Incidence and management of bevacizumab-associated gastrointestinal perforations in patients with recurrent ovarian carcinoma.

OBJECTIVE: The objective of this study was to examine the incidence and management of bevacizumab-associated gastrointestinal (GI) perforations in patients with recurrent ovarian carcinoma.
METHODS: We identified all patients who received bevacizumab off protocol from August 2004-August 2008. We examined their medical records for reports of confirmed GI perforation, associated clinicopathological factors, treatment, and outcomes.
RESULTS: Six (4%) of 160 patients with ovarian carcinoma who had been treated with bevacizumab developed GI perforations, with a median of 4 (range, 2-8) previous cytotoxic regimens. The median serum CA-125 at the start of treatment was 228 U/mL (range, 50-3106 U/mL). The median number of bevacizumab cycles prior to perforation was 10.5 (range, 2-20). The median time from the last bevacizumab dose to diagnosis of GI perforation was 13 days (range, 1-28 days). Four (67%) patients underwent an exploratory surgery. At laparotomy, one had a gastric perforation and one had an appendiceal perforation; the site of perforation could not be identified in the other 2 Two patients (33%) were managed conservatively-one with a PEG tube and the other with supportive care. The median time of death from the date of diagnosis of GI perforation was 27 days (range, 4-326 days). Only two patients-one with a gastric and the other with an appendiceal perforation-survived >65 days. The 30-day mortality rate following a bevacizumab-associated GI perforation was 50%.
CONCLUSION: Bevacizumab-associated GI perforations in patients with recurrent ovarian carcinoma occurred in 4% of our patients. The prognosis of patients diagnosed with bevacizumab-associated GI perforations in this study was poor, and treatment should be individualized.