<scp>PRAME</scp> expression in cutaneous melanoma does not correlate with disease‐specific survival

Abstract

BACKGROUND: Immunohistochemistry-based protein biomarkers can provide useful prognostic information in cutaneous melanoma. The independent prognostic value of Ki-67 has been studied with variable results. PReferentially expressed Antigen in MElanoma (PRAME) immunohistochemistry is a useful new ancillary tool for distinguishing cutaneous nevi from melanoma; however, its prognostic value has not been well studied. We evaluated PRAME as a prognostic marker in cutaneous melanoma, compared to Ki-67. METHODS: We analyzed the immunohistochemical expression of PRAME and Ki-67 in 165 melanocytic lesions, including 92 primary melanomas, 19 metastatic melanomas, and 54 melanocytic nevi using tissue microarrays. PRAME immunostaining was scored based on the percentage of positive nuclei: 0 <1%, 1+ 1%-25%, 2+ 26%-50%, 3+ 51%-75%, and 4+ >75%. The percentage of Ki-67-positive tumor nuclei was used to calculate the proliferation index. RESULTS: PRAME and Ki-67 both showed significantly increased expression in melanomas compared to nevi (p < 0.0001 and p < 0.001, respectively). There was no significant difference in PRAME expression in primary versus metastatic melanomas. By contrast, the Ki-67 proliferation index was higher in metastatic melanoma than in primary melanoma (p = 0.013). Increased Ki-67 index correlated with ulceration (p < 0.001), increased Breslow depth (p = 0.001), and higher mitotic rate (p < 0.0001), whereas increased PRAME expression correlated with higher mitotic rate (p = 0.047) and Ki-67 index (p = 0.007). Increased Ki-67 index correlated with worse disease-specific survival in patients with primary melanoma (p < 0.001), but PRAME expression did not show prognostic significance in disease-specific survival (p = 0.63). In a multivariable analysis of patients with primary melanoma, tumor Breslow depth, ulceration, mitotic rate, and Ki-67 index were each independent predictors of disease-specific survival (p = 0.006, 0.02, 0.001, and 0.04, respectively); however, PRAME expression was not predictive of disease-specific survival (p = 0.64). CONCLUSION: Ki-67 is an independent prognostic marker; although increased PRAME expression correlates with the Ki-67 proliferation index and mitotic rate, PRAME is not an independent prognostic marker for cutaneous melanoma. PRAME and Ki-67 are useful ancillary tools for distinguishing benign from malignant melanocytic lesions.