Comparative Analysis of Tumor Mutational Burden and Copy Number Alterations in WHO-Graded Meningiomas

Background: Meningiomas represent the most common primary intracranial tumor, yet molecular predictors of clinical aggressiveness remain poorly characterized. We conducted a comprehensive multi-omics analysis to investigate whether tumor mutational burden (TMB) and copy number alterations (CNA) correlate with WHO grade classification. Methods: We obtained multi-omics data from 121 meningioma patients (95.0% with somatic mutations, n=115; 100% with copy number data, n=121; 79.3% with protein abundance data, n=96) from the University of Toronto cBioPortal study. Supplementary data included 2 TCGA meningioma samples with whole exome sequencing and 51 meningiomas from Gene Expression Omnibus datasets (GSE43290, n=47; GSE88720, n=14). Clinical annotations included WHO grade (Grade I: n=59; Grade II: n=43; Grade III: n=19; missing: n=2) and recurrence status. Statistical comparisons used the Wilcoxon rank-sum test with significance threshold p<0.05. Machine learning classification was performed using logistic regression, random forest, and XGBoost models on 114 samples with complete genomic data. Results: High-grade meningiomas (WHO Grade II/III, n=62) demonstrated significantly elevated TMB compared to Grade I tumors (n=56): mean 0.450 vs. 0.222 mutations/Mb (fold- change 2.03×, p=1.28×10⁻⁴). Similarly, CNA burden was substantially higher in high-grade tumors: mean fraction genome altered 0.166 vs. 0.098 (fold-change 1.70×, p=1.10×10⁻⁴). Logistic regression achieved the highest classification performance with AUC=0.879 and 76.5% accuracy on the test set (n=34), significantly exceeding the discrimination threshold of AUC>0.70. Conclusion: Genomic variant burden effectively stratifies meningioma aggressiveness and demonstrates clinical utility for grade prediction. These findings support TMB and CNA as quantitative biomarkers for identifying high-risk meningiomas and may guide surveillance protocols and treatment decisions.