Impact of climate and acquired immunity on malaria burden in Senegal and The Gambia: an agent-based modeling study

Immunity against Plasmodium falciparum malaria is acquired slowly and incompletely through repeated infections, leading to differing levels of protection among individuals based on their history of exposure to the disease. Since malaria is a climate sensitive vector-borne disease, immunological history is shaped by the local climate. Nonetheless, compartmental malaria models, which may be used to inform policy, implicitly represent immunity as a population-average trait --- an assumption that can lead to incorrect estimates of clinical prevalence by age, and of the overall burden of malaria. As the rate at which immunity is acquired depends on transmission intensity, and therefore on background climatic conditions, the error introduced by this assumption is expected to be climate-sensitive. We quantify how individual exposure histories shape community-level prevalence, and identify the age groups for which resolving immunity individually matters most across climatic regimes. For this, we introduce VECTRI-ABM, a grid-based framework coupling the climate-driven vector component of the VECTRI model to an agent-based representation of the human population, resolving immunity at the individual level. The framework was calibrated against entomological and epidemiological observations from Dielmo and Ndiop (Senegal, 1990--2003) and evaluated against reported entomological inoculation rates, Malaria Atlas Project estimates of parasite rate in children aged 2--10, and confirmed cases from Senegal's National Malaria Control Program (PNLP). We then applied it across Senegal and The Gambia, which encompass a pronounced Sahelian-to-tropical rainfall gradient, and compared simulations in which immunity develops individually against a counterfactual where agents carry the local population-average immunity, thus mimicking the assumption made by standard models. Results show that the averaging approach misallocates burden across age, underestimating clinical prevalence in children, especially so in the wettest high-transmission regions, and overestimating adult asymptomatic reservoirs, with implications for transmission-targeting interventions. We find that the simpler approach to immunity is adequate for immunologically mature populations, typically above 15 years of age, and that the error it introduces elsewhere is climate-dependent.