Radiomics Nomogram Based on Multiparametric MRI for Predicting the Hormone Receptor Status of HER2-Low Expression...

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• Hormone receptor (HR) status is critical for breast cancer treatment decisions.
• HER2-low breast cancer represents a distinct subtype requiring specific diagnostic approaches.
• Accurate prediction of HR status in HER2-low breast cancer is essential for personalized therapy.

Purpose of the Study:

• To develop and validate a multiparametric magnetic resonance imaging (mpMRI)-based radiomics nomogram.
• To predict the hormone receptor (HR) status in patients with HER2-low breast cancer.
• To evaluate the predictive performance of the developed nomogram compared to existing models.

Main Methods:

• Retrospective analysis of 198 HER2-low breast cancer patients who underwent mpMRI.
• Extraction of radiomics features from T2WI, DWI, and DCE-MRI sequences.
• Construction of clinical-radiological, radiomics, and combined mpMRI radiomics nomogram models.
• Validation using receiver operating characteristic (ROC) curve analysis.

Main Results:

• The mpMRI radiomics nomogram achieved high predictive performance (AUC=0.957 in training, 0.891 in testing sets).
• The nomogram outperformed single-modality radiomics models and clinical-radiological models.
• Key predictors included ADC value, T2SI ratio, and enhancement pattern.

Conclusions:

• An mpMRI-based radiomics nomogram effectively predicts HR status in HER2-low breast cancer.
• The nomogram offers a non-invasive tool for assessing HR status, aiding treatment stratification.
• This approach demonstrates significant potential for improving diagnostic accuracy in HER2-low breast cancer management.