The future of influenza vaccinesBMJ 2009; 339 doi: http://dx.doi.org/10.1136/bmj.b4014 (Published 06 October 2009) Cite this as: BMJ 2009;339:b4014
- Menno D de Jong, professor of clinical virology1,
- Rogier W Sanders, assistant professor of microbiology and immunology2
- 1Department of Medical Microbiology, Centre for Infection and Immunity Amsterdam (CINIMA), Academic Medical Centre of the University of Amsterdam, 1105 AZ Amsterdam, Netherlands
- 2Department of Microbiology and Immunology, Weill Medical College of Cornell University, 1300 York Avenue, New York, NY 10065, USA
The linked case-control study by Garcia-Garcia and colleagues (doi:10.1136/bmj.b3928) shows that, contrary to earlier speculation, some level of cross protection against the new influenza A/H1N1 virus may be provided by the 2008-9 seasonal flu vaccine.1 In a group of 60 confirmed infected and 180 uninfected controls, uninfected people were significantly more likely to have received the seasonal flu vaccine (29% v 13%). Turning the data around, of the 179 unvaccinated people in the study, 52 (29%) became infected with pandemic H1N1 whereas eight of 61 (13%) vaccinated people became infected. Furthermore, all of these eight people survived, whereas 18 of the 52 unvaccinated people who were infected died. Despite the study’s limitations, the data suggest that vaccination with the seasonal flu vaccine confers some protection against pandemic H1N1 virus infection and severe disease from this infection.
Although these observations agree with recently reported low rates of seroconversion to new H1N1 specific antibodies after seasonal vaccination in adults,2 the authors emphasise that their observations do not mitigate the need for pandemic H1N1 vaccines.
Despite the relative success of traditional flu vaccines over the past 60 years, innovation is needed to ensure adequate protection against seasonal and pandemic flu. The current H1N1 pandemic illustrates the shortcomings in our capacity to design and produce vaccines on the scale needed to protect most of the world’s population. Despite massive efforts to produce such vaccines, some models predict that it may be too late, even for countries that procure enough vaccine. Predictions based on the 1957 pandemic indicate the epidemic may peak in late October.3 Vaccines will probably not become available before late October and it takes a few weeks to elicit a robust immune response, so vaccinated people may be protected only after the peak of the pandemic has passed.
To protect against seasonal and pandemic strains, vaccines and vaccine production need to improve. New vaccines take at least five to six months to produce once a new virus is identified. Moreover, although production has increased recently, it is probably not high enough to curtail a pandemic of highly pathogenic influenza variants with low immunogenicity, such as H5N1 and H7N1,4 for which the vaccine dose or number of doses needs to be increased. In this respect we may be lucky with the current H1N1 pandemic, because preliminary studies suggest that single low vaccine doses may induce a protective immune response.5 6 Furthermore, the efficacy of current vaccines is relatively low in elderly people, and cross protection against “drifted” strains is often limited. Several recent advances should help to rejuvenate the development of the flu vaccine.
Most licensed flu vaccines consist of inactivated egg grown virus that is purified and chemically disrupted. Live attenuated virus vaccines delivered via nasal spray are also available. The current seasonal flu vaccines consist of three components: influenza A/H1N1, influenza A/H3N2, and an influenza B strain. The specific strains require periodic updating according to semiannual recommendations from the World Health Organization because of continuous evolution of the haemagglutinin (HA) protein (“antigenic drift”). The introduction of potential pandemic strains (“antigenic shift”) is much less predictable, as shown by the current H1N1 pandemic of swine origin, when pandemic viruses from avian origin, particularly H5N1, were long anticipated.
Most flu vaccines are still produced in embryonated hens’ eggs, but this has several problems. Firstly, not all influenza strains infect chicken eggs efficiently, and this can result in delayed vaccine production or lower yields. Secondly, virus propagation in eggs may select for strains that differ antigenically from the circulating strain. Thirdly, these vaccines are not suitable for people with egg allergy. Fourthly, production of virus strains with pathogenic effects on eggs, such as highly pathogenic avian viruses, is difficult. Finally, egg supplies may be limited. As alternatives, cell based production platforms are being developed, which minimise these problems and can be scaled up.
HA is the most important antigen, and anti-HA antibody titres correlate with protection. Therefore, using subunits of HA to make vaccines is promising. In principle, the production of recombinant HA is relatively fast and easy, and propagation in eggs is not necessary. HA protein subunit vaccines have proved to be effective in phase III trials.7 Early clinical trials with HA encoding DNA vaccines have shown about 50% efficacy, which is promising, but not yet sufficient to compete with conventional flu vaccines or HA protein vaccines.8
The addition of novel adjuvants can improve the immunogenicity of flu vaccines,9 10 so that a smaller dose can be used and more people can be vaccinated. Antibodies generated in this way may also be more protective against drift variants, which is important as pandemic viruses evolve.
Antibodies that recognise multiple influenza strains are rare, which explains the frequent lack of cross protection between vaccines and natural infection. However, antibodies have been identified that may open possibilities of developing a “universal flu vaccine.”11 12
Traditional flu vaccines have a good track record in terms of efficacy and safety, so it will take years to replace them, but the new techniques and vaccines at various stages of testing are both necessary and promising.
Cite this as: BMJ 2009;339:b4014
Competing interests: None declared.
Provenance and peer review: Commissioned; not externally peer reviewed.