Improvement in Treatment Outcome and Identification of a New Prognostic Parameter in Down Syndrome Acute Myeloid Leukemia (DS-AML): Results of the Children’s Oncology Group (COG) Phase III AAML0431 Trial

Background: DS-AML patients (<4 years of age), have favorable outcomes compared to non-DS AML patients. However, DS patients experience significant treatment-related morbidity and mortality due to infectious and cardiac complications. Blast cells from DS-AML patients reportedly have increased sensitivity to cytarabine (araC) and daunorubicin, suggesting that optimization of drug dosing may improve outcomes and/or reduce toxicity.

Objective: The main objectives of the AAML0431 study were to determine whether high-dose (HD)-araC) administered earlier during treatment could improve EFS rates, and whether EFS rates could be maintained despite reducing the cumulative daunorubicin dose and the number of intrathecal treatments. The study also aimed at determining the clinical significance of minimal residual disease (MRD) in this AML subgroup.

Methods: AAML0431 consisted of 4 cycles of induction and 2 of consolidation therapy based on a backbone of the previous COG A2971 trial with several modifications including the use of HD-araC for Induction II rather than during Intensification therapy (fifth cycle of therapy), a 25% reduction in the cumulative daunorubicin dose and a decrease in the number of intrathecal treatments from 7 to 2. The recommended criteria for proceeding to each subsequent cycle of therapy was an absolute neutrophil count ≥1,000/µL and platelets ≥100,000/µL. MRD was measured by a multi-parameter flow cytometry assay capable of detecting 0.01% DS AML blasts.

Results: Between March 2007 and December 2011, 205 children (106 females, 99 males) with DS or DS mosaicisim were enrolled. Of the 204 evaluable patients, 144 were classified as having AML and 60 myelodysplastic syndrome (MDS); median age at diagnosis was 1.57 years. Sixty two patients (40 AML, 28%; 22 MDS, 37%) had a prior diagnosis of the transient myeloproliferative disorder (TMD) and 6 had received chemotherapy for TMD. Congenital heart defects were present in 90 (44%) patients.

Among AML patients, median white blood cell count (WBC) at diagnosis was 6.5 x 10³/µL, peripheral blast percentage 7% and platelet count 34 x 10³/µL; for MDS patients, these were 4.85. x 10³/µL, 0% and 36.5 x 10³/µL, respectively. No patient had CNS leukemia at diagnosis. Institutional reporting of morphology was megakaryocytic leukemia in 85 (42%) of cases.

Event-free survival (EFS) from study entry for all eligible patients was 90% ± 4.4% at 3 years; overall survival (OS) was 92.7% ± 3.8% at 3 years, with equivalent results for MDS patients. There were 19 treatment failures: 2 induction failures, 14 relapses and 3 non-relapse deaths. MRD data on day 28 of Induction I was available in 149 patients. Three-year disease-free survival among patients who were MRD-negative (blasts <0.01%) after Induction I (93.5% ± 4.8%; n=125) was significantly higher than that of MRD-positive patients (70.6% ± 18.7%; n=24; log-rank P <.001). The OS at 3 years for the 16 patients with refractory/relapse leukemia was 30% ± 23.6%.

There were a total of 1045 adverse events (AEs) classified as grade 3 or higher; 66% of these AEs were reported during the Induction II cycle. No life-threatening cardiac toxicities were reported and overall, only 7 cardiac AEs classified as ≥ Grade 3 were identified.

Thirteen patients were electively taken off protocol therapy by request from the treating physician (n=6) or parents (n=7) (Induction I, 4; Induction II, 3; Induction IV, 2; Intensification I, 4), primarily due to experiencing significant toxicities. One patient taken off therapy after Induction II, relapsed and died and another had a non-relapse death after stopping therapy post-Induction I. The remaining patients were reported to be alive.

The earlier use of HD-araC in AAML0431 resulted in improvements in EFS and OS compared to past COG studies, while a 25% reduction of the cumulative daunorubicin dose (compared to COG A2971) and limited intrathecal chemotherapy treatments did not adversely impact outcome. In addition, MRD analysis performed after Induction I was identified as a prognostic factor for DS-AML patients for the first time. MRD provides a prospective tool for risk stratification for future trials to identify patients who may benefit from a further reduction in therapy intensity (e.g., elimination of HD-araC), as well as identifying patients with resistant disease who require new treatment options (e.g. treatment intensification, new agents).