We report the feasibility and efficacy of a fludarabine/busulfan-based dose-reduced conditioning regimen followed by stem cell transplantation from related (n=19) or unrelated HLA-matched donors (n=18) in 37 patients with myelodysplastic syndrome (MDS) or secondary acute myeloid leukemia (sAML) who were not eligible for a standard myeloablative conditioning regimen. The conditioning regimen consisted of fludarabine (120-180 mg/m(2)), busulfan (8 mg/kg p.o. or 6.4 mg/kg i.v.), and antithymocyte globulin (n=25). Graft-versus-host disease (GVHD) prophylaxis consisted of cyclosporine (n=36) and a short course of methotrexate (n=29) or mycophenolate mofetil (n=3). The median age of the patients was 55 years (range: 23-72). The reasons to perform a dose-reduced conditioning were reduced performance status (n=14), age (n=12), prior autologous (n=5) or allogeneic (n=1) transplantation, or prior/active fungal infection (n=5). Diagnoses at transplantation were refractory anemia (RA) (n=8), refractory anemia with excess of blasts (RAEB) (n=6), RAEB in transformation (RAEB-T) (n=13), chronic myelomonocytic leukemia (CMML) (n=3), and sAML (n=7). Stem cell sources were peripheral blood stem cells (PBSC) (n=29) or bone marrow (n=8). One patient received a T-cell-depleted peripheral stem cell graft. Two primary graft failures were observed (6%). Engraftment of leukocytes (>1.0x10(9)/l) and platelets (>20x10(9)/l) was seen after a median of 14 days. Acute GVHD grade II-IV was seen in 37%, while severe grade III/IV GVHD was observed in six patients (17%). Chronic GVHD was seen in 13 patients (48%). There were ten deaths (27%) due to treatment (TRM). The probability of TRM was higher in patients with unrelated donors (45 vs 12%, p=0.03) and in patients with poor cytogenetics in comparison to those with a low or intermediate karyotype (75 vs 20%, p=0.009). During follow-up 12 patients relapsed (32%). Patients without chronic GVHD had a significantly higher probability of relapse compared to those with chronic GVHD (70 vs 15%, p=0.02). After a median follow-up of 20 months, the 3-year estimated disease-free survival (DFS) is 38% [95% confidence interval (CI): 21-55%] and the overall survival (OS) is 39% (95% CI: 22-56%). The OS and DFS after related and unrelated transplantations was 45% (95% CI: 19-71%) vs 31% (95% CI: 9-53%) (n.s.) and 51% (95% CI: 29-73%) vs 25% (95% CI: 4-47%) (n.s.), respectively. We conclude that dose-reduced conditioning followed by allogeneic stem cell transplantation from related or unrelated donors is an effective treatment approach in patients with MDS/sAML and might cure a substantial number of patients who are not eligible for a standard allogeneic transplantation.