October 26, 2011

expert reaction to meta-analysis of seasonal flu vaccines, as published in The Lancet Infectious Diseases

Scientists analysed forty years of data to determine the effectiveness of current flu vaccines.

Professor Wendy Barclay, Chair in Influenza Virology at University College, London, said:

“It is really important to understand what we are trying to achieve when we vaccinate against influenza, and how we aim to do that. We want to save lives and prevent illness. We currently vaccinate the very people who are likely to fare badly after influenza infection, the elderly and infirm but it is a difficult task because they don’t respond well to vaccines. The paper adequately confirms this problem. Overall the paper comes out quite strongly to say that vaccinating children with a live attenuated influenza vaccine is one of the more reliable paths to follow given the currently available armament. However what has been measured is the chances of the people who have been vaccinated ending up in hospital- what we need to know is whether the elderly people can be saved if we reduce influenza in the community by vaccinating children. This is not what has been measured in the studies analysed here and will require longer term analyses on a larger scale. Vaccinating children against influenza virus with a live attenuated vaccine is not a policy we currently apply here in the UK.”

Professor Peter Openshaw, Director of the Centre for Respiratory Infection at Imperial College, said:

“1. This careful analysis of published work focussed on virologically confirmed influenza (using PCR or viral culture), requiring a very careful (and costly) study design.
2.,PCR is a relatively new detection method, and viral culture fails to detect many cases of infection.
3.,The strict criteria used to select publications means that evidence from many well designed studies was excluded.
4.,No study of standard vaccines in subjects over 65y met the criteria, in part because it is judged unethical to withhold vaccination from those who most benefit from it.
5.,The absence of evidence does not provide evidence of an absence of a beneficial effect, especially in patient groups in which it is hard to do studies or to obtain viral samples (studies of children, for example, are much harder than studies of adults).
6.,It is usually not possible to test new formulations of flu vaccines to ensure that they are effective against infection because of the speed with which each annual formulation has to be made. Therefore, it is standard to use the serum antibody responses of volunteers as an indication of likely benefit.
7.,With newer vaccines (e.g. the live attenuated vaccine, given intranasally), the protection against infection occurs in the absence of a good serum antibody response. This clearly shows that we need new and better methods by which to estimate vaccine efficacy.
8.,The authors make the very valid point that current vaccines have limitations, and that new and more effective vaccines should be developed. This is particularly true of vaccines for older persons, who often don’t respond strongly to a standard doses of vaccine.
9.,We need new vaccines that work regardless of the exact strain of flu that is circulating. These cross-protective vaccines are elusive, but are urgently needed for unpredictable future outbreaks of influenza that may have a much higher lethality than currently circulating strains.
10.,Although focused on vaccines in use in the USA, the review is generally supportive of current UK vaccination policies.”

‘Efficacy and effectiveness of influenza vaccines: a systematic review and meta-analysis’ by Michael Osterholm et al., published in The Lancet Infectious Diseases on Wednesday 26th October 2011.