基于蛋白组孟德尔随机化与共定位分析的胃癌风险相关候选蛋白研究

Proteome-wide Mendelian randomization and colocalization analysis identifying candidate proteins associated with gastric cancer risk

摘要:

背景与目的 胃癌早期症状隐匿且缺乏可规模化普及的高效早筛手段，而循环血浆蛋白具备可检测与可干预潜力。研究旨在利用遗传工具系统评估血浆蛋白与胃癌风险的潜在因果关联。方法 以4 907个循环血浆蛋白的pQTL汇总统计作为暴露数据，以FinnGen R12胃癌GWAS数据作为结局，开展双样本孟德尔随机化（MR）筛查。对候选信号进一步进行基于汇总数据的孟德尔随机化（SMR）/依赖工具的异质性分析及贝叶斯共定位分析，以评估蛋白信号与胃癌风险信号共享因果变异的可能性。随后采用独立欧洲人群胃癌GWAS数据集进行外部验证及Meta分析。进一步结合TCGA-STAD、GTEx及GEPIA2数据库，对优先候选基因的表达特征、预后价值及相关生物学过程进行验证。结果 MR筛查共获得84个与胃癌风险相关的候选蛋白，其中36个呈风险增加方向，48个呈保护方向。SMR分析进一步筛选出11个阳性蛋白。贝叶斯共定位分析显示，HGF具有最强共享因果变异支持（PP.H4=0.792）。外部GWAS验证及Meta分析进一步支持遗传预测的循环HGF水平升高与胃癌风险增加相关（OR=1.224，95% CI=1.070~1.401，P=0.003）。转录组分析显示，HGF在胃癌组织中表达升高，且高表达与较差总体生存及无病生存相关。GSEA结果提示，HGF相关基因主要富集于细胞外基质重塑、血管生成及免疫调控等生物学过程。结论 本研究构建了胃癌风险相关循环蛋白的遗传优先级图谱，并在多层次遗传证据链中优先支持HGF作为胃癌风险相关候选蛋白。相关结果为胃癌风险分层、机制研究及后续临床验证提供了新的分子线索。

Abstract:

Background and Aims Gastric cancer lacks highly effective and widely applicable biomarkers for early detection. Circulating proteins are measurable molecular intermediates with potential biological and clinical relevance. This study aimed to systematically evaluate the genetic associations between circulating plasma proteins and gastric cancer risk and to prioritize proteins with convergent genetic evidence.Methods A two-sample Mendelian randomization (MR) analysis was conducted using pQTL summary statistics for 4 907 plasma proteins as exposures and FinnGen Release 12 gastric cancer GWAS data as the outcome. Candidate signals were further evaluated using summary-data-based Mendelian randomization (SMR)/heterogeneity in dependent instruments (HEIDI) analysis and Bayesian colocalization analysis to assess the probability of shared causal variants between protein and gastric cancer signals. Independent European gastric cancer GWAS data were subsequently used for external validation and meta-analysis. TCGA-STAD, GTEx, and GEPIA2 datasets were further used to investigate tissue expression, prognostic relevance, and associated biological processes of the prioritized candidate gene.Results The MR screen identified 84 candidate proteins associated with gastric cancer risk, including 36 risk-increasing and 48 protective signals. SMR analysis further prioritized 11 positive proteins. Bayesian colocalization analysis showed that HGF had the strongest evidence for a shared causal variant (PP.H4=0.792). External GWAS validation and meta-analysis further supported the association between genetically predicted higher circulating HGF levels and increased gastric cancer risk (OR=1.224, 95% CI=1.070-1.401, P=0.003). Transcriptomic analyses demonstrated that HGF was upregulated in gastric cancer tissues, and high HGF expression was associated with poorer overall survival and disease-free survival. GSEA indicated that HGF-related genes were mainly enriched in extracellular matrix remodeling, angiogenesis, and immune regulatory processes.Conclusion This study constructed a genetic prioritization landscape of circulating proteins associated with gastric cancer risk and provided convergent genetic evidence supporting HGF as a prioritized gastric cancer-related candidate protein. These findings offer molecular clues for future risk stratification, mechanistic investigation, and clinical validation in gastric cancer.