Global Vaccine Safety

Global Advisory Committee on Vaccine Safety, report of meeting held 17-18 June 2009

Published in the WHO Weekly Epidemiological Record on 7 August 2009

The Global Advisory Committee on Vaccine Safety, an expert clinical and scientific advisory body, was established by WHO to deal with vaccine-safety issues of potential global importance independently from WHO and with scientific rigour.1 The Committee held its twentieth meeting in Geneva, Switzerland, during 17–18 June 2009.2 The Committee reviewed the safety profile of a malaria vaccine entering phase III clinical development, updated the safety profiles of rotavirus vaccines and human papillomavirus (HPV) vaccines, discussed preparedness for novel influenza virus vaccines and completed a review of the safety of using measles vaccines in people infected with HIV.

Safety of experimental malaria vaccine RTS,S/AS01

The Committee reviewed clinical safety data from phase I and phase II trials of the candidate malaria vaccine RTS,S/AS01 (GlaxoSmithKline Biologicals, Rixensart, Belgium). The designation “RT” refers to approximately 190 amino acids from the carboxy terminus of the Plasmodium falciparum circumsporozoite protein, and “S” refers to the hepatitis-B surface antigen. RTS,S virus-like particles form when the RTS malaria–hepatitis B fusion protein is co-expressed with S antigen alone in yeast cells (Saccharomyces cerevisiae). The adjuvant AS01 consists of liposomes plus MPL plus QS21. An earlier version of the RTS,S vaccine was adjuvanted with AS02 (an oil–water emulsion plus MPL plus QS21). The paediatric dose of RTS,S for the phase III trial is 25 µg using a 3-dose schedule delivered intramuscularly at the ages of 0 months, 1 month and 2 months. The current RTS,S/AS01 vaccine has completed phase II trials in children aged 5–17 months and also in trials using the Expanded Programme on Immunization (EPI) for children aged 6–14 weeks; multicentre phase III trials in infants and young children started in May 2009 in sub-Saharan Africa.

The Committee reviewed clinical safety data from phase I and phase II trials of the candidate malaria vaccine RTS,S/AS01 (GlaxoSmithKline Biologicals, Rixensart, Belgium). The designation “RT” refers to approximately 190 amino acids from the carboxy terminus of the Plasmodium falciparum circumsporozoite protein, and “S” refers to the hepatitis-B surface antigen. RTS,S virus-like particles form when the RTS malaria–hepatitis B fusion protein is co-expressed with S antigen alone in yeast cells (Saccharomyces cerevisiae). The adjuvant AS01 consists of liposomes plus MPL plus QS21. An earlier version of the RTS,S vaccine was adjuvanted with AS02 (an oil–water emulsion plus MPL plus QS21). The paediatric dose of RTS,S for the phase III trial is 25 µg using a 3-dose schedule delivered intramuscularly at the ages of 0 months, 1 month and 2 months. The current RTS,S/AS01 vaccine has completed phase II trials in children aged 5–17 months and also in trials using the Expanded Programme on Immunization (EPI) for children aged 6–14 weeks; multicentre phase III trials in infants and young children started in May 2009 in sub-Saharan Africa.

Phase II studies have shown consistent efficacy for the vaccine, and the Committee evaluated its safety in anticipation of licensure for the prevention of P. falciparum malaria in young children in regions where it is endemic. Safety parameters included reactogenicity observed during 7 days following vaccination and unsolicited symptoms recorded during 30 days after inoculation with doses 1, 2 and 3 in delivery schedules administered at 0 months, 1 month and 2 months and for schedules administered at 0 months, 1 month and 7 months. Comparator vaccines given for control were cell-culture rabies vaccine and pentavalent diphtheria–tetanus–whole-cell pertussis plus hepatitis B vaccine plus Haemophilus influenzae type b vaccine. A total of 1147 study participants (including 340 in the age range recommended by EPI) who received 2 or 3 doses of RTS,S/AS01 were evaluated. In general, injection-induced local inflammatory signs and systemic signs and symptoms occurring during the 7 days and 30 days following vaccination were similar between study and control vaccines. Of those receiving study vaccines, 2 children had convulsions, which were assessed as febrile. One sign – skin rash – appeared in excess (statistically significant) during the 7-day interval in study children, and further detailed description of the nature of the rash is necessary to assess its clinical significance. This observation is particularly important in view of the earlier observations of skin lesions (including urticaria and immunoglobulin E response characteristic of immediate-type hypersensitivity, which may lead to anaphylaxis) and delayed-type hypersensitivity detected by skin testing in a few participants who received other experimental malaria vaccines.3 These other malaria vaccines used different antigens (for example, linear circumsporozoite antigen in doses of 200–2000 µg) and different adjuvants (that is, not AS01 or AS02).

The Committee concluded that the RTS,S/AS01 vaccine has an acceptable safety profile, although data are available only on a relatively small number of children. During the course of presentations, the safety profile of the AS01 adjuvant was reviewed; this adjuvant is delivered with a number of experimental vaccines, mostly in adult volunteers during phase I studies. This review gave additional assurance of the safety of the vaccine under consideration.

Age of administration of rotavirus vaccines

In April 2009, WHO’s Strategic Advisory Group of Experts (SAGE) reviewed recent data on the effectiveness and safety of 2 rotavirus vaccines the pentavalent RV5 (RotaTeq,® Merck & Co., Whitehouse Station, NJ, USA) and the monovalent RV1 (Rotarix,® GlaxoSmithKline Biologicals, Rixensart, Belgium). Based on its review of the evidence, SAGE recommended that vaccination against rotavirus be included for infants in all national immunization programmes. Introduction of the vaccine is strongly recommended in countries where deaths from diarrhoeal diseases account for ≥10% of mortality among children aged <5 years.4

In pivotal trials with RotaTeq, children were vaccinated with dose 1 between 6 weeks and 12 weeks of age; with Rotarix, children were vaccinated between 6 weeks and 14 weeks of age (in Latin America) or between 6 weeks and 15 weeks of age (in Europe). RotaTeq is delivered in a 3-dose schedule; Rotarix requires only 2 doses. The maximum age recommended for the last dose of RotaTeq is 32 weeks; it is 24 weeks and 6 days for Rotarix. In the United States, where RotaTeq has been generally administered at ages corresponding to those studied in the pivotal trials, post-licensure safety data do not suggest an increased risk of intussusception.

Based on these data, and in order to harmonize schedules between the vaccines, SAGE recommended that the first dose of either Rotarix or RotaTeq be administered at between 6 weeks to 15 weeks of age, and the maximum age recommended for administering the last dose of either vaccine should be 32 weeks. SAGE 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%.

The Committee supported SAGE’s recommendation to expand the age eligibility for rotavirus vaccination. Administration of rotavirus vaccines at the recommended ages early in infancy would provide high protection from rotavirus diarrhoea among vaccinated children. However, in many developing countries, not all children receive immunizations according to recommended dosing schedules, and the challenges of delivering timely immunization may be most acute in countries with high rates of early childhood mortality. The data on Rotarix and RotaTeq support the safety and effectiveness of both new rotavirus vaccines. The level of risk of intussusception identified with the previous vaccine (RotaShield,®) can be ruled out with confidence, and there may be no increased risk since neither Rotarix nor RotaTeq has given a signal of increased risk. It has been hypothesized that giving rotavirus vaccines outside of the recommended ages may be associated with an increased risk of intussusception. No adequate data directly support this hypothesis, and the Committee concluded that even if there were a theoretical increase in risk, the benefits of vaccination would exceed any possible risk of intussusception. Therefore, the Committee recommended expanding the age eligibility for vaccination beyond the maximum age recommended by SAGE in order to maximize coverage of the vaccines, especially in countries with high rates of early childhood mortality caused by diarrhoeal illness.

Although data support the safety and effectiveness of both Rotarix and RotaTeq, ongoing evaluation is needed to provide additional data for the expanded age groups. National immunization programmes that elect to extend age eligibility for the first dose of rotavirus vaccine to infants aged >15 weeks or the last dose beyond 32 weeks of age should be encouraged to initiate monitoring for effectiveness and safety. Rates of intussusception in unvaccinated children are higher among older infants, and even without an increased risk associated with vaccination, cases of intussusception in temporal association with vaccination are more likely to occur in infants vaccinated when aged >15 weeks than in younger infants. A systematic approach to monitoring intussusception will be essential in order to interpret the significance of these cases and reported rates. Effective risk-communication strategies will also be essential.

Safety of human papillomavirus vaccines

The Committee reviewed the safety of HPV vaccines. By March 2009, >60 million doses of the quadrivalent or bivalent HPV vaccine had been distributed either as part of national immunization programmes in 21 countries or by private physicians. Data from post-marketing surveillance were reviewed from countries that introduced the vaccine early, from regulatory authorities and from manufacturers. Additional data were reviewed from 4 demonstration studies undertaken in developing countries that were conducted by the nongovernmental organization PATH (the Program for Appropriate Technology in Health) and from recently completed and ongoing studies conducted by manufacturers on the vaccination of young males and the concomitant use of HPV vaccine with other vaccines in young girls.

The accumulating evidence on the safety of HPV vaccines is reassuring. The most common adverse events were reactions at the injection site and muscle pain. Allergic reactions have also been reported. The potential risk of injury after vaccination resulting from dizziness and syncope has been added to the label of 1 of the vaccines. Several different signals were observed in countries introducing HPV vaccination but none, other than syncope, was judged to be causally related to vaccination. The limited data on the inadvertent administration of HPV vaccines shortly before pregnancy or during pregnancy are reassuring. They do not establish a relationship between HPV vaccination and miscarriage, but the data are insufficient to rule out a small effect, in particular if conception occurs shortly after vaccination. The Committee considers that further studies should be encouraged, given the limited data.

The Committee noted with satisfaction that studies on HPV immunization have been initiated in Africa, including some among HIV-positive women. As preparation for introducing HPV vaccines, capacity building for surveillance for adverse events is being addressed. While the safety profile of HPV vaccine is reassuring, the collection of high-quality safety data from different geographical locations and epidemiological settings where the vaccine is being introduced remain a high priority.

Assessing the safety of novel influenza vaccines

The Committee reviewed 2 potential safety issues related to pandemic influenza vaccines: the previous experience with Guillain–Barré syndrome following administration of swine flu vaccines and the adjuvants used in influenza vaccines. The swine influenza vaccines used in the United States in 1976 were associated with a small but significant risk of Guillain–Barré syndrome developing during the 8 weeks after immunization. The attributable risk among those who were vaccinated was approximately 1 case/100 000 people vaccinated. The underlying reasons for the association are unknown. Studies of other influenza vaccines since 1976 have shown either no association with Guillain–Barré syndrome or, in some studies, a very small risk (for example, an attributable risk of approximately 1 case/1 000 000 vaccinations). Because the new H1N1 influenza virus is derived in part from a swine lineage, the Committee discussed the importance of preparing for active surveillance of cases of Guillain–Barré syndrome in individuals vaccinated with the novel H1N1 vaccines. Surveillance should include developing common protocols, case definitions and assessments of suspected cases to learn more about Guillain–Barré syndrome. It will be important to obtain baseline rates of the syndrome occurring among unvaccinated populations, particularly in developing countries, and to be prepared to assess whether there is an association between Guillain–Barré syndrome and vaccination with novel H1N1 vaccines, as well as between Guillain–Barré syndrome and influenza illness caused by the novel H1N1 virus. Health authorities in countries capable of conducting active surveillance for the syndrome should collaborate and communicate to develop common approaches, and results should be shared with WHO so that other countries using similar vaccines may benefit from the information. It may also be possible to expand surveillance for acute flaccid paralysis to all ages in an effort to capture cases of Guillain–Barré syndrome in some low-income and middle-income countries. However, the Committee emphasized that such surveillance should be limited to settings where assessment would not interfere with ongoing polio eradication activities. There was also discussion about which additional laboratory assessments might be conducted as part of clinical trials of novel H1N1 vaccines to potentially elucidate mechanisms of aberrant immune responses that may predispose people to Guillain–Barré syndrome.

There was discussion about adjuvants for influenza vaccines, including the limited safety data for certain populations, such as children aged <3 years and pregnant women; the limited information on interactions between preservatives and oil-in-water adjuvants; and the need to develop post-marketing surveillance requirements for the vaccines. There may be significant challenges to addressing certain potential concerns, such as adverse autoimmune events, if there is a long latency period between vaccination and events. Another issue is whether there might be an increased risk of febrile convulsions occurring in young children, given the increased reactogenicity of adjuvanted vaccines, which highlights the need for monitoring. At the moment, there are no data on the safety or immunogenicity on the interchangeability of adjuvanted and non-adjuvanted influenza vaccine used for either the first or the second dose.

It is anticipated that the range of H1N1 vaccines will be heterogeneous in formulation, antigen presentation and the substrates used for manufacturing, and these vaccines may have significantly different safety profiles. In view of what is likely to be their widespread use in all age groups and the required 2-dose schedule, there may also be differences from the well established safety profiles of licensed seasonal influenza vaccines. The choice of which post-marketing surveillance strategy should be applied to H1N1 vaccines will depend on how much time is available to institute modified monitoring systems for adverse events following immunization and the capacity of countries to conduct active surveillance for selected, potentially serious adverse events in addition to Guillain–Barré syndrome. Baseline rates should also be obtained for other conditions that are anticipated to occur coincidentally in populations targeted for vaccination.

In view of the use of H1N1 vaccines, countries should carefully assess the capacity of their systems to monitor vaccine effectiveness and safety, and to enhance rapid detection of potential signals. It is of paramount importance to develop robust and efficient mechanisms for notifying cases of adverse events following immunization that are linked with global networks for analysis and risk communication. Strengthening the monitoring of seasonal influenza vaccines offers an opportunity to test the functionality of the systems. Timely reporting and analysis of adverse events following immunization are of utmost importance. Auxiliary monitoring systems may be required when H1N1 vaccines are used in subpopulations who may be at enhanced risk; these systems may include perinatal registries that capture pregnancy outcomes in immunized pregnant women and monitoring of responses of HIV-positive people who are vaccinated.

Safety of measles vaccine in children infected with HIV

The Committee commissioned a systematic review and meta-analysis to identify and synthesize evidence about the safety, immunogenicity and efficacy of measles vaccination in children infected with HIV. A total of 8 electronic databases were searched for studies published until February 2009 relating to measles vaccination in HIV-positive children. Altogether, 723 articles were identified, of which 25 studies with comparison groups (involving 4519 vaccinated children) and 1 case-report were eligible for inclusion. Another 13 studies without comparison groups (involving 690 vaccinated children) were also examined for data on adverse events.

Adverse events were not mentioned in 20 of 39 studies. In the 19 studies that described adverse events, 17 reported no serious or severe adverse events. In 2 prospective studies that reported on adverse events and allowed comparative analysis, there was no increased risk of vaccine-related serious adverse events in HIV-positive children when compared with HIV-exposed but uninfected children or children not exposed to HIV. A total of 8 hospitalizations were reported after vaccination in 457 HIV-positive children enrolled in prospective studies, excluding those events explicitly stated by the authors to be unrelated to vaccination. There were 55 deaths in 387 HIV-positive children who had been vaccinated. Serological assessments of measles antibody titres after vaccination showed that measles vaccination at the age of 6 months resulted in similar levels of antibody in HIV-positive children and children who had not been exposed to HIV; by the age of 9 months, fewer HIV-positive children (with severity of disease ranging from no clinical signs of AIDS to groups where 71% were symptomatic) responded to measles vaccine than did children who had not been exposed to HIV. After measles vaccination at age 6 months, children who had initially tested HIV-positive owing to maternal antibodies but were subsequently found to be uninfected, were slightly more likely to have developed antibodies than children who had not been exposed to HIV. Two studies suggested that the antibody response in HIV-positive children waned faster than it did in children who were not infected with HIV. There were scant data about the effects of highly active antiretroviral treatment (HAART) on responses to measles vaccination, and the possibility of comparing vaccinated children to unvaccinated HIV-positive children was limited. Data relating to clinical efficacy against measles were also scarce.

Based on these findings the Committee drew the following conclusions.

  • The evidence does not demonstrate a serious risk in using measles vaccine in HIV-positive children. Although millions of doses of measles vaccine have been administered to HIV-positive children, only 1 case report was identified that suggested possible severe adverse events following immunization. However, ascertainment of such events may be incomplete.
  • The literature review documented higher mortality among HIV-positive children who received measles vaccine than among children who were not infected with HIV and who received measles vaccine. However, it seems plausible that most or all of this effect is due to HIV infection alone rather than to measles vaccination. There are many confounding factors that could explain a higher rate of death or severe adverse events following measles vaccination in this population. One possible approach to resolving this issue would be to recommend systematic follow up of children vaccinated against measles in populations with a high prevalence of HIV and to conduct case–control studies of all cases with severe adverse events following measles vaccination in order to assess the possible part played by HIV infection.
  • Measles vaccine appears to be immunogenic in the majority of HIV-positive children. Important areas for further research include determining the duration of immunity and protection, and whether there is a benefit to administering a second dose of measles vaccine to HIV-positive children.
  • On the basis of the literature reviewed, the Committee considers that there is no need to modify WHO’s recommendation on measles vaccination in HIV-positive children.
  • The recommendation on the use of measles vaccines indicates that it is contraindicated in people who are severely immunocompromised. This reflects the risk–benefit ratio, since children with low CD4 cell counts might derive little benefit from the vaccine.
  • Studies should be conducted to investigate remaining concerns. These include studies to determine the etiology of pneumonia in HIV-positive infants (no studies have examined for measles virus as the etiological agent of pneumonia in HIV-positive children), to compare morbidity and mortality among HIV-positive children with and without measles vaccination, to determine the immunogenicity of measles vaccine in HIV-positive children on HAART, and to evaluate the duration of immunity and whether there is an added benefit in administering a second dose of measles vaccine to HIV-positive children.
  • See No. 41, 1999, pp. 337–338.
  • The committee invited additional experts to present evidence on the safety of influenza, rotavirus, measles, malaria and human papillomavirus vaccines. Depending on the session, these experts were affiliated with the United States Centers for Disease Control and Prevention, Atlanta, GA, USA; the Johns Hopkins University and the University of Maryland, Baltimore, MD, USA; Bern University, Berne, Switzerland; the European Medicines Agency, London, United Kingdom; Merck Research Laboratories, West Point, PA, USA; and GlaxoSmithKline Biologicals, Rixensart, Belgium.
  • Edelman R et al. Immediate-type hypersensitivity and other clinical reactions in volunteers immunized with a synthetic multi-antigen peptide vaccine (PfCS-MAP1NYU) to Plasmodium falciparum sporozoites. Vaccine, 2002, 21:269–280.
  • See, No. 84, 2009, pp. 232–236.
Share