The benefits of secondary cytoreduction for platinum-sensitive relapsed ovarian cancer are still widely debated. We aimed to assess the efficacy of secondary cytoreduction plus chemotherapy versus chemotherapy alone in this patient population.

This multicentre, open-label, randomised, controlled, phase 3 trial (SOC-1), was done in four primarily academic centres in China (two in Shanghai, one in Hangzhou, and one in Guangzhou). Eligible patients were women aged 18 years and older with platinum-sensitive relapsed epithelial ovarian cancer with a platinum-free interval of at least 6 months after the end of first-line platinum-based chemotherapy and were predicted to have potentially resectable disease according to the international model (iMODEL) score and PET-CT imaging. iMODEL score was calculated using six variables: International Federation of Gynecology and Obstetrics stage, residual disease after primary surgery, platinum-free interval, Eastern Cooperative Oncology Group performance status, serum level of cancer antigen 125 at recurrence, and presence of ascites at recurrence. An iMODEL score of 4·7 or lower predicted a potentially complete resection. As per a protocol amendment, patients with an iMODEL score of more than 4·7 could only be included if the serum level of cancer antigen 125 was more than 105 U/mL, but the principal investigators assessed the disease to be resectable by PET-CT. Eligible participants were randomly assigned (1:1) via a permuted block design (block size of six) and stratified by study centre, iMODEL score, residual disease at primary surgery, and enrolment in the Shanghai Gynecologic Oncology Group SUNNY trial, to undergo secondary cytoreductive surgery followed by intravenous chemotherapy (six 3-weekly cycles of intravenous paclitaxel [175 mg/m] or docetaxel [75 mg/m] combined with intravenous carboplatin [area under the curve of 5 mg/mL per min]; surgery group) or intravenous chemotherapy alone (no surgery group). Primary endpoints were progression-free survival and overall survival, analysed in all participants randomly assigned to treatment, regardless of treatment received (intention-to-treat [ITT] population). Here, we report the final analysis of progression-free survival and the prespecified interim analysis of overall survival. Safety was assessed in all participants who received their assigned treatment and had available adverse event data. This study is registered with ClinicalTrials.gov, NCT01611766, and is ongoing but closed to accrual.

Between July 19, 2012, and June 3, 2019, 357 patients were recruited and randomly assigned to the surgery group (182) or the no surgery group (175; ITT population). Median follow-up was 36·0 months (IQR 18·1-58·3). In the no surgery group, 11 (6%) of 175 participants had secondary cytoreduction during second-line therapy while 48 (37%) of 130 participants who had disease progression crossed-over and had surgery at a subsequent recurrence. Median progression-free survival was 17·4 months (95% CI 15·0-19·8) in the surgery group and 11·9 months (10·0-13·8) in the no surgery group (hazard ratio [HR] 0·58; 95% CI 0·45-0·74; p<0·0001). At the interim overall survival analysis, median overall survival was 58·1 months (95% CI not estimable to not estimable) in the surgery group and 53·9 months (42·2-65·5) in the no surgery group (HR 0·82, 95% CI 0·57-1·19). In the safety population, nine (5%) of 172 patients in the surgery group had grade 3-4 surgical morbidity at 30 days, and no patients in either group had died at 60 days after receiving assigned treatment. The most common grade 3-4 adverse events during chemotherapy were neutropenia (29 [17%] of 166 patients in the surgery group vs 19 [12%] of 156 patients in the no surgery group), leucopenia (14 [8%] vs eight [5%]), and anaemia (ten [6%] vs nine [6%]). Four serious adverse events occurred, all in the surgery group. No treatment-related deaths occurred in either group.

Secondary cytoreduction followed by chemotherapy was associated with significantly longer progression-free survival than was chemotherapy alone in patients with platinum-sensitive relapsed ovarian cancer, and patients should be counselled about the option of secondary cytoreduction in specialised centres. Long-term survival outcomes will be assessed using mature data on overall survival.

Zhongshan Development Program.

For the Chinese translation of the abstract see Supplementary Materials section.