The quiet erosion of malaria's front-line treatment is one of global health's most consequential threats — and tracking it across a continent with patchy surveillance has been nearly impossible. A large-scale systematic review and spatiotemporal modelling effort now provides the highest-resolution picture yet of where artemisinin partial resistance (ART-R) and partner-drug resistance markers are entrenched, spreading, or still absent across sub-Saharan Africa, with direct implications for how combination therapies are deployed.
Drawing on data from PubMed, Ovid MEDLINE, and Web of Science searches covering studies published between September 2014 and July 2025, the investigators merged their systematic review findings with 11 unpublished datasets and three major surveillance repositories — WWARN's molecular surveyor, MalariaGEN Pf8, and the WHO malaria threats map — to construct an unusually rich, deduplicated dataset. The analysis focused on three genetic markers: kelch13 (k13) mutations conferring artemisinin partial resistance, and mdr1 N86Y and crt K76T variants associated with reduced susceptibility to ACT partner drugs such as lumefantrine and amodiaquine. Geostatistical modelling then generated continuous spatiotemporal prevalence estimates across the continent, revealing not merely point prevalences but the directionality and pace of spread.
The practical weight of this work is substantial. Artemisinin-based combination therapies remain the cornerstone of malaria treatment for hundreds of millions of people, and validated k13 resistance mutations — well-characterized in Southeast Asia for over a decade — have been accumulating genomic evidence of independent emergence in East and Central Africa. What this modelling effort contributes that individual surveillance studies cannot is the ability to distinguish local de novo emergence from geographic diffusion across borders, a distinction critical for containment strategy. The study's reliance on genotyping data from convenience samples rather than population-representative cohorts is a meaningful limitation, and modelled estimates in regions with sparse sampling carry wider uncertainty. Still, as a synthesis tool informing WHO and national malaria program deployment of next-generation ACTs or triple therapies, this is a genuinely policy-relevant piece of science — confirmatory in parts, and in others, a meaningful advancement in continental-scale resistance surveillance.