Use of PBPK modelling to extrapolate in vitro porcine ear skin permeability to in vivo human dermal absorption

Abstract

Porcine ear skin is widely accepted as a surrogate for human skin in in vitro permeation tests (IVPT); however, differences in permeability between the two have been reported in several studies. Physiologically based pharmacokinetic (PBPK) modelling has been recognised as a valuable tool for supporting the correlation and extrapolation of in vitro data to human in vivo conditions. In the present study, the feasibility of using IVPT data obtained from porcine ear skin was evaluated for the optimisation of model input parameters to improve the prediction of systemic exposure to substances in humans under in vivo conditions. To this end, PBPK models were developed and optimised for three compounds: caffeine, crisaborole, and ketoprofen, using Multi-phase multi-layer mechanistic dermal absorption (MPML MechDermA) model within Simcyp® Simulator version 23. First, transdermal permeation models were constructed and refined based on data from IVPT studies conducted using porcine skin sourced from scientific literature. Subsequently, human in vivo PBPK models were developed, either incorporating or excluding the optimised skin partition coefficients (Kp), and the predictive performance of both approaches was compared. In all cases, the stratum corneum:lipid vehicle Kp was identified as a key parameter for optimisation. The results demonstrated that, with appropriate model optimisation and incorporation of mechanistic parameters, in vitro data from porcine skin can enhance the accuracy in predicting in vivo human transdermal permeation of caffeine, crisaborole, and ketoprofen, and can be effectively used to simulate systemic exposure following topical application.

Keywords: Caffeine; Crisaborole; Human; Ketoprofen; PBPK models; Pig skin; Transdermal delivery.