Separating Signal from Noise: PK/PD modelling of QT-interval prolongation

PowerPoint of poster

Background:  The presence of a prolonged QT-interval has become an identifier for the risk of a unique form of polymorphic ventricular tachycardia, Torsade de Pointes (TdP). Since this finding can be a serious safety issue, policies and guidelines have been proposed to ensure that the effects of non-cardiovascular drugs on QT-interval are accurately characterised. Such policies have assumed that ECG measurements are highly reproducible. However, there is convincing evidence from clinical research that QT-interval assessments can show high variability if considered over a wide time span. Therefore, any meaningful attempt to characterise drug-induced changes in ECG parameters requires identification of variability sources, as they will have major impact on clinical study design and sample size.

Objective:  The primary objective of this investigation was to develop a pharmacokinetic / pharmacodynamic (PK/PD) model to describe the time course and variability of QT-interval in healthy subjects. In addition, it was our aim to establish the relevance of external factors on the accuracy and reproducibility of ECG measurements.

Methods:  30 healthy subjects were given a single oral dose of 160 mg d, l-sotalol, a beta-blocker well known to produce clinically significant QT prolongation, according to a double-blind, randomised, placebo-controlled, crossover study design. Pharmacokinetic sampling was performed at various times up to 24 h after dosing.  12-lead ECG was monitored continuously throughout the study and recordings were made at different time points before and after dosing. QT-intervals were read from automated recordings as well as from manual assessments, as defined by a cardiographist. The pharmacokinetics of sotalol was described according to a two-compartment model with first order absorption. To account for the effect of heart rate on QT-interval, the relation between QT and RR was modelled as y = x^a*b. Drug effect was then characterised as a covariate on the intercept. Various models were explored to define the underlying exposure-response relation for d,l-sotalol. Data analysis was based on non-linear mixed effects modelling (NONMEM v5.1).

Results:  We have derived a population-based correction model to estimate the QT/RR relation and subsequently estimate drug effect on corrected QT-interval. We found clusters causing major discrepancies in the reproducibility of recordings derived from automated measurements. An iterative mixture model was implemented to account for data clustering and estimate a QT/RR relation for each individual subject. Circadian patterns in HR and the QTc/RR relationship were characterized using harmonic Fourier functions. A direct effect model using an Emax link function was sufficient to characterise the QT response on d,l-sotalol exposure.

Conclusion:  High variability exists in QT-interval measurements despite the careful control of the likely sources of noise in a clinical setting. In addition, there are major differences in the measurements and variability from automated readings, as compared to manual ones. A population-based correction factor improves the estimates of drug-induced effect on QT-interval and accurately characterises variability in ECG measurements. In contrast to the currently accepted statistical methodology for assessing drug-induced QTc prolongation, PK/PD modelling identifies the contribution of the various factors to mitigate QTc liability of new chemical entities.