Our aim is to build a physiologically based pharmacokinetic and JAK2 occupancy model (PBPK-JO) to simultaneously predict pharmacokinetic (PK) and pharmacodynamic (PD) changes of baricitinib (BAR) in healthy human when co-administration with kidney transporters OAT3 and MATE2-K inhibitors, and in patients with hepatic and renal impairment. Probenecid and vandetanib were selected as OAT3 and MATE2-K competitive inhibitors, respectively. Here, we have successfully simulated PK and JAK2 occupancy profiles in human by PBPK-JO model. Moreover, this modelling reproduced every observed PK data, and every mean relative deviation (MRD) were below 2. The simulation demonstrated that oral dose of BRA should be reduced to half when co-administration with probenecid. The prediction suggested also vandetanib was unlikely to affect PK and PD of BAR. In simulations of hepatic and renal impairment patients, the predictions suggested that significant changes in PK and PD of BAR occurred. However, there was a lower fold increase in JAK2 occupancies than in PK in patients relative to healthy individuals. In other words, administration dose adjustment of BAR in patients with hepatic or renal impairment should combine PK and PD changes of BAR, instead off only considering PK alteration.

Background and Purpose: Excellent drug therapy is usually characterized by longer duration with sufficiently high occupancy on desired target. Except for classical pharmacokinetics (PK) contribution, binding kinetics (BK) and target kinetics (TK) are emerging as key role. This study aims to establish mathematical simulation model of kinase target engagement by integrating three intrinsic kinetic processes (i.e., PK, BK and kinase target kinetics, KTK) of interaction between kinase inhibitors (KIs) and kinase protein. Experimental Approach: Five half-life ratios, innovative pharmacokinetics-inhibition-kinase kinetics (PIKK) and diagnostic line plots (DLPs) were generated to assess which parameters were playing crucial role in the kinases target occupancy (KTO) decline. PK-BK-TKT model was first developed by a series of differential equations, and using this model simulated time courses of protein kinase occupancy. And the temporal kinase selectivity (TKS) method was firstly developed to judge inhibitor selectivity. Key Results: By three visualization approaches, key rate-limiting factors were first identified in kinase occupancy. We first simulated time course profiles of interaction between 55 approved inhibitors with 55 clinically validated protein kinases. Subsequently, the TKS between 30 inhibitors and 55 protein kinases were generated, and 6 inhibitors were identified as a highly selective chemical probe for their respective primary target. Conclusion and Implications: The mathematic model can describe the entire process of inhibitor-kinase interaction in-depth and shows extensive applications in many aspects.