AUTHOR=Huang Yuhong , Chen Wenben , Zhang Xiaoling , He Shaofu , Shao Nan , Shi Huijuan , Lin Zhenzhe , Wu Xueting , Li Tongkeng , Lin Haotian , Lin Ying 

TITLE=Prediction of Tumor Shrinkage Pattern to Neoadjuvant Chemotherapy Using a Multiparametric MRI-Based Machine Learning Model in Patients With Breast Cancer

JOURNAL=Frontiers in Bioengineering and Biotechnology

VOLUME=Volume 9 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2021.662749

DOI=10.3389/fbioe.2021.662749

ISSN=2296-4185

ABSTRACT=Purpose: After neoadjuvant chemotherapy (NACT), tumor shrinkage pattern is a more reasonable outcome to decide a possible breast-conserving surgery (BCS) than pathological complete response (pCR). We aim to establish a machine learning model combining radiomics features from multi-parametric magnetic resonance imaging (mpMRI) and clinicopathologic characteristics, for early prediction of tumor shrinkage pattern before NACT in breast cancer.
Materials and Methods: This study included 199 breast cancer patients who successfully completed NACT and underwent following surgery. For each patient, 4198 radiomics features were extracted from segmented 3D regions of interest (ROI) in mpMRI sequences including T1-weighted dynamic contrast-enhanced imaging (T1-DCE), fat-suppressed T2-weighting imaging (T2WI) and apparent dispersion coefficient (ADC) map. Feature selection and supervised machine learning algorithms were used to identify the predictors correlated with tumor shrinkage pattern as following: (1) Feature dimension reduction by analysis of variance (ANOVA) and the least absolute shrinkage and selection operator (LASSO); (2) Splitting the dataset into training dataset and testing dataset, and constructing prediction models using 12 classification algorithms; (3) Assessing the model performance via area under the curve (AUC), accuracy, sensitivity and specificity. We also compared the most discriminative model in different molecular subtypes of breast cancer.
Results: The Multi-Layer Perception (MLP) neural network achieved the higher AUC and accuracy than other classifiers. The radiomics model achieved a mean AUC of 0.975 (accuracy = 0.912) on the training dataset and 0.900 (accuracy = 0.828) on the testing dataset with 30-round 5-fold cross validation. When incorporating clinicopathologic characteristics, the mean AUC was 0.985 (accuracy = 0.930) on the training dataset and 0.939 (accuracy = 0.870) on the testing set. The model further achieved good AUC on the testing dataset with 30-round 5-fold cross validation in three molecular subtypes of breast cancer as following: (1) HR+/HER2-: 0.901 (accuracy = 0.816); (2) HER2+: 0.940 (accuracy = 0.865); (3) TN: 0.837 (accuracy = 0.811).
Conclusions: It is feasible that our machine learning model combining radiomics features and clinical characteristics could provide a potential tool to predict tumor shrinkage patterns prior to NACT. Our prediction model will be valuable in guiding NACT and surgical treatment in breast cancer.