3. MODIFICATION OF EXPOSURES BY NONGENETIC FACTORS The primary focus of this chapter will be the effects of genetic variability on associations between risk factors and breast cancer. However, in addition to genetic factors, demographic or lifestyle variables may impact the effects of exposures on ultimate breast cancer risk. The age at which exposures occur may influence breast cancer risk and the impact of risk factors may vary depending on the age at which breast cancer is diagnosed.

3.1. Timing of Exposures The timing of carcinogenic exposures may be critical to risk of breast cancer. In rats, carcinogens administered before a first pregnancy result in twice the tumor load than in rats exposed after mammary cell differentiation (2). As reviewed by Colditz and Frazier (3) and confirmed by Marcus et al. (4,5), studies of the effects of irradiation, alcohol consumption, and cigarette smoking have shown that breast cancer risk is increased by exposure at an early age. Colditz suggests that genetic damage resulting from exposures before a first pregnancy may be immortalized by cell proliferation during breast development and pregnancy, and that the decrease in cell turnover following pregnancy may prevent further genetic damage, thus reducing risk. An understanding of the importance of the age at which exposures occur may be of key importance for strategies in cancer prevention.

3.2. Age at Diagnosis of Breast Cancer It is possible that etiologic pathways may differ for pre- and postmenopausal breast cancer, with the effects of specific risk factors having differential effects by menopausal status. For example, research indicates that there is a crossover effect of BMI by menopausal status (6-9). As reviewed by Hunter and Willett (10), the majority of breast cancer studies have found that higher BMI increases risk for postmenopausal women, but leaner women are at increased risk for premenopausal breast cancer. Relationships between exposures and breast cancer risk may also vary between subsets of individuals depending upon numerous other modifying factors, a few of which include dietary intakes of macro- and micronutrients, smoking, and reproductive histories. Another extensively studied putative effect modifier of breast cancer risk is a history of breast cancer in a first-degree relative.