The planning of a controlled clinical trial requires the knowledge of some basic characteristics. For example, for comparing means in a two-group parallel design the knowledge of the effect size and its variability are required to calculate the sample size needed for achieving a specified power when using a test at a given significance level. Misspecification of the designing parameters yields an underpowered or an overpowered study.
Interim analyses in group sequential trials provide the possibility of reaching conclusions on efficacy and safety prior to the planned end of the trial. The practical use of group sequential designs, however, is limited since they do not allow for data-driven sample size reassessments or other design changes. Assuming that the treatment effect was overestimated and/or the variance turned out to be larger than anticipated, then the study will not have enough power to yield a significant result. Moreover, the classical group sequential methods require a data independent choice of the group sizes, and it is assumed that only external factors may influence the change or the modification of the design. However, in many practical situation it would be very welcome to redesign a study based on the results of an interim analysis. Specifically, it would be desirable to adjust or to reassess the sample size of a study, to skip a treatment arm, or to change the null hypothesis.
New adaptive (flexible) study designs allow for correct data-driven re-estimation of the sample size while controlling the type I error rate. Redesigning the sample size in an interim analysis based on the results observed so far considerably improves the power of the trial since the best available information at hand is used for the sample size adjustment. That's why adaptive study designs are capable of reducing the risk of a false-negative study outcome.
In recent years, several methods have been proposed that enable a flexible design by use of adaptive interim analyses while maintaining the type I error rate. A strategy that copes well with the demands of practice is based on combining the p-values obtained from the separate stages by use of the inverse normal method. This strategy was proposed for two-stage designs by Bauer and Köhne (1994) and more generally for multistage designs by Lehmacher and Wassmer (1999). The resulting flexible designs are a powerful supplement to the classical group sequential test designs.
Adaptive designs were proposed from the U.S. statistical community mainly for the purpose of sample size reassessment. The seminal paper of Bauer and Köhne considered the more general type of design adaptations which is based on the combination testing principle. It allows, e.g., dropping treatment arms, refining the endpoint and hence – basically – changing the hypothesis. This enables much broader kinds of adaptations. For example, selecting treatment arms in a multi-armed trial is straightforward when using the combination testing principle. Other proposals to accomplish multiplicity issues that occur when considering multiple treatment arms within a sequential design exist. The combination testing principle is generally applicable and easy to use. It allows the full range of approaches (model-based, MCPs, Bayesian, etc.) to be applied within an adaptive strategy.
'ADDPLAN Adaptive Designs - Plans and Analyses' is designed for the purpose of planning and conducting a clinical trial based on an adaptive group sequential test design using this combination test principle. In order to evaluate the performance of an adaptive design 'ADDPLAN Adaptive Designs - Plans and Analyses' also allows the simulation of specific adaptation rules. It combines the key features of planning and conducting a sequentially planned clinical trial with a user-friendly interface.
