Abstract
Uveal melanoma (UM) is the most common primary intraocular malignancy in adults. DNA repair genes play a vital role in cancer development. However, there has been very little research about DNA repair genes in UM. This study aimed to evaluate the importance of DNA repair genes and established a signature for predicting prognosis and immune features of UM. In this study, we mined TCGA database through bioinformatics analysis, and the intersect was taken between DNA repair genes and prognosis related genes and yielded 52 genes. We divided 80 UM patients into C1 and C2 subtypes. GSEA results indicated that abundant cancer-promoting functions and signaling pathways were activated in C2 subtype and the proportion of SNVs was higher in C2 than in C1 which suggested a worse prognosis. We built a six DNA repair genes model including ITPA, CETN2, CCNO, POLR2J, POLD1, and POLA1 by LASSO regression to predict prognosis of UM patients and utilized the median value of risk scores as the cutoff point to differentiate high risk and low risk group. The survival analyses and the receiver operating characteristic (ROC) curves in the validation group and entire data set confirmed the accuracy of this model. We also constructed a nomogram based on age and risk scores to evaluate the relationship between risk scores and clinical outcome. The calibration curve of the overall survival (OS) indicated that the performance of this model is steady and robust. Finally, the enrichment analysis showed that there were complex regulatory mechanisms in UM patients. The immune infiltration analysis indicated that the immune infiltration in C2 in the high risk group was different from that in the low risk group. Our findings indicated that the DNA repair genes may be related to UM prognosis and provide new insight into the underlying mechanisms.