977Mechanistic within-host modelling to fast-track the selection of new antimalarial combination therapies.

Background The efficacy of artemisinin-based combination therapies (ACTs), currently the first-line antimalarial treatments, is declining due to the emergence of resistance of malaria parasites to these drugs. This has led drug development initiatives to search for novel combination therapies to replace the failing ACTs. We developed a biologically informed within-host model, validated against data from volunteer infection studies, to guide critical drug development decisions. Methods A within-host model was developed, linking drug concentrations of two novel antimalarial drugs, OZ439 and DSM265, to their combined killing action and accounting for differential killing of these compounds against stages of the parasite's lifecycle. Data collected from malaria-infected volunteers treated with OZ439–DSM265 were used to estimate the model parameters in a hierarchical Bayesian framework. Posterior-predictive simulations of the model were used to determine the dosing regimen required to cure >90% patients. Results The results showed that 800 mg of OZ439 combined with 450 mg of DSM265, which are within the safe and tolerable dose range, can provide day 42 cure rates >90%, despite the estimated antagonistic interaction between the drugs. The importance of accommodating parasite age specificity of drug action was demonstrated. Conclusions The dosing regimens for the combination of OZ439-DSM265 determined from our data-informed in silico model suggest this compound may be a suitable candidate to replace failing ACTs. Key messages Assessing various scenarios within a simulation framework allows discovery of robust dosing regimens, accelerating the drug development process and ensuring efficient allocation of resources for phase 2 and 3 clinical trials. [ABSTRACT FROM AUTHOR]