Multiplicity Control on Drop-the-Losers Designs.
Description of R package of proposed drop-the-losers design
This R package is developed based on the manuscript "Multiplicity Control in Oncology Clinical Trials with a Surrogate Endpoint-Based Drop-The-Losers Design". It includes functions to calculate the correlation boundary for the correlation between the response rate and the log-rank test statistic for the binary surrogate endpoint and the time-to-event primary endpoint, as well as functions to conduct simulation studies to obtain design operating characteristics. Additionally, it features an R Shiny application for interactive implementation of the design.
The proposed design will result in a smaller sample size and shorter study duration compared to the naive and most conservative approach, which sets the correlation between the DTL statistics and the final analysis statistics to 1. We believe this will lead to more efficient oncology trials.