To the Editor: It has been shown that the skin within and adjacent to actinic keratoses may demonstrate a disproportionate increase in melanoma antigen recognized by T-cells 1 melanoma Antigen Recognized by T-cells 1 (MART-1)-positive cells in the absence of other evidence of a melanocytic proliferation, creating a potential for overdiagnosis of melanoma in situ.1 Additionally, “pseudomelanocytic nests” have been described as MART-1 positive, S100-negative nonmelanocytic cellular aggregates at the dermoepidermal junction of sun-damaged skin, often in the setting of interface dermatitis. The aberrant MART-1 staining of such nests may likewise lead to a false impression of melanoma.2,3 Despite these difficulties, we continue to find MART-1 to be valuable in the diagnosis of melanocytic neoplasms, and caution that aberrant immunostaining may occur with any antibody. To illustrate this point, we present 2 cases of junctional proliferations on sun-damaged skin which were MART-1 positive, but S100 negative. Both were clinically and histologically suspicious for melanoma, and proved to be positive for microphthalmia-associated transcription factor microphthalmia-associated Transcription Factor (MITF) and vimentin. CASE 1 A 75-year-old man presented with a several month history of an enlarging, irregularly pigmented brown patch on the left cheek. The clinical diagnosis was lentigo maligna, and a shave biopsy was performed. Hematoxylin and eosin-stained sections demonstrated small, irregularly spaced junctional nests overlying solar elastosis (Fig. 1). MART-1 highlighted these nests, and an increased number of single-unit junctional melanocytes, with areas of follicular extension. S100 did not stain the junctional nests, but appropriately stained internal and external positive controls. Both the single-unit and nested components observed with MART-1 were also positive with MITF, human melanoma black-45 (HMB-45), and vimentin. The nests did not stain with pancytokeratin. Based on clinicopathologic correlation, a diagnosis of lentigo maligna was made.FIGURE 1: Hematoxylin and eosin-stained sections show scattered small junctional nests on sun-damaged skin (A). Increased single-unit and nested junctional melanocytes are largely negative with S100 (B), but positive with MART-1 (C) and MITF (D).CASE 2 A biopsy was taken from a dark brown macule on the upper back of a 72-year-old man. The preoperative differential diagnosis included melanoma. Histologic sections showed vacuolation and increased pigmentation at the dermoepidermal junction (Fig. 2). Immunohistochemistry demonstrated a nearly confluent proliferation of MART-1-positive junctional cells, with occasional small MART-1-positive nests. S100 showed good staining of internal and external positive controls, but little staining of the population highlighted by MART-1. HMB-45 was similarly weak in this population, despite strong staining of positive control tissue on the same slide. MITF and vimentin strongly stained the junctional proliferation, and a diagnosis of melanoma in situ was made.FIGURE 2: Hematoxylin and eosin-stained sections show vacuolation and small nests at the dermoepidermal junction (A). A regionally confluent junctional proliferation is negative with S100 (B), but positive with MART-1 (C) and MITF (D).DISCUSSION It has been demonstrated that in sun-damaged skin, MART-1/Melan-A may stain a greater number of nested and single-unit junctional cells than S100, and that this increase in MART-1 staining does not necessarily imply a junctional melanocytic neoplasm,1,3 making it necessary to exercise caution with these stains to avoid overdiagnosis of melanoma. The cases presented above demonstrate that a converse difficulty may also occur. In both cases, a MART-1-positive junctional proliferation showed little positivity with S100, prompting us to pursue additional immunohistochemical evaluation. The junctional proliferation stained with HMB-45 in case 1, and vimentin and MITF were positive in both cases. Although the latter stains are not specific for melanocytes, the combined immunohistochemical findings together with the hematoxylin-eosin morphology and the available clinical information were supportive of a diagnosis of a melanocytic lesion in both cases. Although S100 is considered a relatively sensitive stain for melanocytes, cases of S100-negative invasive melanoma have been documented, as has variability in staining of melanocytes with different clones of S100.4 The negative S100 staining observed in our cases may have resulted from a lack of expression of S100 protein, or a failure of expressed S100 to stain with the clone utilized in our laboratory (Ventana monoclonal 4C4.9). These cases illustrate the potential for diagnostic confusion based on negative staining with S100 in junctional melanocytic proliferations on sun-damaged skin. Even as caution is necessary to avoid overinterpretation of MART-1 positivity, it also seems unwise to rely solely upon S100 to distinguish MART-1-positive melanocytes from pseudomelanocytic proliferations. Morgan L Wilson, MD Eric W Hossler, MD Tammie C Ferringer, MD Dirk M Elston, MD Geisinger Medical Center, Departments of Dermatology and Pathology, Danville, PA
Search from 250 M+ research papersSearch from 250 M+ research papers
Aug 1, 2024
Huiyan You + 6
Open Access