Abstract
BACKGROUND: Although RT-qPCR remains the gold-standard for COVID-19 diagnosis, anti-SARS-CoV-2 serology-based assays have been widely used during 2020 as an alternative for individual and mass testing, and are currently used for seroprevalence studies. OBJECTIVE: To study the clinical performance of seven commercial serological tests for COVID-19 diagnosis available in South America. METHODS: We conducted a blind evaluation of five lateral-flow immunoassays (LFIA) and two enzyme-linked immunosorbent assays (ELISAs) for detecting anti-SARS-CoV-2 antibodies. RESULTS: We found no statistically significant differences among ELISA kits and LFIAs for anti-SARS-CoV-2 IgG sensitivity (values ranging from 76.4% to 83.5%) and specificity (100% for the seven serological assays). For anti-SARS-CoV-2 IgM, the five LFIAs have a significantly higher sensitivity for samples collected 15 days after the first time RT-qPCR positive test, with values ranging from 47.1% to 88.2%; moreover, the specificity varied from 85% to 100%, but the only LFIA brand with a 100% specificity had the lowest sensitivity. CONCLUSION: The diagnostic performance of the seven serological tests was acceptable for the seven brands tested for anti-SARS-CoV-2 IgG detection for seroprevalence screening purposes. On the other hand, our results show the lack of accuracy of anti-SARS-CoV-2 IgM detection in LFIAs as a tool for SARS-CoV-2 acute-phase infection diagnosis.