The erythromycin breath test reflects P-glycoprotein function independently of cytochrome P450 3A activity

Abstract

The erythromycin breath test (ERBT) has been widely used as a phenotypic measure of cytochrome P450 (CYP) 3A activity in individuals, as well as its modulation by inhibitors or inducers. However, it is not entirely clear what this measure actually reflects because, in addition to CYP3A, animal studies suggest that P-glycoprotein is also involved in erythromycin's hepatic disposition. Thus studies were undertaken to determine the effect of tariquidar, a potent P-glycoprotein inhibitor that does not affect CYP3A activity, on the ERBT and on the CYP3A-mediated metabolism of midazolam, a non-P-glycoprotein substrate. A randomized, double-blind, 2-way crossover trial was performed in 8 healthy subjects involving the intravenous administration of either placebo or tariquidar (150 mg over a period of 30 minutes) on 2 study days 2 weeks apart. On both days, a 1-hour ERBT was performed, followed by determination of midazolam's systemic clearance after a 1-mg intravenous dose. Tariquidar increased the ERBT 1-hour value in all subjects (median, 2.1% [interquartile range (IQR), 1.9% to 3.3%] versus 5.4% [IQR, 3.7% to 7.8%] for placebo and tariquidar, respectively; P = .012), representing a median 2.3-fold (IQR, 1.9- to 3.0-fold) increase. By contrast, midazolam's systemic clearance after tariquidar was unchanged (median change, -4.6% [IQR, -10.2% to 10.7%]; P = .78). Hepatic P-glycoprotein is an important determinant of the ERBT and a potentially confounding factor in interpreting the meaning of the trait measure. In addition, the results demonstrate the dynamic interplay between hepatic drug metabolism and transport of dual CYP3A/P-glycoprotein substrates.