Modeling of The Relationship Between Ustekinumab Exposure, Fecal Calprotectin and Endoscopic Outcomes in Patients with Crohn’s Disease

Background: Ustekinumab is approved for the treatment of patients with moderate-to-severe Crohn’s disease (CD) and ulcerative colitis. In the UNITI endoscopy sub-study, only 17.4% of patients on ustekinumab achieved endoscopic response and 10.9% of patients achieved endoscopic remission at week 44 (w44).1

Aims: We aimed to investigate if improved endoscopic outcomes could be achieved through dose optimization based on a population pharmacokinetic-pharmacodynamic (popPK/PD) modeling and simulation analysis.

Methods: Data were obtained from 83 patients with moderate-to-severe CD (94% multi-refractory) enrolled in a prospective open-label cohort study.2 All received ustekinumab 6 mg/kg IV induction, followed by 90 mg SC every eight weeks (q8w). Ustekinumab serum concentrations were measured during induction (w4) and at trough (w8, w16, w24). Fecal calprotectin (fCal) was measured at baseline, and at w8, w16, w24. Endoscopic response (50% decrease in simple endoscopic score for CD [SES-CD]) and endoscopic remission (SES-CD ≤2) were assessed at w24. Modeling and simulation were performed using NONMEM 7.4.

Results: The PK of ustekinumab was well described by a 2-compartment model with first-order absorption and first-order elimination. Sequentially, an indirect response popPK/PD model was used to describe the inhibitory effect of ustekinumab exposure (Emax model) on the “generation rate” constant of fCal. A logistic regression model was used to predict the probability of achieving endoscopic remission/response at w24 based on observed fCal at w8 through an inhibitory Emax function. Volume of distribution (**1), and clearance (**0.75) were allometrically scaled. Ustekinumab clearance was found to increase with decreasing serum albumin. The terminal half-life of ustekinumab in a median patient (bodyweight 65 kg, serum albumin 42.7 g/L) was 21.8 days. fCal decreased with increasing ustekinumab exposure. The probability of endoscopic remission at w24 increased from 2.1% to 10.0% with fCal at w8 decreasing from 1,800 µg/g to 214 µg/g. The probability of endoscopic response at w24 increased from 10.0% to 17.9% with fCal at w8 decreasing from 1,800 µg/g to 694 µg/g. Simulation-based comparison of q8w and q4w maintenance dosing predicted median trough concentrations at steady state of 1.2 µg/mL and 4.8 µg/mL ustekinumab, and of 833 µg/g and 437 µg/g fCal, respectively. Associated endoscopic remission rates were estimated to be 4.1% on q8w dosing and 6.5% on q4w dosing (using fCal at steady state instead of w8). Associated endoscopic response rates were estimated to be 16.3% on q8w dosing and 21.9% on q4w dosing. Study sample sizes of 1,368 and 774 patients would be required to identify statistically significant differences in the rates of endoscopic remission (NNT=42) and endoscopic response (NNT=18), respectively, between both study arms (1:1 randomization, α=5%, power=80%).

Conclusion: The developed models can be used to guide clinical trial design and to support model-informed dose optimization (stratified or individualized dosing) to improve endoscopic outcomes. Although our analyses showed that q4w dosing resulted in lower fCal concentrations, the predicted increase in proportion of patients achieving endoscopic remission was limited.


  1. Rutgeerts P, Gasink C, Chan D, et al.Efficacy of Ustekinumab for Inducing Endoscopic Healing in Patients With Crohn’s Disease.Gastroenterology2018; 155: 1045–58.
  2. Verstockt B, Dreesen E, Noman M, et al.Ustekinumab Exposure-outcome Analysis in Crohn’s Disease Only in Part Explains Limited Endoscopic Remission Rates. J Crohns Colitis2019; 13: 864–72.