—  SYMPOSIUM #42  —

Soft Tissue Tumors of Borderline Malignancy
Moderators: Dr. Cyril Fisher and Dr. John R. Goldblum

Section 5 - Fibrohistiocytic Tumors of Intermediate Biologic Potential: An Update

Christopher D.M. Fletcher
Department of Pathology
Brigham and Women's Hospital and Harvard Medical School
Boston , MA


Introduction

As the category of so-called fibrohistiocytic tumors is progressively eroded, then the only lesions which would strictly fit the title of this talk, based on the 2002 WHO Classification, would be plexiform fibrohistiocytic tumor and giant cell tumor of soft tissue. The subsequent WHO Classification of skin tumors might possibly have also included dermatofibrosarcoma protuberans (and giant cell fibroblastoma) although this is not explicitly stated. Nevertheless, the title of this presentation allows coverage of a variety of lesions which are often regarded (even if incorrectly) as fibrohistiocytic, thereby enabling a brief review of their current nosologic status.

So-called Fibrohistiocytic Tumors of Intermediate (Rarely Metastasizing) Potential

Plexiform fibrohistiocytic tumor [1, 2, 3] is an uncommon tumor which arises most often in children or adolescents, with a notable predominance in females. The upper limb and limb girdle are most often affected, although occasional cases arise at almost any location. The majority of cases are dermal/subcutaneous, deeper lesions being notably infrequent, and most measure < 3 cm. Local recurrence is frequent after marginal or incomplete excision. In the original, largest series, just 2 cases were associated with nodal metastasis, [1] but subsequent experience has shown that lung metastasis may occur, [3] although the frequency of this phenomenon outside referral centers is unknown.


These lesions are typically centered on the dermal/subcutaneous junction and have a strikingly plexiform/infiltrative architecture. They are composed, in variable proportions, of fascicles of cytologically uniform, palely eosinophilic myofibroblasts (resembling fibromatosis) and nodular aggregates of histocytoid cells, often admixed with osteoclastic giant cells. In cases with a scant histiocytoid component, the latter cells may be scattered singly or in only very small clusters within the myofibroblastic component. There is no atypia or pleomorphism but vascular invasion may be identified in up to 20% of cases, perhaps accounting for the metastatic tendency. Immunostains show SMA positivity in the spindle cell component and variable CD68 positivity in the histiocytoid component, mainly in the osteoclastic cells.

The superficial location, plexiform architecture and at least focally evident histiocytoid component allow distinction from fibromatosis. Some cases may also be confused with cellular neurothekeoma, but the latter is rarely plexiform and lacks histiocytoid nodules, instead being composed of nests and bundles of palely eosinophilic ovoid-to-spindled cells between hyaline dermal collagen bundles.

Giant cell tumor of soft tissue, [4, 5, 6] although first described long ago, [7] was for many years subsumed in the now defunct category of so-called giant cell 'MFH'. These lesions arise most often in dermis or subcutis of the limbs (less often the trunk) over a wide age range but most frequently in adults. They are typically well circumscribed and usually measure < 5 cm. The more infrequent deep-seated lesions may be larger. Tumors of this type are generally benign but may recur locally. Very rare cases (showing nuclear pleomorphism and atypia, often with a high mitotic rate) may metastasize.

Histologically, the majority of these tumors are characterized by multiple small tumor nodules histologically indistinguishable from giant cell tumor of bone, being composed of ovoid mononuclear cells and osteoclastic giant cells. These nodules are distributed in a cellular fibroblastic stroma showing variably prominent hemorrhage. Some cases show prominent reactive bone formation, usually at the periphery. Tumor cells often show frequent mitoses.

Other Tumors of Intermediate Biologic Potential which are Often Regarded as Fibrohistiocytic

Dermatofibrosarcoma protuberans (DFSP) [8, 9, 10] presents most often in adults of either sex between the ages of 30 and 50, but the lesion frequently has been present for many years. This is because DFSP has a prolonged plaque (flat) phase of very slow growth in most cases and only becomes nodular or protuberant at a later stage. Perhaps for these reasons DFSP in childhood is uncommon, although it does occur. [11, 12] The trunk (especially the abdominal wall) is by far the commonest location, while smaller percentages of cases arise in the limbs or head or neck region. Lesional size infrequently exceeds 5 cm. In its classical form DFSP has a 30-50% local recurrence rate unless widely excised, but the risk of metastasis is virtually non-existent. This figure rises substantially in cases showing high-grade ('fibrosarcomatous') transformation (see below).

DFSP is characterized by monomorphic spindle cells, with pale ill-defined cytoplasm (hence the lesion looks basophilic) and tapering nuclei, arranged in a consistent storiform (rush mat) pattern. Dermal involvement is invariable. Focal stromal hyalinization is quite frequent and may blur the storiform arrangement. Epidermal hyperplasia is usually absent and DFSP lacks the inflammatory component, histiocytes and multinucleate cells of fibrous histiocytoma. An important and distinctive feature is diffuse infiltration of subcutaneous fat, either in a honeycomb pattern or else in complex linear array. Stromal myxoid change is infrequent and prominent in no more than 1-2% of cases (myxoid DFSP): [13] such cases may show increased vascularity and relative loss of the storiform pattern. Occasional cases contain localized aggregates of floret-like giant cells, usually close to the deep aspect of the lesion, which appear to represent the subtlest form of differentiation towards giant cell fibroblastoma. Rare cases of DFSP recur as giant cell fibroblastoma [14] and lesions with hybrid features are well recognized [12] (see below).

Immunohistochemically tumor cells in DFSP are consistently CD34 positive (whereas cellular FH is only occasionally positive) and stain negatively for S-100, actin and EMA. Ultrastructurally lesional cells resemble fibroblasts. Most cases of DFSP have ring chromosomes which are composed of material from chromosomes 17 and 22, arranged in a tumor-specific complex translocation, [15] characterized by the gene fusion PDGFβ-COLIA1, [12, 16] which results in activation of the platelet-derived growth factor β receptor (PDGFR-β). Recent data indicate that this is of therapeutic relevance since PDGFR-β in DFSPs is effectively inhibited by Gleevec, a tyrosine kinase inhibitor. [17]

Pigmented DFSP (Bednar tumor) [18] – between 5 and 10% of cases of DFSP contain variably prominent S-100-positive dendritic pigmented melanocytes. Such lesions are commoner in black patients and, while this variant was held previously to support the concept of neuroectodermal differentiation in DFSP, it seems most likely that this phenomenon represents lesional colonization by melanocytes from the epidermis or skin adnexae. Aside from this finding there are no other clinicopathologic differences from usual DFSP.

Fibrosarcomatous DFSP [19, 20, 21] – this term refers to the approximately 5% of DFSPs which show transition to a fascicular pattern or higher-grade morphology with more frequent mitoses. Such transition may be gradual or abrupt and the higher grade component, while most often having the herringbone pattern of classic fibrosarcoma, may also be more eosinophilic (myofibroblastic) or more pleomorphic ("MFH"-like). This phenomenon is more common in long-standing de novo lesions than in recurrences of ordinary DFSP. Cumulative data from the numerous small, published series combined with our own study of 40 cases [21] reveals a significant metastatic rate of at least 15-20% (irrespective of the status of excision margins).

Giant cell fibroblastoma. Because it was realized that some cases of giant cell fibroblastoma (GCF) contained areas indistinguishable from DFSP and that some cases of GCF recur as DFSP, it was thought initially that GCF, when first properly described, [22] was a juvenile counterpart of DFSP but this probably was an over-simplification. GCF [22, 23, 24] occurs mainly in young children and shows the same predilection for the trunk as DFSP. There is a notable predominance in males. A small minority of cases present in adults. Most cases present as a poorly defined plaque-like cutaneous or subcutaneous swelling, usually not more than 5 cm in size, and local recurrence is very common unless the lesion is widely excised. To date GCF has never been reported to metastasize.

GCF is a poorly defined, infiltrative lesion which primarily involves the dermis and extends into subcutis. It has two principal components: rather primitive-looking spindle cells with pale cytoplasm, which often look somewhat neural and tend to diffusely infiltrate fat in a manner reminiscent of DFSP, and so-called angiectoid spaces, which are angulated pseudovascular channels lined by mononuclear and multinucleate (often floret-like) giant cells. Giant cells may also be scattered in the solid spindle cell areas. As in DFSP, tumor cells are almost invariably positive for CD34 but are negative for actin, S-100 and EMA.

From the biologic point of view, the following pieces of evidence point to a close relationship between GCF and DFSP: (1) Hybrid lesions showing areas of both tumor types occur; [12, 25] (2) GCF may recur locally as DFSP [26] and may even progress to the fibrosarcomatous variant of DFSP; (3) DFSP may recur as GCF; [14] (4) GCF shows exactly the same karyotypic aberrations involving chromosomes 17 and 22 as are seen in DFSP. [12, 16, 27, 28]

In practice there is no real differential diagnosis for GCF, at least not when it occurs in the customarily young patient. The bizarre morphology may raise the possibility of a sarcoma (such as myxofibrosarcoma) but the characteristic biphasic growth pattern with angiectoid spaces excludes all else.

Atypical fibroxanthoma (AFX) is the term used since the beginning of the 1960s to describe a histologically highly pleomorphic cutaneous neoplasm that almost always pursues a benign clinical course. [29, 30, 31] As with all the so-called "fibrohistiocytic" tumors, the histogenesis of this lesion always has been controversial. Over the years the "histiocytic" theory has become less tenable due to non-specific electron microscopic and immunohistochemical findings and other potential candidates such as myofibroblasts, fibroblasts and undifferentiated mesenchymal cells have been proposed. As a result of this, and also due to the demise of pleomorphic "MFH" as a specific entity, it is apparent that this tumor no longer can be considered simply a superficial non-metastasizing variant of so-called 'MFH'. It has become clear that AFX is a diagnosis of exclusion and that many cases classified as such in the past, especially those with metastasis, [32] probably represent examples of other high-grade sarcomas or non-mesenchymal malignancies, such as spindle cell squamous cell carcinoma and melanoma. Although initial descriptions of AFX emphasized variants with a prominent spindle-cell population, over the years the presence of pleomorphism became the sine qua non for this diagnosis, thereby overlooking cases with mild pleomorphism and a prominent spindle cell population. This more monomorphic variant has been called spindle-cell non-pleomorphic AFX. [33] Given the distinctive p53 mutations as well as UV photoproducts identified in these lesions, [34, 35] AFX is best regarded as a UV-induced pseudosarcomatous proliferation of fibroblasts, as was first suggested many years ago!

AFX presents as a solitary, polypoid, ulcerated lesion in sun-damaged skin, especially of the head and neck and much less commonly on the dorsum of the hands, of elderly patients. A short history of rapid lesional growth (often simulating a pyogenic granuloma) is common. Those cases reported in the past to occur in non-sun-exposed areas of young adults probably represent examples of atypical fibrous histiocytoma. Behavior is essentially benign and complete excision is generally curative. Local recurrence is very uncommon and should raise at least a little doubt as to the diagnosis.

Histologic criteria for the diagnosis of AFX should be very strict and require the use of immunohistochemistry. Histologically, all lesions are unencapsulated dermal neoplasms, quite often surrounded by an epidermal collarette and usually not extending deeper than adnexal structures. Growth tends to be expansile with only limited infiltration. Solar elastosis is generally prominent. Ulceration is present in most cases and this makes evaluation of epidermal dysplasia or junctional activity to exclude epidermal origin difficult or impossible. Classical AFX is composed of highly pleomorphic histiocyte-like cells and very atypical giant cells interspersed with a variable number of spindle-shaped cells and inflammatory cells. Normal and abnormal mitotic figures are numerous. The spindle cell variant of AFX is composed of fascicles of eosinophilic spindle cells with vesicular nuclei and one or two eosinophilic nucleoli, [33] thus often simulating malignant melanoma. Mitotic figures are common but cytologic pleomorphism is focal and can be absent. I do not accept deep (subcutaneous or further) invasion, necrosis or vascular or perineurial invasion as features of AFX, although there may be focal very superficial invasion into fat. Such features, to me, indicate an unequivocal diagnosis of malignancy (unclassified dermal sarcoma if keratin and S-100 protein are negative). Immunohistochemistry is essential for confirming the diagnosis of AFX. Vimentin is diffusely positive in all cases and a few are positive focally for smooth-muscle actin, suggestive of fibroblastic or myofibroblastic differentiation. Keratin, S-100 protein and desmin are always negative, ruling out the main three differential diagnoses namely, spindle cell squamous cell carcinoma, spindle cell melanoma and leiomyosarcoma.

Angiomatoid "MFH" (so-called) is undoubtedly a distinct, definable entity which has always differed significantly from other so-called 'MFH's by its involvement of a much younger age group. [36] However, the line of differentiation in these lesions is poorly understood. Clinically, so-called angiomatoid "MFH" arises most often in children or adolescents of either sex and usually develops in superficial soft tissue of the limbs more often than the trunk; [36, 37, 38, 39]t he overall age range is wide and a minority of cases are deep-seated. A small but significant proportion of patients have associated systemic features such as pyrexia, anemia, weight loss, or paraproteinemia, which remit following excision of the tumor. The majority of cases are subcutaneous and generally measure less than 3–4cm. Following excision, the rate of locoregional recurrence is 10-15%, but no more than 1-2% of patients develop lymph node or systemic metastases and therefore tumor-related deaths are rare.

These are well-circumscribed, multinodular, cystic masses which often show extensive hemorrhage. Histologically, [36, 37, 38, 39] they consist of multiple nodules and sheets of bland, rather eosinophilic, ovoid to spindle cells with uniform vesicular, somewhat spindled, nuclei showing infrequent mitoses. Nuclear pleomorphism, other than in occasional examples, is mild and giant cells are rare. These nodules are set in a collagenous stroma which usually shows prominent hemosiderin deposition and a prominent lymphoplasmacytic infiltrate, which may simulate a lymph node. Variations to this pattern do occur and in some cases either inflammation or the blood-filled spaces may be inconspicuous or even absent; occasionally the spaces, which lack an endothelial lining, contain only eosinophilic proteinaceous fluid. Immunohistochemically, [39, 40, 41] personal experience in over 100 cases suggests that, in around 40–50%, the eosinophilic tumor cells are desmin positive, while EMA and muscle actin (HHF-35), but not smooth muscle actin, are positive in almost as many; CD68 is also commonly positive, although its specificity is questionable. Electron microscopy in numerous case reports has provided conflicting and inconclusive results. At present, these lesions seem most likely to be myoid in nature and it seems quite possible that the tumor cells are differentiating towards so-called "fibroblastic reticulum cells", the myoid cells which are normally found in lymph nodes. An interesting more recent development has been the recognition that these lesions appear to be characterized by consistent ATF1 gene rearrangement, [42, 43] most often in association with a (12;16) translocation. The differential diagnosis includes mainly aneurysmal benign fibrous histiocytoma which is consistently more superficial, cytologically more polymorphic, and desmin negative.

Myxofibrosarcoma (low grade) [44, 45, 46, 47] is a cohesive, distinct and definable entity which shows a broad spectrum of histologic grade. The high grade, poorly differentiated end of this spectrum resembles lesions formerly known as so-called pleomorphic "MFH" in terms of cellularity, atypia and pleomorphism and this is how the now abandoned term "myxoid MFH" originated. It is the low-grade lesions which have no metastatic potential, but which may advance in grade, that are most relevant to this overview.

Myxofibrosarcoma affects mainly adults in the 6th to 8th decades, with a slight predominance in males. The great majority of these tumors arise in the limbs; retroperitoneal origin is extremely infrequent. Of particular note, about two thirds of cases develop in subcutaneous rather than deeper tissue. There is a tendency for multinodular, diffusely infiltrative growth and the true extent of these tumors, especially the superficial ones, often is underestimated. Survival correlates with histologic grade but overall 5-year survival is 60-70%. The lowest grade lesions have no capacity to metastasize but may become higher grade and acquire metastatic potential in a recurrence. The high-grade lesions have an unusual but distinct tendency to demonstrate lymph node metastasis, in addition to anticipated pulmonary and osseous spread in around 30% of cases.

The principal features shared by tumors of all grades but, most striking in low-grade lesions, are hypocellular areas, containing thin-walled curvilinear vessels, in which hyperchromatic small spindle and stellate cells with poorly defined, sometimes vacuolated (mucin-containing) cytoplasm are found. Most lesions are multinodular and a prominent inflammatory infiltrate is common, especially at the lesional periphery. The cellularity and degree of pleomorphism parallel the histologic grade, but it is the characteristic low-grade areas which define this lesion. The spindle and stellate cells have ultrastructural features of fibroblasts or, occasionally, myofibroblasts. When low-grade lesions recur, they may acquire hypercellular or more solid areas of pleomorphic ("MFH"-like) or spindle cell tumor, indicating progression to an intermediate or high-grade lesion with acquisition of metastatic potential. Cytogenetic analysis often reveals ring chromosomes (of uncertain origin) in low-grade cases.

Principal differential diagnoses in low-grade lesions are myxoid liposarcoma, intramuscular myxoma, cellular myxoma and superficial angiomyxoma. Myxoid liposarcoma does not show such a degree of nuclear atypia and pleomorphism, its vessels have a branching (crow's feet) pattern and it contains lipoblasts. Intramuscular myxoma contains hardly any vessels at all and shows no nuclear atypia. Cellular myxomas are intermediate in their appearance between intramuscular myxoma and low-grade myxofibrosarcoma. They are distinguished by their lack of nuclear atypia or hyperchromasia. Superficial angiomyxoma does not show nuclear atypia or pleomorphism and often has an epithelial component. Low-grade fibromyxoid sarcoma (Evans) is more collagenous, has a swirling growth pattern and shows less nuclear atypia.

References
  1. Enzinger FM, Zhang RY. 1988. Plexiform fibrohistiocytic tumor presenting in children and young adults. An analysis of 65 cases. Am J Surg Pathol 12: 818-826

  2. Hollowood K, Holley MP, Fletcher CDM. 1991. Plexiform fibrohistiocytic tumour: clinicopathological, immunohistochemical and ultrastructural analysis in favour of a myofibroblastic lesion. Histopathology 19: 503-513

  3. Remstein ED, Arndt CA, Nascimento AG. 1999. Plexiform fibrohistiocytic tumor: clinicopathologic analysis of 22 cases. Am J Surg Pathol 23: 662-670

  4. Folpe A L, Morris R J, Weiss S W 1999 Soft tissue giant cell tumor of low malignant potential: a proposal for the reclassification of malignant giant cell tumor of soft parts. Mod Pathol 12: 894-902

  5. Oliveira A M, Dei Tos A P, Fletcher C D M, Nascimento A G 2000 Primary giant cell tumor of soft tissues: a study of 22 cases. Am J Surg Pathol 24: 248-256

  6. O'Connell JX, Wehrli BM, Nielsen GP, Rosenberg AE. 2000. Giant cell tumors of soft tissue: a clinicopathologic study of 18 benign and malignant tumors. Am J Surg Pathol 24: 386-395

  7. Salm R, Sissons H A 1972 Giant cell tumours of soft tissues. J Pathol 107: 27-39

  8. Taylor HB, Helwig EB. Dermatofibrosarcoma protuberans. A study of 115 cases. Cancer 1962; 15:717-725.

  9. Burkhardt BR et al. Dermatofibrosarcoma protuberans. Study of fifty-six cases. Am J Surg 1966; 111:638-644.

  10. Fletcher CDM et al. Dermatofibrosarcoma protuberans: a clinicopathological and immunohistochemical study with a review of the literature. Histopathology 1985; 9:921-938.

  11. McKee PH, Fletcher CDM. Dermatofibrosarcoma protuberans presenting in infancy and childhood. J Cutan Pathol 1991; 18:241-246.

  12. Terrier-Lacombe M-J, Guillou L, Maire G et al. Dermatofibrosarcoma protuberans, giant cell fibroblastoma and hybrid lesions in children: clinicopathologic comparative analysis of 28 cases with molecular data. Am J Surg Pathol 2003; 27:27-39.

  13. Frierson HF, Cooper PH. Myxoid variant of dermatofibrosarcoma protuberans. Am J Surg Pathol 1983; 7:445-450.

  14. Coyne J, Kaftan SM, Craig RD. Dermatofibrosarcoma protuberans recurring as a giant cell fibroblastoma. Histopathology 1992; 21:184-187.

  15. Naeem R, Lux ML, Huang SF, et al. Ring chromosomes in dermatofibrosarcoma protuberans are composed of interspersed sequences from chromosomes 17 and 22. Am J Pathol 1995; 147:1553-1558.

  16. Simon MP, Pedeutour F, Sirvent N, et al. Deregulation of the platelet-derived growth factor B-chain gene via fusion with collagen gene COL1A1 in dermatofibrosarcoma protuberans and giant cell fibroblastoma. Nature Genetics 1997; 15:95-98.

  17. McArthur GA, Demetri GD, van Oosterom A et al. Molecular and clinical analysis of locally advanced dermatofibrosarcoma protuberans treated with imatinib. J Clin Oncol 2005; 23:866-873.

  18. Fletcher CDM et al. Pigmented dermatofibrosarcoma protuberans (Bednar tumour): melanocytic colonisation or neuroectodermal differentiation? Histopathology 1988; 13:631-643.

  19. Ding J, Hashimoto H, Enjoji M. Dermatofibrosarcoma protuberans with fibrosarcomatous areas. Cancer 1989; 64:721-729.

  20. Connelly JH, Evans HL. Dermatofibrosarcoma protuberans. A clinicopathologic review with emphasis on fibrosarcomatous areas. Am J Surg Pathol 1992; 16:921-925.

  21. Mentzel T et al. Fibrosarcomatous ("high grade") dermatofibrosarcoma protuberans: clinicopathologic and immunohistochemical study of a series of 41 cases with emphasis on prognostic significance. Am J Surg Pathol 1998; 22:576-587.

  22. Shmookler BM, Enzinger FM, Weiss SW. Giant cell fibroblastoma. A juvenile form of dermatofibrosarcoma protuberans. Cancer 1989; 64:2154-2161.

  23. Dymock RB et al. Giant cell fibroblastoma. A distinctive recurrent tumor of childhood. Am J Surg Pathol 1987; 11:263-271.

  24. Fletcher CDM. Giant cell fibroblastoma of soft tissue: a clinicopathological and immunohistochemical study. Histopathology 1988; 13:499-508.

  25. Beham A, Fletcher CDM. Dermatofibrosarcoma protuberans with areas resembling giant cell fibroblastoma. Histopathology 1990; 17:165-167.

  26. Alguacil-Garcia A. Giant cell fibroblastoma recurring as dermatofibrosarcoma protuberans. Am J Surg Pathol 1991; 15:798-801.

  27. Craver RD et al. Aggressive giant cell fibroblastoma with a balanced 17;22 translocation. Cancer Genet Cytogenet 1995; 80:20-22.

  28. Dal Cin P et al. Cytogenetic and immuno-histochemical evidence that giant cell fibroblastoma is related to dermatofibrosarcoma protuberans. Genes Chromos Cancer 1996; 15:73-75.

  29. Kempson RL, McGavran MH. Atypical fibroxanthomas of the skin. Cancer 1964; 176:1463-1471.

  30. Fretzin D, Helwig EB. Atypical fibroxanthoma of the skin. A clinicopathological study of 140 cases. Cancer 1973; 39:1541-1552.

  31. Mirza B, Weedon D. Atypical fibroxanthoma: a clinicopathological study of 89 cases. Australas J Dermatol 2005; 46:235-238.

  32. Helwig EB, May D. Atypical fibroxanthomas of the skin with metastasis. Cancer 1986; 57:368-376.

  33. Calonje E et al. Spindle cell non-pleomorphic atypical fibroxanthoma: analysis of a series and delineation of a distinctive variant. Histopathology 1993; 22:247-254.

  34. Dei Tos AP, Maestro D, Doglioni C et al. Ultraviolet-induced p53 mutations in atypical fibroxanthoma. Am J Pathol 1994; 145:11-17.

  35. Sakamoto A, Oda Y, Hakura E et al. Immunoexpression of ultraviolet photoproducts and p53 mutation analysis in atypical fibroxanthoma and superficial MFH. Mod Pathol 2001; 14:581-588.

  36. Enzinger F M 1979 Angiomatoid malignant fibrous histiocytoma. A distinct fibrohistiocytic tumor of children and young adults simulating a vascular neoplasm. Cancer 44: 2147-2157

  37. Costa M J, Weiss S W 1990 Angiomatoid malignant fibrous histiocytoma. A follow-up study of 108 cases with evaluation of possible histologic predictors of outcome. Am J Surg Pathol 14: 1126-1132

  38. Pettinato G, Manivel J C, De Rosa G et al 1990 Angiomatoid malignant fibrous histiocytoma: cytologic, immunohistochemical, ultrastructural and flow cytometric study of 20 cases. Mod Pathol 3: 479-487

  39. Fanburg-Smith J C, Miettinen M 1999 Angiomatoid "malignant" fibrous histiocytoma: a clinicopathologic study of 158 cases and further exploration of the myoid phenotype. Hum Pathol 30: 1336-1343

  40. Fletcher C D M 1991 Angiomatoid "malignant fibrous histiocytoma": an immunohistochemical study indicative of myoid differentiation. Hum Pathol 22: 563-568

  41. Smith M E F, Costa M J, Weiss S W 1991 Evaluation of CD68 and other histiocytic antigens in angiomatoid malignant fibrous histiocytoma. Am J Surg Pathol 15: 757-763

  42. Waters BL, Panagopoulos I, Allen EF. 2000. Genetic characterization of angiomatoid fibrous histiocytoma identifies fusion of the FUS and ATF-1 genes induced by a chromosomal translocation involving bands 12q13 and 16p11. Cancer Genet Cytogenet 121: 109-116

  43. Hallor KH, Mertens F, Jin Y et al. 2005. Fusion of the EWSR1 and ATF1 genes without expression of MITF-M transcript in angiomatoid fibrous histiocytoma. Genes Chromos Cancer 44: 97-102

  44. Merck C, Angervall L, Kindblom L-G, Oden A. 1983. Myxofibrosarcoma. A malignant soft tissue tumour of fibroblastic-histiocytic origin. A clinicopathologic and prognostic study of 110 cases using multivariate analysis. Acta Pathol Microbiol Immunol Scand A 91 (Suppl. 282):1-40.

  45. Weiss SW, Enzinger FM. 1977. Myxoid variant of malignant fibrous histiocytoma. Cancer 39:1672-1685.

  46. Mentzel T, Calonje E, Wadden C et al. 1996. Myxofibrosarcoma: clinicopathologic analysis of 75 cases with emphasis on the low-grade variant. Am J Surg Pathol 20:391-405.

  47. Huang HY, Lal P, Qin J et al. 2004. Low grade myxofibrosarcoma: a clinicopathologic analysis of 49 cases treated at a single institution with simultaneous assessment of the efficacy of 3-tier and 4-tier grading systems. Hum Pathol 35:612-621.