The FDA’s response to the COVID-19 pandemic has been matched by device makers, but the ID Now molecular test by Abbott Park, Ill.-based Abbott Laboratories has been the target of recent criticism. Tim Stenzel, director of the Office of In Vitro Diagnostics and Radiological Health, said at a May 20 town hall meeting that Abbott has agreed to yet another study of the ID Now, the terms of which were under negotiation at the time of the meeting.
Stenzel said the agency receives several queries on a weekly basis, including on whether a serology test should combine the three known antibody isotypes for the SARS-CoV-2 virus. “The agency doesn’t really have a position” on which of the three isotypes is most important, he said, but the minimal performance criteria for serology tests have not changed in the past couple of weeks.
Abbott had pressed the argument that some recent studies of the ID Now were faulty, including the well-known study by researchers at the New York University Langone Medical Center. Abbott’s position was in part that the use of viral transport media (VTM) in conjunction with swab samples for the ID Now is problematic. This position was affirmed by Stenzel, who said the issues seen with the ID Now were at least in part associated with dilution of the sample when the swab is stored in VTM prior to processing.
ID Now label edited for ‘presumed negative’ result
Stenzel said the instructions for use for the ID Now had been modified several weeks ago to remove any mention of VTM, adding that the agency is in possession of some unpublished data suggesting a drop in sensitivity for the ID Now “at the very low end.”
“We have worked collaboratively with Abbott to adjust as we go along,” Stenzel said, including edits to the instructions for use and the intended use statements. One of the edits is to state that a negative result from the ID Now would henceforth be described as a presumed negative result, and thus a negative result that contradicts the patient’s symptoms should be checked against another molecular test for the virus.
The company “has agreed to a number of post-market studies” for the ID Now, Stenzel continued, adding that the two parties are negotiating terms for the endpoints for another such study. He said if the agency is able to determine the sensitivity of the assay in a controlled trial is at least 80%, “we feel like that test has a valuable place going forward in this pandemic.”
Stenzel said the agency is “looking at all sorts of helpful tests in this pandemic,” adding that if a sufficient sensitivity is achieved, that test may play a role in suppressing the pandemic, including point-of-care diagnostics that detect either a protein antigen or a nucleic acid, and which “return a result very quickly.” This assumes the test demonstrates a sensitivity of 80%, even if that POC test is less sensitive than a test conducted at a central lab, a position adopted due to the need to triage patients.
The FDA continues to see some variable performance from home saliva collection kits, and Stenzel said there are few data points suggesting any correlation between the outcomes of tests that are based on saliva and those using a nasopharyngeal swab. He said this does not suggest that saliva is useless as a source of testing material, just that the inherent variability is not yet well characterized. As long as overall performance is good, saliva is seen as adequate.
Stenzel reiterated the agency’s position that antigen tests – which can be reproduced in high volumes and thus may more useful than some other tests types – are relatively short on sensitivity. The FDA is amenable to return-to-work screening tests for those who are asymptomatic, and Stenzel said this assumes the subject of the test is confirmed to be asymptomatic. So long as the developer has gone through the EUA process, “it’s okay by us,” Stenzel said, although he added that there are few data regarding the performance of any of the existing test types in those who are asymptomatic. If a kit developer or lab would like to claim performance characteristics, the FDA would prefer that party contact the agency about the design of the study up front.
FDA open to the use of spiked dry swabs
The FDA may be willing to accept dry swabs that are spiked with positive material if the sponsor is attempting to study a test in an area where COVID-positive participants are difficult to come by. Stenzel said the University of Washington is working on a method to address this scenario, with some emphasis on validation of resuspension of dry swabs. There is a fairly efficient path forward to address this predicament, he continued, adding that one of the prime concerns for the FDA is that the method of resuspension be well characterized and perform reliably.
Stenzel said home-use, over-the-counter tests would be subject to the same performance characteristics that would otherwise apply for that test type, adding that usability studies will be required. “That’s a great dialogue to have,” he said of OTC sales of such a test, advising however that home use would require a different type of study than required for POC use. He offered no specifics, however.
One of the callers to the town hall complained that the FDA was focusing insufficient attention on a test being developed by a small local company, but Stenzel pushed back on the insinuation that the agency was neglecting smaller developers for their more massive brethren. “Everybody gets a fair shake with us,” Stenzel stated, adding, “that’s how we roll.”