Marylène Gaignet, Sylvain Fouliard, Marylore Chenel
Clinical Pharmacokinetics Department, Institut de Recherches Internationales Servier, France
Introduction: According to in vitro – in vivo correlation (IVIVC) guideline [1], in vitro dissolution data can be used early in drug development process to help designing/refining formulations, or at later stages, as a surrogate for bioequivalence studies for qualifying formulation scale-up or post-approval changes.
Objectives: An IVIVC model, based on a population compartmental approach, was implemented to predict the in vivo drug concentration-time profiles (CTP) with double-peak absorption of 3 different modified release (MR) formulations from their respective in vitro dissolution fraction-time profiles (DFTP).
Methods: In vivo CTP were obtained after oral administration of one immediate release (IR) and 3 MR formulations of molecule S (with 3 different release rates: fast, intermediate, and slow, showing double-peak absorption), in a single cross-over trial, performed in 13 healthy volunteers. The IR CTP were fitted with a population PK model. Then, 6 in vitro DFTP of each of the 3 MR formulations were fitted with a non-linear mixed-effect model. At last, IVIVC model consisted in a unique relationship for all MR formulations between in vitro DFTP and in vivo DFTP in which individual IR PK model were plugged. Internal validation (model built with 3 formulations) and external validation (model built using slow and fast formulations only, then used to predict intermediate form) were performed by computing the percentage prediction errors (%PE) on Cmax and AUC.
Results: IR CTP were fitted using a 2-compartment model with a first order absorption rate and a lag-time; inter-individual variability was estimated on clearance and central volume. For each MR formulation, in vitro DFTP were fitted with a Gompertz function, and inter-tablet dissolution variability was estimated. IVIVC model involved a 3-phase non-linear time scaling (no change in first phase, second phase with a stopped dissolution, and third phase with a delayed dissolution) and inter-individual variability on tablet dissolution and relative bioavailability terms. Internal and external validations showed good prediction abilities of the model. The non-linear scaling allowed the description of double-peak absorption.
Conclusions: This work illustrates another application of a population PK approach [2] to IVIVC model building, adding an accurate individual description of complex absorption CTP, which is valuable for evaluation of further formulation in development.
References:
[1] Food and Drug Administration (1997) Guidance for Industry: Extended Release Oral Dosage Forms: Development, Evaluation, and Application of In vitro/In vivo Correlations.
[2] Claire Gaynor, Marylène Gaignet, Marylore Chenel, Developing an In Vitro – In Vivo Correlation Model Using a Population Approach in NONMEM, PAGE 20 (2011) Abstr 2099 [www.page-meeting.org/?abstract=2099]
Reference: PAGE 21 () Abstr 2505 [www.page-meeting.org/?abstract=2505]
Poster: Other Modelling Applications