The latest U.S. FDA town hall for diagnostics included the usual technical questions about test validation, but there are some frustrations among test developers regarding turn-around times for emergency use authorizations (EUAs). Nonetheless, Tim Stenzel, director of the Office of In Vitro Diagnostics and Radiological Health (OIR) at the FDA, said the surge in staff assigned to review EUA filings has worked to some benefit, claiming that the agency has rendered a decision in connection with 65 applications in the two weeks leading up to the Dec. 9 town hall.
Stenzel said at the outset that the OIR is on its second surge of staff to help process EUA applications for the pandemic, a timely move given what he described as “an increasing pace” of received filings. He touted the OIR staff’s pace of work that allowed them to process 65 EUAs for serology tests, but acknowledged that non-pandemic test applications may still be queuing up as the agency grapples with this latest surge of confirmed diagnoses and hospitalizations.
Off-label use of test for pooling possible
Among the recent authorizations are those for pooling of samples, which the agency sees as particularly useful in identifying asymptomatic patients who nonetheless would test positive for the SARS-CoV-2 virus. This is particularly important in the context of congregate living settings, such as colleges and universities. Any tests previously authorized for asymptomatic patients can be used in a pooling regime even if not authorized specifically for pooling, Stenzel said. This applies so long as the associated EUA does not spell out any other restrictions that would make pooling a second off-label use, such as an EUA that explicitly authorizes the test only for symptomatic patients.
Stenzel said the agency continues to prioritize point-of-care (POC) tests, home collection kits and home testing systems, including panel assays that test for one or more influenza variants as well as the SARS-CoV-2 virus.
In the case of the use of software to classify assay results, Stenzel said that his office is not typically asking for as exhaustive a software validation as it would under ordinary circumstances. Instead, developers should focus on the most critical components of the software prior to authorization, a subject that should be taken up before the EAU filing is complete.
The objective “is to have no obvious errors that could lead to false or inaccurate results” at the launch of the algorithm with the test, Stenzel said. Precisely what this portends for software testing depends on several factors, such as whether the test is intended for use in a clinical lab vs. a physician office or home use. Rather than exhaustively validate the software prior to presentation of an EAU submission, the sponsor might want to consult with the agency, he recommended.
Any saliva collection devices that are cleared or approved for non-COVID uses, such as for genetic testing, cannot be bootstrapped to a test authorized for saliva samples just by virtue of the device’s non-EUA marketing authorization. Toby Lowe, associate OIR director, said any test developer that wants to include the use of a tube that does not offer RNA preservation, such as a commercially available, sterile tube, would be liable for demonstrating the use of that tube with the developer’s test.
Conversely, a tube that is marketed with claims of RNA preservation has a more demanding premarket path in front of it. Lowe said that any claims made by the tube regarding use with a test for COVID-19 would have to demonstrate RNA preservation if that manufacturer wants to make the tube generally available to test developers.
No ban on OUS test evaluation
The FDA will accept test performance evaluations for point-of-care (POC) rapid antigen tests when those evaluations take place outside the U.S. (OUS), but Stenzel said the agency would like to see at least some of these evaluations done in the U.S. Another preference for the agency is that the instructions for use be printed and evaluated in both English and Spanish, and the agency would like to encourage developers to evaluate their test performance in minors as well as adults.
Stenzel said the agency is not entirely convinced that pediatric users will necessarily have a more difficult time properly executing a test than adults, but that there is nonetheless a preference for obtaining such data. One of the confounders regarding site of test conduct is that the definition of the term “point of care” is not the same in all nations. In some European nations, the term is at times used to describe a test executed in a lab, depending on the circumstances, Stenzel said.