MO19390 (SAiL): bleeding events in a phase IV study of first-line bevacizumab with chemotherapy in patients with advanced non-squamous NSCLC.

INTRODUCTION: The clinical benefit and safety profile associated with first-line bevacizumab with doublet chemotherapy in patients with advanced non-squamous non-small cell lung cancer (NSCLC) was established in two large phase III studies, E4599 and AVAiL. SAiL, a single-arm phase IV study, was conducted to evaluate bevacizumab with a range of first-line chemotherapy regimens in a routine oncology practice setting.
METHODS: This analysis of the SAiL data was undertaken to specifically evaluate bleeding adverse events (AEs) in this study, and to explore potential associations between bleeding and baseline patient and disease characteristics.
RESULTS: In total, 2212 patients were evaluated. Bleeding AEs (any grade) occurred in 38.2% of patients (grade ≥ 3 bleeding AEs: 3.6%). Grade ≥ 3 pulmonary hemorrhage and central nervous system bleeding events were observed in 0.7% and 0.1% of patients, respectively. The incidence of grade ≥ 3 bleeding AEs was comparable across patient subgroups defined by central tumor location, tumor cavitation, histology, concomitant anticoagulation therapy and age. The majority (88.6%) of bleeding events resolved or improved, 10.2% persisted and 1.3% led to death; 10.2% of bleeding events required bevacizumab interruption or discontinuation.
CONCLUSIONS: This analysis from the SAiL trial reaffirms a comparable incidence of clinically significant bleeding associated with first-line bevacizumab and chemotherapy as previous phase III studies in NSCLC patients despite less stringent first-line selection criteria. Grade ≥ 3 bleeding appears to be comparable when analyzed for patient and tumor characteristics, including tumor cavitation and concomitant anticoagulation therapy. Most bleeding events resolved or improved, and interruption/discontinuation of bevacizumab was infrequent in a standard oncology practice setting.