Observational studies have suggested that physical activity may be associated with improved survival after cancer treatment. However, data from controlled clinical trials are required. We analyzed survival data of 103 patients from a previously published randomized controlled trial in allogeneic stem cell transplant patients who were randomized to either an exercise intervention (EX) or to a social contact control group. EX patients trained prior to hospital admission, during inpatient treatment, and for 6-8 weeks after discharge. Survival analyses were used to compare both total mortality (TM) and non-relapse mortality (NRM) after discharge and transplantation during an observation period of 2 years after transplantation. Analyses were corroborated with Cox and Fine & Gray regression models adjusting for potential confounders. After discharge, EX patients had a significantly lower TM rate than controls (12.0 vs. 28.3%, p 5 0.030) and a numerically lower NRM rate (4.0 vs. 13.5%, p 5 0.086). When the inpatient period was included, absolute risk reductions were similar but not significantly different (TM: 34.0 vs. 50.9%, p 5 0.112; NRM: 26.0 vs. 36.5%, p 5 0.293). The number needed to treat (NNT) to prevent one death with EX was about 6. Furthermore, regression analyses revealed that baseline fitness was protective against mortality. The data suggest that exercise might improve survival in patients undergoing allo-HCT. However, the results should be interpreted with caution as the study was not designed to detect differences in survival rates, and as no stratification on relevant prognostic factors was carried out.Allogeneic hematopoietic stem cell transplantation (allo-HCT) is an effective treatment option for patients with highrisk hematological malignancies such as acute leukemia. However, recipients of allo-HCT may suffer from numerous treatment-related side effects and complications, such as infections and graft-versus-host disease (GvHD), and the transplant-related mortality rate is considerable. 1 Due to the high mortality rates in myeloablative regimens (30-40% 2 ) and even in reduced intensity conditioning (RIC) regimens (20-30% 3 ), clinicians and researchers have attempted to determine variables that can be used to characterize subtypes of patients who are candidates for transplantation. 4 Several prognostic risk scores have been established to predict survival after allo-HCT. The EBMT risk score 5 is a well-known tool that predicts the risk of death after allo-HCT using a score based on age, disease stage, time interval from diagnosis to transplant, donor type and donor-recipient gender combination. Prognostic validity has also been demonstrated by the clinician-assessed Karnofsky Performance Score and by the co-morbidity score by Sorror et al. 6,7 A recently published pilot study by Wood et al. suggests that cardiopulmonary fitness, as defined by either peak oxygen uptake or the 6-minute walk distance Test (6-MWD), is a predictor of symptom severity, quality of life, and mortality in patients undergoi...