10519 Background: Bevacizumab, a monoclonal antibody against vascular endothelial growth factor A (VEGF), has shown significant clinical benefit in a number of advanced or metastatic cancer settings. We assessed the value of circulating VEGF level as a prognostic biomarker for outcome and a predictive biomarker of benefit with bevacizumab-containing treatment across four randomized phase III studies in metastatic colorectal cancer, lung cancer, and renal cell cancer. We also evaluated the overall distribution of circulating VEGF levels across these different tumor types and its association with levels of tumor VEGF expression. Methods: Baseline plasma, collected from 1816 patients enrolled in bevacizumab studies in metastatic colorectal, lung, and renal cell cancers, was analyzed for VEGF using ELISA. Matched archival tumor samples from colorectal and renal cell cancer studies were analyzed for VEGF expression by in situ hybridization. Hazard ratios and 95% confidence intervals for study primary end points were estimated using Cox regression analysis. Results: Across all VEGF subgroups analyzed, the estimated hazard ratios for progression-free survival (PFS) were < 1 (range, 0.59–0.89) in favor of bevacizumab treatment. Higher circulating VEGF levels were associated with shortened PFS and overall survival regardless of bevacizumab treatment, but were not well associated with tumor expression of VEGF. The distributions of circulating VEGF levels were similar across the tumor types, supporting the indirect relationship between tumor VEGF levels and circulating VEGF levels. Conclusions: Measurement of baseline circulating VEGF levels may be useful as a prognostic biomarker, but not as a predictive biomarker for bevacizumab-based treatment benefit in metastatic colorectal, lung, and renal cell cancers. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Genentech, Roche Roche