Impact of different model structure and prior distribution in continual reassessment method

Abstract

Clinical trials are the most effective approach for comparing and examining experimental drugs, medical treatments, or clinical interventions on humans. Safety of the potential treatment is the first determined treatment during phase I trials, but it is continuously monitored through all phases. In this study, we introduced a comparative simulation study that has different model structures and prior distributions in the continual reassessment method (CRM). In order to understand the effect of model structures and prior distributions, a real life story and nine scenarios are implemented. Some of the common phase I methods were compared with the CRM where the model structure and prior distribution were different.

Keywords:

• Clinical trials
• Dose–response models
• Maximum tolerated dose
• Phase I