Combining population with physiologically-based pharmacokinetic (PBPK) models for oral drug absorption: Predicting the segmental bioavailability differences of R-oxybutynin and its main metabolite using a middle out approach.
Andrés Olivares-Morales (1), Avijit Ghosh (2), Amin Rostami-Hodjegan (1,3) and Leon Aarons (1)
(1) Centre for Applied Pharmacokinetic Research, Manchester Pharmacy School, University of Manchester, Manchester, UK.(2) Janssen Pharmaceutica, Spring House, PA, USA. (3) Certara, Blades Enterprise Centre, Sheffield, UK.
Objectives: The minimal Segmented Absorption and Transit (mSAT) model has been recently proposed and applied to mechanistically predict the bioavailability differences observed for oxybutynin’s (OXY) OROS formulation compared to its immediate release (IR) tablet [1]. The aim of the present study was to expand and refine this model for the pharmacokinetic (PK) prediction of both R-OXY and its main metabolite, N-desethyloxybutynin (DEOB), IR and OROS, and to describe the high interindividual variability (IIV) observed in their PK response using the so-called “middle out” approach to develop a population PBPK (pop-PBPK) model for OXY and DEOB
Methods: The mSAT model was expanded to account for the formation of R-DEOB in the gastrointestinal tract and the liver, whereas its disposition was described by a semi-physiological model. The model was implemented in NONMEM 7.3 as a system of 26 ODEs using the ADVAN13 subroutine and then fitted to R-OXY/DEOB plasma concentrations obtained after the administration of three IR tablets (over 24 hours) to 41 healthy volunteers with rich sampling. The parameter estimation was aided with the use of NONMEM’s prior functionality to obtain maximum a posteriori (MAP) estimates [2]. The goodness of fit was evaluated by visual predictive checks (VPCs) and the model performance was evaluated by simulating the PK profiles of R-OXY/DEOB after the administrations of one 10 mg OROS tablet as well as by simulating a drug-drug interaction (DDI) study between OXY OROS and ketoconazole (KTZ) 200mg BID
Results: The model provided a good fit to the observed data of R-OXY/DEOB for the IR tablets, as seen in the VPCs. Using the parameter estimates obtained from the IR fit and the in vitro release profile as input parameters, the mSAT model was able to capture both the PK and IIV observed for R-OXY/DEOB after the administration of a 10mg OROS formulation. The simulated DDI study between OXY OROS and KTZ BID predicted a 2.5 to 3.2 fold increase (mean 2.8) in the exposure of the parent drug, whereas for the metabolite the exposure remained almost unaffected. This was consistent with information provided in the product label [3].
Conclusions: The pop-PBPK model allowed the refinement of the previously developed mSAT model for OXY and the description of the high IIV observed in PK profiles. This approach allowed a better understanding of the PK of OXY and its metabolite.
References:
[1] Olivares-Morales A. Aarons L., Rostami-Hodjegan A. Predicting the Higher Bioavailability Observed for Oxybutynin’s OROS Formulation Compared to the Immediate-Release (IR) Tablet Using a Novel Simplified Absorption PBPK Model. Presented at 2015 AAPS Annual Meeting and Exposition; Poster T33171 [http://abstracts.aaps.org/Verify/AAPS2015/PosterSubmissions/T3317.pdf]
[2] Gisleskog, P., et al. (2002) Use of Prior Information to Stabilize a Population Data Analysis. Journal of Pharmacokinetics and Pharmacodynamics 29(5-6): 473-505.
[3] U.S. Food and Drug Administration (2015). Ditropan XL® Product Label. Retrieved 06/01/2015, 2015, from http://www.accessdata.fda.gov/drugsatfda_docs/label/2003/17577se8-033,18211se8-016,20897slr010_ditropan_lbl.pdf