摘要
目的:挖掘浸润性乳腺癌复发的独立预后基因并构建浸润性乳腺癌复发风险预测模型.方法:应用公共数据库,使用一致性聚类,根据肿瘤血管生成相关基因表达将样本分组并进行差异分析.通过单因素Cox回归和最小绝对收缩和选择算子(Lasso)回归,以无复发生存期(RFS)为观测指标,结合患者临床特征,建立浸润性乳腺癌复发风险预测模型.结果:鉴定了 13个基于肿瘤血管生成的复发独立预后基因(PLS3、IGFBP4、CXCL14、HIST1H2BH、EMC9、H2BFS、S100A9、GJ A1、NID2、ID3、PDZD2、GRP、FMO1),并结合患者 TNM 分期建立了浸润性乳腺癌复发风险预测模型.结论:基于肿瘤血管生成的13个基因复发风险预测模型具有良好的预测性能,可以准确预测浸润性乳腺癌患者的复发风险.
Abstract
Objective:To identify independent prognostic genes for the recurrence of invasive breast cancer and construct a risk prediction model for recurrent invasive breast cancer.Method:Public databases were applied to group samples according to angiogenesis-related gene expression using consistent clustering and performed differen-tial analysis.Differential genes were screened by univariate Cox regression and the least absolute shrinkage and se-lection operator(Lasso)regression,and a recurrence risk prediction model for invasive breast cancer was construc-ted by combining the clinical characteristics of the patients with the recurrence-free survival(RFS)as an observa-tional index.Results:13 angiogenesis-based recurrence-independent prognostic genes(PLS3,IGFBP4,CXCL14,HIST1H2BH,EMC9,H2BFS,S100A9,GJA1,NID2,ID3,PDZD2,GRP,FMO1)were identified.The invasive breast cancer recurrence risk prediction model was established in conjunction with patients'TNM stage.Conclusion:The 13-gene recurrence risk prediction model based on angiogenesis has promising predictive performance and can accurately predict the recurrence risk of invasive breast cancer patients.
基金项目
武汉大学中南医院转化医学及交叉学科研究联合基金(ZNJC202236)
NETL
NSTL
万方数据