Abstract P3-11-16: First Pregnancy Events Determine Future Breast Density Independent of Systemic Sex Steroid Levels

Abstract

Abstract Introduction: Pregnancy outcomes (e.g., preterm birth, preeclampsia, and birthweight) have been associated with altered breast cancer risk. Hypotheses linking pregnancy outcomes and subsequent breast cancer risk include persistent changes in endogenous hormones and/or imprinted changes in mammary gland differentiation and morphology. Limited information exists on the relationship between pregnancy characteristics and mammographic density. The Marin Women's Study (MWS) was initiated in 2006 to examine breast cancer in a mammography population in an area with high breast cancer incidence (Marin County). MWS enrollment occurs at mammography sites in Marin County that participate in the San Francisco Mammography Registry, one of seven registries included in the National Cancer Institute Breast Cancer Surveillance Consortium. The MWS includes ∼14,000 participants, covering ∼80% of the County's mammography screening population. An analysis was undertaken using MWS data to examine the relationship between first pregnancy outcomes and later life mammographic density, controlling for standard breast cancer risk factors. Methods: Reproductive characteristics were self-reported on the MWS questionnaire at the time of mammogram. Volumetric compositional breast density (BDsxa) using single X-ray absorptiometry techniques (SXA) was measured on screening digital mammography examinations measured by the San Francisco Mammography Registry. Systemic sex steroid hormone levels were measured on a single early morning saliva sample. Women were excluded if they reported a history of breast cancer, if their first birth was a multi-fetal gestation, if they reported using antiestrogens within the five years preceding the questionnaire, or if they were missing data on any of the model variables. Robust regression models were controlled for current age, body mass index (BMI), race, education, physical activity, alcohol consumption, smoking, family history of breast cancer, hysterectomy, menopausal status, and hormone use at the time of mammography. Results: The study population included 2162 parous women. Using multivariable regression we found that a history of gestational hypertension during first pregnancy was associated with a statistically significant decrease in BDsxa (b=-0.29, p=0.01), while each month of breastfeeding after first birth was associated with a statistically significant increase in BDsxa (b=0.01, p=0.01). Menopausal status (pre-and peri-vs. postmenopausal) and age at first birth (<30 vs. 30+ y) modified these associations. Reproductive factors explained a greater proportion of the variability in BDsxa in premenopausal than in postmenopausal women (40% vs 31%), and a greater proportion of variability in those with a later age at first birth (45% of age 30+ y vs. 36% of age <30y). Salivary hormone levels (E2, P, T, DHEA) were not consistently associated with breast density. Conclusions: This analysis demonstrates that a woman's first pregnancy outcomes may determine future breast density, and that this association varies by age at first birth and menopausal status at the time of breast density measurement. These findings suggest that pregnancy may exert its influence on future breast cancer risk via changes in breast morphology and density. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P3-11-16.