When a patient with recurrent ovarian cancer (ROC) visits a clinic, a physician must decide whether he or she will perform SCS. If the patient has platinum-resistant or unresectable tumors, SCS is usually abandoned and palliative chemotherapy is administered. However, when the patient has platinum-sensitive and resectable tumors, it is unclear whether we should perform SCS. Numerous studies including the study by Bae et al. has been conducted to find who will benefit from the SCS, and found several factors (residual tumor size after SCS, progression-free survival (PFS) from primary treatment to recurrence, ascites, number of recurrent tumors, good performance status) were associated with prolonged survival after SCS.1 However, these studies did not directly address the issue - 'If the patient has platinum-sensitive and resectable tumors, should we perform SCS?' - because these studies did not compare the patients who had undergone SCS with those who had not. In the study by Bae et al., all patients with resectable tumors underwent SCS. Therefore, based on the study by Bae et al., we do not know whether SCS is beneficial or not. The GOG has recently initiated a randomized trial (GOG 213) addressing the role of SCS in patients with ROC. In GOG 213, the patients with ROC who had resectable tumors will be randomized into surgery vs. no surgery. The GOG 213 is anticipated to provide the answer for the role of SCS.

Many studies on the outcome of patients who underwent SCS suggested that the greatest benefit of SCS is seen if all gross tumor is resected.2 Unless tumors are optimally debulked, the benefit of SCS is unclear. Therefore, attempts to preoperatively identify the patients whose tumor will be optimally debulked have been conducted. In AGO-DESKTOP 2 trial, the value of a criteria (AGO score) in preoperatively predicting complete resection at SCS was investigated. The results were a complete resection was achieved in 76% of patients with AGO score positive (good performance status, complete resection at initial surgery, and absence of ascites >500 ml).3 In the study by Bae et al., their own criteria were used to select surgical candidates. They performed SCS in patients with PFS ≥6 months, GOG performance status ≤2, and no radiographic findings of extra-abdominal metastasis or unresectable intra-abdominal tumors (e.g. peritoneal carcinomatosis, multiple liver metastasis, involvement of the porta hepatis, pancreatic head, abdominal wall, para-aortic lymph node above the renal vein). After SCS, a complete resection was possible in 32 patients (59%), and residual tumor <0.5 cm was achieved in 15 patients (28%). Considering 47 of 54 (87%) patients attained optimal resection (<0.5 cm), we think that the criteria used in the study by Bae et al. is quite accurate in predicting the surgical outcomes after SCS. However, it is unclear whether their criteria is generally applicable to patients with ROC because, in the study be Bae et al., all surgeries were performed by a surgeon in a single institute. Therefore, we suggest that the external validation of their criteria should be attempted.

In selecting surgical candidates for SCS, we believe that the presence of ascites is a critical factor. Generally, patients with ROC who have malignant ascites should not undergo SCS. This is concordant with the AGO score criteria used in the AGO-DESKTOP 2 trial. In the study by Bae et al., it is unclear whether the patients with ascites were excluded from the study. The 'peritoneal carcinomatosis' may actually mean the presence of ascites on our assumption. Otherwise, we suggest the presence of ascites to be included in the selection criteria for SCS.