The Coronavirus Could Get Worse

Delta is far from the last variant. But what shape the virus takes next depends on us.

a rubik's cube representing coronavirus evolution
Matt Chase

If evolution is a numbers game, the coronavirus is especially good at playing it. Over the past year and a half, it’s copied itself quickly and sloppily in hundreds of millions of hosts, and hit upon a glut of genetic jackpots that further facilitate its spread. Delta, the hyper-contagious variant that has swept the globe in recent months, is undoubtedly one of the virus’s most daring moves to date. This variant is the product of unfettered transmission, and will thrive further on it; if allowed to, Delta could morph into something even more formidable. “Delta is already a really strong competitor,” Michal Tal, an immunologist at Stanford University, told me. “It could get significantly worse.”

We can’t precisely predict what worse will look like. There is no playbook for evolution. Delta could continue to ratchet up its rate of spread, or it could be ousted by another super-infectious variant. But the speed that has powered Delta’s transmission for months probably can’t sustain SARS-CoV-2 forever, at least not on its own. Humanity’s collective immunity to the virus is growing, which means the next variants we encounter might be better off taking a tack that relies a lot more on stealth. “There’s some sort of tipping point where immune evasion becomes a bigger fitness advantage than transmission,” Stephen Goldstein, an evolutionary virologist at the University of Utah, told me. No one yet knows exactly where that tipping point is—just that we will probably, eventually, collide with it.

This transition will mark a new stage in our extended parlay with SARS-CoV-2. Viruses depend intimately on their hosts—and the global population no longer looks or acts as it did when this one was a fresh threat. A large fraction of us, especially in vaccine-wealthy countries such as the United States, now have some degree of immunity, simultaneously suppressing the pathogen’s ability to pass among us and pressuring it to circumvent those shields. Our defenses are upping the ante for the virus. And the virus will likely rise to meet it.

The cyclical nature of this game might sound disheartening. But nothing will ever put us back at square one. Even as the virus evolves away from us, we can give chase. As immunity builds, our dalliances with the virus will trend milder, shorter, and less frequent. With vaccination on our side, we’re giving the virus fewer turns at the board, and slowing the pace at which the game is played. Although we can’t yet trounce SARS-CoV-2 for good, we can buy ourselves time to make our next decisive move.


In broad strokes, the rules of evolutionary play are simple enough. Researchers still aren’t sure where, or in whom, most variants arise, but they’re clearly more likely to sprout when the coronavirus is allowed to stick around and make more and more of itself, whether in an individual person or in a whole population. Mutations happen like typos during a virus’s messy replication; the majority are inconsequential, even detrimental to the pathogen. But scattered among these genetic glitches will be the occasional windfall, a mistake that helps one version of the virus outcompete its kin. Those proportionally rare events become more absolutely common when given more opportunities to occur. “The longer the virus persists, the more opportunities it’ll have to sample what makes it more fit,” Oliver Fregoso, a virologist at UCLA, told me.

SARS-CoV-2’s self-xeroxing process isn’t particularly error-prone compared with many of the respiratory viruses we regularly tussle with. All else equal, that’s great news: In the few short days most infected people need to marshal immune responses and purge the pathogen, the coronavirus has barely enough time to tweak its genome once, if at all. “The virus that comes out is going to be basically identical to the virus that goes in,” Goldstein told me. And any variants that do arise have little chance to accumulate in high enough numbers to matter. Most mutants “never make it past the person who’s infected,” Siobain Duffy, an evolutionary virologist at Rutgers University, told me. Many that do exit are doomed to extinguish before they can locate their next host.

But all bets are off when the body’s barriers start to break down. Since the pandemic’s start, several independent research groups have uncovered evidence that variants may have an easier time arising in people with weak immune systems, including those taking immunosuppressive drugs—meaning that “immunocompromised people probably have a role to play” in SARS-CoV-2’s evolution, says Ravindra Gupta, a virologist at the University of Cambridge who has been studying this link. Some struggle to clear the virus for months, giving the pathogen time to spawn a menagerie of mutants. Most of the pathogen’s progeny will still be evolutionary duds. But the more of them that are made, the higher the chance that one will rise above the fray and tumble back out into the world. Though it’s tough to prove definitively, this may be the origin story of Alpha, Delta’s super-transmissible forerunner. Its genome is pockmarked with an unusual number of mutations, the telltale sign that a variant may have been stewing inside a single person.

These prolonged infections can’t explain everything. Delta’s genome, for instance, is relatively clean. Its roots might lie in a different sort of numeric abundance—many brief infections in rapid succession.

A variant’s success is also contingent on the specifics of the board it’s playing on, and which opponent it’s facing. Consider, for instance, the Beta and Gamma variants, which both carry mutations that make them much less recognizable to antibodies, a trait that likely helps them wriggle their way into well-defended hosts. They appear to have gained traction in South Africa and Brazil, respectively, where a somewhat large fraction of the population may have already been infected by an older version of the virus. Delta, however, seems to have sprouted first in India, which was slammed later in the pandemic, and where far fewer people had seen SARS-CoV-2 before. In that environment, Delta didn’t need much covertness to establish itself—just a penchant for seriously speedy spread. That strategy helped Delta rapidly outstrip several of its wilier but more sluggish competitors and hopscotch across the globe.


A virus’s primary objective is to spread, through whatever means it can. So far, Delta has had little reason to switch up its tactics. Although the variant appears to carry at least a couple of mutations that help it evade certain antibodies—a probable perk when it infiltrates someone with immunity—most scientists have been much more concerned about Delta’s ability to hack its way quickly and efficiently into cells. The variant is so transmissibly supercharged that it can crest in the body, and probably hop into new hosts, before many of the most potent immune defenses kick into high gear. A virus doesn’t need to be invisible if it can get in and out before security has time to spot it.

But the more people Delta and its comrades infect, the more they disadvantage themselves. “Delta is leaving behind it people with high antibody titers,” Sarah Cobey, an evolutionary biologist at the University of Chicago, told me. Those fast-acting immune fighters stick around and can rapidly purge the variant should it try its luck in the same person again. Humans are also steadily adding to the ranks of the protected with vaccines, which offer even stronger safeguards. To keep itself going, SARS-CoV-2 will need to dodge these defenses.

This, then, is the inevitable push and pull of coexisting with a virus long-term. Immunity shortens and softens infections; virus evolution stretches them back out. Once a large proportion of the population can thwart the virus, SARS-CoV-2 will need to find new ways to “stick around a day or two more,” Bill Hanage, an epidemiologist at the Harvard School of Public Health, told me. The goal is the same—to keep the coronavirus in circulation—but the virus must take a different route to achieve it. That appears to have driven some of the sneaky changes in flu viruses and common-cold coronaviruses that allow them to reinfect old hosts. The more pressure on a pathogen, the more incentive it has to escape.

In the worst-case scenario, a variant could arise that would “make it like the vaccines did not exist,” Hanage said. But at the moment, “there is no such variant like that.” And it would probably be extraordinarily difficult for one to manifest. Even the most evasive variants we know of—the ones that have stumped certain antibodies—aren’t fully duping vaccinated bodies, which harbor a slew of other immunological guards. Hanage also pointed out that many people’s immune systems have been trained on different triggers—distinct brands of vaccines, unique variants, or some combination thereof. A new version of SARS-CoV-2 would find skirting all of those blockades at once to be nearly impossible.

Viruses aren’t infinitely mutable; sometimes, to keep themselves in contention, they must make sacrifices. Several experts told me they’re hopeful that the coronavirus might struggle to max out both transmission and immune evasion at once, requiring some sort of trade-off between the two. Some of the most powerful anti-coronavirus antibodies target SARS-CoV-2’s spike protein, which the virus uses to unlock and enter our cells. If the virus altered the protein to sidestep those antibodies, it might make itself less recognizable to the immune system. But it could also hurt its ability to infect us at all.

That might help explain why Beta has, so far, remained only a supporting character in the coronavirus’s ensemble cast. Another hint comes from Alpha, which didn’t seem to benefit all that much when it acquired an antibody-eluding mutation last spring, despite widespread fears. There is, in other words, probably a limit to just how bad SARS-CoV-2 can get: Even the most careful dog breeders cannot turn a bulldog into a bear.


What lies ahead might, in some ways, feel never-ending, like a series of checks with no checkmate. Vaccine recipes can be tweaked to accommodate new variants, and boosters can refresh fading immune memories. But that doesn’t make extra shots enjoyable to take.

Vaccines, however, aren’t just reactive. They are also proactive interventions that curb the number of times the virus gets to roll the evolutionary dice, cutting down on the number, intensity, and duration of infections, and the chance that they’ll pass to others. A more vaccinated world creates a more hostile global environment for SARS-CoV-2. Mutations will still occur, but fewer of them will be of consequence; lineages will still splinter, but they’ll do so less often. “The overriding effect of vaccination should be to reduce the rate of [virus] adaptation,” Cobey told me. Variants, after all, can’t adapt when they’re starved of hosts to infect.

Glimmers of early evidence suggest that this slowdown has already begun. One recent study, not yet published in a peer-reviewed journal, found that SARS-CoV-2’s shape-shifting rate is lower in highly immunized countries, the expected outcome of a virus knocking up against new immune walls. Gupta, of the University of Cambridge, also hopes that we’ll someday cook up vaccines that can stamp out infection and transmission to an even greater degree—or ones that direct immune cells to hit the virus in spots that can’t mutate without hamstringing it. “That will force the virus into a corner,” he told me. We’d need those types of inoculations less often, too. “I don’t envision a constant cat-and-mouse game.”

This is not yet our reality. Billions of people around the world have yet to receive a single inoculation; even the vaccine-rich U.S. is deep in a dire summer surge. Delta’s rampant spread is driving more disease, more death—and more opportunities for mutation in the virus. In the absence of vigilant masking and distancing, people are getting battered with gobs of virus, testing the protective limits of even recently inoculated bodies. New variants will continue to appear at unprecedented speeds “until we get to the point where the virus is not allowed to replicate this often, or this quickly,” Jennifer Dien Bard, a clinical virologist at Children’s Hospital Los Angeles, told me. Unabated transmission also raises the risk that some people could become viral mixing vessels: Should two variants come to occupy the same cell, they could swap hunks of their genome with each other, birthing hybrids with the nastiness of both parents in tow. Where transmission occurs unabated, “that is definitely a risk,” Lisa Gralinski, a coronavirologist at the University of North Carolina at Chapel Hill, told me. Viruses don’t want things. But if they did, it would be very close to this.

A continued arms race with the virus is inevitable; hitting the immunological tipping point probably is too. How quickly we reach it, and how disadvantaged we are when we do, are not. Masking, distancing, ventilation, and other interventions can limit viral spread, but vaccines remain our most powerful tools: They put some of the controls back in our hands, allowing us to safely accelerate our acquisition of immunity. Anyone who isn’t inoculated will eventually become infected, likely within the next few years, creating many of the same immunological hurdles for the virus to clear—but with a devastating public-health cost. “There’s no scenario we choose where we don’t impose selective pressure on this virus,” Goldstein, of the University of Utah, told me. “But are we going to do it while we prevent people from dying, or not?”

The Atlantic’s COVID-19 coverage is supported by grants from the Chan Zuckerberg Initiative and the Robert Wood Johnson Foundation.

Katherine J. Wu is a staff writer at The Atlantic.