Lysine-Specific Histone Demethylase, LSD1, (KDM1A) As a Novel Therapeutic Target in Myeloproliferative Neoplasms

Among BCR-ABL-negative myeloproliferative neoplasms (MPN), primary myelofibrosis (PMF) and post PV/ET myelofibrosis (MF) are associated with the highest degree of morbidity and mortality, including progressive bone marrow (BM) fibrosis and resultant BM failure. Although the JAK inhibitor ruxolitinib is now approved for the treatment of MF-associated splenomegaly and systemic symptoms, JAK inhibitor therapy does not reduce the proportion of JAK2-mutant cells in MPN patients. The limited ability of JAK inhibition to induce molecular or clinicopathologic responses in the majority of MPN patients underscores the need for the development of more effective therapies for JAK kinase-dependent malignancies.

Recent studies have shown that the lysine-specific histone demethylase, LSD1 (KDM1A), participates in the balance between proliferation and differentiation in vivo by influencing state-specific gene expression patterns. In physiologic hematopoiesis, LSD1 is essential for normal myeloid differentiation affecting the erythroid, megakaryocytic and granulocytic lineages. Small molecule inhibitors of LSD1 have shown promising results in preclinical models of acute myeloid leukemia (AML) and solid cancers and have recently entered clinical trials in AML. However, the role and requirement for LSD1 in the pathogenesis of MPNs and the therapeutic targeting of LSD1 in MPN has not been investigated.

In this study, we first tested the effects of IMG-98, a potent, selective LSD1 inhibitor, in the MPLW515L-driven ET/MF mouse model. After disease was established, mice were treated with IMG-98 or vehicle for 28 days. LSD1 inhibition in mice markedly suppressed myeloproliferation reducing granulocyte counts and spleen weights compared to mice treated with vehicle thus establishing therapeutic efficacy (Fig. 1a). Pathologic analysis of BM and spleen confirmed a marked reduction in myeloproliferation as well as a reversal of extramedullary hematopoiesis (EMH). Most notably, we observed a marked reduction in reticulin fibrosis with IMG-98 treatment (Fig. 1b). We next investigated the impact of IMG-98 therapy on inflammatory cytokine signaling; in contrast to the broad anti-cytokine effects of JAK1/2 inhibition, we observed a more specific anti-cytokine effect of IMG-98, a significant reduction in the secretion of the inflammatory cytokine Cxcl5 (Fig. 1c), a key participant in pathologic inflammatory states.

We then investigated the in vivo impact of IMG-98 therapy on mutant disease burden. IMG-98 therapy reduced mutant allele burden to a degree not seen with JAK1/2 inhibitor therapy: whereas 74.6% of circulating cells in mice treated with vehicle were GFP-positive cells, only 43.2% of circulating cells were GFP-positive in IMG-98-treated mice (Fig. 1d). Flow cytometry analysis of spleen and BM revealed reduced numbers of CD11b/Gr1-positive myeloid cells and CD41-positive megakaryocytes. The numbers of mutant GFP-positive myeloid cells and megakaryocytes in these tissues were also significantly reduced by IMG-98 treatment. Studies of the impact of LSD1 inhibition on MPN stem cell function and on epigenetic regulation in MPN cells will be presented in detail.

In summary, the LSD1 inhibitor IMG-98 had a highly significant therapeutic effect in an established preclinical model of ET/MF. LSD1 inhibition in diseased mice reduced JAK-STAT-driven myeloproliferation, markedly reversed EMH and BM fibrosis, and reduced the mutant clone burden. These data suggest LSD1 is a valid target in MPN and that clinical studies of LSD1 inhibitor IMG-98 alone and in combination with JAK inhibitors are warranted.

Figure 1. a, b) LSD1 inhibition results in reduced white blood cell counts (WBC) and platelet counts (PLT). (a), and in near-complete elimination of BM fibrosis (b). c) Profound reduction of Cxcl5 serum levels in IMG-98 treated mice compared to vehicle treated mice. d) Significantly lower mutant allele burden in the peripheral blood of IMG-98 treated mice. * P<0.05, n=5.