Early Clinical Development Planning via Biomarkers, Clinical Endpoints, and Simulation: A Case Study to Optimize for Phase III Dose Selection

• Activity Number: 377
• Type: Topic Contributed
• Date/Time: Tuesday, August 5, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Bret Musser*+ and Ghassan Fayad and Yue Shentu and James A. Bolognese and Nitin Patel and Jaydeep Bhattacharyya
• Companies: Merck and Merck and Bausch and Lomb and Cytel and Cytel and Cytel
• Keywords: modeling ; simulation ; proof-of-concept ; drug ; development ; dose selection

Abstract

Dose choice for Phase 3 clinical trials is one of the most difficult but important factors for a successful drug development program. Often the information upon which Phase 3 dose choice is based includes early development biomarker trials and Phase 2 dose-finding trials using a clinically relevant endpoint. This paper investigates a framework that leverages the relationship between early biomarkers and the target clinical endpoint to optimize a Phase 1-2 development plan. The framework is used to assess different biomarker designs for PoC studies to ultimately improve Phase 3 dose choice. A case study using a Bayesian tri-variate normal distribution model illustrates the framework. Among the findings of the simulation study are: (1) at typical sizes of Ph1b and Ph2b trials, biomarkers appear useful for PoC, but not for clinical endpoint dose-finding; (2) even with near perfect correlation between biomarkers and large amounts of Ph1b prior information, improved Ph3 dose selection requires increased Ph2b sample sizes. Thus, the fastest development path may be to move into Ph2b and measure the clinical endpoint of interest soon after PoC.