探索m6A相关lncRNAs预测胃腺癌患者总体生存率

m6A-Related LncRNAs in Predicting the Overall Survival of Gastric Adenocarcinoma Patients

目的:胃腺癌是胃癌最常见的亚型。N6-甲基腺苷(N6-methyladenosine,m6A)和长链非编码RNA(long non-coding RNAs,lncRNAs)在胃腺癌的预后价值中起着至关重要的作用。因此,辨别与m6A相关的lncRNA在胃腺癌研究中至关重要。方法:本研究从TCGA数据库中获取了胃腺癌研究的RNA序列转录组数据、相关临床信息。使用R语言的“limma”包对15 901个lncRNAs按肿瘤组织和正常组织分组进行差异表达分析。采用单因素Cox回归分析从TCGA数据集中的347个与m6A相关lncRNAs中筛选出与预后相关的基因67个(P <0.05)。使用R包“glmnet”对67个入选lncRNAs进行LASSO Cox回归分析,确定12个m6A相关的lncRNAs与胃腺癌患者的总生存期明显相关,并采用多因素Cox回归分析建立起m6A相关的lncRNAs风险评分模型。单因素和多因素回归分析结果表明风险评分模型是患者预后的独立影响因素(HR=3.45,95%CI为2.46~4.99,P <0.001)。分析风险评分与肿瘤免疫微环境细胞浸润的关系。最后,基于独立的预后影响因素(年龄、性别、风险评分、T分期、N分期)建立列线图。结果:确定12个m6A相关的lncRNAs与胃腺癌患者的总生存期明显相关,建立起m6A相关的lncRNAs风险评分模型。并证明该模型与肿瘤微环境细胞浸润存在相关性。基于独立的预后影响因素建立列线图,模型的C-index为0.73,模型预测有效性较高。结论:通过对lncRNAs风险模型的反复验证,证实了12个m6A相关lncRNAs可以作为胃腺癌患者风险预测的指标。

Objective:Gastric adenocarcinoma(GA)is the most frequently seen subtype of gastric cancer.N6-methylade-nosine(m6A)and long non-coding RNAs(IncRNAs)play vital roles in the prognosis of GA.Thus,discerning m6A-associat-ed IncRNAs in GA patients is critical.Methods:The RNA sequencing transcriptome data and relevant clinical information of GA patients were obtained from the TCGA database.The‘limma'package in R was used to perform to analyze the differ-ential expressions of 15,901 lncRNAs in tumor and normal tissues.Univariate Cox regression analysis was applied to select 67 genes associated with prognosis from 347 m6A-related IncRNAs in the TCGA dataset(P<0.05).The glmnet'package in R was used to conduct LASSO Cox regression analysis on the 67 selected lncRNAs,identifying 12 m6A-related IncRNAs significantly associated with the overall survival of GA patients.Multivariate Cox analysis was used to establish a risk scoring model for m6A-related IncRNAs.Univariate and multivariate regression analyses showed that the risk scoring model was an independent prognostic factor(HR=3.50,95%CI:2.46~4.99,P<0.001).The relationship between the risk scores and tumor immune microenvironment cell infiltration was analyzed.Finally,a nomogram was also constructed based on independent prognostic factors such as age,sex,risk score,T stage,N stage,etc.Results:Twelve m6A-related lncRNAs were significantly associated with the overall survival of GA patients,and a risk scoring model for m6A-related IncRNAs was established.It showed that this model was correlated to tumor microenvironment cell infiltration.A nomogram was constructed based on independent prognostic factors,with C-index of 0.73,indicating a high predictive efficacy of the model.Conclusion:Twelve m6A-associated IncRNAs were confirmed as predictors of risks in GA by repeated validation of the lncRNA risk model.