Bird flu virus circulated in cows for four months before outbreak confirmed by USDA, analysis shows

As agricultural authorities and epidemiologists try to get their arms around the scope of the latest confounding chapter in the decades-long story of the H5N1 avian influenza virus — its jump into U.S. herds of dairy cattle — they’re turning to the genetic breadcrumbs the virus leaves behind in the animals’ nose, lungs, and, primarily, milk.

On Wednesday, U.S. Department of Agriculture scientists released a preprint — a study that has not yet been peer-reviewed — describing for the first time what their investigations of 220 viral genomes from infected cows have so far turned up. The study’s authors suggest that the spread in cattle started from a single spillover event from birds in the Texas panhandle that may have happened in early December. The USDA didn’t confirm the presence of H5N1 in a Texas herd until March 25.

“These data support a single introduction event from wild bird origin virus into cattle, likely followed by limited local circulation for approximately 4 months prior to confirmation by USDA,” the authors wrote.

The findings add more precision to what had previously been reported by academic scientists. Reading viral genomes can provide clues to the origins of the outbreak and allows researchers to monitor how the virus, which primarily infects wild and farmed birds, is changing as it finds a foothold in bovine hosts.

In an initial analysis of USDA genome sequence data released last week, academic DNA sleuths had revealed that the outbreak in dairy cows has likely been going on for months longer than previously realized, and has probably spread more widely than official numbers would suggest. So far, the USDA has reported 36 herds in nine states have tested positive for the virus.

The new analysis also offers a window into how the bird flu is changing as it spends time in the bodies of cattle.

In the last few years, H5N1 has spread from wild birds to a variety of carnivorous mammals, including foxes, bears, and seals, but in each of those instances, the virus has hit a dead end. The outbreak in dairy cows represents one of the first times that this bird flu virus has demonstrated the ability to efficiently transmit between mammals, said Thomas Mettenleiter, a virologist who served as the director of the Friedrich Loeffler Institut — Germany’s leading animal disease research center — from 1996 until he stepped down last year. The other instance was a number of outbreaks at mink farms in Spain and Finland in 2022 and 2023, respectively.

“These spillover events don’t usually lead to transmission chains,” he said. “This situation is definitely an eye-opener for me.”

The USDA’s analysis found about two dozen mutations that have arisen in the H5N1 virus as it has circulated in dairy cattle that are known to make influenza viruses more deadly or more likely to be able to infect humans.

“It’s still really difficult to draw a risk map out of that, but there seems to be ongoing evolution,” Mettenleiter said. “This is not surprising but it’s good to know. All these mammal-to-mammal passages, as we would do experimentally, put an evolutionary pressure on the virus to mutate and this is what we see with the increase of these known mammalian adaptation markers.”

Vivien Dugan, director of the influenza division at the Centers for Disease Control and Prevention, told STAT Thursday that the mutations found so far did not raise any immediate red flags for increased risk to human health.

“I think based on our analysis of the consensus and some of that raw [sequence] data — because we have a good data-sharing relationship with USDA — we’ve not seen anything that would be concerning to us for mammalian adaptation, at this point,” Dugan said.

The CDC has been testing existing H5 vaccines in ferrets, and found that vaccination appears to offer cross-protection against the virus from the man who was infected in Texas.

Scientists who have been frustrated by the slow drip of data from the USDA’s investigations hailed the preprint on social media as progress. “Really grateful to this research team for sharing this, though I hope they weren’t holding on to the data solely to ensure they published first,” Angela Rasmussen, a virologist who studies pathogens that jump from animals to people at the Vaccine and Infectious Disease Organization at the University of Saskatchewan, in Saskatoon, Canada, posted on X on Thursday.

For weeks, the agency has been facing criticism from scientists and pandemic experts for a lack of transparency and timely sharing of data about the outbreak that has slowed down efforts to track its progress. When the USDA finally uploaded a large tranche of genetic sequences of the pathogen to a public database, researchers eager to analyze the sequences to determine if the H5N1 virus has been changing as it is transmitted from cow to cow quickly discovered that the sequences didn’t include necessary information about when and where the samples were collected. All are simply labeled with “USA” and “2024.”

The USDA has denied taking that basic information — called metadata — off the sequence files. The agency’s Animal and Plant Health Inspection Service has said it is sharing raw sequence data as quickly as it is available and plans to upload “consensus sequences,” which are more thoroughly edited and contain the metadata scientists are seeking, when they are ready.

Helen Branswell contributed reporting.