Rapid COVID-19 antigen tests demonstrated advantages in two new studies, but laboratory scientists are cautious, given the small sample size and lack of details in news reports. These experts engaged in a deeper discussion on clearing the confusion around COVID-19 testing at MedTech Strategist’s Global Investment & Partnering Virtual Summit (ongoing through Nov. 20). The full discussion is recorded and available to conference attendees.
The state of diagnostic and surveillance testing for COVID-19 is evolving and confusion is rampant, with even experts disagreeing on important details, in part because the science is still evolving and data is limited. Two new studies, albeit very small, support a growing contention by a group of highly visible and increasingly vocal experts that rapid antigen testing, especially when used for surveillance of the general population, has the potential to be a game-changer for re-opening society.
One study, described in the New York Times on October 15, indicated that Abbott Laboratories Inc.’s point of care $5 antigen test, BinaxNOW, is nearly as effective at diagnosing COVID-19 as the gold standard, PCR, but much cheaper, faster, and simpler to use. The study, conducted by scientists from the University of California San Francisco (UCSF) and community groups in San Francisco, is one of the first comparisons of a rapid COVID-19 assay and PCR in a real-world setting (see box).
The researchers tested 878 commuters in San Francisco, using both PCR and BinaxNOW, which received an FDA emergency use authorization (EUA) in August 2020 for use by healthcare professionals in point-of-care and near-patient settings. While not the first rapid POCT to get to market, it is the only FDA-EUA condoned test to date that does not require any equipment for analysis and result reporting, making it ultra-portable and easy to use on a mass scale.
Of those tested, only 26 individuals were identified as positive by PCR, and, of those, BinaxNOW picked up 15. But, interestingly, the rapid test detected almost all of the positive cases with high viral loads. This could be important if, as some scientists theorize, people with high viral loads are more likely to be infectious. No one has proven that however, and studies designed to do so are notoriously difficult to conduct. Moreover, the article and researchers provided few details on some important variables, such as type of nasal swab and the cutoffs used to define ‘high’ viral loads.
Another recent study, conducted by researchers at BD, which makes a rapid antigen platform known as the BD Veritor System, was mentioned briefly in The Times article and also sheds some, but limited, light on the utility of rapid antigen tests. Published as a preprint in Medrxiv on October 5, the BD investigators found that rapid positive COVID-19 antigen-based testing could be a more accurate predictor of an individual’s potential to transmit the disease than PCR (see box).
BD received an FDA EUA for its Veritor System for Rapid Detection of SARS-CoV-2 antigen assay in July–like the Abbott test, it produces results in 15 minutes and can be use in a variety of settings by healthcare professionals, but unlike the Abbott test, it requires use of a small (cellphone sized) instrument; BD has 25,000 of these already on the market for other diagnostic applications.
In this study, the Veritor demonstrated a higher concordance of antigen-positive results for rapid detection of SARS C0v2 antigen test with cultured, infectious virus than did RT-PCR. “The correlation between SARS-CoV-2 antigen and SARS-CoV-2 culture represents a significant advancement in determining the risk for potential transmissibility beyond that which can be achieved by detection of SARS-CoV-2 genomic RNA, the researchers wrote.
The New York Times article was published just days after a recent panel of leading laboratory experts sorted through the thicket of COVID-19 testing for the ongoing, virtual MedTech Strategist Global Investment & Partnering Virtual Summit 2020. Their discussion is available via recording to attendees of the summit, which runs on-demand online through November 20. The speakers don’t claim to be comprehensive, and they avoid rehashing testing basics, some of which by now, for the first time ever in history, perhaps, are ingrained in our collective public conversation. Rather, they offer real-time perspective, as experts who have been immersed in laboratory medicine for years at operational, clinical, or strategic levels. Among the topics they cover are the pragmatic challenges of testing related to gaps in interoperability of data, international standards around testing methodologies, and US test manufacturing capacity, along with the more visible issues of delays in test result reporting and the persistent problem of interpreting negative versus positive test results.
As a testament to how fast the field is moving—even for those closely clued into its dynamics—the subsequent news of the Abbott and BD data led to a lively email exchange among our panelists, who made the following observations about the status of our ability to measure infectiousness, which is after all the key information we need to contain the spread of COVID-19.
Mara Aspinall, Managing Director, Health Catalysts Group and a professor at the College of Health Solutions, Arizona State University: “Absent direct data to the contrary (there is none today), it is reasonable to adopt the working hypothesis that viral load in a patient is a good proxy of their ability to transmit it. This is common to all infectious diseases to some degree or other—although the actual mechanism of transport varies greatly—breath or bodily fluids, etc. But, generally speaking, the more virus there is, the more likely infectious doses of the virus will be exhaled into the area of other individuals. There will not be any international standards on this anytime soon—[there are] too many variations of tests and materials—head to head comparisons are required and they do not exist yet…” Later in the exchange, she explains, “Knowing that a virus is present does not mean that an individual will suffer from the disease. Each different virus has a typical infectivity capability (sometimes called a Q value)—it takes more than one virus to infect a cell, and the number varies a lot by disease. For COVID-19, it seems to take from 1,000 to 10,000 virus particles to infect one cell (for comparison HIV requires 10 million, polio requires 30-50.) This number will also vary by individual, hence why we do not have an absolute number for this infectivity.”
James Crawford, MD PhD, SVP, Laboratory Services, Northwell Health System and Chair of the department of pathology and laboratory medicine at the Zucker School of Medicine at Hofstra University: “The way I parse my thinking out is: If an individual is symptomatic, then the rapid PCR and rapid antigen tests fall under the FDA EUA ‘for use in individuals within seven days of onset of symptoms.’ If the individual is symptomatic, that person shouldn’t be out and about anyway. For the public health activity of screening asymptomatic persons for whether they are infected (and therefore, potentially infective), the FDA EUAs for rapid tests are not approved for this use. So while our society is rapidly galloping towards use of rapid testing for screening of asymptomatic persons, it is off label, however justified.”
He goes on: “The risk of false-negatives in a population health setting is undetected transmission. Given the need for mass screening, this is a tolerable risk, but we must continue to remember that rapid tests use for screening of asymptomatic persons is off label. For the practice of medicine, the risk of a false negative test is much higher, since there may be direct harm potentially emanating from failure to detect the presence of SARS CoV-2—hence the need for the much more sensitive laboratory-based PCR tests. And the national supply for laboratory-based PCR testing remains constrained, with woeful under-prioritization of the near-patient laboratories (hospital based and other in-system clinical laboratories), which need these diagnostics. The bottom line is that there are differing use cases for both laboratory-based PCR testing and for rapid testing. We must remain vigilant in how these tests are matched to the use case.”
Jeff Andrews, MD, Worldwide Clinical Medical Director for Women’s Health & Cancer & Molecular Diagnostics, BD Life Sciences, and a co-author of the BD study: “The UCSF study is interesting, but too little is known from the New York Times article to evaluate it further. The study hasn’t been published or presented, nor has it been peer reviewed. Information is lacking about the types of swabs that were used to collect the samples, which PCR test was used, and how the cutoffs were defined to identify people with high viral loads. What I glean is that they used PCR cycle time (CT) values and arbitrarily chose some cutoff and rejected individuals above that figure, but what’s missing is the evidence that the chosen cut off was correct from a culturability or infectiousness perspective. Finally, whatever PCR method they used would need to be calibrated to the as-yet-not-established international standard. Tim Stenzel [MD] (director of the FDA’s Office of In Vitro Diagnostics and Radiological Health) keeps saying that FDA is observing an international collaboration to develop an international standard that would allow us to convert CT values between PCR tests and to correlate CT values to infectiousness and culturability…but we do not currently have those standards.”
He continues: “Does culture align with infectiousness? Theoretically, yes. There is indirect evidence that antigen test negativity is consistent with non-infectiousness. Nonetheless, when testing a patient with symptoms consistent with COVID-19, a negative antigen test is considered presumptive and the patient has a follow-on RT-PCR test for confirmation.
In order to prove infectiousness, however, we would need to follow a large number of infected people prospectively, and all of their contacts, and swab daily for viral load, antigen, and culture, to see if they infect others, and then try to figure out whether they infected someone on a day when culture was negative but PCR was positive—this would be a very difficult study.”
Trial MyStrategist.com and unlock 7-days of exclusive subscriber-only access to the medical device industry's most trusted strategic publications: MedTech Strategist & Market Pathways. For more information on our demographics and current readership click here.