Knowledge Gaps in the Pursuit of a Universal Influenza Vaccine

Seasonal influenza outbreaks cause between 250,000 to 500,000 deaths word-wide each year (according to 2008 WHO estimates).  What’s more, unpredictable pandemics, of which there were four in the 20^th and the 21^st centuries (1918, 1957, 1968, and 2009) pose a still greater threat. The worst of these pandemics, the 1918 Spanish flu outbreak, claimed an estimated 50 to 100 million lives globally (according to 2014 WHO estimates).

The human population was “only” 1.9 billion individuals during the 1918 pandemic, whereas there now are about 8 billion people inhabiting our planet. Thus, a future pandemic might be far more catastrophic than the 1918 episode. Moreover, since pandemic strains are derived in part from zoonotic influenza viruses, the constant rise in livestock numbers, intensive farming, and the numbers of animals being transported around the world, combine to facilitate the genetic mixing and evolution of influenza viruses, and the chance of an animal influenza virus becoming able to jump to humans and causing a pandemic.

Unlike vaccines against other viruses (e.g., measles), the seasonal flu vaccine needs to be updated each year to keep up with the antigenic changes that continually occur in influenza viruses. This creates several problems, the first of which is that individuals need to be re-vaccinated each year. And, since it takes months to produce a vaccine, when the updated vaccine is at last ready, it may not be a particularly good match against the new season’s strains. The current vaccine is only about 30 percent effective, which accounts at least in part for the unusually severe flu season we are currently experiencing. And while the efficacy of the current vaccine may be atypically low, even in good years the match is less than optimal.

Pandemics present a much greater challenge to vaccine makers, since pandemics may be vastly more severe than seasonal outbreaks, and since an entirely new vaccine is needed against pandemic viruses. That latter is so because, as noted, pandemic strains are derived in part from zoonotic influenza viruses, which by-and-large are antigenically distinct from strains already circulating in the human population, that humans already express immunity against. Consider the example of the 2009 pandemic. Because an entirely new vaccine was needed to meet the threat of the pandemic virus, the vaccine was not available until after the first wave of infection had already occurred. Fortunately, the 2009 pandemic virus was relatively mild.

A “universal” flu vaccine, that could provide lifelong protection against all seasonal strains of influenza, as well as provide protection against a pandemic virus, would be a most crucial and significant breakthrough. A major international workshop, entitled “Pathway to a Universal Influenza Vaccine,” was convened June 28 and 29, 2017, by the U.S. National Institute of Allergy and Infectious Diseases, to identify gaps in our knowledge that need to be addressed to develop such a vaccine (1).

Participants noted shortcomings in our understanding of the epidemiology, transmission, natural history, and pathogenesis of influenza. Among the issues specifically mentioned: “influenza surveillance is lacking in certain regions of the developing world and globally in certain high-risk groups… gaps in knowledge include the relationship between symptoms, viral shedding and transmission, as well as the level of protection needed to interrupt transmission.”

The host factors that influence influenza disease severity were also acknowledged to be poorly understood. To that point, participants addressed the need to better understand how pre-existing immunity—which might result from multiple natural influenza infections, as well as from repeated vaccinations—might affect how that person’s immune response will respond to future infections and, importantly, how past exposures might affect the efficacy of a vaccine. As stated in the meeting report: “Recent data provide strong epidemiologic evidence that infection with the influenza strain circulating during one’s childhood elicits a lifelong immunologic imprint that impacts responses to novel strains and can help protect against unfamiliar HA subtypes from the same phylogenetic group as the original infecting virus…The potential consequences of imprinting infants with vaccines versus natural exposure need to be carefully assessed.” [The HA protein is the so-called hemagglutinin, which serves the virus as its attachment protein.]

Participants also noted gaps in our understanding of the underlying B and T cell immune mechanisms that are induced by both natural infections and vaccinations. Further study of these responses was recommended so that we might be better able to stimulate them.

As might be expected, it is singularly important to identify the antigens that might be the most promising targets of a universal vaccine. To that point: “…most areas of the HA head are subject to antigenic change, and therefore unlikely to yield a broadly protective immune response. Neutralizing antibody responses to conserved regions such as the HA stalk, and non-neutralizing antibodies such as those directed at the neuraminidase (NA), and matrix 2 ectodomain, merit further study…The importance of each site may differ for pandemic versus seasonal influenza.” [The HA head region binds to receptors on the cell surface. After the bound virus is taken into the cell by receptor-mediated endocytosis, the low pH within endosomes triggers a conformational rearrangement of the HA stalk region, which mediates fusion of the viral envelope with the endosomal membrane, thereby releasing the viral cores into the cytosol. Since mutations within the stalk region might disrupt its membrane fusion function, such mutations are generally selected against. Researchers have recently had success developing antibodies that target the neuraminidase protein at the viral surface.]

Participants acknowledged that animal models play an important role in influenza research, especially when studying pandemic viruses. Nonetheless, “Animal models have limitations including the inability to mimic the human experience regarding genetic background, lifetime exposure to natural influenza infection or vaccine, viral susceptibility…and determinants of immune response and protection.” Thus, the participants noted that “a human a challenge model will be a crucial tool for vaccine development, as it can help answer fundamental questions about influenza immunity and serve as a mechanism for rapidly testing the efficacy of new products…Expansion of this resource should be a top priority…”

Day two of the workshop consisted of a rapporteur session on key conclusions, chaired by David Baltimore and Anthony S. Fauci (1)

Participants agreed that a robust collaboration between government agencies, academia, and industry would be needed to translate the fruits of basic research into a universal influenza vaccine. To that point, a January 24, 2018 editorial in Nature (doi: 10.1038/d41586-018-01070-w) asserted: “…advocates rightly argue that the research and development of a universal flu vaccine — ultimately the only effective defense against future pandemics — merits a program equivalent in scale to the Manhattan Project.” Yet the US government last year invested just $75 million on universal flu vaccine research and development.

1. 1. Paules C.I., D. Marston, R. W. Eisinger, D. Baltimore, and A. S.Fauci. (2017), Meeting Report: The Pathway to a Universal Influenza Vaccine. Immunity 47: 599-603. https://doi.org/10.1016/j.immuni.2017.09.007