Let's discuss ways to enhance the clinical relevance of PFS. RECIST has been shown to be a powerful tool to assess anti-cancer activity. However, some trials have shown large effects on PFS (HR<0.50) without any effect on survival. Oncologists have noted that patients are often asymptomatic when they progress. Declaring progression too early might distance progression from death. Under what circumstances is there a place for an alternative to RECIST, or an additional intermediate measure of progression between RECIST and death? What do you think of changes, such as requiring an increase above baseline, or requiring that symptoms accompany progression of non-target or new lesions? What experience have you had with alternative progression criteria, such as the Immune Criteria in melanoma, the MacDonald criteria in glioblastoma, or definitions that include symptoms in ovarian cancer?