With the follow-up of 116,430 women enrolled in the Nurses' Health Study II now supported by a cooperative agreement (UM1CA176726), this competing renewal continues the scientific pursuit of modifiable determinants of breast cancer risk, especially those acting during the premenopausal years. In this unique cohort, exposure information has been collected at two-year intervals, beginning in 1989 when women were 25-42 years of age, with 93% cumulative follow-up. Our focus on younger women with 28 years of follow-up allows examination of early adult exposures and breast cancers diagnosed before and after menopause; through 2017 we expect 6,232 incident breast cancer cases. Our extensive biorepository, with blood and cheek samples, tumor tissue, and mammograms, allows us to thoroughly examine relationships between lifestyle factors, biomarkers, intermediate endpoints, and tumor expression. In this proposal, we aim to identify dietary and metabolic factors associated with breast cancer risk, and to examine the underlying mechanistic pathways by incorporating plasma metabolomic profiles and tumor subtypes. The metabolome, defined as the set of small molecule metabolites that are the final downstream products of the genome, but which is also strongly influenced by diet and other modifiable factors, is critical for growth and maintenance of cells, tissues and entire organisms. Measuring specific metabolites in plasma, we propose to examine branched chain amino acids (valine, leucine, isoleucine) as well as evaluate more broadly metabolomic signatures that may be important in breast carcinogenesis. Building on previous work identifying dietary factors associated with breast cancer and diabetes, we will examine how these influence breast cancer through their effect on the metabolome, including intake of branched chain amino acids, dietary breast cancer risk factors (alcohol, red meat, fruits, vegetables), and a dietary diabetes risk reduction score. We will interrogate the PI3K/Akt/mTOR signaling pathway that regulates cell growth, proliferation, invasion, and survival and is important in breast cancer. Mutations in the PIK3CA gene or loss of PTEN, events present in 50-75% of breast tumors, lead to constitutively active mTOR. Dietary and other modifiable risk factors that influence this pathway would be expected to be associated specifically with tumors that do not already have an activated PI3K pathway due to these genetic changes. To evaluate the possible role of these mechanisms, we will examine metabolomic, dietary, and lifestyle factors in relation to breast cancer risk, incorporating tumor PIK3CA mutations and loss of PTEN. For any significant associations we observe, we will examine breast density as a mediator between exposures and breast cancer. Our aims will substantially enhance our understanding of the role of dietary factors, the metabolome, and the PI3K pathway in the etiology of breast cancer. These aims will identify potential preventive strategies, including dietary changes, and new metabolic targets for risk prediction and chemoprevention.