摘要
目的:利用自噬促进基因构建胃癌预后预测模型并进行验证。方法:利用TCGA胃癌队列筛选存在显著表达差异及与预后显著相关的基因,通过LASSO回归分析建立预后预测模型,并利用GEO胃癌队列对模型的预后预测效能进行验证。结果:139个自噬促进基因中21.6%在胃癌和邻近正常组织中表达存在显著差异,单因素Cox回归分析显示其中33.1%与胃癌患者预后显著相关。构建由12个基因构成的预后预测模型,以中位风险值将病例分为高、低风险组。高风险组OS显著短于低风险组。Cox单因素与多因素分析显示风险评分与患者预后显著相关。时间依赖性ROC肯定了模型的预后预测效能。结论:自噬促进基因在胃癌中具有预后预测价值。
Objective:To construct and validate a prognostic prediction model for gastric cancer using genes which positively regulate autophagy.Methods:TCGA gastric cancer cohort was used to screen the genes with significant expression differences and significant prognostic correlation.LASSO regression analysis was used to establish the prognostic prediction model,and GEO gastric cancer cohort was used to verify the prognostic prediction efficiency of the model.Results:Among the 139 autophagy promoting genes,21.6%were significantly differentially expressed between gastric cancer and adjacent normal tissues.Univariate Cox regression analysis showed that 33.1%of them were significantly associated with the prognosis of gastric cancer patients.A prognostic prediction model consisting of 12 genes was constructed,and the patients were divided into high risk group and low risk group according to the median risk value.The OS of the high risk group was significantly shorter than that of the low risk group.Cox univariate and multivariate analysis showed that risk score was significantly correlated with patient prognosis.The time-dependent ROC confirmed the prognostic efficacy of the model.Conclusion:Autophagy promotion genes have prognostic value for gastric cancer.
作者
刘礼新
蔺瑞江
岳鹏
胡文滕
王栋
马敏杰
LIU Li-xin;LIN Rui-jiang;YUE Peng;HU Wen-teng;WANG Dong;MA Min-jie(Thoracic Surgery Department,The First Hospital of Lanzhou University;The First Clinical Medical College of Lanzhou University,Lanzhou,Gansu 730030)
出处
《赣南医学院学报》
2023年第2期128-134,共7页
JOURNAL OF GANNAN MEDICAL UNIVERSITY
基金
甘肃省青年科技基金计划项目(21JR7RA382)
兰州市科技发展指导性计划项目(2022-ZD-98)
兰州大学第一医院青年基金计划项目(ldyyyn2020-75)
兰州大学学生创新创业行动计划项目(2022)。
关键词
胃癌
自噬
预后预测模型
LASSO回归分析
生存分析
Gastric cancer
Autophagy
Prognostic prediction model
LASSO regression analysis
Survival analysis