Risk of severe pulmonary embolism in cancer patients receiving bevacizumab: Results from a meta-analysis of published and unpublished data.

To evaluate the association between severe pulmonary embolism events and bevacizumab, we conducted the first meta-analysis evaluating the incidence and risk of pulmonary embolism associated with bevacizumab-based therapy. We searched PubMed, EMBASE, Cochrane Library, and ClinicalTrials.gov up to September 2016 for randomized controlled trials comparing bevacizumab with no bevacizumab on cancer patients. Incidence rates, relative risks, and 95% confidence intervals were calculated using fixed- or random-effects models. The primary end point was the association of bevacizumab with pulmonary embolism. Subgroup analyses were performed according to tumor type, dose, and publication status. In total, 23 randomized controlled trials were included. For patients receiving bevacizumab, the overall incidence of severe pulmonary embolism events was 1.76% (95% confidence interval = 1.25%-2.27%). Cancer patients treated with bevacizumab did not increase the risk of pulmonary embolism events (relative risk = 1.00, 95% confidence interval = 0.80-1.25). No significant differences in pulmonary embolism incidence or risk among subgroup analyses were observed. No evidence of publication bias was observed. This study suggested that bevacizumab may not increase the risk of pulmonary embolism in cancer patients.