Comparison of the Enzymatic and Cellular Profiles of Clinical JAK2 Inhibitors for the Treatment of Myelofibrosis

Introduction: Three Janus kinase 2 (JAK2) inhibitors (JAKinibs) have been approved by the FDA for the treatment of myelofibrosis (MF): ruxolitinib, fedratinib, and pacritinib; a fourth one, momelotinib, is currently under FDA review. In addition to JAK2, these drugs also inhibit other kinases with distinct profiles with potential implications in their observed pharmacology. Published data on kinase inhibitory profiles of the four JAKinibs are limited and were determined under differing assay conditions (Wernig G et al. Cancer Cell 2008; Pardanani A et al. Leukemia 2009; Quintás-Cardama A et al. Blood 2010; Singer J et al. J Exp Pharm 2016), which limits head-to-head comparisons and understanding of their JAK2 inhibition unrelated effects. Here, we present full kinome profiling data for these four JAKinibs. Additionally, we have compared their effects on signaling and proliferation in myeloid cells and other human cells.

Methods and Results: Enzymatic assays performed in the presence of 1 mM ATP showed that ruxolitinib is the most potent JAK2 inhibitor (IC₅₀ 4 nM) followed by fedratinib (14 nM), momelotinib (51 nM), and pacritinib (53 nM). Inhibition of other JAK family proteins such as JAK1 and TYK2 were also observed to varying degrees with each drug, and these enzymatic activities were consistent with inhibition of STAT signaling in response to different cytokines in secreted embryonic alkaline phosphatase (SEAP) cellular reporter assays (HEK-Blue Detection, InvivoGen, San Diego, CA).

Full kinome analyses (372 activity assays of 356 unique wild-type protein kinases) were conducted (Reaction Biology, Malvern, PA) with determination of percentage inhibition at 100 nM of each drug with 10 μM ATP. Kinases exhibiting >50% inhibition with any of the drugs were then evaluated for dose-response to each drug with 100 μM ATP. The results shown in the Figure demonstrate that fedratinib, momelotinib, and pacritinib inhibited additional kinases outside of the JAK family with similar potency to JAK2 inhibition, whereas ruxolitinib was relatively selective for JAK2 inhibition. In contrast to published reports on the inhibitory activities of pacritinib and momelotinib against ALK2 (ACVR1), the inhibition of ALK2 observed here was 5-10 times weaker than JAK2 inhibition. Furthermore, pacritinib and momelotinib were both weaker inhibitors of ALK2 activity when compared with their potency towards more than 30 other kinases screened. Fedratinib, pacritinib, and momelotinib, but not ruxolitinib, exhibited high inhibitory potencies against FLT3 and/or KIT kinases, both of which are critical for the self-renewal and differentiation of normal hematopoietic stem cells.

In cellular assays monitoring JAK2 signaling (pSTAT5 level in SET2 cells), ruxolitinib was the most potent agent (IC₅₀ 14 nM), followed by momelotinib (205 nM), pacritinib (429 nM), and fedratinib (672 nM). Similar trends were observed in other cellular assays assessing antiproliferative effects in JAK2-dependent myeloid cell lines.

In cells not dependent on JAK pathway signaling for growth in culture, such as primary human fibroblasts and endothelial cells, ruxolitinib showed weak or no inhibition at 5 μM, while fedratinib, pacritinib and momelotinib showed anti-proliferative effects with low micromolar IC₅₀s. Similar trends were observed in human HEK293 and HepG2 cell lines, which are commonly used to assess drug effects on general cell health.

We analyzed the effects of the JAKinibs on the differentiation of CD34⁺ hematopoietic stem and progenitor cells (HSPCs) derived from healthy donors at concentrations observed in the plasma of MF patients enrolled in clinical trials. Consistent with their activity on multiple kinases, including FLT3 and KIT, fedratinib and pacritinib elicited the most pronounced decrease in overall viability, affecting HSPCs and megakaryocytes. In contrast, ruxolitinib as a more selective JAK2 inhibitor, showed the least effect on the differentiation of normal CD34⁺ HSPCs.

Conclusion: JAKinibs clinically evaluated for treatment of MF exhibit distinct kinase inhibition profiles and cellular activities. Amongst the agents tested in this study, ruxolitinib was the most potent and selective JAK2 inhibitor. At concentrations several fold above clinically relevant concentrations, ruxolitinib had no observable effects on the health of cells not reliant on JAK/STAT mediated signaling.