Skip to Main Content

An experimental influenza vaccine developed using messenger RNA technology appears capable of inducing what should be a protective immune response against all known subtypes of flu, at least in animals. If the work is translated into humans it could turn out to be a version of a long-sought universal vaccine.

This would not be a vaccine that would block all flu infections, nor would it replace the need for an annual flu shot. Instead, it would prime the immune system to better respond to new flu viruses, lowering the risk of hospitalization, death, and social disruption.


Flu pandemics, in effect, would be defanged.

Michael Worobey, a professor of evolutionary biology at the University of Arizona, has long been interested in how the immune system responds to influenza. He hailed the work as a potential game changer.

“I think this is some of the most exciting work in vaccinology in a long, long time,” Worobey, who was not involved in the research, told STAT.


The study was published Thursday in the journal Science. The senior author is Scott Hensley, a professor of microbiology at the University of Pennsylvania. The experimental vaccine that was tested was made by Penn immunologist Drew Weissman, one of the pioneers of mRNA vaccines.

In an interview, Hensley explained how he hopes the vaccine would work in people.

“If an H7 pandemic happens tomorrow, and we’re all vaccinated with this,” he said, referring to the vaccine, which targets the H7 influenza subtype, among others, “well, what we would expect is that we would already have protection against severe disease and death.”

“Imagine if the world’s population was primed against every influenza subtype. It is possible that we prevent severe disease and death from any pandemic influenza, a strain that we might encounter in the future. Wouldn’t that be a great thing?” Hensley asked.

The search for a so-called universal flu vaccine has been underway for years, with various groups using different approaches to try to train human immune systems to recognize a dastardly shape-shifting foe. Two types of influenza A viruses, H1N1 and H3N2, and two lineages of influenza B viruses circulate among people, causing hundreds of millions of illnesses annually. The World Health Organization estimates that during normal flu seasons, between 290,000 and 650,000 people die from flu globally.

But myriad other flu viruses exist in nature, infecting a range of mammals. From time to time at unpredictable intervals, a new virus will break out of nature and start circulating in people, triggering a pandemic. Some have been brutal; the 1918 pandemic caused an estimated 50 million deaths worldwide while others — in 1957 and 1968 — resulted in several million deaths. The most recent, the 2009 H1N1 pandemic, was at the other end of the severity spectrum, causing an estimated quarter million deaths.

Hensley and colleagues worked to develop a vaccine that targeted all 18 known types of hemagglutinin, the protein on the surface of flu viruses that attaches to cells and initiates infection; they are numbered H1 through H18. The vaccine also targeted the two lineages of influenza B, named Victoria and Yamagata.

The group ran a series of experiments with the resulting vaccine in mice and ferrets — the latter is the best model for flu disease in humans — to see if it generated immune responses against all of the targets in the animals. It did. The antibody levels in the vaccinated mice remained virtually constant over a followup period of four months, which is a relatively long time in the life of a mouse, Hensley noted.

That work was done in naïve mice — mice that have not previously been infected with or vaccinated against flu. But few humans are naïve to flu; by age 2 most people will have had at least one infection. So the team also needed to test the vaccine in animals that had been exposed to influenza in the past to see if it worked under circumstances that more closely mimic human immune experience with influenza.

Specifically, researchers wanted to see if the vaccine could override a phenomenon known as original antigenic sin, which is sometimes called imprinting.

Our earliest exposures to flu shape our lifetime experience with this family of viruses. People who were first infected with H3N2 viruses in childhood were more likely to become seriously ill during the 2009 H1N1 pandemic, for example. And H3N2-dominant flu seasons take a heavy toll on older adults whose first exposures to flu were with H1N1 viruses. Imprinting also influences how well we respond to traditional flu vaccines. We get better responses to the viruses we first encountered than to viruses that came later.

To see if imprinting interfered with the development of antibodies to other strains of flu, Hensley and his colleagues gave the vaccine to mice that had been infected previously with H1 flu viruses — in some cases a strain that was matched to the H1 in the vaccine, in others a strain that was genetically distant from the vaccine target. In both cases, the H1 antibody levels in the mice were boosted. But importantly, the animals also developed robust antibody responses to the other targets of the vaccine.

“Thus, the 20-HA … vaccine elicits high levels of antibodies against all 20 encoded HAs in mice with and without prior exposures to influenza virus,” the researchers wrote.

The vaccinated mice were exposed to H1 viruses 28 days after they had been vaccinated. The mice infected with an H1 strain identical to the one in the vaccine showed few signs of illness, and all survived. Mice infected with a distantly related H1 virus had more signs of illness and some did not survive, signaling that if a vaccine is developed using this approach, the degree to which it would mitigate illness might vary, depending on how closely matched the vaccine strain is to the strain causing disease.

The group also ran an experiment in ferrets, giving them two doses of the experimental vaccine. Later the animals were infected with an H1 flu virus that was not well-matched to the one in the vaccine. All of the unvaccinated ferrets in that study lost a significant amount of weight, a common response to infection in ferrets; half of them died. The vaccinated ferrets lost about half as much weight as the unvaccinated animals. And none died.

Henley said the work suggests the vaccine could imprint the immune systems of children who haven’t yet been exposed to flu with all the flu viruses they might face in their lifetimes, and override the imprinting problem that exists for adults. “I like to think of this vaccine as an absolution of original antigenic sin in adults,” he said.

He and his team hope to begin a human trial with a modified version of the vaccine soon, one that would target five hemagglutinins.

But there are significant hurdles ahead.

The first relates to the mRNA platform, specifically how well it is tolerated. The Covid-19 vaccines, the first that use the mRNA delivery system, trigger uncomfortable side effects in a significant portion of people who get them. In the parlance of vaccinology, they are reactogenic. Finding a dose that would induce enough protection in people against all 20 targets while still being tolerable to take will be a challenge, said Sarah Cobey, a professor of viral ecology and evolution at the University of Chicago. Cobey, who was not involved in this research, has collaborated with Hensley on other work.

“Reactogenicity is in some ways an Achilles heel right now for mRNA vaccines, and there is a lot of concern about that potential trade-off between immunogenicity and reactogenicity,” Cobey said.

Hensley agreed the size of the dose needed for people could be an issue. In the animal experiments, a dose of 50 micrograms — 2.5 micrograms per flu virus targeted — was used. To put that in context, that is the size of the Moderna Covid booster, which targets two Covid strains. The Pfizer-BioNTech bivalent booster contains 30 micrograms of antigen.

“How low can we go? We haven’t done those careful experiments. And that’s exactly what we’re doing in the lab right now,” Hensley said.

Should human trials show that the vaccine is safe and potentially effective, figuring out a way to get regulatory approval will also be a substantial challenge.

New vaccines brought to market are typically required to show they will block some portion of infections; this vaccine could have a hard time hurdling that bar. A vaccine that significantly reduced hospitalizations and deaths would be of value, but trials to show those impacts might need to be so large as to be infeasible. Likewise, there would be no way to show the vaccine’s protection against strains that don’t currently circulate in people other than by measuring antibody levels or potentially doing challenge studies, where vaccinated volunteers are exposed to a weakened version of a particular virus. Whether ethical approval to do that work could be obtained is unclear.

“The licensure path does not look straightforward to me,” Cobey acknowledged.

Worobey also suggested regulatory agencies may want evidence that use of the vaccine wouldn’t interfere with the immune responses generated by seasonal flu shots. Ultimately, though, he said regulators will need to figure out how to go forward with more broadly protective flu vaccines if the world is ever to have tools with which to blunt the impact of future flu pandemics.

“Do you want to take something that actually is very strongly indicative of being able to prevent the worst-case scenario, which we all expect could happen any time, versus you could research those questions for decades,” he said. “And so I do think you’ve got to try to move this forward, even if you don’t know every single possible implication, if it is as promising as it looks to be.”

Get your daily dose of health and medicine every weekday with STAT’s free newsletter Morning Rounds. Sign up here.

Create a display name to comment

This name will appear with your comment

There was an error saving your display name. Please check and try again.