This helps account for some cases of false positivity in serological testing for antibodies. In COVID-19, some false positive antibody testing is likely to occur from individuals’ existing antibodies to other coronaviruses that are extremely common in the population.
Based on their overall structure and roles, human antibodies are classified into five different isotypes. During acute infection, B cells typically first produce antibodies of the IgM isotype, but its levels may fall off quickly. IgG antibodies, largely responsible for long-term immunity after infection or vaccination, usually arrive a little later. Compared with IgM, IgG binds with higher specificity, and it may be detected for months or years in the blood.1
Antibodies can also be characterized by whether they are neutralizing or non-neutralizing.
In antibody-mediated viral neutralization, antibodies in cell culture may bind to specific viral domains and prevent entry into the cell. This is the type of antibody interaction desired for long-term immune protection. Other positive antibody responses may help engage other parts of the immune system, such as encouraging phagocytosis.2
Only a small subset of the antibodies formed against the virus will be of the neutralizing type: Antibodies may have other effects. Some antibodies produced have neither positive nor negative effects on a given infection. Others may upregulate pro-inflammatory cytokines, potentially resulting in immunopathology. Even more concerning is a potential problem called antibody-dependent enhancement of infection. In this scenario, which has been documented in other viruses such as dengue, antibodies may bind to the virus and facilitate viral entry into cells.2
Antibody Response in COVID-19
Much remains to be learned about the antibodies that appear in response to SARS-CoV-2. Most commonly, antibodies appear in serum one to three weeks after symptoms start. However, somewhat atypically for viruses, at least in some people IgM antibodies are delayed. Several publications have shown that IgG appears before IgM in COVID-19, and IgG is found at much higher concentrations compared to IgM.
Dr. Kadkhoda
Antibody development in SARS-CoV-2 seems to correspond to a decreased viral load in the respiratory tract.3 However, Kamran Kadkhoda, PhD, D(ABMM), D(ABMLI), medical director of the immunopathology laboratory at the Cleveland Clinic, points out that individuals with critical illness from COVID-19 have higher than average antibody levels. Thus, it is unclear exactly what role the antibodies are playing in recovery. He notes that up to a third of patients either don’t mount a neutralizing antibody response at all or do so only at very low levels.
It is not known how long IgM and IgG antibodies remain detectable following infection, but they may be present when an individual is still shedding viral particles and so still able to infect others.3 Dr. Kadkhoda adds, “I’ve heard anecdotally from some colleagues that individuals lose their titers, the antibody levels, almost five to six weeks after they are positive for the virus by [polymerase chain reaction testing].”
Cassandra Calabrese, DO, trained as a rheumatologist and infectious disease specialist and is associate staff at the Cleveland Clinic. She underscores a key open question surrounding these antibodies, asking, “Is IgG reliable for telling if you are immune to COVID-19? The short answer is no. We are hoping that infection will generate protective and long-lasting IgG antibodies, but now we just don’t know.”
Dr. Calabrese
Some other common coronaviruses possibly provide immunity for at least a limited time. Early primate studies show that rhesus macaques develop neutralizing antibodies that seem to prevent reinfection.4 However, Dr. Kadkhoda points out these animals have only been followed for weeks, not for months or years. “There is a possibility that the immunity, if there is such a thing as immunity at all, is short lived.”
Antibody Serological Testing in SARS-CoV-2
For identifying active SARS-CoV-2 infection, researchers use tests based on polymerase chain reaction (PCR) to identify the viral genetic material present in throat or nasal swabs. Such molecular, nucleic acid-based tests are intended to recognize active SARS-CoV-2 infection but will not identify previously infected individuals who have already cleared the virus.
In contrast, serologic antibody tests are designed to reveal the presence of IgG and IgM. In the early days of an infection, these antibodies are not present at detectable levels. Thus, such tests should never be used as the sole means to diagnose active infection, though they might theoretically help identify previously infected individuals.5
Many such tests have been rapidly developed over the last several months, differing somewhat in basic design and other characteristics. Some assess IgG only; others provide information on both IgG and IgM; some provide “total antibody” readings. Some require a blood draw, and others are designed to be performed with only a pinprick sample. While some can be performed only in labs capable of high complexity testing, others can be done in laboratories with more basic resources.6
These serological antibody tests are typically designed using recombinant (noninfectious) antigen based on the SARS-CoV-2 spike protein or nucleocapsid phosphoprotein (the two most common antibody targets). This allows such testing to be performed in a wide variety of labs, in contrast to studies performed with live virus, which must be performed in highly specialized labs meeting requisite biosafety standards.3
Many are enzyme-linked immunosorbent assays (ELISAs), a common laboratory platform that can be used to assess antibody concentration. The test utilizes a plate coated with a recombinant viral protein. Patient samples are incubated with the protein, and any antibodies to the virus will bind. These antibodies can then be detected by another (anti-human) binding antibody that produces a visible fluorescent or luminescent readout. Similarly, chemiluminescence immunoassays generate light signals proportional to SARS-CoV-2 antibodies. Both tests may be able to provide quantitative or semi-quantitative information about antibody concentration, although currently available commercial tests do not supply information about antibody titers.5
In contrast, lateral flow immunoassays provide positive or negative results without any quantitative information. Antibodies from the serum sample bind to recombinant viral antigens in the test strip and are wicked laterally by capillary action along its length. In the indicator region, these are bound by anti-human antibodies, which are immobilized and held in the indicator region, where they accumulate with some sort of colorant that can be assessed by a reader.1
Such kits are cheap and easy to use, and they can be used as point-of-care assays. However, they are not amenable to mass automation and large-scale testing in the same way that ELISA or chemiluminescence type tests are.1
Critically, these antibody tests are not designed to specifically measure neutralizing antibodies. Assessment of neutralizing antibodies requires much more sophisticated testing such as plaque reduction neutralization tests (PRNT). Such tests assess the effects of the antibodies on the active live virus via cell culture, but few test developers have the capability to perform such tests because of biosafety requirements.
Most of the SARS-CoV-2 antibody tests have not presented data of test validation via PRNT.7 However, some have claimed to find a correlation between titers from ELISA and titers assessed through neutralization tests.8
COVID-19 Antibody Test Proliferation
These antibody tests received a great deal of media attention over the past several months, as concerns rose about test quality of the assays flooding the market.
The U.S. Food & Drug Administration (FDA) oversees issuance of emergency use authorizations (EUAs), intended to help make testing and treatments available to the public for serious or life-threatening conditions when no other approved, adequate alternatives are available. EUA approval is granted by the FDA based on information submitted by the companies, including information about test sensitivity and specificity—even if tests or treatments haven’t gone through the full testing needed for FDA approval.
In March, the FDA loosened its standards around the use of EUAs for serological antibody tests being developed in commercial labs. The FDA began allowing such tests to be marketed without first applying for an EUA, as long as companies notified the FDA and provided certain labeling notifications. For example, labeling had to disclose the tests were not FDA approved and specify they should not be used for diagnosis. The FDA also did not authorize any antibody tests to be released or marketed for home use.9
Some companies applied for EUAs and had them granted. Others began marketing such tests without applying for EUAs. Despite the labeling requirements specified by the FDA, many companies began marketing their products inappropriately. Some companies marketed their products as diagnostic tests for COVID-19.6 Other tests were falsely labeled as being FDA authorized or FDA approved, when an EUA had not been granted. Some were inappropriately advertised for at-home testing.10
Concerns also arose about these tests’ reliability. An independent analysis by the National Institutes of Health revealed that some tests were performing poorly.9
In May, the FDA released new guidance for commercial manufactures. All such companies are now expected to submit official requests for an EUA to the FDA (including information about their validation results). This puts FDA guidance in line with similar parameters given to manufacturers of the molecular tests developed to test for active infection. The FDA also provided specific performance recommendations for specificity and sensitivity that should be met for commercial and non-commercial test developers.9
Non-commercial labs are regulated by a slightly different system, under the Clinical Laboratory Improvement Amendments (CLIA) by the Centers for Medicare & Medicaid Services. No sort of FDA authorization has been required for such tests, although they must be performed in a laboratory certified to perform high-complexity testing (as specified by CLIA). Such labs have been instructed to perform their own test validation and provide notification to the FDA of their tests. At this time, these labs are encouraged but not required to submit EUAs.9
The FDA has announced it will take action against tests falsely advertised as FDA authorized that have not been granted EUA status, and it has issued a limited number of cease and desist letters. The FDA also claims it has turned away illegitimate test kits at the border.11
However, the new policy does not specify additional actions the FDA will take if test developers fail to comply. Instead, the FDA “encourages states, hospitals and consumers to be on high alert and to make informed purchasing decisions regarding these tests.”11
On May 21, the FDA posted a list of antibody tests being removed from their earlier list under the new policy.12 Some manufactures have also voluntarily withdrawn their products from the market. Dr. Kadkhoda adds, “Initially it was a large number of lateral flow assays that had very questionable quality. At least those, for the most part, are gone now.”
Tests with Current EUA Approvals
The FDA has granted EUAs to tests of several types, including ELISAs and lateral flow assays. As of this writing, 39 COVID-19 antibody tests have been granted EUAs. The FDA provides a list of all tests granted EUAs, along with basic information about test characteristics and performance based on data supplied by the manufacturers.13,14
Although the FDA requires that companies validate their results with such parameters as specificity and sensitivity using known COVID-19 samples, data from PRNT tests demonstrating antibody neutralization are not required for EUAs. Dr. Calabrese notes that such tests are ordinarily performed as part of a standard process for FDA approval of a new test.
The FDA is actively collaborating with the National Institutes of Health’s National Cancer Institute, the Centers for Disease Control and Prevention and the Biomedical Advanced Research and Development Authority to independently validate some of these antibody tests. Information about test sensitivity, specificity and other factors from a limited number of such tests is now available (although these also do not include information from neutralizing PRNT studies).15
Utility of Antibody Testing in SARS-CoV-2
Much media attention has focused on the appearance of fraudulent tests and the validity of available tests, and deeper questions remain about when and whether such tests should be used. The issue is not just one of testing quality. Dr. Kadkhoda points out, “Usually when labs start offering a test, we know a lot about it, and we have enough evidence of its clinical usefulness. What happened in SARS-CoV-2, was that we did the reverse. A lot of labs started testing first, before we had that evidence.”
Dr. Kadkhoda adds, “There are some tests out there that are downright bad, and that’s why the FDA is now cracking down on them after its initial complacency. But even the best ones out there still fall short in terms of clinical usefulness—the science is just not there.”
The issue of false positivity is key, even for tests of the highest caliber. Ninety to 100 percent of adults have antibodies to common coronaviruses, a potential source of false positivity. Dr. Kadkhoda notes that other known factors can lead to false positives, such as flu and flu vaccination, syphilis, herpes virus, metapneumovirus and dengue. Of special note to rheumatologists, both patients with high anti-nuclear antibody levels and patients positive for rheumatoid factor show an increased rate of false positivity for these tests.7
Disease prevalence also influences reliability. Dr, Kadkhoda presents a scenario of <2% disease prevalence, a reasonable current estimate in much of the country. He explains, “Even with a sensitivity of 100% and a specificity of 99%, which would be unreasonably high, our positive predictive value would be very low. Up to half of results could be false positives. When the prevalence in an area is very low, the possibility of a false positive result is actually very high.”
Very Limited Diagnostic Role
The Infectious Diseases Society of America emphasizes these antibody tests should never be used as a single diagnostic tool. Dr. Calabrese points out that serological testing has never been routinely used for diagnosing respiratory viral infections like flu or respiratory syncytial virus, because it isn’t particularly useful. If performed early in the disease course, the test is very likely to be negative, because antibodies have not yet developed, particularly in more mild cases. PCR is a much more reliable option.
Dr. Kadkhoda also points out that most hospitalized patients test positive via PCR throughout the course of their hospitalization. He adds, “By the time IgM appears, the vast majority of these patients have already presented with things like septic shock or [acute respiratory distress syndrome], which is too late.”
Dr. Calabrese notes these tests may potentially play a very limited role in diagnosis in specific circumstances. “The duration of PCR positivity is variable and can turn negative a couple weeks into infection,” she explains. “If someone presents late into symptom onset, their PCR might be negative, and there might be a role for complementary antibody testing.” The high risk of false positivity should also be considered in such scenarios, particularly in an area of low disease prevalence in a person with no known disease exposure.
A potential exception is the new multisystem inflammatory syndrome in children (MIS-C) that is potentially linked to SARS-CoV-2 infection. Dr. Kadkhoda suggests that such serological antibody testing might be used in the context of an exclusionary diagnosis, if initial viral PCR were negative. But even here, he adds, a positive antibody marker wouldn’t necessarily establish diagnosis, and the test results would be unlikely to change medical management.
As Immune Status Marker
Some governments have suggested that antibody results might be used for “immunity passports” that might enable individuals to travel because they are presumed to be immune to re-infection. However, the World Health Organization and other health groups have strongly argued against this.16
Dr. Calabrese adds, “Some places are using these antibodies as an immune status check or even a back to work check. But that does not seem to be a good or safe idea. You could have positive IgG to COVID-19 but still be sick and shedding virus.” Patients testing positive for the virus—either as true positives or false positives—may stop taking recommended public health precautions.
Test specificity is an issue, but another problem is that the “correlate of protection” has not been established for COVID-19: We don’t fully understand the measurable signs that can be used to indicate a person is immune. “Compare it with something like hepatitis B virus,” says Dr. Kadkhoda. “We do serology for it; we know which antibody is the correlate of protection, and we know even what level of antibody equals protection.”
Complicating matters still further, at this point researchers cannot rule out the possibility of antibody-dependent immune enhancement, though thus far no data have directly suggested that this is the case.2
Through prospective vaccine trials, we may ultimately discover if antibody is the component of the immune system that might confer protection, and if so, what level might be needed. Potentially other factors, such as cellular immunity, might be important.7
Says Dr. Calabrese, “We have patients reaching out to us all the time asking for antibody tests so they can feel more comfortable going back out into society, but we strongly recommend against using a positive antibody test as a security blanket.” A role for serology may emerge once we know more about the natural history of the virus and the immune response.
Valid Uses: Serosurveys, Plasma Donor Screening, Vaccine Development
Serosurveys are a valid use of the antibody tests, and they provide a helpful avenue for epidemiological studies of disease prevalence. However, Dr. Kadkhoda points out that even for these purposes, health officials should view antibody test results with caution. He points out that in an area of low disease prevalence, even high-quality antibody tests will have a high rate of false positivity, giving an inflated number of individuals who have previously been infected. He says, “As a result, you may arrive at a low case fatality rate, and erroneously people may think this disease is not as dangerous as it is.”
Serology tests are also currently being used in convalescent plasma therapy. In this treatment, plasma containing antibodies from recovered patients is transfused to patients very ill with COVID-19. Serology tests are being used to identify candidates to donate plasma. Though some early small studies have been promising, the most recent randomized controlled trial showed no clinical benefit.17,18 Further controlled trials will be needed to demonstrate that the therapy is effective.
Tests for SARS-CoV-2 antibodies are also important for vaccine design and evaluation as part of clinical trials. This will require a test with high sensitivity—and even more importantly, high specificity—to assess the immune response after vaccination. Here, researchers would specifically need to identify neutralizing type antibodies.
Dr. Kadkhoda explains, “Immunogenicity trials typically should include a neutralizing assay, not just an ELISA or other commercial assays. They might try those also, but it’s very important to have a neutralization assay to make sure those individuals make neutralizing antibodies against that spike protein of SARS-CoV-2.”
Overall Usefulness for Clinicians
With more information gleaned from vaccine development, it may make sense to run these antibody tests as immune markers. But at the present time, these serological antibody tests, even the high-quality ones, offer little utility to clinicians. Clinicians may do well to remember the familiar adage not to test unless the results might impact medical management.
Neither Dr. Kadkhoda nor Dr. Calabrese are choosing to recommend serological antibody assays as a part of routine testing. “I don’t want to offer the test and have the healthcare provider ask me, ‘What does this result mean?’ I wouldn’t have anything helpful to tell them,” says Dr. Kadkhoda.
If clinicians do choose to offer such antibody tests to patients, they should evaluate the regulatory status and performance characteristics of specific antibody tests used, choosing tests that have demonstrated high specificity. Clinicians must also keep in mind factors influencing the positive predictive value of the test, such as disease prevalence. Results can also be improved by focusing on people with a higher pre-test probability of having antibodies, such as people who’ve recently experienced symptoms consistent with COVID-19. Alternatively, clinicians may choose to confirm a positive result with a second test that uses different design characteristics.3
But Dr. Calabrese emphasizes that neither clinicians nor patients should overinterpret results of positive SARS-CoV-2 antibodies. “At the present time, at least until we learn more, positive serology should not cause anyone to ease up on infection prevention measures.”
Ruth Jessen Hickman, MD, is a graduate of the Indiana University School of Medicine. She is a freelance medical and science writer living in Bloomington, Ind.
References
- Jacofsky D, Jacofsky EM, Jacofsky M. Understanding antibody testing for COVID-19. J Arthroplasty. 2020 Apr 27;S0883-5403(20)30442-3.
- Iwasaki A, Yang Y. The potential danger of suboptimal antibody responses in COVID-19. Nat Rev Immunol. 2020;20(6):339–341.
- Coronavirus disease 2019 (COVID-19). Interim guidelines for COVID-19 antibody testing. Centers for Disease Control and Prevention. 2020 May 23 (reviewed).
- Chandrashekar A, Liu J, Martinot AJ, et al. SARS-CoV-2 infection protects against rechallenge in rhesus macaques. Science. 2020 May 20;eabc4776. (Online ahead of print)
- IDSA COVID-19 antibody testing primer. Infectious Diseases Society of America. 2020 May 4.
- Abbasi J. The promise and peril of antibody testing for COVID-19. JAMA. 2020 Apr 17. doi: 10.1001/jama.2020.6170. (Online ahead of print)
- Kadkhoda K. COVID-19 serologic testing: FAQs and caveats. Cleve Clin J Med. 2020 Jun;87(6):329–333.
- Amanat F, Stadlbauer D, Strohmeier S, et al. A serological assay to detect SARS-CoV-2 seroconversion in humans. Nat Med. 2020 May 12. doi: 10.1038/s41591-020-0913-5. (Online ahead of print)
- Policy for coronavirus disease-2019 tests during the public health emergency (revised). U.S. Food & Drug Administration. 2020 May 11.
- Coronavirus (COVID-19) update: FDA alerts consumers about unauthorized fraudulent COVID-19 test kits. U.S. Food & Drug Administration. 2020 Mar 20.
- Shah A, Shuren J. Insight into FDA’s revised policy on antibody tests: Prioritizing access and accuracy. U.S. Food & Drug Administration. 2020 May 4.
- Coronavirus (COVID-19) update: Daily roundup. U.S. Food & Drug Administration. 2020 May 21.
- Emergency use authorizations. In vitro diagnostic EUAs. U.S. Food & Drug Administration. 2020 Jun 8.
- EUA authorized serology test performance. U.S. Food & Drug Administration. 2020 Jun 4.
- Independent evaluations of COVID-19 serological tests. U.S. Food & Drug Administration. (no date).
- ‘Immunity passports’ in the context of COVID-19. World Health Organization. 2020 Apr 24.
- Shen C, Wang Z, Zhao F, et al. Treatment of 5 critically ill patients with COVID-19 with convalescent plasma. JAMA. 2020 Mar 27;323(16):1582–1589. (Online ahead of print)
- Li L, Zhang W, Hu Y, et al. Effect of convalescent plasma therapy on time to clinical improvement in patients with severe and life-threatening COVID-19: A randomized clinical trial. JAMA. 2020 Jun 3. doi: 10.1001/jama.2020.10044. (Online ahead of print)
Updated Aug. 21, 2020, to reflect current number of FDA-approved tests and a new URL for reference No. 13.
