Global Vaccine Safety

Global Advisory Committee on Vaccine Safety, 3-4 December 2003

The Global Advisory Committee on Vaccine Safety (GACVS) was established in 1999 by WHO to respond promptly, efficiently, independently (of WHO), and with scientific rigour to vaccine safety issues of potential global importance.1 The Committee acts as a scientific and clinical advisory body to WHO.

GACVS held its ninth meeting in Geneva, Switzerland, on 3-4 December 2003 and considered the following safety concerns regarding immunization.2

Intranasal influenza virus vaccine

The Committee considered safety issues pertaining to a new generation influenza vaccine, administered by the intranasal route. It is a live vaccine, and there are concerns about viral shedding and thus of the potential risk of person-to-person transmission. In the United States, the press has reported that more than 5 million doses have been manufactured, but probably no more than 400 000 had been distributed by the end of November 2003. Comprehensive surveillance has been put in place for monitoring safety. GACVS will review in detail the post-licensure safety data of the vaccine at its June 2004 meeting.

Influenza vaccines and neurological complications

GACVS considered the October conclusions of the United States Institute of Medicine (IOM) Immunization Safety Review Committee, which reviewed four neurological -complications reported in association with inactivated influenza vaccines: Guillain-Barré syndrome (GBS); multiple sclerosis; optic neuritis; and demyelinating neurological disease in children aged 6-23 months. Besides GBS, which was found to be causally linked with the 1976 Swine flu vaccine, IOM found the available evidence to be inadequate to accept or reject a causal relationship between influenza vaccine and any of these neurological syndromes. The United States health authorities are maintaining vigilance for any possible association between neurological disease and influenza vaccine. They hope to conduct a study examining whether the 1976 Swine flu vaccine may have been inadvertently contaminated by Campylobacter jejuni, an infection endemic in chicken flocks, now shown to be a cause of GBS. Review of the suggested -association, and of the -purported mechanism(s), which might include immune-mediated disease, with or without a genetic basis should continue.

Influenza vaccination of women during pregnancy

GACVS discussed recommendations for the use of inactivated influenza vaccine in women during pregnancy and particularly during the first trimester. It was pointed out that manufacturers, as well as national regulatory authorities, tend to caution against routine use of influenza -vaccine during pregnancy. Despite the paucity of data -related to the use of influenza vaccines during the first -trimester of pregnancy, other inactivated vaccines (e.g. -tetanus) have proved safe in this context. There is concern that influenza during pregnancy carries a significantly higher risk of morbidity, hospitalization, and even of fatal outcome, comparable to that in persons aged 65 years and over. The risk of maternal influenza to the fetus is the same throughout pregnancy.

GACVS concluded that risk–benefit of influenza vaccination during pregnancy, at all stages of pregnancy, should be reconsidered, given the high risk to the mother – and thus to the fetus – of the disease itself, and (as far as is known) the small potential risk to mother and -fetus of the inactivated influenza vaccine. WHO was -advised of the view of the Committee accordingly. Such -advice would not apply to situations where the risk of influenza is low or to live attenuated vaccines, which in any event would not be indicated in pregnancy.

Nonspecific effects of childhood immunization

GACVS had received a series of published and unpublished epidemiological studies dealing with the nonspecific and, in some cases, potentially adverse effects of diphtheria–tetanus–pertussis (DTP), measles and bacille Calmette-Guérin (BCG) vaccines on morbidity and mortality in children in the developing world. The issue has been the subject of previous deliberations and review by GACVS.3 A small team, the majority of whose members will come from outside GACVS to ensure independence from earlier deliberations, will be asked to review the latest information, concentrating initially on the potential adverse and nonspecific effects of DTP in childhood immunization. The review will be convened early in 2004, to report back to GACVS at its June 2004 meeting. This review will be included in a broader and more general research, clinical and theoretical assessment to define whether perturbation of the immune system by infant immunization may occur and whether this might be deleterious in certain circumstances.

Safety issues in the poliomyelitis eradication campaign

GACVS considered safety issues pertaining to the poliomyelitis eradication programme worldwide. Information was presented detailing the efforts of the programme in 6 endemic countries, 3 of which contain 5 hyperendemic states. The Committee noted that there have recently been 7 cases of polio infection with a type 2 polio virus (MEF-1 strain), probably as a result of contamination of a batch of polio vaccine after completion of the manufacturing and filling processes (which were not found to be at fault).4

GACVS was informed of the programme’s decision to stop oral polio vaccine use after certification of eradication in light of the adverse effects associated with its long-term use. It acknowledged that there are four critical elements of work for the period following the global interruption of polio transmission: finalizing the strategy for discontinuing oral polio vaccine after certification; providing country-level guidance on decisions regarding future use of inactivated polio vaccine; ensuring the necessary laboratory capacity for continued surveillance; and “mainstreaming” (integrating into routine services) the highly experienced and competent polio eradication infrastructure and personnel that have been developed for the programme.

International reference laboratory for mumps virus isolates

Parotitis and aseptic meningitis have been described as complications of mumps vaccination with Urabe, Leningrad-Zagreb and Leningrad-3 strains of the vaccine but not with the Jeryl-Lynn strains.4 Now that all mumps virus strains can be characterized by polymerase chain reaction and nucleotide sequencing, it should be possible to address scientifically a number of unresolved questions regarding vaccine safety. These include defining the molecular determinants of virus attenuation, precise genetic determinants of virulence, safety of vaccines in relation to either pure or mixed virus populations in the vaccine and their antigenicity, determining the stage at which point mutations occur in the virus (in passage, during replication in the body, or both), and the presence of sub-variant viruses in different vaccines. A hypothesis that single point mutations at certain positions of the haemagglutinin-neuraminidase protein gene region of the genome might be associated with post-vaccination aseptic meningitis should be further investigated. Such knowledge would improve current understanding of genetic and molecular characteristics of strains used for production of mumps vaccines. Moreover, further investigation may help to define the safety and immunogenicity profile of mumps vaccines. If molecular assays could distinguish wild-type from vaccine strains of the mumps virus, it would improve the quality control of existing, and future, vaccines. GACVS has recommended to WHO that an international reference laboratory for mumps virus isolates from vaccinees should be established.

Yellow fever vaccine safety

Cases of viscerotropic disease following yellow fever vaccination were described in the United States and other countries, and current evidence suggests that these cases were a result of vaccine-type virus and not of reversion of the vaccine virus to wild type. Unlike the viscerotropic complications of yellow fever vaccination, reported neurotropic cases from the United States have not been fatal. The neurotropic cases presented as GBS (thought to be immune- mediated), encephalopathy (thought to be due to active invasion of the central nervous system by the virus) or acute demyelinating encephalomyelitis (thought to be caused either by direct invasion of the brain by the virus or by an immune-mediated response to the virus). All cases of viscerotropic disease to date have been reported in primary vaccinees, starting 2–5 days after vaccination. Whether this reflects the persistence of pre-existing immunity at the time of revaccination or lower rates of revaccination in the study populations cannot be determined. The rate of reported viscerotropic disease varies in different settings. In particular, a lower rate has been reported in Brazil than in the United States. GACVS considered that it would be important to examine the possibility that naturally circulating yellow fever virus in endemic regions may account for some of these differences in risk.

The risk of viscerotropic complications of yellow fever vaccine is age-related; individuals aged 65 years and over who are first recipients of the vaccine are at considerably higher risk than younger subjects (although the young are not excluded from the risk). Recent evidence from the United States also suggests a higher risk of neurotropic disease in older vaccinees. There needs to be a better estimate of the risk, and prediction of subjects at risk and of factors predisposing to risk of both types of serious complication of the yellow fever vaccine. In communicating such risks, consideration needs to be given to highlighting potential contributing factors, such as age or sensitivity of surveillance systems, to risk estimates based on currently available data. Particular care should be taken that the vaccine is received only by those travellers truly at risk for yellow fever exposure. Furthermore, care should be taken that routine yellow fever vaccination programmes are not jeopardized by risk–benefit ratios that may be inapplicable to the target populations in endemic countries.

A critical and unresolved issue is the safety and efficacy of yellow fever vaccine in human subjects infected with immunodeficiency virus (HIV). It remains to be determined whether HIV-positive status materially affects seroconversion, the risk of invasion of the nervous system and of encephalopathy, the stage of HIV disease at which yellow fever vaccination should be contraindicated, and whether there are differences in the incidence of minor and major adverse effects in HIV-positive subjects.

BCG immunization in HIV-positive subjects

GACVS reviewed the available data on the benefits and risks of BCG immunization in infants living in areas where there is a high prevalence of tuberculosis, with and without a concurrent high prevalence of HIV infection. There are few population-based data on the effectiveness, or otherwise, of BCG vaccine in preventing severe tuberculosis in HIV-positive infants. Given the high prevalence of HIV and tuberculosis in certain countries and of the current development of new tuberculosis vaccines, some of which are based on BCG, GACVS advises no change in the current -recommendations for BCG immunization of infants in countries with a high prevalence of tuberculosis and that population-based studies should be undertaken to determine the efficacy and safety of BCG and related vaccines in HIV-negative and HIV-positive children in countries with a high endemic rate of tuberculosis. GACVS supports the -recent WHO initiative to further characterize BCG strains, which will help systematic differentiation of BCG vaccines in terms of safety and efficacy.

Safety of smallpox vaccination

GACVS has previously considered the safety of smallpox vaccination.4 The Committee was provided with an up-dated account of the safety of smallpox vaccination, based on 38 759 persons vaccinated in the United States since January 2003, covering 65% of health care workers and at least one health care worker in 45% of hospitals. There is, in addition, a pregnancy registry of 160 women exposed to smallpox vaccine during or immediately before pregnancy, identifiable by testing. Consistent adverse effects reported in smallpox vaccinees have been myopericarditis and dilated cardiomyopathy; the frequency of each adverse effect exceeds what might have occurred by coincidence.

Switch from rubella vaccine to mumps, measles and rubella vaccine

The attention of the Committee was drawn to the unavailability of a monovalent rubella vaccine in some countries and to the need to provide a rubella-containing combination vaccine to postpartum women seronegative for rubella. GACVS is not aware of any safety issues that would restrict the provision of a rubella-containing combination vaccine in place of single rubella vaccine in those circumstances.

Aluminium-containing vaccines and macrophagic myofasciitis

GACVS reviewed the data of the case-control study performed in France and concluded, in accordance with previous statements,5 that the persistence of aluminium-containing macrophages at the site of a previous vaccination is not associated with specific clinical symptoms or disease.

Conclusions on other vaccine safety issues that are regularly followed by GACVS, and particularly on hepatitis B vaccine and multiple sclerosis and hepatitis B vaccine and leukaemia, can be found on the GACVS web site developed by the WHO secretariat for open access by interested parties (

1 See No. 41, 1999, pp. 337–340.
2 GACVS invited additional experts to present evidence and participate in the discussions on the safety of influenza vaccination during pregnancy as well as on the safety of mumps, BCG, yellow fever and smallpox vaccines.
3 See No. 47, 2002, pp. 389–394 and No. 32, 2003, pp. 282–284.
4 See No. 32, 2003, pp 282–284.
5See No. 41, 1999, pp. 338-341 et No 32, 2003, pp. 282-284.