A Targetable Bone Marrow-Niche Axis for the Treatment of Acute Myeloid Leukemia

Converging lines of evidence show that the endosteal niche within the bone marrow (BM) microenvironment plays a crucial role in the pathogenesis and chemoresistance of myeloid malignancies. Dysplastic cells can exploit niche-dependent, cell non-autonomous pathways to favor their growth. Molecular delineation and targeting of those pathways may help overcome resistance to targeted therapies. In particular, acute myeloid leukemia (AML) remains recalcitrant to conventional chemo- and targeted-therapies leading to relapse and low overall survival rates (5-year survival rate <30%). These shortcomings highlight the urgency of proposing novel and adapted therapies for the disease.

Leveraging genetic mouse models, patient-derived xenografts (PDX), and patient samples, herein we show the mechanism and therapeutic targeting of a cell non-autonomous progression pathway in AML. Specifically, we show that conditional ablation of serotonin receptor 1b (HTR1B) in osteoblasts -during or after AML engraftment- hampers or even prevents AML progression, increasing overall survival. Mass spectrometry-based untargeted metabolomics identified the tryptophan catabolite kynurenine (Kyn) as an oncometabolite secreted by AML cells. Remarkably, the AML proliferative pathway that exploits peripheral serotonin signaling in osteoblasts, is not driven by serotonin but by Kyn, which acts as a previously unrecognized HTR1B ligand.

To explore in vivo the significance of Kyn for leukemia progression, we inhibited its synthesis by suppressing indoleamine 2,3-dioxygenase-1 (IDO1) activity in mouse and human AML cells. CRISPR-Cas9-mediated Ido1 targeting suppressed AML growth in a dose-dependent manner. Next, we sought to identify the downstream molecular targets of Kyn in human osteoblasts that render the BM niche permissive to AML engraftment and support proliferation of leukemia cells. RNAseq analysis of co-cultures between primary human osteoblasts and leukemia cells shows that AML cells induce a pro-inflammatory remodeling of the osteoblastic niche that depends on HTR1B engagement by Kyn. Among the plethora of secreted pro-inflammatory molecules induced in osteoblasts, AML-secreted Kyn (but not the canonical HTR1B ligand serotonin) induces secretion of the acute-phase protein serum amyloid-A (SAA), which in turn acts in a positive feedback-loop on leukemia cells by increasing IDO1 expression, the rate-limiting enzyme for Kyn synthesis. Acting in an autocrine manner -in a mode distinct from its established immunomodulatory properties in cancer- Kyn activates the aryl hydrocarbon receptor (AHR), thereby supporting further AML proliferation.

Highlighting a clinical relevance of these findings, the Kyn-HTR1B-SAA-IDO1 AML-promoting axis identified in leukemia mouse models, is recapitulated in PDX as well as myelodysplastic syndromes (MDS) and AML patients. Preferential and progressive production of Kyn over serotonin by leukemic cells, as well as increased levels of SAA1 in the BM plasma, occur as the disease pathogenesis proceeds from MDS to AML in patients. Moreover, genetic or pharmacological (Epacadostat) IDO1 inhibition hinders AML progression in PDX models. Of note, inhibition of IDO1 synergizes with chemotherapy to reduce AML burden. Overall, these data suggest that the leukemia-osteoblast crosstalk conferred by the Kyn-HTR1B-SAA-IDO1 axis has potential as a therapeutic target. This niche-dependent, cell non-autonomous AML axis, may impact the management of myeloid malignancies, opening also new opportunities for cancer treatment in conjunction with chemo/immunotherapies.