遗传预测的强直性脊柱炎与肾恶性肿瘤风险的因果关联：一项基于FinnGen的两样本孟德尔随机化研究

Causal Association between Genetically Predicted Ankylosing Spondylitis and the Risk of Malignant Neoplasm of the Kidney: A Two-Sample Mendelian Randomization Study Based on FinnGen

摘要: 目的：探讨遗传预测的强直性脊柱炎(ankylosing spondylitis, AS)与肾恶性肿瘤(不含肾盂)风险之间的潜在因果关系。方法：采用两样本孟德尔随机化(Mendelian randomization, MR)设计，暴露为FinnGen数据库的强直性脊柱炎(GWAS ID：finn-b-M13_ANKYLOSPON；病例数1462，对照数164,682，总样本量166,144)，结局为肾恶性肿瘤(不含肾盂) (GWAS ID：finn-b-C3_KIDNEY_NOTRENALPELVIS；病例数971，对照数217,821，总样本量218,792)。以全基因组显著性阈值P < 5 × 10⁻⁸筛选AS相关工具变量，并进行LD去相关(r² = 0.001，窗口10,000 kb)。结局侧允许使用LD代理SNP (r² ≥ 0.8)以提高匹配率。等位基因协调采用action = 2，并进行Steiger方向性过滤。主要分析采用逆方差加权法(IVW, multiplicative random effects)，并辅以加权中位数法(weighted median)及MR-Egger回归验证稳健性。通过Cochran’s Q检验评估异质性，MR-Egger截距检验与MR-PRESSO评估水平多效性，并采用留一法(leave-one-out)检验单个SNP对结果的影响。工具变量强度以F统计量评估。结果：共纳入11个工具变量，均通过协调与方向性过滤。工具变量强度良好(F统计量最小值32.90，均值约308.60)。IVW结果显示，遗传预测的AS与肾恶性肿瘤风险升高显著相关(β = 0.0394, SE = 0.0145, P = 0.0066; OR = 1.040, 95%CI: 1.011~1.070)。加权中位数与MR-Egger结果方向一致且同样达到统计学显著(weighted median: OR = 1.047, 95%CI: 1.004~1.093, P = 0.0319; MR-Egger: OR = 1.068, 95%CI: 1.009~1.130, P = 0.0495)。敏感性分析未提示异质性(Q检验P = 0.777)、方向性水平多效性(MR-Egger截距P = 0.277)或离群工具变量(MR-PRESSO global P = 0.862, outlier = 0)。留一法提示剔除任一单个SNP后总体效应方向保持一致。结论：遗传预测的强直性脊柱炎与肾恶性肿瘤(不含肾盂)风险升高存在统计学相关性，提示AS可能是肾癌风险增加的潜在因果因素之一。仍需在其他人群与独立数据集中进一步验证，并结合机制研究阐明其生物学基础。

Abstract: Objective: To investigate the potential causal association between genetically predicted ankylosing spondylitis (AS) and the risk of malignant neoplasm of the kidney (except renal pelvis). Methods: A two-sample Mendelian randomization (MR) analysis was performed using FinnGen summary statistics: AS (finn-b-M13_ANKYLOSPON; 1462 cases and 164,682 controls) and kidney cancer excluding renal pelvis (finn-b-C3_KIDNEY_NOTRENALPELVIS; 971 cases and 217,821 controls). Instrumental SNPs were selected at P < 5 × 10⁻⁸ and clumped (r² = 0.001, 10,000 kb); outcome-side proxies were allowed (r² ≥ 0.8). Alleles were harmonized (action = 2) and Steiger filtering was applied. IVW multiplicative random-effects was the primary method, with weighted median and MR-Egger as sensitivity analyses; Heterogeneity was assessed using Cochran’s Q test. Horizontal pleiotropy was evaluated using the MR-Egger intercept test and the MR-PRESSO method. A leave-one-out analysis was performed to assess the influence of individual single nucleotide polymorphisms (SNPs) on the overall estimate. Instrument strength was evaluated using the F statistic. Results: A total of 11 instrumental variables were included, all of which passed harmonization and directional filtering. The instruments demonstrated strong strength (minimum F statistic = 32.90; mean F statistic ≈ 308.60). The inverse-variance weighted (IVW) analysis showed that genetically predicted AS was significantly associated with an increased risk of malignant kidney neoplasm (β = 0.0394, SE = 0.0145, P = 0.0066; OR = 1.040, 95% CI: 1.011~1.070). The weighted median and MR-Egger methods yielded consistent effect directions and also reached statistical significance (weighted median: OR = 1.047, 95% CI: 1.004~1.093, P = 0.0319; MR-Egger: OR = 1.068, 95% CI: 1.009~1.130, P = 0.0495). Sensitivity analyses indicated no evidence of heterogeneity (Cochran’s Q test, P = 0.777), directional horizontal pleiotropy (MR-Egger intercept, P = 0.277), or outlier instruments (MR-PRESSO global test, P = 0.862; outliers = 0). The leave-one-out analysis demonstrated that the overall effect estimate remained stable after sequential removal of each SNP. Conclusion: Genetically predicted ankylosing spondylitis was statistically associated with an increased risk of malignant kidney neoplasm (excluding renal pelvis cancer), suggesting that AS may be a potential causal risk factor for renal cancer. Further validation in independent populations and datasets is warranted, and mechanistic studies are needed to elucidate the underlying biological pathways.