NEW YORK – With both interest in, and concerns about, serology testing for coronavirus running high, the US Food and Drug Administration on Wednesday laid out its still-nascent plan on how it will incorporate antibody-based tests to help tackle the COVID-19 pandemic.
Among other things, the agency offered insights into how it may validate commercial serology tests by allowing manufacturers to voluntarily submit them for analysis. The FDA also touched on its authority to deem the complexity of tests, and warned about the interaction between reagents in two commercial test products.
The FDA has laid out four regulatory pathways to support emergency testing for SARS-CoV-2, the virus that causes COVID-19, as well as for immune response indicating a current or past infection. Among the simplest of the pathways, so-called "Policy D," allows manufacturers to notify the agency by email regarding their intent to distribute a validated serology test. As long as the test has certain specific verbiage in its labeling, manufacturers can begin distributing it without having to submit a full Emergency Use Authorization application.
In the town hall Wednesday, Timothy Stenzel, director of FDA's Office of In Vitro Diagnostics and Radiological Health, offered further insights about validation of serology testing.
"We continue to see problematic claims and uses of serology tests, particularly those that have been listed on our [Frequently Asked Questions] page under pathway D," Stenzel said. "These devices are only for high-complexity labs; they are not to be used in moderate complexity labs or in point-of-care, near-patient settings, nor in the home."
So far, there are only four serology tests that have received EUA from the FDA — commercial tests from Cellex, Chembio Diagnostics, and Ortho Clinical Diagnostics, and a lab-developed test from Mount Sinai Laboratory.
The agency announced last week that it has formed an interagency voluntary validation program that is open to the more than 70 serology tests in Policy D, and Stenzel said Wednesday that the program has reached out to all the developers to invite them to participate.
"We have been receiving test kits from some of these developers and we believe testing will begin in the very near future," Stenzel said. "As soon as we can find a way to report results, either through EUA authorization or other means, we will make that available."
The agency is also working on a way to communicate the protocol by which tests will be assessed, and Stenzel encouraged developers considering the validation program to inquire for details.
"At a high level, we are trying to determine whether the results that come from these tests with patient samples can be relied upon and can have a level of accuracy that indicates that these are fit for use in the United States," he said.
The results can potentially be used for an FDA EUA submission, but the way they will be reported has not yet been decided, Stenzel said.
The agency is also looking at "an alternate pathway for EUA authorization when participants have participated in this program," and Stenzel said more details on that will be provided in the future.
The FDA has also engaged with multiple parties to stimulate development of reference panels that labs and developers can use to assess tests, Stenzel said, adding that commercial entities are now collecting patient samples to address this need. "As soon as such material is available for test development, validation, and verification, we will endeavor to make that publicly known on our FAQ page," he said.
Serology tests are particularly vulnerable to specificity issues that can produce false positive results, which can lead people to believe they are immune to COVID-19 when they were in fact still susceptible. The FDA is "looking for ways to additionally highlight these concerns," Stenzel said. "In a setting of very low prevalence, even a very specific test can have a number of false positives, and the positive predictive value is relatively low," he added, noting the prevalence of SARS-CoV-2 in the US is currently unknown.
The FDA has also coordinated with the Centers for Medicare and Medicaid Services, Stenzel said, so that FDA can deem the setting of use for a test in its authorization letter and paperwork.
"CMS has generously agreed to allow this process to occur," he said, adding that FDA now gives a letter, H for tests deemed to be high complexity, M for tests deemed moderate complexity, and W for tests deemed to be CLIA waived. "Tests that have come solely through pathway D are not available for this deeming," he also said.
Sampling spit, and a warning
The FDA authorized the first test that can detect SARS-CoV-2 in saliva samples this week, in a molecular assay from Rutgers University. The workflow uses a saliva collection device from Spectrum Solutions.
"Saliva remains a very interesting sample type," Stenzel said, noting that the Rutgers test had performance that was "quite excellent … in a comparison between either nasopharyngeal swabs or oropharyngeal swabs and saliva."
The test is for use in a healthcare setting under observation, and also uses "PerkinElmer extraction coupled to the Thermo Fisher Scientific RT-PCR reagent, so this is somewhat of a hybrid authorization," Stenzel said. In other instances, the agency has seen more variable results with saliva-based tests.
"We don't know exactly why. This could be a sample stability issue. But we clearly see that at least in one case, this works, whether it is a combination of particular devices or it's resulting from the use of a device that may stabilize the viral RNA immediately."
Stenzel also emphasized that the FDA has not yet authorized home collection or home testing for the coronavirus, but it is actively working with a number of developers and expects to authorize these in the future.
Stenzel also noted in the call that the agency is issuing a warning about two products in the COVID-19 testing workflow that have a potential for a dangerous interaction and should not be used together.
Specifically, the PrimeStore MTM media from Longhorn Vaccines and Diagnostics contains a substance called guanidine that can potentially interact with bleach to produce cyanide gas, Stenzel said.
The only manufacturer the agency is currently aware of that uses bleach in its testing workflow is Hologic, with the Panther system. Jeff Fischer, president of Longhorn, said in an email the bleach is used automatically to decontaminate infectious material after testing is finished.
Fischer said Longhorn became aware of this and discussed it with Hologic, but Hologic determined it could not instruct labs to turn off the disinfecting step because it was part of their IVD authorization for the instrument. Longhorn then immediately put the warning up on its site about potential interaction with samples collected in PrimeStore and used on the Hologic Panther, he said. The company also informed the FDA of the issue.
"To our knowledge no other platform uses bleach in their process," Fischer said in an email, adding, "The bleach issue is not specific to PrimeStore MTM, but to labs that might not know about guanidine [being present] in most molecular tests." He cited extraction kits from Qiagen as another example of a commonly used product that contains guanidine.
Stenzel said the FDA will update its FAQ page with a warning, and that the agency is also looking across all authorized manufacturers to see if any others use bleach in their protocols and may require warnings about use of products containing guanidine.
He also asserted that the agency is responding to concerns about test performance using its tried-and-true strategy of first reaching out to manufacturers.
It was through this approach that Abbott has now been guided to clarify that samples collected for testing in the ID Now do not need to be transported in viral transport media, Stenzel said.
"The test works as intended with a direct sample," he said. "Customers will be notified to not use VTM," he added, saying that customers have reported issues, likely related to dilution that may have reduced the sensitivity of the assay. "Abbott will be updating its labeling and instructions to remove the use of VTM with this test," Stenzel said.
The Abbott ID Now had been authorized for qualitative detection of nucleic acid from the SARS-CoV-2 viral RNA in direct nasal, nasopharyngeal, or throat swabs, and nasal, nasopharyngeal, or throat swabs eluted in viral transport media. The instructions stated that swabs that needed to be transported should be eluted into 0.5 to 3.0 milliliters of saline or viral transport media by rotating the swab in the liquid for 10 seconds within one hour of sample collection, with a list of acceptable transport media also provided. The instructions also contain a note that says, "Minimal dilution of the sample is recommended as dilution may result in decreased test sensitivity."
Sara Brenner, associate director for medical affairs and chief medical officer at FDA, commented that the agency is encouraging stakeholders at universities, academic medical centers, labs, or test developers, to reach out for collaboration.
"We are being very creative in terms of looking at ways that we can work together on improving test quality, expanding our testing capacity across the US, and also to report and track the results that are coming forth," Brenner said. There are likely three tiers of diagnostics data that need to be tracked – national, state or local, and hyper-local – that need to be harmonized and made clinically and epidemiologically useful, she said.
Brenner also described community engagement strategies to address supply chain shortages, calling out 3D printing of nasopharyngeal swabs as a recent example. "If you have those sorts of thoughts, ideas, or capacities that you are trying to build up, and you're looking to engage with FDA for guidance and direction, please go ahead and reach out to us."