Objective Internal tandem duplication (ITD) of fms-like tyrosine kinase 3 (FLT3) is one of the most common mutations in acute myeloid leukemia (AML), occurring in nearly 30% of cases. FLT3-ITD involves in-frame duplication of 3-400 base-pairs at the juxta-membrane domain, resulting in ligand-independent activation of FLT3 signaling. Downstream effectors include activation of ERK/STAT5 via SRC kinase, activation of PI3K/AKT, phosphorylation of FOXO3A, down-regulation of equilibrative nucleoside transporter 1 (ENT1) for cytarabine, and induction of reactive oxygen species (ROS) that may lead to increased DNA damage and defective repair. The present study investigated if the latter can be effectively targeted for the treatment of this AML subtype. Methods Primary samples from patients with FLT3-ITD AML, human cell line carrying FLT3-ITD (MOLM-13) as well as mouse B-lymphoid Ba/F3 cells transduced with human FLT3-ITD were used in this study. Traffic Light Reporter (TLR) assay was used to measure fidelity of double-strand breaks (DSB) repair, either via error-free homologous recombination (HR) or error-prone non-homologous end joining (NHEJ). Percentage of repair events by HR or NHEJ were quantified by flow cytometry. DNA DSB were examined by γ-H2AX foci staining using confocal microscopy. Single-cell DSB were quantified by neutral comet assay and analysed by OpenComet software. The tail moment was calculated as the length of comet tail multiplied by the tail DNA %. MOLM-13 was transplanted into NOD/SCID/IL2Rg-/- (NSG) mice by tail vein injection. Treatment comprised cytarabine (25mg/kg, i.p., day 1-5) and doxorubicin (1.5mg/kg, i.v., day 1-3), with or without olaparib (25mg/kg, i.p., day 1-5). Comparisons between groups of numerical data were evaluated using Student's t-test. P-values less than 0.05 were considered statistically significant. Results To investigate the link between FLT3-ITD AML and DNA damage response (DDR), expression of critical DDR genes in primary AML samples was examined by real-time quantitative PCR. The panel of genes included apical kinase ATM, ATR and DNA-PKcs; DNA damage mediators BRCA1,BRCA2 and PARP1; downstream response kinase CHEK1 and CHEK2 and effectors TP53. Among them, BRCA2 was significantly down-regulated in FLT3-ITD AML(Wild-type FLT3=18 samples; FLT3-ITD=13 samples; Average dCt 9.7 in WT vs. 10.6 in ITD; p<0.05). Down-regulation of BRCA2 in FLT3-ITD AML was further validated in a microarray database GSE15434 from a multicenter study investigating gene expression profiles of normal karyotype AML( FLT3-WT=148; FL3-ITD=86; ; log2 gene expression 5.61 in WT vs 5.45 in ITD; p<0.001). As BRCA2 is an important protein mediating HR, a DSB DNA repair TLR assay was performed. HR was significantly down-regulated in Ba/F3-FLT3-ITD as compared with parental Ba/F3 line by flow cytometry(1.21% HR repair in control vs. 0.44% in ITD; p<0.05) while NHEJ was unaffected (1.90% NHEJ repair in control vs. 2.38% in ITD; p=0.50). Down-regulation of BRCA2 expression and defective HR in FLT3-ITD AML was reminiscent of BRCA mutant breast and ovarian cancers. Therefore, the effects of poly (ADP-ribose) polymerase (PARP) inhibitor (PARPi) olaparib were examined. Ba/F3-FLT3-ITD were more sensitive to olaparib than parental line. Olaparib inhibited base excision repair and increased DSB as indicated by increased number of γ-H2AX foci and increased tail moment in Ba/F3-FLT3-ITD cells when compared to parental line. Olaparib-induced DSB in Ba/F3-FLT3-ITD cells was mainly repaired by NHEJ as shown by colocalization of γ-H2AX foci with 53BP1. Combination of chemotherapy (cytarabine and doxorubicin) and olaparib synergistically reduced leukemic cell growth of MOLM-13 in NSG murine xenograft model. Conclusion Results from the present study supported the use of PARPi in the treatment of FLT3-ITD AML. Its therapeutic benefits in combination with chemotherapy would have to be further evaluated. Acknowledgements: Health and Medical Research Fund (Project number: 04152326) and Li Shu Fan Medical Foundation, LKS Faculty of Medicine, University of Hong Kong. Disclosures No relevant conflicts of interest to declare.