Non-DS-AMKL can be dived into genetic subtypes based on the presence of recurrent oncogenic fusions.

  • Major finding: Non-DS-AMKL can be dived into genetic subtypes based on the presence of recurrent oncogenic fusions.

  • Approach: RNA and/or whole-exome sequencing of 99 patients characterized the genomic landscape of non-DS-AMKL.

  • Impact: The identification of key genetic events in non-DS-AMKL may allow for improved risk stratification.

Acute megakaryoblastic leukemia (AMKL) is a subtype of acute myeloid leukemia (AML) that accounts for 4% to 15% of childhood AML, but rarely occurs in adults. Individuals with Down syndrome have an increased risk of developing AMKL, but AMKL in patients without Down syndrome (non-DS-AMKL) is associated with poorer clinical outcomes. Gene rearrangements resulting in chimeric oncogenes have been identified in many non-DS-AMKLs, but the genetic underpinnings of 30% to 40% of cases have not been determined. De Rooij, Branstetter, and colleagues performed RNA and/or exome sequencing of non-DS-AMKLs from 75 pediatric and 24 adult patients. RNA sequencing revealed genetic alterations predicted to produce fusion protein products in 5.5% of adult patients and 72.4% of pediatric patients. The pediatric fusions included the previously identified CBFA2T3–GLIS2 (18.6% of patients), KMT2A fusions (17.4% of patients), NUP98–KDM5A (11.6% of patients), and RBM15–MKL1 (10.5% of patients). Previously undescribed fusion genes were also identified including several HOX rearrangements occurring in 14% of patients. Fusions were not identified in 21% of pediatric cases, but half of the cases lacking a fusion had a truncating mutation in GATA1. Gene expression profiling revealed that the cases clustered based on the expressed fusions, indicating that fusion proteins determine the gene expression signature. Whole-exome sequencing revealed frequent mutations in GATA1, JAK, or STAT genes, cohesin or CTCF genes, and RAS pathway genes in pediatric patients, whereas adult patients had recurrent mutations in TP53, cohesin genes, splicing factor genes, ASXL, and DNMT3A. Based on these data, the pediatric cohort could be divided into seven subsets that correlated with survival: CBFA2T3–GLIS2, RBM15–MKL1, NUP98–KDM51, KMT2-rearranged, HOX-rearranged, GATA1-mutant, and other. CBFA2T3–GLIS2 cases had the poorest survival, and GATA1-mutant cases had the best survival. The identification of genomic subtypes of pediatric non-DS-AMKL provides insight into their etiology and may allow for improved risk stratification in patients with non-DS-AMKL.

de Rooij JD, Branstetter C, Ma J, Li Y, Walsh MP, Cheng J, et al. Pediatric non-Down syndrome acute megakaryoblastic leukemia is characterized by distinct genomic subsets with varying outcomes. Nat Genet 2017 Jan 23 [Epub ahead of print].