As the U.S. gatekeeper for medical tests, how should the FDA strike a balance between making sure the public has options during a national emergency and making sure those tests actually work? The agency’s stumbles during the earliest stages of the COVID-19 pandemic hold some important lessons, according to a report from the inspector general of the Department of Health and Human Services (HHS).
The watchdog office reviewed the FDA’s embrace of “regulatory flexibility” and its use of emergency use authorizations during the first months of 2020, at a time when no coronavirus tests were available anywhere in the country.
Open doors allowed diagnostic and antibody tests to reach the market quickly, but—whether they came from well-intentioned yet inexperienced product developers looking to contribute to the global pandemic response or others simply chasing a gold rush for millions in public testing contracts—many tests of poor quality ended up providing inaccurate results, ultimately undermining efforts to stop the spread of the virus.
The inspector general’s report starts with the failure of the first diagnostic put forward by the Centers for Disease Control and Prevention (CDC). It received an EUA from the FDA in early February 2020, but, within days, public health laboratories began reporting that some of the supplied reagents could not pass verification steps to make sure they worked correctly.
By the end of that month, the FDA told labs they could proceed with a modified version of the test and skip the use of the faulty reagents entirely. Combined with a limited rollout to begin with, the delays led to weeks where no authorized alternatives were available, the HHS report said.
“Because FDA was only directly engaged with developers through its EUA process, FDA was slow to realize that testing within [public health laboratories] was far more limited than it had initially expected,” the inspector general’s office said in its report. “Regarding authorized tests, FDA communicates directly with developers, not with the labs implementing them.”
“As a result, FDA was unaware that testing within PHLs was very limited and may not have provided sufficient testing,” it continued. “However, FDA learned from media sources that far fewer PHLs were testing than expected. On February 21, when FDA requested clarification from CDC about the use of the authorized COVID-19 test, CDC ‘surprised FDA’ by reporting that only seven PHLs were using the test.”
In March, the FDA opened up the use of CDC’s lab kits to a wider number of labs, and by the third week of that month more than 10 million tests had been released.
The report said the experience with the CDC’s test “underscores the need for a national testing strategy” and that the FDA “needs clear, direct communication with the lab community during an emergency.” The agency concurred with those findings as well as with all of the report’s recommendations—including developing a range of EUA templates and guidance documents for test developers in tracking emerging infections.
As the coronavirus continued to spread across the U.S. in early 2020, the FDA “made calculated decisions to increase availability of COVID-19 testing, but these decisions often came at a potential cost to test quality,” the report said, including allowing test developers to submit evidence at lower standards when they were having trouble accessing clinical samples.
The FDA's moves ultimately led to a record number of submissions, with the agency's device and diagnostic center receiving more than 1,200 products of various quality for review by October 2020. That workload began to displace the regulator's efforts in other diseases, while staff were authorizing coronavirus tests at a rate of about one per day on average.
The first EUA for a commercial test was granted to Roche in mid-March and was handed down within 24 hours of the agency receiving the application.
The FDA also began allowing qualified labs to begin using their own tests in the clinic before any federal review at all, relying on them to self-validate. The agency instead required developers to notify the agency that they planned to submit an EUA request in the near future.
While many laboratories took advantage of the policy—as many as 245 by mid-May 2020, according to the report—relatively fewer commercial test developers opted for a similar avenue. One responder to the inspector general’s survey of companies noted that insurers wouldn’t cover a test without an EUA and that an authorization also granted certain liability protections.
And, while the FDA sought to focus its work on high-throughput and molecular lab tests for the coronavirus, it allowed developers of blood-based antibody tests to move forward with even more relaxed requirements and no EUA required.
This led to poorly performing serology tests flooding the U.S. market—including some that were simply inaccurate and others that falsely claimed they could diagnose active COVID-19 infections or carried a full FDA approval.
The agency began walking back that policy in multiple phases over the year, saying at the time that “flexibility never meant we would allow fraud,” and started intercepting illegitimate tests produced overseas at international mail facilities. By November 2020, the FDA had removed 167 serology tests from store shelves.
FDA officials have previously been upfront that the agency’s experiment with regulatory flexibility had unwanted consequences. In a February 2021 editorial published in The New England Journal of Medicine, the FDA’s device center director, Jeff Shuren, M.D., and the agency’s in vitro diagnostics chief, Timothy Stenzel, M.D., Ph.D., said in hindsight that the open-door policy for antibody tests was flawed.
“This series of events differed from our experience with commercially manufactured diagnostic tests,” Shuren and Stenzel wrote, adding that serological tests had previously played limited roles in past viral outbreaks and were rarely used in immediate clinical care.
“Knowing what we know now, we would not have permitted serology tests to be marketed without FDA review and authorization, even within the limits we initially imposed,” they said.
More recently, FDA Commissioner Robert Califf, M.D., related how the agency’s experiences during COVID helped shape its response to monkeypox while testifying earlier this month on Capitol Hill over the government’s response to the new and ongoing outbreak.
“One of the lessons from COVID was that when the gates were open, a lot of tests that were considered not so good got out there and we had to rein them back in,” Califf said Sept. 14.
This month, the FDA granted its first EUA to a monkeypox test developed by Quest Diagnostics—and said it does not plan to enforce certain regulatory requirements for tests developed by laboratories and used without an emergency green light.
But, unlike in the early stages of the COVID pandemic, this time the FDA is limiting this policy to antibody tests developed and performed by CLIA-certified laboratories operating under academic medical centers, where the test results will be used by professionals alongside other diagnostics for the direct medical care of a patient.
“This should mitigate the potential misuse of monkeypox serology test results while also fostering research from availability of data from serology testing of patients,” the FDA said in announcing the policy. “For example, these laboratories can test their patients daily or frequently over a period of time to monitor immune response and are most likely to benefit from the information learned from serology testing at this stage of the monkeypox outbreak.”