(T-017) Application of an advanced gut PBPK model to characterize intestinal transport and P-glycoprotein-mediated drug-drug interaction: a case study with Digoxin and Clarithromycin

Kazuya Ishida, PhD – Senior Scientist, Gilead; Xiaomin Liang, PhD – Senior Scientist, Gilead; Jia Hao, PhD – Associate Scientist, Gilead; Christine Wan, PhD – Research Associate, Gilead; Yurong Lai, PhD MD FAAPS – Executive Director, Gilead

Objective: The present study assessed the application of a full physiologically-based pharmacokinetic (PBPK) modeling approach with an advanced gut model to characterize the role of intestinal efflux transporters and passive permeability on determining pharmacokinetic profiles and intestinal clearance of Digoxin, a P-glycoprotein (P-gp) substrate, and to predict P-gp mediated drug-drug interaction (DDI) with Clarithromycin, a P-gp inhibitor.

Methods: A full PBPK model with a multi-layer gut wall within an advanced dissolution, absorption, and metabolism (M-ADAM) model was built in SimBiology® (MathWorks). Plasma/serum concentration data of Digoxin in humans after intravenous (IV) and oral (PO) administration with and without Clarithromycin PO coadministration were obtained from the literature^(1-5). The model was fitted to human Digoxin plasma/serum concentration data after IV and PO doses simultaneously to estimate efflux clearance in the intestine and other pharmacokinetic (PK) parameters. The model was then co-fitted with a Clarithromycin PBPK model to estimate the reduction of P-gp-mediated efflux of Digoxin in the intestine by Clarithromycin. In vitro transcellular transport of Digoxin was evaluated using Caco-2 cells.

Results: Human clinical PK profiles of Digoxin were captured well by the model in an IV and PO dose range of 0.25 and 1 mg. Digoxin intestinal clearance was explained by the mechanism of P-gp-mediated efflux and passive diffusion. In vitro Caco-2 assays showed that there was no P-gp efflux saturation present in the tested concentration range (0.3-100 µM). When co-fitted and co-simulated with Clarithromycin, the model captured a 3.8- and 2.8-fold increase in digoxin C_max and AUC, respectively, due to P-gp inhibition following a 0.75 mg PO dose of Digoxin and 250 mg PO twice daily dose of Clarithromycin.

Conclusions: A full PBPK modelling approach with M-ADAM gut description was applied to predict human pharmacokinetic profiles and intestinal P-gp-mediated DDIs. The platform model implemented in SimBiology allows the incorporation and exploration of the interplay between different intestinal transporters through global sensitivity analysis. The model can be further extended to characterize and predict complex DDIs, such as interactions with multiple transporters and more than two drugs.

Citations:
Kramer WG et al., Pharmacokinetics of digoxin: relationship between response intensity and predicted compartmental drug level in man. J Pharmacokinet Biopharm, 7, 47-61 (1979).

Rengelsausen J et al., Contribution of increased oral bioavailability and reduced nonglomerular renal clearance of digoxin to the digoxin-clarithromycin interaction. J Clin Pharmacol, 56, 32-38 (2003).

Ueda H et al., Studies with tritiated digoxin in renal failure and diabeties mellitus. Jap Heart J, 8, 369-377 (1967).

Chu SY et al., Absolute bioavailability of clarithromycin after oral administration in humans. Antimicrob Agents Chemother, 36, 1147-1150 (1992).

Kees F et al., Serum and cellular pharmacokinetics of clarithromycin 500 mg q.d. and 250 mg b.i.d. in volunteers. Infection, 23, 168-172 (1995).