O6-Guanine Targeting Novel DNA Cross-Linker and ATR Inhibitor Combination for MGMT-Silenced IDH1/2 Mutant Acute Myeloid Leukemia

Mutually exclusive mutations in IDH1 and IDH2 account for a total of 20% of AML cases and are commonly found genetic alterations in both de novo and treatment-related AML (t-AML). The use of mutant IDH1/2 inhibitors has demonstrated substantial activity in AML. However, these drugs are not curative, and resistance occurs mostly within the first year of treatment due to either leukemia stemness or selection of mutations in RUNX1/CEBPA or RAS-RTK pathway. Thus, novel biomarker-targeted therapies are needed to target both de novo and refractory/relapsed (r/r) cases of IDH1/2 AMLs. As per TCGA (The Cancer Genome Atlas), around 50-60% of AML exhibit silencing of O6-methylguanine methyltransferase (MGMT), a DNA repair protein that removes alkyl lesions from the O6-position of guanine resulting from alkylation DNA damage. Interestingly, patients with IDH1/2-mutant AML exhibit high MGMT promoter hypermethylation and low protein expression, thus making it a promising targetable biomarker.

Our RNA-seq analysis on 12 AML patient samples (7 wild-type, 5 IDH1/2-mutant) and detailed CpG site-specific investigation of MGMT promoter methylation on 10 AML patient samples (2 wild-type, 8 IDH1/2-mutant) from the Yale repository independently corroborates the TCGA findings. To identify potential therapeutics employing MGMT as a potential biomarker, we first screened a panel of DNA alkylating agents that deposit different alkyl groups (methyl, chloro-ethyl, fluoro-ethyl) onto the O⁶-position of guanine and clinical cross-linkers (nitrogen mustards) against multiple AML cell line pairs isogenic for MGMT. Interestingly, temozolomide (TMZ), which is an FDA-approved MGMT-dependent methylating agent, and our novel first-in-class fluoro-ethylating agent KL50 emerged as top hits imparting the maximum therapeutic index.

Previous MGMT-biomarker based clinical trials in AML using temozolomide (TMZ) highlighted a moderate correlation of MGMT promoter hypermethylation with overall treatment response. A functional mismatch repair (MMR) pathway is the key to the cytotoxic effects of methylating agents like TMZ. We have established both in vitro and in vivo that MMR loss in MGMT-silenced AML often leads to the development of resistance, and evidence suggests a significant loss of MMR in up to 50% of refractory and relapsed cases of AML. Moreover, well-characterized hematological adverse effects (HAEs) associated with TMZ which include aplastic anemia, predisposition to AML or myelodysplastic syndrome (MDS), and severe myelosuppression demand novel MGMT-dependent therapies. Here, we highlight the role of our novel fluoro-ethylating agent (KL50) in targeting MGMT-silenced AML. The time-dependent cross-linking of DNA induced by KL50 selectively activates ATR-dependent DNA damage response pathways independent of MMR. Blocking the ATR-mediated DNA repair axis with AZD6738, an ATR inhibitor (ATRi) currently in clinical trials for myeloid leukemia (NCT03770429) profoundly synergizes with the low doses of KL50 in MGMT-isogenic in vitro and in vivo preclinical AML models. We are now investigating the synergistic efficacy of this novel combination on MGMT-silenced patient-derived xenografts (PDXs) in humanized MISTRG mice models. This clinically translatable study holds tremendous potential for identifying a new class of combinatorial therapy that could target both de novo and refractory/relapsed (r/r) cases of IDH1/2 mutant AMLs.

No relevant conflicts of interest to declare.