A molecularly integrated grade for meningioma.

Area of Science:

• Neuro-oncology
• Genomics
• Cancer grading

Background:

• Meningiomas are the most common primary brain tumors in adults.
• Current World Health Organization (WHO) grading relies on histopathology and surgical extent, but often fails to predict clinical behavior.
• There is a need for improved prognostic information to optimize meningioma patient management.

Purpose of the Study:

• To evaluate if chromosomal copy-number data can enhance the prediction of meningioma recurrence time.
• To develop and validate a novel grading system integrating genomic information for improved prognostic accuracy.

Main Methods:

• Utilized Cox proportional hazards, random survival forest, and gradient boosting models.
• Developed and validated a 3-tiered Integrated Grade incorporating mitotic count and specific chromosomal losses (1p, 3p, 4, 6, 10, 14q, 18, 19, or CDKN2A).
• Tested models on a discovery cohort (527 patients) and two independent validation cohorts (172 patients) using orthogonal genomic platforms.

Main Results:

• The Integrated Grade reclassified 32% of meningiomas to a different risk category compared to the WHO grade.
• The Integrated Grade demonstrated superior accuracy in identifying patients at risk for recurrence.
• Performance was validated using time-dependent area under the curve, average precision, and Brier score.

Conclusions:

• A molecularly integrated grading scheme for meningiomas significantly improves prediction of progression-free survival over the current WHO system.
• This novel framework is broadly adoptable by clinicians using accessible genomic technologies.
• The proposed system represents an advance in meningioma patient care and management.