Background: Outcomes for pediatric patients with relapsed or refractory acute myelogenous leukemia (AML) are poor, with <30% overall survival in recurrent settings. Additionally, patients harboring high-risk molecular features, including KMT2A/MLL-rearrangements or GLIS2 fusion genes, experience poor outcomes. LSD1 inhibition has been evaluated as a potential therapeutic approach in adult patients with AML, but its role in pediatric AML remains limited. In this study, we characterize the anti-leukemic effect of an LSD1 inhibitor, IMG-7289, across a panel of pediatric AML models in vitro and in vivo and evaluate its effect on leukemia initiating cells (LICs). Methods: Dose-response effect of IMG-7289 on cell viability was assessed across 11 pediatric leukemia cell lines. Induction of apoptosis and cell cycle arrest was assessed by flow cytometric analysis of Annexin V and 7-AAD respectively. Flow cytometry was also used to analyze a panel of cell surface markers to assess for differentiation effect. Pediatric AML patient-derived xenograft (PDX) mouse models, including an acute megakaryoblastic leukemia (AMKL) model and a KMT2A-rearranged (KMT2A-r) AML model, established by implanting AML cells into NSG-S mice, were used for in vivo validation studies and limiting dilution secondary transplantation studies. PDX models were treated with IMG-7289 (25 mg/kg PO) or Vehicle daily for 21 days (n=6/arm). Engraftment was determined using flow cytometry to assess expression of human CD45 (hCD45) in the bone marrow. Leukemia initiating cell (LIC) frequency was evaluated by transplanting serial dilutions (2x105, 105, 104, 103, and 102 cells) of bone marrow cells derived from Vehicle- or IMG-7289-treated PDX models into untreated donor mice. The LIC frequency was estimated using the Extreme Limiting Dilution Analysis (ELDA) software (Walter and Eliza Hall Bioinformatics Institute of Medical Research), and log-rank analysis was used to compare event-free survival (EFS), defined as time from secondary transplant to engraftment. Results: Seven pediatric AML cell lines demonstrated sensitivity to IMG-7289, including 2 AMKL cell lines, M07e (IC50: 0.03 μM ± 0.01) and CMK (IC50: 0.065 μM ± 0.003), and 2 KMT2A-r AML cell lines, MV4;11 (IC50: 0.007 μM ± 0.001) and MOLM14. Treatment with IMG-7289 leads to induction of apoptosis (DMSO 8.8% vs IMG-7289 30.3%, P=0.0005) and cell-cycle arrest (G0/G1 phase populations: DMSO 65.5% vs IMG-7289 74.7%, P<0.0001) in MV4;11 cells. To investigate the effect of IMG-7289 on induction of cell differentiation, cell surface marker changes were evaluated over time in IMG-7289-treated AML. MV4;11 and M07e cells treated over 25 days showed increased expression of the monocytic cell surface markers CD14 and CD86 compared to DMSO control. Additionally, decreased expression of megakaryocytic markers CD41 and CD42b were observed in IMG-7289-treated M07e cells. To evaluate the in vivo activity of IMG-7289, a KMT2A-r AML PDX model was treated with IMG-7289 for 21 days. We observed a significant reduction in hCD45+ cells in IMG-7289-treated mice (27.5% hCD45+) vs Vehicle-treated mice (76% hCD45+, P=0.0012). To evaluate effects of IMG-7289 on the LIC population, serial transplantation of cells derived from Vehicle-treated or IMG-7289-treated AMKL PDX model was performed. Assessment of leukemia engraftment (flow cytometric analysis for hCD45+ cells in mouse bone marrow) at 20 weeks post-secondary transplant demonstrated an LIC frequency of 1 in 4.9x104 in IMG-7289-treated mice representing a ~500-fold reduction in LIC number compared to Vehicle control. Furthermore, a significant improvement in EFS was observed in IMG-7289-treated mice (median survival = 33 days) compared to Vehicle control (median survival = 76 days, P=0.0009, log-rank). Conclusions: In conclusion, treatment with IMG-7289 results in decreased viability, as demonstrated by induction of apoptosis and cell cycle arrest, as well as differentiation induction in pediatric AML cell lines. The in vitro and in vivo activity observed in pediatric AML is also observed in models with high-risk phenotypes (KMT2A-r AML and AMKL). Activity of LSD1 inhibition on the LIC population may represent a promising strategy to mitigate relapse or refractory disease.
Read full abstract