Cohort 1 patients
Newly diagnosed patients (n = 381) aged 21–88 years (y) (mean, 53.6 ± 11.4 y) were enrolled. Only 37 patients were aged >70 y. Most patients (n=359, 94.2%) were diagnosed with EOC; eight (2.1%) were diagnosed with TC and 14 (3.7%) with PPC. Of these, 77 (20.2%) received postoperative paclitaxel-carboplatin doublet plus BEV. A total of 147 (38.6%) were classified as having FIGO stages I–II and 234 (61.4%) as having stages III–IV; 116 (30.4%) patients were in the ICON7-defined high-risk group. More than half of the patients (n=203, 53.3%) had serous histology, 108 (28.3%) had clear cell carcinoma (CCC) histology, 39 (10.2%) had endometrioid histology, and 20 (5.2%) had mucinous histology. The median follow-up period was 33.7 m (range 1.7–99.9 m). BEV dosage was 7.5–15.1 (mean, 9.2 ± 1.5) mg/kg, and number of treatment cycles was 1–17 (mean 7.9 ± 5.0). The dosage was <10 mg/kg/cycle in 71.4% of patients, and <10 cycles were used in 76.6% of patients. BEV dosages were similar between stage III–IV (n = 67) and ICON7-defined high-risk patients (n = 37) (mean, 9.3 ± 1.6 and 9.3 ± 1.8 mg/kg, respectively). There was no difference in the number of treatment cycles between these groups (mean, 7.9 ± 5.0 and 7.9 ± 4.6 cycles, respectively). During follow-up, 45 patients (58.4%) in the chemotherapy plus BEV group developed progressive disease, and 22 (28.6%) died; 156 patients (51.3%) in the chemotherapy alone group had progressive disease, and 114 patients (37.5%) died.
Relationships between whether BEV was added to chemotherapy and clinicopathological factors are presented in Table 1 . A significantly higher proportion of women with advanced stage, serous histology, preoperative chemotherapy, residual tumour size ≥1 cm, and triweekly delivery of postoperative chemotherapy received BEV.