Background/Purpose: A number of Janus kinase (JAK) inhibitors are being actively investigated for treatment of rheumatoid arthritis (RA), including tofacitinib, baricitinib, filgotinib (GLPG0634), and decernotinib (VX-509). However, it is unclear how these drugs may differentiate from each other in the clinic based on their profiles of JAK-dependent cytokine inhibition. The aim of this work was to provide an integrated modelling approach using knowledge of both in-vitro whole cell JAK inhibition potencies and plasma pharmacokinetics to better understand profiles of cytokine inhibition for clinical JAK inhibitors in the context of clinically-meaningful doses.

Methods: IC50 values for IFNα, IFNγ, IL-6, IL-15, IL-21, IL-10, IL-27, IL-12, IL-23 and erythropoietin (EPO) signaling of tofacitinib, baricitinib, filgotinib, and decernotinib were measured in total lymphocytes, CD34+ cells (EPO) and CD3+ cells (IL-6) in human whole blood by a flow cytometry-based assay, quantifying the phosphorylation state of various STAT proteins. Human daily average plasma concentrations (C_av) were used as reported or predicted for tofacitinib (5 mg BID; 68 nM), baricitinib (4 mg QD; 32 nM), and filgotinib (200 mg QD; 527 nM). Confidence in the prediction of decernotinib pharmacokinetics was considered low because of high in-vitro metabolic instability and was not assessed further. Percent levels of cytokine inhibition (IC_xx=100*C_av/(IC₅₀ + C_av)) were determined at clinically-meaningful doses.

Results: Each JAK inhibitor showed a relatively similar profile of cytokine inhibition versus type I and II interferons (IFNα, IFNγ), the common γ-chain cytokines (IL-15, IL-21), and IL-6 and IL-27 (Figure 1). Each also showed some decrease in potency for IL-10, IL-12 and 23, and EPO. Comparing between JAK inhibitors, tofacitinib and baricitinib were overall more potent inhibitors than decernotinib and filgotinib. Clinical pharmacokinetics of tofacitinib, baricitinib, and filgotinib were available and used to further compare predicted cytokine profiles of inhibition in patients with RA. The profile of cytokine inhibition for each JAK inhibitor was in general similar at clinically-meaningful doses (Figure 1). While the pharmacokinetics were unavailable for decernotinib, the clinical dose ranges being explored are consistent with filgotinib which showed similar in-vitro inhibitory potencies.

Conclusion: These analyses illustrate the importance of studying a broad range of JAK pairing potencies and clinical concentrations when comparing JAK-inhibitor compounds. Calculated profiles of cytokine inhibition for a number of JAK inhibitors in RA are in general similar when efficacious doses are considered, suggesting limited differentiation of these JAK inhibitors based on JAK pharmacology. Ultimately, only robust clinical testing will determine whether there are clinical differences between JAK inhibitors.