A CpG island methylator phenotype in acute myeloid leukemia independent of IDH mutations and associated with a favorable outcome

Acute myeloid leukemia (AML) causes the most leukemia-related deaths in the United States. Although genetic changes are assessed in the clinic for risk stratification, current classifications are imperfect and epigenetic determinants of AML curability remain poorly understood. To address this gap in knowledge we performed genome-wide DNA methylation analysis using the next-generation sequencing-based Digital Restriction Enzyme Analysis of Methylation (DREAM) assay on 96 clinical AML samples, and 35 normal peripheral blood controls. We identified patterns of aberrant DNA hypermethylation in distinct subsets of cases, and these patterns were associated with unique clinical, and genetic features. Validation an extension of these findings was performed using 194 samples from The Cancer Genome Atlas (TCGA). Digital restriction enzyme analysis of methylation (DREAM) was performed to determine the methylation profile of human genomic DNA from the bone marrow of 96 patients with acute myeloid leukemia (AML) and peripheral blood from 35 healthy controls. Briefly, genomic DNA was sequentially digested by SmaI and XmaI, which both recognize the sequence CCCGGG. SmaI is methylation sensitive, where XmaI is methylation insensitive. Distinct signatures, 5’-GGG at unmethylated sites or 5’-CCGGG at methylated sites were created by enzyme digestion and ultimately high througput sequencing was used to map these sites to the genome. Methylation ratios for each individual CCCGGG site were calculated as a proportion of methylated counts to the sum of unmethylated and methylated counts, and subsequently adjusted for differences in restriction enzyme efficiency using values obtained from spiked in standards.