ORIGINAL STUDY
Primary Surgery or Interval Debulking for Advanced Epithelial Ovarian Cancer: Does It Matter? Algirdas Markauskas, MD,* Ole Mogensen, MD, DMSci,* Rene´ dePont Christensen, PhD,Þ and Pernille Tine Jensen, MD, PhD*
Objective: The aim of the present study was to investigate the surgical complexity, the postoperative morbidity, and the survival of the women after primary debulking surgery (PDS) and neoadjuvant chemotherapy followed by interval debulking surgery (NACT-IDS) for advanced epithelial ovarian cancer. Materials and Methods: We consecutively included all patients who underwent debulking surgery at our institution between January 2007 and December 2012 for stages IIIc and IVof epithelial ovarian cancer. Results: Of the 332 patients included, 165 (49.7%) underwent PDS, and 167 (50.3%) had NACT-IDS. Complete intraperitoneal cytoreduction was achieved in 70.9% after PDS and in 59.9 % after NACT-IDS. Residual disease of greater than 1 cm was left in 18.5% and 27.5% after PDS and NACT-IDS, respectively. Compared with NACT-IDS, PDS was associated with higher surgical complexity (P G 0.001), longer operating time (P G 0.001), greater blood loss (P G 0.001), longer hospitalization (P = 0.001), and a higher rate of major postoperative complications (26.7% vs 16.8%). No statistical difference in the median overall survival (OS) was found between the patients having complete cytoreduction and residual disease of 1 cm or less after NACT-IDS. Furthermore, no statistical difference in the median OS was found between the patients with macroscopic residual disease (e1 vs 91 cm) after NACT-IDS. Patients with residual disease of greater than 1 cm after PDS had a median OS of 15 months. Conclusions: We suggest that NACT-IDS may be a better treatment alternative for the group of highly selected women not suitable for PDS, where expected suboptimal cytoreduction does not have any appreciable survival benefit and exposes them for unnecessary risks. A substantial number of women who receive either PDS or NACT-IDS have greater than 1 cm of tumor tissue left after the operation. These women probably have no survival benefit from the operation, and future studies should focus on how to select these women preoperatively. Key Words: Debulking surgery, Epithelial ovarian cancer, Surgical complexity, Postoperative morbidity, Survival Received May 11, 2014, and in revised form July 15, 2014. Accepted for publication July 15, 2014. (Int J Gynecol Cancer 2014;24: 1420Y1428)
*Department of Gynaecology & Obstetrics, Odense University Hospital; and †Research Unit of General Practice, University of Southern Denmark, Odense, Denmark. Copyright * 2014 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000241
1420
Address correspondence and reprint requests to Algirdas Markauskas, MD, Department of Gynaecology & Obstetrics, Odense University Hospital, Sdr Boulevard 29, 5000 Odense C, Denmark. E-mail: [email protected]. The authors declare no conflicts of interest.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
theoretical benefits of primary cancer surgery are T hemultiple, including removal of resistant clones of tumor
cells, increasing the growth fraction of the tumor, maximizing the effect of chemotherapy, reducing the risk for drug resistance, improving the tumor perfusion, and enhancing the immunological competence of the patient.1,2 In epithelial ovarian cancer (EOC), an inverse relationship between the size of the residual disease after primary surgery, and overall survival (OS) was first established by Griffiths.3 The definition of ‘‘optimal cytoreduction’’ has changed over time, and the term is currently reserved for those with no macroscopic residual disease.4Y6 This is supported by the randomized study of the European Organisation for Treatment and Research of Cancer and the National Cancer Institute of Canada (EORTC-NCIC), showing that optimal cytoreduction should be defined as total tumor resection without macroscopic residuals.7 This often demands extensive upper abdominal surgery, which carries a high risk for postoperative complications. Neoadjuvant chemotherapy (NACT) before cytoreductive surgery of women with advanced EOC (International Federation of Gynecology and Obstetrics [FIGO] stages IIIc and IV) reduces the surgical complexity and shows a tendency to decrease the postoperative morbidity and mortality.7 In addition, NACT may increase the rate of complete cytoreduction. Two recently published randomized trials compared primary debulking surgery (PDS) with NACT and interval debulking surgery (NACT-IDS) in women with advanced EOC.7,8 It was demonstrated that the median OS did not differ in the 2 arms, and NACT-IDS was associated with lesser postoperative morbidity and mortality. The studies prompted a paradigm shift in the treatment for women with advanced EOC, going from primary surgery to NACT-IDS. However, it is a biological concern that NACT in the presence of a large tumor burden may increase the likelihood of mutations leading to chemotherapyresistant clones and potentially shorter progression-free survival (PFS) and OS.9,10 Our center adopted the principles of advanced cytoreductive surgery for the treatment of EOC in 2007.6 The center is population based and serves from 20% (2007Y2010) to 35% (2011 and ongoing) of the Danish population. Some of the data on the present population has been included in a nationwide Danish study of Fago¨-Olsen et al.11 The main purpose of the present study is to analyze the surgical complexity, the postoperative morbidity, and the survival of women with FIGO stages IIIc to IV EOC, treated with PDS and NACT-IDS.
MATERIALS AND METHODS All patients with FIGO stage IIIc to IVepithelial ovarian, fallopian tube, and primary peritoneal cancers treated at our institution between January 1, 2007, and December 31, 2012, were identified. Only patients, who underwent debulking surgery, were included. Patients who underwent laparotomy before NACT were included in the NACT-IDS group, because we considered laparotomy as being diagnostic as no effort to achieve maximal cytoreduction was made. Electronic medical files of the patients were used for data collection. All patients were scheduled to receive 6 cycles of platinum-based chemotherapy with standard regimen being
Primary Surgery or Interval Debulking
carboplatin + paclitaxel. No patients received intraperitoneal (IP) chemotherapy as this treatment is not yet approved for gynecological malignancies in Denmark. The decision to perform either PDS or NACT-IDS was made at a multidisciplinary team conference and based on the probability of complete IP cytoreduction. All patients were preoperatively evaluated by magnetic resonance imaging (MRI) and positron emission tomography and computed tomography (PET-CT) scans. In some cases, open laparoscopy was performed to exclude substantial serosal disease of the small bowel. Patients where complete IP cytoreduction seemed not to be achievable because of disseminated disease (tumor 9 2 cm behind porta hepatis, extensive intestinal serosal disease, bulky lymph nodes above the level of the left renal vein) or poor general condition (combination of age 9 75Y80 years, albumin G 3.0 g/dL, American Society of Anesthesiologists Group [ASA] 9 2) were scheduled to NACT.12 Women with FIGO stage IV disease were treated equally to stage IIIc, and PDS was performed if complete IP cytoreduction was estimated to be possible, unless the patient had multiple metastases in the liver and/or the lungs or substantial malignant pleural effusions. All patients were routinely evaluated clinically and by MRI scans by the same team of experts after 2 cycles of NACT and then offered IDS after the third cycle, unless they had evidence of progressive disease or the general condition of the patient still was too poor for surgery. All operations were performed by a gynecological oncologist. Bowel, hepatopancreatic surgery, and lymphadenectomy above the renal vein were performed in collaboration with experienced abdominal surgeons. Pelvic and paraaortic lymphadenectomy were performed if diagnosed as pathological on imaging or if pathological bulky nodes were identified during surgical exploration. All patients received the same postoperative care, including 24 to 48 hours observation at the intensive care unit, if major peritoneal stripping was performed. Surgical complexity was graded using a surgical complexity scoring system based upon the complexity and the number of surgical procedures performed,13 for example, hysterectomy + salpingo-oophorectomy giving, 1 point; omentectomy, 1 point; paraaortic lymphadenectomy, 1 point; pelvic lymphadenectomy, 1 point; small bowel resection, 1 point; pelvic peritoneum stripping, 1 point; abdominal peritoneum stripping, 1 point; large bowel resection, 2 points; splenectomy, 2 points; diaphragm stripping/resection, 2 points; splenectomy, 2 points; liver resection, 2 points; rectosigmoidectomy with anastomosis, 3 points. This gives 3 surgical complexity score (SCS) groups: 1 (low), e3 points; 2 (intermediate), 4 to 7 points; and 3 (high), Q8 points. Postoperative complications were evaluated using the Memorial Sloan-Kettering Cancer Center surgical secondary events grading system14: 0, no events observed within 30 days after operation 1, use of oral medications or bedside interventions to treat an event 2, use of intravenous medications, total parenteral nutrition, enteral nutrition, or blood transfusion to treat an event 3, interventional radiology, therapeutic endoscopy, or operation required to treat an event
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1421
International Journal of Gynecological Cancer
Markauskas et al
& Volume 24, Number 8, October 2014
TABLE 1. Baseline characteristics of the patients
Age, median (range), years BMI, median (range) Cancer type, n (%) Epithelial ovarian cancer Primary peritoneal cancer Fallopian tube cancer Histology, n (%) Serous Mucinous Clear cell Endometrioid Carcinosarcoma Undifferentiated FIGO stage, n (%) IIIc IV ASA group, n (%) 1 2 3 4
PDS, N = 165
NACT-IDS, N = 167
P
65 (24Y89) 24.6 (15.2Y40.8)
66 (28Y88) 24.8 (17.6Y42.4)
0.814* 0.826* 0.012†
127 (77.0) 20 (12.1) 18 (10.9)
118 (70.7) 39 (23.4) 10 (6.0)
151 (91.5) 9 (5.5) 3 (1.8) 1 (0.6) 1 (0.6) 0 (0.0)
162 (97.0) 1 (0.6) 2 (1.2) 1 (0.6) 0 (0.0) 1 (0.6)
127 (77.0) 38 (23.0)
111 (66.5) 56 (33.5)
0.028†
0.039†
0.193† 61 (37.0) 87 (52.7) 16 (9.7) 1 (0.6)
50 (29.9) 105 (62.9) 12 (7.2) 0 (0.0)
*Median test by Pearson W2. †Fischer exact test. BMI, body mass index; FIGO, International Federation of Gynecology & Obstetrics.
4, residual or lasting disability that requires major rehabilitation or organ resection 5, an event that resulted in the death of the patient Major postoperative complications were defined as grade 3 to 5 events. Chemotherapy was considered postponed if delayed more than 6 weeks after the operation and abandoned if the patient did not receive the full 6 cycles. Patients were followed until September 15, 2013, or death.
Statistical Analysis The OS time was calculated in days from the date of the decision making regarding treatment modality (PDS vs NACT-IDS) to death or to the date of the last follow-up. The PFS time was calculated in months from the date of decision making to the approximate date of recurrence/progression of the disease. The Kaplan-Meier method was used to estimate survival. Multiple Cox regression was used to compare survival between the relevant groups, adjusted for suitable confounders. To assess the differences in the baseline and the surgical characteristics of the PDS and NACT-IDS populations, we used a W2-based median test for continuous variables, for example, age and Fischer exact test for categorical variables, for example, disease stage. We chose not
1422
to perform statistical comparison of OS and PFS between the PDS and NACT-IDS groups, as patients in our retrospective study were highly selected, that is, patients submitted to NACT-IDS had more extensive disease and/or poorer general condition. Software used was Stata 11.2 (StataCorp, College Station, TX).
RESULTS Four hundred seven women with advanced EOC were identified. Seventy-five (18.4%) were excluded, as they had no debulking surgery at any time during their treatment, either because of progressive disease during chemotherapy (n = 59, 78.7%) or poor general condition (n = 16, 21.3%). Three hundred thirty-two women were included in the study of whom165 (49.7 %) underwent PDS and 167 (50.3 %) were treated with NACT-IDS. Twenty-eight (16.8%) in the NACTIDS group underwent diagnostic laparotomy before NACT. The preoperative and surgical characteristics of the study population are summarized in Tables 1 and 2, respectively. Patients submitted to PDS had more complex surgery (P G 0.001) with higher rates of diaphragmatic striping (P = 0.02) and bowel resection (P = 0.005), longer operating time (P G 0.001), greater perioperative blood loss (P G 0.001), and longer hospitalization (P = 0.001). Complete IP cytoreduction was achieved in 117 (70.9%) patients of the PDS group * 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
Primary Surgery or Interval Debulking
TABLE 2. Surgical characteristics of the patients PDS, N = 165 SCS, median (range) SCS groups, n (%) Low SCS Intermediate SCS High SCS Diaphragmatic surgery, n (%) Bowel stoma, n (%) Other radical procedures, n (%) Liver resection Cholecystectomy Pancreas resection Gastric resection Splenectomy Urinary bladder resection Residual disease, n (%), cm 0 e1 91 Blood loss, median (range), mL Length of the operation, median (range), minutes Hospitalization, median (range), days
6 (1Y15) 16 80 69 66 78 35 5 1 5 1 21 2
(9.7) (48.5) (41.8) (40) (47.3) (21.2) (3.0) (0.6) (3.0) (0.6) (12.7) (1.2)
NACT-IDS, N = 167
P
4 (1Y12)
G0.001* G0.001†
65 62 40 46 53 23 3 4 3 1 11 1
(38.9) (37.1) (24.0) (27.5) (31.7) (13.8) (1.8) (2.4) (1.8) (0.6) (6.6) (0.6)
0.020† 0.005† 0.084†
0.030† 117 22 26 1440 210 9
(70.9) (13.3) (15.8) (100Y13000) (30Y496) (3Y54)
100 21 46 800 150 7
(59.9) (12.6) (27.5) (20Y6000) (30Y480) (2Y46)
G0.001* G0.001* 0.001*
*Median test using Pearson W2. †Fischer exact test. RD, residual disease.
and 100 (59.9%) of the NACT-IDS group. The proportion of patients with stage IV disease who had complete IP cytoreduction was 28 (24%) and 39 (39%) in the PDS and NACT-IDS group, respectively. Debulking to less than or equal to 1 cm of residual disease could be achieved in 139 (84.2%) women of the PDS group and 121 (72.5%) of the NACT-IDS group. Residual disease of greater than 1 cm was left in 26 (15.8%) women after PDS and 46 (27.5%) after NACT-IDS. Most patients experienced no or just minor events (grades 0Y2) after debulking surgery (73.7% and 83.3% in the PDS and NACT-IDS group, respectively). However, the rate of major complications (grade 3Y5 events) was greater in the PDS group compared with NACT-IDS (44 [26.7%] vs 28 [16.8%]). The major complications are given in Table 3. Twenty-four (14.5%) in the PDS group needed reoperation to treat an event: 5 having intra-abdominal abscess, 8 having wound dehiscence, 6 having bowel perforation, 4 having fistula, and 1 having urinary tract leakage. In the NACT-IDS group, 15 (9%) required reoperation: 4 having intra-abdominal abscess, 4 having wound dehiscence, 3 having bowel perforation, 1 having bowel obstruction, and 3 having wound infection. Twenty-two (6.6%) patients experienced a long lasting disability (grade 4 event) after debulking surgery, 12 (7.3%) in the PDS group and 10 (6.0%) in the NACT-IDS group, and
it was associated with surgery of high complexity in 8 (66.7%) and 6 (60%) patients, respectively. Complete cytoreduction was achieved in 10 (83%) and 10 (100%) patients, who experienced a grade 4 event after PDS and NACT-IDS, respectively. Six (3.6%) in the PDS group died during the first 30 days after the operation (grade 5 event), and only 2 (33.3%) of these women achieved complete cytoreduction. Surgery of high complexity was performed in only 1 (16.7%) case. Four patients died because of sepsis, one had pulmonary embolism, and one died because of deteriorating general condition after PDS. In the NACT-IDS group, 3 women (1.8%) died during the first 30 days after surgery and all had complete IP cytoreduction. One patient underwent surgery of low complexity (33%). The causes of death were pulmonary embolism, cerebral thrombosis, and sepsis. The median OS and PFS in relation to the residual disease and type of debulking surgery are demonstrated in Figures 1 and 2, respectively. Five (3%) patients had complete radiological and biochemical remission (Cancer Antigen 125 G 35 U/mL) after NACT and all had microscopic tumor residuals only. The median OS of those patients was 31 months, and they had no postoperative events after IDS. The median delay of chemotherapy in patients, where chemotherapy was postponed, was 7 weeks for both groups.
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1423
International Journal of Gynecological Cancer
Markauskas et al
& Volume 24, Number 8, October 2014
TABLE 3. Specification of postoperative events PDS (N = 165), n (%) Vascular events Venous thromboembolism Cardiovascular events Cerebrovascular events Major infections Sepsis Wound infection Intra-abdominal abscess Bleeding/hematoma Pleural effusions Wound dehiscence Bowel perforation (overall) Anastomotic leak Ischemia/necrosis Bowel obstruction Urinary tract leakage Pancreatic leakage Fistula Surgical events Grades 0Y2 Grades 3Y5 Reoperation Postponed chemotherapy Abandoned chemotherapy
NACT-IDS (N = 167), n (%)
13 (7.9) 7 (4.2) 6 (3.6) 0 (0.0) 34 (20.6) 15 (9.1) 2 (1.2) 17 (10.3) 4 (2.4) 13 (7.9) 10 (6.1) 6 (3.6) 2 (1.2) 4 (2.4) 2 (1.2) 2 (1.2) 2 (1.2) 5 (3.0)
7 4 1 2 21 6 8 13 11 6 4 3 1 2 2 1 0 3
(4.2) (2.4) (0.6) (1.2) (12.6) (3.6) (4.8) (7.8) (6.6) (3.6) (2.4) (1.8) (0.6) (1.2) (1.2) (0.6) (0.0) (1.8)
121 (73.3) 44 (26.7) 24 (14.5) 21 (12.7) 23 (13.9)
139 28 15 9 9
(83.3) (16.8) (9.0) (5.4) (5.4)
P* 0.174
0.055
0.110 0.103 0.109 0.335
1.000 0.622 0.246 0.500 0.126
0.127 0.022 0.009
*Fischer exact test.
The median number of chemotherapy cycles in PDS patients where chemotherapy was abandoned was difficult to estimate (mean value, 1.57), as 13 (56.5%) did not receive postoperative chemotherapy at all. Five (55.6%) NACT-IDS patients did not receive postoperative chemotherapy, and the remaining 4 (44.4%) received a median of 4 cycles in total. Patients in the NACT-IDS group received at least 3 cycles of chemotherapy before IDS. A Cox regression analysis of the covariates in relation to survival for the PDS and NACT- IDS groups are given in Table 4. When adjusted for multiple variables, residual disease and age were significantly associated with median OS after PDS. Residual disease of greater than 1 cm was associated with an almost 3-fold increased risk for death after PDS compared with complete cytoreduction (hazard ratio [HR], 2.921; 95% confidence interval [CI], 1.643Y5.195; P = 0.0003). There was a trend that residual disease of 1 cm or less was associated with worse median OS when compared with complete cytoreduction after PDS (HR, 1.78; 95% CI, 0.98Y3.22; P = 0.0564). In the NACT-IDS group, FIGO stage IV (HR, 1.63; 95% CI, 1.01Y2.65), residual disease of greater than 1 cm (HR, 1.92; 95% CI, 1.15Y3.19), and ASA score of greater
1424
than 2 (HR, 3.36; 95% CI, 1.35Y8.35) were associated with significantly worse median OS.
DISCUSSION In the present cohort study, we have analyzed surgical complexity, postoperative morbidity, and OS of women treated at our institution with debulking surgery for FIGO stage IIIc and IV epithelial ovarian, fallopian tube, and primary peritoneal cancers. Although most women in the PDS and NACT-IDS groups (73.7% and 83.3%, respectively) experienced no or just minor events after surgery, our study confirms that the risk for major complications (22%) or even death (2.7%) should not be neglected, while counseling and selecting patients for surgery. The rate of major postoperative complications was higher after PDS compared with NACT-IDS, and there was a positive relationship with surgical complexity. It is problematic to compare postsurgical morbidity between various studies because of potential reporting bias and differences in the selection of patients. Chi et al15 reported 22% of grade 3 to 5 events after upfront extensive upper abdominal surgery for advanced EOC and a mortality rate of 1.4%. The * 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
Primary Surgery or Interval Debulking
likely, they should have been referred to NACT-IDS. Selection of patients for upfront surgery is a major issue as prediction of optimal debulking surgery is very difficult. Use of open laparoscopy together with diffusion-weighted MRI and PET-CT seem to be promising in providing additional information.18 All women in our study had preoperative PET-CT and MRI scans, and the rate of upfront surgery was lower compared with other centers, but the rate of complete cytoreduction was high. However, 28 (16.8%) women underwent laparotomy before NACT and represent failures in preoperative triaging, as they were found with more extensive disease than otherwise expected on imaging. In all cases, no major effort for upfront cytoreduction was made, only sufficient tissue material was secured for the proper histological diagnosis of malignancy, and the patients were scheduled to NACT-IDS. So far, the optimal method for selecting women for PDS is lacking, and improvements are highly warranted as operation should only be performed if a clear survival benefit is expected (ie, complete cytoreduction). Another pending question is if some of the women who were radically operated after NACT-IDS should have been operated
FIGURE 1. OS by residual disease and type of debulking surgery. RD, residual disease; [n to .], no estimation of the upper limit of the confidence interval for the survival median, because of short observational time. overall rate of major complications after PDS at the Memorial Sloan-Kettering Cancer Center was reported to 9%, and the rate of grade 5 complications was 0.7%.16 The rate of major postoperative complications and mortality reported in a retrospective multicenter study of Rafii et al17 were 11.5% and 0.5%, respectively, and the rate of complications was higher after PDS compared with NACT-IDS (OR, 2.17 [1.16Y4.09]). A mortality rate of 2.5% and 0.7% for the PDS and NACTIDS arms, respectively, was reported in the EORTC-NCIC trial, and the corresponding figures were 5.6% and 0.5% in the CHORUS trial.7,8 However, compared with our results, the patients were younger in the EORTC-NCIC trial and the study of Rafii et al, and the proportion of patients with FIGO stage IV disease was lower in all 3 studies cited previously.7,8,17 In the present study, two thirds of the postoperative PDS deaths (n = 4) occurred in patients in whom optimal cytoreduction could not be achieved and despite of surgery of low complexity. Residual disease of greater than 1 cm was left in 26 women (15.8%) after attempting radical surgery at PDS. These women have not gained any survival benefit, and most
FIGURE 2. PFS by residual disease and type of debulking surgery. RD, residual disease; [n to .], no estimation of the upper limit of the confidence interval for the survival median, because of short observational time.
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1425
International Journal of Gynecological Cancer
Markauskas et al
& Volume 24, Number 8, October 2014
TABLE 4. Multivariate Cox regression analysis showing HRs and CIs for OS after PDS and IDS PDS
NACT-IDS
Variables
HR
95% CI
P
HR
95% CI
P
Age BMI ASA = 1 ASA e 2 ASA 9 2 Stage IIIc Stage IV RD = 0 cm RD e 1 cm RD 9 1 cm Observations
1.02 0.99 1 0.74 1.19 1 0.93 1 1.78 2.92 165
1.00Y1.05 0.94Y1.04 Y 0.43Y1.29 0.51Y2.78 Y 0.51Y1.72 Y 0.98Y3.22 1.64Y5.19
0.0438 0.6993 Y 0.2882 0.6908 Y 0.8294 Y 0.0564 0.0003
1.02 0.99 1 1.38 3.36 1 1.63 1 0.93 1.92 167
0.99Y1.05 0.94Y1.04 Y 0.80Y2.40 1.35Y8.35 Y 1.01Y2.65 Y 0.41Y2.10 1.15Y3.19
0.1488 0.7201 Y 0.2496 0.0092 Y 0.0478 Y 0.8626 0.0119
BMI, body mass index; RD, residual disease.
upfront? It remains speculative if the survival of these women would have improved if complete cytoreduction on PDS could have been achieved. A meta-analysis by Bristow and Chi10 showed that each cycle of chemotherapy delivered before surgery decreased median survival by 4.1 months. However, the studies of Vergote et al7 and Kehoe et al8 clearly showed no survival benefit of PDS versus NACT-IDS in their populations of patients with advanced EOC. A higher percentage of women having residual disease of greater than 1 cm after NACT-IDS can partly be explained by that this subgroup also included women who were technically debulkable to less than or equal to 1-cm residual disease but were expected surgical effort to achieve it was judged too risky in relation to the extent of the disease, the general condition of the patient, and the expected survival benefit from incomplete cytoreduction. The rate of upfront cytoreduction to 1 cm or less varies widely among institutions, but different surgical groups have suggested that more than 75% should be acceptable.19 Cytoreduction to 1 cm or less was achieved in 78% in our study. The high rates of complete IP cytoreduction in our study (Q60% in both groups) can be attributed to the strict selection of the patients for debulking surgery (especially PDS) combined with will and skills of the gyne-oncological surgical team. However, we had a lower median OS compared with a study of Chi et al,16 reporting a median OS of 50 months for all women (FIGO stages IIIc-IV) treated with PDS and 78 months for the women without residual disease. The difference in the survival between the 2 studies may partly be explained by an older population and a higher proportion of stage IV disease in our study.16A direct comparison of the results regarding OS from our retrospective study with data from the 2 randomized trials7,8 should be interpreted with caution. The Gynecologic Oncology Group study, GOG 172, showed that IP chemotherapy improves survival of the women with stage III disease having less than or equal to 1-cm
1426
residual tumor after PDS by approximately 17 months.20 This treatment is currently not recognized in Denmark, and it remains speculative whether addition of IP chemotherapy would have translated into more superior OS in the present population. Data on the use of IP chemotherapy after NACTIDS are still limited, as the results of larger prospective studies are pending.21 We chose not to perform statistical comparison of the survival between the PDS and NACT-IDS groups, because the groups were highly selected on the extent of the disease and/or the general condition of the patients. However, the difference in the median OS between the subgroups with complete IP cytoreduction (51.7 vs 33.3 months after PDS and NACT-IDS, respectively) does raise some speculations. The NACT-IDS group had lower rates of postoperative complications and higher rates of completion of chemotherapy. Similar results were found in a study by Fago¨-Olsen et al,11 which showed better OS in patients after complete IP cytoreduction after PDS compared with NACT-IDS (55.5 vs 36.7 months) and increased risk for death after 2 years of follow-up after NACT-IDS. The difference in the survival observed in the present study could be explained by more extensive disease among patients referred to NACT-IDS compared with PDS, for example, 39% versus 24% with stage IV having complete IP cytoreduction in our study. Vergote et al7 have confirmed that complete resection of all macroscopic diseases during debulking surgery is the single most important prognostic factor in advanced EOC. Therefore, it remains unclear why significantly higher rates of complete cytoreduction (19% vs 51%) and lower morbidity after NACT-IDS do not translate into more superior OS in the study of Vergote et al.7 We speculate that some patients who relapse shortly after complete cytoreduction after NACT-IDS may have unidentified macroscopic disease left. This point of view is supported by the observation from our study that the median OS of women achieving complete cytoreduction after * 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
NACT-IDS did not differ statistically significant from those having residual disease of 1 cm or less (33.3 vs 28 months; HR, 0.93; P = 0.8626). It has recently been demonstrated that carcinomatous areas have a benign visual appearance more often after NACT than at PDS and chemotherapy-induced fibrosis may hide vital tumor elements in otherwise obvious fibrotic tissue.22 This may lead to insufficient surgery and overestimation of NACT-IDS women with ‘‘no macroscopic residuals’’ and questions whether macroscopic tumor clearance during IDS is rather illusionary and may not be comparable with the same definition as during PDS. Furthermore, a special attention must be paid to the women in the NACTIDS population, where lacking statistical difference in the median OS between the subgroups having macroscopic residual disease (e1 vs 91 cm) may point toward ‘‘all or nothing’’ solution. The strength of the present study is its population-based nature, avoiding some of the bias that may occur in institutionalbased investigations. No heterogeneity is present regarding the choice of treatment modalities as all decisions were made by the same team of experts at multidisciplinary team conferences. A review of the Electronic Patient Files from our institution allowed inclusion of all cases meeting the inclusion criteria as well as a precise registration of surgical characteristics and postoperative events. Our study is limited by its retrospective nature and relatively short follow-up. Patients treated with chemotherapy alone are not represented in our data, as the main inclusion criterion was debulking surgery. Within the limits of our study, we suggest that NACTIDS may be a better treatment alternative for the group of highly selected women where complete IP cytoreduction cannot be expected upfront, as suboptimal PDS does not have any appreciable survival benefit and exposes the patients for unnecessary postoperative complications. The NACT-IDS is, until further evidence is presented, in our center considered as a possibility of decreasing surgical complexity and postoperative morbidity without compromising survival for a carefully selected group of women in whom alternative surgical outcome otherwise would be suboptimal tumor resection. Complete IP cytoreduction should be aimed in all cases of debulking surgery. A substantial number of women who receive either PDS or NACT-IDS have greater than 1 cm of tumor tissue left after the operation. These women probably have no survival benefit from the operation, and future studies should focus on how to select these women preoperatively.
REFERENCES 1. Bookman MA, Young RC. Principles of chemotherapy in gynecologic cancer. In: Hoskins WJ, Perez CA, Young RC, eds. Principles and Practice of Gynaecologic Oncology. Philadelphia, PA: Lippincott Williams & Wilkins; 2000:404Y407. 2. De Vita VT. The relationship between tumor mass and resistance to chemotherapy. Implications for surgical adjuvant treatment of cancer. Cancer. 1983;51:1209Y1220. 3. Griffiths CT. Surgical resection of tumor bulk in the primary treatment of ovarian cancer. Natl Cancer Inst Monogr. 1978;42:131Y136.
Primary Surgery or Interval Debulking
4. Chi DS, Eisenhauer EL, Lang J, et al. What is the optimal goal of primary cytoreductive surgery for bulky stage IIIc epithelial ovarian carcinoma (EOC)? Gynecol Oncol. 2006;103: 559Y564. 5. Stuart GC, Kitchener H, Bacon M, et al. 2010 Gynecologic Cancer Inter Group (GCIG) consensus statement on clinical trials in ovarian cancer: report from the Fourth Ovarian Cancer Consensus Conference participants of 4th Ovarian Cancer Consensus Conference (OCCC), Gynecologic Cancer Intergroup. Int J Gynecol Cancer. 2011;21: 750Y755. 6. Zivanovic O, Aldini A, Carlson JW, et al. Advanced cytoreductive surgery: American perspective. Gynecol Oncol. 2009;114:S3YS9. 7. Vergote I, Trope CG, Amant FA, et al. Neoadjuvant chemotherapy or primary surgery in stage IIIcYIV ovarian cancer. N Engl J Med. 2010;363:943Y953. 8. Kehoe S, Hook J, Nankivell M, et al. Chemotherapy or upfront surgery for newly diagnosed advanced ovarian cancer: Results from the MRC CHORUS trial. Paper presented at: 2013 American Society of Clinical Oncology Annual Meeting; May 2013; Chicago, IL. 9. Goldie JH, Coldman AJ. A mathematic model for relating the drug sensitivity of tumors to their spontaneous mutation rate. Cancer Treat Rep. 1979;63:1727Y1733. 10. Bristow RE, Chi DS. Platinum-based neoadjuvant chemotherapy and interval surgical cytoreduction for advanced ovarian cancer: a meta-analysis. Gynecol Oncol. 2006;103: 1070Y1076. 11. Fago¨-Olsen CL, Ottesen B, Kehlet H, et al. Does neoadjuvant chemotherapy impair long-term survival for ovarian cancer patients? A nationwide Danish study. Gynecol Oncol. 2014;132: 292Y298. 12. Aletti GD, Eisenhauer EL, Santillan A, et al. Identification of patient groups at highest risk from traditional approach to ovarian cancer treatment. Gynecol Oncol. 2011; 120:23Y28. 13. Aletti GD, Santillan A, Eisenhauer EL, et al. A new frontier for quality of care in gynaecologic oncology surgery: multi institutional assessment of short-term outcomes for ovarian cancer using a risk-adjusted model. Gynecol Oncol. 2007;107:99Y106. 14. Martin RCII, Brennan MF, Jaques DP. Quality of complication reporting in the surgical literature. Ann Surg. 2002;235: 803Y813. 15. Chi DS, Zivanovic O, Levinson KL, et al. The incidence of major complications after the performance of extensive upper abdomen surgical procedures during primary cytoreduction of advanced ovarian, tubal, and peritoneal carcinomas. Gynecol Oncol. 2010;119:38Y42. 16. Chi D, Musa F, Dao F, et al. An analysis of patients with bulky advanced stage ovarian, tubal and peritoneal carcinoma treated with primary debulking surgery (PDS) during an identical time period as the randomized EORTC-NCIC trial of primary debulking surgery vs neoadjuvant chemotherapy (NACT). Gynecol Oncol. 2012;124:10Y14. 17. Rafii A, Stoeckle E, Jean-Laurent M, et al. Multi-center evaluation of post-operative morbidity and mortality after optimal cytoreductive surgery for advanced ovarian cancer. PLoS One. 2012;7:e39415. 18. Fagotti A, Ferrandina G, Fanfani F, et al. Prospective validation of a laparoscopic predictive model for optimal cytoreduction
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1427
Markauskas et al
International Journal of Gynecological Cancer
in advanced ovarian carcinoma. Am J Obstet Gynecol. 2008;199:642Y646. 19. Bristow RE, Eisenhauer EL, Santillan A, et al. Delaying the primary surgical effort for advanced ovarian cancer: a systematic review of neo-adjuvant chemotherapy and interval cytoreduction. Gynecol Oncol. 2007;104:480Y490. 20. Armstrong D, Bundy B, Wenzel L, et al. Phase iii randomized trial of intravenous cisplatin and paclitaxel versus an intensive regimen of intravenous paclitaxel, intraperitoneal cisplatin, and intraperitoneal paclitaxel in stage iii ovarian cancer: a
1428
& Volume 24, Number 8, October 2014
Gynecologic Oncology Group study. N Engl J Med. 2006;354:34Y43. 21. Mackay HJ, Provencheur D, Heywood M, et al. Phase II/III study of intraperitoneal chemotherapy after Neoadjuvant chemotherapy for ovarian cancer: ncic ctgov.21. Curr Oncol. 2011;18:84Y90. 22. Hynninen J, Lavonius M, Oksa S, et al. Is perioperative visual estimation of intra-abdominal tumor spread reliable in ovarian cancer surgery after neoadjuvant chemotherapy? Gynecol Oncol. 2013;128:229Y232.
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
Primary Surgery or Interval Debulking for Advanced Epithelial Ovarian Cancer: Does It Matter? Algirdas Markauskas, MD,* Ole Mogensen, MD, DMSci,* Rene´ dePont Christensen, PhD,Þ and Pernille Tine Jensen, MD, PhD*
Objective: The aim of the present study was to investigate the surgical complexity, the postoperative morbidity, and the survival of the women after primary debulking surgery (PDS) and neoadjuvant chemotherapy followed by interval debulking surgery (NACT-IDS) for advanced epithelial ovarian cancer. Materials and Methods: We consecutively included all patients who underwent debulking surgery at our institution between January 2007 and December 2012 for stages IIIc and IVof epithelial ovarian cancer. Results: Of the 332 patients included, 165 (49.7%) underwent PDS, and 167 (50.3%) had NACT-IDS. Complete intraperitoneal cytoreduction was achieved in 70.9% after PDS and in 59.9 % after NACT-IDS. Residual disease of greater than 1 cm was left in 18.5% and 27.5% after PDS and NACT-IDS, respectively. Compared with NACT-IDS, PDS was associated with higher surgical complexity (P G 0.001), longer operating time (P G 0.001), greater blood loss (P G 0.001), longer hospitalization (P = 0.001), and a higher rate of major postoperative complications (26.7% vs 16.8%). No statistical difference in the median overall survival (OS) was found between the patients having complete cytoreduction and residual disease of 1 cm or less after NACT-IDS. Furthermore, no statistical difference in the median OS was found between the patients with macroscopic residual disease (e1 vs 91 cm) after NACT-IDS. Patients with residual disease of greater than 1 cm after PDS had a median OS of 15 months. Conclusions: We suggest that NACT-IDS may be a better treatment alternative for the group of highly selected women not suitable for PDS, where expected suboptimal cytoreduction does not have any appreciable survival benefit and exposes them for unnecessary risks. A substantial number of women who receive either PDS or NACT-IDS have greater than 1 cm of tumor tissue left after the operation. These women probably have no survival benefit from the operation, and future studies should focus on how to select these women preoperatively. Key Words: Debulking surgery, Epithelial ovarian cancer, Surgical complexity, Postoperative morbidity, Survival Received May 11, 2014, and in revised form July 15, 2014. Accepted for publication July 15, 2014. (Int J Gynecol Cancer 2014;24: 1420Y1428)
*Department of Gynaecology & Obstetrics, Odense University Hospital; and †Research Unit of General Practice, University of Southern Denmark, Odense, Denmark. Copyright * 2014 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000241
1420
Address correspondence and reprint requests to Algirdas Markauskas, MD, Department of Gynaecology & Obstetrics, Odense University Hospital, Sdr Boulevard 29, 5000 Odense C, Denmark. E-mail: [email protected]. The authors declare no conflicts of interest.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
theoretical benefits of primary cancer surgery are T hemultiple, including removal of resistant clones of tumor
cells, increasing the growth fraction of the tumor, maximizing the effect of chemotherapy, reducing the risk for drug resistance, improving the tumor perfusion, and enhancing the immunological competence of the patient.1,2 In epithelial ovarian cancer (EOC), an inverse relationship between the size of the residual disease after primary surgery, and overall survival (OS) was first established by Griffiths.3 The definition of ‘‘optimal cytoreduction’’ has changed over time, and the term is currently reserved for those with no macroscopic residual disease.4Y6 This is supported by the randomized study of the European Organisation for Treatment and Research of Cancer and the National Cancer Institute of Canada (EORTC-NCIC), showing that optimal cytoreduction should be defined as total tumor resection without macroscopic residuals.7 This often demands extensive upper abdominal surgery, which carries a high risk for postoperative complications. Neoadjuvant chemotherapy (NACT) before cytoreductive surgery of women with advanced EOC (International Federation of Gynecology and Obstetrics [FIGO] stages IIIc and IV) reduces the surgical complexity and shows a tendency to decrease the postoperative morbidity and mortality.7 In addition, NACT may increase the rate of complete cytoreduction. Two recently published randomized trials compared primary debulking surgery (PDS) with NACT and interval debulking surgery (NACT-IDS) in women with advanced EOC.7,8 It was demonstrated that the median OS did not differ in the 2 arms, and NACT-IDS was associated with lesser postoperative morbidity and mortality. The studies prompted a paradigm shift in the treatment for women with advanced EOC, going from primary surgery to NACT-IDS. However, it is a biological concern that NACT in the presence of a large tumor burden may increase the likelihood of mutations leading to chemotherapyresistant clones and potentially shorter progression-free survival (PFS) and OS.9,10 Our center adopted the principles of advanced cytoreductive surgery for the treatment of EOC in 2007.6 The center is population based and serves from 20% (2007Y2010) to 35% (2011 and ongoing) of the Danish population. Some of the data on the present population has been included in a nationwide Danish study of Fago¨-Olsen et al.11 The main purpose of the present study is to analyze the surgical complexity, the postoperative morbidity, and the survival of women with FIGO stages IIIc to IV EOC, treated with PDS and NACT-IDS.
MATERIALS AND METHODS All patients with FIGO stage IIIc to IVepithelial ovarian, fallopian tube, and primary peritoneal cancers treated at our institution between January 1, 2007, and December 31, 2012, were identified. Only patients, who underwent debulking surgery, were included. Patients who underwent laparotomy before NACT were included in the NACT-IDS group, because we considered laparotomy as being diagnostic as no effort to achieve maximal cytoreduction was made. Electronic medical files of the patients were used for data collection. All patients were scheduled to receive 6 cycles of platinum-based chemotherapy with standard regimen being
Primary Surgery or Interval Debulking
carboplatin + paclitaxel. No patients received intraperitoneal (IP) chemotherapy as this treatment is not yet approved for gynecological malignancies in Denmark. The decision to perform either PDS or NACT-IDS was made at a multidisciplinary team conference and based on the probability of complete IP cytoreduction. All patients were preoperatively evaluated by magnetic resonance imaging (MRI) and positron emission tomography and computed tomography (PET-CT) scans. In some cases, open laparoscopy was performed to exclude substantial serosal disease of the small bowel. Patients where complete IP cytoreduction seemed not to be achievable because of disseminated disease (tumor 9 2 cm behind porta hepatis, extensive intestinal serosal disease, bulky lymph nodes above the level of the left renal vein) or poor general condition (combination of age 9 75Y80 years, albumin G 3.0 g/dL, American Society of Anesthesiologists Group [ASA] 9 2) were scheduled to NACT.12 Women with FIGO stage IV disease were treated equally to stage IIIc, and PDS was performed if complete IP cytoreduction was estimated to be possible, unless the patient had multiple metastases in the liver and/or the lungs or substantial malignant pleural effusions. All patients were routinely evaluated clinically and by MRI scans by the same team of experts after 2 cycles of NACT and then offered IDS after the third cycle, unless they had evidence of progressive disease or the general condition of the patient still was too poor for surgery. All operations were performed by a gynecological oncologist. Bowel, hepatopancreatic surgery, and lymphadenectomy above the renal vein were performed in collaboration with experienced abdominal surgeons. Pelvic and paraaortic lymphadenectomy were performed if diagnosed as pathological on imaging or if pathological bulky nodes were identified during surgical exploration. All patients received the same postoperative care, including 24 to 48 hours observation at the intensive care unit, if major peritoneal stripping was performed. Surgical complexity was graded using a surgical complexity scoring system based upon the complexity and the number of surgical procedures performed,13 for example, hysterectomy + salpingo-oophorectomy giving, 1 point; omentectomy, 1 point; paraaortic lymphadenectomy, 1 point; pelvic lymphadenectomy, 1 point; small bowel resection, 1 point; pelvic peritoneum stripping, 1 point; abdominal peritoneum stripping, 1 point; large bowel resection, 2 points; splenectomy, 2 points; diaphragm stripping/resection, 2 points; splenectomy, 2 points; liver resection, 2 points; rectosigmoidectomy with anastomosis, 3 points. This gives 3 surgical complexity score (SCS) groups: 1 (low), e3 points; 2 (intermediate), 4 to 7 points; and 3 (high), Q8 points. Postoperative complications were evaluated using the Memorial Sloan-Kettering Cancer Center surgical secondary events grading system14: 0, no events observed within 30 days after operation 1, use of oral medications or bedside interventions to treat an event 2, use of intravenous medications, total parenteral nutrition, enteral nutrition, or blood transfusion to treat an event 3, interventional radiology, therapeutic endoscopy, or operation required to treat an event
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1421
International Journal of Gynecological Cancer
Markauskas et al
& Volume 24, Number 8, October 2014
TABLE 1. Baseline characteristics of the patients
Age, median (range), years BMI, median (range) Cancer type, n (%) Epithelial ovarian cancer Primary peritoneal cancer Fallopian tube cancer Histology, n (%) Serous Mucinous Clear cell Endometrioid Carcinosarcoma Undifferentiated FIGO stage, n (%) IIIc IV ASA group, n (%) 1 2 3 4
PDS, N = 165
NACT-IDS, N = 167
P
65 (24Y89) 24.6 (15.2Y40.8)
66 (28Y88) 24.8 (17.6Y42.4)
0.814* 0.826* 0.012†
127 (77.0) 20 (12.1) 18 (10.9)
118 (70.7) 39 (23.4) 10 (6.0)
151 (91.5) 9 (5.5) 3 (1.8) 1 (0.6) 1 (0.6) 0 (0.0)
162 (97.0) 1 (0.6) 2 (1.2) 1 (0.6) 0 (0.0) 1 (0.6)
127 (77.0) 38 (23.0)
111 (66.5) 56 (33.5)
0.028†
0.039†
0.193† 61 (37.0) 87 (52.7) 16 (9.7) 1 (0.6)
50 (29.9) 105 (62.9) 12 (7.2) 0 (0.0)
*Median test by Pearson W2. †Fischer exact test. BMI, body mass index; FIGO, International Federation of Gynecology & Obstetrics.
4, residual or lasting disability that requires major rehabilitation or organ resection 5, an event that resulted in the death of the patient Major postoperative complications were defined as grade 3 to 5 events. Chemotherapy was considered postponed if delayed more than 6 weeks after the operation and abandoned if the patient did not receive the full 6 cycles. Patients were followed until September 15, 2013, or death.
Statistical Analysis The OS time was calculated in days from the date of the decision making regarding treatment modality (PDS vs NACT-IDS) to death or to the date of the last follow-up. The PFS time was calculated in months from the date of decision making to the approximate date of recurrence/progression of the disease. The Kaplan-Meier method was used to estimate survival. Multiple Cox regression was used to compare survival between the relevant groups, adjusted for suitable confounders. To assess the differences in the baseline and the surgical characteristics of the PDS and NACT-IDS populations, we used a W2-based median test for continuous variables, for example, age and Fischer exact test for categorical variables, for example, disease stage. We chose not
1422
to perform statistical comparison of OS and PFS between the PDS and NACT-IDS groups, as patients in our retrospective study were highly selected, that is, patients submitted to NACT-IDS had more extensive disease and/or poorer general condition. Software used was Stata 11.2 (StataCorp, College Station, TX).
RESULTS Four hundred seven women with advanced EOC were identified. Seventy-five (18.4%) were excluded, as they had no debulking surgery at any time during their treatment, either because of progressive disease during chemotherapy (n = 59, 78.7%) or poor general condition (n = 16, 21.3%). Three hundred thirty-two women were included in the study of whom165 (49.7 %) underwent PDS and 167 (50.3 %) were treated with NACT-IDS. Twenty-eight (16.8%) in the NACTIDS group underwent diagnostic laparotomy before NACT. The preoperative and surgical characteristics of the study population are summarized in Tables 1 and 2, respectively. Patients submitted to PDS had more complex surgery (P G 0.001) with higher rates of diaphragmatic striping (P = 0.02) and bowel resection (P = 0.005), longer operating time (P G 0.001), greater perioperative blood loss (P G 0.001), and longer hospitalization (P = 0.001). Complete IP cytoreduction was achieved in 117 (70.9%) patients of the PDS group * 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
Primary Surgery or Interval Debulking
TABLE 2. Surgical characteristics of the patients PDS, N = 165 SCS, median (range) SCS groups, n (%) Low SCS Intermediate SCS High SCS Diaphragmatic surgery, n (%) Bowel stoma, n (%) Other radical procedures, n (%) Liver resection Cholecystectomy Pancreas resection Gastric resection Splenectomy Urinary bladder resection Residual disease, n (%), cm 0 e1 91 Blood loss, median (range), mL Length of the operation, median (range), minutes Hospitalization, median (range), days
6 (1Y15) 16 80 69 66 78 35 5 1 5 1 21 2
(9.7) (48.5) (41.8) (40) (47.3) (21.2) (3.0) (0.6) (3.0) (0.6) (12.7) (1.2)
NACT-IDS, N = 167
P
4 (1Y12)
G0.001* G0.001†
65 62 40 46 53 23 3 4 3 1 11 1
(38.9) (37.1) (24.0) (27.5) (31.7) (13.8) (1.8) (2.4) (1.8) (0.6) (6.6) (0.6)
0.020† 0.005† 0.084†
0.030† 117 22 26 1440 210 9
(70.9) (13.3) (15.8) (100Y13000) (30Y496) (3Y54)
100 21 46 800 150 7
(59.9) (12.6) (27.5) (20Y6000) (30Y480) (2Y46)
G0.001* G0.001* 0.001*
*Median test using Pearson W2. †Fischer exact test. RD, residual disease.
and 100 (59.9%) of the NACT-IDS group. The proportion of patients with stage IV disease who had complete IP cytoreduction was 28 (24%) and 39 (39%) in the PDS and NACT-IDS group, respectively. Debulking to less than or equal to 1 cm of residual disease could be achieved in 139 (84.2%) women of the PDS group and 121 (72.5%) of the NACT-IDS group. Residual disease of greater than 1 cm was left in 26 (15.8%) women after PDS and 46 (27.5%) after NACT-IDS. Most patients experienced no or just minor events (grades 0Y2) after debulking surgery (73.7% and 83.3% in the PDS and NACT-IDS group, respectively). However, the rate of major complications (grade 3Y5 events) was greater in the PDS group compared with NACT-IDS (44 [26.7%] vs 28 [16.8%]). The major complications are given in Table 3. Twenty-four (14.5%) in the PDS group needed reoperation to treat an event: 5 having intra-abdominal abscess, 8 having wound dehiscence, 6 having bowel perforation, 4 having fistula, and 1 having urinary tract leakage. In the NACT-IDS group, 15 (9%) required reoperation: 4 having intra-abdominal abscess, 4 having wound dehiscence, 3 having bowel perforation, 1 having bowel obstruction, and 3 having wound infection. Twenty-two (6.6%) patients experienced a long lasting disability (grade 4 event) after debulking surgery, 12 (7.3%) in the PDS group and 10 (6.0%) in the NACT-IDS group, and
it was associated with surgery of high complexity in 8 (66.7%) and 6 (60%) patients, respectively. Complete cytoreduction was achieved in 10 (83%) and 10 (100%) patients, who experienced a grade 4 event after PDS and NACT-IDS, respectively. Six (3.6%) in the PDS group died during the first 30 days after the operation (grade 5 event), and only 2 (33.3%) of these women achieved complete cytoreduction. Surgery of high complexity was performed in only 1 (16.7%) case. Four patients died because of sepsis, one had pulmonary embolism, and one died because of deteriorating general condition after PDS. In the NACT-IDS group, 3 women (1.8%) died during the first 30 days after surgery and all had complete IP cytoreduction. One patient underwent surgery of low complexity (33%). The causes of death were pulmonary embolism, cerebral thrombosis, and sepsis. The median OS and PFS in relation to the residual disease and type of debulking surgery are demonstrated in Figures 1 and 2, respectively. Five (3%) patients had complete radiological and biochemical remission (Cancer Antigen 125 G 35 U/mL) after NACT and all had microscopic tumor residuals only. The median OS of those patients was 31 months, and they had no postoperative events after IDS. The median delay of chemotherapy in patients, where chemotherapy was postponed, was 7 weeks for both groups.
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1423
International Journal of Gynecological Cancer
Markauskas et al
& Volume 24, Number 8, October 2014
TABLE 3. Specification of postoperative events PDS (N = 165), n (%) Vascular events Venous thromboembolism Cardiovascular events Cerebrovascular events Major infections Sepsis Wound infection Intra-abdominal abscess Bleeding/hematoma Pleural effusions Wound dehiscence Bowel perforation (overall) Anastomotic leak Ischemia/necrosis Bowel obstruction Urinary tract leakage Pancreatic leakage Fistula Surgical events Grades 0Y2 Grades 3Y5 Reoperation Postponed chemotherapy Abandoned chemotherapy
NACT-IDS (N = 167), n (%)
13 (7.9) 7 (4.2) 6 (3.6) 0 (0.0) 34 (20.6) 15 (9.1) 2 (1.2) 17 (10.3) 4 (2.4) 13 (7.9) 10 (6.1) 6 (3.6) 2 (1.2) 4 (2.4) 2 (1.2) 2 (1.2) 2 (1.2) 5 (3.0)
7 4 1 2 21 6 8 13 11 6 4 3 1 2 2 1 0 3
(4.2) (2.4) (0.6) (1.2) (12.6) (3.6) (4.8) (7.8) (6.6) (3.6) (2.4) (1.8) (0.6) (1.2) (1.2) (0.6) (0.0) (1.8)
121 (73.3) 44 (26.7) 24 (14.5) 21 (12.7) 23 (13.9)
139 28 15 9 9
(83.3) (16.8) (9.0) (5.4) (5.4)
P* 0.174
0.055
0.110 0.103 0.109 0.335
1.000 0.622 0.246 0.500 0.126
0.127 0.022 0.009
*Fischer exact test.
The median number of chemotherapy cycles in PDS patients where chemotherapy was abandoned was difficult to estimate (mean value, 1.57), as 13 (56.5%) did not receive postoperative chemotherapy at all. Five (55.6%) NACT-IDS patients did not receive postoperative chemotherapy, and the remaining 4 (44.4%) received a median of 4 cycles in total. Patients in the NACT-IDS group received at least 3 cycles of chemotherapy before IDS. A Cox regression analysis of the covariates in relation to survival for the PDS and NACT- IDS groups are given in Table 4. When adjusted for multiple variables, residual disease and age were significantly associated with median OS after PDS. Residual disease of greater than 1 cm was associated with an almost 3-fold increased risk for death after PDS compared with complete cytoreduction (hazard ratio [HR], 2.921; 95% confidence interval [CI], 1.643Y5.195; P = 0.0003). There was a trend that residual disease of 1 cm or less was associated with worse median OS when compared with complete cytoreduction after PDS (HR, 1.78; 95% CI, 0.98Y3.22; P = 0.0564). In the NACT-IDS group, FIGO stage IV (HR, 1.63; 95% CI, 1.01Y2.65), residual disease of greater than 1 cm (HR, 1.92; 95% CI, 1.15Y3.19), and ASA score of greater
1424
than 2 (HR, 3.36; 95% CI, 1.35Y8.35) were associated with significantly worse median OS.
DISCUSSION In the present cohort study, we have analyzed surgical complexity, postoperative morbidity, and OS of women treated at our institution with debulking surgery for FIGO stage IIIc and IV epithelial ovarian, fallopian tube, and primary peritoneal cancers. Although most women in the PDS and NACT-IDS groups (73.7% and 83.3%, respectively) experienced no or just minor events after surgery, our study confirms that the risk for major complications (22%) or even death (2.7%) should not be neglected, while counseling and selecting patients for surgery. The rate of major postoperative complications was higher after PDS compared with NACT-IDS, and there was a positive relationship with surgical complexity. It is problematic to compare postsurgical morbidity between various studies because of potential reporting bias and differences in the selection of patients. Chi et al15 reported 22% of grade 3 to 5 events after upfront extensive upper abdominal surgery for advanced EOC and a mortality rate of 1.4%. The * 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
Primary Surgery or Interval Debulking
likely, they should have been referred to NACT-IDS. Selection of patients for upfront surgery is a major issue as prediction of optimal debulking surgery is very difficult. Use of open laparoscopy together with diffusion-weighted MRI and PET-CT seem to be promising in providing additional information.18 All women in our study had preoperative PET-CT and MRI scans, and the rate of upfront surgery was lower compared with other centers, but the rate of complete cytoreduction was high. However, 28 (16.8%) women underwent laparotomy before NACT and represent failures in preoperative triaging, as they were found with more extensive disease than otherwise expected on imaging. In all cases, no major effort for upfront cytoreduction was made, only sufficient tissue material was secured for the proper histological diagnosis of malignancy, and the patients were scheduled to NACT-IDS. So far, the optimal method for selecting women for PDS is lacking, and improvements are highly warranted as operation should only be performed if a clear survival benefit is expected (ie, complete cytoreduction). Another pending question is if some of the women who were radically operated after NACT-IDS should have been operated
FIGURE 1. OS by residual disease and type of debulking surgery. RD, residual disease; [n to .], no estimation of the upper limit of the confidence interval for the survival median, because of short observational time. overall rate of major complications after PDS at the Memorial Sloan-Kettering Cancer Center was reported to 9%, and the rate of grade 5 complications was 0.7%.16 The rate of major postoperative complications and mortality reported in a retrospective multicenter study of Rafii et al17 were 11.5% and 0.5%, respectively, and the rate of complications was higher after PDS compared with NACT-IDS (OR, 2.17 [1.16Y4.09]). A mortality rate of 2.5% and 0.7% for the PDS and NACTIDS arms, respectively, was reported in the EORTC-NCIC trial, and the corresponding figures were 5.6% and 0.5% in the CHORUS trial.7,8 However, compared with our results, the patients were younger in the EORTC-NCIC trial and the study of Rafii et al, and the proportion of patients with FIGO stage IV disease was lower in all 3 studies cited previously.7,8,17 In the present study, two thirds of the postoperative PDS deaths (n = 4) occurred in patients in whom optimal cytoreduction could not be achieved and despite of surgery of low complexity. Residual disease of greater than 1 cm was left in 26 women (15.8%) after attempting radical surgery at PDS. These women have not gained any survival benefit, and most
FIGURE 2. PFS by residual disease and type of debulking surgery. RD, residual disease; [n to .], no estimation of the upper limit of the confidence interval for the survival median, because of short observational time.
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1425
International Journal of Gynecological Cancer
Markauskas et al
& Volume 24, Number 8, October 2014
TABLE 4. Multivariate Cox regression analysis showing HRs and CIs for OS after PDS and IDS PDS
NACT-IDS
Variables
HR
95% CI
P
HR
95% CI
P
Age BMI ASA = 1 ASA e 2 ASA 9 2 Stage IIIc Stage IV RD = 0 cm RD e 1 cm RD 9 1 cm Observations
1.02 0.99 1 0.74 1.19 1 0.93 1 1.78 2.92 165
1.00Y1.05 0.94Y1.04 Y 0.43Y1.29 0.51Y2.78 Y 0.51Y1.72 Y 0.98Y3.22 1.64Y5.19
0.0438 0.6993 Y 0.2882 0.6908 Y 0.8294 Y 0.0564 0.0003
1.02 0.99 1 1.38 3.36 1 1.63 1 0.93 1.92 167
0.99Y1.05 0.94Y1.04 Y 0.80Y2.40 1.35Y8.35 Y 1.01Y2.65 Y 0.41Y2.10 1.15Y3.19
0.1488 0.7201 Y 0.2496 0.0092 Y 0.0478 Y 0.8626 0.0119
BMI, body mass index; RD, residual disease.
upfront? It remains speculative if the survival of these women would have improved if complete cytoreduction on PDS could have been achieved. A meta-analysis by Bristow and Chi10 showed that each cycle of chemotherapy delivered before surgery decreased median survival by 4.1 months. However, the studies of Vergote et al7 and Kehoe et al8 clearly showed no survival benefit of PDS versus NACT-IDS in their populations of patients with advanced EOC. A higher percentage of women having residual disease of greater than 1 cm after NACT-IDS can partly be explained by that this subgroup also included women who were technically debulkable to less than or equal to 1-cm residual disease but were expected surgical effort to achieve it was judged too risky in relation to the extent of the disease, the general condition of the patient, and the expected survival benefit from incomplete cytoreduction. The rate of upfront cytoreduction to 1 cm or less varies widely among institutions, but different surgical groups have suggested that more than 75% should be acceptable.19 Cytoreduction to 1 cm or less was achieved in 78% in our study. The high rates of complete IP cytoreduction in our study (Q60% in both groups) can be attributed to the strict selection of the patients for debulking surgery (especially PDS) combined with will and skills of the gyne-oncological surgical team. However, we had a lower median OS compared with a study of Chi et al,16 reporting a median OS of 50 months for all women (FIGO stages IIIc-IV) treated with PDS and 78 months for the women without residual disease. The difference in the survival between the 2 studies may partly be explained by an older population and a higher proportion of stage IV disease in our study.16A direct comparison of the results regarding OS from our retrospective study with data from the 2 randomized trials7,8 should be interpreted with caution. The Gynecologic Oncology Group study, GOG 172, showed that IP chemotherapy improves survival of the women with stage III disease having less than or equal to 1-cm
1426
residual tumor after PDS by approximately 17 months.20 This treatment is currently not recognized in Denmark, and it remains speculative whether addition of IP chemotherapy would have translated into more superior OS in the present population. Data on the use of IP chemotherapy after NACTIDS are still limited, as the results of larger prospective studies are pending.21 We chose not to perform statistical comparison of the survival between the PDS and NACT-IDS groups, because the groups were highly selected on the extent of the disease and/or the general condition of the patients. However, the difference in the median OS between the subgroups with complete IP cytoreduction (51.7 vs 33.3 months after PDS and NACT-IDS, respectively) does raise some speculations. The NACT-IDS group had lower rates of postoperative complications and higher rates of completion of chemotherapy. Similar results were found in a study by Fago¨-Olsen et al,11 which showed better OS in patients after complete IP cytoreduction after PDS compared with NACT-IDS (55.5 vs 36.7 months) and increased risk for death after 2 years of follow-up after NACT-IDS. The difference in the survival observed in the present study could be explained by more extensive disease among patients referred to NACT-IDS compared with PDS, for example, 39% versus 24% with stage IV having complete IP cytoreduction in our study. Vergote et al7 have confirmed that complete resection of all macroscopic diseases during debulking surgery is the single most important prognostic factor in advanced EOC. Therefore, it remains unclear why significantly higher rates of complete cytoreduction (19% vs 51%) and lower morbidity after NACT-IDS do not translate into more superior OS in the study of Vergote et al.7 We speculate that some patients who relapse shortly after complete cytoreduction after NACT-IDS may have unidentified macroscopic disease left. This point of view is supported by the observation from our study that the median OS of women achieving complete cytoreduction after * 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 8, October 2014
NACT-IDS did not differ statistically significant from those having residual disease of 1 cm or less (33.3 vs 28 months; HR, 0.93; P = 0.8626). It has recently been demonstrated that carcinomatous areas have a benign visual appearance more often after NACT than at PDS and chemotherapy-induced fibrosis may hide vital tumor elements in otherwise obvious fibrotic tissue.22 This may lead to insufficient surgery and overestimation of NACT-IDS women with ‘‘no macroscopic residuals’’ and questions whether macroscopic tumor clearance during IDS is rather illusionary and may not be comparable with the same definition as during PDS. Furthermore, a special attention must be paid to the women in the NACTIDS population, where lacking statistical difference in the median OS between the subgroups having macroscopic residual disease (e1 vs 91 cm) may point toward ‘‘all or nothing’’ solution. The strength of the present study is its population-based nature, avoiding some of the bias that may occur in institutionalbased investigations. No heterogeneity is present regarding the choice of treatment modalities as all decisions were made by the same team of experts at multidisciplinary team conferences. A review of the Electronic Patient Files from our institution allowed inclusion of all cases meeting the inclusion criteria as well as a precise registration of surgical characteristics and postoperative events. Our study is limited by its retrospective nature and relatively short follow-up. Patients treated with chemotherapy alone are not represented in our data, as the main inclusion criterion was debulking surgery. Within the limits of our study, we suggest that NACTIDS may be a better treatment alternative for the group of highly selected women where complete IP cytoreduction cannot be expected upfront, as suboptimal PDS does not have any appreciable survival benefit and exposes the patients for unnecessary postoperative complications. The NACT-IDS is, until further evidence is presented, in our center considered as a possibility of decreasing surgical complexity and postoperative morbidity without compromising survival for a carefully selected group of women in whom alternative surgical outcome otherwise would be suboptimal tumor resection. Complete IP cytoreduction should be aimed in all cases of debulking surgery. A substantial number of women who receive either PDS or NACT-IDS have greater than 1 cm of tumor tissue left after the operation. These women probably have no survival benefit from the operation, and future studies should focus on how to select these women preoperatively.
REFERENCES 1. Bookman MA, Young RC. Principles of chemotherapy in gynecologic cancer. In: Hoskins WJ, Perez CA, Young RC, eds. Principles and Practice of Gynaecologic Oncology. Philadelphia, PA: Lippincott Williams & Wilkins; 2000:404Y407. 2. De Vita VT. The relationship between tumor mass and resistance to chemotherapy. Implications for surgical adjuvant treatment of cancer. Cancer. 1983;51:1209Y1220. 3. Griffiths CT. Surgical resection of tumor bulk in the primary treatment of ovarian cancer. Natl Cancer Inst Monogr. 1978;42:131Y136.
Primary Surgery or Interval Debulking
4. Chi DS, Eisenhauer EL, Lang J, et al. What is the optimal goal of primary cytoreductive surgery for bulky stage IIIc epithelial ovarian carcinoma (EOC)? Gynecol Oncol. 2006;103: 559Y564. 5. Stuart GC, Kitchener H, Bacon M, et al. 2010 Gynecologic Cancer Inter Group (GCIG) consensus statement on clinical trials in ovarian cancer: report from the Fourth Ovarian Cancer Consensus Conference participants of 4th Ovarian Cancer Consensus Conference (OCCC), Gynecologic Cancer Intergroup. Int J Gynecol Cancer. 2011;21: 750Y755. 6. Zivanovic O, Aldini A, Carlson JW, et al. Advanced cytoreductive surgery: American perspective. Gynecol Oncol. 2009;114:S3YS9. 7. Vergote I, Trope CG, Amant FA, et al. Neoadjuvant chemotherapy or primary surgery in stage IIIcYIV ovarian cancer. N Engl J Med. 2010;363:943Y953. 8. Kehoe S, Hook J, Nankivell M, et al. Chemotherapy or upfront surgery for newly diagnosed advanced ovarian cancer: Results from the MRC CHORUS trial. Paper presented at: 2013 American Society of Clinical Oncology Annual Meeting; May 2013; Chicago, IL. 9. Goldie JH, Coldman AJ. A mathematic model for relating the drug sensitivity of tumors to their spontaneous mutation rate. Cancer Treat Rep. 1979;63:1727Y1733. 10. Bristow RE, Chi DS. Platinum-based neoadjuvant chemotherapy and interval surgical cytoreduction for advanced ovarian cancer: a meta-analysis. Gynecol Oncol. 2006;103: 1070Y1076. 11. Fago¨-Olsen CL, Ottesen B, Kehlet H, et al. Does neoadjuvant chemotherapy impair long-term survival for ovarian cancer patients? A nationwide Danish study. Gynecol Oncol. 2014;132: 292Y298. 12. Aletti GD, Eisenhauer EL, Santillan A, et al. Identification of patient groups at highest risk from traditional approach to ovarian cancer treatment. Gynecol Oncol. 2011; 120:23Y28. 13. Aletti GD, Santillan A, Eisenhauer EL, et al. A new frontier for quality of care in gynaecologic oncology surgery: multi institutional assessment of short-term outcomes for ovarian cancer using a risk-adjusted model. Gynecol Oncol. 2007;107:99Y106. 14. Martin RCII, Brennan MF, Jaques DP. Quality of complication reporting in the surgical literature. Ann Surg. 2002;235: 803Y813. 15. Chi DS, Zivanovic O, Levinson KL, et al. The incidence of major complications after the performance of extensive upper abdomen surgical procedures during primary cytoreduction of advanced ovarian, tubal, and peritoneal carcinomas. Gynecol Oncol. 2010;119:38Y42. 16. Chi D, Musa F, Dao F, et al. An analysis of patients with bulky advanced stage ovarian, tubal and peritoneal carcinoma treated with primary debulking surgery (PDS) during an identical time period as the randomized EORTC-NCIC trial of primary debulking surgery vs neoadjuvant chemotherapy (NACT). Gynecol Oncol. 2012;124:10Y14. 17. Rafii A, Stoeckle E, Jean-Laurent M, et al. Multi-center evaluation of post-operative morbidity and mortality after optimal cytoreductive surgery for advanced ovarian cancer. PLoS One. 2012;7:e39415. 18. Fagotti A, Ferrandina G, Fanfani F, et al. Prospective validation of a laparoscopic predictive model for optimal cytoreduction
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1427
Markauskas et al
International Journal of Gynecological Cancer
in advanced ovarian carcinoma. Am J Obstet Gynecol. 2008;199:642Y646. 19. Bristow RE, Eisenhauer EL, Santillan A, et al. Delaying the primary surgical effort for advanced ovarian cancer: a systematic review of neo-adjuvant chemotherapy and interval cytoreduction. Gynecol Oncol. 2007;104:480Y490. 20. Armstrong D, Bundy B, Wenzel L, et al. Phase iii randomized trial of intravenous cisplatin and paclitaxel versus an intensive regimen of intravenous paclitaxel, intraperitoneal cisplatin, and intraperitoneal paclitaxel in stage iii ovarian cancer: a
1428
& Volume 24, Number 8, October 2014
Gynecologic Oncology Group study. N Engl J Med. 2006;354:34Y43. 21. Mackay HJ, Provencheur D, Heywood M, et al. Phase II/III study of intraperitoneal chemotherapy after Neoadjuvant chemotherapy for ovarian cancer: ncic ctgov.21. Curr Oncol. 2011;18:84Y90. 22. Hynninen J, Lavonius M, Oksa S, et al. Is perioperative visual estimation of intra-abdominal tumor spread reliable in ovarian cancer surgery after neoadjuvant chemotherapy? Gynecol Oncol. 2013;128:229Y232.
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
