Mito-FLAG as salvage therapy for relapsed and refractory acute myeloid leukemia

Background: This study was performed to examine the feasibility and toxicit y of the combination of mitoxantrone, fludarabine, cytarabine as bolus (B) or continuous infusion (CI) and granulocyte-colony stimulating factor (G-CS F) in patients with recurrent and refractory acute myeloid leukemia (AML). Patients and Methods: 29 patients with relapsed (n = 17) or refractory (n = 12) AML were treated with the Mito-FLAG protocol consisting of mitoxantron e (7 mg/m(2), days 1/3/5), fludarabine (15 mg/m(2), every 12 h, days 1-5), cytarabine (Ara-C) as bolus infusion (1000 mg/m(2) over 1 h, every 12 h, da ys 1-5) (n = 15) or as continuous infusion (100-150 mg/m(2) over 24 h, days 1-5) (n = 14), and G-CSF (5 mug/ kg/day, day 0 until a neutrophile count o f 0.5 x 10(9)/l). Results: 17 patients (59%) and 1 patient (3%) achieved co mplete remission (CR) and partial remission (PR), respectively; thus the ov erall response rate was 62%. Following Mito-FLAG, 5 patients with CR underw ent high-dose therapy (HDT) with allogeneic (n = 2) or autologous (n = 3) s tem cell transplantation (SCT). With a median follow-up of 28 (range 6-54) months, 4 transplanted patients are alive in CR (n = 2) or in relapse (n = 2). The median duration of event-free survival (EFS) and overall survival ( OS) was 3.2 and 6.8 months, and probabilities of EFS and OS after 1 year we re 14 and 34%, respectively. The 1-year rates for EFS and OS in this group were 18 and 53%, respectively. Median duration of WHO grade 4 granulocytope nia and thrombocytopenia was 20 and 23 days, respectively. Nonhematological side effects were moderate, predominantly reaching WHO grades 1-2. Neutrop enic fever was seen in 85% of courses, with a median duration of 4 (1-38) d ays. Four patients (14%) suffered an early death because of aplasia (n = 2) , pneumonia (n = 1) or progressive AML (1 nonresponding patient). Conclusio ns: Our data suggest that the Mito-FLAG protocol is feasible and can be saf ely performed with both schedules of Ara-C. In this study the regimens have shown high efficacy and acceptable toxicity in patients with relapsed or r efractory AML. We currently examine the importance of bolus versus continuo us infusion of Ara-C as part of the Mito-FLAG regimen in a prospective rand omized multicenter trial.