In randomized phase III trials two anti-vascular endothelial growth factor (VEGF) approaches have yielded survival benefit in patients with metastatic cancer. In one approach, the addition of bevacizumab, a VEGF-specific antibody, to standard chemotherapy improved overall survival in colorectal and lung cancer patients and progression-free survival in breast cancer patients. In the second approach, multitargeted tyrosine kinase inhibitors that block VEGF receptor and other kinases in both endothelial and cancer cells, demonstrated survival benefit in gastrointestinal stromal tumor and renal-cell-carcinoma patients. By contrast, adding bevacizumab to chemotherapy failed to increase survival in patients with previously treated and refractory metastatic breast cancer. Furthermore, addition of vatalanib, a kinase inhibitor developed as a VEGF receptor-selective agent, to chemotherapy did not show a similar benefit in metastatic colorectal cancer patients. These contrasting responses raise critical questions about how these agents work and how to combine them optimally. We summarize three of the many potential mechanisms of action of anti-VEGF agents, and also discuss progress relating to the identification of potential biomarkers for anti-VEGF-agent efficacy in humans.