Preclinical characterization of a novel, wild-type-sparing, JAK2 V617F mutant-selective inhibitor

Abstract Background The JAK2 V617F mutation is the most prevalent molecular abnormality in BCR-ABL-negative myeloproliferative neoplasms, occurring in approximately 95% of patients with polycythemia vera, and 50% of patients with essential thrombocythemia or primary myelofibrosis. This acquired point mutation in the JAK2 pseudokinase (JH2) domain results in the constitutive activation of JAK2 kinase signaling leading to cytokine-independent growth of hematopoietic cells. Clinically approved JAK2 inhibitors, such as ruxolitinib, are kinase domain (JH1) binders that effectively alleviate symptoms and improve outcomes in patients with myeloproliferative neoplasms. However, even with long-term treatment, this therapy rarely leads to significant reductions in variant allele frequency (VAF) or molecular remission because they are dose-limited by hematologic toxicities. A JAK2 V617F mutant-selective inhibitor that spares wild-type (WT) JAK2 has the potential to selectively target and eliminate mutant clones and induce molecular remission while avoiding hematological tolerability issues. Herein, we report the preclinical characterization of CGT1145, a JAK2 V617F JH2 selective mutant inhibitor lead which shows potential best-in-class potency and selectivity against JAK2 WT. Methods/Results The primary SAR for the lead series (including CGT1145) was navigated through JAK2 JH2 WT and JAK2 JH1 WT binding affinity and selectivity assays. A custom HTRF competition binding assay, using purified JH1 and JH2 domains and custom fluorescent tracers, demonstrated preferential binding to the JH2 domain (6300 nM JH1 / 0.3 nM JH2, 21,000-fold selective). Cellular activity was measured using phospho-STAT5 readouts via HTRF in GM-CSF stimulated TF-1 human erythroleukemia cells (JAK2 WT) and HEL 92.1.7 human erythroleukemia cells (homozygous for JAK2 V617F). The data generated across the lead series demonstrates potential best-in-class potency with JAK2 V617F cellular IC50s <100nM and >50-fold selectivity over JAK2 WT. The lead compound, CGT1145, has an IC50 of 91nM in the mechanistic JAK2 V617F cellular assay and is 89-fold selective over JAK2 WT. This compound shows high kinome selectivity with advanceable oral bioavailability across species. Target engagement in vivo was studied using athymic nude mice subcutaneously implanted with HEL 92.1.7 cells. At the 4-hour timepoint, CGT1145 showed robust dose dependent inhibition of tumor phospho-STAT5 via AlphaLISA at low dose levels. In a mouse model of human disease, athymic nude mice were implanted intravenously with Ba/F3-EPOR-JAK2V617F cells. Treatment with CGT1145 showed dose dependent inhibition of splenomegaly equivalent to ruxolitinib. Additionally, CGT1145 was well tolerated over the course of the study. Conclusions CGT1145 is a potential best-in-class JAK2 WT sparing, JAK2 V617F JH2 mutant selective inhibitor that has the potential to provide significantly higher target engagement which may lead to improvements in the ability to eradicate JAK2 V617F MPN-propagating cells and induce molecular remission while avoiding hematological tolerability issues. Further studies are ongoing to assess CGT1145, as well as other advanced close in analogs in our lead series.