Abstract
Aim: We assessed the long-term local disease-free survival (LDFS) and overall survival (OS) of patients with peritoneal sarcomatosis (PS) uniformly-treated with cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CS plus HIPEC). Patients and Methods: Retrospective data of 15 patients who underwent CS plus HIPEC for PS were extracted from a prospectively collected database. DFS and OS were calculated from the date of CS plus HIPEC to local relapse and death, respectively. Results: After a median follow-up of 28 months (range=4-144 months), median LDFS was 15 months (95% Confidence Interval CI=1-40 months). Median OS was 27 (95% CI=24.7-29.3) months. Long-term OS was achieved in three patients (20%) and ranged between 93 and 144 months. Female sex was the only factor significantly correlated with a greater LDFS (p=0.018). Patients with PS of visceral origin seem at lower risk of recurrence and death but the difference did not prove significant. Conclusion: In our series, long-term survival was achieved in 20% of patients after CS plus HIPEC, with a benefit in female patients with PS of visceral origin. The impact of HIPEC after radical surgery for PS remains questionable and still has to be further evaluated in large cooperative multi-institutional studies.
Cytoreductive surgery (CS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) has been adopted by many centers as the standard treatment for patients with pseudomixoma peritonei (PMP) and malignant peritoneal mesothelioma, and in selected cases of colon, ovary and gastric carcinomatosis (1). The systemic metastatic inefficiency of some primary and recurrent abdominal sarcomas suggest a locoregional treatment of these malignancies in humans (2). The spread of a sarcoma throughout the abdomen, in the absence of extra-abdominal sites of localization, is defined as peritoneal sarcomatosis (PS), a condition considered as a potential indication for CS plus HIPEC. PS can be the result of an abdominal diffusion of retroperitoneal sarcomas, gastrointestinal stromal tumors (GIST), pelvic tumors (uterine leiomyosarcomas) and other rare types of mesenchymal cancer (3). All these tumors are rare, associated with a dismal prognosis and (except GIST) without any effective chemotherapy (4). In waiting for more findings on the efficacy of targeted drugs on sarcomas (5), surgery remains the main treatment. In cases of PS, surgery is reserved for more localized and favorable cases, generally in the setting of a multidisciplinary approach of chemo- and radiotherapy (6). After radical surgery (i.e. no macroscopic residual disease), patients with PS have a median survival of 23 months, with a local relapse expected in up to 80% of cases (7). Intraperitoneal chemotherapy after surgery, either normothermic or hyperthermic (HIPEC), has been investigated in some trials, but there is no evidence that these combined approaches improve survival (8). This work investigated the long-term outcome of patients with recurrent PS uniformly-treated at a single institution with CS plus HIPEC, in an attempt to identify a subgroup of patients who may benefit from CS plus HIPEC.
Patients and Methods
Retrospective data were extracted from a prospectively collected database, which includes patients who underwent CS plus HIPEC for PS at the Department of Surgery, Oncology and Gastroenterology of the University of Padova, between 1995-2011. Clinical and pathological features, as well as previous treatments, HIPEC technical details, and patient local disease-free and overall survival (LDFS and OS) were retrieved from the database. Eligibility criteria for CS plus HIPEC were the following: age ≥18 and ≤75 years, Eastern Cooperative Oncology Group (ECOG) performance status ≤2, no significant comorbidities, no liver metastasis or extra-abdominal disease, PS considered amenable to complete CS at preoperative CT, primary tumor pathology slides reviewed by a pathologist with experience in soft-tissue tumors (MCM). Patients included signed a detailed, written informed consent and were uniformly treated according to a protocol approved by the local Institutional Review Board (Protocol number 293/03/L) (9). Tumor spread was scored at the time of laparotomy using the peritoneal cancer index, PCI (range=1-39), as described by Jacquet and Sugarbaker (10). A size of 2.5 mm of the tumor nodules after CS was considered the maximum value amenable to HIPEC after CS (11).
Clinical pathway of patients with peritoneal sarcomatosis treated at with cytoreductive surgery and hyperthermic intraperitoneal chemotherapy.
Surgery consisted of en-bloc resection of the involved organ(s). If a bowel resection was required, anastomosis were performed after HIPEC. If hysterectomy needed to be performed, the vaginal cuff was closed prior to HIPEC. Parietal peritonectomy was performed only in cases of macroscopic tumor involvement. HIPEC was performed with the open abdomen technique, and was started immediately after CS. Peritoneal perfusion was achieved by an open circuit consisting of two inflow and two outflow catheters connected with a pump, supplied by a heater and a heat exchanger. A temperature monitoring system, using four thermal probes placed on different sites of the circuit and in the peritoneal cavity, was necessary for maintaining the desired temperature. Once the target temperature (41.5°C) was reached, the drugs were bolus-injected into the circuit and the perfusion was initiated and maintained for 60 min. The drug dosages were cisplatin at 25 ml/m2/l plus mitomycin-C 3.3 mg/m2/l in two patients and cisplatin at 45 mg/l plus doxorubicin at 15 mg/l in the others.
Outcome evaluation. Postoperative systemic toxicity was classified using the World Health Organization scale (12). Patient outcome was evaluated as LDFS and OS. Local or distant disease progression was established by abdominal/thoracic (CT) scan performed every six months for the first two years after treatment, then annually.
Statistical analysis. Clinicopathological factors were categorized and analyzed as follows: patient age (<47 vs. ≥47 years) and sex, primary tumor site (visceral vs. non visceral), previous treatment (performed, not performed) and PCI (<6 vs. ≥6). LDFS and OS were calculated as the interval between the first surgery performed at our Department and recurrence or death due to sarcoma. Survival curves were constructed according to the Kaplan Meier method, and compared with the log-rank test. Statistical analyses were conducted with SPSS 17.0 (SPSS Statistics, Release 17.0.0) considering a 95% confidence interval (CI) and an alpha error of 0.05.
Results
From May 1995 to June 2004, 15 patients (9 males) with PS were treated with CS plus HIPEC (Table I). All but one patient (14/15) had recurrent disease at the time of CS plus HIPEC. After (c-KIT) testing introduction, a systematic histopathological and molecular revision of all the cases was carried out and two patients with leiomyosarcoma were re-categorized as having a GIST.
The median time between the first surgery (performed within or outside our Institution) and CS plus HIPEC was 21.5 (range=6-103) months, with a median of one (range=1-3) previous surgical resections.
Survival curves for peritoneal sarcomatosis treated with cytoreductive surgery and hyperthermic intraperitoneal chemotherapy.
Ten patients underwent radio- and/or chemotherapy (six had chemotherapy only, two radiotherapy only, and two patients received both treatments). None of the patients with a revised diagnosis of GIST were treated with inhibitor of the tyrosine kinases (imatinib). Two patients were re-treated with CS plus HIPEC 10 and 17 months after the first CS plus HIPEC, respectively. The primary site of origin of the tumor was retroperitoneal in eight cases (53 %) and visceral in six (40%). In one case, PS was secondary to a metastatic sarcoma of the lower limb. Median PCI at the time of surgery was 5.5 (range=2-15). Surgery was macroscopically complete (CC0) in 14 patients (93.3%). In one patient, residual tumor <2.5 mm (CC1) was obtained. No intra- or postoperative mortality was recorded. Incidence of grade III-IV adverse events was 46.6%. After a median follow-up of 28 months (range=4-144 months), isolated peritoneal disease progression occurred in three patients (20%), distant metastases in seven (47%), and three patients (20%) experienced local and distant progression simultaneously. Median LDFS was 15 months (95% CI=0.1-40 months). LDFS after 1, 3 and 5 year was 64.8%, 17.4% and 17.4%, respectively (Figure 1). Median OS was 27.0 months (95%CI 24.7-29.3). OS after 1-, 3- and 5-year was 80%, 36%, and 29%, respectively (Figure 1). Long-term survival was achieved in three patients (20%, one with ovarian stromal sarcoma, one with retroperitoneal leiomyosarcoma, one with gastric GIST), who had a survival range of 93-144 months. After exploratory analysis of prognostic factors potentially related to LDFS and OS, only sex seemed significantly correlated with LDFS (Table II) (p-value=0.016). There was a trend towards significance for visceral origin of PS for LDFS and OS, with a median OS for patients with visceral site of 53.6 months compared to 26.4 months for those with tumor in other sites (p-value=0.28).
Prognostic factors.
Discussion
Sarcomas originating within the abdominal cavity account for 18% of all sarcomas (3). The majority of them (15%) arise in the retro-peritoneum, whereas 3% have a visceral origin. Progression of retroperitoneal and visceral sarcomas can result from different patterns of spread: local infiltration, peritoneal implantation, hematogenous and seldom lymphatic dissemination (2). A variable proportion of sarcomas tends to be confined in the abdomen or in the retroperitoneal space for a long time (so-called metastatic inefficiency), directly invading with time the visceral organs or diffusing in the abdominal cavity as multiple nodules, a condition named PS (13). PS may occur at presentation or, more often, after surgical treatment and represents a clinical challenge because no effective treatment has been developed yet. Traditional therapy of PS (debulking surgery and systemic chemoradiotherapy) does not translate into a clear survival advantage, with a median survival of 12-13 months (6, 7, 14). Radical surgery (i.e. no macroscopic residual disease) remains the only chance of cure of PS. In a large group of patients (N=72) with PS, Karakousis et al. reported a median survival of 23 months after radical surgery compared with only nine months for the group that underwent debulking/palliative surgery only (7). The impact of radical surgery on PS was confirmed by Bonvalot et al., who demonstrated that an aggressive surgical approach can achieve a median and 5-year survival of 29 months and 40%, respectively (15). The benefit of complete surgical removal of macroscopic disease on the natural history of PS gave the theoretical basis for including intraperitoneal chemotherapy (IPC) in a multimodal treatment plan (15-25). The results of studies on IPC in PS are difficult to interpret because of the high variability of indications, radicality of surgery, employed drugs and adopted techniques (postoperative normothermic vs. intraoperative hyperthermic). Early postoperative intraperitoneal chemotherapy (EPIC) with cisplatin, doxorubicin and mitoxantrone has been tested in some uncontrolled series, with a 5-year OS of 7-30% (16-18). In a small-single center prospective randomized trial comparing radical surgery with (N=19) and without EPIC (N=19), no statistical difference in LDFS, metastatic relapse-free survival or OS were identified (12.5, 18 and 29 months, respectively) (15). EPIC is probably efficient for eliminating microscopic disease, but postoperative adhesions and chemical sclerosis around the catheter cause a heterogeneous drug distribution throughout the abdomen, voiding the therapeutic effect of intraperitoneal drug delivery (16). Theoretically, HIPEC allows for a more homogenous drug perfusion and can overcome these issues, although its efficacy has not been tested in prospective randomized trials, as only small phase I-II studies have been conducted (Table III). After CS plus HIPEC, the grade of cytoreduction (22), tumor grading (22, 23), tumor load (17, 25) and locoregional disease-free interval (25) are the prognostic factors affecting the outcome. Berthet et al. found that the only factors related to survival were the number of regions affected by PS and the number of anatomical sites involved (17). In the same study, the PCI, which combines the number of abdominal quadrants and the tumor load of each quadrant, had prognostic value only at univariate analysis. Some as previous other reports (22, 23), our study, did not find any correlation between PCI and survival in thoroughly resected patients. The lack of prognostic impact of PCI could be explained by the fact that the mean and median PCI values reported in all these studies were always low (e.g. <15), probably reflecting the selection of patients with a lower tumor load.
Literature review on cytoreductive surgery and hyperthermic intraperitoneal chemotherapy in peritoneal sarcomatosis. Only series with more than 10 patients are included.
Higher tumor load may alter the prognosis, as shown by Lim et al. who investigated two HIPEC regimes (cisplatin vs. cisplatin plus mitoxantrone), finding a median survival of 17 months for both arms (21). The authors explained this very poor outcome with the high tumor burden which affected patients of both arms (median size of nodules 9 and 12.2 cm and median number of resected nodules of 100 and 83, respectively).
The negative impact of an extensive peritoneal involvement on patient outcome after CS plus HIPEC has been confirmed for other treatments. In 51 patients with PS treated with palliative/debulking surgery plus palliative radio- and/or chemotherapy, Billimoira et al. identified two groups at different risk of death (high vs. low tumor volume) with the presence of >10 nodules with a single tumor mass >5 cm or >20 separate nodules of any size (6). In the low-volume group, the 2-year OS was 82% compared with the 24% in the high volume group. No impact on prognosis was detected for the different sarcoma histotypes, tumor grade and non-operative treatments. The number of nodules was a prognostic factor, even in a retrospective analysis of Anaya et al. on retroperitoneal sarcomas (26). In this study, multifocality was a prognostic factor for OS, independently of other known adverse prognostic features (e.g. recurrence, grading, thoroughness of surgery). These data suggested that the number of tumor nodules reflects a more aggressive disease independently of the therapeutic approach proposed for PS, and the presence of multifocal disease should drive the treatment planning for these patients.
In our small series, patients with PS of visceral origin were found to have a trend towards a better outcome both in terms of local control and survival, although not significant. The possibility of a more favorable prognosis for visceral PS is controversial in the literature (15, 23). An interesting working hypothesis arises from the study of Baratti et al. which found a potential benefit of CS plus HIPEC treatment in female patients with PS derived from uterine leiomyosarcoma (23). Uterine leiomyosarcoma is characterized by a high tendency of intra-peritoneal diffusion, as shown after tumor morcellation and spillage during surgery (27). PS from uterine leiomyosarcoma may be a condition suitable for CS plus HIPEC, but additional data are needed to confirm whether it represents an indication for such procedure.
Although there is no agreement on a possible correlation between sex and prognosis (13, 23), in our study, female patients seemed to be at lower risk of disease progression. The prognostic role of sex has been investigated in other types of carcinomatosis, such as peritoneal mesothelioma, where a better survival was detected for female patients after CS plus HIPEC (28). Peritoneal mesothelioma, which is characterized by distinct histological entities (epithelial, sarcomatoid or mixed), has a different pattern of expression of estrogen and progesterone receptors in males and females, supporting the hypothesis of a hormone-driven mechanism of tumor growth (29). Although the hormonal profile of sarcomas is poorly defined, recent immunohistochemical analysis of estrogen and progesterone receptor expressions have shown their significant influence on the outcome of patients with non-GIST soft tissue sarcomas (30) and uterine leymiosarcomas (31). The potential role of sexual hormones in sarcomas is also supported by a prolonged disease-free survival observed for patients with endometrial stromal sarcoma and uterine leiomyosarcoma treated with aromatase inhibitors (32). Further investigations are needed to evaluate the role of sex as a prognostic factor and selection criterion for PS.
In conclusion, we have reported the results of an inhomogenous and small group of patients with PS uniformely treated at our Institution. Due to the lack of information on these patients with very poor prognosis, our experience may be relevant for institutions treating patients with sarcoma. Our study confirms the results reported in literature on CS plus HIPEC treatment of PS. Radical surgery with no microscopic residual disease seems to be the only valid therapeutic option for sarcoma diffusion throughout the abdomen, taking into account that multifocality is probably associated with a poorer prognosis and only patients with lower tumor load seem to benefit from an aggressive surgical approach. The role of additional prognostic factors, such as tumor of visceral origin, female sex and uterine leiomyosarcoma histology should be further tested. The impact of HIPEC after radical surgery for PS remains questionable and still needs to be further evaluated in large cooperative multi-institution studies.
Footnotes
-
↵† Oral presentation at the 32nd Congress of the European Society of Surgical Oncology, Valencia, Spain, 19-21 September 2012.
-
Conflicts of Interest
The Authors have nothing disclore.
- Received May 30, 2013.
- Revision received July 1, 2013.
- Accepted July 3, 2013.
- Copyright© 2013 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
