Explainer: What is antigen testing?

By Shawn Cunningham
© 2020 Telegraph Publishing LLC

Antigen testing for the coronavirus has received a lot of attention in Vermont, when in early July more than 50 positive results came from a clinic in Manchester.

This type of rapid testing was approved by the Food and Drug Administration in May on an emergency basis, and seemed to hold a lot of promise for identifying those infected with the virus and isolating them to prevent spread. It could also serve as a basis for contact tracing more quickly than the standard polymerase chain reaction (PCR) test, which must be processed in a lab and can take days or longer to get results.

Until recently, the antigen test was considered to be on par with the PCR test for identifying those who are positive for the coronavirus. Where the antigen test has problems is in identifying false negatives. The antigen test finds false negatives at a rate of 12 percent where the PCR test has a six percent false negative rate.

Early in July, Manchester Town Manager John O’Keefe and Town Health Officer Dr. Tom Sterling announced on Facebook that tests performed at the Manchester Medical Clinic had resulted in dozens of people testing positive for Covid-19. That led to the Vermont Health Department, using the PCR tests, to conduct days of testing at Flood Brook School in Londonderry and at Riley Rink in Manchester, produing more than 1,500 negative results. And nearly all who tested positive with the antigen test at the clinic subsequently tested negative under the PCR test.

In Facebook posts around the time of that testing, Manchester Medical Clinic co-owner Dr. Janel Kitteridge-Sterling repeatedly asserted that the antigen testing platform the clinic was using is reliable for revealing coronavirus positives saying, “If you test POSITIVE, you are POSITIVE!” Repeated phone calls to Kitteridge-Sterling have not been returned.

While the large testing sites set up by the state came up negative for an outbreak, officials took the antigen tests seriously and were careful to say that it was not a false alarm.

With the FDA and Quidel, the San Diego-based manufacturer of this antigen test, investigating the circumstances around what appear to be false positives, it seemed like a good time to ask what is an antigen test, how is it done, how does it differ from the PCR test used by the state, and what might explain the divergent results in Manchester.

The process

An antigen is a molecule that triggers an immune response from the body. The antigen is on the outside of a pathogen like the coronavirus and the immune response to it produces antibodies to fight the pathogen.

Tamara Ranalli, a scientist who is also vice president of North American marketing for Quidel, told The Telegraph in a recent interview that the test uses lab-created antibodies that will bind to the virus’s antigen. These antibodies have a fluorescent marker that can be detected by Quidel’s analyzing machine, known as Sofia. When Sofia finds that a fluorescent marker is present, that indicates the virus is also, thus a positive test for Covid-19.

The procedure is relatively simple, but there are some critical timing steps.

• Swab the patient using a direct nasal swab or the longer nasopharyngeal swab.
• Place the nasal swab into a liquid buffer that is in a tube – everything is provided in the kit.
• Swirl it around and let it sit for a minute. That allows the virus to be transferred into the liquid.
• Transfer the liquid containing the virus onto a ‘cassette’ which which will go into the machine.
• Incubate for 15 minutes and put the cassette into the Sofia, which will read positive or negative.

Ranalli says that using the machine is more accurate than just looking at the sample. “Doing this visually can lead to subjective judgments,” she says, adding, “but the machine can detect the fluorescent light vs. the background precisely at very low levels.”

“This is not a brand new machine, this is not our first rodeo. We’ve been in the respiratory diagnostic market for 30 plus years,” says Ranalli. “We had some of the first visually read rapid flu tests, and released the Sofia flu tests six or seven years ago.”

What’s the problem?

When the large number of positive tests appeared in the Londonderry-Manchester corridor, the Vermont Department of Health did not report them on its daily Covid 19 Dashboard, leading to questions and consternation in Southern Vermont. Dr. Mark Levine, Vermont’s health commissioner, explained that, to be added to the state count, the positives first would have to be confirmed with a PCR test. Then the retests of the Manchester positives came back almost entirely negative. The number of days between the first test and the retest varied, but Levine has said that the time elapsed between the tests was inconsequential.

On Tuesday, July 21, Ranalli said that Quidel had only heard about the situation in Manchester that morning and that a quality assurance team was beginning a “proper investigation.” She said it was too early to speak about the investigation, but that she could address the clinical data in that “…we’ve matched up 100 percent of the time against molecular (a.k.a. PCR) when we were testing patients within the first seven days from symptom onset. We never picked up a … false positive – compared to PCR in our clinical studies. We’re all surprised that they would have a discrepancy like that.”

A week later, on Tuesday, July 28, a spokesperson for the Food and Drug Administration told The Telegraph that the “FDA is aware of the issue and takes seriously our responsibility to monitor device safety and performance. However, the FDA, by policy, does not discuss possible or ongoing investigations.”

According to Levine, “They’ve both (Quidel and the FDA) taken a lot of data from the machine – lot numbers, serial numbers, etc. and we’ll just have to wait on understanding exactly what they find.”

Ranalli noted that there are a number of situations that could lead to a false result and that Covid-19 is a brand new disease with myriad symptoms that not everyone gets that the scientific community is just learning about as the pandemic continues. “Some people don’t develop respiratory symptoms, some people develop GI issues,” says Ranalli. “So this is what’s different about (Covid-19) from flu: Flu is primarily a respiratory disease. Now we’ve come to find out that Covid can attack all different parts of your body.” That leaves the question of whether you are testing the right place at the right time and when it is retested whether the virus is still present.

Ranalli, who trained as a scientist and has a PhD in biochemistry, added, “I would want to evaluate things side by side to see what’s going on. I would like to get to the root cause of this. If there is a problem, we would want to know what’s going on.”

“If you separate (testing and retesting), even if it’s a couple of days, especially if (people) were asymptomatic, you may not have a viral load high enough … to detect via a molecular (PCR) test. If you are testing them three days later run both tests at the same time. If both come out negative they do not have virus present. It could have potentially been a false positive but they also could have cleared the virus by then.”

On Tuesday, July 28, Levine agreed, saying, “The time interval between should obviously be as short as possible because you are trying to confirm a test … not days down the road.”

‘A role in the future’

Levine has been cautious in his remarks about the use of antigen testing. “It’s hard to be overly critical of choosing to have this test available,” said Levine at a press conference earlier in May, “Manchester Medical took that upon their shoulders in choosing this test and they’ve shown an understanding of the test characteristics and it’s an approved test. The problem is that it’s been approved very recently and this is why so many questions are coming up now.” And he emphasized, however, that this was “not a false alarm.”

On May 24, Levine said he believes there is an appropriate place in the state’s testing strategy for antigen tests and that he is concerned that people will lose faith in this machine that does the testing. “I do think it has a role in the future.”

And on Tuesday, he reiterated his support: “I actually feel very positively about the equipment and if the problem isn’t too severe, I feel they could play a significant role in testing in the future.”

Asked if he felt the antigen test has a part to play in rapidly finding coronavirus in a school setting, Levine noted that the test seems to work best in the first five days of symptoms and in an area of high prevalence of the disease. The rapid test – in those circumstances – could allow quick decisions to be made about contact tracing and isolation. But he said that a school setting would not present those conditions and so it would be less useful in schools.

While the investigations of the testing discrepancies continue, we may not see an explanation for some time. Quidel says it is continuing to work with the FDA and another visit to Manchester is in the works. Neither was able to say when answers could be expected.