If the SARS-CoV-2 virus has achieved anything useful in the world of in vitro diagnostics, it’s that the associated pandemic has shone a bright and unsparing light on the respective merits of diagnostic and surveillance testing. Harvard University’s Michael Mina, an assistant professor of epidemiology, was one of several academic researchers who took up the gauntlet yet again in opposition to what they characterized as a gross misunderstanding of the respective roles of these types of tests, a misunderstanding they said must be addressed if the pandemic is to be corralled.
The latest leg in this long-running dispute began with an editorial in The BMJ, formerly known as the British Medical Journal, which took up the use of lateral flow tests (LFTs) by U.K. government agencies as a method of screening for the virus. The authors, Angela Raffle of the University of Bristol and Jonathan Deeks of the University of Birmingham, blasted the U.K. government for deploying a “society-wide ‘Moonshot’ testing” program before a thorough field evaluation of the underlying tests had been completed. Among the tests used in this program was an unidentified LFT by Innova Medical Group Inc., of Pasadena, Calif., which recently reported that its test can reliably detect at least two of the novel variants of the SARS-CoV-2 virus.
U.K. described as ‘a mess’ after pilot testing program
Deeks and Raffle further alleged that the surveillance testing campaign was conducted with the aid of “selective emphasis of unrealistic performance estimates” for LFT testing, a practice they allege has created confusion and thus U.K. society is “now in a mess.”
Depending on the circumstances of the deployment of the Innova and other LTF tests, the sensitivity ranged from 79% to 58% among those who were symptomatic at the time of test administration, while sensitivity among those who were asymptomatic fell to as low as 50%.The authors said the “poor sensitivity” of the Innova test in those who are asymptomatic “should have been expected,” adding that the World Health Organization has taken the position that a negative result from a rapid antigen test should not suffice to eliminate quarantine requirements. They argued that the U.K. government “ignored key evidence and dismissed expert international advice,” concluding that the use of LFT tests represent “an unhelpful diversion from the important task of vaccination roll-out.”
The dispute seems to highlight a contrast between expectations and reality, something highlighted in the editorial response by Mina and several others in the Feb. 17, 2021, online issue of The Lancet. Mina, et al said that the debate over the U.K. government’s use of the Innova test “risks confusing policymakers internationally, and potentially stalling deployment” of these tests in other nations.
Mina has repeatedly challenged the conventional thinking about rapid assays in the months since the beginning of the pandemic. He pointed out in August 2020 that molecular testing handled via PCR is simply not suited to the task of surveillance testing, describing the FDA’s approach to the question as “an archaic view” of the use of LFT. In the editorial in The Lancet, Mina and his co-authors said that the repeated cycling of test material makes PCR a powerful diagnostic test, but that running a sample through 30-40 cycles might do nothing more than allow the identification of insignificantly miniscule copies of viral RNA, and indeed may yield only one copy of the viral genetic code. The problem remains, however, that traces of viral RNA may be detectable for weeks after the patient has cleared all whole viruses, a scenario that may capture those who were both symptomatic and asymptomatic at the time of testing.
The period of contagiousness is still pegged at 4-8 days, with nearly all contagiousness coming to a halt by day nine post-infection. The editorial notes that this characterization seems to map accurately onto the observed patterns of viral transmission, which typically peak at two days prior to the onset of symptoms and eases by day five after symptoms emerge. The issue with PCR testing is that it can pick up a case over a period of 22 to 23 days in most instances, thus demonstrating that most positive diagnoses via PCR are for those who are no longer likely to pass the virus to others. The authors said that RNA copies can continue to circulate inside the human body for months after infection, further detracting from the utility of PCR outside of its use as a diagnostic.
No standard understanding of cycle threshold
The authors characterized the criticism of the U.K. testing program as misinterpreting the data from an interim report on a pilot testing program conducted in Liverpool. They said that the associated pairing of PCR and LFT was conducted at a time when the epidemiological curve was declining in Liverpool, a circumstance that should have led the reasonable observer to expect that a non-PCR test would fare poorly against PCR in those who are asymptomatic and/or without any known exposure.
Mina and his co-authors said there is considerable confusion over viral load, infectiousness and the cycle threshold (Ct) – the number of amplification cycles needed to make the viral RNA detectable – used in PCR testing. They said that there is no standard understanding of the number of amplification cycles needed to isolate between 100,000 and 1 million RNA copies per milliliter of specimen, a volume below which viral cultures typically detect nothing and the risk of transmission is seen as low. The labs that processed the samples in the Liverpool pilot study may have used a Ct value of 25 or lower, but there are labs in some nations that use a Ct threshold of 30, and the lack of concordance on this point is yet another source of confusion.
The Innova LFT test picked up 19 out of 24 samples with Ct values lower than 20 and 10 out of 11 samples with Ct below 18. However, the test did not return similar nominal figures in a study at the University of Birmingham, a predicament engendered by the fact that Ct values in the Birmingham study ranged to 29 and higher. The conclusion, Mina said, was that the viral loads seen in the Birmingham study were considerably lower than in other testing sites elsewhere in the U.K., and thus the Innova test was “working as expected.”
Innova claims test worked as advertised
Robbin Xu, chief scientist at Innova, told BioWorld that in the U.K. Porton Down study, the company’s antigen test sensitivity was 94.2% (194 out of 206) in subjects with Ct values of less than 28. Sensitivity was 97.0% (130 of 134) in subjects with Ct values of less than 25, which demonstrates that Innova tests “were very sensitive for patients with high viral load, in line with expectations.”
Xu noted that the pilot testing program in Liverpool has been accompanied by erroneous information in the media regarding the accuracy and purpose of lateral flow testing, and that an interim report dated Dec. 23, 2020, demonstrated that the test “was working as expected, identifying at least two-thirds of those people with the virus who were likely to pass it on.” Xu noted that Iain Buchan, of the University of Liverpool, said the pilot effort provide a test to nearly half a million residents in the area, which identified more than 12,000 as highly likely to be infectious when they were unaware that they were infected. These individuals were obviously also unaware that they could have passed the virus on to others, but that the Innova test allowed them to isolate in a time frame that is not practicable with lab-based tests.
Xu reiterated Mina’s point about the variability between the volume of viral copies and the number of cycles needed to obtain that volume, but he said that the U.K. government was not under a false impression about the performance or utility of the company’s LFT test. “With frequent testing, the rapid antigen test can help break the chain of transmission, which is not possible with slower, expensive lab-based tests,” he said, adding that the company’s test performed as advertised.