Beyond morphology: imaging the glioblastoma microenvironment in the era of quantitative neuro-oncology.
Authors
Affiliations (4)
Affiliations (4)
- Centre for Medical Sciences (CISMed), University of Trento, Trento, Italy.
- Department of Economics and Management, University of Trento, Trento, Italy.
- Unit of Artificial Intelligence and Computer Systems, Department of Engineering, Università Campus Bio-Medico di Roma, Roma, Italy.
- Department of Nuclear Medicine, University Hospital Leipzig, Leipzig, Germany.
Abstract
Glioblastoma (GBM) is the most aggressive primary brain tumor in adults and is characterized by rapid progression, marked spatial and molecular heterogeneity, and poor prognosis despite multimodal treatment strategies. Tumor behavior is not determined solely by intrinsic genetic alterations but also by dynamic interactions within the tumor microenvironment, including hypoxia, aberrant angiogenesis, immune modulation, and metabolic reprogramming, which are major drivers of treatment resistance and disease recurrence. Magnetic resonance imaging (MRI) remains the cornerstone of diagnosis and treatment planning; advanced quantitative techniques have expanded its role beyond structural assessment, enabling <i>in vivo</i> characterization of tissue cellularity, vascular architecture, perfusion, and microenvironmental dynamics. Positron Emission Tomography (PET) provides complementary metabolic and molecular information, improving tumor delineation, detection of infiltrative disease, and assessment of hypoxia and treatment response. PET-MRI integration within a multimodal framework enables spatially resolved mapping of tumor heterogeneity and microenvironmental niches that cannot be adequately assessed by either modality alone. Radiomics and radiogenomics approaches further enhance this paradigm by extracting quantitative imaging features that reflect underlying biological processes and linking imaging phenotypes with molecular and clinical outcomes. Artificial Intelligence enables automated feature extraction, multimodal data integration, and predictive modeling for diagnosis, prognosis, and treatment response assessment. Despite these advances, clinical translation remains limited by methodological heterogeneity, lack of standardized acquisition protocols, and insufficient prospective validation. Future research should prioritize harmonized multicenter studies and biologically informed multimodal analytical frameworks to enable the routine implementation of microenvironment-oriented precision imaging in GBM management.