Abstract
Streptococcus pneumoniae (the pneumococcus) is a common nasopharyngeal commensal capable of infecting normally sterile anatomical sites, resulting in invasive pneumococcal disease (IPD). Effective vaccines preventing IPD exist, but each of the antigens they contain typically induces protective immunity against only one of the approximately 100 pneumococcal serotypes, which are differentiated by immunogenically-distinct polysaccharide capsules. Serotypes vary in their propensity to cause IPD, quantified as their invasiveness. Vaccines are designed to include serotypes commonly isolated from IPD, but the immunity they induce is sufficiently strong to also eliminate vaccine serotypes from carriage. This enables their replacement by non-vaccine serotypes in the nasopharynx. The emergence of invasive non-vaccine serotypes has undermined some vaccination programmes’ benefits. Recent advances in genomics and modeling have enabled forecasting of which non-vaccine serotypes will be successful post-vaccination. Here, we demonstrate that vaccines optimised using this framework can minimise IPD and antibiotic-resistant disease more effectively than existing formulations in the model, through mitigating the consequences of serotype replacement. The simulations also demonstrate that tailoring vaccines to the pre-vaccine bacterial population is likely to have a substantial impact on reducing IPD, highlighting the importance of epidemiological data, genomics and ecological models as tools for vaccine design and evaluation.