Contralesional grey matter volume as an index of macrostructural plasticity in patients with brain tumors
Lucía Manso-Ortega1, Sandra Gisbert-Muñoz2, Lucia Amoruso3
1BCBL, Basque Center of Cognition, Brain and Language, Donostia-San Sebastian, 20009, Spain; University of the Basque Country (UPV/EHU), Bilbao, 48013, Spain.
Background:
Brain tumors can disrupt the functional and structural architecture of the brain, with the damage not limited to the lesioned tissue or its vicinity. Whether they induce coordinated, network-level changes in the macrostructure remains unresolved.
Methods:
We analyzed structural MRI data from a cohort of 107 adults, including patients diagnosed with left-hemisphere brain tumors and healthy participants. To characterize brain-wide alterations in contralesional grey matter volume, we applied a data-driven approach that combined Voxel-Based-Morphometry (VBM) and Principal Component Analysis (PCA). Specifically, PCA was used as an anomaly detection method to model the normative covariance structure of the contralesional hemisphere to identify patient-specific deviations from this pattern. A subset of patients was evaluated longitudinally approximately four months after surgery to track changes over time.
Results:
Patients exhibited significantly greater contralesional GMV than healthy participants, and the PCA-based structural covariance model distinguished patients from controls with an AUC of 0.78 pre-surgically and 0.93 post-surgically. Longitudinal analysis revealed that structural alterations are not static: post-surgical divergence from the normative structural covariance pattern increased substantially. Correlation analysis identified a negative association between contralateral GMV expansion and post-surgical cognitive performance in language and intelligence domains.
Conclusion:
Left-hemisphere brain tumors induce widespread, evolving, and spatially coordinated structural adaptations throughout the contralesional hemisphere. Our findings provide evidence that the brain responds to focal pathology through distributed and coordinated network changes, likely reflecting altered structural covariance. Structural plasticity is tightly linked to function, and structural covariance represents a promising biomarker for predicting cognitive outcomes and guiding personalized neuro-rehabilitation strategies.
