Bidirectional Two-Sample Mendelian Randomization Study Reveals Causal Associations Between Aging and Endometriosis.

Previous studies have suggested that aging may influence reproductive functions of female. Nonetheless, the causal relationship between aging and endometriosis has yet to be completely understood. This study aims to determine whether aging had a causal association with the incidence of endometriosis. We conducted bidirectional MR analyses to evaluate the causal relationship between aging biomarkers, particularly leukocyte telomere length (LTL), and endometriosis risk. Instrumental variables for LTL were derived from the UK Biobank GWAS, while endometriosis-associated variants were obtained from the FinnGen GWAS dataset. Subgroup analyses were performed to investigate the association between LTL and endometriosis subtypes. Additionally, validation was performed using independent GWAS meta-analysis datasets. Inverse variance-weighted (IVW) analysis revealed a significant association between longer LTL and an increased risk of endometriosis (OR-IVW = 1.276, 95% CI: 1.143 to 1.424, FDR-adjusted P = 7.00E-5), with consistent findings across multiple MR methods. Sensitivity analysis using an independent GWAS meta-analysis dataset did not confirm the LTL-endometriosis association (OR-IVW = 1.128, 95% CI: 0.140 to 9.115, P = 0.910). Bidirectional MR analysis found no causal relationship between endometriosis and LTL. Subgroup analyses indicated that longer LTL was significantly associated with endometriosis of the ovary (OR-IVW = 1.343, 95% CI: 1.143 to 1.577, P = 3.00E-4) and endometriosis of the rectovaginal septum and vagina (OR-IVW = 1.336, 95% CI: 1.064 to 1.676, P = 0.013), while no significant association was found with endometriosis of the pelvic peritoneum. Our findings suggest that longer LTL may contribute to an increased risk of endometriosis, particularly in ovarian and rectovaginal subtypes. However, no causal effect of endometriosis on aging was observed. The lack of replication in independent datasets highlights the potential influence of population heterogeneity and dataset-specific factors, warranting further validation in diverse cohorts.