Breakthroughs don’t happen overnight. Years of research, millions of dollars, and uncountable work hours go into even seemingly overnight successes. For example, the COVID-19 mRNA vaccines went from sequence to injected into arms in a lightning-fast year, but they stood on the shoulders of decades of work. Influenza, on the other hand, has managed to dodge the vaccine strategies that researchers have used to target other viruses, despite having lingered in human lungs for thousands of years. 

Every year, flu vaccines need to be updated since the virus mutates at such a rapid clip that each successive year’s vaccine is rendered mostly ineffective by the following season. Using data from infection patterns in various parts of the world, virologists and epidemiologists make educated estimates of the type of influenza virus to vaccinate against. This is difficult guesswork for the biologists, a logistical challenge for vaccine manufacturers who must produce billions of doses every year, and a game of whack-a-mole for healthcare workers who struggle to get 40% of the US to take the vaccine, blunting its effectiveness even further.

A universal flu vaccine that could be used for years without requiring updates would solve many of these problems. In June, a group of researchers led by Jeffrey Taubenberger of the National Institute of Allergy and Infectious Diseases (NIAID) announced in Science Translational Medicine a successful test of a vaccine in mice and ferrets that protected against a wide variety of influenza subtypes (1).

Flu viruses are broadly sorted into two categories based on the type of hemagglutinin protein (HA or H) and the type of neuraminidase protein (NA or N) they carry. HA proteins dot the outside of the virus on the head of a stalk, like a sunflower, and help the virus attach to target cells. NA destroys the receptor that the virus uses on the way into a cell, which helps newly created virus particles burst out to infect new cells (2). 

Since these proteins are the most visible to the immune system, they are under the most evolutionary pressure to mutate and evade detection. HA and NA are also encoded on different parts of the virus’s segmented genome, allowing each of the 18 different HAs and 11 different NAs to mix and match. On top of that, influenza’s RNA polymerase is famously inaccurate and constantly introduces mutations into new viruses. These factors all contribute to the flu’s constantly shifting identity: H1N1, the 1918 pandemic flu, H5N8, an avian flu that rarely infects humans, and H5N1, another avian flu that frequently infects humans. are a few of the variants commonly in the news. 

For the NIAID vaccine, Taubenberger’s team used specifically chosen subtypes with the hope of generating immune responses against viruses with different H- and N-types. The team created a cocktail of inactivated viruses containing the viral subtypes H1N9, H3N8, H5N1, and H7N3, all inactivated using beta-propiolactone, a common preparation for creating inactivated viral vaccines. They then tested the ability of this cocktail to protect mice and ferrets against a variety of genetically dissimilar viral subtypes and monitored other markers of health such as weight loss, lung damage, and detectable viral RNA.

H5N8, an avian influenza variant, killed 10% of the test animals that received the vaccine intramuscularly and 30% that received it intranasally. Other than that, all animals survived all other viral challenges for the 14-day duration of the trial, including an H2N7 strain that shared only about 45% of the HA sequence with any strains in the vaccine, and an H3N2 strain that only shared 43% of the vaccine’s NA sequences. 

Vaccinated test subjects were also far more likely to maintain weight after being infected. During the course of their illness, most animals dropped at most 10% of their initial weight, and most returned to a healthy weight by the end of the test round, although animals infected with an H10N7 strain or an H6N1 strain lost between 15-20% of their weight. Even those animals had a 100% survival rate. 

Vaccinated animals had 90-100% less detectable viral RNA in their blood than animals that were inoculated with buffer without a vaccine. Unvaccinated animals sustained damage to their lungs when infected. For example, H1N1, the 1918 pandemic flu strain, produced visible swaths of necrotized bronchioles and pulmonary edema. In comparison, immunized animals showed “minimal histopathological changes, including mild, focal bronchiolitis, an absence of alveolitis, and no viral antigen in alveolar epithelial cells,” according to the authors.

Buoyed by these results, researchers conducting a Phase I clinical trial have begun using a vaccine similar to the one tested by Taubenberger’s group in mice and ferrets. The trial is expected to be complete in March 2023. But what happens if the results are positive? Will a universal human flu vaccine become available?

“A universal flu vaccine could be anything, right?” said Richard Webby, a virologist at St. Jude Children's Research Hospital in Tennessee and the director of the World Health Organization Collaborating Centre for Studies on the Ecology of Influenza in Animals who wasn’t involved with the study. “When the term was first floated, it was really meaning something you got once and it protected you from everything. That was always sort of the Holy Grail.

“But flu is a very, very different virus than those where vaccines are one and done in terms of how it changes, in terms of the protective efficacy, in terms of how fast it replicates. So, I think that's still a very, very tall ask. And I haven't seen anything that would suggest we’re close to having a vaccine like that.”

Current vaccines are potent, offering strong antibody-based protection when they match the correct H- and N-subtypes of that season’s viruses. Since those targets frequently change, other universal vaccine approaches have tried targeting more stable regions of the virus, a common one being the stalk region that holds the HA protein. Unfortunately, the stalk is simply harder for the immune system to spot so the vaccines just don’t work as well. “Most of the universal approaches that target something else in the virus come with a little bit of drop of potency,” said Webby.

With these challenges, some researchers are reconsidering what it means for a vaccine to be universal. Given the frequently shifting nature of influenza viruses, asking just one vaccine, even one that can protect against multiple strains, to cover every possible patient at every point during life might be asking too much.

“There're lots of different strains of influenza,” said Nicholas Heaton, a virologist at the Duke University School of Medicine who was not involved with the study. “There're people who have different exposure histories. There are types of vaccines that brand new babies need, versus young adults, versus the elderly. Those are probably all different populations that all need different vaccines based on their exposure histories, based on the status of their immune responses, based on the strains that are circulating at that time.”

Since influenza is so variable and many different viruses cause flu-like diseases, another illness could make an apt comparison. “Maybe the best analogy I can give you is the ‘cure for cancer,’” said Heaton. The phrase “cure for cancer” has been thrown about since Richard Nixon was president, with seemingly each successive administration promising a “cure” or a “moonshot.”

“The idea was we could cure cancer. And the reality is that there're different types of cancers with different types of mutations and different people. And it's probably not a one-size-fits-all-solution. And I think those types of ideas are going to apply to influenza vaccines.”

Calling a vaccine universal or simply better or stronger matters. Vaccine uptake hovers around 40% of the population in the US, when 80% should be getting it (3). So even creating a 100% protective influenza vaccine would still not really achieve universality. A better goal, according to Heaton, may be to stop aiming for moonshots and simply aim for better.

“We would hate to set ourselves up for a situation where we make a flu vaccine that’s 10 times better than what we have now. It's still not universal, and then that's seen as a failure,” he said. 

References

1. Park, J. et al. An inactivated multivalent influenza A virus vaccine is broadly protective in mice and ferrets. Science Translational Medicine  14, (2022). 
2. Gamblin, S. J. & Skehel, J. J. Influenza hemagglutinin and neuraminidase membrane glycoproteins. Journal of Biological Chemistry  285, 28403–28409 (2010). 
3. Vardavas, R., Breban, R. & Blower, S. A universal long-term flu vaccine may not prevent severe epidemics. BMC Research Notes  3, (2010).