New questions are emerging about how best to use COVID-19 tests as the country grapples with a surge of breakthrough cases in vaccinated individuals caused by the omicron variant.
Some evidence suggests omicron may produce more false negative results — or at least seemingly delayed positive results — in rapid antigen tests, with medical experts calling for more studies.
With omicron’s presumed shorter incubation period and increased transmissibility, detecting the virus quicker is becoming more important.
The issue is primarily with rapid antigen tests, which are less sensitive than traditional lab-based PCR tests but generally considered likely to identify people when they’re contagious. Preliminary evidence points to the possibility that rapid antigen tests can now miss a positive case in the first few days of infectiousness.
“The key idea here is that people are becoming symptomatic earlier after exposure and getting tested at that point, when their viral levels are still low enough to slip under the radar,” said Robert Colgrove, a physician who chairs the Diagnostics Committee at the Infectious Diseases Society of America. “This is less of an issue for PCR since it is more sensitive.”
A study of 30 people published in the preprint journal medRxiv found that rapid antigen tests lagged behind saliva-based PCR lab tests by roughly three days in identifying people with infectious levels of omicron.
The study’s sample is extremely small and has not been reviewed by experts yet, but the data are the latest addition to a growing body of anecdotal evidence that the new variant could spread more quickly in the throat than the nose. In recent weeks, social media users have been sharing positive test results from using a throat swab on their at-home antigen tests, which the tests are not authorized for, after testing negative with the appropriate nasal swab.
“We need a comprehensive study done of just how well those lateral flow tests pick up infection status throughout the course of being infected,” said Michael Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota and a former member of President Joe Biden’s transition team.
Studies are underway at the National Institutes of Health after the Food and Drug Administration released a statement in late December saying rapid antigen tests had reduced sensitivity to omicron.
Saliva vs. nasal samples
One hypothesis, although some experts dispute its premise, is that saliva samples may be better suited to detect omicron in rapid antigen tests since the virus is believed to show up sooner in saliva than in nasal samples.
While some PCR tests use saliva samples, the FDA has not authorized any rapid antigen tests for use with saliva, despite the throat swabbing featured on social media.
Bruce Tromberg, director of the NIH National Institute of Biomedical Imaging and Bioengineering and head of the agency’s RADx initiative to accelerate more COVID-19 tests, said the agency is conducting experiments to compare nasal and saliva samples in both PCR and antigen tests.
He told CQ Roll Call he expects more data from partners at Emory University to be released in the coming weeks.
While several scientific experts interviewed by CQ Roll Call generally approved of using saliva for rapid antigen tests, pending approval from the FDA, Tromberg threw cold water on the idea because he said the results are often inconsistent. He's overseen by acting NIH Director Lawrence Tabak, who literally “wrote the book” on the topic, Tromberg said.
“There is no rapid antigen test that has been authorized for saliva, and that's no accident,” he said. “It’s because the rapid antigen tests, the limits of detection are just not good enough to detect unamplified virus in the saliva.”
Additionally, saliva makeup changes based on what part of the mouth or throat it’s collected from, he added.
“If you swab in your cheek, or if you swab under your tongue, or if you swab at the back of your throat, you're activating different glandular structures and there is just a different composition,” he said.
The Centers for Disease Control and Prevention faced blowback in December from the public and medical groups after shortening recommended isolation guidance for asymptomatic COVID-19 cases to five days without requiring a negative test. The move was widely seen as an acknowledgment of the testing shortages the nation continues to face.
CDC Director Rochelle Walensky defended the decision by noting that the tests are not authorized specifically as a one-time pass for leaving isolation. For people without symptoms, they’re typically authorized to be used serially over several days.
When she was a Harvard Medical School professor in 2020, she advocated for the use of saliva antigen tests in surveillance screening rather than more-sensitive PCR tests, although she acknowledged at the time that saliva tests were missing up to 30 percent of cases.
It’s not clear yet how well the at-home antigen tests are catching omicron cases on the tail end of their infectiousness. In the 30-person study, one person had a false negative while still harboring presumed contagious levels of the virus. But the study was also done by comparing antigen test results with the number of amplifications it takes to recognize the virus on a PCR test, which only approximates infectiousness.
“It's still not a perfect marker, particularly with people who have some preexisting immunity,” said Celine Gounder, a clinical assistant professor of medicine and infectious diseases at New York University's medical school and a former Biden transition team member. “It's not going to be a perfect marker of infectiousness, and I think to have that kind of data would be really helpful here.”
Michael Mina, a former Harvard professor and vocal antigen testing advocate who serves as the chief scientific officer at eMed, said rapid antigen tests are performing comparatively well to PCR as they did with previous variants. People should quarantine once they become symptomatic, he said, then test the following day and again two to three days later.
“It's kind of the same story, where before we said wait a few days after exposure to take a test,” Mina said. “Now we're pretty much saying the same thing, but just know that you’re going to become symptomatic potentially within a day after exposure now because your immune system kicks in so early.”
Mina also said there are some “worrying signs” that people are transmitting the virus at lower viral loads, which could potentially undercut PCR tests too. But Tromberg cautioned that viral load is difficult to calculate, especially since the antigen and PCR tests detect different parts of the virus.
Sanjat Kanjilal, associate medical director of clinical microbiology at Brigham and Women’s Hospital, has seen no unusual uptick in the rate of false negatives with PCR tests thus far and said a longer isolation time or a PCR test should be used to clear people from isolation, even though PCRs can test positive long after a person is no longer infectious.
But people could still have high viral loads on day five, given the earlier onset of symptoms, he argued, and antigen tests are 20 percent less likely than PCR tests to pick up cases even then.
“So you’re willing to miss 20 percent of people that may have potential high viral loads going back to work and just leading to more transmission,” he said. “I think that's insane, personally.”
Melissa Miller, director of the clinical molecular microbiology lab at the University of North Carolina at Chapel Hill School of Medicine, fears the public will lose sight of the second part of the CDC’s guidance that recommends people wear a mask for five more days following isolation. People should still isolate for the full 10 days if possible, Miller said.
“I don't sense that that is the message that is coming out,” she said.