Lack of correlation between in vitro inhibition of CYP3A-mediated metabolism by a PPAR-gamma agonist and its effect on the clinical pharmacokinetics of midazolam, an in vivo probe of CYP3A activity

RG 12525 (2-[[4-[[2-(1H-tetrazole-5-ylmethyl)phenyl]methoxy]phenoxy]methyl] quinolone) is a novel peroxisome proliferator-activated receptor gamma (PPAR-gamma) agonist. In vitro microsomal inhibition assays indicated that RG 12525 is a potent inhibitor of CYP3A4, with a Ki value of 0.5 microM. With the conservative assumption that the total plasma concentration of drug was available to metabolic enzymes following RG 12525 oral administration, marked inhibition of CYP3A4 was expected to substantially reduce the systemic clearance of compounds metabolized by this enzyme. The possibility also existed for inhibition of intestinal and hepatic CYP3A4 by RG 12525 to reduce "first-pass" metabolism and increase absolute bioavailability of CYP3A4 substrates orally coadministered. Consequently, an in vivo drug-drug interaction study was performed to evaluate the effects of orally administered RG 12525 on in vivo CYP3A4 activity in healthy male subjects. The pharmacokinetics of oral midazolam, a probe for intestinal and hepatic CYP3A activity, was not influenced by either the low (100 mg qd for 4 days) or high (600 mg qd for4 days) RG 12525 dosing regimen despite the resulting total plasma concentrations of inhibitor that were well above in vitro Ki values. The point estimates and 90% confidence intervals for the ratios of mean midazolam AUC for subjects administered 100 mg RG 12525 (110.6; 98.7-124.1) and 600 mg RG 12525 (98.4; 84.4-114.7) versus midazolam alone were within 80% to 125%. To explain these results, factors that could limit the accuracy of in vitro models in predicting metabolic drug interactions, mainly the high degree of RG 12525 protein binding (> 99.9%), were considered. The lack of correlation between the in vitro inhibition of CYP3A4 by RG 12525 and the inconsequential effects of this compound on midazolam pharmacokinetics accentuate the need to recognize factors other than plasma drug concentrations and potency of in vitro enzyme inhibition when extrapolating in vitro data to predict in vivo drug-drug interactions.
CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				L. S. Lee, J. S. Bertino Jr, and A. N. Nafziger Limited Sampling Models for Oral Midazolam: Midazolam Plasma Concentrations, Not the Ratio of 1-Hydroxymidazolam to Midazolam Plasma Concentrations, Accurately Predicts AUC as a Biomarker of CYP3A Activity J. Clin. Pharmacol., February 1, 2006; 46(2): 229 - 234. [Abstract] [Full Text] [PDF]