Breast cancer is the most commonly diagnosed malignancy in American women with ~229,000 new diagnoses and ~39,000 deaths expected in 2012. Further, breast cancer incidence is projected to increase by ~45% over the next two decades. Although tamoxifen is considered the standard chemopreventive strategy for women at high-risk of breast cancer, it is medically contradicted in up to ~30% of high-risk women, associated with several adverse toxicities, and only effective for prevention of estrogen receptor positive (ER+) breast cancer. Thus, there is an urgent need for new, non-toxic strategies to prevent both ER+ and ER negative (ER-) breast cancer. Observational studies indicate that regular exercise is consistently associated with a 20%-30% reduction in both ER+ and ER- breast cancer incidence in both pre-and postmenopausal women. However, the host and/or tumor-related mechanisms that underline this association as well as the optimal 'dose' to modulate these effects have not been elucidated. Against this background, the objective of this grant is to elucidate the 'dose-response' effect of exercise on inhibition of breast cancer initiaton and the role of growth factor withdrawal to regulate this effect in preclinical and clinical studie. Our central hypothesis is that exercise, in a dose-dependent manner, will reduce proinflammatory growth factor availability in the host (peripheral circulation) and mammary tissue that, in turn, will effectively inhibit 'core' cell signaling pathways that induce the transtion to the malignant phenotype. This central hypothesis will be addressed by three translational aims that span laboratory to clinical studies. Aim 1 (preclinical): Determine the 'dose-response' of exercise on tumor inhibition in genetically-distinct transgenic models of breast cancer; Aim 2 (clinical): Determine the 'dose-response' of exercise training on mammary tissue growth factor profile in a randomized trial among women at high-risk of breast cancer (i.e., cytological confirmed atypia); and Aim 3 (mechanistic): Identification and validation of growth factor and signaling pathway modulation using targeted molecular analyses (IHC, RT-PCR) in combination with unbiased approaches [i.e., Reverse Phase Protein Microarray]. To test the mechanistic role of exercise-induced growth factor withdrawal, we will over express growth factor availability or growth factor receptor activity using genetic (knock-in models) or pharmacological (specific agonistic antibodies) strategies. This sub-aim will be addressed using in vitro experiments using human premalignant cell lines exposed to serum from patient samples obtained in Aim 2 as well recombinant growth factors using 'add back' experiments.