ABO and LINC00339 as shared causal genes between endometriosis, endometriosis subtypes and epithelial ovarian cancer: a genome-wide association study
Description
We analyzed genome-wide association study (GWAS) data to explore the shared genetic structure and causality association of EM and EM subtypes with EOC. Linkage disequilibrium score regression (LDSC) was used to assess genetic correlations, while pleiotropy and enrichment were evaluated through the pleiotropic analysis under composite null hypothesis (PLACO), the multi-marker analysis of genoMic annotation (MAGMA), the Functional Mapping and Annotation of Genetic Associations (FUMA) and colocalization analyses. Mendelian randomization (MR) was applied to assess causal relationships. The positive genetic associations of EM, adenomyosis (AM), ovarian endometriosis (OE), pelvic peritoneal endometriosis (PPE) and unspecified/other endometriosis (UOE) with EOC were identified, while no correlations of deep endometriosis (DE), fallopian tube endometriosis (FTE), intestinal endometriosis (IE) and endometriosis of the rectovaginal septum and vagina (ERSV) with EOC were found. PLACO and FUMA analysis identified 6 independent genomic risk loci for EM and EOC (EM-EOC), 2 for AM and EOC (AM-EOC), 5 for OE and EOC (OE-EOC), 6 for PPE and EOC (PPE-EOC), and 2 for UOE and EOC (UOE-EOC), 2 of which had strong evidence of colocalization (ABO and LINC00339). Additionally, PLACO and MAGMA analyses revealed that pleiotropic genes associated with EM-EOC, OE-EOC, PPE-EOC, and UOE-EOC were particularly enriched in ovarian tissues. Meanwhile, pleiotropic genes linked to OE-EOC were also specifically enriched in fallopian tubes. Pathway analysis determined the key role of the proteoglycans in cancer and pathways in cancer for EM-EOC, OE-EOC, and PPE-EOC. Finally, the MR analysis demonstrated that EM and EM subtypes causally increase the risk of EOC, while no evidence supported reverse causation.