Though it is still too early to close the book on the 2017 to 2018 flu season, it has already proven to be historically severe on numerous metrics. For example, the Centers for Disease Control and Prevention (CDC) collects data on outpatient visits for influenza or other similar illness through the Outpatient Influenza-like Illness Surveillance Network, or ILINet. The national baseline percentage of outpatient visits that participate in ILINet is 2.2%, but the real percentage for the 2017 to 2018 season has exceeded that baseline – peaking at 7.7% -- for eleven consecutive weeks since late November. This ILINet activity is the highest level observed since the H1N1 “swine flu” pandemic of 2009, though ILINet only captures part of the disease’s impact. Preliminary studies on the efficacy of this year’s vaccine in the southern hemisphere paint a similarly grim picture. Is it time to rethink the way we develop flu vaccines?

In short, yes. Though some of the factors contributing to the severity of the 2017 to 2018 season are out of our control, such as increased prevalence of the H3N2 strain relative to others, our current mode of vaccine development sets us up for failure when these factors occur. Because we have not yet discovered a universal vaccine for the flu, our efforts currently center around identifying and neutralizing viral surface proteins (hemagglutinins, or HAs) that change over time and are different from strain to strain. The World Health Organization (WHO) collects viral data from around the world for five to eight months before issuing guidance on how best to update vaccines for the coming season.  Armed with this WHO data, vaccine makers have between six to nine months to update, synthesize, and deploy their vaccines for the year. The combined expertise of the WHO and CDC is a formidable weapon, but viral evolution at the molecular level is a worthy opponent – and it hinders both our efforts to characterize existing strains and accurately recreate them in the lab.

H3N2 is a particularly difficult strain for researchers to counter because its HAs evolve quite rapidly in the wild, rendering vaccines created from months-old WHO guidance unlikely to be a match. Yet there is another unique issue that exposes a fundamental flaw in a long-running tradition of the vaccine industry. Its HAs mutate to adapt to its host, and many vaccines are grown in chicken eggs. The inevitable viral mutations that take place in the egg would be of great use to us if we were developing vaccines for chickens, but the changes give no such benefit to humans. In fact, there is data to suggest just the opposite – that mutations stemming from incubation in eggs caused vaccines to be less effective in the 2016 to 2017 season than if they were grown using recombinant, egg-free techniques.

This entire annual research, development, and production pipeline exists because we have not yet found the universal flu vaccine. What if we could? In the wake of a particularly severe flu season, the subject deserves and indeed receives more attention. In this way, a bad year for the flu vaccine can somewhat paradoxically mean a good year for flu vaccine research and development. Many research groups hope to capitalize on increased public awareness and discontent with the existing process to create a single vaccine that neutralizes all flu strains, or provides more extended protection than the current approximate year of immunity. Identifying parts of the flu virus that do not vary, or at least vary less than the rapidly evolving HAs, is key to developing something that can end our arms race against viral evolution.

Note: Do not let this discussion dissuade you from getting vaccinated if you have not yet done so this year! It takes about two weeks for relevant antibody concentrations to increase after vaccination, and this season could last for many more weeks. The flu shot, however imperfect, leads to a measurable reduction in infections, severity of symptoms, and deaths even in bad years. The industry is gradually transitioning away from outdated egg-based production methods toward more modern recombinant techniques that are much more effective.

References:

https://www.cdc.gov/flu/about/season/flu-season-2017-2018.htm

http://www.nejm.org/doi/full/10.1056/NEJMp1714916#

https://www.vox.com/science-and-health/2018/2/1/16960758/flu-vaccine-effectiveness

http://www.sciencemag.org/news/2017/09/why-flu-vaccines-so-often-fail