Abstract Background: The role of androgens in breast cancer etiology has been hypothesized but not confirmed. Androgen receptors (AR) have important physiological effects in normal breast development and are expressed in malignant breast tissue. Clinical studies indicate that AR may be a prognostic factor for breast cancer survival; however, studies addressing its role in breast cancer development are limited. Alcohol consumption, a confirmed breast cancer risk factor, has been hypothesized to exert its effect, at least partially, through the AR signaling pathway. We previously reported that alcohol consumption during adult life was positively associated with breast cancer risk (per 10 grams/day: hazard ratio [HR] = 1.10, 95% confidence interval [CI] = 1.07-1.12), and the association appeared stronger for estrogen receptor (ER) positive tumors. However, no epidemiologic studies have so far investigated alcohol and breast cancer risk according to AR status. Methods: We assessed this association in a prospective study of 105,858 women enrolled in the Nurses' Health Study. Alcohol consumption was first assessed through food frequency questionnaire at baseline (1980) and updated six times during follow-up; cumulative average alcohol use was calculated by averaging alcohol intake over time starting at baseline. Tumor AR status was determined through immunohistochemical staining using sections from a tumor tissue microarray (TMA); ER and progesterone receptor (PR) status was obtained from the TMA, or secondarily from the medical record pathology report. We applied a constrained competing risk survival model that allows for assessment of the alcohol and breast cancer association by AR status while controlling for ER and PR status, as well as other breast cancer risk factors. Results: During over 2.2 million person-years of follow-up from 1980 to 2006, 2706 invasive breast cancer cases with complete tumor marker data (i.e. AR, ER and PR) contributed to the analysis. When considering AR separately without adjustment of ER and PR status, alcohol consumption was positively associated with risk of invasive breast cancer among tumors expressing AR but not among AR-negative tumors (per 11 grams/day [equivalent to 1 drink/day] HR [95%CI]: 1.10 [1.05 - 1.16] and 1.02 [0.93 - 1.12], respectively; P-value for heterogeneity test = 0.14). After taking into account both ER and PR status, alcohol intake remained positively associated with risk of invasive breast cancer among AR-positive tumors but not among AR-negative tumors (per 11 grams/day HR [95%CI]: 1.10 [1.05 - 1.15] and 1.04 [0.96 - 1.12], respectively). However, the difference in relative risks between AR-positive and AR-negative tumors was not statistically significant (P-value for heterogeneity = 0.24). Conclusion: Our data suggest a positive association between alcohol intake and breast cancer risk for AR-positive tumors, but larger studies will be needed to confirm this. Citation Format: Jun Wang, Xuehong Zhang, Andrew H. Beck, Laura C. Collins, Wendy Chen, Rulla M. Tamimi, Aditi Hazra, Bernard Rosner, Susan E. Hankinson. Lifetime alcohol consumption and risk of breast cancer, by tumor androgen receptor expression. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 1270. doi:10.1158/1538-7445.AM2014-1270