Tertiary Cytoreduction for Recurrent Epithelial Ovarian Cancer: a Multicenter Study in Turkey.

BACKGROUND
The purpose of this study was to determine the benefit of tertiary cytoreductive surgery (TC) for secondary recurrent epithelial ovarian cancer (EOC), focusing on whether optimal cytoreduction has an impact on disease-free survival, and whether certain patient characteristics could identify ideal candidates for TC.


MATERIALS AND METHODS
Retrospective analysis of secondary recurrent EOC patients undergoing TC at three Turkish tertiary institutions from May 1997 to July 2014 was performed. All patients had previously received primary cytoreduction followed by intravenous platinum-based chemotherapy and secondary cytoreduction for first recurrence. Clinical and pathological data were obtained from the patients' medical records. Survival analysis was caried out using the Kaplan Meier method. Actuarial curves were compared by the two tailed Logrank test with a statistical significance level of 0.05.


RESULTS
Median age of the patients was 49.6 years (range, 30-67) and thirty-eight (72%) had stage III-IV disease at initial diagnosis. Twenty six (49%) had optimal and 27 (51%) suboptimal cytoreduction during tertiary debulking surgery . Optimal initial cytoreduction, time to first recurrence, optimal secondary cytoreduction, time interval between secondary cytoreduction and secondary recurrence, size of recurrence, disease status at last follow-up were found to be significant risk factors to predict optimal TC. Optimal cytoreduction in initial and tertiary surgery and serum CA-125 level prior to TC were independent prognostic factors on univariate analysis.


CONCLUSIONS
Our results and a literature review clearly showed that maximal surgical effort should be made in TC, since patients undergoing optimal TC have a better survival. Thus, patients with secondary recurrent EOC in whom optimal cytoreduction can be achieved should be actively selected.

The chances of developing recurrent disease after a second clinical remission is nearly 100 percent (Shih et al., 2010). Hitherto data regarding tertiary cytoreduction are limited with a few retrospective studies.
The objective of this study was to determine the benefit of TC in secondary recurrent EOC, whether optimal cytoreduction had an impact on disease-free survival (DFS), and whether certain patient characteristics could identify ideal candidates for TC.

Materials and Methods
After obtaining approval from the Institutional Review Board at Istanbul University Cerrahpasa School of Medicine (Date: 02.06.2015/Number: A-07), retrospective analysis of secondary recurrent EOC patients underwent TC at three Turkish tertiary institutions from May 1997 to July 2014 was performed. The data were collected from the medical records. Patients' age at initial diagnosis, EOC histologic subtype, stage and tumor grade at initial diagnosis, outcomes of initial, secondary and TC (optimal vs suboptimal), time to first recurrence, time from secondary cytoreduction, number of lines of chemotherapy prior to TC, platinum sensitivity, serum CA-125 level prior to TC, size of the largest recurrence, sites of recurrence (single vs multiple), status at last follow-up and length of survival from TC were captured. All patients had previously undergone primary cytoreduction followed by intravenous platinum-based chemotherapy and secondary cytoreduction for first recurrence. Staging was performed according to the FIGO guidelines (Benedet et al., 2000). Patients were considered eligible for tertiary cytoreduction if they fulfill the following criteria: good performance status (PS 0-1 ECOG), no extra-abdominal metastasis and/or unresectable intra-abdominal tumors (peritoneal carcinomatosis, multiple liver metastasis, involvement of porta hepatis, pancreatic head, abdominal wall or para-aortic lymph nodes above renal vein) and/or ascites at diagnostic preoperative instrumental procedures. All surgeries were performed via midline laparotomy by gynecologic oncologists.
Optimal cytoreduction was defined as residual disease of less than 1 cm for initial cytoreduction and 0.5 cm for secondary and tertiary cytoreduction. Patients with a platinum-free interval of six months or longer and less than six months were considered to have platinum-sensitive and platinum-resistant disease, respectively.
Outcomes of interest were overall survival (OS) and disease-spesific survival (DSS). OS was defined as the time from initial diagnosis to death or last followup. DSS was defined as the time from tertiary surgical cytoreduction to death or last follow-up.
Statistical analyses were performed using the SAS statistical software (version 9.1.3; The SAS Institute, Cary, NC, USA). A two-sided p-value <0.05 was considered statistically significant. Survival analysis was caried out using Kaplan Meier method. Acturial curves compared by the two tailed Logrank test with a statistical significance level of 0.05. Among the single factor analysis results, the independent variables of P<0.2 (to rule confounding more effectively, we used liberal criterion p <0.2 for inclusion of covariates in the model (Maldonado et al., 1993)) were selected as the potential predictors of the survival timeafter TC, then the multivariate proportional hazards regression model described by Cox was performed. The estimates of the models are given as adjusted hazard ration with 95 % confidence interval.

Results
The clinicopathologic characteristics of the patients and comparison of patients with regard to outcome of TC were presented in Table 1. Median age of the patients was 49.6 years (range, 30-67) and thirty-eight patients (72%) had stage III-IV disease at initial diagnosis. Seven patients had grade I (13.2%), sixteen patients had grade II (30.2%) and thirty patients had grade III disease (56.6%). Thirtyfive patients (66%) had serous histology, five patients had mucinous (9%), five patients had endometrioid (9%), one patient had clear cell histology (2%) and the remaining seven patients had mixed tumor histologies (14%). Time interval for the appearance of first recurrence after the initial surgery was ≤32 months in 31 patients (58%) and >32 months for the remaining 22 patients (42%). Secondary cytoreductive surgery for the initial recurrent disease was optimal (≤0.5 cm) in 32 patients (60%) and  DOI:http://dx.doi.org/10.7314/APJCP.2016.17.4.1909  Thirty-six (68%) had ≤20 months treatment free interval (TFI) and seventeen patients (32%) patients had >20 months TFI before the tertiary cytoreduction. Number of lines of chemotherapy regimens prior to TC was ≤3 in 48 patients (90%) and platinum sensitivity was observed in 39 patients (73%). CA125 level at TC was ≤190 IU/L in 32 patients (60%) and >190 IU/L in 21 patients (40%). Largest size of recurrent disease was ≤8 cm in 41 patients (77%) and >8cm in 12 patients (23%). Twenty six patients (49%) had optimal cytoreduction during tertiary debulking surgery while twenty seven had suboptimal cytoreduction (51%). Optimal initial cytoreduction, time to first recurrence, optimal secondary cytoreduction, time interval between secondary cytoreduction and secondary    Optimal initial cytoreduction, time to first recurrence, optimal secondary cytoreduction, time interval between SC and TC, size of the largest recurrence, status at last follow-up TC: Tertiary cytoredution. TFI: Treatment-free interval. NR: Not reported. NA: Not available; a Two retrospective studies authored by Leitao MM et al. 2004(9) and Fotopoulou et al. 2011(12) were excluded from the table while patients in those studies included in the following studies authored by Shih KK et al. 2010(8) and Fotopoulou et al. 2013 (15), respectively.; b Treatment-free survival, instead of DFS; c Overall survival, instead of DFS.
recurrence, size of recurrence, disease status at last followup were found to be significant risk factors to predict optimal TC (Table 1).
With a median follow up of 27.6 months after TC, disease specific survival (DSS) was 27 months (Table 2). It seems that probability of survival beyond 40 months is less than 0.3. In univariate analysis, outcomes of initial and tertiary cytoreduction (Figure 1, 2), time to first recurrence and serum CA-125 level prior to TC (Figure 3) were found to be significant prognostic factors. However, when we adjust for other factors in multivariate analysis, none of the variables were significant (Table 2).

Discussion
The role of TC in patients with recurrent EOC was first assessed in 2004 by Leitao et al and great benefit in patients who had undergone optimal TC and with a treatment free interval longer than 12 months was reported (Leitao et al., 2004). Following studies questioned about the best candidates for TC, prognostic effect of TC and the impact of residual disease after TC (Karam et al., 2007;Gultekin et al., 2008;Shih et al., 2010;Fotopoulou et al., 2011;Hızlı et al., 2012;Tang et al., 2013;Fotopoulou et al., 2013). Results of the literature review regarding TC are presented in Table 3. Only 8 retrospective studies were published to date with different results (Leitao et al., 2004;Karam et al., 2007;Gultekin et al., 2008;Shih et al., 2010;Fotopoulou et al., 2011;Hızlı et al., 2012;Tang et al., 2013;Fotopoulou et al., 2013). Therefore, the questions still remain unanswered.
In the first study in this area, Leitao et al. reported 26 patients with TC in 2004, from Memorial Sloan-Kettering Cancer Center (Leitao et al., 2004. In the following study from the same center, 77 patients who underwent TC were evaluated, including the 26 patients in their previous report (Shih et al., 2010). They suggested that TC for recurrence after secondary cytoreduction may be a very reasonable option especially in patients in whom an optimal (≤0.5 cm) cytoreduction, or preferably no gross residual disease, is thought to be feasible (Shih et al., 2010). In addition, single site recurrence was found to be the only predictor of complete tumor resection. Karam et al. similarly showed a survival benefit for patients who were able to undergo optimal TC (Karam et al., 2007). However, their findings indicated that the advantage of TC is limited to those patients with limited disease. Size of the largest tumor (<5 cm) was determined as a significant predictor of optimal TC (Karam et al., 2007).
In the study of Gultekin M et al, including 20 patients with TC, optimal primary and secondary cytoreductive surgeries were found to have significant favorable effect on survival in univariate analysis (Gultekin et al., 2008). However, multivariate analysis did not reveal any significant prognostic factor. It has been suggested that TC may not be helpful for survival in patients with secondary recurrence of EOC (Gultekin et al., 2008). In a subsequent study, the most pessimistic results were found with treatment-free survival of 14 and 0 months after optimal and suboptimal TC, respectively (p=0.018) (Hızlı et al., 2012). They could not find any clinicopathologic factor associated with TC outcome (Hızlı et al., 2012).
In a retrospective study from China, 83 patients were evaluated and complete tumor clearance was found to be associated with better prognosis (Tang et al., 2013). It has been shown that patients with carcinomatosis or tumor sites in the middle and upper abdomen have lower chance to undergo optimal TC (Tang et al., 2013). They concluded that a prospective randomised study is needed to determine the role of TC in secondary recurrence of EOC (Tang et al., 2013).
The largest, unicenter report came from Charite´ University Medical Center Berlin in 2011 (Fotopoulou et al., 2011). Unlike other studies, which patients who had undergone palliative surgery were excluded, Fotopoulou et al. evaluated all consecutive patients without any exclusion (Fotopoulou et al., 2011). Complete TC, serous papillary histology and 3 or more years interval to primary diagnosis were found to have significant positive effect on survival (Fotopoulou et al., 2011). Tumor involvement of the upper and middle abdomen and peritoneal carcinomatosis were the only two parameters affecting tumor resection (Fotopoulou et al., 2011).
A recent, international multicenter study including 406 patients showed that even in the last stage of the disease, complete macroscopic tumor clearance has a significant favorable impact on overall survival (Overall survival: 49 vs 12 months) (Fotopoulou et al., 2013). Unlike previous studies, they showed for the first time that after controlling for resiual status, benefit of complete TC is clear, overruling the presence of peritoneal carcinomatosis which no prognostic significance was found. In addition, postoperative systemic chemotherapy was shown for the first time to have a significant impact on overall survival. However, a selection bias was suspected while only those patients who were fit enough and able to tolerate chemotherapy had already more favorable prognosis than patients who were not able to tolerate or even so advanced that chemotherapy was not indicated (Fotopoulou et al., 2013). Progression-free interval was the most important survival determinant.
In the present analysis, we assessed the value of TC in recurrent EOC patients. Optimal cytoreduction was achieved in 26 out of 53 patients (49%). Optimal initial cytoreduction, time to first recurrence, optimal secondary cytoreduction, time interval between SC and TC, size of the largest recurrence, status at last follow-up were the most important predictors of the optimal TC. Optimal cytoreduction in initial and tertiary surgery and CA-125 level prior to TC were independent prognostic factors in univariate analysis.
Our study has some limitations, similar to previous studies. The main limitations are its retrospective design, the relatively small number of patients included and the lack of systemic evaluation of the indication for TC.
Including our data, the literature review clearly shows that maximal surgical effort should be made in TC, while patients underwent optimal TC have a better survival. Thus, to select the patients with secondary recurrent EOC in whom optimal cytoreduction can be achieved, comes into question. Prospective trials and further studies with larger patient populations are needed to provide an answer.