Genomic architecture and treatment outcome in pediatric acute myeloid leukemia: a Children's Oncology Group report

Abstract

Childhood acute myeloid leukemia (AML) is frequently characterized by chromosomal instability. Approximately 50% of patients have disease relapse, and novel prognostic markers are needed to improve risk stratification. We performed genome-wide genotyping in 446 pediatric patients with de novo AML enrolled in Children's Oncology Group (COG) studies AAML0531, AAML03P1, and CCG2961. Affymetrix and Illumina Omni 2.5 platforms were used to evaluate copy-number alterations (CNAs) and determine their associations with treatment outcome. Data from Affymetrix and Illumina studies were jointly analyzed with ASCAT and GISTIC software. An average of 1.14 somatically acquired CNAs per patient were observed. Novel reoccurring altered genomic regions were identified, and the presence of CNAs was found to be associated with decreased 3-year overall survival (OS), event-free survival (EFS), and relapse risk from the end of induction 1 (hazard ratio [HR], 1.7; 95% confidence interval [CI], 1.2-2.4; HR, 1.4; 95% CI, 1.0-1.8; and HR, 1.4; 95% CI, 1.0-2.0, respectively). Analyses by risk group demonstrated decreased OS and EFS in the standard-risk group only (HR, 1.9; 95% CI, 1.1-3.3 and HR, 1.7; 95% CI, 1.1-2.6, respectively). Additional studies are required to test the prognostic significance of CNA presence in disease relapse in patients with AML. COG studies AAML0531, AAML03P1, and CCG2961 were registered at www.clinicaltrials.gov as #NCT01407757, #NCT00070174, and #NCT00003790, respectively.

© 2017 by The American Society of Hematology.

Figures

Figure 1.

CNAs in pediatric AML. Broad copy number landscape of pediatric AML, shown as a heatmap of CNAs as seen in the COG cancer cohort. Deletions are shown in blue, amplifications in red, and copy-neutral (CN) LOHs in green.

Figure 2.

Focal recurrent aberrations. Significant focal CNAs in the COG pediatric AML cohort. GISTIC analysis revealed significantly recurring regions of focal CNAs, stratified according to amplifications (A), deletions (B), and CN LOHs (C). The vertical green line shows a false-discovery rate Q P > .25, which is considered significant.

Figure 3.

CNAs in pediatric AML. (A) OS and (B) EFS.

Figure 4.

Stratified survival analysis according to risk group. (A,C,E) OS and (B,D,F) EFS for standard- (A-B), favorable- (C-D), and poor-risk groups (E-F).