Abstract
Cytoreductive surgery is accepted as a major treatment of primary ovarian cancer. The role in recurrent ovarian cancer remains a field of discussion and controversy, mainly owing to missing data from prospective randomized trials and lack of universal definitions. Retrospective data indicate that complete resection of recurrent tumor formations should be aimed for, since survival prolongation is mainly seen for patients with no residual disease. Thus, it is most important to find predictors of complete resection, on the one hand to offer the best therapeutic chances to patients, but on the other hand to protect patients with limited life expectancy from additional surgical burden. The first prospective surgical trial in recurrent ovarian cancer, AGO-DESKTOP II validated a score (‘AGO score’) for complete resection. It was shown that patients with a good general condition (ECOG 0), no residual disease after surgery for primary ovarian cancer and absence of ascites in presurgical diagnostics have a 76% likelihood of undergoing complete resection. In this article, further recent data regarding surgery for recurrent ovarian cancer are going to be discussed and the advantages of incorporating these patients into randomized trials are highlighted.
Keywords
Epithelial ovarian carcinoma is the second most common genital malignancy in females and accounts for the majority of deaths from gynecologic malignancies. To date, no screening methods to detect ovarian cancer at an early stage have been developed, thus approximately 60–70% of women with newly diagnosed ovarian cancer present with advanced disease [1,2]. Nevertheless, complete clinical remission can be achieved in approximately 80% of these patients with the use of maximal surgical cytoreduction and platinum-based combination chemotherapy [1]. Today a combination of carboplatin and paclitaxel is the standard primary chemotherapy regime [3,4]. Unfortunately, upwards of 75% of patients with clinical complete response will develop recurrent disease. Recent studies have reported encouraging results in women with platinum-sensitive relapsed ovarian cancer (platinum-free interval >6 months) treated with pegylated liposomal doxorubicin [5], paclitaxel [6] or gemcitabine [7] in combination with carboplatin. For patients with platinum-resistant relapsed ovarian cancer, many drugs are available but pegylated liposomal doxorubicin [8], topotecan [9,10] or gemcitabine [11,12] are the most commonly used as monotherapy.
General considerations regarding surgery in recurrent ovarian cancer
While cytoreductive surgery in primary epithelial ovarian cancer is considered the standard of care [13], there are still many open questions in surgery for recurrent ovarian cancer and the benefit remains a matter of controversy. Currently there in no level I evidence for surgery in recurrent ovarian cancer. Furthermore, the definition of secondary cytoreductive surgery is not consistently used, and different studies included different groups of patients, namely patients with recurrent disease and those with persistent disease. Moreover, even patients with either progressive disease at the end of chemotherapy or patients with persisting, but not progressing, ovarian cancer, as well as both patients with small residual tumors that responded to systemic treatment and patients suffering from recurrence after a disease-free period of some weeks or several years, have been included [14,15]. For these subgroups, survival times of up to 9 months were reported, not justifying cytoreductive surgery with a morbidity rate of up to 24% in this setting. Therefore, it is necessary to have clear definitions of different types of surgery in ovarian cancer
This article focuses on cytoreductive surgery for recurrent ovarian cancer, which is defined as an operation performed in patients with recurrent disease after completion of primary treatment (surgery with or without chemotherapy) and a period without any evidence of disease.
What should be the aim of surgery for recurrent ovarian cancer?
The phrase of ‘optimal debulking’ has been introduced for primary cytoreductive surgery in advanced ovarian cancer. Retrospective studies reported a threshold above which cytoreduction did not result in a more favorable outcome and defined all lesions with a maximum diameter of ≤1 cm of the residual tumor as cut-off for inclusion criteria [16,17] or as stratum [3,18,19]. Nowadays this paradigm has changed towards complete resection of all visible tumor manifestations, owing to recently published reports [20–22], and the Gynecologic Cancer Interstudy Group (GCIG) has changed the official nomenclature to that effect [4]. However, the concept of “optimal debulking” has not been established in cytoreductive surgery for recurrent disease.
Surgical procedures in recurrent ovarian cancer.
This type of surgery could be performed at any time in the course of ovarian cancer (e.g., to get a histological diagnosis). Second-look surgery belongs to this group of procedures. It is an operation performed in patients who are clinically, biochemically and radiologically free of disease after the completion of a defined course of chemotherapy with the purpose to confirm the response status (in principle, the removal of the remaining tumor at second-look passes the border of diagnostic procedures)
This surgery should be performed in patients with macroscopically early ovarian cancer limited to the ovaries or the pelvis. The aim of this surgery is the detection of tumor spread
Surgery with the aim of complete resection of all macroscopic tumors in patients with first diagnosis of ovarian cancer before any other treatment modalities (e.g., chemotherapy)
An operation performed in patients after chemotherapy, usually two or three cycles, with an attempt to remove any remaining tumor that has not been removed by chemotherapy, followed by additional cycles of chemotherapy
An operation with the purpose of removing obviously resistant tumors that have not responded to chemotherapy and progressed during primary chemotherapy
Surgery aiming for complete resection of all macroscopic tumor in patients with recurrent ovarian cancer after completion of primary therapy including a subsequent period without any signs of disease
An operation performed in patients with symptoms caused by progressive disease or sequelae from prior treatment. These operations are performed in an effort to relieve symptoms and do not aim primarily at survival prolongation
The definition of optimal debulking is varying and ranges between “removal of all visible tumor” and “small residuals” with different dimensions of maximum diameters (0.5–2.0 cm).
Complete debulking rates varied between 9 and 82% in a systematic review comprising retrospective studies with more than 20 patients [23] and between 9 and 100% in a meta-analysis using all published data between 1983 and 2007 [24]. All but one series were collected retrospectively and were obviously exposed to selection bias. Unfortunately, information regarding selection criteria and proportions of patients who were not intended for cytoreductive surgery was lacking in most reports. If reported, the rate of patients not being offered surgery varied between 7 and 64%.
Even larger series (>100 included patients) dealing with cytoreductive surgery for recurrent disease provided controversial findings concerning the impact of the scale of surgical intervention. Eisenkop
In conclusion, the aim of surgery in recurrent ovarian cancer should be complete resection, as the described benefit of so-called ‘optimal debulking’ in some series is very low and probably a false finding. Therefore, if complete resection is not feasible, the aim of surgery should change from cytoreduction to palliation with the aim to minimize surgical comorbidity and to start chemotherapy as soon as possible.
Is it possible to predict complete resection in recurrent ovarian cancer?
The presence of symptoms, localization of disease, number of disease sites and treatment-free interval were reported as predictive factors for complete resection in univariate analyses. Cancer-Antigen (CA)-125 elevation was also found to be predictive in univariate analyses, and most recently it has been shown that the rate of complete resection declines by approximately by 3% per week after first CA-125 elevation was noticed and no surgery was performed [30].
Four retrospective series reported multivariate analyses of predictive factors associated with favorable surgical outcome. Absence of preoperative salvage chemotherapy, good performance status and a size of recurrent disease of less than 10 cm were predictors for complete debulking [25]. In another series with 38 patients, the number of disease sites (solitary vs multiple) was an independent factor for complete resection [31] and absence of ascites and residual disease after primary surgery were reported as predictors for complete resection [32]. The Descriptive Evaluation of preoperative Selection KriTeria for OPerability in recurrent OVARian cancer (DESKTOP OVAR) I trial conducted by the Arbeitsgemeinschaft Gynäkologische Onkologie (AGO) identified a combination of predictive parameters for complete resection: good performance status (Eastern Cooperative Oncology Group 0), no residual disease after surgery for primary ovarian cancer (alternatively, if unknown, early initial Fédération Internationale de Gynécologie et d'Obstétrique [FIGO] stage) and absence of ascites in presurgical diagnostics. Complete resection was achieved in 79% of patients scoring all these factors. If not all factors were positive, a complete resection was achieved in only 43% [33]. The latter group could be further differentiated: complete resection was achieved in 74% of this subgroup if there was no peritoneal carcinomatosis found intraoperatively, otherwise only 26% could be completely debulked [34].
In the subsequent AGO-DESKTOP II trial the ‘AGO score’ was validated in a prospective multicenter study. A total of 512 patients with platinum-sensitive relapse were screened and 261 patients (51%) presented with good performance status, complete resection at primary surgery and absence of ascites, and were defined as ‘AGO score’ positive. From these, 129 patients (49.4%) had their first relapse and underwent surgery for recurrent disease. A positive ‘AGO score’ resulted in a complete resection rate of 76%, thus the score was successfully validated [35]. In conclusion, the ‘AGO score’
Does complete resection translate into survival benefit in recurrent ovarian cancer?
The median survival of completely debulked patients ranges from 16 to 100 months [23] and overlaps with the median survival described in recently reported large prospective trials in recurrent platinum-sensitive ovarian cancer (i.e., ICON4/AGO-OVAR 2.2 [6] and the GCIG study AGO-OVAR 2.5 [7]). These studies reported a median survival of 18 and 29 months in the respective superior arms. The majority of series on cytoreductive surgery for recurrent ovarian cancer did not report median survival exceeding the ICON4/AGO-OVAR 2.2 results.
In conclusion, the lack of randomized trials makes it impossible to conclude whether a more favorable outcome in series with high rates of complete debulking could be attributed to biology (i.e., selection bias) or to surgical results.
Which prognostic factors are associated with prolonged survival in patients who received cytoreductive surgery for recurrent ovarian cancer?
Almost all series reported a relationship between survival and surgical outcome in univariate analysis. Complete debulking was one of the strongest predictors for survival in all multivariate analyses performed. All other analyzed factors provided controversial results. Treatment-free interval before cytoreductive surgery showed no significant impact on outcome in univariate analyses in approximately half of the series, but others reported a significant role. However, only few patients with rather short treatment-free survival were included in the respective series and the proportion of patients with less than 6 months treatment-free survival ranged from 0 to 13.5%. Therefore, the data regarding a possible impact of treatment-free interval periods are mainly valid for different periods beyond 6 months [23]. Eisenkop
Summary of series with >100 patients for surgery of recurrent ovarian cancer, reporting smallest achieved residual tumor.
2-year survival rates for patients with no macroscopic disease, 56% of patients had a disease-free interval of 7–12 months and 91% had a disease-free interval of 13–24 months.
ECOG: Eastern Cooperative Oncology Group; FIGO: Fédération Internationale de Gynécologie et d'Obstétrique.
‘AGO score’ for patients with epithelial ovarian cancer recurrence.
Good performance status (ECOG0)
No residual disease after surgery for primary treatment (alternatively, if unknown: early initial FIGO stage)
Absence of ascites in presurgical diagnostics
If all three items are existent, a complete resection of recurrent disease is feasible in 76% of all patients.
In addition, platinum-based chemotherapy should be administered after recovery from surgery to eradicate minimal residual disease. The value of hyperthermic intraperitoneal chemotherapy was looked at in a small retrospective analysis, but results were not convincing [36], thus the intravenous route should be the route of chemotherapy application.
Morbidity & mortality in surgery of recurrent ovarian cancer
Depending on the experiences and surgical capabilities, postoperative morbidity and mortality rates are varying, but complication rates in surgery for recurrent ovarian cancer are not significantly higher, compared with primary debulking surgery.
Mean 30-day morbidity after primary cytoreductive surgery for advanced stage ovarian cancer ranges between 22% [37] and 34% [38], while the morbidity rate in a meta-analyses of surgery in recurrent ovarian cancer ranged between 0 and 88.8%, with a weighted mean of 19.2% [24]. In the AGO-DESKTOP II trial, 33% of patients had at least one complication in the postoperative period [35]. Mean 30-day postoperative mortality in primary debulking surgery ranges between 0.7% [39] and 2.8% [40], while the mortality rate of surgery in recurrent ovarian cancer ranges between 0 and 5.5%, with a weighted mean of 1.2% [24], 7.8% in single centers [26] and 0.8% in the AGO-DESKTOP II trial [35]. Nevertheless, these data are very prone to be affected by selection and publication bias, since there is a strict definition neither for morbidity, nor for the observed time after surgery.
Future perspective
Currently, there is no level I evidence for cytoreductive surgery in recurrent ovarian cancer. However, even the most active chemotherapy regimens provided only limited activity with a median survival of 29 months (ICON4/AGO-OVAR 2.2) and improvement is clearly needed for these patients. The series of surgery for recurrent disease reported survival rates of up to 100 months in patients with complete resection, thus far exceeding the median survival rates reported after chemotherapy alone. These findings result from highly selected patient cohorts and, therefore, the main question might be “How can we select suitable patients for cytoreductive surgery in recurrent ovarian cancer?” The available information is far from being conclusive, but some factors were repeatedly cited as predictors for successful surgery. The AGO-DESKTOP II trial was the only one that successfully prospectively validated a score (good performance status, complete resection at primary surgery, absence of ascites) for complete resection in a multicenter setting.
Most recently, the AGO-DESKTOP III trial, “A prospective randomized trial comparing surgery and chemotherapy versus chemotherapy alone in recurrent ovarian cancer”, was launched (NCT01166737) [101]. Another ongoing study investigating the role of cytoreductive surgery for platinum-sensitive recurrent ovarian cancer is being performed by the Gynecologic Oncology Group (GOG) (NCT00565851) [102]. Until conclusive results that define the role of surgery in recurrent ovarian cancer are available, it is feasible to counsel patients regarding their surgical options in recurrent ovarian cancer outside of clinical trials based on the ‘AGO score’
Executive summary
Approximately 75% of patients with clinical complete response after first-line therapy will develop recurrent disease.
Different definitions have been used to describe surgery for recurrent ovarian cancer. A reasonable definition reads as follows: “surgery aiming for complete resection of all macroscopic tumor in patients with recurrent ovarian cancer after completion of primary therapy including a subsequent period without any signs of disease.”
The aim of surgery in recurrent ovarian cancer should be complete resection.
‘Arbeitsgemeinschaft Gynäkologische Onkologie (AGO) score’: good performance status (Eastern Cooperative Oncology Group [ECOG] 0), no residual disease after surgery for primary ovarian cancer and absence of ascites in presurgical diagnostics.
Prospective AGO-DESKTOP II trial: if the ‘AGO score’ is positive, complete resection is feasible in 76% of patients with recurrent ovarian cancer.
Morbidity and mortality rates in recurrent ovarian cancer do not exceed rates of primary surgery, if conducted in experienced centers.
Footnotes
