Differentiating neurotized melanocytic nevi from neurofibromas using Melan-A (MART-1) immunohistochemical stain.
The Melan-A (Mart-1) gene was cloned from melanoma cell lines in 1994 by 2 research groups independently. (9,10) Chen et al (11) generated a monoclonal antibody against Melan-A in 1996, which is a relatively sensitive and specific marker for benign and malignant melanocytes, except those of desmoplastic melanomas. (12-13) However, mixed success was reported in staining a few neurotized melanocytic nevi with Melan-A. (12,14,15) Several of the cited articles stated that neurotized melanocytic nevus was negative for Melan-A. Given the melanocytic origin of nevus cells, we propose that Melan-A (MART-1) may be useful for identifying melanocytes even in the background of extensive neurotization.
Neurofilaments are a family of the 10-nm, intermediate filaments found specifically in the axons of neurons. Neurofilaments are heterodimers of 3 different subunits, neurofilament-light (70 kDa), neurofilament-medium (150-160 kDa), and neurofilament-heavy (200 kDa) chains, which are encoded by 3 different genes located at band 8p21 (NEFL and NEFM) and band 22q12.2 (NEFH). (16,17) Antibodies to neurofilament protein, which presumably stain the intratumoral axons in neurofibroma, stain positive in benign nerve sheath tumors but are not useful in differentiating neurofibromas from Schwannomas. (18) Because neurofibromas contain Schwann cells and frequent nerve twigs, we hypothesize that antibodies to neurofilament protein may be supplemental to Melan-A in differentiating neurofibromas from neurotized melanocytic nevi.
MATERIALS AND METHODS
Forty-nine consecutive specimens of partially neurotized melanocytic nevi (43 patients) and 49 consecutive specimens of neurofibromas (43 patients) were selected from the surgical pathology archives at Washington University Medical Center (St Louis, Missouri) from 2006 through 2008. The cases with little material left in the blocks were excluded.
The archival slides stained with hematoxylin-eosin were reviewed by 3 pathologists independently. Additionally, 5-[micro]m sections were taken from the paraffin blocks and stained using antibodies against Melan-A (prediluted, monoclonal, catalog number 790-2990; Ventana Medical Systems, Tucson, Arizona), S100 protein (prediluted, polyclonal, catalog number 760-2523; Ventana Medical), and neurofilament (prediluted, clone 2F11, monoclonal, catalog number 760-2661; Ventana Medical) following the manufacturer's protocols in a BenchMark XT automated slide stainer (Ventana Medical). In brief, heat-induced epitope retrieval with ethylenediaminetetraacetic acid-Tris base buffer (pH 8.0; catalog number 950-124; Ventana Medical) were used for Melan-A immunohistochemical stain; primary antibody incubation time was 16 minutes; an Ultra View Universal alkaline phosphatase red detection kit (catalog number 760-501; Ventana Medical) was used for Melan-A and an Ultra View Universal DAB detection kit (catalog number 760-500; Ventana Medical) was used for S100 and neurofilament.
Patient Demographics and Histopathology
Of the 49 specimens of neurotized melanocytic nevi (Figure 1, A and B), 28 (57%) were from the trunk, and 21 (43%) were from the head and neck region. Nine (18%) were compound melanocytic nevi, and 36 (73%) were intradermal melanocytic nevi. The patient population was composed of 14 men and 29 women, and the median age was 40 years (range, 14-84 years). Neurotized areas of the melanocytic nevi ranged from less than 25% to greater than 75% of the entire lesion, with most lesions (34 of 45; 76%) composed of less than 50% neurotized areas. All nevi showed congenital features, except the 4 specimens mentioned below. Rare individual or patchy, pigmented melanocytes and melanophages were seen in 26 specimens (53%), most of which were superficially located. Pigment was associated with neurotized areas in only 4 specimens (8%), all of which was very focal. Fatty infiltrates were seen in 10 cases (20%) and all were seen in association with the neurotized areas.
Four biopsies, diagnosed previously as melanocytic nevi with near-complete neurotization, were included among the melanocytic nevi. Upon reevaluation of the hematoxylin-eosin slides, a diagnosis of neurofibroma was suspected before immunohistochemical staining.
Of the 49 specimens of neurofibromas (Figure 2, A and B), 33 (67%) were from the trunk, 8 (16%) from the head and neck region, and 8 (16%) from the upper (n = 7; 14%) and the lower extremities (n = 1; 2%). The patient population was composed of 17 men and 26 women, and the median age was 56 years (range, 32-84 years). All neurofibromas were well-circumscribed, nonencapsulated, dermal or subcutaneous neoplasms. There was a diffuse growth of loosely arranged, wavy spindle cells in a pale-staining stroma. Increased small-caliber ectatic vessels and mast cells were present in the stroma. Fatty infiltrates were seen in 9 cases (18%). No plexiform features were noted.
Forty-five of the melanocytic nevi (92%) showed MelanA staining within the neurotized areas. In most cases (42 of 45; 93%), the neurotized areas showed strong positive signal intensity (Figure 3, A through C). Only 3 (7%) of the cases showed a weak signal and stained in less than 25% (n = 2; 4%) or between 25% and 50% (n = 1; 2%) of the neurotized areas. Fifteen cases (33%) showed positive staining of more than 75% of the neurotized areas; 12 cases (27%) showed staining of 50% to 75% of the neurotized areas, 9 cases (20%) had 25% to 50% staining, and 9 cases (20%) showed less than 25% staining. None of the neurofibromas (0 of 49; 0%) stained for Melan-A. Normal numbers of Melan-A+ melanocytes were observed in the overlying epidermis in all cases of neurofibromas (Figure 3, D).
All of the melanocytic nevi, including the neurotized areas, stained strongly and diffusely for S100 with an intense, coarse, and globoid appearance (Figure 4, A and B). The cytoplasm stained more diffusely than the nuclei. The neurofibromas showed a distinctive, sharp, wavy pattern of staining (Figure 5, A and B), with mostly cytoplasmic and some nuclear pattern. The 4 cases (8%) previously diagnosed as melanocytic nevi but suspected to be neurofibromas after review of routine stains (Figure 6, A and B), showed complete absence of Melan-A staining and the S100 pattern of staining was similar to the other neurofibromas (Figure 6, C and D).
Neurofilament protein antibody stained almost all cases of neurofibromas (48 of 49; 98%), with occasional cytoplasmic positivity in both nerve twiglike areas and random, single cells; about half of the cases of neurotized melanocytic nevi (25 of 49; 51%) showed occasional cytoplasmic positive staining within random tumor cells. The entrapped, normal nerve fibers within both tumors all stained positively.
Melanocytic nevi and neurofibromas both originate from neural crest-derived stem cells. These stem cells are capable of differentiating into components of the peripheral nervous system, including axons and Schwann cells, somatic, and autonomic nervous systems. They can also differentiate into various other cell types including, neuroendocrine and amine precursor uptake and decarboxylation cells, (19,20) melanocytes, mesenchymal cells, and chondrocytes. (21) Neurotization of melanocytic nevi refers to the presence of elongated or slender melanocytes (exaggeration of the type C melanocytes) with fibrillary cytoplasm, often in the mid to deep portions of some nevi. (22,23) The melanocytes within neurotized areas are arranged in short bundles or structures resembling Meissner corpuscles or Verocay bodies. (15) Some authors prefer to use the term melanocytic nevi with peripheral nerve sheath differentiation. (24) Given the histologic and immunologic similarities to neuroid tissue, type C melanocytes were initially thought to result from metaplasia or maturation of melanocytes into Schwannian cells with acquisition of a peripheral nerve phenotype. However, most current evidence supports that type C cells are melanocytes rather than Schwann cells. (2,25,26)
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Melanocytic nevi are often clinically distinct from neurofibromas because of the presence of pigment. However, pigment has been reported in neurofibromas, (27) and senescent nevi are often nonpigmented. Thus, melanocytic nevi are commonly included in the clinical differential diagnosis of neurofibromas. The distinction under the microscope may also pose a diagnostic challenge. The most common scenario is when the specimens are small and the melanocytic nevi exhibit large areas of neurotization. In addition, pigmented neurofibromas have been reported, (28) and small melanocytic nevi have been shown to be associated with neurofibromas within the same specimens in neurofibromatosis, type 1. (29) In our study, the following histologic features were useful in distinguishing between these 2 entities: (1) neurotized melanocytic nevi had congenital and nested growth patterns, type A and type B melanocytes, in the nonneurotized areas and pigment; and (2) neurofibromas had diffuse growth patterns with the characteristic feature of more ectatic, small-caliber vessels. A fatty infiltrate was noted in similar numbers of neurotized melanocytic nevi (10 of 45; 22%) and neurofibromas (9 of 53; 17%). Thus, we consider fatty infiltrate to not be a useful feature in distinguishing these 2 entities.
Several previous studies have mainly used neurilemmal markers to differentiate these 2 entities. (4-8) Type C melanocytes showed identical staining pattern when compared with type A and type B melanocytes using S100 protein, Leu-7, glial fibrillary acid protein, and myelin-basic protein. (4) Melan-A, a marker of melanocytic differentiation, has been established in routine clinical use to label melanocytes specifically but has not yet, to our knowledge, been evaluated in the literature as useful in differentiating neurotized melanocytic nevi from neurofibromas. There is a misconception that Melan-A stains negative in neurotized melanocytic nevi, but it has been evaluated in only a handful of cases and showed mixed results. (12,14,15) Our study demonstrated positive Melan-A staining in all melanocytic nevi with partial neurotization. About 60% of the cases showed more than 50% of the neurotized areas staining positive for Melan-A. Scattered, positively stained melanocytes were usually seen throughout the neurotized area. None of the neurofibromas were positive for Melan-A. We used a red chromogen for immunohistochemistry detection, thus, eliminating the possibility of false-positive staining of pigmented melanophages or dendritic cells. In our opinion, any Melan-[A.sup.+] dermal cells support a diagnosis of a neurotized melanocytic nevus rather than a neurofibroma.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Although most of the neurotized areas in melanocytic nevi stained for Melan-A, there were areas devoid of staining. In congenital melanocytic nevi, the area without detectable Melan-A staining can be large (D. Lu MD, PhD, unpublished data, February 2011). If a small biopsy was taken from those areas only, the negative Melan-A staining could be misleading. This may be the reason that previous reports (12,14) stated that neurotized melanocytic nevi were negative for Melan-A. Positive S100 protein staining has been demonstrated in both neurotized melanocytic nevi and neurofibromas. (2,4,30) However, the different staining pattern has not been specified between these 2 entities. Our study showed an intense, globoid staining pattern of S100 protein among the neurotized areas as in the previous report. (2) There is a distinctive fine, sharp, and wavy staining pattern in all the neurofibromas. Thus, when a differential diagnosis of melanocytic nevus with complete neurotization and neurofibroma is proposed in a small biopsy, characterizing the S100 staining pattern may aid in the diagnosis if the area lacks Melan-A staining.
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
Neural twigs are frequently present in neurofibromas, but whether these axonlike structures are present in neurotized melanocytic nevi is not well known. Our neurofilament antibody detected scattered positive cells in neurotized melanocytic nevi as well as in neurofibromas without morphologically identifiable neural twigs. This finding is similar to that of previous studies (6,30,31) and supports neurofilament protein not being a useful antibody to aid in the differential diagnosis between neurotized melanocytic nevi and neurofibromas.
In our study, Melan-A was expressed in virtually all melanocytic nevi, with generally strong staining retained in areas of neurotization, whereas all cases of neurofibromas were negative for Melan-A staining. Additionally, neurotized melanocytic nevi and neurofibromas showed a distinctive S100 staining pattern. Both neurotized melanocytic nevi and neurofibromas showed similar occasional staining for neurofilament. Thus, Melan-A immunohistochemical staining can be helpful in differentiating neurotized melanocytic nevi from neurofibromas in cases in which it is difficult to distinguish between them based on histomorphology alone. S100 protein staining pattern can also be used to aid in this distinction. In contrast, neurofilament immunohistochemical stain is not helpful in this matter.
We thank Jianping Li, BS, from the research laboratory at the Division of Anatomic Pathology, Department of Pathology and Immunology, Washington University School of Medicine (St. Louis, Missouri), for her technical contributions. This study was supported by the Department of Pathology and Immunology, Washington University School of Medicine.
(1.) Maize JC, Foster G. Age-related changes in melanocytic naevi. Clin Exp Dermatol. 1979; 4(1):49-58.
(2.) Van Paesschen MA, Goovaerts G, Buyssens N. A study of the so-called neurotization of nevi. Am J Dermatopathol. 1990; 12(3):242-248.
(3.) Lund HZ, Stobbe GD. The natural history of the pigmented nevus; factors of age and anatomic location. Am J Pathol. 1949; 25(6):1117-1155, incl 4 pl.
(4.) Gray MH, Smoller BR, McNutt NS, Hsu A. Neurofibromas and neurotized melanocytic nevi are immunohistochemically distinct neoplasms. Am J Dermatopathol. 1990; 12(3):234-241.
(5.) Gray MH, Smoller BR, McNutt NS, Hsu A. Immunohistochemical demonstration of factor XIIIa expression in neurofibromas: a practical means of differentiating these tumors from neurotized melanocytic nevi and schwannomas. Arch Dermatol. 1990; 126(4):472-476.
(6.) Prieto VG, McNutt NS, Lugo J, Reed JA. Differential expression of the intermediate filament peripherin in cutaneous neural lesions and neurotized melanocytic nevi. Am J Surg Pathol. 1997; 21(12):1450-1454.
(7.) Hwang SM, Choi EH, Lee WS, Choi SI, Ahn SK. Nevus spilus (speckled lentiginous nevus) associated with a nodular neurotized nevus. Am J Dermatopathol. 1997; 19(3):308-311.
(8.) Penneys NS, Mogollon R, Kowalczyk A, Nadji M, Adachi K. A survey of cutaneous neural lesions for the presence of myelin basic protein: an immunohistochemical study. Arch Dermatol. 1984; 120(2):210-213.
(9.) Coulie PG, Brichard V, Van Pel A, et al. A new gene coding for a differentiation antigen recognized by autologous cytolytic T lymphocytes on HLA-A2 melanomas. J Exp Med. 1994; 180(1):35-42.
(10.) Kawakami Y, Eliyahu S, Delgado CH, et al. Cloning of the gene coding for a shared human melanoma antigen recognized by autologous T cells infiltrating into tumor. Proc Natl Acad Sci USA. 1994; 91(9):3515-3519.
(11.) Chen YT, Stockert E, Jungbluth A, et al. Serological analysis of Melan-A(MART-1), a melanocyte-specific protein homogeneously expressed in human melanomas. Proc Natl Acad Sci USA. 1996; 93(12):5915-5919.
(12.) Orosz Z. Melan-A/Mart-1 expression in various melanocytic lesions and in non-melanocytic soft tissue tumours. Histopathology. 1999; 34(6):517-525.
(13.) Kucher C, Zhang PJ, Pasha T, et al. Expression of Melan-A and Ki-67 in desmoplastic melanoma and desmoplastic nevi. Am J Dermatopathol. 2004; 26(6):452-457.
(14.) Busam KJ, Jungbluth AA. Melan-A, a new melanocytic differentiation marker. Adv Anat Pathol. 1999; 6(1):12-18.
(15.) Kroumpouzos G, Cohen LM. Intradermal melanocytic nevus with prominent schwannian differentiation. Am J Dermatopathol. 2002; 24(1):39-42.
(16.) Hurst J, Flavell D, Julien JP, Meijer D, Mushynski W, Grosveld F. The human neurofilament gene (NEFL) is located on the short arm of chromosome 8. Cytogenet Cell Genet. 1987; 45(1):30-32.
(17.) Mattei MG, Dautigny A, Pham-Dinh D, Passage E, Mattei JF, Jolles P. The gene encoding the large human neurofilament subunit (NF-H) maps to the q121-q131 region on human chromosome 22. Hum Genet. 1988; 80(3):293-295.
(18.) Nascimento AF, Fletcher CD. The controversial nosology of benign nerve sheath tumors: neurofilament protein staining demonstrates intratumoral axons in many sporadic schwannomas. Am J Surg Pathol. 2007; 31(9):1363-1370.
(19.) Pearse AG, Polak JM, Rost FW, Fontaine J, Le Lievre C, Le Douarin N. Demonstration of the neural crest origin of type I (APUD) cells in the avian carotid body, using a cytochemical marker system. Histochemie. 1973; 34(3): 191-203.
(20.) Takor Takor T, Pearse AG. Cytochemical identification of human and murine pituitary corticotrophs and somatotrophs as APUD cells. Histochemie. 1973; 37(3):207-214.
(21.) Wong CE, Paratore C, Dours-Zimmermann MT, et al. Neural crest-derived cells with stem cell features can be traced back to multiple lineages in the adult skin. JCell Biol. 2006; 175(6):1005-1015.
(22.) Mooi WJ, Krausz T. Pathology of Melanocytic Disorders, 2nd ed. London, England: Hodder Arnold; 2007.
(23.) Miescher G, von Albertini A. Histologie de 100 cas de naevi pigmentaire d'apres les methodes de Masson. Bull Soc Fr Dermatol Syphiligr. 1935; 42:1265 1273.
(24.) Misago N. The relationship between melanocytes and peripheral nerve sheath cells (part I): melanocytic nevus (excluding so-called "blue nevus") with peripheral nerve sheath differentiation. Am J Dermatopathol. 2000; 22(3):217-229.
(25.) Thorne EG, Mottaz JH, Zelickson AS. Tyrosinase activity in dermal nevus cells. Arch Dermatol. 1971; 104(6):619-624.
(26.) Niizuma K. Electron microscopic study of nevic corpuscle. Acta Derm Venereol. 1975; 55(4):283-289.
(27.) Schaffer JV, Chang MW, Kovich OI, Kamino H, Orlow SJ. Pigmented plexiform neurofibroma: distinction from a large congenital melanocytic nevus. J Am Acad Dermatol. 2007; 56(5):862-868.
(28.) Fetsch JF, Michal M, Miettinen M. Pigmented (melanotic) neurofibroma: a clinicopathologic and immunohistochemical analysis of 19 lesions from 17 patients. Am J Surg Pathol. 2000; 24(3):331-343.
(29.) Ball NJ, Kho GT. Melanocytic nevi are associated with neurofibromas in neurofibromatosis, type I, but not sporadic neurofibromas: a study of226 cases. J Cutan Pathol. 2005; 32(8):523-532.
(30.) Kawahara E, Oda Y, Ooi A, Katsuda S, Nakanishi I, Umeda S. Expression of glial fibrillary acidic protein (GFAP) in peripheral nerve sheath tumors: a comparative study of immunoreactivity of GFAP, vimentin, S-100 protein, and neurofilament in 38 schwannomas and 18 neurofibromas. Am J Surg Pathol. 1988; 12(2):115-120.
(31.) Aso M, Hashimoto K, Hamzavi A. Immunohistochemical studies of selected skin diseases and tumors using monoclonal antibodies to neurofilament and myelin proteins. J Am Acad Dermatol. 1985; 13(1):37-42.
Yumei Chen, MD, PhD; Paul W. Klonowski, MD; Anne C. Lind, MD; Dongsi Lu, MD, PhD
Accepted for publication November 4, 2011.
From the Department of Pathology and Immunology, Washington University School of Medicine, St Louis, Missouri. Dr Chen is now with the Department of Pathology, Community Hospital Anderson, Anderson, Indiana. Dr Klonowski is now with the Department of Pathology, University of Calgary and Calgary Laboratory Services, Calgary, Alberta, Canada.
The authors have no relevant financial interest in the products or companies described in this article.
Presented in part at the 45th Annual Meeting of the American Society of Dermatopathology, San Francisco, California, October 16-19, 2008.
Reprints: Dongsi Lu, MD, PhD, Department of Pathology, Washington University School of Medicine, Campus Box 8118, 660 S Euclid Ave, St Louis, MO 63110 (e-mail: firstname.lastname@example.org).