Frederico Martins1, Anis Krache 2, Eric Helmer1,Florence Namour1 ,Amit Taneja1
1: Galapagos SASU, 102 avenue Gaston Roussel, 93230 Romainville, France :2 : Université Toulouse III - Paul Sabatier - 118 route de Narbonne 31062 Toulouse, France
Objectives: GLPG1205 is a novel, potent and selective antagonist of G protein-coupled receptor 84 (GPR84), activated by medium chain fatty acids. It is more than 100 fold selective for GPR84 over 123 other GPCRs, including free fatty acid homologs. Physiologically based pharmacokinetic (PBPK) modelling is a key component in the movement toward in vitro-based risk assessment, providing a tool to integrate diverse experimental data and mechanistic information to relate in vitro effective concentrations to equivalent human exposures [1]. The objective of the work was to compare the predictive performance of allometric model (AM) with a (PBPK) model to predict clearance, area under the concentration-time curve (AUC) and maximum concentration (Cmax) and Tmax ( time to maximum concentration ), following administration of single oral doses to healthy human volunteers.
Methods: A whole-body PBPK model of GLPG1205 was established with PK-Sim® modelling software (Version 7.3.0) [2]. Physicochemical parameters incorporated from experimental data were logP = 2.38, pka= 1.34, thermodynamic solubility pH 7.4 = 5.18 µg/mL, fu=0.029, the absorption rate was determined in vitro by Caco-2 permeability and clearance from in vitro human hepatocytes. The volume of distribution was calculated based on Poulin & Theil model [3]. Allometric scaling of clearance and volume of distribution obtained from a single dose study in monkeys was performed. Data from a first-in-human volunteer study were used for performance verification and model refinement. Pharmacokinetic (PK) parameters (AUC, Cmax, Tmax) were calculated using PKsim (version 7.3). Overall prediction performance was evaluated based on performance indicators (Absolute average fold error-AAFE).
Results: A PBPK model was developed for GLPG1205 to describe the tissue-specific absorption, distribution, metabolism, and excretion. Model based simulations of GLPG1205 plasma concentrations in humans showed that Tmax was attained in 2-4 hr, with dose proportional increase in Cmax and AUC over 10 to 600 mg single dose . The PBPK model was able to predict the prolonged elimination half-life (observed 32±12hr, predicted 47±20hr) against 3.7 hr of classical allometry . Predictions with the PBPK model showed 0.5 to 2-fold prediction error compared to 4 to 10-fold prediction error with allometric scaling, compared to observed clinical data.
Conclusions: Allometric model entails scaling of clearance and volume of distribution and resulted in a lower accuracy in human PK prediction than PBPK model, which include the mechanistic parametrization of biologic processes. This case study is an example of in vitro to in vivo extrapolation (IVIVE) approach for PBPK model development. While Allometric and PBPK models may at times be comparable, PBPK is of particular value for drugs with scaling challenges due to poor ADME properties.
References:
[1] Zhao, P. , Zhang, L. , Grillo, J. , Liu, Q. , Bullock, J. , Moon, Y. , Song, P. , Brar, S. , Madabushi, R. , Wu, T. , Booth, B. , Rahman, N. , Reynolds, K. , Gil Berglund, E. , Lesko, L. and Huang, S. (2011), Applications of Physiologically Based Pharmacokinetic (PBPK) Modeling and Simulation During Regulatory Review. Clinical Pharmacology & Therapeutics, 89: 259-267.
[2] Stefan Willmann, Jörg Lippert, Michael Sevestre, Juri Solodenko, Franco Fois, Walter Schmitt,PK-Sim®: a physiologically based pharmacokinetic ‘whole-body’ model,BIOSILICO,4: 121-124.
[3]Patrick Poulin, Frank-Peter Theil.A (2000) Priori Prediction of Tissue:Plasma Partition Coefficients of Drugs to Facilitate the Use of Physiologically-Based Pharmacokinetic Models in Drug Discovery,Journal of Pharmaceutical Sciences,Volume 89, ,16-35.
Reference: PAGE 28 (2019) Abstr 8825 [www.page-meeting.org/?abstract=8825]
Poster: Drug/Disease Modelling - Absorption & PBPK