Steroid Hormones and Risk of Breast Cancer
Barbara S. Hulka, M.D., M.P.H.,* Edison T. Liu, M.D.,t and Ruth A. Lininger, M.D., M.P.H.$
Background. More than three decades of epidemiologic studies have identified numerous risk factors for breast cancer. These factors have been estimated to account for only 20-25% of disease occurrence. However, among these factors, several are related to sex steroid hormones: sex of the affected individuals [women), early age of menarche and late age of menopause, parity, late age at first pregnancy, and obesity in postmenopausal women. Methods. Theoretical models and laboratory data support hormonal mechanisms of carcinogenesis, particularly as they relate to proliferation of breast ductal epithelium and terminal end bud growth and differentiation in the lobules of the breast. The recent introduction of biologic markers and molecular epidemiology allows for studies that use laboratory technology in the context of epidemiologic research.
Results. This paper summarizes the epidemiologic literature on exogenous hormones, addresses the issue of endogenous steroid hormone levels and estrogen metabolism in serum and breast tissue in premenopausal and postmenopausal women with and without cancer, speaks to the cellular mechanisms of action of estrogen and progesterone, and highlights some of the biologic markers relevant to studies of breast cancer and precursor lesions, with particular emphasis on those that may be hormonally induced or altered.
Conclusions. These markers must be better defined in terms of breast cancer pathogenesis. Studies are needed to evaluate the direct effects of behavioral/environmental risk factors on relevant biomarkers as well as
Presented at the Conference on Breast Cancer Rcsearch: Current
Issues-Future Directions, Atlanta, Georgia, April 25-28, 1993.
From the 'Department of Epidemiology, School of Public
Health, and the tLineberger Comprehensive Cancer Ccnter, University of North Carolina, Chapel Hill; and the $Department of Pathology, Duke University Medical Center, Durham, North Carolina.
The authors thank Ms. Kathy Sutton for her word processing, editing, and bibliographic work.
Address for reprints: Barbara S. Hulka, M.D., M.P.H., Department of Epidemiology, School of Public Health, University of North
Carolina, CB 7400, McGavran-Greenberg Hall, Chapel Hill, NC
Accepted for publication April 14, 1994.
to assess the interactions of epidemiologic factors and biomarkers on risk of breast cancer. Cancer 1994;74:111124.
Key words: steroid hormones, breast cancer, biomarkers, epidemiology. Sex steroid hormones are undoubtedly involved in the genesis and progression of breast cancer. Data to support this statement are available from animal models in which estrogen administration increases the number of mammary neoplasms formed and the rapidity of their development. In vitro systems of cell lines derived from human breast tissue can be induced to develop malignant properties through estrogen administration alone or in conjunction with other hormone^.^,' This information is important from a clinical and public health standpoint to the extent that it is also applicable to women. We focus our attention on women, because the estimated number of new breast cancer cases for 1993 in the United States is 183,000 in women compared with 1000 in men, a ratio approaching 200 to l.5
Hormonal Risk Factors
The female breast is subject to a lifetime of hormonal controls, whose effect is evident at the time of menarche and during the menstrual cycle, pregnancy, and lactation. After menopause, breast tissue becomes quiescent when circulating sex steroid hormone levels drop to low
(estrogen) or nonmeasurable levels (progesterone).
Epidemiologists have studied and recorded multiple risk factors for breast cancer, some of which are a reflection of hormonally mediated events. The