Predicting Optimal Antimalarial Drug Combinations from a Standardized Plasmodium falciparum Humanized Mouse Model

The development of new combinations of antimalarial drugs is urgently needed to prevent the spread of parasites resistant to drugs in clinical use and contribute to the control and eradication of malaria. In this work, we evaluated a standardized humanized mouse model of erythrocyte asexual stages of Plasmodium falciparum (PfalcHuMouse) for the selection of optimal drug combinations. The development of new combinations of antimalarial drugs is urgently needed to prevent the spread of parasites resistant to drugs in clinical use and contribute to the control and eradication of malaria. In this work, we evaluated a standardized humanized mouse model of erythrocyte asexual stages of Plasmodium falciparum (PfalcHuMouse) for the selection of optimal drug combinations. First, we showed that the replication of P. falciparum was robust and highly reproducible in the PfalcHuMouse model by retrospective analysis of historical data. Second, we compared the relative value of parasite clearance from blood, parasite regrowth after suboptimal treatment (recrudescence), and cure as variables of therapeutic response to measure the contributions of partner drugs to combinations in vivo. To address the comparison, we first formalized and validated the day of recrudescence (DoR) as a new variable and found that there was a log-linear relationship with the number of viable parasites per mouse. Then, using historical data on monotherapy and two small cohorts of PfalcHuMice evaluated with ferroquine plus artefenomel or piperaquine plus artefenomel, we found that only measurements of parasite killing (i.e., cure of mice) as a function of drug exposure in blood allowed direct estimation of the individual drug contribution to efficacy by using multivariate statistical modeling and intuitive graphic displays. Overall, the analysis of parasite killing in the PfalcHuMouse model is a unique and robust experimental in vivo tool to inform the selection of optimal combinations by pharmacometric pharmacokinetic and pharmacodynamic (PK/PD) modeling.