Myelodysplastic syndromes (MDS) are a heterogeneous group of hematopoietic disorders and characterized by dysplasia, ineffective hematopoiesis and predisposition to secondary acute myeloid leukemia (sAML). Azacitidine (AZA) is the standard care for high-risk MDS patients not eligible for allogenic bone marrow transplantation. However, only half of the patients respond and all patients eventually relapse. Response predicting biomarkers and combinatorial drugs targets enhancing response and its duration are needed. Here, we have taken a dual approach. First, we have evaluated genes encoding chromatin regulators for their capacity to modulate AZA response. We were able to validate several genes, whose genetic inhibition affected the cellular AZA response, including four genes encoding components of ISWI chromatin remodelling complex pointing towards a specific function and co-vulnerability. Second, we have used a classical cohort analysis approach measuring the association of expression of a gene panel in bone marrow samples collected at diagnosis of 36 MDS patients. The gene panel included the identified AZA modulators, genes known to be involved in AZA metabolism and previously identified candidate modulators. In addition to confirming a number of previously made observations, we were able to identify several new associations, such as SMARCA5 as positive indicator for AZA response, or NSUN3 that correlated with increased overall survival. Taken together we have identified a number of genes associated with AZA response in vitro and in patients. These groups of genes are largely non-overlapping suggesting that different gene sets need to be exploited for the development of combinatorial drug targets and response-predicting biomarkers.