Quantitative Susceptibility Mapping: methods to detect longitudinal and spatial glioblastoma infiltration

Glioblastoma is the most malignant type of primary brain tumor. Because of its highly infiltrative nature, current treatments fail to remove the tumor completely. The standard tool to determine disease extent and guide surgical planning is conventional Magnetic Resonance Imaging (MRI). However, it lacks the specificity needed to detect hidden tumor cell infiltration, particularly within the peritumoral edema. To overcome this limitation, this study employed Quantitative Susceptibility Mapping (QSM). QSM helps to identify tissue composition by associating positive susceptibility values with paramagnetic materials (e.g., iron) and negative values with diamagnetic materials (e.g., myelin and calcifications). Pre- and post-operative MRI scans of 20 glioblastoma patients were acquired using a 3D multi-echo gradient-echo sequence, to generate quantitative maps of brain tissue magnetic susceptibility. The analysis performed in this study identified spatial patterns of altered susceptibility within the edema region, highlighting potential areas of active tumor infiltration. Furthermore, longitudinal monitoring was conducted to track this predicted infiltration throughout disease progression. These findings were corroborated by an independent voxelwise analysis, while the spatial pathways of tumor’s infiltration were investigated using a standard white matter tractography atlas. Lastly, the microstructural integrity of these infiltrated regions was further evaluated using a standard Fractional Anisotropy (FA) atlas. This study demonstrates how QSM can effectively map magnetic susceptibility alterations that may reflect hidden tumoral activity. By providing a quantitative assessment of tissue magnetic properties, this work aims to contribute to the future validation of QSM as a non-invasive imaging biomarker, capable of supporting surgical resection strategies and personalized treatments.