Design and analysis considerations for thorough QT studies employing conventional (10 s, 12-lead) ECG recordings

The QT interval from the ECG cannot be measured precisely. The relationship of the QT interval to the RR interval within individuals across time and different RR values, and across individuals eludes complete understanding. Intrinsic beat-to-beat variability in QT interval corrected for heart rate (QTc interval) is not trivial. Therefore, it is difficult to determine a valid and reliable estimate of the time for ventricular repolarization based on the QTc interval. Yet, it must be demonstrated that a drug does not result in an increase in the QTc interval that exceeds 5 ms with some reasonable degree of certainty to be quite confident that the drug does not convey some risk of ventricular tachydysrhythmia due to delayed ventricular repolarization. This demonstration can be a Herculean task due to the magnitude of variability in the QTc interval. Design features and analytical methods that might be used in the thorough QT study to improve the chances of demonstrating the true relationship between a drug and QTc interval are reviewed.

Effects of Subcutaneous Pasireotide on Cardiac Repolarization in Healthy Volunteers: a Single‐Center, Phase I, Randomized, Four‐Way Crossover Study

The aim of this study was to evaluate the effects of subcutaneous pasireotide on cardiac repolarization in healthy volunteers. Healthy volunteers were randomized to one of four treatment sequences (n = 112) involving four successive treatments in different order: pasireotide 600 µg (therapeutic dose) or 1,950 µg (maximum tolerated dose) bid by subcutaneous injection (sc), placebo injection and oral moxifloxacin. Maximum ΔΔQTcI occurred 2 hours post-dose for both doses of pasireotide. Mean ΔΔQTcI was 13.2 milliseconds (90% CI: 11.4, 15.0) and 16.1 milliseconds (90% CI: 14.3, 17.9) for the 600 and 1,950 µg bid doses, respectively. Maximal placebo-subtracted change in QTcI from baseline for moxifloxacin was 11.1 (90% CI: 9.3, 12.9) milliseconds. Both pasireotide doses caused a reduction in heart rate: maximal heart rate change compared with placebo occurred at 1 hour for pasireotide 600 µg bid and at 0.5 hours for pasireotide 1,950 µg bid, with heart rate reductions of 10.4 and 14.9 bpm, respectively. At the therapeutic dose of 600 µg, pasireotide has a modest QT-prolonging effect. The relatively small increase of ∼3 milliseconds in ΔΔQTcI in the presence of a 3.25-fold increase in dose suggests a relatively flat dose–effect relationship of pasireotide on ΔΔQTcI in healthy volunteers. No safety concerns for pasireotide were identified during the study.

Effects of atomoxetine on the QT interval in healthy CYP2D6 poor metabolizers

AIM

The effects of atomoxetine (20 and 60 mg twice daily), 400 mg moxifloxacin and placebo on QT(c) in 131 healthy CYP2D6 poor metabolizer males were compared.

METHODS

Atomoxetine doses were selected to result in plasma concentrations that approximated expected plasma concentrations at both the maximum recommended dose and at a supratherapeutic dose in CYP2D6 extensive metabolizers. Ten second electrocardiograms were obtained for time-matched baseline on days -2 and -1, three time points after dosing on day 1 for moxifloxacin and five time points on day 7 for atomoxetine and placebo. Maximum mean placebo-subtracted change from baseline model-corrected QT (QT(c)M) on day 7 was the primary endpoint.

RESULTS

QT(c)M differences for atomoxetine 20 and 60 mg twice daily were 0.5 ms (upper bound of the one-sided 95% confidence interval 2.2 ms) and 4.2 ms (upper bound of the one-sided 95% confidence interval 6.0 ms), respectively. As plasma concentration of atomoxetine increased, a statistically significant increase in QT(c) was observed. The moxifloxacin difference from placebo met the a priori definition of non-inferiority. Maximum mean placebo-subtracted change from baseline QT(c)M for moxifloxacin was 4.8 ms and this difference was statistically significant. Moxifloxacin plasma concentrations were below the concentrations expected from the literature. However, the slope of the plasma concentration-QT(c) change observed was consistent with the literature.

CONCLUSION

Atomoxetine was not associated with a clinically significant change in QT(c). However, a statistically significant increase in QT(c) was associated with increasing plasma concentrations.

Validation of QT Interval Correction Methods When a Drug Changes Heart Rate

The QT interval is correlated with heart rate; therefore, the QT interval is usually corrected by heart rate when drug-induced QT effect is studied. Currently, there are many correction methods that use either fixed or data-driven approaches. The effectiveness of correction methods depends on many factors and varies from study to study. Statistical validation and comparisons need to be performed to determine the most appropriate correction method for each study. We examined different validation methods and explored a new approach to use when the testing drug changes heart rate.

Methodologies to characterize the QT/corrected QT interval in the presence of drug-induced heart rate changes or other autonomic effects

This White Paper, written collaboratively by members of the Cardiac Safety Research Consortium from academia, industry, and regulatory agencies, discusses different methods to characterize the QT effects for drugs that have a substantial direct or indirect effect on heart rate. Descriptions and applications are provided for individualized QT-R-R correction, Holter bin, dynamic QT beat-to-beat, pharmacokinetic-pharmacodynamic modeling, and QT assessment at constant heart rate. Most of these techniques are optimally performed using continuous electrocardiogram data obtained in clinical studies designed to characterize a drug's effect on the QT interval. An important study design element is the collection of drug-free data over a range of heart rates seen on treatment. The range of heart rates is increased at baseline by using ambulatory electrocardiogram recordings in addition to those collected under semisupine, resting conditions. Discussions in this study summarize areas of emerging consensus and other areas in which consensus remains elusive and provide suggestions for additional research to further increase our knowledge and understanding of this topic.

A 5-week study of the pharmacokinetics and pharmacodynamics of LY2189265, a novel, long-acting glucagon-like peptide-1 analogue, in patients with type 2 diabetes

AIM

To investigate the safety, tolerability, pharmacokinetics and pharmacodynamics of LY2189265 (LY), a novel, long-acting glucagen-like peptide-1 analogue, administered once weekly to subjects with type 2 diabetes.

METHODS

This was a placebo-controlled, parallel-group, subject- and investigator-blind study of LY in subjects (N = 43) with type 2 diabetes mellitus controlled with diet and exercise alone or with a single oral antidiabetic medication. Subjects taking metformin or thiazolidinediones continued on their therapy. Subjects receiving sulfonylurea, acarbose, repaglinide or nateglinide were switched to metformin prior to enrollment. Subjects received five once-weekly doses of 0.05, 0.3, 1, 3, 5 or 8 mg. Effects on glucose, insulin and C-peptide concentrations were determined during fasting and following standard test meals. The pharmacokinetics of LY and its effects on HBA1c, glucagon, body weight, gastric emptying and safety parameters were assessed.

RESULTS

Once-weekly administration of LY significantly reduced (p < 0.01) fasting plasma glucose, 2-h post-test meal postprandial glucose and area under the curve (AUC) of glucose after test meals at doses ≥1 mg. These effects were seen after the first dose and were sustained through the weekly dosing cycle. Most doses produced statistically significant increases in insulin and C-peptide AUC when normalized for glucose AUC. Statistically significant reductions in HBA1c were observed for all dose groups except 0.3 mg. The most commonly reported adverse effects (AEs) were nausea (35 events), headache (20 events), vomiting (18 events) and diarrhoea (8 events).

CONCLUSIONS

LY showed improvement in fasting and postprandial glycaemic parameters when administered once weekly in subjects with type 2 diabetes. The pharmacokinetics and safety profiles also support further investigation of this novel agent.

Mixed models for data from thorough QT studies: part 2. One-step assessment of conditional QT prolongation

We investigate mixed analysis of covariance models for the 'one-step' assessment of conditional QT prolongation. Initially, we consider three different covariance structures for the data, where between-treatment covariance of repeated measures is modelled respectively through random effects, random coefficients, and through a combination of random effects and random coefficients. In all three of those models, an unstructured covariance pattern is used to model within-treatment covariance. In a fourth model, proposed earlier in the literature, between-treatment covariance is modelled through random coefficients but the residuals are assumed to be independent identically distributed (i.i.d.). Finally, we consider a mixed model with saturated covariance structure. We investigate the precision and robustness of those models by fitting them to a large group of real data sets from thorough QT studies. Our findings suggest: (i) Point estimates of treatment contrasts from all five models are similar. (ii) The random coefficients model with i.i.d. residuals is not robust; the model potentially leads to both under- and overestimation of standard errors of treatment contrasts and therefore cannot be recommended for the analysis of conditional QT prolongation. (iii) The combined random effects/random coefficients model does not always converge; in the cases where it converges, its precision is generally inferior to the other models considered. (iv) Both the random effects and the random coefficients model are robust. (v) The random effects, the random coefficients, and the saturated model have similar precision and all three models are suitable for the one-step assessment of conditional QT prolongation.

Mixed models for data from thorough QT studies: part 1. assessment of marginal QT prolongation

We investigate mixed models for repeated measures data from cross-over studies in general, but in particular for data from thorough QT studies. We extend both the conventional random effects model and the saturated covariance model for univariate cross-over data to repeated measures cross-over (RMC) data; the resulting models we call the RMC model and Saturated model, respectively. Furthermore, we consider a random effects model for repeated measures cross-over data previously proposed in the literature. We assess the standard errors of point estimates and the coverage properties of confidence intervals for treatment contrasts under the various models. Our findings suggest: (i) Point estimates of treatment contrasts from all models considered are similar; (ii) Confidence intervals for treatment contrasts under the random effects model previously proposed in the literature do not have adequate coverage properties; the model therefore cannot be recommended for analysis of marginal QT prolongation; (iii) The RMC model and the Saturated model have similar precision and coverage properties; both models are suitable for assessment of marginal QT prolongation; and (iv) The Akaike Information Criterion (AIC) is not a reliable criterion for selecting a covariance model for RMC data in the following sense: the model with the smallest AIC is not necessarily associated with the highest precision for the treatment contrasts, even if the model with the smallest AIC value is also the most parsimonious model.

Statistical characterization of QT prolongation

The appropriate assessment of QT prolongation remains controversial. We suggest that before the relative merits of various methods can be evaluated, we must state what we assume an assessment of QT prolongation should be about. As a general framework for the assessment of QT prolongation we propose that an assessment of "absolute" or "uncorrected" QT prolongation is properly carried out through a between-treatment (active versus placebo) comparison of the marginal distributions of QT data; an assessment of "heart rate corrected" QT prolongation is carried out through a between-treatment comparison of the conditional distributions of QT data (conditional on RR interval or heart rate). Under this general framework, conditional QT prolongation is, in general, a function of RR interval, and we discuss three possible summary characteristics for that function. We show how current procedures for the assessment of QT prolongation relate to the general approach (that is, to between-treatment contrasts of the marginal and conditional expectation of the QT interval), and to each other. It transpires that only the so-called "one-step procedure" can provide a complete characterization of conditional QT prolongation. We show that the "two-step procedure" with data-driven correction provides an unbiased estimate of expected conditional QT prolongation, which may, from a clinical point of view, be a more satisfactory characteristic than the conventional characteristic, QT prolongation at the reference RR interval. We strongly suggest that two-step procedures with fixed correction be abandoned in the analysis of thorough QT/QTc studies: Fixed correction is either redundant (when a drug has no effect on average RR interval), or systematically biased (when a drug does affect average RR interval).

Statistical models for heart rate correction of the QT interval

The analysis of QT interval data is now an essential part of the assessment of drug safety. As the QT interval is inversely associated with heart rate, an appropriate correction must be applied in order to evaluate QT data in clinical trials. The aim is to characterize changes in QT interval at a standard heart rate, taking into account the correlation between these two variables to adjust for heart rate changes during the course of the trial. It has been shown that the relationship between the RR interval (=1/heart rate) and the QT interval is highly variable between individuals but stable over time within each individual.Many mathematical models have been developed to describe the QT-RR relationship. However, there has been less emphasis on the derivation of suitable statistical models that account for the multilevel structure of the ECG data.An important example is the interpretation of the so-called population-specific heart rate corrections, which are based on data pooled from different subjects. Often, simple regression techniques are used to quantify the population correction, disregarding the subject level and leading to biased parameter estimates. Instead, population-based corrections that account for individual intercepts should be used, in order to distinguish within-subject-effects from between-subject effects. Therefore, population-specific corrections cannot be derived solely from the cross-sectional data. The impact of the different statistical models is illustrated by data from the baseline periods of six clinical QT studies.

Bias and variance evaluation of QT interval correction methods

There is an increasing regulatory emphasis on assessing drug-induced QT interval prolongation. Since QT interval is correlated with heart rate (HR), assessment of drug-induced QT interval prolongation should be made at a standardized HR, resulting in the need to correct QT interval (QTc) for HR. This study investigates the statistical properties of QTc intervals using individual based correction (IBC), population based correction (PBC), or fixed correction (FC) methods under both the linear and log-linear regression models for the QT-RR relationship where RR is the time elapsing between two consecutive heart beats (inversely related to HR through RR = 60/HR). This study shows that QTc intervals using PBC and FC methods are conditionally biased. The QTc interval using the IBC method is conditionally unbiased under the linear regression model, but is conditionally biased under the log-linear regression model. It also shows that under both the linear and log-linear regression models, the conditional variances of the QTc intervals using the three correction methods satisfy the order FC < or = PBC < or = IBC. Suggestions for analyzing QT intervals based on these findings are discussed.

QT effects of duloxetine at supratherapeutic doses: a placebo and positive controlled study

BACKGROUND

The electrophysiological effects of duloxetine at supratherapeutic exposures were evaluated to ensure compliance with regulatory criteria and to assess the QT prolongation potential.

METHODS

Electrocardiograms were collected in a multicenter, double-blind, randomized, placebo-controlled, crossover study that enrolled 117 healthy female subjects aged 19 to 74 years. Duloxetine dosages escalated from 60 mg twice daily to 200 mg twice daily; a single moxifloxacin 400 mg dose was used as a positive control. Data were analyzed using 3 QT interval correction methods: mixed-effect analysis of covariance model with RR interval change from baseline as the covariate, the QT Fridericia's correction method, and the individual QT correction method. Concentrations of duloxetine and its 2 major metabolites were measured.

RESULTS

Compared with placebo, the mean change from baseline in QTc decreased with duloxetine 200 mg twice daily. The upper limits of the 2-sided 90% confidence intervals for duloxetine vs. placebo were <0 msec at each time point by any correction method. No subject had absolute QT Fridericia's correction values >445 msec with duloxetine, and the change in QT Fridericia's correction from baseline with duloxetine did not exceed 36 msec. No relationship was detected between QTc change and plasma concentrations of duloxetine or its metabolites even though average duloxetine concentrations ranged to more than 5 times those achieved at therapeutic doses. Moxifloxacin significantly prolonged QTc at all time points, regardless of correction method.

CONCLUSIONS

Duloxetine does not affect ventricular repolarization as assessed by both mean changes and outliers in QT corrected by any method.

A one-step approach to the analysis of the QT interval in conscious telemetrized dogs

INTRODUCTION

To account for heart rate-induced changes in the QT interval, correction formulas are generally applied to normalize the QT interval for heart rate. None of these formulas is entirely accurate because correction or normalization of any parameter in biology may introduce an additional source of variation in estimating the parameter. In this article, a one-step approach for the statistical analysis of the QT interval was proposed based on modeling the functional relationship between the QT interval and heart rate.

METHODS

The QT-HR relationship was incorporated into the statistical analysis to provide a model-based correction. This was accomplished by including HR as a covariate in the QT interval analysis. The approach was demonstrated using data generated from Lilly Research Laboratories. We compared the false positive rate and statistical power of QT, QTcF, and the proposed one-step method.

RESULTS

We found the one-step method demonstrated the greatest sensitivity in detecting a QT interval change without an increase in the false positive rate. It was shown that the one-step QT analysis could detect a 5%-6% increment of the QT interval. This is approximately equivalent to an increase of 11-13 ms in QT interval in beagle dogs.

DISCUSSION

Several advantages and unique features of the one-step method are discussed. These include evaluating treatment effect on QT without applying a heart rate correction formula and estimating QT difference flexibly at any selected heart rate. In addition to the linear QT-HR relationship, other functional relationships can be easily implemented to this approach.

The Combined Use of Ibutilide as an Active Control With Intensive Electrocardiographic Sampling and Signal Averaging as a Sensitive Method to Assess the Effects of Tadalafil on the Human QT Interval

OBJECTIVES

This study was designed to evaluate effects of tadalafil, a phosphodiesterase-5 inhibitor used for the treatment of erectile dysfunction (ED), on the QT interval.

BACKGROUND

Cardiovascular disease is common in men with ED. Men with cardiovascular disease and ED may have decreased cardiac repolarization reserve.

METHODS

Effects of tadalafil (100 mg by mouth), ibutilide (0.002 mg/kg intravenously), and placebo on the QT interval in healthy men were compared (placebo and tadalafil [n = 90], with a subset [n = 61] receiving all treatments; mean age 30 years, range 18 to 53 years). Electrocardiographic sampling was done for two days before treatment and on treatment days. The QT was corrected for RR interval with five correction methods, including an individual correction (QTcI). Plasma concentrations of tadalafil were measured to evaluate concentration-QT effect relationships.

RESULTS

At the time corresponding to maximum plasma concentration of tadalafil, the mean difference in the change in QTcI between tadalafil and placebo was 2.8 ms; tadalafil was equivalent to placebo (a priori, upper limit of 90% confidence interval < 10 ms [actual = 4.4 ms]; post hoc, upper limit of 95% confidence interval < 5 ms [actual = 4.8]). The active control, ibutilide, significantly increased QTcI by 6.9 and 8.9 ms compared with tadalafil and placebo, respectively. Similar statistical results were obtained with four additional QT correction methods. No subject had a QTcI > or = 450 ms or an increase in QTcI > or = 30 ms with any treatment.

CONCLUSIONS

Based on the a priori statistical test of equivalence, placebo and high-dose tadalafil produced equivalent effects on the QT interval. This study reliably discerned 5- to 10-ms changes in corrected QT in the ibutilide active control group.