Ex vivo susceptibility to dihydroartemisinin in Plasmodium falciparum patient isolates from Eastern Rwanda, 2025

Abstract Background The emergence of artemisinin partial resistance (AR) in Plasmodium falciparum , characterized by mutations in the Plasmodium falciparum Kelch 13 ( PfK13 ) gene and delayed parasite clearance, represents a significant threat to malaria control in Africa. While the Ring-Stage Survival Assay (RSA) is the phenotypic gold standard for AR detection, its technical complexity limits scalability. The Extended Recovery Ring-Stage Survival Assay (eRRSA), utilizing quantitative PCR (qPCR), has been proposed as a more scalable alternative for molecular surveillance. Methods In early 2025, 19 clinical P. falciparum isolates were collected and analyzed from the Kirehe district of Rwanda. AR phenotypes were characterized using both ex vivo RSA (72-h microscopy) and eRRSA (120-h qPCR). Parasite genotypes were determined via Sanger sequencing of the PfK13 propeller domain. Statistical analyses, including Spearman correlation and Receiver Operating Characteristic (ROC) curves, were applied to compare assay performance and define resistance thresholds. Results Non-synonymous PfK13 mutations were identified in 52.6% (10/19) of isolates, predominantly the validated R561H marker (31.6%). The N490T mutation (2/19) was detected for the first time in Rwanda. RSA-confirmed AR (survival rate >1%) was present in 38.9% of valid assays. The R561H mutation was significantly associated with elevated RSA survival ( P =0.02), and the N490T isolates had high survival rates. Two further rare variants, P667S (2/18) and F699C (1/18), did not show elevated survival rates. eRRSA recovery rates correlated with RSA survival ( P =0.003, ρ=−0.7); however, eRRSA exhibited moderate diagnostic accuracy (AUC = 0.86). A recovery rate threshold of 59.5 yielded high sensitivity (100%) but limited specificity (67%) for identifying RSA-defined resistance. Conclusion These findings confirm the continued expansion of the PfK13 R561H lineage in Rwanda and provide a first ex vivo indication that the N490T variant confers an AR phenotype. While the eRRSA shows potential as a high-throughput screening tool for surveillance due to its high sensitivity, further refinement of recovery thresholds and clinical validation are required to improve its predictive specificity.