RESUMEN
BACKGROUND: Bevacizumab, a humanized monoclonal antibody targeting the vascular endothelial growth factor, is a therapeutic agent used in a variety of neoplasms. We did a meta-analysis of randomized controlled trials to fully characterize the arterial thromboembolic events (ATEs) risk with bevacizumab in certain patients' subgroups. MATERIALS AND METHODS: We carried out a literature search on Medline for randomized trial reported from January 1966 to December 2009. Abstracts presented at the American Society of Clinical Oncology held between 2004 and 2009 were also searched for relevant clinical trials. Summary incidence, relative risks (RRs) and 95% confidence intervals (CIs) were calculated using random-effects or fixed-effects models based on the heterogeneity of included studies. RESULTS: A total of 13,026 patients from 20 randomized trials were included in the meta-analysis. Overall RR for ATE with bevacizumab-based therapy versus controls was 1.46 (95% CI 1.11-1.93, P = 0.007). On subgroup analysis, no significant risk differences were found based on the type of malignancy, type of clinical trial (phase II or III trials), type of publication (full papers versus presentations), high- versus low-dose bevacizumab and early versus advanced disease trials. When stratified by concomitant therapies, we found that gemcitabine-based regimens had a significant lower ATE risk compared with non-gemcitabine regimens (P = 0.01). CONCLUSIONS: Bevacizumab treatment is associated with a significant increase in the risk of arterial thrombosis. Our results seem to be generalizable to the vast majority of patients receiving bevacizumab in multiple settings.