Markets that sold animals—some dead, some alive—in December 2019 have emerged as a probable source of the coronavirus pandemic in a major investigation organized by the World Health Organization (WHO).

That investigation winnowed out alternative hypotheses on when and where the pandemic arose, concluding that the virus probably didn’t spread widely before December or escape from a laboratory. The investigation report, released today, also takes a deep look at the likely role of markets—including the Huanan market in Wuhan, to which many of the first known COVID-19 infections are linked.

“We could show the virus was circulating in the market as early as December 2019,” says the WHO’s Peter Ben Embarek, who co-led the investigation. He adds that this investigation is far from the last. “A lot of good leads were suggested in this report, and we anticipate that many, if not all of them, will be followed through because we owe it to the world to understand what happened, why and how to prevent it from happening again”.

Eddie Holmes, a virologist at the University of Sydney in Australia, says that the report does a good job of laying out what’s known about the early days of the pandemic—and notes that it suggests next steps for study. “There was clearly a lot of transmission at the market,” he says. “To me, looking at live-animal markets and animal farming should be the focus going forward.”

Nevertheless, exactly what happened at the Huanan market remains unknown. Genomic analyses and inferences based on the origins of other diseases suggest that an intermediate animal—possibly one sold at markets—passed SARS-CoV-2 to humans after becoming infected with a predecessor coronavirus in bats.

After the report’s publication, the WHO director-general, Tedros Adhanom Ghebreyesus, who was not directly involved with the investigation, posted a statement saying that he looks forward to future studies of the coronavirus’s animal origins—but that he wasn’t content with the examination of a potential laboratory leak. “I do not believe that this assessment was extensive enough,” he wrote. “This requires further investigation, potentially with additional missions involving specialist experts, which I am ready to deploy.”

Huanan-market outbreak

In late January and early February, 34 scientists from nations including China, Japan, the United States and the United Kingdom gathered in Wuhan and assessed data. Today, the team published its findings in a 300-page report.

Much of it is devoted to COVID-19 cases occurring in December 2019 and January 2020. Two-thirds of the 170-odd people who had symptoms in December reported having been exposed to live or dead animals shortly beforehand, and 10% had travelled outside Wuhan.

Chinese researchers sequenced the genomes of SARS-CoV-2 from some of the people in this group, finding that eight of the earliest sequences were identical, and that infected people were linked to the Huanan market. This suggests an outbreak there, according to the report.

However, researchers also found that these genomes varied slightly from those in a few other early cases. Some linked to the market; others did not. This means that the coronavirus might have been spreading under the radar in communities, evolving along the way, and coincidentally occurring in people linked to the market, says the report.

Another possibility is that an outbreak occurred at a farm that provided animals to the market, suggests Holmes. Several infected animals—with slightly different variations of SARS-CoV-2—might have then been sold at markets in Wuhan, sparking multiple infections in humans.

Plenty of animals were sold at the Huanan market. December 2019 records list poultry, badgers, rabbits, giant salamanders, two kinds of crocodiles and more. Chinese officials said that the market didn’t sell live mammals or illegal wildlife, the report adds, but also references unverified media reports suggesting that it did, along with photos that Holmes published after a trip there in 2014, of animals such as live raccoon dogs.

Chinese researchers collected nearly 1,000 samples from the Huanan market in early 2020, swabbing doors, rubbish bins, toilets, stalls that sold vegetables and animals, stray cats and mice. The majority that tested positive were from stalls that sold seafood, livestock and poultry. The researchers also took samples from 188 animals from 18 species at the market, all of which tested negative.

But these animals don’t represent everything sold in the market, notes WHO team member Peter Daszak, president of the non-profit research organization Ecohealth Alliance in New York City. “A thousand samples is a great start, but there’s more to do,” he says. He points out that researchers traced farmed animals at the market back to three provinces in China where pangolins and bats carrying coronaviruses similar to SARS-CoV-2 had been found. Although the pangolin and bat viruses proved too distant to be the direct progenitors of SARS-CoV-2, Daszak says that the animals might provide a clue that outbreaks among animals started in those places.

Market or lab?

The WHO report also concludes that it’s highly unlikely that the coronavirus escaped from a lab at the Wuhan Institute of Virology. Most scientists say that evidence overwhelmingly favours SARS-CoV-2 having spilled over from animals into humans, but a few have backed the idea that the virus was intentionally or accidentally leaked from a lab.

When the report authors visited the institute, its scientists told them that no one in the lab had antibodies against SARS-CoV-2, ruling out the notion that someone there had been infected in an experiment, and had spread it to others.

The Wuhan researchers also said that they hadn’t kept any live virus strains similar to SARS-CoV-2. And in their discussions with the investigative team, they pointed out a Nature Medicine paper showing that similar viruses exist in animals in China, rather than in their lab. They further explained that everyone in the lab has safety training and psychological evaluations, and that their physical and mental health are continuously monitored.

“We were allowed to ask whatever questions we wanted, and we got answers,” says Daszak, who collaborates with researchers at the Wuhan institute. “The only evidence that people have for a lab leak is that there is a lab in Wuhan,” he adds.

Nevertheless, the findings are likely to be contested by some. A small group of scientists have sent letters to the media saying that they wouldn’t trust the outcome of the investigation because it was closely overseen by China’s government.

But others say that the WHO team’s conclusions seem solid. “I’m sure people will say that the Chinese researchers are lying, but it strikes me as honest,” argues Holmes. Matthew Kavanagh, a global-health researcher at Georgetown University in Washington, DC, says that he’s heard no evidence pointing to a lab escape. “But the sceptics are going to want a deeper investigation than the Chinese government allowed,” he says.

He adds that it’s challenging for the WHO to carry out such studies. “The WHO is in a completely impossible position because they are being criticized for not holding China accountable, but they are given almost no tools to compel any country to cooperate,” he says. China holds information closely, and “in that context, the WHO’s team has gotten a good look at a lot of data—but it can only get so far”.

Narrowing down the timing

Some studies have suggested that COVID-19 was spreading among people before December 2019. To explore that possibility, the report authors looked at analyses of SARS-CoV-2 sequences collected from people in January 2020, and estimated that they evolved from a common ancestor between mid-November and early December of 2019. That estimate roughly corroborates the findings of a report published in Science this month.

The researchers also looked at death certificates in China, and found a steep increase in the number of weekly deaths in the week beginning 15 January 2020. They found that the death rate peaked first in Wuhan, and then, two weeks later, in the wider province of Hubei, suggesting that the outbreak began in Wuhan. The report also publishes data on people seeking care for respiratory infections, which similarly suggests that COVID-19 didn’t begin taking off until January.

As for reports of SARS-CoV-2 circulating in Italy and Brazil in October and November 2019, the report calls these studies inconclusive because they were based on partial sequences of SARS-CoV-2, and therefore could be a case of mistaken viral identity.But inconclusive doesn’t mean impossible. And Tedros indicates that there will be more work to come. “This report is a very important beginning, but it is not the end.”

This article is reproduced with permission and was first published on March 30 2021.