Pubertal timing and incident uterine cancer in the Sister Study cohort

Abstract

Younger age at menarche is an established uterine cancer risk factor. Age at onset of breast development (thelarche), the earliest marker of pubertal “unopposed” estrogen exposure, may also be relevant to uterine cancer risk, particularly considering rapid declines in age at thelarche over time in parallel with increasing uterine cancer incidence rates. Using data from 34,152 participants with an intact uterus when they enrolled in the US Sister Study cohort (2003–2009; ages 35–74 years), we examined associations of self-reported ages at thelarche and menarche with incident uterine cancer using multivariable-adjusted Cox proportional hazards regression. We stratified by birth cohort, race, weight relative to peers in childhood, and body mass index (BMI) at enrollment to explore potential effect measure modification and tested for statistical heterogeneity across strata. During follow-up (median = 13.3 years), 445 women reported an incident uterine cancer diagnosis. Ages at thelarche (hazard ratio [HR] per 1-year older: 0.91, 95% confidence interval [CI]: 0.85–0.97) and menarche (HR per 1-year older: 0.90, 95% CI: 0.84–0.96) were inversely associated with uterine cancer incidence. Associations were similar in non-Hispanic Black and White women and did not vary by relative childhood weight or BMI in adulthood. Inverse associations of thelarche and menarche were limited to women born in 1950 or later (p-het