Researchers who have examined the genomes of two coronavirus infections in Washington State say the similarities between the cases suggest that the virus may have been spreading in the state for weeks.
Washington had the United States’ first confirmed case of coronavirus, announced by the Centers for Disease Control and Prevention on Jan. 20. Based on an analysis of the virus’s genetic sequence, another case that surfaced in the state and was announced on Friday probably was descended from that first case.
The two people live in the same county, but are not known to have had contact with one another, and the second case occurred well after the first would no longer be expected to be contagious. So the genetic findings suggest that the virus has been spreading through other people in the community for close to six weeks, according to one of the scientists who compared the sequences, Trevor Bedford, an associate professor at the Fred Hutchinson Cancer Research Center and the University of Washington.
Dr. Bedford said it was possible that the two cases could be unrelated, and had been introduced separately into the United States. But he said that was unlikely, however, because in both cases the virus contained a genetic variation that appears to be rare — it was found in only two of the 59 samples whose sequences have been shared from China, where the virus originated.
A scientist who was not involved in the analysis said he agreed with the conclusion that the second case was connected to the original Washington case. “I think he’s right,” said Andrew Rambaut, professor of molecular evolution at the University of Edinburgh, referring to Dr. Bedford. “It’s extremely unlikely that two viruses coming from outside the U.S.A. independently would arrive in the same geographical area and be genetically related unless they were connected.”
State and local health officials have been hamstrung in their ability to test widely for the coronavirus. Until very recently, the C.D.C. had insisted that only its test could be used, and only on patients who met specific criteria — those who had traveled to China within 14 days of developing symptoms or had contact with a known coronavirus case.
If the virus has been spreading undetected in Washington since mid-January, that could mean that anywhere from 150 to 1,500 people may have it, with about 300 to 500 people the most likely range, said Dr. Mike Famulare, a principal research scientist at the Institute for Disease Modeling in Bellevue, Wa., who performed the analysis. These people “have either been infected and recovered, or currently are infected now,” he said.
Many of those people would now be in the early stages of incubating the virus, and might not yet be contagious, Dr. Famulare said.
Dr. Famulare’s estimate was based on a simulation using what scientists have learned about the incubation period and transmissibility of the virus. He called his figures a “best guess, with broad uncertainty.” Another method, based on the size of the local population, the number of tests performed and the proportion of those that were positive, produced similar estimates of how widely the virus may have spread in the community.
The scientists immediately reported the genomic sequence and their findings to state and federal health officials. The scientists said they had already been working closely with government health officials. “There’s an enormous collaborative spirit right now,” Dr. Famulare said.
Neither the C.D.C. nor the state or local health departments immediately responded to a request for comment.
The first patient, a man in his 30s, has recovered after being treated in a hospital isolation unit. The later patient, a teenager, had a mild enough illness to recuperate at home.
According to a statement by the Snohomish Health District, the teenager was unaware that he was being tested for the coronavirus. His case came to light on Friday because he went to a clinic on Feb. 24 to be tested for the flu, and his sample was shared with the Seattle Flu Study, which tested it for a variety of pathogens including the new coronavirus.
“I do think, as more community cases start popping up in the United States, this approach and technique could prove very useful to figuring out the extent of community transmission we currently are having,” Dr. Bedford said of the genetic analysis.
Similar analyses have helped public health officials trace cases and fight outbreaks of Ebola in West Africa and the Democratic Republic of Congo.
On Saturday, local health officials in Seattle said that delays in being able to test for the virus had slowed identification of community cases, meaning those who did not travel to places with major outbreaks or have contact with known patients. “If we had the ability to test earlier, I’m sure we would have been able to identify patients earlier,” said Dr. Jeffrey Duchin, health officer for Seattle and King County.
Two more confirmed cases in the state were announced on Sunday.
The genomic technique used to compare the viruses is akin to constructing a family tree. “As a virus passes from person to person, there will be errors that occur” as copies of the virus are made, Dr. Bedford said. To explain, he compared the tiny mutations in the genetic sequence to mistakes made during a game of telephone. “Those can link up,” he said.
The first case had one genetic difference from the original virus that was detected in Wuhan. The new case had that mutation, plus three additional ones. More than 125 genomes derived from samples taken from coronavirus patients around the world have been shared among scientists thus far, providing data for the analysis.
In the first case in Washington, the man in his 30s had been traveling in Wuhan, China, and returned home to Snohomish County, Wash., on Jan. 15. He sought medical care a few days later after developing symptoms and suspecting that he might have the coronavirus, officials have said, and tests later came back positive.
Health officials then scrambled to retrace his history, tracking down eight people he had socialized with at a group lunch and 37 more who were in the clinic when he showed up for medical help. They also reached out to people on his flight back to the United States.
But as the man remained in hospital isolation, and then later returned home, officials reported no new cases in Washington state. They tested two dozen people over a span of five weeks, and all came back negative.
That changed this week, when the state laboratory became able to test for the virus. Officials reported two new confirmed cases Friday night, and then more, including the first patient to die of the virus in the United States. They are now working to trace how the cases in the state might be linked, and who else might have been exposed.
Two cases have been detected at a skilled nursing facility in Kirkland, Wash., where officials said dozens of other people also had symptoms that could be a sign of coronavirus infection but could also be symptoms of flu.
Gov. Jay Inslee of Washington has declared a state of emergency, and said officials may need to take steps like canceling sporting events and closing schools to slow the spread of the virus in the community.