Leukemic stem cells and therapy resistance in acute myeloid leukemia

Abstract

A major obstacle in the treatment of acute myeloid leukemia (AML) is refractory disease or relapse after achieving remission. The latter arises from a few therapy-resistant cells within minimal residual disease (MRD). Resistant cells with long-term self-renewal capacity that drive clonal outgrowth are referred to as leukemic stem cells (LSC). The cancer stem cell concept considers LSC as relapse-initiating cells residing at the top of each genetically defined AML subclone forming epigenetically controlled downstream hierarchies. LSC display significant phenotypic and epigenetic plasticity, particularly in response to therapy stress, which results in various mechanisms mediating treatment resistance. Given the inherent chemotherapy resistance of LSC, targeted strategies must be incorporated into first-line regimens to prevent LSC-mediated AML relapse. The combination of venetoclax and azacitidine is a promising current strategy for the treatment of AML LSC. Nevertheless, the selection of patients who would benefit either from standard chemotherapy or venetoclax + azacitidine treatment in first-line therapy has yet to be established and the mechanisms of resistance still need to be discovered and overcome. Clinical trials are currently underway that investigate LSC susceptibility to first-line therapies. The era of single-cell multi-omics has begun to uncover the complex clonal and cellular architectures and associated biological networks. This should lead to a better understanding of the highly heterogeneous AML at the inter- and intra-patient level and identify resistance mechanisms by longitudinal analysis of patients' samples. This review discusses LSC biology and associated resistance mechanisms, potential therapeutic LSC vulnerabilities and current clinical trial activities.

Figure 1.

Evolution and relapse of acute myeloid leukemia. (A) Illustration of minimal residual disease (MRD) and leukemic stem cell (LSC)-mediated therapy resistance in acute myeloid leukemia (AML). Drug-tolerant persister cells, persisting over treatment and fueling relapse, are illustrated (LSC and non-LSC). (B) Scheme of AML evolution illustrating normal hematopoiesis, clonal hematopoiesis (an age-dependent pre-leukemic state) and clonal outgrowth (overt AML). Pre-LSC, in contrast to LSC, maintain their differentiation ability capable of giving rise to mature blood and immune cells. These mutation-bearing progenitors favor an inflammatory environment, thereby contributing to cardiovascular disease and probably also to clonal expansion. Additional mutations in pre-LSC or mutated multipotent progenitor cells then result in LSC fueling clonal outgrowth. HSC: hematopoietic stem cell, MPP: multipotent progenitor cell, pre-LSC: pre-leukemic stem cell.

Figure 2.

Leukemic stem cell vulnerabilities and targeted therapeutic approaches. The figure illustrates a leukemic stem cell (LSC) and highlights phenotypic characteristics, vulnerabilities and potential therapeutic approaches. LSC are considered metabolically inflexible and uniquely reliant on amino acids and fatty acids to fuel oxidative phosphorylation. BCL-2 and MCL-1 are anti-apoptotic members of the BCL-2 family residing in the outer mitochondrial membrane (OMM). BAX and BAK are pore-forming proteins and the BH3-only proteins are pro-apoptotic. All BCL-2 family proteins interact to maintain the integrity of the OMM. Upon cellular stress, the cell is committed to induce apoptosis via upregulation of the pro-apoptotic BH3-only proteins and downregulation of BCL-2/MCL-1. A shift in the BCL-2 family interactome releases the effector proteins BAX/BAK and promotes homo-oligomerization to form cytotoxic pores in the OMM. Hence, BCL-2 inhibitors (called BH3-mimetics) induce apoptosis. Dihydroorotate dehydrogenase localized at the outer layer of the inner mitochondrial membrane is crucial for de novo pyrimidine synthesis thereby providing substrates for nucleic acid synthesis. DNMT3A and TET2 have opposite effects on DNA methylation. DNMT3A catalyzes de novo methylation of cytosine residues (CpG dinucleotide), while TET2 catalyzes the conversion of 5-methylcytosine to 5-hydroxymethylcytosine, the initial step of DNA demethylation. FLT3: FMS-like tyrosine kinase 3; TKD: tyrosine kinase domain; Brd4: member of the BET (bromodomain and extra-terminal motif) family; MYC/p53/EVI1: transcription factors; LSD1: lysine-specific histone demethylase; PARP1: poly-ADP-ribose polymerase 1, me: methylation; DNMT3A: DNA methyltransferase 3A; TET2: tet methylcytosine dioxygenase 2; BCAA: branched-chain amino acids, BCAT1: BCAA transaminase 1; α-KG: alpha-ketoglutarate; 2-HG: 2-hydroxyglutarate; ROS: reactive oxygen species; OXPHOS: oxidative phosphorylation; TCA: tricarboxylic acid cycle; DHODHi: dihydroorotate dehydrogenase inhibitors; 5-AZA: 5-azacitidine; NK cell: natural killer cell; NKG2D ligand: natural killer group 2D ligand; IDH: isocitrate dehydrogenase.