Earlier this year, U.S. regulators settled on a new strategy for COVID-19 vaccines. Like the annual flu shot, the vaccines will be updated each year based on the virus’ evolution, then rolled out in the fall. Accordingly, on 15 June, advisers to the U.S. Food and Drug Administration will weigh which strain or strains of SARS-CoV-2 should make up the next iteration of vaccine, so that the agency can greenlight a version for companies to mass-produce.
Regulators may well jettison the original SARS-CoV-2 strain that emerged in China and is long extinct—but which people are still being vaccinated against today. Many scientists favor eliminating it. The ancestral strain “should go out of the formulation,” says William Messer, an infectious disease specialist and viral immunologist at Oregon Health & Science University. Last week, the World Health Organization (WHO) agreed. But other questions loom, including whether to bundle multiple virus strains into the vaccine or just one.
To date, COVID-19 vaccines have been modified only once, when a bivalent version based on both the original strain and the BA.5 Omicron variant was introduced in September 2022. Uptake was disappointing: Only 17% of people in the United States have rolled up their sleeves. (By comparison, about 50% get an annual flu shot.) Furthermore, many researchers say the bivalent vaccine packed less of a punch than it could have. The decision to preserve the ancestral strain sprang from worries that if an entirely new variant emerged, an Omicron-only vaccine might falter against it.
This hedging proved unwarranted: All major new variants have flowed from Omicron, which was first detected in South Africa in November 2021. And evidence increasingly shows that a vaccine split between a current strain and one that’s extinct makes it harder for people to mount a strong immune response to the virus.
On 4 May, for example, David Ho, a virologist at Columbia University, and his colleagues posted a preprint study of 72 people, including some who had received four doses of the original vaccine and others who’d gotten three doses and a bivalent booster. Those who got the booster didn’t produce antibodies that were notably better at neutralizing Omicron. The reason, Ho explains, is a phenomenon called immunological imprinting, in which repeatedly exposing the immune system to one strain—in this case, the ancestral one—skews the immune response in that direction. When the decision was made to keep the ancestral strain in COVID-19 vaccines, Ho says, imprinting “was probably not a dominant consideration, but it is now.”
Florian Krammer, a virologist at the Icahn School of Medicine at Mount Sinai, agrees. He and his colleagues published a study this month in The Lancet Microbe in which they studied blood from 16 people 1 month before and about 2 weeks after they got a bivalent booster. After the booster, antibodies in the blood did a slightly better job of neutralizing the ancestral strain than BA.5. Krammer says his team also couldn’t find “specific” antibodies solely reactive to BA.5, which could be especially protective if they’re plentiful.
Last week, a WHO advisory group said in a statement that although current COVID-19 vaccines guard against severe disease, “protection against symptomatic disease is limited and less durable.” In place of a bivalent shot, the group recommended a single-strain fall vaccine based on the XBB.1 lineage now dominating across continents, although it left the door open to other effective vaccine recipes.
Whether a single-strain XBB.1 vaccine is the best bet or whether multiple Omicron strains should be included is a point of debate. In the past few months, two closely related XBB substrains, XBB.1.5 and XBB.1.16, have crowded out other Omicron variants. “We’re basically trying to guess what the next generation of variants will be, descending from which lineage,” Ho says.
“From what we know now, matching the vaccine to whatever circulating variants you’re trying to protect against probably does best,” says Angela Branche, an infectious disease specialist at the University of Rochester. She co-leads a study called
COVAIL that’s examining immune responses spurred by different boosters. It has found that monovalent vaccines against Omicron perform somewhat better than those that include the ancestral strain.
An important question is whether vaccines better matched to current strains could reduce not just severe illness, but also transmission—something current vaccines appear to do poorly. An April study in The New England Journal of Medicine showed that after BA.5 faded and other Omicron strains surged, the bivalent vaccine’s ability to prevent transmission peaked at about 30% 2 weeks after someone got the shot and fell to 0% at 16 weeks. “It’s not an unreasonable supposition” that a closer match could perform a bit better, though the effect is still unlikely to persist, Messer says.
Some researchers also think the updated vaccines should not be limited to the messenger RNA formulations made by Pfizer and Moderna. Novavax makes a protein subunit vaccine, the technology used in hepatitis B and human papillomavirus vaccines. “It would be good to have protein vaccines for the fall,” as those may give more durable protection, Branche says. But it’s unclear whether the company would be able to mass-produce a new vaccine in time.
Robert Frenck, who directs the Vaccine Research Center at Cincinnati Children’s Hospital and helped conduct trials of Pfizer’s COVID-19 vaccine, points out that most vaccines for other infectious diseases “use one methodology,” without causing concern. The strategy against COVID-19 need not be any different, he says.
Messer hopes regulators and companies will stay flexible as COVID-19 knowledge continues to grow—and urges targeting the new vaccines to people at highest risk. In the fall, “vaccine fatigue, COVID fatigue, is still going to persist,” he says, and “triaging your efforts to get good vaccine uptake” will be vital.