Serological testing for COVID-19 is intended to identify individuals already exposed to—and potentially immune to reinfection from—the virus. However, the FDA’s approach to regulating the tests has resulted in inconsistent oversight. AgencyIQ explains the FDA’s current approach, current issues, and potential paths forward.
By Laura DiAngelo, MPH
Serological tests are those that can detect antibodies to a disease, rather than the disease itself, indicating that a person has had a virus and presumably may be immune to catching it again.
As US policymakers seek to find ways to safely re-open the economy, they are looking to widespread serological testing as a means to identify those persons least at risk of both becoming infected and infecting others. However, even as the tests are seen as a critical linchpin to public health strategies, currently marketed tests are by themselves unlikely to be sufficient to protect the public health.
Three issues currently ail these tests: Their validity, a lack of information about the significance of their findings, and a lack of consistent FDA oversight.
In this analysis, AgencyIQ looks at the science behind serological testing, the policies that have led the market to be flooded with potentially inaccurate tests, responses from policymakers, and the FDA’s limited authority to pull back on the bad tests.
What is a serological, or “antibody” test?
Serologic tests are those intended to detect antibodies to a certain disease, rather than the presence of the disease itself. As compared to a molecular diagnostic for COVID-19, which will detect nucleic acid from the SARS-CoV-2 viral RNA, a serologic test is designed to identify immunoglobulin antibodies (IgM or IgG) against the virus itself.
The validity of a serologic test is not as well defined as other types of diagnostics. Serological tests require researchers to identify “clean, readily interpretable titers” (or levels) “for the immune response to the chosen antigen.” While the pathogen will look the same in every human body, different individuals at different disease stages will have different immune responses, making it difficult to accurately identify and validate the test itself, and potentially even more difficult for a provider or laboratory to interpret.
While serologic tests can be used as the single diagnostic to identify a condition if a molecular diagnostic is unavailable or infeasible (for example, percutaneous allergies such as latex allergy), evidence of their accuracy varies. One acute concern is the rate at which a test returns false positives. For example, research from the World Health Organization (WHO) estimated that serological tests for tuberculosis (which are not recommended by the WHO) would identify return false positives for 8.6 patients for every one correct positive returned.
Depending on how well-understood the immunoreceptors to a virus are, serological tests can also product inconsistent results. They are often used as screening for certain conditions—for example, celiac disease—that can inform more invasive but more accurate diagnostic procedures, such as a biopsy. The recommended use of a serologic test, then, is often weighed against the potential risk to a patient if there is a false positive. Notably, the US Preventive Services Task Force (USPSTF), the entity that determines whether a product or service is free to patients under the Affordable Care Act, recommends against serologic screening for some conditions (e.g., genital herpes) due to the high rate of false positives and low specificity. However, it does recommend serologic screening for some other conditions, especially if there is a treatment available (e.g., hepatitis B).
How are serology tests for COVID-19 regulated?
By some estimates, there are about 108 serological tests marketed for COVID-19. However, only eight have actually been reviewed by the FDA.
In March 2020, the FDA issued comprehensive guidance for diagnostic developers of both molecular diagnostics and serologic tests. Under that guidance, serologic test developers could pursue two pathways to market access:
- Pathway C: A serological test developer can submit an Emergency Use Authorization (EUA) to the agency for formal authorization.
- Pathway D: A serological test developer can submit notification to the agency that they have completed validation of a test.
While tests that are distributed under Pathway D face significantly lower barriers to market than those under Pathway C, as they must only notify the FDA of their validation efforts rather than submit an EUA, there are significant limits placed on their scope of use. Tests that have not been authorized for use under an EUA under Pathway C may not be used as the sole basis to diagnose or exclude COVID-19, and would therefore need to be followed up with a molecular diagnostic to confirm a diagnosis of the novel coronavirus. Additionally, there are limits to where validated tests can be used based and the laboratory in which they could be performed.
In additional to FDA regulatory oversight under Pathways C and D, tests are also subject to oversight under the Clinical Laboratory Improvement Amendments, or CLIA.
Under CLIA, Congress granted authority over clinical laboratory testing procedures to the Centers for Medicare and Medicaid Services (CMS), rather than the FDA. While the FDA does have authority over diagnostics that are manufactured and distributed through traditional processes, tests that are developed and used within a single laboratory or health system and don’t enter into interstate commerce are subject to regulatory oversight by CMS.
CLIA establishes standards for the laboratories in which tests are developed and/or performed–not measures of validity for the tests themselves. The FDA’s role for tests that are not approved is to determine the level of complexity of a test, which controls the types of laboratory in which it may be used.
There are four levels of CLIA certificates for laboratories, based on the level of complexity of tests that they are permitted to conduct and their oversight standards:
- CLIA Waivers: Laboratories may perform only waived tests and are exempt from surveys and personnel requirements.
- Provider-performed microscopy (PPM) certificates: Laboratories may perform both waived tests and PPM tests, are subject to midlevel practitioner requirements, and are exempt from surveys.
- Certificates of compliance: Laboratories may perform waived, PPM, moderate and high complexity testing, and are subject to both surveys and personnel requirements.
- Certificates of accreditation: Laboratories can perform waived and PPM testing but must perform moderate and high complexity testing. Compliance is determined through an Accrediting Organization, not CMS directly, although CMS may issue validation and complain surveys.
Tests that are deemed to be “high complexity,” then, can only be used in a laboratory with a certificate of compliance or a certificate of accreditation.
However, these types of laboratories represent the significant minority of all clinical laboratories in the US: Of the total number of clinical laboratories, 15,732 have a certificate of accreditation, 17,480 have a certificate of compliance, 40,570 have a PPM, while 192,332 have a waiver.
When the FDA authorizes a serological test through an EUA, it also will determine the level of complexity for the test. To date, all serological tests authorized for use at the point of care under an EUA are considered “waived,” which means that they can be performed outside other CLIA-regulated laboratory setting.
However, Pathway D tests, which are not reviewed by the agency, are all considered “high complexity” by default—meaning that they are only eligible for use by a limited set of laboratories. According to the law firm Hyman, Phelps & McNamara, the FDA’s policy to consider all non-authorized, validated serological tests to be “high complexity” significantly limits their use, as a high-complexity test can not be performed at point of care, but rather only in a high-complexity certified laboratory.
As the FDA doesn’t have insight into the availability of these tests, however, it also has little insight into their use, including the locations of their use. It remains possible that laboratories or sites without the appropriate accreditation are performing these high-complexity serological tests.
Can we re-open with the serology tests we have?
Ensuring the validity, accuracy and appropriate use of serological tests is a cornerstone of policymakers’ recent plans to use widespread serological testing to identify who has had the virus and who may be immune in order to re-start the economy. Plans referencing this idea include those from the White House, former FDA Commissioners Scott Gottlieb and Mark McClellan, other clinical researchers and experts and state governors.
But even as widespread serological testing may be needed, public health officials concede the tests’ lack of accuracy may hinder their use as part of a short-term strategy or in health care settings. For example, formal guidance from the Minnesota Department of Health cites “specific concerns regarding the use of serological tests in the diagnosis and management of COVID-19,” recommending that providers “should be cautious if asked to interpret the results of serological testing.”
A new report from the National Governor’s Association (NGA) issued on April 21 also casts doubt on the current serological tests’ reliability to support accurate exposure data.
“Many key scientific questions have yet to be answered,” the report states. “These questions include the accuracy of the many tests on the market, as the U.S. Food and Drug Administration (FDA) has yet to review data on the vast majority of marketed tests. In addition, it is still unknown the extent to which a measured antibody response corresponds directly to immunity to disease and the duration of such immunity.”
Although widespread serology testing remains one of the steps that the NGA report recommends as part of any re-opening plan, the organization warns against relying on currently available tests.
The FDA has meanwhile said that it is “not aware of an antibody test that has been validated for diagnosis of COVID-19 infection.”
“While FDA remains open to submissions of these tests for such uses, based on the underlying scientific principles of antibody tests, we do not expect that an antibody test can be shown to definitively diagnose or exclude COVID-19 infection,” it said in an April 2020 FAQ document.
The FDA said that any tests validated under Pathway D and marketed without an EUA “should be ordered only by clinicians who are familiar with the use and limitations of the test”—a far cry from allowing widespread testing of every individual to clearly determine if they can return to work.
Altogether, the effects of the FDA’s policies raise significant concerns about the validity of some serological tests, the accuracy of their results, and the ability of the personnel in certain laboratories to perform serological tests.
According to the Director of the FDA Office of IVDs and Radiological Health, Timothy Stenzel, the FDA has “seen a huge number” of serological tests come into the US through Pathway D. Some anecdotal reports have placed the number at about 100. Comparatively, the FDA has only issued eight EUAs for serological tests to identify the novel coronavirus as of April 27.
Recent actions have indicated that the FDA is seeking to increase oversight of these tests, pulling them back into line with regulatory requirements. The FDA issued a statement in April noting that while it allows for the distribution (under Pathway D) of serological tests that are neither approved, cleared, nor authorized, these tests may not include a claim that they have been reviewed by the agency. The FDA considers these false claims and has pushed back against the developers of Pathway D tests that are incorrectly claiming that they can diagnose COVID-19.
However, with increased media attention, the FDA’s statements on test validity may be of limited help to providers and policymakers—and limit the public’s trust.
While Dr. Stenzel said at a town hall meeting that the FDA was looking into the creation of an inter-agency task force to oversee the myriad tests on the market under Pathway D, no formal announcement has been made. Recently, however, the FDA’s serological testing FAQ document was updated with an initiative with the National Institutes of Health (NIH), Biomedical Advanced Research and Development Authority (BARDA) and the CDC “to assess the performance of serological tests offered under” Pathway D.
Under this inter-agency process, developers can submit their serological test to the NIH, which will perform validation testing on their behalf using a standard performance assessment protocol. Evidence from that testing can be used in an EUA submission, reducing burden for developers that choose to participate. However, this process remains voluntary. Although the FDA is encouraging developers with a validated, marketed serological test to submit an EUA, there is currently no requirement that they do so, and the FDA has limited authority to force their removal or validation (unlike an EUA, which can be withdrawn).
Can the FDA actually fix the problem with the tests?
Each re-opening plan hinges on the idea that serological testing could identify both the actual prevalence of the virus in the population and also determine who has functional immunity to the virus after being infected, even if they did not experience symptoms.
But even that approach depends on additional information validating whether someone with antibodies is actually immune from a second round of the virus and how long that immunity could last. Those data are still not available. Without accurate data on measures that could prevent a second infection, the FDA can’t approve any tests that would be indicated for that purpose.
According to a sharply worded scientific brief from the WHO, “at this point in the pandemic, there is not enough evidence about the effectiveness of antibody-mediated immunity to guarantee the accuracy” of a policy that would allow individuals with antibodies back into public life. No matter how quickly the FDA moves to review tests, without accurate validation measures their use will remain limited.
Questions about the validity and accuracy of serological testing is also likely to persist for months, recalling the issues with molecular diagnostic availability and accuracy from February and March.
Already, Congressional lawmakers have been questioning the FDA’s oversight and ability to ensure accuracy of the serological tests. According to a preliminary report from the House Committee on Oversight and Reform, lawmakers found that “FDA is unable to validate the accuracy of antibody tests that are already on the market, and companies are ignoring requests from [HHS] to voluntarily submit their tests for validation.” While the tests marketed under Pathway D are not “fraudulent,” as the House report contends, the report reflects growing concern in Congress about the FDA’s oversight authority. Going forward, Congress could pressure the agency to take action to remove Pathway D tests from the market, or even alter the agency’s authority to do so.”
Even without data on immunity measures, potentially invalid tests that have never been reviewed by the FDA could remain on the market and provide misleading evidence of safety or protection.
“It’s a disaster,” said Severin Schwan, CEO of Roche, in a statement. “These tests are not worth anything, or have very little use.” Roche has developed its own serological test that will reportedly be submitted for an EUA in May.
While the FDA can’t take enforcement actions against manufacturers of Pathway D tests unless they’re found to be misrepresenting themselves, there is an opportunity for these tests to be removed from market through the EUA process itself. Under the EUA regulations, one of the conditions for authorization is that there be “no adequate, approved, and available alternative to the product for diagnosing, preventing, or treating such disease or condition.” If another product for the same indication is approved, the FDA may terminate the authorizations of unapproved products, or move to rescind its “Pathway D” policy.
What’s next for serology tests?
States, which are moving to make use of serology testing, are likely to move ahead on testing despite limited evidence. For example, New York state released preliminary results indicating that 13.9% of the population have COVID antibodies, according to a sample of 3,000 individuals. In running that study, the state used serological tests developed by its flagship public health laboratory, the Wadsworth Center. Going forward, NY Gov. Andrew Cuomo (D) has said that “the state will continue working with the federal government to assist with the supply chain and coordinate private labs to ramp up diagnostic testing,” but it’s unclear if the Governor will be ordering EUA-authorized tests from manufacturers.
However, other states without a sophisticated public health laboratory system like New York will likely have to rely on tests from other developers. It may be more likely that state governors will place large orders for tests that have been authorized by the FDA rather than those that have just been validated—although widespread testing plans have not yet been clearly outlined or implemented, beyond the call from the NGA.
In the near-term, the FDA is expected to issue a new EUA template for a new type of COVID-19 diagnostic: antigen tests, which are intended to identify the virus itself. With an increased capacity to test for the virus in a less complex way, some of the difficulties with widespread patient testing may be alleviated. However, it will take time to develop, validate, submit, authorize, manufacture and distribute these tests.