In a move it described as a step forward in getting more COVID-19 tests to more Americans more quickly, the Food and Drug Administration has provided developers with new Emergency Use Authorization guidance on the validation processes it expects for pooling samples and for screening of asymptomatic people.
As noted on the FDA's website FAQ page, there are no SARS-CoV-2 diagnostic tests currently authorized for the broad screening of asymptomatic individuals for COVID-19. While it has always been a healthcare provider's discretion to test asymptomatic individuals using EUA tests, Jeffrey Shuren, the director of FDA's Center for Radiological Health said in a statement that using COVID-19 diagnostic tests in this way is not the same as using such tests as a broad screening tool.
Shuren said the FDA recognizes that organizations may wish to conduct screening of asymptomatic individuals as part of a broader strategy to help ensure the safety of employees, patients, students, and others.
Likewise, a lab is typically free to conduct bridging studies with an EUA test in order to modify how it is used, but the FDA had not previously advised developers on ways to get a pooling indication on a test.
Pooling is a technique to conserve testing resources, the lack of which was a major problem during the height of the pandemic. It involves mixing aliquots of samples together and testing them with a single test. If the entire pool comes back positive, the individual samples can then be retested, but if the pool is negative, then each sample is also presumed to be negative.
Pooling dilutes the viral nucleic acids, however, so developers must validate that using this technique does not produce false negative results.
"The FDA has always been open to working with diagnostic developers on novel testing ideas, like asymptomatic screening and specimen pooling, and has encouraged all test developers to reach out to the agency to discuss appropriate validation approaches," Shuren said. "Today, the FDA is taking another step forward by updating templates for test developers that outline the validation expectations for these testing options to help facilitate the preparation, submission, and authorization under an" EUA, he added. "The steps taken today by the FDA further demonstrate how we are proactively working with diagnostic test developers to facilitate new approaches and get more tests to more Americans more quickly."
On its website, the FDA noted that the Centers for Disease Control and Prevention issued guidance relating to worker testing, suggesting organizations may want to conduct broad screening of asymptomatic individuals as part of a strategy to assure the safety of their employees, patients, students, and others.
"Broad screening using a highly sensitive test, especially given the asymptomatic testing pool, leads to the most accurate results, which is why the FDA has provided validation recommendations designed to establish high sensitivity for tests intended for broad screening in the updated templates," the agency said.
In the updated template for molecular diagnostic tests, there are two pathways to the asymptomatic and pooling indications – one for a previously unauthorized test, and one for a test that already has an EUA.
For an asymptomatic screening claim, FDA recommends in its template that the test be evaluated in a clinical study of the intended population. The number of patients should be enough to detect 20 positive samples, with a positive percent agreement, or PPA, with a comparator test of 95%, and negative percent agreement, or NPA, of 98%.
To add asymptomatic population screening to a test that already has an EUA, a post-authorization study may be appropriate, with a minimum of 20 asymptomatic positive specimens and at least 100 negative specimens. The FDA expectation is that PPA should be 95% or greater and NPA should be 98% or greater.
For pooling, developers with a test that has not previously been authorized should establish the pool size for the claim and perform a clinical validation study large enough to ensure at least 30 samples test positive with a comparator method. All samples must also be tested individually.
Adding pooling to an already authorized test requires a clinical study large enough to include 20 positive samples.
For both claims, the agency recommends developers collect samples at a minimum of three geographically diverse sites, but it will consider data from one or two sites in the context of an EUA. Archived, rather than prospectively collected, samples could also potentially be used.
In a virtual town hall with developers on Wednesday, Timothy Stenzel, director of the office of in vitro diagnostics and radiological health at CDRH, said that the agency "obviously" wants accurate testing for pooling and for asymptomatic people. "We want to make it clear that EUAs are required for pooling and they are also required if you want to make a claim about the ability of your test or assay to detect asymptomatic viral positive patients."
Labs running authorized commercial tests that want to add either asymptomatic testing or pooling can work with the commercial test manufacturer, Stenzel said. "We are encouraging kit manufacturers to submit an EUA amendment – if we can facilitate you working with a commercial manufacturer to add these additional claims, we will do our part to do that," Stenzel said. Labs can also submit their own EUA amendments if a manufacturer provides a right of reference document.
"It has been our experience that most, if not all, manufactures are very open to doing this for their customers," he said.
This story first appeared in our sister publication, Genomeweb.