American Statistical Association New York City Metropolitan Area Chapter Memorial Sloan-Kettering Cancer Center Biostatistics Seminar

While clinical trials have been the mainstream method for evaluating new interventions for the past several decades, there are still many issues in design and analysis that need further discussion. Among these are the design of non-inferiority trials and the use of adaptive designs. For non-inferiority designs, one of the challenges is how to define the margin of indifference between the new and the control intervention. For the adaptive design case, there are many widely used adaptive designs but the altering sample size based on emerging trends has received the most attention recently. Some of the recent interest has been stimulated in part by political and regulatory issues. We shall explore these two issues and comment on current thinking and future directions.