Peripubertal Anti-Mullerian Hormone Levels Are Associated With Hyperandrogenemia During Adolescence: The Avon Longitudinal Study of Parents and Children

Polycystic ovary syndrome (PCOS) is among the most common endocrine disorders in women and is associated with negative reproductive and metabolic outcomes including subfertility, pregnancy complications, metabolic syndrome, and type 2 diabetes. The diagnosis of PCOS cannot be made until reproductive maturity, when the diagnostic criteria of hyperandrogenemia and oligomenorrhea develop. However, studies have described early metabolic and reproductive characteristics of the disorder in girls at increased risk, suggesting the pathogenesis starts much earlier. Indeed, studies in animal models suggest that exposure to excessive androgen or anti-Mullerian hormone (AMH) levels during critical developmental periods can program the offspring to develop the metabolic and reproductive features of PCOS during reproductive maturity. We investigated early maternal or peripubertal factors associated with hyperandrogenemia during adolescence using data from the Avon Longitudinal Study of Parents and Children (ALSPAC), a United Kingdom birth cohort which has been ongoing since 1991. We performed linear regression to test for an association of testosterone levels at 15 years with peripubertal reproductive and metabolic phenotypes and with maternal measures of insulin sensitivity. Peripubertal phenotypes included AMH levels at 7, 9, and 11 years, and androstenedione, DHEAS, SHBG, IGF-1, fasting insulin, QUICKI, post-glucose insulin, leptin and adiponectin at age 8 years. Maternal phenotypes included fasting insulin levels and QUICKI at a post-partum visit. Unadjusted and adjusted analyses including the covariates pubertal stage, ethnicity, maternal and daughter BMI were performed. Testosterone levels at age 15 years were significantly positively associated with AMH levels at ages 7(N=299), 9(N=295), and 11(N=300) years in both the adjusted and unadjusted models (Age 7 unadjusted P<0.0001, adjusted P=0.01; Age 9 unadjusted P<0.0001, adjusted P=0.003; Age 11 unadjusted P<0.0001, adjusted P=0.02). Testosterone at age 15 years was also associated with DHEAS at age 8 years using the unadjusted (P<0.0001) but not the adjusted model. There was no significant association between any of the other peripubertal metabolic and reproductive phenotypes or the maternal metabolic phenotypes of fasting insulin and QUICKI with testosterone level at age 15 years. We have found a persistent and significant positive association of AMH levels at pre- or peri-pubertal ages with testosterone levels during adolescence, a developmental stage at which a clinical diagnosis of PCOS can be made. It remains unclear if this early elevation in AMH contributes to the pathogenesis of hyperandrogenemia or is an early marker of PCOS. Nonetheless, these findings suggest there are early differences in the reproductive phenotype in girls with hyperandrogenemia, even before the onset of puberty.