Optimal reinduction therapy for patients with relapsed pediatric acute myeloid leukemia (AML) remains undefined. Liposomal daunorubicin (DaunoXome), which offers the potential for reduced cardiotoxicity compared with traditional daunorubicin, is effective in this setting. Gertjan J.L. Kaspers, MD, PhD, and colleagues in the International Berlin-Frankfurt-Münster Study Group recently reported what may be the first randomized trial and the largest trial conducted to date in pediatric relapsed AML.¹

In this investigator-initiated international phase III trial, published in the Journal of Clinical Oncology, the addition of liposomal daunorubicin to FLAG (fludarabine, cytarabine, and granulocyte colony-stimulating factor [filgrastim (Neupogen)]) improved early bone marrow status in patients with relapsed pediatric AML. Overall survival was similar with the two treatments, although subgroup analysis suggested a potential advantage with the liposomal daunorubicin-containing regimen in patients with better prognostic characteristics.

Study Details

The trial was performed by 13 groups in 20 countries, with patient enrollment occurring from November 2001 through April 2009. A total of 394 patients aged less than 21 years with relapsed or primary refractory AML were randomly assigned to liposomal daunorubicin plus FLAG or FLAG alone followed by FLAG, given in 28-day cycles.

Granulocyte colony-stimulating factor was given at 200 µg/m² on days 0 to 5 and immediately followed by fludarabine 30 mg/m² on days 1 to 5; liposomal daunorubicin was given at 60 mg/m² after fludarabine on days 1, 3, and 5, and cytarabine 2,000 mg/m² was started 4 hours after fludarabine on days 1 to 5. Patients with complete response after two courses went on to intensive or low intensive consolidation therapy prior to stem cell transplantation. The primary endpoint of the trial was 28-day bone marrow status, assessed shortly before the second course of chemotherapy.

Patients in the liposomal daunorubicin and control groups were well balanced for sex (57% and 60% male), age (median 9 and 10 years), proportion of patients with less than 20% blasts (16% and 20%), nonfavorable cytogenetics (79% and 81%), and disease status (early relapse, late relapse, and primary refractory disease in 48%, 44%, and 8%, respectively, and 46%, 48%, and 6%, respectively). Overall, 20% of patients had core-binding factor AML. Patient groups were balanced for types of consolidation, stem cell transplantation, and allogeneic stem cell transplantation, and for time to transplant from complete response (median 105 and 107 days). 

Response Rates

The rate of good response (≤ 20% blasts) at 28 days was significantly higher in the liposomal daunorubicin group (80% vs 70%, odds ratio [OR] = 0.60, P = .04). Among patients with less than 20% blasts at baseline, 10% of those in the liposomal daunorubicin group vs 26% of those in the FLAG group had more than 20% blasts at day 28 (P = .088). Response rates were 70% vs 54% (P = .04) in patients with early relapsed AML, 90% vs 84% in those with late relapsed disease, 100% vs 94% in those with core-binding factor AML, and 76% vs 64% (P = .048) in those with unfavorable
cytogenetics.

Complete response rates were 69% in daunorubicin patients and 59% in control patients (P = .07), including rates of 97% vs 91% in patients with core-binding factor AML and 62% vs 55% in patients with other cytogenetics. The estimated cumulative incidence of relapse at 4 years was higher in the liposomal daunorubicin group (29% vs 19%, P = .02). Probability of survival at 4 years did not differ between the two groups (40% vs 36%, P = .54). A post hoc subgroup analysis showed that among patients with core-binding factor AML, probability of survival at 4 years was higher in the daunorubicin group (82% vs 58%, P = .04), with estimated cumulative incidence of relapse at 4 years being nonsignificantly lower (17% vs 22%, P = .59).

Treatment group was not a significant prognostic factor for overall survival on univariate or multivariate analysis. On regression analysis for overall survival including treatment group, time to relapse, and cytogenetic risk, there was a significant interaction of favorable cytogenetic risk and liposomal daunorubicin treatment indicating a survival benefit of liposomal daunorubicin in the patient subgroup with favorable cytogenetics (hazard ratio [HR] = 0.3, 95% confidence interval [CI] = 0.1-0.9, P = .034).

Similar Toxicity

Grade 3 or 4 toxicity was similar in the two treatment groups apart from skin toxicity, which occurred in 4% of liposomal daunorubicin patients and 1% of control patients (P = .04). Among grade 3 or 4 toxicities, thrombocytopenia, neutropenia, and anemia occurred in 94%, 97%, and 85% of the liposomal daunorubicin/FLAG group, respectively, and 98%, 98%, and 89% of the FLAG group, respectively; infections occurred in 40% vs 31%, nausea in 21% vs 20%, and fever in 18% vs 17%. Grade 3 or 4 acute cardiotoxicity occurred in five patients in the liposomal daunorubicin group (2.7%; one with grade 4) vs one patient in the control group (0.6%). Cardiotoxicity always coincided with fever and infections, and none of the episodes was fatal. Long-term cardiotoxicity data are not yet available.

Conclusions

With regard to the overall findings of the trial, the authors stated: “We report the first (to our knowledge) randomized study in pediatric relapsed AML. In this multinational setting, we achieved the best outcome for these children reported thus far, with a 4-year [probability of overall survival] of 38% in [the entire] group of 394 children…. International collaboration is feasible in the treatment of pediatric relapsed AML and should be pursued….”

With regard to the findings on the addition of liposomal daunorubicin, they noted that the nature of the trial prohibited close monitoring of all details of treatment and outcome, limiting the analysis of such factors as types of consolidation, quality of response according to minimal residual disease monitoring, and differences in first-line treatment. Nevertheless, they stated:

[T]he main study end point is robust and can be used in future studies in relapsed AML, even though more detailed immunophenotyping of blasts to discriminate them from regenerating normal blasts and information on [minimal disease monitoring] will be useful in addition. One may argue that the clinical benefit of [liposomal daunorubicin] is unknown at best, because [overall survival] did not improve significantly. However, it should be taken into account that therapy after induction was heterogeneous and may have been intensified in patients with poor early treatment response.

A better early treatment response and an improved complete response rate are clinically positive findings, they noted, as is the significantly higher probability of overall survival with liposomal daunorubicin plus FLAG for core-binding factor AML. ■

Disclosures: Drs. Kaspers and Reinhardt have received compensation from Galen for consultant or advisory roles.

Reference

1. Kaspers GJL, Zimmermann M, Reinhardt D, et al: Improved outcome in pediatric relapsed acute myeloid leukemia: Results of a randomized trial on liposomal daunorubicin by the International BFM Study Group. J Clin Oncol 31:599-607, 2013.