Objective Evidence on recurrence patterns after bevacizumab in epithelial ovarian cancer is still insufficient. The aim of this study was to evaluate recurrence patterns after treatment with bevacizumab as second-line treatment in patients with platinum-sensitive, recurrent epithelial ovarian cancer.

Methods We retrospectively identified epithelial ovarian cancer patients who relapsed ≥6 months after primary treatment consisting of surgery and platinum-based chemotherapy between January 2008 and June 2019. Only those who received platinum-based doublet chemotherapy with bevacizumab or without bevacizumab as second-line treatment were included (n=192). To adjust confounders, we conducted 1:2 propensity score matching for platinum-free interval and secondary debulking surgery. Imaging studies were performed to locate newly developed or enlarged pre-existing tumors. Recurrence patterns were compared between bevacizumab users (study group) and non-users (control group).

Results After matching, the study group (n=52) and control group (n=104) showed similar baseline clinicopathologic characteristics including platinum-free interval (median (range) 15.3 (6.2–87.3) vs 14.0 (6.2–143.5) months; p=0.29) and patient age at the time of first recurrence (median (range) 55.5 (33.7–72.4) vs 55.0 (35.7–84.2) years; p=0.56). Initially, FIGO stage III disease was the most common in both two groups (55.8% vs 66.3%; p=0.20). Bevacizumab users were less likely to develop disease recurrence in the retroperitoneal lymph nodes (13.5% vs 34.6%; p=0.005), pelvis (17.3% vs 35.6%; p=0.018), and abdomen (40.4% vs 61.5%; p=0.012). However, no difference in distant metastasis was observed between the groups (23.1% vs 24.0%; p>0.99). Multivariate analyses adjusting for stage, histologic type, grade, platinum-free interval, and secondary debulking surgery revealed that the use of bevacizumab significantly reduced risks of nodal (adjusted HR (aHR) 0.24; 95% CI 0.10 to 0.56; p=0.001), pelvic (aHR 0.32; 95% CI 0.15 to 0.68; p=0.003), and abdominal recurrences (aHR 0.43; 95% CI 0.26 to 0.71; p=0.001). Nevertheless, use of bevacizumab did not influence risk of distant metastasis (aHR 0.70; 95% CI 0.35 to 1.40; p=0.32).

Conclusions In patients with platinum-sensitive, recurrent epithelial ovarian cancer, second-line chemotherapy with bevacizumab is associated with reduced risks of nodal, pelvic, and abdominal recurrences, but similar risks of distant metastases.