Antibody responses to SARS-CoV-2 nucleocapsid and spike proteins in hospitalized patients with COVID-19: A multicenter,retrospective, cross-sectional study in Japan

BackgroundThere are many commercially available automated assays for assessing coronavirus disease 2019 (COVID-19) immune responses; however, owing to insufficient data, their validities remain unknown. Here, we examined antibody responses during acute-phase COVID-19 using four assays that detect anti-spike protein IgM (S-IgM), anti-nucleocapsid protein IgG (N-IgG), anti-spike protein total Ig (S-total Ig), and anti-spike protein IgG (S-IgG).MethodsWe measured antibody levels in 1154 serum samples collected from 286 hospitalized patients with confirmed COVID-19 by a gene amplification method between February and December 2020 in Japan. Sera from 860 healthcare workers were used as negative controls.ResultsThe antibody positivity rates increased on week 2, peaked, and then started to plateau by the beginning of week 3 after symptom onset. On week 1, there were some significant differences in seropositivity rates between assays (p = 0.032): 14.9% (11.0%–19.4%) for S-IgM and 8.9% (6.0%–12.7%) for N-IgG. The seropositivity for the S-total Ig (10.6% 7.3%–14.6%]) assay was considerably better than that for the S-IgG (6.9% 4.3%–10.4%]) assay, although the difference was not statistically significant (p = 0.150). The levels of S-IgM antibodies and the three others peaked on weeks 3 and 5, respectively. All four assays showed high specificities (>99%).ConclusionsAll four assays had good specificities and were suitable for seropositivity detection after week 3 of symptom onset. Assays of IgM alone or total Ig (containing IgM) were better than those of IgG alone as an adjunct serological test for early-stage COVID-19 diagnosis, albeit the use of a serological assay alone is insufficient.