Abstract 4142520: Causal associations between age of menarche or sex hormones and the risk of cardiovascular mortality: A Mendelian randomization analysis

Background: Recently, observational studies have reported several sex-specific factors associated with risk of cardiovascular (CV) disease and mortality. In addition, endogenous sex hormone levels in both men and women are related to the risk of CV disease and mortality. We explored the causal relationships between several sex-specific factors or sex hormones and CV mortality. Methods: Among 219,320 women or 187,297 men registered in the UK biobank database, we investigated two-sample Mendelian randomization analyses for outcome data (CV mortality). Exposure data were extracted from published genome-wide association studies (GWAS) catalog database. We selected each SNP with sex-specific (age at menarche, age at menopause in women) factors and sex hormones (total testosterone, sex hormone-binding globulin adjusted for BMI, Estradiol, dehydroepiandrosterone sulfate (DHEAS), androsterone sulfate in both men and women), to use it as an instrumental variable. Mendelian randomization analysis was performed using the inverse variance weighted (IVW), MR-Egger regression, and weighted median methods. Results: Among the sex-specific factors, we found that genetically predicted age at menarche was causally associated (IVW, beta=−0.134, SE=0.062, P=0.031), but not the age at menopause (IVW, beta=-0.028; SE=0.019, P=0.131) with the risk of CV mortality in women. Among the several sex hormones, androsterone sulfate has a causal relationship with CV mortality (IVW, beta=0.272, SE=0.089, P=0.002), but not other sex hormones (IVW, P=0.498 for total testosterone; P=0.469 for sex hormone-binding globulin adjusted for BMI; P=0.441 for Estradiol (men); P=0.099 for DHEAS). Conclusions: These results provide evidence that the earlier age at menarche in women and higher androsterone sulfate levels are associated with increased CV mortality. Key words: sex specific factors, sex hormones, cardiovascular mortality, Mendelian randomization , genetic variants