Poster Presentations (original) (raw)

Clinical Trials, 2009

Abstract

Using survival status at T months as the primary endpoint in a phase II trial with interim looks is sometimes challenging. The primary issue is censoring at the time of the interim analysis – not all subjects are followed for T months. The simple proportion of all subjects surviving the required time is a biased estimate of the survival probability if some have incomplete follow-up. However, suspending accrual to allow all the subjects to be followed for the required length of time is impractical. Case and Morgan (2003) developed a two-stage design for evaluating survival probabilities with a planned interim analysis without interrupting the trial. In this report, the authors review the design, and describe development of analysis tools for calculating the test statistic for the interim analysis. Both a SAS macro and R-code were developed. The SAS macro is convenient if SAS is used as the analysis platform for the trial. The R-code is also easy to use if R is the preferred analysis tool. Development of both tools allowed for cross-verification of the programs. A cancer trial example is used to demonstrate the procedure and discuss practical issues related to rapid accrual. The SAS macro is available upon request.

anil mudda hasn't uploaded this paper.

Let anil know you want this paper to be uploaded.

Ask for this paper to be uploaded.