Bone mineral density and the risk of breast cancer: the Rotterdam Study.

Estrogens play an important role in the development of breast cancer, but studies on serum levels of estrogens have shown inconsistent results. Bone mineral density is considered to be a marker for lifetime estrogen exposure. Some studies have suggested that a higher bone mass is associated with an increase in breast cancer risk. We investigated the association between bone mineral density (BMD), as measured at the lumbar spine and femoral neck, and the risk of breast cancer in women age 55 or older in the Rotterdam Study, a population-based cohort study in the Netherlands. Information on baseline lumbar spine and femoral neck BMD, as measured by DEXA (Lunar DPX-L), and cancer incidence was available for 3,107 women. The Rotterdam Cancer Registry provided information on follow-up of incident cancer. After an average follow-up time of 6.5 years, 74 new cases of breast cancer occurred. Z-scores of lumbar spine and femoral neck BMD were divided into tertiles and risk estimates for breast cancer were computed by the Cox proportional hazards model, using the middle tertile as a reference. Breast cancer risk in the upper tertile of lumbar spine BMD was doubled compared to the reference after adjustment for age, body mass index, and age at menopause (hazards ratio = 2.1 [1.1-3.7]), whereas risk estimates for women in the lower tertile did not significantly differ from the reference (hazards ratio = 1.5 [0.8-2.9]). Women with either a low intertrochanteric or femoral neck BMD appeared to have a somewhat decreased breast cancer risk, although this was not statistically significant. The results of this study suggest that in elderly women an association between especially lumbar spine BMD and incident breast cancer exists. Stimulating effects of estrogen on both trabecular bone and mammary cells may be responsible.