Clinical Report: Machine Learning Prognostic Model for Metastatic HR-Positive Breast Cancer

Overview

This study investigates the prognostic value of postoperative radiotherapy (RT) for overall survival (OS) in patients with de novo metastatic hormone receptor (HR)-positive breast cancer. A machine learning-based prognostic model was developed and validated.

Background

Breast cancer is the most prevalent malignancy among women, with HR-positive tumors accounting for over 70% of cases. Patients with de novo metastatic HR-positive breast cancer have a significantly worse prognosis compared to those with early-stage disease. The role of postoperative radiotherapy in improving survival outcomes remains controversial, as some studies suggest it does not improve OS in the context of effective systemic therapy.

Data Highlights

Key Findings

• Postoperative RT was associated with improved OS in patients with de novo metastatic HR-positive breast cancer.
• The LR model outperformed other machine learning models with an AUC of 0.721.
• Multivariate analysis identified RT as an independent factor associated with OS.
• The study included a training cohort of 2,266 patients and an external validation cohort of 79 patients.
• The predicted outcomes reflect 3-year prognostic risk rather than causal benefits of RT.

Clinical Implications

The findings indicate the need for further investigation into the role of postoperative RT in prognostic assessments for patients with de novo metastatic HR-positive breast cancer.

Conclusion

Postoperative RT is identified as a factor associated with OS in this patient population. The developed prognostic model provides estimates based on observed data.

Related Resources & Content

1. Li et al., Frontiers in Oncology, 2026 -- Development and validation of a machine learning model to evaluate survival in patients with newly diagnosed breast cancer with liver metastasis
2. Li et al., Frontiers in Oncology, 2026 -- Development and validation of an interpretable machine learning-based predictive model for breast cancer bone metastasis
3. Li et al., Frontiers in Medicine, 2026 -- Development and validation of an interpretable machine learning model for predicting 5-year recurrence in breast cancer
4. ASCO, Journal of Clinical Oncology, 2025 -- Endocrine and Targeted Therapy for Hormone Receptor–Positive, Human Epidermal Growth Factor Receptor 2–Negative Metastatic Breast Cancer—Capivasertib-Fulvestrant: ASCO Rapid Recommendation Update
5. ECOG-ACRIN, Journal of Clinical Oncology, 2021 -- Early Local Therapy for the Primary Site in De Novo Stage IV Breast Cancer: Results of a Randomized Clinical Trial (E2108)
6. the asco post — External Validation Confirms Ability of AI Model to Stratify Recurrence Risk in Early-Stage Lung Cancer
7. Machine Learning-Based Prognostic Model for De Novo Metastatic HR-Positive Breast Cancer
8. Endocrine and Targeted Therapy for Hormone Receptor–Positive, Human Epidermal Growth Factor Receptor 2–Negative Metastatic Breast Cancer—Capivasertib-Fulvestrant: ASCO Rapid Recommendation Update | Journal of Clinical Oncology
9. Early Local Therapy for the Primary Site in De Novo Stage IV Breast Cancer: Results of a Randomized Clinical Trial (E2108) | Journal of Clinical Oncology
10. Study Results | NCT02364557 | Testing Whether Treating Breast Cancer Metastases With Surgery or High-Dose Radiation Improves Survival | ClinicalTrials.gov
11. AGO Recommendations for the Diagnosis and Treatment of Patients with Locally Advanced and Metastatic Breast Cancer: Update 2025 - PMC
12. Primary Surgery with Systemic Therapy in Patients with de Novo Stage IV Breast Cancer: 10-year Follow-up; Protocol MF07-01 Randomized Clinical Trial
13. Options for postoperative radiation therapy in patients with de novo metastatic breast cancer - ScienceDirect