Fan Yu-Xiang, Lu Di, Yang Cheng-Bin, Song Zi-Hao, Chen Yi-Guang, Ma Yong-Jie, Li Jing-Wei, Zhang Hong-Qi
Department of Neurosurgery, Xuanwu Hospital, Capital Medical University, Beijing, China.
China International Neuroscience Institute (China-INI), Xuanwu Hospital, Capital Medical University, Beijing, China.
CNS Neurosci Ther. 2025 May;31(5):e70430. doi: 10.1111/cns.70430.
The absence of pharmaceutics poses challenges in preventing intracranial aneurysm (IA) progression and rupture. This research emphasized identifying drug targets for IA through a druggable genome-wide Mendelian randomization (MR) analysis.
A two-sample MR analysis was performed leveraging cis-expression quantitative trait loci in the blood (n = 31,684) and arteries (n = 584) aligned with 5883 druggable genes as exposure and the largest IA summary statistics (n = 7495) as outcome. Bayesian colocalization analysis, plasma cis-protein quantitative trait loci (n = 35,559), and external IA cohorts (FinnGen, n = 2582; Zhou, n = 380) were used for validation. A phenome-wide MR (Phe-MR) incorporating 783 diseases uncovered side effects. Multivariable MR addressed unmeasured pleiotropy.
Five druggable genes in blood and one in the coronary artery showed significant association with IA risk (p- ≤ 0.05). NT5C2, PRCP, and CRMP1 shared a common variant with IA (PPH4 ≥ 0.8). The external validation cohorts confirmed the effects of NT5C2 on IA (FinnGen cohort, Odds Ratio [OR], 0.81, 95% Confidential Interval [95% CI] 95% CI, 0.707-0.930; p = 0.003; Zhou cohort, OR, 0.68, 95% CI, 0.469-0.983; p = 0.041). The genetically predicted protein level of PRCP validated an inverse association with IA risk (OR, 0.734; 95% CI, 0.561-0.959; p = 0.023). The Phe-MR revealed insignificance for NT5C2 or PRCP. Direct causal effects on IA were 0.60 (95% CI, 0.457-0.797; p = 1.36E-05) for PRCP and 0.67 (95% CI, 0.527-0.860; p = 0.002) for NT5C2 after adjusting for IA modifiable risk factors.
NT5C2 and PRCP were identified as potential drug targets for IA, with effects independent of known modifiable risk factors. Targeting NT5C2 and PRCP appeared exclusively effective and safe.
缺乏药物制剂给预防颅内动脉瘤（IA）进展和破裂带来了挑战。本研究着重通过全基因组可药物化孟德尔随机化（MR）分析来确定IA的药物靶点。
进行了一项两样本MR分析，利用血液（n = 31684）和顺式表达数量性状位点的动脉（n = 584）与5883个可药物化基因作为暴露因素，并将最大的IA汇总统计数据（n = 7495）作为结果。使用贝叶斯共定位分析、血浆顺式蛋白质数量性状位点（n = 35559）和外部IA队列（芬兰基因队列，n = 2582；周，n = 380）进行验证。纳入783种疾病的全表型MR（Phe-MR）揭示了副作用。多变量MR解决了未测量的多效性问题。
血液中的5个可药物化基因和冠状动脉中的1个基因与IA风险显示出显著关联（p值≤0.05）。NT5C2、PRCP和CRMP1与IA共享一个常见变异（PPH4≥0.8）。外部验证队列证实了NT5C2对IA的影响（芬兰基因队列，优势比[OR]，0.81，95%置信区间[95%CI]，0.707 - 0.930；p = 0.003；周队列，OR，0.68，95%CI，0.469 - 0.983；p = 0.041）。PRCP的遗传预测蛋白水平验证了与IA风险的负相关（OR，0.734；95%CI，0.561 - 0.959；p = 0.023）。Phe-MR显示NT5C2或PRCP无显著意义。在调整IA可改变的风险因素后，PRCP对IA的直接因果效应为0.60（95%CI，0.457 - 0.797；p = 1.36E - 05），NT5C2为0.67（95%CI，0.527 - 0.860；p = 0.002）。
NT5C2和PRCP被确定为IA的潜在药物靶点，其作用独立于已知的可改变风险因素。靶向NT5C2和PRCP似乎是唯一有效且安全的。
