The combination of imatinib with mitoxantrone, etoposide or cytarabine were shown to be additive to highly synergistic on BCR-ABL-positive leukemias in vitro by several investigators, including our group. Therefore, we initiated a phase I/II trial for patients with myeloid blast crisis of chronic myelogenous leukemia. Patients were treated in four cohorts starting from mitoxantrone 10 mg/m2/d and etoposide 100 mg/m2/d for 2 or 3 consecutive days and start of imatinib 600 mg/d from day 15 (cohorts #1 and #2) or from day 1 (cohorts #3 and #4). Cytarabine was given at a dose of 10 mg/m2/d s.c. as maintenance treatment. Seventeen patients were included in the study: 8 pts in cohort 1, 1 pt in cohort 2, 4 pts in cohort 3, and 4 pts in cohort 4. Median age was 59.6 years (range 42.8–75.4), 10 male/7 female. Median time from diagnosis of CML to inclusion into the study was 3.0 yrs (range 0–13.6). Five patients had received prior therapy for blast crisis. Fifteen patients are evaluable (for 1 patient no follow-up data are available, the diagnosis of another patient was later revized as lymphatic blast crisis). Four patients received an allogeneic stem cell transplant at a median of study treatment duration of 3 months (range 2–6 months) after achieving hematological response: 2 pts are still alive at 367 and 379 days after study entry; 2 pts died of transplant-related complications (acute GVHD and sepsis). Out of 11 pts who received conventional treatment one is still alive and in a complete cytogenetic remission at day 924 after study entry (coh. #1). Causes of death were progressive disease (n=8: 4 pts in coh. #1, 1 patient in coh. #2 and coh. #3, and 2 pts in coh. #4), cerebral hemorrhage (n=1 in coh. #1), and pneumonia (n=1 in coh. #4). When analyzed separately, patients of coh. #1 and #2 (n=9) showed a trend to a shorter median survival than the patients of coh. #3 and #4 (n=6): median survival 142 days (10–928) vs. 377 days (50–467), t=0.072. The study treatment was well tolerated. Patients required a median of 8 (range 2–18) red blood cell concentrates, a median of 5 (1–8) platelet concentrates and therapeutic antibiotic treatment for 10d (0–38). Of note, WHO grade III or grade IV mucositis did not occur. Pts. who received imatinib on day 1 (coh. #3 and #4) had a significantly longer duration of severe leukopenia (WBC <1/nl): median of 19 days (0–23) vs. 2 days (0–14) in coh. #1 and #2. Interestingly, this did not require a longer therapeutic application of antibiotics in the cohorts #3 and #4. We conclude that the combination of mitoxantrone/etoposide and imatinib is a well tolerated protocol with mild non-hematological toxicity even in older patients. Patients who received more aggressive treatment schedules (imatinib started on day 1: coh. #3 and #4) had a trend to a longer survival. Eligible patient may have an advantage from an allogeneic stem cell transplantation following the induction treatment.