Abstract TP217: Cerebrovascular Events in Glioma Patients Treated with Bevacizumab

Abstract

Introduction: Bevacizumab, a humanized monoclonal anti-vascular endothelial growth factor antibody, was FDA approved in 2009 for progressive glioblastoma. Phase II clinical trials suggested an increased risk of ischemic stroke (IS) and intracranial hemorrhage (ICH) while on bevacizumab. The incidence, clinical characteristics, and neuroimaging of glioma patients who developed cerebrovascular events while on treatment is lacking. We report our experience using bevacizumab for glioma patients. Methods: A retrospective review of glioma patients treated with bevacizumab at our institution from July 2005 to June 2011 was studied. Patients with MRI-confirmed IS and/or ICH while on bevacizumab was investigated and compared to historical data. Patient demographics, vascular risk factors, clinical presentations, tumor characteristics, treatments (surgery, chemotherapy, and radiation), and treatment duration were collected. Results: A total of 162 patients (65% male) received bevacizumab and 23 (14%) developed a cerebrovascular event while on treatment, with 3 (2%) IS and 20 (12%) ICH. All IS and ICH patients received prior brain radiation. In the IS group, 2 (66%) patients were symptomatic, with 1 (33%) cardiembolic and 2 (66%) lacunar strokes. None had risk factors besides hypercoagulable state from malignancy. In the ICH group, 3 (15%) patients had a symptomatic bleed and 6 (30%) had associated hypertension. All ICHs were intratumoral and 16 (80%) associated with tumor progression. Median survival after stroke was 9.8 and 3.7 months in the IS and ICH groups, respectively. Length of bevacizumab treatment was not significantly associated with development of IS ( p = 0.6) and ICH ( p = 0.3). Conclusion: Glioma patients have an inherently elevated risk of IS and ICH because of disease- and treatment-related effects. In our study, 78% of the events were asymptomatic and diagnosed on serial imaging. ICHs were more common, but all were small intratumoral bleeds, mostly in the setting of tumor progression. Development of cerebrovascular events was not associated with the duration of bevacizumab treatment. Our study, however, cannot determine causality and randomized controlled studies are needed to determine the risk of IS and ICH with bevacizumab use.