Yang Xiaolei, Lin Xuejing
Department of Neurosurgery, The First Hospital of China Medical University, Shenyang, China.
Department of Neurology, The First Hospital of China Medical University, Shenyang, China.
Medicine (Baltimore). 2025 Aug 1;104(31):e43636. doi: 10.1097/MD.0000000000043636.
Glioblastoma (GBM) is one of the most lethal central nervous system malignancies, characterized by complex inflammatory responses and metabolic reprogramming. However, the causal relationships between inflammatory factors, metabolites, and GBM risk remain largely unclear. We conducted a 3-stage Mendelian randomization analysis using data from the FinnGen biobank (341 GBM cases and 314,193 controls) and genome-wide association studies of 91 inflammatory factors and 1400 metabolites. For instrument variable selection, we applied a stringent threshold (P < 1 × 10-5) and performed linkage disequilibrium pruning (r2 < 0.001). We employed inverse variance weighted method as the primary analysis, supplemented with weighted median, simple mode, and weighted mode methods. Sensitivity analyses included heterogeneity tests, pleiotropy assessments, and leave-one-out analyses. Mediation analysis was performed to explore the metabolic pathways linking inflammatory factors to GBM risk. We identified 4 inflammatory factors causally associated with GBM risk: CCL25 (OR = 1.24, 95% CI = 1.02-1.51), M-CSF1 (OR = 1.70, 95% CI = 1.09-2.68), and IL-33 (OR = 1.61, 95% CI = 1.01-2.56) showed risk-increasing effects, while FGF21 (OR = 0.57, 95% CI = 0.37-0.89) demonstrated protective effects. Among 1400 metabolites, 23 showed significant causal associations with GBM risk across multiple statistical methods. The strongest evidence was found for phospholipid metabolism (1-stearoyl-2-oleoyl-GPE, OR = 1.21, 95% CI = 1.08-1.36) and energy metabolism (cAMP/taurocholate ratio, OR = 0.85, 95% CI = 0.77-0.94). Mediation analysis revealed 21 inflammatory factor-metabolite-GBM pathways, with FGF21 showing the most extensive metabolic regulatory network (12 pathways), followed by M-CSF1 (7 pathways) and CCL25 (2 pathways). Our study establishes a causal framework linking peripheral inflammatory factors and metabolic reprogramming to GBM risk, with FGF21 emerging as a potential protective factor. These findings provide novel therapeutic targets and contribute to the development of precision medicine strategies for GBM.
胶质母细胞瘤（GBM）是最致命的中枢神经系统恶性肿瘤之一，其特征为复杂的炎症反应和代谢重编程。然而，炎症因子、代谢物与GBM风险之间的因果关系仍 largely 不清楚。我们使用来自芬兰基因生物银行的数据（341例GBM病例和314,193名对照）以及91种炎症因子和1400种代谢物的全基因组关联研究进行了三阶段孟德尔随机化分析。对于工具变量选择，我们应用了严格的阈值（P < 1×10-5）并进行了连锁不平衡修剪（r2 < 0.001）。我们采用逆方差加权法作为主要分析方法，并辅以加权中位数、简单模式和加权模式方法。敏感性分析包括异质性检验、多效性评估和留一法分析。进行中介分析以探索将炎症因子与GBM风险联系起来的代谢途径。我们确定了4种与GBM风险因果相关的炎症因子：CCL25（OR = 1.24，95%CI = 1.02 - 1.51）、M-CSF1（OR = 1.70，95%CI = 1.09 - 2.68）和IL-33（OR = 1.61，95%CI = 1.01 - 2.56）显示出风险增加效应，而FGF21（OR = 0.57，95%CI = 0.37 - 0.89）表现出保护作用。在1400种代谢物中，23种在多种统计方法中显示出与GBM风险有显著的因果关联。在磷脂代谢（1-硬脂酰-2-油酰-GPE，OR = 1.21，95%CI = 1.08 - 1.36）和能量代谢（cAMP/牛磺胆酸盐比率，OR = 0.85，95%CI = 0.77 - 0.94）方面发现了最有力的证据。中介分析揭示了21条炎症因子-代谢物-GBM途径，其中FGF21显示出最广泛的代谢调控网络（12条途径），其次是M-CSF1（7条途径）和CCL25（2条途径）。我们的研究建立了一个将外周炎症因子和代谢重编程与GBM风险联系起来的因果框架，FGF21成为一种潜在的保护因子。这些发现提供了新的治疗靶点，并有助于GBM精准医学策略的发展。
