Key Points• C/EBPa directly represses the leukemia maintenance program; however, the pattern of repressed genes is specific for each type of AML.• Overexpression of C/EBPa does not globally displace these proteins from their binding sites, but overrides their repressive activity.Acute myeloid leukemia (AML) is a heterogeneous disease caused by recurrent mutations in the transcription regulatory machinery, resulting in abnormal growth and a block in differentiation. One type of recurrent mutations affects RUNX1, which is subject to mutations and translocations, the latter giving rise to fusion proteins with aberrant transcriptional activities. We recently compared the mechanism by which the products of the t(8;21) and the t(3;21) translocation RUNX1-ETO and RUNX1-EVI1 reprogram the epigenome. We demonstrated that a main component of the block in differentiation in both types of AML is direct repression of the gene encoding the myeloid regulator C/EBPa by both fusion proteins. Here, we examined at the global level whether C/EBPa is able to reverse aberrant chromatin programming in t(8;21) and t(3;21) AML. C/EBPa overexpression does not change oncoprotein expression or globally displace these proteins from their binding sites. Instead, it upregulates a core set of common target genes important for myeloid differentiation and represses genes regulating leukemia maintenance. This study, therefore, identifies common CEBPA-regulated pathways as targets for therapeutic intervention.