As Covid-19 began spreading in early 2020, one of Linfa Wang’s first ideas was to test the blood of people who’d survived a previous coronavirus outbreak. The virologist, who works out of Duke-NUS Medical School, a collaboration between Duke and the National University of Singapore, has been studying bat-borne viruses for decades. He’d helped show that SARS-CoV-1, which killed almost 800 people in 2003, likely jumped to humans from horseshoe bats. Wang’s new theory was that people who’d recovered from the original SARS might harbor antibodies that could help fight the new one, SARS-CoV-2.
Initially, the experiment was a bust. The patients Wang tested had antibodies only against the older version of SARS. But as a number of Covid variants began spreading early this year, he decided to test the patients again. By this point, many of the Singaporean SARS survivors had also been vaccinated against Covid, providing a rare set of immune systems that had been exposed to proteins from these related coronaviruses.
What Wang found astonished him. After getting the Covid shot, the SARS patients had developed something akin to super-antibodies, which blocked both SARS viruses and a multitude of other coronaviruses. All eight patients had antibodies that, in test-tube experiments, neutralized five different bat and pangolin coronavirus strains that had never infected humans. The results, published in the New England Journal of Medicine in August, offered one of the strongest bodies of evidence that a universal coronavirus vaccine is possible.
The need is growingas public health officials know all too well, three new coronavirus diseases have emerged in just 20 years: first SARS, then MERS in 2012, and now Covid. At his 13th-floor lab a few kilometers from Singapore’s central business district, Wang is working on a prototype vaccine that could generate the same type of wide-ranging immune response he saw in the Covid-vaccinated SARS survivors. His regimen combines a first shot containing the Covid spike protein with a second shot containing a hybrid SARS protein. If it worksWang says experiments in mice are promisingthe vaccine could be deployed in the event of a Covid-26 or a SARS-3.
“We want something that is broadly protective, so that when the next one jumps from animals to humans, we already have a vaccine in hand,” says Melanie Saville, head of vaccine research and development at the Coalition for Epidemic Preparedness Innovations, a global nonprofit based in Oslo. CEPI plans to spend $200 million to develop broad-acting coronavirus vaccines over the next five years. One of its first grants under the program was awarded in November to Israel’s MigVax Ltd., a crowdfunded startup working on a “variant-proof” Covid vaccine in tablet form.
A broad-acting vaccine could provide a ready-to-use weapon against threats such as omicron, which has far more mutations than any previously identified variant and which researchers and governments are scrambling to understand and to develop boosters for. “After delta there is going to be something else, until we run out of Greek letters,” University of Pennsylvania researcher Drew Weissman told me in the days before omicron was named. “By making a booster, you are always a step behind.”
Weissman, who pioneered key technology used in both the Pfizer and Moderna messenger RNA vaccines, is among those working on pancoronavirus shots. Initially, such vaccines are likely to focus on fairly close relatives of Covid, but the more ambitious goal is protecting against a wide array of coronaviruses, including several strains that cause the common cold. Given the plethora of bat coronaviruses lurking in nature, there’s every reason to expect more Covid-like epidemics. Future ones could be “as bad as or even worse than what we are going through right now,” says Anthony Fauci, director of the U.S. National Institute of Allergy and Infectious Diseases and President Joe Biden’s top medical adviser. “Rather than responding to the next outbreak, it is critical to develop a vaccine that would protect against all iterations of coronavirus.”
In September, Fauci’s agency announced $36.3 million in funding for research into pancoronavirus vaccines by scientists at Harvard, Duke, and the University of Wisconsin. More than a dozen academic teams, along with a handful of biotech companies, are working on the problem. Labs at Duke and a few other U.S. universities have already created prototypes demonstrating potent cross-virus immune responses in animals, including against SARS-CoV-1, SARS-CoV-2, and some related bat coronaviruses. The Walter Reed Army Institute of Research also has a shot that’s shown promise against multiple coronaviruses: Its formulation is in a Phase 1 human trial, one of the first such shots to make it that far.
Important questions remain unanswered, such as which parts of the virus to target, which technology works best, and how broad-acting the shots should be. Pfizer Inc., Moderna Inc., and many other major Covid vaccine companies aren’t investing heavily so far, instead waiting as the academic research plays out. Mikael Dolsten, chief scientific officer of Pfizerwhich is researching an omicron booster and is developing ones for the beta and delta variants with its mRNA vaccine partner, BioNTech SE says that, given that existing vaccines work and that mRNA shots can be quickly updated, it could be “a dangerous game” to switch tracks to a pancoronavirus shot. “We are following it, but it’s more of an academic approach at the moment,” Dolsten says. “I would say stay with what works.” He posits, too, that waning effectiveness of vaccines over time could turn out to be a bigger problem than variants, something universal shots wouldn’t necessarily solve.
Stephane Bancel, chief executive officer of Moderna, which is also developing mRNA boosters against the beta, delta, and omicron variants, calls universal vaccines “a good idea” and says he’d be happy to strike partnerships to develop them when viable options emerge. But he cautions that researchers have been working on universal flu vaccines for years without a breakthrough.
It’s true that scientists have long dreamed of developing a vaccine that would eliminate the need for an annual flu shot. There’s been no breakthrough, though several trials are under way. The difficulty has been the pace of mutation: Influenza evolves rapidly. The speed at which coronaviruses evolve is intensely debated, but the case for broad-acting shots is nonetheless compelling, given the promising results from early attempts, the increasing frequency with which coronaviruses have been crossing over from other species, and the staggering human and financial costs of Covid. “It would be crazy not to do something,” says Pamela Bjorkman, a structural biologist at the California Institute of Technology who’s working on a universal vaccine. “For the world to act like ‘Wow, we have solved this’ is really shortsighted.”
The history of flu vaccines offers an object lesson. Early shots were pioneered in the 1940s by University of Michigan virologist Thomas Francis with help from the U.S. Army, which was worried about an epidemic ravaging troops in crowded barracks. Francis and his protege, Jonas Salk, grew the vaccine in fertilized chicken eggs, a method still in wide use today. The first shots showed strong efficacy in many studies. One influenza B shot given to troops receiving specialized training at the university in 1945 proved 88% effective, not far from the 90%-plus efficacy reported for the mRNA Covid shots. The researchers quickly realized that strains could evolve from year to year, resulting in a mismatch between vaccine and virus. But they were optimistic it would be manageable. “The outlook for increasingly broad and effective prophylactic immunization against the range of influenza viruses is extremely promising,” Francis concluded in a 1953 publication.
The early flu shots contained two viral strains, then three in the 1970s, and finally four in the past decade. Research focused heavily on attacking the most abundant protein found on a virus’s surface, hemagglutinin, with flu shots selected each year based on their ability to generate antibodies against it. But for much of the vaccine’s history, doctors couldn’t accurately measure how shots were performing in the real world. After modern viral load tests the same kind of PCR tests used to definitively diagnose Covid todaystarted being used in the 1990s, enabling researchers to study real-world flu vaccine performance more carefully, it slowly became clear that the results varied widely and topped out at middling. Since 2004, even as more of the U.S. population has been getting vaccinated annually in keeping with expanded government recommendations, the shot’s real-world effectiveness has ranged between only 10% and 60%, according to the Centers for Disease Control and Prevention. Most years, it’s around 40% to 50%.
Over the years researchers have come out with flu vaccines that use more modern, faster production methods, including a genetically engineered flu shot grown in insect cells, but there have been no breakthroughs in efficacy. Efforts to develop a universal vaccine that would work against all strains didn’t seriously accelerate until after the H1N1 version of influenza swept the globe in 2009. They’re still years away from fruition. “That’s what I’m calling the mistake of the past,” says Matthew Memoli, a respiratory disease researcher at the NIH. “We really didn’t make any effort. We got comfortable and said we have a vaccine, it works OK.” In May 2020, Memoli and two colleagues penned an essay in NPJ Vaccines calling for immediate research into universal coronavirus vaccines. “I don’t want it to take 50 years to make the same decision,” he says.
Covid vaccines employ more advanced technology than most flu shots, but like flu shots they’ve thus far focused largely on the most obvious targetin their case, the spike protein SARS-CoV-2 uses to get into cells. The mRNA vaccines cause the body to produce millions of copies of the protein, stimulating a potent immune response. That’s obviously been a smart strategy, given the impressive efficacy shown by the Moderna and Pfizer-BioNTech vaccines. (Shots using slightly more established methods, formulated by Johnson & Johnson, AstraZeneca, and others, have also shown strong results.) Both Pfizer and Moderna are testing strain-specific boosters that could be rolled out periodically, just as the flu shot is updated every year. Their mRNA technology allows researchers to easily combine shots against multiple variants into one vaccine, and mRNA shots can be updated faster than the technologies used to make the flu shot. Researchers are cautiously optimistic that the boosters will help provide more durable protection against symptomatic Covid from any variant.
Long term, there’s still a risk that relying on updates to existing vaccines could leave people vulnerable. The coronavirus variants currently scourging the world already appear to be somewhat more able to escape vaccine-induced antibodies than the primary strain, even as overall protection remains strong. In theory, a few more mutations in key portions of the virus could hobble the vaccines, forcing a scramble to catch up. Not to mention that a new bat coronavirus will probably emerge someday.
Bjorkman, the Caltech structural biologist, spent much of her career trying in vain to develop vaccines against HIV, one of the fastest-mutating viruses ever discovered. When Covid appeared, she started reading old papers from the few scientists who did coronavirus research pre-2020. “Every paper I read from coronavirus researchers said it is only a matter of time before we have a huge pandemic caused by a coronavirus,” she recalls. Yet, as far as she could tell, no funding agency had made a sustained investment to come up with a broad-acting vaccine.
Now Bjorkman has turned her attention to the task. Working with researchers at Oxford, she’s been testing a nanoparticle a harmless, cage-like protein structure about the size of a virusthat incorporates key portions of the spike proteins from as many as eight coronaviruses. When she tried the vaccine in mice, it did something remarkable: It generated antibodies that could neutralize coronaviruses not included in the vaccine. She published the results in January, but hardly anyone paid attention. Some who did asked why the approach was needed. Once delta became pervasive in the summer, though, calls from potential collaborators began flooding in.
Bjorkman’s team is one of at least four working on pan-coronavirus vaccines using nanoparticles, which can be adorned with dozens of fragments from different coronaviruses. The proteins assemble into symmetrical balls with 24 or more sides to which immune-stimulating fragments from one or more viruses are attached. Some existing vaccines, including Merck & Co.’s shot for human papillomaviruses, are based on empty virus particles, a form of nanoparticle. But newer, modular nanoparticle vaccines are constructed from a variety of nonviral proteins. Bjorkman’s mix-and-match nanoparticle is a 60-sided cage based on a synthetic design developed at Oxford.
The beauty of nanoparticles is that the immune system recognizes them as a virus, laying the groundwork for a more far-reaching immune response than with single-protein approaches to vaccines. And because they allow spikes from multiple coronaviruses to be placed on a single particle, they can, in theory, trigger immune cells to produce cross-acting antibodies capable of neutralizing numerous similar viruses. With nanoparticles, an antibody-producing cell “is seeing multiple spikes in a very small space at the same time,” says Kayvon Modjarrad, a vaccine researcher at the Walter Reed Army Institute of Research. “We think it’s probably pumping out a lot broader responses against all parts of the spike protein and making lots of each of those types of antibodies.” He says researchers may not need to put every known coronavirus onto a nanoparticle to induce broad protection, just a representative sample.
Modjarrad and his colleagues have developed a nanoparticle-based vaccine using the iron-storage protein ferritin, to which eight Covid spike proteins are attached. In tests on lab animals, it generated neutralizing antibodies against SARS and Covid viruses; results from Phase 1 trials sponsored by the U.S. government are due soon. A third nanoparticle shot, from VBI Vaccines Inc. in Cambridge, Mass., has been found in lab testing to generate strong antibodies against numerous Covid variants and a cold virus called OC43. And two other nanoparticle shots, one out of Duke and another from the University of Washington, showed promise in animal experiments whose results were published this year in top scientific journals. The Duke team and its collaborators got $17.5 million from the NIH in September, the biggest chunk of the agency’s pancoronavirus shot funding to date; this, after outside reviewers rejected a government grant proposal in 2020, because they didn’t see the need, according to Barton Haynes, director of the Duke Human Vaccine Institute.
Then there’s another major promising technology for pancoronavirus shots: mRNA, which would see the body’s cells do the work of producing the vaccine, as they do in the Covid shots available now. Unlike with most other options, manufacturing mRNA vaccines doesn’t involve growing proteins or viruses in live cells, so it’s simpler to do on a large scale. Earlier this year, University of North Carolina virologist David Martinez and his colleagues took a big step toward a multi-coronavirus mRNA shot, by designing prototype vaccines that instruct cells to produce hybrid spike proteins. These proteins blend together the genetic code for chunks of multiple human and bat coronaviruses, thereby inducing a far-reaching immune response. “You can take pieces from a cousin coronavirus and pop it into a vaccine,” Martinez explains. “Now you have made the spike vaccine more broad.”
One vaccine the UNC researchers tested instructs cells to produce a spike protein consisting of one-third of the SARS spike, one-third of the original Covid-19 spike, and one-third of the spike from a bat coronavirus called HKU3-1. Lab mice that received the vaccine were protected against SARS, Covid, and another bat coronavirus strain that hadn’t been included, according to results the researchers reported in Science in June.
Many antibodies made in response to infections and to mRNA vaccines target the business end of the spike protein, the part at the top that binds to receptors on human cells. But this portion mutates quickly, so many scientists are attempting to direct immune response toward parts of the protein that are less likely to differ between related viruses. Duane Wesemann, a biologist at Harvard Medical School, has been comparing blood samples from people who’ve been vaccinated or had mild Covid to try to identify the best strategy for designing universal vaccines. He’s found that the former Covid patients tend to produce proportionately fewer antibodies at first but a wider array of them, and that the antibodies that act against multiple coronaviruses tend to target the bottom portion of the spike and to be less frequent. Bolstering this theme, researchers at the University of Washington, working with San Francisco-based Vir Biotechnology Inc., have already identified some rare antibodies that target the bottom stem of the Covid spike and can neutralize numerous other coronaviruses, including SARS 1, the MERS virus, and even some common cold viruses, according to results reported in Science in August. In addition to providing clues for how to design pancoronavirus vaccines, this work could also lead to monoclonal antibody treatments for those who’ve been diagnosed with Covid.
And, finally, some researchers are trying not only to hit the spike protein, but also to stimulate other aspects of immune response. Gritstone Bio Inc., a biotechnology company based near San Francisco, is focusing on vaccines that also bolster the immune system’s other main line of defense against viruses, killer T-cells, which can identify and destroy virus-infected cells. Because T-cells home in on internal viral proteins that are less likely to mutate, the protection they provide may be less susceptible to viral variants than purely spike-based shots. Gritstone is testing several vaccines that use mRNA to direct cells to produce the spike protein from Covid, as existing vaccines do, but that also add a second coronavirus protein, the nucleocapsid, that’s a target for T-cells. The goal is a shot that stimulates their deployment against protein regions that don’t differ widely between related viruses. One trial is under way in the U.K. using the strategy as a booster, while another human trial is slated to begin in South Africa in late 2021.
In the long run, the hardest part of developing a universal coronavirus shot may turn out to be the economics, not the science. As it stands, the most obvious business incentives run against creating and manufacturing pancoronavirus vaccines. Big companies such as Moderna and Pfizer are expected to make tens of billions of dollars this year and next selling their Covid vaccines and boosters. If variants such as delta and omicron keep emerging, the companies stand to continue profiting by tweaking their existing products with fourth, fifth, and sixth booster shots. That may give them less incentive to prioritize a panvariant shot that might just end the problem for good.
“Pfizer and Moderna aren’t going to fund it,” says Corey Casper, who heads the Infectious Disease Research Institute in Seattle, which is also working on pancoronavirus shots. (In response to a request for comment on this assertion, Pfizer said boosters remain “the best strategy currently available for protecting against circulating variants of concern.” Moderna denied that it would delay moving ahead on a universal vaccine to protect its existing business.) Casper estimates it would cost $600 million for large-scale trials and manufacturing, calling that “a small investment for what would be such a huge return.”
Even so, as with most every conflict between public need and commercial reality, large dollops of government and nonprofit funding would likely be required. It’s not yet clear where all the money would come from or which countries would provide it, though CEPI says it’s committed to bringing a pancoronavirus shot forward. Ultimately, countries are also the likeliest buyers, and companies may not go forward if there’s no assurance that the shots will be bought and stockpiled. Since omicron, Casper says, he’s been “on the phone nonstop” with government officials who seem well aware of the need. Yet where funding is concerned, he’s mostly heard “lip service.”
If governments were to stockpile, says Jason McLellan, a structural biologist at the University of Texas at Austin who’s also working on the vaccines, they could roll them out to the most vulnerable populations quickly, buying time until virus-specific interventions can be developed. “You could get a vaccine that is maybe 60% to 70% efficacious against many existing coronaviruses or future ones that could be used early on in a pandemic to blunt that first wave,” McLellan explains. “Imagine if we had, back in February or March 2020, a vaccine that was 60% to 70% efficacious, how that would have helped with the hospitalizations.”
Wang, the Duke-NUS Medical School virologist, remains optimistic that governments and companies will see the wisdom of investing. He’s planning to spend months or years in his high-rise lab in Singapore, perfecting his vaccine candidate. If it looks promising in animal testing, he hopes more money will be there to bring it through human trials and to manufacture it. “Before Covid, it would have been unthinkable,” Wang says. “The world has changed. This is called preemptive vaccine development.”