Performance Comparison of Five SARS-CoV-2 Antibody Assays for Seroprevalence Studies

Abstract

Background: Seroprevalence studies of coronavirus disease 2019 (COVID-19) cases, including asymptomatic and past infections, are important to estimate the scale of the disease outbreak and to establish quarantine measures. We evaluated the clinical performance of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibody assays available in Korea for use in seroprevalence studies. Methods: The sensitivity, specificity, cross-reactivity, and interference of five SARS-CoV-2 antibody assays were evaluated using the following: 398 serum samples from confirmed COVID-19 patients, 510 negative control samples from before 2018 (pre-pandemic), 163 serum samples from patients with SARS, Middle East respiratory syndrome (MERS), and other viral infections, and five samples for the interference study. Results: The sensitivities of the five assays ranged from 92.2% to 98%, and their specificities, including cross-reactivity and interference, ranged from 97.5% to 100%. The agreement rates were excellent (kappa >0.9). Adjustment of the cutoff values could be considered through ROC curve analysis. The positive predictive values of the individual assays varied from 3.5% to 100% at a 0.1% prevalence but were as high as >= 95% when two assays were combined. Conclusions: The prevalence of COVID-19 in Korea is considered to be exceptionally low at present; thus, we recommend using a combination of two or more SARS-CoV-2 antibody assays rather than a single assay. These results could help select SARS-CoV-2 antibody assays for COVID-19 seroprevalence studies in Korea.