Purpose: In meningiomas, TERT promotor mutations are rare but qualify the diagnosis of anaplasia, directly impacting adjuvant therapy. Effective screening for patients at risk for promotor mutations could enable more targeted molecular analyses and improve diagnosis and treatment. Methods: Semiautomatic segmentation of intracranial grade 2/3 meningiomas was performed on preoperative magnetic resonance imaging. Discriminatory power to predict TERT promoter mutations was analyzed using a random forest algorithm with an increasing number of radiomic features. Two final models with five and eight features with both fixed and differing radiomics features were developed and adjusted to eliminate random effects and to avoid overfitting. Results: A total of 117 image sets including training (N = 94) and test data (N = 23) were analyzed. To eliminate random effects and demonstrate the robustness of our approach, data partitioning and subsequent model development and testing were repeated a total of 100 times (each time with repartitioned training and independent test data). The established five- and eight-feature models with both fixed and different radiomics features enabled the prediction of TERT with similar but excellent performance. The five-feature (different/fixed) model predicted TERT promotor mutation status with a mean AUC of 91.8%/94.3%, mean accuracy of 85.5%/88.9%, mean sensitivity of 88.6%/91.4%, mean specificity of 83.2%/87.0%, and a mean Cohen’s Kappa of 71.0%/77.7%. The eight-feature (different/fixed) model predicted TERT promotor mutation status with a mean AUC of 92.7%/94.6%, mean accuracy of 87.3%/88.9%, mean sensitivity of 89.6%/90.6%, mean specificity of 85.5%/87.5%, and a mean Cohen’s Kappa of 74.4%/77.6%. Of note, the addition of further features of up to N = 8 only slightly increased the performance. Conclusions: Radiomics-based machine learning enables prediction of TERT promotor mutation status in meningiomas with excellent discriminatory performance. Future analyses in larger cohorts should include grade 1 lesions as well as additional molecular alterations.

Background: Our aim was to investigate the predictability of postoperative meningioma resection status based on clinical features. Methods: We examined 23 clinical features to assess their effectiveness in distinguishing gross total resections (GTR) from subtotal resections (STR). We analyzed whether GTR/STR cases are better predictable if the classification is based on the Simpson grading or the postoperative operative tumor volume (POTV). Results: Using a study cohort comprising a total of 157 patients, multivariate models for the preoperative prediction of GTR/STR outcome in relation to Simpson grading and POTV were developed and subsequently compared. Including only two clinical features, our models showed a notable discriminatory power in predicting postoperative resection status. Our final model, a straightforward decision tree applicable in daily clinical practice, achieved a mean AUC of 0.885, a mean accuracy of 0.866, a mean sensitivity of 0.889, and a mean specificity of 0.772 based on independent test data. Conclusions: Such models can be a valuable tool both for surgical planning and for early planning of postoperative treatment, e.g., for additional radiotherapy/radiosurgery, potentially required in case of subtotal resections.

BACKGROUND AND OBJECTIVES: Simpson grading remains the most widely used clinical classification of extent of resection in meningioma surgery, yet its relevance in modern neurosurgical practice remains debated. The aim was to evaluate whether Simpson grading is associated with recurrence risk after gross total resection (GTR) of intracranial meningiomas within the context of tumor biology and anatomic location. METHODS: We analyzed 4074 patients undergoing GTR (Simpson grades 1-3) for primary intracranial meningioma, compiling data from 42 institutions worldwide. Cumulative incidence of recurrence was estimated using Aalen-Johansen methods. Multivariable inverse probability of censoring weighted logistic regression models were used to estimate adjusted recurrence risks at 5 and 10 years, adjusting for age, sex, World Health Organization (WHO) grade, Ki-67 proliferation index, and healthcare system setting. RESULTS: Higher Simpson grades were consistently associated with increased recurrence risk in WHO-1 and WHO-2. Among WHO-1, Simpson grade 2 had more than twice the recurrence odds compared with grade 1 at 5 years, whereas grade 3 showed even higher odds ratio of 3.47 (95% CI, 1.98-6.08). Similar patterns were observed in WHO-2 tumors. Interaction analyses suggested that differences between Simpson grades increase with higher Ki-67 proliferation indices. The prognostic significance of Simpson grade varied across anatomic locations, with clear associations observed for convexity and skull-base meningiomas, but limited relevance for parasagittal/falx meningiomas irrespective of WHO grade. CONCLUSION: Simpson grading remained strongly associated with recurrence after GTR. These findings suggest that the extent of dural and bony clearance remains clinically relevant; however, Simpson grading should be contextualized within tumor biology and anatomic constraints. Future recurrence models integrating molecular and surgical variables may refine prognostic frameworks rather than current standalone classification systems.