A non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRI

A non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRI

Abstract Introduction Breast cancer is the leading cause of cancer-related death among women, with metastasis accounting for the majority of these deaths. Sentinel lymph node (SLN) status is crucial for staging and treatment planning. This study aims to develop a non-invasive preoperative model for...

Full description

Saved in:
Bibliographic Details
Main Authors: Yunqing Yang, Zhulin Wang, Haidong Wang, Yun Wang, Yang Fu
Format: Article
Language:English
Published: BMC 2025-08-01
Series:BMC Medical Imaging
Subjects:
Breast cancer
Sentinel lymph node
Dynamic contrast-enhanced MRI (DCE-MRI)
Nomogram
Decision curve analysis
Recursive partitioning analysis
Online Access:https://doi.org/10.1186/s12880-025-01890-z
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Introduction Breast cancer is the leading cause of cancer-related death among women, with metastasis accounting for the majority of these deaths. Sentinel lymph node (SLN) status is crucial for staging and treatment planning. This study aims to develop a non-invasive preoperative model for predicting SLN metastasis using clinical data and preoperative MRI. Methods A retrospective study included 4,276 breast cancer patients who underwent surgery were enrolled. After exclusions, 999 patients were analyzed. Univariable and multivariable logistic regression identified significant predictors of SLN metastasis, which were used to construct nomograms. Calibration curves and decision curve analysis (DCA) validated the model’s accuracy. Recursive partitioning analysis (RPA) was used to create a risk stratification system. Results Significant predictors of SLN metastasis included tumor size on MRI, multifocality, MRI-BIRADS classification, ADC value, short axis, and cortical thickness (P < 0.05). The nomogram showed excellent discriminatory power with an AUC of 0.847. The RPA stratified patients into low-, intermediate-, and high-risk groups, with respective SLN metastasis probabilities of 15.8%, 28.6%, and 69.8%. Conclusions This non-invasive SLN metastasis prediction model and risk stratification system provide a valuable tool for personalized clinical decision-making, potentially reducing the need for SLN biopsy in low-risk patients. Further studies are needed to validate these findings.
ISSN:1471-2342

Similar Items