authorities. Consideration should be given to updating the SmPC if the results have
clear implications for the advice given (e.g. on the need for additional doses to maintain protection).
4.3. Special considerations for vaccine development
4.3.1. Immune interference
- Vaccines that contain more than one antigen
The inclusion of multiple antigens in a vaccine may lead to immune interference (i.e. enhancement or
suppression of immune responses to one or more of the antigens). The potential for immune
interference should be evaluated in clinical studies that compare immune responses to the individual
antigens when given separately and when given as components of the combined vaccine. The design
and interpretation of such studies must be tailored to the antigens involved and should take into
account any relevant experience about the possible effects of their combination.
- Concomitant administration of vaccines
The concomitant administration of vaccines may lead to immune interference (i.e. enhancement or
suppression of immune responses to one or more of the antigens). The potential for immune
interference should be evaluated in clinical studies that compare immune responses to the individual
antigens when given separately and when given concomitantly. The design and interpretation of such
studies must be tailored to the antigens involved and should take into account any relevant experience
about the possible effects of their co-administration.
For some vaccines, such as those intended for the primary series in infants, it may be necessary to
ensure that all subjects in a clinical trial receive all the required antigens before reaching a certain age.
To address this need, trials may need to compare concomitant administration with separate
administrations made in a staggered fashion (e.g. to compare concomitant administration at 2 and 4
months with administration of routine infant vaccines at 2 and 4 months and the candidate vaccine at
3 and 5 months). In older age groups, it is more likely possible to find populations in which coadministration can be compared with separate administrations since it may be less critical to achieve
protection against all antigens in a short timeframe. For some types of vaccine, such as those generally
given before travel, it would also be important to assess immune interference at the most concentrated
schedule that might be needed.
If any co-administration studies identify important reductions in immune responses, further trials could
explore the minimum dose interval that does not lead to any impact.
4.3.2. Cross-reacting immune responses
The immune response to an antigen may cross-react with antigen(s) of one or more other species or
subtypes within a species. The potential for cross-reacting immune responses should be evaluated in
clinical studies that compare immune responses to the individual antigens when given separately and
when given as components of the combined vaccine or when given concomitantly. The design and
interpretation of such studies must be tailored to the antigens involved and should take into account
any relevant experience about the possible effects of their combination or co-administration.
4.3.3. Using different vaccines to prime and to boost
The use of different vaccines to prime and to boost may lead to immune interference (i.e.
enhancement or suppression of immune responses to one or more of the antigens). The potential for
immune interference should be evaluated in clinical studies that compare immune responses to the
individual antigens when given as a homologous prime-boost regimen and when given as a
heterologous prime-boost regimen. The design and interpretation of such studies must be tailored to
the antigens involved and should take into account any relevant experience about the possible effects of
their combination or co-administration.
4.3.4. Vaccine lots and lot-to-lot consistency studies
The manufacturing process for vaccines is complex and may lead to variability in the final product. The
potential for variability should be evaluated in clinical studies that compare immune responses to the
individual antigens when given as components of different lots of the vaccine. The design and
interpretation of such studies must be tailored to the antigens involved and should take into account
any relevant experience about the possible effects of lot-to-lot variability.
A lot-to-lot consistency trial is not routinely required but may be considered useful under certain
circumstances that should be considered on a case by case basis. If such a trial is conducted it is
important to consider and justify the number of lots to be compared and the method of lot selection
(e.g. consecutively produced or chosen at random). Careful consideration needs to be given to the
primary immune response endpoint and the pre-defined acceptance criteria.
It is recommended that several lots of the candidate vaccine with a formulation comparable to that of
the final product intended for marketing should be tested during the clinical development programme.
If this is not possible due to late stage manufacturing changes, the sponsor should justify the
relevance of the clinical trial data to the lots intended for marketing based on quality attributes and/or
should conduct a clinical comparison between lots.
4.3.5. Bridging studies
Bridging studies may be used to extrapolate data from one population to another (e.g. from adults to
children, from one ethnic group to another, from one geographical region to another). The design and
interpretation of such studies must be tailored to the antigens involved and should take into account
any relevant experience about the possible effects of the population differences.
4.3.6. Circumstances in which approval might be based on very limited data
In some circumstances, it may be possible to generate only very limited data for new vaccines intended
to prevent rare infections that carry considerable morbidity and mortality. The extent of the data that
might be acceptable to support a marketing authorisation requires consideration on a case by case
basis. Applicants are advised to seek scientific advice from EU Competent Authorities at an early
stage.
4.4 Clinical safety and pharmacovigilance requirements
The extent of the safety data that can be provided pre-authorisation will depend on the overall content
of the clinical development programme, such as whether or not protective efficacy studies have been
performed. There are also some special considerations for the collection of vaccine safety data
depending on such factors as route of administration, recording of solicited signs and symptoms in
addition to all other adverse events, definitions of some adverse events and the determination of their
relationship to vaccination. Detailed guidance on post-authorisation vaccine pharmacovigilance will
be provided in a separate guideline.
REFERENCES (SCIENTIFIC AND / OR LEGAL)
Directive 2001/83/EC of the European Parliament and of the Council of 6 November 2001 on
the Community code relating to medicinal products for human use, as amended.
Commission Directive 2003/63/EC of 25 June 2003 amending Directive 2001/83/EC of the
European Parliament and of the Council on the Community code relating to medicinal products
for human use.
Note for Guidance on Clinical Evaluation of New Vaccines (CPMP/EWP/463/97).
Note for Guidance on Preclinical Pharmacological and Toxicological Testing of Vaccines
(CPMP/SWP/465/95).
Note for Guidance on the Choice of the Non-Inferiority Margin (CHMP/EWP/2158/2005).
Guideline on Similar Biological Medicinal Products Containing Biotechnology-Derived Proteins
as Active Substance: Non-Clinical and Clinical Issues (EMEA/42832/2005).
Note for Guidance on Statistical Principles for Clinical Trials (CPMP/ICH/363/96).
Note for Guidance on Planning Pharmacovigilance Activities (CPMP/ICH/5716/03).
Note for Guidance on Clinical Safety Data Management: Definitions and Standards for
Expedited Reporting (CPMP/ICH/377/95).
Note for Guidance on General Considerations for Clinical Trials (CPMP/ICH/291/95).
Note for Guidance on Clinical Investigation of Medicinal Products in the Paediatric
Population (CPMP/ICH/2711/99).
Note for Guidance on Studies in Support of Special Populations: Geriatrics (CPMP/ICH/379/95).
Note for Guidance on Influenza Vaccines (CPMP/BWP/214/96).
Note for Guidance on Live Recombinant Viral Vaccines (CPMP/BWP/3088/99).
Note for Guidance on Validation of Analytical Procedures: Text and Methodology
(CPMP/ICH/381/95).
Note for Guidance on Missing Data in Confirmatory Clinical Trials (CPMP/EWP/1776/99).
Note for Guidance on Switching between Superiority and Non-Inferiority (CPMP/EWP/482/99).
Note for Guidance on Multiplicity Issues in Clinical Trials (CPMP/EWP/908/99).
Note for Guidance on Application with 1. Meta-Analyses 2. One Pivotal Study (CPMP/EWP/2330/99).
Note for Guidance on Adjustment for Baseline Covariates in Clinical Trials (CPMP/EWP/2863/99).
Note for Guidance on Investigation of Subgroups in Confirmatory Clinical Trials
(CPMP/EWP/3094/02).
Reflection Paper on Methodological Issues in Confirmatory Clinical Trials Planned with an
Adaptive Design (CHMP/EWP/2459/02).
Guidance on Format of the Risk-Management Plan (RMP) in the EU – in Integrated Format
(EMEA/192632/2006).
Note for Guidance on Risk Management Systems for Medicinal Products for Human Use
(EMEA/CHMP/96268/2005).
Note for Guidance on Good Pharmacovigilance Practices (GVP) (EMEA/813938/2005).
Note for Guidance on Good Pharmacovigilance Practices (GVP) Annex I - Definitions
(EMEA/876333/2006).
Note for Guidance on Good Pharmacovigilance Practices (GVP) Module V – Risk Management
Systems (EMEA/838713/2005).
