Abbott Laboratories’ ID Now test for the SARS-CoV-2 virus made waves when it was introduced thanks to the rapid turnaround time the molecular test promised. But questions about the test’s performance have surfaced several times in recent weeks. The latest criticism comes from a study conducted at the New York University Langone Medical Center, which claims the diagnostic compared poorly against a competitor. However, the U.S. FDA said that it has questions about the swabs and transport media used in connection with the test. As a result, it will stand pat regarding the ID Now’s place in the emergency use authorization (EUA) program.
The FDA’s director of the Office of In Vitro Diagnostics and Radiological Health, Tim Stenzel, mentioned the predicament in the agency’s April 15 testing EUA town hall meeting. Stenzel remarked that Abbott, of Abbott Park, Ill., would update the labeling for the ID Now to remove any mention of the use of viral transport media (VTM), adding that the test “works as intended with a direct simple use.” Stenzel said at the time that the issue surrounding the test appears to “be primarily a dilutional [one] that reduces the sensitivity of the assay.”
A May 14 update from the FDA said that “early data” has emerged that suggests false negative results associated with the ID Now test. Stenzel said the agency is working with Abbott to “create additional mechanisms” for studying the test. He said the diagnostic “can still be used and can correctly identify many positive results.” However, negative findings may require the use of a confirmatory, high-sensitivity molecular test.
FDA: Studies of ID Now are themselves flawed
The agency said that it is aware of studies suggesting accuracy issues with the ID Now, but noted that it is focused on whether the problem is related to the types of swabs used to collect patient samples and/or the type of transport media used with specimens. The studies in question are described as suffering from limitations, including small sample sizes, potential design biases or a failure to execute the test per manufacturer’s instructions, the FDA said.
The FDA has received 15 adverse event reports in connection with the ID Now. However, it did not cite the total test volume denominator for that 15-adverse event numerator. Abbott said in a May 11 press release that it has shipped more than 1.7 million tests to various locations in the U.S., and that it was manufacturing 50,000 tests per day on that date. The company plans to produce 2 million tests per month by June, and the FDA said Abbott has agreed to conduct a study of the test for at least 150 patients who have been confirmed as positive for COVID-19. These tests will be handled in a “variety of clinical settings,” the agency added.
The Langone study appears in the biorxiv.org database, and as is the case with a large volume of the content therein, has not been peer-reviewed. The volume of pandemic-associated content that lacks peer review has caused concern among clinicians who fear that this approach to publishing has led to hasty and, in some instances, ultimately repudiated conclusions about several aspects of the virus’s behavior, both in terms of patient physiological response and human-to-human transmission.
The authors of the Langone study said their clinical site uses the Roche Cobas SARS-CoV-2 and the Cepheid Xpert Xpress SARS-CoV-2 tests, the former of which requires a run time of more than three hours, while the latter typically needs 45 minutes. The authors said the ID Now “missed a third of the samples detected positive” by the Cepheid test “regardless of method of collection and sample type” when using nasopharyngeal swabs in VTM. The miss rate for the ID Now is said to have been 48% when using dry nasal swabs. One of the confounders may be that dry swabs are commonly associated with the acquisition of lower viral loads, which was seen in an earlier study of testing for influenza.
Abbott urges avoidance of transport/storage media
Abbott advised April 15 that some hospitals and academic labs have been validating the ID Now with VTM samples, citing the prospect that the associated dilution can affect test sensitivity. The company added that the ID Now is designed to be used with patient samples collected and applied to the instrument directly, “near the patient and without the use of VTM.” The statement goes on to recommend that the swabs should be “placed directly in the instrument for testing and not placed in any other media.”
In addition, Abbott asserted May 13 that the Langone results “are not consistent with other studies,” including one in Washington state that is said to have demonstrated 91% sensitivity and 100% specificity. This study was of roughly 1,000 subjects that yielded 23 positive results on a lab test, with the ID Now test detecting 21 of those.
The company provided a May 14 update saying it is aware of studies that are deploying the ID Now test in a manner inconsistent with its intended use. It reiterated that the ID Now is intended to be used “near the patient with a direct swab test method,” and it noted that it will provide users with additional clarification about the appropriate use of the test. Any negative results with the ID Now that are inconsistent with patient presentation should be evaluated via an alternative molecular assay, the company said, adding that it would reinforce proper sample collection and handling methods to users of the ID Now.