A pharmacokinetic-pharmacodynamic model for ECG pattern changes in dog and monkey
F. Jonsson(1), A. Fleury(2), T. Lavé(2)
(1) Pharsight, a Certara company, St. Louis, MO, USA; (2) F. Hoffmann-La Roche Ltd., Modeling and Simulation Group, Drug Metabolism and Pharmacokinetics Department, Basel, Switzerland
Objectives: Drug-induced prolongation of cardiac repolarization is a significant reason for early-phase failure of drug candidates. Changes in PR interval and QRS complex that raised safety concerns were observed following administration of potential CNS compound R1551 to dog and monkey. The aim of the study was to investigate the use of non-linear mixed-effects PK-PD modeling to relate R1551 exposure to the changes of the ECG pattern.
Methods: Using a crossover design, eight beagle dogs received single oral doses of R1551 (ranging 0-100 mg/kg) and seven cynomolgus monkeys received oral doses of 0-30 mg/kg once daily for 5 days. Telemetry was used to measure ECG parameters (PR interval and QRS complex) and heart rate up to ≥24 h after the final dose. Pharmacokinetic parameters were estimated by fitting a two-compartment model to the data. This estimation included data from 32 additional dogs and 9 additional monkeys given peroral or intravenous doses. The model-predicted individual concentration-time profiles of R1551 were then linked to the PR interval and QRS complex via an effect compartment.
Results: A linear concentration-effect model provided a good fit to the observed data for both ECG parameters in dogs and QRS data in monkeys. In monkeys, there was no effect delay and the effect compartment was removed. For the PR interval in monkeys, fit was improved by the addition of a placebo effect compartment to the linear model, to account for a suspected vehicle effect. The estimated slope factors for the PR interval were 0.0093 and 0.0934 ms mg-1 kg-1 l-1 in dog and monkey, respectively. For the QRS prolongation, the corresponding slope factors were 0.00274 and 0.00200 ms mg-1 kg-1 l-1.
Conclusions: The modeling approach used in our study shows the value of characterizing the complete time-course of a drug when assessing its safety profile. The use of a separate effect compartment accounts for the possible presence of active metabolites and for the distribution to the active site. The similarity of the slope factors between dog and monkey indicates that the effect on the ECG pattern is similar in magnitude between species, even if its timing is different. This is an indication that a similar magnitude of ECG change would be expected in humans.
