P1-14-05: Surgical Complications from the GeparQuinto Trial of Patients Receiving Preoperative Bevacizumab.

Abstract

Abstract Background: Bevacizumab has been reported to increase the risk of surgical complications, especially in pts treated with neoadjuvant chemotherapy. Methods: In January 2010 it was decided to collect surgical complications prospectively on a specifically developed from. All patients received chemotherapy with epirubicin/cyclophosphamide (EC) followed by docetaxel (T) with or w/o bevacizumab. Patients not responding to the first 4 cycles of EC+/−B were enrolled in another setting. All patients receiving bevacizumab were supposed to be operated at 5 weeks after the last infusion of bevacizumab. Results: Data from 699 patients have been prospectively collected, 329 received EC-T and 370 ECB-TB. Median age was 48 years in both arms. Median tumor size was 30mm in the ET-C arm vs 35mm in the ECB-TB arm (p=0.13). Multifocal or multicentric disease was present in 21% in the EC-T arm vs 22.4% (p=0.74). 36% of the patients received a sentinel node biopsy prior to start of neoadjuvant chemotherapy in both groups. 68.7% of patients received breast conserving therapy in the EC-T arm compared to 71.1% in the ECB-TB arm (p=0.54). The rate of pathological complete response was 19.5% with EC-T compared to 23.2% with ECB-TB (p=0.26). The median time from last CHT+/− bevacizumab to surgery was 29 days in EC-T and 34 days in ECB-TB. Bleeding or hematoma were reported in 6.7% of the patients treated by EC-T and 7.1% with ECB-TB (p=0.88). Wound infections or abscess were reported in 2.5% with EC-T and 3.9% with ECB-TB (p=0.39). Delayed wound healing was reported in 3.4% with EC-T and 6.3% with ECB-TB (p=0.11). Necrosis was reported in 1.9% with EC-T and 2.5% with ECB-TB (p=0.80). Any complication was reported in 11% with EC-T compared to 15.3% with ECB-TB (p=0.12). Conclusion: In general the rate of surgical complications was low. Patients treated with bevacizumab had a numerical increase of surgical complications. However, none of the differences were statistically significant. The intervals from last infusion to surgery as defined by the study protocol appear to be safe. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-14-05.