FRI0465 TOFACITINIB POPULATION PHARMACOKINETICS IN CHILDREN WITH JUVENILE IDIOPATHIC ARTHRITIS: A POOLED ANALYSIS OF DATA FROM THREE CLINICAL STUDIES

Abstract

Background: Tofacitinib is an oral JAK inhibitor that is being investigated for juvenile idiopathic arthritis (JIA). Objective(s): To describe tofacitinib pharmacokinetics (PK) in patients with JIA, identify potential covariates accounting for variability in exposure, assess the formulation effect of oral solution vs tablet and propose a simplified dosing regimen. Method(s): This was a pooled analysis of data from 3 tofacitinib clinical studies in patients with JIA aged 2-<18 years: a Phase 1, open-label (OL), non-randomised study (NCT01513902); a Phase 3, randomised, double-blind, placebo-controlled, withdrawal study (NCT02592434); and an OL long-term extension study (NCT01500551). Tofacitinib was dosed at 5 mg twice daily (BID) in patients >=40 kg or at body weight (BW)-based lower doses BID in patients <40 kg, to achieve average concentrations (Cavg) comparable with those in patients receiving 5 mg BID. A sparse PK sampling scheme was applied, and the plasma samples were assayed using a validated, sensitive and specific high-performance liquid chromatography tandem mass spectrometric method (lower limit of quantification = 0.100 ng/mL). A nonlinear mixed-effects modelling approach was used for the population PK model, and population parameter variability was assumed to be log-normally distributed. Covariates relating to patient demographics, disease characteristics, concomitant medications and formulation (oral solution vs tablet) were selected using a stepwise covariate modelling approach, and parameter-covariate relationships were evaluated using stepwise forward-inclusion (p<0.05) backward-deletion (p<0.001) procedures. The effect of time-varying BW on oral clearance (CL/F) and apparent volume of distribution (V/F) was characterised using an allometric model. Final model quality was assessed by Visual Predictive Checks (VPCs). Result(s): Of 246 patients in the analysis, 74.0% were female; 87.8% were white, 2% were black, 10.2% were 'other' races and no patients were Asian. Median (range) BW was 46.3 (11.1-121.8) kg. Initially, 100 patients received oral solution and 146 patients received tablets; 11 patients switched formulations during the studies. A one compartment disposition model with first-order absorption and a lag time sufficiently described the data. Final estimates for CL/F, V/F and the first-order absorption rate constant (ka) for tablets were 26.1 L/hr, 89.2 L and 2.78 hr-1, respectively. The only statistically significant covariate was a formulation effect on ka. All parameters were estimated adequately. Estimated allometric exponents were 0.310 for CL/F and 0.537 for V/F. Absorption was described with an estimated lag time of 0.186 hr, and the oral solution had a 1.64-fold faster absorption rate vs the tablet. VPCs sufficiently described the observed data over time, across BWs and ages. Given the PK characterisation and variability in patients with JIA, a simplified dosing scheme was proposed, targeting Cavg values equivalent to those in patients receiving 5 mg BID: 3.2 mg BID solution in patients 10-<20 kg; 4 mg BID solution in patients 20-<40 kg; and 5 mg BID tablet or solution in patients >=40 kg. Conclusion(s): Tofacitinib population PK in patients with JIA were adequately described by a one compartment model parameterised in terms of CL/F, V/F and first-order absorption with a lag time. Drug absorption from the oral solution was faster than from the tablet. Tofacitinib does not require dose modification or restrictions for any covariates, except BW, to account for differences in Cavg. Based on the results of this analysis, a simplified BW-based dosing regimen was proposed.