EGFR-TKI获得性耐药相关基因标志物筛选及其对肺腺癌预后预测的价值  

Identification of an acquired EGFR-TKI resistance-related gene signature and its value of prognosis prediction in lung adenocarcinoma

作　　者：周灵巧 唐雄[2] 莫梦春 邱少绵 ZHOU Lingqiao;TANG Xiong;MO Mengchun;QIU Shaomian(The People's Hospital of Bishan District,Chongqing Bishan 402760,China;Guangdong Provincial People's Hospital,Guangdong Guangzhou 510080,China)

机构地区：[1]重庆市璧山区人民医院,重庆璧山402760 [2]广东省人民医院,广东广州510080

出　　处：《现代肿瘤医学》2022年第3期436-442,共7页Journal of Modern Oncology

摘　　要：目的:寻找有效的表皮生长因子酪氨酸激酶受体抑制剂(epidermal growth factor receptor-tyosine kinase inhibitor,EGFR-TKI)获得性耐药相关基因标志物,用于肺腺癌(lung adenocarcinoma,LUAD)的预后预测。方法:通过单因素和多因素Cox比例风险回归分析筛选出EGFR-TKI获得性耐药相关基因标志物,并构建LUAD预后风险评分模型。采用基因本体论(gene ontology,GO)进行功能注释。结果:筛选出EGFR-TKI获得性耐药相关基因MYO1D、PCDH7、TNFSF15、COL4A6、BEX2、WFDC21P、CADPS2作为LUAD预后的基因标志物,并基于标志基因构建了LUAD预后风险评分模型。针对建立的风险评分模型,通过Keplan-Meier生存曲线和受试者工作特征曲线分析可知,训练组中高风险患者5年生存率(37.8%)明显低于低风险患者(71.5%)(P<0.001),且受试者工作特征(receiver operating characteristic,ROC)曲线下面积(area under curve,AUC)为0.743,验证组中高风险患者5年生存率(25.6%)明显低于低风险患者(49.8%)(P<0.001),其AUC为0.719,说明构建的风险评分模型具有较好的预测能力。同时,将风险评分与其他临床信息进行Cox比例风险回归分析表明风险评分为独立预测指标,进一步说明筛选出的基因标志物为独立预后因子。GO功能注释显示,筛选出的标志基因可能从调控肌动蛋白、细胞外基质、凋亡等方面调控LUAD预后。结论:筛选出的EGFR-TKI获得性耐药相关基因标志物为独立预后因子并可能成为LUAD的潜在治疗靶点。Objective:To search for an effective acquired epidermal growth factor receptor-tyosine kinase inhibitor(EGFR-TKI)resistance-related gene signature,and use it to predict prognosis of lung adenocarcinoma.Methods:Univariate and multivariable Cox proportional hazards regression were applied to identify an acquired EGFR-TKI resistance-related gene signature,and gene signature-based prognostic risk score model was constructed.The function of gene signature was annotated by gene ontology(GO).Results:Acquired EGFR-TKI resistance genes(MYO1D,PCDH7,TNFSF15,COL4A6,BEX2,WFDC21P,CADPS2)were screened out as a gene signature of prognosis for lung adenocarcinoma(LUAD),and gene signature-based prognostic model was constructed.For the constructed prognostic risk score model,according to the analysis of Kaplan-Meier and ROC,5-year survival rate of high-risk patients(37.8%)was lower than that of low-risk patients(71.5%)in training dataset(P<0.001),and the area under the curve(AUC)was 0.743,5-year survival rate of high-risk patients(25.6%)was lower than that of low-risk patients(49.8%)in validation dataset(P<0.001),and the AUC was 0.719.The results showed that the risk score model had a powerful predictive ability.At the same time,Cox proportional hazards regression analysis of risk score and other clinical information verified that risk score was an independent prognostic factor,and it further showed that the selected gene signature was an independent prognostic factor.GO functional analysis revealed that the selected gene signature may regulate the prognosis of LUAD from actin,extracellular matrix,apoptosis and the likes.Conclusion:The selected acquired EGFR-TKI resistance gene signature is an independent prognostic factor and might be a therapeutic target for LUAD.

关 键 词：LUAD EGFR-TKI获得性耐药 预后 基因标志物 

分 类 号：R734.2[医药卫生—肿瘤]