Association between treatment effects on disease progression (DP) endpoints and overall survival (OS) in patients with metastatic renal cell carcinoma (mRCC)

Abstract

5105 Background: Access to safe and effective novel therapies may be expedited if DP endpoints (progression-free survival [PFS], time to progression [TTP], or event-free survival) are established as valid surrogates for OS in pivotal studies in mRCC. While the association between DP endpoints and OS has been established in other cancers, it has not been rigorously examined in patients with mRCC. We assessed the association between treatment effects on DP endpoints and treatment effects on OS in controlled trials of patients with mRCC. Methods: A systematic literature search was conducted (Medline, conference abstracts, references of retrieved studies/ reviews) to identify studies meeting the following criteria: controlled trials in mRCC of IL-2, IFN-α, sunitinib, sorafenib, pazopanib, bevacizumab, temsirolimus, or everolimus; English language; publication date≥1997; median time to DP and OS reported for 2 or more treatment groups. For each treatment group comparison, we calculated the differences in median time to DP and median OS and analyzed the association between the differences in median time to DP and differences in median OS using weighted ordinary least-squares (OLS) regression. Results: A total of 21 studies representing 6182 patients, 52 treatment groups, and 35 comparisons were identified. The median difference between treatment groups in time to DP averaged 1.3 months; the median difference between treatment groups in OS averaged 2.8 months. The correlation between differences in median time to DP and differences in median OS was 0.69 (p < 0.0001). In weighted OLS regression, a 1 month difference in DP was associated with a 1.4 month difference in OS (p < 0.0001, adjusted R-sq = 0.46). Conclusions: In patients receiving treatment for mRCC, treatment effects on DP endpoints are predictive of treatment effects on OS. [Table: see text]