Oncometabolite-Independent Anti-Apoptotic Feature Via Phospholipid Metabolic Adaptation in IDH2 Mutant AML Cells

Isocitrate dehydrogenase (IDH) gene mutation is found in approximately 20% of adult acute myeloid leukemia (AML) patients. IDH gene mutations give a pathological enzymatic activity producing oncometabolite, which impair hematopoietic differentiation while promoting leukemogenesis via epigenetic changes. Targeting mutant (mut) form of IDH enzymes, small molecule inhibitors have been developed. Among these inhibitors, AG-221 (enasidenib), which is the first-in-class, selective and orally available mut IDH2 inhibitor (Yen et al., Cancer Discov. 2017), has been tested in clinical trial. The clinical trial showed overall response rate was approximately 40% and more than half of the patient didn't respond to the treatment. This clinical study also showed clear correlation between clearance of IDH2 mut clones and complete clinical response rate (Stein et al., Blood 2019). Relapse cases under the AG-221 treatment, in which additional mutation in IDH2 or IDH1 gene occurred in the residual IDH2 mut clone have been reported (Intlekofer et al., Nature 2018, Harding et al., Cancer Discov. 2018). These reports suggested that IDH2 mut cells have an oncometabolite-independent survival advantage which contributes to refractoriness and relapse. However, most of the studies on IDH mut AML have been focused on oncometabolite-related mechanisms and oncometabolite-independent mechanism has not been investigated in detail so far.

IDH gene mutations frequently coexist with other oncogenic mutations in AML patient cells, which make it difficult to analyze the precise role of IDH gene mutation in leukemogenesis. To address this issue, we utilized the cytokine-dependent AML cell line, TF-1 cells and its isogenic cell line with IDH2 gene mutation, which showed cytokine independent growth. Though AG-221 suppressed the cytokine independent growth in IDH2 mut cells, the growth of AG-221 treated IDH2 mut cells remained significantly faster than IDH2 wild type (WT) cells. These data indicated that IDH2 mut cells have a growth advantage machinery in an oncometabolite-independent manner.

Metabolome analysis clearly separated AG-221 treated IDH2 mut cells from IDH2 WT cells and identified that inositol phosphate metabolism pathway showed significant difference between these two groups. High throughput drug screening using metabolic inhibitors also revealed that inositol phosphate related enzyme, phospholipase C (PLC) specific inhibitor promoted IDH2 mut cell proliferation. These data suggested that suppressed PLC activity induced oncometabolite-independent growth advantage in IDH2 mut AML cells. Since PLC indirectly release free arachidonic acid (AA) from phospholipid bilayer of cellular membrane and intracellular AA level has positive correlation to apoptosis (Cao et al. PNAS 2000), we analyzed intracellular AA and apoptosis in IDH2 mut cells. As expected, these analyses showed significantly lower intracellular AA and reduced Annexin V⁺ apoptotic cells in IDH2 mut cells irrespective of AG-221 treatment. Supporting these data, PLC activity related gene sets were negatively enriched and anti-apoptosis related genes were positively enriched in IDH2 mut cells in RNA seq analysis.

Based on these novel mechanisms, we tested FDA-approved anti-inflammatory drugs, COX-2 and 5-LOX specific inhibitors (celecoxib and zileuton) to upregulate intracellular AA and induce apoptosis of TF-1 cells in vitro and in vivo xenograft. These drugs in combination with AG-221 successfully suppressed IDH2 mut cell proliferation and induced apoptosis both in vitro and in vivo. The mice treated with these triple drugs showed no sign of cytopenia nor therapy-associated complication, securing safety of this combination therapy.

In this study, we identified unappreciated oncometabolite-independent anti-apoptotic feature acquisition via phospholipid metabolic adaptation in IDH2 mut AML cells. Moreover, targeting this metabolic adaptation, we identified that FDA-approved anti-inflammatory drugs in combination with mut IDH2 inhibitor could eradicate IDH2 mut AML cells by efficiently inducing apoptosis. Since these anti-inflammatory drugs have a track record of safety in clinical practice, the triple drug combination therapy could be an alternative therapeutic option for the IDH2 mut AML patients who couldn't tolerate existing therapy.

Takahashi:Celgene/BMS: Consultancy; GSK: Consultancy; Agios: Consultancy; Ostuka Pharmaceuticals: Honoraria; Novartis: Consultancy; Symbio Pharmaceuticals: Consultancy; Mission Bio: Honoraria; Illumina: Honoraria.