Multi-Objective Approach to Assist in Selection of Efficient Clinical Trial Operation Design with Respect to Minimizing Expected Recruitment Time and the Geographic Site Footprint

• Activity Number: 217 - Theory and Algorithms for Adaptive Clinical Trial Design with Multiple Objectives
• Type: Invited
• Date/Time: Tuesday, August 9, 2022 : 8:30 AM to 10:20 AM
• Sponsor: WNAR
• Author(s): Austin Matt* and Vlad Anisimov
• Companies: Amgen and Amgen
• Keywords: Multi-objective optimization; Constrained optimization; Pareto efficiency; Poisson-gamma; Clinical trials

Abstract

Operational considerations in clinical trials have competing constraints such as time, complexity (number of sites and countries), and cost. Early in the planning stage, study teams may not be sure which constraint has the most weight. To help teams understand the potential trade-offs inherit in these constraints, we developed a multi-objective optimization (MOO) approach that provides teams with a candidate set Pareto efficient choices to consider. We will discuss the evolutionary inspired approach to the MOO and the development of faster approximation methods for evaluating the lower confidence bound of the expected enrollment completion time based on a Poisson-gamma process.