基于EMT相关基因构建宫颈癌患者的预后模型  

Construction of a Prognostic Model for Cervical Cancer Patients Based on EMT-related Genes

作　　者：王妍 赵邹宇 杨萍[1] WANG Yan;ZHAO Zou-yu;YANG Ping(Department of Gynecology,the First Affiliated Hospital of Shihezi University,Xinjiang Shihezi,832008)

机构地区：[1]石河子大学第一附属医院妇科,新疆石河子832008

出　　处：《农垦医学》2024年第6期481-487,共7页Journal of Nongken Medicine

基　　金：国家自然科学基金面上项目(82072893);新疆生产建设兵团重点领域科技攻关计划项目(2023AB055);天山英才科技创新领军人才—高层次领军人才项目(CZ001201)。

摘　　要：目的:利用生物信息学方法构建基于EMT(Epithelial-Mesenchymal Transition,EMT)相关基因的宫颈癌预后模型,并分析其与免疫浸润和药物敏感性的相关性。方法:通过TCGA数据库获取304名宫颈癌患者的表达谱数据及临床信息,利用GSEA网站下载Hallmark-EMT基因集的200个基因进行最小绝对收缩和选择算法(Least Absolute Shrinkage and Selection Operator,LASSO)和COX回归分析,构建宫颈癌患者的预后模型。根据风险评分的中位数分为高、低风险组,生存曲线分析2组患者的预后情况,受试者工作特征(Receiver Operating Characteristic,ROC)评估预测效能。风险评分结合临床特征构建列线图,一致性指数(Concordance Index,C-index)和ROC曲线评价准确性。CIBERSORT算法、免疫检查点分析和ESTIMATE分析不同风险组间免疫微环境的差异;通过IC50值区分两组在化疗药物敏感性方面的差异。结果:共筛选出6个具有独立预后价值的EMT相关基因构建宫颈癌患者的预后模型。K-M生存曲线表明高风险组宫颈癌患者预后生存期显著缩短;ROC曲线下面积均>0.7,说明该模型的预测性能较强;筛选出分期与风险评分构建列线图,ROC曲线均提示该列线图的预测能力优于其他临床病理特征(P<0.05);多数免疫细胞和免疫检测点在低风险组中比例升高,且低风险组免疫评分显著高于高风险组(P<0.05);大部分化疗药物在低风险组中IC50值较低。结论:本研究构建EMT相关基因的预后模型可独立预测宫颈癌患者的生存期,有望成为宫颈癌治疗的新靶标。Objective:To develop a prognostic model for cervical cancer utilizing bioinformatics techniques centered on EMT-associated genes,and to investigate its association with immune infiltration and drug responsiveness.Methods:The expression profile data and clinical information of 304 cervical cancer patients were obtained from the TCGA database.Furthermore,a set of 200 genes associated with Hallmark-EMT was downloaded from the GSEA website.In order to establish prognostic models,Least Absolute Shrinkage and Selection Operator and COX regression analyses were performed.Patients were stratified by median risk score into high and low risk.Afterwards,Kaplan-Meier survival analysis was conducted to examine the prognostic differences between these two groups.Additionally,the predictive ability of the model was assessed using ROC analysis.Risk scores were combined with clinical characteristics to construct column-line plots,and the accuracy was assessed using the concordance index and ROC curves.The immune microenvironment differences between two risk groups were analyzed utilizing the CIBERSORT algorithm,immune checkpoint analysis,and ESTIMATE.Finally,IC50 values were employed to discern disparities in sensitivity to chemotherapeutic drugs between the two groups.Results:A total of six EMT-related genes with independent prognostic value were identified and utilized to construct the prognostic model.The Kaplan-Meier survival curves demonstrated a significantly shorter survival in cervical cancer patients belonging to the high-risk group,while all areas under the ROC curves were greater than 0.7,indicating a robust predictive performance of the model.Furthermore,a column line graph was constructed using staging and risk scores,which exhibited superior predictive ability compared to other clinicopathological features(P<0.05),as suggested by the ROC curves.Most immune cells and immune detection sites showed proportional elevation in the low-risk group.Additionally,immune scores were significantly higher in this group compared t

关 键 词：宫颈癌 上皮间充质转化 预后模型 

分 类 号：R737.33[医药卫生—肿瘤]