rpact (R Package for Adaptive Clinical Trials) is a comprehensive package that enables the design, simulation, and analysis of confirmatory adaptive group sequential designs. Particularly, the methods described in the recent monograph by Wassmer and Brannath (published by Springer, 2016) are implemented. It also comprises advanced methods for sample size calculations for fixed sample size designs incl., e.g., sample size calculation for survival trials with piecewise exponentially distributed survival times and staggered patients entry.
rpact includes the classical group sequential designs (incl. user spending function approaches) where the sample sizes per stage (or the time points of interim analysis) cannot be changed in a data-driven way. Confirmatory adaptive designs explicitly allow for this under control of the Type I error rate. They are either based on the combination testing or the conditional rejection probability (CRP) principle. Both are available, for the former the inverse normal combination test and Fisher's combination test can be used.
Specific techniques of the adaptive methodology are also available, e.g., overall confidence intervals, overall p-values, and conditional and predictive power assessments. Simulations can be performed to assess the design characteristics of a (user-defined) sample size recalculation strategy. Designs are available for trials with continuous, binary, and survival endpoint.
For more information please visit www.rpact.org. If you are interested in professional services round about the package or need a comprehensive validation documentation to fulfill regulatory requirements please visit www.rpact.com.
rpact is developed by
Gernot Wassmer, Friedrich Pahlke
Wassmer, G., Brannath, W. (2016) Group Sequential and Confirmatory Adaptive Designs in Clinical Trials (Springer Series in Pharmaceutical Statistics; doi:10.1007/978-3-319-32562-0)
Report bugs at https://github.com/rpact-com/rpact/issues