Grading of soft tissue sarcomas: the challenge of providing precise information in an imprecise world.

By identifying patients at greatest risk for distant metastasis and, hence, most likely to benefit from adjuvant therapy, the grading of sarcomas has been one of the most important contributions pathologists have made to the treatment of sarcomas. Over the years, many grading schemes have been proposed and validated as efficacious. The three-tier system proposed by the French Federation of Cancer Centres is precisely defined, easy to use, and is the most widely employed. However, no system performs perfectly on all sarcomas. Sarcomas that do not lend themselves well to grading include (i) those in which grade provides no incremental information over histological subtypes (e.g. well-differentiated liposarcoma/atypical lipomatous neoplasm, Ewing's sarcoma); (ii) tumours traditionally considered "ungradable" (e.g. epithelioid sarcoma, clear cell sarcoma, angiosarcoma); and (iii) sarcomas that customarily have been graded but in which grade has recently been shown not to correlate well with outcome (e.g. malignant peripheral nerve sheath tumour). Consequently, several sarcoma-specific risk stratification schemes have been proposed. The future may well witness a synthesis of these two approaches. Nomograms, which incorporate clinical, histological and demographic findings, have proved accurate in predicting disease-specific survival in sarcomas. Diagnosis and grading are increasingly based on tissue obtained by core needle biopsy, which poses new challenges for pathologists, particularly if neoadjuvant therapy is to be given. Grading on needle biopsies may require a two-tier grading system (i.e. low versus high grade) and a close dialogue with clinicians to resolve ambiguities.

Lingual alveolar soft part sarcoma; 14 cases: novel clinical and morphological observations.

AIMS: Alveolar soft part sarcoma (ASPS) is a rare sarcoma in the buttocks or thigh of young adults, often with metastases to lung, brain, or bone. This study examines the morphological and clinical features of lingual ASPS. METHODS AND RESULTS: Fourteen cases, in eight males and six females (ages 3-21 years, median 5 years), ranged from 8 to 50 mm, median 25 mm. All tumours were intramuscular, circumscribed and multinodular. Tumours from all but the oldest patient exhibited a predominantly solid (non-alveolar) growth pattern. Vascular invasion was common. Crystals varied in number from none or extremely rare to nearly 100% of tumour cells. Immunohistochemical results: Fifty percent desmin positive, all focally smooth muscle antigen (SMA) positive; negative for vimentin, neural/melanocytic, myoid, histiocytic, and epithelial markers. All tumours were surgically excised; only two patients received chemotherapy. Follow-up on 10 patients showed that all patients were alive without disease (4-32, median 22 years). Only one patient had a microscopic metastasis to lung (3 years) but was without disease at 11 years. CONCLUSIONS: Lingual ASPS is a tumour of childhood with a distinctive, predominantly solid growth pattern. Despite typical vascular invasion, the early diagnosis and small tumour size may explain its relatively good outcome.

Intra-abdominal synovial sarcoma: a clinicopathological study.

AIMS: To evaluate a series of synovial sarcomas arising in the abdomen, pelvic cavity, or retroperitoneum. Synovial sarcoma is rare within the abdomen. In this location, it can be confused with other biphasic tumours and with other spindle and round cell sarcomas. METHODS AND RESULTS: Cases were retrieved from archives. There were 11 intra-abdominal tumours among 300 synovial sarcomas in two referral practices (3.7%). Three were pelvic (two midline, one sidewall) and eight were retroperitoneal. They occurred in six males and five females aged from 25 to 75 years (mean 49 years, median 46 years), and ranged in diameter from 65 to 470 mm (mean 210 mm, median 150 mm). Six examples were biphasic, five were monophasic and seven had poorly differentiated areas. Monophasic tumours displayed at least one epithelial marker. One biphasic tumour had a SYT-SSX2 fusion gene. Seven sarcomas were high-grade and four of intermediate grade malignancy. Follow-up data were available in 10 patients. In all but one case, tumour recurred or metastasized within the abdomen. The pelvic sarcomas also metastasized outside the abdomen. Eight of 10 patients (80%) died of disease with survival from 4 to 36 months (mean 17 months, median 18 months). Two patients were alive with disease at 43 and 48 months. CONCLUSIONS: Synovial sarcomas rarely arise within the abdomen and pelvis. They occur mainly in middle age, attain a large size, are difficult to excise and recur locally. Pelvic tumours metastasize distantly. Retroperitoneal tumours remain confined to the abdomen and, unlike synovial sarcomas elsewhere, do not metastasize remotely, although mortality is high.

Do leiomyomas of deep soft tissue exist? An analysis of highly differentiated smooth muscle tumors of deep soft tissue supporting two distinct subtypes.

There is a prevailing view that leiomyomas of deep soft tissue are rare or nonexistent, but there are limited data on this subject in the form of large clinical studies with long follow-up information. We reviewed 36 consultation cases that had been diagnosed as leiomyoma or probable leiomyoma based on absence of nuclear atypia, necrosis, and no/minimal mitotic activity. Follow-up information was obtained to determine whether these stringent histologic criteria could identify a biologically benign group of smooth muscle tumors of deep soft tissue. The tumors occurred in two distinct locations. The first (n = 13) occurred in deep somatic soft tissue of the lower extremity (7), upper extremity (2), trunk (2), axilla (1), and back (1) and affected the sexes equally (7 male, 6 female). Composed of a circumscribed mass of mature smooth muscle cells, they were frequently calcified with a mean mitotic activity of <1 mitosis/50 high power fields (HPF) (range 1-4 mitoses/50 HPF). Estrogen receptor and progesterone receptor proteins were negative in the three cases tested. No tumors recurred or metastasized (mean follow-up 58.7 months, range 5-97 months). The second group (n = 23) occurred within the retroperitoneum (20) or abdominal cavity (3) of women (1 male, 22 female). Resembling uterine leiomyomas, they were always distinct from the uterus, occasionally multiple (n = 4), and sometimes occurred up to years after hysterectomy (n = 3). Four cases occurred with synchronous uterine leiomyomas. In the six cases tested, five of six were positive for the estrogen receptor protein and all were positive for progesterone receptor protein. Mean mitotic activity was 1 mitosis/50 HPF (range <1-10 mitoses/50 HPF). None developed metastasis within the follow-up period (mean 42.5 months, range 6-120 months); one tumor with a positive margin recurred at 10 months. We conclude that clinically benign smooth muscle tumors of deep soft tissue are rare but can be identified using stringent histologic criteria. They comprise two distinct subtypes: leiomyomas of somatic soft tissue and retroperitoneal-abdominal leiomyomas. The latter probably arise from hormonally sensitive smooth muscle. Although similar to uterine leiomyomas, they are located at sites removed from the uterus and are likely independent soft tissue primaries rather than parasitic leiomyomas of the uterus. We suggest that these two groups of smooth muscle tumors be diagnostically approached in a site-specific fashion.

CD31 expression in intratumoral macrophages: a potential diagnostic pitfall.

CD31 (platelet endothelial adhesion molecule, PECAM-1) is generally regarded to be the most sensitive and specific endothelial marker in paraffin sections. We have recently encountered several cases in which intratumoral CD31-positive macrophages were misinterpreted as evidence of a vascular sarcoma. We therefore reviewed our last 1950 consultation cases with respect to cases in which CD31 immunostains were performed, to determine the frequency of CD31 expression in macrophages in formalin-fixed, paraffin-embedded tissue and how often the presence of these cells was a source of diagnostic confusion. CD31 immunohistochemistry had been performed on 59 of 1950 (3%) of cases. These 59 cases consisted of both vascular (20 cases) and nonvascular tumors (39 cases). CD31-positive macrophages were distinguished from endothelial or tumor cells by correlation with the morphologic features and the immunohistochemical staining pattern of the cells of interest. In no case was CD31 positivity seen in the lesional cells of a nonvascular tumor. CD31-positive macrophages were identified in 48 of 59 (81%) cases. CD31-positive macrophages were present in 34 of 39 (87%) nonvascular tumors. A vascular tumor was diagnosed or favored by the referring pathologist in 15 of these 39 cases (38%). In 14 of these 15 cases CD31 immunostains were performed by the referring pathologist; 13 (93%) showed CD31-positive macrophages. In 4 of these 14 cases (29%) the misdiagnosis of a vascular tumor was based primarily or in part on the misinterpretation of CD31-positive macrophages as tumor cells. In all cases with CD34 and CD68 immunostains, the CD31-positive macrophages were CD34 negative and CD68 positive. We conclude that CD31 expression is very common in macrophages. Misinterpretation of CD31-positive macrophages as tumor cells may result in the erroneous diagnosis of a primary vascular neoplasm. Recognition of the characteristic granular, membranous pattern of CD31 expression in macrophages and careful distinction from tumor cells should allow the accurate interpretation of CD31 immunohistochemistry in possible vascular neoplasms. CD31 may also be useful as a nonlysosomal marker of macrophages in formalin-fixed, paraffin-embedded sections.

Consultative (expert) second opinions in soft tissue pathology. Analysis of problem-prone diagnostic situations.

We reviewed 500 consecutive soft tissue lesions referred for expert consultation to determine types of lesions and/or situations in which major discrepancies occur. Of 266 cases (53.2%) accompanied by a diagnosis, essential agreement with the second opinion was noted in 68%, minor discrepancy in 7%, and major discrepancy in 25%. The 65 major discrepancies were distributed proportionally to the referring sources and could be divided into 4 groups: benign mesenchymal lesions diagnosed as sarcomas (45%), sarcomas diagnosed as benign tumors (23%), nonmesenchymal lesions diagnosed as sarcoma (20%), and major grading discrepancies (12%). Relatively few lesions accounted for a major proportion of major discrepancies. Problematic lesions were lipoma and fasciitis and their variants and desmoplastic-neurotropic melanoma. Needle biopsy specimens were somewhat more likely to be associated with a discrepant opinion. With the exception of nonmesenchymal lesions, the diagnosis for all major discrepant cases could be made on the basis of the H&E-stained slides, suggesting that failure to perform immunostains did not account for discrepancies. Lack of familiarity with rare or unusual lesions is probably more significant in explaining diagnostic discrepancies than is the increasing use of needle biopsy or the failure to perform immunohistochemical analysis.

Diffuse dermal angiomatosis of the breast: response to isotretinoin.

Dermal angiomatosis of the breast is an extremely rare disorder of unknown origin characterized by increased angiomatosis and ulceration. We report a case of a young woman whose disorder responded to isotretinoin. Our findings have potential relevance to the treatment of skin disorders in which ulceration is a prominent feature.

The generation and characterization of a cell line derived from a sporadic renal angiomyolipoma: use of telomerase to obtain stable populations of cells from benign neoplasms.

Angiomyolipomas are benign tumors of the kidney derived from putative perivascular epithelioid cells, that may undergo differentiation into cells with features of melanocytes, smooth muscle, and fat. To gain further insight into angiomyolipomas, we have generated the first human angiomyolipoma cell line by sequential introduction of SV40 large T antigen and human telomerase into human angiomyolipoma cells. These cells show phenotypic characteristics of angiomyolipomas, namely differentiation markers of smooth muscle (smooth muscle actin), adipose tissue (peroxisome proliferator-activator receptor gamma, PPARgamma), and melanocytes (microophthalmia, MITF), thus demonstrating that a single cell type can exhibit all of these phenotypes. These cells should serve as a valuable tool to elucidate signal transduction pathways underlying renal angiomyolipomas.

Expression of Fli-1, a nuclear transcription factor, distinguishes vascular neoplasms from potential mimics.

Fli-1 protein, a member of the ETS family of DNAbinding transcription factors, is involved in cellular proliferation and tumorigenesis. Approximately 90% of Ewing's sarcoma/primitive neuroectodermal tumors (ES/PNET) have a specific translocation, t(11;22)(q24;q12), which results in fusion of EWS to Fli-1, and production of an EWS-Fli-1 fusion protein. We have recently shown that immunohistochemistry for the carboxy terminal of Fli-1 protein is sensitive and highly specific for the diagnosis of ES/PNET. In our earlier study we noted that among normal tissues only endothelial cells and small lymphocytes expressed Fli-1. Fli-1 expression in vascular neoplasms has not been previously studied. Formalin-fixed paraffin-embedded tissue from 54 vascular tumors and 75 nonvascular tumors were immunostained for Fli-1 (1:120, Sc 356, Santa Cruz Biotechnology, Santa Cruz, CA), after steam heat-induced epitope retrieval. Only cases with >10% of cells showing nuclear staining were accepted as positive. Cases without positive internal controls (endothelium and small lymphocytes) were not scored. Positive internal controls were present in 122 of 129 cases (95%). One vascular tumor (Kaposi's sarcoma) and 7 nonvascular tumors (2 epithelioid sarcomas and 5 carcinomas) without internal controls were not scored. Fli-1 was expressed by 50 of 53 vascular tumors scored (94%), including 20 of 22 angiosarcomas, 11 of 12 hemangioendotheliomas, 7 of 7 hemangiomas, and 12 of 12 Kaposi's sarcomas. In contrast, Fli-1 expression was absent in the 68 nonvascular tumors scored (0 of 68), including 16 sarcomas, 7 melanomas, and 45 carcinomas. The results of this study strongly suggest a role for Fli-1 as a novel marker of both benign and malignant vascular tumors. The sensitivity (94%) and specificity (100%) of Fli-1 with regards to the cases evaluated in this study equal or exceed those of the established vascular markers, CD31, CD34, and von Willebrand factor. As the first nuclear, rather than cytoplasmic or membranous marker of endothelium, Fli-1 immunostaining also generally lacks cytoplasmic staining artifacts that are the result of endogenous peroxidases or biotin.

Gardner-associated fibromas (GAF) in young patients: a distinct fibrous lesion that identifies unsuspected Gardner syndrome and risk for fibromatosis.

Gardner syndrome (GS), caused by mutations in the adenomatous polyposis coli (APC) gene, is characterized by polyposis coli, osteomas, and various soft-tissue tumors. If undetected or untreated, virtually all patients develop colonic carcinoma at a young age. Early detection, while essential, can be difficult because of attenuated phenotypes or spontaneous mutations. We present the clinicopathologic features of 11 identical fibromatous lesions that we have termed Gardner-associated fibroma (GAF), which not only appear to be a part of the spectrum of lesions associated with GS but, in some cases, represent the sentinel event leading to its detection. The GAFs occurred in 11 patients (5 boys and 6 girls; age range, 3 months-14 years), were solitary (n = 7) or multiple (n = 4), and occurred in the superficial and deep soft tissues of the paraspinal region (n = 7), back (n = 3), face (n = 2), scalp (n = 2), chest wall (n = 2), thigh (n = 1), neck (n = 1), and flank (n = 1). Histologically, GAFs resemble nuchal-type fibromas (NFs), consisting of thick, haphazardly arranged collagen bundles between which are found occasional bland fibroblasts, and having margins that frequently engulf surrounding structures including adjacent fat, muscle and nerves. After surgical excision, four patients developed recurrences that were classic desmoid fibromatoses (DFs). In one patient with multiple GAFs, one lesion had the features of GAF and DF in the absence of surgical trauma. A family history of GS or polyposis (n = 6) or DF (n = 1) was known at the time of surgery in seven patients. In three patients, the diagnosis of GAF resulted in the diagnosis of unsuspected APC in older family members, with the detection of an occult colonic adenocarcinoma in one parent. In the family of the remaining patient, no stigmata of GS were present. Genetic analysis of this child was performed to investigate the presence of a spontaneous (new) mutation; however, no abnormalities were detected. The significance of GAF is that it serves as a sentinel event for identifying GS kindreds, including those with a high risk for the development of DF, and it may potentially identify children with spontaneous mutations of the APC gene. Because NFs and GAFs resemble one another, we suggest that a subset of NF occurring in multiple sites, unusual locations, or children may be GAF.

Differential expression of active mitogen-activated protein kinase in cutaneous endothelial neoplasms: implications for biologic behavior and response to therapy.

BACKGROUND: Tumors of endothelium range from benign hemangiomas of infancy to highly malignant angiosarcomas of the elderly. Hemangiomas are the most common tumors in infants and may affect up to 10% of all children. The biologic behavior of these lesions ranges from self-resolving, in the case of hemangiomas and pyogenic granulomas, to lethal metastatic neoplasms in the case of angiosarcoma. Although the clinical outcomes of these diseases are easily distinguished, the biologic basis for these differences is not well understood. Activation of mitogen-activated protein kinase (MAPK) is an important signal transduction mechanism that may predict response of a tumor to chemotherapy. OBJECTIVE: Our purpose was to examine expression of phosphorylated (activated) MAPK in hemangiomas of infancy, pyogenic granulomas, hemangioendotheliomas, and angiosarcomas to determine whether phosphorylated MAPK was expressed in endothelial tumors. In addition, we examined endothelial tumors of infectious origin, Kaposi's sarcoma, and verruga peruana. METHODS: Skin sections from benign and malignant endothelial tumors, including hemangioma of infancy, angiosarcoma, and infectious endothelial lesions (Kaposi's sarcoma, verruga peruana) were stained with an antibody specific for phosphorylated MAPK. RESULTS: We demonstrated strong expression of phosphorylated MAPK in benign endothelial tumors, including capillary hemangioma of infancy and pyogenic granuloma, and greatly decreased expression in angiosarcoma. In addition, infectious endothelial tumors stained strongly with this antibody, similar to benign tumors. The presence of immunoreactive phosphorylated MAPK appears to be inversely correlated with degree of malignancy. CONCLUSION: We demonstrate that the use of antibodies specific for signal transduction pathways is feasible in paraffin-fixed tissue. Thus the activity of a given signal transduction pathway can be ascertained in a biopsy specimen. Immunohistochemistry for phosphorylated MAPK may help the pathologist distinguish benign from malignant endothelial processes and thus guide therapy.

Atypical and malignant glomus tumors: analysis of 52 cases, with a proposal for the reclassification of glomus tumors.

Occasional glomus tumors display unusual features, such as large size, deep location, infiltrative growth, mitotic activity, nuclear pleomorphism, and necrosis. Although a small number of purportedly malignant glomus tumors have been described, histologic criteria for malignancy in glomus tumors have never been elaborated. The authors studied 52 unusual glomus tumors (retrieved from their consultation files) previously diagnosed as "atypical" or "malignant" by virtue of nuclear atypia, infiltrative growth, or mitotic activity. They evaluated size, depth, growth pattern, cellularity, nuclear grade, number of mitotic figures per 50 high-power fields (HPF), atypical mitotic figures, vascular space involvement, and necrosis to define criteria for malignancy in glomus tumors. Estimated relative risk was calculated and the Fisher exact test was used for statistical analysis. The 27 female patients and the 25 male patients ranged in age from 8 to 83 years (median age, 43 years). The tumors measured from 0.2 to 12 cm (median size, 2 cm) and occurred predominantly in the extremities, in both the superficial (n = 35) and deep (n = 17) soft tissues. Atypical features were usually observed centrally with a rim of benign-appearing glomus tumor. Follow-up information (n = 35; range, 5 months-23 years; mean 5.5 years) showed seven recurrences, eight metastases, and seven deaths from disease. Five-year cumulative metastatic risk increased significantly for tumors with a deep location (p = 0.005), with a size of more than 2 cm (p = 0.004), and with atypical mitotic figures (p = 0.004). Mitotic activity of more than 5 mitoses/50 HPF, high cellularity, the presence of necrosis, and moderate to high nuclear grade approached but did not reach significance. High nuclear grade alone, infiltrative growth, and vascular space involvement were not associated with metastasis. The authors propose the following classification scheme and criteria. Malignant glomus tumor: Tumors with a deep location and a size of more than 2 cm, or atypical mitotic figures, or moderate to high nuclear grade and > or =5 mitotic figures/50 HPF. Symplastic glomus tumor: Tumors with high nuclear grade in the absence of any other malignant feature. Glomus tumor of uncertain malignant potential: Tumors that lack criteria for malignant glomus tumor or symplastic glomus tumor but have high mitotic activity and superficial location only, or large size only, or deep location only. Glomangiomatosis: Tumors with histologic features of diffuse angiomatosis and excess glomus cells. Using this classification scheme, metastasis was observed in 38% of tumors fulfilling the criteria for malignancy. In contrast, metastatic disease was not seen in any specimen classified as symplastic glomus tumor, glomus tumor of uncertain malignant potential, or glomangiomatosis.

Microphthalmia transcription factor and melanoma cell adhesion molecule expression distinguish desmoplastic/spindle cell melanoma from morphologic mimics.

Desmoplastic/spindle cell melanoma is a rare variant of melanoma. A number of factors complicate the diagnosis of desmoplastic/spindle cell melanoma, including the variable absence of a lentiginous component, its spindle cell morphology, and its many morphologic mimics, including scars, malignant peripheral nerve sheath tumor, neurofibroma, atypical fibroxanthoma, and spindled carcinoma. The immunohistochemical confirmation of desmoplastic/spindle cell melanoma may also be difficult, because the majority of tumors are negative for specific melanocytic markers such as HMB-45 and Melan-A, despite their usual expression of S-100 protein. Two new and potentially promising melanocytic markers, microphthalmia transcription factor (MiTF) and melanoma cell adhesion molecule (Mel-CAM), have been shown to be sensitive markers of epithelioid melanoma, but have not been tested in desmoplastic/spindle cell melanoma or in other rare melanocytic neuroectodermal tumors such as clear cell sarcoma. We immunostained 79 tumors (20 desmoplastic/spindle cell melanomas, 10 scars, 10 neurofibromas, 12 malignant peripheral nerve sheath tumors, 10 atypical fibroxanthomas, 10 clear cell sarcomas, 3 melanotic schwannomas, and 4 cellular blue nevi) for MiTF and Mel-CAM. MiTF expression was seen in 11 of 20 desmoplastic/spindle cell melanomas, 0 of 10 scars, 2 of 10 neurofibromas, 0 of 12 malignant peripheral nerve sheath tumors, 1 of 10 atypical fibroxanthomas, 7 of 10 clear cell sarcomas, 3 of 3 melanotic schwannomas, and 3 of 4 cellular blue nevi. Mel-CAM expression was present in 14 of 17 desmoplastic/spindle cell melanomas, 0 of 10 scars, 4 of 10 neurofibromas, 3 of 11 malignant peripheral nerve sheath tumors, 0 of 10 atypical fibroxanthomas, 9 of 10 clear cell sarcomas, 3 of 3 melanotic schwannomas, and 0 of 4 cellular blue nevi. MiTF and Mel-CAM were coexpressed in 6 of 17 desmoplastic/spindle cell melanomas and in no other tumor. Regarding desmoplastic/spindle cell melanoma, scar, neurofibroma, malignant peripheral nerve sheath tumor, and atypical fibroxanthoma, the sensitivity and specificity of MiTF for desmoplastic/spindle cell melanoma were 55% and 91%, respectively. For this same group of tumors, Mel-CAM had a sensitivity of 82% and a specificity of 83%. We conclude that the sensitivity and specificity of MiTF for desmoplastic melanoma equals or exceeds that of such markers as HMB-45 or Melan-A, and that MiTF should be part of the initial immunohistochemical panel for the work-up of such cases. Mel-CAM, while very sensitive, is relatively nonspecific, because it is also expressed in a variety of mesenchymal tumors and carcinomas. Mel-CAM is best reserved for cases morphologically suspected to be desmoplastic/ spindle cell melanoma, in which S-100 is positive and MiTF and other melanocytic markers are negative. These markers may also be helpful in certain other differential diagnoses, such as distinguishing clear cell sarcomas from epithelioid malignant peripheral nerve sheath tumors.

Immunohistochemical detection of FLI-1 protein expression: a study of 132 round cell tumors with emphasis on CD99-positive mimics of Ewing's sarcoma/primitive neuroectodermal tumor.

The histologic and immunohistochemical differentiation of Ewing' s sarcoma/primitive neuroectodermal tumor (ES/PNET) from other small, blue, round cell tumors may be difficult. Despite initial promise, CD99 (MIC2) has not proven to be a specific marker. Approximately 90% of ES/PNET have a specific t(11; 22)(q24;q12) that results in fusion of the EWS and FLI-1 genes, and overexpression of FLI-1 protein. A recent study has shown immunohistochemical FLI-1 expression in five of seven of the ES/PNET cases tested. We evaluated FLI-1 expression in 132 well-characterized small, blue, round cell tumors. All tumors were immunostained for FLI-1 (1:40, Sc 356 polyclonal, Santa Cruz Biotechnology) using steam heat for epitope retrieval. Only nuclear staining was accepted as positive. Endothelial cells were strongly positive in all cases and served as an internal control. In many cases, a subset of lymphocytes also stained positive. No staining was seen in any other normal tissue. FLI-1 expression was seen in 29 of 41 (71%) ES/PNET, 7 of 8 (88%) lymphoblastic lymphomas, 0 of 8 poorly differentiated synovial sarcomas (PDSS), 0 of 32 rhabdomyosarcoma (RMS), 0 of 30 neuroblastomas, 0 of 8 esthesioneuroblastomas, 0 of 3 Wilms' tumors, 0 of 1 mesenchymal chondrosarcoma, and in 1 of 1 desmoplastic round cell tumor. This last case was known to have an EWS/WT-1 fusion. Although the EWS/FLI-1 fusion gene is specific for ES/PNET, FLI-1 protein expression is not. Significantly, the great majority of lymphoblastic lymphomas (also CD99-positive) are strongly FLI-1-positive. Immunohistochemical detection of FLI-1 may be valuable in confirming the diagnosis of ES/ PNET in cases in which molecular genetic evaluation is not feasible. FLI-1 protein expression is also helpful in distinguishing ES/PNET from other tumors that may be CD99-positive, such as PDSS and RMS. It is not surprising that some ES/ PNET are FLI-1-negative, because not all ES/PNET have the classic EWS/FLI-1, and some cases of ES/PNET may produce either low levels of protein or idiotypically different protein.

Cutaneous angiosarcoma arising in a gouty tophus: report of a unique case and a review of foreign material-associated angiosarcomas.

Rare cases of angiosarcoma have been reported to arise in the setting of retained foreign material or in association with arteriovenous fistulae. No previous case of angiosarcoma, or any other malignancy, has been reported to arise with a gouty tophus. We present a case of an 86-year-old man with a high-grade angiosarcoma that arose within a long-standing tophus.

Low-grade fibromyxoid sarcoma and hyalinizing spindle cell tumor with giant rosettes: a clinicopathologic study of 73 cases supporting their identity and assessing the impact of high-grade areas.

Low-grade fibromyxoid sarcoma (LGFMS) is a rare sarcoma characterized by bland histologic features and a paradoxically aggressive clinical course. The hyalinizing spindle cell tumor with giant rosettes (HSCT) is a closely related tumor characterized by the presence of giant collagen rosettes. Only a single example of a metastasizing HSCT has been reported. A small subset of both LGFMS and HSCT display areas of increased cellularity and atypia which qualify as intermediate- to high-grade sarcoma; the significance of these features has not been definitively assessed. We present the clinicopathologic features of 77 cases of LGFMS and HSCT to determine the degree of overlap of these two lesions, their biologic behavior, and the significance of the occasional presence of intermediate- to high-grade sarcoma within both. The patients (33 female, 40 male) ranged from 3 to 78 years of age (median, 34 yrs). Fourteen cases occurred in patients less than 18 years of age. The tumors measured from 1 to 23 cm (median, 4.5 cm) and occurred predominantly in the trunk and lower extremities in both the deep (66 cases) and superficial (7 cases) soft tissues. In 15 cases, the tumor was present > 1 year before diagnosis. All tumors showed predominantly the typical hypocellularity and bland cytologic features of typical LGFMS; however, areas of hypercellularity and nuclear enlargement and hyperchromatism were identified in 12 of 73 (16%) and 7 of 73 (10%), respectively. Necrosis and mitotic activity >5/50 high-powered fields (HPF) were present in 6 of 73 (8%) and 5 of 73 (7%), respectively. Epithelioid areas were present in 33 of 73 (45%) and rosettes in 22 of 73 (30%). Follow up (54 cases; range, 2-192 mos; median, 24 mos; mean, 38 mos) showed 5 recurrences, 3 metastases, and 1 death. The diagnosis of LGFMS or HSCT was made prospectively in 51 patients; none had metastatic disease. Two of the metastatic tumors were LGFMS and one was a HSCT. LGFMS may occur more often in the pediatric population and show a much wider histologic spectrum than previously thought. A significant number of LGFMS possess inconspicuous collagen rosettes characteristic of HSCT, indicating that these two tumors are ends of a common spectrum rather than distinct entities. HSCT, like LGFMS, are low-grade sarcomas with metastatic potential. The presence of focal areas of intermediate- to high-grade sarcoma does not portend a worse outcome in the short term. The better prognosis reflected in this study compared with previous ones might reflect the fact that all were initially diagnosed as sarcomas and treated with aggressive surgery. The fact that the only three patients to develop metastatic disease were patients whose LGFMS or HSCT was identified retrospectively supports this concept.

Clear cell myomelanocytic tumor of the falciform ligament/ligamentum teres: a novel member of the perivascular epithelioid clear cell family of tumors with a predilection for children and young adults.

The perivascular epithelioid cell family of tumors (PEComas), defined by their co-expression of melanocytic and muscle markers, includes angiomyolipoma, lymphangioleiomyoma, and clear cell "sugar" tumors of the lung, pancreas, and uterus. We present seven cases of a unique and previously unrecognized tumor of children and young adults, which represents a new addition to the PEComa group of tumors. Culled from three institutions over a 50-year period, all cases occurred in or immediately adjacent to the ligamentum teres and falciform ligament. Six patients were female and one male; their ages ranged from 3 to 21 years (median, 11 yrs). Tumor sizes ranged from 5 to 20 cm (median, 8 cm). All cases consisted of clear to faintly eosinophilic spindled cells arranged in fascicular and nested patterns. The cells had small but distinct nucleoli and low mitotic activity. Immunohistochemically, all cases were positive with antibodies to gp100 protein (HMB-45) and negative for S-100 protein. In three of the seven cases studied immunohistochemically, the tumors expressed smooth muscle actin, melan-A, microphthalmia transcription factor (MiTF), and myosin, but not desmin. No expression of the TSC2 gene product, tuberin, was seen in three cases. One case studied cytogenetically disclosed a t(3;10). Follow-up data, available in six of seven cases (median duration, 18 mos), showed five patients to be free of disease and one to have a radiographically presumed lung metastasis. We think these tumors comprise a new entity for which we propose the term "clear cell myomelanocytic tumor of the falciform ligament/ligamentum teres." The differential diagnosis of these tumors includes clear cell sarcoma of tendons and aponeuroses, leiomyosarcoma, and angiomyolipoma.

Sarcomas arising in dermatofibrosarcoma protuberans: a reappraisal of biologic behavior in eighteen cases treated by wide local excision with extended clinical follow up.

There is a prevailing view that sarcomas arising in dermatofibrosarcoma protuberans (DFSP) have a higher risk of metastasis than ordinary DFSP, but these data are based on cases with variable and often suboptimal treatment. There has not been a large study of sarcomas arising in DFSP in which all cases were treated by wide local excision, thereby arguably altering outcome. Clinicopathologic features of 18 cases of sarcomas arising in DFSP treated by wide local excision and having follow up of at least 5 years were analyzed. An estimate of the proportion of sarcoma and DFSP was made. The number of mitotic figures and degree of CD34 immunoreactivity were assessed in each case. The cohort included 13 females and 5 males (age, 23-87 yrs; median, 47 yrs). The tumors involved the trunk (7), scalp (4), extremities (4), and inguinal region (3), and ranged from 1.5 to 7 cm (median, 4 cm). Sarcoma occurred de novo in 15 cases and in a recurrence in three. Sarcomas resembled fibrosarcoma (17) or malignant fibrous histiocytoma (1) and occupied between 20% and 80% of the tumor (median, 60%). Mitotic activity ranged from 2 to 16 per 10 high-power field (HPF; median 7 per 10 HPF) in the sarcomatous component and 0 to 3 per 10 HPF (median, 1 per 10 HPF) in the DFSP component. All tumors expressed CD34 in the DFSP component but only nine (50%) in the sarcomatous component. All patients were treated by wide local excision with negative margins; three additionally received radiation. Four patients (22%) developed recurrences, but none developed metastasis during the follow-up period of 62 months to 17 years (median, 81.5 mos). In contrast to earlier studies, we demonstrate that patients with sarcomas arising in DFSP do not have an increased risk of distant metastasis within a 5-year follow-up period, provided they are treated by wide local excision with negative margins. This probably reflects the fact that wide local excision results in eradication of local tumor, thereby eliminating the source for subsequent dissemination. However, we cannot completely exclude the possibility that tumors in which clear margins are achieved represent a less aggressive subset, as has been suggested for high-grade extremity sarcomas. Previous studies showing increased metastasis for sarcomas arising in DFSP should be re-evaluated to determine if, with treatment stratification, metastatic rate varies.