MRI-Based Radiomics Nomogram: Prediction of Axillary Non-Sentinel Lymph Node Metastasis in Patients With Sentinel...

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• Overtreatment of axillary lymph node dissection (ALND) is a concern for breast cancer patients with positive sentinel lymph nodes (SLN) but negative non-SLN (NSLN).
• Accurate prediction of NSLN metastasis is crucial for optimizing surgical strategies and avoiding unnecessary morbidity.

Purpose of the Study:

• To develop and validate a magnetic resonance imaging (MRI)-based radiomics nomogram for predicting axillary NSLN metastasis in SLN-positive breast cancer patients.
• To improve personalized axillary surgical decision-making and reduce overtreatment.

Main Methods:

• Retrospective analysis of 285 patients with SLN-positive breast cancer.
• Extraction of MRI-based radiomic features and assessment of MRI morphologic findings of primary tumors and axillary lymph nodes.
• Development and comparison of four predictive models: radiomics signature, MRI-clinical nomogram, and two MRI-clinical-radiomics nomograms using 5-fold cross-validation and ROC analysis.

Main Results:

• The MRI-clinical-radiomics nomograms demonstrated high predictive performance, with the highest AUC reaching 0.868 (with CA 15-3) and 0.852 (without CA 15-3).
• Optimal predictors included DWI texture features, clinicopathologic characteristics, and MRI morphologic findings.
• The developed nomograms showed good discrimination, calibration, and clinical usefulness.

Conclusions:

• MRI-clinical-radiomics nomograms offer high predictive accuracy for axillary NSLN status in SLN-positive breast cancer patients.
• These nomograms can guide pre-operative surgical planning and personalize axillary treatment strategies.
• The findings support the use of advanced imaging techniques for optimizing breast cancer management.