兰州大学机构库 >第二临床医学院
基于代谢相关lncRNA构建膀胱癌预后预测模型的研究
Alternative TitleA Study On The Construction Of A Prognostic Prediction Model For Bladder Cancer Based On Metabolism-Related LncRNA
周安安
Subtype硕士
Thesis Advisor田俊强
2023-06-01
Degree Grantor兰州大学
Place of Conferral兰州
Degree Name医学硕士
Degree Discipline外科学
Keyword膀胱癌 Bladder cancer 代谢相关lncRNA Metabolism 预后模型 LncRNA TCGA Prognostic signature TCGA
Abstract

背景:膀胱癌(Bladder cancer,BLCA)是泌尿系统最常见的恶性肿瘤,尽管目前基于分类和分级膀胱癌拥有丰富的治疗手段,但一部分接受根治性手术的膀胱癌患者的预后依然不是非常理想。肿瘤发生过程通常伴随着代谢模式的改变,也称为代谢重编程,被广泛认为是癌细胞的新兴标志。越来越多的研究结果证明了长链非编码RNA(Long non-coding RNA,lncRNA)可能从多层面、多途径调控基因表达,进而影响肿瘤的发生、发展。

目的:通过筛选代谢相关lncRNA,并以此为基础构建预后模型,绘制列线图,评估膀胱癌患者预后。

方法:本研究从癌症基因组图谱(The cancer genome atlas, TCGA)数据库下载膀胱癌患者的RNA测序(RNA sequencing, RNA-seq)数据(包括lncRNA和mRNA表达数据)以及临床信息,并从分子特征数据库(The molecular signatures database, MSigDB)中获得代谢相关基因,通过在R语言(Version 4.0.3)中使用Pearson相关性分析得到与代谢相关的lncRNA。使用最小绝对值收敛和选择算子(Least absolute shrinkage and selection operator, LASSO)回归,以及多变量比例风险(Cox proportional hazards model, Cox模型)回归来鉴定、筛选膀胱癌中和预后相关的代谢性lncRNA,通过这些lncRNA的表达水平以及多变量Cox回归系数构建一个基于代谢性lncRNA的膀胱癌统计学预测模型,并计算每个膀胱癌患者的风险评分。通过Wilcoxon符号秩检验来评估临床病理变量与风险评分之间的相关性,并利用受试者工作特征曲线(Receiver operating characteristic curve, ROC)曲线评估该风险评分模型预测膀胱癌患者预后的准确性。结合上述预后模型以及临床病理变量,绘制预测膀胱癌预后的列线图,并通过ROC曲线和一致性指数(Concordance index, C-index)评估其预测潜力。最后利用筛选出的lncRNA,使用Cytoscape软件(Version 3.9.0)构建了lncRNA-mRNA共表达网络来区分这些lncRNA对膀胱癌的作用,并通过基因集富集分析(Gene set enrichment analysis, GSEA)揭示了该风险评分模型的潜在分子机制。

结果:本研究从TCGA数据库中下载了412例膀胱癌患者信息,从中筛选出403例具有完整生存数据的病例进行分析。从MSigDB数据库中获得了940个代谢相关基因。通过使用Pearson相关性分析、Kaplan-Meier分析、单变量Cox回归分析以及多变量Cox回归分析获得了8个与膀胱癌患者预后相关的代谢性lncRNA,并以此构建了膀胱癌预后风险评分计算模型。在测试队列验证得到这个预后模型对膀胱癌患者的预后有较好的预测能力,并在训练队列和完整队列中验证了该模型的有效性。通过将风险评分与膀胱癌患者临床病例特征对比分析,验证了风险评分是患者总生存期的一个高危因素,并且该评分可以作为膀胱癌患者的独立预后因素。同时,将预后模型联合临床病理特征(包括年龄、性别、临床分期、T、N、M分期)构建列线图,用来预测膀胱癌患者的生存概率,该列线图的ROC曲线下面积和C指数分别为0.761和0.729,表明该列线图有较好的预测能力。此外,本研究还利用Cytoscape和桑葚图可视化了筛选出的8种代谢相关lncRNA与相应的mRNA的共表达网络,发现AP002884.1和MAFG−DT是膀胱癌患者的危险因素,而其余6个代谢相关lncRNA(AC062017.1、AC073534.1、AC099518.2、AC104564.3、LINC01637、USP30−AS1)是膀胱癌患者的保护因素。最后,为了研究基于风险评分预测模型的潜在分子机制,本研究通过GSEA识别高危和低危组膀胱癌患者的京都基因和基因组百科全书(Kyoto encyclopedia of genes and genomes, KEEG)通路,结果发现癌症和代谢相关的途径在高危组中富集。

结论:我们以8个代谢相关lncRNA为基础构建的风险评分预测模型能够可靠地预测膀胱癌患者的预后,并有可能指导临床决策。代谢相关lncRNA有望成为膀胱癌患者诊断与预后的新的生物标志物。

Other Abstract

Background: Bladder cancer (BLCA) is the most common urological tumor. Although there are currently abundant treatment methods for bladder cancer based on classification and grading, the prognosis of some patients undergoing radical surgery for bladder cancer is still unsatisfactory. An increasing number of research results have proved that long non-coding RNA (lncRNA) can influence the tumorigenesis and tumor development by regulating gene expression from multiple levels and ways. Generally, the tumorigenesis process is accompanied by changes in metabolic patterns, known as metabolic reprogramming, which is widely regarded as an emerging hallmark of cancer cells now.

Objective: we constructed a metabolic-related lncRNA prognostic model and drew a nomogram to evaluate the prognosis of bladder cancer patients.

Methods: We downloaded RNA-Seq data (including expression data of lncRNA and mRNA) as well as clinical information of BLCA patients from TCGA database. The metabolism-related genes were obtained from the Molecular Signatures Database (MSigDB). Using Pearson correlation analysis obtained metabolism-related lncRNAs in R (Version 4.0.3). Cox regression and least absolute shrinkage and selection operator (LASSO) regression were utilized to identify metabolism-related lncRNAs related to prognosis in BLCA. Cox regression and LASSO were utilized to identify metabolism-related lncRNAs related to prognosis. Then, a prognostic signature based on the expression levels ofmetabolism-related lncRNAs and multivariate Cox regression coefficient was developed, and each patient's risk score was calculated. The Wilcoxon signed-rank test was to assess the correlation between clinicopathological variables and risk score. We further constructed the prognostic nomogram and evaluated the predictive ability by ROC curve and C-index. Finally, we utilized Cytoscape (Version 3.9.0) to construct a lncRNA-mRNA co-expression network to distinguish the effects of these lncRNAs on bladder cancer and performed GSEA to reveal the potential molecular mechanism.

Result: In this study we identified 8 metabolism-related lncRNA (AC062017.1, AC073534.1, AC099518.2, AC104564.3, AP002884.1, LINC01637, MAFG-DT, USP30-AS1), and then constructed a prognostic signature. In both the TCGA training cohort and TCGA whole cohorts, the metabolism-related lncRNA prognostic signature was verified. The risk score based on prognostic signature was substantially correlated with advanced clinical stage, T stage, N stage, and M stage. The prognostic signature accurately categorized BLCA patients into high- and low-risk groups by median risk score, and it was an independent predictor for the prognosis of BLCA. Then, using metabolism-related lncRNA prognostic signature and clinicopathologic variables to construct a nomogram, which has high predictive accuracy (area under ROC curve and C-index of 0.761 and 0.729, respectively) and the ability to accurately predict 1-year, 3-year, and 5-year survival probability of BLCA patients. Furthermore, we built an mRNA-lncRNA co-expression network with 14 mRNA-lncRNA pairs, AP002884.1 and MAFG−DT were found to be risk factors for bladder cancer patients, while the remaining six metabolic-related lncRNAs were protective factors for bladder cancer patients. Finally, we utilized GSEA and discovered that the pathways related to cancer and metabolism were enriched in high-risk BLCA patients.

Conclusion: 8-lncRNA prognostic signature can reliably predict the prognosis of BCLA patients and guide clinical decisions, and metabolism-related lncRNAs are promising to act as novel diagnostic biomarkers for BCLA patients.

Subject Area泌尿外科
MOST Discipline Catalogue医学 - 临床医学 - 外科学(含:普外、骨外、泌尿外、胸心外、神外、整形、烧伤、野战外)
URL查看原文
Language中文
Other Code262010_220200904911
Document Type学位论文
Identifierhttps://ir.lzu.edu.cn/handle/262010/537773
Collection第二临床医学院
Affiliation
兰州大学第二临床医学院
Recommended Citation
GB/T 7714
周安安. 基于代谢相关lncRNA构建膀胱癌预后预测模型的研究[D]. 兰州. 兰州大学,2023.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Altmetrics Score
Google Scholar
Similar articles in Google Scholar
[周安安]'s Articles
Baidu academic
Similar articles in Baidu academic
[周安安]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[周安安]'s Articles
Terms of Use
No data!
Social Bookmark/Share
No comment.
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.