PRAME predicting outcome in uveal melanoma

Abstract

Abstract Uveal melanoma (UM) is the most common primary intra‐ocular malignancy in adults and carries a high risk of metastatic disease, affecting approximately 50% of patients despite advances in local tumour control. PRAME (PReferentially expressed Antigen in MElanoma) has emerged as a marker for metastatic risk; however, evidence is largely derived from mRNA expression studies rather than routine immunohistochemistry (IHC). This study evaluates PRAME protein expression by IHC in a large UM cohort and examines its association with clinicopathological features and survival outcomes. A total of 403 UM cases were included. PRAME IHC was performed on tissue microarrays (TMAs) and/or whole tumour sections and scored as positive based on nuclear staining in >10% of tumour cells. Associations were assessed using chi‐square and Mann–Whitney U tests. Survival outcomes were analysed using Kaplan–Meier curves and Cox proportional hazards models. PRAME positivity was identified in 5% of TMA cases and 40% of whole‐section cases. On TMAs, PRAME expression was associated with worse overall survival (median 26 versus 50 months; p = 0.03) but not metastasis‐free survival (MFS). In contrast, whole‐section PRAME expression was strongly associated with shorter MFS (median 5 versus 20 months; p < 0.001) and worse overall survival (median 13 versus 52 months; p = 0.04). In multivariate analysis including age, American Joint Committee on Cancer stage and BAP1 status, whole‐section PRAME positivity remained an independent adverse predictor of MFS (hazard ratio = 10.5, p = 0.002). PRAME IHC is a clinically meaningful prognostic biomarker in UM, particularly when assessed on whole tumour sections, where it provides strong independent prognostic information.