Target-mediated drug disposition model of ustekinumab in patients with ulcerative colitis
Armando Tratenšek1, Jurij Aguiar Zdovc2, Iztok Grabnar1, Igor Locatelli1, David Drobne3,4, Tomaz Vovk1
1University of Ljubljana, Faculty of Pharmacy, 2Pharmetheus AB, 3University Medical Centre Ljubljana, Department of Gastroenterology, 4University of Ljubljana, Medical Faculty
Objectives: Ustekinumab (UST) is a fully human IgG1 monoclonal antibody that specifically binds to the common p40 subunit of interleukin-12 (IL-12) and interleukin-23 (IL-23), both of which are involved in inflammatory and immune responses [1]. It is approved for the treatment of inflammatory bowel disease, with indications for both moderately to severely active Crohn’s disease (since 2016) [1], [2] and, more recently, ulcerative colitis (since 2019) [1], [3]. Standard weight-based intravenous induction dose followed by subcutaneous injections of 90 mg every 8 weeks has been shown to be insufficient to achieve endoscopic remission for many patients with inflammatory bowel disease [4], [5], [6]. Therefore, population pharmacokinetic-pharmacodynamic (PK-PD) models for UST have been developed for Crohn’s disease to support model-informed dose optimization and improve endoscopic outcomes [7], [8]. Target-mediated drug disposition models (TMDD) are commonly used to describe the nonlinear PK of therapeutic antibodies, as they account for the interactions between the drug, its target antigen, and the resulting immune complex [9]. Aguiar Zdovc et al. have previously developed a two compartment quasi-equilibrium approximation of the TMDD model for UST in Crohn’s disease patients that incorporates two-compartment target kinetics [8]. However, PK studies of UST in ulcerative colitis remain limited. The aim of this study was to investigate the PK of ustekinumab in patients with ulcerative colitis, focusing on identification and characterisation of nonlinear kinetics. Methods: Patients with ulcerative colitis aged 18 years or older who initiated treatment with UST at a single tertiary referral centre between June 2020 and February 2024 were included. Patients received a weight-based intravenous induction dose (=55 kg: 260 mg; 55–85 kg: 390 mg; >85 kg: 520 mg) infused over one hour at week 0 (baseline), followed by a fixed maintenance dose of 90 mg, administered subcutaneously every 8 weeks (standard of care). Serum UST concentration was prospectively measured at baseline, 1 hour after intravenous infusion and subsequently at weeks 2, 4, 8, 8+1 day, 9, 10, 12, 16, 20, and 24. Demographic characteristics, clinical data, quality of life and biochemical markers were recorded. The population PK model was developed using NONMEM software. Structural models with one- and two compartments were evaluated followed by the inclusion of various nonlinear elimination models (Models 1-6). First-order absorption was assumed after subcutaneous administration. The following models were fitted to the UST serum concentrations: time-dependent elimination rate approximated by an exponential decrease in clearance (Model 1); concentration-dependent (Michaelis-Menten type) elimination (Model 2); irreversible binding of UST to the latent target without describing target turnover (Model 3); target-mediated elimination with irreversible binding of UST to the latent target and zero-order antigen input (Model 4), as described by Ternant et al. [10]; quasi-equilibrium approximation of the full TMDD model as described by Gibiansky et al. (Model 5) [9], [11]. In addition, a peripheral compartment for the target distribution was added to Model 5, as described by Aguiar Zdovc et al. (Model 6) [8]. The comparison between the models was based on the objective function value (OFV) and the Akaike information criterion (AIC). Results: Data were available from 35 patients, including 337 measurements of serum UST concentrations. The data were best described by a two-compartment model (OFV 1626.43, AIC 1642.34). Introduction of IIV on CL and V1 and F1 improved both OFV and AIC (1187.201 and 1209.201, respectively). In addition, nonlinear Models 1-6 with IIV on CL, V1 and F1 improved model performance as follows: Model 1 (OFV 1176.63, AIC 1202.63), Model 2 (OFV 1182.43, AIC 1208.43), Model 3 (OFV 1187.357, AIC 1213.357), Model 4 (OFV 1187.188, AIC 1215.188), Model 5 (OFV 1178.611, AIC 1206.517), and Model 6 (OFV 1160.897, AIC 1196.897). Based on OFV, AIC and goodness-of-fit plots, PK was best described by a quasi-equilibrium approximation of the TMDD model extended by the distribution of unbound UST and unbound target into peripheral compartments (Model 6). IIV was introduced on CL, V1 and F1. A combination of additive and proportional error models was used for the residual variability. Conclusion: We demonstrated that the PK of UST in patients with ulcerative colitis is nonlinear. A two-compartment TMDD model using the quasi-equilibrium approximation was further extended with a peripheral target compartment to best describe the concentration-time profile of UST.
[1] “STELARA-pi.pdf.” Accessed: Mar. 10, 2025. [Online]. Available: https://www.janssenlabels.com/package-insert/product-monograph/prescribing-information/STELARA-pi.pdf [2] B. G. Feagan et al., “Ustekinumab as Induction and Maintenance Therapy for Crohn’s Disease,” N. Engl. J. Med., vol. 375, no. 20, pp. 1946–1960, Nov. 2016, doi: 10.1056/NEJMoa1602773. [3] B. E. Sands et al., “Ustekinumab as Induction and Maintenance Therapy for Ulcerative Colitis,” N. Engl. J. Med., vol. 381, no. 13, pp. 1201–1214, Sep. 2019, doi: 10.1056/NEJMoa1900750. [4] J. Hanžel et al., “Ustekinumab concentrations shortly after escalation to monthly dosing may identify endoscopic remission in refractory Crohn’s disease,” Eur. J. Gastroenterol. Hepatol., vol. 33, no. 1S Suppl 1, pp. e831–e836, Dec. 2021, doi: 10.1097/MEG.0000000000002275. [5] B. Verstockt et al., “Ustekinumab Exposure-outcome Analysis in Crohn’s Disease Only in Part Explains Limited Endoscopic Remission Rates,” J. Crohns Colitis, vol. 13, no. 7, pp. 864–872, Jul. 2019, doi: 10.1093/ecco-jcc/jjz008. [6] V. Solitano et al., “Effectiveness of Ustekinumab for Patients With Moderate-to-Severe Ulcerative Colitis: A Multicenter Real-World Canadian Study,” Am. J. Gastroenterol., Nov. 2024, doi: 10.14309/ajg.0000000000003212. [7] Z. Wang et al., “Population pharmacokinetic-pharmacodynamic model-based exploration of alternative ustekinumab dosage regimens for patients with Crohn’s disease,” Br. J. Clin. Pharmacol., vol. 88, no. 1, pp. 323–335, 2022, doi: 10.1111/bcp.14971. [8] J. Aguiar Zdovc et al., “Ustekinumab Dosing Individualization in Crohn’s Disease Guided by a Population Pharmacokinetic-Pharmacodynamic Model,” Pharmaceutics, vol. 13, no. 10, p. 1587, Sep. 2021, doi: 10.3390/pharmaceutics13101587. [9] L. Gibiansky and E. Gibiansky, “Target-mediated drug disposition model: approximations, identifiability of model parameters and applications to the population pharmacokinetic-pharmacodynamic modeling of biologics,” Expert Opin. Drug Metab. Toxicol., vol. 5, no. 7, pp. 803–812, Jul. 2009, doi: 10.1517/17425250902992901. [10] D. Ternant et al., “Nonlinear pharmacokinetics of rituximab in non-Hodgkin lymphomas: A pilot study,” Br. J. Clin. Pharmacol., vol. 85, no. 9, pp. 2002–2010, Sep. 2019, doi: 10.1111/bcp.13991. [11] L. Gibiansky, E. Gibiansky, T. Kakkar, and P. Ma, “Approximations of the target-mediated drug disposition model and identifiability of model parameters,” J. Pharmacokinet. Pharmacodyn., vol. 35, no. 5, pp. 573–591, Oct. 2008, doi: 10.1007/s10928-008-9102-8.
