Prognostic Significance of Grading and Staging Systems using MIB-1 Score in Adult Patients with Soft Tissue Sarcoma of the Extremities and Trunk

843 Prognostic Significance of Grading and Staging Systems using MIB-1 Score in Adult Patients with Soft Tissue Sarcoma of the Extremities and Trunk Tadashi Hasegawa, M.D. 1 Seiichiro Yamamoto, Ph.D. 2 Ryohei Yokoyama, M.D. 3 Toru Umeda, M.D. 3 Yoshihiro Matsuno, M.D. 4 Setsuo Hirohashi, M.D. 1 1 Pathology Division, National Cancer Center Research Institute, Tokyo, Japan. 2 Cancer Information and Epidemiology Division, National Cancer Center Research Institute, Tokyo, Japan. 3 Orthopedic Division, National Cancer Center Hospital, Tokyo, Japan. 4 Clinical Laboratory Division, National Cancer Center Hospital, Tokyo, Japan. BACKGROUND. The predictive value of histologic grading and staging systems for overall survival in different types of adult soft tissue sarcoma of the extremities and trunk is unclear. METHODS. Histologic slides from 193 patients with primary tumors were reviewed for diagnosis, and Ki-67 (MIB-1) immunostaining was performed for grading in all patients. Univariate and multivariate analyses were conducted to analyze the results from patients with soft tissue sarcomas as a group and among the six main histologic categories: malignant fibrous histiocytoma (n 49 patients), liposarcoma (n 48 patients), synovial sarcoma (n 30 patients), spindle cell sarcoma (n 24 patients), small round cell sarcoma (n 15 patients), and others (n 27 patients). The median follow-up was 50 months. RESULTS. Univariate analysis of soft tissue sarcomas showed that tumor size and depth, histologic type, MIB-1 score, grades based on three criteria (tumor differentiation/histologic type, necrosis, and either mitosis or MIB-1 score) and disease stage, as assessed by tumor size, depth, and grade, were associated with overall survival. Among these variables, grading and staging systems using the MIB-1 score had better predictive value compared with the MIB-1 score and standard grading and staging models in the main histologic categories. Because survival curves for the different tumor grades and stages showed similar trends between the different histologic types, multivariate analysis was conducted adjusting for age, gender, disease site, surgical margin, tumor size and depth, grade, stage, and histologic type. Consequently, Grade 3 emerged as the most significant adverse prognostic factor. Additional adverse prognostic factors were Stage III, Grade 2, a histologic type of spindle cell sarcoma, and patient age 50 years at the time of presentation. The histologic type liposarcoma was identified as a favorable prognostic factor. CONCLUSIONS. The current results indicate that grading and staging systems using the MIB-1 score are very strong prognostic factors in patients with the main histologic types of soft tissue sarcoma. Specific assessment of histologic type should be carried out before deciding on treatment strategies. Cancer 2002;95: 843 51. 2002 American Cancer Society. DOI 10.1002/cncr.10728 KEYWORDS: soft tissue sarcoma, adult, prognosis, MIB-1, histologic type, grade, stage. Address for reprints: Tadashi Hasegawa, M.D., Pathology Division, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan; Fax: 81-3-3248-2463; E-mail: tdhasega@gan2.ncc.go.jp. Received September 24, 2001; revision received March 12, 2002; accepted March 21, 2002. Soft tissue sarcomas in adults are rare malignant tumors. Better histologic diagnosis and grading are essential for correct treatment decisions and improved patient outcome. Among several different grading schemes for patients with sarcomas, 1,2 a grading system has been proposed based on three criteria: tumor differentiation/ histologic type, necrosis, and Ki-67 (MIB-1) score. 3 This system 2002 American Cancer Society

844 CANCER August 15, 2002 / Volume 95 / Number 4 replaces the mitotic count of the French Federation of Cancer Centers (FNCLCC) Sarcoma Group grading system 2 with an MIB-1 labeling index that, in a multivariate analysis, was the most significant independent prognostic factor for patients with soft tissue sarcomas overall as well as for patients with some specific histologic types, such as retroperitoneal liposarcoma and synovial sarcoma. 4,5 In a previous study, the validity and reproducibility of this system, when it was tested by four pathologists, was high for the histologic diagnosis of synovial sarcoma, small round cell sarcoma, and liposarcoma, substantial for grading, and higher compared with the FNCLCC grading system. 6 The pathologic TNM (ptnm) staging system, proposed by the American Joint Committee on Cancer (AJCC) and the International Union Against Cancer (UICC), is the standard sarcoma staging system and consists of the removal and pathologic evaluation of the primary tumor and regional lymph nodes or distant metastases. The tumor size and depth, which have recently been incorporated into the modified AJCC/UICC staging system, 7 and the grade are major variables in the staging system. The incorporation of histologic grade into the stage groups is unique to the staging classification of soft tissue sarcoma and is distinct from that of other carcinomas, because the biologic behavior of soft tissue sarcomas depends mainly on the histologic grade. In addition, the grade itself is based on the tumor differentiation score according to the histologic type of soft tissue sarcoma, which was the most valid and reproducible parameter among the three components of the grading system. 6 The prognostic significance of histologic type and grading and staging systems for adult patients with soft tissue sarcomas of the extremities who were treated at a single institution were investigated previously in a large series. 8,9 However, soft tissue sarcomas in adults show great variations in their histologic type and grade. 10,11 The estimated range of grade and expected biologic behavior for each histologic type are so variable that the ability of the grade and stage to predict the patient s prognosis within each histologic type of soft tissue sarcoma was investigated in this study. MATERIALS AND METHODS Patients Records of 193 patients with soft tissue sarcomas of the extremities and trunk who were diagnosed and treated between January, 1975 and December, 1998 were retrieved from the pathology files of the National Cancer Center (NCC), Tokyo. Lesions of the trunk included superficial, chest, or body wall lesions. Patients with distant metastases or regional lymph node involvement at the time of diagnosis (Stage IV) were excluded from this analysis. There was no difference in the survival rate between patients who were treated with adjuvant chemotherapy or radiation therapy compared with patients who were treated with surgery alone, as described previously. 3 The clinical details, including follow-up information, were obtained by reviewing all medical charts. No patients were lost to follow-up. Follow-up began on the date of primary surgery. The median follow-up was 50 months. Overall survival was recorded as the time to death due to any cause. Pathology Review, Grading, and Staging Histologic slides of primary tumors from all patients were reviewed for diagnosis by two experts (T.H. and Y.M.) at the NCC who had developed the grading system (Japanese grading system). Whenever necessary, immunohistochemistry was used for the confirmation of diagnosis or tumor typing according to the classification system described by Enzinger and Weiss. 11 MIB-1 immunostaining was performed for the grading of all tumors. In this study, soft tissue sarcomas were classified into six histologic types: malignant fibrous histiocytoma (MFH), liposarcoma, synovial sarcoma, spindle cell sarcoma, small round cell sarcoma, and others. The histologic grade is a threegrade system obtained by adding the scores for tumor differentiation, tumor necrosis, and MIB-1 score, each of which is given a score of 0, 1, 2, or 3. 6 An MIB-1 score of 1 is assigned to lesions with an MIB-1 labeling index (LI) of 0 9%, an MIB-1 score of 2 is assigned to lesions with an MIB-1 LI of 10 29%, and an MIB-1 score of 3 is assigned to lesions with an MIB-1 LI 30%. The standard (FNCLCC) grading system also was used in this study. According to the description in the AJCC Cancer Staging Manual, 7 histologic Grade 3 and 4 tumors can be combined into a single grade (poorly differentiated to undifferentiated; Grade 3 4). Therefore, Grade 1 (well differentiated) and Grade 2 (moderately differentiated) tumors that were assessed by using the grading systems were grouped as low grade, and Grade 3 tumors were high grade in this staging system. Stages I and II lesions are subdivided into five categories: Stage I lesions are either low grade and small ( 5 cm; Stage IA), or low grade and large ( 5 cm) and superficial (Stage IB); Stage II lesions are either low grade, large, and deep (Stage IIA), high grade and small (Stage IIB), or high grade, large, and superficial (Stage IIC). Stage III tumors are high grade, large, and deep. A three-tiered staging system was used in accordance with the grading system in this study. To define depth,

Grade and Stage in Soft Tissue Sarcoma/Hasegawa et al. 845 superficial lesions did not involve the superficial fascia, and deep lesions were deep to or invaded the superficial fascia. Statistical Analysis The following parameters were considered for their prognostic value: age at presentation, gender, anatomic site, tumor size, depth, surgical margin, histologic type, MIB-1 score, grade, and stage. Univariate analysis was performed by comparing Kaplan Meier survival curves and log-rank tests from all 193 patients with soft tissue sarcoma and each histologic type. The relative risk (RR) of each variable was estimated by with a Cox proportional hazards model in univariate and multivariate analyses. If there was no interaction between each histologic type and grade or stage, then a multivariate analysis that included histologic type as a covariate was conducted with variable selection by backward elimination ( 0.2). All analyses were conducted using SAS Software (version 6.12; SAS Institute, Cary, NC). RESULTS Patients and Tumor Characteristics Of 193 patients with soft tissue sarcomas, 109 patients were male (56.5%), and 84 patients were female (43.5%) (Table 1). The median age at the time of presentation was 47 years (range, 15 87 years), and 86 patients (44.6%) were age 50 years. Tumor locations included the extremities in 130 patients (67.4%) and the trunk in 63 patients (32.6%). The median tumor size was 7 cm, and 98 tumors (50.8%) measured 7. One hundred forty-six tumors (75.6%) were situated deeply, and 47 tumors (24.4%) were superficial. Surgical procedures consisted of wide excision, amputation, or disarticulation with adequate surgical margins in 101 patients (52.3%) and marginal or intralesional excision with marginal or inadequate margins in 92 patients (47.7%). The 193 tumors were classified into 6 main histologic categories as follows: MFH in 49 patients (25.4%), liposarcoma in 48 patients (24.9%), synovial sarcoma in 30 patients (15.5%), spindle cell sarcoma in 24 patients (12.4%; 10 fibrosarcomas, 5 leiomyosarcomas, and 9 malignant peripheral nerve sheath tumors [MPNST]), small round cell sarcoma in 15 patients (7.8%; 6 rhabdomyosarcomas and 9 primitive neuroectodermal tumors), and others in 27 patients (14.0%; 2 alveolar soft part sarcomas, 3 angiosarcomas, 12 epithelioid sarcomas, 5 extraskeletal myxoid chondrosarcomas, and 5 extraskeletal osteosarcomas). The 49 MFH lesions were subdivided into the following histologic subtypes: storiform-pleomorphic in 24 patients and myxoid in 25 patients. Similarly, the 48 liposarcomas were subclassified into myxoid/round cell in 25 patients, well differentiated in 19 patients, and dedifferentiated in 5 patients. The MIB-1 LI ranged from 1% to 90% (median, 30%). Based on the MIB-1 LI, an MIB-1 score of 1 was assigned to 59 tumors (30.5%), an MIB-1 LI score of 2 was assigned to 37 tumors (19.2%), and an MIB-1 LI score of 3 was assigned to 97 tumors (50.3%). According to the grading systems that used mitosis and the MIB-1 score, 49 tumors (25.4%) and 44 tumors (22.8%) were Grade 1, 63 tumors (32.6%) and 68 tumors (35.2%) were Grade 2, and 81 tumors (42.0%) and 81 tumors (42.0%) were Grade 3, respectively. According to a staging system based on both grading systems, 52 tumors (26.9%) and 53 tumors (27.5%) were Stage I, 87 tumors (45.1%) and 84 tumors (43.5%) were Stage II, and 54 tumors (28.0%) and 56 tumors (29.0%) were Stage III, respectively. There were 42 Stage IA tumors, 11 Stage IB tumors, 60 Stage IIA tumors, 16 Stage IIB tumors, and 8 Stage IIC tumors (Table 1). The distribution of histologic types and the corresponding grade using the MIB-1 score and stage are listed in Table 2. Univariate Analysis for Overall Survival in 193 Patients with Soft Tissue Sarcoma In the univariate analysis (Table 1), tumor depth (P 0.01), tumor size (P 0.007), histologic type (P 0.001), MIB-1 score (P 0.001), both grades (P 0.001), and both stages (P 0.001) all had a significant impact on overall survival. When the histologic type was considered, patients with liposarcoma had better overall survival, with a 5-year survival rate of 89.4% (95% confidence interval [95%CI], 80.6 98.2%) compared with survival rates of 64.2% (95%CI, 49.9 78.5%) for patients with MFH, 57.4% (95%CI, 37.8 76.9%) for patients with synovial sarcoma, 47.9% (95%CI, 27.1 68.7%) for patients with spindle cell sarcoma, 60.0% (95%CI, 35.2 84.8%) for patients with small round cell sarcoma, and 60.0% (95%CI, 36.0 84.1%) for patients with other tumors (Fig. 1A). Among the liposarcomas, dedifferentiated liposarcoma is a high-grade tumor and should be separated from the well-differentiated histology. In our series, however, the patients with well-differentiated and dedifferentiated subtypes, with very few of the latter, who had a 5-year survival rate of 100%, had a better overall survival compared with patients who had the myxoid/ round cell subtype (P 0.02), with a 5-year survival rate of 80.0% (95%CI, 64.3 95.7%). Among the patients with MFH, those with the myxoid subtype, who had a 5-year survival rate of 75.8% (95%CI, 57.0 94.5%), tended to have a better overall survival curve (P 0.15) compared with patients who had the sto-

846 CANCER August 15, 2002 / Volume 95 / Number 4 TABLE 1 Patient and Tumor Characteristics and Univariate Analysis for Overall Survival in 193 Adult Patients with Soft Tissue Sarcoma of the Extremities and Trunk Variable No. of patients (%) Percent 5-year survival rate (95% confidence interval) Log-rank P value Relative risk (95% confidence interval) Age (yrs) 50 107 (55.4) 67.2 (57.4 76.9) 0.31 1.0 50 86 (44.6) 65.0 (54.0 75.9) 1.3 (0.8 2.0) Gender Female 84 (43.5) 62.1 (50.7 73.6) 0.57 1.0 Male 109 (56.5) 69.2 (59.9 78.7) 0.9 (0.6 1.4) Site Extremities 130 (67.4) 69.6 (61.1 78.1) 0.25 1.0 Trunk 63 (32.6) 73.9 (64.9 82.9) 1.3 (0.8 2.1) Surgical margin Inadequate 92 (47.7) 60.1 (49.4 70.9) 0.21 1.0 Adequate 101 (52.3) 73.9 (64.9 82.9) 0.7 (0.5 1.2) Histologic type Liposarcoma 48 (24.9) 89.4 (80.6 98.2) 0.001 1.0 Malignant fibrous histiocytoma 49 (25.4) 64.2 (49.9 78.5) 4.9 (2.0 12.0) Synovial sarcoma 30 (15.5) 57.4 (37.8 76.9) 4.8 (1.9 12.6) Spindle cell sarcoma 24 (12.4) 47.9 (27.1 68.7) 7.8 (3.0 20.2) Small round cell sarcoma 15 (7.8) 60.0 (35.2 84.8) 5.2 (1.7 16.2) Others 27 (14.0) 60.0 (36.0 84.1) 4.4 (1.6 12.2) Tumor depth Superficial 47 (24.4) 83.7 (71.4 96.0) 0.01 1.0 Deep 146 (75.6) 60.8 (52.3 69.4) 2.3 (1.2 4.5) Tumor size (cm) 5 65 (33.7) 77.5 (65.5 89.6) 0.007 1.0 5 10 82 (42.5) 63.0 (51.8 74.1) 2.2 (1.2 3.9) 10 46 (23.8) 56.1 (40.9 71.4) 2.6 (1.3 5.0) MIB-1 score 1 59 (30.5) 97.4 (92.3 100) 0.001 1.0 2 37 (19.2) 71.8 (56.8 86.7) 4.0 (1.5 10.5) 3 97 (50.3) 44.3 (32.9 55.6) 11.9 (5.1 28.1) Grade (mitosis) 1 49 (25.4) 94.9 (87.8 100) 0.001 1.0 2 63 (32.6) 71.7 (59.3 84.2) 4.2 (1.6 11.3) 3 81 (42.0) 44.0 (32.1 55.9) 11.5 (4.5 29.1) Stage (mitosis) I 52 (26.9) 86.4 (75.2 97.6) 0.001 1.0 II 87 (45.1) 73.3 (63.1 83.6) 1.4 (0.7 2.8) III 54 (28.0) 35.3 (21.6 48.9) 6.5 (3.3 12.7) Grade 1 44 (22.8) 96.7 (90.2 100) 0.001 1.0 2 68 (35.2) 77.1 (66.0 88.2) 3.9 (1.3 11.6) 3 81 (42.0) 39.6 (27.6 51.5) 16.4 (5.9 45.9) Stage I 53 (27.5) 86.8 (75.8 97.7) 0.001 1.0 IA 42 IB 11 II 84 (43.5) 76.2 (66.1 86.3) 1.3 (0.6 2.6) IIA 60 IIB 16 IIC 8 III 56 (29.0) 31.5 (18.2 44.8) 7.9 (4.0 15.5)

Grade and Stage in Soft Tissue Sarcoma/Hasegawa et al. 847 TABLE 2 Histologic Type and Corresponding Grade using MIB-1 Score and Stage Grade Stage Histologic type 1 2 3 I II III Total Liposarcoma 33 9 6 5 38 5 48 Malignant fibrous histiocytoma 5 26 18 19 16 14 49 Synovial sarcoma 0 12 18 9 9 12 30 Spindle cell sarcoma 3 8 13 8 7 9 24 Small round cell sarcoma 0 4 11 2 4 9 15 Others 3 9 15 10 10 7 27 Total 44 68 81 53 84 56 193 riform-pleomorphic subtype, with a 5-year survival rate of 53.0% (95%CI, 32.5 73.4%). Among the patients with sarcomas of different grades and stages, Grade 2 (RR, 4.2; 95%CI, 1.6 11.3) (RR, 3.9; 95%CI 1.3 11.6), Grade 3 (RR, 11.5; 95%CI, 4.5 29.1) (RR, 16.4; 95%CI, 5.9 45.9), and Stage III (RR, 6.5; 95%CI, 3.3 12.7) (RR, 7.9; 95%CI, 4.0 15.5) were associated significantly with overall survival (Table 1). An MIB-1 score of 2 (RR, 4.0; 95%CI, 1.5 10.5) and an MIB-1 score of 3 (RR, 11.9; 95%CI, 5.1 28.1) also were associated with overall survival. In the grading and staging systems that used the MIB-1 score, there was no significant difference in survival curves between patients with Stage I tumors and patients with Stage II tumors, but patients with Grade 2 tumors fared worse than patients with grade 1 tumors (Fig. 1B,C). Survival Difference for Grade and Stage in Each Histologic Type of Soft Tissue Sarcoma The survival differences for MIB-1 score, grades, and stages were estimated with a log-rank test for each histologic type of soft tissue sarcoma. Grading and staging systems using the MIB-1 score had better predictive value compared with the MIB-1 score and standard grading and staging models in the main histologic categories, except for other subtypes. Figures 2 7 show that Kaplan Meier survival curves by grade using the MIB-1 score and disease stage, respectively, showed similar trends for patients with MFH (P 0.03 and P 0.001), liposarcoma (P 0.001 and P 0.001), synovial sarcoma (P 0.001 and P 0.01), spindle cell sarcoma (P 0.02 and P 0.005), small round cell sarcoma (P 0.09 and P 0.05), and other sarcoma types (P 0.61 and P 0.78). Multivariate Analysis for Overall Survival in 193 Patients with Soft Tissue Sarcoma Because survival curves of grades and stages were similar between the histologic types, multivariate analysis was conducted adjusting for age, gender, disease site, surgical margin, tumor size and depth, grade using the MIB-1 score and stage, and histologic type, without including the interaction between histologic type, grade, and stage. Among the variables (Table 3), Grade 3 emerged as the most significant adverse prognostic factor, with an RR of 4.7 (95%CI, 1.3 17.2). Additional adverse prognostic factors were Stage III (RR, 3.4; 95%CI, 1.6 7.1), Grade 2 (RR, 2.5; 95%CI, 0.8 8.2), histologic type of spindle cell sarcoma (RR, 1.8; 95%CI, 1.0 3.2), and age 50 years at the time of presentation (RR, 1.8; 95%CI, 1.1 2.9). The histologic type liposarcoma (RR, 0.5; 95%CI, 0.2 1.3) was identified as a favorable prognostic factor. After excluding patients with other sarcoma types (n 27 patients), we also conducted same analysis based on 113 patients in high-grade categories, i.e., MFH (n 49 patients); synovial sarcoma (n 30 patients); small round cell sarcoma (n 15 patients); high-grade spindle cell sarcoma (n 13 patients); high-grade liposarcoma (n 6 patients), including round cell (n 4 patients) and dedifferentiated (n 2 patients) subtypes; and 53 patients in the intermediate grade and low-grade categories, including the remaining patients with liposarcoma (n 42 patients) and spindle cell sarcoma (n 11 patients). Even adjusting for other variables, the most significant adverse prognostic factor was Grade 3 in the high-grade categories (RR, 4.4; 95%CI, 2.0 9.4; P 0.001), and Grade 2 only was high risk in the intermediate grade and low-grade categories (RR, 18.7; 95%CI, 2.3 155.3; P 0.007). DISCUSSION In the current study, the prognostic value of the histologic grading scheme using the MIB-1 score and staging system combined with other clinicopathologic parameters was investigated in adult patients with soft tissue sarcomas. Using univariate analysis, tumor size and depth, histologic type, MIB-1 score, grade, and stage were associated significantly with overall survival, in accordance with previous studies. 8,9 Among

848 CANCER August 15, 2002 / Volume 95 / Number 4 FIGURE 2. Kaplan Meier survival curves for grade (A) and stage (B) in 49 patients with malignant fibrous histiocytoma. FIGURE 1. Kaplan Meier survival curves for histologic type (A), grade (B), and stage (C) in 193 adult patients with soft tissue sarcoma. these variables, grading and staging systems using the MIB-1 score had better predictive value than the MIB-1 score and standard grading and staging models in the main histologic categories. In another multivariate analysis using the standard grading and staging model, tumor size 10 cm, Grade 3, tumor size of 5 10 cm, deep location, histologic type of spindle cell sarcoma, age at presentation 50 years, and tumor site in the trunk were adverse prognostic factors, and the histologic type of liposarcoma was a favorable prognostic factor (data not shown). Thus, the grading and staging systems using the MIB-1 score appear to be useful for estimating the risk of death by smaller numbers of variables (Table 3). In patients with soft tissue sarcomas, Kaplan Meier survival curves differed significantly between patients with Grade 1 tumors and patients with Grade 2 tumors, whereas patients with Stage I lesions did not differ from patients with Stage II tumors, probably due to the fact that the staging system no longer recognizes any difference between Grade 1 tumors and Grade 2 tumors. This staging system may have the advantage of discriminating Stage III lesions (high grade, large [ 5 cm], and deep) from other lesions. Although the cut-off size of the grading system is 5 cm, it has been shown that patients with tumors measuring 10 cm have an even worse prognosis. 8,9,12 How-

Grade and Stage in Soft Tissue Sarcoma/Hasegawa et al. 849 FIGURE 3. Kaplan Meier survival curves for grade (A) and stage (B) in 48 patients with liposarcoma. ever, in this study, little difference was seen in the survival curves of patients with high-grade and deep lesions that measured 5 10 cm or 10 cm. The predictive value of grading and staging in the main histologic categories was almost identical to that of soft tissue sarcomas as a whole; survival curves of grades and stages showed similar trends among patients with MFH, liposarcoma, synovial sarcoma, spindle cell sarcoma, and small round cell sarcoma. In patients with MFH, a difference in histologic subtype between myxoid and storiform-pleomorphic may have an impact on overall survival, 13 but patients with Stage III lesions were at the greatest risk of mortality, in keeping with the results of other large series. 14 However, our group of patients with spindle cell sarcoma was too small to be studied separately. Coindre et al. reported that histologic grade for metastasis development had no predictive value in patients with MPNST. 15 In agreement with others, 9 approximately 70% of liposarcomas were Grade 1 in our series and behaved FIGURE 4. Kaplan Meier survival curves for grade (A) and stage (B) in 30 patients with synovial sarcoma. less aggressively than other histologic types. Among these, well-differentiated and dedifferentiated liposarcomas comprised one subgroup, and myxoid and round cell liposarcomas make up a second subgroup. 11 Highgrade lesions (dedifferentiated liposarcomas), as a result of the progression of well-differentiated liposarcomas, occur most frequently in the retroperitoneum 4 but also occur in the extremities and trunk. Within the latter location, a survival rate of 100% for patients with welldifferentiated and even dedifferentiated liposarcoma was found compared with a survival rate of 80% for patients with myxoid and round cell liposarcomas. Grade 3 and Stage III lesions, although they are rare among liposarcomas, were associated significantly with a poor outcome. From a practical viewpoint, it seems useful to apply grading and staging to myxoid and round cell liposarcomas, which have the potential to metastasize, to predict their behavior.

850 CANCER August 15, 2002 / Volume 95 / Number 4 FIGURE 5. Kaplan Meier survival curves for grade (A) and stage (B) in 24 patients with spindle cell sarcoma. FIGURE 6. Kaplan Meier survival curves for grade (A) and stage (B) in 15 patients with small round cell sarcoma. In a previous study, 5 multivariate analysis showed that Grade 3 was the most adverse prognostic factor in patients with synovial sarcomas. Synovial sarcomas generally are viewed as high-grade sarcomas with a high metastatic risk. In contrast, patients with Grade 2 and Stage I tumors appear to comprise a fairly low-risk group. The trend toward different survival curves for patients with different grades and stages of small round cell sarcomas, such as rhabdomyosarcomas and primitive neuroectodermal tumors, although they are rare in patients age 15 years, was similar to the trend seen for patients with synovial sarcomas. In patients with other subtypes of soft tissue sarcomas, who were lumped together into a single heterogeneous group, it is believed generally that tumor size has high prognostic value, especially in patients with epithelioid sarcomas. 16 In addition, clear cell sarcoma, epithelioid sarcoma, and alveolar soft part sarcoma may have histologic specific behavior rather than behavior that depends on the tumor grade. 15 Because the survival curves for different grades and stages were similar in each main histologic category, multivariate analysis showed that Grades 2 and 3 were significantly associated with a poor outcome (patients who had Grade 3 tumors had an approximately 12 times greater risk compared with patients who had Grade 2 tumors). Among patients with Grade 3 tumors, patients with large and deep sarcomas (Stage III) were at increased risk of mortality. Even if we analyzed two separate groups high-grade and intermediate grade or low-grade categories grading had the greatest predictive value for overall survival in each group; thus, it seems necessary for patients in the high-grade categories. These grading and staging systems can contribute to the identification of individual patients with high-grade sarcomas who are at high risk of death and can help stratify patients for neoadjuvant chemotherapy trials (unpublished data). In conclusion, grading and staging systems using the MIB-1 score are very strong prognostic factors in patients with the main histologic types of soft tissue sarcoma. Multivariate analysis demonstrated that,

Grade and Stage in Soft Tissue Sarcoma/Hasegawa et al. 851 and the specific assessment of histologic type should be included with the assessment of tumor size, depth, and grade when deciding on treatment strategies for adult patients with soft tissue sarcomas. 8 FIGURE 7. Kaplan Meier survival curves for grade (A) and stage (B) in 27 other patients. TABLE 3 Multivariate Analysis with Grading and Staging Systems using MIB-1 Score for Overall Survival in 193 Adult Patients with Soft Tissue Sarcoma of the Extremities and Trunk Variable P value Relative risk 95% Confidence interval Age 50 yrs 0.02 1.8 1.1 2.9 Liposarcoma 0.14 0.5 0.2 1.3 Spindle cell sarcoma 0.05 1.8 1.0 3.2 Grade 2 0.12 2.5 0.8 8.2 Grade 3 0.02 4.7 1.3 17.2 Stage III 0.001 3.4 1.6 7.1 when adjusting for all other factors, the histologic type liposarcoma was a favorable prognostic factor, and spindle cell sarcoma was an adverse prognostic factor. These findings may reinforce the previously appreciated idea that the various histologic types of soft tissue sarcoma are associated with different risks of death, REFERENCES 1. Costa J, Wesley RA, Glatstein E, Rosenberg SA. The grading of soft tissue sarcomas. Results of a clinicohistopathologic correlation in a series of 163 cases. Cancer. 1984;53:530 541. 2. Guillou L, Coindre JM, Bonichon F, et al. Comparative study of the National Cancer Institute and French Federation of Cancer Centers Sarcoma Group grading systems in a population of 410 adult patients with soft tissue sarcoma. J Clin Oncol. 1997;15:350 362. 3. Hasegawa T, Yokoyama R, Lee YH, Shimoda T, Beppu Y, Hirohashi S. Prognostic relevance of a histological grading system using MIB-1 for adult soft-tissue sarcoma. Oncology. 2000;58:66 74. 4. Hasegawa T, Seki K, Hasegawa F, et al. Dedifferentiated liposarcoma of retroperitoneum and mesentery: varied growth patterns and histological grades a clinicopathologic study of 32 cases. Hum Pathol. 2000;31:717 727. 5. Hasegawa T, Yokoyama R, Matsuno Y, Shimoda T, Hirohashi S. Prognostic significance of histologic grade and nuclear expression of beta-catenin in synovial sarcoma. Hum Pathol. 2001;32:257 263. 6. Hasegawa T, Yamamoto S, Nojima T, et al. Validity and reproducibility of histologic diagnosis and grading for adult soft-tissue sarcomas. Hum Pathol. 2002;33:111 115. 7. Fleming ID, Cooper JS, Henson DE, et al. Soft tissue sarcoma AJCC cancer staging manual, 5th ed Philadelphia: Lippincott-Raven, 1997:149 156. 8. 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 1689. 9. Ramanathan RC, A Hern R, Fisher C, Thomas JM. Modified staging system for extremity soft tissue sarcomas. Ann Surg Oncol. 1999;6:57 69. 10. Hashimoto H, Daimaru Y, Takeshita S, Tsuneyoshi M, Enjoji M. Prognostic significance of histologic parameters of soft tissue sarcomas. Cancer. 1992;70:2816 2822. 11. Weiss SW, Goldblum JR. Enzinger and Weiss s soft tissue tumors. 4th ed. St. Louis: Mosby, 2001. 12. Brennan MF. Staging of soft tissue sarcomas. Ann Surg Oncol. 1999;6:8 9. 13. Le Doussal V, Coindre JM, Leroux A, et al. Prognostic factors for patients with localized primary malignant fibrous histiocytoma: a multicenter study of 216 patients with multivariate analysis. Cancer. 1996;77:1823 1830. 14. Salo JC, Lewis JJ, Woodruff JM, Leung DH, Brennan MF. Malignant fibrous histiocytoma of the extremity. Cancer. 1999;85:1765 1772. 15. Coindre JM, Terrier P, Guillou L, et al. Predictive value of grade for metastasis development in the main histologic types of adult soft tissue sarcomas: a study of 1240 patients from the French Federation of Cancer Centers Sarcoma Group. Cancer. 2001;91:1914 1926. 16. Hasegawa T, Matsuno Y, Shimoda T, Umeda T, Yokoyama R, Hirohashi S. Proximal-type epithelioid sarcoma: a clinicopathologic study of 20 cases. Mod Pathol. 2001;14:655 663.