Medical Student Brigham and Women's Hospital/Harvard Medical School Brookline, Massachusetts, United States
Introduction: The anatomic location of meningiomas has been shown to correlate with distinct clinical manifestations, histopathological subtypes, and surgical risk. However, meningioma anatomic origin sites can be obscured in large tumors and those crossing compartments. We therefore sought to apply unbiased lesion mapping to localize intracranial meningioma distributions and their association with biology and grade.
Methods: MR images, World Health Organization (WHO) grade, and a molecularly Integrated Grade (IG) derived from cytogenetics were analyzed from 882 adults with pathology-confirmed intracranial meningiomas from a tertiary academic medical center. Semi-automated tumor segmentation was performed on pre-treatment T1-weighted contrast-enhanced MRI. We used a voxel-based lesion mapping technique to generate a meningioma atlas with spatial distribution frequency. Further, we correlated this with tumor grade and location classification by anatomic origin.
Results: Of 882 patients (mean age of 57±14 years, 68.8% female), 589 were WHO grade 1 (66.8%), 266 WHO grade 2 (30.1%), and 27 WHO grade 3 (3.1%). After molecular reclassification, 585 were IG-1 (66.3%), 160 were IG-2 (18.2%), and 137 were IG-3 (15.5%). Benign tumors were heavily concentrated in and around the midline anterior skull base. Malignant meningiomas were enriched for the falcine/parasagittal and the right sphenoid region. The spatial clustering of meningiomas was far more striking when stratified by the molecular Integrated Grade than by WHO grade. WHO grade 2 meningiomas divided equally across IG 1-3 by molecular criteria, with corresponding partition of spatial distribution in the midline anterior skull base (in WHO grade 2, IG 1) and falcine/parasagittal and sphenoid (WHO grade 2, IG3).
Conclusion : Using an unbiased mapping approach, we demonstrate the preferential locations of intracranial meningiomas, with specific locations associated with more aggressive tumors. Distinct tumor distribution patterns emerged across both histopathologic and molecularly defined grades. Molecular grading associated with sharper spatial clusters of meningioma, supporting a phenotype-genotype association in meningioma.
