Introduction:
The rise of multidrug resistant malaria demands the urgent development of novel antimalarial compounds. For a drug to achieve registration and availability in the market, new treatments must undergo a series of rigorous evaluations, from preclinical studies in human volunteers experimentally infected with malaria, through to Phase 3 clinical trials in malaria patients. Pharmacokinetic-pharmacodynamic (PK-PD) models, that relate antimalarial drug concentrations with the parasite-time profile, aid in assessing dosing schedules for new antimalarial treatments. We performed a simulation study to evaluate the utility of a Bayesian hierarchical mechanistic PK-PD model for predicting parasite-time profiles for a Phase 2 study of a new antimalarial drug, cipargamin.
Methods/Approach:
We simulated cipargamin concentration and parasitaemia profiles that reproduce the observed profiles for 8 patients enrolled in a Phase 2 volunteer infection study (McCarthy J et. al. AAC 2021). The volunteer infection study inoculated patients with malaria parasites and measured parasite burden frequently before administering low-dose cipargamin 7 days post-inoculation. We simulated these pre- and post-treatment parasite profiles using our biologically informed PD model, which captures the cyclical growth and death processes of the parasite during blood-stage infection. PK-PD parameters were estimated using a Bayesian hierarchical model with STAN, and the process repeated for each of 1000 simulated datasets of 8 patients, each dataset mimicking the Phase 2 study design.
Results:
The population PK model parameters describing the absorption, distribution and clearance of cipargamin were estimated with minimal bias (relative bias ranged from 1.7 to 8.4%) and the posterior predictive checks captured the simulated PK profiles. The PD model was fitted to the parasitaemia profiles of each simulated dataset using the estimated PK parameters derived from the hierarchical PK modelling. The bias of the estimated population average PD parameters was low-to-moderate in magnitude. However, the posterior predictive checks demonstrate that our PK-PD model successfully captures the central trend and variability of both the pre- and post-treatment simulated PD profiles.
Discussion:
This simulation study demonstrates the utility of our Bayesian PK-PD model in predicting parasitological outcomes in Phase 2 volunteer infection studies. This work will help inform the dose-effect relationship of cipargamin, guiding decisions on dosing regimens to be evaluated in Phase 3 trials.
References
McCarthy J, et. al. Defining the Antimalarial Activity of Cipargamin in Health Volunteers Experimentally Infected with Blood-Stage Plasmodium falciparum. Antimicrobial Agents & Chemotherapy, 2021.