Acute myeloid leukemia (AML) are heterogeneous malignancies characterized by the uncontrolled proliferation of immature myeloid cells in the bone marrow. FLT3 mutations, present in approximately 30% of patients, drive leukemic cell survival by enhancing glucose metabolism, amino acid utilization, and redox homeostasis. These metabolic adaptations contribute to therapy resistance, leading to high relapse rates and poor prognosis. Therefore, targeting the specific metabolic vulnerabilities of this patient subgroup could offer new therapeutic opportunities.

To investigate novel FLT3-specific non-genetic mechanisms of therapeutic resistance, we performed a combination of bulk and single-cell multiomics analyses with functional approaches, alongside in vivo experiments, using FLT3-mutant AML cell lines and primary specimens from AML FLT3-mutated patients.

In this context, we identified the transcription factor C/EBPα as a key regulator of lipid metabolism and FLT3 inhibitors response in FLT3-mutated AML. We demonstrated that its inhibition perturbs lipid homeostasis by reducing phospholipid production and increasing PUFA accumulation, sensitizing leukemic cells to ferroptosis, a lipid peroxidation-driven cell death. Notably, we found that ferroptosis inducers potentiate the effects of FLT3 inhibitors through the C/EBPα-SCD pathway, uncovering a metabolic vulnerability that could be therapeutically exploited. Beyond resistance to FLT3 inhibitors, FLT3-mutated AML also exhibits poor response to venetoclax/azacitidine (VEN/AZA), the current frontline therapy for newly diagnosed patients ineligible for intensive chemotherapy. However, the mechanisms underlying the response to VEN/AZA remain poorly understood in this mutational context. Our current study reveals that mitochondrial transaminases play a critical role in VEN/AZA resistance in FLT3-mutated cells. Mechanistically, C/EBPα regulated mitochondrial function and amino acid metabolism through glutamate-pyruvate transaminase 2 (GPT2) expression. Since amino acid metabolism is crucial in VEN/AZA response, we further showed that genetic and chemical inhibition of mitochondrial transaminases GPT2 or glutamate-oxaloacetate transaminase 2 (GOT2) (and not cytosolic transaminase GOT1) triggered mitochondrial oxidative stress and markedly sensitized FLT3-mutant AML cells to VEN/AZA treatment in vivo.

Together, our findings establish a novel role of C/EBPα in regulating both lipid and amino acid metabolism, affecting therapeutic response in FLT3-mutant AML. By identifying lipid biosynthesis enzymes and mitochondrial transaminases as key metabolic drivers of FLT3 inhibitors or VEN/AZA resistance, respectively, our study highlights promising targets to enhance treatment efficacy in this high-risk patient subgroup.