Links between breast cancer and birth weight: an empirical test of the hypothesized association between size at birth and premenopausal adult progesterone concentrations.

Some studies have reported that birth size is a risk factor for breast cancer, but the reasons for this observation are unknown. Ovarian hormone concentrations may be a link between birth size and breast cancer, but the few tests of this hypothesis are inconsistent, perhaps because of differences in sample composition, inclusion of anovulatory cycles, or use of one hormonal measurement per woman. We present results from the first study to use daily hormonal measurements throughout a woman's complete ovulatory cycle to test the hypothesized relationship between birth size and adult progesterone concentrations. We used a study sample and accompanying data set previously obtained for another research project in which we had collected daily urine samples from 63 healthy premenopausal women throughout a menstrual cycle. Multivariate regression was used to test for trends of individual progesterone indices (from 55 ovulatory cycles) with birth weight or ponderal index, while controlling for age, adult BMI, and age at menarche. Our main finding was that neither birth weight nor ponderal index was associated with biologically significant variation in luteal progesterone indices; the best-estimated effect sizes of birth size on these progesterone indices were small (3.7-10.2%). BMI was the only significant predictor of mean peak urinary progesterone, but it explained <6% of the variance. Our findings, in light of what is currently known regarding associations of breast cancer risk with birth size and adult size, suggest that environmental factors (particularly those that vary by socioeconomic status and affect growth) may underlie associations between birth size and cancer risks without there being any association of birth size with adult ovarian hormone concentrations.