Most allogeneic haematopoietic stem cell transplants now use peripheral blood progenitor cell transplantation (PBPCT) instead of bone-marrow transplantation (BMT). Long-term data on outcome and late effects of PBPCT compared with BMT are scarce. Here we present long-term data from a randomised study comparing PBPCT with BMT. Between February, 1995, and September, 1999, 329 patients with leukaemia received either PBPCT (n=163) or BMT (n=166) from HLA-identical sibling donors after central randomisation accounting for stratification criteria. Follow-up data were collected via questionnaires from 87% (176 of 202; 84 PBPCT, 92 BMT) patients who survived for more than 3 years (median of 9.3 years) after transplantation. Efficacy analyses included all patients who received treatment. This study is registered with ClinicalTrials.gov, number NCT01020175. 10-year overall survival was 49.1% for patients who underwent PBPCT and 56.5% for patients who underwent BMT (HR 0.83, 95% CI 0.60-1.15; p=0.27). Leukaemia-free survival was 28.3% with BMT versus 13.0% with PBPCT (0.61, CI 0.32-1.16; p=0.12) for acute lymphoblastic leukaemia; 62.3% with BMT versus 47.1% with PBPCT for acute myeloid leukaemia (0.67, 0.39-1.16; p=0.16); and 40.2% with BMT versus 48.5% with PBPCT for chronic myeloid leukaemia (1.12, 0.73-1.74; p=0.60). More patients developed chronic graft-versus-host disease after PBPCT (n=56, 73%) than after BMT (n=46, 56%; p=0.021), with more frequent involvement of skin, liver, and oral mucosa, and more patients who underwent PBPCT needed immunosuppressive treatment 5 years after transplantation (n=20, 26%) than patients who had BMT (n=10, 12%; p=0.024). Nonetheless, there was no difference in performance status, return to work, incidence of bronchiolitis obliterans, and haematopoietic function between the two groups. 14 cases of secondary malignancies occurred (five after BMT, nine after PBPCT), resulting in a cumulative incidence of 3% and 7% after BMT and PBPCT (p=0.17), respectively. More than 9 years after transplantation, overall and leukaemia-free survival remain similar in patients who underwent BMT and PBPCT. Differences in the incidence of chronic graft-versus-host disease and the duration of immunosuppression exist, but do not affect survival, general health status, or late events. No external funding was received.