Science’s COVID-19 reporting is supported by the Pulitzer Center and the Heising-Simons Foundation.
News from U.S. manufacturer Moderna that its COVID-19 vaccine is still “expected to be protective” against a virus variant first detected in South Africa came as a relief to scientists and the public. But the 25 January announcement included a caveat: Antibodies triggered by the vaccine appear to be a little less potent against the new variant, named B.1.351, than the one the vaccine was developed for. So researchers were perhaps even more relieved to hear the company will start development of booster shots tailored to B.1.351 and other variants.
“These are exactly the steps that I hoped to see,” says virologist Trevor Bedford of the Fred Hutchinson Cancer Research Center. “It may well not be necessary to have a vaccine update in the fall, but taking these steps now is the right course of action.” Other vaccinemakers are also contemplating updates.
Scientists have grown increasingly concerned that new coronavirus variants may worsen the pandemic. B.1.1.7, first detected in England and now spreading globally, has been shown to be more transmissible; on 22 January, the U.K. government said it may be deadlier as well. B.1.351 and a very similar variant named P.1 that originated in Brazil’s Amazonas state are suspected of evading immunity in people who were vaccinated or previously infected.
Now, researchers from Moderna and the Vaccine Research Center at the U.S. National Institutes of Health have tested the potency of antibodies from eight people who had received the company’s vaccine against a retrovirus modified to express the mutated spike proteins of B.1.351 and B.1.1.7. In a preprint, they report that antibodies neutralized the virus in both cases. But for B.1.351, the levels needed were six times higher than for virus expressing the original protein.
A similar study by virologist David Ho of Columbia University, under review at Nature and posted as a preprint on bioRxiv, found that the serum of 22 people vaccinated with Moderna’s vaccine or a similar one from Pfizer was six to nine times less potent against B.1.351, and serum from 20 previously infected people was 11 to 33 times less potent. Researchers in South Africa, meanwhile, have found that antibodies from six recovered patients were six to 200 times less effective at neutralizing B.1.351.
Such drops sound alarming, but the vaccines produced by Pfizer and Moderna trigger very high levels of antibodies, which likely compensates for the decline in potency, says Florian Krammer, a vaccine researcher at the Icahn School of Medicine at Mount Sinai. Besides, antibodies are only one part of the immune response; the vaccines also trigger T cells. Krammer is “quite optimistic” that both vaccines will still protect against B.1.351 and P.1. “However, this is worrisome for vaccines that are not as potent in inducing neutralizing antibodies as the two mRNA [messenger RNA] vaccines.”
Others agree the results don’t spell doom yet. “Given the high starting point, it’s conceivable [vaccine efficacy] could drop only slightly,” Bedford says. Immunity is not binary, adds Jeremy Farrar, head of the Wellcome Trust: “It doesn’t suddenly turn on and turn off.” A drop in antibody potency could have more subtle effects, such as immunity waning a bit faster, he says. The results with sera from recovered patients also suggest the risk of reinfection with COVID-19 may be rising, especially for people who produced low levels of antibodies during their first encounter with the virus, says Stephen Goldstein, a virologist at the University of Utah. “Most of these people I expect to still have good protection from serious disease. It’s on a spectrum, though.”
It may well not be necessary to have a vaccine update in the fall, but taking these steps now is the right course of action.
Moderna says it will start phase I trials of two booster strategies: a third dose of its current vaccine, or of a slightly different one in which the mRNA has been tweaked to incorporate B.1.351’s mutations. They may be given to volunteers 6 to 12 months after the initial immunization, Moderna Chief Medical Officer Tal Zaks said in a call with investors. Pfizer, in an email to Science, wrote that it, too, is “laying the groundwork to respond quickly if a future variant of SARS-CoV-2 is unresponsive to existing vaccines.” Novavax, which is in late-stage trials with a vaccine based the spike protein, says it is “testing sera against the new strains.”
Georgetown University virologist Angela Rasmussen says it’s “very wise” to start to prepare boosters now. “It’s also wise to begin thinking about how they will be distributed,” she adds. “For example, will they be allocated to regions with evidence that B.1.351 is circulating?” Regulators still need to spell out what trials they would require for updated vaccines. At a press conference on Monday, World Health Organization official Bruce Aylward said work to define a regulatory pathway was “kicking off right now.”
Scientists also need to agree on faster ways to address any concerns about immune escape variants, says Farrar, and standardize the way they test antibodies’ potency: “We need harmonization of the assays, so we can compare the results and it doesn’t matter which lab you’re in.” Animal experiments need to be coordinated as well. Vincent Munster, a virologist at the U.S. National Institute of Allergy and Infectious Diseases, says he has already vaccinated hamsters and will challenge them with virus variants in the next couple of weeks. “These studies take a lot of coordination and we are discussing the need for a more planned approach to prepare for other novel variants emerging,” Munster says.
The most timely answers on B.1.351 may come from humans, however. Efficacy trials of several vaccines, including the Pfizer one, are ongoing in South Africa; Tulio de Oliveira, a virologist at the University of KwaZulu-Natal, says researchers are now sequencing the virus from 150 study participants who became infected. “We’re going have the results in 36 hours,” he says. But only after the trial is unblinded next week will researchers know how many of these infections occurred in people who received the vaccine instead of a placebo.
Ho’s paper also sheds some light on how B.1.351 escapes the immune response. The team produced retroviruses with spike proteins incorporating each of B.1.351’s nine mutations separately, as well as all at once. A mutation named E484K accounted for much of the effect, they found. “E484K is really the bad boy here,” Goldstein says. Brazil’s P.1 variant has the same mutation, which might be a sign that the virus has few other tricks to evade immunity, he says: “The virus has a lot of room to evolve but not infinite room. We may have come upon one of the worst possible mutations already.”
But other researchers say the plethora of recent changes is a warning sign that the coronavirus may have more surprises in store—and that the world needs to administer existing vaccines as fast as possible. “I think we need to stop the virus from replicating however we can,” Ho says. “Otherwise, it will keep accumulating more mutations.”
With reporting by Meredith Wadman.