We studied the clinical course of 130 chronic myeloid leukemia (CML) patients (89 males and 41 females) in the European Bone Marrow Transplantation Group (EBMT) registry who received transplants before January 1, 1988 and who subsequently had evidence of recurrent leukemia. All patients had received a pretransplant conditioning regimen including total body irradiation (TBI). The first evidence of relapse was cytogenetic only in 74 (57%) patients and hematologic in 56 (43%). The overall actuarial survival from relapse was 36% at 6 years, with a significantly higher proportion of survivors among female patients (53% v 30%; P < .002). In univariate analysis, the 6-year probability of survival was 52% for patients with cytogenetic relapse and 30% for patients relapsing in chronic phase (CP), while no patient who relapsed in advanced phase (AP or BC) survived more than 3.5 years from relapse (P < .0001). The actuarial survival of patients relapsing before 6 months, between 6 and 12 months, and later than 12 months after transplant was 27%, 26%, and 45%, respectively (P < .002). Among patients with cytogenetic relapse, partial or complete disappearance of Ph-positive cells occurred in 40% of untreated patients and in 42% of those treated with interferon (IFN). However, IFN therapy significantly delayed progression toward hematologic disease. Cytogenetic responses were observed in 25% of patients who received IFN for relapse into CP, while only one minor cytogenetic response was reported in patients on conventional chemotherapy. For patients presenting with cytogenetic relapse as well as for those in hematologic relapse, IFN therapy significantly improved the 2-year probability of survival. However, long-term survival for IFN-treated patients in either group was not different from long-term survival in comparable patients not receiving IFN therapy. Twenty-nine patients of this series underwent a second bone marrow transplant (BMT) and the projected survival at 4 years after the second transplant is 28%. In multivariate Cox regression analysis, four factors remained significantly associated with survival: disease phase at relapse (P < .0001), duration of time interval from BMT to relapse (P = .0001), interferon therapy at relapse (P = .0024), and patient sex (P = .0032). This retrospective study provides evidence that some patients who relapse after BMT may benefit from treatment with IFN; a second BMT may offer the chance of cure. Data from this analysis may be useful in designing future prospective trials on posttransplant CML relapse.