To examine the effects of repeat oral dosing of rosiglitazone on the pharma
cokinetics of nifedipine, a prototype CYP3A4 substrate, a randomized, open-
label, crossover study was performed with two treatment phases separated by
a washout period of at least 14 days. Twenty-eight healthy male volunteers
received either a single 20 mg oral nifedipine dose or rosiglitazone 8 mg
orally once daily for 14 days with a single 20 mg oral nifedipine dose admi
nistered on day 14. Plasma nifedipine concentrations were determined over t
he 24-hour period following administration of the nifedipine doses. Lack of
effect was defined as the demonstration that the 90% CI was contained enti
rely within a symmetrical 30% range either side of unity on the log(e)-scal
e. Following rosiglitazone + nifedipine administration, the area under the
nifedipine concentration-time curve from time zero to infinity (AUC((0-infi
nity))) was 13% lower than that after administration of nifedipine alone. T
his difference in nifedipine AUC((0-infinity)) was not deemed to be clinica
lly significant since the 90% CI was contained within the protocol-defined
30% range (point estimate for ratio of geometric means 0.87; 90% CI: 0.79,
0.96). Rosiglitazone had no marked effect on nifedipine peak plasma concent
ration (point estimate: 0.99; 90% CI: 0.73, 1.34) or time to peak concentra
tion compared with nifedipine alone. Rosiglitazone coadministration produce
d a small decrease in the mean nifedipine half-life (point estimate: -0.77;
90% CI: mean difference -1.29 h, -0.25 h). Both treatment regimens were we
ll tolerated and associated with a favorable safety profile. Rosiglitazone,
at the highest dose used in clinical studies, produced a small, clinically
insignificant decrease in nifedipine exposure. The very small effect on ni
fedipine pharmacokinetics suggests that rosiglitazone is an extremely weak
inducer of CYP3A4, a characteristic that distinguishes rosiglitazone from t
roglitazone. (C) 1999 the American College of Clinical Pharmacology.