Pharmacokinetic consequences of time-dependent inhibition using the isolated perfused rat liver model

Mathematical models exist to describe the pharmacokinetic changes caused by time-dependent inhibition (TDI) and these have been reported to predict accurately for several marketed drugs. However, their robustness using in-depth, carefully controlled pre-clinical studies has yet to be established. In the current study, the isolated perfused rat liver was employed to investigate the effects of TDI under carefully controlled conditions. A five-compartmental model was used to describe the observed data and bring context to in vitro TDI data (k_inact and K_I). Co-administration of midazolam with troleandomycin, mifepristone, erythromycin and the discovery compound, AZ-X, increased midazolam area under the curve (AUC) 3.2-, 2.5-, 1.6- and 1.0-fold, respectively, compared with AUC increases of 1.8-, 1.4-, 1.2- and 1.1-fold predicted by the model. These experimental findings, whilst modest in overall effect, support the use of this model in the rat and it is proposed that projections can be made for the likely clinical impact of novel time-dependent inhibitors in man based on predicted human pharmacokinetics and TDI potency determined in vitro.