August 31, 2022 | The relative merits and drawbacks of using antigen tests rather than molecular assays during outbreaks of emerging infectious diseases was the topic of a lively debate held at last week’s Next Generation Dx Summit in Washington, D.C. The team arguing against antigen testing ultimately won the debate by swaying a healthy percentage of the audience to their side.
On the winning team were Susan Butler-Wu, Ph.D., associate professor of clinical pathology at the Keck School of Medicine of the University of Southern California, and Linoj P. Samuel, Ph.D., head of clinical microbiology at Henry Ford Hospital. Their opponents, arguing in favor of antigen testing, were Richard Noel, director of North America marketing for LumiraDx, and Nicole Zitterkopf, vice president of the laboratory service line at Optum.
Moderating the event was Matthew Binnicker, Ph.D., consultant in the division of clinical microbiology and vice chair of practice in the department of laboratory medicine and pathology at the Mayo
Clinic. The debaters didn’t necessarily agree with the side they were arguing for, he notes at the outset.
In her opening statement, Zitterkopf points out that a screening type test is needed to break the chain of transmission at the community level. This means that at two to three R0, (pronounced “R naught”)—the average number of people who will contract a contagious disease from one infected person—daily testing would need to be done at a high volume and, as a practical matter, the method would have to be both inexpensive and easy to perform.
Molecular testing not only takes longer to do than antigen testing, Zitterkopf adds. It is also more challenging to implement by users who have “never even touched an instrument like this before.” Antigen tests are more forgiving as well as easier to roll out because they are simple and don’t require a lot of electronics. Test results are also available in less than 15 minutes and “anything longer is a nonstarter in the community.”
Testing should not be all about convenience, counters Butler-Wu. The contest comes down to what would work best in an emerging public health threat and “antigen testing is barely better than a flip of a coin.” It requires a “flaming amount of virus” to meet the performance bar of 80% sensitivity and 97% specificity set by the World Health Organization, she adds, and thanks to some creatively structured trials, some of the 150-plus antigen tests authorized for emergency use during the COVID-19 pandemic effectively had 0% sensitivity.
If testing is needed daily, compliance is bound to be an issue based on behavior during the current pandemic, Butler-Wu continues. To make matters worse, antigen testing is not one test but a two- or three-test regimen.
It is incorrect to say that someone with a low viral load can’t transmit disease, says Butler-Wu. “There is no difference in the Ct [cycle threshold] value of people known to be spreaders and people who are not spreaders. A test needs to be accurate, not just rapid.”
Antigen tests are not all created equal, and the perfect test does not exist, says Noel. Based on PCR results, people can test positive for up to 90 days, which would give it poor sensitivity during a 10-day contagion period. Antigen-based tests, on the other hand, perform best closest to infectiousness and better snag the true positives.
Most of the recent emerging outbreaks—Ebola, MERS, Zika, Nipah—have not seen the use of rapid antigen tests, says Samuel. That is because PCR tests are in fact more forgiving in settings where neither the optimum specimen type nor time to test are known.
An assay target is not always easy to identify, and PCR tests can multiplex easily enough, Samuel adds. Further, $5 antigen tests don’t exist, and the newest microfluidic ones have grown ever more costly. They may be “quick and cheap, but too often wrong,” he says, quoting from a 2020 article in Science.
The Food and Drug Administration (FDA) now recommends a three-test regime, but “if you keep repeating assays you get a lot of false positives,” he continues. That can represent a steep cost of adoption for a family of four, equal to if not greater than the cost of PCR testing, which is more trustworthy. Laboratory developed antigen tests that have a certificate of waiver from the Clinical Laboratory Improvement Amendment (CLIA) may be accessible and produce results in 20 minutes, but also provide a “false sense of security.”
Samuel kicks off round two of the debate addressing the top dangers of antigen testing, naming interpretation of test results and a false sense of safety that contributes to disease spread among family members and work colleagues.
In the real world, Zitterkopf counters, people tend not to do as they’re told (i.e., isolate) during the wait time for PCR tests results and are thus still driving infectious diseases processes.
The specificity level of an antigen test matters, concedes Noel, including false positives causing people to unnecessarily isolate. But PCR tests can also produce positive results in individuals far beyond the period of infectivity.
Zitterkopf fielded a question from an audience member about misuse of the word “screening” to describe antigen tests. She clarifies that the term is applicable in the context of a public health emergency where a high volume of testing is being routinely done before individuals unknowingly spread the virus to others.
Done at scale, the false-positive count with antigen testing is “astronomical,” Butler-Wu retorts. Moreover, sensitivity gets knocked down 30% to 40% with self-testing versus an assay done by a trained professional.
Noel makes the point that who has and hasn’t been infected with a disease like COVID is not known. PCR tests are “shots in the dark” and will find people, but antigen tests are done at “a more actionable point of time to impact behavior.”
It is not certain that “persistent positivity” will be the situation with viruses generally, notes Butler-Wu, furthering her case for molecular tests with high upfront sensitivity.
But for an outbreak that collectively impacts population health, rapid and low-cost antigens tests are what will drive human behavior, says Noel.
At Henry Ford Hospital, Samuel notes, turnaround time for PCR test results has routinely been under 24 hours. Rapid results were in fact commonplace on the front lines of the pandemic.
At Optum, Zitterkopf says, the wait for PCR tests results often exceeded 14 days. “At a certain point, in medically underserved areas and hot spots we were not breaking the chain of transmission.”
In response to a comment that undocumented test results dilute the value of rapid diagnostics, Zitterkopf says this is true only of home tests.
Rapid self-test kits were provided to nursing homes free of charge with no reporting requirement, which Samuel cites as a “huge issue.” Even if Ct values were being reliably tracked, it is questionable whether people would respond accordingly. “There is not enough focus on human behavior.”
Replying to an audience remark that the limit of detection with rapid COVID tests has no equivalency with the new variants, Noel counters that “no statistical difference” has been found with certain antigen tests.
In his closing statements, Samuel mentions the relative advantage of PCR testing in terms of better utilization of antivirals and limited bed space. Molecular tests also offer high consistency of case detection, while the quality of antigen tests varies by brand, patient population, and sample type. “How confident are we that no one in this room has COVID?” he rhetorically asks.
“Don’t blame antigens,” says Noel, citing highly sensitive antigen tests that can help “bend the curve” during infectious disease outbreaks. “It’s an optimal starting point.”
No evidence is possible or exists to back the future state Noel refers to, says Butler-Wu. “We need a better boat, not a bigger one.” The latest iterations of pandemic preparedness plans talk about molecular testing. “We already have the tool to do this. We have at-home molecular tests available now, and hospitals have acquired the equipment to do them.”
South Korea successfully flattened the curve on COVID-19 by committing government resources to the problem, including validating tests for manufacturers, Butler-Wu says. In contrast, the response in the U.S. resulted in a flood of Emergency Use Authorizations filings to an overwhelmed FDA.
“Why settle for less when you can have the best?” says Zitterkopf, referencing clinicians’ desire to give answers at the time of testing. She acknowledges that many “terrible” tests—both antigen- and PCR-based—were released at the beginning of the pandemic, highlighting the importance of validation against some sort of standard.
As for the nursing homes receiving the free diagnostic tests, Zitterkopf questions to whom exactly they were to report results. “They would have sat [in one public health office or another] for weeks.” Total control is simply unrealistic in the early stages of an outbreak, she says. Communities need to be responsible for their own mitigation activities, and antigen tests are the best tools to equip them for that task because they’re cheap, fast, and easy to access.
But, in the end, the case for molecular testing won over audience members with the winning team swaying minds against the use of rapid antigen tests from 12% pre-debate to 57% at its conclusion.