Population Pharmacokinetics and Exposure-Response Analyses for Risankizumab in Patients with Active Psoriatic Arthritis

Abstract

Introduction: Risankizumab is an anti-IL23 monoclonal antibody approved for the treatment of moderate to severe plaque psoriasis and active psoriatic arthritis (PsA). This work characterizes the pharmacokinetics of risankizumab in PsA compared with psoriasis and evaluates the efficacy and safety exposure-response relationships in PsA.

Methods: The population pharmacokinetic analyses included data from 1527 participants that originated from one phase 1 healthy participant study, one phase 2 dose-ranging study in patients with PsA with an open-label extension study, and two pivotal phase 3 studies in patients with PsA, where the clinical regimen of risankizumab 150 mg administered subcutaneously (SC) at weeks 0, 4, and every 12 weeks thereafter was compared with placebo. Pharmacokinetics were analyzed using nonlinear mixed-effects modeling. Simulation analyses using the final model were conducted to evaluate the impact of covariates on exposure. Data from 1407 patients with PsA from the phase 3 studies were included in the exposure-response analyses. Graphical analyses were used to evaluate efficacy and safety exposure-response relationships, and logistic regression was conducted for further assessment of efficacy exposure-response relationships.

Results: Risankizumab pharmacokinetics were well described by a two-compartment model with first-order SC absorption and elimination. None of the evaluated covariates showed clinically relevant impact on exposure. On the basis of the final model, systemic clearance, steady-state volume of distribution, and terminal phase elimination half-life were estimated to be ~ 0.31 L/day, 11.1 L, and 26.3 days, respectively, for a typical 90 kg patient with PsA. Absolute SC bioavailability was estimated to be 83.5%. Exposure-response quartile analyses suggested that exposures associated with the clinical regimen maximized efficacy across the endpoints evaluated. No exposure dependency was observed for key safety endpoints.

Conclusions: Risankizumab exhibited linear and time-independent pharmacokinetics in patients with PsA and was comparable to patients with plaque psoriasis. Efficacy and safety exposure-response analyses support that the clinical regimen achieved robust efficacy with a favorable safety profile for patients with active PsA.

Clinical trials: NCT02596217, NCT02719171, NCT02986373, NCT03671148, and NCT03675308.

Clinical trials: NCT02596217, NCT02719171, NCT02986373, NCT03671148, and NCT03675308.

Keywords: ACR20/50/70; Efficacy; Minimal disease activity; PASI 90/100; Pharmacokinetics; Psoriatic arthritis; Risankizumab.

Fig. 1

Comparison of risankizumab pharmacokinetics in psoriasis and psoriatic arthritis to support the use of the psoriasis population pharmacokinetic model. Graphical comparisons of risankizumab exposure levels for psoriasis and PsA patient populations on the clinical regimen of 150 mg SC at weeks 0, 4, and q12w thereafter are shown for a week 28 predose observed risankizumab concentrations in patients with PsA and psoriasis, b simulated risankizumab concentrations as a function of time since first dose using psoriasis population pharmacokinetic model and observed concentration in patients with PsA in the phase 2 study, and c simulated and observed risankizumab predose concentrations at week 28 in the phase 3 studies. The shaded regions show the 95% CIs around median and 5th and 95th percentiles of simulated risankizumab concentrations. The gray dots denote the observed risankizumab concentration in the PsA studies, and the lines/dashed lines in (b) and the crosses in (c) show the median and 5th and 95th percentile of the observed risankizumab concentration in the PsA studies. PsA, psoriatic arthritis; SC, subcutaneous; q12w, every 12 weeks

Fig. 2

Simulated impact of statistically significant covariates identified in the population pharmacokinetic analyses on risankizumab exposures for patients with psoriatic arthritis. Points represent medians and error bars represent 95% confidence intervals of the normalized exposure ratios across 200 simulation replicates. The vertical black dashed line shows exposure ratio of 1 relative to the reference group, and the shaded area represents the 0.8–1.25 default equivalence boundaries. AUC, area under the concentration–time curve between weeks 16 and 28 (AUC_tau); C_trough, concentration after a dosing interval at week 28; hsCRP, high-sensitivity C-reactive protein

Fig. 3

Exposure–response relationships for efficacy endpoints at week 24 from phase 3 studies. Values on the x-axis represent range of the observed phase 3 risankizumab C_trough at week 28 for each quartile. Plots show %response (using nonresponder imputation) and n/N, where n represents number of responders and N represents total number of patients in each exposure-quartile bin. For ACR, PASI, and MDA analyses (at week 24), 52, 26, and 52 patients (N), respectively, from phase 3 studies were excluded from these analyses owing to missing C_trough values at week 28. Of these 52 patients for ACR, 28 (53.8%), 19 (36.5%), and 9 (17.3%) were ACR20 (a), ACR50 (b), and ACR70 (c) responders, respectively. Of the 26 patients for the PASI endpoint, 11 (42.3%) and 9 (34.6%) were PASI 90 (d) and PASI 100 (e) responders, respectively. Of these 52 patients for the MDA endpoint, 14 (26.9%) were MDA (f) responders at week 24. ACR20/50/70, at least 20%/50%/70% improvement in American College of Rheumatology response criteria; C_trough, concentration at the end of a dosing interval; PASI 90/100, at least 90%/100% improvement in Psoriasis Area and Severity Index relative to baseline; MDA, minimal disease activity

Fig. 4

Exposure–response quartile analyses for safety events of interest over weeks 0–24 in phase 3 studies. Values on the x-axis represent range of the observed phase 3 studies risankizumab C_trough at week 28 for each quartile. Plots show %response (using nonresponder imputation) and n/N, where n represents number of responders and N represents total number of patients in each exposure-quartile bin. In this analysis, 52 patients (N) from the phase 3 studies who had missing C_trough at week 28 were excluded. Of these 52 patients, 9 (17.3%), 2 (3.8%), 24 (46.2%), and 3 (5.8%) had a infections, b serious infections, c adverse events, and d serious adverse events, respectively, by week 24. C_trough, concentration at the end of a dosing interval