Efficacy and safety results from GEICO 1205, a randomized phase II trial of neoadjuvant chemotherapy with or without bevacizumab for advanced epithelial ovarian cancer

Background Bevacizumab is an approved treatment after primary debulking surgery for ovarian cancer. However, there is limited information on bevacizumab added to neoadjuvant chemotherapy before interval debulking surgery. Objective To evaluate neoadjuvant bevacizumab in a randomized phase II trial. Methods Patients with newly diagnosed stage III/IV high-grade serous/endometrioid ovarian cancer were randomized to receive four cycles of neoadjuvant chemotherapy with or without ≥3 cycles of bevacizumab 15 mg/kg every 3 weeks. After interval debulking surgery, all patients received post-operative chemotherapy (three cycles) and bevacizumab for 15 months. The primary end point was complete macroscopic response rate at interval debulking surgery. Results Of 68 patients randomized, 64 completed four neoadjuvant cycles; 22 of 33 (67%) in the chemotherapy-alone arm and 31 of 35 (89%) in the bevacizumab arm (p=0.029) underwent surgery. The complete macroscopic response rate did not differ between treatment arms in either the intention-to-treat population of 68 patients (6.1% vs 5.7%, respectively; p=0.25) or the 55 patients who underwent surgery (8.3% vs 6.5%; p=1.00). There was no difference in complete cytoreduction rate or progression-free survival between the treatment arms. During neoadjuvant therapy, grade ≥3 adverse events were more common with chemotherapy alone than with bevacizumab (61% vs 29%, respectively; p=0.008). Intestinal (sub)occlusion, fatigue/asthenia, abdominal infection, and thrombocytopenia were less frequent with bevacizumab. The incidence of grade ≥3 adverse events was 9% in the control arm versus 16% in the experimental arm in the month after surgery. Conclusions Adding three to four pre-operative cycles of bevacizumab to neoadjuvant chemotherapy for unresectable disease did not improve the complete macroscopic response rate or surgical outcome, but improved surgical operability without increasing toxicity. These results support the early integration of bevacizumab in carefully selected high-risk patients requiring neoadjuvant chemotherapy for initially unresectable ovarian cancer.