Atypical intraepidermal melanocytic proliferations (AIMP) have random cytologic atypia and other histologic features that are concerning for malignancy and often require immunohistochemistry to differentiate from melanoma in situ. Immunostaining with S100, Melan-A, and microphthalmia-associated transcription factor (MITF) was performed for 49 morphologically well-characterized AIMP lesions. The percentage of cells in the basal layer of the epidermis that were identified as melanocytes by immunohistochemistry was compared with the percentage observed by morphology on hematoxylin and eosin staining, which is the gold standard stain for identifying cytologic atypia within an AIMP. Melan-A estimated the highest percentage of melanocytes and S100 the fewest in 47 of the 49 lesions examined. The estimated percentage of melanocytes was 23.3% (95% confidence interval: 18.6–28.1; P < 0.001) higher for Melan-A compared with hematoxylin and eosin staining. Melanocyte estimates were similar for hematoxylin and eosin and MITF (P = 0.15) although S100 estimated 21.8% (95% confidence interval: –27.2 to –16.4; P < 0.001) fewer melanocytes than hematoxylin and eosin. Melan-A staining produces higher estimates of epidermal melanocytes than S100 and MITF, which may increase the likelihood of diagnosing melanoma in situ. In contrast, melanoma in situ may be underdiagnosed with the use of S100, which results in lower estimates of melanocytes than the other 2 immunostains. Therefore, the best immunohistochemical marker for epidermal melanocytes is MITF.
Departments of *Pathology and Laboratory Medicine
†Dermatology University of Pennsylvania Perelman School of Medicine, Philadelphia, PA.
Reprints: Xiaowei Xu, MD, PhD, Department of Pathology and Laboratory Medicine, Hospital of University of Pennsylvania, Philadelphia, PA 19104 (e-mail: email@example.com).
With respect to design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the article, this study was funded in part by grants from the National Institutes of Health (CA-116103, CA-093372, and AR-054593 to X.X.).
The authors X.X., M.S., and G.E.N. had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
G.E.N. and M.S. contributed equally to this article.
The authors have no relevant conflicts of interest concerning the content or message of this article.