Comprehensive Genetic Assessment of Endometriosis: Integrating a GWAS Approach with Replication and Sub-Phenotype Evaluation

Endometriosis is a chronic, estrogen-dependent condition affecting approximately 10% of women of reproductive age, characterized by ectopic endometrial-like tissue that causes pelvic pain, dysmenorrhea, dyspareunia, gastrointestinal (GI) discomfort, and infertility. This study integrated comprehensive genotype data, high-quality imputation, and detailed phenotypic characterization in an Italian cohort of surgically and histologically confirmed endometriosis cases, aiming to elucidate the genetic architecture underlying both the overall disease and its diverse sub-phenotypes. Rigorous quality control (including checks for sex discrepancies, missingness rates, heterozygosity outliers, relatedness, and population stratification) yielded a robust post-imputation dataset encompassing over seven million single nucleotide polymorphisms (SNPs). Genome-wide association analyses were performed using logistic regression models adjusted for principal components, initially contrasting all endometriosis cases versus controls. While no SNP reached genome-wide significance in the overall disease phenotype, specific sub-phenotype analyses revealed two SNPs—rs185338542 near ACOT7 on chromosome 1 and rs138188726 within PCDH7 on chromosome 4—that achieved genome-wide significance in patients reporting GI pain. These findings suggest that alterations in lipid metabolism and cell-cell adhesion pathways contribute to symptom-specific manifestations, particularly GI-related complaints, in endometriosis. Additional efforts replicated 42 endometriosis-associated SNPs reported by a recent large-scale meta-analysis. Twelve of these showed nominal significance in the Italian cohort, and among them two loci—KDR/4q12 (VEGFR2) and CD109/6q13—remained robustly associated following multiple-testing corrections, underscoring the roles of angiogenesis and TGF-β signaling in endometriosis pathogenesis. Further stratification by rASRM stage (I/II vs. III/IV) showed that KDR retained significance across early and advanced disease, whereas certain loci, including CDKN2B-AS1/9p21.3, were implicated only in more severe forms. Likewise, distinct variants emerged when comparing lesion subtypes: WT1 and CEP112 were exclusive to ovarian endometriomas, while GREB1, ABO, RNLS, and IGF1 were specific to deep infiltrating endometriosis. Only a small subset of variants, including FSHB and CD109, exhibited cross-phenotype relevance, indicating shared pathogenic processes across phenotypic manifestations. These findings reinforce the significance of sub-phenotype-focused approaches, revealing how certain genetic factors are masked when endometriosis is treated as a homogeneous entity. By clarifying variant associations across GI pain, disease stages, and lesion types, this research refines the genetic landscape of endometriosis and emphasizes the complex interplay of angiogenesis, inflammatory pathways, and hormonal regulation. The identification of symptom-specific genetic determinants and the confirmation of core biological loci, such as KDR and CD109, highlight a nuanced disease architecture that may inform more targeted diagnostic and therapeutic strategies, ultimately improving patient outcomes in this clinically heterogeneous disorder.