B1395
Title: Advancing clinical trial design in syndromic diseases with observational data
Authors: Alisa Stephens Shields - University of Pennsylvania (United States) [presenting]
Abstract: Motivated by inconclusive past trials in chronic pain conditions such as chronic pelvic pain syndrome and psoriatic arthritis, it is demonstrated how exploratory data analysis and simulation based on observational data provide evidence for addressing the shortcomings of previous trials of these syndromic, heterogeneous conditions. It discussed how consultation of a prospective cohort study and observational data embedded within a trial led to key findings for refining eligibility criteria and recruitment procedures of future clinical trials, including demonstrating the need for rigorous patient phenotyping during screening and identification of subgroups more likely to respond to certain classes of therapies. Lessons learned from observational data are also discussed for defining primary outcomes by establishing clinically important differences. Considerations for advanced statistical approaches are further presented, including adaptive clinical trial designs, sequential multiple assignment randomized trials, and methods for handling multiple outcomes, that are particularly suitable for heterogeneous conditions and may expedite the discovery of effective therapies relative to standard approaches.
