Projection of doses for QT-prolongation studies based on modeling of the worst-case inhibition of CYP3A

Abstract

Aims To project the probability of “worst-case scenario” for CYP3A-mediated ketoconazole (K) inhibition of CYP3A drugs based on modeling of pharmacokinetic inhibition effect to select doses for high-dose QT prolongation studies for a low extraction ratio (ER) Drug A (range: 0.01–0.03) and high ER Drug B (range: 0.5–1). Both of these drugs are orally absorbed and cleared by CYP3A by unknown fractions (fm). Methods The predicted K portal vein and systemic concentration were used to drive its inhibitory effect on intrinsic clearance and bioavailability of Drugs A and B. Simulations were performed to test worst-case scenarios, including the duration of K dosing and the range of drug's fm and ER. Results Performance of the model was assessed by comparing simulated trials to actual clinical trial results for A and B given with 200 mg K. The model well predicted the 2-fold (A) and 6-fold (B) AUC increases observed in the trials. Assuming worst-case fm and bioavailability, 400 mg K produced less than doubling of AUC compared to results at 200 mg K for both drugs; maximum inhibition was obtained following 3 daily K doses. Doses for QT studies were then selected to achieve the 90th or the 50th quantile of the distribution of AUC ratios, or a 4- and 8-fold exposure-multiple from the target clinical doses, for Drugs A and B, respectively. Conclusions Model-based projection of the probable worst-case CYP3A inhibition can be used for high-dose QT prolongation study design. Clinical Pharmacology & Therapeutics (2005) 77, P31–P31; doi: 10.1016/j.clpt.2004.12.012