A year and a few days after the World Health Organization declared Covid-19 a pandemic, there’s a palpable sense that the pendulum is swinging back: Vaccines have been approved, countries are receiving them through their own purchases or via the international collaboration called Covax, people are making plans to take up their lives again.
Not to be a downer, but: not so fast. A small cadre of scientists is warning that we have not paid sufficient attention to the possibility that SARS-CoV-2, the virus that causes Covid, may be with us long-term. That is not just because it might become an endemic disease, surging up periodically when population immunity dips low enough to let it gain a foothold. It’s also because we haven’t dealt adequately with the implications of the coronavirus being a zoonotic infection, one that leapt between species to cause illness in the human world.
To the degree we’ve approached that problem, it has been by investigating—via an official WHO-sponsored mission and also via conspiracy theorizing—how the coronavirus accomplished its spillover from an asymptomatic bat pathogen to a lethal human one. We haven’t yet tackled the dimensions of a second phenomenon, what researchers are calling spillback. That is the process by which the novel coronavirus jumps from humans into additional animal species, giving it new territory in which to survive and mutate, and maybe jump again. There are already signs that may be happening—and we have not yet begun to set up the systems that will tell us what the virus is doing in its new home.
“Covid-19 is first and foremost a medical public health crisis,” says Christine Kreuder Johnson, a veterinary epidemiologist and professor at the UC Davis School of Veterinary Medicine who directs the EpiCenter for Emerging Infectious Disease Intelligence, a project funded by the National Institutes of Health to detect wildlife-to-human spillovers. “But there are other professions that need to be engaged in this, from the veterinary and agriculture side as well as the environment, in terms of policy and surveillance and ongoing monitoring. We need to understand that this is going to be a long-term problem.”
We’ve always known that Covid-19 had an animal connection. The discovery that the coronavirus causing it was bat-associated came early in the pandemic, and scientists have subsequently theorized that a second, still-unknown species helped the virus make the evolutionary adaptations that allow it to infect humans.
All of that happened before Covid-19 began spreading among people in China in December 2019, breaking out into the world’s notice just before the turn of the year. But within a few months, as the coronavirus spread rapidly through the world’s human population, it leapt from people back into an animal species: minks, being raised in confinement on fur farms.
In April, workers on two fur farms in the Netherlands unknowingly passed the virus to minks being raised there. As it spread from farm to farm, health authorities decided drastic action was necessary, and hundreds of thousands of the animals were slaughtered to prevent the virus from spreading. But by July, SARS-CoV-2 was also in mink farms in Spain. By October, it had landed in Denmark, the largest producer of mink in the world outside China—and by November, the Danish government decided to kill every mink in the country, all 17 million, in order to forestall any evolution of the virus as it passed among them.
This went badly. The minks were asphyxiated and buried in giant trenches, and within a month, gases from decomposition started to push the decaying bodies out of the ground, leading to claims of mink zombies. (They were not zombies.) The new minister of agriculture—the previous one was forced to resign over the mink slaughter—vowed to have the dead minks dug up and incinerated instead.
By then, five more European Union nations, plus Canada, had also recorded cases on mink farms—but they were no longer the only places affected. In August, the coronavirus was identified in minks on fur farms in Utah, and by October 10,000 of them had died. By December, the virus had also invaded farms in Michigan, Oregon, and Wisconsin. (Unlike in Europe, American fur producers did not kill their minks.)
The original spillover had been the initial transfer from bat to mystery animal to humans. The spillback went the other way, from humans back into animals—into hundreds of thousands of members of a different species than the ones that had previously given the virus a place to adapt. That many minks, living in close quarters, could provide SARS-CoV-2 with a huge opportunity to mutate in unpredictable ways. By the end of last year that possibility was confirmed. Among Danish mink, a viral variant appeared that possessed a cluster of mutations not recorded before, changes that allowed the virus to evade some of the immune protection conferred by neutralizing antibodies.
Spillback feels inherently troubling—but it might not be a public health danger if a virus moves back into an animal population but doesn’t spread further from there. An investigation in the Netherlands last summer found a small number of farm workers carrying versions of the virus that, on genomic analysis, had clearly passed through the minks first. Minks on those farms, stressed by confinement and crowding, might be uniquely vulnerable to the virus, and therefore they and their handlers together might form a unique hotspot. Anticipating that, two European health agencies recommended early this month that farmed mink and farm workers undergo regular frequent testing to see which viruses might be circulating on farms.
But what if the virus passed not between one confined species and its handlers but into wild members of that species or other unrelated ones? That scenario haunts veterinarians and public health officials, and it may have come to pass.
In December, the USDA found the virus in a wild mink in Utah that was trapped near a fur farm. Presumably, it acquired the virus by coming into contact with the minks held on the farm, or with farm debris, or even with an escaped animal; USDA officials said no other wildlife trapped and tested in the area were carrying the virus. But the possibility that other wild species could acquire the virus troubles scientists. That could include ones that are closely related to minks (like ferrets), other animals in the same family (such as weasels or otters), or even unrelated ones.
“This is something we have to be very concerned about, and not only because it could establish an alternative reservoir that could then be a source for humans,” says Raina Plowright, a disease ecologist and veterinarian, and associate professor at Montana State University. “In every reservoir, there are going to be different selective pressures on the pathogen, so the virus will evolve in different ways to overcome whichever barriers are present within that species. If we started to have coronavirus circulating in different species, all having slightly different genotypes, then we also have the possibility for new coronaviruses emerging that may be sufficiently different from the current one that they may evade vaccine-induced immunity.”
Those alternative hosts might be bats, the coronaviruses’ apparent original home. Last September, a team of researchers from several institutions estimated that up to 40 species of North American bats might be susceptible to infection and could serve as viral reservoirs. It also might mean nonhuman primates: Johnson, whose NIH-funded project works in South America, worries about possible viral traffic between humans and forest-dwelling monkeys.
But it also could mean species so small that we don’t take notice of them, even though they already live among us and bring diseases near us. Last summer, a team of Canadian researchers showed in lab experiments that North American deer mice—which live everywhere from forests to suburbs and play a role in transmitting Lyme disease and hantavirus—can become infected with SARS-CoV-2, harbor it without symptoms, and pass it to other mice. Whether this would translate to mice in the wild is unknown.
Pretty much everyone who is thinking about the problem of spillback—and the possibility of what scientists are coming to call “secondary spillover” back into the human world—calls for more funding to create some sort of monitoring system: on animal farms, among farm workers, of free-living wildlife. There is a model for how that might work, an existing surveillance system that keeps track of one persistent and intermittently deadly disease: the flu.
Influenza’s ancestral terrain is wild water birds, which pick up the virus, carry it with them as they migrate across the globe, and poop it down onto human society—including onto farms, where it can find fresh hosts, adapt, and produce strains of swine and avian flu that find their way to humans. Because of that persistent risk, an elaborate surveillance network has been built up, which includes scientists from the World Health Organization, the health agencies of various nations, and academic research groups. They sample viruses in waterfowl, monitor pathogens in wild birds and poultry, and track the evolution of the seasonal flu strains that infect humans, looking for the emergence of something threatening and new.
That system didn’t come out of nowhere, though. It didn’t even, entirely, come out of the perception that influenza is a profound public health burden. In its journey from the wild to the human world, the flu passes through an industry that is eager to avoid outbreaks in domesticated animals—like the 2015 avian flu epidemic that devastated the Midwestern turkey industry—that are enormously costly to quell. “The entry point for influenza is agriculture,” says Colin J. Carlson, a global change biologist, assistant professor at Georgetown University Medical Center, and principal investigator for a consortium called the Viral Emergence Research Initiative. “The reason that we’re able to have the global preparedness system that we have is because we know the spillover interface. It’s one that we regularly monitor, and it’s one that has the financial and organizational resources to do that.”
Since before the pandemic began, scientists have been trying to create better detection of pathogens emerging from the wild world into humans. The Trump administration famously defunded the best-known of those, the Predict Network, about two months before the first Covid-19 infections were discovered. Because the pandemic made it belatedly obvious that it’s a mistake to ignore zoonotic threats, other moves to create monitoring have revved up since, from academic viral databases, to a federally supported network of labs focused on threats in different geographical regions, to a commission brought together by The Lancet that scooped up Predict personnel, to an envisioned multinational agency that would shine a global spotlight on new threats.
Almost all of those efforts, like Predict before it was canceled, focus on detecting the first emergence of a virus from wildlife, or clues that one might be ready to jump. The existence of spillback and secondary spillovers makes the task of identifying pathogens as they move between animals and humans much more complicated. In those cases, the viruses are already known and identified; it’s the possible new host animals that require advance identification.
Deciding which animals are vulnerable, and whether they would offer an evolutionary cul-de-sac or a highway onramp to rapid spread, requires a surveillance system that would be much more comprehensive, potentially sampling many more animal species, farmed and wild, before viruses move into them, and understanding what in their contact with the human world make them vulnerable—or make us vulnerable to them. Parts of this might exist now, in animal agriculture and wildlife management agencies, but knitting them together feels so complex that no one can quite articulate yet how it would work.
One approach might be to admit that we don’t know as much as we should, from the macro to the microscopic level, about how an intact ecosystem maintains itself. Anna Fagre, a veterinarian and postdoctoral fellow at Colorado State University who collaborates with Carlson’s project, proposes that we start by developing comprehensive measures for what a healthy wildlife population looks like. With those baselines established and monitored, we could develop a monitoring system for when a new pathogen enters, whether it’s spilling over from other wildlife or spilling back from the human world.
“If we can start to detect when things are going wrong, so we can potentially predict when they may be more immunocompromised or more able to transmit pathogens, we would not only by keeping an eye on their populations,” she says. “We would be safeguarding human health.”
- 📩 The latest on tech, science, and more: Get our newsletters!
- How to find a vaccine appointment and what to expect
- The perplexing psychology of returning to “normal”
- “We’re vaccinated. Everyone wants to visit. Now what?”
- How to remember a disaster without being shattered by it
- Premature babies and the lonely terror of a pandemic NICU
- Read all of our coronavirus coverage here