自噬相关基因的表达量化及其对肾透明细胞癌的预后价值分析

Quantification of the expression of autophagy-related genes and its prognostic value in renal clear cell carcinoma

目的采用生物信息学方法分析预后相关的自噬相关基因在泛癌中的表达并构建肾透明细胞癌(KIRC)的预后模型以准确评估患者的预后。方法于GEPIA数据库筛选与预后相关的自噬相关基因,得到的基因用ssGSEA计算其在各癌症中的自噬评分,以量化其在癌症中的表达。单因素、多因素Cox回归分析筛出自噬评分与KIRC患者预后的相关性。edgeR包分析KIRC患者差异表达的自噬基因,再进行逐步、LASSO和多因素Cox回归分析以构建风险预后模型,生存曲线、受试者工作特征曲线(ROC)评估模型的预测价值。CIBERSORT计算高低风险组间免疫细胞的浸润情况,采用CellMiner数据库筛选模型基因的敏感性药物。结果与预后相关的自噬相关基因在各癌症中的表达存在差异,其中在KIRC中表达最高,且自噬评分与其预后密切相关。经逐步回归分析、LASSO分析筛选出4个KIRC的预后模型基因,并经多因素Cox回归分析显示,BIRC5和EIF4EBP1高表达是影响KIRC预后的危险因素[HR(95%CI)=1.021(1.010~1.033)、1.003(1.001~1.005)],而BAG1和BNIP3高表达是影响KIRC预后的保护因素[HR(95%CI)=0.963(0.942~0.985)、0.997(0.994~0.999)]。模型基因风险评分、M分期和N分期可作为KIRC预后的独立影响因素(P<0.05)。ROC曲线显示,模型基因预测KIRC患者1、3、5年生存率的曲线下面积(AUC)分别为0.743、0.740、0.699。高低风险组患者在多种免疫细胞浸润中存在差异,与低风险组比较,高风险组中CD8^(+)T细胞、辅助性滤泡T细胞、调节性T细胞和M0巨噬细胞等比例显著增加,而静息杀伤细胞、单核细胞和静息肥大细胞等比例显著降低。模型基因与多种癌细胞系药物的敏感性相关,尤其是在肾癌细胞系中,相关性系数可达到0.9以上。结论自噬相关基因分析可为KIRC患者提供独立、可靠的生物标志物和治疗靶点,为探索KIRC个性化治疗提供新思路。

Objective Using bioinformatics methods to analyze the expression of prognostic autophagy related genes in pancancer and construct a prognostic model for renal clear cell carcinoma(KIRC)to accurately assess the prognosis of patients.Methods The autophagy related genes related to prognosis were screened from the GEPIA database,and the obtained genes were used to calculate their autophagy scores in various cancers using ssGSEA to quantify their expression in cancer.Univariate and multivariate Cox regression analysis screened the correlation between autophagy scores and prognosis in patients with KIRC.The edgeR package analyzed autophagy genes differentially expressed in patients with KIRC,and then conducted stepwise,LASSO,and multivariate Cox regression analysis to construct a risk prognosis model.The survival curve,and subject work characteristic curve(ROC)were used to evaluate the predictive value of the model.CIBERSORT calculates the infiltration of immune cells between high-risk and high-risk groups,and uses the CellMiner database to screen sensitive drugs for model genes.Results The expression of autophagy related genes related to prognosis varies among various cancers,with the highest expression in KIRC,and autophagy scores are closely related to prognosis.Four prognostic model genes for KIRC were selected through stepwise regression analysis and LASSO analysis.Multivariate Cox regression analysis showed that the high expression of BIRC5 and EIF4EBP1 was a risk factor affecting the prognosis of KIRC[HR(95%CI)=1.021(1.010-1.033),1.003(1.001-1.005)],while the high expression of BAG1 and BNIP3 was a protective factor affecting the prognosis of KIRC[HR(95%CI)=0.963(0.942-0.985),0.997(0.994-0.999)].Model gene risk score,M stage,and N stage can be independent influencing factors for the prognosis of KIRC(P<0.05).The ROC curve showed that the area under the curve(AUC)of the model gene predicting the 1,3,and 5 year survival rates of KIRC patients were 0.743,0.740,and 0.699,respectively.There are differences in the infiltration of multiple immune cells in patients with high and low risk.Compared with the low risk group,the proportion of CD8^(+)T cells,auxiliary follicular T cells,regulatory T cells,and M0 macrophages in the high risk group significantly increased,while the proportion of resting killer cells,monocytes,and resting mast cells significantly decreased.Model genes are associated with the sensitivity of various cancer cell lines to drugs,especially in renal cancer cell lines,with a correlation coefficient of more than 0.9.Conclusions Autophagy related gene analysis can provide independent and reliable biomarkers and treatment targets for patients with KIRC,and provide new ideas for exploring personalized treatment of KIRC.