Update on Secondary Cytoreduction in Recurrent Ovarian Cancer

Catherine John, MD, and Katherine Tucker, MD

January 24, 2023

Recurrent ovarian cancer is difficult to treat; it has high recurrence rates and poor targeted treatment options. Between 60% and 75% of patients initially diagnosed with advanced-stage ovarian cancer will relapse within 2-3 years.1 Survival for these patients is poor, with an average overall survival (OS) of 30-40 months from the time of recurrence.2 Historically, immunotherapy has shown poor efficacy for recurrent ovarian malignancy, leaving few options for patients and their providers. Given the lack of effective treatment options, secondary cytoreductive surgery (surgery at the time of recurrence) has been heavily studied as a potential therapeutic option.

Dr Catherine John

The initial rationale for cytoreductive surgery (CRS) in patients with advanced ovarian cancer focused on palliation of symptoms from large, bulky disease that frequently caused obstructive symptoms and pain. Now, cytoreduction is a critical part of therapy. It decreases chemotherapy-resistant tumor cells, improves the immune response, and is thought to optimize perfusion of the residual cancer for systemic therapy. The survival benefit of surgery in the frontline setting, either with primary or interval debulking, is well established, and much of the data now demonstrate that complete resection of all macroscopic disease (also known as an R0 resection) has the greatest survival benefit.3 Given the benefits of an initial debulking surgery, secondary cytoreduction has been studied since the 1980s with mixed results. These data have demonstrated that the largest barrier to care has been appropriate patient selection for this often complex surgical procedure.

The 2020 National Comprehensive Cancer Network guidelines list secondary CRS as a treatment option; however, the procedure should only be considered in patients who have platinum sensitive disease, a performance status of 0-1, no ascites, and an isolated focus or limited focus of disease that is amenable to complete resection. Numerous retrospective studies have suggested that secondary CRS is beneficial to patients with recurrent ovarian cancer, especially if complete cytoreduction can be accomplished. Many of these studies have similarly concluded that there are benefits, such as less ascites at the time of recurrence, smaller disease burden, and a longer disease-free interval. From that foundation, multiple groups used retrospective data to investigate prognostic models to determine who would benefit most from secondary cytoreduction.

Dr Katherine Tucker

The DESKTOP Group initially published their retrospective study in 2006 and created a scoring system assessing who would benefit from secondary CRS.4 Data demonstrated that a performance status of 0, FIGO stage of I/II at the time of initial diagnosis, no residual tumor after primary surgery, and ascites less than 500 mL were associated with improved survival after secondary cytoreduction. They created the AGO score out of these data, which is positive only if three criteria are met: a performance status of 0, R0 after primary debulk, and ascites less than 500 mL at the time of recurrence.

They prospectively tested this score in DESKTOP II, which validated their findings and showed that complete secondary CRS could be achieved in 76% of those with a positive AGO score.5 Many believed that the AGO score was too restrictive, and a second retrospective study performed by a group at Memorial Sloan Kettering showed that optimal secondary cytoreduction could be achieved to prolong survival by a median of 30 months in patients with a longer disease-free interval, a single site of recurrence, and residual disease measuring less than 5 mm at time of initial/first-line surgery.6 Many individuals now use this scoring system to determine candidacy for secondary debulking: disease-free interval, number of sites of recurrence (ideally oligometastatic disease), and residual disease less than 5 mm at the time of primary debulking.

Finally, the iMODEL was developed by a group from China and found that complete R0 secondary CRS was associated with a low initial FIGO stage, no residual disease after primary surgery, longer platinum-free interval, better Eastern Cooperative Oncology Group performance status, lower CA-125 levels, as well as no ascites at the time of recurrence. Based on these criteria, individuals received either high or low iMODEL scores, and those with a low score were said to be candidates for secondary CRS. Overall, these models demonstrate that the strongest predictive factor that suggests a survival benefit from secondary CRS is the ability to achieve a complete R0 resection at the time of surgery.

Secondary debulking surgery has been tested in three large randomized controlled trials. The DESKTOP investigators and the SOC-1 trial have been the most successful groups to publish on this topic with positive results. Both groups use prognostic models for their inclusion criteria to select candidates in whom an R0 resection is believed to be most feasible. The first randomized controlled trial to publish on this topic was GOG-213,7 which did not use prognostic modeling for their inclusion criteria. Patients were randomized to secondary cytoreduction followed by platinum-based chemotherapy with or without bevacizumab versus chemotherapy alone. The median OS was 50.6 months in the surgery group and 64.7 months in the no-surgery group (P = .08), suggesting no survival benefit to secondary cytoreduction; however, an ad hoc exploratory analysis of the surgery arm showed that both overall and progression-free survival were significantly improved in the complete cytoreduction group, compared with those with residual disease at time of surgery.

The results from the GOG-213 group suggested that improved survival from secondary debulking might be achieved when prognostic modeling is used to select optimal surgical candidates. The SOC-1 trial, published in 2021, was a phase 3, randomized, controlled trial that used the iMODEL scoring system combined with PET/CT imaging for patient selection.8 Patients were again randomized to surgery followed by platinum-based chemotherapy versus chemotherapy alone. Complete cytoreduction was achieved in 73% of patients with a low iMODEL score, and these data showed improved OS in the surgery group of 58.1 months versus 53.9 months (P < .05) in the no-surgery group. Lastly, the DESKTOP group most recently published results on this topic in a large randomized, controlled trial.9 Patients were again randomized to surgery followed by platinum-based chemotherapy versus chemotherapy alone. Inclusion criteria were only met in patients with a positive AGO score. An improved OS of 7.7 months (53.7 vs. 46 months; P < .05) was demonstrated in patients that underwent surgery versus those exposed to only chemotherapy. Again, this group showed that overall survival was further improved when complete cytoreduction was achieved.

Given the results of these three trials, the Society for Gynecologic Oncology has released a statement on secondary cytoreduction in recurrent ovarian cancer (see Table).10 While it is important to use caution when comparing the three studies as study populations differed substantially, the most important takeaway the difference in survival outcomes in patients in whom complete gross resection was achieved versus no complete gross resection versus no surgery. This comparison highlights the benefit of complete cytoreduction as well as the potential harms of secondary debulking when an R0 resection cannot be achieved. Although not yet evaluated in this clinical setting, laparoscopic exploration may be useful to augment assessment of disease extent and possibility of disease resection, just as it is in frontline ovarian cancer surgery.

The importance of bevacizumab use in recurrent ovarian cancer is also highlighted in the SGO statement. In GOG-213, 84% of the total study population (in both the surgery and no surgery cohort) were treated with concurrent followed by maintenance bevacizumab with an improved survival outcome, which may suggest that this trial generalizes better than the others to contemporary management of platinum-sensitive recurrent ovarian cancer.

Overall, given the mixed data, the recommendation is for surgeons to consider all available data to guide them in treatment planning with a strong emphasis on using all available technology to assess whether complete cytoreduction can be achieved in the setting of recurrence so as to not delay the patient's ability to receive chemotherapy.

Dr. John is a gynecologic oncology fellow at the University of North Carolina at Chapel Hill. Dr. Tucker is assistant professor of gynecologic oncology at the university.


1. du Bois A et al. J Natl Cancer Inst. 2003;95:1320-9.

2. Wagner U et al. Br J Cancer. 2012;107:588-91.

3. Vergote I et al. N Engl J Med. 2010;363:943-53.

4. Harter P et al. Ann Surg Oncol. 2006;13:1702-10.

5. Harter P et al. Int J Gynecol Cancer. 2011;21:289-95.

6. Chi DS et al. Cancer. 2006 106:1933-9.

7. Coleman RL et al. Lancet Oncol. 2017;18:779-1.

8. Shi T et al. Lancet Oncol. 2021;22:439-49.

9. Harter P et al. N Engl J Med 2021;385:2123-31.

10. Harrison R, et al. Gynecol Oncol. 2021;163:448-52.

This article originally appeared on, part of the Medscape Professional Network.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.