Global Vaccine Safety

Global Advisory Committee on Vaccine Safety
11-12 June 2014

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 30th meeting in Geneva, Switzerland, on 11–12 June 2014.2 The Committee reviewed 4 specific issues:

  • the safety profiles of 2 novel vaccines:
    – a live attenuated rotavirus vaccine and,
    – a recombinant hepatitis E vaccine.
  • a study of the safety of meningococcal A conjugate vaccine among pregnant women in Ghana;
  • and issues related to the monitoring of the safety of future anti-malarial vaccines.

During the meeting, the Committee also reflected on the accomplishments and opportunities after 15 years since the establishment of GACVS.

Safety profile of a novel live attenuated rotavisrus vaccine

Rotavirus gastroenteritis remains a major cause of morbidity and mortality among young children. Two rotavirus vaccines are currently widely used in national immunization programmes. A new rotavirus vaccine, Rotavac, was licensed in India in January 2014. The vaccine is derived from a naturally attenuated human neonatal strain containing 1 bovine segment (G9P[11]) that was originally isolated from an asymptomatic infant at the All India Institute of Medical Sciences in 1988. The strain was further studied by Indian and U.S. investigators, with safety trials conducted in adults and children in Cincinnati, United States of America. In 2000, the strain was then licensed to Bharat Biotech International (Hyderabad, India), which developed the vaccine with clinical testing in Phase 1, 2 and 3 studies in India. The current vaccine formulation requires -20°C storage, although the product may be stored at +2 to +8 °C forthe final 6 months prior to expiry. Several different presentations have been licensed, but the manufacturer expects the main presentation to be in a 5-dose vial, which could be used for 8 hours after opening. An oral antacid buffer is currently administered prior to administration of the vaccine.

The randomized, double-blind, placebo-controlled Phase 3 study was carried out with active monitoring for safety in 4532 infants who received the rotavirus vaccine and 2267 placebo recipients. The study was conducted during their routine childhood immunization series, and the subjects were followed up to the age of 2 years. There was no imbalance noted between the Rotavac and placebo groups with respect to adverse events, death or intussusception.

There were 11 confirmed cases of intussusception; importantly, no case occurred in proximity to the time of vaccination. The earliest case following receipt of placebo was 36 days post dose 3. The earliest case following receipt of Rotavac was 112 days post dose 3. The lack of a temporal association argues strongly against a causative relationship between Rotavac and intussusception since most rotavirus vaccine-attributable cases are expected to occur within the first week following vaccination. The observed incidence of confirmed intussusception was 94 per 100 000 child-years (95% confidence interval [CI]: 41–185) among vaccinated infants and 71 per 100 000 child-years (95% CI:15–206) among those who received placebo. This incidence is similar to that measured, in the absence of vaccine, from countries with active surveillance systems for intussusception.

As the available safety data support further use of the vaccine, a post-licensure study of at least 45 000 vaccinated infants is planned. However, based on the experience with similar vaccines, it will be important that additional data be continuously collected in order to assess the risk of intussusception as well as to identify any other rare adverse events that may occur. Based on the experience with other rotavirus vaccines, the infrastructure of sentinel sites that exists in India should be utilized for continued intussusception surveillance in order to fully characterize the safety profile of this new rotavirus vaccine.

Safety profile of a recombinant hepatitis E vaccine

The recombinant hepatitis E vaccine (Hecolin), also designated HEV 239, encompasses amino acids 368-606 of the HEV open reading frame 2 (ORF2) capsid protein from HEV genotype 1 and is expressed as a non-fusion protein in Escherichia coli. The purified HEV 239 assembles as homodimers resulting in viruslike particles. The vaccine contains 30 μg of the purified antigen and 0.8 mg aluminium hydroxide suspended in 0.5 ml buffered saline. It is manufactured by Xiamen Innovax Biotech, Xiamen China. It was approved by the Chinese Food and Drug Administration in 2011 and has been available since October 2012.

Pre-licensure safety and immunogenicity data were presented. The vaccine was evaluated in Phase 1 and 2 prelicensure trials designed to evaluate the optimal dose and regimen of Hecolin. The efficacy and safety of the vaccine was evaluated in a randomized, double-blind, controlled Phase 3 clinical trial in >112 000 healthy subjects 16–65 years of age irrespective of anti-HEV antibody status. That study was conducted between August 2007 and June 2009 in the Jiangsu Province of China, a region where HEV genotype 4 is the predominant genotype, with genotype 1 also circulating. Study participants were randomized 1:1 to receive either Hecolin (n = 56 302) or a licensed hepatitis B vaccine (n = 56 302) administered through intramuscular injection in 3 doses (0, 1 and 6 months). Participants were followed for 19 months.

Data showed that the vaccine was immunogenic and efficacious. To assess local and systemic solicited and unsolicited adverse events a subset consisting of 1316 participants in the Hecolin group and 1329 participants in the control group (participants from one township) were observed for 30 minutes and then assessed by home visits at 6, 24, 48 and 72 hours, as well as 7, 14 and 28 days after each dose. Other study participants were asked to report any adverse events to nearby clinics within 1 month after each dose without active follow-up visits. In addition, the investigators used data obtained from the local medical insurance system covering the whole study area (11 townships) to identify hospitalizations and deaths among the trial participants during the 19-month follow-up period. After reviewing the hospital records related to these serious adverse events, the investigators categorized them according to the Medical Dictionary for Regulatory Activities (MedDRA). Overall, safety data derived from Phase 1, 2 and 3 clinical trials suggested that the vaccine was well tolerated. Short-term (72 hours) local and systemic solicited adverse event data obtained from the “reactogenicity subset” that participated in the Phase 3 clinical trial showed more frequent local adverse events in the Hecolin group compared to the active control group. The solicited systemic adverse events and unsolicited adverse events occurred at similar rates between study groups. There appeared to be no difference in serious adverse events or death identified in the Hecolin group compared to the active control group. Because there was no active follow-up the number of severe adverse events and deaths among trial participants could not be ascertained.

In the Phase 3 clinical trial there were 37 women in the Hecolin group and 31 women in the placebo group who were inadvertently administered vaccine during pregnancy. The vaccine appeared to be well tolerated in pregnant women with rates of adverse events similar to those observed in matched non-pregnant women. Nineteen of the pregnant women in the Hecolin group and 14 in the active placebo group underwent elective abortion. There were 18 and 17 live births in the Hecolin and active control groups, respectively. Weight, body lengths and gestational age of the babies were comparable in the 2 groups. However, the overall sample size was too small to allow a conclusive statement on the safety of Hecolin in pregnant women and their babies. The safety of Hecolin was also evaluated in HBsAgpositive persons and the data are reassuring; however the analysis subset did not include persons with ongoing liver disease as this was an exclusion criterion for the trial.

The Chinese CDC has established an online reporting system to collect post-marketing safety information and to date has not identified any safety concerns. A small Phase 4 trial in the elderly (>65 years of age) is ongoing, as well as an extended follow-up trial of the Phase 3 study cohort. In addition, Hecolin is being used as an active control arm in an ongoing Phase 3 study of a human papilloma virus (HPV) vaccine in approximately 7300 healthy women.

In summary, available safety data on Hecolin derived from Phase 1, 2 and 3 clinical trials in healthy subjects are reassuring. However, GACVS noted that there are no safety data in paediatric subjects (<16 years of age), the elderly (>65 years of age), persons with underlying diseases or conditions such as those who are immunosuppressed persons or have liver disease and thus recommended that studies be conducted to assess the safety of Hecolin in these subpopulations. Any follow-up of those inadvertently vaccinated in pregnancy during the HPV trial should be useful to assess safety in this group. The Committee also noted that there are as yet no studies to evaluate the safety and immunogenicity of Hecolin when given concomitantly with other vaccines. In addition, GACVS recommended that a Phase 4 post-marketing study be conducted once the vaccine is in more widespread use to further assess the safety profile of Hecolin, in particular with regard to serious and rare adverse events.

Meningococcal A conjugate vaccine during pregnancy

The Committee was presented with results from an open label observational evaluation of the safety of a meningococcal A conjugate vaccine in pregnancy (MenAfriVac, manufactured by the Serum Institute of India), conducted in Ghana. The vaccine is a lyophilized group A conjugate vaccine developed under the Meningitis Vaccine Project.3 It contains PsA10 μg, TT conjugate 10–33 μg, aluminium phosphate adjuvant 0.3mg Al3+ and thiomersal 0.01%, in each 0.5 ml dose.

The GACVS had been following the vaccine from its initial Phase 1 and 2/3 clinical trials with just over 1100 subjects, through to licensure and the first mass immunization campaigns conducted in early adopter countries.4 By the time of the last update, in June 2011, over 50 million doses had been administered. At each update, the GACVS continued to be reassured of its ongoing safety while recommending specific actions be taken to help ensure continued safety vigilance and attention to key aspects of safety. As is common, clinical trials during vaccine development and licensure did not target pregnant women, however inadvertent vaccination in pregnancy throughout the early phases had not revealed any concerns.

GACVS supported WHO’s technical guidance5 that MenAfriVac should be offered to pregnant and lactating women from the African meningitis belt during any stage of pregnancy or lactation, while recommending that a plan be developed to follow up women in antenatal or obstetric clinics, and to monitor pregnancy outcomes by making appropriate comparisons with unvaccinated pregnant women.6

This study was developed in response to those recommendations. The Navrongo Health Research Centre, Ghana, is part of the INDEPTH network that collects continuous longitudinal demographic and health data and outcomes within its populations. In the surveillance area (covering 2 districts – Kassena-Nankana East and Kassena-Nankana West), >156 000 individuals are part of this demographic surveillance, and receive visits by study teams about 3–4 times per year to update their demographic and health status. A mass vaccination campaign was held in Ghana between 9 and 19 October 2012, targeting individuals between 1 and 29 years of age including pregnant women. While pregnancy was not considered a contraindication during this campaign given the benefit of vaccination, some pregnant women elected not to receive the vaccine. Given the participation of the districts in the INDEPTH network, this provided an opportunity to evaluate the safety of the meningococcal A vaccine by comparing the rates of pregnancy- related outcomes in vaccine recipients, with rates among unvaccinated pregnant women. In addition, a second age and season matched historical control group was assembled to document pregnancy outcomes in a time period before the immunization campaign. Outcomes included overall maternal, fetal and neonatal mortality, overall rates of spontaneous abortions, still births, perinatal deaths, prematurity, low birth weight, small for gestational age, and rates of caesarean section.

A total of 1730 pregnant women were vaccinated during the campaign, while 919 pregnant women elected not to be vaccinated. A total of 3551 pregnant women were in the historical unvaccinated control group. Comparing the outcomes, there was no significant difference in any of the pre-specified outcomes between women who had received the meningococcal A conjugate vaccine and those who had not, either in the concurrent or historical comparison groups. Mean birth weights were over or near 2900g in each group, and gestational age over 37 weeks. Rates of miscarriage and stillbirth were 1.8% in the vaccinated and 2.2% in both control groups, with prematurity 3.1% among the concurrent controls, 3.6% in the vaccinated and 5.6% among the comparison controls.

The Committee noted the quality of this study and its reassuring results. It highlighted the potential for this study methodology to examine the safety of vaccines in mass immunization contexts, especially in outcomes as complex as pregnancy. Previous attempts to study pregnancy outcomes have been less robust. Some details were more difficult to ascertain in this study, including reasons for spontaneous abortion, impact of vaccination during lactation and other factors that the demographic survey questions had not collected. However, future studies could be planned that would include additional variables added to the survey visits to address these questions.

Given that mass immunization campaigns have been staggered and another 100 million individuals are in line for vaccination over the next 2 years, opportunities for additional evaluation of meningococcal A conjugate vaccine in pregnancy will be available. This may include an opportunity to evaluate the vaccine’s safety during lactation. This present study was conducted using existing infrastructure in the Navrongo Health Research Centre and thus may provide a powerful tool in these evaluations.

Overall, in the almost 4 years since MenAfriVac was rolled out in the first mass campaigns, and beginning even earlier with the clinical trials, no concerns have been identified regarding its use in pregnancy. As with other inactivated vaccines, neither pregnancy nor lactation are contraindications for vaccination in situations of increased disease risk. Given the emerging evidence of the effectiveness of this meningococcal A conjugate vaccine in controlling disease in the countries of the meningitis belt in Africa, more permissive language in the package insert may be warranted.

Preparing for malaria vaccine introduction

The most recent WHO malaria mortality estimate is 627 000 deaths for the year 2012. While this represents an estimated 42% reduction in global malaria mortality rates since 2000 in association with a large scaling up of WHO recommended preventive, diagnostic and treatment measures, there remains a need for additional preventive measures including vaccines. As one candidate malaria vaccine has reached the regulatory evaluation stage, GACVS considered the need for post- licensure safety assessment for when malaria vaccines become available for public use.

GACVS considers that the development of recommendations for post-licensure safety assessment of malaria vaccines is an important preparatory step, in order toprovide early implementing sites with sufficient time for planning, training and improving or developing surveillance systems. Early identification of sites would also have the benefit of allowing the establishment of active surveillance for events of special interest, thereby providing background rates for those events prior to vaccine introduction. GACVS noted that the safety guidance would be developed alongside effectiveness and impact guidance and that it was important to ensure harmonisation with this guidance, as it is likely that studies could be designed to examine the impact of both safety and effectiveness. GACVS also noted the guidance was intended for use by the public sector of implementing countries to assist them to conduct independent studies and be prepared to assess data obtained by the manufacturer.

GACVS discussed the principal elements of such recommendations and suggested that the main components should cover on-going strengthening of routine systems for reporting adverse events following immunization (AEFIs), stimulated passive reporting in selected settings, such as health demographic surveillance system sites and active follow-up for specific events of interest using suitable epidemiological designs – such as casecontrol, self-controlled case series and cohort event monitoring – to enable testing of hypotheses and quantification of risks. These components would allow detection and evaluation of signals for rare unexpected events as well as assessment of events of interest from clinical trials – in particular febrile convulsions and meningitis. It was noted that it is important that lessons are learnt from the experience in Africa of safety studies for meningococcal serogroup A vaccine introduction, but that a key difference was that the vaccine would probably be introduced with a routine schedule rather than through large mass campaigns. It was also noted that it is important that the guidance balances the need for high quality studies with what is feasible in the settings where these studies are likely to be done. This includes, for example, use of case definitions adapted to local clinical practice. Further discussion focused on the importance of considering rare but serious events and possible mechanisms for following up vaccine recipients such as diary cards, issuing of mobile telephones or identifying patients through hospital admissions. It is expected that a guidance document will be available in mid-2015.

Fifteen years of GACVS: challenges and opportunities

On the occasion of its 15th anniversary, the Committee reviewed its accomplishments and reflected on new challenges in view of the evolving public health environment. The first GACVS meeting took place on 14–15 September 1999, and the Committee’s first report addressed macrophagic myofasciitis.7 Since then, the Committee has met regularly twice yearly and has also been convened by telephone conference more frequently when needed. The Committee’s regular reports are published soon after each meeting in the WHO Weekly Epidemiological Record, while urgent reports are posted separately on-line, and a compendium is available on the GACVS website maintained by WHO.8 Since its establishment the Committee has produced >100 reports related to vaccine safety issues. The role of the GACVS is primarily to assess risks related to vaccine use in order to assist policy-makers in identifying benefit and risks as part of evidence-based vaccination policies. GACVS risk assessments are regularly used by WHO advisory bodies, including the Strategic Advisory Group of Experts (SAGE), the Expert Committee on Biological Standards (ECBS) as well as regional technical advisory groups related to immunization.

In addition to its regular committee reports, GACVS also issues statements in response to urgent vaccine safety issues. If required, the Committee can be convened urgently by conference call in order to respond to alerts of international significance. More recently, GACVS has also looked into capacity building aspects of global vaccine pharmacovigilance. The Committee has, in particular, provided advice on the development of the Global Vaccine Safety Blueprint9 – WHO’s strategy to optimize the safety of vaccines through effective use of pharmacovigilance principles and methods in all countries – and is now advising on the development of specific tools for vaccine safety monitoring, including the classification of AEFI, core data elements, and indicators for surveillance systems.

A total of 39 experts have served on GACVS to date, the current Committee being composed of 15 members. Current and past members represent all WHO regions, although a majority (26) originate from industrialized countries in Europe, North America or Australia. They provide expertise in multiple fields related to vaccine safety including epidemiology, statistics, clinical medicine, pharmacology and toxicology, infectious diseases, public health, immunology, vaccinology, pathology, ethics and health product regulation. GACVS members, in addition to participating in bi-annual in-person meetings also contribute to the work of the Committee through various subgroups which develop statements on selected topics between regular meetings.

Perspectives were presented on the relevance of the work of the GACVS, including views from an immunization programme, a regulatory authority, a vaccine clinician, a vaccine communication expert, a pharmacovigilance collaborating centre, and a WHO advisory committee. The discussion highlighted several specific examples where GACVS provided timely and useful guidance. Those have addressed some time-limited issues such as the risk of Bell’s palsy following intranasal vaccination in 2002,10 transmissible spongiform encephalopathy raised in 200411 and conjugate meningococcal vaccines and Guillain-Barre syndrome in 2005.12 Other vaccines were reviewed and re-visited over time as new evidence accumulated or in response to new concerns. An example concerned the vaccine preservative thiomersal which was first discussed in 2002; additional reassuring evidence became available over time and a more comprehensive review was provided in preparation for the United Nations Environmental Program development of a global legally binding instrument on mercury, now known as the Minamata Convention on Mercury (2013).13 Similarly, evolving evidence related to the risk of intussusception associated with rotavirus vaccines has been continuously reviewed since 2005.14

The discussion highlighted the needs for an evolving approach in several domains. Particular considerations were given to: i) the evolving technical aspects of vaccine pharmacovigilance; ii) process issues related to GACVS operations; and iii) communication of GACVS findings. With respect to technical aspects, the main needs relate to the increasing number of new vaccine products that are becoming available for immunization programmes and their rapid availability for populations that are not served with robust safety monitoring systems. In some instances, some vaccines are specifically designed for rapid roll-out in parts of the world that report very few AEFIs and have not developed strong expertise for investigating specific concerns or actively monitoring them. This requires additional guidance from GACVS, not only in assessing available evidence but in identifying gaps in knowledge and proposing approaches that could be reasonably expected to answer the most pressing questions in those particular settings (e.g. the safety of meningococcal A conjugate vaccine used during pregnancy, as reviewed during this meeting). In addition, the methods by which vaccines are developed and produced is evolving, relying on newer technologies and processes. Likewise, individual susceptibility to vaccine reactions varies and new methodologies, including genomics, could potentially provide useful insights with respect to predisposing factors and ways to minimize such risks.

With respect to the GACVS process, the Committee’s independence – including from the WHO secretariat – and high level of individual expertise are the main features that can maintain the credibility and impact of its advice. Participants also highlighted the need to maintain the highest possible standards with respect to the review of scientific evidence and adjusting with evolving methodologies. This implies a greater use of systematic and graded reviews when an association between a vaccine and a particular health event has been studied in many parts of the world. However this requirement will not affect most of GACVS’ work, since it has increasingly focused on accompanying the early post-licencing use of new vaccine products for which the scientific evidence is usually available from a limited number of sources. Throughout its existence, GACVS has operated through closed sessions. This confidential process was deemed necessary in order to ensure that all available data, including proprietary information, could be considered. Ensuring that committee deliberations could be protected from undue influence was another consideration. In view of evolving standards relating to committees of public importance, it was recommended that a transparency policy be developed with a view to providing more specific information on how conclusions were reached.

GACVS communication is currently directed to WHO technical audiences through the WHO Weekly Epidemiological Record and website. These audiences are immunization managers and policy makers in health ministries, regulatory authorities, professional organizations and immunization advisory groups in Ministries of Health. One GACVS initiative to reach out to broader audiences, and help counteract anti-immunization groups, was launched in 2003 through the creation of the Vaccine Safety Net. In order to assess information on vaccines publicly available on the internet, GACVS developed 4 categories of criteria for good information practices – regarding credibility, content, accessibility and design – to which sites providing information on vaccine safety should adhere. WHO evaluates websites for their adherence to these criteria and provides a list of resources in multiple languages. Vaccine safety communication, however, should be developed further and it was proposed that WHO examine approaches taken in other sciences where public risk is an important consideration, in order to develop and promote more effective practices.

The Committee concluded that although its work is well established and recognized, it was critical to remain mindful of current vulnerabilities. The evolving global vaccination landscape requires a continuous adjustment of methods and processes. A detailed report in preparation will provide a more comprehensive account of this analysis and propose a way forward to ensure that independent advice on vaccine safety issues for WHO remains relevant and timely.

1 See No. 41, 1999, pp. 337–338.

2 GACVS invited additional experts to present and discuss evidence related to particular topics. These experts included persons affiliated with: Bahrat Biotech, Hyderabad, India; Cincinnati Children’s Hospital Medical Centre, Cincinnati OH, USA; Dalhousie University, Halifax, Canada; Food and Drugs Administration, Manila, Philippines; Innovax Biotech, Xiamen, China; London School of Hygiene and Tropical Medicine, London, United Kingdom; Navrongo Health Research centre, Navrongo, Ghana; Postgraduate Institute of Medical Sciences, Lucknow, India; University of Colorada, Aurora CO, USA; University of Ghana, Accra, Ghana.

3 See

4 See No. 30, 2011, pp. 321–324.

5 See Women__during_MenA__conj_Campaigns_22Nov10_april_2011.pdf

6 See No. 30, 2010, pp. 285–291

7 See No. 41, 1999, pp. 338–340.

8 See

9 WHO 2012. Global Vaccine Safety Blueprint. WHO/IVB/12.07

10 See No. 47, 2002, p. 393.

11 See No. 1, 2005, pp. 4–5.

12 See No. 2, 2006, p. 18.

13 See

14 See