Analytical and diagnostic performances of a high-throughput immunoassay for SARS-CoV-2 IgM and IgG

Background Reliable SARS-CoV-2 serological assays are required for diagnosing infections, for the serosurveillance of past exposures and for assessing the response to future vaccines. In this study, the analytical and clinical performances of a chemiluminescent immunoassays for SARS-CoV-2 IgM and IgG detection (Mindray CL-1200i), targeting Nucleocapsid (N) and receptor binding domain (RBD) portion of the Spike protein, were evaluated. Methods Precision and linearity were evaluated using standardized procedures. A total of 157 leftover serum samples from 81 hospitalized confirmed COVID-19 patients (38 with moderate and 43 with severe disease) and 76 SARS-CoV-2 negative subjects (44 healthcare workers, 20 individuals with rheumatic disorders, 12 pregnant women) were included in the study. In an additional series of 44 SARS-CoV-2 positive, IgM and IgG time kinetics were also evaluated in a time-period of 38 days. Results Precision was below or equal to 4% for both IgM and IgG, in all the studied levels, whilst a slightly significant deviation from linearity was observed for both assays in the range of values covering the manufacturer’s cut-off. Considering a time frame ≥ 12 days post symptom onset, sensitivity and specificity for IgM were 92.3% (95%CI:79.1%-98.4%) and 92.1% (95%CI:83.6%-97.0%). In the same time frame, sensitivity and specificity for IgG were 100% (95%CI:91.0%-100%) and 93.4% (95%CI:85.3%-97.8%). The assays agreement was 73.9% (Cohen’s kappa of 0.373). Time kinetics showed a substantial overlapping of IgM and IgG response, the latter values being elevated up to 38 days from symptoms onset. Conclusions Analytical imprecision is satisfactory as well as the linearity, particularly when taking into account the fact that both assays are claimed to be qualitative. Diagnostic sensitivity of IgG was excellent, especially considering specimens collected ≥12 days post symptom onset. Time kinetics suggest that IgM and IgG are detectable early in the course of infection, but the role of SARS-CoV-2 antibodies in clinical practice still requires further evaluations.