A Snohomish County man who was the country’s first confirmed COVID-19 patient was probably not the source of the coronavirus outbreak in Washington state, according to a new genetic analysis by an international group of scientists.
The finding upends one of the most compelling scientific narratives of the pandemic’s arrival in the United States, but it also showcases the power of quick public-health action, said Michael Worobey, lead author of the report and head of the University of Arizona’s Department of Ecology and Evolutionary Biology.
As researchers learn more about the pandemic’s roots, it’s becoming clear that the virus entered the United States via multiple paths and at multiple times. But Washington still seems to be the place where it first took hold in this country and flared into a community outbreak, said co-author Joel Wertheim, a molecular epidemiologist at the University of California, San Diego.
The Snohomish County man, sometimes erroneously referred to as “patient zero,” started feeling sick after he returned Jan. 15 from a visit to Wuhan, China — the pandemic’s birthplace. He was confirmed positive Jan. 20. Public-health officials tested and isolated everyone they could identify who came into contact with him, and found no other infections.
So it was baffling when a second case emerged Feb. 28 and genetic analysis showed it was similar to the first, differing by only two mutations. Trevor Bedford, a computational biologist at the Fred Hutchinson Cancer Research Center, and his colleagues, concluded that the two cases were linked — with the Snohomish County man as the original source. They also estimated the virus had been spreading silently for six weeks.
But the new analysis, which was posted Monday on the preprint site bioRxiv and has not been peer-reviewed, says it’s more likely that quick action by public-health officials succeeded in stamping out any spread from the first infection, turning it into a dead end. The outbreak that eventually flared in late February and early March was probably the result of a separate introduction from China around Feb. 13, either directly or by way of British Columbia, Worobey and his colleagues argue.
To reach that conclusion, the team examined many more viral genomes than had been available earlier in the outbreak and found none in Washington that exactly matched the earliest known infection. They also found none of the expected “missing link” genomes, intermediate between the first case and subsequent infections.
They then used computer simulations to, as Worobey put it, “rerun the tape of evolution over and over again.” Thousands of times, in fact, to see how the viral genomes would be expected to change and evolve. Again, they found no indication that the viral strain carried by the Snohomish County man was the source of the state’s spreading outbreak.
To understand how a closely related strain might have been introduced separately, the team examined travel patterns and found that the Feb. 2 ban on air travel from China was actually quite leaky, with an estimated 40,000 U.S. residents returning to the United States from China via airports including Seattle-Tacoma International Airport.
“The remaining influx likely provided ample opportunity for a second introduction to Washington State,” the report says. “It is also possible that the virus entered via nearby Vancouver, British Columbia, which is closely linked to both China and Washington state.”
But Dr. Jared Roach, a senior research scientist at the Institute for Systems Biology in Seattle, said the new analysis is not likely to be the final word. Among other things, the model the researchers used for their simulations didn’t appear to account for super-spreader events, like a choir practice where many people were infected, which could affect the virus’ evolutionary path, he said. Nor did it account for other possible variations in the way the virus spread, with some lineages fizzling out and others catching fire.
“They are far too confident their answer is the only answer,” Roach said. “I think there are other explanations they didn’t consider.”
In a series of 18 tweets Monday, Bedford said he now agrees a second introduction was probably responsible for Washington’s outbreak. However, he’s still convinced that second seed was planted fairly early, sometime between Jan. 18 and Feb. 9.
The exact timing is difficult to determine, Worobey said. But the implications are important.
If the virus had a six-week head start, spreading unseen and, thus, untouchable by public-health measures, then it was essentially hopeless — there’s not much anyone could have done to prevent the eventual flare-up.
But if, as the new analysis suggests, the virus didn’t take hold until mid-February, then control might have been possible — if the country hadn’t fumbled the rollout of testing and tracing capacity.
“The timing suggests strongly that those weeks that were lost were pretty consequential weeks, and that we did have more of a chance (of stopping it) than we realized,” Worobey said.
The paper also criticized the U.S. Food & Drug Administration for halting the Seattle Flu Study, which could have provided valuable insights into the early spread of the virus. The project, funded by Bill Gates, had been collecting nasal swabs from volunteers across King County to study the spread of respiratory disease. Their samples represented a potential wealth of information, but researchers were initially barred from accessing that information because the project lacked specific approval to test the samples for the coronavirus.
The timeline of the pandemic’s spread is being revealed in bits and pieces, and a full picture probably won’t emerge for quite some time. At least two Washington residents who were sick in December later tested positive for antibodies to the new coronavirus, though it’s not clear when their infections occurred. The first recorded death in the United States occurred in Snohomish County on Feb. 26, but posthumous analyses in California confirmed two earlier fatalities, the first on Feb. 6.