Diminishing oncometabolic havoc: Approved IDH1 and IDH2 inhibitors in relapsed or refractory acute myeloid leukemia

Activating somatic mutations in isocitrate dehydrogenase (IDH) isoforms 1 and 2 in acute myeloid leukemia 
(AML) have been shown to contribute in wreaking intracellular oncometabolic havoc that adversely affects 
cellular growth and differentiation. Novel developments in IDH1 and IDH2 inhibitors have led to the recent 
U.S. Food and Drug Administration (FDA) approvals of these targeted agents in relapsed or refractory AML 
harboring these respective driver mutations. Their promising efficacy and well-tolerated toxicity profiles 
render them welcomed additions in the treatment landscape of AML. However, the IDH1 and IDH2 
inhibitors ivosidenib and enasidenib, respectively, have unique class-effect toxicities that warrant early 
recognition in order for prompt management and re-institution of an otherwise effective class of agents.


A B S T R A C T
Activating somatic mutations in isocitrate dehydrogenase (IDH) isoforms 1 and 2 in acute myeloid leukemia (AML) have been shown to contribute in wreaking intracellular oncometabolic havoc that adversely affects cellular growth and differentiation. Novel developments in IDH1 and IDH2 inhibitors have led to the recent U.S. Food and Drug Administration (FDA) approvals of these targeted agents in relapsed or refractory AML harboring these respective driver mutations. Their promising efficacy and well-tolerated toxicity profiles render them welcomed additions in the treatment landscape of AML. However, the IDH1 and IDH2 inhibitors ivosidenib and enasidenib, respectively, have unique class-effect toxicities that warrant early recognition in order for prompt management and re-institution of an otherwise effective class of agents.
© 2018 Jun Gong. Hosting by Science Repository. All rights reserved.

Function of Isocitrate Dehydrogenase 1 and 2
Isocitrate dehydrogenase (IDH) isoforms 1 and 2 are nicotinamide adenine dinucleotide phosphate (NADP)-dependent enzymes involved in human cellular NADPH/NADP+ redox reactions [1]. Encoded by the IDH1 gene on 2q33.3, IDH1 catalyzes the oxidative decarboxylation of isocitrate (ICT) to 2-ketoglutarate or α-ketoglutarate (2KG) to generate NADPH from NADP+ as well as the reverse reaction, reductive carboxylation, of 2KG to ICT that oxidizes NADPH to NADP+. In contrast to the reactions carried out by IDH1 in the cytosol and peroxsomes, IDH2, encoded by the IDH2 gene on 15q26.1, catalyzes the same reversible reaction in the mitochondria. IDH1 and IDH2 are structural homologs and function as homodimers with roles in cellular defense against oxidative damage. Evidence is increasing to suggest that activating somatic mutations in IDH1/2 act as driver mutations across a spectrum of malignancies through gain of function, rather than loss of tumor suppression, that result in the intracellular production of the oncometabolite, (D) 2-hydroxyglutarate ((D)-2HG), which wreaks oncometabolic havoc and adversely affects several key enzymes important in cellular growth and differentiation.
IDH1/2 mutations likely affect prognosis in AML depending on the specific mutation and context in which they co-occur with other mutations [4,7]. For example, IDH1(R132) mutations have been shown to predict poor outcome in subsets of molecular low-risk AML, whereas IDH2(R172) mutations significantly confer a worse prognosis than IDH2(R140) mutations in patients with AML [4,7]. As such, favorable risk AML can often be defined by association with NPM1 or CEBPA mutation with neither FLT3/ITD or IDH1 mutations [6]. When IDH2 mutations occur in AML, the majority are IDH2(R140) in 80% of cases while 20% of cases are IDH2(R172) [7]. Relapse in wild-type FLT3/ITD, mutated NPM1, and mutated IDH2(R140) patients was lower than in favorable-risk cytogenetics AML patients [7]. Relapse of AML in patients with IDH2(R172) mutations but wild-type FLT3/ITD and wild-type NPM1 was comparable with adverse-risk cytogenetics [7].

Approved IDH1/2 Inhibitors in AML
Despite many similarities of IDH1 and IDH2 mutations, it is possible that they represent distinct molecular or clinical subgroups of acute myeloid leukemia for targeting [4]. The FDA also approved the Abbott RealTime IDH1 assay for detection of IDH1 mutations.

Toxicities of Interest
The recent FDA approvals of the IDH1/2 inhibitors ivosidenib and enasidenib have afforded additional therapeutic options in the treatment paradigm of relapsed/refractory AML. As these agents are likely to undergo more widespread implementation in clinical practice, it would be prudent for clinicians to recognize the unique toxicities observed in registration trials. For example, non-infectious leukocytosis appears to be a class-effect with the majority of patients experiencing this in the first 30 days of treatment [9]. Prolongation of the QT interval was seen in 44 patients (24.6%) of patients treated with ivosidenib 500 mg daily (18 or 10.1% with grade ≥3), which resulted in dose interruptions for 13 patients (7.3%) and dose reductions in 2 (1.1%). Notably, concurrent medications known to prolong the QT were allowed in this study [9]. indicated [9]. The rate of IDH differentiation syndrome appears to be similar between ivosidenib and enasidenib [8,9].

Conclusion
The IDH1/2 inhibitors ivosidenib and enasidenib represent recent practice-changing approvals in the treatment of relapsed/refractory AML. In unselected patients with relapsed/refractory AML, nontargeted agents have historically produced complete remission (CR) rates of approximately 15% and median OS of <4 months [9]. To this end, they represent exciting developments in treating an otherwise poor prognosis cohort of patients with AML. These novel agents are well-tolerated though warrant special recognition of class-effect toxicities that are manageable if identified early. The future holds bright for these agents and future directions likely involve their integration into combination regimens such as integration into regimens with demethylating agents decitabine and azacitidine given the frequency of DNMT3A along with IDH1/2 mutations in AML [5,10].