Multiple Comparisons of Repeated Measured Response: Issues of Assessment of Prolongation of QT Interval in Thorough QT Trials

Abstract

The primary objective of a thorough QT clinical trial is to demonstrate the lack of QT prolongation induced by the test treatment. The ICH E14 guidance defined drug-induced prolongation of QT interval as evidenced by an upper bound of the 95% one-sided confidence interval around the mean effect on QTc of 10 ms. Furthermore, it defined a negative thorough QT/QTc study as one in which the upper bound of the 95% one-sided confidence interval for the largest time-matched mean effect of the drug on the QTc interval excludes 10 ms. Conventionally, this objective is tested with the intersection–union test by testing the mean difference between the test treatment and placebo of QTc changes from baseline at each of the matched time points. The multiple-comparison nature of the test leads to higher false positive rate when comparisons are made repeatedly at multiple time points. Many approaches have been proposed in the last 5 years in order to improve the efficiency of the test. In this article, we survey and discuss some of the approaches.

Key Words:

• Multiple comparisons
• QT interval
• Repeated measured responses

ACKNOWLEDGMENTS

We thank Drs. Joanne Zhang of FDA and Frank Bretz of Novartis Pharma AG for reading an earlier version of the article. Their discussion and comments led to many improvements to the current version.

The views expressed in this presentation represent the opinions of the presenters, and do not necessarily represent the views of the U.S. Food and Drug Administration.

Notes

Note: : Conventional ICH E14 method. : Normal approximation with bias adjustment method. : Parametric bootstrap bias-correction method. : Nonparametric bootstrap bias-correction method.