肝细胞癌预后相关的炎症反应关键基因的筛选及其预后价值

Development and prognostic significance of inflammation-related key genes model in hepatocellular carcinoma

目的:筛选与肝细胞癌(hepatocellular carcinoma,HCC)预后密切相关的炎症反应关键基因,构建预后风险评分模型,并评估该模型在HCC中的预后价值。方法:从TCGA数据库获取HCC患者肝肿瘤组织及正常肝组织的mRNA表达数据作为训练集,从GEO数据库中获取HCC患者的mRNA表达数据作为验证集。筛选出与HCC预后相关的炎症反应相关基因,运用LASSO回归和随机生存森林(RSF)方法筛选与HCC预后相关的炎症反应关键基因。基于这些关键基因构建预后风险评分模型并验证,应用Cox比例风险回归评估该模型对患者预后的影响。构建列线图并进行一致性分析。结果:共筛选出15个与HCC预后相关的炎症反应基因,经LASSO回归和RSF分析,确定11个炎症反应关键基因,即IL18RAP、MEP1A、RIPK2、CYBB、SLC7A1、ADM、IL7R、P2RX4、ACVR2A、SERPINE1、SLC7A2。构建的预后风险评分模型在训练集和验证集上预测1、3、5年生存率,AUC值均超过0.60;Kaplan-Meier分析显示高风险组总生存期(OS)显著低于低风险组(P<0.01);PCA和t-SNE分析显示该模型能有效区分高、低风险患者。Cox回归分析提示预后风险评分是独立的预后因素,且与OS显著相关(P<0.01)。列线图模型C-index为0.672,具有较高的预测精度,1年和3年校准曲线与标准曲线吻合度好,5年吻合度稍弱。结论:本研究成功构建并验证了一个基于炎症反应相关基因的风险评分模型,筛选出的11个炎症反应关键基因(IL18RAP、MEP1A、RIPK2、CYBB、SLC7A1、ADM、IL7R、P2RX4、ACVR2A、SERPINE1、SLC7A2)与HCC进展密切相关,且在模型中显示出较强的预后预测能力。

Objective:To screen key inflammatory genes closely related to the prognosis of hepatocellular carcinoma(HCC),construct a prognostic risk score model,and evaluate the prognostic value of this model in HCC.Methods:The mRNA expression data of liver tumor tissues and normal liver tissues of HCC patients were obtained from the TCGA database as the training set,and the mRNA expression data of HCC patients from the GEO database as the validation set.LASSO regression and Random survival forest(RSF)method were used to screen the key genes of inflammatory response related to HCC prognosis.Based on these key genes,a prognostic risk score model was constructed and validated.Cox proportional risk regression was applied to evaluate the effect of this model on patient prognosis.Construct a nomogram and perform a consistency analysis.Results:The study identified fifteen inflammation-related genes associated with HCC prognosis.After conducting LASSO regression and RSF analysis,eleven key inflammation-related genes were determined:IL18RAP,MEP1A,RIPK2,CYBB,SLC7A1,ADM,IL7R,P2RX4,ACVR2A,SERPINE1,and SLC7A2.The prognostic risk scoring model predicted 1-year,3-year,and 5-year survival rates with AUC values exceeding 0.60 in both the training and validation sets.Kaplan-Meier analysis revealed that the high-risk group exhibited significantly lower overall survival rates compared to the low-risk group(P<0.01).Furthermore,the PCA and t-SNE analyses demonstrated the model's effectiveness in distinguishing high-and low-risk patients.The Cox regression analysis showed that the prognostic risk score was an independent prognostic factor significantly correlated with overall survival(P<0.01).The nomogram model attained a C-index of 0.672,indicating high predictive accuracy.The calibration and standard curves for 1-year and 3-year predictions demonstrated good concordance,although slightly weaker concordance was observed for 5-year predictions.Conclusion:This study successfully developed and validated a risk scoring model based on inflammation-related genes.The eleven identified genes(IL18RAP,MEP1A,RIPK2,CYBB,SLC7A1,ADM,IL7R,P2RX4,ACVR2A,SERPINE1,and SLC7A2)are closely associated with HCC progression and demonstrate robust prognostic predictive power within the model.