Background: Tumor cells of malignant melanoma, the "great imitator," may morphologically mimic almost any cell, including histiocytes. Immunohistochemical stains for histiocytes are often used to distinguish histiocytic lesions that resemble melanomas, but we have noted and others have reported that these markers may be immunoreactive in melanomas.
Methods: We evaluated 43 primary and metastatic melanomas with traditional markers for melanomas (S100, HMB45, and NKI-C3) and common markers used for histiocytes (alpha-1-antitrypsin or AAT, CD68/KP1, HAM56, Mac387, and Muramidase). The extent (<5%, 5 to 30%, 30 to 60%, 60 to 90%, >90%) and intensity (1+ to 4+) of staining were recorded semi-quantitatively.
Results: Melanoma immunoreactivity (>5% of tumor cells) was as follows: S100, 100%; HMB45, 91%; NKI, 91%; AAT, 95%; CD68, 86%; HAM56, 26%; Mac387, 7%; and Muramidase, 30%. Among the histiocytic markers, staining by AAT and CD68 was typically diffuse but weak. Staining by HAM56, Mac387, and Muramidase was usually focal. In contrast, the traditional melanoma markers showed diffuse and strong staining. Interpretation of the histiocytic markers was complicated by scattered atypical histiocytes and pigmented tumor cells.
Conclusion: Melanomas are commonly immunoreactive for histiocytic markers. AAT and CD68 immunostains are diffusely positive almost as frequently as traditional melanoma markers, although with weaker intensity. HAM56, Mac387, and Muramidase are less commonly positive and exhibit focal staining. Therefore, depending on the context, histiocytic markers may not be helpful in differentiating histiocytes and histiocytic tumors from melanomas.