Global Vaccine Safety

Global Advisory Committee on Vaccine Safety,
6-7 June 2018

The Global Advisory Committee on Vaccine Safety (GACVS), an independent expert clinical and scientific advisory body, provides WHO with scientifically rigorous advice on vaccine safety issues of potential global importance.1 GACVS held its 38th meeting in Geneva, Switzerland, on 6-7 June 2018.2 The Committee discussed 2 vaccine safety issues: pharmacovigilance in the RTS,S malaria vaccine pilot study and data on dengue vaccine from the Philippines. It also reviewed three generic issues: progress in the Global Vaccine Safety Initiative (GVSI), communication about vaccine safety and new developments in the Vaccine Safety Net (VSN).

Pharmacovigilance in pilot use of malaria vaccine

Following a joint review convened by the African Vaccine Regulatory Forum (AVAREF), the national regulatory authorities of Ghana, Kenya and Malawi granted special authorization in May 2018 for use of the RTS,S malaria vaccine in the planned pilot implementation programme. It is anticipated that introduction will commence later this year. GACVS has assessed the safety profile of RTS,S throughout its development and clinical trials and will continue to assess safety data arising from the pilot implementation.3 Safety data will be derived from: (i) post-marketing monitoring of cohort events by the manufacturer GlaxoSmithKline (GSK), with detailed active follow-up; (ii) surveillance of mortality throughout the pilot area and surveillance of meningitis and cerebral malaria in sentinel hospitals in both control and RTS,S areas; (iii) active surveillance of adverse events of special interest (AESI); and (iv) pharmacovigilance through passive reports of adverse events following immunization (AEFI) with all vaccines from each country. A communications strategy and mechanisms for sharing these data during implementation are being developed to ensure that data flow between and within each country (as part of the pilot study and as part of national pharmacovigilance) and from GSK to the African Regional Safety Committee and the coordinators of the pilot implementation study, AVAREF and GACVS.

In June 2017, GACVS endorsed criteria to ensure that each country had a functioning system for reporting and assessing AEFI in time for introduction of the pilot project.4 The criteria are: (i) a minimum of 10 AEFI reports per 100 000 surviving infants; (ii) a functioning AEFI committee that meets regularly; (iii) trained, resourced AEFI investigation teams; (iv) safety communications plans evaluated and tested; (v) an identified focal person in each country’s expanded programme on immunization to oversee and ensure optimal reporting and training; and (vi) methods for active surveillance of AESI developed and data collection initiated.

Each country reported progress at the GACVS meeting in December 2017 and have now provided further updates. Ghana is meeting all the criteria. Regular train¬ing is provided to health care workers in reporting AEFI, and further training is planned for both the AEFI committee and the investigation teams. Reporting targets are shared with regions, and job aids have been developed. A communications plan is ready to be tested. In Malawi, although the reporting rates for 2017 are on target, more training is required in the pilot areas to sensitize health care workers and for investigation teams to assess AEFI and to use the reporting tools. Communications plans have been developed and have been pilot-tested, and training is planned. In Kenya, although the reporting rate is below 10 per 100 000, sensitization projects were begun in May 2018, and a second round of sensitization is planned just before introduction of RTS,S. Electronic reporting is being pilot-tested in one county. A communications plan is being drafted and members of the national AEFI committee appointed. Training for AEFI investigation teams is planned.

GACVS encouraged each country to ensure training of investigation teams and sensitization for reporting as soon as possible, so that AEFI surveillance would be functioning when pilot implementation began. It also stressed the importance of ensuring the timely availability of individual data on AEFI in order that the levels and quality of reporting can be monitored regularly throughout the pilot project and any improvements made when necessary. GACVS also noted that barriers to reporting, such as the belief of health care workers that a report indicates an error on their part, should be addressed in training.

For criterion (vi), on active surveillance, a working group with representatives from each country, WHO and the Centers for Disease Control and Prevention (Atlanta, GA, USA) has produced a manual that can be adapted in each country to its protocols for AESI surveillance. The manual, which covers 11 adverse events with both Brighton case definitions and simplified working case definitions, was presented for comment to GACVS. Surveillance would be limited to the duration of the pilot implementation project and to the target age group in both the control areas and those receiving RTS,S. Each country will identify which health care workers and health care facilities are to be responsible for active surveillance. Cases will be identified by regular review and extraction of case details at the facilities on reporting forms. The data will then be entered into a dedicated AESI database. GACVS agreed that development of country protocols, training and testing should proceed as soon as possible, and these activities are planned for the third quarter of 2018. GACVS noted that, if pilot implementation of RTS,S starts later in 2018, AESI surveillance may not be fully in place. Surveillance is an important component of the safety evaluation that allows comparison of control areas with those in which RTS,S is implemented and should be initiated as soon as possible.

Safety of dengue vaccine in the Philippines

GACVS last reviewed the CYD-TDV dengue vaccine at its meeting on 6–7 December 2017.5 The Committee noted that long-term follow-up in clinical efficacy trials indicated that, overall, vaccinated trial participants had a reduced risk of virologically confirmed severe dengue and hospitalization; however, a subset of trial participants who had not been infected with dengue virus before vaccination (i.e. dengue-naïve, seronegative according to the NS1 assay) had a higher risk of severe dengue and hospitalization. The new evidence presented at that meeting was based on a reanalysis of the clinical trial data by the manufacturer, with a new test that distinguishes individuals with and without previous exposure to wild dengue virus retrospectively.6 The WHO Strategic Advisory Group of Experts (SAGE) on immunization previously identified research on vaccine safety in this seronegative population as a priority.7 Following the December 6–7 meeting in 2017, GACVS recommended that CYD-TDV not be administered to individuals who have not been previously infected with wild dengue virus. GACVS also noted that no data are currently available to allow an analysis of risk according to the number of vaccine doses received by people who are seronegative at baseline.

At its meeting on 17–18 April 2018, SAGE advised countries considering CYD-TDV vaccination as part of their dengue control programme to include pre-vaccination screening, so that only dengue-seropositive persons are vaccinated; the limitations of such screening should be clearly communicated to those offered vaccination.8

WHO will release a revised position paper on dengue vaccine in September 2018. The purposes of an update of the GACVS statement on dengue vaccine are: (i) to review the reports on vaccine safety received by the Philippines Ministry of Health after announcement of the risk for severe dengue of vaccine recipients who were dengue-naïve at the time of vaccination; (ii) to review difficulties in determining whether, apart from vaccine failure, the cases of severe dengue in vaccine recipients who were dengue-naive at the time of CYD-TDV vaccination were due to vaccine-related immune enhancement; and (iii) to review the updated safety profile of CYD-TDV.

The Philippines Food and Drug Administration approved use of CYD-TDV in December 2015, and the Disease Prevention and Control Bureau proposed its introduction as part of the National Dengue Prevention and Control Program. Vaccine administration began in 2016, first as part of a school programme in highly endemic regions and then extended to community programmes in October 2016. Surveillance of the safety of all vaccines is well established in the country, as a part of integrated disease surveillance and response. Should a serious AEFI or cluster be detected, the epidemiology bureau of the Department of Health is notified within 24–48 h. Serious cases are investigated, and the results of the investigations are compiled and sent to the regional and national AEFI committees. Before the programme was suspended, over 875 000 children had received at least 1 dose, almost 350 000 had received all 3 doses, and about 400 000 had received 2 doses.

Post-marketing data were presented to GACVS by the manufacturer. CYD-TDV is registered in 20 countries, and most doses are distributed in Brazil (where it is used in a public programme in Parana State) and the Philippines. In Brazil, dengue cases are reported through a national reportable disease information system, and data on AEFI are collected through passive surveillance in a national immunization programme. Guidelines for enhanced reporting and training of vaccine centre workers were provided by local authorities in Parana State.

The 14 fatal case reports in the Philippines were first reviewed by the national AEFI committees and the Dengue Investigative Task Force (DITF). The reports included 3 cases of dengue shock syndrome and 6 cases with other clinical diagnoses and no clear causal link other than a temporal association. The other cases were coincidental (3) or unclassifiable (2). A further review of 12 cases (8 fatal and 4 non-fatal) was undertaken by the DITF after training in AEFI methodology by international specialists. Although the DITF found that most cases were indeterminate, coincidental or unclassifiable, it recognized several cases of dengue disease. GACVS maintained its earlier recommendation that CTD-TDV should not be administered to people who have not previously been infected with wild dengue virus. It concluded that, in the absence of criteria for distinguishing vaccine failure from vaccine-related immune enhancement, individual cases cannot be attributed to one or the other. As a result, such cases should be classified as indeterminate, irrespective of the time since vaccination.

Between December 2015 and March 2018, 1876 adverse events were reported to the manufacturer, mainly from Brazil and the Philippines; reporting was consistent with the pattern of dose distribution in both countries. The most frequently reported adverse events were fever, headache, dizziness, vomiting and rash. Of the 211 serious AEFI reported, most were consistent with an underlying infectious disease, including dengue fever. By 20 March 2018, 87 cases of dengue infection had been reported after vaccination with CYD-TDV; 23 were serologically confirmed, 61 suspected with no virological confirmation and 3 with negative virological tests. Of the 87 dengue cases, 14 were fatal. Of the 14 cases, 6 had completed the vaccination schedule, 3 had received 2 doses and 5 had received only 1 dose. All 9 cases for which the interval between vaccination and disease onset was known occurred within 6 months of the last vaccination.

Progress was reported in cohort event monitoring, sponsored by the manufacturer to obtain information on selected AEFI and serious adverse events in people vaccinated with CYD-TDV over 5 years in Brazil, Mexico and the Philippines. The target for enrolment in the study of post-authorization safety is 30 000 vaccinated participants. As of 5 April 2018, 12 573 participants had been enrolled and had received at least 1 dose of CYD-TDV.

One of the challenges in conducting post-market surveillance after vaccination with CYD-TDV is determining whether the vaccine gives rise to vaccine-related immune enhancement. An increasing number of AEFI were reported after suspension of the vaccination programme in the Philippines and media coverage. A task force was established by the Department of Health to review all fatal cases, and guidelines on AEFI reporting and response to vaccine recipients were issued by the Department of Health. In addition, the National AEFI Committee, established in 2012, was charged with reviewing all non-fatal AEFI.

GACVS also examined the possible risk of viscerotropic or neurotropic disease associated with the yellow fever backbone of the CYD-TDV vaccine. Although this remains a theoretical possibility, non-clinical and clinical evaluations do not provide evidence of an association. Viscerotropic and neurotropic diseases are rare serious reactions to yellow fever vaccination and occur only in close temporal association with vaccination. As severe dengue may also be accompanied by haemorrhagic systemic phenomena, a differential diagnosis can be made only if the vaccine strain is isolated from affected organs and if such syndromes occur within the accepted interval between vaccination and symptom onset (8 days).

Progress in the Global Vaccine Safety Initiative

The Global Vaccine Safety Blueprint, a framework of 8 objectives for enhancing global vaccine safety activities, was last discussed by the Committee in 2012. The Blueprint was later used as the basis for the vaccine safety strategy in the Global Vaccine Action Plan. GACVS provided input to the Blueprint and implemented it in the GVSI. At its meeting in December 2012, the Committee reviewed the GVSI work plan and examined areas of interaction between its own mandate for vaccine safety issues of global importance and that of the GVSI to support global vaccine pharmacovigilance capacity.9 GACVS was notified of progress made through the GVSI in achieving its objectives 6 years after launch of the Initiative and was informed about a programme to strengthen global monitoring of vaccine safety, the Global Vaccine Safety Observatory. It also discussed the global vaccine safety strategy in the context of development of the Global Vaccine Action plan after 2020.

The Blueprint vision of effective vaccine pharmacovigilance systems established in all countries has progressed steadily. Countries are reporting AEFI and are meeting indicators of improvement in safety surveillance capacity. Six annual GVSI meetings have brought partners and countries together to build collaborations and plan future activities. Resources, training packages on basic vaccine safety, guidelines, AEFI surveillance and management, signal detection and communications are integral to robust building and maintenance of capacity for vaccine pharmacovigilance and trust in immunization programmes.

GACVS has advocated for the GVSI and supported its objectives on several fronts, from assisting in develop¬ment of tools and helping to identify priorities to responding to safety concerns raised by countries either during the regular 6-monthly meetings or ad hoc. Five of the 8 strategic objectives of the GVSI benefit directly from input by GACVS: AEFI monitoring, investigation, harmonized tools and methods, technical support platforms and expert advice.

The concept of the Global Vaccine Safety Observatory was discussed. It was conceived as a clearinghouse for data on vaccine safety systems to assist member coun¬tries in achieving the Blueprint objectives. The Observatory will start with 4 regional nodes that provide academic, programmatic, regulatory and technical expertise. The expected outputs of the Observatory include presentation and analysis of relevant data, a website to provide indicators of vaccine safety capacity and links to relevant activities for vaccine vigilance, and an annual report. The products of the Observatory will be disseminated through several activities, some of which have had to evolve and recruit more than mini-mal capacity to deal with emerging safety issues. Each node will present relevant specialized data to allow members to track and compare progress over time, aggregate more sensitive data regionally, share regulatory recall and safety alerts, map globally reported safety concerns and make links to relevant experience.

Finally, as the Decade of Vaccines will be completed by 2020, a new vaccine strategy is being developed, which will be aligned with the recently approved WHO General Programme of Work 2019–2023 in support of the sustainable development goal for health. GACVS therefore recommends close collaboration to ensure that the global vaccine safety strategy is well positioned in the new global approach to immunization.

Vaccine safety communication

A new GACVS subcommittee on vaccine safety communication has been established in order to integrate safety assessments with better capacity to communicate them. It is proposed that a framework and templates for communication on vaccine safety be prepared by mapping vaccine safety communication activities throughout the life cycle of products, examining current vaccine safety communication tools and identifying gaps, and proposing approaches to fill the gaps. The first task of the subcommittee was to prepare a more detailed action plan, with case studies to illustrate how safety is communicated under various circumstances.

The Committee noted the extensive strategies and education resources already available for avoiding and mitigating crises in communicating vaccine safety and highlighted two. 1. The new Council for International Organizations of Medical Sciences (CIOMS)10 Guide to Vaccine Safety Communication covers strategic communication issues, particularly for regulators but also for immunization programmes and other stake¬holders. It stresses the importance of personnel with appropriate skills in safety communication and provides guidance and templates for a communications plan. 2. The WHO Vaccination and Trust Library includes overviews of how concerns arise, describes the role of communication in mitigating crises, provides resources to help immunization stakeholders to avoid and manage crises and outlines a training programme.

Strategies, guidelines and resources are available for mitigating and managing vaccine safety crises. A case study was presented that highlighted the value of trust, clear communication at all levels and credible sources. GACVS considers that vaccine safety communication requires coordination among many stakeholders in 5 areas: (i) a framework that includes scenarios and proposes common principles for addressing specific situations; (ii) common messaging of vaccine safety issues for global partners; (iii) sharing of existing communications resource materials through an e-library; (iv) quality standards for planning and implementing vaccine safety communications; and (v) collabo-ration and coordination of partners so that each stakeholder has opportunities to make actionable contributions.

Vaccine safety net

The Vaccine Safety Net (VSN) is a WHO initiative initially launched to identify trustworthy information on vaccine safety and immunization on the Internet.11 GACVS supports the VSN by providing advice and criteria for website quality and content, thereby facilitating access by public health authorities, health professionals and the public to reliable information on vaccine safety. There are currently 58 member websites in 16 languages, covering the 6 WHO regions.

VSN members met on 4–5 June 2018 in Veyrier-du-Lac (France), for the second time in less than 2 years, to review the status of their activities, reflect on recent advances in social media and the Web and further discuss approaches, strategies and challenges in managing digital information and communication on vaccine safety. A preliminary report of the meeting was presented to GACVS. Despite increasing recognition worldwide, VSN members identified a number of challenges, including additional investment. The Net requires more partnerships and collaborations, qualitative research based on the experience of the VSN websites, communication research involving VSN members, engagement of young professionals and students in vaccination communication to stimulate more engagement by advocates and champions, and engagement of global and regional foundations in building vaccine acceptance and addressing vaccine hesitancy.

The good alignment of VSN members provides new opportunities for research. A recently explored area is web analytics to document patterns of web-searching on specific vaccine safety issues around the globe and at each VSN site. Web analytics could also be used to monitor the effects of digital communication strategies in real time. Research on measuring, understanding, tracking and addressing vaccine confidence was identified as another important area. A digital toolkit or newsletter would provide updates, tips, lessons learnt and risk communication guidance and resources for responding to vaccine safety events that occur locally in member countries.

During the 2-day meeting, participants were presented with preliminary results from the VSN web analytics project and plans for digital communication models for vaccine safety. GACVS continues to seek improved communication of vaccine safety information to the public and to its partners and therefore welcomes the contribution of the VSN and supports the work presented. In the overloaded web communications environment, where information competes for attention, easy access to reliable, trustworthy content on vaccina¬tion and immunization remains of paramount importance.

1 See No. 41, 1999, pp. 337–338.

2 GACVS invited additional experts to present and discuss evidence related to particular topics. The experts included people affiliated with: Department of Health, Manila, the Philippines; Food and Drugs Authority and Expanded Programme on Immunization, Ghana Health Service, Accra, Ghana; Pharmacy and Poisons Board and National Vaccine and Immunization Programme, Nairobi, Kenya; Pharmacy, Medicines and Poisons Board and Expanded Programme on Immunization, Ministry of Health, Lilongwe, Malawi; Centers for Disease Control and Prevention, Atlanta (GA), USA; Ospedale Pediatrico Bambino Gesu, Rome, Italy; Monash Children’s Hospital, Melbourne, Australia; and Sanofi Pasteur, Lyon, France.

3 See No. 28, 2017, pp. 393–396.

4 See No. 3, 2018, pp. 17–19.

5 See No. 3, 2018, pp. 21–25.

6 Sridhar S et al. Effect of dengue serostatus on dengue vaccine efficacy. N Engl J Med 2018. doi: 10.1056/NEJMoa1800820.

7 See No. 21, 2016, pp. 282–284.

8 See No. 23, 2018, pp. 337–340.

9 See No. 6, 2013, pp. 69–70.

10 See

11 See WHO, Vaccine safety net (, accessed June 2018).