Because the prognostic value of rare recurrent chromosomal aberrations still has to be elucidated, these data will contribute to future risk stratification for the treatment of pediatric AML.

Relapse remains the major cause of treatment failure in pediatric acute myeloid leukemia (AML). We analyzed the clinical characteristics, treatment response to relapse treatment and overall survival (OS) of 379 children with AML relapse treated according to three consecutive frontline protocols of the AML-Berlin/Frankfurt/Muenster study group (AML-BFM-87/-93/-98). Of 313 treated patients with data on remission status, 198 children (63%) achieved a second complete remission (CR2). There were no significant differences in remission rates and OS for the intensive reinduction treatment schedules used.The 5-year OS rate was 23% for the total group and 29% for patients treated with curative intent. OS rates increased with study periods from 18 to 34% (P log rank ¼ 0.012), whereas the proportion of patients receiving only palliative treatment decreased from 23 to 11% (P CMH ¼ 0.005). Late relapse, no allogeneic stem cell transplantation (SCT) in CR1, age o10 years and favorable cytogenetics were independent favorable prognostic factors for survival. Achievement of CR2 was the most important prognostic factor (OS 44 vs 3%; P log rank o0.0001). Overall, one-third of children with relapsed AML can be cured today. SCT in CR2 is recommended for most patients, although its impact on CR2 is discussed.

Key Points• AML induction with liposomal daunorubicin (80 mg/m 2 per day for 3 days) shows antileukemic activity comparable to idarubicin (12 mg/m 2 per day for 3 days).• Liposomal daunorubicin promises to be more active in the t(8;21) subgroup and causes less treatment-related toxicity.Outcomes of patients with acute myeloid leukemia (AML) improve significantly by intensification of induction. To further intensify anthracycline dosage without increasing cardiotoxicity, we compared potentially less cardiotoxic liposomal daunorubicin (L-DNR) to idarubicin at a higher-than-equivalent dose (80 vs 12 mg/m 2 per day for 3 days) during induction. In the multicenter therapy-optimization trial AML-BFM 2004, 521 of 611 pediatric patients (85%) were randomly assigned to L-DNR or idarubicin induction. Five-year results in both treatment arms were similar (overall survival 76% 6 3% [L-DNR] vs 75% 6 3% [idarubicin], P logrank 5 .65; event-free survival [EFS] 59% 6 3% vs 53% 6 3%, P logrank 5 .25; cumulative incidence of relapse 29% 6 3% vs 31% 6 3%, P (Gray) 5 .75), as were EFS results for standard (72% 6 5% vs 68% 6 5%, P logrank 5 .47) and high-risk (51% 6 4% vs 46% 6 4%, P logrank 5 .45) patients. L-DNR resulted in significantly better probability of EFS in patients with t(8;21). Overall, treatment-related mortality was lower with L-DNR than idarubicin (2/257 vs 10/264 patients, P 5 .04). Grade 3/4 cardiotoxicity was rare after induction (4 L-DNR vs 5 idarubicin). Only 1 L-DNR and 3 idarubicin patients presented with subclinical or mild cardiomyopathy during follow-up. In conclusion, at the given dose, L-DNR has overall antileukemic activity comparable to idarubicin, promises to be more active in subgroups, and causes less treatment-related mortality. This trial was registered at www.clinicaltrials.gov as #NCT00111345. (Blood. 2013;122(1):37-43)

Infants o1 year of age have a high prevalence of prognostically unfavorable leukemias and a presumed susceptibility to treatment-related toxicities. A total of 125 infants with acute myeloid leukemia (AML) were treated in studies AML-BFM-98 (n ¼ 59) and -2004 (n ¼ 66). Treatment regimens of both studies were comparable, consisting of intensive induction followed by four courses (mainly high-dose cytarabine and anthracyclines). Allogeneic-hematopoietic stem-cell-transplantation (allo-HSCT) in 1st remission was optional for high-risk (HR) patients. Most infants (120/125 ¼ 96%) were HR patients according to morphological, cytogenetic/molecular genetic and response criteria. Five-year overall survival was 66±4%, and improved from 61±6% in study-98 to 75±6% in study-2004 (P logrank 0.14) and event-free survival rates were 44±6% and 51±6% (P logrank 0.66), respectively. Results in HR infants were similar to those of older HR children (1-o2-or 2-o10-year olds, P logrank 0.90 for survival). Survival rates of HSCT in 1st remission, initial partial response and after relapse were high (13/14, 2/8 and 20/30 patients, respectively). The latter contributes to excellent 5-year survival after relapse (50±8%). Despite more severe infections and pulmonary toxicities in infants, treatment-related death rate was identical to that of older children (3%). Our data indicate that intensive frontline and relapse AML treatment is feasible in infants, toxicities are manageable, and outcome is favorable.