P440: enhancer hijacking in complex-karyotype aml

Background: The total number of structural rearrangements in acute myeloid leukemkia (AML) is generally much lower than in other cancer types. However, AML with a complex karyotype (ckAML) is a rare AML subtype (10-15% of all cases) characterized by the presence of at least three cytogenetic alterations and has a dismal prognosis. It is still poorly understood. Many commonly deleted regions have been identified, for example in 5q, 7q, 12p and 17p, but it is still unclear how they may drive leukemia. In parallel, it has been discovered in the last decade that structural rearrangements can lead to aberrant expression of genes, for example of EVI1 in inv(3) or t(3;3), but no systematic search for such events has been undertaken. Since ckAML cases have many structural rearrangements, it is likely that some of them could lead to enhancer hijacking. Aims: We systematically searched for genes, which can be activated by enhancer hijacking in ckAML. Methods: We performed whole genome sequencing (WGS) and RNAseq of 42 ckAML cases. For each gene, we identified samples which had breakpoints located in the same topologically-associating domain, using the WGS. Then, we tested if the samples with breakpoints in the vicinity of the gene had an outlier high expression for this gene. Enhancer hijacking should lead to the expression of only the rearranged allele, so we filtered for genes with monoallelic expression, based on SNPs detected in WGS and RNAseq. Results: We detected many putative enhancer hijacking events in our cohort of 42 ckAML cases. Among the top 20 candidates were three genes, which have already been reported in the literature as being activated by enhancer hijacking in AML: EVI1, BCL11B and MNX1. This supports the reliability of our method. In addition, we detected several novel interesting candidates and we could for some of them find recurrence of these events in other cohorts. Several enhancers were hijacked, including strong hematopoietic enhancers like MYC, GATA2 or CDK6. Summary/Conclusion: We performed for the first time an exhaustive search for enhancer hijacking events in ckAML, and found both known and novel genes. This provides a better understanding of this deadly and poorly understood disease. Since these events are rare, sequencing even larger cohorts could lead to the identification of even more genes and could help estimate the frequency of these events.Keywords: AML, EVI1, TP53, Complex aberrant karyotype