Track

Clinical Studies

Abstract

Early-stage melanoma diagnosis followed by surgical resection is associated with favorable survival rates. However, diagnostic discordance exists among dermatopathologists for certain indeterminate melanocytic tumors. We hypothesize that a weighted interpretation of a panel of immunohistochemical (IHC) markers (p16, p21, Ki67, HMB-45, and PRAME) may improve classification of challenging melanocytic lesions compared with reliance on a single marker. A retrospective review identified benign nevi, nevi with atypia, and melanomas stained with p16, p21, Ki67, HMB-45, and/or PRAME. Fisher’s exact test assessed differences in marker expression. Multivariate proportional odds ordinal logistic regression evaluated the combined ability of markers to classify lesions. False discovery rate was controlled with the Benjamini-Hochberg method (adjusted p < 0.05). 92 lesions were analyzed: 28 benign nevi, 28 atypical nevi, and 36 melanomas. For p16, benign and atypical nevi more often showed higher staining (>25% of melanocytes positive); melanomas more often showed absent or low staining (<25% positive); differences were not statistically significant (p = 0.11, q = 0.34). p21 and PRAME showed variable distribution without significance after correction (all q ≥ 0.30). In multivariate analysis, no marker significantly classified lesions. Neither individual IHC markers nor their combined use achieved statistically significant classification of melanocytic lesions. Nonetheless, patterns observed suggest meaningful biological variability. These findings reinforce that no single stain should be relied upon in isolation. Interpretation should include a weighted, multi-marker approach integrated with histopathologic and clinical features. The trends observed support further investigation into optimized multi-marker scoring systems in larger, prospective cohorts.