Predictability of Overall Survival by Progression-Free Survival: A Simulation Study
View Presentation View Presentation
CHIA-WEN KO, FDA  Rajeshwari Sridhara, FDA  Shenghui Tang, FDA  *Lijun Zhang, FDA 


In the evaluation of cancer therapeutics, particularly involving solid tumors, it is common to use endpoints or outcomes based on tumor measurements. One such endpoint is Progression-Free Survival (PFS). PFS is defined as the time from randomization until objective tumor progression or death from any cause. Overall survival (OS) is defined as the time from randomization until death from any cause. OS is generally considered as the gold standard in evaluating cancer treatments. PFS is often used as a surrogate endpoint reasonably likely to predict OS in evaluating oncology drug products. It is important to investigate when PFS benefit would be translated to OS benefit. This study assessed the predictability of PFS for OS through simulations, with OS expressed as the sum of PFS and survival post-progression (SPP). A series of scenarios in a setting of metastatic cancers were simulated with various combinations of SPP medians, post-progression cross-over rates, and PFS improvement magnitudes. The conditions (such as magnitude, SPP duration, etc) under which PFS benefit could be translated to OS benefit were explored.