Roberta Bartolucci(1)(2), Anne-Gaëlle Dosne(2), Dénes Csonka(2), Juan Jose Perez Ruixo(2), Paolo Magni(1), Italo Poggesi(2)
(1) University of Pavia - Italy, (2) Janssen Research & Development
Introduction: Macitentan is an orally active, nonpeptide, potent, dual endothelin receptor antagonist, which is approved for the treatment of Pulmonary Arterial Hypertension (PAH) in adults at the dose of 10 mg qd (Opsumit®)[1]. A population pharmacokinetic (popPK) model of macitentan and its active metabolite (ACT-132577) was already reported in the literature[2] and it was based on steady state data from a phase III study, following the administration of macitentan at the dose of 10 mg once daily. In the current work, a popPK model is developed to describe the pharmacokinetics of macitentan and ACT-132577 integrating several clinical studies, in which the drug was given at different dose levels. Therefore, the aim is to provide the most comprehensive description as possible of the pharmacokinetics of macitentan and its metabolite.
Methods: A total of 7597 plasma concentration samples for macitentan and its ACT-132577 were collected from 421 subjects across 8 studies (7 phase I studies on 135 subjects characterized by rich sampling and 1 phase III study in 286 patients with PAH with trough concentrations and a full PK profile at the steady state in a subset of 20 subjects), spanning a dose range between 0.2 and 600 mg.
The development of the popPK model was performed in NONMEM (ICON plc, version 7.3) using the First-Order Conditional Estimation method with INTERACTION (FOCE INTER). Exploratory analysis, diagnostic graphics and post-processing of NONMEM results were performed in R (version 3.4.1). The choice of the best model was based on the assessment of several diagnostics, including the objective function value, the stability of the estimation process, the goodness-of fit (GOF) plots and Visual Predictive Check (VPC) plots.
Results: The final model was an open 1-compartment disposition model for both macitentan and ACT-132577, with first order elimination for the parent and first order formation and elimination for the metabolite. The mean (CV%) macitentan and ACT-132577 elimination half-lives were approximately 15(24%) and 47(12%) hours, respectively, and they are consistent with the current label[3]. Macitentan and ACT-132577 plasma accumulation was 2- and 10-fold when macitentan was given every day at the dose of 10 mg. The estimated systemic clearance of macitentan was 0.048 L/h with a coefficient of variation of 35% and, based on in vitro findings, 62% of the clearance accounted for the formation of ACT-132577.
A complex absorption model was implemented to account for the decrease in macitentan bioavailability and the multiple peaks observed at the higher doses. The absorption phase was therefore described by two parallel first-order absorption pathways for macitentan: a faster with kaf=0.3 h-1 and a slower with kas=0.06 h-1, after a tlag of 13.6 h. The fraction of dose going through each absorption pathway was assumed to be dose-dependent: at lower doses most of the absorption took place via the faster absorption (89% at 10 mg dose), whereas the slower absorption pathway becomes gradually more relevant at higher doses (52% at 600 mg dose). The dose-dependent relative bioavailability was described with an exponential function using 10 mg dose as reference (F=1). Therefore, the estimated bioavailability at 600 mg was about 30%. To account for the pre-systemic formation of the active metabolite, a fraction of the dose (estimated to be 6.6%) was assumed to be directly absorbed as metabolite.
The model adequately described the PK profile of both macitentan and ACT-132577 following single and multiple doses across the entire dose range. The parameter estimations were stable, with relative standard errors less than 30%. The GOF plots did not show evident trends, neither at population nor at individual level, and prediction-corrected VPCs indicated a good adherence of the model to the data.
Conclusion: Several Phase I and Phase III studies were integrated to develop a comprehensive popPK model that was able to describe the pharmacokinetics of macitentan and ACT-132577 in healthy and PAH subjects, following oral administration of macitentan at a wide range of doses. In particular, the model included semi-mechanistic features and it was able to capture the complex absorption processes of macitentan.
Since this population PK model is based on a large dataset, which includes the results of numerous dose levels, it is more suited for predicting the PK behaviour at different dosing regimens and it can be used for a more meaningful analysis of covariates.
References:
[1] T. Patel and K. McKeage. Macitentan: First Global Approval. Drugs. 2014; 74:127–133.
[2] A. Krause, J. Zisowsky, J. Dingemanse. Modeling of pharmacokinetics, efficacy, and hemodynamic effects of macitentan in patients with pulmonary arterial hypertension. Pulmonary Pharmacology and Therapeutics. 2018; 49:140-146.
[3] https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/204410s017lbl.pdf
Reference: PAGE () Abstr 9438 [www.page-meeting.org/?abstract=9438]
Poster: Drug/Disease Modelling - Absorption & PBPK