Analysis of prognostic factors and construction of prognostic prediction models for hormone receptor-positive/human epidermal growth factor receptor 2-positive breast cancer
10.3760/cma.j.cn115355-20240307-00107
- VernacularTitle:激素受体阳性/人表皮生长因子受体2阳性乳腺癌预后因素分析及预后预测模型的构建
- Author:
Xinli WANG
1
;
Yan XUE
1
Author Information
1. 西安国际医学中心医院肿瘤科，西安 710100
- Publication Type:Journal Article
- Keywords:
Breast neoplasms;
Receptor, erbB-2;
Prognosis;
Nomograms
- From:
Cancer Research and Clinic
2025;37(5):321-327
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the clinicopathological characteristics and prognostic factors of hormone receptor-positive/human epidermal growth factor receptor 2-positive (HR +/HER2 +) breast cancer, and to construct a nomogram model for predicting the prognosis of patients. Methods:SEER*Stat 8.3.5 software was used to screen out the data of 11 664 HR +/HER2 + and 5,054 HR -/HER2 + breast cancer patients from 2010 to 2015 from the Surveillance, Epidemiology, and End Results (SEER) database. The clinicopathological data of the both groups were compared and the overall survival of the both groups was analyzed by using the Kaplan-Meier method and log-rank test was also performed; the factors influencing the overall survival in HR +/HER2 + patients with breast cancer were analyzed by using the Cox proportional hazards model. According to multivariate Cox regression analysis results, the median of the probability values of combined variables (linear predictive values) was calculated by using MedCalc software, with ≥ the median classified as high-risk group and < the median as low-risk group; the risk factor affinity diagrams were constructed by using R4.0.3 software. Based on the independent influencing factors of the overall survival in HR +/HER2 + breast cancer patients, a nomogram model to predict the 1-, 3-, 5-year overall survival of patients was established; the receiver operating characteristic (ROC) curve and calibration curve were plotted, and the predictive efficacy of the model was verified. Results:The differences in the proportions of HR +/HER2 + and HR -/HER2 + breast cancer patients with different age, race, histological grade, surgical method, whether receiving radiotherapy or chemotherapy, T stage, N stage, and M stage were statistically significant (all P < 0.05). The overall survival of HR +/HER2 + patients was better than that of HR -/HER2 + patients ( P < 0.001). Cox regression analysis showed that age, race, histological grade, surgical method, whether receiving radiotherapy or chemotherapy, first primary tumor or not, T stage, N stage, and M stage were independent influencing factors of overall survival in HR +/HER2 + patients (all P < 0.05). MedCalc software was used to obtain the median of the linear predictive values (0.394), with ≥ 0.394 classified as the high-risk group (5 838 cases) and < 0.394 as the low-risk group (5 826 cases). The risk factor affinity diagrams showed that all risk factors were more highly expressed in high-risk group compared with those in low-risk group; the overall survival of patients in high-risk group was poorer than that in low-risk group ( P < 0.001). The area under the ROC curve of a nomogram model for predicting the 1-, 3-, and 5-year overall survival rates was 0.900, 0.760, 0.760, respectively; and C index was 0.799, and the calibration curve was highly matched with the ideal curve. Conclusions:The prognosis of HR +/HER2 + patients is better than that of HR -/HER2 + patients. The constructed risk factor affinity diagrams and nomograms can well predict the prognosis of HR +/HER2 + breast cancer patients, which have a high clinical reference value.
