遗传预测血液代谢物与中耳、鼻腔及鼻窦良性肿瘤的因果关系——两样本双向孟德尔随机化研究

Genetic Prediction of Causality between Blood Metabolites and Benign Tumors of the Middle Ear, Nasal Cavity, and Paranasal Sinuses—Two-Sample Bidirectional Mendelian Randomization Study

摘要: 目的：利用孟德尔随机化分析(Mendelian randomization, MR)方法，探究1400种血液代谢物和中耳、鼻腔及鼻窦良性肿瘤的因果关系。方法：本研究采用两样本双向孟德尔随机化分析，利用全基因组关联研究(Genome-wide Association study, GWAS)数据库获得血液代谢物和中耳、鼻腔及鼻窦良性肿瘤相关数据。使用R软件和TwoSampleMR软件包进行分析。研究采用逆方差加权法(IVW)为主，结合MR-Egger回归、加权中位数法(WM)、简单模型法(Simple mode)及加权模型法(Weighted mode)作为补充分析血液代谢物和中耳、鼻腔及鼻窦良性肿瘤的因果关系。为进一步增强结果的可靠性和稳定性通过Cochran Q检验、MR-Egger回归检验、MR-PRESSO综合检验以及MR Egger截距检测异质性及水平多效性。由于样本过大，为了结果更加严谨对结果进行错误发现率(FDR)矫正。反向MR分析以中耳、鼻腔及鼻窦良性肿瘤为暴露因素，将正向筛选得到的血液代谢物作为结局变量进行效应分析和敏感性分析。结果：分析结果显示，发现1种血液代谢物为棕榈油酸(16:1n-7)升高与中耳、鼻腔及鼻窦良性肿瘤风险增高显著相关(IVW:OR = 1.971, 95% CI: 1.392~2.789, P < 0.001)，而反向MR提示中耳、鼻腔及鼻窦良性肿瘤与棕榈油酸(16:1n-7)无显著相关性(IVW:OR = 1.027, 95% CI: 0.980~1.076, P = 0.269)。通过Cochran Q检验、MR-Egger回归检验、MR-PRESSO综合检验以及MR Egger截距检测结果显示工具变量之间不存在异质性及水平多效性，同时，通过留一法检验分析证实，单个SNPs对整体结果没有显著影响，进一步增强了结果的可靠性和稳定性。结论：在1400种血液代谢物种发现1种棕榈油酸(16:1n-7)与中耳、鼻腔及鼻窦良性肿瘤发病存在正向因果关联，可为中耳、鼻腔及鼻窦良性肿瘤发病机制及早期筛检和治疗提供参考。

Abstract: Objective: To explore the causal relationship between 1400 blood metabolites and benign tumors of middle ear, nasal cavity and paranasal sinus by using Mendelian randomization (MR) method. Methods: Two-sample bi-directional Mendelian randomization analysis was used in this study, and the data related to blood metabolites and benign tumors of middle ear, nasal cavity and paranasal sinus were obtained from Genome-wide Association study (GWAS) database. R software and Two Sample MR software package were used for analysis. The inverse variance weighting (IVW) method was used as the main method, and MR-Egger regression, weighted median (WM), simple mode and weighted mode were used as supplementary methods to analyze the causal relationship between blood metabolites and benign tumors of middle ear, nasal cavity and paranasal sinus. To further enhance the reliability and stability of the results, Cochran Q test, MR-Egger regression test, MR-PRESSO comprehensive test and MR Egger intercept were used to detect heterogeneity and horizontal pleiotropy. Due to the large sample size, the results were corrected by false detection rate (FDR) for more rigorous results. The reverse MR analysis took benign tumors of middle ear, nasal cavity and paranasal sinus as the exposure factor, and the blood metabolites obtained by positive screening were used as outcome variables for effect analysis and sensitivity analysis. Results: The results showed that the increase of palmitic acid (16:1n-7) was significantly associated with the risk of benign tumors of middle ear, nasal cavity and paranasal sinus (IVW:OR = 1.971, 95% CI: 1.392~2.789, P < 0.001), while reverse MR showed no significant correlation between benign tumors of middle ear, nasal cavity and paranasal sinus and palmitic acid (16:1n-7) (IVW:OR = 1.027, 95% CI: 0.980~1.076, P = 0.269). The results of Cochran Q test, MR-Egger regression test, MR-PRESSO comprehensive test and MR Egger intercept test showed that there was no heterogeneity and horizontal pleiotropy between the instrumental variables. At the same time, the analysis of one-way ANOVA confirmed that a single SNP had no significant effect on the overall results, further enhancing the reliability and stability of the results. Conclusions: Among 1400 blood metabolites, there is a positive causal relationship between palmitic acid (16:1n-7) and the incidence of benign tumors of middle ear, nasal cavity and paranasal sinus, which can provide reference for the pathogenesis, early screening and treatment of benign tumors of middle ear, nasal cavity and paranasal sinus.