Efficacy of POC Antibody Assays After COVID-19 Infection and Potential Utility for "Immunity Passports"

Akram Shalaby, MD; Hansini Laharwani, MD; John T. Bates, PhD; Patrick B. Kyle, PhD

Disclosures

Lab Med. 2022;53(3):262-265. 

In This Article

Discussion

Point-of-care antibody tests for SARS-CoV-2 have several advantages: They are inexpensive, easy to perform, and offer rapid results. Antibody tests for SARS-CoV-2 antibodies may provide an insight into the prevalence of COVID-19 in specific geographic locations. This concept was shown in a pilot study in Chelsea, MA, involving 200 apparently healthy individuals, of whom 64 (approximately 30%) were positive for antibodies to SARS-CoV-2. These findings provided a valuable snapshot in a community known to have a high prevalence of disease at the time.[10] The Florida Department of Health reported that 4.4% of 123,552 healthcare workers, firefighters, police officers, and first responders were positive for antibodies.[11,12]

In this study, 2 of the immunoassays, ELISA and BioMedomics, exhibited false-positive results that reduced their specificities to 93.3% and 96.7%, respectively. Assays with very high specificities are required for population screening because when prevalence is low, even a few false-positive results cause a significant overestimation of disease.[13] For example, if the prevalence of the disease in the population is 5%, the positive predictive value of a test that exhibits 95% sensitivity and 95% specificity will be 50.0%, essentially a coin toss. Meanwhile, tests that exhibit 95% sensitivity but 99% and 99.5% specificity would exhibit positive predictive values of 83.3% and 90.8%, respectively. Therefore, positive results from assays that exhibit low specificity should be followed with another test whenever the prevalence of disease is low.[14] Although these antibody tests correlated well with RT-PCR results, they did not meet the 99% specificity recommended for population surveillance. In this dataset, a follow-up test would have eliminated all false-positive results.

Interestingly, 3 patients with COVID-19 who tested positive by RT-PCR were negative on all 4 serological assays. This outcome could have resulted from antibody concentrations below the detection limit, which could occur with a weak immune response to the virus or during the early stages of disease before sufficient antibody concentrations have formed. A study of sailors infected during the USS Theodore Roosevelt outbreak found that only 90% of infected individuals who reported a positive RT-PCR test before the study made detectable levels of SARS-CoV-2 spike-specific antibody after infection,[15] which generally agrees with the 87% seroconversion we observed. Reasons for the low seroconversion rate among RT-PCR–confirmed infections remain unknown but are likely related to the length of time after infection.

Our results showed sensitivities of 50% and 100% for both the BioMedomics and Premier Biotech assays in specimens collected <7 days and >14 days after the RT-PCR testing, respectively. Others have attributed negative SARS-CoV-2 antibody results after confirmed infection to delayed specific antibody responses in patients with severe illness.[16,17] The 3 specimens that were negative by both lateral flow assays, ELISA, and the Abbott ARCHITECT but collected from patients who were RT-PCR–confirmed positive were collected at 7, 9, and 10 days after the onset of symptoms. These specimens would have been expected to contain sufficient concentrations of IgM, if not IgG.[18] Given the reported variability of the antibody response to SARS-CoV-2 infection, the agreement between these assays raises the possibility that these specimens represent true biological negatives rather than false negatives.

Antibodies are likely to offer protection against reinfection with SARS-CoV-2; therefore, antibody assays may have a role in identifying immune individuals. One study involving 3.2 million people who had undergone SARS-CoV-2 antibody testing concluded that seropositive individuals have a significantly decreased risk for future SARS-CoV-2 infection.[19] Animal studies investigating immune response in rhesus macaque monkeys[20,21] also showed humoral and cellular immune response to SARS-CoV-2, suggesting that some level of protective immunity may occur. The transfer of sera from immunized primates to hACE2 transgenic mice in another study protected against a challenge with SARS-CoV-2.[22] Antibodies are the correlate of protection for the great majority of viral infections and seem to correlate with protection against SARS CoV-2 infection.[23]

There is an ongoing debate about the use of "immunity passports" and their role in travel restrictions.[24] Several European countries began issuing certificates of travel to verify vaccination against SARS-CoV-2, receipt of a negative test result, or recovery from the virus.[25] Italy, Iceland, Spain, Greece, and other countries were opening their borders to travelers who have been vaccinated or recently tested negative for COVID-19.[26] Note that at this time, the U.S. Food & Drug Administration does not recommend antibody tests to assess immunity or protection from vaccination.[27] Because the presence of antibodies to COVID-19 likely indicates immunity,[19] antibody tests could play a role as one modality to prove immunity in previously infected individuals. This testing could benefit travelers from underserved countries with limited vaccination rates or limited COVID-19 testing. Given the results observed in this study, antibody tests may provide variable results within the first 10 days after the development of symptoms of SARS-CoV-2 infection. Point-of-care tests are inexpensive, easy to perform, offer rapid results, and may have a utility in screening for immunity and resuming international travel even at the point of travel. A limitation of antibody tests could involve the potential misinterpretation of results in that (1) days or weeks are required for seroconversion after infection and (2) assay sensitivities and specificities may vary considerably.

processing....