# A Priori Prediction of Neoadjuvant Chemotherapy Response in Breast Cancer Using Deep Features from Pre-Treatment MRI and CT

## TL;DR

This study uses deep learning on MRI and CT scans to predict how breast cancer patients will respond to chemotherapy before treatment begins, potentially allowing for more personalized treatment plans.

## Contribution

The novel use of deep learning features from pre-treatment MRI and CT scans to predict neoadjuvant chemotherapy response in breast cancer.

## Key findings

- ResNet34 deep features outperformed handcrafted radiomic features in predicting pathologic complete response and treatment response.
- Deep learning-based models achieved balanced accuracies of 81.6% and 73.5% for predicting pCR and responders, respectively.
- Combining deep features with clinical data improved prediction of chemotherapy response in breast cancer.

## Abstract

Early identification of breast cancer patients who are unlikely to respond to neoadjuvant chemotherapy (NAC) is critical for potentially guiding alternative therapeutic strategies. In this study, routinely acquired pre-treatment MRI and CT scans were analyzed using deep learning-based features in combination with clinical information. Deep features were extracted from intratumoral and peritumoral regions using ResNet architectures pre-trained on large-scale medical imaging datasets. Among the models tested, ResNet34 demonstrated the best performance, exceeding both handcrafted radiomic models and other ResNet backbones. These findings suggest that deep features extracted from standard-of-care imaging can complement established clinical predictors and may facilitate more personalized treatment planning in breast cancer.

Background: Response to neoadjuvant chemotherapy (NAC) is a key prognostic indicator in breast cancer, yet current assessment relies on postoperative pathology. This study investigated the use of deep features derived from pre-treatment MRI and CT scans, in conjunction with clinical variables, to predict treatment response a priori. Methods: Two response endpoints were analyzed: pathologic complete response (pCR) versus non-pCR, and responders versus non-responders, with response defined as a reduction in tumor size of at least 30%. Intratumoral and peritumoral segmentations were generated on contrast-enhanced T1-weighted (CE-T1) and T2-weighted MRI, as well as contrast-enhanced CT images of tumors. Deep features were extracted from these regions using ResNet10, ResNet18, ResNet34, and ResNet50 architectures pre-trained with MedicalNet. Handcrafted radiomic features were also extracted for comparison. Feature selection was conducted with minimum redundancy maximum relevance (mRMR) followed by recursive feature elimination (RFE), and classification was performed using XGBoost across ten independent data partitions. Results: A total of 177 patients were analyzed in this study. ResNet34-derived features achieved the highest overall classification performance under both criteria, outperforming handcrafted features and deep features from other ResNet architectures. For distinguishing pCR from non-pCR, ResNet34 achieved a balanced accuracy of 81.6%, whereas handcrafted radiomics achieved 77.9%. For distinguishing responders from non-responders, ResNet34 achieved a balanced accuracy of 73.5%, compared with 70.2% for handcrafted radiomics. Conclusions: Deep features extracted from routinely acquired MRI and CT, when combined with clinical information, improve the prediction of NAC response in breast cancer. This multimodal framework demonstrates the value of deep learning-based approaches as a complement to handcrafted radiomics and provides a basis for more individualized treatment strategies.