Abstract
BACKGROUND: Limited data is available to guide the choice of the conditioning regimen for patients with acute myeloid leukemia (AML) undergoing transplant with persistent disease. METHODS: We retrospectively compared outcome of fludarabine-treosulfan (FT), thiotepa-busulfan-fludarabine (TBF), and sequential fludarabine, intermediate dose Ara-C, amsacrine, total body irradiation/busulfan, cyclophosphamide (FLAMSA) conditioning in patients with refractory or relapsed AML. RESULTS: Complete remission rates at day 100 were 92%, 80%, and 88% for FT, TBF, and FLAMSA, respectively (p = 0.13). Non-relapse mortality, incidence of relapse, acute (a) and chronic (c) graft-versus-host disease (GVHD) rates did not differ between the three groups. Overall survival at 2 years was 37% for FT, 24% for TBF, and 34% for FLAMSA (p = 0.10). Independent prognostic factors for survival were Karnofsky performance score and patient CMV serology (p = 0.01; p = 0.02), while survival was not affected by age at transplant. The use of anti-thymocyte globulin (ATG) was associated with reduced risk of grade III-IV aGVHD (p = 0.02) and cGVHD (p = 0.006), with no influence on relapse. CONCLUSIONS: In conclusion, FT, TBF, and FLAMSA regimens provided similar outcome in patients undergoing transplant with active AML. Survival was determined by patient characteristics as Karnofsky performance score and CMV serology, however was not affected by age at transplant. ATG appears able to reduce the incidence of acute and chronic GVHD without influencing relapse risk.