Department of Ophthalmology, Eye, Ear, Nose, and Throat Hospital of Fudan University, Shanghai, China.
Laboratory of Myopia, NHC Key Laboratory of Myopia (Fudan University), Chinese Academy of Medical Sciences, Shanghai, China.
Math Biosci Eng. 2021 Sep 15;18(6):8045-8063. doi: 10.3934/mbe.2021399.
Uveal melanoma (UM) is the most aggressive intraocular tumor worldwide. Accurate prognostic models are urgently needed. The present research aimed to construct and validate a prognostic signature is associated with overall survival (OS) for UM patients based on metabolism-related genes (MRGs).
MRGs were obtained from molecular signature database (MSigDB). The gene expression profiles and patient clinical data were downloaded from The Cancer Genome Atlas (TCGA) database. In the training datasets, MRGs were analyzed through univariate Cox regression analyses and least absolute shrinkage and selection operator (LASSO) Cox analyses to build a prognostic model. The GSE84976 was treated as the validation cohort. In addition, time-dependent receiver operating characteristic (ROC) and Kaplan-Meier survival curve analyses the reliability of the developed model. Then, gene set enrichment analysis (GSEA) was used for gene enrichment analysis. Nomogram that combined the five-gene signature was used to evaluate the predictive OS value of UM patients.
Five MRGs were identified and used to establish the prognostic model for UM patients. The model was successfully validated using the testing cohort. Moreover, ROC analysis demonstrated a strong predictive ability that our prognostic signature had for UM prognosis. Multivariable Cox regression analysis revealed that the risk model was an independent predictor of prognosis. UM patients with a high-risk score showed a higher level of immune checkpoint molecules.
We established a novel metabolism-related signature that could predict survival and might be therapeutic targets for the treatment of UM patients.
葡萄膜黑色素瘤（UM）是全球最具侵袭性的眼内肿瘤。迫切需要准确的预后模型。本研究旨在构建和验证一个与 UM 患者总生存期（OS）相关的代谢相关基因（MRGs）预后特征。
MRGs 从分子特征数据库（MSigDB）中获得。基因表达谱和患者临床数据从癌症基因组图谱（TCGA）数据库下载。在训练数据集中，通过单变量 Cox 回归分析和最小绝对值收缩和选择算子（LASSO）Cox 分析对 MRGs 进行分析，以构建预后模型。GSE84976 被视为验证队列。此外，时间依赖性接收器操作特征（ROC）和 Kaplan-Meier 生存曲线分析了所开发模型的可靠性。然后，进行基因集富集分析（GSEA）进行基因富集分析。联合五个基因特征的列线图用于评估 UM 患者的预测 OS 值。
确定了五个 MRGs 并用于建立 UM 患者的预后模型。该模型在测试队列中得到了成功验证。此外，ROC 分析表明，我们的预后特征对 UM 预后具有很强的预测能力。多变量 Cox 回归分析显示，风险模型是预后的独立预测因子。风险评分较高的 UM 患者表现出更高水平的免疫检查点分子。
我们建立了一个新的代谢相关特征，可以预测生存，并且可能成为治疗 UM 患者的治疗靶点。
