|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Localized Extremity Soft Tissue Sarcoma: Improved Knowledge With Unchanged Survival Over Time
From the Departments of Surgery and Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY. Address reprint requests to Murray F. Brennan, MD, FACS, Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021; email: brennanm{at}mskcc.org.
Purpose: The objective of this study was to define whether survival of patients with extremity soft tissue sarcoma (STS), stratified for known risk factors, has improved over the last 20 years. Patients and Methods: From January 1982 to December 2001, 1,706 patients with primary and recurrent STS of the extremities were treated at our institution and were prospectively followed. From this cohort, we selected 1,261 patients who underwent complete macroscopic resection and had one of the following histopathologies: fibrosarcoma, liposarcoma, leiomyosarcoma, malignant fibrous histiocytoma, or synovial sarcoma. Median follow-up was 55 months. Patient, tumor, and treatment factors were analyzed as prognostic factors. Results: The 5-year disease-specific actuarial survival was 79% (78% for patients treated from 1982 to 1986, 79% for patients treated from 1986 to 1991, 79% for patients treated from 1992 to 1996, and 85% for patients treated from 1997 to 2001; P = not significant). For high-risk patients (high-grade, > 10 cm, deep tumors; n = 247), 5-year disease-specific survival was 51% (50% for patients treated from 1982 to 1986, 45% for patients treated from 1986 to 1991, 52% for patients treated from 1992 to 1996, and 61% for patients treated from 1997 to 2001; P = not significant). Tumor depth, size, grade, microscopic margin status, patient age, presentation status (primary tumor versus local recurrence), location (proximal versus distal), and certain histopathologic subtypes were significant prognostic factors for disease-specific survival on multivariate analysis; however, time period of treatment was not. Conclusion: Prognosis of patients with extremity STS, stratified for known risk factors, has not improved over the last 20 years, indicating that current therapy has reached the limits of efficacy.
IN RECENT years, major advances in the knowledge regarding extremity soft tissue sarcoma (STS) have been achieved. Based on large prospective databases, prognostic clinical factors have been defined that help to stratify patients regarding their risk for local and distant recurrence and death from disease. Tumor grade, size, depth, completeness of resection, and presentation status (primary tumor v local recurrence) are among the independent prognostic factors defined for STS.1–3 It has also become apparent that tumor biology rather than radical surgery is the primary determinant of the fate of the patient, which has led to the acceptance of limb- and function-preserving surgical procedures.4,5 Multimodality treatment has been increasingly used for the management of STS.6 Radiotherapy has been shown to decrease local recurrence of STS in randomized controlled trials without an improvement of survival.7,8 Adjuvant chemotherapy may slightly improve overall survival for extremity STS, with a meta-analysis demonstrating a 7% absolute benefit at 10 years.9 The most recent Cochran meta-analysis also demonstrated an improvement of the 10-year overall survival of 7% for patients with extremity STS treated with adjuvant chemotherapy.10 Besides a better understanding of prognostic factors and an evolution in the clinical management of patients with STS, insights from molecular studies promise to allow an even better risk stratification and the development of targeted molecular therapies.11,12 In light of these advances, the question arises whether this improved knowledge has impacted prognosis of patients with extremity STS over the last two decades. The objective of this study was to evaluate whether there has been significant improvement in the survival results after resection of the five most common histologic subtypes of STS of the extremities by an analysis of a cohort of 1,261 patients over a 20-year period, taking into account currently established clinical prognostic factors.
Patients, Treatment, and Follow-Up This study was approved by the Institutional Review Board of Memorial Sloan-Kettering Cancer Center (New York, NY) with waiver of the informed consent. From January 1982 to December 2001, 1,706 adult patients with primary and recurrent STS of the extremities were admitted and treated at Memorial Sloan-Kettering Cancer Center, entered into a database, and prospectively followed. From this cohort, we selected 1,261 patients who underwent complete macroscopic resection of a localized tumor (with no evidence of distant metastases) and had one of the following common histopathologies: fibrosarcoma, liposarcoma, leiomyosarcoma, malignant fibrous histiocytoma (MFH), and synovial sarcoma. No attempt was made to reclassify MFH from earlier time points. Patient demographics, tumor characteristics, treatment, and postoperative outcome were analyzed. Histopathology was reviewed at a weekly multidisciplinary Soft Tissue Sarcoma Group conference and entered into the database. All patients had a complete resection defined as the absence of macroscopic visible residual disease after excision of the tumor. Some patients received preoperative or adjuvant chemotherapy or radiation either as part of a clinical trial or at the discretion of the multidisciplinary Soft Tissue Sarcoma Group. The length of follow-up was calculated from the date of operation at our institution. Guidelines for follow-up schedule were available; however, follow-up was performed at the discretion of the treating physician and varied according to individual risk of tumor recurrence. Occurrence of distant metastases and local recurrences, follow-up time, and reason for death were obtained for each patient to assess distant recurrence-free survival, local recurrence-free survival, relapse-free survival, and disease-specific survival. Because desmoid tumors of the extremities do not metastasize or lead to death, patients with desmoid tumors (n = 68) were excluded from the analyses for distant recurrent-free survival and disease-specific survival.13,14
Definitions
Statistical Analysis
Clinical, Pathologic, and Treatment Factors Out of the prospective database, 1,261 patients were identified with localized STS (fibrosarcoma, liposarcoma, leiomyosarcoma, MFH, and synovial sarcoma) of the extremities who underwent complete macroscopic resection between 1982 and 2001. Patient age ranged from 16 to 95 years, with a median age of 53 years. Table 1  lists the clinical, pathologic, and treatment factors of the patients. Table 2 stratifies these factors according to time intervals of treatment. In the more recent time periods, patients were older, presented with larger and deeper tumors, were more likely to have primary than recurrent sarcomas, and were less likely to have a liposarcoma. Radiotherapy was more often used more recently, and chemotherapy was most often administered during the first time period.
Follow-Up and Outcome The median follow-up time for all patients was 55 months, with an interquartile range (IQR) of 23 to 103 months; for surviving patients, the median follow-up time was 63 months (IQR, 31 to 120 months). The median follow-up time stratified according to time interval of treatment was 113 months (IQR, 47 to 180 months), 101 months (IQR, 44 to 137 months), 61 months (IQR, 37 to 80 months), and 22 months (IQR, 11 to 39 months) for patients treated from 1982 to 1986, 1987 to 1991, 1992 to 1996, and 1997 to 2001, respectively. The median follow-up time for survivors for the same time intervals was 166 months (IQR, 110 to 194 months), 126 months (IQR, 100 to 146 months), 69 months (IQR, 55 to 86 months), and 25 months (IQR, 12 to 40 months), respectively. Local recurrences occurred in 235 patients (19%), and distant recurrences occurred in 322 patients (26%). At time of the last follow-up, 819 patients (65%) were without evidence of disease (Table 3 ).
Prognostic Factors for Relapse-Free Survival The actuarial 5-year relapse-free survival rate was 62%, and the 10-year survival rate was 55%. Table 4 depicts the univariate and multivariate analysis of prognostic factors for relapse-free survival. Age greater than 50 years, presentation with locally recurrent tumor, tumor size greater than 5 cm, high tumor grade, histologic subtype leiomyosarcoma, and positive microscopic margin were independent adverse prognostic factors. Year of treatment was not an independent prognostic factor for relapse-free survival.
Prognostic Factors for Local Recurrence-Free Survival The actuarial 5- and 10-year local recurrence-free survival rates were 79% and 74%, respectively. Age greater than 50 years, presentation with locally recurrent tumor, tumor size greater than 5 cm, histologic subtype fibrosarcoma, limb-sparing resection, and positive microscopic margin, but not year of treatment, were independent adverse prognostic factors (Table 5 ).
Prognostic Factors for Distant Recurrence-Free Survival The actuarial 5-year distant recurrence-free survival was 72%, and the 10-year distant recurrence-free survival was 67%, excluding patients with desmoid tumors. Age greater than 50 years, presentation with locally recurrent tumor, tumor size greater than 5 cm, deep tumor location, high tumor grade, histologic subtype leiomyosarcoma, and positive microscopic margin emerged as independent adverse prognostic factors. Again, year of treatment was not an independent prognostic factor (Table 6 ).
Prognostic Factors for Disease-Specific Survival The actuarial 2-, 5-, and 10-year disease-specific survival for all patients, excluding patients with desmoid tumors, was 89% (95% confidence interval [CI], 87% to 91%), 79% (95% CI, 76% to 82%), and 73% (95% CI, 70% to 77%), respectively. The actuarial 2-year disease-specific survival was 90% (95% CI, 86% to 93%), 89% (95% CI, 86% to 93%), 90% (95% CI, 86% to 93%), and 89% (95% CI, 85% to 93%) for patients treated from 1982 to 1986, 1987 to 1991, 1992 to 1996, and 1997 to 2001, respectively (P = not significant). The actuarial 5-year disease-specific survival was 78% (95% CI, 73% to 84%), 79% (95% CI, 74% to 84%), 79% (95% CI, 74% to 84%), and 85% (95% CI, 79% to 90%) for the same time intervals. For high-risk patients (high-grade, > 10 cm, deep tumors, n = 247), the actuarial 5-year disease-specific survival was 51% (50% for patients treated from 1982 to 1986, 45% for patients treated from 1986 to 1991, 52% for patients treated from 1992 to 1996, and 61% for patients treated from 1997 to 2001; P = not significant). In the multivariate analysis, age greater than 50 years, presentation with locally recurrent tumor, tumor size greater than 5 cm, deep tumor location, high tumor grade, proximal tumor location, amputation, histologic subtype leiomyosarcoma, and positive microscopic margin, but not year of treatment, emerged as independent adverse prognostic factors (Table 7 ).
This study investigated whether survival of patients with extremity STS, stratified for known risk factors, has improved over the last 20 years because substantial advances in the knowledge regarding extremity STS have been achieved in the recent years.6,12 This study is based on a prospective, well-matured database of over 1,700 adult patients with STS of the extremity. Using this database, clinical prognostic factors have been well characterized, allowing the stratification of patients according to risk of tumor recurrence and death of disease.1,3 Inclusion of these factors in the multivariate analysis of prognostic factors should diminish the influence of patient selection, thereby allowing the detection of any true advance in the efficacy of therapy over time. The comparison of clinical, pathologic, and treatment-related factors between the different treatment time periods demonstrates that such a multivariate analysis is essential because these factors have changed considerably over time (Table 2  ). In the more recent time period, we more frequently treated primary and low-grade tumors, which are factors associated with improved survival. Table 2 also demonstrates that several other factors associated with a worse outcome were more frequently observed during the more recent time periods, such as older patient age, larger tumor size, deep tumor depth, and nonliposarcoma histologies. These data should not be interpreted as true changes in the biology of this disease because they most likely reflect referral pattern to our institution.
Treatment philosophy has changed over the years at our institution. Clearly, surgery is performed less radically, with a major focus on function-preserving procedures seen more recently. This is reflected by the decrease of the amputation rate from 13% in the first time period to 5% in the most recent time period (P = .0001). In parallel, there is an increase in the rate of positive microscopic margins over time, although this did not reach statistical significance (P = .06, Table 2
The multivariate analysis of prognostic factors for distant recurrence-free survival, local recurrence-free survival, relapse-free survival, and disease-specific survival confirmed our previous published data on prognostic factors in patients with extremity STS.1 Age greater than 50 years, presentation with locally recurrent tumor, tumor size greater than 5 cm, deep tumor location, high tumor grade, proximal tumor location, certain histologic subtypes, and positive microscopic margin are important adverse prognostic factors for patients with extremity STS. Time period of treatment was not a significant prognostic factor of survival on either univariate or multivariate analysis (Table 8
One limitation of this study is the difference in follow-up data for the different defined groups, which is unavoidable when analyzing time period of treatment. Because actuarial survival is commonly longer than actual survival, and we did not see a significant improved survival in the most recent time interval, we do not think that our results were biased by this effect. We arbitrarily categorized the patients into four 5-year groups according to the time point of first operation at our institution. Any other categorization would have been equally justified, provided a sufficient patient number in each category. Multiple tests using different time categories must be avoided because of the higher chance of obtaining a false-positive statistical result. A further possibility would be to include time point of treatment as a continuous variable in the multivariate model. We performed such an analysis, and time point of treatment was not a significant prognostic factor. One could argue that treatment-related factors should not be included in the multivariate analysis of survival because inclusion of these factors could interfere with the effect of treatment year in the model because treatment has changed over time. Therefore, we repeated the analysis without inclusion of treatment-related factors (chemotherapy, radiotherapy, amputation v limb-sparing surgery, and margin status). This model defined the same risk factors for all four different survival categories, and again, time period of treatment was not a significant prognostic factor (data not shown). The role of radiotherapy warrants further discussion. Radiotherapy was not associated with an improved local control in our analysis, which is in contrast to the results of randomized trials.7,8 In our patient cohort, radiotherapy was partly administered in the setting of a randomized trial and partly at the discretion of the multidisciplinary Soft Tissue Sarcoma Group at our institution. As expected, radiotherapy was used selectively; when analyzing the association with known prognostic factors, radiotherapy was used significantly more often in proximal, deep, large, high-grade, and margin-positive tumors (data not shown). Radiotherapy was associated with a worse outcome on univariate analysis, which again stresses the point that this modality was more often used in high-risk patients. On multivariate analysis, radiotherapy was not a prognostic factor for local recurrence-free survival, demonstrating that radiotherapy was used so selectively that its effect could not be demonstrated by our analysis. The multivariate analysis apparently was not able to correct for all the factors that led to the decision to administer radiation by the treating physicians. The lack of improvement in survival over time for extremity STS demonstrated by our analysis indicates that current therapy, including risk stratification based on clinical prognostic criteria, surgical and perioperative management, and chemotherapy and radiotherapy, has reached the limits of efficacy. It is also unlikely that implementation of screening programs will be of any benefit because of the rarity of this disease. Improved knowledge regarding the molecular disease mechanisms will most likely be the basis of new effective therapeutic regimens, as has been demonstrated recently for certain sarcoma subtypes.18–20
We thank Madhu Mazumdar, PhD, (Department of Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, NY) for statistical advice.
Supported by grant no. CA47179 (M.F.B.) from the National Institutes of Health, Bethesda, MD.
1. Pisters PWT, Leung DH, Woodruff J, et al: Analysis of prognostic factors in 1041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol 14:1679–1689, 1996 2. Stojadinovic A, Leung DH, Hoos A, et al: Analysis of the prognostic significance of microscopic margins in 2,084 localized primary adult soft tissue sarcomas. Ann Surg 235:424–434, 2002[CrossRef][Medline]
3. Kattan MW, Leung DH, Brennan MF: Postoperative nomogram for 12-year sarcoma specific death. J Clin Oncol 20:791–796, 2002
4. Heslin MJ, Woodruff J, Brennan MF: Prognostic significance of a positive microscopic margin in high-risk extremity soft tissue sarcoma: Implications for management. J Clin Oncol 14:473–478, 1996 5. Rosenberg SA, Tepper J, Glatstein E, et al: The treatment of soft-tissue sarcomas of the extremities. Ann Surg 196:305–315, 1982[Medline] 6. Pisters PWT: Combined modality treatment for extremity soft tissue sarcomas. Ann Surg Oncol 5:464–472, 1998[Abstract]
7. Pisters PWT, Harrison LB, Leung DH, et al: Long-term results of a prospective randomized trial of adjuvant brachytherapy on soft tissue sarcoma. J Clin Oncol 14:859–868, 1996
8. Yang JC, Chang AE, Baker AR, et al: Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the extremity. J Clin Oncol 16:197–203, 1998 9. Tierney JF, Steward LA, Parmar MKB: Adjuvant chemotherapy for localised resectable soft-tissue sarcoma of adults: Meta-analysis of individual data. Lancet 350:1647–1654, 1997[CrossRef][Medline] 10. Sarcoma Meta-analysis Collaboration (SMAC): Adjuvant chemotherapy for localised soft tissue sarcoma in adults (Cochrane Review). Oxford, United Kingdom, The Cochrane Library, Update Software, 2003
11. Ladanyi M, Antonescu CR, Leung DH, et al: Impact of SYT-SSX fusion type on the clinical behavior of synovial sarcoma: A multi-institutional retrospective study on 243 patients. Cancer Res 62:135–140, 2002 12. Singer S, Demetri GD, Baldini EH, et al: Management of soft-tissue sarcomas: An overview and update. Lancet Oncol 1:75–85, 2000[CrossRef][Medline] 13. Lewis JJ, Boland PJ, Leung DH, et al: The enigma of desmoid tumors. Ann Surg 229:866–873, 1999[CrossRef][Medline] 14. Merchant NB, Lewis JJ, Woodruff J, et al: Extremity and trunk desmoid tumors. Cancer 86:2045–2052, 1999[CrossRef][Medline] 15. Hajdu SI, Shiu MH, Brennan MF: The role of the pathologist in the management of soft tissue sarcomas. World J Surg 12:326–331, 1988[CrossRef][Medline] 16. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457–481, 1958[CrossRef] 17. Cox DR: Regression models and life-tables. J R Stat Soc (B) 34:187–220, 1972
18. DeMatteo R: The GIST of targeted cancer therapy: A tumor (gastrointestinal stromal tumor), a mutated gene (c-kit), and a molecular inhibitor (STI571). Ann Surg Oncol 9:831–839, 2002 19. Joensuu H, Fletcher C, Dimitrijevic S, et al: Management of malignant gastrointestinal stromal tumours. Lancet Oncol 3:655–664, 2002[CrossRef][Medline]
20. Rubin BP, Schuetze SM, Eary JF, et al: Molecular targeting of platelet-derived growth factor B by imatinib mesylate in a patient with metastatic dermatofibrosarcoma protuberans. J Clin Oncol 20:3586–3591, 2002 Submitted February 6, 2003; accepted April 26, 2003.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 test. Survival was estimated according to the Kaplan-Meier method and compared using the log-rank test.
.1 at univariate analysis).
