Digital Quantification of Intratumoral CD8+ T-Cells Predicts Relapse and Unfavorable Outcome in Uveal Melanoma

SIMPLE SUMMARY: In this study, we used digital pathology and image analysis for the prediction of the prognosis in uveal melanoma (UM). We retrospectively evaluated a total of 404 histopathological slides of 101 patients. Immune biomarkers for CD4, CD8, CD68, and CD163 were performed and evaluated by digital image acquisition and quantitative analysis. Our results showed that a higher intratumoral CD8 positive cell density imposed a negative impact on RFS and OS. Our study also showed that older age and stage III were independent negative prognostic factors for both RFS and OS. Our results demonstrate that a specific distribution profile of CD8 may imply a higher risk of relapse and death in UMs, and further studies could elaborate specific subgroups that are amenable to various treatment regimens. ABSTRACT: Although it is a disease that occurs mainly in the Caucasian population, uveal melanoma (UM) is the most common primary intraocular tumor in adults. Here, we used digital pathology and image analysis for the diagnosis of UM and the prediction of the prognosis. Our retrospective study included a total of 404 histopathological slides from 101 patients. A digital image acquisition and quantitative analysis of tissue immune biomarkers (CD4, CD8, CD68, CD163) were performed. A negative impact of the intratumoral CD8 positive cell density higher than 13.3 cells/mm(2) was detected for both RFS (HR 2.08, 95% Cl 1.09 to 3.99, p = 0.027) and OS (HR 3.30, 95% CI 1.58 to 6.88, p = 0.001). Moreover, we confirmed that older age and stage III were independent negative prognostic factors for both RFS and OS. Our results suggest that a specific distribution profile of CD8 in UM might predict the risk of relapse and death, with potential implications for determining which subgroups of UMs are amenable to specific pharmacological treatment regimens.