基于4个自噬相关LncRNA表达信息学分析构建肺鳞癌预后风险评分模型

ESTABLISHMENT OF A PROGNOSTIC RISK SCORING SYSTEM FOR LUNG SQUAMOUS CELL CARCINOMA BASED ON INFORMATICS ANALYSIS OF THE EXPRESSION OF FOUR AUTOPHAGY-RELATED LONG NON-CODING RNAS

目的利用癌症基因组图谱(TCGA)数据库基于自噬相关长链非编码RNA(LncRNA)表达信息学分析构建肺鳞癌病人预后的风险评分模型。方法从TCGA数据库和人类自噬基因数据库(HADb)下载肺鳞癌和正常肺组织的自噬基因表达谱数据及临床相关资料,计算肺鳞癌自噬基因的差异表达,筛选出与自噬基因相关LncRNA。采用单因素Cox风险回归和LASSO回归的方法筛选和建立自噬相关LncRNA预后模型,并通过实验验证筛选出的自噬相关LncRNA在肺鳞癌中的表达。结果从TCGA数据库中下载得到502例肺鳞癌组织和49例正常肺组织的自噬基因表达谱,使用R软件中的Limma包进行差异基因分析(log 2(fold change)的绝对值>1.5,伪发现率<0.05),得到差异表达的自噬基因30个,与自噬基因相关的LncRNA有89个。单因素Cox分析及LASSO回归分析得到基于4个LncRNA的预后风险模型:风险评分=0.26×AC245060.2+0.18×AL390719.2+0.23×AC012181.1+0.21×AL365356.4。该模型可较准确区分高、低风险病人,并且可独立于其他变量作为肺鳞癌预后的预测因子。实时定量PCR检测显示,AC245060.2、AL390719.2、AC012181.1、AL365356.4在肺鳞癌组织中的表达较正常肺组织高(t=17.35~69.28,P<0.05)。结论本研究构建了自噬相关LncRNA预后模型,可为临床评估肺鳞癌病人的预后提供参考。

Objective To establishan autophagy-related long non-coding RNA(LncRNA)riskscoringsystem for evaluating the prognosis of patients with lung squamous cell carcinoma(LUSC)by using The Cancer Genome Atlas(TCGA)database.Methods The TCGA database andthe Human Autophagy Database(HADb)were used to download the data on autophagy gene expression profile and related clinical data in LUSCtissue and normal lung tissue,and the differential expression of autophagy genes in LUSC was calculated to screen for the LncRNAs associated with autophagy genes.The univariate Cox proportional-hazards regression modeland LASSO regression analysis were used for screening and establishing a prognostic model based onautophagy-rela-ted LncRNAs.Results The autophagy gene expression profiles of 502 LUSC tissuesamples and 49 normal lung tissuesamples were obtained from the TCGA database.The Limma package in R software was used to perform an analysis of the differentially expressed genes(the absolute value of log 2(fold change)>1.5 andafalse discovery rate of<0.05),which obtained 30 differentially expressedautophagy genes and 89 LncRNAs associated with these autophagy genes.The univariate Cox analysis and LASSO regression analysis obtained a prognostic risk model based on four LncRNAs,i.e.,risk score=0.26×AC245060.2+0.18×AL390719.2+0.23×AC012181.1+0.21×AL365356.4.This model could accurately distinguish high-risk LUSC patients from low-risk LUSC patients and could be used as an independent predictive factor forthe prognosis of LUSC.Quantitative real-time PCR showedthat the expression levels of AC245060.2,AL390719.2,AC012181.1,and AL365356.4 in LUSC tissue were significantlyhigher than thosein normal lung tissue(t=17.35-69.28,P<0.05).Conclusion Thisstudy establishes a prognostic model based on autophagy-related LncRNAs,which provides a reference for evaluating the prognosis of LUSC in clinical practice.