Histopathology 2016, 68, 286–296. DOI: 10.1111/his.12756
Cutaneous epithelioid malignant peripheral nerve sheath tumour: a clinicopathological analysis of 11 cases Bostjan Luzar, Rebecca Shanesmith,1 Rathi Ramakrishnan,2 Cyril Fisher3 & Eduardo Calonje4 Institute of Pathology, Medical Faculty University of Ljubljana, Ljubljana, Slovenia, 1Division of Dermatology, East Carolina University, Greenville, NC, USA, 2Imperial College Hospitals NHS Trust, London, UK, 3Department of Histopathology and Sarcoma Unit, The Royal Marsden NHS Foundation Trust, London, UK, and 4Department of Dermatopathology, St John’s Institute of Dermatology, St Thomas’ Hospital, London, UK Date of submission 28 April 2015 Accepted for publication 5 June 2015 Published online Article Accepted 10 June 2015
Luzar B, Shanesmith R, Ramakrishnan R, Fisher C & Calonje E (2016) Histopathology 68, 286–296. DOI: 10.1111/his.12756
Cutaneous epithelioid malignant peripheral nerve sheath tumour: a clinicopathological analysis of 11 cases Aims: Epithelioid malignant peripheral nerve sheath tumour (E-MPNST) is a distinctive variant of malignant peripheral nerve sheath tumour characterized by the predominance of epithelioid cells, diffuse S100 positivity and infrequent association with neurofibromatosis type 1. The aim of this study was to further delineate clinicopathological features of cutaneous E-MPNST, correlate them with disease outcome and discuss differential diagnosis. Methods and results: We analysed 11 cutaneous E-MPNSTs (six males, five females, median age 49 years, median size 1.6 cm). Tumours showed a predilection for lower extremities (45%) and trunk (45%), followed by upper extremity (9%). Follow-up was available for nine of 11 patients (range 24–100 months, median 52 months). Four patients
had an uneventful clinical course (44%), two developed local recurrence(s) (22%) and three died due to disseminated disease (33%). No histological parameters were found to predict local recurrence(s), development of distant metastases or disease outcome, including size, percentage of epithelioid component, number of mitoses per 10 high-power fields, degree of nuclear atypia or site of occurrence (dermis, dermis/ subcutis, subcutis) (P > 0.05). Immunohistochemically, all tumours were diffusely S100-positive, with a subset displaying loss of integrase interactor 1 (INI1) expression (50%). Conclusions: Cutaneous E-MPNST has the potential to pursue an aggressive clinical course, associated with wide dissemination and unfavourable disease outcome.
Keywords: clinicopathological analysis, cutaneous, immunohistochemistry, INI1 loss, malignant epithelioid peripheral nerve sheath tumour, malignant epithelioid schwannoma, S100 positivity, superficial, survival
Introduction The term ‘malignant epithelioid neurilemmoma’ (schwannoma) was coined by McCormack et al.1 in 1954 for a tumour arising within the peroneal nerve of a 30-year-old male patient, which pursued an aggressive clinical course resulting in disseminated Address for correspondence: B Luzar, Institute of Pathology, Medical Faculty University of Ljubljana, Korytkova 2, 1000 Ljubljana, Slovenia. e-mail: [email protected] © 2015 John Wiley & Sons Ltd.
disease and death of the patient. Nevertheless, it appears likely that three similar tumours were reported originally some 20 years earlier as a distinctive variant of the so-called neurogenic sarcoma by Stewart and Copeland.2 The ultrastructural features of the entity were first delineated by Alvira et al.3, confirming its neural crest origin by demonstrating characteristics of Schwann cell differentiation in tumour cells. The preferred designation, epithelioid malignant peripheral nerve sheath tumour (E-MPNST), was suggested subsequently by Enzinger
Cutaneous epithelioid malignant peripheral nerve sheath tumour
287
and Weiss.4 However, the term ‘malignant epithelioid schwannoma’ continues to be used interchangeably with E-MPNST for malignant tumours with peripheral nerve sheath differentiation composed predominantly of epithelioid cells. In the latest WHO 2013 classification of tumours of soft tissue and bone,5 E-MPNST has been classified as a variant of MPNST with distinctive histological and immunohistochemical characteristics, representing fewer than 5% of MPNSTs. As the name implies, the predominant (more than 50%) or exclusive component of the tumour is represented by polygonal atypical epithelioid cells generally displaying diffuse and at least moderate degrees of cytological atypia.1,4,6–8 Using immunohistochemistry, E-MPNST distinctively displays strong and diffuse nuclear S100 positivity, unlike classical MPNSTs, which typically only show focal S100 positivity in a limited number of cells in 50–80% of the tumours.4 In contrast to classical MPNST, occurrence from or within a pre-existing benign peripheral nerve sheath tumour is uncommon in E-MPNST, and the association with neurofibromatosis type 1 (von Recklinghausen’s disease) is exceptional.6–11 E-MPNST can be subdivided further on the basis of its occurrence/localization into the superficial (cutaneous) and deep variants.4,7 Cutaneous E-MPNST arises within the dermis and/or subcutis. The study by Laskin et al., as well as a few additional case reports, suggested a more indolent clinical behaviour of cutaneous E-MPNSTs due to their superficial location and thus probably earlier detection.7,9,12–14 However, recent analysis of 63 E-MPNSTs at cutaneous, deep soft tissue and visceral sites failed to confirm any association between the anatomical site and depth of the tumour with the disease outcome.8 Herein, we report the series of 11 cutaneous EMPNSTs and demonstrate that their biological behaviour is not as favourable as considered previously. Two of nine patients (22%) in the present study developed local recurrence and another three patients (33%) died due to the distant metastases. Unfortunately, no clinical or morphological parameters were found to predict local recurrence, development of distant metastases and/or the disease outcome, necessitating long-term follow-up for these patients.
previous or concurrent melanoma should be excluded by detailed clinical investigations. A total of 11 cases reported as epithelioid MPNSTs fulfilling the above criteria were retrieved from the referral archives of two of the authors (E.C. and C.F.). All cases were stained routinely with haematoxylin and eosin. Immunohistochemical studies were carried out on formalin-fixed, paraffin-embedded tissue using a streptavidin–biotin peroxidase detection system (Dakocytomation, Cambridge, UK). The antibodies used, their dilutions and sources are summarized in Table 1. Appropriate positive and negative controls were also included. The histological and immunohistochemical slides were reviewed and reassessed when necessary. Special emphasis was given to the following histological parameters: location of tumour within the dermis, subcutis or both, the percentage of the epithelioid component, the number of mitoses per 10 high-power fields (high-power field size = 0.25 mm2), the presence of atypical mitoses, degree of cytological atypia (mild, moderate and severe) and the presence/ absence of tumoral necrosis. The mitotic activity was calculated by selecting the area with highest mitotic activity and expressed as the number of mitoses per 10 high-power fields. Clinical information, including gender, age, size, location, clinical diagnosis and follow-up data featuring development of local recurrence, distant metastases and survival, were requested from the referring physicians in cases in which this information was missing. Statistical analysis was performed by using statistical programme SPSS version 16.0 for Windows (SPSS Inc., Chicago, IL, USA). Differences among different clinicopathological parameters were analysed by the Kruskal–Wallis test. P-values 0.05).
Discussion In the present study, we analysed 11 cutaneous E-MPNSTs. In agreement with previous reports,7–9,12– 18 the cutaneous E-MPNST arises typically on the lower extremities and trunk (45 and 45%, respectively), and shows a slight predilection for males (M: F = 1.2:1). In addition to the superficial and deep soft tissues, E-MPNST has also been reported exceptionally in visceral organs such as kidneys,19 urinary bladder,20 ovary,21 trachea,22 lungs,23 ileum,8 prostate,8 pleura,8 mediastinum8 and retroperitoneum.8 Although the majority of the cutaneous E-MPNSTs develop in the fifth decade of life (mean age = 48.6 years), we have noticed an infrequent occurrence in infancy and childhood (6 months and 9 years, respectively). Cutaneous E-MPNST does not appear to have a distinctive clinical presentation, as the majority of the lesions generally present either as a painful or a nontender, slowly growing mass, or a nodule of variable duration. As expected, due to their superficial location their size is much smaller (median size of 1.6 cm in the present study) in comparison to their deep softtissue counterparts (median size of 5 cm in the study by Laskin et al.)7. As also confirmed in the present © 2015 John Wiley & Sons Ltd, Histopathology, 68, 286–296.
293
study, the association between cutaneous and deep soft-tissue E-MPNSTs and neurofibromatosis type 1 (NF1) is exceptional.6–8,11 In contrast, approximately 50% of ordinary MPNST arise in the setting of NF124 and the lifetime risk of MPNST development in NF1 patients has been estimated to be between 5 and 10%.25 The defining histological feature of E-MPNST is epithelioid morphology of neoplastic cells, which should represent the predominant component (e.g. in excess of 50%) of the tumour cell population.1,2,4 The polygonal epithelioid cells are characterized by round to oval nuclei, vesicular chromatin, variably prominent basophilic nucleoli and abundant eosinophilic or amphophilic cytoplasm with well-defined cell membranes. The epithelioid cells display variable degrees of nuclear pleomorphism, which is usually at least moderate and widespread.8 Mitotic activity is easily identifiable. We found the mean number of seven mitoses per 10 high-power fields, while atypical mitoses were uncommon. The number of mitoses appears to reflect the degrees of nuclear atypia/pleomorphism. Tumour cells typically grow in cords, strands, nests or, alternatively, with formation of larger sheets. According to Laskin et al.7, cutaneous E-MPNST is more prone to form a single nodule being partially surrounded by a capsule of a pre-existing nerve, while deeper counterparts characteristically grow as multinodular lesions. The present study failed to demonstrate the presence of pre-existing nerves in cutaneous E-MPNST in all but one tumour, which was entirely intraneural. The tendency of recurrent lesions to acquire a higher percentage of epithelioid phenotype has already been observed previously.4,13 As in the previous study by Jo and Fletcher,8 we found no histological parameters to be predictive of local recurrence(s), development of distant metastases or disease outcome in patients with cutaneous E-MPNST, including size of the tumour, the percentage of epithelioid component, the number of mitoses per 10 high-power fields, degree of nuclear atypia or occurrence in either dermis, dermis/subcutis or subcutis. Heterologous elements, including rhabdomyosarcomatous, cartilaginous and chondro-osseous differentiation have been observed infrequently in E-MPNSTs7,8,15,17 but have, in contrast, been reported in up to 15% of classical malignant schwannomas.4,26 In one study, rhabdomyosarcomatous differentiation was associated with dismal survival of the patients.17 In addition, a single patient with cartilaginous differentiation occurring in the cutaneous E-MPNST died of disseminated disease 5 years after
292
B Luzar et al.
Table 3. Summary of the immunohistochemical features of cutaneous epithelioid malignant peripheral nerve sheath tumours in the current series and of the four previously published series containing the largest numbers of epithelioid malignant peripheral nerve sheath tumours Antibody
Current study 2015
Jo and Fletcher8
Laskin et al.7
Lodding et al.6
Morgan and Gray12
S100
11/11 (100%)
63/63 (100%)
20/25 (80%)
7/14 (50%)
3/3 (100%)
INI1
3/6 (50%)
35/52 (67%)
ND
ND
ND
GFAP
1/11
24/40 (60%)
ND
ND
ND
Melan A
1/11
0/58
ND
ND
ND
HMB45
0/11
0/57
0/15
ND
0/3
Tyrosinase
0/11
ND
ND
ND
ND
CK-MNF116
0/11
0/33
0/15
ND
0/3
MITF
0/8
0/9
ND
ND
ND
CD31
0/7
ND
ND
ND
ND
Smooth muscle actin
0/6
0/3
ND
ND
0/3
Desmin
0/6
ND
ND
ND
ND
Myogenin
0/6
ND
ND
ND
ND
CK-AE1/AE3
0/2
0/33
ND
ND
ND
Number of positive cases/number of tested cases; % percentage of positive cases; ND, Not done.
nuclear integrase interactor 1 (INI1) expression (three of six, 50%). TREATMENT AND OUTCOME
Figure 8. Cutaneous epithelioid malignant peripheral nerve sheath tumour. Diffuse and strong nuclear and cytoplasmic S100 positivity is typical of the tumour.
small peripheral nerve. Furthermore, EMA also highlighted entrapped yet non-involved small nerve endings in two additional primary intradermal/subcutaneous tumours. Three tumours displayed focal NSE positivity (three of 11, 27%) and one for GFAP (one of 11, 9%). Three of six tumours displayed loss of
Follow-up information was available for nine of the 11 cutaneous E-MPNSTs (range from 24 to 100 months, mean: 57.3 months, median: 52 months) (Table 1). All patients were treated initially with conservative surgical excision, which was marginal in two of the cases, necessitating re-excision to achieve free margins in all 11 cases. Two patients were subsequently lost to follow-up. Four patients showed no recurrence or metastatic disease in the follow-up period (four of nine, 44%), and their clinical course was uneventful. Two patients developed local recurrence (two of nine, 22%). One patient had a single local recurrence (case 8) at 36 months from the initial diagnosis, respectively, and one had two recurrences (case 2) at 36 and 60 months, with no evidence of disease after follow-up of 86 months. While a patient with a single local recurrence was treated with complete conservative re-excision (case 8), the patient with two local recurrences (case 2) was treated with a below-the-knee amputation following the second recurrence due to the © 2015 John Wiley & Sons Ltd, Histopathology, 68, 286–296.
Cutaneous epithelioid malignant peripheral nerve sheath tumour
children also display loss of INI1 expression in about 10 and 40%, respectively.29,32 The markers of epithelial, e.g. low and high molecular weight cytokeratins (CK-MNF116, CK-AE1/AE3), myogenic/myofibroblastic, e.g. desmin, myogenin and smooth muscle actin, and vascular differentiation, e.g. CD31 are consistently negative in E-MPNST. The summary of immunohistochemical features of E-MPNST in the four largest studies on E-MPNST is represented in Table 3. The main differential diagnosis of cutaneous E-MPNST includes melanoma, poorly differentiated carcinoma and other types of sarcomas with predominantly or exclusively epithelioid morphology, including epithelioid sarcoma and myoepithelial carcinoma. In addition, distinction from epithelioid schwannoma can be difficult. E-MPNST displays considerable immunohistochemical overlap with melanoma, especially considering its consistent S100 positivity,6,7 and discrimination between the two entities may be difficult. Unlike melanoma with epithelioid morphology, E-MPNST generally lacks expression of second-line markers of melanocytic differentiation, for example melan A, HMB45, tyrosinase and MITF.6,7 In particular, MITF positivity as an adjunct to S100 positivity, a combination of the two not demonstrated in E-MPNSTs, has been regarded as highly specific for melanoma.33 Furthermore, melanoma consistently retains expression of INI1 protein while loss of INI1 expression occurs commonly in E-MPNST.8,29 It is also very important to exclude any previous history of melanoma or the presence of concurrent melanoma on clinical grounds. The single tumour in our study with focal weak melan A positivity lacked expression of other markers of melanocytic differentiation, including HMB45, tyrosinase and MITF. In addition, the possibility of previous and concurrent melanoma was excluded by meticulous clinical examination and clinical history. Metastatic carcinoma can be separated easily from E-MPNST by cytokeratin positivity in the former and its consistent negativity in the latter. Epithelioid sarcoma typically shows loss of INI1 expression, positivity for both low and high molecular weight cytokeratins and is consistently negative for markers of melanocytic differentiation, including S100.29,34–36 Myoepithelial carcinoma shares similar morphological features with E-MPNST, but can be distinguished from the latter by consistently staining for cytokeratins.8,29 Epithelioid schwannomas are generally well-circumscribed and encapsulated, and lack significant atypia.8 Nevertheless, the presence of diffuse and/or significant nuclear atypia/pleomorphism, vesicular nuclei or atypical mitoses favour E-MPNST.8 © 2015 John Wiley & Sons Ltd, Histopathology, 68, 286–296.
295
In conclusion, we analysed the clinicopathological features of 11 cutaneous E-MPNSTs and correlated them with survival. Although the majority of the patients were adults in their fifth decade of life, infrequent occurrence in infancy and childhood was observed. On follow-up, 22% of the patients developed local recurrence and an additional 33% of the patients died due to disease dissemination. None of the clinicopathological parameters, including gender, age, size of the tumour, location (dermis, subcutis or both), percentage of epithelioid component, the number of mitoses per 10 high-power fields, and degree of cytological atypia had any predictive value for local recurrence(s), development of distant metastases and overall survival of patients. Immunohistochemically, all tumours were diffusely S100-positive, with a subset displaying loss of INI1 expression (50%). The results of our study confirm that cutaneous E-MPNST has the potential to pursue an aggressive clinical course, associated with wide dissemination and unfavourable disease outcome, necessitating long-term follow-up.
Acknowledgements The authors thank the following colleagues for their contributions in cases and/or clinical data and followup information: Dr S. Agarwal, Queen Elizabeth II Hospital, Welwyn Garden City, UK; Dr L. Melly, Royal Free Hospital, Pond Street, UK; Dr L. Wheen, Royal Devon and Exeter Hospital, Church Lane, UK; Dr D. E. Hughes, Royal Hallamshire Hospital, UK; Dr M. Khan, Queen Mary’s Hospital, Kent, UK; Dr Muller, Glenfield Hospital, UK; Dr C. D’Arrigo, The National Hospital for Neurology and Neurosurgery, UK; Dr C. Scott, Salisbury District Hospital, Wiltshire, UK; and Dr M. Fallowfield, Mid Essex Hospital, Chelmsford, UK.
Conflict of interest The authors declare that there are no conflicts of interest.
References 1. McCormack LJ, Hazard JB, Dickson JA. Malignant epithelioid neurilemmoma (schwannoma). Cancer 1954; 7; 725–728. 2. Stewart FW, Copeland MM. Neurogenic sarcoma. Am. J. Cancer 1931; 15; 1235–1320. 3. Alvira MM, Mandybur TI, Menefee MG. Light microscopic and ultrastructural observations of a metastasizing malignant epithelioid schwannoma. Cancer 1976; 38; 1977–1982.
296
B Luzar et al.
4. Weiss SW, Goldblum JR. Malignant tumors of the peripheral nerves. In Weiss SW, Goldblum JR eds. Enzinger & Weiss’s soft tissue tumors. 5th ed. Philadelphia, PA: Mosby Elsevier, 2008; 903–944. 5. Nielsen GP, Antonescu CR, Lothe RA. Malignant peripheral nerve sheath tumour. In Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F eds. WHO classification of tumours of soft tissue and bone. Lyon: IARC Press, 2013; 187–189. 6. Lodding P, Kindblom L-G, Angervall L. Epithelioid malignant schwannoma. A study of 14 cases. Virchows Arch A Pathol. Anat. Histopathol. 1986; 409; 433–451. 7. Laskin WB, Weiss SW, Bratthauer GL. Epithelioid variant of malignant peripheral nerve sheath tumor (malignant epithelioid schwannoma). Am. J. Surg. Pathol. 1991; 15; 1136– 1145. 8. Jo VY, Fletcher CDM. Epithelioid malignant peripheral nerve sheath tumor. Clinicopathological analysis of 63 cases. Am. J. Surg. Pathol. 2015; 39; 673–682. 9. Manganoni AM, Farisoglio C, Lonati A, Tucci G, Pinton PG. Cutaneous epithelioid malignant schwannoma: review of the literature and case report. J. Plast. Reconstr. Aesthet. Surg. 2009; 62; 318–321. 10. 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. 11. Molenaar WM, Ladde BE, Schraffordt Koops H, Dam-Meiring A. Two epithelioid malignant schwannomas in a patient with neurofibromatosis. Cytology, histology and DNA-flow-cytometry. Pathol. Res. Pract. 1989; 184; 529–534. 12. Morgan KG, Gray C. Malignant epithelioid schwannoma of superficial soft tissue? A case report with immunohistology and electron microscopy. Histopathology 1985; 9; 765–775. 13. Suster S, Amazon K, Rosen LB, Ollague JM. Malignant epithelioid schwannoma of the skin. A low-grade neurotropic malignant melanoma? Am. J. Dermatopathol. 1989; 11; 338–344. 14. Shimizu S, Teraki Y, Ishiko A et al. Malignant epithelioid schwannoma of the skin showing partial HMB-45 positivity. Am. J. Dermatopathol. 1993; 15; 378–384. 15. Yamamoto T, Minami R, Ohbayashi C. Subcutaneous malignant epithelioid schwannoma with cartilaginous differentiation. J. Cutan. Pathol. 2001; 28; 486–491. 16. Bleach NR, Keen CE, Dixon JA. Superficial malignant schwannoma of the face: a case for early radical surgery. J. Laryngol. Otol. 1989; 103; 316–318. 17. Morgan MB, Stevens L, Patterson J, Tannenbaum M. Cutaneous epithelioid malignant nerve sheath tumor with rhabdoid features: a histologic, immunohistochemical, and ultrastructural study of three cases. J. Cutan. Pathol. 2000; 27; 529–534. 18. Tsuchiya D, Takamura H, Saito K, Kashiwa H, Maeda K, Yamashita H. Immunohistochemical diagnosis of a rare case of epithelioid malignant peripheral nerve sheath tumor with multiple metastases. Jpn. J. Ophtalmol. 2004; 48; 565–569. 19. Sokolova IN, Gurariĭ LL. Malignant epithelioid schwannoma of the kidney. Arkh. Patol. 1988; 50; 59–64. 20. Eltoum IA, Moore RJ III, Cook W, Crowe DR, Rodgers WH, Siegal GP. Epithelioid variant of malignant peripheral nerve
21.
22.
23. 24.
25.
26.
27.
28.
29.
30.
31. 32.
33.
34.
35.
36.
sheath tumor (malignant schwannoma) of the urinary bladder. Ann. Diagn. Pathol. 1999; 3; 304–308. Ivaskevics K, S L aszlo A, api Z. Malignant epithelioid ovarian schwannoma: a case report. Int. J. Gynecol. Cancer 2006; 16; 360–362. Fukai I, Masaoka A, Yamakawa Y, Niwa H, Eimoto T. Mediastinal malignant epithelioid schwannoma. Chest 1995; 108; 574–575. Kitamura H, Kitamura H. Primary epithelioid malignant schwannoma of the lung. Pathol. Int. 1994; 44; 317–324. Dunn GP, Spiliopoulos K, Plotkin SR et al. Role of resection of malignant peripheral nerve sheath tumors in patients with neurofibromatosis type 1. J. Neurosurg. 2013; 118; 142–148. Thway K, Fisher C. Malignant peripheral nerve sheath tumor: pathology and genetics. Ann. Diagn. Pathol. 2014; 18; 109– 116. Hruban RH, Shiu MH, Senie RT, Woodruff JM. Malignant peripheral nerve sheath tumors of the buttock and lower extremity. A study of 43 cases. Cancer 1990; 66; 1253–1265. Thewes M, Jungfer-Weber B, Wiebecke B, Engst R, Ring J. Malignant epithelioid schwannoma with melanocytic differentiation: a rare tumor with unusual feature. Acta Derm. Venereol. 1997; 77; 493–494. Pizem J, Nicholson KM, Mraz J, Prieto VG. Melanocytic differentiation is present in a significant proportion of nonpigmented diffuse neurofibromas. A potential diagnostic pitfall. Am. J. Surg. Pathol. 2013; 37; 1182–1191. Hornick JL, Dal Cin P, Fletcher CD. Loss of INI1 expression is characteristic of both conventional and proximal-type epithelioid sarcoma. Am. J. Surg. Pathol. 2009; 33; 542–550. Carter JM, O’Hara C, Dundas G et al. Epithelioid malignant peripheral nerve sheath tumor arising in a schwannoma, in a patient with ‘neuroblastoma-like’ schwannomatosis and a novel germline SMARCB1 mutation. Am. J. Surg. Pathol. 2012; 36; 154–160. Roberts CW, Orkin SH. The SWI/SNF complex-chromatin and cancer. Nat. Rev. Cancer 2004; 4; 133–142. Gleason BC, Fletcher CD. Myoepithelial carcinoma of soft tissue in children: an aggressive neoplasm analyzed in a series of 29 cases. Am. J. Surg. Pathol. 2007; 31; 1813–1824. Miettinen M, Fernandez M, Franssila K, Gatalica Z, Lasota J, Sarlomo-Rikala M. Microphtalmia transcription factor in the immunohistochemical diagnosis of metastatic melanoma: comparison with four other melanoma markers. Am. J. Surg. Pathol. 2001; 25; 205–211. Lin L, Skacel M, Sigel JE et al. Epitheliod sarcoma: an immunohistochemical analysis evaluating the utility of cytokeratin 5/6 in distinguishing superficial epithelioid sarcoma from spindled squamous cell carcinoma. J. Cutan. Pathol. 2003; 30; 114–117. Miettinen M, Fanburg-Smith JC, Virolainen M, Shmookler BM, Fetsch JF. Epithelioid sarcoma: an immunohistochemical analysis of 112 classical and variant cases and a discussion of the differential diagnosis. Hum. Pathol. 1999; 30; 934–942. Guillou L, Wadden C, Coindre JM, Krausz T, Fletcher CD. ‘Proximal-type’ epithelioid sarcoma, a distinctive aggressive neoplasm showing rhabdoid features. Clinicopathologic, immunohistochemical, and ultrastructural study of a series. Am. J. Surg. Pathol. 1997; 21; 130–146.
© 2015 John Wiley & Sons Ltd, Histopathology, 68, 286–296.
Cutaneous epithelioid malignant peripheral nerve sheath tumour: a clinicopathological analysis of 11 cases Bostjan Luzar, Rebecca Shanesmith,1 Rathi Ramakrishnan,2 Cyril Fisher3 & Eduardo Calonje4 Institute of Pathology, Medical Faculty University of Ljubljana, Ljubljana, Slovenia, 1Division of Dermatology, East Carolina University, Greenville, NC, USA, 2Imperial College Hospitals NHS Trust, London, UK, 3Department of Histopathology and Sarcoma Unit, The Royal Marsden NHS Foundation Trust, London, UK, and 4Department of Dermatopathology, St John’s Institute of Dermatology, St Thomas’ Hospital, London, UK Date of submission 28 April 2015 Accepted for publication 5 June 2015 Published online Article Accepted 10 June 2015
Luzar B, Shanesmith R, Ramakrishnan R, Fisher C & Calonje E (2016) Histopathology 68, 286–296. DOI: 10.1111/his.12756
Cutaneous epithelioid malignant peripheral nerve sheath tumour: a clinicopathological analysis of 11 cases Aims: Epithelioid malignant peripheral nerve sheath tumour (E-MPNST) is a distinctive variant of malignant peripheral nerve sheath tumour characterized by the predominance of epithelioid cells, diffuse S100 positivity and infrequent association with neurofibromatosis type 1. The aim of this study was to further delineate clinicopathological features of cutaneous E-MPNST, correlate them with disease outcome and discuss differential diagnosis. Methods and results: We analysed 11 cutaneous E-MPNSTs (six males, five females, median age 49 years, median size 1.6 cm). Tumours showed a predilection for lower extremities (45%) and trunk (45%), followed by upper extremity (9%). Follow-up was available for nine of 11 patients (range 24–100 months, median 52 months). Four patients
had an uneventful clinical course (44%), two developed local recurrence(s) (22%) and three died due to disseminated disease (33%). No histological parameters were found to predict local recurrence(s), development of distant metastases or disease outcome, including size, percentage of epithelioid component, number of mitoses per 10 high-power fields, degree of nuclear atypia or site of occurrence (dermis, dermis/ subcutis, subcutis) (P > 0.05). Immunohistochemically, all tumours were diffusely S100-positive, with a subset displaying loss of integrase interactor 1 (INI1) expression (50%). Conclusions: Cutaneous E-MPNST has the potential to pursue an aggressive clinical course, associated with wide dissemination and unfavourable disease outcome.
Keywords: clinicopathological analysis, cutaneous, immunohistochemistry, INI1 loss, malignant epithelioid peripheral nerve sheath tumour, malignant epithelioid schwannoma, S100 positivity, superficial, survival
Introduction The term ‘malignant epithelioid neurilemmoma’ (schwannoma) was coined by McCormack et al.1 in 1954 for a tumour arising within the peroneal nerve of a 30-year-old male patient, which pursued an aggressive clinical course resulting in disseminated Address for correspondence: B Luzar, Institute of Pathology, Medical Faculty University of Ljubljana, Korytkova 2, 1000 Ljubljana, Slovenia. e-mail: [email protected] © 2015 John Wiley & Sons Ltd.
disease and death of the patient. Nevertheless, it appears likely that three similar tumours were reported originally some 20 years earlier as a distinctive variant of the so-called neurogenic sarcoma by Stewart and Copeland.2 The ultrastructural features of the entity were first delineated by Alvira et al.3, confirming its neural crest origin by demonstrating characteristics of Schwann cell differentiation in tumour cells. The preferred designation, epithelioid malignant peripheral nerve sheath tumour (E-MPNST), was suggested subsequently by Enzinger
Cutaneous epithelioid malignant peripheral nerve sheath tumour
287
and Weiss.4 However, the term ‘malignant epithelioid schwannoma’ continues to be used interchangeably with E-MPNST for malignant tumours with peripheral nerve sheath differentiation composed predominantly of epithelioid cells. In the latest WHO 2013 classification of tumours of soft tissue and bone,5 E-MPNST has been classified as a variant of MPNST with distinctive histological and immunohistochemical characteristics, representing fewer than 5% of MPNSTs. As the name implies, the predominant (more than 50%) or exclusive component of the tumour is represented by polygonal atypical epithelioid cells generally displaying diffuse and at least moderate degrees of cytological atypia.1,4,6–8 Using immunohistochemistry, E-MPNST distinctively displays strong and diffuse nuclear S100 positivity, unlike classical MPNSTs, which typically only show focal S100 positivity in a limited number of cells in 50–80% of the tumours.4 In contrast to classical MPNST, occurrence from or within a pre-existing benign peripheral nerve sheath tumour is uncommon in E-MPNST, and the association with neurofibromatosis type 1 (von Recklinghausen’s disease) is exceptional.6–11 E-MPNST can be subdivided further on the basis of its occurrence/localization into the superficial (cutaneous) and deep variants.4,7 Cutaneous E-MPNST arises within the dermis and/or subcutis. The study by Laskin et al., as well as a few additional case reports, suggested a more indolent clinical behaviour of cutaneous E-MPNSTs due to their superficial location and thus probably earlier detection.7,9,12–14 However, recent analysis of 63 E-MPNSTs at cutaneous, deep soft tissue and visceral sites failed to confirm any association between the anatomical site and depth of the tumour with the disease outcome.8 Herein, we report the series of 11 cutaneous EMPNSTs and demonstrate that their biological behaviour is not as favourable as considered previously. Two of nine patients (22%) in the present study developed local recurrence and another three patients (33%) died due to the distant metastases. Unfortunately, no clinical or morphological parameters were found to predict local recurrence, development of distant metastases and/or the disease outcome, necessitating long-term follow-up for these patients.
previous or concurrent melanoma should be excluded by detailed clinical investigations. A total of 11 cases reported as epithelioid MPNSTs fulfilling the above criteria were retrieved from the referral archives of two of the authors (E.C. and C.F.). All cases were stained routinely with haematoxylin and eosin. Immunohistochemical studies were carried out on formalin-fixed, paraffin-embedded tissue using a streptavidin–biotin peroxidase detection system (Dakocytomation, Cambridge, UK). The antibodies used, their dilutions and sources are summarized in Table 1. Appropriate positive and negative controls were also included. The histological and immunohistochemical slides were reviewed and reassessed when necessary. Special emphasis was given to the following histological parameters: location of tumour within the dermis, subcutis or both, the percentage of the epithelioid component, the number of mitoses per 10 high-power fields (high-power field size = 0.25 mm2), the presence of atypical mitoses, degree of cytological atypia (mild, moderate and severe) and the presence/ absence of tumoral necrosis. The mitotic activity was calculated by selecting the area with highest mitotic activity and expressed as the number of mitoses per 10 high-power fields. Clinical information, including gender, age, size, location, clinical diagnosis and follow-up data featuring development of local recurrence, distant metastases and survival, were requested from the referring physicians in cases in which this information was missing. Statistical analysis was performed by using statistical programme SPSS version 16.0 for Windows (SPSS Inc., Chicago, IL, USA). Differences among different clinicopathological parameters were analysed by the Kruskal–Wallis test. P-values 0.05).
Discussion In the present study, we analysed 11 cutaneous E-MPNSTs. In agreement with previous reports,7–9,12– 18 the cutaneous E-MPNST arises typically on the lower extremities and trunk (45 and 45%, respectively), and shows a slight predilection for males (M: F = 1.2:1). In addition to the superficial and deep soft tissues, E-MPNST has also been reported exceptionally in visceral organs such as kidneys,19 urinary bladder,20 ovary,21 trachea,22 lungs,23 ileum,8 prostate,8 pleura,8 mediastinum8 and retroperitoneum.8 Although the majority of the cutaneous E-MPNSTs develop in the fifth decade of life (mean age = 48.6 years), we have noticed an infrequent occurrence in infancy and childhood (6 months and 9 years, respectively). Cutaneous E-MPNST does not appear to have a distinctive clinical presentation, as the majority of the lesions generally present either as a painful or a nontender, slowly growing mass, or a nodule of variable duration. As expected, due to their superficial location their size is much smaller (median size of 1.6 cm in the present study) in comparison to their deep softtissue counterparts (median size of 5 cm in the study by Laskin et al.)7. As also confirmed in the present © 2015 John Wiley & Sons Ltd, Histopathology, 68, 286–296.
293
study, the association between cutaneous and deep soft-tissue E-MPNSTs and neurofibromatosis type 1 (NF1) is exceptional.6–8,11 In contrast, approximately 50% of ordinary MPNST arise in the setting of NF124 and the lifetime risk of MPNST development in NF1 patients has been estimated to be between 5 and 10%.25 The defining histological feature of E-MPNST is epithelioid morphology of neoplastic cells, which should represent the predominant component (e.g. in excess of 50%) of the tumour cell population.1,2,4 The polygonal epithelioid cells are characterized by round to oval nuclei, vesicular chromatin, variably prominent basophilic nucleoli and abundant eosinophilic or amphophilic cytoplasm with well-defined cell membranes. The epithelioid cells display variable degrees of nuclear pleomorphism, which is usually at least moderate and widespread.8 Mitotic activity is easily identifiable. We found the mean number of seven mitoses per 10 high-power fields, while atypical mitoses were uncommon. The number of mitoses appears to reflect the degrees of nuclear atypia/pleomorphism. Tumour cells typically grow in cords, strands, nests or, alternatively, with formation of larger sheets. According to Laskin et al.7, cutaneous E-MPNST is more prone to form a single nodule being partially surrounded by a capsule of a pre-existing nerve, while deeper counterparts characteristically grow as multinodular lesions. The present study failed to demonstrate the presence of pre-existing nerves in cutaneous E-MPNST in all but one tumour, which was entirely intraneural. The tendency of recurrent lesions to acquire a higher percentage of epithelioid phenotype has already been observed previously.4,13 As in the previous study by Jo and Fletcher,8 we found no histological parameters to be predictive of local recurrence(s), development of distant metastases or disease outcome in patients with cutaneous E-MPNST, including size of the tumour, the percentage of epithelioid component, the number of mitoses per 10 high-power fields, degree of nuclear atypia or occurrence in either dermis, dermis/subcutis or subcutis. Heterologous elements, including rhabdomyosarcomatous, cartilaginous and chondro-osseous differentiation have been observed infrequently in E-MPNSTs7,8,15,17 but have, in contrast, been reported in up to 15% of classical malignant schwannomas.4,26 In one study, rhabdomyosarcomatous differentiation was associated with dismal survival of the patients.17 In addition, a single patient with cartilaginous differentiation occurring in the cutaneous E-MPNST died of disseminated disease 5 years after
292
B Luzar et al.
Table 3. Summary of the immunohistochemical features of cutaneous epithelioid malignant peripheral nerve sheath tumours in the current series and of the four previously published series containing the largest numbers of epithelioid malignant peripheral nerve sheath tumours Antibody
Current study 2015
Jo and Fletcher8
Laskin et al.7
Lodding et al.6
Morgan and Gray12
S100
11/11 (100%)
63/63 (100%)
20/25 (80%)
7/14 (50%)
3/3 (100%)
INI1
3/6 (50%)
35/52 (67%)
ND
ND
ND
GFAP
1/11
24/40 (60%)
ND
ND
ND
Melan A
1/11
0/58
ND
ND
ND
HMB45
0/11
0/57
0/15
ND
0/3
Tyrosinase
0/11
ND
ND
ND
ND
CK-MNF116
0/11
0/33
0/15
ND
0/3
MITF
0/8
0/9
ND
ND
ND
CD31
0/7
ND
ND
ND
ND
Smooth muscle actin
0/6
0/3
ND
ND
0/3
Desmin
0/6
ND
ND
ND
ND
Myogenin
0/6
ND
ND
ND
ND
CK-AE1/AE3
0/2
0/33
ND
ND
ND
Number of positive cases/number of tested cases; % percentage of positive cases; ND, Not done.
nuclear integrase interactor 1 (INI1) expression (three of six, 50%). TREATMENT AND OUTCOME
Figure 8. Cutaneous epithelioid malignant peripheral nerve sheath tumour. Diffuse and strong nuclear and cytoplasmic S100 positivity is typical of the tumour.
small peripheral nerve. Furthermore, EMA also highlighted entrapped yet non-involved small nerve endings in two additional primary intradermal/subcutaneous tumours. Three tumours displayed focal NSE positivity (three of 11, 27%) and one for GFAP (one of 11, 9%). Three of six tumours displayed loss of
Follow-up information was available for nine of the 11 cutaneous E-MPNSTs (range from 24 to 100 months, mean: 57.3 months, median: 52 months) (Table 1). All patients were treated initially with conservative surgical excision, which was marginal in two of the cases, necessitating re-excision to achieve free margins in all 11 cases. Two patients were subsequently lost to follow-up. Four patients showed no recurrence or metastatic disease in the follow-up period (four of nine, 44%), and their clinical course was uneventful. Two patients developed local recurrence (two of nine, 22%). One patient had a single local recurrence (case 8) at 36 months from the initial diagnosis, respectively, and one had two recurrences (case 2) at 36 and 60 months, with no evidence of disease after follow-up of 86 months. While a patient with a single local recurrence was treated with complete conservative re-excision (case 8), the patient with two local recurrences (case 2) was treated with a below-the-knee amputation following the second recurrence due to the © 2015 John Wiley & Sons Ltd, Histopathology, 68, 286–296.
Cutaneous epithelioid malignant peripheral nerve sheath tumour
children also display loss of INI1 expression in about 10 and 40%, respectively.29,32 The markers of epithelial, e.g. low and high molecular weight cytokeratins (CK-MNF116, CK-AE1/AE3), myogenic/myofibroblastic, e.g. desmin, myogenin and smooth muscle actin, and vascular differentiation, e.g. CD31 are consistently negative in E-MPNST. The summary of immunohistochemical features of E-MPNST in the four largest studies on E-MPNST is represented in Table 3. The main differential diagnosis of cutaneous E-MPNST includes melanoma, poorly differentiated carcinoma and other types of sarcomas with predominantly or exclusively epithelioid morphology, including epithelioid sarcoma and myoepithelial carcinoma. In addition, distinction from epithelioid schwannoma can be difficult. E-MPNST displays considerable immunohistochemical overlap with melanoma, especially considering its consistent S100 positivity,6,7 and discrimination between the two entities may be difficult. Unlike melanoma with epithelioid morphology, E-MPNST generally lacks expression of second-line markers of melanocytic differentiation, for example melan A, HMB45, tyrosinase and MITF.6,7 In particular, MITF positivity as an adjunct to S100 positivity, a combination of the two not demonstrated in E-MPNSTs, has been regarded as highly specific for melanoma.33 Furthermore, melanoma consistently retains expression of INI1 protein while loss of INI1 expression occurs commonly in E-MPNST.8,29 It is also very important to exclude any previous history of melanoma or the presence of concurrent melanoma on clinical grounds. The single tumour in our study with focal weak melan A positivity lacked expression of other markers of melanocytic differentiation, including HMB45, tyrosinase and MITF. In addition, the possibility of previous and concurrent melanoma was excluded by meticulous clinical examination and clinical history. Metastatic carcinoma can be separated easily from E-MPNST by cytokeratin positivity in the former and its consistent negativity in the latter. Epithelioid sarcoma typically shows loss of INI1 expression, positivity for both low and high molecular weight cytokeratins and is consistently negative for markers of melanocytic differentiation, including S100.29,34–36 Myoepithelial carcinoma shares similar morphological features with E-MPNST, but can be distinguished from the latter by consistently staining for cytokeratins.8,29 Epithelioid schwannomas are generally well-circumscribed and encapsulated, and lack significant atypia.8 Nevertheless, the presence of diffuse and/or significant nuclear atypia/pleomorphism, vesicular nuclei or atypical mitoses favour E-MPNST.8 © 2015 John Wiley & Sons Ltd, Histopathology, 68, 286–296.
295
In conclusion, we analysed the clinicopathological features of 11 cutaneous E-MPNSTs and correlated them with survival. Although the majority of the patients were adults in their fifth decade of life, infrequent occurrence in infancy and childhood was observed. On follow-up, 22% of the patients developed local recurrence and an additional 33% of the patients died due to disease dissemination. None of the clinicopathological parameters, including gender, age, size of the tumour, location (dermis, subcutis or both), percentage of epithelioid component, the number of mitoses per 10 high-power fields, and degree of cytological atypia had any predictive value for local recurrence(s), development of distant metastases and overall survival of patients. Immunohistochemically, all tumours were diffusely S100-positive, with a subset displaying loss of INI1 expression (50%). The results of our study confirm that cutaneous E-MPNST has the potential to pursue an aggressive clinical course, associated with wide dissemination and unfavourable disease outcome, necessitating long-term follow-up.
Acknowledgements The authors thank the following colleagues for their contributions in cases and/or clinical data and followup information: Dr S. Agarwal, Queen Elizabeth II Hospital, Welwyn Garden City, UK; Dr L. Melly, Royal Free Hospital, Pond Street, UK; Dr L. Wheen, Royal Devon and Exeter Hospital, Church Lane, UK; Dr D. E. Hughes, Royal Hallamshire Hospital, UK; Dr M. Khan, Queen Mary’s Hospital, Kent, UK; Dr Muller, Glenfield Hospital, UK; Dr C. D’Arrigo, The National Hospital for Neurology and Neurosurgery, UK; Dr C. Scott, Salisbury District Hospital, Wiltshire, UK; and Dr M. Fallowfield, Mid Essex Hospital, Chelmsford, UK.
Conflict of interest The authors declare that there are no conflicts of interest.
References 1. McCormack LJ, Hazard JB, Dickson JA. Malignant epithelioid neurilemmoma (schwannoma). Cancer 1954; 7; 725–728. 2. Stewart FW, Copeland MM. Neurogenic sarcoma. Am. J. Cancer 1931; 15; 1235–1320. 3. Alvira MM, Mandybur TI, Menefee MG. Light microscopic and ultrastructural observations of a metastasizing malignant epithelioid schwannoma. Cancer 1976; 38; 1977–1982.
296
B Luzar et al.
4. Weiss SW, Goldblum JR. Malignant tumors of the peripheral nerves. In Weiss SW, Goldblum JR eds. Enzinger & Weiss’s soft tissue tumors. 5th ed. Philadelphia, PA: Mosby Elsevier, 2008; 903–944. 5. Nielsen GP, Antonescu CR, Lothe RA. Malignant peripheral nerve sheath tumour. In Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F eds. WHO classification of tumours of soft tissue and bone. Lyon: IARC Press, 2013; 187–189. 6. Lodding P, Kindblom L-G, Angervall L. Epithelioid malignant schwannoma. A study of 14 cases. Virchows Arch A Pathol. Anat. Histopathol. 1986; 409; 433–451. 7. Laskin WB, Weiss SW, Bratthauer GL. Epithelioid variant of malignant peripheral nerve sheath tumor (malignant epithelioid schwannoma). Am. J. Surg. Pathol. 1991; 15; 1136– 1145. 8. Jo VY, Fletcher CDM. Epithelioid malignant peripheral nerve sheath tumor. Clinicopathological analysis of 63 cases. Am. J. Surg. Pathol. 2015; 39; 673–682. 9. Manganoni AM, Farisoglio C, Lonati A, Tucci G, Pinton PG. Cutaneous epithelioid malignant schwannoma: review of the literature and case report. J. Plast. Reconstr. Aesthet. Surg. 2009; 62; 318–321. 10. 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. 11. Molenaar WM, Ladde BE, Schraffordt Koops H, Dam-Meiring A. Two epithelioid malignant schwannomas in a patient with neurofibromatosis. Cytology, histology and DNA-flow-cytometry. Pathol. Res. Pract. 1989; 184; 529–534. 12. Morgan KG, Gray C. Malignant epithelioid schwannoma of superficial soft tissue? A case report with immunohistology and electron microscopy. Histopathology 1985; 9; 765–775. 13. Suster S, Amazon K, Rosen LB, Ollague JM. Malignant epithelioid schwannoma of the skin. A low-grade neurotropic malignant melanoma? Am. J. Dermatopathol. 1989; 11; 338–344. 14. Shimizu S, Teraki Y, Ishiko A et al. Malignant epithelioid schwannoma of the skin showing partial HMB-45 positivity. Am. J. Dermatopathol. 1993; 15; 378–384. 15. Yamamoto T, Minami R, Ohbayashi C. Subcutaneous malignant epithelioid schwannoma with cartilaginous differentiation. J. Cutan. Pathol. 2001; 28; 486–491. 16. Bleach NR, Keen CE, Dixon JA. Superficial malignant schwannoma of the face: a case for early radical surgery. J. Laryngol. Otol. 1989; 103; 316–318. 17. Morgan MB, Stevens L, Patterson J, Tannenbaum M. Cutaneous epithelioid malignant nerve sheath tumor with rhabdoid features: a histologic, immunohistochemical, and ultrastructural study of three cases. J. Cutan. Pathol. 2000; 27; 529–534. 18. Tsuchiya D, Takamura H, Saito K, Kashiwa H, Maeda K, Yamashita H. Immunohistochemical diagnosis of a rare case of epithelioid malignant peripheral nerve sheath tumor with multiple metastases. Jpn. J. Ophtalmol. 2004; 48; 565–569. 19. Sokolova IN, Gurariĭ LL. Malignant epithelioid schwannoma of the kidney. Arkh. Patol. 1988; 50; 59–64. 20. Eltoum IA, Moore RJ III, Cook W, Crowe DR, Rodgers WH, Siegal GP. Epithelioid variant of malignant peripheral nerve
21.
22.
23. 24.
25.
26.
27.
28.
29.
30.
31. 32.
33.
34.
35.
36.
sheath tumor (malignant schwannoma) of the urinary bladder. Ann. Diagn. Pathol. 1999; 3; 304–308. Ivaskevics K, S L aszlo A, api Z. Malignant epithelioid ovarian schwannoma: a case report. Int. J. Gynecol. Cancer 2006; 16; 360–362. Fukai I, Masaoka A, Yamakawa Y, Niwa H, Eimoto T. Mediastinal malignant epithelioid schwannoma. Chest 1995; 108; 574–575. Kitamura H, Kitamura H. Primary epithelioid malignant schwannoma of the lung. Pathol. Int. 1994; 44; 317–324. Dunn GP, Spiliopoulos K, Plotkin SR et al. Role of resection of malignant peripheral nerve sheath tumors in patients with neurofibromatosis type 1. J. Neurosurg. 2013; 118; 142–148. Thway K, Fisher C. Malignant peripheral nerve sheath tumor: pathology and genetics. Ann. Diagn. Pathol. 2014; 18; 109– 116. Hruban RH, Shiu MH, Senie RT, Woodruff JM. Malignant peripheral nerve sheath tumors of the buttock and lower extremity. A study of 43 cases. Cancer 1990; 66; 1253–1265. Thewes M, Jungfer-Weber B, Wiebecke B, Engst R, Ring J. Malignant epithelioid schwannoma with melanocytic differentiation: a rare tumor with unusual feature. Acta Derm. Venereol. 1997; 77; 493–494. Pizem J, Nicholson KM, Mraz J, Prieto VG. Melanocytic differentiation is present in a significant proportion of nonpigmented diffuse neurofibromas. A potential diagnostic pitfall. Am. J. Surg. Pathol. 2013; 37; 1182–1191. Hornick JL, Dal Cin P, Fletcher CD. Loss of INI1 expression is characteristic of both conventional and proximal-type epithelioid sarcoma. Am. J. Surg. Pathol. 2009; 33; 542–550. Carter JM, O’Hara C, Dundas G et al. Epithelioid malignant peripheral nerve sheath tumor arising in a schwannoma, in a patient with ‘neuroblastoma-like’ schwannomatosis and a novel germline SMARCB1 mutation. Am. J. Surg. Pathol. 2012; 36; 154–160. Roberts CW, Orkin SH. The SWI/SNF complex-chromatin and cancer. Nat. Rev. Cancer 2004; 4; 133–142. Gleason BC, Fletcher CD. Myoepithelial carcinoma of soft tissue in children: an aggressive neoplasm analyzed in a series of 29 cases. Am. J. Surg. Pathol. 2007; 31; 1813–1824. Miettinen M, Fernandez M, Franssila K, Gatalica Z, Lasota J, Sarlomo-Rikala M. Microphtalmia transcription factor in the immunohistochemical diagnosis of metastatic melanoma: comparison with four other melanoma markers. Am. J. Surg. Pathol. 2001; 25; 205–211. Lin L, Skacel M, Sigel JE et al. Epitheliod sarcoma: an immunohistochemical analysis evaluating the utility of cytokeratin 5/6 in distinguishing superficial epithelioid sarcoma from spindled squamous cell carcinoma. J. Cutan. Pathol. 2003; 30; 114–117. Miettinen M, Fanburg-Smith JC, Virolainen M, Shmookler BM, Fetsch JF. Epithelioid sarcoma: an immunohistochemical analysis of 112 classical and variant cases and a discussion of the differential diagnosis. Hum. Pathol. 1999; 30; 934–942. Guillou L, Wadden C, Coindre JM, Krausz T, Fletcher CD. ‘Proximal-type’ epithelioid sarcoma, a distinctive aggressive neoplasm showing rhabdoid features. Clinicopathologic, immunohistochemical, and ultrastructural study of a series. Am. J. Surg. Pathol. 1997; 21; 130–146.
© 2015 John Wiley & Sons Ltd, Histopathology, 68, 286–296.
