Atypical fibroblastic neoplasm with pseudoangiomatous, cystic/microcystic and myxoid features, possibly solitary fibrous tumor variant Janez Lamovec, M.D. Institute of Oncology Ljubljana, Slovenia

Case History: An 81-year-old woman presented with a subcutaneous tumor of the nape of the neck. The tumor was first noted 3 years previously when it measured 3 cm. Recently, it enlarged considerably and prompted the patient to seek medical attention. Slide 1 Click to view with ImageScope Click to view with a Web-Based Viewer Slide 2 Click to view with ImageScope Click to view with a Web-Based Viewer Slide 3 Click to view with ImageScope Click to view with a Web-Based Viewer Slide 4 Click to view with ImageScope Click to view with a Web-Based Viewer On physical examination, a subcutaneous tumor measuring 10 x 10 cm was seen; it was covered by livid skin. Tumor was relatively soft on palpation, and not fixed to skin and spine. It was marginally excised. Pathology: Gross specimen was represented by a segment of skin and subcutaneous tissue with a tumor measuring 8.5 x 6 cm; overlying skin was stretched by a bulging tumor but intact, at deep margin there was a segment of muscle fascia and thin muscle layer. All margins were free of tumor. On cut surface, the tumor appeared fleshy, red-brown, soft, well circumscribed. Microscopically, tumor shows a thin capsule and expansive margins toward dermis and underlying fascia without any infiltration of these structures. It shows innumerable cysts filled with blood or proteinaceous fluid; smaller cysts appear empty. In addition, many spaces are confluent and form larger blood lakes. The cellular tissue among the cysts and lakes is for the most part loose, focally myxoid; denser cellular areas are also present. The neoplastic cells are spindle, oval or stellate with relatively normochromic oval, elongated or spindle nuclei of different sizes with small or indistinct nucleoli. Some binucleated cells are also seen while there are no distinct multinucleated giant cells. Rare nuclear cytoplasmic inclusions are also seen. The cytoplasm is generally sparse or moderate in amount, eosinophilic. Scattered cells with more abundant cytoplasm may be found; in some neoplastic cells there are intracytoplasmic vacuoles and in others the cytoplasm appears foamy. No lipoblasts or lipocytes are found. Mitoses are rare; there is < 1 mitosis per HPF, some atypical forms may be identified. Majority of the tumor shows disordered growth of cells although in some parts there is some fascicular arrangement of spindle cells, in other foci a storiform pattern of growth is suggested and a hemangiopericytoid appearance may also be appreciated focally. The vast majority of cystic spaces, small and large, are lined by tumor cells and not by endothelium. True vascular structures are for the most part thin-walled; thick-walled vessels are rare. The interstitial matrix contains amorphous or vaguely fibrillary proteinaceous fluid,some extravasated erythrocytes and relatively numerous reticulin fibers revealed by Gomori staining; collagen fibers as highlighted by van Gieson stain are extremely scant except for a few foci; mostly hyalinized. Inflammatory cells are sparse, mostly small lymphocytes. Immunohistochemically, tumor cells were diffusely positive for vimentin, CD34 and bcl-2 while positivity for alpha smooth muscle actin and calponin was observed in a number of cells although unevenly distributed. Nuclei of many cells showed positive reaction for progesterone receptors. Only occasional cells, mostly endothelial, were positive for CD68. CD31 and factor VIII delineated focally quite dense network of mostly thin-walled and compressed delicate blood vessels that were not so easily appreciated on H&E. Proliferation marker MIB-1 decorated less than 2% of neoplastic nuclei. The reaction to numerous other antigens was negative, including desmin, caldesmon, EMA, MNF116, estrogen receptors, CD21, CD99, CD117, and ALK. Ultramicroscopical examination of tissue material obtained from formalin fixed paraffin embedded tissue showed short fascicles and disorderly arranged plump and spindle shaped neoplastic cells with fibroblastic characteristics; no identifiable myofibroblasts were found. In extracellular matrix, there was a relatively abundant granulo-fibrillary material and no collagen fibers. Flow cytometric DNA measurements showed no abnormal DNA content. Diagnosis: Atypical fibroblastic neoplasm with pseudoangiomatous, cystic/microcystic and myxoid features, possibly solitary fibrous tumor variant Follow-up: Five years and a half following surgery the patient is alive, 86 years old, and shows no evidence of local recurrence or metastatic disease. Comment: This case of a soft tissue tumor is difficult to classify into any existing and well-known category of soft tissue neoplasms. The overall relative cytological uniformity, thin peripheral capsule, expansive margins and a scarcity of the inflammatory component are arguments for neoplastic rather than pseudoneoplastic nature of the lesion. The morphology of the tumor and also a relatively long follow-up without evidence of recurrence are in favor of its benign or at most borderline biological nature. It is possible that the present soft tissue tumor is an unusual variant of a solitary fibrous tumor (SFT). It has been known for quite a while that there are many faces to this neoplasm, either inside an individual case or among different cases. Similarly to pleural and to extrapleural SFTs of various sites, there were several growth patterns also described in SFT of soft tissues: haphazard, hemangiopericytoid, myxoid, alternating hypercellular/hypocellular spindle cell pattern, storiform pattern [1, 2]; tumors may contain fat cells and multinucleated giant cells being similar or identical to so-called lipomatous hemangiopericytomas [3] and giant cell angiofibromas or giant fibroblastomas [4, 5] . Microcystic changes, myxoid appearance and pseudovascular spaces were also described in SFT [6]. Several of the patterns listed above are also seen, at least focally, in the presented tumor, but the general aspect of it, including cytological features and extensive, almost diffuse microcystic pattern does not quite fit into the picture. Alternating hypercellular and hypocellular sclerotic foci, bland looking spindly cells, intimate intertwining of thin and thick collagen with spindly cells, and also the characteristic haphazard, storiform or fascicular arrangement of cells are not so obvious; these features are essential for the firm diagnosis of SFT [4]. Giant cell angiofibroma [7] and giant cell fibroblastoma [8] are even less probable possibilities for our case. Pseudovascular spaces are observed in both of the former tumors, and CD 34 positive stromal cells may also be somewhat similar to those in the presented case. However, in both of them, multinucleated giant cells lining pseudovascular spaces are characteristically present; they were not found in our case. Cellular angiofibroma, a tumor of superficial soft tissue, mostly in external genital areas, is a very remote differential diagnostic possibility [9]. It is altogether different tumor with numerous evenly spaced small and medium sized thick-walled blood vessels set in a relatively uniform stroma composed of bland, CD34 positive spindle cells with short bundles of collagen. No pseudovascular or microcystic spaces were observed in cases described. Somewhat morphologically overlapping and perhaps to cellular angiofibroma related tumor of the genital area, angiomyofibroblastoma, a tumor with many prominent vessels and loose stroma with round to spindle fibroblastic and myofibroblastic cells, strongly positive for desmin [10], is even less probable diagnosis; the tumor has not been described outside genital region neither in females nor in males. Another two, subcutaneous, CD34 positive lesions, such as mammary-type fibroblastoma of soft tissues [11] and spindle cell lipoma [12], two closely related lesions with only subtle differences in appearance, may also be considered in differential diagnosis. They are composed of uniform spindle cells with oval or tapered nuclei, often wavy in contours, and usually arranged in fascicles of different length. Admixed lipocytes are almost always present. Hyalinized bands of collagen in the former or ropey collagen in a latter is a prominent feature. Vascularity of both tumors is not marked but pseudovascular spaces were described in both of them [13]. Although rare microscopic foci in our tumor may vaguely resemble these two tumors, it is generally morphologically quite different from both of them. Aneurysmal benign fibrous histiocytoma, is a cutaneous tumor, rarely with subcutaneous involvement, with blood filled aneurysmal spaces but with polymorphous cellular composition, with numerous giant cells, more extensive hemosiderin deposition and CD34 negative cells [14]. Several morphologic features and cutaneous location most strongly argue against such a diagnosis for the presented case. Immunohistochemical results in our case do not substantially contribute to more exact classification of it. Positivity for CD34 is shared by a plethora of soft tissue tumors; that for smooth muscle actin may potentially show to some myofibroblastic differentiation, and positive reaction for progesterone receptors may not be so easily explained; it may or may not indicate some hormonal effects on tumor cells; estrogen receptors were not present. It is interesting that the positivity for progesterone receptors has been demonstrated previously in some SFT cases while no estrogen receptors positivity was found [15, 16]. Thus, the exact nature of our tumor remains unclear. The closest one can get with differential diagnostic possibilities is to consider it as a fibroblastic tumor, perhaps somewhat related to SFT. In previous ultramicroscopical studies, the cells composing SFT showed considerable heterogeneity: in addition to fibroblasts, numerous perivascular undifferentiated cells and pericytes were demonstrated. A minority of cells were myofibroblasts [17]. In our case, EM examination of a relatively poorly preserved tissue showed only fibroblastic cells; no myofibroblasts were identified. In regard to latter cells, it has been disputed how strict the criteria for ultramicroscopical identification of myofibroblasts should be and whether a tumor could be diagnosed at all as a myofibroblastic by light microscopical and immunohistochemical methods only [18, 19, 20]. In addition, it has not been clearly stated what should be the proportion of neoplastic myofibroblasts in a given tumor to call it myofibroblastic, fibroblastic/myofibroblastic or simply fibroblastic. It appears that many transitional forms exist but are not clearly defined. Moreover, a number of tumors listed above are morphologically and biologically related and there is a considerable overlapping among them. It is not certain if the presented tumor may also be in a gray zone between different fibroblastic and myofibroblastic neoplasms. Future cytogenetic and molecular studies might potentially elucidate the nature of some of such tumors. References: Suster S, Nascimento AG, Miettinen M, et al. Solitary fibrous tumor of soft tissue. A clinicopathologic and immunohistochemical study of 12 cases. Am J Surg Pathol 1995; 19: 1257-66 Nielsen GP, O'Connell JX, Dickersin R, et al. Solitary fibrous tumor of soft tissue: a report of 15 cases, including 5 malignant examples with light microscopic, immunohistochemical, and ultrastructural data. Mod Pathol 1997; 10: 1028-37 Gengler C, Guillou L. Solitary fibrous tumor and hemangiopericytoma: evolution of a concept. Histopathology 2006; 48: 63-74 Chan JKC. Solitary fibrous tumor – everywhere, and a diagnosis in vogue. Histopathology 1997; 31: 568-76 Sigel JE, Fisher C, Vogt D, et al. Giant cell angiofibroma of the inguinal region. Ann Diagn Pathol 2000; 4: 240-44 Hasegawa T, Matsuno Y, Shimoda T, et al. Extrathoracic solitary fibrous tumors: their histological variability and potentially aggressive behavior. Hum Pathol 1999; 30: 1464-73 Guillou L, Gebhard S, Coindre JM. Orbital and extraorbital giant cell angiofibroma: a giant cell-rich variant of solitary fibrous tumor. Clinicopathologic and immunohistochemical analysis of a series in favor of a unifying concept. Am J Surg Pathol 2000; 24: 971-79 Fletcher CDM. Giant cell fibroblastoma of soft tissue: a clinicopathological and immunohistochemical study. Histopathology 1988; 13: 499-508. Iwasa Y, Fletcher CDM. Cellular angiofibroma. Clinicopathologic and immunohistochemical analysis of 51 cases. Am J Surg Pathol 2004; 28: 1426-35 Fletcher CDM, Tsang WY, Fisher C, et al. Angiomyofibroblastoma of vulva. A benign neoplasm distinct from aggressive angiomyxoma. Am J Surg Pathol 1992; 16: 373-82 McMenamin ME, Fletcher CDM. Mammary-type myofibroblastoma of soft tissue: a tumor closely related to spindle cell lipoma. Am J Surg Pathol 2001; 25: 1022-29 Angervall L, Dahl I, Kindblom LG, et al. Spindle cell lipoma. Acta Pathol Microbiol Scand 1976; 84: 477-87 Hawley IC, Krausz T, Evans DJ, et al. Spindle cell lipoma – a pseudoangiomatous variant. Histopathology 1994; 24: 565-69 Calonje E, Fletcher CDM. Aneurysmal benign fibrous histiocytoma: a clinicopathological analysis of 40 cases frequently misdiagnosed as vascular neoplasm. Histopathology 1995; 26: 323-31 Fukunaga M, Naganuma H, Nikaido T, et al. Extrapleural solitary fibrous tumor: a report of seven cases. Mod Pathol 1997; 10: 443-50 Bhargava R, Shia J, Hummer AJ, et al. Distinction of endometrial stromal sarcoma from "hemangiopericytomatous" tumors using a panel of immunohistochemical stains. Mod Pathol 2005; 18: 40-47 Obara IF, Mishima K, Saito I, et al. Ultrastructural spectrum of solitary fibrous tumor: a unique perivascular tumor with alternative lines of differentiation. Virchows Arch 2005; 446: 646-52 Schürch W, Seemayer TA, Gabbiani G. The myofibroblast. A quarter century after its discovery. Am J Surg Pathol 1998; 22: 144-47 Legace R, Seemayer TA, Gabbiani G, et al. Myofibroblastic sarcoma. Letter to the editor. Am J Surg Pathol 1999; 23; 1432-33 Fletcher CDM. Myofibroblastic sarcoma. Letter to the editor. Am J Surg Pathol 1999; 1433-35