To investigate the prognostic value of postoperative radiotherapy (RT) for overall survival (OS) in patients with de novo metastatic hormone receptor (HR)-positive breast cancer, and to develop and externally validate a machine learning-based prognostic prediction model.
Approach:
Study Design: Retrospective cohort study using SEER data and an independent external validation cohort from Guangxi Medical University Cancer Hospital.
Data Analysis: Kaplan–Meier analysis and log-rank test for OS comparison; multivariate Cox regression for identifying independent prognostic factors; construction of four machine learning models (KNN, LR, RF, XGBoost) for predicting 3-year OS.
Key Findings:
The RT group showed significantly superior OS compared to the non-RT group (HR = 0.52, P < 0.001).
RT was confirmed as an independent protective factor for OS (HR = 0.657, P < 0.001), validated in the external cohort (HR = 0.171, P = 0.015).
The LR model outperformed others with an AUC of 0.721 and an AP of 0.429.
Interpretation:
Postoperative RT is an independent protective factor for OS in patients with de novo metastatic HR-positive breast cancer. The LR-based model provides reliable prognostic risk estimates.
Limitations:
The model predicts observed survival risk, not the causal benefit of RT.
The study is retrospective and may have inherent biases.
Conclusion:
The LR-based prognostic model should not be used as a standalone tool for treatment decisions.
