Treosulfan + Fludarabine +/− Thymoglobulin - An Effective Low Toxicity Conditioning Regimen for Allogeneic Hematopoietic Stem Cell Transplantation in Chronic Myeloid Leukemia

Allogeneic hematopoietic stem cell transplantation (alloHSCT) is the only treatment of proved long-term efficacy in chronic myeloid leukemia (CML). However, high procedure-related toxicity observed after oral busulfan- or TBI-based conditioning limits its applicability and deteriorates outcome. Between 2003–2005 a phase II study was conducted to evaluate the feasibility of a new preparetive regimen consisting of Treosulfan (a soluble alkylyting agent) 14 g/m2/d. on days -6, -5, -4, Fludarabine 30 mg/m2/d on days -5, -4, -3, -2, -1, and, in case of unrelated donor transplants (URD-HSCT), anti-thymocyte globulin (Thymoglobulin) at a total dose of 6 mg/kg. Results were compared to those from the historical control group of CML patients treated with oral Busulfan (16 mg/kg) + Cyclophosphamide (120 mg/kg) (BuCy) in the same institution between 2000–2003. 35 patients (age 35, range 16–52 years) with CML in the 1st chronic phase (n=33) or in 2nd chronic phase (n=2) were included in the study. Median interval from diagnosis to alloHSCT equaled 10 (6–144) months. 22 (63%) patients were given transplant from an unrelated donor, 13 (37%) - from an HLA identical sibling. Bone marrow was used a source of stem cells in 29 patients, peripheral blood - in 6 cases. GVHD prophylaxis consisted of Cyclosporin A and short-course Methotrexate. All patients engrafted after a period of absolute agranulocytosis. Median time to neutrophil recovery >0.5 G/L was 24 (10–42) days, and to PLT >50 G/L - 21 (13–38) days. 1/35 patient experienced grade 3 mucositis; no severe (grade 3–4) neutropenic infection nor VOD was observed. The incidence of grade II acute GVHD was 17%, grade III–IV - 3%. The cumulative incidence of non-relapse mortality (NRM) at 2 years equaled 14% (4/35). Causes of death were: EBV-LPD, late neuroinfection, late fungal infection, acute GVHD. At 2 years the probability of the overall survival and hematological relapse-free survival equaled 86% (+/−7%) and 83% (+/−7%). Respective rates for the control BuCy group (n=78) were significantly lower: 55% (+/−6%), p=0.02, and 54% (+/−6%), p=0.03. Seven patients in the Treosulfan+Fludarabine group required immunosuppression taper and additional interferone or imatinib treatment because of incomplete donor chimerism or molecular/cytogenetic relapse after initial response. We conclude that Treosulfan+Fludarabine+/−Thymoglobulin myeloablative conditioning is associated with low organ toxicity, low incidence of acute GVHD and NRM. The regimen is feasible for CML patients and appears superior in comparison with BuCy.