A woman is tested for COVID-19 in a parking lot in Richmond where the health department set up a temporary walk-up testing center. Fewer Virginians are being tested for COVID-19 through local health departments, making it difficult to track statewide case rates. (2020 photo by Ned Oliver/Virginia Mercury)
For months, whenever a lab in Virginia sequences a sample of the virus that causes COVID-19, it’s almost unfailingly been linked to the delta variant.
“It’s been really hard to see anything but delta,” said Dr. Amy Mathers, director of the clinical laboratory at UVA Health. Nationally, it accounts for 99 percent of new coronavirus cases. Over the month of September, Mathers’ lab analyzed more than 1,000 samples from Virginia and didn’t find a single other variant. But after months of concern over the highly infectious form of the virus, public health experts have a new variant to worry about.
There’s much that’s still unknown about omicron, a new version of the SARS-CoV-2 virus first detected in South Africa and now identified in two U.S. states. Virginia has yet to report a case. But the emergence of the lineage — classified as a “variant of concern” by the World Health Organization and U.S. Centers for Disease Control and Prevention — is forcing Virginia to revamp its detection strategy. That responsibility largely falls to the state’s public health laboratory, which has been analyzing samples of the virus since early in the pandemic.
More accurately known as whole genome sequencing, it’s a way of unraveling the entire genetic code of the virus. And as SARS-CoV-2 has evolved into new and sometimes more dangerous variants, it’s been a crucial tool in identifying cases linked to those lineages. COVID-19 tests can verify if a person’s contracted the disease, but only sequencing can confirm whether it’s a version of the virus caused by a variant of concern.
But Virginia, like many states, is still limited in its sequencing capacity. While there have been improvements over the course of the pandemic, the state has only sequenced around 2.7 percent of its total coronavirus cases, according to data from the CDC. It’s higher than some neighboring states (Kentucky and Tennessee, for instance, have sequenced less than 2 percent of their total cases), but lower than neighbors like West Virginia and far lower than national leaders including Vermont and Wyoming, which have each sequenced more than 20 percent of their cumulative cases.
Virginia, of course, has a much higher population than either state and has recorded more than eight to 10 times the number of COVID-19 cases. But Dr. Denise Toney, director of the state’s public health lab, has acknowledged she’d like to boost Virginia’s sequencing numbers.
Currently, the state has the capacity to sequence between 800 and 900 samples a week through its public health laboratory and two contracted partners — Mathers’ lab at UVA and Virginia Tech’s Fralin Center. However, Toney said Virginia would need to sequence between 1,100 and 1,400 samples a week to be 95 percent confident that it could pick up omicron if it was circulating in the community.
“That’s based on some of the UVA modeling that’s been shared with us,” she said. And while the public health lab is sequencing every positive COVID-19 test result it processes, it’s also conducting far less testing than it’s able to. The state’s laboratory network, including UVA and Virginia Tech, has the capacity to test more than 5,000 samples a day, according to Toney. Currently, though, it’s only running between 500 and 600.
“Many hospitals now have the capacity to do their own testing so they don’t need to send it to us,” Toney said. “Lots of school systems, correctional facilities have all brought on the point-of-care tests. There’s so much testing out there that there’s not a need to send samples to the state laboratory anymore.”
In some ways, it’s a major advancement over the early days of the pandemic, when Virginia’s public health lab was the only source of COVID-19 testing in the state. But as officials increasingly recognize the importance of detecting new versions of the virus, the abundance of at-home tests has, in some ways, become a barrier to variant surveillance.
Omicron, for example, has multiple mutations along its spike protein, a target of some PCR-based tests. When a sample linked to the variant is tested through those platforms, it can register as a target failure, or S-gene dropout — a key clue to send the sample for sequencing.
The state lab is currently conducting all of its testing on a platform that targets the S-gene to better isolate potential cases of suspicious variants (the B.1.1.7 lineage, sometimes better known as the UK variant, had the same testing failure). Over time, though, PCR tests have increasingly been supplanted by antigen or at-home test kits, with samples (and results) that generally aren’t passed on to the state.
“A lot of community testing is now antigen-based, for example,” Tomey said. “And those don’t have residual samples to be used for genetic sequencing.”
Even with those limitations, expanding capacity is largely a function of funding. Mathers said the CDC has recognized the importance of better variant surveillance, but her lab doesn’t currently have the funding to significantly boost its numbers. The state’s public health lab has added additional sequencing equipment over the pandemic, including, most recently, a robotic system that will automate much of the prep work. Still, Toney said it will likely only boost capacity by around 100 samples a week.
“Of course, if we could spend a lot of money to get 20,000 samples sequenced a day, that would be even better,” said Madhav Marathe, director of the Network Systems Science and Advanced Computing Division of UVA’s Biocomplexity Institute. Responsible for much of the pandemic-related modeling and predictions across the state, the institute helped develop target sequencing numbers based on an equation that includes the prevalence of new cases across Virginia.
Still, Marathe said a fundamental shift in strategy could be just as helpful as boosting capacity. Like many states, Virginia has largely focused its sequencing efforts on cases with potential public health consequences — including those linked to fast-spreading outbreaks or vaccine breakthroughs. Those can be a useful indicator of possible variants, but with little real information on how the omicron lineage behaves, detecting it might require casting a wider net.
“We’re shifting attention from prevalence and outbreaks to a variant detection program,” he said. “So, what gets sent in has to change in some form.” Both the state lab and the Virginia Department of Health have asked testing providers to send in any samples that register as an S-gene dropout on PCR platforms. They’ve also asked hospitals and commercial labs to submit at least 50 percent of their positive samples to the state’s laboratory network for sequencing. (The state lab first made the request in July, as the delta variant was becoming prevalent, but the number of samples has dropped off over the last few months, according to Toney.)
“We’re also looking to expand our capacity by bringing on new partners or establishing partnerships with other laboratories,” she said. “So if there were an instance where we had to ramp up significantly, we would have that in place.”
Better sequencing — in Virginia and nationwide — can offer essential information on how new variants behave once they’re circulating in a community. Omicron has attracted global attention for both the number and location of mutations along its genetic sequence, which have been associated with increased transmissibility and the potential to evade existing immunity through vaccines or previous infections. But Mathers said it’s still unclear if the variant will actually demonstrate those traits, or if it’s transmissible enough to outpace the delta variant.
“I don’t know if I’m a big fan of the level of hype it’s getting,” she said. “I think it’s a little early.” But expansive sequencing could help answer fundamental questions if the virus begins to circulate widely, such as whether omicron might be linked to more severe symptoms of disease.
“If you were to predict how this virus might behave, without much data, you would say it’s going to misbehave,” Mathers said. “But right now, we can’t say anything for sure.”
Our stories may be republished online or in print under Creative Commons license CC BY-NC-ND 4.0. We ask that you edit only for style or to shorten, provide proper attribution and link to our web site. Please see our republishing guidelines for use of photos and graphics.