Implementation of innovative adaptive designs in early phase oncology trials: from theory to practice
*Inna Perevozskaya, Pfizer 

Keywords: Phase I trials , oncology, dose-escalation, adaptive design, continual re-assessment method, modified toxicity probability interval method

The methodology development for first innovative dose-escalation designs dates back to a couple of decades ago. Today many methods are available for statisticians to use and their statistical efficiency over traditional 3+3 design is well documented by numerous publications. Despite these methodological advances, the general level of acceptance of innovative adaptive designs in Phase 1 oncology remains low. The barrier is often not the methodology but rather cautious attitudes towards novel designs and implementation challenges. In this presentation we will review 2 case studies utilizing adaptive design in a phase 1 oncology trial: one implementing a Continual Re-assessment Method (CRM) and the other one using modified toxicity probability interval (mTPI) method. A review of study design, implementation and lessons learned will be given. We will focus on practical aspects and challenges of implementation for both of these methods.