|
Abstract:
|
Both objective response rate (ORR) and progression-free survival (PFS) as defined by RECIST criteria have been shown to be only weakly correlated to overall survival (OS) in trials evaluating immunotherapy drugs. This has widely been attributed to unique patterns of response observed with immunotherapies, including delayed decrease in tumor response after an initial increase (i.e., 'psuedoprogression'). Alternative intermediate endpoints (IME) to define clinical response have been explored in an effort to improve associations with OS, but have only shown modest improvements (Gao et all, 2017). In this talk, we explore the heterogeneity within the PFS endpoint, showing that different subgroups within the broad RECIST-based PFS criteria show differential OS, which may explain at least in part the weak associations observed between PFS and OS. Further, we define novel IMEs based on the observed PFS heterogeneity, and examine associations of these IMEs with OS at both the patient and trial level. We highlight potential improvements of these IMEs over the current surrogate measures of ORR and PFS based on RECIST.
|
