The ability to distinguish malignant from benign lesions using MRI imaging can significantly enhance diagnostic and therapeutic management. Artificial Intelligence (AI) can play a key role in this process. Dynamic Contrast-Enhanced (DCE) MRI has proven effective in improving malignancy prediction by providing valuable features related to cellularity and neoangiogenesis. In this study, we present our findings on the classification of malignant and benign lesions using the publicly available Advanced MRI Breast Lesions dataset from The Cancer Imaging Archive. This dataset includes T2-weighted and DCE-MRI sequences, along with segmentations of all suspicious lesions, from 200 patients. Given the limited dataset size, deep learning algorithms must be used with caution, as training models with a high number of parameters is not feasible. Instead, we adopted a radiomic approach to evaluate whether an XGBoost classification model could leverage information from DCE-MRI. After extracting radiomic features from robustly-scaled images, we trained and assessed an XGBoost model, experimenting with different feature combinations in a stratified 5-fold cross-validation. Using only T2-weighted MRI, the best model achieved an Area Under the Curve (AUC) score of 0.83 ± 0.02, while when incorporating both T2-weighted and DCE-MRI data, the model performance improved to an AUC of 0.87 ± 0.03. Our future work will focus on identifying additional discriminative features within DCE-MRI that could further enhance diagnostic accuracy.