Anthracyclines induce epigenetic changes in ALDH1A that promote chemoresistance in AML, revealing a potential treatment strategy co-targeting ALDH1A inhibition with hypomethylating agents

Abstract Background: Relapse/Refractory Acute Myeloid Leukemia (R/R AML) is a condition where patients do not achieve complete remission after chemotherapy and thus develop chemo-resistance. Aldehyde dehydrogenase 1A (ALDH1A) plays a critical role in leukemic cell survival by maintaining stemness, proliferation, and chemoresistance through aldehyde detoxification and retinoic acid synthesis. Elevated ALDH1 expression has been linked to poor prognosis in acute myeloid leukemia (AML), but the underlying mechanisms remain poorly defined. Methods: Molecular and cellular techniques, including GFP-ALDH1 knock-in models, site directed mutagenesis, ChIP-seq, luciferase reporter assays, RT-qPCR, protein expression analysis, capillary electrophoresis-based protein analysis, ALDH enzymatic tests, ATP measurements. drug interaction studies, and colony formation assays were employed. Experiments were conducted using human AML cell lines, murine MLL-AF9 leukemia models, murine and human AML mouse models, and R/R AML patient samples. In addition, ALDH1 expression was analysed in an independent cohort of 116 AML patients and using data from BEAT AML public database. Results: While analysing ALDH1 in an independent cohort of 116 AML patients and integrated data from the BEAT AML public database, we observed that refractory AML blasts exhibited significantly higher ALDH1 levels than non-refractory cases, particularly in patients with prior myelodysplastic syndrome (MDS), MDS/myeloproliferative neoplasms, or therapy-related AML. Longitudinal analysis revealed that as the chemotherapy cycles progressed, there was an upregulation of ALDH1 on circulating blasts, thus suggesting an ALDH1-high phenotype induced by the treatment. High ALDH1A1 expression strongly correlated with reduced overall survival, particularly in R/R AML and patients with prior anthracycline exposure. To evaluate the role of anthracyclines in the transcription of ALDH1A1 and ALDH1A2, we carried out genetic studies using GFP-ALDH1 knock-in models, site-directed mutagenesis, ChIP-seq, luciferase reporter assays, RT-qPCR, and protein expression analysis. Our results demonstrated that anthracyclines directly activate ALDH1A1 and ALDH1A2 transcription via drug-inducible enhancer elements, regulated by JNK and STAT transcription factors. This was further supported by a combinational treatment of ALDH1 inhibition with Daunorubicin in a KG1a mouse model. We then evaluated the role of ALDH1A in promoting therapeutic resistance by performing functional studies using capillary electrophoresis-based protein analysis, ALDH enzymatic tests, and ATP measurements. These results demonstrated the role of ALDH1A in oxidative stress adaptation, and promoting therapeutic resistance. We then carried out drug response assays to determine the effect of ALDH1A inhibition in combination with 10 standard-of-care chemotherapeutic drugs in a panel of AML cell lines, we observed that ALDH1A inhibition in combination with hypomethylating agents has a synergistic role on AML cell lines. The effect of ALDH1A inhibition in combination with hypomethylating agents was further explored using colony formation assays and the synergistic interaction of ALDH1A inhibition with 5-Azacytidine was further supported by transplanting murine MLL-AF9 leukemic cells in C57BL/6J mice. To further understand the synergistic effect of ALDH1A inhibition with hypomethylating agents on primary R/R AML, we performed colony formation assays. Lastly, we determined the molecular mechanism of ALDH1A inhibition by capillary electrophoresis which further revealed that ALDH1A inhibition disrupts key proliferative pathways and induces extrinsic apoptosis via the Noxa/Mcl-1 axis, remaining effective in R/R AML. Conclusion: Our findings identify ALDH1A as a key driver of adaptive resistance in anthracycline-exposed AML, highlighting its potential as a therapeutic target. Preclinical in vivo models and patient-derived R/R AML samples demonstrate that ALDH1A inhibition enhances anthracycline sensitivity when sequentially administered. Alternatively, co-targeting ALDH1A and DNA methylation offers a promising salvage strategy in R/R AML cases. These results support the clinical development of ALDH1A inhibitors for high-risk AML subsets.