Targeting Defective DNA Damage Response Pathways in IDH1/2-mutant AML

Heterozygous mutations in two key metabolic genes, isocitrate dehydrogenase-1 and -2 (IDH1/2), are
present in up to 20% of newly diagnosed AML patients. IDH1/2 enzymes convert isocitrate to a-ketoglutarate
(aKG) in the tricarboxylic acid (TCA) cycle. IDH1/2 mutations impart a neomorphic enzyme activity, leading to
the conversion of aKG to the oncometabolite, 2-hydroxyglutarate (2HG). 2HG competitively inhibits aKG-
dependent dioxygenases, which induces profound epigenetic alterations and impaired hematopoietic
differentiation. IDH1/2 inhibitors are now FDA-approved for AML, although these agents typically are not
curative, with complete response (CR) rates and median overall survival (OS) ranging between 20-30% and
~8-12 months, respectively. In addition, primary and acquired resistance to mutant IDH1/2 inhibitors commonly
occurs. The inability to achieve a cure with these drugs as a monotherapy in part can be attributed to their
mechanism of action. Specifically, these drugs act in a cytostatic manner via the induction of differentiation,
which is highlighted by persistence of mutant IDH1/2 clones in the majority of patients, even those who achieve
a CR. These data underscore the need to develop alternative approaches to target IDH1/2-mutant AML.
Our team recently discovered that IDH1/2 mutations induce a DNA damage response (DDR) defect which
confers sensitivity to poly(ADP)-ribose polymerase (PARP) inhibitors. Mechanistically, we demonstrated that
2HG-induced inhibition of the lysine demethylase, KDM4B, results in aberrant hypermethylation of histone 3
lysine 9 (H3K9) at loci surrounding DNA breaks, masking a local H3K9 trimethylation signal that is essential for
the proper execution of homologous recombination (HR), a key DNA double-strand break (DSB) repair
pathway. We also extended these findings to other TCA gene mutations which create oncometabolites, which
we have collectively termed “oncometabolite-induced BRCAness”. Our work suggests that oncometabolite-
induced BRCAness is tumor type-agnostic, and we are now directly translating this work into multiple clinical
trials, which currently are testing the efficacy of PARP inhibitors against IDH1/2-mutant cancers, including AML
(NCT03953898; the PRIME trial; PI: Prebet).
It is now well-established that IDH1/2 mutations induce DDR defects in AML, and here we propose to
study: (a) the impact of common, co-occurring mutations in AML on the associated DDR defect, which will be
critical for therapeutic targeting; (b) which DDR inhibitors will be most effective, and whether combinations with
other systemic agents in AML will increase efficacy; and (c) the extent to which our DDR inhibitor-based
strategies will be effective against tumors with intrinsic or acquired resistance to therapy. These studies have
the potential to establish an entirely new therapeutic approach for newly diagnosed and relapsed IDH1/2-
mutant AML, which exploits DDR defects identified by our team. By focusing on drugs which are either FDA-
approved or in clinical trials, our work can be rapidly translated into the clinic.

Grant Number: 5R01CA266604-05
NIH Institute/Center: NIH
Principal Investigator: Ranjit Bindra