Prognostic Model of ER-positive, HER2-negative Breast Cancer Predicted by Clinically Relevant Indicators

Purpose: To study the clinicopathological variables connected with disease-free survival (DFS) as well as overall survival (OS) in patients who are ER-positive or HER2-negative and to propose nomograms for predicting individual risk.

Methods: In this investigation, we examined 585 (development cohort) and 291 (external validation) ER-positive, HER2-negative breast cancer patients from January 2010 to January 2014. From January 2010 to December 2014, we retrospectively reviewed and analyzed 291 (external validation) and 585 (development cohort) HER2-negative, ER-positive breast cancer patients. Cox regression analysis, both multivariate and univariate, confirmed the independence indicators for OS and DFS.

Results: Using cox regression analysis, both multivariate and univariate, the following variables were combined to predict the DFS of development cohort: pathological stage (HR = 1.391; 95% CI = 1.043-1.855; P value = 0.025), luminal parting (HR = 1.836; 95% CI = 1.142-2.952; P value = .012), and clinical stage (HR = 1.879; 95% CI = 1.102-3.203; P value = 0.021). Endocrine therapy (HR = 3.655; 95% CI = 1.084-12.324; P value = 0.037) and clinical stage (HR = 6.792; 95% CI = 1.672-28.345; P value = 0.009) were chosen as predictors of OS. Furthermore, we generated RS-OS and RS-DFS. According to the findings of Kaplan-Meier curves, patients who are classified as having a low risk have considerably longer DFS and OS durations than patients who are classified as having a high risk.

Conclusion: To generate nomograms that predicted DFS and OS, independent predictors of DFS in ER-positive/HER2-negative breast cancer patients were chosen. The nomograms successfully stratified patients into prognostic categories and worked well in both internal validation and external validation.