Fig. 2: Serum anti-S and anti-N antibody levels in the presence or absence of new infections.

a, b The antibody fold increase for participants without hybrid immunity from baseline (Dec. 2022) to the end of observation (Feb. 2023). a Anti-N antibody levels and b anti-S antibody titers. Each data point (dots) and the box plots are shown. The gray dotted line indicates the cutoff value (four-fold) based on Fig. 1d. n = 3506. c Correlation between serum anti-S and anti-N antibody levels for all participants. Data points (blue dots), correlation coefficient, (P value, regression line, and 95% confidence interval (ribbon) are shown. n = 4496. d, e The antibody fold increases in individuals with hybrid immunity from baseline (Dec. 2022) to the end of observation (Feb. 2023). d Anti-N antibody levels and e anti-S antibody titers. Each data point (dots) and the box plots are shown. The gray dotted line indicates the cutoff value (four-fold) based on Fig. 1d. n = 990. f–n Antibody fold increase before and after infection in subsampled specimens for neutralizing antibody response analysis. Anti-N antibody levels (f, j), anti-S antibody titers (g, k), and neutralizing antibody titers against the ancestral strain (h, l) and the BA.5 variant (i, m) are shown for individuals with (f–i) and without hybrid immunity (j–m), with and without new infections, at the time of the December 2022 survey and the February 2023 survey. Each data point (dots) and the box plots are shown. The gray dotted line indicates the cutoff value (four-fold) based on Fig. 1d. n = 80 participants. n–u Spearman correlation matrix of antibody titers in the subsampled specimens. The square size and a heat scale indicate Spearman correlation coefficients. The statistical significance level, corrected using the false discovery rate (FDR), is shown in the square; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. n = 80 participants.