Low-grade fibromyxoid sarcoma: Clinical, morphologic and genetic features

Low-grade fibromyxoid sarcoma: Clinical, morphologic and genetic features

Annals of Diagnostic Pathology 28 (2017) 60–67 Contents lists available at ScienceDirect Annals of Diagnostic Pathology journal homepage: www.elsevi...

1MB Sizes 239 Downloads 108 Views

Annals of Diagnostic Pathology 28 (2017) 60–67

Contents lists available at ScienceDirect

Annals of Diagnostic Pathology journal homepage: www.elsevier.com/locate/anndiagpath

Low-grade fibromyxoid sarcoma: Clinical, morphologic and genetic features Mustafa Mohamed, Cyril Fisher, Khin Thway

MARK



Sarcoma Unit, Royal Marsden Hospital, London, UK

A R T I C L E I N F O

A B S T R A C T

Keywords: Low-grade fibromyxoid sarcoma Hyalinizing spindle cell tumor with giant rosettes Sclerosing epithelioid fibrosarcoma Genetics Pathology Gene rearrangement Translocation FUS-CREB3L2 FUS-CREB3L1 EWSR1-CREB3L2

Low-grade fibromyxoid sarcoma (LGFMS) is a bland spindle cell neoplasm that typically arises in the deep soft tissues of the proximal extremities or trunk of young adults. The majority of LGFMS are characterized by a recurrent (7;16)(q34;p11) translocation, resulting in the FUS-CREB3L2 fusion gene, which generates a chimeric protein with transcriptional regulatory activity. Small numbers harbor a FUS-CREB3L1 fusion resulting from t (11;16)(p11;p11), whilst rare cases harbor the EWSR1-CREB3L1 fusion. LGFMS is of low to moderate cellularity and consists of bland spindle cells with small, angulated nuclei and scant, wispy cytoplasm, arranged in a whorled growth pattern and typically showing abrupt transition from myxoid to fibrous areas. Immunohistochemical expression of MUC4 is a consistent finding. Hyalinized spindle cell tumor with giant rosettes (HSCTGR) is a morphological variant of LGFMS that shares the same balanced translocation, and is also immunoreactive for MUC4. A potential relationship between LGFMS and sclerosing epithelioid fibrosarcoma (SEF), a rare fibroblastic neoplasm that most commonly arises in the deep soft tissues of the lower extremities, limb girdles or trunk, has also been suggested. SEF is classically composed of nests and cords of epithelioid cells with clear or eosinophilic cytoplasm embedded within densely sclerotic stroma. In some cases, areas indistinguishable from LGFMS are present, and these have been shown to contain FUS-CREB3L2 fusion transcripts. The majority of pure SEF tumors harbor EWSR1 rearrangements, with EWSR1-CREB3L1 and more rarely EWSR1-CREB3L2 gene fusions more common than those involving FUS. MUC4 immunoreactivity is also seen in approximately 70% of SEF. Surgical resection of these tumors with clear margins is the treatment of choice. Correct diagnosis is important because of the significant potential for recurrence and late metastatic spread. We review LGFMS and SEF, discussing morphology and immunohistochemistry, genetics and molecular findings, and the differential diagnosis.

1. Introduction Low-grade fibromyxoid sarcoma (LGFMS) is a malignant, often latemetastasizing tumor with a misleadingly bland histological appearance that typically arises in the deep soft tissues of the proximal extremities or trunk of young adults. It was first proposed as a distinct entity by Evans in 1987 [1] who described two variably cellular neoplasms composed of alternating fibrous and myxoid areas containing bland spindle or stellate cells showing a swirling, whorled pattern of growth. Both tumors occurred in women in their late twenties and were located in the shoulder/trunk area. Lung metastases were present in both cases. Six years later, Evans expanded his original series with 10 more cases, also noting that ‘dedifferentiation’ occurred in one of the cases [2]. In 1997, Lane et al. described a tumor composed of hypocellular hyalinizing collagen cores rimmed by minimally atypical spindle cells in a fibromyxoid background [3]. This was initially thought to be a distinct neoplastic entity and was termed hyalinizing spindle cell tumor with



giant rosettes (HSCTGR). However, it was noted that both tumors shared similar morphologic features and it was therefore proposed that HSCTGR should be considered a histological variant of LGFMS. This notion was further supported by a large series of LGFMS cases that found collagenous rosettes in tumors with otherwise classic histological features of LGFMS [4]. The majority of LGFMS and HSCTGR cases have since been shown to harbor a common t(7;16)(q34;p11), resulting in a chimeric fusion protein derived from the fused in sarcoma (FUS) gene of chromosome 16p11 and the cAMP responsive element-binding protein 3like 2 (CREB3L2) gene of 17q33 [5]. A minority of cases have been shown to display a FUS-CREB3L1 fusion resulting from t(11;16) (p11;p11) [6] whilst the EWSR1-CREB3L1 fusion has been reported in two cases [7]. It has also been suggested that there may be a relationship between LGFMS and sclerosing epithelioid fibrosarcoma (SEF) [8]. This is a rare fibroblastic neoplasm, first described in 1995 by Meis-Kindblom et al. [9], that most commonly arises in the deep soft tissues of the lower

Corresponding author at: Sarcoma Unit, The Royal Marsden NHS Foundation Trust, 203 Fulham Road, London SW3 6JJ, UK. E-mail address: [email protected] (K. Thway).

http://dx.doi.org/10.1016/j.anndiagpath.2017.04.001

1092-9134/ © 2017 Elsevier Inc. All rights reserved.

Annals of Diagnostic Pathology 28 (2017) 60–67

M. Mohamed et al.

follow-up for all patients with LGFMS. Superficial LGFMS has generally been associated with a good prognosis, which is better than that for deep-seated neoplasms [20]. No specific histologic features of LGFMS, other than ‘dedifferentiation’, have been shown to correlate with prognosis, although small tumor size (< 3.5 cm) might represent a favorable prognostic factor [36]. SEF affects patients in an older age group with mean and median ages reported as 46 and 48 years, respectively (age range 10 to 78 years) [37]. While no significant gender difference in incidence has been observed in large series [9,12,37], the recent series of PrietoGranada et al. has shown a striking female predominance (90% of patients) in tumors with morphology of pure SEF [38]. The most common anatomic site is the lower limb, followed by the trunk, retroperitoneum, paravertebral region, neck, upper limb, abdominal cavity and thoracic cavity [37,39]. Primary renal [40-42] and bone involvement (most frequently the long bones of the extremities) [43] have also been documented. Surgical excision with clear resection margins is the treatment of choice. The outcome of patients with SEF is highly variable. Like LGFMS, some neoplasms are characterized by a relatively protracted clinical course with metastases developing several years after surgical excision [9], whilst others exhibit more aggressive behavior with a high metastatic rate and a high proportion of tumorrelated deaths [8]. The mortality rate between patients with pure SEF and those with hybrid SEF/LGFMS has been shown to be roughly similar, at 44% and 37% respectively (with a mean overall follow-up period of 66 months) [38].

extremities, limb girdles or trunk, where it is often closely associated with fascia or periosteum. Infiltration into adjacent structures, including bone, is common. Less frequently, SEF can arise in bone, usually in the head and neck region [10]. It typically occurs in middle-aged adults and can pursue an aggressive clinical course with more than half of patients developing local recurrences and distant metastases [9]. SEF is classically composed of nests and cords of epithelioid cells with clear or eosinophilic cytoplasm embedded within a densely sclerotic stroma. In some cases, areas indistinguishable from LGFMS are present [11]. The earliest observation of LGFMS cases with areas resembling what is now known as SEF was provided by Evans in 1993 [2], which was interpreted as evidence of ‘dedifferentiation.’ A subset of SEF, including some with hybrid LGFMS features, has been shown to contain FUSCREB3L2 fusion transcripts or FUS gene rearrangements [11,12]. Furthermore, the immunohistochemical expression of MUC4, which is a consistent finding in LGFMS, is also seen in approximately 70% of SEF [13], lending further weight to the possibility of a close relationship between these tumors. 1.1. Clinical features and management The incidence of LGFMS has been reported as 0.18 per million [14]. It can affect patients of all ages but has a peak incidence in young adults, with a mean age of 33 years and a median of 32.5 years (age range 10 to 69 years) [15]. 13–19% of cases occur in patients younger than 18 years of age [4,16]. LGFMS is extremely rare in children younger than 5 years of age [17] with only nine cases reported in the literature to date [18]. The male to female ratio has been reported to be 7:2 in infants (0–5 years of age) and 8:5 in young adolescents (6–15 years of age) [18]. In adult cases, the male to female ratio was reported to be either equal [6,15] or 3:1 [19]. The majority of LGFMS cases present as painless, slow growing and deep seated masses within the proximal extremities and trunk, although superficial LGFMS is reported to be more common in the pediatric population [20]. Rarely, LGFMS can also arise as primary disease at other anatomic sites, including the head and neck (with reported cases in the malar area, palate, masseter muscle, face, thyroid and brain) [21-26], and visceral organs such as small and large bowel [27,28], heart [29,30] and kidney [31]. LGFMS can present with late metastases, most commonly to the lungs [32], with rare reported cases of metastases to the prostate [33] and liver [34]. Patients presenting with metastatic disease may have a decadeslong history of a primary mass in the extremity or trunk [4]. It has been postulated by Maretty-Nielsen et al. [14] that LGFMS is not expected to be very chemo- or radiosensitive due to its low nuclear grade and infrequent mitotic activity. In their study, the best response to chemotherapy was short-term stabilization of disease progression with trabectedin. Another study by le Cesne et al. [35] suggested that trabectedin could offer some benefit in translocation-related soft tissue sarcomas such as LGFMS. However, surgical excision with clear resection margins remains the first line treatment option. In one series of patients in whom medium term follow-up was achieved [19], no instances of local recurrence or metastasis occurred during the follow up period, even though two thirds of patients had a marginal resection. In a larger series, Folpe et al. [4] reported a local recurrence rate of 9%, metastasis rate of 6%, and 1% of patients dying of LGFMS at a mean of 38 months and median of 24 months follow-up. Guillou et al. [11] reported a smaller series with substantially longer follow-up. Their recurrence rate and metastasis rate were both 21% for those cases presenting with only local disease, with an overall metastasis rate of 27%. Their median times to local recurrence and metastasis were 276 months and 132 months respectively, with 83% of metastatic cases occurring beyond nine years follow up [11]. Moreover, in Evans' most recent comprehensive study of 33 LGFMS cases with long term follow-up (mean of 14 years), half of the patients developed metastases and 42% died of disease [36]. Thus, the potential for late recurrences and metastatic spread is high, necessitating long-term

1.2. Ultrastructure LGFMS consists of slender spindle cells with long, narrow, delicate and mostly non-branching cell processes, embedded in a variable amount of collagenous stroma. The tumor cells have irregular nuclear outlines with finely clumped chromatin and inconspicuous nucleoli. Intra-nuclear cytoplasmic invaginations are found in some nuclei. Most cells have a variable number of cytoplasmic organelles, including free ribosomes, intermediate filaments and rough endoplasmic reticulum cisternae, which may be poorly, mildly or moderately well developed. The cisternae of the endoplasmic reticulum are often dilated and contain finely granular contents. Some cells may consist of ovoid cytoplasmic structures with a fibrillary content, consistent with angulate lysosomes [44,45]. Nielsen et al. [46] reported the ultrastructural findings of 3 cases of HSCTGRs. The neoplastic cells showed fibroblastic features with long branching rough endoplasmic reticulum complexes. In all tumors, deposits of amorphous basal lamina-like substance were admixed with abundant extracellular collagen fibers and were also seen within the dilated cisternae. Ultrastructurally, the tumor cells of SEF have fibroblastic features, such as well-developed networks of rough endoplasmic reticulum, abundant cytoplasmic intermediate filaments, and lack of basement membranes [9]. 1.3. Genetics LGFMSs are characterized in the majority of cases by a balanced translocation, t(7;16)(q34;p11), resulting in fusion of the FUS and CREB3L2 genes, with a small minority of cases showing a variant FUSCREB3L1 fusion resulting from t(11;16)(p11;p11) [6,11,47-49] The FUS gene, located at the chromosomal band 16p11, encodes an RNAbinding protein, and is fused to another transcription factor gene, CREB3L2, which is a member of the OASIS DNA-binding and basicleucine zipper family located at 7q34. The resulting fusion gene exhibits both transcription activating and oncogenic properties [50]. Gene sequencing has shown that fusion points can vary within FUS, CREB3L1 and CREB3L2 genes, with the insertion of variable numbers of intron sequences or nucleotides from the FUS and CREB3L2 genes, or whose origin is unknown [11]. Exon 6 of FUS and exon 5 of CREB3L2 are the most involved breakpoints [11]. In addition to the translocation t(7;16) 61

Annals of Diagnostic Pathology 28 (2017) 60–67

M. Mohamed et al.

(q34;p11), the FUS-CREB3L2 fusion gene may also result from more complex rearrangements involving 7q and 16p, or may be present within supernumerary ring chromosomes [51,52]. Additionally, two cases have been shown to harbor the EWSR1-CREB3L1 fusion, without morphologically different features from classical LGFMS [7]. The FUS and EWSR1 genes are highly homologous [53], and serve as alternative binding partners to a range of translocation-associated neoplasms. Proteins encoded by FUS and EWSR1 are RNA-binding proteins of the FET family, which have roles in maintaining genomic integrity, regulating gene expression and in processing mRNA and microRNAs [7,54]. These gene fusions can be detected in formalin-fixed paraffin-embedded sections by reverse transcriptase-polymerase chain reaction and fluorescence in situ hybridization (FISH) [55-57]. However, as molecular diagnostics laboratories may only have primers for the common FUS-CREB3L2 fusion, FUS rearrangement should also be sought by FISH. Additionally, FISH for EWSR1 rearrangement should be performed in tumors with the morphology of LGFMS in which FUS rearrangements and FUS-CREB3L2 and FUS-CREB3L1 fusion transcripts are undetectable. LGFMS has been shown to have a distinct gene expression profile from morphologically similar neoplasms in its differential diagnosis [58] with top regulated genes including MUC4, FOXL1 and CD24 [58]. Similar to LGFMS, in hybrid tumors with histologic features of both LGFMS and SEF, rearrangements of FUS accompanied by the CREB3L2 rearrangement account for the vast majority of fusions [38] although EWSR1 rearrangements can be present in smaller numbers [37,59]. In contrast, the majority of pure SEF tumors harbor EWSR1 rearrangements, with EWSR1-CREB3L1 and to a much lesser extent, the EWSR1CREB3L2 fusions more common than those involving FUS [38,60,61]. As rearrangements of EWSR1 and CREB3L1 predominate over those of FUS and CREB3L2 in pure SEF, this had led some authors to conclude that pure SEF and LGFMS (with either pure or mixed histologic features) are indeed separate entities [38,61].

LGFMS [62]. Beside giant rosettes and focal hypercellularity, LGFMS may display some unusual, potentially misleading features such as marked nuclear pleomorphism and/or epithelioid morphology (mimicking SEF), cyst formation, calcification, small foci of bone formation, heterotopic ossification, multinucleated giant cells, nuclear palisading (Fig. 3C), prominent hemangiopericytoma-like vasculature (resembling solitary fibrous tumor), hyalinization and tumor necrosis [11,36,63,64]. The presence of these misleading features does not rule out the diagnosis of LGFMS [15]. Tumors can show a variety of ‘high-grade’ morphologic features (hypercellularity, nuclear enlargement and hyperchromasia, tumor necrosis and high mitotic index) [4], but it is unusual for neoplasms to display all four of these parameters. These features are not thought to be of prognostic significance and none has as yet been seen to be associated with recurrent or metastatic behavior [4]. Recurrent LGFMS can also show areas with features of SEF [36]. In cases with metastatic disease, primary tumors can show typical features or can show increased cellularity and cytologic atypia [4]. Metastatic LGFMS may show classical features or can be high-grade sarcomas with either primitive or anaplastic round cell appearances [4,36], although areas of conventional LGFMS are typically present, including giant rosettes [4]. Evans has described large, moderately anaplastic round cell features (associated with high mitotic indices and focal tumor necrosis) in recurrent LGFMS as dedifferentiation [36]. Immunohistochemically, LGFMS characteristically shows strong and diffuse granular cytoplasmic immunoreactivity with MUC4 (Fig. 3D), which is a highly sensitive marker in its diagnosis, labeling up to 100% of LGFMS and has been shown to be absent in most other soft tissue neoplasms [63]. MUC4 is a high molecular weight transmembrane glycoprotein normally expressed on many epithelial surfaces [65,66] and is overexpressed in a wide range of adenocarcinomas [6770]. The mucin 4 (MUC4) gene, located on the long arm of chromosome 3 (3q29), is upregulated in LGFMS and exhibits corresponding overexpression at the protein level. When upregulated, the tumorigenic effect of MUC4 is thought to be elicited through interactions with the ERBB2 (HER2) family of growth receptors to enhance the proliferation and survival of malignant epithelial and fibroblastic cells [65,71]. However, MUC4-negative LGFMS with FUS-CREB3L2 fusion has been described, so while immunohistochemistry for MUC4 is a highly useful screening tool for neoplasms with morphologic features of LGFMS, its absence in a tumor with otherwise characteristic features should prompt molecular investigations to assess for FUS or EWSR1 rearrangements. Although MUC4 immunohistochemistry is highly sensitive for LGFMS [63], MUC4 expression is seen in other spindle cell or epithelioid neoplasms in its differential diagnosis, including 90% of biphasic synovial sarcomas, predominantly in the glandular areas [37], and some monophasic synovial sarcomas [63]. Focal expression has also been reported in 5/17 ossifying fibromyxoid tumors, 2/10 epithelioid gastrointestinal stromal tumors and 1/10 myoepithelial carcinomas [37]. With the exception of MUC4, the immunohistochemical findings in LGFMS are relatively nonspecific. The first immunohistochemical studies of LGFMS showed that most tumor cells were strongly positive for vimentin [72,73]. Since then, epithelial membrane antigen (EMA) expression has been shown to be a consistent finding [11]. CD99 and Bcl-2 expression has also been demonstrated in the majority of LGFMS [11]. However, these markers are commonly expressed in other soft tissue neoplasms such as synovial sarcoma and solitary fibrous tumor and are therefore of little diagnostic value. Focal positivity for smooth muscle actin (SMA), desmin, CD34, and cytokeratin is rarely seen, whilst S100 protein, GFAP, h-caldesmon, beta-catenin, MDM2 and CD117 are typically negative [6,11]. Rare cases demonstrating immunoreactivity with claudin-1 and DOG-1 have also been reported [74,75].

1.4. Pathologic and immunohistochemical findings Macroscopically, LGFMS appears as a well-defined mass ranging in size from 1 cm up to 20 cm, with a white fibrous cut surface, often with glistening myxoid areas. On histological examination, these tumors can be circumscribed or infiltrative, and are predominantly composed of bland-appearing spindle cells with small, angulated nuclei with inconspicuous nucleoli and scant, wispy cytoplasm (Figs. 1–F, 2A–C). Mitotic figures tend to be absent or sparse, although a mitotic index of > 5/50 high power fields and tumor cell necrosis can be seen in < 10% of cases [4]. The tumor cells are arranged in a patternless or whorled growth pattern (Figs. 1A–D), usually showing abrupt transition from myxoid to densely collagenized areas (Fig. 1A). A fascicular growth pattern is sometimes seen (Figs. 1D–F), which can be associated with cellular atypia and increased cellularity [4], and herringbone or storiform patterns can be focally present (Fig. 1F) [36]. A curvilinear capillary network (which can be similar to that seen in low-grade myxofibrosarcoma) is frequently seen in the myxoid areas (Figs. 2D–E) [11,62]. While most neoplasms are of moderate or relatively sparse cellularity, areas of high cellularity have been noted in approximately 15% of primary tumors [4]. HSCTGR, which is a morphological variant of LGFMS, is also characterized by a proliferation of bland-appearing spindle cells, with fibromyxoid areas. Scattered throughout the tumor are hyalinized, eosinophilic acellular islands surrounded by oval, epithelioid and spindle cells in a palisading pattern, producing a distinctive pattern referred to as giant rosettes (although these can vary in size and miniature forms can be present) (Figs. 2F, 3A–B) [4]. Perivascular aggregates of these cells may form incipient rosettes [11]. Intranuclear inclusions may be seen in a few of the rounded cells at the edge of the rosettes. The rosettes often coalesce into long cords or bands of dense hyalinization. Elsewhere, there may be areas indistinguishable from 62

Annals of Diagnostic Pathology 28 (2017) 60–67

M. Mohamed et al.

Fig. 1. A–F: Low-grade fibromyxoid sarcoma (LGFMS). These tumors can be circumscribed or infiltrative, and are predominantly composed of bland-appearing spindle cells with small, angulated nuclei with inconspicuous nucleoli and scant, wispy cytoplasm. Tumors are most frequently arranged in patternless or whorled growth patterns (Panels A–D), or less frequently have more fascicular architectures (Panel E). A slightly more storiform pattern is discernible in Panel F. The stroma ranges from variably collagenous (Panel C) to more delicately fibrous (Panel D) to variably myxoid (Panels A–B), and there is often abrupt transition from myxoid to collagenized areas (Panel A). Because of the marked spectrum of morphological features, LGFMS can frequently be mistaken for non-malignant tumors, and can resemble fibromatosis (Panel E) or a nerve sheath neoplasm (Panel F).

variable positivity for EMA in a proportion of cases, with a reported frequency of up to 50% [8,9]. Focal and weak S100 protein expression has been reported in a minority of cases [8,9,76]. Expression of cytokeratin was detected in 2 of 14 cases examined in the original series by Meis-Kindblom et al. [9], and in another case described by Reid et al. [77]; however, most series have reported absence of expression [10,78]. SMA, desmin, and CD34 are typically negative in SEF [8-10,78]. The immunophenotype of SEF is therefore nonspecific and somewhat similar to that of LGFMS. Immunostains are thus predominantly useful for excluding other neoplasms in the differential diagnosis.

On gross assessment, SEF has been described as a circumscribed, nodular mass often involving the underlying periosteum and ranging in size from 5 to 10 cm. It has a firm gray to white cut surface, sometimes with calcification or ossification. Microscopically, SEF consists of nests, cords and single files of epithelioid or clear cells embedded in sclerotic to fibrohyaline, eosinophilic extracellular matrix (Figs. 3E–F) [8,9], imparting an appearance resembling carcinoma, lymphoma, chondrosarcoma or even paraganglioma. Focal spindle cell morphology can be a relatively frequent feature [38]. Mitotic activity can be relatively low at 4–5 mitotic figures per 10 high power fields, and necrosis is generally present in fewer than half of cases [38]. Hybrid SEF/low-grade fibromyxoid sarcoma (SEF containing areas indistinguishable from LGFMS) also occur [43], and hybrid SEF/LGFMS can show either predominantly SEF or LGFMS morphology, and may include giant collagenous rosettes [38]. Doyle et al. reported MUC4 expression in 78% of SEFs, including 100% of hybrid LGFMS-SEFs [37]. Of the ‘pure’ SEFs, MUC4 expression was detected in 69%. Similar to LGFMS, MUC4 staining in SEF was found to be strong and diffuse. In their study, the presence of MUC4 expression did not predict the presence of FUS rearrangement, with only 40% of MUC4-positive SEFs showing FUS rearrangement. Moreover, FUS rearrangement was not detected in any MUC4-negative cases of SEF, leading the authors to conclude that MUC4 expression is a more consistent finding in SEF than FUS gene rearrangement. SEF shows

1.5. Differential diagnosis The morphologic diagnosis of LGFMS may be challenging, owing to its typically low cellularity, abundant collagen and bland cytomorphology. It is important to distinguish it from benign or low-grade fibromyxoid lesions, because of the significant potential for recurrence and late metastatic spread [2,4,73]. The differential diagnosis of LGFMS includes desmoid fibromatosis, nodular fasciitis, perineurioma, neurofibroma, myxoma, ossifying fibromyxoid tumor, dermatofibrosarcoma protuberans and low-grade myxofibrosarcoma. Areas of LGFMS can have paucicellular areas with a fascicular pattern of growth reminiscent of fibromatosis [20]. Additionally, intra-abdominal fibromatosis can 63

Annals of Diagnostic Pathology 28 (2017) 60–67

M. Mohamed et al.

Fig. 2. A–E Low-grade fibromyxoid sarcoma (LGFMS). The blandness of the cell population is discernible at low magnification (Panel A). Because this tumor can often have a relatively nondescript appearance it can be confused for a variety of tumors, including often for benign neoplasms (Panel A); even so this example shows a subtle, alternating pattern of more myxoid and more collagenous areas typical of LGFMS. The constituent cells can vary in appearance, but most typically they have small, angulated nuclei with an almost ‘rhomboid’ shape, with inconspicuous nucleoli, even chromatin and scant, wispy cytoplasm (Panel B). Sometimes the cells have more spindled nuclei (Panel C), although the cytology remains bland. Mitotic figures are generally absent or sparse, and necrosis is relatively unusual. A curvilinear capillary network (which can be similar to that seen in low-grade myxofibrosarcoma) is frequently seen in the myxoid areas (Panel D). However, the vascular network can be composed of small thin or thicker-walled vessels (Panel E), with this example resembling the vessels seen in fibromatosis.F: Hyalinizing spindle cell tumor with giant rosettes (HSCTGR): HSCTGR is a morphological variant of LGFMS, characterized by a proliferation of bland-appearing spindle cells, with fibromyxoid areas. Scattered throughout the tumor are hyalinized, eosinophilic acellular islands surrounded by oval, epithelioid and spindle cells in a palisading pattern, producing a distinctive pattern referred to as giant rosettes (although these can vary markedly in size).

show myxoid areas. However, unlike LGFMS, fibromatosis is a poorly circumscribed tumor with infiltrative borders and exhibits myofibroblastic differentiation morphologically and immunohistochemically [79]. Architecturally, fibromatosis shows distinct long sweeping fascicles of elongated myofibroblasts, with uniform cellularity and it does not demonstrate abrupt transition from myxoid areas to fibrous zones that is characteristic of LGFMS. Furthermore, cases of fibromatosis show nuclear positivity for beta-catenin and do not have the curvilinear vasculature commonly seen in the myxoid lobules of LGFMS, instead showing a distinct vascular pattern of small thick-walled vessels with larger, compressed thin-walled vessels. Like LGFMS, nodular fasciitis can have variably myxoid and fibrous zones and is well circumscribed [20]. However, histological features including a loose storiform pattern of growth, myxoid degeneration and extravasation of lymphocytes and erythrocytes, which are characteristic of nodular fasciitis, are absent in LGFMS. The vasculature in nodular fasciitis is reactive and more prominently at the periphery of the lesion, whereas in LGFMS, the curvilinear vasculature is prominent throughout the myxoid areas. Collagen rosettes, when present, can also help distinguish LGFMS from nodular fasciitis. Nodular fasciitis is associated with characteristic MYH9-USP6 gene fusions [80,81], which are not

seen in LGFMS. Soft tissue perineurioma may mimic LGFMS, as the tumor cells have bland morphology, often grow in a whorled pattern, and may show myxoid change. However, the branching curvilinear vessels of LGFMS are usually absent in perineuriomas. Although perineuriomas are almost always positive for EMA, expression of EMA is found in a smaller, albeit significant percentage of LGFMS [74]. Perineuriomas may also express Glut-1, CD34, claudin-1, and SMA, which are rarely positive in LGFMS, whilst MUC4 positivity is seldom seen in perineuriomas [82]. Desmoplastic fibroblastomas (collagenous fibromas) are also sparsely to paucicellular lesions composed of patternless distributions of spindle and stellate fibroblasts within hypovascular, prominently fibrous stroma. These lesions can be focally positive for SMA [83-86], and are associated with a characteristic translocation (2;11)(q31;q12) [87]. Superficial angiomyxomas can show variable cellularity and contain prominent vasculature and resemble superficial LGFMS;(20) however, in contrast, these can contain acellular mucin pools and muciphages, and often neutrophils amongst their sparse inflammatory mixed infiltrate, as well as sometimes intermingled epithelial structures, including squamous epithelial strands, epidermoid cysts and buds 64

Annals of Diagnostic Pathology 28 (2017) 60–67

M. Mohamed et al.

Fig. 3. A–B: Hyalinizing spindle cell tumor with giant rosettes: The rosettes can appear less distinct and more subtle (Panel A). Bland, ovoid and epithelioid cells are seen to palisade around acellular collagenous nodules (Panel B).C: LGFMS can display more unusual features, such as nuclear palisading.D: LGFMS characteristically shows strong and diffuse granular cytoplasmic immunoreactivity with MUC4. This is a highly sensitive marker in its diagnosis, labeling up to 100% of LGFMS and has been shown to be absent in most other soft tissue tumors.E–F: Sclerosing epithelioid fibrosarcoma (SEF): SEF consists of nests, cords and single files of epithelioid or clear cells with variable atypia, embedded in sclerotic to fibrohyaline, eosinophilic extracellular matrix, imparting an appearance that can mimic a variety of neoplasms including carcinoma, lymphoma, chondrosarcoma and paraganglioma.

fibromyxoid tumor (OFMT). Both tumors can appear morphologically similar, with bland ovoid cells within a fibrous to myxoid stroma. MUC4 immunoreactivity can also be seen in some cases of OFMTs. However, in OFMT, there is frequent expression of S100 protein and EAAT4, often loss of expression (deletion) of INI1 (typically in a mosaic pattern) and no FUS-CREB3L2 fusion transcripts or FUS rearrangements detected by RT-PCR or FISH, respectively [92]. LGFMS can arise in the abdomen, retroperitoneum or pelvis, and because these can express DOG1 [93] and can be relatively cellular and composed of fascicular distributions of bland spindle cells, these can be confused with gastrointestinal stromal tumor (GIST), particularly in core biopsies in which there is limited diagnostic material and as LGFMS may not be anticipated at these sites. Most GISTs show diffuse and strong CD117 expression, and harbor KIT or less frequently PDGFRA mutations, which are not associated with LGFMS. Dedifferentiated liposarcoma (DDL) can show morphologically low-grade features, and be of relatively sparse to moderate cellularity with bland spindle cells within collagenous stroma, mimicking LGFMS [94,95]. However, DDL may arise in a patient with a history of previous well-differentiated liposarcoma (WDL), or histologically may show areas of WDL. Immunohistochemically DDL typically shows diffuse and strong expression of CDK4, MDM2 and p16, and shows high-level MDM2 amplification with FISH. The two most important malignant entities in the differential diagnosis of LGFMS are dermatofibrosarcoma protuberans (DFSP) and

of basaloid cells [88,89]. Patients with superficial angiomyxomas may have Carney complex (atrial myxoma, psammomatous melanotic schwannoma, pigmented dermal lesions and endocrine overactivity), which is not associated with LGFMS [20]. Focally, LGFMS can resemble neurofibroma, which is characterized by a random arrangement of spindled cells in myxoid stroma, with fine and delicate or chunky and refractile collagen. However, unlike neurofibroma, LGFMS lacks mast cells and the tumor cells of LGFMS do not have elongated and wavy ‘nerve sheath-type’ nuclei. Neurofibroma also tends to be relatively uniform and does not show abrupt transition from myxoid to fibrous zones. Neurofibroma, while sometimes predominantly myxoid, does not have the prominent curvilinear vasculature seen in LGFMS. Collagen rosettes are not present in neurofibroma. Moreover, lesional cells in neurofibroma are positive for S100 and CD34, both of which are typically negative in LGFMS. Similar to LGFMS, myxoma consists of small, bland spindled or stellate cells, with small hyperchromatic nuclei and minimal cytoplasm. However, myxoma is far less cellular than LGFMS as cells are separated by abundant myxoid stroma containing very sparse blood vessels. If a myxoma is intramuscular, the stroma may focally extend between muscle fibers at the tumor periphery, unlike LGFMS, which is often well circumscribed. Myxomas are frequently strongly CD34 positive, in contrast to LGFMS. Rare cases of LGFMS have a surrounding rim of mature, metaplastic bone [90,91] which can create diagnostic confusion with ossifying 65

Annals of Diagnostic Pathology 28 (2017) 60–67

M. Mohamed et al.

aberration of low-grade fibromyxoid sarcoma. Am J Surg Pathol 2013;37:734–8. [8] Antonescu CR, Rosenblum MK, Pereira P, Nascimento AG, Woodruff JM. Sclerosing epithelioid fibrosarcoma: a study of 16 cases and confirmation of a clinicopathologically distinct tumor. Am J Surg Pathol 2001;25:699–709. [9] Meis-Kindblom JM, Kindblom LG, Enzinger FM. Sclerosing epithelioid fibrosarcoma: a variant of fibrosarcoma simulating carcinoma. Am J Surg Pathol 1995;19:979–93. [10] Folk GS, Williams SB, Foss RB, Fanburg-Smith JC. Oral and maxillofacial sclerosing epithelioid fibrosarcoma: report of five cases. Head Neck Pathol 2007;1:13–20. [11] Guillou L, Benhattar J, Gengler C, Gallagher G, Ranchere-Vince D, Collin F, et al. Translocation-positive low-grade fibromyxoid sarcoma: clinicopathologic and molecular analysis of a series expanding the morphologic spectrum and suggesting potential relationship to sclerosing epithelioid fibrosarcoma: a study from the French sarcoma group. Am J Surg Pathol 2007;31:1387–402. [12] Wang WL, Evans HL, Meis JM, Liegl-Atzwanger B, Bovee JV, Goldblum JR, et al. FUS rearrangements are rare in “pure” sclerosing epithelioid fibrosarcoma. Mod Pathol 2012;25:846–53. [13] Kindblom LG, Mertens F, Coindre JM, et al. Sclerosing epithelioid fibrosarcoma. In: CDM Fletcher, Bridge JA, PCW Hogendoorn, editors. WHO classification of Tumours of Soft Tissue and Bone. Lyon: IARC Press; 2013. p. 97–8. [14] Maretty-Nielsen K, Baerentzen S, Keller J, Dyrop HB, Safwat A. Low-grade fibromyxoid sarcoma: incidence, treatment strategy of metastases, and clinical significance of the FUS gene. Sarcoma 2013;2013:256280. [15] Rekhi B, Deshmukh M, Jambhekar NA. Low-grade fibromyxoid sarcoma: a clinicopathologic study of 18 cases, including histopathologic relationship with sclerosing epithelioid fibrosarcoma in a subset of cases. Ann Diagn Pathol 2011:303–11. [16] Tang Z, Zhou ZH, Lv CT, Qin LY, Wang Y, Tian G, et al. Low-grade fibromyxoid sarcoma: clinical study and case report. J Oral Maxillofac Surg 2010;68:873–84. [17] Dobin SM, Malone VS, Lopez L, Donner LR. Unusual histologic variant of a lowgrade fibromyxoid sarcoma in a 3-year-old boy with complex chromosomal translocations involving 7q34, 10q11.2, and 16p11.2 and rearrangement of the FUS gene. Pediatr Dev Pathol 2013;16:86–90. [18] Kurisaki-Arakawa A, Suehara Y, Arakawa A, Takagi T, Takahashi M, Mitani K, et al. Deeply located low-grade fibromyxoid sarcoma with FUS-CREB3L2 gene fusion in a 5-year-old boy with review of literature. Diagn Pathol 2014;9:163. [19] Rose B, Tamvakopoulos GS, Dulay K, Pollock R, Skinner J, Briggs T, et al. The clinical significance of the FUS-CREB3L2 translocation in low-grade fibromyxoid sarcoma. J Orthop Surg Res 2011;6:15. [20] Billings SD, Giblen G, Fanburg-Smith JC. Superficial low-grade fibromyxoid sarcoma (Evans tumor). A clinicopathologic analysis of 19 cases with a unique observation in the pediatric population. Am J Surg Pathol 2005;29:204–10. [21] Lee JH, Choi HJ, Jung HY. Low-grade fibromyxoid sarcoma of the malar area. Arch Plast Surg 2016;43:110–2. [22] Soma S, Bhat S, Shetty SK. Low grade fibromyxoid sarcoma of the palate: a case report. J Clin Diagn Res 2015;9:XD01–2. [23] Lee EJ, Hwang HJ, Byeon HK, et al. A low grade fibromyxoid sarcoma originating from the masseter muscle: a case report. J Med Case Reports 2015;9:176. [24] Abe Y, Hashimoto I, Nakanishi H. Recurring facial low-grade fibromyxoid sarcoma in an elderly patient: a case report. J Med Invest 2012;59:266–9. [25] Dong W, Zhang H. Low-grade fibromyxoid sarcoma of the thyroid: a case report. Ann Acad Med Singapore 2013;42:55–6. [26] Lee EJ, Hwang HJ, Byeon HK, Park HS, Choi HS. Primary intracranial low-grade fibromyxoid sarcoma (Evans tumor). J Clin Neurosci 2008;15:1298–301. [27] Laurini JA, Zhang L, Goldblum JR, Montgomery E, Folpe AL. Low-grade fibromyxoid sarcoma of the small intestine: report of 4 cases with molecular cytogenetic confirmation. Am J Surg Pathol 2011;35:1069–73. [28] Mendoza AS, O'Leary MP, Peng SK, Petrie BA, Li AI, French SW. Low-grade fibromyxoid sarcoma of the sigmoid colon. Exp Mol Pathol 2015;98:300–3. [29] Jakowski JD, Wakely Jr. PE. Primary intrathoracic low-grade fibromyxoid sarcoma. Hum Pathol 2008;39:623–8. [30] Ferlosio A, Doldo E, Polisca P, et al. Low-grade fibromyxoid sarcoma: an unusual cardiac location. Cardiovasc Pathol 2013;22:e15–7. [31] Alevizopoulos A, Mygdalis V, Tyritzis S, Stravodimos K, Constantinides CA. Lowgrade fibromyxoid sarcoma of the renal pelvis: first report. Case Rep Nephrol Urol 2012;2:87–91. [32] Chang E, Lee A, Lee E, Shin O, Kang C, Kim JM, et al. Hyalinizing spindle cell tumor with giant rosettes with pulmonary metastasis after a long hiatus: a case report. J Korean Med Sci 2004;19:619–23. [33] Baydar DE, Aki FT. Low-grade fibromyxoid sarcoma metastatic to the prostate. Ann Diagn Pathol 2011;15:64–8. [34] Konecna J, Liberale G, Haddad J, de Saint-Aubain N, El Nakadi I. Diffuse intraabdominal low grade fibromyxoid sarcoma with hepatic metastases: case report and review of the literature. Int J Surg Case Rep 2015;14:40–3. [35] Le Cesne A, Cresta S, Maki RG, Blay JY, Verweij J, Poveda A, et al. A retrospective analysis of anti-tumour activity with trabectedin in translocation-related sarcomas. Eur J Cancer 2012;48:3036–44. [36] Evans HL. Low-grade fibromyxoid sarcoma: a clinicopathologic study of 33 cases with long-term follow-up. Am J Surg Pathol 2011;35:1450–62. [37] Doyle LA, Wang WL, Dal Cin P, Lopez-Terrada D, Mertens F, Lazar AJ, et al. MUC4 is a sensitive and extremely useful marker for sclerosing epithelioid fibrosarcoma: association with FUS gene rearrangement. Am J Surg Pathol 2012;36:1444–51. [38] Prieto-Granada C, Zhang L, Chen HW, et al. A genetic dichotomy between pure sclerosing epithelioid fibrosarcoma (SEF) and hybrid SEF/low-grade fibromyxoid sarcoma: a pathologic and molecular study of 18 cases. Genes Chromosomes Cancer 2015;54:28–38.

low-grade myxofibrosarcoma. Areas within superficial LGFMS can resemble the loose storiform pattern seen in some cases of myxoid DFSP. However, superficial LGFMS lacks the cellular, tight storiform pattern or infiltration into subcutaneous adipose tissue with ‘honeycombing’ seen in classic DFSP. Moreover, tumor cells in DFSP are positive for CD34 and negative for MUC4, whilst LGFMS is negative for the former and positive for the latter. Low-grade myxofibrosarcoma is often confused with LGFMS, often on account of similar nomenclature. In contrast to LGFMS, low grade myxofibrosarcoma typically arises in the subcutis of older adults (most common in the sixth to eighth decades), shows perivascular tumor growth along prominent, thinwalled curvilinear vessels and a greater degree of nuclear atypia, and lacks the sharply demarcated alternating fibrous and myxoid lobules of LGFMS [96]. Focal nuclear atypia, when seen in LGFMS on core biopsy may make the distinction more difficult on morphological grounds. SEF can be difficult to distinguish from tumors with epithelioid/ round cell morphology, such as poorly differentiated carcinomas, lymphomas, melanomas, small round cell tumors, and myoepithelial tumors. Immunohistochemistry is useful in this instance primarily to exclude these histologic mimics. The prominent sclerotic extracellular collagenous matrix of primary SEF of bone can be mistaken for tumoral osteoid leading to diagnostic confusion with osteosarcoma [43], but osteosarcomas express SATB2 [97,98] and are negative for MUC4, while most SEFs of bone are MUC4 and SATB2 negative [43]. 2. Conclusions LGFMS is a deceptively banal spindle cell neoplasm predominating in young adults and associated in the majority of cases with a characteristic FUS-CREB3L2 gene fusion. HSCTGR is a morphological variant of LGFMS that shares the same immunophenotype and genetic profile, whilst SEF is considered by some authors to be a closely related tumor. The morphologic diagnosis of LGFMS may be challenging, owing to its typically low cellularity, abundant collagen and bland cytomorphology. However, correct recognition is crucial as this tumor has significant potential for late recurrence and metastatic spread. Confirmatory cytogenetic and molecular diagnosis with FISH and RTPCR, respectively, in the context of appropriate morphologic and immunohistochemical findings is optimal. Disclosures The authors have no conflicts of interest or funding to disclose. Acknowledgments We acknowledge support from the NIHR Royal Marsden/ICR Biomedical Research Centre. References [1] Evans HL. Low-grade fibromyxoid sarcoma. A report of two metastasizing neoplasms having a deceptively benign appearance. Am J Clin Pathol 1987;88:615–9. [2] Evans HL. Low-grade fibromyxoid sarcoma: a report of 12 cases. Am J Surg Pathol 1993;17:595–600. [3] Lane KL, Shannon RJ, Weiss SW. Hyalinizing spindle cell tumor with giant rosettes: a distinctive tumor closely resembling low-grade fibromyxoid sarcoma. Am J Surg Pathol 1997;21:1481–8. [4] Folpe AL, Lane KL, Paull G, Weiss SW. 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. Am J Surg Pathol 2000;24:1353–60. [5] Reid R, de Silva MV, Paterson L, Ryan E, Fisher C. Low-grade fibromyxoid sarcoma and hyalinizing spindle cell tumor with giant rosettes share a common t(7;16) (q34;p11) translocation. Am J Surg Pathol 2003;27:1229–36. [6] Mertens F, Fletcher CD, Antonescu CR, Coindre JM, Colecchia M, Domanski HA, et al. Clinicopathologic and molecular genetic characterization of low-grade fibromyxoid sarcoma, and cloning of a novel FUS/CREB3L1 fusion gene. Lab Invest 2005;85:408–15. [7] Lau PP, Lui PC, Lau GT, Yau DT, Cheung ET, Chan JK. A novel alternative molecular

66

Annals of Diagnostic Pathology 28 (2017) 60–67

M. Mohamed et al. [39] Righi A, Gambarotti M, Manfrini M, Benini S, Gamberi G, Cocchi S, et al. Sclerosing epithelioid fibrosarcoma of the thigh: report of two cases with synchronous bone metastases. Virchows Arch 2015;467:339–44. [40] Baydar DE, Kosemehmetoglu K, Aydin O, Bridge JA, Buyukeren B, Aki FT. Primary sclerosing epithelioid fibrosarcoma of kidney with variant histomorphologic features: report of 2 cases and review of the literature. Diagn Pathol 2015;10:186. [41] Ohlmann CH, Brecht IB, Junker K, van der Zee JA, Nistor A, Bohle RM, et al. Sclerosing epithelioid fibrosarcoma of the kidney: clinicopathologic and molecular study of a rare neoplasm at a novel location. Ann Diagn Pathol 2015;19:221–5. [42] Argani P, Lewin JR, Edmonds P, et al. Primary renal sclerosing epithelioid fibrosarcoma: report of 2 cases with EWSR1-CREB3L1 gene fusion. Am J Surg Pathol 2015;39:365–73. [43] Wojcik JB, Bellizzi AM, Dal Cin P, Bredella MA, Fletcher CD, Hornicek FJ, et al. Primary sclerosing epithelioid fibrosarcoma of bone: analysis of a series. Am J Surg Pathol 2014;38:1538–44. [44] Zamecnik M, Michal M. Low-grade fibromyxoid sarcoma: a report of eight cases with histologic, immunohistochemical, and ultrastructural study. Ann Diagn Pathol 2000;4:207–17. [45] Antonescu CR, Baren A. Spectrum of low-grade fibrosarcomas: a comparative ultrastructural analysis of low-grade myxofibrosarcoma and fibromyxoid sarcoma. Ultrastruct Pathol 2004;28:321–32. [46] Nielsen GP, Selig MK, O'Connell JX, Keel SB, Dickersin GR, Rosenberg AE. Hyalinizing spindle cell tumor with giant rosettes. Am J Surg Pathol 1999;23:1227–32. [47] Matsuyama A, Hisaoka M, Shimajiri S, Hayashi T, Imamura T, Ishida T, et al. Molecular detection of FUS-CREB3L2 fusion transcripts in low-grade fibromyxoid sarcoma using formalin-fixed, paraffin-embedded tissue specimens. Am J Surg Pathol 2006;30:1077–84. [48] Panagopoulos I, Storlazzi CT, Fletcher CD, Fletcher JA, Nascimento A, Domanski HA, et al. The chimeric FUS/CREB3l2 gene is specific for low-grade fibromyxoid sarcoma. Genes Chromosomes Cancer 2004;40:218–28. [49] Panagopoulos I, Moller E, Dahlen A, Isaksson M, Mandahl N, Vlamis-Gardikas A, et al. Characterization of the native CREB3L2 transcription factor and the FUS/ CREB3L2 chimera. Genes Chromosomes Cancer 2007;46:181–91. [50] Storlazzi CT, Mertens F, Nascimento A, Isaksson M, Wejde J, Brosjo O, et al. Fusion of the FUS and BBF2H7 genes in low grade fibromyxoid sarcoma. Hum Mol Genet 2003;12:2349–58. [51] Bartuma H, Moller E, Collin A, Domanski HA, Von Steyern FV, Mandahl N, et al. Fusion of the FUS and CREB3L2 genes in a supernumerary ring chromosome in lowgrade fibromyxoid sarcoma. Cancer Genet Cytogenet 2010;199:143–6. [52] Mezzelani A, Sozzi G, Nessling M, Riva C, Della Torre G, Testi MA, et al. Low grade fibromyxoid sarcoma. A further low-grade soft tissue malignancy characterized by a ring chromosome. Cancer Genet Cytogenet 2000;122:144–8. [53] Aman P, Panagopoulos I, Lassen C, Fioretos T, Mencinger M, Toresson H, et al. Expression patterns of the human sarcoma-associated genes FUS and EWS and the genomic structure of FUS. Genomics 1996;37:1–8. [54] Law WJ, Cann KL, Hicks GG. TLS, EWS and TAF15: a model for transcriptional integration of gene expression. Brief Funct Genomic Proteomic 2006;5:8–14. [55] Vroobel K, Gonzalez D, Wren D, Thompson L, Swansbury J, Fisher C, et al. Ancillary molecular analysis in the diagnosis of soft tissue tumours: reassessment of its utility at a specialist centre. J Clin Pathol 2016;69:505–10. [56] Thway K, Wren D, Lee J, Thompson L, Fisher C, Gonzalez D. Evaluation of the optimal provision of formalin-fixed, paraffin-embedded material for reverse transcription-PCR in soft-tissue tumour diagnosis. J Clin Pathol 2017;70:20–4. [57] Patel RM, Downs-Kelly E, Dandekar MN, Fanburg-Smith JC, Billings SD, Tubbs RR, et al. FUS (16p11) gene rearrangement as detected by fluorescence in-situ hybridization in cutaneous low-grade fibromyxoid sarcoma: a potential diagnostic tool. Am J Dermatopathol 2011;33:140–3. [58] Moller E, Hornick JL, Magnusson L, Veerla S, Domanski HA, Mertens F. FUSCREB3L2/L1-positive sarcomas show a specific gene expression profile with upregulation of CD24 and FOXL1. Clin Cancer Res 2011;17:2646–56. [59] Doyle LA, Hornick JL. EWSR1 rearrangements in sclerosing epithelioid fibrosarcoma. Am J Surg Pathol 2013;37:1630–1. [60] Fisher C. The diversity of soft tissue tumours with EWSR1 gene rearrangements: a review. Histopathology 2014;64:134–50. [61] Arbajian E, Puls F, Magnusson L, Thway K, Fisher C, Sumathi VP, et al. Recurrent EWSR1-CREB3L1 gene fusions in sclerosing epithelioid fibrosarcoma. Am J Surg Pathol 2014;38:801–8. [62] Vernon SE, Bejarano PA. Low-grade fibromyxoid sarcoma: a brief review. Arch Pathol Lab Med 2006;130:1358–60. [63] Doyle LA, Moller E, Dal Cin P, Fletcher CD, Mertens F, Hornick JL. MUC4 is a highly sensitive and specific marker for low-grade fibromyxoid sarcoma. Am J Surg Pathol 2011;35:733–41. [64] Lee AF, Yip S, Smith AC, Hayes MM, Nielsen TO, O'Connell JX. Low-grade fibromyxoid sarcoma of the perineum with heterotopic ossification: case report and review of the literature. Hum Pathol 2011;42:1804–9. [65] Bafna S, Kaur S, Batra SK. Membrane-bound mucins: the mechanistic basis for alterations in the growth and survival of cancer cells. Oncogene 2010;29:2893–904. [66] Chaturvedi P, Singh AP, Batra SK. Structure, evolution, and biology of the MUC4 mucin. FASEB J 2008;22:966–81. [67] Andrianifahanana M, Moniaux N, Schmied BM, Ringel J, Friess H, Hollingsworth MA, et al. Mucin (MUC) gene expression in human pancreatic adenocarcinoma and chronic pancreatitis: a potential role of MUC4 as a tumor marker of diagnostic significance. Clin Cancer Res 2001;7:4033–40. [68] Park HU, Kim JW, Kim GE, Bae HI, Crawley SC, Yang SC, et al. Aberrant expression

[69]

[70]

[71]

[72] [73]

[74] [75]

[76]

[77] [78]

[79] [80]

[81]

[82]

[83]

[84]

[85] [86]

[87]

[88]

[89]

[90]

[91]

[92] [93] [94]

[95]

[96]

[97] [98]

67

of MUC3 and MUC4 membrane-associated mucins and sialyl Le(x) antigen in pancreatic intraepithelial neoplasia. Pancreas 2003;26:e48–54. Rakha EA, Boyce RW, Abd El-Rehim D, Kurien T, Green AR, Paish EC, et al. Expression of mucins (MUC1, MUC2, MUC3, MUC4, MUC5AC and MUC6) and their prognostic significance in human breast cancer. Mod Pathol 2005;18:1295–304. Singh AP, Chauhan SC, Bafna S, Johansson SL, Smith LM, Moniaux N, et al. Aberrant expression of transmembrane mucins, MUC1 and MUC4, in human prostate carcinomas. Prostate 2006;66:421–9. Bafna S, Singh AP, Moniaux N, Eudy JD, Meza JL, Batra SK. MUC4, a multifunctional transmembrane glycoprotein, induces oncogenic transformation of NIH3T3 mouse fibroblast cells. Cancer Res 2008;68:9231–8. Fukunaga M, Ushigome S, Fukunaga N. Low-grade fibromyxoid sarcoma. Virchows Arch 1996;429:301–3. Goodlad JR, Mentzel T, Fletcher CD. Low grade fibromyxoid sarcoma: clinicopathological analysis of eleven new cases in support of a distinct entity. Histopathology 1995;26:229–37. Thway K, Fisher C, Debiec-Rychter M, Calonje E. Claudin-1 is expressed in perineurioma-like low-grade fibromyxoid sarcoma. Hum Pathol 2009;40:1586–90. Thway K, Ng W, Benson C, Chapman J, Fisher C. DOG1 expression in low-grade fibromyxoid sarcoma: a study of 11 cases, with molecular characterization. Int J Surg Pathol 2015;23:454–60. Gisselsson D, Andreasson P, Meis-Kindblom JM, Kindblom LG, Mertens F, Mandahl N. Amplification of 12q13 and 12q15 sequences in a sclerosing epithelioid fibrosarcoma. Cancer Genet Cytogenet 1998;107:102–6. Reid R, Barrett A, Hamblen DL. Sclerosing epithelioid fibrosarcoma. Histopathology 1996;28:451–5. Eyden BP, Manson C, Banerjee SS, Roberts IS, Harris M. Sclerosing epithelioid fibrosarcoma: a study of five cases emphasizing diagnostic criteria. Histopathology 1998;33:354–60. Fisher C, Thway K. Aggressive fibromatosis. Pathology 2014;46:135–40. Erickson-Johnson MR, Chou MM, Evers BR, Roth CW, Seys AR, Jin L, et al. Nodular fasciitis: a novel model of transient neoplasia induced by MYH9-USP6 gene fusion. Lab Invest 2011;91:1427–33. Amary MF, Ye H, Berisha F, Tirabosco R, Presneau N, Flanagan AM. Detection of USP6 gene rearrangement in nodular fasciitis: an important diagnostic tool. Virchows Arch 2013;463:97–8. Yang EJ, Hornick JL, Qian X. Fine-needle aspiration of soft tissue perineurioma: a comparative analysis of cytomorphology and immunohistochemistry with benign and malignant mimics. Cancer Cytopathol 2016;124:651–8. Hasegawa T, Shimoda T, Hirohashi S, Hizawa K, Sano T. Collagenous fibroma (desmoplastic fibroblastoma): report of four cases and review of the literature. Arch Pathol Lab Med 1998;122:455–60. Miettinen M, Fetsch JF. Collagenous fibroma (desmoplastic fibroblastoma): a clinicopathologic analysis of 63 cases of a distinctive soft tissue lesion with stellateshaped fibroblasts. Hum Pathol 1998;29:676–82. Nielsen GP, O'Connell JX, Dickersin GR, Rosenberg AE. Collagenous fibroma (desmoplastic fibroblastoma): a report of seven cases. Mod Pathol 1996;9:781–5. Sciot R, Samson I, van den Berghe H, Van Damme B, Dal Cin P. Collagenous fibroma (desmoplastic fibroblastoma): genetic link with fibroma of tendon sheath? Mod Pathol 1999;12:565–8. Bernal K, Nelson M, Neff JR, Nielsen SM, Bridge JA. Translocation (2;11)(q31;q12) is recurrent in collagenous fibroma (desmoplastic fibroblastoma). Cancer Genet Cytogenet 2004;149:161–3. Calonje E, Guerin D, McCormick D, Fletcher CD. Superficial angiomyxoma: clinicopathologic analysis of a series of distinctive but poorly recognized cutaneous tumors with tendency for recurrence. Am J Surg Pathol 1999;23:910–7. Allen PW, Dymock RB, MacCormac LB. Superficial angiomyxomas with and without epithelial components. Report of 30 tumors in 28 patients. Am J Surg Pathol 1988;12:519–30. Thway K, Chisholm J, Hayes A, Swansbury J, Fisher C. Pediatric low-grade fibromyxoid sarcoma mimicking ossifying fibromyxoid tumor: adding to the diagnostic spectrum of soft tissue tumors with a bony shell. Hum Pathol 2015;46:461–6. Hisaoka M, Matsuyama A, Aoki T, Sakamoto A, Yokoyama K. Low-grade fibromyxoid sarcoma with prominent giant rosettes and heterotopic ossification. Pathol Res Pract 2012;208:557–60. Schneider N, Fisher C, Thway K. Ossifying fibromyxoid tumor: morphology, genetics, and differential diagnosis. Ann Diagn Pathol 2016;20:52–8. Ng W, Lee J, Vroobel K, Thway K. DOG1 expression in soft tissue tumors. Int J Surg Pathol 2016;24:274–6. Thway K, Jones RL, Noujaim J, Zaidi S, Miah AB, Fisher C. Dedifferentiated liposarcoma: updates on morphology, genetics, and therapeutic strategies. Adv Anat Pathol 2016;23:30–40. Henricks WH, Chu YC, Goldblum JR, Weiss SW. Dedifferentiated liposarcoma: a clinicopathological analysis of 155 cases with a proposal for an expanded definition of dedifferentiation. Am J Surg Pathol 1997;21:271–81. Oda Y, Takahira T, Kawaguchi K, Yamamoto H, Tamiya S, Matsuda S, et al. Lowgrade fibromyxoid sarcoma versus low-grade myxofibrosarcoma in the extremities and trunk: a comparison of clinicopathological and immunohistochemical features. Histopathology 2004;45:29–38. Conner JR, Hornick JL. SATB2 is a novel marker of osteoblastic differentiation in bone and soft tissue tumours. Histopathology 2013;63:36–49. Ordonez NG. SATB2 is a novel marker of osteoblastic differentiation and colorectal adenocarcinoma. Adv Anat Pathol 2014;21:63–7.