Comparison of Different Choices of Baseline Values in Parallel Thorough QTc Studies

• Activity Number: 665
• Type: Topic Contributed
• Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Qianyu Dang*+ and Joanne Zhang
• Companies: FDA and FDA
• Keywords: TQT studies ; Drug Safety ; Nonparametric

Abstract

In most parallel thorough QTc (TQT) studies, one normally uses a time-matched baseline method to account for diurnal variability. Recently, some researchers proposed to use the average of the whole day baseline for the statistical analyses. The results based on the two baseline definitions can be very different. In order to evaluate these two baseline definitions, we analyzed the baseline profile of QTcF for all treatment groups based on 25 parallel TQT studies submitted to FDA. It is indicated that there is a non-flat pattern of QTcF on the baseline day in most parallel studies. Using the mean of whole day baseline will eliminate such a pattern and may lead to biased results. We also analyzed profiles in placebo group and found the similar pattern. A simulation study was used to compare the performance of two choices of baseline under different scenarios.