A more contagious variant of the coronavirus first found in Britain is spreading rapidly in the United States, doubling roughly every 10 days, according to a new study.
Analyzing a half-million tests and hundreds of genomes, a team of researchers predicted that in a month this variant could become predominant in the U.S., potentially bringing a surge of new cases and increased death risk.
The new research offers the first nationwide look at the history of the variant, known as B117, since it arrived in the U.S. late last year. Last month, the CDC warned that B117 could become predominant by March if it behaved the way it did in Britain. The new study confirms that projected path.
"Nothing in this paper is surprising, but people need to see it," said Kristian Andersen, a study co-author and a virus expert at the Scripps Research Institute in La Jolla, Calif. "We should probably prepare for this being the predominant lineage in most places in the United States by March."
Andersen's team estimated that the transmission rate of B117 in the U.S. is 30% to 40% higher than that of more common variants, although those figures may rise as more data comes in, he said. The variant has already been implicated in surges in other countries, including Ireland, Portugal and Jordan.
"There could indeed be a very serious situation developing in a matter of months or weeks," said Nicholas Davies, a public health researcher at the London School of Hygiene and Tropical Medicine who was not involved in the study. "These may be early signals warranting urgent investigation by public health authorities."
Davies cautioned that U.S. data is patchier than that in Britain and other countries that have national variant monitoring systems. Still, he found results from some parts of the U.S. especially worrisome. In Florida, where the new study indicates the variant is spreading particularly quickly, Davies fears that a new surge may hit even sooner than the rest of the country.
"If these data are representative, there may be limited time to act," he said.
Andersen and his colleagues posted their study online Sunday. It has not yet been published in a scientific journal.
When the British government announced the discovery of B117 on Dec. 20, Andersen and other researchers in the U.S. began checking for it in U.S. coronavirus samples. The first case turned up Dec. 29 in Colorado, and Andersen found another soon after in San Diego. In short order it was spotted in many other parts of the country.
But it was difficult to determine just how widespread the variant was. B117 contains a distinctive set of 23 mutations scattered in a genome that is 30,000 genetic letters long. The best way to figure out if a virus belongs to the B117 lineage is to sequence its entire genome — a process that can be carried out only with special machines.
The CDC contracted with Helix, a lab testing company, to examine their COVID-19 samples for signs of B117. The variant can deliver a negative result on one of the three tests that Helix uses to find the coronavirus. For further analysis, Helix sent these suspicious samples to Illumina to have their genomes sequenced. Last month, Helix reached out to Andersen and his colleagues to help analyze the data.
Analyzing 212 U.S. B117 genomes, Andersen's team concluded that the variant most likely first arrived in the U.S. by late November, a month before it was detected.
The variant was separately introduced into the country at least eight times, most likely as a result of people traveling to the U.S. from Britain between Thanksgiving and Christmas.
The researchers combined data from the genome sequencing with Helix's overall test results to come up with an estimate of how quickly the variant had spread. It grew exponentially more common over the past two months.
In Florida, the scientists estimate that more than 4% of cases are now caused by B117. The national figure may be 1% or 2%, according to his team's calculations.
If that is true, then 1,000 or more people may be getting infected with the variant every day. The CDC has recorded only 611 B117 cases, attesting to the inadequacy of the country's genomic surveillance.
In parts of the country where Helix does not do much testing, it is likely delivering an underestimate of the spread, Andersen cautioned.
"I can guarantee you that there are places where B117 might be relatively prevalent by now that we would not pick up," he said.
"There's still a lot that we have to learn," said Nathan Grubaugh, a virus expert at Yale University who was not involved in the study. "But these things are important enough that we have to start doing things now."
It is possible that chains of B117 transmission are spreading faster than other viruses. Or it might be that B117 was more common among incoming travelers starting new outbreaks.
Vaccinations can also be part of the strategy to fight B117. In Israel, where the variant is now predominant, new cases, severe illnesses and hospitalizations have already dropped significantly in people older than 65, a group that was given top priority for vaccines.