摘要
目的探讨铁死亡基因(FRGs)和肝癌预后的关系,构建FRGs的肝癌预后评估模型。方法从癌症基因组图谱(TCGA)及国际癌症基因组联盟(ICGC)数据库下载肝癌的转录组及临床数据,应用R和Perl软件对数据进行分析整理。TCGA肝癌数据集中,采用Wilcoxon检验获取肝癌和正常组织差异表达的FRGs,单因素Cox回归和最小绝对值收敛和选择算子(Lasso)分析挑选与生存相关的基因并构建肝癌预后模型。根据模型评分将患者分为高低评分两组,Wilcoxon检验获取两组间的差异表达基因(DEGs),并对DEGs进行基因本体论(GO)和京都基因与基因组百科全书(KEGG)功能富集分析,同时进行单样本基因集富集分析(ssGSEA)探究预后差异的可能机制。在ICGC肝癌数据集中验证模型的可靠性。生存数据比较采用Kaplan-Meier法和Log-Rank检验,独立t检验比较高低评分组间的免疫细胞浸润。结果对比肝癌和癌旁组织,从292个FRGs中,筛选8个(AURKA、LOX、FOXM1、G6PD、MAPT、SLC7A11、NQO1和STMN1)与肝癌患者总生存时间(OS)显著相关,基于此构建了肝癌预后评估模型。高评分组患者OS明显短于低评分组患者(3.148年比5.838年,χ^(2)=15.307,P<0.01),且模型[风险比(HR)=3.02,95%可信区间(CI):2.15~4.23,P<0.05]能独立于患者性别,年龄,肿瘤分级及分期,预测肝癌患者的预后。模型预测患者1,2,3年生存率的受试者工作特征(ROC)曲线下面积在测试集和验证集中分别为0.794,0.726,0.691和0.740,0.772,0.766。模型的高低评分组间DEGs的GO富集主要与免疫功能相关;KEGG分析主要富集于白细胞介素(IL)-17信号通路,肿瘤坏死因子(TNF)信号通路等。ssGSEA分析提示高评分组高于低评分组的肿瘤浸润的免疫细胞有树突状细胞(DC,0.627比0.602,t=-4.522,P<0.01)、巨噬细胞(0.753比0.727,t=-4.944,P<0.01)、Th2细胞(0.537比0.506,t=-3.733,P<0.01),Treg细胞(0.777比0.766,t=-4.719,P<0.01),而高评分组低于低评分组的肿瘤浸润免疫细胞有NK细胞(0.510比0.544,t=4.121,P<0.01);在免疫功能中,高评分组的免疫检查点(0.656比0.641,t=-2.880,P<0.01)及HLA-Ⅰ类抗原(0.978比0.973,t=-4.382,P<0.01)高于低评分组。结论FRGs预后评估模型可独立预测肝癌患者预后,其可能机制是免疫抑制的肿瘤微环境。
Objective To explore the relationship between the ferroptosis-related genes(FRGs)and the prognosis of liver cancer,and construct the prognosis assessment model of the FRGs for liver cancer.Methods Transcriptomic and clinical data of hepatocellular carcinoma were downloaded from The Cancer Genome Atlas(TCGA)database and the International Cancer Genome Consortium(ICGC)database,and analyzed by R and Perl software.In TCGA database,the differential expression genes of ferroptosis-related in liver cancer and normal tissues were obtained by Wilcoxon test in.And,Univariate Cox regression analysis and least absolute value convergence and selection operator(Lasso regression)analysis were used to select survival-related genes and constructed a prognostic model of liver cancer.According to the model score,the patients were divided into two groups with high and low scores,and the differentially expressed genes(DEGs)between the two groups were obtained by the Wilcoxon test.DEGs were analyzed by gene ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG).At the same time,a single-sample gene set enrichment analysis(ssGSEA)was performed to explore the possible mechanism of prognostic differences.Validating the reliability of the model in the ICGC liver cancer dataset.Survival data were compared using Kaplan-Meier method and Log-Rank test,and independent t test was used to compare immune cell infiltration between high and low score groups.Results Comparing hepatocellular carcinoma and paraneoplastic tissue,8 FRGs(AURKA,LOX,FOXM1,G6PD,MAPT,SLC7A11,NQO1 and STMN1)were identified from 292 which were associated with liver cancer patients′overall survival(OS)time.The prognostic assessment model about FRGs was constructed.The OS time of liver cancer patients with high risk scores in this model was significantly shorter than that of patients with low risk scores(3.148 years vs.5.838 years,χ^(2)=15.307,P<0.01).And the prognostic assessment model[hazard ratio(HR)=3.02,95%confidence interval(CI):2.15-4.23,P<0.05]independently predicted the prognosis of patients with hepatocellular carcinoma independently of patient gender,age,tumour grade and stage.The area under the Receiver Operating Characteristic(ROC)curve of the model for predicting the survival rates of patients among 1,2,3 years in the testing set and validating set were 0.794,0.726,0.691 and 0.740,0.772,0.766,respectively.The GO enrichment of DEGs in high and low scoring groups was mainly associated with immune function.KEGG analysis were mainly enriched in interleukin(IL)-17 signalling pathway,tumor necrosis factor(TNF)signalling pathway,etc.ssGSEA analysis showed that the tumor-infiltrating immune cells in the high-score group were higher than those in the low-score group,including dendritic cells(DC,0.627 vs.0.602,t=-4.522,P<0.01),macrophages(0.753 vs.0.727,P<0.01),Th2 cells(0.537 vs.0.506,t=-3.733,P<0.01),Treg cells(0.777 vs.0.766,t=-4.719,P<0.01),while NK cells(0.510 vs.0.544,t=4.121,P<0.01)in the high-score group was lower than that in the low-score group.In the immune function,the immune checkpoints in the high score group(0.656 vs.0.641,t=-2.880,P<0.01)and HLA-Ⅰantigens(0.978 vs.0.973,t=-4.382,P<0.01)higher than the low score group.Conclusion The prognostic assessment model of FRGs constructed in this study can independently predict the prognosis of liver cancer patients,and its possible mechanism is the immunosuppressive tumor microenvironment,which provides a new direction for the selection of precise strategies for the prevention and treatment of liver cancer.
作者
范顺利
王政禄
史源
张赛
王昊
郑虹
Fan Shunli;Wang Zhenglu;Shi Yuan;Zhang Sai;Wang Hao;Zheng Hong(The First Central Clinical School,Tianjin Medical University,Tianjin 300070,China;Organ Transplant Department,Tianjin First Central Hospital,Tianjin Medical University,Tianjin 300192,China;School of Medicine,Nankai University,Tianjin 300074,China)
出处
《中华实验外科杂志》
CAS
北大核心
2022年第10期1875-1878,共4页
Chinese Journal of Experimental Surgery
关键词
铁死亡基因
肝癌
预后
生物信息学
Ferroptosis-related genes
Liver cancer
Prognosis
Bioinformatics
