Calling the current turnaround time for COVID-19 test results “unacceptable,” Gov. Ralph Northam announced last week that Virginia would join a consortium of other states to pursue a mass purchase of 500,000 rapid antigen tests — a relatively new technology when it comes to diagnosing cases of the virus.
The potential purchasing agreement is still preliminary, but a total of 10 states are now involved, from neighboring Maryland and North Carolina to Utah and Arkansas, according to Northam’s spokeswoman, Alena Yarmosky.
The intent is to use interest by the states to encourage wide-scale production of the new tests. Currently, only two companies have received approval from the U.S. Food and Drug Administration for the method, which can deliver results within 15 to 20 minutes. Antigen tests are often referred to as “point of care” because doctors can deliver results at the time patients come in for testing.
It’s a dramatically different timeline than the one for PCR (polymerase chain reaction) tests, currently considered the gold standard when it comes to diagnosing active cases of coronavirus. In Virginia, the average result for a PCR test from a commercial lab can take between seven to nine days (though some patients are reporting even longer waits). That’s a big problem when it comes to COVID-19, a disease that’s widely contagious — ranked somewhere between the common cold and smallpox, according to Harvard epidemiologist Michael Mina — and potentially deadly for certain patients.
“When we talk about tracing, it makes it so much more difficult when you don’t have immediate results,” Northam said at a news briefing last month. Like its partners, Virginia is hoping antigen testing can help to fill in those existing gaps. Yarmosky said last week that Northam “has had several conversations with [Maryland] Governor Hogan on how to collaboratively increase testing, given the lack of a national testing strategy.”
But as multiple states pursue the method as one possible fix for the current PCR backlog, public health experts are also urging caution over viewing antigen tests as the ultimate answer — or a possible replacement for PCR tests. And Northam’s administration hasn’t said exactly how they’ll be deployed.
“I definitely would not point to this as the solution that’s going to fix all of our problems,” said Dr. Danny Avula, the director of the Richmond and Henrico Health Districts. “This is a useful tool and will help solve a gap, but as a country, we have a much bigger supply and capacity issue that has to be addressed.”
First, what is an antigen test?
Since the beginning of the COVID-19 pandemic, three dominant types of testing have emerged: PCR, antibody, and now, antigen. PCR is considered the most accurate — a molecular test that’s usually conducted on samples taken collected on a long nasopharyngeal swab (some labs can also process nose and throat swabs, but the results are less reliable).
PCR tests can tell patients if they currently have the virus — unlike antibody tests, which analyze blood samples for proteins that build up to fight off the infection. An antibody test can tell you if you’ve been exposed to COVID-19 at some point in time, but can’t be used to diagnose a current case.
Enter antigen tests, the first of which was approved by the FDA in early May. The currently available models use nasal swabs to test for — of course — antigens, another type of protein found within or on the surface of the virus.
“What it’s doing is, it’s looking for bits and pieces of the virus,” said Virginia Health Commissioner Dr. Norman Oliver. A positive result on an antigen test can be used to identify an active infection. And results are fast — generally under 20 minutes.
That sounds great. What’s the problem?
Antigen tests just aren’t as reliable as PCR tests. It comes down to an issue of specificity versus sensitivity. Specificity measures how well a test can rule out healthy people who don’t have the disease. Sensitivity determines how well it can detect patients who are actually infected.
Both of the currently approved antigen test models, from companies called Quidel and Becton, Dickinson and Company, have 100 percent specificity, according to the manufacturers — meaning that a positive result is considered highly accurate. But the sensitivity is only 80 to 85 percent, which means that a negative result isn’t as reliable. “The basic idea that if you had 100 people who had the disease, an antigen test might only catch 80 of them, or maybe even a little bit less,” Avula said.
That’s a big problem when it comes to a virus as potentially dangerous as COVID-19. Oliver said he doesn’t consider antigen tests truly diagnostic because they can miss so many cases. Because the results aren’t as accurate, VDH doesn’t currently report them on its public COVID-19 dashboard, though Oliver said antigen results do make up roughly 17 to 18 percent of Virginia’s “probable cases” (a total of 4,194 as of Thursday).
Antigen testing doesn’t currently appear to be widespread across the state — VDH spokeswoman Marian Hunter said Wednesday that the agency has received a total of 1,151 positive and 198 negative results.
The department is currently in the process of developing a new data portal to report antigen test results separately. Oliver said that right now, they’re mostly being conducted at private practices where doctors can use their clinical judgement to help interpret the results. Dr. Sandy Chung, a Northern Virginia-based pediatrician and president of the Virginia chapter of the American Academy of Pediatrics — who conducts antigen tests at her own office — says there’s leeway in that setting to confirm a result through a PCR test, or reschedule another antigen test a few days later when a patient may have developed a higher viral load.
“That’s something we’ve done traditionally for things like flu and strep,” she added. “But the benefit is that it’s positive right away, I can tell you what to do next.”
Does that mean that doctors will get the tests?
It’s currently unclear how exactly the 500,000 tests will be deployed if the state makes the purchase. Yarmosky wrote Wednesday that “VDH will conduct a scan of current capabilities and needs to determine priorities for deployment.”
But given the limitations of antigen testing, Avula said there are multiple situations where it shouldn’t be used. Those include any mass screening event, including colleges or K-12 schools using the tests to determine who’s safe to return to campus.
One possible use is in nursing homes and assisted living facilities. Last month, the U.S. Centers for Medicare and Medicaid Services announced it would deliver rapid antigen testing devices to every licensed nursing home in the country — part of a sweeping new mandate requiring facilities to test residents and staff on a weekly basis. Avula said antigen testing is most useful in high-risk situations where rapid results can guide quick decision-making. Those include nursing homes with symptomatic individuals, where fast results can help administrators decide which residents to isolate.
But Amy Mathers, an infectious disease physician and associate director of microbiology at the University of Virginia, was more skeptical. While she agreed the tests might be useful to quickly test symptomatic residents, she said the risk of false negatives made them a highly unreliable way to screen an entire facility, or to test residents before they’re admitted to the facility. She said the same is true for other medical settings, including hospitals using antigen testing on admitted patients or those coming in for elective procedures.
“If you’re using it as a diagnostic in the hospital, I would not want to use a low sensitivity test,” Mathers added. “The biggest fear that I have is you’re going to have false negatives in very high-risk populations.”
Resources are also a question
The lag in PCR testing is linked to shortages of reagents, machinery and other components needed to make the tests work. But there’s no indication that antigen tests are any more widely available. Chung said there’s currently a waiting list for the devices needed to run them. Quidel spokeswoman Jeannine Mason said the company is currently prioritizing shipments to nursing homes and “plans to produce upwards of 1.8 million tests per week by September.” But if the multistate purchasing agreement is finalized, each state would commit to buying 500,000 tests — a total of 5 million.
“Current production of antigen tests are certainly below needed volumes, which is why we are banding together to demonstrate demand,” Yarmosky wrote in an email on Thursday. “The goal is to influence manufacturers to scale up their supply accordingly.”
When it comes to antigen testing, Mathers said supply is important because the tests are most useful from a public health perspective when they’re widely deployed. Even if antigen tests are less reliable than PCR, they turn around results more quickly and may be more accurate in cases of patients with a high viral load. If the tests are conducted on a large scale, it increases the likelihood that potential spreaders will learn they have the disease more quickly and isolate before they can infect anyone else.
“The issue is, you have to have enough testing, and then you would want to deploy it in high-risk settings where people are more likely to transmit the virus to each other readily,” Mathers added. “Like, they live together in a building.”
But even if antigen tests are widely deployed, they won’t totally eliminate the need for PCR testing. VDH recently issued interim recommendations for antigen testing in nursing homes, which advise facilities to confirm negative results with a molecular assay. Oliver said there are other cases when results should be confirmed, particularly if symptomatic and high-risk patients test negative.
“But then that’s two tests for one person and right now we don’t have enough for one test for one person,” Mathers said. It’s an issue that also comes down to cost. LeadingAge, a national association of nonprofit aging services, recently estimated that nursing homes with 100 or more employees could spend at least $14,250 a week on PCR retesting.
“The whole value that we’re banking on is the value of the positive,” Mather added. Oliver acknowledged that antigen tests aren’t perfect, but said adding them to the state’s COVID-19 response would still be an improvement over the current testing situation.
“In the context of not having adequate numbers of PCR tests, a less sensitive test is still better than no test at all,” he said.