ABSTRACT
This chapter identifies special issues that come up in clinical trials along with solutions or paths for consideration. The comparison of several distinct treatments with a common control occurs frequently in clinical trials. To optimize and potentially limit sample size, adaptive designs with the ability to evaluate interim efficacy and/or safety data and also potentially drop inferior arms are often proposed. The critical values for the classical Dunnett’s pairwise tests are applied to the classical dose-finding studies with the common control. The chapter argues that allowing multiple arms to carry forward to the final analysis is often important because the more efficacious arms may also be associated with greater toxicity. Sometimes the highest efficacy arm has to be dropped at an interim analysis because intolerable toxicities are observed. The chapter discusses how to use the blinded information to update the knowledge about the treatment effect using the maximum likelihood method and Bayesian Method.
