ABCA3作为非小细胞肺癌的预后生物标志物及其与免疫细胞浸润的相关性

ABCA3 as a prognostic biomarker in non-small cell lung cancer and its correlation with immune cell infiltration

目的 本研究旨在基于生物信息学分析探讨ABCA3在非小细胞肺癌——肺腺癌(lung adenocarcinoma,LUAD)中的临床意义和生物学功能。方法 基于癌症基因组图谱(the Cancer Genome Atlas,TCGA),提取LUAD中ABCA3表达和相应的临床信息。使用R4.2.2软件进行统计分析。采用Wilcoxon符号秩和检验比较TCGA LUAD队列中肿瘤组织与对照组之间ABCA3的表达,以及临床病理参数与ABCA3表达水平之间的关系。根据ABCA3表达的中位数,将TCGA LUAD患者分为高表达组和低表达组。使用survival包和survminer包分析ABCA3与LUAD患者总生存(overall survival,OS)期和无进展生存(progression free survival,PFS)期的关系,并进行单因素和多因素Cox回归分析确定与LUAD预后相关的危险因素。使用rms包构建预测LUAD患者1年、2年和3年OS率的列线图模型。使用DESeq2包对高表达组和低表达组患者进行差异表达基因(differential expressed genes,DEGs)分析,并使用cluster Profil er包对DEGs进行基因本体(Gene Ontology,GO)数据库和基因组京都百科全书(Kyoto Encyclopedia of Genes and Genomes,KEGG)富集分析。利用CIBERSORT算法预测LUAD肿瘤样本和正常样本中22种浸润性免疫细胞的丰度,并分析ABCA3表达与免疫浸润细胞和免疫检查点之间的相关性。结果 ABCA3基因在LUAD肿瘤组织中的表达均低于配对和非配对样本正常组织(P<0.05)。ABCA3高表达LUAD患者的OS (P=0.031)和PFS (P=0.044)明显高于低表达患者。T2分期(P=0.004)和T3分期(P=0.044)患者ABCA3的表达显著低于T1分期患者。多因素Cox回归分析显示,T分期、N分期和ABCA3表达是影响LUAD预后的独立影响因素(P<0.05)。基于ABCA3表达、T分期和N分期构建的列线图预测模型的C-index为0.86,提示模型与实际结果较为一致。校准曲线中的偏差修正线接近理想曲线,提示列线图模型的预测结果与观测结果较吻合。基因集富集分析显示ABCA3高表达组主要与花生四烯酸代谢、补体和凝血级联、肠道免疫网络免疫球蛋白A产生等有关。ABCA3的表达与NRP1、TRFRSF14、TNFSF15、CD40LG、HHLA2、CD28呈明显正相关,而与IDO1、CD70、TNFSF4、CD276和VTCN1呈明显负相关(P<0.05)。12种免疫浸润细胞在高表达组和低表达组存在明显差异(P<0.05)。此外,14种免疫浸润细胞与ABCA3表达呈现明显相关性(P<0.05),包括巨噬细胞、CD4+T细胞和树突细胞。结论 ABCA3的表达与LUAD患者预后和免疫细胞浸润有关,可能是预测非小细胞肺癌预后的有价值生物标志物。

Objective The aim of this study is to investigate the clinical significance and biological functions of ABCA3 in non-small cell lung cancer-lung adenocarcinoma(LUAD)based on bioinformatics analysis.Methods Based on the Cancer Genome Atlas(TCGA) database,ABCA3 expression and corresponding clinical information in LUAD were extracted.Statistical analysis was performed using R4.2.2 software.Wilcoxon rank sum assay was used to compare ABCA3 expression between tumor tissues and controls in the TCGA LUAD cohort,and the relationship between clinicopathological parameters and ABCA3 expression levels.TCGA LUAD patients were divided into high and low expression groups based on the median ABCA3 expression.The relationship between ABCA3 and overall survival(OS)and progression-free survival(PFS)of LUAD patients was analyzed by using the survival package and survminer package,and univariate and multifactorial Cox regression analyses were performed to identify risk factors associated with LUAD prognosis.Columnar line graph models predicting the probability of 1-year,2-year and 3-year OS in LUAD patients were constructed by using the rms package.Differentially expressed genes(DEGs)analysis was performed for patients in the high and low expression groups by using the DESeq2 package.The CIBERSORT algorithm was used to predict the abundance of 22 infiltrating immune cells in LUAD tumor samples and normal samples,and to analyze the correlation between ABCA3 expression and immune infiltrating cells and immune checkpoints.Results ABCA1 gene expression was lower in LUAD tumor tissues than in normal tissues of both paired and unpaired samples(P<0.05).OS(P=0.031)and PFS(P=0.044)were significantly higher in patients with high ABCA3 expression LUAD than in patients with low expression.ABCA3 expression was significantly lower in T2-staged(P=0.004)and T3-staged(P=0.044)patients than in T1-staged patients.Multi-factor Cox regression analysis showed that T-stage,N-stage and ABCA3 expression were independent factors affecting the prognosis of LUAD(P<0.05).The C-index of the prediction model of column line graph constructed based on ABCA3 expression,T stage and N stage was 0.86,which suggesting that the model was more consistent with the actual results.The deviation correction line in the calibration curve was close to the ideal curve,which suggesting that the prediction results of the column line graph model were in better agreement with the observed results.GSEA analysis showed that the ABCA3 high expression group was mainly associated with arachidonic acid metabolism,complement and coagulation cascades,and Ig A production in the intestinal immune network.ABCA3 expression showed a significant positive correlation with NRP1,TRFRSF14,TNFSF15,CD40LG,HHLA2,and CD28,and a significant negative correlation with IDO1,CD70,TNFSF4,CD276,and VTCN1(P <0.05).Twelve immune infiltrating cells were significantly different between the high and low expression groups(P<0.05).In addition,14 immune infiltrating cells showed significant correlation(P <0.05)with ABCA3 expression,including macrophages,CD4+T cells and dendritic cells.Conclusions ABCA3 expression was associated with prognosis and immune cell infiltration in patients with LUAD and may be a valuable biomarker for predicting the prognosis of non-small cell lung cancer.