基于生物信息学分析细胞周期蛋白B1与肺腺癌患者预后及浸润性免疫细胞的关系

Analysis of relationship between cyclin B1 and prognosis and infiltrating immune cells in patients with lung adenocarcinoma based on bioinformatics

目的通过生物信息学探索细胞周期蛋白B1(CCNB1)与肺腺癌患者预后及浸润性免疫细胞的关系。方法从TCGA下载肺腺癌患者CCNB1基因转录表达数据及临床信息。通过临床蛋白质组学肿瘤分析联盟(CPTAC)进一步研究肺腺癌中的CCNB1蛋白表达情况。受试者工作特征(ROC)曲线获取区分肺腺癌与邻近正常组织的CCNB1基因最佳截断值。利用单因素及多因素COX及Kaplan-Meier方法评估CCNB1与预后的关系。利用STRING构建蛋白质-蛋白质相互作用(PPI)网络。利用肿瘤免疫估计资源(TIMER)评估CNNB1基因与肿瘤浸润性免疫细胞的关系。结果CNNB1基因及蛋白在肺腺癌组织中的表达显著上调。CCNB1在男性、淋巴结转移、高TNM分期、吸烟、吸烟超过40年的患者中高表达,在年龄超过60岁的患者中低表达(P均<0.05)。Kaplan-Meier曲线表明,CCNB1高表达组的肺腺癌患者的预后比低表达组的患者更差[42.2个月vs 54.4个月,HR(95%CI):1.586(1.186~2.120),P=0.002],各亚组分析也显示了相同的结果。ROC曲线表明,区分肺腺癌与正常组织的最佳CCNB1值为3.483,AUC值为0.980(95%CI:0.967~0.993),其敏感性、特异性、阳性预测值、阴性预测值及约登指数分别为98.3%、93.5%、62.4%、99.8%、91.8%。相关分析表明,CCNB1表达与CD4^(+)T细胞(r=-0.2,P<0.001)、B细胞(r=-0.255,P<0.001)及树突状状细胞(r=-0.124,P=0.006)的浸润丰度具有负相关性,与中性粒细胞(r=0.173,P<0.001)及CD8^(+)T细胞(r=0.109,P=0.015)浸润丰度具有正相关性。结论高表达的CCNB1与肺腺癌预后不良和免疫浸润显著相关,可作为预后生物标志物及免疫治疗的潜在靶点。

Objective To explore the relationship between cyclin B1(CCNB1)and the prognosis and infiltrating im⁃mune cells of patients with lung adenocarcinoma based on bioinformatics.Methods The CCNB1 gene transcription ex⁃pression data and clinical information of patients with lung adenocarcinoma were downloaded from TCGA.The expression of CCNB1 protein in lung adenocarcinoma was further analyzed by the Clinical Proteomics Tumor Analysis Consortium(CPTAC).The best cut-off value of CCNB1 gene that distinguishes lung adenocarcinoma from adjacent normal tissues was obtained by receiver operating characteristic(ROC)curve.Single-factorand multi-factor COX and Kaplan-Meier methods were used to evaluate the relationship between CCNB1 and prognosis.STRING were used to build a protein-protein interaction(PPI)network.The tumor immunity estimation resource(TIMER)were used to evaluate the relationship between CCNB1 gene and tumor infiltrating immune cells.Results The expression of CCNB1 gene and protein in lung adenocarcinoma tissue was significantly up-regula⁃ted.CCNB1 was highly expressed in men,lymph node metastasis,high TNM staging,smoking,and patients who had smoked for more than 40 years,and was low in patients over 60 years of age(all P<0.05).The Kaplan-Meiercurve shows that the prognosis of lung adenocarcinoma patients in the CCNB1 high expression group was worse than that of the low expression group[42.2 months vs 54.4 months,HR(95%CI):1.586(1.186-2.120),P=0.002],the analy⁃sis of each subgroup also showed the same result.The ROC curve shows that the best CCNB1 value for distinguishing lung adenocarcinoma from normal tissue is 3.483,and the AUC value is 0.980(95%CI:0.967⁃0.993),and its sen⁃sitivity,specificity,positive predictive value,negative predictive value and Yorden index were respectively 98.3%,93.5%,62.4%,99.8%,91.8%.Correlation analysis showed that CCNB1 expression was negatively correlated with infiltration abundance of CD4^(+)T cells(r=-0.2,P<0.001),B cells(r=-0.255,P<0.001)and dendritic cells(r=-0.124,P=0.006),and positively correlated with infiltration abundance of neutrophils(r=0.173,P<0.001)and CD8^(+)T cells(r=0.109,P=0.015).Conclusion Highly expressed CCNB1 is significantly related to poorprog⁃nosis and immune infiltration of lung adenocarcinoma,and can be used as a prognostic biomarkerand a potential target for immunotherapy.