基于泛凋亡构建甲状腺癌预后预测模型及潜在中药预测  

Construction of a Prognosis Prediction Model and Potential TCM Prediction for Thyroid Cancer Based on Pan-apoptosis

作　　者：胡尧 伍涛 余会 郭佳佳 黄洪 代芳 HU Yao;WU Tao;YU Hui;GUO Jiajia;HUANG Hong;DAI Fang(Guizhou University of Traditional Chinese Medicine,Guiyang 550025,Guizhou,China;Department of En-docrinology,The First Affiliated Hospital of Guizhou University of Traditional Chinese Medicine,Guiyang 550001,Guizhou,China.)

机构地区：[1]贵州中医药大学,贵州贵阳550025 [2]贵州中医药大学第一附属医院内分泌科,贵州贵阳550001

出　　处：《贵州中医药大学学报》2025年第1期49-61,共13页Journal of Guizhou University of Traditional Chinese Medicine

基　　金：贵州省中医药管理局中医药、民族医药科学技术研究项目(QZYY-2014-019)。

摘　　要：目的分析泛凋亡在甲状腺癌发生发展过程中的生物学机制以及对THCA患者预后的预测价值。方法从癌症基因组图谱(TCGA)数据库中获取THCA样本和正常样本的基因表达谱及临床数据。进行泛凋亡基因差异分析,寻找预后相关泛凋亡基因并将样本分为两种泛凋亡亚型。筛选两种亚型间的差异基因,并通过单变量Cox分析和LASSO回归分析寻找特征基因构建预后预测模型。将患者按风险评分划为高、低风险组并进行肿瘤微环境、免疫细胞和免疫检查点的差异分析。通过医学本体信息检索平台Coremine Medical数据库预测对THCA具有潜在治疗作用的中药。结果共有47个泛凋亡基因在THCA组和正常组间存在差异表达,从中筛选出4个预后相关泛凋亡基因(BNIP3L、ELANE、E2F1、IL18)。根据A、B两种泛凋亡亚型间的差异基因确定了4个特征基因(ZNF367、PAPSS2、CILP、FGF7),通过这4个特征基因构建的预后预测模型表明,风险评分越高,预后越差,ROC曲线分析证明该模型具有优良的预测性能。差异分析得到高、低风险组之间肿瘤微环境、免疫细胞和免疫检查点表达差异显著。通过Coremine Medical数据库预测出灵芝、赤芍、地榆等对THCA具有潜在治疗作用的中药。结论本研究构建模型可准确预测THCA患者的预后,筛选出的相关基因可能是靶向治疗的潜在靶点,可为THCA的中药治疗提供参考。Objectivee To analyze the biological mechanisms of PANoptosis in the occurrence and development of thyroid cancer(THCA)and its predictive value for the prognosis of THCA patients.MethodssThe gene expression profiles,clinical data of THCA samples and normal samples were obtained from the cancer genome atlas(TCGA)database.Pan-apoptotic gene differential analysis was performed to search for prognosticrelated pan-apoptotic genes and to classify the samples into two pan-apoptotic subtypes.Select the dfferential genes between the two subtypes and find the prognosis prediction model by univariate Cox analysis and LASSO regression analysis.Patients were classified as high and low risk groups by risk score and underwent differential analysis of tumor microenvironment,immune cells and immune checkpoints.Traditional Chinese medicines with potential therapeutic effects on THCA were predicted through the Coremine Medical database of the medical ontology information retrieval platform.Results A total of 47 PANoptosis genes exhibited differential expression between THCA and the normal group,from which 4 prognostic-related PANoptosis genes(BNIP3L,ELANE,E2F1,IL18)were selected.Based on the differential genes between two PANoptosis subtypes A and B,four characteristic genes(ZNF367,PAPSS2,CILP,FGF7)were identified.Four characteristic genes(ZNF367,PAPSS2,CILP,FGF7)were identified according to the differential genes between the two pan-apoptotic subtypes A and B.The prognostic prediction model constructed by these four characteristic genes showed that the higher the risk score,the worse the prognosis,and the ROC curve analysis proved that the model had excellent predictive performance.Differential analysis yielded significant differences in tumor microenvironment,immune cells and immune checkpoint expression between the high-and low-risk groups.Traditional Chinese medicines with potential therapeutic effects of Ganoderma lucidum,red peony,and elm on THCA were predicted by Coremine Medical database.Conclusion The model constructed in

关 键 词：甲状腺癌 泛凋亡 预后预测模型 中药预测 灵芝 赤芍 地榆 

分 类 号：R736.1[医药卫生—肿瘤] R273[医药卫生—临床医学]