Peiying Zuo, Mark Bush, Malcolm Young
Clinical Pharmacology Modeling & Simulation, GlaxoSmithKline, Research Triangle Park, North Carolina, USA
Objectives: Drugs A and B are structurally similar molecules in Phase I development for the treatment of conditions associated with delayed gastric emptying. As part of early safety assessment, probabilistic modeling was used to provide quantification of risk for QTc prolongation in healthy volunteers within and above the range of predicted therapeutic doses.
Methods: Data were collected from first time in human studies of drug A (N=48, 1-150mg) and B (N=17, 10-125mg). A nonlinear mixed-effect analysis of the concentration-QT relationship was performed in NONMEM 6.0 (ICON Development Solutions) with time-matched PK and ECG parameters over 48 hours [1, 2]. Simulations were conducted for predicted therapeutic doses of both molecules in ModelRisk 4.0 (Vose Software) to calculate the risk profile, i.e. percentage of subjects with upper 90%CI of ddQTci (QT corrected individually by RR, baseline and placebo)>10msec. Random simulations (N=1000) was conducted in 100 subjects to generate QTci based on the final PK/QT model and correlation of RR at baseline and Tmax for active doses and placebo and at Cmax values randomly sampled from observed distribution. These results were then used to calculate ddQTci used for risk assessment and compared to the predicted QT prologation directly calculated from 90%CI of NONMEM slope estimate at observed Cmax for both molecules.
Results: The probabilistic approach suggested that the likelihood of having upper 90%CI of ddQTci above 10msec is minimal at all doses tested for drug A with a median upper 90%CI of 1.3msec at the top dose studied. The likelihood for drug B is minimal at 10mg (0.3%, median upper 90%CI of 3.3msec) but increased to 40% at 80mg (median upper 90%CI of 9.2msec). The calculation for QTc based on 90%CI of slope and observed mean Cmax suggested an upper 90%CI of 2.5msec at the top dose for drug A, 1.8msec for drug B at 10mg and 20msec at 80mg. Overall, probabilistic simulations are in agreement with conclusions derived from the slope-Cmax calculations.
Conclusions: The probabilistic approach was used to characterize the risk of QT prolongation for two Phase I molecules from the same class. It provided a more thorough evaluation of QTc prolongation and allowed the clinical team to evaluate and compare the safety profiles, contributing to candidate progression decisions.
References:
[1] ICH E14. Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-Antiarrhythmic Drugs. October 2005.
[2] V. Piotrovsky. Pharmacokinetic-Pharmacodynamic Modeling in the Data Analysis and Interpretation of Drug-induced QT/QTc Prolongation. AAPS J 2005; 7 (3): 610-622.
Reference: PAGE 21 () Abstr 2650 [www.page-meeting.org/?abstract=2650]
Poster: Safety (e.g. QT prolongation)