PO-514 High-throughput drug screening to identify therapies reversing aberrant cell differentiation in AML

Introduction Acute myeloid leukaemia (AML) is a malignancy characterised by impaired cell differentiation and uncontrollable accumulation of immature myeloid progenitor cells in the bone marrow. Although most patients respond to the first line chemotherapeutic treatment, the majority of patients eventually relapse and therefore the overall survival from AML remains very poor. A therapy based on inducing the immature leukemic cells to differentiate may turn AML into a curable disease, as has been shown in the case of Acute Promyelocytic Leukaemia (APL), the M3 subtype of AML. However, AML is a heterogeneous disease and we still lack knowledge of how different genetic alterations disturb normal hematopoiesis and how the cell differentiation blockade may be lifted in the various subtypes of AML. Thus, our aims are 1) to identify therapies reversing aberrant cell differentiation in AML by using flow cytometry-based high-throughput drug screening 2) to establish links between molecular subtypes of AML and responsiveness to identified cell differentiating therapies. Material and methods Our approach is to explore drug-induced cell differentiation in primary samples from AML patients and genetically engineered AML mouse models using high-throughput flow cytometry. Specifically, we characterise samples by their expressed cell differentiation-related cell surface markers with and without drug treatment to identify compounds capable of inducing cell differentiation ex vivo . The primary AML patient samples are also profiled by exome- and RNA-sequencing. Results and discussions Preliminary results with a Flt3/Npm1 –mutated mouse model suggest that several compounds elicit cell differentiation-state specific responses, some of which could be indicative of a dynamic shift towards cell differentiation. For example, treatment with mTOR-kinase inhibitors sapanisertib and vistusertib resulted in a concentration dependent increase of differentiating cells marked by CD11b expression and decrease of stem cells marked by CD34 expression. Conclusion We have established a high-throughput flow cytometric screening platform to identify drugs that are able to induce differentiation in primary AML samples and genetically engineered mouse models of AML. By integrating drug response data with genetic background information from samples we aim to reveal novel genotype-phenotype relationships in the level of cell differentiation that could eventually be translated into clinically relevant biomarkers and treatment options for AML patients.