与胰腺癌免疫预后相关的自噬相关基因预测

Prediction of autophagy-related genes associated with immune prognosis in pancreatic cancer

目的通过表征自噬相关基因(ARGs)在胰腺癌(PAAD)中的表达模式和预后,寻找与胰腺癌免疫治疗预后不良相关的候选基因。方法从人类自噬数据库(HADb)和基因集富集分析(GSEA)网站获得566个自噬相关基因,通过GEPIA、Kaplan-Meier Plotter和cBioPortal数据库检测胰腺癌患者中ARGs的RNA-seq表达和生存分析数据,筛选出与无复发生存期(RFS)和总生存期(OS)显著相关的差异表达基因(DEGs)。利用DAVID及STRING数据库进行GO和KEGG分析,预测相关信号通路。利用TISIDB分析ARGs的表达量与肿瘤浸润淋巴细胞及免疫检查点抑制剂的相关性。结果GEPIA分析结果显示大部分DEGs在胰腺癌中表达高于正常胰腺组织(P<0.05)。生存分析结果表明CTTN、MET、MTDH、RALB、RIPK2、BCL2L1、CHMP2B、EIF2S1、ITGA3、NCKAP1、SH3GLB1表达高的患者有更低的OS和RFS。ATP6V0D1、ATP6V0A1、ATP6V0E2、MAP1LC3A表达高的患者有更高的OS和RFS。KEGG表明NOD样受体信号通路、PI3K-AKT信号通路和RAS信号通路是预后不良相关基因的主要的信号通路。TISIDB分析表明,候选ARGs与肿瘤浸润淋巴细胞之间的相关性趋势同ARGs与免疫检查点抑制剂之间的趋势存在相似性。结论自噬相关基因MET、RALB、RIPK2、BCL2L1、ITGA3在胰腺癌中表达升高,是潜在的影响免疫治疗预后不良的候选基因。

Objective To characterize the expression and correlation of autophagy-related genes (ARGs) to prognosis in pancreatic cancer for finding the candidate genes associated with poor immunotherapy. Methods A total of 566 ARGs were obtained from the Human Autophagy Database (HADb) and Gene Set Enrichment Analysis (GSEA) website.The RNA-seq expression and survival data of ARGs in patients with pancreatic cancer were detected by GEPIA, Kaplan– Meier Plotter, and c BioPortal databases. Differentially expressed genes (DEGs)that significantly correlated with relapse free survival (RFS) and overall survival (OS) were screened. DAVID and STRING databases were used to analyze Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) and predict relevant signal pathways. TISIDB was used to analyze the correlation between the expression of ARGs and tumor infiltrating lymphocytes and immune checkpoint inhibitors. Results GEPIA analysis showed that most of DEGs expression in pancreatic cancer was higher than that in normal pancreatic tissue (P<0.05).Survival analysis showed that patients with high expression of CTTN, MET, MTDH, RALB, RIPK2, BCL2L1,CHMP2B, EIF2S1, ITGA3, NCKAP1 and SH3GLB1 had lower OS and RFS. Patients with high expression of ATP6V0D1, ATP6V0A1, ATP6V0E2 and MAP1LC3A had higher OS and RFS. KEGG showed that NOD like receptor signaling pathway, PI3K-AKT signaling pathway and RAS signaling pathway were the main signaling pathways of genes related to poor prognosis. TISIDB analysis showed that the correlation trend between candidate ARGs and tumor infiltrating lymphocytes was similar to that between ARGs and immune checkpoint inhibitors.Conclusion Autophagy-related genes MET, RALB, RIPK2, BCL2L1 and ITGA3 increase in pancreatic cancer, and are potential candidate genes for poor prognosis in immunotherapy.