Study AML-BFM 2004: Improved Survival In Childhood Acute Myeloid Leukemia without Increased Toxicity

Abstract 181

Acute Myeloid Leukemia - Therapy, excluding Transplantation: Pediatric and Adult AML Therapy

Study AML-BFM 2004 was designed to improve outcome of children and adolescents with AML without increasing toxicity. Patients were stratified into a standard- (SR)* or high-risk (HR)** group according to morphology, cyto-/molecular genetics including FLT3-ITD, and therapy response on day 15. Notably, reclassification of SR patients to the HR group in case of FLT3-ITD positivity was newly established in this study. Improvement of prognosis was attempted by intensification of chemotherapy: (1) Randomized introduction of liposomal daunorubicin (L-DNR) in a higher equivalent dose than idarubicin during induction in both risk groups (L-DNR 80mg/m²/day/3x) in comparison to standard induction using idarubicin 12mg/m²/day/3x, each combined with cytarabine and etoposide (L-DNR may offer an increased therapeutic window due to lower cardiotoxicity) and (2) randomised introduction of 2-chloro-2-deoxyadenosine (2-CDA, 2×6mg/m²) as intensification during the cytarabine/idarubicin (AI) consolidation in HR patients only.

Overall results improved compared to the previous study AML-BFM 98: Survival estimates at 5 year (pOS) in patients (excluding Myeloid leukemia in Down syndrome) were 72% + 3% vs. 64% + 2% (AML-BFM 04 n=566 vs. AML-BFM 98 n=472), p_logrank=0.02; the 5-year event-free survival (pEFS): was 54% + 3% vs. 50% + 2%, p_logrank=.40. Results in the 197 SR patients were excellent: pOS 88% + 3% vs. 78% + 3% (n=182), p_logrank=.01, EFS 69%, + 4% vs. 64% + 4%, p_logrank=.40. Results in the 368 HR patients also improved: pOS 63% + 4% vs. 56% + 3% (n=290), p_logrank=.07, EFS 46%, + 3% vs. 41% + 3%, p_logrank=.43. OS improvement was partly due to better results after treatment of relapse or nonresponse (3-year pOS after nonresponse/relapse in 171 patients of study 2004 40% + 5% vs. 32% + 4% in 198 patients in AML-BFM 98, p_logrank=.017).

Results for the 1^st randomization L-DNR vs. idarubicin during induction were similar (pOS 78% + 4% vs. 70% + 4%, p_logrank=.15, pEFS 60% + 4% vs. 54% + 4%, p_logrank=.17). There were less early deaths (4 vs. 8 patients) and less treatment related deaths in remission in the L-DNR group (2 vs. 5 patients). The rate of severe infection was slightly lower with L-DNR (p_Fisher 0.15). Two L-DNR vs. 6 idarubicin patients showed grade III/IV cardiotoxicity after induction. Results of the 2^nd randomization in HR patients (AI/2-CDA vs. AI) were also similar: p=OS 75% + 5% vs. 65% + 5%, p_logrank=.18, pEFS 51% + 5% vs. 51% + 5%, p_logrank=.98. Toxicity rates of the intensification with 2-CDA were tolerable.

In conclusion, compared to the previous study AML-BFM 98, results of study AML-BFM 2004 show significant improvement in both risk groups. This appears attributable to a combination of factors including therapy intensification, better supportive care and improved treatment of patients with relapse or nonresponse. Given the reduced toxicity of L-DNR and a trend towards better survival rates by adding L-DNR during induction and 2-CDA during HR consolidation, these agents will be further used in the forthcoming AML-BFM study.

*Standard risk group definition: FAB M1/M2 with Auer rods, FAB M4eo or favorable cytogenetics [t(8;21) or inv(16)] and blasts in the bone marrow on day 15 <5%, and FAB M3 (all patients)

**High-risk group definition: all others.