2019 - Stockholm - Sweden

PAGE 2019: Drug/Disease modelling - Other topics
Mats Magnusson

Population PK and PASI exposure-response modelling for Certolizumab pegol in patients with chronic plaque psoriasis

Celine Sarr (1), Mats O. Magnusson (1), Pavan Vajjah (2), Miren Zamacona (2)

(1) Pharmetheus, Sweden (2) UCB Pharma, Slough, UK

Objectives: The overall objective of this analysis was to develop a population pharmacokinetic (PK) model to describe the PK characteristics of certolizumab pegol (CZP) in subjects with moderate to severe chronic plaque psoriasis (PSO) and to describe the exposure-response relationship between CZP and the PSO efficacy variable Psoriasis Area and Severity Index (PASI)[1].

Methods: The data for the model development originated from plasma concentrations of CZP and PASI observations from three Phase 3 clinical studies combining data from 834 subjects with PSO. In the three Phase 3 trials, subjects were randomized at baseline to treatment with either CZP 200 mg dosing every 2 weeks (Q2W) (loading dose of CZP 400 mg at Weeks 0, 2 and 4), CZP 400 mg Q2W or placebo (PBO). CZP was dosed subcutaneously. Subjects with an inadequate PASI response (PASI50 or PASI75 non-responders) at Week 16 were transitioned to different CZP treatments. Data up to Week 48 were included in the population PK analysis and up to Week 16 for the exposure-response analysis. The starting point for the population PK model was a population PK model developed for the rheumatoid arthritis population. PASI was treated as a continuous variable. Since there is both an upper and a lower bound of the PASI scale a logit-transformation was used. The development of a structural model for PASI was guided by a graphical analysis. The exposure-response model included a component to describe the PASI baseline observations, a placebo effect component, a drug effect component, and a component describing the time-course of PASI response. The covariates tested included demographics, anti-CZP antibodies, region, disease duration and prior biologic therapy. Nonlinear mixed effects modelling was conducted with NONMEM v 7.3.0.

PK characteristics of CZP in patients with PSO were well described by a one compartment model with first-order absorption and a first order elimination from the central compartment. No deviation form dose proportionality could be identified. Consistent with CZP PK knowledge, in PSO subjects both apparent clearance and apparent volume of distribution increase with body weight with heavier subjects achieving lower CZP plasma concentration. The presence of anti-CZP antibodies increases the apparent clearance.

The relationship between CZP exposure and PASI was described with an indirect response model where a placebo effect could result in both an increase or a decrease in PASI, and where a sigmoidal maximum effect model inhibiting the production of the response described the relationship between the individually predicted CZP plasma concentrations at the time of the pharmacodynamic observations and PASI response. The exposure-response model was parametrized to estimate an EC90 (CZP concentration resulting in 90% inhibition of the PASI production rate) as the two dose regimens tested were at the top of the dose-exposure-response curve. The Emax was estimated to be almost 100% and the typical value for EC90 was 11.1 µg/mL. The typical value for PASI t1/2 was 22.5 days leading to a time to maximum response of 16 weeks in the typical subjects, which matches with the clinical observed data. The covariate analysis revealed that body weight was a significant covariate of PASI half-life with heavier subjects taking longer to achieve maximum response. There were other statistically significant covariates such as region on PASI half-life or baseline PASI, and baseline PASI on maximum effect, but none were deemed to be clinically relevant.

The proposed models were able to characterize the PK and the exposure response of CZP versus PASI in subjects with moderate to severe chronic plaque PSO. The two tested regimens were at the top of the dose-concentration-response curve.

[1] C. Hu, Z. Yao, Y. Chen , B. Randazzo, L. Zhang, Z. Xu, A. Sharma, H. Zhou, A comprehensive evaluation of exposure-response relationships in clinical trials: application to support guselkumab dose selection for patients with psoriasis, J Pharmacokinet Pharmacodyn. 2018 Aug;45(4):523-535. 

Reference: PAGE 28 (2019) Abstr 9118 [www.page-meeting.org/?abstract=9118]
Poster: Drug/Disease modelling - Other topics
Click to open PDF poster/presentation (click to open)