基于免疫相关长链非编码RNA及基因建立膀胱癌预后模型并筛选靶向分子药物

Establishment of a prognostic model based on immunerelated long non-coding RNAs and immune-related genes and screening of targeted molecular drugs for bladder cancer

目的:膀胱癌(bladder cancer,BLCA)是具有高发病率的泌尿系统肿瘤之一,临床常规诊疗方法可提高患者的生存率,但肿瘤的复发和转移致使患者的预后仍然较差。长链非编码RNA(long non-coding RNA,lncRNA)在调控肿瘤发生、发展及癌症免疫中具有重要作用,可以作为一种新型的标志物预测患者预后及免疫应答。本研究基于生物信息学分析,构建BLCA预后模型,并筛选出BLCA的靶向分子药物。方法:从癌症基因组图谱(The Cancer Genome Atlas,TCGA)数据库下载RNA测序数据和临床数据,并将BLCA患者分为训练集和验证集。基于加权基因共表达网络分析(weighted gene co-expression network analysis,WGCNA)的方法提取免疫相关lncRNAs(immune related lncRNAs,IRlncRNAs),通过最小绝对收缩和选择算子(least absolute shrinkage and selection operator,LASSO)回归的方法构建预后模型,利用美国国家生物技术信息中心(National Center for Biotechnology Information,NCBI)数据库和泛癌数据进行预后基因集的肿瘤相关性分析,并依据Kaplan-Meier分析和受试者操作特征(receiver operating characteristic,ROC)曲线评价该模型的预后准确性。根据风险打分将样本分为高风险组与低风险组,并在此基础上通过风险差异基因筛选出高风险基因。最后,通过CMap数据库筛选抑制高风险基因表达的药物,从而降低BLCA患者的发病风险,提高患者的生存率。结果:基于TCGA数据库中的414例肿瘤样本和19例癌旁样本,预处理得到1342个免疫相关基因(immune related genes,IRGs)和409个临床数据,并通过WGCNA得到927个生存数据。根据Pearson相关分析和单因素Cox分析,得到74个与预后相关的IR-lncRNAs(|R|>0.4,P<0.05)。运用LASSO回归构建预后模型,得到12个IR-lncRNAs和21个IRGs。通过后续分析可知:该预后模型可预测高、低风险组患者的生存及预后情况,并验证了预后模型的独立预测能力及预测准确性,最后筛选了11种BLCA的潜在靶向药物。结论:本研究为BLCA患者构建了一个基于IRGs和IR-lncRNAs的预后模型,筛选了11种潜在的靶向分子药物,为BLCA患者的治疗提供了新思路。

Objective:Bladder cancer(BLCA)is one of the urinary system tumors with high incidence rate.Routine clinical diagnosis and treatment methods have improved the survival rate of patients,but the prognosis of patients remains poor due to tumor recurrence and metastasis.Long non-coding RNA(lncRNA)plays an important role in regulating tumorigenesis and cancer immunity,and can be used as a novel marker to predict patient prognosis and immune response.Based on bioinformatics analysis,this study constructs a BLCA prognostic model and screens for targeted molecular drugs in BLCA.Methods:RNA-sequencing(RNA-seq)data and the corresponding clinical data were downloaded and organized from the database of The Cancer Genome Atlas(TCGA),and the patient cohorts were assigned into a training cohort and a testing cohort.The immunerelated lncRNAs(IR-lncRNAs)were extracted based on the method of weighted gene coexpression network analysis(WGCNA).The model was constructed by least absolute shrinkage and selection operator(LASSO)regression analysis,and tumor correlation of prognostic gene set was analyzed by National Center for Biotechnology Information(NCBI)database and pan-cancer data.The prognostic accuracy of the model was evaluated according to Kaplan-Meier analysis and receiver operating characteristic(ROC)curve.In addition,the samples were classified into a low-risk group and a high-risk group according to risk scores,and then high-risk genes were identified by differentially expressed genes.Finally,the CMap was used to screen for small molecule drugs that could suppress the expression of high-risk genes,so as to reduce the risk of BLCA patients and improve the survival rate of patients.Results:Based on 414 tumor samples and 19 normal samples in TCGA database,a total of 1342 immune-related genes(IRGs)and 409 clinical data were pretreated,and 927 survival data were obtained through WGCNA.Pearson correlation analysis and univariate Cox regression analysis showed that 74 IR-lncRNAs linked to prognosis were screened out(|R|>0.4,P<0.05).Twelve IR-lncRNAs and 21 IRGs were performed through LASSO regression analysis.According to the follow-up analysis,the prediction model predicted the survival and prognosis of patients in high-risk and low-risk group,verified the independent prediction ability and prediction accuracy of the prognostic model,and finally screened 11 potential targeted drugs of BLCA.Conclusion:A prognostic model based on IRGs and IR-lncRNAs for BLCA patients is constructed,and 11 potential targeted molecular drugs are screened,which may provide new ideas for the precise treatment of BLCA patients.