Shared Genetic Architecture and Pleiotropic Loci Linking Endometriosis and Inflammatory Bowel Disease: An Integrative GWAS, Colocalization, and Mendelian Randomization Study
Background Endometriosis (EMS) and inflammatory bowel disease (IBD) are both chronic inflammatory disorders with overlapping clinical features, suggesting a potentially shared etiology. Although epidemiological studies have reported an association between the two conditions, the genetic basis of this relationship remains poorly understood. This study is aimed at characterizing the shared genetic architecture of EMS and IBD, identify pleiotropic loci implicated in both conditions, and explore the potential causal relationships and underlying biological mechanisms. Methods We analyzed large‐scale GWAS summary statistics to investigate the genetic relationship between EMS and IBD. Linkage‐disequilibrium score regression (LDSC) was used to estimate genetic correlation. The PLACO method was applied to identify pleiotropic loci, which were further evaluated by Bayesian colocalization. Functional implications were assessed using MAGMA gene‐set analysis, tissue enrichment analysis, and functional annotation. Bidirectional two‐sample Mendelian randomization (MR) was performed to examine causal inference, accompanied by leave‐one‐out sensitivity analyses and statistical power calculations. Results A statistically significant but modest positive genetic correlation was observed between EMS and IBD (rg = 0.0957, p = 0.0042). Twenty‐four genomic loci showed evidence of pleiotropic effects, and six of these carried strong evidence of a shared causal variant (posterior probability: PP.H4 > 0.7), including loci at 2p23.3 (ADCY3/DNAJC27), 4q12, 5q23.3, 7p15.2, 11q13.1 (CCDC88B), and 1q22. Functional analyses consistently implicated eight pleiotropic genes across multiple platforms: ADCY3, CCDC88B, GREB1, NOD2, RSPO3, SYNE1, THADA, and TRAIP. Pathway enrichment highlighted “positive regulation of gene expression” as a key shared mechanism (p < 0.05). Tissue‐specific analyses revealed strong signals in whole blood, spleen, and colon, consistent with the immunoinflammatory etiology of both diseases. Bidirectional MR did not support a causal relationship in either direction (IBD on EMS: OR = 1.007, 95% CI: 0.986–1.028, p = 0.543; EMS on IBD: OR = 1.054, 95% CI: 0.991–1.122, p = 0.092). Leave‐one‐out analyses confirmed that no single variant drove these estimates, and power analysis indicated adequate power to detect moderate causal effects. Conclusions The findings provide evidence for a shared, albeit modest, and genetic basis between EMS and IBD that is likely mediated through immune regulation, hormonal signaling, and tissue repair. The absence of significant causal effects in MR suggests that the observed comorbidity reflects common genetic susceptibility rather than a direct causal link. These results offer new insight into the mechanisms underlying the co‐occurrence of the two conditions and may inform future investigation of shared therapeutic targets.