FLT3-N676K drives acute myeloid leukemia in a xenograft model of KMT2A-MLLT3 leukemogenesis

Activating signaling mutations are common in acute leukemia with KMT2A (previously MLL) rearrangements. Herein, we show that co-expression of FLT3-N676K and KMT2A-MLLT3 in human CD34+ cord blood cells primarily cause acute myeloid leukemia (AML) and rarely acute lymphoblastic leukemia (ALL) in immunodeficient mice. By contrast, expression of KMT2A-MLLT3 alone cause ALL, double-positive leukemia (DPL, expressing both CD33 and CD19), or bilineal leukemia (BLL, comprised of distinct myeloid and lymphoid leukemia cells), and rarely AML. Further, AML could only be serially propagated with maintained immunophenotype in secondary recipients when cells co-expressed KMT2A-MLLT3 and FLT3-N676K. Consistent with the idea that activated signaling would allow myeloid cells to engraft and maintain their self-renewal capacity, in a secondary recipient, a de novo KRAS-G13D was identified in myeloid cells previously expressing only KMT2A-MLLT3. Gene expression profiling revealed that KMT2A-MLLT3 DPL had a highly similar gene expression profile to ALL, with both expressing key lymphoid transcription factors and ALL cell surface markers, in line with the DPL cells being ALL cells with aberrant expression of CD33. Taken together, our results highlight the need for constitutive active signaling mutations for driving myeloid leukemia in a human xenograft model of KMT2A-R acute leukemia. Overall design: mRNA sequencing of various immunophenotypic populations from KMT2A-MLLT3 xenograft leukemias with or without FLT3-N676K generated using Illumina NextSeq 500.

31

Hyrenius-Wittsten A, Pilheden M, Falqués-Costa A, Eriksson M, Sturesson H, Schneider P, Wander P, Garcia-Ruiz C, Liu J, Ågerstam H, Hultquist A, Lilljebjörn H, Stam RW, Järås M, Hagström-Andersson AK