The guidance applies to COVID-19 test developers who are requesting Emergency Use Authorization (EUA), whose tests already have an EUA, and who offer their tests as outlined in FDA’s testing guidance. Note, in regard to laboratory developed tests (LDTs), the Department of Health and Human Services’ August 19, 2020, announcement restricting FDA’s ability to regulate premarket review of LDTs is still in effect. Thus, while the recommendations in the guidance document are informative for all test developers, the scope of the guidance document likely does not cover LDTs unless the developer is seeking or maintaining an EUA for their LDT.

When reviewing EUA requests, FDA intends to consider a test’s performance across all known viral variants and the developer’s plans for post-authorization monitoring. FDA recommends that EUA requests address whether labeling should convey limitations considering the time period, geography, and genetic variety of specimens used in test development. FDA will work with developers of already-authorized tests to determine the need for similar labeling statements. FDA recommends that developers of tests offered under the testing guidance consider such statements as well.

The guidance also makes recommendations for development of COVID-19 molecular, antigen, and serology tests. Mutations, or individual changes in a genomic sequence, can affect the performance of COVID-19 tests because of how the tests operate. Molecular tests detect COVID-19 by targeting specific regions of the SARS-CoV-2 viral genome. Mutations can cause molecular tests to produce false negative results if the mutations are in the part of the genome assessed by the test. Antigen tests target specific viral proteins that are coded for in the viral genome. If mutations cause changes in the structure of viral proteins, the antigen test may not detect the proteins and produce false negative results. Serology tests target human antibodies that the body produces in an immune response to viral proteins. Mutations that change these viral proteins may trigger changes in the resulting antibodies, and serology tests may not detect these altered antibodies.

Recommendations for molecular test developers

Most of the guidance focuses on molecular tests because the potential impact of a genetic mutation on a molecular test’s performance is more straightforward, and this will be the focus of our discussion below. While FDA recommends that developers of serology and antigen tests consider the impact of current mutations and variants, the agency acknowledges that monitoring and assessing the impact of mutations on these tests is less straightforward than for molecular tests.

FDA’s recommendations for molecular test developers fall into three areas: test design, monitoring, and labeling. First, developers should design tests to minimize the impact of future mutations. Future EUA requests should discuss how the test design mitigates the risk of future viral mutations and should describe how a developer evaluated their test performance across all known variants with mutations in the targeted region. The guidance document notes that FDA intends to update its EUA templates to include conditions for authorization related to test performance across variants.

Second, molecular test developers should monitor for mutations that may impact test performance. Developers should conduct periodic sequence alignment of their test’s primer/probe sequences with publicly available SARS-CoV-2 genomes. FDA indicated that it may be problematic for a test’s performance to drop 5 percent from the previously established performance or to drop below the performance level recommended in the EUA template. The agency said it intends to revise the applicable EUA template(s) to include information and recommendations for monitoring sequence databases, and to update the EUA template(s) as databases are improved and a more systematic collection of viral sequences becomes available.

Third, molecular test developers should draft labeling that identifies any limitations on the test’s performance. Limitations may stem, for example, from the time period, geography, and genetic variety of specimens used in test development. Developers should discuss the need for such limitations in their EUA request.

The guidance also provides background on how FDA is regularly monitoring authorized molecular tests by assessing them against SARS-CoV-2 genome databases, including GISAID. GISAID (initially an acronym for “global initiative on sharing avian influenza data”) is a global science initiative and primary source that provides open-access to genomic data of influenza viruses and the novel coronavirus responsible for COVID-19. In the guidance, FDA pledged to continue to monitor signals concerning variants and COVID-19 authorized tests and to provide additional information to stakeholders and the public as more information becomes available.

FDA suggests that developers of molecular tests should also periodically conduct sequence alignment of their primer/probe sequences with publicly available SARS-CoV-2 genomes, such as in the GISAID database, to determine the extent to which mutations may impact test performance. Although FDA does not require authorized test developers to conduct monitoring, the guidance indicates that FDA may ask authorized test developers to conduct additional mutation performance testing and that such testing should be performed in a timely manner. Importantly, the results of such testing may lead to labeling changes or possibly revocation of a test’s EUA if performance across mutations is not acceptable.

FDA intends this policy to remain in effect during the COVID-19 public health emergency.

Authored by Blake Wilson, Susan Tiedy-Stevenson, and Gabrielle Field