Preclinical Validation of EZH2 and HDAC I Dual Inhibition As a Potent Therapy for Refractory Myeloid Leukemia Associated with Down Syndrome

Children with Down Syndrome who are less than 5 years old have approximately 400-fold increased risk of developing myeloid leukemia. Even though children with myeloid leukemia associated with Down syndrome (DS-ML) are more sensitive to chemotherapy than their non-DS-ML counterpart, relapse and treatment-related toxicity needs to be eliminated to achieve better outcome for DS-ML children. Epigenetic alterations including DNA methylation and histone modifications play a key role in the pathogenesis of DS-ML. Enhancer of zeste homolog 2 (EZH2) is one of the epigenetic regulators that harbors high frequency of mutations in DS-ML. EZH2 is a catalytic subunit of PRC2 complex which suppresses gene expression by di- and tri-methylation of lysine27 on histone 3 (H3K27me2/3). EZH2 recruits class I histone deacetylases (HDAC, which remove acetyl groups from histones), on target genes for further suppression of gene expression. Using patient-derived xenograft models of DS-ML, we showed previously that combination of DNA hypomethylating agent azacitidine with HDAC inhibitor panobinostat or BCL2 inhibitor venetoclax showed significant improvement in median survival. Here we aimed to show synergistic effect of EZH2 inhibitor, GSK126, and class I HDAC inhibitor, romidepsin.

The viability of CMK, a DS-ML cell line, and NTPL-60 and NTPL-386, in house generated DS-ML PDX lines, was determined by exclusion of propidium iodide-stained cells by flow cytometry. IC50 values for GSK126 and romidepsin were calculated using GraphPad Prism (dose-response curve). The IC50 of GSK126 was 40.5, 18.0, and 17.2 mM, while that of romidepsin was 10.2, 9.2, and 3.0 nM respectively. To determine synergy, the viability of cells exposed to a variety of concentrations of GSK126 and romidepsin alone or in combination was measured. Synergy index was calculated using relative risk ratio (RRR) formula as described previously (Quagliano et al, 56:36, Leuk Res, 2017). All the 3 DS-ML lines showed RRR value of less than 1 indicating synergistic inhibition of cell death by GSK126 and romidepsin treatment in ex vivo culture (Figure 1A).

NTPL-386, a refractory DS-ML PDX line, which showed minimal response to chemotherapy (11-day improvement in median survival, Barwe et al., 134:2683, Blood, 2019), was chosen to determine the in vivo efficacy of GSK126 and romidepsin combination. NTPL-386 engrafted mice (n=5 each) were treated with GSK126 (100 mg/Kg) and/or romidepsin (2 mg/Kg) twice weekly for three weeks. Thus, one treatment cycle consisted of six doses. A separate cohort of mice were treated with four cycles of the combination, with two weeks rest between cycles. Disease progression was monitored by percentage of human cells in peripheral blood by flow cytometry. Kaplan-Meier survival estimates were plotted based on the time when mice were euthanized because they met pre-determined experimental endpoints. Compared to untreated mice, GSK126 or romidepsin treatment showed an increase in median survival by 8 days and 11 days respectively (Figure 1B, P <0.005). The combination of these drugs showed a remarkable improvement in survival (41 days). The difference in the median survival of the mice treated with the combination was statistically significant when compared to GSK126 or romidepsin (P <0.005). Further treatment for three additional cycles showed sustained remission upto 150 days post cell injection, when the study was terminated. Prolonged treatment showed a statistically significant improvement in survival compared to one treatment cycle (P <0.05). Further studies to characterize the mechanism of synergy are in progress. In summary, our data show a synergistic effect of EZH2 and HDAC class I inhibition on DS-AML growth and pave the path for clinical evaluation of this combination.

No relevant conflicts of interest to declare.