The first report of secondary cytoreductive surgery was demonstrated by Dr. Berek in 1983 [3]. Then several retrospective studies and few prospective but non-randomized trials were reported [4567]. Four randomized trials were initiated in the 2010s and the patient recruitment lasted for a long period, ranging from 4.5–9.5 years [8]. Despite one trial dropped because of the delayed enrollment, all the other 3 trials completed the patient enrollment, and 2 of them were presented at the 2017 and 2018 American Society of Clinical Oncology Annual Meetings respectively [2]. In countries such as Netherlands and China, patients and the gynecologic oncological surgeons have a strong belief in secondary cytoreductive surgery because of several publications. This explains why these trials have lasted for so long time or being earlier stopped.

In this issue of the journal, Dr. So and their colleagues [9] show in their retrospective study that women with recurrent ovarian cancer may derive significant survival benefit from secondary cytoreductive surgery. A set of patients with platinum-sensitive recurrent ovarian cancer (treatment-free interval more than 6 months) treated by either secondary cytoreductive surgery or chemotherapy alone were reviewed in So et al.'s study [9], and 52 patients with low-risk iMODEL (also called TIAN model [10]) at a median score of 2.3 were evaluated. Of them, 22 patients received surgery, and other 22 patients with second-line chemotherapy alone were matched using the propensity-score matching method. The survival in patients with recurrence was greater in those treated by secondary surgical resection than in those with chemotherapy alone. The limitations of the study are the small sample size and the large span from 2004 to 2016 with the heterogeneity among patients either in improvement of surgical techniques or updated standard care.

The progression-free survival in this study was quite similar to that of DESKTOP 3 study. In recent studies on platinum-sensitive recurrent ovarian cancer, the median progression-free survival increased from 14.0 m to 19.6 m (DESKTOP 3), 16.5 m to 18.2 m (GOG 213), when compared with 5.5 m to 11.9 m (AVANOVA2) in the combination of niraparib-bevacizumab, so called chemo-free therapy [11]. So, surgery is not worse than combined target therapy, but it is more cost-effective than the target therapy. In an international collaborative pooled analysis, the median overall survival of patients with recurrent ovarian cancer undergoing secondary surgery was 57.7 m, 27.0 m, 15.6 m in R0, R1 (residual disease ≤1 cm), R2 (residual disease >1 cm) groups respectively [5].

The rate of R0 increased from 39.4% (earlier pooled data) to 78.6% (recently closed trial) in non-selected or selected patients with recurrent ovarian cancer [25679]. Neither progression-free survival nor overall survival in GOG 213 was significantly improved in patients who underwent secondary cytoreductive surgery compared to chemotherapy alone [12]. As stated by the authors, more than 80% of the patients received combined bevacizumab. While recognizing the difficulty in evaluating the effect of bevacizumab on the study results, more studies are needed to confirm whether bevacizumab or other target therapy may reduce the potential survival benefit associated with surgery. Notably, in the previous report (objective #1) of GOG 213, the median overall survival in the chemotherapy and bevacizumab group was only 42.2 m (95% confidence interval [CI]=37.7–46.2) and 37.3 m (95% CI=32.6–39.7) (NCT00565851) [13]. Although there existed large variation in patient population, the authors had difficulty to understand the huge survival difference, 42.4 m versus 65.5 m, in patients treated with the same bevacizumab.

Which patient with recurrent ovarian cancer can benefit from secondary cytoreductive surgery? Which surgeon can most successfully operate on these patients? These are 2 critical questions when we make the subsequent surgical effort. Since surgical experience was an important factor determining the survival in recurrent ovarian cancer, we considered it as a factor when we established the iMODEL. In our previous study, all the participating centers had at least 10 years of experience with a series of publications. Thus, we finally eliminated the factor of surgical experience in the final iMODEL. According to the data, the rate of complete surgical resection (R0) was 39.4%, and it increased from 8.3% to 40.7% over the period of 1986–1997, 2002–2006 in the leading center [5].

There are two steps to consider secondary cytoreductive surgery as the standard care for women with recurrent ovarian cancer. Since the patients with Arbeitsgemeinschaft Gynäkologische Onkologie (AGO) positive score are part of the iMODEL (Fig. 1), the first step is to confirm the survival benefit in DESKTOP 3 study, and then in Shanghai Gynecologic Oncology Group (SGOG) SOC-1 study [2]. The second step is the early detection of the recurrences. Patients' compliance of follow-up, a combination of progression-free interval and the serum level of CA125, should also be considered during the recruitment. Some patients had longer progression-free interval, but their prognosis was worse because of the late recurrence diagnosis. Those patients will be less likely to benefit from secondary cytoreductive surgery even R0 resection. Most studies have confirmed the survival benefit of surgery for solitary recurrence. In the current study, solitary recurrence was associated with excellent outcome. All 6 patients (100%) with solitary disease were alive and 5 out of 6 patients (83.3%) survived without disease at the median follow-up of 60.2 m. Positron emission tomography-computed tomography scan has more positive findings in patients with suspect recurrence who have rising CA125 levels despite negative results on standard magnetic resonance imaging or computed tomography; it can modify the management by accurately mapping the distribution of recurrent disease, and it is definitely associated with a real R0 status after secondary cytoreductive surgery [14].