Metabolomic and Lipidomic Characterization of Meningioma Grades Using LC-HRMS and Machine Learning

Area of Science:

• Neuro-oncology
• Metabolomics
• Lipidomics

Background:

• Meningiomas are common central nervous system tumors, with WHO classification including benign, atypical, and anaplastic subtypes.
• Some histologically benign meningiomas exhibit aggressive clinical behavior, necessitating better diagnostic tools.

Purpose of the Study:

• To compare metabolomic and lipidomic profiles of low-grade and high-grade meningiomas.
• To identify potential biomarkers for discriminating between benign and malignant meningioma tumors.

Main Methods:

• Untargeted metabolomics and lipidomics analyses were performed on 85 meningioma tumor biopsy samples.
• High-resolution mass spectrometry coupled to liquid chromatography was employed.
• Feature selection and machine learning techniques were applied to identify diagnostic markers.

Main Results:

• Three distinct biomarkers were identified capable of differentiating low-grade from high-grade meningioma brain tumors.
• Metabolomic and lipidomic profiles varied significantly between tumor grades.

Conclusions:

• Mass spectrometry-based metabolomics and lipidomics, combined with machine learning, can identify biomarkers for meningioma grading.
• These findings may lead to novel therapeutic and prognostic strategies for meningioma patients.