皮肤黑素瘤缺氧相关lncRNA预后风险评分模型的构建  

Construction of Hypoxia-Related lncRNA Prognostic Risk Score Model for Cutaneous Melanoma

作　　者：汤世崇 尹锐 TANG Shichong;YIN Rui(Department of Dermatology,the First Affiliated Hospital of Army Medical University(the Third Military Medical University),Chongqing 400038,China)

机构地区：[1]陆军军医大学(第三军医大学)第一附属医院皮肤科,重庆400038

出　　处：《中国皮肤性病学杂志》2023年第8期893-901,共9页The Chinese Journal of Dermatovenereology

摘　　要：目的 挖掘皮肤黑素瘤(cutaneous melanoma,CM)中与缺氧相关的长非编码RNA(hypoxia-related lncRNA,HYlncRNA),筛选潜在的预后生物标志物,构建预后风险评分模型。方法 从TCGA和GTEx数据库下载皮肤黑素瘤和正常皮肤的转录组数据和相应临床数据,从MsigDB数据库下载缺氧相关基因(hypoxia-related genes,HYGs)数据集。通过差异性分析和相关性分析筛选差异表达的缺氧相关lncRNAs(differentially expressed hypoxia-related lncRNAs,DEHYlncRNAs)。使用单因素COX和LASSO回归分析构建预后风险评分模型。应用单因素和多因素COX回归分析筛选皮肤黑素瘤的独立预后因素,并进一步构建诺莫图(nomogram)。分析皮肤黑素瘤高低风险组之间免疫细胞浸润及免疫功能的差异。结果 共筛选出19个DEHY-lncRNAs构建预后风险评分模型。风险模型图和Kaplan-Meier分析显示模型中高风险组患者的生存时间显著短于低风险组(P<0.05),且模型在验证集和整体数据集中均表现出良好的预测效能。COX回归分析表明风险评分是皮肤黑素瘤的独立预后因素(P<0.01)。皮肤黑素瘤高风险组与肿瘤免疫细胞浸润呈负相关。结论 本研究认为19个DEHYlncRNAs可能成为皮肤黑素瘤潜在的预后生物标记物,成功构建的预后风险评分模型能够帮助提高预测皮肤黑素瘤患者预后的水平。Objective To explore hypoxia-related long non-coding RNA(HYlncRNA) in cutaneous melanoma(CM),screen potential prognostic biomarkers,and construct a prognostic risk score model.Methods RNA-Seq data and corresponding clinical data of CM and normal skin were downloaded from TCGA and GTEx databases,and hypoxia-related gene data set from MsigDB database.The differentially expressed hypoxia-related lncRNAs(DEHYlncRNAs) were identified by difference and correlation analysis.Univariate COX and LASSO regression analysis were used to construct the prognostic risk score model.Then,we obtained the independent prognostic factors of CM based on univariate and multivariate COX regression analysis,with the Nomogram plotted subsequently.At the end,we analyzed the differences of immune cell infiltration and immune function between high and low risk groups of CM.Results A total of 19 DEHYlncRNAs were identified to construct the prognostic risk score model.Risk model plot and Kaplan-Meier analysis showed that the survival time of CM patients in the high risk group was significantly shorter than that in the low risk group(P<0.05),and this model demonstrated good prediction efficiency for CM in both the validation set and the overall dataset.COX regression analysis indicated that risk score was an independent prognostic factor for CM(P<0.01).The high risk group of CM was negatively correlated with tumor immune cell infiltration.Conclusion This study suggests that these 19 DEHYlncRNAs may be potential prognostic biomarkers of CM,and the prognostic risk score model constructed successfully contributes to better predicting the prognosis of CM patients.

关 键 词：黑素瘤 皮肤 长非编码RNA 生物信息学 预后模型 免疫浸润 

分 类 号：R739.5[医药卫生—肿瘤]