Impact of Age-Related Clonal Hematopoiesis on Evolution of Atherosclerosis in CML

Recent data suggest that hematopoietic ageing is associated with an accumulation of somatic mutations in myeloid stem cells and that acquisition of such mutations is associated with an increased risk a) to develop a myeloid malignancy and b) to develop a severe cardiovascular disease. Correspondingly, the incidence of cardiovascular events increases with age and is higher in patients suffering from certain myeloid neoplasms. However, little is known about cellular interactions and molecular mechanisms contributing to cardiovascular changes in these patients. One hypothesis is that clonal leukocytes mediate inflammation and vascular damage and thus contribute to atherosclerosis. Another hypothesis is that vascular endothelial cells are derived from pluripotent neoplastic stem cells and that the acquired mutations alter endothelial cell function which in turn facilitates the development of atherosclerosis. We and others have recently shown that patients with chronic myeloid leukemia (CML) treated with nilotinib or ponatinib have an increased risk to develop severe cardiovascular events. We also found that nilotinib and ponatinib exert pro-atherogenic and growth-inhibitory effects on endothelial cells. More recently we detected age-related loss-of-function (LOF) mutations (TET2, DNMT3A, ASXL1) in CML patients and found that these mutations cluster in patients who develop severe cardiovascular events during nilotinib-therapy. In the current project, the functional impact of age-related LOF-mutations and their distribution among hematopoietic and non-hematopoietic cells in CML will be investigated. In a first step, clonality of various leukocyte-subsets and endothelial cells will be determined at diagnosis and at the time of deep molecular response. Next, the effects of age- related LOF-mutations on macrophage- and endothelial cell function will be examined. The pro- atherogenic effects of LOF-mutations will be analyzed by measuring expression of pro-atherogenic receptors and mediators in endothelial cells and macrophages after CRISPR Cas 9 or shRNA-mediated knock-down of LOF-target-genes. In addition, the impact of LOF-mutations will be studied in Ldlr-/- mice crossed with LOF gene-deficient mice. In order to define early stages of pluripotent stem cells and their angiogenic potential in CML, we will generate induced pluripotent stem cells (iPSC) from CML cells. We will also explore the effects of various drugs, including nilotinib and ponatinib, on differentiation of iPSC into endothelial cells and macrophages. Long term aims are to 1) identify target-pathways triggering differentiation of CML stem cells into endothelial cells and macrophages, to 2) define the pro-atherogenic and anti-angiogenic impact of LOF-mutations, and to 3) identify drugs capable of interfering with stem cell differentiation into endothelial cells and macrophages and with the pro-atherogenic activity of LOF mutations. Our study should contribute to a better understanding of mechanisms underlying vascular changes in CML patients during nilotinib or ponatinib and may pave the way for new therapeutic strategies through which cardiovascular events can be prevented in these patients.

Age-related clonal hematopoiesis (ARCH) is frequently detected in myeloid leukemias and is associated with an increased risk to develop cardiovascular events (CVE). In patients with chronic myeloid leukemia (CML) receiving nilotinib or ponatinib, the prevalence of CVE is particularly high. The current project analyzed the mechanisms underlying drug-induced CVE in these patients and asked for alternative drugs that would not cause CVE. The results of the project show that ARCH mutations are detectable in CML patients before and during therapy. ARCH mutations were detected in CML stem cells as well as in monocytes and sometimes even in endothelial cells. The most frequently mutated genes were DNMT3A, TET2 and ASXL1. Nilotinib and ponatinib were found to affect endothelial cell growth and function, whereas asciminib, a novel anti-CML drug, showed no effects on these cells. Whereas nilotinib induced the expression of proatherogenic molecules in endothelial cells, no such effects were observed with ponatinib or asciminib. These data would favor asciminib as less toxic drug. Next, the study team asked whether targeting of niche cells may be a potential therapeutic strategy to overcome niche-induced resistance of leukemic cells in CML. In these studies the osteoblast was identified as a primary site of resistance of CML cells against nilotinib and ponatinib. Drugs directed against the PI3 kinase (PI3K) were found to suppress the proliferation of osteoblasts and to overcome osteoblast-mediated resistance of CML cells against nilotinib and ponatinib. Since the PD-1/PD-L1 axis has recently been implicated in drug resistance and vascular repair, and PD-1/PD-L1 inhibition may lead to CVE, the project also examined expression of these checkpoint molecules on niche cells and CML cells. It was found that endothelial cells and osteoblasts express PD-L1 and that interferon-gamma (IFN-G) and tumor necrosis factor-alpha (TNF-A) promote expression of PD-L1 in these cells. The BRD4/MYC blocker JQ1 and the BRD4 degraders dBET1 and dBET6 suppressed cytokine-induced upregulation of PD-L1 in these cells. Even in cells exposed to both cytokines, JQ1, dBET1 and dBET6 suppressed the upregulation of PD-L1. CML cells were also found to express PD-L1. In addition, the study team found that IFN-G and TNF-A augment the expression of PD-L1 in CML cells, and that BRD4/MYC-targeting drugs counteract cytokine-induced expression of PD-L1 in these cells. In summary, the data obtained in the current project show that nilotinib and ponatinib exert pro-atherogenic effects in endothelial cells, whereas asciminib did not affect endothelial function. Moreover, it was found that osteoblasts mediate drug resistance in CML cells and that PI3K inhibitors overcome osteoblast-mediated drug resistance. Finally, the data show that endothelial cells, osteoblasts and CML LSC express the immune checkpoint molecule PD-L1.