Bevacizumab with peri-operative epirubicin, cisplatin and capecitabine (ECX) in localised gastro-oesophageal adenocarcinoma: a safety report

Background: Pen-operative chemotherapy and surgery is a standard treatment of localised oesophagogastric adenocarcinoma; however, the outcomes remain poor. Patients and methods: ST03 is a multicentre, randomised, phase II/III study comparing pen-operative ECX with or without bevacizumab (ECX-B). The primary outcome measure of phase II (n = 200) was safety, specifically gastrointestinal (GI) perforation rates and cardiotoxicity. Results: Two hundred patients were randomised between October 2007 and April 2010. Ninety-one/101 (90%) ECX and 86/99 (87%) ECX-B patients completed pre-operative chemotherapy; 7 ECX and 9 ECX-B patients stopped due to toxicity. Gastrointestinal perforations (3 ECX, 1 ECX-B), cardiac events (1 ECX, 4 ECX-B) and venous thromboembolic events (VEs, 8 ECX, 7 ECX-B) were uncommon. Arterial thromboembolic events (ATEs, myocardial infarction (MI) or cerebrovascular accident) were more frequent with ECX-B (5 versus 1 with ECx). Delayed wound healing, anastomotic leaks and GI bleeding rates were similar.More asymptomatic left ventricular ejection fraction (LVEF) falls (>= 15% and/or to <50%) occurred with ECX-B (21.2% versus 11.1% with ECX). Clinically significant falls (>= 10% to below lower limit of normal, LLN) occurred in (15.3%) and (8.9%) respectively, with no associated cardiac failure (median 22 months follow-up). Conclusions: Addition of bevacizumab to pen-operative ECX chemotherapy is feasible with acceptable toxicity and no negative impact on surgical outcomes.