Minimal Residual Disease in Acute Myeloid Leukemia Following Induction Chemotherapy Can be Effectively Eradicated By Targeting Mitochondrial Metabolism

Acute myeloid leukemia (AML) stem cells (AMLSCs) AMLSCs and residual cytarabine (AraC)-resistant AML cells (constituting minimal residual disease, MRD) thought to be responsible for chemoresistance and treatment failure, were shown to be highly dependent on mitochondrial function for survival and thus are vulnerable to pharmacological blockade of the oxidative phosphorylation (OXPHOS) (Farge et al. Cancer Discov, 2017). Efficacy of OXPHOS inhibitor IACS-010759 (OXPHOS-i) was previously reported, demonstrating potent inhibition of mitochondrial complex I, OXPHOS suppression and growth inhibition of AML cells (Molina, et al. Nat Med, 2018). Here we evaluated OXPHOS dependency of AML MRD cells and determined impact of OXPHOS blockade on residual AML cells surviving standard chemotherapy (Doxorubicin/AraC, DA). Our results demonstrated that AML cell lines treated with AraC or DA induced accumulation of reactive oxygen species, mitochondrial superoxides, increased mitochondrial mass and mitochondrial membrane potential. AraC- and DA therapies in vitro were significantly enhanced by OXPHOS-i. OXPHOS dependency shown as a significantly increased basal and maximal oxygen consumption rate after AraC and DA treatment, was fully inhibited by OXPHOS-i, leading to complete mitochondrial collapse. OXPHOS inhibition in combination with DA translated into reduction of viable cell numbers, induction of apoptosis and differentiation in AraC-sensitive and AraC-resistant cell lines models. In vitro efficacy was also observed in engineered p53-mutated MOLM13 model, indicating that combination of OXPHOS-i and DA might successfully overcome p53 mutation-driven chemoresistance. These effects were further validated in a subset of AraC-resistant primary patient samples. Mechanistically, induction of ROS caused by OXPHOS-i addition upon DA contributed to differentiation and cell death, and was partially reversed by ROS scavengers.

Next, the efficacy of IACS-010759 together with DA chemotherapy was evaluated in several chemotherapy-sensitive and -resistant animal models in vivo. DA/IACS-010759 combination significantly reduced leukemia burden and extended survival in OCI-AML3/Luc/GFP model and in FLT3-ITD⁺ AML PDX model. In the latter, IACS-010759 led to the reduction of leukemia burden and delayed leukemia recurrence when administered post completion of DA. At the single-cell level, CyTOF analysis demonstrated that this combination reduced frequency of CD34⁺CD38^lowCD123⁺AML LSCs and facilitated differentiation of immature subpopulations (Fig.1A). Furthermore, addition of OXPHOS-i to DA extended survival of mice inoculated with chemoresistant PDX AML models. Finally, OXPHOS-i administered during consolidation phase significantly extended mice survival compared to standard of care arm, supporting clinical utility of OXPHOS inhibitors in AML (Fig.1B).

In conclusion, our findings indicate that chemotherapy fosters mitochondrial respiration in AML, which could be abrogated by OXPHOS inhibitor at the LSCs and MRD level, in vitro and in vivo. While IACS-010759 (Yap et al. Nat Med 2023) showed toxicities impeding its clinical utility, our data advocate for combining mitochondrial targeting strategies with chemotherapy as a part of induction and consolidation treatment for improved control of MRD, eradication of AMLSC and extended response duration. Thus, further studies to identify compounds with improved safety profile are warranted.