Low-grade soft tissue sarcomas of the extremities. Analysis of risk factors for metastasis

The Feasibility Study of Hypofractionated Radiotherapy with Regional Hyperthermia in Soft Tissue Sarcomas

Simple Summary

The recommended management of marginally resectable or unresectable soft tissue sarcomas is an attempt of neoadjuvant therapy. The use of neoadjuvant chemotherapy is limited in low-grade tumors, sarcomas with chemoresistant pathology or in unfit patients. There is a growing evidence on hypofractionated radiotherapy in soft tissue sarcomas, but its efficacy may be limited by radioresistance that is frequently associated with chemoresistance. Regional hyperthermia is a potent and minimally invasive radiosensitizer. We aimed to investigate the feasibility of moderately hypofractionated radiotherapy combined with regional hyperthermia in aforementioned clinical situations. Our findings indicate that proposed combination is feasible while maintaining good short-term local efficacy and tolerance. It could serve as a basis for further studies on radiotherapy with hyperthermia in soft tissue sarcomas.

Abstract

Introduction: Management of marginally resectable or unresectable soft tissue sarcomas (STS) in patients who are not candidates for neoadjuvant chemotherapy due to chemoresistant pathology or contraindications remains a challenge. Therefore, in these indications, we aimed to investigate a feasibility of 10x 3.25 Gy radiotherapy combined with regional hyperthermia (HT) that could be followed by surgery or 4x 4 Gy radiotherapy with HT. Materials and methods: We recruited patients with locally advanced marginally resectable or unresectable STS who (1) presented chemoresistant STS subtype, or (2) progressed after neoadjuvant chemotherapy, or (3) were unfit for chemotherapy. The primary endpoint was the feasibility of the proposed regimen. Results: Thirty patients were enrolled. All patients received the first part of the treatment, namely radiotherapy with HT. Among them, 14 received the second part of radiotherapy with HT whereas 13 patients underwent surgery. Three patients did not complete the treatment protocol. The feasibility criteria were fulfilled in 90% of patients. Two patients developed distant metastases. One patient died due to distant progression. One patient developed rapid local recurrence after surgery. Conclusions: Hypofractionated radiotherapy with HT is a feasible treatment for marginally resectable or unresectable STS in patients who are not candidates for chemotherapy. Results of this clinical trial support the further validation of RT and HT combinations in STS.

Why do patients with low-grade soft tissue sarcoma die?

INTRODUCTION

The patterns of failure and mechanisms of sarcoma-specific death are poorly characterized among the minority of patients with low-grade soft tissue sarcoma (STS) who succumb to disease.

METHODS

Between 1982 and 2006, 2,041 patients aged >or=16 years with low-grade STS of all sites were treated with curative intent and prospectively followed at a single institution.

RESULTS

Among this cohort, 181 (9%) died from disease (DOD). Overall, 105 patients (58%) died from locally recurrent disease (DOLR), and 59 (32%) died from distant disease (DODR). In 17 patients (9%), the mechanism of sarcoma-related death could not be verified. DOD occurred at a median of 62 months, while median disease-specific survival for the entire cohort was not reached. Median follow-up was 66 months (range 2-431 months). On multivariate analysis, DOD was associated with site, size, and less than R0 resection. For DOLR, site, size, positive margins, liposarcoma histology, and local recurrence (by definition) were significant factors. For DODR, site, histology, and positive margins were not significant factors, while size and local recurrence were. Of DOLR, 80% were retroperitoneal, 68% were liposarcoma, and only 2% were extremity. Conversely, of DODR, extremity (47%) and trunk (18%) were the most common sites, but histology was more variable [liposarcoma 35%, malignant fibrous histiocytoma (MFH) 20%, fibrosarcoma 12%, extraskeletal myxoid chondrosarcoma 10%]. High-grade recurrence rates were comparable between DOLR (27%) and DODR (25%).

CONCLUSION

Among patients with low-grade STS, DOD occurs in approximately 9% of patients. Nonextremity site, larger size, and less than R0 resection are the most important risk factors for DOD, and distinct patterns of recurrence and death are predicted by primary tumor site.

Clinical features and outcome of initially unresected nonmetastatic pediatric nonrhabdomyosarcoma soft tissue sarcoma

PURPOSE

To describe the clinical features, response to therapy, and outcome of pediatric patients with initially unresected nonmetastatic nonrhabdomyosarcoma soft tissue sarcoma (NRSTS).

PATIENTS AND METHODS

We retrospectively reviewed the presenting clinical features and tumor characteristics of all 40 pediatric patients with initially unresected nonmetastatic NRSTS who were seen at our institution between March 1962 and December 1996. A subset of 27 patients for whom complete treatment information was available was analyzed to determine whether response to therapy was associated with local disease control and event-free and overall survival.

RESULTS

More than 70% of the 40 patients had tumors with high-risk features (tumor size > 5 cm, high grade, invasiveness). For the 27 patients included in the outcome analysis, 5-year event-free survival and survival estimates were 33% +/- 9% and 56% +/- 10%, respectively. Ten (37%) of these patients had a complete or partial response to neoadjuvant chemotherapy and/or radiotherapy, and only two of the 10 had residual tumor after surgery. Combined chemotherapy and radiotherapy seemed more effective than either modality alone in inducing a response, but the response to neoadjuvant therapy did not predict outcome. Most treatment failures were local, and postrelapse survival was poor (19% +/- 10%).

CONCLUSION

Initially unresected NRSTS constitutes a unique subgroup of pediatric sarcomas that commonly present with high-risk features and respond poorly to neoadjuvant therapy. Only about one third of patients treated with multimodal therapy remain disease-free, and local control is the major limiting factor in achieving cure. More effective risk-directed treatments are needed for this unique subgroup of patients.

Soft tissue sarcoma of the hand

A retrospective review of our institute's tumor registry from January 1972 to January 1996 revealed 24 patients with a diagnosis of primary soft tissue sarcoma of the hand, from a total of 570 extremity soft tissue sarcomas (4%). The most frequent histologic type was malignant fibrous histiocytoma, which occurred in 9 (38%) of the 24 patients. The second most common histologic type was epithelioid sarcoma, which occurred in 6 (25%) patients. There was a statistically significant difference in the rate of local recurrence based on the type of treatment in which amputation was superior to the other forms of treatment. There was no statistically significant difference in the rate of distant failures between treatment groups. The estimated cumulative 5- and 10-year overall survival rates for all patients were 59% and 53%, respectively. Stage II patients had estimated cumulative 5- and 10-year survival rates of 68% and 59%, respectively. Stage III patients had a cumulative 5-year survival rate of 20%. Factors that were statistically significant in predicting survival were the size of the primary tumor, with tumors smaller than 5 cm having a better prognosis, and stage of the tumor at presentation, with stage I and II tumors having the highest survival rate. In selected patients with a primary hand sarcoma, aggressive limb-sparing surgery with adjuvant therapy offered equivalent survival compared with amputation.

Soft tissue sarcoma of the upper extremity: descriptive data and outcome in a population-based series of 108 adult patients

We report a population-based, 29-year review of 108 adult patients with soft tissue sarcoma of the upper extremity who had a minimum follow-up period of 3 years. Fifty tumors were localized in the upper arm, 8 in the elbow, 40 in the forearm, and 10 in the hand. Thirty-four tumors were subcutaneous and 74 were deep-seated. The median tumor size was 6 cm. Malignant fibrous histiocytoma was the most common histotype (n = 43), and grade IV (on a 4-grade scale) was the most common malignancy grade (n = 54). All but 6 patients underwent surgery; limb-sparing surgery of the primary tumor was performed in 89 patients. Twenty-four patients were given adjuvant radiotherapy; 11 received adjuvant chemotherapy. Inadequate local treatment was more common in patients treated outside the tumor center. Local recurrence occurred in 15 of 28 with inadequate local treatment and in 16 of 74 patients with adequate local treatment (20 of the 39 patients treated outside the center and 11 of the 63 patients treated at the center). At the latest follow-up visit, 32 patients had developed metastases, giving a 5-year metastasis-free survival rate of 0.72. In a multivariate analysis, tumor size larger than 5 cm and vascular invasion emerged as independent prognostic factors. Patients without these 2 factors had excellent survival. When compared with soft tissue sarcoma of the lower extremity or trunk wall, tumors in the upper extremity were smaller at the time of diagnosis and had a higher 5-year metastasis-free survival rate.

Modified staging system for extremity soft tissue sarcomas

BACKGROUND

The establishment of a universally acceptable staging system for soft tissue sarcomas has been hampered by the low incidence, various grading systems, and lack of consensus regarding the value of different prognostic factors. We aimed to evaluate prognostic factors in patients with extremity soft tissue sarcomas and to test the validity of the AJCC/UICC staging system.

METHODS

Prospectively collected data from 316 previously untreated patients with primary extremity soft tissue sarcomas treated at a single institution between 1989 and 1995 were studied. The influence of clinical and pathological factors on local recurrence, distant metastasis, and disease-specific survival was analyzed by univariate and multivariate techniques.

RESULTS

Large tumor size and high histological grade were independent adverse prognostic factors for distant metastasis. Large size, high grade, and positive microscopic surgical margins were independent adverse prognostic factors, and liposarcoma histology was an independent favorable prognostic factor for disease-specific survival. Within each histological grade, there was a progressive decline in survival with increasing tumor size, as reflected by an almost linear increase in hazard ratios. Similarly, there was a progressive fall in survival with increasing grade within each size group (<5 cm, 5 to 10 cm, 10 to 15 cm, and > 15 cm). AJCC staging did not correlate well with prognosis. Survival for intermediate-grade tumors smaller than 5 cm (stage IIA) was better than that for low-grade tumors larger than 5 cm (stage IB) (86% vs. 73%). Survival for high-grade tumors smaller than 5 cm (stage IIIA) was better than that for intermediate-grade tumors larger than 5 cm (stage IIB) (72% vs. 57%). A modified staging system was formulated based on the additive influence of size and grade on the estimated hazard ratios for disease-specific survival, as follows: stage IA, G1T1; stage IB, G1T2 or G2T1; stage IIA, G1T3 or G2T2 or G3T1; stage IIB, G1T4 or G2T3 or G3T2; stage IIIA, G2T4 or G3T3; stage IIIB, G3T4; and stage IV, M1 (G1, G2, G3 = low, intermediate, and high grade; T1, T2, T3, T4 = tumor size < 5 cm, 5-10 cm, 10-15 cm, and > 15 cm, respectively). The 5-year disease-specific survivals of stages IA, IB, IIA, IIB, IIIA, and IIIB were 100%, 83%, 74%, 61%, 39%, and 18%, respectively. The 5-year disease-specific survival for stages I, II, III, and IV were 90%, 67%, 31%, and 6% respectively. The survival difference between each stage was statistically significant (P < .001).

CONCLUSION

Histological grade and tumor size are equally important determinants of distant metastases and survival. The AJCC/UICC staging system is based primarily on the grade of the tumor, with size used to subgroup each stage. A staging system for extremity soft tissue sarcomas with equal emphasis on grade and size is proposed that correlates extremely well with prognosis.

Resection of brain metastases from sarcoma

BACKGROUND

Brain metastases from sarcoma are rare, and data concerning the treatment and results of therapy are sparse.

METHODS

We retrospectively reviewed 25 patients with brain metastases from sarcoma of skeletal or soft-tissue origin, surgically treated in a single institution during 20 years.

RESULTS

In 18 patients the brain lesion was located supratentorially, and in 7 patients infratentorially. Median age at brain metastasis diagnosis was 25 years. Median time from primary diagnosis to diagnosis of brain metastasis was 26.7 months. Lung metastases were present in 19 patients and in 8 patients they were synchronous with the brain lesion. Pulmonary metastases were resected in 12 patients (48% of total, and 63% of those with pulmonary lesions). The overall median survival from diagnosis of the primary sarcoma was 38 months and from craniotomy was 7 months. The presence or absence of lung lesions did not alter the median survival as calculated from diagnosis of brain metastasis. Overall percent survival was 40% at 1 year and 16% at 2 years.

CONCLUSIONS

Because brain metastases from sarcoma are refractory to alternative treatment, surgical excision is indicated when feasible. Brain metastases from sarcoma are uncommon, usually occurring with or after lung metastasis. Long-term survival is possible in some patients.

Treatment of soft-tissue sarcomas of the hand

We studied the clinical features, radiographic and pathological findings, treatment, and results for twenty-three patients who had been managed for a soft-tissue sarcoma of the hand between 1982 and 1990. The ages of the patients ranged from sixteen to seventy-six years (median age, thirty-one years). The most common clinical finding was a small, painless soft-tissue mass. Twenty of the tumors were high-grade, and eighteen were less than five centimeters in diameter. The most common diagnosis was synovial sarcoma, which was identified in eight patients. Leiomyosarcoma, rhabdomyosarcoma, and malignant fibrous histiocytoma developed in three patients each; epithelioid sarcoma, in two patients; and angiosarcoma, liposarcoma, neuroectodermal tumor, and clear-cell sarcoma, in one patient each. Curative wide excision or amputation was attempted in twenty-two patients; the margins were positive for tumor cells in eight, and local recurrence was seen in nine. Of the twenty-three patients, fourteen had survived, without evidence of disease, after a median duration of follow-up of forty-nine months, and nine had died of disease. The median rate of survival did not differ significantly on the basis of the size or grade of the tumor or the use of adjuvant treatment. However, the rate of survival of the patients who had a soft-tissue sarcoma of the hand that was less than five centimeters in diameter was significantly lower (p = 0.0008) than that of 152 patients who had a similar tumor at another site in an extremity.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparison of charges related to radiotherapy for soft-tissue sarcomas treated by preoperative external-beam irradiation versus interstitial implantation

BACKGROUND

We compared treatment-related charges associated with external beam irradiation and interstitial implantation for soft-tissue sarcoma of the extremity.

METHODS

Charges related to radiotherapy in 35 patients with soft-tissue sarcoma of the extremity were reviewed. Preoperative external beam irradiation (EB) delivering 50 Gy in 25 fractions with 6 MV photons was administered to 12 of the patients evaluated. The remaining 23 patients were treated with interstitial implantation (IR) as the only radiotherapeutic intervention. The anatomic distribution of the sarcomas treated by IR included 14 lower-extremity (LE) and nine upper-extremity (UE) lesions. The average length of iridium wire used for IR was 78 cm. Because LE lesions tend to be larger, the average length equaled 109.5 cm as compared with the 47 cm for UE implants.

RESULTS

The radiotherapeutic approach represented the only difference in treatment-related charges because the operative procedure of wide local excision was performed in each group. No difference in perioperative complications was observed between the two treatment approaches. Charges were stratified according to hospital-based and professional services. Radiotherapy-based hospital charges for the administration of EB averaged $6,515 compared with $4,050 for IR (p < 0.0001). Professional services also were significantly different, totaling $4,390 for EB and $3,240 for IR (p < 0.0001). The total of these charges for radiotherapy procedures and professional fees equaled $10,905 for EB compared with $7,290 for IR (p < 0.0001). Incorporating the necessary operating-room time for implant placement ($750) and five additional hospital days ($1,800), the costs associated with IR totaled $9,840; using chi-square analysis, the cost for IR remained significantly (p < 0.0001) less expensive than the $10,905 associated with EB. Because a large component of the radiotherapy cost for IR is related to the length of iridium 192 wire required, charges were stratified according to the location of the tumor. The total charge for IR of the UE equaled $9,345 compared with $10,335 for LE implants. Chi-square comparison for both UE and LE implants continued to show significant differences (p < 0.0001) when related to EB therapy.

CONCLUSION

Cost-analysis comparison of brachytherapy versus external beam irradiation found lower charges for patients undergoing adjuvant irradiation with brachytherapy for soft-tissue sarcoma. To optimize the cost-benefit ratio, prospective studies are necessary to define the application of these radiotherapeutic approaches based on clinical criteria.

Image analysis derived ploidy and proliferation indices in soft tissue sarcomas: comparison with clinical outcome

AIMS

To compare prognostic information obtained by image analysis cytometry of paraffin wax embedded soft tissue sarcomas with conventional assessment.

METHODS

A CAS 200 image analyser was used to determine DNA content of Feulgen stained cytology preparations and tissue sections and to quantify immunostaining by Ki67 and PC10 antibodies. A mitotic count in 50 high power fields was undertaken and histological grade assigned by the Trojani system. Clinical details including follow up and outcome were obtained by case note review. The Kruskal-Wallis one way analysis test, Spearman rho significance test, Kaplan-Meier method, and log-rank test were applied in statistical analysis.

RESULTS

Ploidy status, DNA index, 2.5c exceeding rate, 5c exceeding rate, mitotic count and Trojani grade all correlated significantly with clinical outcome. The relation between Ki67 index and outcome did not reach significance. The PC10 index and outcome were not related. Only 2.5c exceeding rate, 5c exceeding rate, and mitotic count correlated significantly with Trojani grade.

CONCLUSIONS

DNA content determination of soft tissue sarcomas by image analysis provides quantifiable information of benefit in prediction of outcome. Larger series are required to determine the independent value of ploidy. In this study quantification of anti-Ki67 and anti-PC10 immunostaining was not of prognostic benefit) by contrast with mitotic count and Trojani grade.

Limb-sparing treatment with surgery and intraoperative radiotherapy (IORT) for a second local recurrence of myxoid liposarcoma in the popliteal region, after previous surgery and high-dose radiation

A 46-year-old man with a second local recurrence of a myxoid liposarcoma in the fossa poplitea, after surgery and high-dose external beam radiotherapy, was successfully treated by a limb-sparing procedure combining marginal excision, intraoperative radiotherapy and reconstructive surgery. After a 48-month follow-up, there was no evidence of disease, and the function of the leg and knee was normal.

Does malignant transformation of benign soft-tissue tumours occur? A clinicomorphological study of ten initially misdiagnosed soft-tissue sarcomas

Ten soft-tissue sarcomas out of a consecutive series of 201 neoplasms are described, in which the clinical histories were suggestive of malignant transformation of a benign lesion excised from the same location 2-97 months before. Each of the 10 preceding soft-tissue neoplasms was reviewed histologically and reclassified as either highly (n = 8) or moderately differentiated (n = 2) sarcomas. Irrespective of the primary treatment, these initially misdiagnosed tumours showed an overall better prognosis than the whole group of soft-tissue sarcomas (5- and 10-year survival rates 76% versus 44%, and 52% versus 30% respectively). Their clinical course was, however, characterized by up to 6 (mean 2.9) local recurrences. From the total series of 201 tumours the authors conclude that soft-tissue sarcomas other than malignant schwannomas in the setting of von Recklinghausen's disease only exceptionally, if ever, arise from benign precursors.

The role of multimodality therapy in soft-tissue sarcoma

Soft-tissue sarcomas are uncommon malignancies. The development during a period of 8 years, in one institution, of a prospective data base incorporating more than 1600 patients with these tumors is described. The most common sites for occurrence are the extremities, but they can occur in any of the soft tissues of the body. Liposarcoma and leiomyosarcoma are the most common histopathologic conditions identified. Prognostic factors for both recurrence and survival include site, histopathology, size, grade, and adequacy of resection. A prospective randomized trial of the use of adjuvant radiation by the brachytherapy technique in extremity lesions has shown a decrease in local recurrence, but no impact on survival. Eligible patients not randomized to the trial show no difference in local recurrence or survival, regardless of whether they received adjuvant radiation.

Soft tissue sarcomas of the head and neck in adults

We reviewed the clinical records and pathologic material of 176 adults with primary soft tissue sarcomas treated at Memorial Sloan-Kettering Cancer Center between 1950 and 1985. Seventy-two patients (41%) had low-grade sarcomas and 104 (59%) had high-grade sarcomas. All but 18 patients underwent some form of excision as initial therapy. Adjuvant radiotherapy and chemotherapy combined with surgical excision showed no significant effect. A significantly increased risk of treatment failure was associated with large tumor size, positive surgical margins, bone involvement, local recurrence, metastatic spread, and high histologic grade. Except for recurrence, the p value by univariate analysis in the log-rank test for comparison of survival according to these clinical and pathologic characteristics was p less than 0.0001. Although the overall survival was 75% at 2 years, 55% at 5 years, and 46% at 10 years, only 20% of the patients with high-grade sarcomas were alive 10 years after treatment. Most patients with rhabdomyosarcoma, high-grade peripheral nerve tumor, and high-grade fibrous histiocytoma and all patients with high-grade angiosarcoma died of disease less than 5 years after diagnosis. New therapeutic strategies are needed to improve the survival of adult patients with high-grade soft tissue sarcomas of the head and neck.

Multifactorial analysis of survival in primary extremity liposarcoma

The treatment of extremity soft tissue sarcoma is now directed at limb preservation with the addition of various adjuvant therapies to improve treatment results. To achieve this goal, a knowledge of prognostic factors for extremity soft tissue sarcoma becomes increasingly critical. The object of this study was to analyze prognostic factors for survival in patients with extremity liposarcoma. Eighty-three patients with primary localized extremity liposarcoma, admitted from 1968 to 1978, were retrospectively reviewed. Surgical resection was the primary mode of treatment. Eleven prognostic factors were analyzed. Tumor factors included: histologic subtype, tumor grade, size, depth, invasion of vital structures, and site; operative factors included: type of operation, and surgical margins; and patient factors included: symptoms, age at diagnosis, and sex. Kaplan-Meier survival curves, and univariate and stratified log-rank tests of association were performed. Independent factors for predicting survival were identified using the Cox model stepwise regression technique. In univariate analysis of the entire group of patients, 5 factors were significant for tumor mortality: tumor grade (p = 0.00005), histologic subtype (p = 0.00025), tumor size greater than or equal to 5 cm (p = 0.005), type of surgery/margin (p = 0.0001), and invasion of vital structures (p = 0.008). When associations among all factors were analyzed, it was found that many factors were interdependent. The independent prognosticators were, therefore, determined according to the Cox model technique. For the multivariate analysis, well-differentiated and lipoblastic liposarcomas were excluded because of lack of survival variation within each group. The former group had no tumor-related deaths and the latter group showed 80% tumor mortality.(ABSTRACT TRUNCATED AT 250 WORDS)

Extremity soft tissue sarcomas: analysis of prognostic variables in 300 cases and evaluation of tumor necrosis as a factor in stratifying higher-grade sarcomas

From July 1975 to February 1985, 300 patients with extremity soft tissue sarcomas (STS) were entered into treatment protocols at the National Cancer Institute. Using the Cox proportional hazards model, the STS were analyzed for independent prognostic variables that had significant association with disease-free and overall survival. The effects of adjuvant treatment were also included in the analysis. Percent tumor necrosis was the basis for separating intermediate and high-grade STS (less than 15% necrosis = grade 2; greater than 15% necrosis = grade 3). The most important prognostic variable was histologic grade, which was highly associated with relative risk of metastasis and fatal outcome (P less than .0001). Histologic type of STS was not a significant prognostic variable in predicting metastases. Patients with tumors located on the proximal or middle part of the extremity had a fivefold increased risk of death (P = .02) compared with patients who had STS located more distally. Mitotic activity (P = .0005) and vascular invasion (P = .0034) were also associated with malignant behavior, but histologic grade was the most significant predictor of outcome. Histologic grade of malignancy was based upon a combination of tumor typing and visual estimation of extent of tumor necrosis; using these two features, one can reliably separate low-grade (i.e., grade 1) from high-grade STS and also stratify the more aggressive (i.e., grade 2 and 3) tumors. Over the 9.5-year period of study, the tumor-related mortality was as follows: grade 1 (1/38), 3%; grade 2 (19/142), 13%; and grade 3 (45/120), 38%.

Management of extremity soft-tissue sarcoma

Soft-tissue sarcomas remain relatively uncommon neoplasms, but they present major problems for both the patient and clinician. The identification of prognostic variables based on both clinical and histologic factors has allowed comparisons to be made between various therapeutic regimens. In extremity sarcoma, a more conservative operation utilizing adjuvant radiotherapy has decreased the need for radical surgery and maximized function without loss of survival benefit. The identification of high-risk groups, that is, patients with high-grade, large lesions, has highlighted the inadequacies of treatment in some areas. Similarly, this has led to more aggressive and innovative developmental programs. The progress, understanding, and therapeutic management of extremity lesions have not been matched in visceral and retroperitoneal tumors. Current efforts are directed at more appropriate use of radical surgery, maximizing conservative function-sparing operations with the use of adjuvant local therapy, and development of new regimens directed at prevention and treatment of metastatic disease.

Prognostic factors in soft tissue sarcomas. A multivariate analysis of 109 cases

Prognostic factors were evaluated in 109 soft tissue sarcomas of the extremities, walls of the trunk, head, and neck. All lesions were graded according to the systems proposed by the National Cancer Institute (NCI) and the French Federation of Cancer Centers (FNCLCC), and a correlation was found between tumor grade and prognosis. Univariate analysis selected the following variables as unfavorable prognostic factors: invasive tumor margins, extra-compartmental status, deep tumors, tumor diameters over 5 cm, inadequate excision, presence of necrosis, high mitotic count, histologically undifferentiated tumors, and blood vessel invasion. These variables were found to be interdependent. Multivariate analysis selected quality of surgery as the most important variable for predicting local recurrences. The factors selected with regard to overall and metastasis-free survival were tumor size, tumor margins, necrosis, and adequacy of excision. These results permitted classification of patients into four prognostic groups: two with good and two with bad prognosis. Five-year survival for the four groups was 100%, 83%, 53%, and 0%; 5-year metastatic rates were 0%, 12%, 67%, and 100%. Similar groups were obtained when the variables of tumor margins and size were combined with an adaptation of the NCI grading (low-grade tumors/high-grade tumors without necrosis/high-grade tumors with necrosis). Comparative analysis showed that patients with tumors of the same histologic grade or type were not necessarily classed in the same prognostic groups. A better clinicopathologic correlation was obtained using a combination of prognostic factors than with histologic grading or typing alone.