摘要
目的筛选出三阴性乳腺癌(TNBC)免疫相关LncRNA基因,构建TNBC预后预测模型。方法①从癌症基因组图谱(TCGA)项目下载TNBC组织和癌旁组织的转录组RNA表达谱原始数据,结合Immport数据库,筛选TNBC免疫相关的LncRNA。进一步应用单因素及多因素Cox分析筛选TNBC与预后关系密切的免疫相关LncRNA,然后再根据最佳AIC值确定与预后关系密切的免疫相关LncRNA。②根据预后关系密切的免疫相关LncRNA,构建TNBC预后预测模型。③根据TNBC预后风险模型测算的风险评分值的中位值将TNBC患者样本分为高风险组和低风险组,比较高低风险组生存率。④采用ROC、主成分分析(PCA)评估构建的TNBC预后预测模型的预测效能。⑤采用多因素Cox分析TNBC预后预测模型预测效能的独立性。结果共计65例份TNBC组织样本,得到369个免疫相关LncRNA,通过单因素和多因素Cox分析,共得到3个与预后关系密切的免疫相关LncRNA(AC090181.2、LINC01235、LINC01943),并构建TNBC预后预测模型,该模型表达公式如下:风险评分(RS)=(-6.24904×A C090181.2)+(0.240562×LINC01235)+(-2.18153×LINC01943)。高风险组和低风险组患者生率差异有统计学意义(P<0.001)。TNBC预后预测模型对患者预后预测效能好(AUC=0.910;高、低风险组投影点分布有区别,区分度高)。TNBC预后预测模型能独立预测TNBC预后(HR=1.028,P<0.05)。结论成功构建了由3个与预后关系密切的免疫相关LncRNA(AC090181.2、LINC01235、LINC01943)组成的TNBC预后预测模型,其能较好预测TNBC患者的预后。
Objective To screen out the immune-related LncRNA genes in triple-negative breast cancer(TNBC)and to construct a prognostic prediction model of TNBC.Methods①We downloaded the transcriptome RNA expres⁃sion profiling raw data of TNBC tissues and para-cancer tissues from the Cancer Genome Atlas(TCGA)project,and com⁃bined with the Immport database to screen TNBC immune-related lncRNAs.Furthermore,univariate and multivariate Cox analyses were used to screen immune-related lncRNAs closely related to TNBC and prognosis,and then immune-related ln⁃cRNAs closely related to prognosis were determined according to the optimal AIC value.②Based on the immune-related lncRNAs,the prognostic prediction model of TNBC was constructed.③The TNBC patient samples were divided into highrisk and low-risk groups based on the median of the risk score values calculated from the TNBC prognostic risk model,and survival was compared between the high-and low-risk groups.④ROC and principal component analysis(PCA)were used to evaluate the predictive efficacy of the TNBC prognostic prediction model.⑤Multivariate Cox analysis was used to ana⁃lyze the independence of TNBC prognostic prediction model in predicting efficacy.Results A total of 369 immune-related lncRNAs were obtained from 65 TNBC tissue samples.Three immune-related lncRNAs(AC090181.2,LINC01235,and LINC01943)were obtained by univariate and multivariate Cox analyses,and the prognostic prediction model of TNBC was constructed.The formula of the model was as follows:risk score(RS)=(-6.24904×AC090181.2)+(0.240562×LINC01235)+(-2.18153×LINC01943).There was significant difference between the high-risk group and the low-risk group(P<0.001).The predictive model of TNBC had a good predictive effect on the prognosis of patients(AUC=0.910;the distribution of projection points was different between the high-and low-risk groups,and the discrimi⁃nation degree was high).TNBC prognostic prediction model could predict TNBC prognosis independently(HR=1.028,P<0.05).Conclusion We have successfully constructed a prognostic prediction model consisting of three immune-relat⁃ed LncRNAs(AC090181.2,LINC01235,and LINC01943)closely related to prognosis,which can better predict the prognosis of TNBC patients.
作者
曹雷雨
刘伟
高艳
马晓丽
魏瑜
渠成程
努尔斯曼古丽·买买提明
张莉
CAO Leiyu;LIU Wei;GAO Yan;MA Xiaoli;WEI Yu;QU Chengcheng;Nuersimanguli Maimaitiming;ZHANG Li(Cadre Health Care Center,The First Affiliated Hospital of Xinjiang Medical University,Urumqi 830054,China;不详)
出处
《山东医药》
CAS
2023年第15期28-32,共5页
Shandong Medical Journal
基金
新疆维吾尔自治区重点研发计划项目(2020B03003-3,2020B03003)。
