Sex Hormones and Risk of Estrogen Receptor (ER)-Positive and ER-Negative Breast Cancer.

Abstract

Abstract Background: There is a need for improved breast cancer risk prediction by tumor estrogen receptor (ER) status. Endogenous sex hormone levels are associated with risk of overall breast cancer and the ER-positive subtype. However, the role of estradiol (E2) and testosterone (T) in the development of ER-negative tumors remains unclear. We investigated the associations of circulating levels of endogenous E2 and T with the risk of ER-positive as well as ER-negative breast cancer in postmenopausal women.Methods: We performed a case-cohort study within the Women's Health Initiative- Observational Study of postmenopausal women not taking exogenous hormones (age range: 50-79 years). Baseline endogenous levels of bioavailable E2 and T were measured using radioimmunoassays (University of Southern California, LA, CA) in 196 cases of invasive ER-positive breast cancer, 107 cases of invasive ER-negative cancer, and 560 randomly selected controls. The average follow-up time was 7.3 years.Results: After adjustment for putative risk factors, women with E2 levels in the upper three quartiles had an approximately 2-fold increased risk of ER-positive breast cancer, compared to those with E2 in the lowest quartile. These associations were only modestly decreased after adjustment for T. Higher T levels also indicated increased risk of ER-positive breast cancer; women in the third quartile of T had a 1.7-fold significantly higher risk, compared to those in the first quartile. However, this association was substantially diminished and not significant after controlling for E2.No association was observed between E2 levels and ER-negative cancer. However, women with T levels in the second, third, and fourth quartiles had lower risks of ER-negative cancer in the order of 54% (p= 0.018), 38% (p= 0.13), and 46% (p= 0.05), respectively, compared to women with T in the lowest quartile. These associations did not change materially after adjusting for E2.Table 1. Risk (Hazard Ratio (HR) and 95% CI) of ER-positive and ER-negative breast cancer by E2 and T quartiles ER-positive (HR (95% CI))ER-negative (HR (95% CI)) Model 1*Model 2**Model 1*Model 2*E2 Q11.001.001.001.00Q22.19 (1.25-3.84)2.12 (1.16-3.87)0.57 (0.29-1.12)0.72 (0.34-1.53)Q31.92 (1.10-3.35)1.74 (0.92-3.33)0.75 (0.40-1.42)1.09 (0.51-2.3)Q42.11 (1.21-3.68)1.86 (0.97-3.56)0.88 (0.48-1.62)1.36 (0.60-3.08)P for trend0.020.150.840.44T Q11.001.001.001.00Q21.01 (0.57-1.78)0.82 (0.44-1.53)0.46 (0.24-0.88)0.45 (0.22-0.92)Q31.72 (1.04-2.84)1.36 (0.75-2.48)0.62 (0.34-1.15)0.56 (0.27-1.18)Q41.45 (0.85-2.46)1.16 (0.62-2.17)0.54 (0.29-1.00)0.44 (0.20-1.00)P for trend0.050.280.100.09*Model1: adjusted for age, race, age at menopause, alcohol use, physical activity, history of needle aspiration, lifetime use of estrogen+progestin, and time since quitting hormone therapy use. **Model2: adjusted for Model1 + sex hormones.Conclusion: Higher endogenous E2 levels were associated with increased risk of ER-positive breast cancer, independent of risk factors and T. In contrast, higher concentrations of endogenous T were related to lower risk of ER-negative breast cancer, independent of risk factors and E2. This is the first study to report an association of testosterone with ER-negative breast cancer. Further studies are needed to confirm this association. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 907.