Allogeneic bone marrow transplantation (BMT) is the only curative therapy for chronic myelogenous leukemia (CML), though several studies indicate that prolonged survival can result from interferon-alpha (IFN-alpha) treatment. IFN-alpha is now often used as initial therapy for CML, before donor availability is known. Because identifying potential donors can take several weeks to months, it is important to know whether IFN-alpha adversely affects outcome of a subsequent BMT. If it does, initiation of IFN-alpha therapy might be delayed until donor availability is determined and avoided in patients for whom BMT is planned. We studied 873 patients who received HLA-identical sibling BMT for chronic-phase CML in 153 centers participating in the International Bone Marrow Transplant Registry. The object was to compare outcome in the 664 who received only hydroxyurea before BMT with outcome in the 209 who received IFN-alpha with or without hydroxyurea. The median duration of IFN-alpha therapy was 2 months (range, 1 to 39 months). Cox proportional hazards analysis was used to compare engraftment, graft-versus-host disease (GVHD), nonrelapse mortality, relapse, survival, and leukemia-free survival after adjustment for other prognostic variables. We found a higher risk of nonengraftment among patients given IFN-alpha than among those given hydroxyurea alone (2% versus 0.2%; P = 0.01). Patients who received IFN-alpha had a lower risk of relapse (relative risk, 0.17; 95% confidence interval, 0.04-0.70). Probabilities of GVHD, nonrelapse mortality, survival, and leukemia-free survival were similar in the two treatment groups. These results suggest that a short course of IFN-alpha does not adversely affect survival after a subsequent HLA-identical sibling BMT for chronic-phase CML. (Blood. 2000;95:410-415)