Deep learning predicts breast cancer marker from brain MRI scans
Deep Learning-Based Prediction of HER2 Status from Whole-Brain MRI in Breast Cancer Brain Metastases
Machine Learning
Summary
Knowing whether a breast cancer has a specific marker called HER2 helps doctors decide on the best treatment. The authors tested if computers can look at MRI brain scans of breast cancer spread and guess the HER2 status. They found that deep learning models can predict HER2 better than older methods but not very accurately. Adding information about where the tumor is in the brain improved predictions only a little. This means MRI images alone are not enough to reliably tell HER2 status with current methods, but the approach shows some promise.
What this means in practice
- •For medical imaging software teams: Integrate whole-brain MRI analysis using deep learning to support non-invasive estimation of HER2 status in breast cancer brain metastases.
- •For oncology data science teams: Add tumor segmentation masks to MRI-based deep learning models to slightly improve HER2 status predictions in metastatic breast cancer.
Tested on one dataset.
Authors
Sara Early, Chris Kang, Vibujithan Vigneshwaran, Kimberly Amador, Roger Y. Tsang, Nils D. Forkert
Abstract
Purpose: Human epidermal growth factor receptor 2 (HER2) status is a key biomarker relevant for prognostication and treatment decision making in metastatic breast cancer. Thus, the aim of this work was to explore whether HER2 status can be predicted directly from whole-brain post-contrast T1-weighted (T1w) MRI scans in breast cancer brain metastases (BCBM) using deep learning, and whether providing explicit tumour segmentation information improves prediction. Methods: Using the publicly available BCBM-Radiogenomics dataset, we analysed 135 patients, including 70 HER2+ and 65 HER2- patients. Three approaches were implemented and evaluated: a conventional radiomics model, a simple fully convolutional network (SFCN) using T1w MRI scans alone, and an SFCN using T1w MRI scans with an expert-derived tumour mask as an additional input channel. Results: The radiomics model achieved an area under the receiver operating characteristic curve (ROC-AUC) of 0.602 (95% confidence interval [CI]: 0.528--0.677) while the SFCN achieved a ROC-AUC of 0.667 (95% CI, 0.63--0.71) using T1w MRI scans alone, which increased to 0.692 (95% CI, 0.58--0.81) when an explicit tumour mask was additionally provided. Conclusion: However, the modest performance across all approaches demonstrates the difficulty of predicting HER2 status from structural MRI scans alone. The difference between the segmentation-free and mask-guided SFCN models suggests that much of the predictive signal captured by the deep-learning model may be learned directly from whole-brain T1w MRI without requiring explicit tumour delineation. These findings support further investigation of segmentation-free deep-learning approaches as a potential non-invasive strategy to support HER2 status determination in BCBM.