#TWO STAGE SEQUENTIAL TESTING TRIAL#

#TWO STAGE SEQUENTIAL TESTING TRIAL#

Phase III clinical trial that evaluated an experimental treatment for the prevention The Blood and Marrow Transplant Clinical Trials Network (BMT CTN) study 0402, a Reduces the risk of occurence of a specific type of event. When competing risks are present, interest may lie in whether a treatment The latter permitting covariate adjustment. Independent increments structure ( Tsiatis, In the survivalĭata setting, the log-rank test and the Cox proportional hazards model ( Cox, 1972) each have been shown to have an Risks outcomes, where they might experience one of multiple events. Survival outcomes, where patients are at risk of a single event, and for competing Techniques, group sequential design and covariate adjustment. Strong need exists for time to event analysis methods that incorporate these two Provides better overall survival for patients compared to usual care. ForĮxample, an oncology trial may evaluate a new treatment to determine whether it Time to event outcomes are often the main interest of a clinical trial. May still occur between treatment groups purely by chance ( Ciolino et al., 2015). In randomized clinical trials, where sizable imbalances on influential covariates Outcome of interest and might confound treatment evaluation. They improve efficiency by taking into account covariates that impact the Specify appropriate boundaries, including Pocock and O’Brien-Fleming testsĪnd error spending functions ( Pocock, 1977)Ĭovariates can also deliver more efficient assessment of the treatment effect.Īnalysis approaches that offer this benefit include regression models and stratified When thisĭistribution has an independent increments structure ( Jennison and Turnbull, 1999), standard methods can be utilized to Requires knowledge of the joint distribution of the test statistics. But, choosing the stoppingīoundaries in a way that will provide control of the type I error rate for the trial One method is group sequential design, whichĬan provide substantial reductions in sample size and trial duration by permittingĮarly stopping of the trial for efficacy and/or futility. Because these trials can beĮxtremely expensive and lengthy, means for improving the efficiency of this Moreover, using a simulation study of randomized group sequential trials, weĭemonstrate that the proposed method preserves the type I error rate and powerĪt their nominal levels in the presence of influential covariates.Ī late phase clinical trial is conducted with the purpose of comparingĮfficacy of an experimental treatment to a control. The prevention of adverse outcomes following blood and marrow transplant.

We demonstrate the test in a reanalysis of BMT CTNĠ402, a phase III clinical trial that evaluated an experimental treatment for Structure, which allows standard techniques such as O’Brien-Flemingĭesigns and error spending functions to be employed to meet type I error rateĪnd power specifications. Of test statistics has an asymptotic distribution with an independent increments Model, permits adjustment for covariates. Sequential test for treatment effect that, because it is based on the Fine-Gray Use in group sequential trials adjust for covariates. But, none of the competing risks methods currently available for Imbalances in patient characteristics, improving the power to detect treatmentĭifferences. Regression models can be used in clinical trials to adjust for residual Of treatment on a primary cause of failure in a clinical trial setting. Analyzing these outcomes allows one to assess the direct benefit Competing risks endpoints arise when patients can fail therapy from

Two stage sequential testing