Abstract
Glioma is the most common malignant tumor in the central nervous system with significant challenges for its treatment and prognosis. Based on publicly available genome-wide association study data, this study employed a bidirectional, two-sample Mendelian randomization (MR) analysis, combined with Bayesian weighted MR, to investigate the causal effect of 233 circulating metabolic traits on glioma and its subtypes, with the expectation of discovering new diagnostic and therapeutic targets. The MR study revealed that the Total cholesterol to total lipids ratio in large VLDL and the Ratio of polyunsaturated fatty acids to total fatty acids (PUFAbyFA) are risk factors for glioma, while free cholesterol or phospholipids in small HDL are protective against glioma. The free cholesterol to total lipids ratio in IDL, the ratios of total cholesterol or cholesteryl esters to total lipids in small or medium VLDL, as well as linoleic acid (LA 18:2), phosphatidylcholine, and sphingomyelins levels are risk factors for non-GBM glioma. Free cholesterol in small HDL is identified as a protective factor for non-GBM glioma. The results were robust to sensitivity analyses and Bayesian weighted mendelian randomization. The causal effects of glioma on circulating metabolites were explored through reverse mendelian randomization. The results provide potential biomarkers for the early diagnosis and treatment of glioma.