Abstract Background: The JAK2 V617F mutation is the most frequent recurring mutation noted in all three types of myeloproliferative disorders (polycythemia vera, essential thrombocytosis, myelofibrosis) and occurs less frequently in other myeloid malignancies. JAK2 V617F occurs in the JH2 pseudo kinase domain of JAK2 and results in hyperactivation of this kinase. The JH2 domain of JAK family kinase members are distinct whereas little difference exists between the JH1 kinase domain. Current available JAK2 inhibitor therapies such as ruxolitinib and fedratinib target the JH1 domain and provide symptomatic relief but are limited by non-selective JAK inhibition and disruption of WT JAK2 signaling. ZE74-2882 is a novel, small molecule designed to selectively inhibit pathologically activated JAK2—such as JAK2V617F—while sparing the wild-type (WT) JAK2 required for normal hematopoietic function described herein. Methods/Results: Utilizing both in-silico and artificial intelligence modeling derived from MolSoft LLC, we identified an optimal small molecule that binds preferentially to JH2 domain of JAK2 versus JH1 domain and to JH2 domain with V617F mutation versus wild type JH2 domain. ZE74-0282 demonstrated 500-fold selectivity to JAK2 V617F JH2 domain versus JAK2 JH2 WT protein in cytoxicity experiment in BaF3 JAK2 V617F cells versus BaF3 JAK2 WT cells. In vitro, ZE74-0282 inhibited BaF3 JAK2 WT cells pSTAT5 phosphorylation and growth at 50% (1456 nM and not reached at 10,000 nM) in contrast to ruxolitinib (30.3 nM and 444.3 nM). In contrast, ZE74-0282 inhibited BaF3 JAK2 V617F cell pSTAT5 phosphorylation and growth at 50% (52.1 nM and 40.6 nM) versus ruxolitinib (3.66 nM and 19.5 nM). SET-2 JAK2 V617F cell line demonstrated similar pSTAT5 inhibition between Ruxolitinib (4.97 nM) and ZE74-0282 (3.46 nM). In vitro, the CC50 of Ruxolitinib and ZE74-0282 was similar against SET-2 JAK2 V617F (10.9 and 24.3 nM) To evaluate the selectivity and functional activity of ZE74-282 we utilized human JAK2V617F+ -MPN whole blood where myeloid cells will have this mutation and lymphoid cells lack it along with murine JAK2V617F and WT cells for colony-forming assays. In JAK2V617F-positive human blood myeloid cells ZE74-282 reduced pSTAT5 levels with an EC50 of 25 nM. Importantly pSTAT5/3 levels were not affected in lymphoid cells, indicating the specificity. In contrast, pSTAT5/3levels were significantly reduced both in myeloid and lymphoid cells upon Ruxolitinib treatment. Viability was equally lost between ZE74-282 and ruxolitinib in human myeloid cells whereas only ruxolitinib resulted in loss of lymphoid cells. In murine cytokine containing colony forming assays, ZE74-282 selectively suppressed colony formation in JAK2ASXL1 mutant progenitors without affecting colonies from WT JAK2-expressing progenitors. In contrast to ruxolitinib, which reduced colony numbers in both genotypes. Pharmacology studies in mice demonstrated linear increases over time with a t1/2 of 7 hours at 28 days. Utilizing a subcutaneous xenograft model of the BaF3 JAK2 617F mouse model, treatment with vehicle, fedratinib (100 mg/kg BID) or ZE74-0282 (100 mg/kg BID) beginning day 14. Versus control, fedratinib modestly delayed growth (p=0.02) whereas growth was inhibited greater with ZE74-0282 (p=0.0001). Superior inhibition of pSTAT5 in vivo was observed with ZE74-0282 as compared to both control and fedratinib treatment. ADME properties of this molecule including mouse GLP toxicity have been favorable for clinical translation to date and will be presented Conclusion: ZE74-282 is a first-in-class, JH2 domain specific JAK2 inhibitor with potent and selective activity against mutant JAK2V617F while preserving normal JAK2-mediated function. Compared with JH1 directed JAK2 inhibitors ZE74-2882 offers a superior therapeutic index in preclinical models, pharmacology suggesting twice daily dosing, and exceptional “drug properties” with ADME studies completed to date. These findings support its further development as a potentially disease-modifying therapy for JAK2 V617F mutated diseases.
Khvat et al. (Mon,) studied this question.