Philippe Pierrillas (1), Pietro Scalfaro (2), Patrice André (3), Raphaël Darteil (2), Elise Roy (2), Diane Sampson (2), Jacky Vonderscher (2), Christian Laveille (1)
(1) Calvagone, Liergues, France / (2) ENYO Pharma SA, Lyon, France / (3) Centre International de Recherche en Infectiologie (CIRI) - INSERM U1111 – CNRS UMR5308 - Université Lyon 1- ENS de Lyon
Objectives: EYP001a is a modulator of the nuclear farnesoid X receptor (FXR) and part of the agonist family which was originally discovered for a therapy of non-alcoholic steato-hepatitis, primary biliary cholangitis and metabolic syndrome. It was also found to have anti-viral activity on Hepatitis B virus (HBV) [1].
The objective of this work was to develop a population Pharmacokinetic-Pharmacodynamic model (PK-PD) using biomarker data to assess the effect of EYP001a on bile acids pathway in both healthy volunteers and HBV infected patients.
Methods: Data from four phase 1 studies (including a 4-arm cross-over study to evaluate the impact of food intake and a potential nycthemeral rhythm and a drug-drug interaction study) conducted in healthy volunteers and HBV-infected patients were included in this analysis. Plasma samples from 180 individuals after single and repeated administrations of EYP001a at different dose levels (from 30 to 800 mg) and different dosing regimen (once a day and twice a day) and placebo were analysed for EYP001a and C4 concentrations (i.e. 7-alpha-hydroxy-4-cholesten-3-one, intermediate in the synthesis of bile acids from cholesterol located in the liver). A covariate analysis, using a stepwise approach, was performed to assess the impact of covariates and also to investigate the potential differences between healthy volunteers and HBV infected patients.
Parameters were estimated with the First-Order Conditional Estimation method with Interaction (FOCE-I method) implemented in NONMEM 7.3 (ICON) and model development was guided by residual- and simulation-based diagnostics.
Results: EYP001a PK was best described using a 2-compartment disposition model and an absorption phase modelled using 4 transit compartments. Relative bioavailability appeared to be nonlinear with time and dose: bioavailability decreases when dose increases after repeated administration. HBV infected patients were found to have a lower clearance (~25%) compared to healthy volunteers, and administration of EYP001a under fed condition decreased the absorption rate by a 2-fold factor but with a similar exposure.
C4 time-course in the placebo arm was modelled using a turn-over model and a cosine function was used to describe the nycthemeral rhythm of C4 production. EYP001a inhibitory effect was modelled using an effect compartment [2] and a steep sigmoidal function (coefficient of sigmoïdicity >2) on the C4 production.
Model evaluation by goodness-of-fit plots and Visual Predictive Checks, were satisfactory.
Simulations performed from the final population PK-PD model showed that administration of either 200 mg twice a day or 400 mg once a day would be suitable in order to reduce C4 plasma levels.
Conclusions: EYP001a and C4 concentrations were adequately described by the elaborated model and confirm the impact of EYP001a on bile acids pathway. This model might be expanded to safety aspects as they will be available in order to assess the benefice-risk balance and better identify the recommended dosing regimen.
References:
[1] Radreau, P. et al. Reciprocal regulation of farnesoid X receptor activity and hepatitis B virus replication in differentiated HepaRG cells and primary human hepatocytes. FASEB J. 30, 2016 Sep;30(9):3146-54.
[2] Sheiner, L.B. et al, Simultaneous modeling of pharmacokinetics and pharmacodynamics: application to d-tubocurarine. Clin. Pharmacol. Ther. 1979; 25:358–371.
Reference: PAGE 28 (2019) Abstr 8911 [www.page-meeting.org/?abstract=8911]
Poster: Drug/Disease Modelling - Other Topics