2056 Background: The only approved 2nd-line treatment for glioblastoma is Bevacizumab (Avastin), with estimated 6-month progression-free survival (PFS) rate of 42.6% and overall survival (OS) of 9.2 months. Laboratory data show Bortezomib (Velcade) is an effective agent in glioma models. We evaluated the combination of Avastin and Velcade in patients with recurrent GBM in a phase II, 2 center, open-label trial.Eligible patients were over 18 years of age, with WHO grade 4 glioma, Karnofsky score 70%. Bevacizumab was given 15 mg/kg IV every 3 weeks, and Bortezomib was given on days 1, 4, 8, 11, 22, 25, 29, and 32 of a 42 day cycle, at 1.7 mg/m2 for patients on non-enzyme-inducing antiepileptic drugs (EIAED) and 2.5 mg/m2 for patients on EIAED. Primary end points were 6-month PFS; secondary endpoints were toxicity, PFS, OS and radiographic response.There were a total of 61 patients enrolled, 32 patients on non-EIAEDs, and 29 patients on EIADs. Median age was 55.2. Six-month PFS was 40.6% for patients on non-EIAEDs, and 24.1% for patients on EIAEDs (p=0.17). Eleven patients are still alive; median OS was 9.1 months, with 8.8 months for non-EIAED patients, and 9.6 months for EIAED patients (p = 0.9). Of the patients on non-EIAEDs, best radiographic response included 11 patients (34.4%) with partial response, 13 with stable disease (40.6%), and 8 with progressive disease. Of patients on EIAEDs, 3 patients (10.3%) had a partial response, 18 (62.1%) with stable disease, and 6 with progression. On comparison of partial response rates between the 2 groups, p=0.03 was obtained. Most frequent grade 3 non-hematologic toxicities included fatigue (15.9%), sensory neuropathy (10.2%), diarrhea (9%), and thrombosis (8%). There were 2 episodes of grade 4 hyponatremia and 1 episode of epistaxis from a nasal septal ulcer. 14 patients eventually had to be taken off of Velcade and continued Avastin alone, due to sensory neuropathy.Bevacizumab and Bortezomib appear to be an effective treatment for recurrent malignant gliomas with a moderate toxicity profile. However, this phase II clinical study suggests that the combination is not superior to single-agent Bevacizumab and that the response may be blunted in the presence of EIAEDs.