ORIGINAL STUDY
Ovarian Cancer Sites of Recurrence Pascale Amate, MD,* Cyrille Huchon, MD, PhD,Þ# Anne Lucie Dessapt, MD,* Che´razade Bensaid, MD,* Jacques Medioni, MD, PhD,þ§ Marie-Aude Le Fre`re Belda, MD, PhD,þ|| Anne-Sophie Bats, MD, PhD,*þ¶ and Fabrice R. Le´curu, MD, PhD*þ¶
Introduction: Improved knowledge of recurrence sites after contemporary surgical management of ovarian cancer is needed. Material and Methods: We retrospectively reviewed consecutive patients managed for epithelial ovarian or tubal cancer with surgery and platinum-based chemotherapy between January 1, 2005, and December 31, 2009, in a tertiary teaching hospital. The site of first recurrence was recorded. Univariate analysis was performed to identify factors associated with site-specific recurrence. Overall survival and progression-free survival were computed using the Kaplan-Meier method, and log-rank tests were performed to assess the impact on survival of the variables of interest. Results: Recurrences were noted in 3 (20%) of 15 International Federation of Gynecologists and Obstetricians stage I to IIa patients and 36 (62.1%) of 58 International Federation of Gynecologists and Obstetricians IIb to IV patients, and the median progression-free survival was 21.6 (2.5Y71) and 19.3 (1.8Y67.6) months, respectively. In the advanceddisease group, 75% of recurrences involved the peritoneum and 40% were confined to the peritoneum; peritoneal recurrences developed at both treated and untreated sites. Peritoneal recurrence was associated with greater initial peritoneal involvement (Sugarbaker score, 12.1 T 8.2 vs 7.1 T 7.4; P = 0.01) and residual postoperative tumor. Nodal recurrences were noted in 38% of all recurrences, usually in combination with peritoneal recurrence and in the abdominal territories. Isolated distant metastasis was a rare mode of recurrence (8%). Conclusions: The peritoneum is the main recurrence site in both early and advanced ovarian cancer. Initial disease spread and extent of surgery are associated with the recurrence risk. This article supports the view that more attention should be directed toward extensive treatment of the peritoneum. Key Words: Ovarian cancer, Recurrence Received April 4, 2013. Accepted for publication August 21, 2013. (Int J Gynecol Cancer 2013;23: 1590Y1596)
*Service de Chirurgie Cance´rologique Gyne´cologique et du Sein, Hoˆpital Europe´en Georges Pompidou, Assistance Publique Hoˆpitaux de Paris, Paris, France; †Service de Gyne´cologie & Obste´trique, CHI PoissySt-Germain, Poissy, France; ‡Universite´ Paris Descartes, Sorbonne Paris Cite´, Faculte´ de Me´decine, Paris, France; §Service d’Oncologie Me´dicale, Hoˆpital Europe´en Georges Pompidou, Assistance Publique Hoˆpitaux de Paris, Paris, France; ||Service d’Anatamopathologie, Hoˆpital Europe´en Georges Pompidou, Assistance Publique Hoˆpitaux Copyright * 2013 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000007
1590
de Paris, Paris, France; ¶INSERM UMR-S 747, Universite´ Paris Descartes, Sorbonne Paris Cite´, Paris, France; and #Equipe d’accueil EA 7285 ‘‘Risques, cliniques et se´curite´ en sante´ des femmes et en sante´ pe´rinatale,’’ Universite´ Versailles-Saint-Quentin en Yvelines, Versailles, France. Address correspondence and reprint requests to Fabrice R. Le´curu, MD, PhD, Service de Chirurgie Cance´rologique Gyne´cologique et du Sein, Hoˆpital Europe´en Georges Pompidou, 20 Rue Leblanc, 75015, Paris, France. E-mail: [email protected]. The authors declare no conflicts of interest.
International Journal of Gynecological Cancer
& Volume 23, Number 9, November 2013
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 23, Number 9, November 2013
surgical management of advanced ovarian cancer has T hechanged radically in recent years. The importance of
maximal cytoreductive surgery1 leaving no gross residual disease2,3 is now recognized throughout the international community of oncologists. Whether primary or interval surgery is best in patients with extensive disease remains debated.4 In patients with advanced disease, adding an angiogenesis inhibitor to platinum-based chemotherapy further improves the outcomes.5,6 In parallel, improving the management of recurrences has become a major objective, and many studies of surgical and pharmacological treatments are under way.7,8 Improved knowledge of recurrence sites after proactive surgery9 for ovarian cancer is needed to determine the best secondary management strategy. The primary objective of this study was to describe ovarian cancer recurrence sites after contemporary cytoreductive surgery and platinum-based chemotherapy. The secondary objective was to identify factors associated with recurrence at each site.
MATERIALS AND METHODS We conducted a retrospective study of patients managed at the European Georges Pompidou Hospital University in Paris, France. According to French law on retrospective studies, institutional review board approval was not required.
Patients The inclusion criteria were early-stage ovarian cancer (International Federation of Gynecologists and Obstetricians [FIGO] IYIIa) or advanced ovarian, tubal, or peritoneal cancer (FIGO IIbYIV) managed at our center with primary or interval surgery aimed at achieving complete gross tumor resection, combined with platinum-based chemotherapy. The first surgical procedure had to have been performed between January 1, 2005, and December 31, 2009. The exclusion criteria were borderline or nonepithelial tumor, conservative treatment, absence of a complete response to surgery plus chemotherapy, incomplete treatment, and treatment conducted in part at another center.
Management The routine preoperative workup included computed tomography (CT) of the thorax, the abdomen, and the pelvis. Positron emission tomography (PET)YCT was also obtained in some patients. Exploratory laparoscopy was then performed routinely to obtain histological documentation of the diagnosis and to choose between primary and interval surgery. Cytoreductive surgery was performed through a midline laparotomy with the objective of achieving complete gross resection. The Sugarbaker peritoneal cancer index was used to measure the extent of peritoneal carcinomatosis before and after resection.10 Some patients with pleural effusions underwent thoracoscopy.11 Chemotherapy was with paclitaxel (175 mg/m2) and carboplatin (AUC 5) given intravenously at least 6 times, at 21-day intervals. The evaluation performed at chemotherapy included a clinical examination; a serum CA125 assay; and CT of the chest, the abdomen, and the pelvis. Follow-up consisted of regular clinical evaluations; serum marker assays; and CT of the chest, the abdomen, and
Recurrence Site After Ovarian Cancer
the pelvis and/or PET-CT. Recurrences were diagnosed on the basis of serum marker elevation with CT abnormalities of the chest, the abdomen, and the pelvis. In addition, laparoscopy was performed when surgery was considered.12,13
Study Data The following data were recorded for each patient: initial epidemiological information, results of the preoperative workup, intraoperative findings during the first procedure, postoperative complications, histological findings, chemotherapy characteristics, and follow-up data. For each patient with recurrent cancer, we recorded the sites involved, the number of recurrent tumors, and the time to recurrence.
Statistical Analysis Ordinal variables were compared between the patients with and without recurrences using the W2 test or the Fisher exact test in the event of small sample sizes. Continuous variables were compared using the Student t test when distribution was normal and analysis of variance otherwise. Progression-free survival (PFS) and overall survival (OS) were computed on the basis of the dates of diagnostic surgery (study inclusion date), last follow-up (outpatient clinic visit or admission), recurrence, and death. Progression-free survival and OS were analyzed using the Kaplan-Meier method and compared using the log-rank test. Values of P e 0.05 were considered significant. The 95% confidence intervals
TABLE 1. Main patient characteristics Age at diagnosis, mean T SD, y BMI, mean T SD, kg/m2 History of abdominal and/or pelvic surgery, n/N (%) Initial mean CA125, mean T SD, IU/mL FIGO stage I II III IV Histology (main component), n/N (%) Serous Mucinous Endometrioid Clear-cell Transitional cell Mixed epithelial Undifferentiated Second cell component, n/N (%) Clear-cell Other
62 T 12.5 24.2 T 4 40/73 (55%) 1842 T 3712 13/73 5/73 40/73 15/73
(18%) (7%) (55%) (21%)
49/70 5/70 10/70 2/70 0/70 2/70 2/70 15/73 10/15 5/15
(70%) (7%) (14%) (3%) (0%) (3%) (3%) (21%) (67%) (33%)
BMI, Body mass index; n, number of patients with the relevant feature; N, number of patients with available data on the relevant feature.
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1591
International Journal of Gynecological Cancer
Amate et al
were computed. All statistical analyses were performed using the Stata SE 11.0 software (Stata Corporation, College Station, TX).
RESULTS Patients Of 146 patients identified during the study period, 73 were excluded, for the following reasons: borderline tumor, n = 34; conservative treatment, n = 8; nonepithelial tumor, n = 12; treatment performed in part at another center, n = 11; and
& Volume 23, Number 9, November 2013
absence of surgical cytoreduction, n = 8. The remaining 73 patients were included in the study. Table 1 reports their main characteristics. There were 15 patients with early cancer (IYIIa) and 58 with advanced cancer (IIbYIV). The mean T SD Sugarbaker index at the beginning of cytoreduction surgery was 2.4 T 2.7 in the early-stage group and 10.5 T 8.1 in the advanced-stage group. Table 2 details the procedures performed during cytoreduction surgery. Of the patients with advanced cancer, 58% had primary surgery and 42% had interval surgery. Most patients had hysterectomy, omentectomy, and dissection of the pelvic and aortocaval nodes (starting in December 2008, the patients were eligible
TABLE 2. Procedures performed during initial cytoreductive surgery Surgical Procedure Hysterectomy Bilateral adnexectomy Resection of both ovarian suspensory ligaments Omentectomy Douglassectomy Resection of the anterior vesical peritoneum Resection of the right paracolic gutter Resection of the left paracolic gutter Stripping of the right hemidiaphragm (mean, 79 cm2; median, 70 cm2) Stripping of the left hemidiaphragm (mean, 71 cm2; median, 57.5 cm2) Appendectomy Cholecystectomy Rectosigmoidectomy Left colectomy Right colectomy Transverse colectomy Small bowel resection Aortocaval node dissection No. nodes, mean T SD Pelvic node dissection No. nodes, mean T SD Splenectomy Resection of a liver nodule Partial resection of Glisson capsule Resection of the ligamentum teres hepatis Partial gastrectomy Resection of the diaphragm Partial cystectomy Destruction of a mesenteric lesion Destruction of a mesocolic lesion Destruction of a small bowel lesion Ostomy Intraoperative blood transfusion
Early-Stage Disease (IYIIa)
Advanced Disease (IIbYIV)
10/12 (83%) 10/10 (100%) 8/15 (53%) 13/13 (100%) 1/15 (6%) 0/15 0/15 0/15 0/15 0/15 7/13 (53%) 0/15 0/15 0/15 0/15 0/15 0/15 14/15 (93%) 22 T 11.7 14/15 (93%) 15 T 6 0/15 0/15 1/15 (6%) 4/15 (26%) 0/15 0/15 0/15 0/15 0/15 0/15 0/15 3/14 (21%)
54/58 (93%) 55/56 (98%) 45/58 (77%) 53/58 (93%) 38/58 (65%) 24/58 (41%) 5/58 (8%) 5/58 (8%) 26/58 (44%) 13/58 (22%) 32/54 (59%) 0/58 23/58 (39%) 2/58 (3%) 3/58 (5%) 2/58 (3%) 4/58 (7%) 50/58 (86%) 26.9 T 12.8 50/58 (86%) 11.7 T 6.8 5/58 (8%) 1/58 (1%) 5/58 (8%) 30/58 (51%) 1/58 (1%) 4/58 (7%) 1/58 (1%) 7/58 (12%) 1/58 (1%) 1/58 (1%) 15/58 (25%) 28/57 (49%)
The denominators vary because some patients had hysterectomy, adnexectomy, or appendectomy before the treatment of ovarian cancer.
1592
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 23, Number 9, November 2013
Recurrence Site After Ovarian Cancer
TABLE 3. Sites of the first recurrence Early Stage Peritoneum Nodes Distal metastases (other than supradiaphragmatic nodes) Peritoneum only Nodes only Distant metastases only Peritoneum + nodes + distant metastases Peritoneum + distant metastases + nodes Nodes + distant metastases + peritoneum Peritoneum + nodes + distant metastases for inclusion in the CARACO trial involving randomization to surgery with vs without dissection of the pelvic and aortocaval nodes).14 The mean numbers of pelvic and aortocaval nodes are given in Table 2. More than 40% of the patients required resection of at least 1 bowel segment; 44%, stripping of the right hemidiaphragm; and 22%, stripping of the left hemidiaphragm. Complete gross resection was achieved in 46 (79.3%) of the 58 patients with advanced disease; in the remaining 12 patients, the residual tumor was less than 1 cm. In the group with advanced disease, tumor spread according to the classification system of Hamilton et al15 was confined to the pelvis and the retroperitoneum in 9 (15.5%) of 58 patients, spared the upper abdomen in 19 (29.3%) of 58 patients, and involved the upper abdomen in 32 (55.2%) of 58 patients.
Outcomes At the time of our retrospective study, the median follow-up was 29 months (range, 2Y71 months). Recurrences were diagnosed in 3 of the 15 patients who had early cancer, with a median PFS of 21.6 months (range, 2.5Y71 months), and in 36 of the 58 patients who had advanced cancer, with a median PFS of 19.3 months (range, 1.8Y67.6 months) (P = 0.03). Of the 15 patients with early-stage disease, none died during follow-up, whose median duration in this group was 27.8 months (range, 2.5Y71 months). Of the 58 patients with advanced disease, 15 died during follow-up, whose median duration was 30.0 months (range, 1.8Y69.3 months). Of the 39 patients with recurrences, 29 (74%) had had complete gross resection and accounted for 47.5% (29/61) of all patients with complete gross resection and 10 (26%) had had gross residual tumor less than 1 cm and accounted for
3/3 0/3 0/3 3/3 0/3 0/3 0/3 0/3 0/3 0/3
Advanced Stage 27/36 14/36 17/36 11/36 5/36 3/36 3/36 8/36 1/36 5/36
(75%) (38%) (47%) (30%) (13%) (8%) (8%) (22%) (2%) (13%)
P (Fisher Test) 1.0 0.54 0.24 0.04 1.0 1.0 1.0 1.0 1.0 1.0
83.3% (10/12) of all patients in this resection group. The difference was statistically significant (P = 0.02). Initial involvement of the aortocaval nodes was significantly associated with a shorter PFS (log-rank; P = 0.03). In contrast, initial pelvic node involvement had no significant impact on PFS (log-rank; P = 0.51). Overall survival was influenced neither by initial aortocaval node involvement (log-rank; P = 0.12) nor by initial pelvic node involvement (log-rank; P = 0.63).
Recurrence Sites Table 3 reports the recurrence sites. The recurrences were confined to the peritoneum in all 3 patients with earlystage disease. In the 36 patients with advanced disease, the peritoneum was the most common site of recurrence (75%). In 40% of the patients with peritoneal recurrences, the recurrences were confined to the peritoneum. Nodal involvement was noted in 38% of all patients with recurrences; among the patients with nodal involvement, 87% had involved nodes below the diaphragm; 40%, both below and above the diaphragm; and 13%, only above the diaphragm. Finally, 47% of the patients had distant metastases at the first recurrence, although only 8% of the patients with recurrences had isolated distant metastases without concomitant peritoneal or nodal disease. According to the classification system developed by Hamilton et al,15 the first recurrence usually involved the upper abdomen. The peritoneal recurrences were confined to the pelvis in 5 (14.2%) of 35 patients, involved the pelvis and the lower abdomen in 5 (14.2%) of 35 patients, and extended to the upper abdomen in 25 (71.4%) of 35 patients. Finally, most of the peritoneal recurrences arose at the treated sites and extended to the untreated sites (Table 4). The
TABLE 4. Peritoneal recurrence sites according to initial sites of peritoneal involvement Site of Peritoneal Recurrence Treated sites only Untreated sites only Treated and untreated sites
Early Stage
Advanced Stage
1/3 1/3 1/3
8/36 7/36 21/36
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1593
Amate et al
International Journal of Gynecological Cancer
recurrence site had no impact on time to recurrence (log-rank; P = 0.15).
Risk Factors Table 5 reports the factors associated with recurrence at each site in the group with advanced disease. The risk for peritoneal recurrence was associated with completeness of initial resection and with peritoneal stripping (left paracolic gutter, right or left hemidiaphragm). Nodal recurrences were significantly more common in the patients with than without initial nodal involvement (10/33 [30.3%] vs 4/40 [10%]; P = 0.04). The mean number of removed aortocaval nodes was significantly greater in the patients with than without nodal recurrences (32.4 T 13.1 vs 24.0 T 12.3, P = 0.04), whereas no differences were found between these 2 patient subgroups for mean number of involved aortic nodes, mean number of removed pelvic nodes, or mean number of involved pelvic nodes. Finally, nodal involvement was significantly more common after resection of the anterior vesical peritoneum. Organ involvement was associated with thoracoscopy during the initial workup and stripping of the left paracolic gutter or the right hemidiaphragm. Histologically, the risk for peritoneal recurrence was associated with initial involvement of 1 of the hemidiaphragms, the omentum, and the ligamentum teres hepatis.
& Volume 23, Number 9, November 2013
Nodal recurrences were associated with initial involvement of the omentum. Finally, distant metastases were associated with initial involvement of the tube, the paracolic gutter, the right hemidiaphragm, and the omentum. The initial mean Sugarbaker index was significantly higher in the patients with recurrences (at any site) than in the other patients (11 T 8.5 vs 6.4 T 6.8, P = 0.01). However, when we compared the patients with and without nodal recurrences, we found no significant difference in the initial mean Sugarbaker index (11.2 T 8.2 vs 8.4 T 8.0, P = 0.27). In contrast, significant differences were found for the patients with versus without peritoneal recurrences (12.1 T 8.2 vs 7.1 T 7.4, P = 0.01) and for the patients with versus without distant metastases (13.5 T 8.3 vs 7.7 T 7.6, P = 0.01).
DISCUSSION Most patients treated for advanced ovarian cancer experience a recurrence within the next 24 months.5,6 The recent change in the surgical paradigm with a shift toward complete primary cytoreduction2 and the use of surgery to treat recurrences8,12 requires an analysis of recurrence sites after modern initial surgical management. The peritoneum was the main site of recurrence in our patients (75%). Among the peritoneal recurrences, 40% were
TABLE 5. Risk factors by recurrence site in the patients with advanced cancer Any Site Peritoneum Nodes Distant Metastases Residual tumor 0/e1 cm Age G / 9 60 y History of abdominal/pelvic surgery Hypertension Diabetes Previous cancer Family history of cancer High PET uptake of initial tumor Carcinomatosis with high PET uptake Omental carcinomatosis by PET Carcinomatosis of diaphragmatic peritoneum by CT only Carcinomatosis of diaphragmatic peritoneum with high PET uptake Lymphadenopathy by CT or MRI Lymphadenopathy with high PET uptake Bowel involvement by CT Bowel involvement by PET Hypermetabolic metastases Thoracoscopy during initial workup After neoadjuvant chemotherapy Ascites Radiological carcinomatosis Omental carcinomatosis by PET Carcinomatosis of diaphragmatic peritoneum by CT only
1594
0.11 0.19 0.74 0.25 0.63 0.41 0.72 0.41 0.01 0.02 0.29 0.09 0.05 0.12 0.69 0.50 1.0 1.0
0.004 0.68 0.99 0.13 0.61 0.11 0.65 1.0 0.10 0.05 0.07 0.05 0.01 0.13 0.13 0.21 0.48 1.0
1.0 0.09 0.45 0.52 1.0 1.0 0.08 1.0 0.31 0.21 0.29 1.0 0.71 0.35 0.17 1.0 0.19 0.61
0.15 0.36 0.90 0.34 0.31 0.66 0.55 1.0 0.16 0.21 0.19 0.23 0.50 0.40 0.23 0.47 0.14 0.04
0.43 0.38 0.54 0.38
1.0 0.68 1.0 0.68
0.67 1.0 1.0 1.0
0.69 0.68 1.0 0.68 * 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 23, Number 9, November 2013
isolated and 58.3% involved both treated and untreated areas (Table 4). The risk factors of peritoneal recurrences reflected the initial disease spread: residual tumor after initial surgery and initial major peritoneal involvement requiring resection procedures. The predominance of peritoneal recurrences and their frequent location at both treated and untreated sites indicate the persistence of tumor deposits even when gross resection is considered complete. These data support the use of extensive peritoneal surgery or methods designed to treat the entire peritoneal cavity such as intraperitoneal chemotherapy with or without hyperthermia.16 Nodal recurrences were noted in 38% of the patients with advanced disease followed by recurrences, usually concomitantly with the involvement of another site and usually in the abdominal compartment. The risk factors of nodal recurrences were nodal involvement identified during the preoperative workup and diffuse peritoneal involvement. Nodal recurrences were also more common among the patients with initial nodal involvement (Table 5). Aortocaval initial involvement was associated with a shorter PFS but without impact on OS. This
Recurrence Site After Ovarian Cancer
emphasizes the debate about the real benefit of lymphadenectomy in these patients.14 Distant metastases were often diagnosed concomitantly with peritoneal and/or nodal recurrences. These were isolated in only 8% of the patients with advanced disease and recurrences. The risk factors were thoracoscopy (which usually indicated stage IV disease), diffuse carcinomatosis (initial mean T SD Sugarbaker index, 13.5 T 8.3 vs 7.7 T 7.6, P = 0.01), and number of neoadjuvant chemotherapy cycles. Thus, despite the greater aggressiveness of the initial surgical procedure, which often included peritoneal resections, the peritoneum remained the main site of recurrence. Peritoneal recurrences involved both treated and untreated sites and often developed in the upper abdomen, indicating that complete gross resection leaves in fact microscopic tumor residuals. This supports the concept of treating the entire peritoneal cavity with intraperitoneal chemotherapy with or without hyperthermia or photodynamic therapy.16 The frequent involvement of the peritoneum also has implications for patient selection to surgery because surgical treatment of recurrences is beneficial
TABLE 5. (Continued)
Carcinomatosis of diaphragmatic peritoneum by PET Lymphadenopathy by CT Lymphadenopathy by PET Bowel involvement by CT Bowel involvement by PET Distant metastases At initial cytoreductive surgery Ascites Resection of Douglas pouch Resection of prevesical peritoneum Resection of right paracolic gutter Resection of left paracolic gutter Resection/stripping of right hemidiaphragm Resection/stripping of left hemidiaphragm Rectosigmoidectomy Left colectomy Right colectomy Transverse colectomy Resection of small bowel Lumboaortic node dissection Pelvic node dissection Splenectomy Fulguration of carcinomatous lesions On the mesentery On the mesocolon On the hemidiaphragms On the small bowel
Any Site
Peritoneum
Nodes
Distant Metastases
0.55 0.29 1.0 1.0 1.0 1.0
0.60 0.60 1.0 0.39 1.0 0.69
1.0 0.04 0.05 0.63 1.0 0.05
1.0 0.63 1.0 0.36 1.0 1.0
0.88 0.17 0.09 0.64 0.15 0.008 0.10 0.13 0.52 0.28 0.52 1.0 1.0 1.0 0.64
0.39 0.07 0.13 0.66 0.02 0.002 0.01 0.22 1.0 0.10 0.21 0.33 1.0 1.0 0.17
0.43 0.34 0.05 1.0 0.59 0.43 0.49 0.36 0.43 1.0 1.0 1.0 1.0 1.0 1.0
0.73 0.26 0.57 0.14 0.02 0.01 0.49 0.66 1.0 0.55 1.0 1.0 0.41 0.41 1.0
1.0 0.38 0.29 0.38
0.69 1.0 0.33 1.0
1.0 1.0 0.24 1.0
0.66 1.0 0.57 1.0
MRI, Magnetic resonance imaging.
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1595
International Journal of Gynecological Cancer
Amate et al
only when complete resection is achieved.13,17,18 Thus, the location of the peritoneal recurrences may limit the potential indications for surgical treatment. To the best of our knowledge, this is the first study whose primary objective was to describe and to analyze recurrence sites after initial complete gross resection of ovarian cancer. The main limitation of our study is the retrospective design, which can result in biases because of patient selection and missing data. Another limitation is that follow-up was shorter than 60 months in some patients. The patients without recurrences, who constituted the control group for our riskfactor analysis, may have developed recurrences later on. However, the sample size and the single-center design limit the potential impact of these sources of bias.
CONCLUSIONS Our data show that the peritoneum is the main recurrence site after contemporary cytoreductive surgery for advanced epithelial cancer involving the ovary, the tube, or the peritoneum. Initial tumor spread and residual tumor after surgery are risk factors of peritoneal recurrence. These data support the use of extensive treatment of the peritoneum.
REFERENCES 1. Bristow RE, Tomacruz RS, Armstrong DK, et al. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol. 2002;20:1248Y1259. 2. Chi DS, Eisenhauer EL, Zivanovic O, et al. Improved progression-free and overall survival in advanced ovarian cancer as a result of a change in surgical paradigm. Gynecol Oncol. 2009;114:26Y31. 3. du Bois A, Reuss A, Pujade-Lauraine E, et al. Role of surgical outcome as prognostic factor in advanced epithelial ovarian cancer: a combined exploratory analysis of 3 prospectively randomized phase 3 multicenter trials: by the Arbeitsgemeinschaft Gynaekologische Onkologie Studiengruppe Ovarialkarzinom (AGO-OVAR) and the Groupe d’Investigateurs Nationaux Pour les Etudes des Cancers de l’Ovaire (GINECO). Cancer. 2009;115:1234Y1244. 4. Vergote I, Trope CG, Amant F, et al. Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer. N Engl J Med. 2010;363:943Y953.
1596
& Volume 23, Number 9, November 2013
5. Burger RA, Brady MF, Bookman MA, et al. Incorporation of bevacizumab in the primary treatment of ovarian cancer. N Engl J Med. 2011;365:2473Y2483. 6. Perren TJ, Swart AM, Pfisterer J, et al. A phase 3 trial of bevacizumab in ovarian cancer. N Engl J Med. 2011;365: 2484Y2496. 7. CHIPOR. Available at: http://clinicaltrials.gov/show/NCT01376752. 8. DESKTOP III. Available at: http://www.clinicaltrials.gov/show/ NCT01166737. 9. Chang SJ, Bristow RE, Ryu HS. Impact of complete cytoreduction leaving no gross residual disease associated with radical cytoreductive surgical procedures on survival in advanced ovarian cancer. Ann Surg Oncol. 2012;19:4059Y4067. 10. Portilla AG, Shigeki K, Dario B, et al. The intraoperative staging systems in the management of peritoneal surface malignancy. J Surg Oncol. 2008;98:228Y231. 11. Klar M, Farthmann J, Bossart M, et al. Video-assisted thoracic surgery (VATS) evaluation of intrathoracic disease in patients with FIGO III and IV stage ovarian cancer. Gynecol Oncol. 2012;126:397Y402. 12. Harter P, Sehouli J, Reuss A, et al. Prospective validation study of a predictive score for operability of recurrent ovarian cancer: the Multicenter Intergroup Study DESKTOP II. A project of the AGO Kommission OVAR, AGO Study Group, NOGGO, AGO-Austria, and MITO. Int J Gynecol Cancer. 2011;21: 289Y295. 13. Harter P, Heitz F, du Bois A. Surgery for relapsed ovarian cancer: when should it be offered?Curr Oncol Rep. 2012;14:539Y543. 14. CARACO. Available at: http://www.e-cancer.fr/recherche/ recherche-clinique/registre-des-essais-cliniques/registre-des-essaiscliniques. 15. Hamilton CA, Miller A, Miller C, et al. The impact of disease distribution on survival in patients with stage III epithelial ovarian cancer cytoreduced to microscopic residual: a Gynecologic Oncology Group study. Gynecol Oncol. 2011;122: 521Y526. 16. OVHIPEC. Available at: http://clinicaltrials.gov/ct2/show/ NCT00426257. 17. Bristow RE, Puri I, Chi DS. Cytoreductive surgery for recurrent ovarian cancer: a meta-analysis. Gynecol Oncol. 2009;112: 265Y274. 18. Chi DS, McCaughty K, Diaz JP, et al. Guidelines and selection criteria for secondary cytoreductive surgery in patients with recurrent, platinum-sensitive epithelial ovarian carcinoma. Cancer. 2006;106:1933Y1939.
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
Ovarian Cancer Sites of Recurrence Pascale Amate, MD,* Cyrille Huchon, MD, PhD,Þ# Anne Lucie Dessapt, MD,* Che´razade Bensaid, MD,* Jacques Medioni, MD, PhD,þ§ Marie-Aude Le Fre`re Belda, MD, PhD,þ|| Anne-Sophie Bats, MD, PhD,*þ¶ and Fabrice R. Le´curu, MD, PhD*þ¶
Introduction: Improved knowledge of recurrence sites after contemporary surgical management of ovarian cancer is needed. Material and Methods: We retrospectively reviewed consecutive patients managed for epithelial ovarian or tubal cancer with surgery and platinum-based chemotherapy between January 1, 2005, and December 31, 2009, in a tertiary teaching hospital. The site of first recurrence was recorded. Univariate analysis was performed to identify factors associated with site-specific recurrence. Overall survival and progression-free survival were computed using the Kaplan-Meier method, and log-rank tests were performed to assess the impact on survival of the variables of interest. Results: Recurrences were noted in 3 (20%) of 15 International Federation of Gynecologists and Obstetricians stage I to IIa patients and 36 (62.1%) of 58 International Federation of Gynecologists and Obstetricians IIb to IV patients, and the median progression-free survival was 21.6 (2.5Y71) and 19.3 (1.8Y67.6) months, respectively. In the advanceddisease group, 75% of recurrences involved the peritoneum and 40% were confined to the peritoneum; peritoneal recurrences developed at both treated and untreated sites. Peritoneal recurrence was associated with greater initial peritoneal involvement (Sugarbaker score, 12.1 T 8.2 vs 7.1 T 7.4; P = 0.01) and residual postoperative tumor. Nodal recurrences were noted in 38% of all recurrences, usually in combination with peritoneal recurrence and in the abdominal territories. Isolated distant metastasis was a rare mode of recurrence (8%). Conclusions: The peritoneum is the main recurrence site in both early and advanced ovarian cancer. Initial disease spread and extent of surgery are associated with the recurrence risk. This article supports the view that more attention should be directed toward extensive treatment of the peritoneum. Key Words: Ovarian cancer, Recurrence Received April 4, 2013. Accepted for publication August 21, 2013. (Int J Gynecol Cancer 2013;23: 1590Y1596)
*Service de Chirurgie Cance´rologique Gyne´cologique et du Sein, Hoˆpital Europe´en Georges Pompidou, Assistance Publique Hoˆpitaux de Paris, Paris, France; †Service de Gyne´cologie & Obste´trique, CHI PoissySt-Germain, Poissy, France; ‡Universite´ Paris Descartes, Sorbonne Paris Cite´, Faculte´ de Me´decine, Paris, France; §Service d’Oncologie Me´dicale, Hoˆpital Europe´en Georges Pompidou, Assistance Publique Hoˆpitaux de Paris, Paris, France; ||Service d’Anatamopathologie, Hoˆpital Europe´en Georges Pompidou, Assistance Publique Hoˆpitaux Copyright * 2013 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000007
1590
de Paris, Paris, France; ¶INSERM UMR-S 747, Universite´ Paris Descartes, Sorbonne Paris Cite´, Paris, France; and #Equipe d’accueil EA 7285 ‘‘Risques, cliniques et se´curite´ en sante´ des femmes et en sante´ pe´rinatale,’’ Universite´ Versailles-Saint-Quentin en Yvelines, Versailles, France. Address correspondence and reprint requests to Fabrice R. Le´curu, MD, PhD, Service de Chirurgie Cance´rologique Gyne´cologique et du Sein, Hoˆpital Europe´en Georges Pompidou, 20 Rue Leblanc, 75015, Paris, France. E-mail: [email protected]. The authors declare no conflicts of interest.
International Journal of Gynecological Cancer
& Volume 23, Number 9, November 2013
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 23, Number 9, November 2013
surgical management of advanced ovarian cancer has T hechanged radically in recent years. The importance of
maximal cytoreductive surgery1 leaving no gross residual disease2,3 is now recognized throughout the international community of oncologists. Whether primary or interval surgery is best in patients with extensive disease remains debated.4 In patients with advanced disease, adding an angiogenesis inhibitor to platinum-based chemotherapy further improves the outcomes.5,6 In parallel, improving the management of recurrences has become a major objective, and many studies of surgical and pharmacological treatments are under way.7,8 Improved knowledge of recurrence sites after proactive surgery9 for ovarian cancer is needed to determine the best secondary management strategy. The primary objective of this study was to describe ovarian cancer recurrence sites after contemporary cytoreductive surgery and platinum-based chemotherapy. The secondary objective was to identify factors associated with recurrence at each site.
MATERIALS AND METHODS We conducted a retrospective study of patients managed at the European Georges Pompidou Hospital University in Paris, France. According to French law on retrospective studies, institutional review board approval was not required.
Patients The inclusion criteria were early-stage ovarian cancer (International Federation of Gynecologists and Obstetricians [FIGO] IYIIa) or advanced ovarian, tubal, or peritoneal cancer (FIGO IIbYIV) managed at our center with primary or interval surgery aimed at achieving complete gross tumor resection, combined with platinum-based chemotherapy. The first surgical procedure had to have been performed between January 1, 2005, and December 31, 2009. The exclusion criteria were borderline or nonepithelial tumor, conservative treatment, absence of a complete response to surgery plus chemotherapy, incomplete treatment, and treatment conducted in part at another center.
Management The routine preoperative workup included computed tomography (CT) of the thorax, the abdomen, and the pelvis. Positron emission tomography (PET)YCT was also obtained in some patients. Exploratory laparoscopy was then performed routinely to obtain histological documentation of the diagnosis and to choose between primary and interval surgery. Cytoreductive surgery was performed through a midline laparotomy with the objective of achieving complete gross resection. The Sugarbaker peritoneal cancer index was used to measure the extent of peritoneal carcinomatosis before and after resection.10 Some patients with pleural effusions underwent thoracoscopy.11 Chemotherapy was with paclitaxel (175 mg/m2) and carboplatin (AUC 5) given intravenously at least 6 times, at 21-day intervals. The evaluation performed at chemotherapy included a clinical examination; a serum CA125 assay; and CT of the chest, the abdomen, and the pelvis. Follow-up consisted of regular clinical evaluations; serum marker assays; and CT of the chest, the abdomen, and
Recurrence Site After Ovarian Cancer
the pelvis and/or PET-CT. Recurrences were diagnosed on the basis of serum marker elevation with CT abnormalities of the chest, the abdomen, and the pelvis. In addition, laparoscopy was performed when surgery was considered.12,13
Study Data The following data were recorded for each patient: initial epidemiological information, results of the preoperative workup, intraoperative findings during the first procedure, postoperative complications, histological findings, chemotherapy characteristics, and follow-up data. For each patient with recurrent cancer, we recorded the sites involved, the number of recurrent tumors, and the time to recurrence.
Statistical Analysis Ordinal variables were compared between the patients with and without recurrences using the W2 test or the Fisher exact test in the event of small sample sizes. Continuous variables were compared using the Student t test when distribution was normal and analysis of variance otherwise. Progression-free survival (PFS) and overall survival (OS) were computed on the basis of the dates of diagnostic surgery (study inclusion date), last follow-up (outpatient clinic visit or admission), recurrence, and death. Progression-free survival and OS were analyzed using the Kaplan-Meier method and compared using the log-rank test. Values of P e 0.05 were considered significant. The 95% confidence intervals
TABLE 1. Main patient characteristics Age at diagnosis, mean T SD, y BMI, mean T SD, kg/m2 History of abdominal and/or pelvic surgery, n/N (%) Initial mean CA125, mean T SD, IU/mL FIGO stage I II III IV Histology (main component), n/N (%) Serous Mucinous Endometrioid Clear-cell Transitional cell Mixed epithelial Undifferentiated Second cell component, n/N (%) Clear-cell Other
62 T 12.5 24.2 T 4 40/73 (55%) 1842 T 3712 13/73 5/73 40/73 15/73
(18%) (7%) (55%) (21%)
49/70 5/70 10/70 2/70 0/70 2/70 2/70 15/73 10/15 5/15
(70%) (7%) (14%) (3%) (0%) (3%) (3%) (21%) (67%) (33%)
BMI, Body mass index; n, number of patients with the relevant feature; N, number of patients with available data on the relevant feature.
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1591
International Journal of Gynecological Cancer
Amate et al
were computed. All statistical analyses were performed using the Stata SE 11.0 software (Stata Corporation, College Station, TX).
RESULTS Patients Of 146 patients identified during the study period, 73 were excluded, for the following reasons: borderline tumor, n = 34; conservative treatment, n = 8; nonepithelial tumor, n = 12; treatment performed in part at another center, n = 11; and
& Volume 23, Number 9, November 2013
absence of surgical cytoreduction, n = 8. The remaining 73 patients were included in the study. Table 1 reports their main characteristics. There were 15 patients with early cancer (IYIIa) and 58 with advanced cancer (IIbYIV). The mean T SD Sugarbaker index at the beginning of cytoreduction surgery was 2.4 T 2.7 in the early-stage group and 10.5 T 8.1 in the advanced-stage group. Table 2 details the procedures performed during cytoreduction surgery. Of the patients with advanced cancer, 58% had primary surgery and 42% had interval surgery. Most patients had hysterectomy, omentectomy, and dissection of the pelvic and aortocaval nodes (starting in December 2008, the patients were eligible
TABLE 2. Procedures performed during initial cytoreductive surgery Surgical Procedure Hysterectomy Bilateral adnexectomy Resection of both ovarian suspensory ligaments Omentectomy Douglassectomy Resection of the anterior vesical peritoneum Resection of the right paracolic gutter Resection of the left paracolic gutter Stripping of the right hemidiaphragm (mean, 79 cm2; median, 70 cm2) Stripping of the left hemidiaphragm (mean, 71 cm2; median, 57.5 cm2) Appendectomy Cholecystectomy Rectosigmoidectomy Left colectomy Right colectomy Transverse colectomy Small bowel resection Aortocaval node dissection No. nodes, mean T SD Pelvic node dissection No. nodes, mean T SD Splenectomy Resection of a liver nodule Partial resection of Glisson capsule Resection of the ligamentum teres hepatis Partial gastrectomy Resection of the diaphragm Partial cystectomy Destruction of a mesenteric lesion Destruction of a mesocolic lesion Destruction of a small bowel lesion Ostomy Intraoperative blood transfusion
Early-Stage Disease (IYIIa)
Advanced Disease (IIbYIV)
10/12 (83%) 10/10 (100%) 8/15 (53%) 13/13 (100%) 1/15 (6%) 0/15 0/15 0/15 0/15 0/15 7/13 (53%) 0/15 0/15 0/15 0/15 0/15 0/15 14/15 (93%) 22 T 11.7 14/15 (93%) 15 T 6 0/15 0/15 1/15 (6%) 4/15 (26%) 0/15 0/15 0/15 0/15 0/15 0/15 0/15 3/14 (21%)
54/58 (93%) 55/56 (98%) 45/58 (77%) 53/58 (93%) 38/58 (65%) 24/58 (41%) 5/58 (8%) 5/58 (8%) 26/58 (44%) 13/58 (22%) 32/54 (59%) 0/58 23/58 (39%) 2/58 (3%) 3/58 (5%) 2/58 (3%) 4/58 (7%) 50/58 (86%) 26.9 T 12.8 50/58 (86%) 11.7 T 6.8 5/58 (8%) 1/58 (1%) 5/58 (8%) 30/58 (51%) 1/58 (1%) 4/58 (7%) 1/58 (1%) 7/58 (12%) 1/58 (1%) 1/58 (1%) 15/58 (25%) 28/57 (49%)
The denominators vary because some patients had hysterectomy, adnexectomy, or appendectomy before the treatment of ovarian cancer.
1592
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 23, Number 9, November 2013
Recurrence Site After Ovarian Cancer
TABLE 3. Sites of the first recurrence Early Stage Peritoneum Nodes Distal metastases (other than supradiaphragmatic nodes) Peritoneum only Nodes only Distant metastases only Peritoneum + nodes + distant metastases Peritoneum + distant metastases + nodes Nodes + distant metastases + peritoneum Peritoneum + nodes + distant metastases for inclusion in the CARACO trial involving randomization to surgery with vs without dissection of the pelvic and aortocaval nodes).14 The mean numbers of pelvic and aortocaval nodes are given in Table 2. More than 40% of the patients required resection of at least 1 bowel segment; 44%, stripping of the right hemidiaphragm; and 22%, stripping of the left hemidiaphragm. Complete gross resection was achieved in 46 (79.3%) of the 58 patients with advanced disease; in the remaining 12 patients, the residual tumor was less than 1 cm. In the group with advanced disease, tumor spread according to the classification system of Hamilton et al15 was confined to the pelvis and the retroperitoneum in 9 (15.5%) of 58 patients, spared the upper abdomen in 19 (29.3%) of 58 patients, and involved the upper abdomen in 32 (55.2%) of 58 patients.
Outcomes At the time of our retrospective study, the median follow-up was 29 months (range, 2Y71 months). Recurrences were diagnosed in 3 of the 15 patients who had early cancer, with a median PFS of 21.6 months (range, 2.5Y71 months), and in 36 of the 58 patients who had advanced cancer, with a median PFS of 19.3 months (range, 1.8Y67.6 months) (P = 0.03). Of the 15 patients with early-stage disease, none died during follow-up, whose median duration in this group was 27.8 months (range, 2.5Y71 months). Of the 58 patients with advanced disease, 15 died during follow-up, whose median duration was 30.0 months (range, 1.8Y69.3 months). Of the 39 patients with recurrences, 29 (74%) had had complete gross resection and accounted for 47.5% (29/61) of all patients with complete gross resection and 10 (26%) had had gross residual tumor less than 1 cm and accounted for
3/3 0/3 0/3 3/3 0/3 0/3 0/3 0/3 0/3 0/3
Advanced Stage 27/36 14/36 17/36 11/36 5/36 3/36 3/36 8/36 1/36 5/36
(75%) (38%) (47%) (30%) (13%) (8%) (8%) (22%) (2%) (13%)
P (Fisher Test) 1.0 0.54 0.24 0.04 1.0 1.0 1.0 1.0 1.0 1.0
83.3% (10/12) of all patients in this resection group. The difference was statistically significant (P = 0.02). Initial involvement of the aortocaval nodes was significantly associated with a shorter PFS (log-rank; P = 0.03). In contrast, initial pelvic node involvement had no significant impact on PFS (log-rank; P = 0.51). Overall survival was influenced neither by initial aortocaval node involvement (log-rank; P = 0.12) nor by initial pelvic node involvement (log-rank; P = 0.63).
Recurrence Sites Table 3 reports the recurrence sites. The recurrences were confined to the peritoneum in all 3 patients with earlystage disease. In the 36 patients with advanced disease, the peritoneum was the most common site of recurrence (75%). In 40% of the patients with peritoneal recurrences, the recurrences were confined to the peritoneum. Nodal involvement was noted in 38% of all patients with recurrences; among the patients with nodal involvement, 87% had involved nodes below the diaphragm; 40%, both below and above the diaphragm; and 13%, only above the diaphragm. Finally, 47% of the patients had distant metastases at the first recurrence, although only 8% of the patients with recurrences had isolated distant metastases without concomitant peritoneal or nodal disease. According to the classification system developed by Hamilton et al,15 the first recurrence usually involved the upper abdomen. The peritoneal recurrences were confined to the pelvis in 5 (14.2%) of 35 patients, involved the pelvis and the lower abdomen in 5 (14.2%) of 35 patients, and extended to the upper abdomen in 25 (71.4%) of 35 patients. Finally, most of the peritoneal recurrences arose at the treated sites and extended to the untreated sites (Table 4). The
TABLE 4. Peritoneal recurrence sites according to initial sites of peritoneal involvement Site of Peritoneal Recurrence Treated sites only Untreated sites only Treated and untreated sites
Early Stage
Advanced Stage
1/3 1/3 1/3
8/36 7/36 21/36
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1593
Amate et al
International Journal of Gynecological Cancer
recurrence site had no impact on time to recurrence (log-rank; P = 0.15).
Risk Factors Table 5 reports the factors associated with recurrence at each site in the group with advanced disease. The risk for peritoneal recurrence was associated with completeness of initial resection and with peritoneal stripping (left paracolic gutter, right or left hemidiaphragm). Nodal recurrences were significantly more common in the patients with than without initial nodal involvement (10/33 [30.3%] vs 4/40 [10%]; P = 0.04). The mean number of removed aortocaval nodes was significantly greater in the patients with than without nodal recurrences (32.4 T 13.1 vs 24.0 T 12.3, P = 0.04), whereas no differences were found between these 2 patient subgroups for mean number of involved aortic nodes, mean number of removed pelvic nodes, or mean number of involved pelvic nodes. Finally, nodal involvement was significantly more common after resection of the anterior vesical peritoneum. Organ involvement was associated with thoracoscopy during the initial workup and stripping of the left paracolic gutter or the right hemidiaphragm. Histologically, the risk for peritoneal recurrence was associated with initial involvement of 1 of the hemidiaphragms, the omentum, and the ligamentum teres hepatis.
& Volume 23, Number 9, November 2013
Nodal recurrences were associated with initial involvement of the omentum. Finally, distant metastases were associated with initial involvement of the tube, the paracolic gutter, the right hemidiaphragm, and the omentum. The initial mean Sugarbaker index was significantly higher in the patients with recurrences (at any site) than in the other patients (11 T 8.5 vs 6.4 T 6.8, P = 0.01). However, when we compared the patients with and without nodal recurrences, we found no significant difference in the initial mean Sugarbaker index (11.2 T 8.2 vs 8.4 T 8.0, P = 0.27). In contrast, significant differences were found for the patients with versus without peritoneal recurrences (12.1 T 8.2 vs 7.1 T 7.4, P = 0.01) and for the patients with versus without distant metastases (13.5 T 8.3 vs 7.7 T 7.6, P = 0.01).
DISCUSSION Most patients treated for advanced ovarian cancer experience a recurrence within the next 24 months.5,6 The recent change in the surgical paradigm with a shift toward complete primary cytoreduction2 and the use of surgery to treat recurrences8,12 requires an analysis of recurrence sites after modern initial surgical management. The peritoneum was the main site of recurrence in our patients (75%). Among the peritoneal recurrences, 40% were
TABLE 5. Risk factors by recurrence site in the patients with advanced cancer Any Site Peritoneum Nodes Distant Metastases Residual tumor 0/e1 cm Age G / 9 60 y History of abdominal/pelvic surgery Hypertension Diabetes Previous cancer Family history of cancer High PET uptake of initial tumor Carcinomatosis with high PET uptake Omental carcinomatosis by PET Carcinomatosis of diaphragmatic peritoneum by CT only Carcinomatosis of diaphragmatic peritoneum with high PET uptake Lymphadenopathy by CT or MRI Lymphadenopathy with high PET uptake Bowel involvement by CT Bowel involvement by PET Hypermetabolic metastases Thoracoscopy during initial workup After neoadjuvant chemotherapy Ascites Radiological carcinomatosis Omental carcinomatosis by PET Carcinomatosis of diaphragmatic peritoneum by CT only
1594
0.11 0.19 0.74 0.25 0.63 0.41 0.72 0.41 0.01 0.02 0.29 0.09 0.05 0.12 0.69 0.50 1.0 1.0
0.004 0.68 0.99 0.13 0.61 0.11 0.65 1.0 0.10 0.05 0.07 0.05 0.01 0.13 0.13 0.21 0.48 1.0
1.0 0.09 0.45 0.52 1.0 1.0 0.08 1.0 0.31 0.21 0.29 1.0 0.71 0.35 0.17 1.0 0.19 0.61
0.15 0.36 0.90 0.34 0.31 0.66 0.55 1.0 0.16 0.21 0.19 0.23 0.50 0.40 0.23 0.47 0.14 0.04
0.43 0.38 0.54 0.38
1.0 0.68 1.0 0.68
0.67 1.0 1.0 1.0
0.69 0.68 1.0 0.68 * 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 23, Number 9, November 2013
isolated and 58.3% involved both treated and untreated areas (Table 4). The risk factors of peritoneal recurrences reflected the initial disease spread: residual tumor after initial surgery and initial major peritoneal involvement requiring resection procedures. The predominance of peritoneal recurrences and their frequent location at both treated and untreated sites indicate the persistence of tumor deposits even when gross resection is considered complete. These data support the use of extensive peritoneal surgery or methods designed to treat the entire peritoneal cavity such as intraperitoneal chemotherapy with or without hyperthermia.16 Nodal recurrences were noted in 38% of the patients with advanced disease followed by recurrences, usually concomitantly with the involvement of another site and usually in the abdominal compartment. The risk factors of nodal recurrences were nodal involvement identified during the preoperative workup and diffuse peritoneal involvement. Nodal recurrences were also more common among the patients with initial nodal involvement (Table 5). Aortocaval initial involvement was associated with a shorter PFS but without impact on OS. This
Recurrence Site After Ovarian Cancer
emphasizes the debate about the real benefit of lymphadenectomy in these patients.14 Distant metastases were often diagnosed concomitantly with peritoneal and/or nodal recurrences. These were isolated in only 8% of the patients with advanced disease and recurrences. The risk factors were thoracoscopy (which usually indicated stage IV disease), diffuse carcinomatosis (initial mean T SD Sugarbaker index, 13.5 T 8.3 vs 7.7 T 7.6, P = 0.01), and number of neoadjuvant chemotherapy cycles. Thus, despite the greater aggressiveness of the initial surgical procedure, which often included peritoneal resections, the peritoneum remained the main site of recurrence. Peritoneal recurrences involved both treated and untreated sites and often developed in the upper abdomen, indicating that complete gross resection leaves in fact microscopic tumor residuals. This supports the concept of treating the entire peritoneal cavity with intraperitoneal chemotherapy with or without hyperthermia or photodynamic therapy.16 The frequent involvement of the peritoneum also has implications for patient selection to surgery because surgical treatment of recurrences is beneficial
TABLE 5. (Continued)
Carcinomatosis of diaphragmatic peritoneum by PET Lymphadenopathy by CT Lymphadenopathy by PET Bowel involvement by CT Bowel involvement by PET Distant metastases At initial cytoreductive surgery Ascites Resection of Douglas pouch Resection of prevesical peritoneum Resection of right paracolic gutter Resection of left paracolic gutter Resection/stripping of right hemidiaphragm Resection/stripping of left hemidiaphragm Rectosigmoidectomy Left colectomy Right colectomy Transverse colectomy Resection of small bowel Lumboaortic node dissection Pelvic node dissection Splenectomy Fulguration of carcinomatous lesions On the mesentery On the mesocolon On the hemidiaphragms On the small bowel
Any Site
Peritoneum
Nodes
Distant Metastases
0.55 0.29 1.0 1.0 1.0 1.0
0.60 0.60 1.0 0.39 1.0 0.69
1.0 0.04 0.05 0.63 1.0 0.05
1.0 0.63 1.0 0.36 1.0 1.0
0.88 0.17 0.09 0.64 0.15 0.008 0.10 0.13 0.52 0.28 0.52 1.0 1.0 1.0 0.64
0.39 0.07 0.13 0.66 0.02 0.002 0.01 0.22 1.0 0.10 0.21 0.33 1.0 1.0 0.17
0.43 0.34 0.05 1.0 0.59 0.43 0.49 0.36 0.43 1.0 1.0 1.0 1.0 1.0 1.0
0.73 0.26 0.57 0.14 0.02 0.01 0.49 0.66 1.0 0.55 1.0 1.0 0.41 0.41 1.0
1.0 0.38 0.29 0.38
0.69 1.0 0.33 1.0
1.0 1.0 0.24 1.0
0.66 1.0 0.57 1.0
MRI, Magnetic resonance imaging.
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1595
International Journal of Gynecological Cancer
Amate et al
only when complete resection is achieved.13,17,18 Thus, the location of the peritoneal recurrences may limit the potential indications for surgical treatment. To the best of our knowledge, this is the first study whose primary objective was to describe and to analyze recurrence sites after initial complete gross resection of ovarian cancer. The main limitation of our study is the retrospective design, which can result in biases because of patient selection and missing data. Another limitation is that follow-up was shorter than 60 months in some patients. The patients without recurrences, who constituted the control group for our riskfactor analysis, may have developed recurrences later on. However, the sample size and the single-center design limit the potential impact of these sources of bias.
CONCLUSIONS Our data show that the peritoneum is the main recurrence site after contemporary cytoreductive surgery for advanced epithelial cancer involving the ovary, the tube, or the peritoneum. Initial tumor spread and residual tumor after surgery are risk factors of peritoneal recurrence. These data support the use of extensive treatment of the peritoneum.
REFERENCES 1. Bristow RE, Tomacruz RS, Armstrong DK, et al. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol. 2002;20:1248Y1259. 2. Chi DS, Eisenhauer EL, Zivanovic O, et al. Improved progression-free and overall survival in advanced ovarian cancer as a result of a change in surgical paradigm. Gynecol Oncol. 2009;114:26Y31. 3. du Bois A, Reuss A, Pujade-Lauraine E, et al. Role of surgical outcome as prognostic factor in advanced epithelial ovarian cancer: a combined exploratory analysis of 3 prospectively randomized phase 3 multicenter trials: by the Arbeitsgemeinschaft Gynaekologische Onkologie Studiengruppe Ovarialkarzinom (AGO-OVAR) and the Groupe d’Investigateurs Nationaux Pour les Etudes des Cancers de l’Ovaire (GINECO). Cancer. 2009;115:1234Y1244. 4. Vergote I, Trope CG, Amant F, et al. Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer. N Engl J Med. 2010;363:943Y953.
1596
& Volume 23, Number 9, November 2013
5. Burger RA, Brady MF, Bookman MA, et al. Incorporation of bevacizumab in the primary treatment of ovarian cancer. N Engl J Med. 2011;365:2473Y2483. 6. Perren TJ, Swart AM, Pfisterer J, et al. A phase 3 trial of bevacizumab in ovarian cancer. N Engl J Med. 2011;365: 2484Y2496. 7. CHIPOR. Available at: http://clinicaltrials.gov/show/NCT01376752. 8. DESKTOP III. Available at: http://www.clinicaltrials.gov/show/ NCT01166737. 9. Chang SJ, Bristow RE, Ryu HS. Impact of complete cytoreduction leaving no gross residual disease associated with radical cytoreductive surgical procedures on survival in advanced ovarian cancer. Ann Surg Oncol. 2012;19:4059Y4067. 10. Portilla AG, Shigeki K, Dario B, et al. The intraoperative staging systems in the management of peritoneal surface malignancy. J Surg Oncol. 2008;98:228Y231. 11. Klar M, Farthmann J, Bossart M, et al. Video-assisted thoracic surgery (VATS) evaluation of intrathoracic disease in patients with FIGO III and IV stage ovarian cancer. Gynecol Oncol. 2012;126:397Y402. 12. Harter P, Sehouli J, Reuss A, et al. Prospective validation study of a predictive score for operability of recurrent ovarian cancer: the Multicenter Intergroup Study DESKTOP II. A project of the AGO Kommission OVAR, AGO Study Group, NOGGO, AGO-Austria, and MITO. Int J Gynecol Cancer. 2011;21: 289Y295. 13. Harter P, Heitz F, du Bois A. Surgery for relapsed ovarian cancer: when should it be offered?Curr Oncol Rep. 2012;14:539Y543. 14. CARACO. Available at: http://www.e-cancer.fr/recherche/ recherche-clinique/registre-des-essais-cliniques/registre-des-essaiscliniques. 15. Hamilton CA, Miller A, Miller C, et al. The impact of disease distribution on survival in patients with stage III epithelial ovarian cancer cytoreduced to microscopic residual: a Gynecologic Oncology Group study. Gynecol Oncol. 2011;122: 521Y526. 16. OVHIPEC. Available at: http://clinicaltrials.gov/ct2/show/ NCT00426257. 17. Bristow RE, Puri I, Chi DS. Cytoreductive surgery for recurrent ovarian cancer: a meta-analysis. Gynecol Oncol. 2009;112: 265Y274. 18. Chi DS, McCaughty K, Diaz JP, et al. Guidelines and selection criteria for secondary cytoreductive surgery in patients with recurrent, platinum-sensitive epithelial ovarian carcinoma. Cancer. 2006;106:1933Y1939.
* 2013 IGCS and ESGO
Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
