Abstract:
|
In the treatment of cancer which include long-term interventions, patients often switch lines of treatments due to disease progression or unacceptable toxicity. In clinical trials, the clinical endpoint, Overall Survival (OS) representing the actual clinical benefits, requires a long term follow up for evaluation. Progression Free Survival (PFS) is a popular and commonly used surrogate which allows shorter and smaller trials to meet the clinical objective. In conventional statistical methods, we study PFS and OS separately without considering their intertwining relationship and relationship with subsequent therapies. In this presentation, we will discuss a multi-state counting process methodology to study the relationship of PFS, OS and study drug exposure simultaneously. The joint relationship may help to answer how to obtain the projected longer term OS effect from the shorter term PFS effect and many other important questions. Simulations will be used to demonstrate the effectiveness of the proposed method.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.