Chemotherapy can produce excellent palliation for many patients with metastatic breast cancer. Survival impact is, however, limited, and permanent remission is extremely rare. There is increasing evidence that dose and dose intensity may be important determinants of outcome in the chemotherapy of breast cancer. Single courses of chemotherapy in doses requiring autologous bone marrow support produce high rates of objective response in patients with metastatic disease that was refractory to prior standard-dose therapy. When used as first chemotherapy for metastases or as consolidation in patients whose disease is responding to lower-dose therapy, high-dose chemotherapy can result in prolonged disease-free survival for some patients. The major cause of treatment failure is relapse from a chemotherapy-induced complete response. Kinetic models suggest that multiple, rapidly cycled courses of high-dose chemotherapy might be superior to single applications or to multiple treatments that are widely spaced in time. Heretofore, the substantial toxicity of high-dose chemotherapy (up to 20% mortality in some early trials) has largely precluded the consideration of timely retreatment; however, the risk appears to have been reduced through the use of hematopoietic growth factors and peripheral blood progenitor cells. Our group has used these new technologies to develop regimens consisting of multiple cycles of high-dose chemotherapy that are rapidly administered. We are currently refining these regimens in preparation for phase II and III studies.