Preferentially expressed Antigen in MElanoma immunohistochemistry as an adjunct for evaluating ambiguous melanocytic proliferation

The histopathological assessment of diagnostically ambiguous melanocytic proliferation remains one of the biggest challenges in the dermatopathology field. Preferentially expressed Antigen in MElanoma (PRAME) immunostaining has been shown highly specific for distinguishing unequivocal malignant melanocytic proliferation from benign ones. Knowledge on its utility for evaluating ambiguous melanocytic neoplasms remains limited. We retrieved in our institutional database all cases of diagnostically ambiguous melanocytic neoplasms from January 2016 to January 2021. Each case was subclassified into “favor benign” or “favor malignant” neoplasm using all collected data. Immunohistochemical expression of PRAME was assessed and correlated with the final subclassification. Using a previously proposed scoring system, diffuse immunopositivity (>75% of tumor cells) was considered positive. Furthermore, for ambiguous melanocytic proliferation occurring on a pre-existing nevus, the staining was considered positive if more than 75% of the morphologically atypical neoplastic cells were labeled, excluding morphologically unambiguous benign nevocytes. Fifty-five cases of ambiguous melanocytic proliferation were examined. Thirty-one cases were finally subclassified as “favor malignant” neoplasms and 24 as “favor benign” neoplasms. Thirty-one tumors showed immunopositivity for PRAME, representing, respectively, 8.3% and 93.5% of “favor benign” and “favor malignant” neoplasms. The specificity and sensitivity of PRAME immunohistochemistry for benign/malignant distinction were, respectively, 91.7% and 93.5%.PRAME IHC shows high sensitivity and specificity for distinguishing malignant challenging melanocytic proliferations from benign ones and could be used as an everyday tool. However, PRAME immunoreactivity should be interpreted cautiously, knowing that rare benign melanocytic neoplasms could show diffuse positivity.