Abstract
Introduction Acute myeloid leukaemia (AML) is a potentially fatal haematological malignancy characterized by the uncontrolled growth of leukemic cells (myeloid blasts), causing bone marrow failure. Despite recent advances in targeted therapeutics, high mortality rates persist across all age groups particularly the patients above the age of 60 years, emphasizing the need for novel and less toxic therapeutic strategies.
KAT2A is a histone acetyltransferase involved in transcriptional regulation as part of two multisubunit complexes: Spt-Ada-Gcn5 acetyltransferase (SAGA) and Ada-Two-A-containing (ATAC), modulating lineage specificity and metabolic/cell-cycle pathways respectively. KAT2A has been shown to be upregulated in AML, where it plays a pivotal role in leukemic stem cell maintenance and chemotherapy resistance.
Pharmacological inhibition of KAT2A has been shown to trigger differentiation in AML blasts while largely sparing normal haematopoietic progenitors, indicating that KAT2A blockade is a mechanistically selective therapeutic approach. We investigated the efficacy and mechanism of action of a novel proteolysis-targeting chimera (PROTAC) degrader; MDI0117218 which targets KAT2A as a differentiation-inducing therapeutic strategy in primary AML blasts.
Methods AML cell lines (MOLM-13, KG1a, OCI-AML3, NB4, HL60) and a large cohort (n=50) of primary AML samples were assessed across a range of assays including: ATP-based proliferation assays (Cell-titre Glo), cytospin morphology, PKH26 cell division tracking and flow cytometric evaluation of differentiation-associated surface markers. KAT2A degradation and subsequent acetylation reduction was monitored by Western blot and intracellular flow for sub-population analysis. Long-term co-culture efficacy with HS5 stroma and colony assays were established to evaluate PROTAC effects on drug resistant adherent and putative stem cell fractions. Synergistic interaction with standard of care agents Cytarabine, Venetoclax and Azacitidine were assessed in combination and pretreatment-washout experiments.
Results MDI0117218 showed nanomolar efficacy in AML cell lines (MOLM13 and OCI-AML3) and 48% of primary AML samples (n=50, Average EC50 33nM +/-324nM in sensitive primary AML samples, Ave 6.6uM +/- 1.4uM in resistant primary AML samples). This effect was observed across both differentiated and undifferentiated FAB types of AML.
Western blotting in primary AML samples confirmed potent and rapid target engagement, with complete and sustained KAT2A degradation observed from 5nM within 30 minutes of incubation. Intracellular flow cytometry confirmed uniform reduction of KAT2A across blast sub-populations in sensitive patient samples compared to those with resistant EC50s. H3K9 acetylation reduction also correlated with KAT2A knockdown in sensitive samples providing potential clinical biomarker readouts. Assessment of MDI0117218 mechanism of action revealed variable cytostatic and differentiation responses with induction of CD11b +/- terminal CD14 differentiation in responders accompanied by cell cycle arrest without apoptotic induction consistent with myeloid differentiation which was confirmed by morphological analysis. Drug efficacy and differentiation was sustained in longer-term stromal co-culture and Colony forming assays, with significant reduction in all blast sub-populations including stromal adherent drug-resistant blasts and colony forming units (p<0.02, n=10). Combination drug assays in ex vivo primary AML blasts showed moderate to additive synergistic interaction between MDI0117218 and cytarabine (Average CI at ED50 0.9 +/- 0.85) and good synergy with Venetoclax (Average CI at ED50 of 0.25 +/- 0.09) across a range of doses. Pre-treatment and sequential washout protocols confirmed synergy to standard agents highlighting the rapid and sustained knockdown effects of MDI0117218.
Conclusion MDI0117218 is an effective and rapid KAT2A degrader across multiple models of cell line and primary AML. Differentiation and anti-proliferative effects were observed across a wide range of differentiated and non-differentiated AML patients, highlighting the potential for a broad acting agent with potential for combination with both Cytarabine and Venetoclax; two mainstays of treatment for transplant-ineligible or relapsed AML. Taken together with favourable pre-clinical toxicology profiles, these data support the rationale for this agent as a strong candidate for early-phase clinical trials.
