On May 5, 2009, the U.S. Food and Drug Administration granted accelerated approval to bevacizumab injection (Avastin; Genentech, Inc., South San Francisco, CA) as a single agent for patients with glioblastoma multiforme (GBM) with progressive disease following prior therapy. The approval was based on durable objective responses (independent radiologic review with stable or decreasing corticosteroid use). Two trials evaluating bevacizumab, 10 mg/kg by i.v. infusion every 2 weeks, were submitted. One trial also randomized patients to bevacizumab plus irinotecan treatment. All patients had received prior surgery, radiotherapy, and temozolomide. Patients with active brain hemorrhage were excluded. One trial enrolled 78 independently confirmed GBM patients. Partial responses were observed in 25.9% (95% confidence interval [CI], 17.0%-36.1%) of the patients. The median response duration was 4.2 months (95% CI, 3.0-5.7 months). The second trial enrolled 56 GBM patients. Partial responses were observed in 19.6% (95% CI, 10.9%-31.3%) of the patients. The median response duration was 3.9 months (95% CI, 2.4-17.4 months). Safety data were provided for the first study. The most frequently reported bevacizumab adverse events of any grade were infection, fatigue, headache, hypertension, epistaxis, and diarrhea. Grade 3-5 bevacizumab-related adverse events included bleeding/hemorrhage, central nervous system (CNS) hemorrhage, hypertension, venous and arterial thromboembolic events, wound-healing complications, proteinuria, gastrointestinal perforation, and reversible posterior leukoencephalopathy. The attribution of certain adverse events (e.g., CNS hemorrhage, wound-healing complications, and thromboembolic events) to either bevacizumab, underlying disease, or both could not be determined because of the single-arm, noncomparative study design.