Over the past two decades, the investigation of high-dose chemotherapy and autologous hematopoietic stem cell transplantation for breast cancer has followed the path of a roller-coaster ride. Initial results from numerous initial Phase II studies were remarkable: the response rate in patients with metastatic disease who underwent high-dose consolidation were impressively high, and a subset of up to 30% remained disease-free for a prolonged period. 1–5 Larger prospective comparisons of conventional dose therapy to one or two cycles of high-dose therapy with transplant were initiated in the early 1990s but accrued slowly, with up to a decade from initiation to reporting of results. Reasons for the slow accrual included the hesitance of many third party payers to cover the clinical costs of patients enrolled on these trials, the biases of many physicians, and strident patient advocacy at times for and at times against the procedure. The first reports from randomized trials at a single institution in both early stage and metastatic disease showed dramatic survival benefits but were ultimately shown to be based on fraudulent data. 6 In the interim, however, the treatment became almost standard-of-care, several US states enacted legislation requiring insurance companies to provide reimbursement, and many patients received high-dose therapy off protocol, reducing accrual to ongoing randomized comparisons. The results of those randomized studies were eventually presented beginning in 1999 but showed little benefit, and the number of procedures has now fallen from thousands yearly to only a few.

Six randomized studies have examined the effect of highdose chemotherapy in patients with metastatic disease: the North American ECOG/Philadelphia Bone Marrow Transplant Group PBT-1 Trial, the Canadian MA-16 study, the French PEGASE 03 and 04 studies, and two trials from Duke University, one of which appears in this issue of Bone