For glioblastoma (GBM), current therapeutic approaches focus on the combination of several therapies, each of them individually approved for GBM or other tumor types. Many efforts are made to decipher the best sequence of treatments that would ultimately promote the most efficient tumor response. There is therefore a strong interest in developing new clinical in vivo imaging procedures that can rapidly detect treatment efficacy and allow individual modulation of the treatment. In this preclinical study, we propose to evaluate tumor tissue changes under combined therapies, tumor vascular normalization under antiangiogenic treatment followed by radiotherapy, using a voxel-based clustering approach. This approach was applied to a rat model of glioma (F98). Six MRI parameters were mapped apparent diffusion coefficient, vessel wall permeability, cerebral blood volume fraction, cerebral blood flow, tissue oxygen saturation and vessel size index. We compared the classical region of interest (ROI)-based analysis with a cluster-based analysis. Five clusters, defined by their MRI features, were sufficient to characterize tumor progression and tumor changes during treatments. These results suggest that the cluster-based analysis was as efficient as the ROI-based analysis to assess tumor physiological changes during treatment, but also gave additional information regarding the voxels impacted by treatments and their localization within the tumor. Overall, cluster-based analysis appears to be a powerful tool for subtle monitoring of tumor changes during combined therapies.