Annals of Diagnostic Pathology 18 (2014) 109–116

Contents lists available at ScienceDirect

Annals of Diagnostic Pathology

Review Article

Malignant peripheral nerve sheath tumor: pathology and genetics☆ Khin Thway, MBBS, BSc, FRCPath ⁎, Cyril Fisher, MD, DSc, FRCPath Department of Histopathology, Sarcoma Unit, Royal Marsden NHS Foundation Trust, 203 Fulham Road, London SW3 6JJ, UK

a r t i c l e

i n f o

Keywords: Malignant peripheral nerve sheath tumor Neurofibromatosis type 1 Neurofibrosarcoma Sarcoma Triton tumor

a b s t r a c t Malignant peripheral nerve sheath tumors are soft tissue neoplasms that show differentiation toward cells of the nerve sheath. They often arise from peripheral nerves or preexisting benign nerve sheath tumors and are generally high-grade neoplasms, which behave aggressively with high incidence of distant metastases. Malignant peripheral nerve sheath tumor can be histologically diverse and is difficult to diagnose because of its morphological overlap with a variety of other sarcomas and its lack of specific immunohistochemical markers or genetic profile. We review the pathology of malignant peripheral nerve sheath tumor, with reference to etiology, molecular genetics, and clinical factors. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Malignant peripheral nerve sheath tumors (MPNSTs) are soft tissue neoplasms that usually arise from peripheral nerves and show variable differentiation toward one of the cellular components of the nerve sheath (Schwann cells, fibroblasts, and perineurial cells). They can occur sporadically or in patients with neurofibromatosis type 1 (NF1) and arise either de novo or from a preexisting neurofibroma or, rarely, schwannoma. They form a heterogeneous group of neoplasms with a range of morphology and are often aggressive tumors with a tendency to recur and metastasize. Effective targeted molecular treatments are not available, and surgical resection remains the mainstay of treatment. Because of their morphologic heterogeneity and the lack of specific immunohistochemical or molecular markers, histologic diagnosis is challenging. Malignant peripheral nerve sheath tumors mainly affect adults with a roughly equal sex distribution, and although the age range is wide, they tend to occur at a younger mean age in patients with NF1. More rarely, MPNST can develop during childhood [1,2]. Approximately 10% of tumors are associated with previous radiation exposure, either therapeutic or environmental, and developing after a latent period [3,4]. A specific “cell of origin” is as yet unknown, although it has been postulated that they may develop from neural crest cells. Malignant peripheral nerve sheath tumor occurs most frequently in the extremities, particularly proximally, followed by the trunk and head and neck. Most arise in major nerve trunks such as the sciatic nerve. Patients may present with a painful or rapidly enlarging mass

☆ Disclosures: The authors have no conflicts of interest or funding to disclose. ⁎ Corresponding author: Department of Histopathology, The Royal Marsden NHS Foundation Trust, 203 Fulham Road, London SW3 6JJ, United Kingdom. Tel.: +44 207 808 2631; fax: +44 207 808 2578. E-mail address: [email protected] (K. Thway). 1092-9134/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.anndiagpath.2013.10.007

with associated neurologic deficits. Poorer prognosis is associated with large tumors (with size varying from N5 to N7 cm in different studies) and those associated with NF1 as well as those of higher grade and with truncal location [5,6]. Other unfavorable features include a mitotic index of greater than 6/10 high-power fields and incomplete resection [6-9]. The recurrence rate is up to 40%, and approximately two-thirds metastasize, usually hematogenously to the lungs and bone. Five-year survival has varied in series from 26% to 60%, and 10-year survival is approximately 45% [5,9,10]. Radical surgery continues to be the mainstay of current management, as these tumors have limited sensitivity to chemotherapy and radiation. The role of adjuvant treatment is as yet unclear. Radiotherapy may be used to control local disease and reduce recurrence, but it appears to have little effect on long-term survival [11,12]. Chemotherapy is generally not effective, although some studies have shown that it may benefit patients with high-grade histology [13] or children with unresectable tumors [12,14]. There are currently no effective targeted therapies for MPNST, although there are currently several preclinical and clinical studies. Potential targets include the mammalian target of rapamycin (mTOR) pathway using the mTOR inhibitor rapamycin alone or in combination with AKT inhibitors, which is demonstrating promising preclinical results for treatment of MPNST [15-19]. Patients should be referred to centers with experience in treating soft tissue sarcomas and the facility to enroll patients in clinical trials [13]. Neurofibromatosis type 1 (von Recklinghausen disease) is an autosomal dominantly inherited disorder, with a diverse range of clinical signs that affects approximately 1 in 3500 newborns. It is caused by a germ line mutation causing inactivation of the NF1 tumor suppressor gene on 17q11.2, which arises as a de novo mutation in approximately 50% of affected individuals. Although NF1 is completely penetrant, its range of manifestations is highly variable. The hallmark of NF1 is the presence of multiple dermal or plexiform neurofibromas, but the disease can have multiple manifestations in

110

K. Thway, C. Fisher / Annals of Diagnostic Pathology 18 (2014) 109–116

different organ systems. Abnormalities associated with NF1 include pigmented lesions (café au lait spots, Lisch nodules, and axillary freckling); bony dysplasias; and the occurrence of a variety of neoplasms such as optic gliomas, higher grade astrocytic neoplasms, and pheochromocytomas as well as learning disabilities [20]. The biallelic inactivation of the NF1 gene through a “second hit” seems to be of crucial importance to the development of certain manifestations, such as neurofibromas and café au lait macules, in which the second hit appears to involve only 1 cell type that is subsequently clonally expanded in a discrete lesion [21]. Patients with NF1 are at higher risk for developing soft tissue sarcomas than the general population, and MPNST is the most common malignancy associated with this disorder. The lifetime risk of MPNST in NF1 is approximately 5% to 10% (compared with 0.001% in the general population) [13], with 1% to 2% of NF1 patients developing the tumor, and studies have suggested that MPNSTs associated with NF1 are associated with a significantly poorer outcome than sporadic cases. Malignant peripheral nerve sheath tumor is the leading cause of mortality in NF1. Neurofibromatosis type 1 patients with MPNST are significantly younger at diagnosis (with a peak incidence in the fourth decade) are more often males with tumors in an axial site, may have multiple neoplasms, and show both shorter survival times and times to local recurrence and metastasis [22]. Many neoplasms arise in preexisting classical or plexiform neurofibromas or in a nerve trunk. Malignant peripheral nerve sheath tumors in NF1 patients also more frequently display divergent differentiation. Most NF1 patients carry a constitutional mutation of the NF1 gene. Its loss, as a tumor suppressor gene, predisposes to tumor development. The protein product of NF1, neurofibromin, is a ubiquitous Ras GTPase-activating protein and is a negative regulator of the Ras protooncogene and its signal transduction pathway [23]. RAS proteins (HRAS, NRAS, and KRAS) are guanosine-binding proteins that act as molecular switches controlling and regulating intracellular signaling networks, including those involved in proliferation, differentiation, migration, and apoptosis. Neurofibromatosis type 1 is considered one of the “Rasopathies,” and NF1 gene deficiency leads to Ras hyperactivation, and the increase in Ras signaling to its downstream effectors leads to the disruption of multiple pathways including Ras/ mitogen-activated protein kinase (MAPK)/ERK and Akt/mTOR. The activation profiles of the AKT/mTOR and MAPK pathways in MPNSTs still remain to be better elucidated. 2. Neurofibroma Most tumors of the human peripheral nervous system derive from Schwann cells or their precursors. Neurofibromas are benign peripheral nerve tumors that occur either sporadically or in approximately 10% are associated with NF1. The primary neoplastic cellular component of neurofibroma is the Schwann cell [24], but neurofibromas also contain a mixture of nonneoplastic peripheral nerve components, including axons, fibroblasts, perineurial cells, and inflammatory cells such as mast cells [25]. The types of neurofibroma specific for NF1 are multiple localized cutaneous neurofibroma, plexiform neurofibroma, and massive soft tissue neurofibroma [26]. However, only plexiform and localized intraneural neurofibromas are significant precursors of MPNST [26]. Although all neurofibromas carry a risk of transformation to MPNST, those that transform are usually deeply sited, and the risk is much greater in plexiform neurofibromas. Plexiform neurofibromas usually present in early childhood and are multinodular lesions involving multiple nerves or nerve branches. These generally affect small nerves but can affect the entire extremity, giving rise to “elephantiasis neuromatosa.” The risk of transformation of plexiform neurofibroma to MPNST is approximately 5% [6], and the evidence from clinical studies suggests that most MPNSTs in patients with NF1 develop from preexisting plexiform neurofibromas [27]. Atypical neurofibromas show foci of

increased cellularity, with interspersed enlarged and hyperchromatic nuclei. These tumors may be precursors of MPNST and are usually seen in patients with NF1. Atypical neurofibroma and low-grade MPNST represent a morphologic continuum, so differentiating the two is difficult, but low-grade MPNST shows more generalized nuclear atypia and increased cellularity with low levels of mitotic activity [28,29]. Neurofibromas can have features of “ancient change” similar to that seen in schwannomas, with scattered cells showing enlarged and hyperchromatic nuclei with smudgy chromatin and degenerative-type atypia. These, however, lack fascicular growth pattern, increased cellularity, or mitotic activity [25], and these features are insufficient for atypical neurofibroma. Weiss and Goldblum [28] suggest that, if there is no mitotic activity, low-grade MPNST can be diagnosed if cellularity (with almost back-to-back cells forming sheets and fascicles) and atypia are marked. Schwannoma can also rarely undergo malignant transformation [30], usually to epithelioid MPNST, although occasionally to angiosarcoma [31-33]. Malignant peripheral nerve sheath tumor has also been reported to arise from traumatic neuroma [34]. 3. Genetics, including transformation from neurofibroma Malignant peripheral nerve sheath tumors have complex karyotypes, with multiple numeric and structural abnormalities. No specific balanced gene rearrangements have been identified. The most frequent gene alterations include loss of NF1 on 17q11 and of p53 on 17q13, and these can include inactivation of the NF1 tumor suppressor gene, both in sporadic cases and NF1 patients. The molecular basis for transformation of neurofibroma to MPNST is still poorly understood. Gene expression analyses of both tumors have shown a major trend in malignant transformation consisting of loss of expression of several genes, rather than marked increase in gene expression [35]. Tumorigenesis in NF1 is thought to require the somatic loss of the second NF1 allele (biallelic NF1 gene inactivation) [36], and inactivation of both alleles has been shown in both neurofibromas and MPNST [37,38] compared with heterozygosity within nonneoplastic cells. This loss of heterozygosity in the progenitor cell, which may be a Schwann cell or its precursor, is combined with haploinsufficiency in multiple supporting cells [15]. In addition to loss of NF1, for malignant transformation to arise, it is thought that there must be a multistep accumulation of additional mutations of multiple tumor suppressor, cell cycle, and signaling regulation genes including CDKN2A, P53, RB1, SOX9, and MET with resulting abnormalities of their respective signal cascades [15] and receptor tyrosine kinase amplification (eg, epidermal growth factor receptor) [36,38-40]. Although there are no defined molecular signatures for MPNST development [36], most show a gene expression signature indicating p53 inactivation [35] as well as a relative downregulation of miR-34a compared with neurofibromas. Relatively few genes are expressed at higher levels in MPNSTs and include those involved in cell proliferation and tumor metastasis [35]. Molecular determinants of prognosis are not as yet established, but studies have shown that, for NF1-associated patients, there was a clear association between nuclear expression of p53 and poor survival [41]. Recent array comparative genomic hybridization-based studies have shown highly complex copy number alterations, with 4 regions of copy number gain associated with poor patient survival. These regions include candidate genes SOX5 (12p12.1), NOL1 and MLF2 (12p13.31), FOXM1 and FKBP1 (12p13.33), and CDK4 and TSPAN31 (12q14.1). CDK4 gain/amplification and increased FOXM1 protein expression were found to be the most significant independent predictors for poor survival in MPNST patients [42]. p-AKT, p-mTOR, and p-S6RP expressions have been associated with poor prognosis in MPNST, and mTOR inhibition by everolimus in vitro has shown antitumor activity in MPNST cell lines. Mammalian target of rapamycin inhibition is a potential treatment option for both NF1-related and sporadic MPNST [43].

K. Thway, C. Fisher / Annals of Diagnostic Pathology 18 (2014) 109–116

There is constitutive activation of several critical cell signaling cascades, and these may define therapeutic targets [36]. There is recent evidence to suggest that the malignant transformation of plexiform neurofibroma is associated with distinct changes in the expression of kinase genes. Expression profiling of 519 kinase genes in 12 matched MPNSTs arising in plexiform neurofibroma and plexiform neurofibroma samples showed that the expression pattern of kinase genes can separate MPNST from preexisting plexiform neurofibromas and that most kinase genes were down-regulated rather than overexpressed in MPNST. The patterns of expression changes were, however, complex and heterogeneous without simple single unifying recurring alterations [44]. Corresponding immunohistochemistry for mitotic regulators BUB1B, PBK, and NEK2 showed higher expression levels at the protein level, and kinases involved in mitotic regulation were particularly enriched in the pool of differentially expressed kinases. Some of these are overexpressed and are therefore possible targets for kinase inhibitors [44]. 4. Histopathology The diagnosis of MPNST can be difficult because of the absence of specific morphological, immunohistochemical, and molecular criteria. Diagnosis of MPNST requires good correlation with the clinical history, including knowledge of any history of NF1, and exclusion of morphologically similar neoplasms by immunohistochemistry ± molecular genetic analysis. This is of a sarcoma that arises from a major nerve, in which other diagnoses for neoplasms that can occur in nerves (eg, synovial sarcoma) have been excluded or which can be shown to have arisen from a preexisting nerve sheath tumor. Grossly, tumors may be within or attached to a nerve trunk, neurofibroma, or plexiform neurofibroma. They often have a fusiform appearance and can extend within a nerve. Lesions tend to be white, solid and fleshy, sometimes with myxoid change, with frequent necrosis and hemorrhage. Microscopically, MPNSTs are usually highly infiltrative lesions that display a varied range of cell morphologies (including spindle, epithelioid, pleomorphic, or small round cell) and architectural patterns. Spindle cell MPNST often shows a pattern of long fascicles or a fibrosarcoma-like herringbone pattern (Fig. 1A-E). The cells have elongated, tapered, buckled, or wavy nuclei and scanty amphophilic cytoplasm. The nuclei may be hyperchromatic or vesicular, the latter with coarse chromatin, and cells often have a blunt end and a pointed end, typical of cells of nerve sheath differentiation (Fig. 1F). Frequent morphological patterns include a “marbled” effect with alternating cellular and myxoid areas (Fig. 2A), subendothelial tumor infiltration of vessel walls, poorly defined nuclear palisading, and neuroid whorls (Fig. 2B). Hyalinized cords with surrounding cell nuclei giving a rosette-like appearance are a less common but well-recognized finding. Tumors are frequently heterogeneous in morphology, harboring a variety of patterns within the same lesion, including pleomorphic (Fig. 2C) or small round cell areas. Some neoplasms consist of sheets of frankly anaplastic spindle, ovoid, and polygonal cells typical of undifferentiated pleomorphic sarcomas. Mitoses, hemorrhage (Fig. 2D), and necrosis are frequent, and necrosis can be extensive, with islands of viable tumor condensed around vessels (Fig. 2E). The small cell variant of MPNST is rare and comprises sheets of nests of small round cells with primitive neuroepithelial differentiation [45], occasionally with rosettes. Grading can be applied according to the 3-step grading system of the French Federation of Cancer Centres Sarcoma Group [46,47]. However, most MPNSTs behave as high-grade sarcomas, and grading has not been shown to be of prognostic value [48]. Approximately 10% to 15% of MPNST are of “low-grade” morphology [26], and patterns vary from cellular/atypical neurofibroma like, low grade fibromyxoid sarcoma like, low grade epithelioid to hemangiopericytoma like [49]. When they resemble cellular or atypical neurofibroma, there is more

111

widespread nuclear atypia and mitotic activity. Rodriguez et al [25] apply the term low-grade MPNST to less anaplastic tumors arising in transition from a neurofibroma precursor. Immunohistochemically, there are no diagnostic markers of MPNST. As Schwannian differentiation in tumors is highly variable and often incomplete, S-100 protein is of limited diagnostic utility. Its expression is usually focal (Fig. 2F), with staining seen in only 50% to 60% of MPNST nuclei, and correlates with the extent of Schwannian differentiation as observed ultrastructurally [50-52]. In the negative cases, diagnosis requires the association of tumor with neurofibroma or a large nerve as well as the exclusion of other lesions in the differential diagnosis. Diffuse expression of S-100 protein is not typical of MPNST (other than the epithelioid form) and should prompt consideration of other diagnoses such as melanoma and cellular schwannoma. Other markers supportive of nerve sheath differentiation may be inconsistent, such as glial fibrillary acidic protein, or nonspecific, such as neuron-specific enolase. Malignant peripheral nerve sheath tumor can rarely express epithelial markers. Although a subset stain focally for pancytokeratin such as AE1/AE3 and for epithelial membrane antigen (EMA), most do not express cytokeratin 7 or 19, helping to differentiate these from monophasic synovial sarcoma [53]. CD34 is expressed in about a quarter of tumors. Transducin-like enhancer of split 1 (TLE1) expression was commonly seen in peripheral nerve sheath tumors, including 33% of neurofibromas, 100% of schwannomas, and 30% of MPNSTs [54]. Nestin, an intermediate filament protein expressed in neuroectodermal stem cells that is expressed primarily in mammalian nervous tissue during embryonic development [55], shows strong cytoplasmic staining in MPNST with more extensive pattern than other neural markers, compared with weak or no expression in benign nerve sheath tumors and other sarcomas [56]. It has been shown to be more sensitive for MPNST than other neural markers and, when used in combination with other markers, could be useful in establishing a diagnosis of MPNST [56]. Nuclear expression of Sox10, a neural crest transcription factor necessary for specification, maturation, and maintenance of Schwann cells and melanocytes, is found in most melanomas and up to 49% of MPNST. Sox10 expression is not seen in most nonSchwannian, nonmelanocytic tumors, and it appears to show increased specificity for tumors of neural crest origin compared with S-100 protein [57,58]. There is also preferential expression of high mobility group AT-hook 2 (HMGA2) in MPNST compared with its close morphologic mimic synovial sarcoma, and hence, immunohistochemistry for HMGA2 may be a useful marker to separate MPNST from this neoplasm. A subset of MPNSTs shows areas of divergent differentiation. This is usually to mesenchymal elements (rhabdomyosarcoma [Fig. 3A], heterologous osteoid [Fig. 3B], or osteosarcoma, chondroid, or chondrosarcoma or rarely angiosarcoma) [31,59] or rarely to epithelial elements (Fig. 3C) [60]. The most common divergent line of differentiation is toward skeletal muscle. Malignant peripheral nerve sheath tumor with divergent skeletal muscle is termed Triton tumor, but the term is not exclusive to malignant nerve sheath and refers to any benign or malignant neoplasm showing differentiation toward both neural and skeletal muscle lineages. Malignant triton tumor comprises MPNST with focal and variable numbers of rhabdomyoblasts, which may be polygonal, spindled, or bizarrely shaped. This should be distinguished from invasion of adjacent skeletal muscle. The rhabdomyoblasts often show cross striations, and there is at least focal desmin expression (Fig. 3A) as well as focal expression of myogenic nuclear transcription factors myogenin or MyoD1. Malignant Triton tumor has a worse prognosis than classical MPNST, in contrast to the other forms of heterologous differentiation (including epithelial), which are not thought to alter prognosis [6]. Epithelial differentiation is very rare and mostly occurs in patients with NF1. The epithelial elements can include mucin-secreting glands (Fig. 3C) [61] and squamous differentiation. The glandular structures

112

K. Thway, C. Fisher / Annals of Diagnostic Pathology 18 (2014) 109–116

Fig. 1. (A-E) Spindle cell MPNST often shows a pattern of long fascicles of spindle cells, with variable atypia ranging from mild to marked. (A) This tumor arose from a preexisting neurofibroma and still shows circumscription. (B) Tumors most frequently show a pattern of long fascicles and can resemble other spindle cell sarcomas such as synovial sarcoma and leiomyosarcoma. Like fibrosarcoma, MPNST is a diagnosis of exclusion so it needs a comprehensive immunohistochemical workup as well as correlation with clinical information such as any history of NF1 or of origin from a nerve or preexisting neurofibroma. (C) This neoplasm shows a striking fibrosarcoma-like herringbone pattern. (D) Spindle cell MPNST is seen spreading intraneurally within the nerve on the right, whereas an adjacent normal nerve is seen on the right. (E) MPNST arising from preexisting neurofibroma. Typical, sparsely cellular neurofibroma is seen on the right, whereas on the left, there is transformation to MPNST, comprising hypercellular distributions of moderately atypical spindle cells. (F) The nuclei of MPNST may be hyperchromatic or vesicular, the latter with coarse chromatin, and cells often have a blunt end and a pointed end, typical of cells of nerve sheath differentiation.

may express neuroendocrine markers such as chromogranin. These are usually focal and discrete and may be histologically benign or more rarely malignant. The presence of glandular elements is not thought to affect the behavior of MPNST. Another uncommon histologic variant is epithelioid MPNST (Fig. 3D), which accounts for approximately 5% of cases, and presents predominantly in superficial or deep soft tissues of the extremity, following a distribution similar to that of conventional MPNST. This can arise in a preexisting benign nerve sheath tumor, typically schwannoma, and although they can occur in patients with NF1, this is a rarer event than for classical MPNST. Development of atypical change in schwannoma has been considered as a precursor to epithelioid MPNST [33]. These tumors typically comprise vague nodules of cords, strands, or clusters of large, rounded cells with prominent nucleoli (Fig. 3D). Spindle cells typical of usual MPNST are often also present, merging with the epithelioid cells. Cells can have clear cytoplasm or a rhabdoid appearance. Myxoid changes may be seen in the stroma. Most epithelioid MPNST express S-100 protein strongly and diffusely (in contrast to conventional MPNST) and

neuron-specific enolase, but all lack melanoma-associated antigens such as HMB45. Cytokeratin [62] expression is very rare. INI1, which is ubiquitously expressed in the nuclei of most cells, is lost in 50% of epithelioid MPNST (as well as approximately 90% of epithelioid sarcomas and all malignant extrarenal rhabdoid tumors). The INI1/ SMARCB1 gene at chromosomal band 22q11.2 encodes a member of the SWI/SNF chromatin remodeling complex, which is a negative regulator of the cell cycle and functions as a tumor suppressor gene [63,64]. Epithelioid MPNST in superficial soft tissues is associated with a more favorable course than deep tumors, with a metastatic rate of only 12% as opposed to 30% for deeply sited tumors in one large series [62]. Although most MPNST differentiate toward Schwann cells, a small proportion of tumors shows perineurial cell differentiation, and these are not generally associated with NF1 [65]. These can arise on the extremities and trunk as well as the mediastinum and retroperitoneum, and this subtype shows less of a tendency to arise from a nerve than conventional MPNST and is associated with a more favorable prognosis. They may be either high or low grade and show a whorled

K. Thway, C. Fisher / Annals of Diagnostic Pathology 18 (2014) 109–116

113

Fig. 2. (A) Frequent morphological patterns include a “marbled” effect with alternating cellular and myxoid areas. (B) Neuroid whorls are a morphologic variant that can be seen focally. (C) Some MPNSTs show marked pleomorphism and may have a morphologic pattern resembling undifferentiated pleomorphic sarcoma. (D) Hemorrhage can be a prominent feature. This can closely resemble foci of heterologous angiosarcomatous differentiation, which is seen rarely in MPNST. (E) Necrosis is a frequent finding, and islands of viable tumor are often seen condensed around vessels. (F) There are no diagnostic immunohistochemical markers for MPNST. S-100 protein is of limited diagnostic utility. Its expression is supportive of the diagnosis in the correct clinical and immunohistochemical context. Expression is typically scanty and focal, and diffuse expression of S-100 protein is not typical of MPNST other than in the epithelioid form and should prompt consideration of other diagnoses such as melanoma and cellular schwannoma.

or storiform architecture within a myxoid stroma. Similar to the immunophenotype of normal perineurial cells and perineurioma, malignant perineuriomas characteristically express EMA, claudin 1, and GLUT 1 and sometimes CD34 but are usually S-100 protein negative. Ultrastructurally, they are composed of perineurial-like cells or occasionally cells intermediate between perineurial and Schwann cells [66]. 5. Differential diagnosis As MPNSTs are morphologically heterogeneous tumors that do not show a diagnostic immunoprofile, it is important first to exclude other tumors that exhibit similar histologic features or that may occur in similar clinical contexts. Other malignant neoplasms that may arise within a nerve include synovial sarcoma [67] and angiosarcoma [31]. Cytokeratin or EMA expression is usually seen focally in the spindle areas of synovial sarcoma but generally is not seen in the spindle cells of MPNST. TLE1 shows diffuse nuclear expression in virtually all synovial sarcomas such that negativity for TLE1 makes synovial sarcoma unlikely but is sometimes also positive, although more focally, in MPNST [54,68]. Synovial sarcoma also has a specific translocation, t(X;18), fusing SSX1, 2 or 4 with SS18, and the detection

of SYT-SSX fusion transcripts is diagnostically useful, but these are lacking in MPNST [69,70]. Malignant peripheral nerve sheath tumor with glands may be difficult to distinguish morphologically from biphasic synovial sarcoma, carcinosarcoma, and mixed/myoepithelial tumor. The glands of MPNST often resemble enteric epithelium and may show neuroendocrine differentiation, in contrast to those of synovial sarcoma. Myoepithelial tumors more frequently occur subcutaneously rather than deeply, are usually fairly circumscribed, and simultaneously express a range of myoepithelial-associated markers such as AE1/AE3, EMA, smooth muscle actin, and calponin in addition to S-100 protein. Unlike in MPNST, many myopeithelial tumors are associated with gene fusions involving EWSR1 with a variety of partner genes including POU5F1, PBX1, and ZNF444 [71-74]. Patients with carcinosarcoma may have a history of previous or primary carcinoma. Epithelioid MPNST can be mistaken for carcinoma or melanoma, and conventional spindle cell MPNST can appear similar to spindle cell melanoma. Epithelioid MPNSTs lack cytokeratin, HMB45, and MelanA expression. Clear cell sarcoma has a predilection for the acral extremities and shows a nested architecture of uniform spindle to epithelioid cells with small prominent nucleoli. Most clear cell sarcomas are associated with reciprocal t(12;22)(q13;q12) translocations

114

K. Thway, C. Fisher / Annals of Diagnostic Pathology 18 (2014) 109–116

Fig. 3. (A) A subset of MPNSTs shows areas of divergent differentiation, usually to mesenchymal elements. This example shows foci of rhabdomyosarcomatous differentiation (malignant Triton tumor). Note the patchy, heterogeneous expression of desmin. (B) This case shows heterologous osteoid on the left, merging with spindle cell MPNST. (C) Heterologous epithelial elements are a rare finding. These can be mucin-secreting glands, seen here, or squamous islands. Epithelial elements can be histologically benign or more rarely malignant, although the presence of glandular elements is not thought to affect the behavior of MPNST. (D) Epithelioid MPNST is a rare variant that can arise in a preexisting benign nerve sheath tumor, typically schwannoma. These neoplasms typically comprise vague nodules of cords, strands, or clusters of large, rounded cells with prominent nucleoli. Spindle cells typical of usual MPNST are often also present, merging with the epithelioid cells. Most epithelioid MPNST express S-100 protein strongly and diffusely, in contrast to conventional MPNST.

resulting in EWSR1-ATF1 fusions, with a smaller subset bearing EWSR1CREB1 fusions [75,76], and these are lacking in MPNST. Embryonal rhabdomyosarcoma, which is composed of fascicles of ovoid to spindle cells with moderately atypical hyperchromatic nuclei, can closely resemble MPNST. It often contains rhabdomyoblasts (mimicking malignant Triton tumor) but is more diffusely positive for desmin, myogenin, and MyoD1, in contrast to the focal expression of these markers in areas of typical MPNST. Patients with NF1 have a greater propensity to develop other sarcomas including gastrointestinal stromal tumor, and gastrointestinal stromal tumor needs exclusion with CD117 and DOG1 in intra-abdominal sites. The abdominopelvic region is another common site for MPNST, and dedifferentiated liposarcoma is another spindle or pleomorphic sarcoma that can be a morphological mimic. Dedifferentiated liposarcomas may have adjacent well-differentiated liposarcomatous components and show immunoreactivity for CDK4, p16, and MDM2 [77], and most harbor MDM2 gene amplifications that are detectable by fluorescence in situ hybridization [78,79]. 6. Conclusions Most malignant peripheral nerve sheath tumors are high-grade neoplasms, which behave aggressively and whose therapeutic options are limited. Their histologic features are diverse making diagnosis difficult, but as their pathogenetic mechanisms become clearer, there may emerge genetic and proteomic markers that can be exploited for diagnostic, therapeutic, and prognostic use. Acknowledgments We acknowledge NHS funding to the NIHR Biomedical Research Centre.

References [1] Coffin CM, Dehner LP. Peripheral neurogenic tumors of the soft tissues in children and adolescents: a clinicopathologic study of 139 cases. Pediatr Pathol 1989;9: 387–407. [2] deCou JM, Rao BN, Parham DM, et al. Malignant peripheral nerve sheath tumors: the St. Jude Children's Research Hospital experience. Ann Surg Oncol 1995;2: 524–9. [3] Ducatman BS, Scheithauer BW. Postirradiation neurofibrosarcoma. Cancer 1983;51:1028–33. [4] Foley KM, Woodruff JM, Ellis FT, et al. Radiation-induced malignant and atypical peripheral nerve sheath tumors. Ann Neurol 1980;7:311–8. [5] Stucky CC, Johnson KN, Gray RJ, et al. Malignant peripheral nerve sheath tumors (MPNST): the Mayo Clinic experience. Ann Surg Oncol 2012;19: 878–85. [6] Ducatman BS, Scheithauer BW, et al. Malignant peripheral nerve sheath tumors. A clinicopathologic study of 120 cases. Cancer 1986;57:2006–21. [7] Sordillo PP, Helson L, Hajdu SI, et al. Malignant schwannoma—clinical characteristics, survival, and response to therapy. Cancer 1981;47:2503–9. [8] Guccion JG, Enzinger FM. Malignant schwannoma associated with von Recklinghausen's neurofibromatosis. Virchows Arch A Pathol Anat Histol 1979;383: 43–57. [9] LaFemina J, Qin LX, Moraco NH, et al. Oncologic outcomes of sporadic, neurofibromatosis-associated, and radiation-induced malignant peripheral nerve sheath tumors. Ann Surg Oncol 2013;20:66–72. [10] Dozois EJ, Wall JC, Spinner RJ, et al. Neurogenic tumors of the pelvis: clinicopathologic features and surgical outcomes using a multidisciplinary team. Ann Surg Oncol 2009;16:1010–6. [11] Ferner RE, O'Doherty MJ. Neurofibroma and schwannoma. Curr Opin Neurol 2002;15:679–84. [12] Carroll SL, Ratner N. How does the Schwann cell lineage form tumors in NF1? Glia 2008;56:1590–605. [13] Ferrari A, Bisogno G, Carli M. Management of childhood malignant peripheral nerve sheath tumor. Paediatr Drugs 2007;9:239–48. [14] Carli M, Ferrari A, Mattke A, et al. Pediatric malignant peripheral nerve sheath tumor: the Italian and German soft tissue sarcoma cooperative group. J Clin Oncol 2005;23:8422–30. [15] Gottfried ON, Viskochil DH, Couldwell WT. Neurofibromatosis type 1 and tumorigenesis: molecular mechanisms and therapeutic implications. Neurosurg Focus 2010;28:E8. [16] Bhola P, Banerjee S, Mukherjee J, et al. Preclinical in vivo evaluation of rapamycin in human malignant peripheral nerve sheath explant xenograft. Int J Cancer 2010;126:563–71.

K. Thway, C. Fisher / Annals of Diagnostic Pathology 18 (2014) 109–116 [17] Johansson G, Mahller YY, Collins MH, et al. Effective in vivo targeting of the mammalian target of rapamycin pathway in malignant peripheral nerve sheath tumors. Mol Cancer Ther 2008;7:1237–45. [18] Zou CY, Smith KD, Zhu QS, et al. Dual targeting of AKT and mammalian target of rapamycin: a potential therapeutic approach for malignant peripheral nerve sheath tumor. Mol Cancer Ther 2009;8:1157–68. [19] Carroll SL. Molecular mechanisms promoting the pathogenesis of Schwann cell neoplasms. Acta Neuropathol 2012;123:321–48. [20] National Institutes of Health Consensus Development Conference Statement: neurofibromatosis. Bethesda, Md., USA, July 13-15, 1987. Neurofibromatosis 1988;1:172–8. [21] Jouhilahti EM, Peltonen S, Heape AM, et al. The pathoetiology of neurofibromatosis 1. Am J Pathol 2011;178:1932–9. [22] Hagel C, Zils U, Peiper M, et al. Histopathology and clinical outcome of NF1associated vs. sporadic malignant peripheral nerve sheath tumors. J Neurooncol 2007;82:187–92. [23] Lynch TM, Gutmann DH. Neurofibromatosis 1. Neurol Clin 2002;20:841–65. [24] Perry A, Roth KA, Banerjee R, et al. NF1 deletions in S-100 protein-positive and negative cells of sporadic and neurofibromatosis 1 (NF1)–associated plexiform neurofibromas and malignant peripheral nerve sheath tumors. Am J Pathol 2001;159:57–61. [25] Rodriguez FJ, Folpe AL, Giannini C, et al. Pathology of peripheral nerve sheath tumors: diagnostic overview and update on selected diagnostic problems. Acta Neuropathol 2012;123:295–319. [26] Woodruff JM. Pathology of tumors of the peripheral nerve sheath in type 1 neurofibromatosis. Am J Med Genet 1999;89:23–30. [27] Tucker T, Wolkenstein P, Revuz J, Zeller J, Friedman JM. Association between benign and malignant peripheral nerve sheath tumors in NF1. Neurology 2005;65: 205–11. [28] Weiss SW, Goldblum JR. Malignant tumors of the peripheral nerves. In: Weiss SW, Goldblum JR, editors. Enzinger and Weiss’s Soft Tissue Tumors. Philadelphia, PA: Mosby Elsevier; 2008. p. 903–44. [29] Scheithauer BW, Woodruff JM, Erlandson RA. Primary malignant tumors of peripheral nerve. In: Rosai J, editor. Atlas of Tumor Pathology. Tumors of the Peripheral Nervous System. Washington, DC: Armed Forces Institute of Pathology; 1999. p. 303–58. [30] Endo M, Yamamoto H, Harimaya K, et al. Conventional spindle cell-type malignant peripheral nerve sheath tumor arising in a sporadic schwannoma. Hum Pathol 2013;44:2845–8. [31] Mentzel T, Katenkamp D. Intraneural angiosarcoma and angiosarcoma arising in benign and malignant peripheral nerve sheath tumours: clinicopathological and immunohistochemical analysis of four cases. Histopathology 1999;35: 114–20. [32] Ruckert RI, Fleige B, Rogalla P, et al. Schwannoma with angiosarcoma. Report of a case and comparison with other types of nerve tumors with angiosarcoma. Cancer 2000;89:1577–85. [33] McMenamin ME, Fletcher CD. Expanding the spectrum of malignant change in schwannomas: epithelioid malignant change, epithelioid malignant peripheral nerve sheath tumor, and epithelioid angiosarcoma: a study of 17 cases. Am J Surg Pathol 2001;25:13–25. [34] Kos Z, Robertson SJ, Purgina BM, et al. Malignant peripheral nerve sheath tumor arising in a traumatic neuroma: a case report. Hum Pathol 2013;44:2360–4. [35] Subramanian S, Thayanithy V, West RB, et al. Genome-wide transcriptome analyses reveal p53 inactivation mediated loss of miR-34a expression in malignant peripheral nerve sheath tumours. J Pathol 2010;220:58–70. [36] Upadhyaya M. Genetic basis of tumorigenesis in NF1 malignant peripheral nerve sheath tumors. Front Biosci 2011;16:937–51. [37] Sawada S, Florell S, Purandare SM, et al. Identification of NF1 mutations in both alleles of a dermal neurofibroma. Nat Genet 1996;14:110–2. [38] Menon AG, Anderson KM, Riccardi VM, et al. Chromosome 17p deletions and p53 gene mutations associated with the formation of malignant neurofibrosarcomas in von Recklinghausen neurofibromatosis. Proc Natl Acad Sci U S A 1990;87:5435–9. [39] Kourea HP, Cordon-Cardo C, Dudas M, et al. Expression of p27(kip) and other cell cycle regulators in malignant peripheral nerve sheath tumors and neurofibromas: the emerging role of p27(kip) in malignant transformation of neurofibromas. Am J Pathol 1999;155:1885–91. [40] Nielsen GP, Stemmer-Rachamimov AO, Ino Y, et al. Malignant transformation of neurofibromas in neurofibromatosis 1 is associated with CDKN2A/p16 inactivation. Am J Pathol 1999;155:1879–84. [41] Brekke HR, Kolberg M, Skotheim RI, et al. Identification of p53 as a strong predictor of survival for patients with malignant peripheral nerve sheath tumors. Neuro Oncol 2009;11:514–28. [42] Yu J, Deshmukh H, Payton JE, et al. Array-based comparative genomic hybridization identifies CDK4 and FOXM1 alterations as independent predictors of survival in malignant peripheral nerve sheath tumor. Clin Cancer Res 2011;17: 1924–34. [43] Endo M, Yamamoto H, Setsu N, et al. Prognostic significance of AKT/mTOR and MAPK pathways and antitumor effect of mTOR inhibitor in NF1-related and sporadic malignant peripheral nerve sheath tumors. Clin Cancer Res 2013;19: 450–61. [44] Stricker TP, Henriksen KJ, Tonsgard JH, et al. Expression profiling of 519 kinase genes in matched malignant peripheral nerve sheath tumor/plexiform neurofibroma samples is discriminatory and identifies mitotic regulators BUB1B, PBK and NEK2 as overexpressed with transformation. Mod Pathol 2013;26:930–43.

115

[45] Abe S, Imamura T, Park P, et al. Small round-cell type of malignant peripheral nerve sheath tumor. Mod Pathol 1998;11:747–53. [46] Coindre JM. Grading of soft tissue sarcomas: review and update. Arch Pathol Lab Med 2006;130:1448–53. [47] Guillou L, Coindre JM, Bonichon F, et al. Comparative study of the National Cancer Institute and French Federation of Cancer Centers Sarcoma Group grading systems in a population of 410 adult patients with soft tissue sarcoma. J Clin Oncol 1997;15:350–62. [48] Coindre JM, Terrier P, Guillou L, et al. Predictive value of grade for metastasis development in the main histologic types of adult soft tissue sarcomas: a study of 1240 patients from the French Federation of Cancer Centers Sarcoma Group. Cancer 2001;91:1914–26. [49] Yamaguchi U, Hasegawa T, Hirose T, et al. Low grade malignant peripheral nerve sheath tumour: varied cytological and histological patterns. J Clin Pathol 2003;56: 826–30. [50] Arpornchayanon O, Hirota T, Itabashi M, et al. Malignant peripheral nerve tumors: a clinicopathological and electron microscopic study. Jpn J Clin Oncol 1984;14: 57–74. [51] Hirose T, Hasegawa T, Kudo E, et al. Malignant peripheral nerve sheath tumors: an immunohistochemical study in relation to ultrastructural features. Hum Pathol 1992;23:865–70. [52] Fisher C, Carter RL, Ramachandra S, et al. Peripheral nerve sheath differentiation in malignant soft tissue tumours: an ultrastructural and immunohistochemical study. Histopathology 1992;20:115–25. [53] Smith TA, Machen SK, Fisher C, et al. Usefulness of cytokeratin subsets for distinguishing monophasic synovial sarcoma from malignant peripheral nerve sheath tumor. Am J Clin Pathol 1999;112:641–8. [54] Kosemehmetoglu K, Vrana JA, Folpe AL. TLE1 expression is not specific for synovial sarcoma: a whole section study of 163 soft tissue and bone neoplasms. Mod Pathol 2009;22:872–8. [55] Krupkova Jr O, Loja T, Zambo I, et al. Nestin expression in human tumors and tumor cell lines. Neoplasma 2010;57:291–8. [56] Shimada S, Tsuzuki T, Kuroda M, et al. Nestin expression as a new marker in malignant peripheral nerve sheath tumors. Pathol Int 2007;57:60–7. [57] Karamchandani JR, Nielsen TO, van de Rijn M, et al. Sox10 and S100 in the diagnosis of soft-tissue neoplasms. Appl Immunohistochem Mol Morphol 2012;20:445–50. [58] Nonaka D, Chiriboga L, Rubin BP. Sox10: a pan-schwannian and melanocytic marker. Am J Surg Pathol 2008;32:1291–8. [59] Morphopoulos GD, Banerjee SS, Ali HH, et al. Malignant peripheral nerve sheath tumour with vascular differentiation: a report of four cases. Histopathology 1996;28:401–10. [60] Ducatman BS, Scheithauer BW. Malignant peripheral nerve sheath tumors with divergent differentiation. Cancer 1984;54:1049–57. [61] Woodruff JM. Peripheral nerve tumors showing glandular differentiation (glandular schwannomas). Cancer 1976;37:2399–413. [62] Laskin WB, Weiss SW, Bratthauer GL. Epithelioid variant of malignant peripheral nerve sheath tumor (malignant epithelioid schwannoma). Am J Surg Pathol 1991;15:1136–45. [63] Isakoff MS, Sansam CG, Tamayo P, et al. Inactivation of the Snf5 tumor suppressor stimulates cell cycle progression and cooperates with p53 loss in oncogenic transformation. Proc Natl Acad Sci U S A 2005;102:17745–50. [64] Versteege I, Medjkane S, Rouillard D, et al. A key role of the hSNF5/INI1 tumour suppressor in the control of the G1-S transition of the cell cycle. Oncogene 2002;21:6403–12. [65] Hirose T, Sumitomo M, Kudo E, et al. Malignant peripheral nerve sheath tumor (MPNST) showing perineurial cell differentiation. Am J Surg Pathol 1989;13: 613–20. [66] Hirose T, Scheithauer BW, Sano T. Perineurial malignant peripheral nerve sheath tumor (MPNST): a clinicopathologic, immunohistochemical, and ultrastructural study of seven cases. Am J Surg Pathol 1998;22:1368–78. [67] Fisher C. Synovial sarcoma. Ann Diagn Pathol 1998;2:401–21. [68] Foo WC, Cruise MW, Wick MR, et al. Immunohistochemical staining for TLE1 distinguishes synovial sarcoma from histologic mimics. Am J Clin Pathol 2011;135:839–44. [69] van de Rijn M, Barr FG, Collins MH, et al. Absence of SYT-SSX fusion products in soft tissue tumors other than synovial sarcoma. Am J Clin Pathol 1999;112:43–9. [70] Ladanyi M, Woodruff JM, Scheithauer BW, et al. Re: O'Sullivan MJ, Kyriakos M, Zhu X, et al: malignant peripheral nerve sheath tumors with t(X;18). A pathologic and molecular genetic study. Mod Pathol 2000;13:1336-1346. Mod Pathol 2001;14: 733–7. [71] Antonescu CR, Zhang L, Chang NE, et al. EWSR1-POU5F1 fusion in soft tissue myoepithelial tumors. A molecular analysis of sixty-six cases, including soft tissue, bone, and visceral lesions, showing common involvement of the EWSR1 gene. Genes Chromosomes Cancer 2010;49:1114–24. [72] Balogh Z, Deak L, Sapi Z. Malignant myoepithelioma of soft tissue: a case report with cytogenetic findings. Cancer Genet Cytogenet 2008;183:121–4. [73] Brandal P, Panagopoulos I, Bjerkehagen B, et al. Detection of a t(1;22)(q23;q12) translocation leading to an EWSR1-PBX1 fusion gene in a myoepithelioma. Genes Chromosomes Cancer 2008;47:558–64. [74] Brandal P, Panagopoulos I, Bjerkehagen B, et al. t(19;22)(q13;q12) Translocation leading to the novel fusion gene EWSR1-ZNF444 in soft tissue myoepithelial carcinoma. Genes Chromosomes Cancer 2009;48:1051–6. [75] Hisaoka M, Ishida T, Kuo TT, et al. Clear cell sarcoma of soft tissue: a clinicopathologic, immunohistochemical, and molecular analysis of 33 cases. Am J Surg Pathol 2008;32:452–60.

116

K. Thway, C. Fisher / Annals of Diagnostic Pathology 18 (2014) 109–116

[76] Wang WL, Mayordomo E, Zhang W, et al. Detection and characterization of EWSR1/ATF1 and EWSR1/CREB1 chimeric transcripts in clear cell sarcoma (melanoma of soft parts). Mod Pathol 2009;22:1201–9. [77] Thway K, Flora R, Shah C, et al. Diagnostic utility of p16, CDK4, and MDM2 as an immunohistochemical panel in distinguishing well-differentiated and dedifferentiated liposarcomas from other adipocytic tumors. Am J Surg Pathol 2012;36: 462–9.

[78] Sirvent N, Coindre JM, Maire G, et al. Detection of MDM2-CDK4 amplification by fluorescence in situ hybridization in 200 paraffin-embedded tumor samples: utility in diagnosing adipocytic lesions and comparison with immunohistochemistry and real-time PCR. Am J Surg Pathol 2007;31:1476–89. [79] Coindre JM, Hostein I, Maire G, et al. Inflammatory malignant fibrous histiocytomas and dedifferentiated liposarcomas: histological review, genomic profile, and MDM2 and CDK4 status favour a single entity. J Pathol 2004;203:822–30.