Fig. 2: Association between maternal antibodies and oral rotavirus vaccine response.

A Geographic differences in maternal antibody concentrations. Groups were compared by ANOVA with post-hoc Tukey tests. The dotted lines at 20 IU/ml indicate the standard cut-off for RV-IgA seropositivity. B Geographic differences in maternal breastmilk/serum RV-IgA ratios. Groups were compared by Dunn’s test. Ratios were calculated using log-transformed antibody concentrations. See Fig. 1 legend for box plot parameters. C Association between maternal antibodies and infant RV-IgA formation. Log-transformed concentrations were compared using Pearson’s correlation coefficient (r) with two-sided hypothesis testing. Infant samples for RV-IgA measurement were collected at the time of dose 1 (week of life 6 in India/Malawi; week of life 8 in the UK) and 4 weeks after dose 2 (week of life 14 in India/Malawi; week of life 16 in the UK). D Correlation between rotavirus-specific antibody concentrations and rotavirus shedding in Indian infants with complete data (n = 298). For shedding variables, 1/Ct was used such that higher values correspond to higher rotavirus quantities. Shedding after week of life 1 was determined based on the group A rotavirus VP6 gene assay (Ct range 23.5–35.0) while shedding after dose 1 was based on the Rotarix-specific NSP2 gene assay (Ct range 20.7–40.0). Variables were compared using Spearman’s rank correlation coefficient (rho) with two-sided hypothesis testing. neo+, infected with rotavirus neonatally (defined by detection of rotavirus shedding in week of life 1 or baseline seropositivity); neo−, uninfected with rotavirus neonatally; ns, not significant; ORV, oral rotavirus vaccine; RV, rotavirus; *p < 0.05; **p < 0.005; ***p < 0.0005.