Biostatistics short course: Some Key Concepts in Design and Analysis of Early Development Clinical Trials – 1/15/2019

Led by Steven Piantadosi, MD, PhD, associate senior biostatistician, surgical oncology, Brigham and Women’s Hospital, this course will cover basic principles underlying translational trials and dose-finding methods. The course will discuss the role of uncertainty in the design and interpretation of these trials, the use of the continual reassessment method (CRM) for dose finding, and a generalization of the CRM that relaxes most of the required assumptions underlying the cytotoxic model. A mid-day interlude will discuss some of the historic evolution in clinical trials. After the session, participants have the option to attend a lecture on late developmental failures and pipeline optimization.

This course will cover basic principles underlying translational trials and dose-finding methods. Translational clinical trials (which might be viewed technically as pre-developmental) are often nested in later developmental studies. They rely on established biological models of disease, treatment mechanism, and efficacy signals. Information gained from them guides subsequent experimental steps but does not alter clinical practice. The course will discuss their role and the formal utility of uncertainty in design and interpretation. Dose-finding methods have historically been of value in cytotoxic drug development, but the same designs have been uncritically applied to many drugs and biologicals, and in current times to immunotherapies. The course will cover the use of the continual reassessment method (CRM) for dose finding. It will also discuss a generalization of the CRM that relaxes most of the required assumptions underlying the cytotoxic model, and therefore might be useful for dose-finding/optimization of immunotherapy agents and other therapies with non-sigmoidal dose-response functions. In addition to these technical topics, a mid-day interlude will discuss some of the historic evolution in clinical trials. There will also be an optional lecture at the end of the day on late developmental failures and pipeline optimization to help illustrate the criticality of well-designed translational trials.

Steven Piantadosi, MD, PhD
Associate Senior Biostatistician, Surgical Oncology, Brigham and Women’s Hospital