Evaluation of Circulating Proteins in Thyroid Cancer: Proteome-wide Mendelian Randomization and Colocalization Analyses

<b>Background: </b>Amidst a concerning upward trend in thyroid cancer (TC) cases worldwide. Investigating the shared genetic components between proteins and TC enhances our understanding of the disease's pathogenesis.<b>Methods:</b> We employed five large-scale multi-ethnic circulating proteomic genome-wide association studies (GWASs) with up to 100,000 participants and a TC Meta-GWAS (n_Case=3418, n_Control=292703) for bidirectional Mendelian Randomization (MR) and Bayesian colocalization analysis. Thyroid tissue-specific protein and gene expression were validated using data from our multi-omic cohort and public databases.<b>Results:</b> A total of 1525 proteins replicated in both discovery and replication cohorts were included. Through MR analysis, we identified 26 circulating proteins with a putative causal relationship with TCs, with NANS protein passing multiple corrections in both cohorts (<i>P</i>_BH=3.28e-5, 0.05/1525). These proteins were enriched in amino acids, carboxylic acids, sulfur-containing compounds metabolism, and organic acid synthesis pathways. Bayesian colocalization analysis further identified six proteins with moderate to strong evidence (PPH4 &gt; 0.5) of association with TCs (VCAM1, LGMN, NPTX1, PLEKHA7, TNFAIP3, and BMP1). Tissue-specific validation confirmed the differential expression of several identified proteins, notably BMP1, LGMN, and PLEKHA7, between normal and TC tissues.<b>Conclusions: </b>Our study indicates limited evidence for a link between circulating proteins and the risk of TCs. We have highlighted the contribution of circulating proteins, particularly those involved in amino acid metabolism pathways and genes, to TCs. This research has revealed several putative causal relationships between specific proteins and TC risk.背景：全球甲状腺癌（thyroid cancer, TC）病例数正呈现令人担忧的上升态势。解析蛋白质与甲状腺癌共有的遗传组分，有助于深化对该疾病发病机制的认知。 方法：本研究纳入5项大规模多族裔循环蛋白质组全基因组关联研究（Genome-Wide Association Studies, GWAS），参与者规模最高达100000例，同时纳入1项甲状腺癌荟萃全基因组关联研究（病例组n=3418，对照组n=292703），用于开展双向孟德尔随机化（Mendelian Randomization, MR）分析与贝叶斯共定位分析。借助本研究多组学队列数据及公共数据库资源，对甲状腺组织特异性蛋白质与基因表达情况进行验证。 结果：最终纳入1525种在发现队列与验证队列中均得到重复验证的蛋白质。通过孟德尔随机化分析，本研究共鉴定出26种与甲状腺癌存在潜在因果关联的循环蛋白质，其中NANS蛋白在两个队列中均通过多重校正（P_BH=3.28×10^-5，校正阈值为0.05/1525）。上述蛋白质显著富集于氨基酸、羧酸、含硫化合物代谢及有机酸合成通路。贝叶斯共定位分析进一步鉴定出6种与甲状腺癌存在中等到强关联证据（PPH4>0.5）的蛋白质，分别为VCAM1、LGMN、NPTX1、PLEKHA7、TNFAIP3及BMP1。组织特异性验证结果证实，部分鉴定得到的蛋白质在正常甲状腺组织与甲状腺癌组织间存在差异表达，其中以BMP1、LGMN及PLEKHA7最为显著。 结论：本研究表明，循环蛋白质与甲状腺癌患病风险之间的关联证据较为有限。本研究强调了循环蛋白质（尤其是参与氨基酸代谢通路的蛋白质及相关编码基因）在甲状腺癌发生发展中的作用。本研究揭示了特定蛋白质与甲状腺癌患病风险之间的数种潜在因果关联。