Gene expression profiling has indicated that the up-regulation of specific subsets of chemokines/cytokine/ growth factors is often linked to poorer prognosis for women with invasive mammary carcinoma. Chemokines are important for the recruitment of leukocytes to the tumor microenvironment (TME) which can affect invasion and metastasis. It has been proposed that specific chemokines released by circulating tumor cells signal to leukocytes in the pre-metastatic niche and entrain myeloid cells to exhibit an anti-tumorigenic phenotype. The entrainment process is thought to require activation of neutrophil NADPH oxidase to ultimately generate extracellular H2O2, which is toxic to tumor cells. This process is reported to be dependent upon activation of PI3 kinase. The goal of this research is evaluate mechanisms of entrainment and to characterize the role of inhibition of PI3 kinase in chemokine mediated entrainment of leukocytes. We hypothesize that perisurgical delivery of entraining chemokines to cancer patients may reduce recurrence and metastasis. Moreover, current PI3K inhibitor therapies for breast cancer patients might fail if they reverse this entrainment, counteract the effect of anti-tumor leukocytes, and thus promote metastasis. There are three specific aims for this study: 1) Using an immunocompetent mouse model we will determine whether manipulation of CCL2, CCL5, CXCL1, CXCL12 and/or TGF-beta levels will differentially effect the recruitment of anti- or pro-tumor leukocytes into the primary tumor and the pre-metastatic niche. Aim1B. We will determine whether blocking the receptors for each of these chemokines will facilitate or disrupt the metastatic capacity of mammary carcinoma; 2) Using immunocompetent mice, intravital imaging, and ex-vivo micro-bioreactors we will determine whether current therapeutics for breast cancer that inhibit PI3K are able to block or facilitate chemokine-mediated 'entrainment' of myeloid cells in tumor bearing mice; We will deliver PI3K inhibitors to tumor bearing mice, isolate myeloid cells, and determine whether neutrophils respond to entraining chemokines with induction of H2O2 and have the capacity to kill tumor cells; 3) We will evaluate changes in the myeloid 'entrainment' status of cancer patients before and after chemotherapy and determine whether leukocytes can be entrained 'ex vivo'. Using a humanized mouse model and implanted patient tumors, we will determine whether drug induced alterations in myeloid entrainment that result in reduced tumor cell killing and/or enhanced metastasis correlates with time to progression in the patient. In humanized mice, we will also compare the effects of ex vivo treatment with chemokines or PI3K inhibitors on 'entrainment' properties of circulating myeloid cells of cancer patients prior to surgery, after surgery, or perisurgery to determine the appropriate time to deliver entraining chemokines or PI3K inhibitors.