Two Stage Phase I Designs: Benefits and Pitfalls

Two stage Phase I designs combine algorithmic designs as an initial stage when limited data are available and switch to model based designs after heterogeneity in the responses is observed. Model based dose finding designs such as the Continual Reassessment Method (CRM), rely on some basic working model. These often take the form of “guess estimates” of the probabilities of toxicity at each level. These initial estimates are often referred to as the model skeleton. In this presentation, we study how use can be made of stage 1 data to inform the model parameters through an illustrative example of a clinical trial. We evaluate the interplay between prior assumptions and skeleton choice in the context of two stage CRM designs. We consider the behavior of a two-stage design at the point of transition from a 3+3 design to CRM. This is joint work with John O’Quigley.


This program is open to all.

Date & Time(s)


Memorial Sloan Kettering Cancer Center
307 East 63rd Street
Room 331
New York, NY 10065


Biostatistics Seminar Series
Department of Epidemiology and Biostatistics
Memorial Sloan Kettering Cancer Center