Pneumococcal population structure influences the effects of air pollution on invasive disease risk in South Africa

Abstract Streptococcus pneumoniae is highly diverse, comprising over 100 serotypes and hundreds of genomic lineages amid widespread vaccination. While it can cause invasive pneumococcal disease (IPD) which exhibits pronounced seasonal spikes, the interplay between pneumococcal diversity and environmental drivers remains unexplored. Here we analysed 59,017 IPD cases over 19 years from South Africa, incorporating 4,350 genome-sequenced isolates, using Bayesian spatiotemporal models to link environmental exposure and pneumococcal diversity. Cumulatively, across an 8-week period, moderate relative humidity (33–49%) and cold minimum temperatures (4–10 °C) increased IPD risk by 5% and 4%, respectively. Conversely, warm maximum temperatures (27–38 °C) were associated with up to a 10% increased risk within a week of exposure. There was a positive association between air pollution (PM 2.5 ) and IPD, although it varied by age, disease presentation, and most notably serotype and lineage. Specifically, the lag time between PM 2.5 exposure and disease onset varied by serotype, with only serotypes 4, 8 and 23F conferring an immediate IPD risk. High prevalence of GPSC21 lineage (serotype 19F) also modified the pollution response, shifting the lag structure to produce immediate risk of disease following high PM 2.5 exposure. Our results demonstrate that pneumococcal population structure shapes air quality risk which in turn can shape the fitness landscape of microbial populations. Integration of these data may inform public health policy.