"The new mathematical model we have developed allows us to predict how combinations may behave in patients before clinical trials"

2021
Dr Nathalie Gobeau

ACPR28 modelling for combinations

Antimalarial drugs can be combined to increase efficacy, delay resistance and prolong their clinical utility, but how can we better understand how individual drugs will interact when administered together? The World Health Organization (WHO) and regulatory authorities assess drug combinations by looking at the number of people who achieve an adequate clinical and parasitological response 28 days from the start of treatment (ACPR28). For a drug to be approved, ≥ 95% of patients must achieve ACPR28. To help MMV decide which drug combinations have such potential, pharmacometric1 scientists use in vitro2 experiments, laboratory models and clinical trials.

Mathematical models are built from this data to characterize drug combinations and predict the drug dose required. This information helps MMV and partners make decisions driven by data, prioritizing combinations that can feasibly be combined to reach the required target efficacy. These predictive models are becoming an important tool for efficiently identifying new antimalarial drug combinations. For example, in 2020, ACPR28 modelling was successfully used by MMV’s Malaria Drug Development Catalyst to analyse and rank six drug combinations. Notably, this helped prioritize four promising combinations, saving resources and time.

In this interview, Dr Nathalie Gobeau, Director of Pharmacometrics at MMV, discusses the use of ACPR28 modelling in investigating antimalarial drug combinations.

1. Why do we need these new modelling tools? What are their advantages and what challenges do they help us overcome?

The new mathematical model we have developed allows us to predict how combinations may behave in patients before clinical trials, which would be much more expensive than establishing and running a model. These models enable us to incorporate all the knowledge available on compounds from different sources and predict the clinical outcome. Based on simulations for six combinations, four are being progressed to clinical trials. We hope that these tools will help select the most promising combinations for further development.

2. Who did MMV partner with to develop them?

We worked with the mathematical modelling company IntiQuan based in Basel, Switzerland, to develop our algorithm, with additional support from Daniel Lill, a student from the University of Freiburg, Germany. We built our pharmacokinetic/pharmacodynamic models in collaboration with Prof. Sebastian Wicha from the University of Hamburg. To obtain information on the pharmacological interaction, Claudia Demarta-Gatsi from the MMV team worked closely with the Swiss Tropical and Public Health Institute to set up in vitro assays, and with The Art of Discovery in Spain to set up laboratory experiments.

3. Now that you and your team have developed these tools, what are the next projects in the pipeline?

In the future, we aim to look at combinations of three or more compounds using similar models. This would allow us to explore the combination of additional characteristics in one medicine, such as transmission-blocking and parasite clearance. We would also like to extend the model to look at chemoprophylaxis.


 

1. The methods and use of models for disease and pharmacological measurement.

2. A laboratory process that is not conducted in a living organism.