Non-invasive prediction model of axillary lymph node status in patients with early-stage breast cancer: a feasibility study based on dynamic contrast-enhanced-MRI radiomics

Weiyue Chen ^1,2,3, Guihan Lin ^1,2,3, Chunli Kong ^1,2,3

⁰Key Laboratory of Imaging Diagnosis and Minimally Invasive Intervention Research, Institute of Imaging Diagnosis and Minimally Invasive Intervention Research, The Fifth Affiliated Hospital of Wenzhou Medical University, Lishui 323000, China.

The British Journal of Radiology +

|

February 2, 2024

View abstract on PubMed

Summary

This study developed a non-invasive radiomics model using dynamic contrast-enhanced MRI to accurately assess axillary lymph node status in early breast cancer patients, aiding personalized treatment.

Area Of Science

Oncology
Radiology
Medical Imaging

Background

Accurate axillary lymph node (ALN) evaluation is crucial for breast cancer prognosis and treatment planning.
Current methods may be invasive or lack precision in preoperative staging.

Purpose Of The Study

To develop and validate a dynamic contrast-enhanced MRI (DCE-MRI)-based radiomics model for preoperative ALN status evaluation in early-stage breast cancer.
To assess the model's ability to predict ALN metastasis and burden.

Main Methods

Retrospective analysis of 410 early-stage breast cancer patients.
Radiomics features extracted from DCE-MRI images.
Construction of a radiomics signature and a combined nomogram incorporating clinical predictors.

Main Results

A radiomics signature comprising 14 features was developed.
The nomogram, incorporating radiomics score and clinical factors, demonstrated strong performance in discriminating metastatic from non-metastatic ALNs (AUCs 0.859-0.881).
The model accurately differentiated between high and low ALN metastatic burden.

Conclusions

A DCE-MRI-based radiomics nomogram offers a potential non-invasive tool for accurate preoperative ALN burden assessment.
This approach can assist in personalized axillary treatment strategies for early breast cancer.
The developed model is non-invasive and user-friendly for preoperative evaluation.

Keywords:

axillary lymph node bagnetic resonance imaging breast cancer nomogram radiomics