Evaluation of in vitro/in vivo correlations for three fentanyl transdermal delivery systems using in vitro skin permeation testing and human pharmacokinetic studies under the influence of transient heat application

Abstract

The value of developing an in vitro/in vivo correlation (IVIVC) is substantial in biopharmaceutical drug development because once the model is developed and validated, an in vitro method may be used to efficiently assess and predict drug product performance in vivo. In this study, three bioequivalent, matrix-type, fentanyl transdermal delivery systems (TDS) were evaluated in vitro using an in vitro permeation test (IVPT) and dermatomed human skin, and in vivo in human pharmacokinetic (PK) studies under harmonized study designs to evaluate IVIVC. The study designs included 1 h of transient heat application (42 ± 2°C) at either 11 h or 18 h after TDS application to concurrently investigate the influence of heat on drug bioavailability from TDS and the feasibility of IVPT to predict the effects of heat on TDS in vivo. Level A (point-to-point) and Level C (single point) IVIVCs were evaluated by using PK-based mathematical equations and building IVIVC models between in vitro fraction of drug permeation and in vivo fraction of drug absorption. The study results showed that the three differently formulated fentanyl TDS have comparable (p > 0.05) heat effects both in vitro and in vivo. In addition, the predicted steady-state concentration (Css) from in vitro flux data and the observed Css in vivo showed no significant differences (p > 0.05). However, the effects of heat on enhancement of fentanyl bioavailability observed in vivo were found to be greater compared to those observed in vitro for all three drug products, resulting in a weak prediction of the impact of heat on bioavailability from the in vitro data. The results from the current work suggest that while IVPT can be a useful tool to evaluate the performance of fentanyl TDS in vivo with a relatively good predictability at a normal temperature condition and to compare the effect of heat on drug delivery from differently formulated TDS, additional testing measures would enhance the ability to predict the heat effects in vivo with a lower prediction error.