Anne Kümmel (1), Daniel Kaschek (1), Bart E. Laurijssens (2), Nathalie Gobeau (3), Fiona MacIntyre (3)
(1) IntiQuan GmbH, Basel, Switzerland (2) BEL Pharm Consulting, Chambonas, France, (3) MMV, Geneva, Switzerland
Introduction: The World Health Organization recommends artemisinin-based combination therapies against malaria. However, the increasing emergence of resistance jeopardizes the efficacy of the current treatments, calling for the continuous generation of new drugs. Medicines for Malaria Venture (MMV) and its partners are developing new drugs and testing new combinations. MMV employs model-based approaches to support selection and progression of individual compounds at different stages of drug development. Recently, the combination of a new compound, artefenomel, with a marketed compound, piperaquine, was tested in patients. For this study, a logistic regression model was built for the fraction of patients with no recrudescence at day 28 as a function of the drug concentrations at day 7 [1]. The model was used to identify covariates driving efficacy and to simulate non-tested dose combinations. However, this model neither provides any information about the time course of parasitemia nor is easily related back to preclinical pharmacology. Also, the ability to extrapolate to different dosing regimens is limited.
Objectives:
- Develop a longitudinal PKPD model, describing the time-course of the parasitaemia after treatment with artefenomel and piperaquine
- Identify significant covariates driving efficacy
Methods: Data from three Phase 2 trials with either artefenomel monotherapy or artefenomel and piperaquine combination therapy was pooled for a stepwise, nonlinear, mixed-effects modeling. First, a PK model for the pooled data was developed and the individual parameter estimates used as regression parameters for the subsequent PD modeling. The PD model described the drug effect with a sigmoidal model linking drug concentrations with parasite clearance. The effect of the combination was the sum of the single drug effects, however, accounting for mutual impact on the maximum effect (Emax) or the potency for one drug (EC50) by the other [2]. The PD model parameters for artefenomel were estimated, and subsequently fixed, based on data from patients treated with artefenomel alone. Interaction parameters were estimated while also estimating the PD parameters for piperaquine or fixing these to values determined in a previous controlled human malaria infection study. Covariates from a preselected set were tested for all PD parameters but the interaction parameters.
Results: The developed longitudinal model described well the observed individual profiles and the fraction of recrudescent patients over time. Region was identified as covariate on the drug effect. Since region was correlated with age and weight, the covariate effect might likewise also be associated with age or weight. Visual predictive checks supported the model’s accuracy in prediction of fractions of recrudescent patients both for the overall and for subpopulations of interest. Fixing the PQP PD parameters on values obtained from the PQP monotherapy challenge study led to biologically plausible parameter estimates and simulations.
Conclusions: The good predictive behavior of the developed model demonstrates the power of longitudinal modeling, over other approaches, for supporting dose selection for clinical study design and different patient populations. Two critical data aspects were highlighted by the analysis: a) availability of monotherapy data for both compounds for combination therapy model building, and b) presence of recrudescence events (treatment failures). In the future, MMV is planning to test both mono- and combination therapy in controlled human malaria infection studies prior to going into patients. This will enable better model-based support of dose and combination selection for Phase II as subtherapeutic doses can be considered, enriching the data with information about treatment failures. Finally, the model will be enriched with patient data as Phase II studies are conducted, and will allow adjustment of doses in specific populations, e.g., children, if required.
References:
[1] Macintyre et al. BMC Medicine 2017 15:181
[2] Wicha et al, Nature Communications (2017) 8, 2129
Reference: PAGE 27 (2018) Abstr 8747 [www.page-meeting.org/?abstract=8747]
Poster: Drug/Disease Modelling - Infection