Heterotropic Allosteric Modulation of CYP3A4 In Vitro by Progesterone: Evidence for Improvement in Prediction of Time Dependent Inhibition for Macrolides.

Abstract

Predictions of drug-drug interactions resulting from time-dependent inhibition (TDI) of CYP3A4 have consistently overestimated or mis-predicted (i.e. false positives) the interaction that is observed in vivo. Recent findings demonstrated that the presence of the allosteric modulator progesterone (PGS) in the in vitro assay could alter the in vitro kinetics of CYP3A4 TDI with inhibitors that interact with the heme moiety, such as metabolic-intermediate complex (MIC) forming inhibitors. The impact of the presence of 100 µM PGS on the TDI of molecules in the class of macrolides typically associated with MIC formation was investigated. Presence of PGS resulted in varied responses across the inhibitors tested. The TDI signal was eliminated for five inhibitors, and unaltered in the case of one, fidaxomicin. The remaining molecules erythromycin, clarithromycin, and troleandomycin, were observed to have a decrease in both potency and maximum inactivation rate ranging from 1.7-fold to 6.7-fold. These changes in TDI kinetics led to a >90% decrease in inactivation efficiency. In order to determine in vitro conditions that could reproduce in vivo inhibition, varied concentrations of PGS were incubated with clarithromycin and erythromycin. Resulting in vitro TDI kinetics were incorporated into dynamic physiologically-based pharmacokinetic (PBPK) models to predict clinically observed interactions. The results suggested that a concentration of ~45 µM PGS would result in TDI kinetic values that could reproduce in vivo observations and could potentially improve predictions for CYP3A4 TDI. Significance Statement The impact of the allosteric heterotropic modulator progesterone on the CYP3A4 time-dependent inhibition kinetics was quantified for a set of metabolic-intermediate complex forming mechanism-based inhibitors. We identify the in vitro conditions that optimally predict time-dependent inhibition for in vivo drug-drug interactions through dynamic physiologically-based pharmacokinetic modeling. The optimized assay conditions improve in vitro to in vivo translation and prediction of time-dependent inhibition.