Combining bevacizumab with irinotecan is a new chemotherapy regimen for patients with recurrent glioblastoma multiforme (GBM). Recent phase II trials suggest that this combined chemotherapy is beneficial to patients, but the subsequent adverse events may lead to treatment discontinuation. No comparison has yet demonstrated conclusively that the combined chemotherapy is more beneficial than single-agent chemotherapy. Thus, a meta-analysis was conducted to assess the efficacy and safety of bevacizumab compared to bevacizumab combined with irinotecan for the treatment of recurrent GBM. A total of 480 patients were included in the study, with 183 patients (38.1%) in the bevacizumab group and 297 patients (61.9%) in the bevacizumab plus irinotecan group. The median overall survival was 8.63 months (95% confidence interval [CI], 8.54-8.72 months) and 8.91 months (95% Cl, 8.69-9.13 months), respectively. The mean objective response rate (complete response plus partial response rate) was 33.9% (95% Cl, 18.1-52.1%) and 45.8% (95% Cl, 28.2-66.7%), respectively. The 6-month progression-free survival rates (PFS-6) were 38.8% (95% Cl, 18.8-57.0%) and 48.3% (95% Cl, 25.4-54.3%), respectively. The rate of discontinuation was 5.5% and 20.0%, respectively. Compared with patients treated with bevacizumab only, those in the bevacizumab plus irinotecan group had higher PFS-6 (p = 0.046), objective response (p = 0.013) and rate of discontinuation (p = 0.000) but there was no statistically significant difference in overall survival between the groups (p = 0.487). Thus, although the combination of bevacizumab and irinotecan may increase the rate of discontinuation, it provided no obvious improvement in overall survival in patients with recurrent GBM. Therefore, the benefits of drug combination are outweighed by the treatment discontinuity and quality of life effects of drug toxicity and should be considered on an individual patient basis only. Crown Copyright (C) 2012 Published by Elsevier Ltd. All rights reserved.