Serotype-specific immunity and association with pneumococcal carriage following 3+1, 3+0 and 2+1 schedules of 10-valent pneumococcal conjugate vaccine.

Abstract

Correlates of protection for pneumococcal carriage, a precursor for invasive pneumococcal disease, are not well defined. The relationship between memory B cells (Bmem) with carriage remains unclear. We assessed serotype-specific Bmem, IgG and opsonophagocytic (OPA) responses following different infant schedules of 10-valent pneumococcal conjugate vaccine (PCV10, Synflorix) and their association with pneumococcal carriage. We analysed secondary outcome data from a phase II/III randomised controlled trial in Viet Nam (trial registration NCT01953510). Infants received PCV10 in a 3+1 (2-3-4-9 months), 3+0 (2-3-4 months) or 2+1 (2-4-9.5 months) schedule. Serotype-specific Bmem, IgG and OPA were measured at 10 months (all) and 18 months (Bmem and IgG only). Carriage was assessed from nasopharyngeal swabs at 12-18-24 months with additional carriage episodes identified by seroincidence. Associations between immunological parameters at 10 months and subsequent carriage were evaluated using multivariable logistic regression. At 10 months, Bmem levels were similar across vaccine arms, but IgG and OPA were higher in the 3+1/2+1 than 3+0 arm. At 18 months, Bmem and IgG waned for most serotypes in all vaccine arms. There were minimal associations between the immunological parameters and subsequent carriage. Only IgG was negatively associated with serotype 19A carriage, when defined as carriage and/or seroincidence ≥2-fold (odds ratio: 0.31, 95% confidence interval: 0.10-0.94, p=0.038), and carriage and/or seroincidence of any increase (odds ratio: 0.13, 95% confidence interval: 0.04-0.40, p<0.001). We found no substantive evidence linking Bmem, IgG and OPA with pneumococcal carriage. Larger studies are needed to identify associations between different immunological parameters and carriage.