Clinical Course of Nonvisceral Soft Tissue Leiomyosarcoma in 225 Patients from the Scandinavian Sarcoma Group

282 Clinical Course of Nonvisceral Soft Tissue Leiomyosarcoma in 225 Patients from the Scandinavian Sarcoma Group Catarina Svarvar, MD 1 Tom B ohling, MD, PhD 2 Orjan Berlin, MD, PhD 3 Pelle Gustafson, MD, PhD 4 Gunnar Follerås, MD 5 Bodil Bjerkehagen, MD 6 Henryk A. Domanski, MD, PhD 7 Kirsten Sundby Hall, MD, PhD 8 Erkki Tukiainen, MD, PhD 1 Carl Blomqvist, MD, PhD 9 and the Scandinavian Sarcoma Group Leiomyosarcoma Working Group 1 Department of Plastic Surgery, Helsinki University Central Hospital, Helsinki, Finland. 2 Department of Pathology, HUSLAB, Helsinki University Central Hospital and University of Helsinki, Helsinki, Finland. 3 Department of Orthopedics, Sahlgrenska University Hospital, Goteborg, Sweden. 4 Department of Orthopedics, University Hospital, Lund, Sweden. 5 Department of Surgery, Rikshospitalet-Radiumhospitalet Medical Center, Oslo, Norway. 6 Department of Pathology, Rikshospitalet- Radiumhospitalet Medical Center, Oslo, Norway. 7 Department of Pathology and Cytology, University Hospital, Lund, Sweden. 8 Department of Oncology, Rikshospitalet-Radiumhospitalet Medical Center, Oslo, Norway. 9 Department of Oncology, Helsinki University Central Hospital, Helsinki, Finland. BACKGROUND. Leiomyosarcoma of nonvisceral soft tissues is an uncommon malignant tumor; thus, only small numbers of cases have been reported. This study was based on a large series of patients from the Scandinavian Sarcoma Group Register acquired during a 15-year period (from 1986 to 2001). Follow-up information was available for all patients. METHODS. The authors analyzed the clinical features of 225 patients with cutaneous, subcutaneous, or deep-seated leiomyosarcoma of the extremities, trunk wall, and superficial parts of the head and neck region to determine the natural course of the disease. Only patients who received their treatment at a specialist sarcoma center were included. Re-evaluation of histopathology was performed. RESULTS. The age of the patients (121 women and 104 men) ranged from 20 years to 98 years (median, 70 years), and the tumors ranged in size from 0.6 cm to 35 cm (median, 4.0 cm). Eighty-two percent of the tumors were classified as high grade. The median follow-up for survivors was 5.5 years. The local treatment was adequate in 154 of 206 patients (75%) who were without metastasis at presentation. At 10 years, 84% of the 206 patients with localized disease at presentation were free from local recurrence, 66% remained metastasis free, and 49% were alive. Multivariate analysis showed that higher malignancy grade (P ¼.006), larger tumor size (P ¼.003), and deeper tumor location (P ¼.002) were correlated significantly with decreased metastasis-free survival, inadequate local treatment was correlated with local recurrence (P ¼.007), and high malignancy grade was correlated with decreased overall survival (P ¼.007). CONCLUSIONS. The long-term prognosis for patients with subcutaneous and deepseated soft tissue leiomyosarcoma remains poor despite the ability to achieve adequate local control through nonmutilating surgery with or without radiotherapy. Cancer 2007;109:282 91. Ó 2006 American Cancer Society. KEYWORDS: leiomyosarcoma, metastasis, prognosis, soft tissue sacroma. Leiomyosarcoma (LMS) of soft tissues is a relatively uncommon malignant tumor that has the phenotypic features of smooth muscle differentiation and may occur anywhere in the body. 1 Several recent reports have indicated that a reliable and reproducible diagnosis of Supported by grants from Finska Lakaresallskapet and K. Albin Johansson Foundation in Finland. The Scandinavian Sarcoma Group Leiomyosarcoma Working Group for this article included the following investigators: Sweden: Thor A. Alvegard, MD, PhD (Department of Cancer Epidemiology, Oncological Center, University Hospital, Lund) and Jonas Bergh, MD, PhD (Department of Oncology, Karolinska Institute and Karolinska University Hospital, Stockholm); Norway: Clement S. Trovik, MD, PhD (Musculo-Skeletal Tumor Center, Haukeland University Hospital, Bergen); and Finland: Paula Lindholm, MD, PhD (Department of Oncology and Radiotherapy, Turku University Central Hospital, Turku). Address for reprints: Catarina Svarvar, MD, PhD, Department of Plastic Surgery, Helsinki University Central Hospital, PO Box 266, FI-00029 HUS, Helsinki, Finland; Fax: (011) 358 9-47176331; E-mail: catarina.svarvar@hus.fi Received August 30, 2006; revision received October 10, 2006; accepted October 13, 2006. ª 2006 American Cancer Society DOI 10.1002/cncr.22395 Published online 7 December 2006 in Wiley InterScience (www.interscience.wiley.com).

Clinical Course of Leiomyosarcoma/Svarvar et al. 283 LMS is difficult without immunohistochemistry, but the available markers are not completely specific or sensitive. 2,3 Most previous studies of LMS have not included immunohistochemical analysis as a diagnostic criterion. LMS in general is considered to occur mainly in the uterus and, according to the literature, also in the retroperitoneal space and abdominal cavity, whereas LMS of nonvisceral soft tissues is more rare and accounts for approximately 10% of soft tissue sarcomas in adults. 1,4,5 However, recent reviews of pathologic soft tissue sarcoma material have changed the proportions of different sarcoma types in favor of LMS. 2,6 8 Methodological advances in immunohistochemistry and molecular genetics have improved diagnostic accuracy vastly, and the majority of intraabdominal LMS have been reclassified as gastrointestinal stromal tumors. LMS that occurs outside the body cavities, in the extremities, trunk wall, and superficial parts of the head and neck region, has not been characterized well, and the experience of any single sarcoma institution usually is limited. Only a few LMS studies 4,8 13 of any significant size included clinical and follow-up data, and several others 14 28 analyzed limited numbers of patients with primary LMS at these sites. In addition, many of the previous studies are difficult to interpret, because heterogeneous tumor groups were studied without specifying the localization relative to skin or fascia or the anatomic site. Histologic malignancy grade, tumor size, tumor depth, and surgical margin have been accepted as factors of prognostic importance in patients with soft tissue sarcoma. 29 31 LMS is considered an aggressive histopathologic type of soft tissue sarcoma. 11,29,31 A significant number of LMS of soft tissue have been reported that arose from small vessels, indicating that this relation may be important for predicting clinical behavior and the risk of metastasis. 11,17 Conversely, it is believed that cutaneous LMS originates in the dermis from the arrectores pilorum muscles of the hair follicles and from the smooth muscle surrounding sweat glands, which may explain the more benign nature of this form of LMS. In the current study we focused on the clinical factors of soft tissue LMS of the extremities, trunk wall, and superficial parts of the head and neck region, including cutaneous, subcutaneous, and deep-seated tumors. We report a series of 225 patients from the Scandinavian Sarcoma Group (SSG) Register in which multivariate analysis was used to identify independent factors that affected patient outcome. To our knowledge, this is the largest study of nonvisceral LMS with follow-up data ever performed to investigate the relation between clinical characteristics and outcome. TABLE 1 Participating Sarcoma Centers and the Number of Patients (%) from Each Hospital Sarcoma center City, country No. of patients (%) University Hospital Lund, Sweden 50 (22) Sahlgrenska University Hospital Goteborg, Sweden 21 (9.3) Karolinska University Hospital Stockholm, Sweden 19 (8.4) University Hospital Linkoping, Sweden 1 (0.4) University Hospital Umea, Sweden 1 (0.4) Norwegian Radium Hospital Oslo, Norway 57 (25) Haukeland University Hospital Bergen, Norway 11 (4.9) University Hospital Trondheim, Norway 3 (1.3) Helsinki University Central Hospital Helsinki, Finland 57 (25) Turku University Central Hospital Turku, Finland 5 (2.2) MATERIALS AND METHODS Criteria for Inclusion Patients with soft tissue sarcoma who were treated at sarcoma centers in Sweden, Norway, and Finland have been reported prospectively along with follow-up data to the SSG Register since 1986. 32 The current study was based on a series of adult patients who were diagnosed with LMS of the extremities, trunk wall, or superficial parts of the head and neck region and who were reported to the SSG Register during a 15-year period from 1986 to 2001. Patients with cutaneous, subcutaneous, and deep-seated LMS were included. Ten university sarcoma centers contributed to the study (Table 1). All tumors were re-evaluated by the experienced sarcoma pathologists of the SSG Peer-Review Committee. 33 A standardized protocol was used for the review process. The LMS diagnoses for all tumors that were included were confirmed based on morphology that showed smooth muscle differentiation and positive immunohistochemical staining (IHC) for a-smooth muscle actin and desmin in most patients. If it was judged necessary, then new sections were obtained, and IHC was performed. Reclassified tumors were excluded from the study. Malignancy grading was based on a 4-tiered grading scheme (see Malignancy grade, below). The depth of the superficial tumors (cutaneous vs subcutaneous) was determined either microscopically or according to pathology records. All patients who were included had nearly complete prospective clinical data regarding the anatomic site, tumor depth, tumor size, preoperative diagnostic procedures, referral pattern to sarcoma center, primary treatment (including surgery and adjuvant therapy), and follow-up (including tumor-related cause of death). Patients who were not referred to a sarcoma center for treatment and patients who had missing follow-up data in the SSG Register were excluded from the study. However, 23 patients who were referred to a sarcoma center after local recurrence

284 CANCER January 15, 2007 / Volume 109 / Number 2 were included. Compared with patients who were treated for primary tumors, as demonstrated previously, 34 patients who were treated for local recurrences did not exhibit a statistically significant difference in the risk of local recurrence (P ¼.972) or metastasis (P ¼.281) after treatment at a sarcoma center. Patients who had visceral, retroperitoneal, gastrointestinal, gynecologic, genitourinary, breast, intrathoracic, and major vessel LMS were excluded. In summary, 225 patients with LMS constituted the basis for the current study, including 202 patients (90%) with primary LMS and 23 patients (10%) who had been referred to a sarcoma center after local recurrence. Nineteen patients (8.4%) had metastasis at the time of diagnosis or at the time they were referred to a sarcoma center. Definitions Anatomic site According to the registration policy of the SSG Register, the head and neck region was considered a single site; the upper extremities were divided into the upper arm, elbow, lower arm, and hand; and the lower extremities were divided into the thigh, knee, lower leg, and foot. The upper part of the trunk wall, the shoulder region, and the lower part of the trunk wall (including the gluteal region and the groins) were assigned to the trunk wall. Tumor depth Cutaneous tumors were defined as tumors in the dermis that only subsequently invaded the subcutis. Subcutaneous tumors were located in the subcutaneous space without involvement of the muscular fascia. All other tumors were considered deep seated, extramuscular,or intramuscular. A subcutaneous tumor with subfascial extension was considered extramuscular. Cutaneous, subcutaneous, and deep-seated tumors external to the skeletal boundary constituted the entity of tumors in the superficial parts of the head and neck region. The definition of the trunk wall covered thoracic, abdominal, and pelvic walls external to the body cavities. Tumor size Tumor size (in centimeters) was defined as the greatest dimension of the tumor in the surgical specimen reported by the original pathologists. The median tumor size was 4 cm (range, 0.6 35 cm). According to the TNM classification system, the tumors were classified as small (5 cm) or large (>5 cm). Malignancy grade According to the SSG protocols, histologic malignancy of the tumors was graded according to a 4-tiered grading scale modified from Broders et al. and Angervall TABLE 2 Comparison Between Clinical Factors of All 225 Patients and 206 Patients Without Metastasis at Presentation Clinical parameter No. of patients (%) All patients (N 5 225) Localized sarcomas (N 5 206) Age, y Median 70.0 69.5 Mean 65.3 65.3 Range 20 98 20 97 Sex Men 104 (46) 97 (47) Women 121 (54) 109 (53) Tumor size, cm Median 4.0 4.0 Mean 5.9 5.5 Range 0.6 35.0 0.6 35.0 5 cm 129 (59) 126 (63) >5 cm 91 (41) 75 (37) Not determinable 5 5 Tumor site Head and neck 6 (2.7) 6 (2.9) Trunk wall 63 (28) 59 (29) Upper trunk 25 (11) 24 (12) Shoulder 10 (4.4) 10 (4.9) Lower trunk 4 (1.8) 4 (1.9) Gluteal region 13 (5.8) 11 (5.3) Groin 11 (4.9) 10 (4.9) Extremities 156 (69) 141 (68) Upper arm 15 (6.7) 15 (7.3) Elbow 2 (0.9) 2 (1.0) Lower arm 7 (3.1) 7 (3.4) Hand 1 (0.4) 1 (0.5) Thigh 77 (34) 66 (32) Knee 14 (6.2) 13 (6.3) Lower leg 35 (16) 32 (16) Foot 5 (2.2) 5 (2.4) Tumor depth Cutaneous 36 (16) 36 (17) Subcutaneous 101 (45) 97 (47) Deep-seated 88 (39) 73 (35) Tumor grade Grade 1 6 (2.7) 6 (3) Grade 2 33 (15) 32 (16) Grade 3 75 (34) 70 (34) Grade 4 108 (49) 95 (47) Not determinable 3 3 et al. 35,36 Grades1and2wereconsideredlow grade,and grades 3 and 4 were considered high grade. Thetumor grade was based on an estimate of the degree of cellularity, the degree of differentiation, the number of mitosis per high-power field, the amount of cellular atypia, and the amount of necrosis (data not shown). 36,37 Surgical margins The surgical margins were classified in the SSG Register as intralesional, marginal, or wide according to the SSG guidelines, which basically are from the surgical

Clinical Course of Leiomyosarcoma/Svarvar et al. 285 TABLE 3 Comparison of Treatments Between All 225 Patients and 206 Patients Without Metastasis at Presentation All patients (N 5 225) Localized sarcomas (N 5 206) TABLE 4 Estimated 5-Year and 10-Year Survival According to Significant Prognostic Factors by Multivariate Statistical Analysis of 206 Patients Without Metastasis at Presentation All patients (N 5 225) Localized sarcomas (N 5 206) Treatment modality No. % No. % Prognostic factor 5-Year, % 10-Year, % 5-Year, % 10-Year, % Surgical treatment All procedures 219 97 205 99.5 Amputation 8 3.6 8 3.9 RT 68 30 61 30 Chemotherapy 11 4.9 5 2.4 Final local treatment Wide surgical margin With RT 19 8.4 19 9.2 Without RT 110 49 106 51 Marginal surgical margin With RT 31 14 28 14 Without RT 38 17 34 17 Intralesional surgical margin With RT 15 6.7 13 6.3 Without RT 6 2.7 5 2.4 No surgery With RT 3 1.3 1 0.5 Without RT 3 1.3 0 0 Adequate local treatment 163 72 154 75 RT indicates radiotherapy. staging system reported by Enneking et al. 38 An intralesional surgical margin is achieved when macroscopic or microscopic tumor tissue is left at the margins of the wound at the surgical procedure. A marginal surgical margin is achieved when the dissection is close to the tumor at any site, irrespective of how much healthy tissue is included elsewhere. The plane of dissection is through the pseudocapsule or reactive tissue around the lesion. A wide surgical margin is achieved by a procedure in which the lesion and a surrounding cuff of normal tissue are resected as a single block, which, in cutaneous and subcutaneous tumors, includes the deep fascia beneath the tumor. A wide margin for a tumor located between muscles includes the muscles around the tumor. In this study, we assessed the cases registered with compartmental excision and myectomy as wide surgical margin. Myectomy is a wide resection with fascial containment of the surgical specimen when the muscle in which the tumor is located is removed unopened. A compartmental surgical margin is obtained when the tumor-involved compartment is removed unopened. Adequacy of local treatment In this study, the local treatment was considered adequate when the surgical margin was wide or when the LRFS Entire group 86 84 86 84 Local treatment Adequate 90 88 Inadequate 72 72 MRFS Entire group 68 60 74 66 Tumor size 5 cm 87 74 >5 cm 52 52 Tumor depth Cutaneous 100 100 Subcutaneous 83 74 Deep-seated 50 39 Malignancy grade Grade 1 100 100 Grade 2 97 97 Grade 3 85 71 Grade 4 58 49 OS Entire group 64 46 69 49 Malignancy grade Grade 1 100 100 Grade 2 93 81 Grade 3 80 45 Grade 4 50 37 LRFS indicates local recurrence-free survival; MRFS, metastatic recurrence-free survival; OS, overall survival. marginal surgical margin was complemented with adjuvant radiation therapy. An intralesional surgical margin with or without adjuvant radiation therapy and a marginal surgical margin alone (independent of tumor depth) were considered inadequate local treatment. Postoperative radiotherapy was recommended after marginal or intralesional surgery, generally to a dose of 50 grays (Gy) in 2-Gy fractions, followed by a boost to areas with microscopic or macroscopic residual tumor. Length of Follow-Up The length of follow-up was calculated from the date of the histologic diagnosis (biopsy or excision) to the date of last follow-up or death. For patients who were referred to a sarcoma center for treatment of a local recurrence, the date of the histologic diagnosis of the local recurrence was used as the date of diagnosis.

286 CANCER January 15, 2007 / Volume 109 / Number 2 TABLE 5 Prognostic Factors by Univariate and/or Multivariate Statistical Analyses With P Values and Relative Risk for Local Recurrence, Metastatic Recurrence, and Death for 206 Patients Without Metastasis at Presentation LRFS MRFS OS Univariate Multivariate Univariate Multivariate Univariate Multivariate Factor P RR P RR P RR P RR P RR P RR Adequacy of local treatment.004* 0.32.007* 0.33.352 0.76.047* 0.63.073 0.65 Tumor size.005* 1.08.229 1.05 <.0001* 1.12.003* 1.07 <.0001* 1.08.072 1.04 Tumor depth.015* 2.13.299 1.51 <.0001* 4.06.002* 2.54 <.0001* 1.91.245 1.28 Malignancy grade.028* 1.99.233 1.47 <.0001* 3.30.006* 2.08 <.0001* 1.98.007* 1.59 LRFS indicates local recurrence-free survival; MRFS, metastatic recurrence-free survival; OS, overall survival; RR, relative risk. * Significant P value. Follow-up information was available for all patients. At the time of this analysis, 118 of 225 patients (52%) were alive. The median follow-up of the surviving patients was 5.5 years (range, from 11 months to 18 years). Statistics Survival curves were plotted using the Kaplan-Meier method. The prognostic factors that were tested were tumor size, tumor depth (cutaneous vs subcutaneous vs deep-seated), malignancy grade (grades 1, 2, 3, and 4), and adequacy of local treatment (adequate vs inadequate), which correspond to the factors that are included in most staging systems. Patients who presented with metastasis at the time of diagnosis and/or were referred to a specialist sarcoma center after metastasis was diagnosed (n ¼ 19 patients) were excluded from the analysis of local control and metastasis-free survival. Univariate analyses of possible prognostic factors for local or distant failure or for death were done using a log-rank test for dichotomous variables and Cox regression analysis for continuous or ordinal variables. Tumor depth was coded as 1, 2, or 3 for cutaneous, subcutaneous, and deep-seated tumors, respectively. Multivariate analysis of the factors that were significant in univariate analyses was done by using Cox regression analysis. A P value <.05 was considered statistically significant. All statistical analyses for the current study were performed using the SPSS Statistical Package on a Macintosh computer. RESULTS Most of the 225 patients in this study were elderly; the median age was 70 years. The sex distribution was even. Sixty-nine percent of tumors were located on the extremities, and the most common site was the thigh (n ¼ 77 tumors). Only 16% of tumors were strictly cutaneous. The majority of the tumors were high grade (82%). Clinical characteristics are summarized in Table 2. Two hundred nineteen of 225 patients (97%) had undergone at least one surgery. All but 1 of 206 patients who presented with localized tumor underwent surgery, and 8 of those patients (3.9%) underwent amputation, which also was the number of amputations for the whole group. Five of 19 patients (26%) who had metastasis at presentation did not undergo surgical treatment because of advanced disease, or old age, or both. The referral pattern to the sarcoma centers showed that 97 of 225 patients (43%) had undergone surgery before referral, but only 1 patient had undergone wide excision. The final surgical margin after treatment at a sarcoma center was intralesional in 21 of 219 patients (9.6%), 69 patients (32%) had marginal surgical margins, and 129 patients (59%) had wide surgical margin. The local treatment (surgery alone or surgery and adjuvant radiation therapy) was considered adequate in 154 of 206 patients (75%) who had localized tumors. Only 4 of 68 patients (6%) who had received radiation therapy received preoperative radiation. Less than 5% of patients received adjuvant cytotoxic treatment. Treatment modalities are summarized in Table 3. Follow-Up Results The results from the statistical analyses are shown in Tables 4 and 5, and Kaplan-Meier survival curves are shown in the 8 Figures. The 5-year and 10-year overall survival rates for the whole group of 225 patients were 64% and 46%, respectively; and the same rates for the 206 patients without metastasis at presentation were 69% and 49%, respectively (Table 4, Fig. 1). The strongest prognostic factor for survival was the presence of

Clinical Course of Leiomyosarcoma/Svarvar et al. 287 FIGURE 1. Overall survival rates are illustrated for the 206 patients who presented without metastasis (thin line) and for the 19 patients who presented with metastasis (thick line). At 0 years, 5 years, and 10 years, the numbers of patients at risk were 225, 96, and 39, respectively. metastasis, either at the time of presentation or occurring subsequently. The 5-year survival rate for the patients who had metastasis at presentation was 16% (Fig. 1). Multivariate analysis showed that high malignancy grade was the main risk factor for decreased overall survival (Table 5, Fig. 2). Altogether, 76 (34%) of 225 patients had disease that metastasized. Sixty-one percent (46 of 76 patients) of metastatic disease occurred within 1 year, and 87% (66 of 76 patients) occurred within 3 years from diagnosis. During follow-up, 69 of 107 patients (64%) died of a tumor-related cause, which was metastatic disease in all but 4 of patients (3.7%). Twenty-nine patients (27%) died without recurrence of their disease, and 9 patients (8.4%) died of unknown causes. None of the cutaneous or grade 1 tumors metastasized during follow-up. The rates of freedom from metastatic recurrence at 5 years and 10 years were 74% and 66%, respectively, for the 206 patients who presented without metastasis (Table 4, Fig. 3). In univariate and multivariate analyses, higher malignancy grade, larger tumor size, and deeper tumor location were associated with a poorer outcome, whereas inadequate local treatment did not correlate with metastasis (Table 5, Figs. 4 6). In the group of 206 patients who presented with localized tumor, the respective local control rates at FIGURE 2. Overall survival is illustrated according to histologic malignancy grade. The dashed line indicates grade 1 tumors, and the solid, progressively thicker lines indicate grade 2, grade 3, and grade 4 tumors, respectively. None of patients with grade 1 tumors died of tumor-related causes. At 0 years, 5 years, and 10 years, the numbers of patients at risk were 203, 92, and 37, respectively. FIGURE 3. The metastatic recurrence-free survival rate is illustrated for the 206 patients who presented without metastasis. At 0 years, 5 years, and 10 years, the numbers of patients at risk were 206, 83, and 33, respectively.

288 CANCER January 15, 2007 / Volume 109 / Number 2 FIGURE 4. Metastatic recurrence-free survival is illustrated according to histologic malignancy grade. The dashed line indicates grade 1 tumors, and the solid, progressively thicker lines indicate grade 2, grade 3, and grade 4 tumors, respectively. None of the grade 1 tumors metastasized. At 0 years, 5 years, and 10 years, the numbers of patients at risk were 203, 83, and 33, respectively. 5 years and 10 years were 86% and 84% (Table 4, Fig. 7). Multivariate analysis showed, as expected, that inadequate local treatment was the main risk factor for local recurrence (Table 5, Fig. 8). DISCUSSION LMS is among the most common types of soft tissue sarcoma, accounting for >10% of all nonvisceral soft tissue sarcomas. 1,2,5 To our knowledge, nonvisceral soft tissue LMS has not been described in any large previous study with complete follow-up and multivariate analysis. Here, we report a register-based series of patients who were treated according to the SSG guidelines at specialized sarcoma centers. In the current study, we observed that tumor depth was a significant prognostic factor for metastasis in both multivariate and univariate analyses, and it also was a significant prognostic factor for local recurrence and death in univariate analysis. The importance of tumor depth as a prognostic factor is in line with earlier studies on nonvisceral soft tissue LMS and other histopathologic types of sarcoma. 8,11,13,18,29,31,32 The problem with previous reports is that cutaneous LMS and subcutaneous LMS have been analyzed as a single FIGURE 5. Metastatic recurrence-free survival is illustrated according to tumor size. The thin line indicates small tumors (5 cm), and the thick line illustrates large tumors (>5 cm). At 0 years, 5 years, and 10 years, the numbers of patients at risk were 201, 82, and 33, respectively. entity, and definitions of tumor depth vary between different studies. In our series, none of the cutaneous tumors metastasized during follow-up, and none of the patients with cutaneous tumors died from tumorrelated causes, consistent with previous studies, which have shown that strictly cutaneous LMS hardly ever metastasizes. 9,10,18,19,23,26,27,37 At the histopathologic review, we designated 8 tumors (3 in patients who had been referred after recurrence) that extended into both cutis and subcutis as subcutaneous tumors, because their growth pattern denied classification as cutaneous tumors. However, none of these tumors metastasized during follow-up. Excluding the cutaneous tumors, 24 of 101 subcutaneous tumors (24%) and 52 of 88 deepseated tumors (59%) metastasized, ie, 76 of all 189 noncutaneous LMS (40%) metastasized. Similar overall metastatic rates from 40% to 50% have been reported in unselected patients with LMS. 4,10,11 Tumor size was shown to be a significant prognostic factor for metastasis by multivariate analysis and by univariate analysis also for local recurrence and death. The importance of increasing tumor size as an adverse prognostic factor is in line with earlier studies on soft tissue LMS and other histopathologic sarcoma types. 8,10,11 13,18,29 31,39,40,41 Histologic malignancy grade is considered to be the most important prognostic factor in adult soft tis-

Clinical Course of Leiomyosarcoma/Svarvar et al. 289 FIGURE 6. Metastatic recurrence-free survival is illustrated according to tumor depth. The thin line indicates cutaneous tumors, and the solid, progressively thicker lines show subcutaneous and deep-seated tumors, respectively. None of the cutaneous tumors metastasized. At 0 years, 5 years, and 10 years, the numbers of patient at risk were 206, 83, and 33, respectively. sue sarcomas. Our multivariate analysis showed that histopathologic malignancy grade is a significant prognostic factor for metastatic disease and death. The results were in accordance with earlier studies on LMS and other types of soft tissue sarcoma. 10,11,13,29,30,31,39,40 We used a 4-tiered grading scale and observed that none of the grade 1 tumors and only 2 of 33 grade 2 tumors (6.1%) metastasized. However, 19 of 75 grade 3 tumors (25%) and 53 of 108 grade 4 tumors (49%) metastasized. The results from our study require further analyses of the histopathologic parameters to define more specific prognostic factors. The strongest determining factor related to death was the presence of metastases, as expected. Nineteen of 76 patients (25%) in our study who had tumors that metastasized presented with metastasis, which is similar to the previously reported rate of 27%. 12 Almost two-thirds of all metastases were detected within the first year. During follow-up, 57 of 206 patients (28%) who presented with localized tumors developed metastatic disease. Altogether, 34% of all tumors metastasized, which is similar to the rates reported for soft tissue sarcoma in general. 30 The combination of surgery and radiation therapy, depending on surgical margins, currently is the standard treatment in Scandinavia and defines the term FIGURE 7. The local recurrence-free survival rate is illustrated for the 206 patients who presented without metastasis. At 0 years, 5 years, and 10 years, the numbers of patients at risk were 206, 85, and 35, respectively. adequate local treatment, which was the only statistically significant factor for local control in our multivariate analysis. It is well known that an adequate surgical margin is the most important factor for local control. 9,10,13,29,31,39,40 Our multivariate analysis identified no statistical correlation between local treatment and metastasis or overall survival. This finding is in line with other reports. 4,5,13,42 However, 5 of 14 subcutaneous tumors (36%) and 11 of 13 deep-seated tumors (85%) that recurred locally developed metastasis, indicating aggressive behavior. None of the 36 patients who had cutaneous tumors developed a local recurrence after treatment at a sarcoma center. Eight of those patients (22%) were referrals after 1 or several subsequent local recurrences after inadequate surgery outside of a sarcoma center, although this referral pattern did not appear to have any significant influence on the risk of metastatic recurrence or death in the cutaneous group, in accordance with previous studies. 9,19 In summary, we observed that, in contrast to subcutaneous and deep-seated soft tissue LMS, cutaneous LMS showed an indolent course of disease despite the risk of local recurrence that seemed to be dependent of adequacy of local treatment irrespective of the depth of the tumor. The strongest risk factor related to death was the presence of metastases either at the time of primary diagnosis or subsequently. The current results confirm the importance of tumor depth, tumor

290 CANCER January 15, 2007 / Volume 109 / Number 2 FIGURE 8. Local recurrence-free survival is illustrated according to the adequacy of local treatment. The thin line indicates adequate local treatment, and the thick line indicates inadequate local treatment. At 0 years, 5 years, and 10 years, the numbers of patients at risk were 206, 85, and 35, respectively. size, malignancy grade, and surgical margins in defining the outcome of patients with nonvisceral soft tissue LMS. The results from this study underline the need for centralization and multidisciplinary integration of the diagnostics and treatment of these tumors. The results from the multivariate analysis showed that prognostic factors for metastatic recurrence and, consequently, for tumor-related death differ from the prognostic factors for local recurrence. The results from our study also emphasize the need for a reproducible prognostication system for the identification of high-risk patients and the development of molecularly targeted therapies as standard treatment for these patients. REFERENCES 1. Weiss SW, Goldblum JR. Leiomyosarcoma. In: Weiss SW, Goldblum JR, eds. Enzinger and Weiss s Soft Tissue Tumors. 4th ed. Philadelphia, Pa: Mosby Harcourt; 2001:727 748. 2. Daugaard S. Current soft-tissue sarcoma classifications. Eur J Cancer. 2004;40:543 548. 3. Coindre JM. Immunohistochemistry in the diagnosis of soft tissue tumours. Histopathology. 2003;43:1 16. 4. Gustafson P, Willen H, Baldetorp B, Ferno M, Akerman M, Rydholm A. Soft tissue leiomyosarcoma. A population-based epidemiologic and prognostic study of 48 patients, including cellular DNA content. Cancer. 1992;70:114 119. 5. Gustafson P. Soft tissue sarcoma. Epidemiology and prognosis in 508 patients. Acta Orthop Scand. 1994;65(suppl 259): 1 31. 6. Fletcher CD, Gustafson P, Rydholm A, Willen H, Akerman M. Clinicopathologic re-evaluation of 100 malignant fibrous histiocytomas: prognostic relevance of subclassification. J Clin Oncol. 2001;19:3045 3050. 7. Lajer CB, Daugaard S, Hansen HS, Kirkegaard J, Holmgaard S, Christensen ME. Soft tissue sarcomas of the head and neck: a single-centre experience. Clin Otolaryngol. 2005;30:176 182. 8. Miyajima K, Oda Y, Oshiro Y, et al. Clinicopathological prognostic factors in soft tissue leiomyosarcoma: a multivariate analysis. Histopathology. 2002;40:353 359. 9. Fields JP, Helwig EB. Leiomyosarcoma of the skin and subcutaneous tissue. Cancer. 1981;47:156 169. 10. JensenML,JensenOM,MichalskiW,NielsenOS,KellerJ.Intradermal and subcutaneous leiomyosarcoma: a clinicopathological and immunohistochemical study of 41 cases. J Cutan Pathol. 1996;23:458 463. 11. Farshid G, Pradhan M, Goldblum J, Weiss SW. Leiomyosarcoma of somatic soft tissues. A tumor of vascular origin with multivariate analysis of outcome in 42 cases. Am J Surg Pathol. 2002;26:14 24. 12. Mankin HJ, Casas-Ganem J, Kim JI, Gebhardt MC, Hornicek FJ, Zeegen EN. Leiomyosarcoma of somatic soft tissues. Clin Orthop Relat Res. 2004;1:225 231. 13. Massi D, Beltrami G, Mela MM, Pertici M, Capanna R, Franchi A. Prognostic factors in soft tissue leiomyosarcoma of the extremities: a retrospective analysis of 42 cases. Eur J Surg Oncol. 2004;30:565 572. 14. Stout AP, Hill WT. Leiomyosarcoma of the superficial soft tissues. Cancer. 1958;11:844 854. 15. Dahl I, Angervall L. Cutaneous and subcutaneous leiomyosarcoma. A clinicopathologic study of 47 patients. Pathol Eur. 1974;9:307 315. 16. Wile AG, Evans HL, Romsdahl MM. Leiomyosarcoma of soft tissue: a clinicopathologic study. Cancer. 1981;48:1022 1032. 17. Berlin O, Stener B, Kindblom LG, Angervall L. Leiomyosarcomas of venous origin in the extremities. A correlated clinical, roentgenologic, and morphologic study with diagnostic and surgical implications. Cancer. 1984;54:2147 2159. 18. Hashimoto H, Daimaru Y, Tsuneyoshi M, Enjoji M. Leiomyosarcoma of the external soft tissues. A clinicopathologic, immunohistochemical, and electron microscopic study. Cancer. 1986;57:2077 2088. 19. Swanson PE, Stanley MW, Scheithauer BW, Wick MR. Primary cutaneous leiomyosarcoma. A histological and immunohistochemical study of 9 cases, with ultrastructural correlation. J Cutan Pathol. 1988;15:129 141. 20. Neugut AI, Sordillo PP. Leiomyosarcomas of the extremities. J Surg Oncol. 1989;40:65 67. 21. Oliver GF, Reiman HM, Gonchoroff NJ, Muller SA, Umbert IJ. Cutaneous and subcutaneous leiomyosarcoma: a clinicopathological review of 14 cases with reference to antidesmin staining and nuclear DNA patterns studied by flow cytometry. Br J Dermatol. 1991;124:252 257. 22. Bernstein SC, Roenigk RK. Leiomyosarcoma of the skin. Treatment of 34 cases. Dermatol Surg. 1996;22:631 635. 23. Kaddu S, Beham A, Cerroni L, et al. Cutaneous leiomyosarcoma. Am J Surg Pathol. 1997;21:979 987. 24. Reix T, Sevestre H, Sevestri-Pietri MA, Szychta P, Pietri J. Primary malignant tumors of the venous system in the lower extremities. Ann Vasc Surg. 1998;12:589 596. 25. AuroyS,ContessoG,SpatzA,etal.Primarycutaneousleiomyosarcoma: 32 cases. Ann Dermatol Venereol. 1999;126:235 242.

Clinical Course of Leiomyosarcoma/Svarvar et al. 291 26. Porter CJ, Januszkiewicz JS. Cutaneous leiomyosarcoma. Plast Reconstr Surg. 2002;109:964 967. 27. Bellezza G, Sidoni A, Cavaliere A, Scheibel M, Bucciarelli E. Primary cutaneous leiomyosarcoma: a clinicopathological and immunohistochemical study of 7 cases. Int J Surg Pathol. 2004;12:39 44. 28. Tsutsumida A, Yoshida T, Yamamoto Y, Itoh T, Minakawa H, Sugihara T. Management of superficial leiomyosarcoma: a retrospective study of 10 cases. Plast Reconstr Surg. 2005;116: 8 12. 29. 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. 30. 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. 31. Koea JB, Leung D, Lewis JJ, Brennan MF. Histopathologic type: an independent prognostic factor in primary soft tissue sarcoma of the extremity? Ann Surg Oncol. 2003;10:432 440. 32. Bauer HC, Alvegard TA, Berlin O, et al. The Scandinavian Sarcoma Group Register. Acta Orthop Scand. 1999;70(suppl 285):41 44. 33. Meis-Kindblom JM, Bjerkehagen B, Bohling T, et al. Morphologic review of 1000 soft tissue sarcomas from the Scandinavian Sarcoma Group (SSG) Register. The peer-review committee experience. Acta Orthop Scand. 1999;70(suppl 285):18 26. 34. Collin CF, Friedrich C, Godbold J, Hajdu S, Brennan MF. Prognostic factors for local recurrence and survival in patients with localized extremity soft-tissue sarcoma. Semin Surg Oncol. 1988;4:30 37. 35. Broders AC, Hargrave R, Meyerding HW. Pathological features of soft tissue fibrosarcoma. Surg Gynecol Obstet. 1939;69:267 280. 36. Angervall L, Kindblom LG, Rydholm A, Stener B. The diagnosis and prognosis of soft tissue tumors. Semin Diagn Pathol. 1986;3:240 258. 37. Hornick JL, Fletcher CD. Criteria for malignancy in nonvisceral smooth muscle tumors. Ann Diagn Pathol. 2003;7:60 66. 38. Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop Relat Res. 1980;153:106 120. 39. Zagars GK, Ballo MT, Pisters PW, et al. Prognostic factors for patients with localized soft-tissue sarcoma treated with conservation surgery and radiation therapy: an analysis of 225 patients. Cancer. 2003;97:2530 2543. 40. Trovik CS, Bauer HC, Alvegard TA, et al. Surgical margins, local recurrence and metastasis in soft tissue sarcomas: 559 surgically-treated patients from the Scandinavian Sarcoma Group Register. Eur J Cancer. 2000;36:710 716. 41. Peabody TD, Monson D, Montag A, Schell MJ, Finn H, Simon MA. A comparison of the prognoses for deep and subcutaneous sarcomas of the extremities. J Bone Joint Surg Am. 1994;76-A:1167 1173. 42. Trovik CS. Local recurrence of soft tissue sarcoma. A Scandinavian Sarcoma Group project. Acta Orthop Scand. 2001; 72(suppl 300):1 31.