Prototype universal influenza vaccines such as this one, a nanoparticle with a protein scaffold (blue) studded with influenza hemagglutinin proteins (green), could provoke an immune response to all human influenza subtypes.Credit: NIAID
Before anyone knew a pandemic was looming, the virus spread steadily on Mexico’s pig farms. At some point, it acquired a mutation. That mutation gave the virus, which belonged to an influenza subtype known as H1N1, the ability to replicate and spread in humans. In the spring of 2009, reports from Veracuz, Mexico, showed that it had made the leap. Then it spread like wildfire, and within a year it had claimed the lives of as many as half a million people worldwide, the US Centers for Disease Control and Prevention (CDC) estimates.
As devastating as the 2009 swine flu pandemic was, it could have been worse — far worse even than the ongoing battle with COVID-19, which has so far killed more than 2 million people. The infamous “Spanish flu” pandemic of 1918 — also caused by an H1N1 strain — killed more than 17 million people, and was among the deadliest plagues in human history. Yet despite decades of warnings from experts about the dire threat of an influenza pandemic, no vaccine exists to prevent another just like it.
Now emerging insights from a wide range of fields, from structural biology to immunology, and bioinformatics to bioengineering, are raising hopes for a universal influenza vaccine (UIV) — a potentially pandemic-stopping vaccine that could train the human immune system to recognize a broad spectrum of influenza viruses, says Stacey Knobler, senior director at the Sabin Vaccine Institute. “There’s a sense of emerging optimism because of what we’re seeing from the science.”
Evasive maneuvers
Influenza vaccines have been around since the 1940s, and they offer important protection from seasonal flu. Yet the protection is far from complete. Influenza vaccines are regarded as good if they protect roughly 60% of recipients from infection, and some years that number dwindles to less than 20%. Partly as a result, influenza kills hundreds of thousands of people each year.
Influenza is tough to stop because both major types of influenza virus — type A and type B — encompass a vast number of distinct strains, and immunity to one strain often does not confer protection from another. Researchers make educated guesses as to which strains are likely to spread in a given year, then develop a vaccine meant to stop them. But influenza evolves so quickly that an annual vaccine may not even confer full protection against that season’s strains.
Predictions are even harder for pandemic strains, which jump without warning from animals to humans. “Our ability to predict the next pandemic strain is almost nonexistent,” says Keith Klugman, a specialist in respiratory pathogens who directs the pneumonia programme at the Bill and Melinda Gates Foundation.
Researchers reason that an effective influenza vaccine should spur the immune system to target antigens that are both widely shared among diverse influenza strains and able to provoke a strong immune response. “The key barrier is that we do not have an effective target to induce long-term, cross-protective immunity,” Klugman says.
Finding common ground
As the swine flu spread in 2009, University of Chicago immunologist Patrick Wilson uncovered some of the first clues that cross-protective immunity might be possible. He was investigating how the immune system responds when it encounters a novel subtype of influenza — in this case the H1N1 strain that caused that pandemic. He and his colleagues found that people who had previously been infected with or vaccinated against one strain of influenza generated antibodies against common features of the new pandemic strain.1
This suggested that a broadly cross-protective immune response was possible. “These were just the kind of antibodies you would want for a universal vaccine,” Wilson says.
Wilson is focusing on antibodies to a viral surface protein called hemagglutinin (HA) that mediates the initial steps of an influenza infection, helping the virus bind to and penetrate host cells. By identifying enough antibody binding sites, or epitopes, that are both strongly immunogenic and highly conserved, he and others could develop mosaic antigens containing several potent epitopes that together invoke broad cross-protection.
New computational tools can also help develop mosaic vaccines, by identifying common features among different influenza strains. Ted Ross, a microbiologist and immunologist at the University of Georgia, is using them to scan HA protein sequences. “We’re using multiple rounds of consensus building and design on strains of flu that represent diverse time periods or regions of the world,” says Ross. Shared HA segments could be combined in mosaic HA proteins that elicit more robust protection than natural HA against emerging strains, including pandemic strains.2
But HA evolves more readily than other influenza proteins, and new versions of HA can camouflage virus strains and help them escape immune detection.
Influenza pandemics are caused by strains that make the jump from animals to humans.Credit: Honglei Sun
Immune reinforcements
Rather than targeting HA, Neil King, a biochemist at the University of Washington, and his UW colleague David Baker are targeting neuraminidase (NA), a key protein that helps influenza escape from infected cells.
To elicit protective antibodies against NA, they are using a powerful program called Rosetta to design new versions of NA that are both more robust and more immunogenic. “Traditional flu vaccines have done a terrible job of eliciting anti-neuraminidase antibodies because they’re not designed to do that,” King says. But a vaccine that provokes a strong response to both HA and NA would fight the virus on two fronts at once.
Besides blocking viral proteins, a successful antibody response to infection triggers other immunological effects that could be essential in stopping pandemic influenza, says Galit Alter of the Ragon Institute of Massachusetts General Hospital, MIT and Harvard. To decipher these effects, her team uses sophisticated analytical techniques to perform what she calls “systems serology.”
Alter and her colleagues reported recently that effective protection depends heavily on signals transmitted by parts of the antibody that don’t bind viral proteins.3 These signals activate the innate arm of the immune system, which mounts a rapid but non-specific defense against invading pathogens and infected cells. “It’s not just the antibody’s reactivity that matters,” Alter says. For an antibody-focused vaccine to achieve its full protective potential, “we also have to think about it as a molecule that can harness the immune system.”Beyond the antibody response, the T-cell-mediated cellular arm of the adaptive immune system can halt disease progression by clearing away infected cells. To stimulate this response, Sarah Gilbert’s team at the University of Oxford has been using harmless viruses to ferry various influenza proteins into human cells.4 The cells would then trigger an anti-influenza T-cell response that targets these proteins and provides lasting protection against diverse flu strains.
Outsmarting the virus
It remains unclear how universal any single vaccine formulation can be — though that’s no reason to shy away from trying for universal protection, experts say. “The virus is going to continue to evolve, and we have zoonotic sources that will keep introducing variants into our population,” Ross says. For these reasons, “I think it’s unlikely that we can make a single vaccine now and be done.” Still, he says, broadly cross-protective vaccines are within reach. These ‘super-seasonal’ vaccines could provide extended protection against non-pandemic flu strains for years on end — and could induce immunity to potential pandemic subtypes.
Sophisticated intel on individual immune responses could also help boost vaccine efficacy. The immune system trains itself to react to the virus variants it encounters early in life — a phenomenon called imprinting— which lessens its ability to respond to new strains of influenza encountered as we age. “If we can't overcome imprinting, your universal flu vaccine may have to be given as a first exposure to a newborn infant,” says Klugman.
More intel could be collected using the Alter group’s systems serology approach, which can help reveal how different factors shape an individual’s response to both infection and vaccination. This knowledge could help develop vaccination strategies tailored for different populations. “By creating these tailored vaccine responses, we can then utilize preexisting immunity to shape the quality of the response,” Alter says.
Speeding delivery
No matter how well designed a vaccine is, it must be produced at scale and delivered. Today influenza virus is grown at scale in live cells or chicken eggs, extensively purified, then processed to isolate key viral antigens. It can take up to six months before sufficient vaccine is available for population-wide inoculation — a timetable that could cost countless lives in a pandemic.
Alternative design and formulation strategies could streamline the production process. King and Baker are using Rosetta software to design viral proteins that self-assemble into compact nanoparticles. “They’re tailored to present influenza antigens in the optimal way to get the best possible immune response, and that's something that you can only do with computational protein design,” says King. In a recent preprint, the pair claimed proof of concept that this approach could match the protection provided by existing annual flu vaccines.5
RNA vaccines offer hope as well, since they are cheaper and faster to manufacture than protein-based biologic drugs, Ross says. New RNA sequences can easily be swapped into a production pipeline, helping upgrade vaccine designs as researchers learn more about antiviral immunity.
As discoveries emerge from the labs, funders and vaccine advocates have taken notice, and they’ve begun marshalling their forces for a major push toward a UIV. In 2018, the Global Funders Consortium for Universal Influenza Vaccine Development was formed as a joint effort between government agencies, non-profits, and research foundations in North America, Europe and Asia. In 2019 the Gates Foundation and Flu Lab issued $12 million in funding to support innovative strategies for “universal” influenza vaccine development.
Also in 2019, the National Institute of Allergy and Infectious Diseases (NIAID) launched a network of Collaborative Influenza Vaccine Innovation Centers, part of a seven-year program, launched with $51 million its first year, to advance new solutions for universal vaccine development. “This is one of the largest programmes that NIAID has ever launched,” says Ross.
Meanwhile, Sabin continues to promote emerging technologies to achieve new vaccine breakthroughs. Knobler is encouraged by how rapidly COVID-19 vaccines were developed, and hopes they will serve as inspiration for the post-pandemic battle against influenza. “There is a redoubling of effort toward a universal influenza vaccine,” she says. “This is an enemy we know, and defences should be shored up now.”