Last Tuesday I attended a Royal Society discussion meeting titled ‘Pandemic influenza: frontiers in research‘ which I hoped might provide some answers as to where research is headed in the wake of the emergence of a pandemic influenza strain. Organized by Neil Ferguson and John Skehel it was a packed schedule, with the highlights for me being talks on the origins of the H1N1 influenza A strain that has swept the globe, the uncertainties that modellers grapple with when attempting to provide the best possible picture of the dynamics of an emergent epidemic (focusing of course on H1N1 influenza A), research into a universal ‘flu vaccine, surveillance of serological responses over the course of the pandemic in the UK by the Fluwatch project and a final word from Charles Penn discussing the WHO response to the H1N1 influenza A outbreak.
Gavin Smith discussed research published in June of last year in which a team of researchers carried out an evolutionary genomics study that showed that reassortment of swine ‘flu virus genes can result in emergence of a pandemic strain. The team of authors, scattered in Oxford, Edinburgh, Hong Kong and Arizona, collaborated using a wiki to share their data and write a paper. Apparently we don’t have anywhere near enough data on ‘flu strains circulating in pigs, or how ‘flu viruses are transported by the pig trade. One big issue for policymakers and researchers alike is garnering the funding and political will to carry out surveillance of ‘flu strains in domestic pigs — clearly farmers might not be happy about such initiatives and moreover, as pointed out by Nick Phin, Head of pandemic flu planning at the HPA in the UK, would we be able to spot a new strain that had pandemic potential? Policymakers would need to decide whether allocating resources to routine ‘flu surveillance in animals was warranted, according to Dr. Smith, but he speculated that it might be possible (if monitoring in animals and humans) to ‘spot unusual gene flow in humans’. Avian ‘flu on the other hand is transmitted by wild birds and its politically easier to establish surveillance in wild bird flocks, but, Robert Webster, also speaking at the meeting, cautioned that a global network of surveillance was needed and that current levels of surveillance are inadequate either to spot emergent strains, or to provide the level of genomic data needed to accurately determine the origins of new pathogenic strains that emerge in humans.
Neil Ferguson highlighted the differences between what the ‘flu modellers expected from a ‘flu pandemic (high attack rates, and emergence of a H1N1 strain from somewhere in Asia) with what they got (lower attack rates and emergence in Latin America) and pointed out that a lack of information on seroconversion rates in the first wave of the epidemic affected the ability of modellers to accurately predict the severity of the second wave of H1N1 infection. He also pointed out that making predictions is much easier with severe strains than with mild strains of ‘flu because surveillance often fails to spot the majority of mild cases. The biggest obstacles to accurate predictions during the early days of the H1N1 outbreak were limited serosurveillance (essentially only for those seeking care, who were by definition likely to have more severe symptoms), lack of information on H1N1 seasonality, and lack of mechanisms for sharing epidemiological data. Tellingly, he asked ‘what is a reasonable worst case scenario’ and of course its very important to remember that messaging changing outputs, by both scientists, politicians and the media, is a key challenge that need to be addressed. Similar views were echoed by John Edmunds at a CRAASH meeting (Healthy Futures: Medical Regulation and human agency) that I went to in January of this year. Regarding the thorny issue of widespread serological surveillance, Ferguson felt that whilst baseline data on serological responses to different ‘flu subtypes might have been helpful, he was concerned over whether gathering such data was cost-effective.
Are we anywhere near a universal ‘flu vaccine? Peter Palese has been working towards this goal for a long time now and reported on recent work in which his team generated cross-subtype specific antibodies (published in PLoS Pathogens) and a new mouse study investigating the protection offered against death and disease caused by a variety of subtypes by a headless HA vaccine, published in the new open access journal from the American Society for Microbiology called mBIO.
Charles Penn drew the meeting to a close and discussed general themes of the pandemic commenting that the only region currently showing an upwards trend in cases is West Africa so he felt that there was insufficient information to say that ‘we’re over the pandemic’. He also commented that the word ‘pandemic’ is loaded as a pandemic can be severe or mild and that the WHO will soon be publishing revised guidance for the clinical management of H1N1 influenza. This has apparently been derived from clinicians sharing clinical notes, although the mechanism for data-sharing wasn’t clear to me. In response to some strong questions regarding recent data that contended that there was a paucity of evidence supporting the cost effectiveness and safety of oseltamivir (Tamiflu), and questions over whether advice received by the WHO was impartial, Dr. Penn was very clear that WHO actively manage conflicts of interest, whether ‘real or perceived’.