New Insights on Assessing Intra-Family Selectivity for Jak2 Inhibitors

Abstract 5150

Essential thrombocythemia (ET), polycythemia vera (PV) and myelofibrosis (MF) are myeloproliferative disorders (MPDs) characterized by a chronic over-production of cells of one or more blood cell lineages and/or bone marrow fibrosis which may, on occasion, progress to acute myeloid leukemia. The V617F gain of function mutation in the pseudokinase domain of Jak2, which results in constitutive activation of Jak2, is the most frequent mutation associated with MPD. Constitutively activated Jak2 can lead to dysregulated downstream signaling pathways (STAT, MAP kinase, and PI3 kinase) which in turn trigger abnormal growth, survival and differentiation of hematopoietic progenitors. Therefore, inhibition of constitutively activated Jak2 may offer therapeutic potential. Designing a Jak2^V617F specific inhibitor encounters challenges due to the lack of enzymatic activity of the pseudokinase domain of Jak2. In lieu of a Jak2^V617F mutant selective inhibitor, a highly selective inhibitor of Jak2 is likely an attainable goal.

Jak2 is a member of the Jak family of kinases including Jak1, Jak3, and Tyk2. Highly selective Jak2 inhibitors may provide a better safety margin in chronic dosing settings in ET and PV patients since inhibiting other Jak family members could cause side-effects such as immunosuppression. Attaining the desired selectivity of Jak2 inhibition versus the other family members has been challenging and few compounds have been reported to date that have the desired Jak2 selectivity. This can be attributed to the high homology of the ATP binding pocket among Jak family members, but is also hampered by a lack of assays capable of distinguishing the Jak-selectivity profile in a physiologically relevant setting.

We compared the potency and selectivity of compounds tested in a pSTAT5 AlphaScreen® assay panel consisting of isogenic Ba/F3 cell lines individually expressing translocated ETS leukemia (TEL) fusions of each Jak-family member (Ba/F3-TEL-Jak) with data from corresponding Jak enzyme assays. Here we report that the selectivity of inhibitor compounds illustrated in enzyme assays did not correlate with the selectivity profile in cell lines due to different shifts in potency for each family member between enzyme and cells (Figure 1). As a consequence the selectivity of compounds for Jak2 against Jak1 observed in enzyme assays may be reduced or reversed in cellular assays. On the other hand, Jak2 selectivity over Jak3 seen in the enzyme assays was conserved in the cellular assay. Thus, we propose that compounds that exhibit greater potency on Jak2 compared to Jak1 in the enzyme assays are needed and should be the main focus of medicinal chemistry efforts in order to attain Jak2 selectivity over Jak1 in a cellular context.

View largeDownload slide

View largeDownload slide

Close modal

We also compared the potency and selectivity of compounds in the isogenic Ba/F3-TEL-Jak cell lines with data obtained with cytokine stimulated peripheral blood mononuclear cells (PBMCs). The potency and selectivity of compounds in PBMCs are determined by measuring the inhibition of phosphorylation of STAT5 in TPO or GM-CSF stimulated platelets or monocytes (mediated by Jak2) and in IL-2 stimulated lymphocytes (mediated by Jak1 and Jak3). We found that potency correlated well between PBMCs and Ba/F3-TEL-Jak2 cells, and the rank order of compounds based on IC₅₀ values obtained with Ba/F3-TEL-Jak cell lines were conserved well in PBMCs; the compound selectivity profiles derived from the Ba/F3-TEL-Jak cell assays were predictive of Jak2 selectivity profiles obtained in the PBMC assays. Therefore, inclusion of Ba/F3-TEL-Jak pSTAT5 cellular assays may be useful for Jak family inhibitor development. Our results also suggest that relying solely on enzyme potency and selectivity data can be misleading, and that evaluating cellular selectivity in a biologically relevant context may provide a more meaningful understanding of selectivity and lead to the development of more selective Jak2 compounds.