Among four high-throughput assays for detecting antibodies against SARS-CoV-2 in patient samples, those from Roche and Siemens provide superior sensitivity and specificity, according to head-to-head testing. "One of the more surprising results was the overlap of some samples with false-positive results across more than one platform, indicating that consideration of assay designs should be taken into account if confirmation of serological testing is desirable," Dr. David M. Manthei of the University of Michigan, in Ann Arbor, told Reuters Health by email. Early on in the COVID-19 pandemic, serologic testing was fraught with problems, including low quality that led to removal of some assays from the U.S. Food and Drug Administration (FDA) list of authorized assays. There have been multiple evaluations of SARS-CoV-2 serology assays, but more-robust evaluation of recently released tests is lacking. Dr. Manthei and colleagues compared 311 patient serum samples across four commercial assays as part of their evaluation for potential implementation of high-throughput testing. The assays included EUROIMMUN's anti-SARS-CoV-2 IgG ELISA (which captures the S1 subunit of the spike protein), DiaSorin's LIAISON SARS-CoV-2 S1/S2 IgG (which captures both S1 and S2 subunits), Roche's Elecsys SARS-CoV-2 (which analyzes total immunoglobulins against the nucleocapsid), and Siemens's SARS-CoV-2 Total (which analyzes total immunoglobulins against the receptor-binding domain within the S1 subunit). The samples included 131 serum samples from 68 patients who tested positive for SARS-CoV-2, including 95 samples collected less than one month after symptom onset and 35 samples collected mostly beyond one month. All assays were highly reproducible, with coefficients of variation of less than 5% for positive controls, the researchers report in the American Journal of Clinical Pathology. Neither the Roche nor the Siemens assay had any false-positive results, yielding specificities of 100%, whereas the specificities of the DiaSorin assay ranged from 96.1% to 97.0% and those of the EUROIMMUN assay ranged from 86.3% to 96.4%. The specificity of the Roche and Siemens assays was significantly better than those of the DiaSorin and EUROIMMUN assays, and the specificity of the DiaSorin assay was significantly better than that of the EUROIMMUN assay. Assay sensitivities improved as time from symptom onset to sample collection increased. Overall sensitivity was 84.5% for the Roche assay, 78.0% for the Siemens assay, 70.2% for the DiaSorin assay, and from 74.1% to 79.4% for the EUROIMMUN assay (with indeterminate results considered as negative and positive, respectively). When the analysis was limited to samples collected at least 14 days after symptom onset, the sensitivity of the DiaSorin assay was inferior to that of the other assays. Among patients with serial samples, repeat chronologic samples tended to become positive earlier on the Roche assay, but only the difference with the DiaSorin assay proved to be statistically significant. Estimated positive predictive values increased as expected with increasing prevalence for the DiaSorin and EUROIMMUN assays, but were 100% for the Roche and Siemens assays across prevalences from 1% to 20%. "Both Roche and Siemens assays perform well and similarly to their stated claims, although there are still implementation issues with both platforms," the authors conclude. "The best choice of a serology assay is likely an individualized decision for each clinical laboratory, depending on practical issues and plans to manage potential false-positive or false-negative test results." "I hope that clinicians take away the importance of understanding how disease prevalence or pre-test probability impacts the interpretation of a test result," Dr. Manthei said. "Tests with good or excellent performance characteristics may still result in a relatively low positive predictive value if the prevalence of disease is low. Although tests may have similar performance characteristics, even minor differences can have a large impact in such a low-prevalence setting." "Especially for serological assays, the timing of when a sample is collected in relation to disease onset will make a difference in whether antibodies - and class of antibody - would be expected to be present," he said. "We are now far enough out from early cases of COVID-19 to see that levels of antibodies may decline naturally over time to the point of no longer being detected. This could potentially mean that if someone tests negative for antibodies to SARS-CoV-2, reasons could include: never being infected, testing too soon after infection, or testing too long after infection." Dr. Manthei added, "Ideally the relevance of serological assay results would be linked to outcomes such as immunity or protection from COVID-19, but for a new infectious disease this is difficult in such a short timeframe where the natural course of disease and its sequelae are still unknown. We will continue to learn more about the relevance of antibody assays as research studies characterize the natural responses and outcomes to infection, and hopefully soon vaccination studies as well." —Will Boggs MD Source
