Radiomic Profiling of GBM Reveals Histopathological Correlations Associated with Tumour Response to Ionizing Radiation

Glioblastoma (GBM) is a particularly malignant brain tumour which carries a poor prognosis and presents limited treatment options. MRI is standard practice for differential diagnosis at initial presentation of GBM and can assist in both treatment planning and responses. MRI radiomics can generate mineable databases from radiological images providing sub-visual feature analysis capable to augment the morphological and functional tumour characteristics beyond traditional imaging techniques. Given that radiotherapy is part of the standard of care for GBM patients, establishing a platform for phenotyping radiation treatment responses using non-invasive methods is of high relevance. In this study, we modelled the responses to ionizing radiation across four orthotopic models of GBM using diffusion and perfusion radiomics. We have identified the optimal set of radiomic features that demonstrate significant correlations with tumour cellularity, microvascular organisation and blood flow alterations when directly compared with endpoint histopathological analysis. We showed that the selected radiomic features can quantify textural information and pixel interrelationships of tumour response to radiation therapy, revealing subtle image patterns that may reflect intra-tumoural spatial heterogeneity. When compared to GBM patients, similarities of selected radiomic features were noted between orthotopic tumours and non-enhancing central tumour areas of patients regarding tumour homogeneity, along with several differences in gray level nonuniformity and image complexity metrics. As the field evolves, radiomic profiling of GBM may enhance the evaluation of targeted therapeutic strategies, accelerate the development of new therapies and could be acting as a potential virtual biopsy tool.