The End of Evusheld

If you’re immunocompromised, this … isn’t great.

For the first couple of years of the coronavirus pandemic, the crisis was marked by a succession of variants that pummeled us one at a time. The original virus rapidly gave way to D614G, before ceding the stage to Alpha, Delta, Omicron, and then Omicron’s many offshoots. But as our next COVID winter looms, it seems that SARS-CoV-2 may be swapping its lead-antagonist approach for an ensemble cast: Several subvariants are now vying for top billing.

In the United States, BA.5—dominant since the end of spring—is slowly yielding to a slew of its siblings, among them BA.4.6, BF.7, BQ.1, and BQ.1.1; another subvariant, XBB, threatens to steal the spotlight from overseas. Whether all of these will divvy up infections in the next few months, or whether they’ll be pushed aside by something new, is still anyone’s guess. Either way, the forecast looks a little grim. None of the new variants will completely circumvent the full set of immune defenses that human bodies, schooled by vaccines or past infections, can launch. Yet all of them seem pretty good at dodging a hefty subset of our existing antibodies.

For anyone who gets infected, such evasions could make the difference between asymptomatic and feeling pretty terrible. And for the subset of people who become sick enough to need clinical care, the consequences could get even worse. Some of our best COVID treatments are made from single antibodies tailored to the virus, which may simply cease to work as SARS-CoV-2 switches up its form. Past variants have already knocked out several such concoctions—among them, REGEN-COV, sotrovimab, and bamlanivimab/etesevimab—from the U.S. arsenal. The only two left are bebtelovimab, a treatment for people who have already been infected, and Evusheld, a crucial supplement to vaccination for those who are moderately or severely immunocompromised; both are still deployed in hospitals countrywide. But should another swarm of variants take over, these two lone antibody therapies could also be obsolete within months, if not weeks. “It seems like the writing is on the wall,” says Erin McCreary, an infectious-disease pharmacist at the University of Pittsburgh Medical Center. “I live constantly low-key worried that I’m not going to have an active therapy for my patients, and I won’t be able to help them.”

All of this bodes poorly for this winter and beyond. In the near term, millions of immunocompromised people could be left without viable options either to keep SARS-CoV-2 at bay or to temper its blaze once an infection begins to burn. And that loss would set a troubling precedent for seasons to come. The business end of the virus “is now adapting so rapidly that I don’t know how it’s going to be possible for monoclonals to keep up,” says Jeanne Marrazzo, an infectious-disease physician at the University of Alabama at Birmingham. Experts may need to revamp the strategies they use to bring new therapies to market—or find themselves, once again, in a serious bind. “I worry,” Marrazzo told me, “that we’re on a razor’s edge.”


Whatever happens this winter, doctors will still have some options to treat COVID patients. Experts don’t think the virus will develop widespread resistance to our antiviral drugs—molnupiravir, remdesivir, and Paxlovid—“anytime soon,” Marrazzo said. But the vanishing of effective antibody therapies would still leave a massive hole that other treatments can’t fill. The benefits of molnupiravir seem lackluster at best; remdesivir offers a few more perks but is a hassle to administer, requiring several days of infusions. And although Paxlovid has worked wonders for people in high-risk groups, one of its ingredients can screw with a long list of other drugs. McCreary has seen many patients hospitalized, she told me, because their physicians prescribed Paxlovid without properly adjusting their regular meds. “Plus,” she added, “Paxlovid tastes awful.”

[Read: Paxlovid mouth is real—and gross]

Monoclonal antibodies aren’t perfect. But at their best, they’re astoundingly effective and safe, and often the first thing McCreary reaches for when caring for newly infected people. Some patients are also “just more comfortable with monoclonal antibodies than they are with antivirals,” says Mari Nakamura, an infectious-disease specialist at Boston Children’s Hospital. And Evusheld remains the only COVID treatment that is authorized to guard people before they encounter the virus at all. People who don’t mount much of a response to vaccines can sign up for a pair of injections—one into each gluteal muscle—and expect to have their defenses buoyed for a good six months. “I see it as an extension of vaccines for those who are vulnerable,” says Jonathan Abraham, an immunologist and physician at Harvard Medical School.

The greatest strength of these treatments, however, also happens to be their most glaring weakness. Monoclonal antibodies work their magic by glomming so tightly onto SARS-CoV-2’s surface that the virus can’t dock onto our cells. Their grip is ultra precise—enough so that it can be nullified by just one viral mutation in exactly the right spot. Those genetic changes have already booted antibody treatments from our lineup. Now the data hint that bebtelovimab might not work against BQ.1 or BQ1.1. The list of subvariants that might be able to resist Evusheld is even longer: BQ.1, BQ.1.1, BA.4.6, BA.2.75.2, BF.7, and XBB.

Soon health-care providers will have to start making tough calls about when to retire these two antibody treatments—and with few hard rules to guide them. Resistance can be a pretty murky concept: Viral mutations sometimes soften an antibody’s grasp without totally obliterating it. With antibiotics, for example, doctors can respond to some forms of low-level drug resistance just by increasing the dose, McCreary told me. But COVID monoclonal antibodies are still new to the scene. Even when an antibody cocktail has clearly become functionally useless against a given set of variants, there’s no universal standard for deciding when those variants have become so common that the cocktail should be shelved. (When I asked the FDA about this, it declined to comment on specifics.) So the choice is often left up to individual hospitals, Nakamura told me, which can create a bit of a patchwork in how experts are approaching COVID treatment—and put a burden on surveillance efforts to deliver hyperlocal data in real time.

In Pittsburgh, McCreary’s team has, in prior seasons, pulled monoclonals when they stop working against just 20 to 30 percent of the reported variant milieu. Alpana Waghmare, a physician at the Fred Hutchinson Cancer Center and Seattle Children’s Hospital, told me her threshold may be closer to about 50 percent, though she pointed out that the more the options dwindle, the more willing health-care workers may be to keep using a variant-mismatched antibody. Alfred Kim, a rheumatologist at Washington University in St. Louis, told me he’d need to see resistant variants make up “the majority in a region” before he’d even consider putting an antibody out to pasture. There’s little downside to administering the treatments, he said, and for his patients, the potential cost of withholding them is just too immense.


Should bebtelovimab and Evusheld be forced from the stage in the coming months, they might, at least, have a few understudies waiting in the wings. Regeneron, the maker of the late REGEN-COV, has two antibody treatments in Phase 1 trials, according to a spokesperson; AstraZeneca, Evusheld’s parent, also has replacements in development, though a spokesperson declined to provide more details on where in the pipeline they sat. Eli Lilly, which manufactures bebtelovimab and the now-gone bamlanivimab/etesevimab, didn’t respond to my questions about whether they were cooking up new recipes for future use. Vir, which makes sotrovimab—still available overseas—is working on “several highly potent” new antibodies “that have shown activity against all COVID-19 variants tested to date including BQ1.1,” according to a spokesperson.

Clearing drugs for human use remains a plodding process; all of those options could be months away from regular use. “The virus may have moved on” by then, Abraham told me. Already, experts are grappling with whether once-a-year shots will be enough to keep pace with coronavirus evolution; updates on the treatment side may have to come much faster. The problem could get worse as SARS-CoV-2 lineages continue to jockey for control. For the moment, at least, the leading variants are invalidating antibody treatments in relatively similar ways. But if variants diverge further, pharmaceutical companies could have an even tougher time devising broadly effective antibody therapies.

Some experts are also concerned that the market for monoclonals may be going dry. Antibodies are expensive to produce, and with a turnover rate this high, the industry may not have much incentive to stay involved, McCreary told me. Marrazzo, too, thinks the urgency may have lessened with the advent of oral antivirals, and the rush to return to “normal.” If anything, though, the need for good monoclonal options may be growing in urgency. Treatments such as REGEN-COV and bamlanivimab/etesevimab once had clearance to be used in people right after they were exposed to SARS-CoV-2—a sort of emergency antiviral contraceptive. Now no monoclonals are available for so-called postexposure prophylactic use. Kids, too, could use more treatment options. Children under 12 are eligible for three-day courses of remdesivir, given by IV infusion—but those are a tough ask for many families who don’t have the time or means to make such frequent trips to the hospital, Nakamura told me. “And that’s pretty much it.”

Yet no one would feel the loss of antibody-based COVID treatments more than the immunocompromised, Waghmare told me. “It’s this horrible nexus,” Marrazzo said: The most vulnerable people will lose their best options first. Many of those who received Evusheld in the spring will soon be due for their second set of injections, scheduled six months after the first. As of right now, “we’re still telling patients to come in,” McCreary told me. But that may not be the advice she gives next month, or the next. Robyn Ruth, of Augusta County, Virginia, is at that decision point now. Her first experience with the treatment, in April, was momentous: “I had my first hug since the beginning of the pandemic,” Ruth told me. “I just remember my knees buckled, because I hadn’t touched another human being in so long.” In the weeks after, Ruth felt safe enough to go to a couple of doctor appointments and visit a few friends, even garden in their company—activities she hadn’t engaged in since the start of 2020. But as variants continue to chip away at Evusheld’s efficacy, Ruth is steeling herself for the possibility that another dose won’t bring the same relief.

Caregivers and patients alike must now strategize for what could be a very difficult winter stretch. Many immunocompromised people can still benefit from vaccines, even if not as much as others. Marrazzo also cautiously pointed out that if things get bad enough, some providers might go back to convalescent plasma—a treatment with just so-so effectiveness that’s hard to roll out in large quantities, and that doesn’t deliver consistent results—as a desperate stopgap. Other than that, though, it’ll come down to the behavioral measures that many Americans have long since abandoned: isolation, quarantine, masking, distancing.

[Read: It’s gotten awkward to wear a mask]

Nakamura told me she’s been struggling to deliver optimistic advice. “All they can do is try not to get the virus,” she said. She also worries about what might happen should her young patients actually fall ill. “Our hospitals are already overflowing,” she said, amid an early seasonal surge of respiratory viruses, including RSV, and a massive mental-health crisis. McCreary, too, knows many tough conversations are ahead. “There’s nothing worse than one day having something safe and highly effective,” she told me, “and the next day, it’s, ‘Sorry, we don’t have that anymore.’”

For some, the simultaneous disappearance of bebtelovimab and Evusheld could almost rewind the clock to the pandemic’s start. Sara Anne Willette, a data analyst in Ames, Iowa, has a condition called common variable immunodeficiency that keeps her from making certain types of protective antibodies. She also has a history of anaphylaxis to antivirals, potentially making bebtelovimab her only postinfection treatment option should she fall ill. Willette’s second dose of Evusheld is scheduled for December, but she’s not sure whether, by that point, risking the trip will even be practical. “It feels like we’re back at square one,” she told me. “I get COVID, and it’s ‘go it alone.’”