Epidemiologic and Genetic Associations of Endometriosis With Depression, Anxiety, and Eating Disorders

Abstract

ABSTRACT Endometriosis is a chronic gynecologic disease that causes pelvic pain and is often seen in association with depression and anxiety. The association with these mental health conditions was previously linked to chronic pain, although some evidence suggests this is not the sole factor. A genetic correlation has shown a potential causal effect of depression on endometriosis; however, further analyses are needed to determine whether pleiotropy observed is due to cause-effect relationships or to shared biological pathways. This genetic association study (GWAS) aimed to conduct phenotypic and genetic association analyses on the comorbidity of endometriosis with depression, anxiety, and eating disorders. Individual-level data were obtained from the UK Biobank (UKB) and combined with genome-wide association statistics from data sets including the Psychiatric Genomics Consortium, the Million Veteran Program, the FinnGen study (Finland), and the CHARGE consortium. Individuals with endometriosis were identified by the International Classification of Diseases Tenth Revision codes as well as self-reported diagnoses from UKB surveys, and individuals with depression, anxiety, and eating disorders were identified using the International Classification of Diseases Tenth Revision codes. Logistic regression models were used to evaluate the association, and bidirectional 1-sample Mendelian randomization was used to test causal associations between endometriosis and anxiety, depression, and eating disorders. The GWAS studies were conducted for each of the mental health disorders being studied, and the Scalable Genetic Correlation Estimator (SCORE) method was used to compute single-nucleotide variant (SNV)–based heritability and genetic correlation from individual genotype and phenotype data. The linkage disequilibrium score regression method was used to calculate SNV-based heritability and genetic correlation using GWAS data. A total of 8276 participants with endometriosis (mean age, 53.1 years) and 194,000 controls (mean age, 56.7 years) were included in this analysis. Regression models accounting for demographic and obstetric covariates, as well as psychiatric comorbidities, found endometriosis to be associated with increased odds of depression (odds ratio [OR], 3.61; 95% confidence interval [CI], 3.32–3.92), eating disorders (OR, 2.94; 95% CI, 1.96–4.41), and anxiety (OR, 2.61; 95% CI, 2.30–2.97). The GWAS studies showed endometriosis to be correlated with depression (genetic correlation [rg] = 0.36, P = 1.5 × 10−9), anxiety (rg = 0.33, P = 1.17 × 10−5), and eating disorders (r = 0.61, P = 0.03). The SCORE analysis of UKB phenotypic and genetic data calculated female-specific SNV-based heritability (SNV-h2) for endometriosis (mean [SE] SNV-h2, 0.086 [0.015]), depression (SNV-h2, 0.019 [0.003]), anxiety (SNV-h2, 0.012 [0.002]), and eating disorders (SNV-h2, 0.004 [0.002]). When applying linkage disequilibrium score regression to UKB genome-wide association statistics, endometriosis showed a consistent genetic correlation with anxiety (rg = 0.36, P = 3 × 10−4) and depression (rg = 0.34, P = 1 × 10−5). Mendelian randomization found significant associations for depression (OR, 1.09; 95% CI, 1.08–1.11) and anxiety (OR, 1.39; 95% CI, 1.13–1.65) with endometriosis, but no association with eating disorders (OR, 0.73; 95% CI, 0.04–1.43) was observed. Only the pleiotropic variant DGKB rs12666606 between endometriosis and depression survived genome-wide multiple testing correction. The results of this GWAS show that eating disorders, depression, and anxiety are all associated with endometriosis and that genetic correlations were in line with phenotypic associations.