摘　要：为了探究铜死亡相关基因在宫颈癌（CC）中的预后价值，从TCGA数据库下载CC患者的临床资料，随机分为训练组和验证组。通过单因素Cox、LASSO-Cox和多因素Cox分析筛选出铜死亡相关基因，构建风险模型。通过Kaplan-Meier曲线分析两个亚组和整个队列的总生存期（OS）、受试者工作特征曲线（ROC）和主成分分析（PCA）验证模型的预后价值。通过单因素和多因素分析来评价临床特征和风险评分的独立预后价值。利用基因本体（GO）和京都基因与基因组百科全书（KGEE）富集分析两个亚组间的生物学功能和途径，并进一步分析了两个亚组对药物的敏感性。最终构建了5个与铜死亡相关基因（FXD1、ARF1、APP、HSF1、MT1A）的预后模型。从风险评分的生存曲线来看，低风险组的OS远超过高风险组，且预后良好（P<0.05）。单因素和多因素Cox分析表明，风险评分是独立的预后因素（P<0.001）。根据ROC和PCA证明了预后模型的预测能力。利用ROC曲线分析评估风险评分和其他临床特征（如年龄、分级和分期）的敏感性和特异性，结果表明，风险评分的预后价值优于其他临床特征。富集分析结果表明，基因功能主要富集于细胞外基质、细胞外结构。根据肿瘤免疫功能障碍与排斥（TIDE）算法，低风险组患者的免疫治疗疗效优于高风险组。此外还发现24种药物的敏感性在两个亚组中的显著差异。本研究建立了5个铜死亡相关基因组成的预后风险模型，并证明了该模型可以准确预测患者的预后，且低风险评分的患者更易从免疫治疗中获益，为临床个体化治疗提供理论依据。
关键词：宫颈癌；铜死亡相关基因；风险模型；免疫治疗；预后
中图分类号：R737.3　　　　　文献标志码：A 　　　　　DOI：10.3969/j.issn.1007-7146.2023.03.009

Abstract: To explore the prognostic value of cuproptosis-related genes in cervical cancer, the clinical data of cervical cancer patients were downloaded from the TCGA database and randomly divided into training group and validation group. Cuproptosis-related genes were screened out by univariate Cox, LASSO-Cox and multivariate Cox analysis, and the risk model was constructed. Overall survival (OS) of the two subgroups and the entire cohort was analyzed by Kaplan-Meier curve, and the prognostic value of the model was verified by ROC curve and PCA. Univariate and multivariate analyses were performed to evaluate the independent prognostic value of clinical features and risk scores. The biological functions and pathways between the two subgroups were analyzed by gene ontology (GO) and Kyoto encyclopedia enrichment of genes and genomes (KGEE), and the sensitivities of the two subgroups to drugs were further analyzed. Finally, prognostic models of five cuproptosis-related genes (FDX1, ARF1, APP, HSF1, MT1A) were constructed. From the survival curve of risk score, OS in the low-risk group was much higher than that in the high-risk group, and the prognosis was good (P<0.05). By univariate and multivariate Cox analysis, risk score an independent prognostic factor (P<0.001). The predictive power of the prognostic model was demonstrated by receiver operating characteristic curve (ROC) and principal component analysis (PCA). ROC curve analysis was used to assess the sensitivity and specificity of risk scores and other clinical features, such as age, grade, and stage. And the results showed that the prognostic value of risk scores was superior to other clinical features. The results of enrichment analysis showed that gene function was mainly concentrated in extracellular matrix and extracellular structure. According to TIDE algorithm, immunotherapy efficacy of patients in the low-risk group was superior to that in the high-risk group. In addition, significant differences were found in the sensitivity of 24 drugs between the two subgroups. The study established a prognostic risk model composed of 5 cuproptosis-related genes, and proved that the model can accurately predict the prognosis of patients. And patients with low risk score are more likely to benefit from immunotherapy, which provides theoretical basis for clinical individualized therapy.
Key words: cervical cancer; cuprotosis related genes; risk model; immunotherapy; prognosis
（Acta Laser Biology Sinica, 2023, 32(3): 259-271）