Limited Capability of MRI Radiomics to Predict Primary Tumor Histology of Brain Metastases in External Validation

Abstract

Growing research demonstrates the ability to predict histology or genetic information of various malignancies using radiomic features extracted from imaging data. This study aimed to investigate MRI-based radiomics in predicting the primary tumor of brain metastases through internal and external validation, using oversampling techniques to address class imbalance. This IRB-approved retrospective multicenter study included brain metastases from lung cancer, melanoma, breast cancer, colorectal cancer, and a combined heterogenous group of other primary entities (five-class classification). Local data were acquired between 2003 and 2021 from 231 patients (545 metastases). External validation was performed with 82 patients (280 metastases) and 258 patients (809 metastases) from the publicly available Stanford BrainMetShare and the University of California San Francisco Brain Metastases Stereotactic Radiosurgery datasets, respectively. Pre-processing included brain extraction, bias correction, co-registration, intensity normalization, and semi-manual binary tumor segmentation. 2528 radiomic features were extracted from T1w (± contrast), FLAIR, and wavelet transforms for each sequence (eight decompositions). Random forest classifiers were trained with selected features on original and oversampled data (five-fold cross-validation) and evaluated on internal/external holdout test sets using accuracy, precision, recall, F1-score, and AUC. Oversampling did not improve the overall unsatisfactory performance on the internal and external test sets. Incorrect data partitioning (oversampling before train/validation/test split) lead to a massive overestimation of model performance. Radiomics models' capability to predict histologic or genomic data from imaging should be critically assessed; external validation is essential.