Genetically Validated Immune Susceptibility Markers for Crohn's Disease: A Multi-Omics Mendelian Randomization Analysis.

Background: Crohn's disease (CD) is a chronic inflammatory bowel disorder with a multifactorial genetic, immune, and environmental basis, yet robustly validated therapeutic targets remain limited. We applied a multi-omics Mendelian randomization (MR) framework and colocalization to prioritize genetically supported immune susceptibility loci for CD. Methods: A two-sample MR framework was used, integrating protein quantitative trait loci (pQTL) and expression quantitative trait loci (eQTL) datasets with genome-wide association study (GWAS) data on CD from the Finnish biobank. After identifying 994 pQTL genes, 610 genes overlapping with known druggable targets were selected. Causal associations with CD were evaluated using inverse variance weighted (IVW) MR. Significant hits were further validated through colocalization analysis and summary-data-based Mendelian randomization (SMR). Results: IVW analysis identified 56 pQTL genes significantly associated with CD risk, with six genes (including HLA-A, MST1, and CDH5) passing false discovery rate (FDR) correction. Colocalization analysis indicated that HLA-A and MST1 shared causal variants with CD. Subsequent eQTL-based MR and SMR analysis confirmed the causal association of HLA-A expression with increased CD risk (SMR OR = 1.434, p = 4.10 × 10-5), with no evidence of heterogeneity. Conclusion: These findings suggest that HLA-A represents a genetically validated immune susceptibility marker in CD. By integrating pQTL, eQTL, colocalization, and SMR analyses, it highlights the utility of multi-omics MR in uncovering novel genetic contributors to complex diseases. Further experimental and clinical validation is warranted to explore the translational potential of targeting HLA-A in CD treatment.