A Bottom-up Approach for Mutant and Wild Type Collies Using Physiologically Based Pharmacokinetic (PBPK) Modeling: A Case Study Using Loperamide.

Abstract

A bottom-up physiologically based pharmacokinetic (PBPK) model was developed to predict the pharmacokinetics of loperamide, a substrate for multidrug resistance 1 (Mdr1) encoded P-glycoprotein (P-gp), in wild-type dogs (WT) and dogs that are homozygous for a base-pair deletion in the Mdr1 gene encoding for P-gp (Mu, Δ-Mdr1). In vitro-to-in vivo extrapolation (IVIVE) techniques were employed where in vitro data describing loperamide absorption, distribution, metabolism, and elimination (ADME) were extrapolated to in vivo dose exposure predictions. Importantly, by applying system parameters extrapolated from other breeds and published information on Collie-specific physiology, for the first time, a breed-specific whole-body PBPK model for the Collie was developed. Using our loperamide IVIVE-PBPK model (Simcyp Animal Simulator), the observed plasma concentration-versus-time profiles after intravenous and oral loperamide administration were successfully captured. The overall model performance for the WT (n = 7) and Mu (n = 10) Collies was within 1.40 and 1.24, and 1.18 and 1.51 AAFE for the Area under the plasma concentration-time-profile curve (AUC)_0-24 h and maximal plasma concentration (C_max) predictions, respectively. Predicted C_max values were within ± 25% of observed values for 67% of all doses for the WT dogs. For the Mu dogs, the predicted AUC_0-24 h was within 50% for all doses. Our work provides the first example of a systematic approach for Collies and illustrates its use to describe the impact of a known genetic variation in the canine Mdr1 gene. Furthermore, we describe the general workflow for establishing, verifying, and applying an IVIVE-PBPK framework for predicting in vivo drug behavior within a specific canine breed.