Patterns and prognostic implications of distant metastasis in breast Cancer based on SEER population data

作者：Maoquan zhang, Hanqing Deng, Renling Hu, Fuwei Chen, Shiwen Dong, Shiyi Zhang, Wenyan Guo, Wenwen Yang, Wenxin Chen · 发表于：Scientific Reports · 年份：2025 · DOI：10.1038/s41598-025-12883-x · 被引用次数：33 · 研究领域：Medical Imaging Techniques and Applications、Breast Cancer Treatment Studies、Radiomics and Machine Learning in Medical Imaging

Distant metastasis remains the leading cause of mortality in breast cancer, yet comprehensive population-based evaluations of metastatic site combinations and their survival implications are limited. This study aimed to explore the clinicopathological determinants and prognostic outcomes of site-specific and multi-organ metastases in breast cancer using SEER data. A total of 200,558 female breast cancer patients diagnosed between 2014 and 2023 were extracted from the SEER database. Logistic regression was used to assess associations between clinicopathological features and metastatic patterns. Kaplan-Meier analysis and Cox proportional hazards models were applied to evaluate overall survival (OS) across different metastatic site combinations. Among patients with distant metastasis classified into 15 common metastatic patterns, bone was the most common metastatic site (21.3%), followed by lung (16.1%), liver (9.2%), and brain (2.9%). Molecular subtypes showed distinct organotropism: HR+/HER2 - tumors were prone to bone-only metastasis, whereas HER2-positive and triple-negative subtypes were more likely to involve visceral and brain metastases. Multi-organ metastases, especially combinations including the brain (e.g., brain + liver + lung), were associated with the poorest prognosis (median OS: 4.0 months). Younger age (≤ 40 years), higher histological grade (Grade III), and tumor location in the axillary tail or unspecified regions were independently associated with increased ...