Annals of Surgical Oncology 10:432-440 (2003)
© 2003 Society of Surgical Oncology
Histopathologic Type: An Independent Prognostic Factor in Primary Soft Tissue Sarcoma of the Extremity?
Jonathan B. Koea, MD,
Dennis Leung, PhD,
Jonathan J. Lewis, MD and
Murray F. Brennan, MD
From the Departments of Surgery (JBK, JJL, MFB) and Biostatistics (DL), Memorial Sloan-Kettering Cancer Center, New York, New York.
Correspondence: Address correspondence and reprint requests to: Murray F. Brennan, MD, Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Ave., New York, NY 10021; Fax: 212-794-3184; E-mail: brennanm{at}mskcc.org
 |
ABSTRACT
|
|---|
Background: We attempted to define the effect of tumor histotype on local recurrence, distant metastasis, and disease-specific survival in patients with surgically treated primary extremity sarcoma.
Methods: A total of 951 patients with primary, localized soft tissue extremity sarcoma were followed up prospectively. Patient- and tumor-related variables, including histopathologic type, were used to identify independent prognostic factors for the study end points of local recurrence, distant recurrence, and disease-specific survival.
Results: There were 137 local recurrences, and significant adverse prognostic factors for local recurrence were patient age >50 years, microscopically positive margins, and malignant peripheral nerve tumor. Adverse prognostic factors for distant recurrence (200 patients) were tumor size >5 cm, tumors beneath the investing fascia, high tumor grade, and leiomyosarcoma. Of the 199 patients who died of disease-related causes, patient age >50 years, tumors beneath the investing fascia, high tumor grade, microscopically positive margin, tumor size >5 cm, leiomyosarcoma, and malignant peripheral nerve tumor were adverse prognostic factors.
Conclusions: These data suggest that differences in biological behavior may exist between sarcoma histotypes and deserve further study.
Key Words: Soft tissue sarcoma • Histopathology • Mortality • Prognostic factors
|
INTRODUCTION
|
|---|
Soft tissue sarcoma is a rare tumor, and approximately 8000 new cases are diagnosed annually in the United States. These tumors arise in mesenchymal tissues and are histopathologically classified on the basis of their tissue of origin. Several investigations have consistently identified the presence of large tumors, high tumor grade, and recurrent disease as adverse prognostic factors for local recurrence, distant metastasis, and tumor-related mortality.1–8 The effect of histopathologic subtype on prognosis remains unclear, although there is a persisting clinical impression that some tumors, principally malignant peripheral nerve tumors, are associated with poor clinical outcomes, whereas others, such as fibrosarcoma, are associated with more favorable outcomes. A number of the investigations that have attempted to define prognostic factors in extremity sarcoma have been based on small numbers of patients,8 and factors such as histopathologic subtype could not be evaluated with sufficient statistical power. In addition, some of these investigations are based on retrospective chart review,5 making the results subject to the limitations of retrospective investigations, whereas other studies have included patients with both primary and recurrent disease, further obscuring the effect of histopathologic type.1 An earlier investigation from our institution identified the different outcomes for subtypes of extremity liposarcoma,9 but the numbers involved were small, and related factors of tumor grade, size, and depth were difficult to evaluate independently.
Patients with primary extremity sarcomas represent an optimal study population to identify the effect of histopathologic type on outcome in soft tissue sarcoma. By using prospectively collected data, this investigation was undertaken to define the effect of tumor histotype on local recurrence, distant metastasis, and disease-specific survival in patients with surgically treated primary sarcoma of the extremity.
|
PATIENTS AND METHODS
|
|---|
A prospective database for patients with a diagnosis of sarcoma was begun at Memorial Sloan-Kettering Cancer Center in July 1982. Between July 1, 1982, and July 1, 1998, 951 patients older than 16 years were admitted to Memorial Sloan-Kettering Cancer Center with a diagnosis of localized (without evidence of regional or distant metastases) primary soft tissue sarcoma of the extremity and form the basis of this study. The diagnosis of extremity sarcoma was verified by review of outside pathologic material or after biopsy (either percutaneous or open) performed under local anesthesia during the initial clinic visit. Because of the low disease-related mortality associated with desmoid tumors,10 patients with a diagnosis of desmoid tumor were not included in this investigation. All patients presented with primary disease, and no patient had undergone any surgical procedure other than biopsy. No patient had been previously treated with either chemotherapy or radiotherapy.
A tumor was considered to be a localized primary tumor if it was previously untreated and if there was no evidence of metastasis. Upper extremity tumors were defined as those at, or distal to, the shoulder joint, whereas lower extremity tumors were defined as those at, or distal to, the hip joint. Axillary and iliac fossa tumors were not included in this series. Tumor size was defined as the maximum diameter of the tumor at pathologic analysis. Tumors were classified as small (<5 cm in diameter), intermediate (5 to 10 cm in diameter), or large (>10 cm in diameter). Tumor grade was classified as either high or low on the basis of the degree of cellularity, degree of differentiation, number of mitoses per high-power field, amount of stromal necrosis, and degree of vascularity.11 The anatomical depth of tumors was defined relative to the investing fascia of the extremity, with tumors defined as either superficial or deep. A tumor was defined as invading bone or a neurovascular structure if there was either gross or microscopic involvement by tumor at operation or pathologic analysis. A microscopically positive margin was defined as tumor within less than one half of a x10 microscopic field (1 mm) from the inked margin.
Nearly all patients were treated with surgical resection with either limb-sparing surgery or amputation. All patients were reviewed by the Multidisciplinary Soft Tissue Sarcoma Group, and some patients received adjuvant chemotherapy, radiation, or both. The decision to treat patients with chemotherapy, radiotherapy, or both was based on the presence of adverse prognostic factors. For example, patients with large (>10-cm-diameter) high-grade tumors were usually offered preoperative chemotherapy, whereas small tumors (<5 cm in diameter) were treated with surgery alone. Any difference in outcome between patients who received chemotherapy and those who did not may not necessarily reflect the effectiveness of preoperative chemotherapy. Inclusion of any treatment-related variables as covariates in an analysis designed to identify prognostic factors could be misleading. Therefore, treatment-related factors were not included in the multivariate analysis.
The following factors were correlated with rates of tumor-related mortality: local recurrence; distant recurrence; patient age, sex, and presentation symptoms; tumor depth, grade, and size; the presence of neurovascular or bony invasion; the presence of positive surgical margins; and histopathologic type. The end points for multivariate analysis were local recurrence, distant recurrence, and disease-specific survival. The time of the first recurrence was calculated from the date of first presentation to Memorial Sloan-Kettering Cancer Center.
Summary statistics were obtained by using standard methods. Only deaths that were confirmed to be related to the sarcoma were considered when tumor-related mortality was studied. All other deaths were censored. The rate of recurrence or death was estimated with the method of Kaplan and Meier, and the effect of each variable on the event rate was examined with the log-rank test. Comprehensive analysis of the prognostic values of all covariates jointly was performed for each end point by using the Cox proportional hazards model.12 Covariates were selected into the model only if they contributed significantly to the fit of the model. For variables found to have independent prognostic value by multivariate analysis, relative risks (RR) with confidence intervals (CI) were calculated. In defining statistical differences between different histopathologic types, patients with malignant fibrous histiocytoma (MFH) were used as a comparison, because this group contained the largest number of patients and occupied an intermediate position with respect to all study end-point measures. Outcomes are therefore expressed relative to outcomes for MFH. It is possible that as MFH is redefined, these data may change; it is important to consider MFH as a fibrosarcoma/myxosarcoma variant. However, given the diagnosis at the point of entry into this prospective database, these data provide information that should be a baseline for future critique or validation.
|
RESULTS
|
|---|
Demographics
The median patient age was 51 years, and patient age ranged from 16 to 92 years. There were 428 males (45%) and 523 females (55%). The median follow-up was 35 months (range, 2–201 months). The clinical and pathologic features of the primary tumors are presented in Table 1.
Surgical Management
Forty patients were treated with symptomatic care alone because significant comorbidities rendered them unsuitable for a surgical resection of any type. Fourteen of these patients underwent biopsy under local anesthesia to verify their histopathologic diagnosis, whereas the diagnosis in the remaining 26 patients was confirmed after review of outside pathologic specimens. The remaining 911 patients underwent surgical resection (limb-sparing resection, n = 852; amputation, n = 59).
Multivariate Analysis of Prognostic Factors
Local Recurrence
There were 137 local recurrences (14%), and the median time to local recurrence was 17 months. In the multivariate analysis, age >50 years (RR, 2.3; 95% CI, 1.1–4.4) and microscopically positive margins (RR, 2.0; 95% CI, 1.3–2.9) were significant adverse prognostic factors (Table 2). Patient sex, tumor size, tumor depth, tumor grade, and the presence of bone or neurovascular invasion were not significant prognostic factors on either univariate or multivariate analysis (Table 2). With respect to histopathologic type, malignant peripheral nerve tumor was an independent adverse prognostic factor (RR, 2.2; 95% CI, 1.3–2.9), whereas leiomyosarcoma was associated with a more favorable outcome (RR, .45; 95% CI, .2–1.0; Table 3; Fig. 1).
View this table:
[in this window]
[in a new window]
|
TABLE 3. Summary of the effect of tumor histotype on local recurrence, distant recurrence, and disease-related death
|
|
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 1. Local recurrence–free survival in 951 primary extremity sarcoma patients stratified by histopathologic subtype. MFH, malignant fibrous histiocytoma; MPNT, malignant peripheral nerve tumor.
|
|
Distant Recurrence
Two hundred thirty-seven patients (25%) developed distant metastases (Table 4). The median time to development of distant metastases was 14 months. In the Cox regression analysis, tumor size of 
5 cm remained an adverse prognostic factor for distant recurrence (intermediate lesions: RR, 2.1; 95% CI, 1.4–3.0; large lesions: RR, 4.0; 95% CI, 2.8–5.9). Deep tumors (RR, 2.4; 95% CI, 1.4–4.0) and high tumor grade (RR, 4.2; 95% CI, 2.6–6.7) also emerged as adverse prognostic factors for distant recurrence (Table 4). In terms of histopathologic subtype, leiomyosarcoma was an adverse prognostic factor (RR, 1.7; 95% CI, 1.1–2.5), whereas liposarcoma (RR, .6; 95% CI, .4–.9) and fibrosarcoma (RR, .5; 95% CI, .2–1.0) carried a more favorable prognosis (Table 3; Fig. 2). Malignant peripheral nerve tumors were not statistically associated with distant recurrence; however all 17 patients who died with this tumor developed metastatic disease.
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 2. Distant recurrence–free survival in 951 primary extremity sarcoma patients stratified by histopathologic subtype. MFH, malignant fibrous histiocytoma; MPNT, malignant peripheral nerve tumor.
|
|
Disease-Specific Survival
A total of 199 patients (22%) died of disease-related causes during the follow-up period (Table 5). In the multivariate analysis, patient age >50 years (RR, 1.4; 95% CI, 1.0–2.0), tumors deep to the investing fascia (RR, 2.0; 95% CI, 1.2–3.6), high tumor grade (RR, 3.8; 95% CI, 2.3–6.4), and microscopically positive margins (RR, 1.5; 95% CI, 1.1–2.1) were adverse prognostic factors. Tumor size was again a significant adverse prognostic factor for both intermediate (RR, 2.5; 95% CI,1.6–3.8) and large (RR, 4.6; 95% CI, 3.0–7.0) tumors. Both leiomyosarcoma (RR, 1.6; 95% CI, 1.0–2.5) and malignant peripheral nerve tumor (RR, 1.7; 95% CI, 1.0–3.1) were adverse prognostic factors, whereas liposarcoma (RR, .6; 95% CI, .4–.9) and fibrosarcoma (RR, .3; 95% CI, .1–.9) were associated with improved outcome (Table 3; Fig. 3).
View this table:
[in this window]
[in a new window]
|
TABLE 5. Analysis of disease-specific survival (DSS) in 956 patients with primary localized sarcoma of the extremity
|
|
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 3. Disease-specific survival in 951 primary extremity sarcoma patients stratified by histopathologic subtype. MFH, malignant fibrous histiocytoma; MPNT, malignant peripheral nerve tumor.
|
|
|
DISCUSSION
|
|---|
The traditional approach to the clinical evaluation of soft tissue sarcoma has emphasized tumor size, tumor grade, the presence of recurrent disease, and positive surgical margins in the assessment for the risk of recurrence and disease-specific mortality. The specific histopathologic type has been considered to be of secondary importance because it has previously been thought that individual histopathologic subtypes of comparable size and grade behave similarly. This investigation was specifically undertaken to attempt to quantify the effect of histopathologic type on the risk of local and distant recurrence, as well as disease-specific survival, in soft tissue sarcoma. For this reason, patients with extremity lesions who presented with primary localized disease were used as a study population. No patient had evidence of metastatic disease at presentation, and all patients were managed and followed up prospectively at Memorial Sloan-Kettering Cancer Center. However, this study can be criticized because not all treatments were accounted for in the analysis; we believed that any differences in outcome between patients who received chemotherapy or radiation and those who did not may not necessarily reflect the effectiveness of these therapies rather than a biological phenomenon.
The local recurrence rate in this study was 14%. and this is slightly lower than the local recurrence rate of 17% reported by Pisters et al.1 from our institution. However, that investigation included patients with both primary and locally recurrent disease, and this would result in a slightly higher local recurrence rate than in our series. Patient age >50 years was also correlated with an increased risk of local recurrence. We have previously shown that this effect of age is at least partly explained by a higher incidence of microscopic margin positivity in very old patients because the surgeon sought to avoid amputation and maintain patient functional status and independence.1 This investigation has confirmed the finding that in patients with primary sarcoma, positive surgical margins are an independent predictor of local recurrence, and this is also reflected in an increased rate of tumor-related mortality in patients with microscopically positive surgical margins.1,3,6
The development of local recurrence seemed to be more likely to occur in malignant peripheral nerve tumors. The reasons for this are unclear. It is tempting to speculate that tumors of peripheral nerves were more likely to be resected with positive margins, explaining the increased rates of local recurrence. However, of the 10 recurrences in this patient group, only 2 had pathologically positive margins, suggesting that this is not the factor responsible for this observation, although the numbers are small. The investigation of Pisters et al.1 previously demonstrated that fibrosarcoma was also an independent adverse prognostic factor for local recurrence. However, in that series, 65 of the 104 fibrosarcomas were of the desmoid subtype.1 These tumors have a distinctive propensity to recur locally,9 and when these locally aggressive lesions were excluded, fibrosarcomas would no longer be significant predictors of local recurrence.
Nearly 25% of the patients in this series developed distant recurrence; this is a similar percentage to that reported previously.1 Large, high-grade tumors that were located deep to the investing fascia were most likely to develop distant recurrence, confirming previous observations that prognostic factors for distant recurrence are different from those for local recurrence. Similarly, the histopathologic types most likely to develop distant recurrence were different from those most likely to develop local recurrence. Although leiomyosarcomas were less likely to develop local recurrence, this histopathologic type was an independent predictor of distant disease, suggesting that extremity leiomyosarcoma have a higher metastatic potential than do other histotypes. This observation may be explained by the higher propensity of deep leiomyosarcoma to metastasize and the lower risk of subcutaneous (superficial and small) leiomyosarcoma to recur locally. The propensity for metastatic recurrence is also reflected in an increased disease-specific mortality for leiomyosarcoma. Correspondingly, extremity liposarcoma and fibrosarcoma are less likely to develop local recurrence, suggesting a more indolent biology.
One hundred ninety nine patients died of disease-related causes during clinical follow-up. All patients who died had documented metastatic disease, and no patient died with only a local recurrence. This phenomenon is reflected in the observation that both leiomyosarcoma and malignant peripheral nerve tumor were independent adverse prognostic factors for disease-specific survival. Although malignant peripheral nerve tumor pathology did not prove to be associated with increased metastatic disease, this probably reflected the relatively small numbers involved, because all 17 patients who died with this tumor died with metastatic disease. Correspondingly, both extremity liposarcoma and fibrosarcoma were less likely to be fatal, emphasizing their more indolent biology. This investigation also confirms the importance of patient age, tumor size >5 cm, tumor depth, tumor grade, and surgical margin in defining outcome after the treatment of primary soft tissue sarcoma of the extremity.1
Collectively, these results represent interesting observations of the biological behavior of primary soft tissue sarcoma of the extremity. Adverse prognostic factors for local recurrence are different from those for distant metastases and disease-specific survival, although the median follow-up in this study was only 35 months. It is possible that, with more prolonged follow-up, the differences recorded may change. However, with Kaplan-Meier analysis, the prognostic implications of tumor histotype were significant and persistent over 10 years (Figs. 1–3) from initial treatment, suggesting that this reflects a true biological observation. The differences were small, and the difficulty in quantifying biological differences is shown in the broad CIs for the hazard ratios. It is also possible that the observed differences in behavior are representative of features other than histotype. For example, we have used a two-grade pathologic classification system for staging because this has been shown to be the most accurate predictor of outcome in patients with primary sarcoma of the extremity,13 although further pathologic classification of tumors by use of alternative systems may yield different results. Similarly, etiological differences may account for some of the differences in tumor behavior.14 Thus, although malignant peripheral nerve tumor may be associated with a higher rate of local recurrence and tumor-related mortality and leiomyosarcomas seem less likely to develop local recurrence but seem prone to metastatic spread and tumor-related death, these results should be interpreted with caution. However, the information presented here should be used to form the basis of further studies to define the biological nature of extremity sarcomas.
|
Footnotes
|
|---|
Presented at the 53rd Annual Cancer Symposium of the Society of Surgical Oncology, New Orleans, Louisiana, March 2000.
The effect of histopathologic type on outcome in primary soft tissue sarcoma of the extremity was investigated in 951 patients. Differences in outcome exist between different histotypes and may form the basis for further investigations.
Received for publication May 10, 2002.
Accepted for publication October 30, 2002.
|
REFERENCES
|
|---|
- Pisters PW, Leung DH, Woodruff J, Shi W, Brennan MF. Analysis of prognostic factors in 1041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol 1996; 14: 1679–89.[Abstract/Free Full Text]
- Collin C, Godbold J, Hajdu S, et al. Localized extremity soft tissue sarcoma: an analysis of factors affecting survival. J Clin Oncol 1988; 5: 601–12.
- Gaynor JJ, Tan CC, Casper ES, et al. Refinement of clinicopathologic staging for localized soft tissue sarcoma of the extremity: a study of 423 adults. J Clin Oncol 1992; 10: 1317–29.[Abstract/Free Full Text]
- Heise HW, Myers MH, Russell WO, et al. Recurrence-free survival time for surgically treated soft tissue sarcoma patients. Multivariate analysis of five prognostic factors. Cancer 1986; 57: 172–7.[CrossRef][Medline]
- Rydholm A, Berg NO, Gullberg B, et al. Prognosis for soft tissue sarcoma in the locomotor system. A retrospective population-based follow-up study of 237 patients. Acta Pathol Microbiol Immunol Scand 1984; 92: 375–86.
- Singer S, Corson JM, Gonin R, et al. Prognostic factors predictive of survival and local recurrence for extremity soft tissue sarcoma. Ann Surg 1995; 219: 165–73.
- Markhede G, Angervall L, Stener B. A multivariate analysis of the prognosis after surgical treatment of malignant soft-tissue tumors. Int J Cancer 1982; 3: 1721–33.
- Mandard AM, Petiot JF, Marnay J, et al. Prognostic factors in soft tissue sarcoma. A multivariate analysis of 109 cases. Cancer 1989; 63: 1427–51.
- Chang HR, Gaynor J, Tan C, Hajdu SI, Brennan MF. Multifactorial analysis of survival in primary extremity sarcoma. World J Surg 1990; 14: 610–8.[CrossRef][Medline]
- Merchant NB, Lewis JJ, Woodruff JM, Leung DH, Brennan MF. Extremity and trunk desmoid tumors: a multifactorial analysis of outcome. Cancer 1999; 86: 2045–52.[CrossRef][Medline]
- Hajdu SI, Shu MH, Brennan MF. The role of the pathologist in the management of soft tissue sarcomas. World J Surg 1988; 12: 326–31.[CrossRef][Medline]
- Cox D. Regression models and life tables (with discussion). J R Stat Soc 1972; 34: 187–220.
- Wunder JS, Healey JH, Davis AM, Brennan MF. A comparison of staging systems for localized extremity soft tissue sarcoma. Cancer 2000; 88: 2721–30.[CrossRef][Medline]
- Ferner RE, Gutmann DH. International consensus statement on malignant peripheral nerve sheath tumors in neurofibromatosis. Cancer Res 2002; 62: 1573–7.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
A. Sternheim, X. Jin, B. Shmookler, J. Jelinek, and M. M. Malawer
'Telangiectatic' Transformation in Soft Tissue Sarcomas. A Clinicopathology Analysis of an Aggressive Feature of High-grade Sarcomas
Ann. Surg. Oncol.,
January 1, 2008;
15(1):
345 - 354.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Ferrari, M. Casanova, P. Collini, C. Meazza, R. Luksch, M. Massimino, G. Cefalo, M. Terenziani, F. Spreafico, S. Catania, et al.
Adult-Type Soft Tissue Sarcomas in Pediatric-Age Patients: Experience at the Istituto Nazionale Tumori in Milan
J. Clin. Oncol.,
June 20, 2005;
23(18):
4021 - 4030.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Y.-F. Lee, M. John, A. Falconer, S. Edwards, J. Clark, P. Flohr, T. Roe, R. Wang, J. Shipley, R. J. Grimer, et al.
A Gene Expression Signature Associated with Metastatic Outcome in Human Leiomyosarcomas
Cancer Res.,
October 15, 2004;
64(20):
7201 - 7204.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
