Key result
New statistical methods for sample size calculation in crossover thorough QT studies were developed to ensure adequate power for detecting assay sensitivity across various covariance structures.
The paper provides updated statistical methodologies for calculating sample size and ensuring adequate power in thorough QT studies, addressing limitations of existing linear mixed models.
May refine TQT study powering across covariances; extends methods but leaves validation and adoption open.
The cost for conducting a "thorough QT/QTc study" is substantial and an unsuccessful outcome of the study can be detrimental to the safety profile of the drug, so sample size calculations play a very important role in ensuring adequate power for a thorough QT study. Current literature offers some help in designing such studies, but these methods have limitations and mostly apply only in the context of linear mixed models with compound symmetry covariance structure. It is not evident that such models can satisfactorily be employed to represent all kinds of QTc data, and the existing literature inadequately addresses whether there is a change in sample size and power for more general covariance structures for the linear mixed models. We assess the use of some of the existing methods to design a thorough QT study through data arising from a GlaxoSmithKline (GSK)-conducted thorough QT study, and explore newer models for sample size calculation. We also provide a new method to calculate the sample size required to detect assay sensitivity with adequate power.
No takes yet. Share an insight, caveat, or question.
Anand et al. (2010) studied Thorough QT studies. Newer models for sample size calculation vs. Existing methods (linear mixed models with compound symmetry) was evaluated on Sample size and power for assay sensitivity. New statistical methods for sample size calculation in crossover thorough QT studies were developed to ensure adequate power for detecting assay sensitivity across various covariance structures.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: