ABSTRACT
A survival trial usually takes a long time to finish, and data are accumulated gradually over the course of the study. For ethical reasons, such trials are usually monitored for early stopping. The stopping rule is devised based on the planned interim analysis, which is a statistical analysis conducted during a clinical trial. A primary goal of the interim analysis is to determine whether the trial should stop if the efficacy and/or futility of the treatment being evaluated is established. Because of the nature of a human clinical trial, it is usually not feasible to monitor the trial for each patient. Thus, interim analysis is usually planned for a group of patients. The statistical methodology developed to monitor a group of patients at each interim look time point is called a group sequential procedure. Group sequential monitoring methods have been developed in the last few decades by Haybittle (1971), Pocock (1977), O’Brien and Fleming (1979), Lan and DeMets (1983), Whitehead and Stratton (1983), Xiong (bib128), Jennison and Turnbull (1997), and many others. Several comprehensive reviews of these methods are available (Jennison and Turnbull 2000, references therein).
