Abstract
Background: Aggressive surgery has been suggested for improving local tumor control in patients with retroperitoneal sarcoma (RS). This study aimed at investigating local disease-free and overall survival after complete surgery, in patients with RS. Patients and Methods: Retrospective data of patients submitted to complete surgery for RS were extracted from a prospectively-maintained database. Results: Forty-three out of 78 patients (55%) presented with primary RS. Infiltrated organs were resected in 42 patients (54%). Patients presenting with recurrent (hazard ratio (HR)=5.57, p=0.002) and high-grade (HR 3.47, p=0.041) tumors were at higher risk of local recurrence. Microscopically-involved tumor resection margins (HR=3.47, p=0.04) and recurrent tumor at presentation (HR=2.49, p=0.008) were independent predictors of poor survival. Conclusion: Patients presenting with primary RS had longer local disease-free survival and overall survival than those with recurrent tumor after complete surgery. Complete surgery remains the standard-of-care for patients with primary RS.
Soft tissue sarcomas are rare malignant tumors that develop from mesenchymal tissue and occur in the retroperitoneal space in 15% of patients with a soft tissue sarcoma (1). The incidence of retroperitoneal sarcoma (RS) is less than 1/100,000 people per year and accounts for approximately one-third of all retroperitoneal masses. Currently, more than 50 histological types of soft tissue sarcoma have been identified, but the most common adult types are liposarcoma (17%), leiomyosarcoma (16%), pleomorphic sarcoma/malignant fibrous histiocytoma (3.5%), myxofibrosarcoma (3%) and synovial sarcoma (2%) (2).
Waiting for patient- and histotype-tailored therapies for RS, surgery still represents the standard treatment (2-4). The optimal goal of surgery is complete resection of the tumor mass and the achievement of microscopically-negative margins (R0), including the resection of multifocal disease when detected [i.e. the presence of two or more tumor nodules within the retroperitoneal space], which is observed in 6-20% of patients (5, 6) and adjacent and vital structures, such as the inferior vena cava, when clinically involved (7, 8).
A thorough excision of retroperitoneal sarcoma is feasible in up to 70% of patients, leading to a five-year local recurrence-free survival and overall survival of 55-80% and 39-90%, respectively (9-11).
Considering the relatively high incidence of local relapse and the infiltrative pattern of growth of RS, some authors have advocated wider resection to improve the thoroughness of surgery (6, 12-14). This approach has been called aggressive surgery and consists of the resection of the tumor mass with wide excisional margins, which includes adjacent organs located around the tumor although not clinically (i.e. macroscopically) involved. Aggressive surgery for RS has led to promising outcomes as local recurrence were observed in 21-22% of patients and 5-year overall survival was 65-68% (12-14). In two retrospective studies, patients treated with aggressive surgery seemed to have a longer local disease-free interval than patients who underwent complete surgery, while no differences were shown for overall survival (13, 14).
To date, complete surgery which is the current standard approach, as suggested by several guidelines (3, 4), and aggressive surgery have never been compared in randomized trials to our knowledge.
In this study, local tumor control and the prognosis of patients with primary or recurrent RS who underwent complete surgery, performed by removing organs adjacent to tumor mass only when involved, were investigated.
Patients and Methods
Case series. The data for patients with RS treated between 1989 and 2010 at the Surgery Branch of the Department of Oncology and Surgical Science of the University of Padova and the Melanoma and Sarcomas Unit of the Veneto Institute of Oncology were extracted from a prospectively-maintained database, using the following selection criteria: I) patients with RS who underwent complete surgery; II) if a previous surgery had been performed at another center, a pathological review performed by our pathologist with expertise in soft tissue sarcoma (M.C.M.) had to be available; III) >6 months of follow-up.
Exclusion criteria were the presence of distant metastases at the time of RS diagnosis or disease not amenable to complete surgery.
Patients were evaluated within the frame of our multidisciplinary team, which is composed of surgeons, radiologists, medical and radiation oncologists and pathologists in order to assess tumor extension and possible involvement of adjacent organs and to plan tumor resection and consider neo/adjuvant radio- and chemo-therapy.
Patients included in this study underwent complete surgery. Our surgical policy is to remove the tumor with adjacent organs when not dissectible or when organs are macroscopically infiltrated to obtain clear excision margins. Clinically-uninvolved but adjacent organs were not resected.
The following clinical and pathological features were recorded for each patient: sex, age, whether tumor presented as primary or recurrent RS; tumor size, grade and histotype; presence of multifocal disease; type of resection; and number and type of resected organs. The tumor grade was determined using the French Federation of Cancer Centers Sarcoma Group Grading System (FNCLCC) (15).
Microscopically-clear margins (R0) were defined as no microscopic residual disease after complete surgery; R1 was defined as microscopically-involved margins but macroscopically-complete resection, and finally R2 was defined as gross residual disease.
Multifocal disease was defined as the presence of two or more non-contiguous tumor nodules in both the abdomen and the retroperitoneal space.
During follow-up, sarcoma recurrences were recorded as local (defined as occurring within the retroperitoneal space) and distant. Sarcoma-specific overall survival was considered. Patients were submitted to a regular follow-up schedule consisting of a physical examination and a whole-body computed tomographic scan every six months for the first five years, then yearly.
Statistical methods. The following clinicopathological factors were considered in the prediction of local disease-free survival and overall survival: age (continuous variable), sex, tumor presentation (primary or recurrent), number of recurrences (continuous variable), tumor size (continuous variable), histological subtype (liposarcoma, leiomyosarcoma and other), tumor grade (G1, G2 and G3 according to FNCLCC), tumor resection margins (R0, R1, R2), number of resected organs (continuous variable), presence of multifocal disease (absent, present).
Analyses were conducted for patients who achieved a R0-1 resection to assess prognostic factors for local disease-free survival and overall survival. Local disease-free survival was calculated from the date of complete surgery to local recurrence or last follow-up. Overall survival was calculated from the time of surgery to time of disease-specific death or the last follow-up.
A univariate analysis was performed with the Kaplan Meier limit method and the Cox proportional hazard model for categorical and continuous variables, respectively. The survival curves were generated and compared with the Kaplan Meier method and the log-rank test, respectively.
A multivariate analysis was conducted with the Cox proportional hazard model using a stepwise procedure in order to select significant variables by considering parameters with at least p≤0.15.
The results were considered statistically significant when p-value <0.05. All data were analyzed with SPSS 18.0 (SPSS Statistics, Release 18.0.0, Chicago, Il).
Results
Case load. During the considered time period (1989-2010), 97 patients underwent complete surgery for RS. Out of these, 19 were excluded from analyses for the following reasons: 10 patients were lost to follow-up; 9 patients presenting with recurrent tumor did not have a pathologicy review of the first surgery performed at our institution. Seventy-eight patients who underwent complete surgery for primary (N=43, 55%) or recurrent (N=35, 45%) RS were thus considered for the analysis. Detailed patient features are reported in Table I.
Forty-two patients (54%) underwent resection of one or more adjacent organs (8/10 presented multifocal disease). Out of these, 15 (36%) required the resection of one organ and 14 (33%) of two organs. The kidney was the most often resected organ (19 cases, 24%), followed by the colon (7 cases, 9%).
Morbidity. The overall post-operative morbidity rate was 17% (13/78). Five patients with primary RS (12%) developed an adverse event, while a greater proportion of patients with recurrent RS had post-surgery complications (8/36, 22%).
In cases with primary RS, femoral nerve injuries were observed in 2 patients as the tumor mass involved the psoas muscle; 2 patients developed post-operative bleeding requiring transfusion and 1 had a biliary tract leakage.
Eight patients with recurrent RS had a major morbidity event: three cases had post operative bleeding requiring transfusion, 2 patients developed intra-abdominal abscess requiring for percutaneous placement of a catheter under radiological guidance, 1 had a biliary tract leakage, 1 patient a dehiscence of the small bowel anastomosis, and in 1 patient, a deep venous thrombosis occurred. Re-operation was never required.
Local disease-free survival. Only patients with R0-1 tumor resection margins were considered for the final analysis (N=73), as five patients had an R2 resection, although R0-1 surgery had been considered feasible at the pre-operative work-up.
After a median follow-up of 49 (interquartile range (IQR)=19-103) months, 32 patients developed an intra-abdominal recurrence (44%), 11 of those presenting with a primary RS (34%) and 21 of those presenting with a recurrent RS (65.5%). The 5-year local disease-free survival rate was 60% (standard error, SE=6%).
Clinical and pathological features of patients with primary (N=43, 55%) and recurrent retroperitoneal soft tissue sarcoma (N=35, 45%).
At univariate analysis, the following parameters were correlated with local disease-free survival: presentation (5-year local disease-free survival for patients with primary and those with recurrent tumor was 86% and 21%, respectively; p<0.001, Figure 1), histotype (5-year local disease-free survival for leiomyosarcoma and other tumor histotypes was 42% and 65%, respectively; p=0.027), tumor grade (5-year local disease-free survival of for G1 and G2-3 was 73% and 50%, respectively; p=0.037) and the number of resected organs [hazard ratio (HR)=0.62; 95% confidence interval (CI)=0.40-0.99; p=0.045].
Sex (p=0.810), age (HR=1.01, 95% CI=0.99-1.04, p=0.243), tumor size (HR=0.97, 95% CI=0.93-1.01; p=0.110), multifocal disease at diagnosis (p=0.430), and resection margin (p=0.458) were not significant prognosticators for local disease-free survival.
At multivariate analysis for prognostic factors for local tumor control (including presentation, histotype, tumor grade, tumor size and number of resected organs), tumor presentation (HR=5.57, 95% CI=1.93-16.10; p=0.002) and tumor grade (HR=3.47, 95% CI=1.05-11.45; p=0.041) were the only independent prognostic factors.
By splitting the patient series according to presentation, none of the considered variables was able to predict for local disease-free survival among patients with recurrent tumor. In contrasts, in patients with primary RS, histotype was the only prognostic factor for local control (5-year local disease-free survival for leiomyosarcoma and other tumor histotypes was 62% and 93%, respectively; p=0.003).
Interestingly, and considering the subgroup of patients who underwent resection of organs adjacent to the tumor because of its macroscopic involvement, the number of resected organs achieved a borderline significance as a predictor of local disease-free survival in univariate analysis (HR=0.63; 95% CI=0.40-0.99; p=0.045). At multivariate analysis, the number of resected organs was no longer significant, while tumor grading (HR=3.52; 95% CI=1.17-10.61; p=0.025) and presentation (HR=6.48; 95% CI=2.2-19.05; p=0.001) predicted local disease-free survival independently.
Overall survival. During follow-up, 37 patients (51%) died because of tumor progression. The 5-year sarcoma-specific survival was 54% (SE=6%).
At univariate analysis, the following parameters were significantly correlated with overall survival: presentation (5-year overall survival for patients with primary and those with recurrent tumor was 70% and 36%, respectively; p=0.002; Figure 2), tumor resection margins (5-year overall survival for patients with R0 and R1 resections was 93% and 45%, respectively; p=0.013; Figure 3), tumor grade (p=0.017) and tumor size (HR=0.96; 95% CI=0.92-0.99; p=0.047).
Sex (p=0.864), age (HR=1.01; 95% CI=0.99-1.03, p=0.467), multifocal disease at diagnosis (p=0.804), histotype (p=0.082) and number of resected organs (HR=0.85; 95% CI=0.58-1.27; p=0.445) were not significant prognostic factors for overall survival.
Microscopically-involved tumor resection margins (HR=3.47; 95% CI=1.06-11.41; p=0.04) and recurrent tumor at presentation (HR=2.49; 95% CI=1.27-4.89; p=0.008) were the only independent prognostic factors for poorer overall survival upon multivariate analysis.
Considering the subgroup of patients with primary RS, overall survival was predicted by sex (5-year overall survival for males and females was 89% and 56%, respectively; p=0.019), grading (5-year overall survival for those with G1 and those with G2-3 was 92% and 53%, respectively; p=0.018), histotype (5-year overall survival for those with leiomyosarcoma and those with other tumor histotypes was 50% and 76%, respectively; p=0.045) and tumor size (HR=0.9, 95% CI=0.82-0.98; p=0.014) at univariate analysis. Tumor size (HR=0.91, 95% CI=0.84.0.98; p=0.016) and sex (HR=4.89; 95% CI=1.08-22.21, p=0.04) were the only independent predictors of overall survival in patients with a primary RS.
Considering the subgroup of patients with recurrent RS, overall survival was predicted by tumor resection margins (5-year overall survival for R0 and R1 resections was 100% and 27%, respectively; p=0.016) at univariate analysis. Tumor size had a borderline significance (HR=1.06; 95% CI=0.99-1.12, p=0.056).
The number of resected organs was not a significant prognostic factor for survival, neither in cases of primary tumor nor in cases of recurrent tumor.
Discussion
The guidelines for the treatment of RS recommend complete surgery consisting of the resection of the localized tumor mass with clinically-negative excision margins (3, 4). As histopathological margin status is recognized as being the most important prognostic factor contributing to long-term local disease free survival (16), aggressive surgery, consisting of the excision of organs and viscera adjacent to the tumor mass although clinically uninvolved, has been proposed to improve local tumor control in patients with primary RS (13, 14). Complete and aggressive surgeries have been directly compared in retrospective series (13, 14), suggesting a possible improvement in tumor control after aggressive surgery, especially in patients with low-grade tumors (17). Several concerns exist regarding the retrospective design of such studies (including a limited length of follow-up), as well as a lack of standardization of the aggressive surgical technique, and the absence of prospective studies designed to compare complete and aggressive surgery (18, 19).
Here, we reported on the outcome of 78 patients treated with complete surgery at a single institution. The patients had been evaluated pre-operatively in multidisciplinary team decision meetings in order to ensure adherence to guidelines and offer the best multidisciplinary management. Resection was performed according to the policy of complete surgery of tumor mass and adjacent organs when clinically affected by the tumor.
Morbidity. In the present series, we recorded a 17% post-operative morbidity rate, and no re-operation for the management of post-surgery morbidity was performed. Gronchi et al. did not report any complications after aggressive surgery (14), while Bonvalot et al. reported that 11% of patients submitted to resection of adjacent uninvolved organs along with tumor mass underwent re-operation (13).
Local disease-free survival. Local recurrence was developed by 44% of patients (34% of patients with primary RS and 66% of patients with recurrent RS) following a median follow-up of 49 months. Overall, local tumor control was achieved in 60% of patients after five years following complete surgery. In patients with primary and recurrent RS, five-year tumor local control was 86% and 21%, respectively (p<0.001). Tumor presentation and grade were independent predictors of local recurrence after adjustment for tumor histotype and number of resected organs, which were the other significant prognostic factors in univariate analysis.
According to two retrospective studies, one from a single institution series (14) and the other from a population-based data collection (The French Association of Surgery) (13), patients treated with aggressive surgery had a longer local disease-free interval than patients submitted to complete surgery. After five years, 78% of patients treated with the aggressive technique were free from local relapse (12). This achievement did not translate into an overall survival benefit, as the prognosis of patients is similar for the two surgical approaches (13, 14). Surgical re-intervention was required in 12% of patients as major post operative morbidity occurred (e.g. anastomotic leakage, post-operative bleeding, retroperitoneal abscess and wound dehiscence) (12).
These findings have been criticized, as patients submitted to aggressive surgery, belonging to a more recent period, have a much shorter follow-up time than those treated with complete surgery (Table II) (18, 19). Moreover, no standardized methods of conducting and reporting pre-operative imaging, operative technique, processing of pathology specimens on pathology findings have been used.
Overall survival. A total of 54% of patients are alive five years after complete surgery. Among those who had a primary RS, the 5-year survival rate, at 70%, is similar to the one reported after aggressive (65-68%) and complete (48-70%) surgeries (Tables II and III). Considering the potential impact of wide-organ resection performed with aggressive surgery on the quality of life, as well as the similar overall survival rate after aggressive and complete surgery and the uncertain impact on local tumor control, it is still to be determined whether or not such an invasive approach provides patients with a better treatment opportunity. Moreover, the similar survival, after complete and aggressive surgeries of patients with primary RS, leads to the consideration of the metastatic potential of RS, rather than the impact of different surgical treatment modalities, as being the main determinant of patient outcome.
Local disease-free survival according to tumor presentation (primary versus recurrent); log-rank test, p<0.001.
Multifocal disease was present at surgery in 13% of our patients and did not affect local tumor control on survival. In contrast, in a recent study by Anaya et al., the presence of multiple sarcoma nodules within the retroperitoneal space (incidence 20%) correlated with worse survival, particularly in the case of more than seven localizations (5). In our series, multifocal disease occurred more frequently in recurrent than primary RS (25.7% versus 2.3%, p=0.004), thus suggesting multifocality as a marker of a more advanced disease rather than of malignant biological potential. Whether the presence of a multiple non-contiguous tumor mass is a marker of more aggressive tumor biology or simply an index of advanced disease, is difficult to address as multifocality could occur at several stages of a tumor's progress.
Overall survival according to tumor presentation (primary versus recurrent); log-rank test, p=0.002.
Overall survival according to tumor excision margins (R0 versus R1); log-rank test, p=0.013.
Conclusion
Overall, these findings support the referral of a patient with a primary RS to a highly-trained cancer center, where pre-defined therapeutic policies are employed. The evaluation of patients in multidisciplinary meetings to elaborate the treatment strategy following guidelines, or within a cancer network, have been shown to be meaningful independent prognostic factors for patients with soft tissue sarcomas (13, 20-22).
Studies reporting 5-year local disease-free survival (LDFS) and 5-year overall survival (OS) of patients with primary RS treated with aggressive surgery.
Studies reporting 5-year local disease-free survival (LDFS) and 5-year overall survival (OS) of patients with primary RS treated with complete surgery.
Complete surgery remains the treatment of choice for patients with RS unless a prospective trial is designed to test the hypothesis of a better outcome for patients treated with aggressive surgery. Aggressive surgery could be evaluated in patients with recurrent RS who have a high probability of local tumor relapse.
Acknowledgements
The Authors are thankful to Mrs Valentina Rossi for editing the manuscript.
- Received June 30, 2013.
- Revision received July 29, 2013.
- Accepted July 30, 2013.
- Copyright© 2013 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.