Identification of co-expressed gene networks promoting CD8 T cell infiltration and having prognostic value in uveal melanoma.

Current immunotherapies are unsatisfactory against uveal melanoma (UM); however, elevated CD8 T cell infiltration level indicates poor prognosis in UM. Here, we aimed to identify co-expressed gene networks promoting CD8 T cell infiltration in UM and created a prognostic hazard model based on the identified hub genes. Raw data and clinical information were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Stromal-immune comprehensive score (ESTIMATE) was used to evaluate the immune-infiltration landscape of the tumor microenvironment. Single-Sample Gene Set Enrichment Analysis (ssGSEA) and Weighted Correlation Network Analysis (WGCNA) were used to quantify CD8 T cell infiltration level and identify hub genes. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed to analyze the biological processes. Least absolute shrinkage and selection operator (LASSO) Cox regression were used to establish a prognostic model, which was further validated. Finally, pan-cancer analysis evaluated these genes to be associated with CD8 T cell infiltration in other tumors. In conclusion, the proposed four-gene (PTPN12, IDH2, P2RX4, and KDELR2) prognostic hazard model had satisfactory prognostic ability. These hub genes may promote CD8 T cell infiltration in UM through antigen presentation, and CD8 T cell possibly function as Treg, resulting in poor prognosis. These findings might facilitate the development of novel immunotherapies.