铁死亡相关基因与分化型甲状腺癌患者预后相关性研究

Ferroptosis-related gene and its relationship with prognosis of differentiated thyroid carcinoma patients

目的基于TCGA数据库分析分化型甲状腺癌(differentiated thyroid carcinoma,DTC)铁死亡标志物及其预后意义。方法本研究从TCGA数据库中下载甲状腺癌患者及甲状腺正常样本的RNA表达数据及生存信息。基因差异分析中增加653例来自GETx数据库的正常甲状腺样本。选择22个铁死亡相关基因进行单因素COX分析。使用LASSO回归进一步在TCGA队列中筛选与预后相关基因并构建预后模型。Adjusted P<0.05且|log2FC>1"为阈值筛选出22个差异表达基因。将筛选出的基因进行多因素COX分析,筛选预后相关基因,并构建列线图。结果纳入的24个铁死亡相关基因中22个在肿瘤和正常组织中差异表达,其中高表达13个,低表达9个,1个基因表达差异无统计学意义,1个基因在半数样本中无表达。单因素COX分析发现DPP4和TFRC与疾病风险程度相关(HR分别为<1和>1)。将GPX4、TFRC和DPP4纳入LASSO模型筛选后,以公式lambda.min=0.0027,Riskscore=(0.7316)×TFRC+(-0.2539)×DPP4,将患者进行危险程度分组后发现与预后相关(Log rank P=0.00635)。多因素COX分析发现DPP4和TFRC与疾病风险程度相关(HR分别为<1和>1)。结论铁死亡相关差异表达基因TFRC和DPP4构建的模型可能是DTC患者潜在的预后标志物,仍需进一步验证,将为患者的临床治疗提供理论依据。

Objective To analyze the marker of ferroptosis-related genes in differentiated thyroid carcinoma(DTC)based on TCGA database.Methods The mRNA expression profiles and survival information of thyroid cancer patients and normal thyroid samples were downloaded from the TCGA database.The genetic difference analysis added 653 normal thyroid samples from the GETx database.Twenty-two ferroptosis-related genes were selected for univariate Cox regression analysis.Genes associated with prognostic in the TCGA cohort were further screened and prognostic models using LASSO regression was constructed.Adjusted P<0.05 and|log2FC>1"as the threshold,22 differentially expressed genes were selected.The genes were screened by multivariate Cox regression analysis for prognosis-related genes and displayed in a line diagram.Results 22 of the 24 ferroptosis-related genes were differentially expressed between the tumor and normal tissues,with13 high expression,9 low expression,1 gene expression without difference and 1 gene not expressed in half of the samples.Univariate Cox regression analysis found that DPP4 and TFRC were associated with the degree of disease risk(HR was<1 and>1,respectively).When integrating GPX4,TFRC and DPP4 into the LASSO model screening,it was found to be related to prognosis after dividing the patients into risk groups according to lambda.min=0.0027,Riskscore=(0.7316)*TFRC+(-0.2539)*DPP4(Log rank P=0.00635.Multivariate Cox regression analysis found that DPP4 and TFRC were related to the degree of disease risk(HR were<1 and>1,respectively).Conclusion The model of TFRC and DPP4 constructed by ferroptosis-related differential expression genes may be potential predictive markers of DTC patients,which still needs further verification and will provide theoretical basis for further clinical treatment.