Supplementary Material for: A novel glycolysis-related signature for predicting the prognosis and immune infiltration of uveal melanoma

Introduction As the most common aggressive intraocular cancer in adults, uveal melanoma (UVM) threatens the survival and vision of many people. Glycolysis is a novel hallmark of cancer, but the role of glycolysis-related genes in UVM prognosis remains unknown. The purpose of the study was to establish a glycolysis-related gene signature (GRGS) to predict UVM prognosis. Methods Raw data were obtained from TCGA-UVM and GSE22138 datasets. The GRGS was established by univariate, LASSO and multivariate Cox regression analyses. Kaplan‒Meier survival and time-dependent receiver operating characteristic curves were used to evaluate the predictive ability of the GRGS. The relationships of the GRGS with infiltrating immune cell levels and mutations were analyzed with CIBERSORT and maftools. Results A novel GRGS (risk score = 0.690861*ISG20 +0.070991*MET -0.227520*SDC2 +0.690223*FBP1 +0.048008*CLN6-0.128520* SDC3) was developed for predicting UVM prognosis. The GRGS had robust predictive stability in UVM. Enrichment annotation suggested that the high-risk group had stronger adaptive immune responses and that the low-risk group had more innate immune cell infiltration. Moreover, BAP1 mutation was related to high risk, and SF3B1 mutation was related to low risk. Conclusions This study developed and validated a novel GRGS to predict UVM prognosis and immune infiltration. The signature revealed an association between glycolysis-related genes and the tumor microenvironment, providing new insights into the role of glycolysis in UVM.

引言 作为成人最常见的侵袭性眼内恶性肿瘤，葡萄膜黑色素瘤（uveal melanoma, UVM）严重威胁众多患者的生存与视力。糖酵解是癌症的新型特征，但糖酵解相关基因在UVM预后中的作用仍未明确。本研究旨在构建一种糖酵解相关基因特征（glycolysis-related gene signature, GRGS）以预测UVM患者的预后。
方法 原始数据取自TCGA-UVM与GSE22138数据集。本研究通过单因素、最小绝对收缩与选择算子（Least Absolute Shrinkage and Selection Operator, LASSO）以及多因素Cox回归分析构建GRGS。采用卡普兰-迈耶（Kaplan-Meier）生存曲线与时依受试者工作特征（time-dependent receiver operating characteristic）曲线评估GRGS的预测效能。借助CIBERSORT与maftools工具分析GRGS与免疫细胞浸润水平及基因突变的关联。
结果 本研究构建了一种新型GRGS，其风险评分公式为：风险评分=0.690861×ISG20 + 0.070991×MET - 0.227520×SDC2 + 0.690223×FBP1 + 0.048008×CLN6 - 0.128520×SDC3。该GRGS在UVM中展现出稳定可靠的预测性能。富集注释分析显示，高风险组患者呈现更强的适应性免疫应答，而低风险组则存在更多的先天免疫细胞浸润。此外，BAP1基因突变与高风险状态相关，SF3B1基因突变则与低风险状态相关。
结论 本研究构建并验证了一种新型GRGS，可用于预测UVM患者的预后与免疫浸润情况。该基因特征揭示了糖酵解相关基因与肿瘤微环境之间的关联，为阐明糖酵解在UVM中的作用提供了新的研究视角。