Global Vaccine Safety

Global Advisory Committee on Vaccine Safety, report of meeting held 7-8 December 2011

Published in the WHO Weekly Epidemiological Record on 10 February 2012

The Global Advisory Committee on Vaccine Safety (GACVS), an expert clinical and scientific advisory body, was established by WHO to provide independent, scientifically rigorous advice on vaccine-safety issues of potential global importance.1 GACVS held its 25th meeting in Geneva, Switzerland, on 7–8 December 2011.2 The Committee reviewed the following:

1. New data related to (i) the safety profile of influenza A(H1N1)pdm09 vaccines and (ii) the risk of intussusception related to the use of second generation rotavirus vaccines.

2. Generic issues concerning vaccine pharmacovigilance: (i) safety of immunization during pregnancy and lactation (ii) progress with the global vaccine safety network and (iii) implementation of the Global Vaccine Safety Blueprint.

Safety of influenza A(H1N1)pdm09 vaccines

Overall, the safety information for the pandemic influenza vaccines continues to be reassuring. Other than the association of narcolepsy/cataplexy with an adjuvanted pandemic influenza vaccine in people aged 4–19 years, seen mainly in Finland and Sweden,3 there were no new proven safety signals from passive surveillance systems.

Population-based epidemiological studies on the association of narcolepsy with the adjuvanted pandemic vaccine (Pandemrix) have been completed in Sweden4 and Finland. Adhering to the precautionary principle, the European Medical Agency’s Committee for Medicinal Products for Human Use concluded “that the bene-fits of Pandemrix continue to outweigh its risks but that it may only be used in people <20 years of age if the recommended annual seasonal trivalent influenza is not available and if immunization against A(H1N1) is still needed, for instance in people at risk of the complications from infection”. Pandemrix marketing authorization has now expired and this vaccine is no longer in use. Epidemiological studies in other countries in Europe and Canada are under way. Immunological and animal model studies to help elucidate the biological mechanism of the observed association are also in progress.

Although preliminary analyses of active surveillance studies for Guillain-Barré syndrome in the United States, which evaluated both adjuvanted and non-adjuvanted pandemic influenza vaccines, indicated an increased risk, this finding has not been replicated elsewhere to date. Available data do not provide conclusive evidence for this increased risk. Were the risk to be confirmed, it would be much lower than that observed following the 1976 swine influenza vaccination campaign in the United States and would be similar to that observed with use of seasonal influenza vaccines.

GACVS reviewed results from additional epidemiological studies on a possible association between influenza A(H1N1)pdm09 vaccines and certain autoimmune and other clinical syndromes. No major safety concerns appeared, although it was noted that the sample size or methodology of these studies might not have been optimal for establishing causal relations. The Committee acknowledged the need for further analysis in this area.

The safety of pandemic vaccines when administered to pregnant women was reassuring. Data from an observational cohort study in Canada and from a birth and infant health registry in the United States did not point to any safety concerns related to pandemic vaccines among women during gestation or their offspring. Several studies on the safety of pandemic vaccines among pregnant women are still being completed in other regions.

Rotavirus vaccines and intussusception

Rotavirus vaccines (RotaTeq and Rotarix) are being introduced into immunization programmes in many countries. In Brazil and Mexico, use of Rotarix has been associated with large reductions in all-cause diarrhoeal deaths. However, because a previous rotavirus vaccine (Rotashield) was associated with a ~30-fold increase in risk of intussusception during the week following receipt of the first vaccine dose, large clinical trials involving more than 70 000 children were carried out with RotaTeq and Rotarix. In the trials of RotaTeq, children were vaccinated with dose 1 at age 6–12 weeks; with Rotarix, children were vaccinated at age 6–14 weeks (in Latin America) or age 6–15 weeks (in Europe). RotaTeq is delivered in a 3-dose schedule; Rotarix requires only 2 doses. After licensure, the maximum age indicated for the last dose of vaccine was 32 weeks for RotaTeq or 24 weeks and 6 days for Rotarix. No evidence of increased risk of intussusception was identified during clinical trials with administration at those recommended ages.

Based on these data, and in order to align schedules between the vaccines, the Strategic Advisory Group of Experts (SAGE) on Immunization and GACVS recommended in 2009 that the first dose of either Rotarix or RotaTeq should be administered at age 6–15 weeks, and the maximum age for administering the last dose of either vaccine should be 32 weeks. It was noted that this expansion of the age range for use of these vaccines might potentially increase coverage of the first dose in developing countries from about 57% to 70% and full-course coverage from about 36% to 54%.

Not all countries are able to deliver early childhood immunization to most children by the recommended ages; delays in timing of vaccination are noted in many countries, in particular in countries with high rates of rotavirus-associated infant mortality.5 Therefore, there has been consideration of liberalizing the age restrictions for use of both Rotarix and RotaTeq, in order to extend the benefits of vaccination to children who otherwise would not receive the vaccines.

Since the recommendation in 2009, post-licensure safety investigations in some countries revealed an increased risk of intussusception after use of Rotarix (Australia and Mexico) and RotaTeq (Australia). The risks – where they have been found – are substantially lower than those previously associated with Rotashield. Studies with active case finding indicated that, with the current vaccines in use, 4- to 6-fold increases in risk were clustered within the first week after the first dose. In other populations, no increase in risk has been observed or reported. An updated risk–benefit analysis predicted that use of current rotavirus vaccines without age restriction would prevent an additional 49 500 (range: 35 000–67 000) rotavirus deaths while potentially causing about 300 (range: 180–400) excess intussusception deaths, compared with the current age-restricted strategy. This analysis incorporated updated estimates of risk of intussusception (similar to those hypothesized in the 2009 published analysis), updated estimates of rotavirus mortality for the year 2008, and newly available data on efficacy of rotavirus vaccines in low-income settings in Africa and Asia.

Because Rotarix and RotaTeq are now generally being used in accordance with the recommended age restrictions, data on the risk of intussusception for administration of either vaccine at older ages are not available. However, an analysis of intussusception risk following Rotashield in the USA found no significant difference in the relative risk of intussusception after administration of the first dose at 7–13 weeks, 14–21 weeks, or 22–52 weeks, although the confidence limits of those estimates were wide.

The Committee concluded that available data suggest that both Rotarix and RotaTeq continue to exhibit a good safety profile, but may be associated with an increased (up to 6-fold) risk of intussusception after the first dose of vaccine in some populations. The levels of risk observed are substantially less than those observed with the previous vaccine, Rotashield. The Committee reiterates that, based on available evidence, the benefits of rotavirus vaccination to all infants, without age restriction, would greatly exceed the risks, particularly in developing countries with moderate and high mortality from rotavirus disease. Active surveillance of intussusception in African and Asian countries that plan to introduce rotavirus vaccines should be seriously considered, because the data accrued would eventually provide additional benefit–risk information related to these important vaccines.

Vaccine safety in pregnancy and lactation

Significant morbidity due to vaccine-preventable diseases among women and infants could be prevented by immunization of pregnant women. Policy formulation regarding vaccination during pregnancy is challenging because the evidence base to guide decisions is extremely limited. SAGE recently requested GACVS to provide guidance on the safety of vaccines used among pregnant and lactating women.

The concerns include risks to the fetus because there have been a small number of demonstrated instances of live vaccine transmission. For other live vaccines and all inactivated vaccines no evidence of such risk exists. For newer vaccines, several of which are used primarily in low- and middle-income countries, there is no evidence of pregnancy-related harm. Despite of the lack of apparent safety issues for many vaccines, precautions and contraindications are often included in product labelling with respect to immunization during pregnancy and lactation, thereby limiting their benefit to women. In developing countries, pregnancy is one of the few opportunities for women to be in contact with the health care system.

Several important questions need to be addressed. These include the possibility of assessing safety of vaccines in pregnancy and lactation during the vaccine development process and additional opportunities for generating post-licensure data. Adverse events may occur with vaccination only coincidentally, giving rise to a false perception of increased risk from vaccination during pregnancy. Better understanding and analysis of the frequency of occurrence of adverse events that occur during pregnancy in the absence of vaccination («background rates») would be of great benefit. An agenda for continued monitoring, better definitions, availability of background rates, inclusion of pregnant women in cohort studies, and collaboration with regulatory authorities could be developed and proposed. Likewise, since labelling of vaccines package inserts is highly variable around the world, there is a possible role for WHO to help harmonize practices in this area.

Based on the presentation and discussion, GACVS recognizes the need to address safety issues related to the use of vaccines during pregnancy and lactation. GACVS proposes to review the relevant evidence. It will also consider including methodological points for planning and analysis of clinical trials and post-marketing studies.

Global network for post-marketing surveillance and AEFI monitoring

GACVS was presented with an update on the progress of the global network for postmarketing surveillance (PMS Network) of adverse events following immunization (AEFI), a WHO-led pilot project aimed at enhancing the monitoring, reporting and sharing of vaccine safety data for countries introducing new prequalified vaccines.

Started in 2007, the aim of the project was to establish a network of at least 20 countries that have introduced newly prequalified vaccines, to stimulate the reporting of AEFI to a central database (Vigibase) located at the WHO Collaborating Centre for international drug monitoring (Uppsala Monitoring Centre; UMC), in Sweden. By 2011, 12 countries had been included in the network: Albania, Brazil, China, India, Islamic Republic of Iran, Kazakhstan, Mexico, Senegal, Sri Lanka, Tunisia, Uganda, and Viet Nam. Countries enrolled in the network have had a baseline evaluation of their national vaccine regulatory process and their AEFI surveillance system capability. In addition, countries were trained in the use of Vigiflow, a software platform to report and upload AEFI cases to UMC, and on the methods of causality assessment for vaccine AEFI classification.

The PMS Network has improved the reporting of AEFI for vaccines in most participating countries. Participating countries have recognized the need for and benefits of the network, but operational challenges exist. Due to lack of harmonization of current surveillance and reporting systems, including case report forms, software systems, and type of AEFI reported, the heterogeneous data accumulated at UMC is likely to provide only limited vaccine safety signals globally. Even though the network primarily focused on newly prequalified vaccines, data collected relates mainly to other, more traditional, vaccines prequalified or not. In addition, the network countries have identified the need for a simpler data processing tool that could be more specific for vaccines and that could be operated offline given the internet connectivity limitations in most countries.

GACVS also recognized the limited value of spontaneous reports to generate comprehensive data that can later be used to determine vaccine safety/risk profiles via standard methods. Nevertheless, spontaneous reports of AEFI are important to generate signals for vaccine safety monitoring systems and can inform the design and conduct of careful epidemiological studies to assess potential risks.

GACVS also acknowledges the importance of a global and centralized database for all drugs and vaccines. A more comprehensive and active database for vaccines could allow countries, regions, and investigators to detect global vaccine safety signals that could go unrecognized at a country level, as well as provide background data on common and uncommon adverse events. UMC has the capability and know-how to provide the data infrastructure needed for such an endeavour. UMC could develop a simple data entry tool, based on a minimal dataset, to serve the perceived needs of lower and middle income countries.

As part of the PMS Network activities, a dictionary for prequalified vaccines has been developed. This vaccine dictionary is considered an essential tool for the countries and AEFI systems to determine the components of vaccines that could be implicated in serious or relevant adverse events. It is therefore paramount to expand this dictionary to ensure the inclusion of other licensed vaccines in use, transparent participation from manufacturers, and free availability to countries that need it most.

GACVS also emphasized the need for strengthening AEFI surveillance systems at country and regional levels to improve the current reporting of safety signals following immunization. The development of regional networks could be the next step towards improving the reporting of AEFI and vaccine safety signals at the global level.

Implementing the Global Vaccine Safety Blueprint

At the 24th meeting of GACVS, the Global Vaccine Safety Blueprint was presented in order to obtain committee input into this strategic plan. The wide-ranging discussions are documented in previous reports. In September 2011, a number of stakeholders, including GACVS members, met in Geneva to discuss this Blueprint. The purpose of this GACVS session was to review the Global Vaccine Safety meeting and the Blueprint discussions, which occurred at the SAGE meeting. The expected outcome was to clarify the role of GACVS in implementation of the Blueprint through a Global Vaccine Safety Initiative (GVSI).

GACVS members who attended the meeting commented that the Blueprint provides a common platform for all stakeholders and recognized the clear need to strengthen vaccine safety surveillance in low- and middle-income countries. The Blueprint is very ambitious and is likely to be achievable only by step-wise increments and prioritization of goals. It was thought that developing novel pilot or demonstration projects to achieve surveillance goals may be more achievable in terms of resources and attraction of funding.

GACVS concurs with SAGE in emphasizing the critical importance of country ownership in monitoring vaccine safety and responding to safety concerns. SAGE also suggested focusing the Vision Statement on vaccine pharmacovigilance and listing specific objectives. While this focus was recognized as important in order to provide specific direction to the Blueprint, it would be important that vaccine safety monitoring was not seen in isolation from other surveillance systems and data. There was also general agreement that methods of surveillance that are alternative and complementary to passive surveillance should be evaluated. The importance of surveillance for burden of vaccine-preventable disease is a critical component of a full evaluation of vaccine risks and benefits.

The proposed terms of reference for the management structure of the GVSI were presented. GACVS members recommended that this structure should be designed to deliver the programmatic changes required of the GVSI. In particular, the lines of reporting and evaluation and the relationship with WHO immunization advisory bodies including GACVS and SAGE should be defined such that the proposed structure aligns its efforts with those of other WHO advisory groups. GACVS identifies its main roles in supporting the GVSI in relation to the development of safety communication plans, the development of internationally harmonized tools, and the provision of international expert advice. GACVS could also have a role in prioritization of GVSI activities and could also review and guide any demonstration surveillance projects.

WHO and its partners should lead the Blueprint implementation. It should be aligned with other related WHO capacity-building efforts. This includes in particular the strengthening of immunization programmes and national regulatory authorities, together with the development of national expert advisory bodies. SAGE suggested that a mechanism be developed to enable prioritization of both activities and countries in the implementation of the Blueprint and SAGE invited the GAVI Alliance and other partners to support this implementation.6

  • See No. 41, 1999, pp. 337–338.
  • GACVS invited additional experts to present and discuss evidence related to particular topics. These experts included persons affiliated with: Medical Products Agency, Sweden; Erasmus University, Rotterdam, The Netherlands; Mount Sinai Hospital Toronto, Canada; Naval Health Research Center; San Diego, United States of America; Institut National de santé publique du Québec, Beauport, Canada; Children’s Hospital Boston, United States; Centers for Disease Control and Prevention, Atlanta, United States; Program for Applied Technologies in Health, Seattle, United States; London School of Hygiene and Tropical Medicine, London, United Kingdom.
  • Statement on narcolepsy and Pandemrix. Geneva, World health Organization, 2011 (; accessed January 2012).
  • Bardage C et al. Neurological and autoimmune disorders after vaccination against pandemic influenza A (H1N1) with a monovalent adjuvanted vaccine: population based cohort study in Stockholm, Sweden. BMJ, 2011, 343:d5956.
  • Clark A, Sanderson C. Timing of children’s vaccinations in 45 low-income and middle-income countries: an analysis of survey data. Lancet, 2009, 373:1543–1549.
  • See No. 1, 2012, pp. 1–16.