Combining Intratumoral and Peritumoral Multimodal MRI to Predict the HER-2 Expression Level in Breast Cancer

Motivation: Understanding the prognostic factors in breast cancer is crucial for improving clinical diagnosis and treatment. Goal(s): This study aims to evaluate the predictive value of multimodal MRI parameters for determining human epidermal growth factor receptor 2 (HER-2) status in breast cancer. Approach: A retrospective analysis was conducted using multimodal MRI techniques, including DCE-MRI, MUSE-IVIM, IDEAL-IQ, and MAGiC, on patients diagnosed with breast cancer. Quantitative parameters were extracted and analyzed. Results: Significant differences in MRI parameters between HER-2 positive and negative groups were found, with combined models showing high diagnostic performance (AUC=0.822) for predicting HER-2 expression. Impact: This study's findings could enhance non-invasive HER-2 status assessment in breast cancer, aiding clinicians in personalized treatment planning. It opens avenues for further research on multimodal MRI's role in predicting other molecular characteristics, improving diagnostic precision.