Radiation therapy is a critical component of treatment of the majority of cancer patients. For many years, quantitative models of the action of this therapy have been developed that define the effect of radiation on tumors in terms of the fraction of cells surviving a given radiation treatment. However, recent research has shown that metastatic, circulating tumor cells may return to the parent tumor and "re-seed" it, in a process that has been termed "tumor self-seeding". We hypothesize that tumor self-seeding may provide a mechanism for tumors to regrow after radiotherapy, through a process that is stimulated by radiation. This idea is supported by in vitro and in vivo data from our group showing that irradiated tumors attract migratory tumor cells through the radiation-inducible expression of the cytokine GM-CSF. The objective of this research is to evaluate this clinically important hypothesis, in order to determine whether radiation may ultimately contribute to tumor regrowth after treatment through attraction of circulating tumor cells. We will pursue this goal through fou specific aims. The first will be to rigorously and quantitatively characterize radiation-induced tumor self- seeding through the use of novel models of tumor metastasis and radiotherapy, utilizing subcutaneous, orthotopic, and spontaneous mouse models of cancer in conjunction with bioluminescence imaging and conformal small animal radiotherapy. We will assess the sensitivity of this process to radiation dose, tumor type, tumor location, and timing. In the second aim, we will investigate the molecular and cellular mechanisms by which GM-CSF facilitates this process, including receptor-mediated signaling in tumor cells and cooperation with macrophages. With an understanding of this mechanism, specific aim 3 will then engineer therapies that attempt to interfere with it in a clinically-relevant manner. Finally, we will investigate GM-CSF and GM-CSF receptor signaling in human breast and lung cancer patients through a pilot clinical study, in order to assess the clinical significance of this process. This work represents a new direction in the study of radiotherapy for cancer and may shed new light on why some tumors recur following radiobiologically curative courses of radiation. Understanding this process will allow more effective prescription of radiotherapy in consideration of patient's metastatic profile, improving control of cancer in these patients. Furthermore, identification and development of therapeutics that counteract this process may further enhance outcomes following radiation treatment.