Read our COVID-19 research and news.

Masked pedestrians during a surge in coronavirus cases in the San Francisco Bay area in December 2020. A new study suggests a prevalent viral variant there may be more transmissible and lethal.

AP Photo/Jeff Chiu

California coronavirus strain may be more infectious—and lethal

Sciences COVID-19 reporting is supported by the Heising-Simons Foundation.

A new strain of the pandemic coronavirus, first identified and now spreading in California, appears to be somewhat more transmissible and heighten patients’ risk of admission to the intensive care unit (ICU) and death, according to a preprint reporting lab studies and epidemiological data.

The variant is also present in other states, but its prevalence among more than 2000 samples collected in California swelled from 0% to greater than 50% between September 2020 and late January, according to researchers at the University of California, San Francisco (UCSF). “This variant is concerning because our data shows that it is more contagious, more likely to be associated with severe illness, and at least partially resistant to neutralizing antibodies,” says senior author Charles Chiu, an infectious diseases physician and sequencing expert at UCSF. The data suggest the new strain “should likely be designated a variant of concern warranting urgent follow-up investigation,” the authors write in their preprint, which has not been peer reviewed and which they say is expected to be posted online soon.

The findings “warrant taking a much closer look at this variant,” says Angela Rasmussen, a virologist at Georgetown University’s Center for Global Health Science and Security who was not involved with the research. They “underscore the importance of pulling out all the stops in terms of both exposure reduction and increased vaccine distribution and access.”

But other coronavirus experts say more data are needed before conclusions are drawn, noting that among patients with the variant, the study included fewer than 10 who were admitted to the ICU and fewer than 10 who died. “If I were a reviewer, I would want to see more data from more infected people to substantiate this very provocative claim,” says David O’Connor, a viral sequencing expert at the University of Wisconsin, Madison, who was not part of the research.

For their study, the authors sequenced 2172 genomes from virus samples captured from patients in 44 California counties between 1 September 2020 and 29 January. The new variant, which comes in two forms labeled B.1.427 and B.1.429 that carry slightly differing mutations, accounted for 21.3% of these sequences overall. (Under a different naming scheme, the variant is sometimes referred to as 20C/L452R.)

The scientists also studied the medical records of 324 people with COVID-19 who were cared for at UCSF clinics or its medical center. The researchers adjusted the data to account for differences in age, gender, and ethnicity, and found that, compared with patients who had other viral strains, those carrying the variant were 4.8 times more likely to be admitted to the ICU and more than 11 times more likely to die.

Other data suggest the variant is more contagious. The scientists found that people infected with the variant harbored about twice as much virus in their noses, an index of viral shedding, which may make them more infectious to others. In the lab, viruses engineered to carry a key mutation found in the variant were better than control viruses at infecting human cells and lunglike structures called organoids. And in one nursing home where the variant took hold, it spread severalfold faster than in four other nursing home outbreaks caused by other viral variants. “The evidence is growing that this [variant] is more transmissible than [its] immediate competitors,” although not as transmissible as some other variants of concern, says William Hanage, an expert on viral evolution at the Harvard T.H. Chan School of Public Health. (Variants of concern are coronaviruses with mutations that make them more likely to spread, evade vaccines, or make people sicker.)

In lab studies, B.1.429 also impacted the effectiveness of antibodies: It was four times less susceptible than the original coronavirus to neutralizing antibodies from the blood of people who recovered from COVID-19, and two times less susceptible to antibodies from the blood of people vaccinated with the Pfizer or Moderna vaccines. That diminished potency is “moderate but significant,” the researchers wrote.

Robert Schooley, an infectious disease physician and virologist at UC San Diego, praised the paper’s ambition and noted its findings of high viral loads in infected people’s noses. “The biology of having a higher level of virus … would certainly fit the thesis that people would not do as well,” he says. That comports with the fact that “we are seeing here in Southern California more people … for a longer period of time in our ICUs.”

The patient data suggest the variants may be linked to worse outcomes. But although the ICU and mortality findings reached statistical significance, the numbers were small: Eight of 61, or 13%, of hospitalized patients with the variants were admitted to the ICU, compared with seven of 244, or 2.9%, of hospitalized patients who did not harbor the variants. Seven of 62 people (or 11.3%) with the variants died, versus five of 246 (or 2%) of people without the variants.

The authors admit it is not possible to tell whether the variants actually make people sicker or whether, for instance, most of the patients with the variant got sick during the worst months of the pandemic, when health care systems were overloaded and patient care may have been suboptimal. All the variant-infected patients in the study who died at UCSF did so between 22 December 2020 and 28 January, when the area was experiencing a surge of infections.

“Could any of the seven individuals who died with this variant have survived if they received treatment when the state wasn’t in the midst of a surge?” O’Connor asks. “It’s really impossible to know, as the authors acknowledge.”

The real evidence will be seeing if, when introduced elsewhere, these lineages start to take off in similar fashion.

William Hanage, Harvard T.H. Chan school of Public Health

In addition to other mutations, B.1.427 and B.1.429 each have an identical trio of mutations in the coronavirus spike protein, which allows the virus to invade human cells. One of those mutations, dubbed L452R, is thought to stabilize the interaction between the spike protein and the receptor it uses to attach to and invade human cells, increasing infectivity. None of those three spike mutations is found in the three other variants of concern, which emerged in the United Kingdom, South Africa, and Brazil.

Evolutionary biologists also caution against overinterpreting the study. “The work is definitely worth reporting, but I don’t buy that on its own this is sufficient to categorize these as variants of concern,” Hanage says. He notes that B.1.427 and B.1.429 likely emerged in July and June 2020, respectively, but infections have not exploded in the exponential curves seen with the three identified variants of concern. “The real evidence will be seeing if, when introduced elsewhere, these lineages start to take off in similar fashion.”

The paper also offers another cautionary tale about the United States’s subpar effort to sequence coronavirus samples nationwide. It’s “worrisome” that a state like Nevada, which borders California, has fewer than 500 sequences in GISAID, the leading coronavirus sequence repository, O’Connor says. The limited data from Nevada currently suggest the variant represents 27% of collected sequences, according to a database created by Scripps Research using GISAID data.