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:
Online Access:https://doi.org/10.1186/s12880-025-01890-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849225867358633984
author Yunqing Yang
Zhulin Wang
Haidong Wang
Yun Wang
Yang Fu
author_facet Yunqing Yang
Zhulin Wang
Haidong Wang
Yun Wang
Yang Fu
author_sort Yunqing Yang
collection DOAJ
description 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.
format Article
id doaj-art-c6cc8633c236404badd03898aae8b311
institution Kabale University
issn 1471-2342
language English
publishDate 2025-08-01
publisher BMC
record_format Article
series BMC Medical Imaging
spelling doaj-art-c6cc8633c236404badd03898aae8b3112025-08-24T11:57:35ZengBMCBMC Medical Imaging1471-23422025-08-012511910.1186/s12880-025-01890-zA non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRIYunqing Yang0Zhulin Wang1Haidong Wang2Yun Wang3Yang Fu4Department of Breast Surgery, The First Affiliated Hospital of Zhengzhou UniversityDepartment of Thoracic Surgery, Affiliated Hospital of Southwest Medical UniversityDepartment of General Surgery, Anyang Dengta HospitalDepartment of Magnetic Resonance Imaging, The First Affiliated Hospital of Zhengzhou UniversityDepartment of Gastrointestinal Surgery, The First Affiliated Hospital of Zhengzhou UniversityAbstract 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.https://doi.org/10.1186/s12880-025-01890-zBreast cancerSentinel lymph nodeDynamic contrast-enhanced MRI (DCE-MRI)NomogramDecision curve analysisRecursive partitioning analysis
spellingShingle Yunqing Yang
Zhulin Wang
Haidong Wang
Yun Wang
Yang Fu
A non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRI
BMC Medical Imaging
Breast cancer
Sentinel lymph node
Dynamic contrast-enhanced MRI (DCE-MRI)
Nomogram
Decision curve analysis
Recursive partitioning analysis
title A non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRI
title_full A non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRI
title_fullStr A non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRI
title_full_unstemmed A non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRI
title_short A non-invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and MRI
title_sort non invasive preoperative model for predicting sentinel lymph node metastasis in breast cancer using clinical data and mri
topic Breast cancer
Sentinel lymph node
Dynamic contrast-enhanced MRI (DCE-MRI)
Nomogram
Decision curve analysis
Recursive partitioning analysis
url https://doi.org/10.1186/s12880-025-01890-z
work_keys_str_mv AT yunqingyang anoninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT zhulinwang anoninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT haidongwang anoninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT yunwang anoninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT yangfu anoninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT yunqingyang noninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT zhulinwang noninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT haidongwang noninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT yunwang noninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri
AT yangfu noninvasivepreoperativemodelforpredictingsentinellymphnodemetastasisinbreastcancerusingclinicaldataandmri