肾恶性横纹肌样瘤免疫浸润分析及预后模型构建  

Prognostic modeling and immune infiltration analysis for malignant rhabdoid renal tumor

作　　者：韦佶秀 许琼倩 陈鹏 黄静茹 陈嘉波[1] Wei Jixiu;Xu Qiongqian;Chen Peng;Huang Jingru;Chen Jiabo(Department of Pediatric Surgery,First Affiliated Hospital,Guangxi Medical University,Nanning 530021,China;Department of Pediatric Surgery,Qilu Hospital,Shandong University,Jinan 250012,China)

机构地区：[1]广西医科大学第一附属医院小儿外科,南宁530021 [2]山东大学齐鲁医院小儿外科,济南250012

出　　处：《中华小儿外科杂志》2024年第3期217-223,共7页Chinese Journal of Pediatric Surgery

基　　金：广西自然科学基金面上项目(2020GXNSFAA297130)。

摘　　要：目的基于生物信息学分析肾恶性横纹肌样瘤免疫浸润水平,鉴定与预后免疫相关基因,并构建预后模型。方法从TARGET数据库获取50例肾恶性横纹肌样瘤高通量测序数据和临床参数,使用estimate、CIBERSORT、ssGSEA分析免疫浸润特征及免疫浸润结果聚类亚型(高免疫浸润组和低免疫浸润组),通过亚型间的基因差异分析获取差异免疫相关基因。单因素COX、LASSO回归分析筛选预后相关免疫基因并构建预后模型。结果高免疫浸润组较低免疫浸润组具有较高的估计得分、免疫评分和基质核心水平,且M2巨噬细胞浸润程度最高。筛选出4个预后免疫相关基因(FGF12、REG1A、TSLP、ANGPTL7)并以此构建预后风险模型。Kaplan-Meier生存曲线和ROC评估预后模型显示出较高风险分层能力,单因素和多因素COX回归分析显示预后模型风险特征值差异有统计学意义。结论以预后免疫相关基因FGF12、REG1A、TSLP、ANGPTL7构建的免疫相关基因风险预后模型可信度较高,为儿童肾恶性横纹肌样瘤治疗提供新的视角。Objective To explore immune infiltration level of malignant rhabdoid renal tumor based upon bioinformatics to identify the genes related to immune prognosis and to construct a prognostic model.Methods The high-throughput sequencing data and clinical parameters of 50 samples of malignant rhabdoid renal tumor were obtained from the database of TARGET.Immune infiltration characteristics and clustering subtypes(Immunity_High/Low)of immune infiltration result were analyzed with"estimate"R package,"CIBERSORT"R package and ssGSEA algorithm.Differentially expressed immunity genes were obtained through differential gene analysis among clustering subtypes.Univariate COX regression and LASSO regression were utilized for screening prognosis-associated immunity genes and constructing a prognostic model.Results The group of Immunity_High had higher scores of ESTIMATE,Immune and Stromal than Immunity_Low and the highest degree of infiltration of macrophages was M2.Four prognosis-associated immunity genes(FGF12,REG1A,TSLP&ANGPTL7)were screened for constructing a prognostic risk model.Kaplan-Meier survival curve and SvivalROC prognostic model showed a high risk stratification capability.Univariate and multivariate COX regression analysis revealed statistically significant difference in risk value of prognostic models.Conclusions Based upon FGF12,REG1A,TSLP and ANGPTL7,the risk prognostic model of immune-related genes is highly reliable.It may provide new therapeutic rationales for malignant rhabdoid renal tumor in children.

关 键 词：肾肿瘤 肾横纹肌样瘤 肿瘤免疫学 浸润 预后 

分 类 号：R737.11[医药卫生—肿瘤]