Construction and evaluation of prognostic nomogram of breast cancer based on risk scoring model of endoplasmic reticulum stress related genes
10.3969/j.issn.1009-9905.2024.12.002
- VernacularTitle:基于内质网应激相关基因风险评分模型的乳腺癌预后列线图的构建与评估
- Author:
Jia-wei PEI
1
;
Xiao HUANG
;
De-yuan FU
Author Information
1. 大连医科大学(辽宁 大连 116000)
- Publication Type:Journal Article
- Keywords:
Breast cancer;
ERSRGs;
Prognosis;
Risk scoring model
- From:
Chinese Journal of Current Advances in General Surgery
2024;27(12):932-937
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the potential value of endoplasmic reticulum stress-related genes(ERSRGs)in assessing the prognosis of breast cancer patients,and to construct a risk scoring model and a prognostic nomogram.Methods:ERSRGs were retrieved from the Genecards data-base,and three machine learning algorithms,namely"LASSO","RF",and"Xgboost",were used in the TCGA database to screen for core prognostic genes and establish a corresponding risk scoring model.The model was externally validated using the GSE20685 and GSE86166 datasets.Based on the risk scoring model,nomograms were further constructed to predict the 3-year and 5-year overall survival(OS)rates of breast cancer patients.The expression of the model genes at the pro-tein and single-cell levels was assessed through the Human Protein Atlas(HPA)and a single-cell sequencing dataset(GSE176078).Results:A total of 1301 ERSRGs were screened from the Genecards database.Among them,237 ERSRGs were differentially expressed between breast cancer and normal tissues.Univariate Cox regression analysis revealed that 22 ERSRGs were as-sociated with the prognosis of breast cancer patients.Through the three machine learning algo-rithms of LASSO,RF,and Xgboost,five ERSRGs(MAP2K6,IFNG,VIM,TH,CEMIP)were identi-fied as core genes related to the prognosis of breast cancer,and an ERSRGs-related risk scoring model and prognostic nomogram were constructed.Both internal and external dataset validations showed significant differences among patients in different risk groups.The C-index,ROC curve,and calibration curve all indicated that the nomogram had good predictive performance.Conclu-sion:In this study,the nomogram constructed based on the ERSRGS-related risk score model can accurately predict the prognosis of breast cancer patients,which can provide certain reference for clinical diagnosis,treatment and the development of new tumor therapeutic targets.
