Anti-Mullerian Hormone and Cardiometabolic Disease in Women: A Two-Sample Mendelian Randomization Study

Background: Higher age-specific circulating anti-Mullerian hormone (AMH) levels have been linked to a lower risk of cardiometabolic outcomes. However, whether AMH has a casual role in the etiology of these diseases is unknown. The objective of this study was therefore to explore if circulating AMH levels have a causal effect on risk of coronary artery disease (CAD), ischemic stroke and type 2 diabetes (T2D) in women, using a two-sample Mendelian randomization (MR) approach. Methods: We used four single nucleotide polymorphisms (SNPs) from the most recent AMH GWAS meta-analysis as instrumental variables. Summary-level data for CAD (n=149,752; 11,802 cases), ischemic stroke (n=17,541; 4678 cases) and T2D (n=464,389; 30,052 cases) were extracted from the UK Biobank, the Stroke Genetics Network, and DIAMANTE consortia, respectively. To assess the presence of potential pleiotropy we tested the association of the four AMH SNPs, both individually and combined in a weighted genetic risk score, with a range of cardiovascular risk factors and intermediate traits using UK Biobank data. Results: MR estimates, i.e., inverse variance-weighted odds ratios (ORIVW), did not support a causal effect of circulating AMH levels on CAD (ORIVW=1.13, 95% CI: 0.95-1.35), ischemic stroke (ORIVW=1.11, 95% CI: 0.83-1.49), and T2D (ORIVW=0.98, 95% CI: 0.87-1.10). After adjustment for multiple testing, we observed associations between genetically predicted AMH and age at menopause, and age at menarche, but not with intermediate traits on the causal pathway between AMH and cardiometabolic health, such as atherosclerosis or glucose levels. Conclusions: This study does not provide evidence for a causal effect of circulating AMH levels on CAD, ischemic stroke and T2D in women, although weak instrument bias cannot be excluded.