Estrogen and
progesterone are key factors in the development of
breast cancer, but it remains unclear whether these
hormones are associated with mammographic density phenotypes in premenopausal women. We measured percent mammographic density, nondense area, and absolute mammographic density using computer-assisted breast density readings (Madena) from digitized mammograms taken on a scheduled day of the menstrual cycle (day 7-12) among 202 healthy, premenopausal women (Energy Balance and
Breast cancer Aspects Study-I). Daily salivary concentrations of 17β-estradiol and
progesterone throughout an entire menstrual cycle and fasting morning serum concentrations of
hormones on 3 specific days of the menstrual cycle were assessed. Salivary and serum 17β-estradiol and
progesterone were positively associated with percent mammographic density, we observed by 1 SD increase in overall salivary
estradiol (β-value equal to 2.07, P=0.044), luteal salivary
progesterone (β-value equal to 2.40, P=0.020). Women with above-median percent mammographic density had a 20% higher mean salivary 17β-estradiol level throughout the menstrual cycle. The odds ratio for having above-median percent mammographic density (>28.5%) per 1 SD increase in overall salivary 17β-estradiol was 1.66 (95% confidence interval 1.13-2.45). Women in the top tertile of the overall average daily 17β-estradiol concentrations had an odds ratio of 2.54 (confidence interval 1.05-6.16) of above-median percent mammographic density compared with women in the bottom tertile. Our finding of a relationship between
estrogen,
progesterone, and percent mammographic density and not with other mammographic density phenotypes in premenopausal women is biologically plausible, but needs to be replicated in larger studies.