Evaluation of hair follicle involvement in malignant melanoma using nestin and cytokeratin 15 immunohistochemistry

AIM: Hair follicle involvement (F) by melanoma cells, a finding that is not included in current guidelines for the classification and grading systems in malignant melanoma (MM), has become an object of interest recently. In this study, our aim is to evaluate the histopathological characteristics of F in MM and to investigate the possible role of stem cells in melanoma pathogenesis and its spread in tissues to see whether there is an association with F patterns. MATERIALS AND METHODS: One hundred five cases of MM, diagnosed with excisional skin biopsies during period of January 2006-March 2018 were evaluated. Cases were evaluated for histopathological and prognostic parameters, for the presence of F, and if present, for the characteristics of it. The tissues of aforementioned cases were then evaluated with nestin and cytokeratin 15 (CK15) immunohistochemistry. Thirty cases of melanocytic nevi with F and 10 nonneoplastic skin specimens that contain hair follicles were also added for comparison. RESULTS: Among 105 cases of MM, 57 cases (%54,3) were F(+), 48 cases (%45,7) were F(-). Comparisons of F(+) and F(-) cases revealed no significant differences for sex, age, tumor localization and diameter, histologic subtype, presence of ulceration, Breslow thickness, Clark invasion level, mitotic count, status of tumor infiltrating lymphocytes and pT stage of the tumor between the groups. Most of the F(+) cases (%66,7) were of head and neck localization. Evaluation of histopathological characteristics of F (the number of follicles that were involved, type of involvement, involvement patterns and destruction of hair follicle by F) revealed no significant associations when compared with each other and other histopathological parameters. CK15 staining patterns showed similarities in cases of MM, nevi and in specimens of nonneoplastic skin; cytoplasmic positivity of basal cells of epidermis and hair follicle epithelium was observed. In tumor cells, cytoplasmic nestin positivity was observed, but its expression patterns showed no significant differences between cases of F(+) and F(-) MM or cases of MM and nevi. In addition, in cases of MM, Nestin expression levels showed no correlation with histopathological or prognostic parameters of the lesion. CONCLUSIONS: According to current guidelines, Clark level and Breslow depth of the tumor is necessary to be reported for the thickness measurement in MM, but there are no recommendations regarding follicular involvement to be included, thus resulting this feature to be neglected at standard reporting. In our study, we observed that involvement of hair follicle by neoplastic cells is a finding seen not infrequently in MM cases. Although our evaluation revealed no associations that are statistically significant between F status and histopathological and prognostic parameters, reporting of this particular feature that can provide insights for extending our horizon in terms of understanding this disease's nature and behavioral properties, and for predicting the prognosis and metastatic potential is recommended. Evaluation with stem cell markers nestin and CK15 failed to provide statistically significant data to prove whether there is an association between melanocytic-keratinocytic stem cells and MM pathogenesis, the tumor's spread in tissues or F status. But it is important to note that cancer stem cells hypothesis is needed to be further studied with different markers for determining their impact on tumor behavior. In order to F status evaluation and reporting of it to be standardized, a need to revise the current guidelines to include this finding and further exploration of the subject to collect more data to help in determining its significance by larger scale studies seems to be a necessity. KEYWORDS: Malignant melanoma, hair follicle involvement, nestin, cytokeratin 15