A Clinical Prediction Model to Determine Probability of Response to Certolizumab Pegol for Crohn’s Disease

Pavine L.C. Lefevre, Parambir S. Dulai, Zhongya Wang, Leonardo Guizzetti, Brian G. Feagan, Anca Pop, Mohamed Yassine, Lisa M. Shackelton, Vipul Jairath, William J. Sandborn, Niels Vande Casteele*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Background: Certolizumab pegol (CZP) is effective for moderately to severely active Crohn’s disease (CD). Higher plasma concentrations are associated with better outcomes and increased drug clearance is the driver of subtherapeutic CZP concentrations. Objective: We aimed to develop a prediction model incorporating predicted CZP clearance and patient variables to allow estimation of the probability for remission prior to initiating therapy. Methods: A population pharmacokinetic model estimated baseline CZP clearance in patients with CD from nine phase II and III trials. Multivariable prediction models were developed and validated using the PRECiSE 1 and PRECiSE 2 datasets to identify candidate predictors for a composite remission outcome (Crohn’s Disease Activity Index ≤ 150 and fecal calprotectin concentration ≤ 250 μg/g) at Weeks 6 or 26. An online clinical decision support tool (CDST) was developed. Results: Baseline predicted CZP clearance ≥ 0.5 L/day was associated with subtherapeutic Week 6 CZP plasma concentrations. Baseline weight (odds ratio [OR] 1.04; 95% confidence interval [CI] 1.02–1.07), calculated CZP clearance (OR 0.92; 95% CI 0.87–0.96), hematocrit (OR 2.55; 95% CI 1.43–4.54), and fecal calprotectin (OR 0.66; 95% CI 0.54–0.80) were associated with Week 6 remission (p ≤ 0.0015 for all predictors). Baseline weight (OR 1.04; 95% CI 1.02–1.07), calculated CZP clearance (OR 0.93; 95% CI 0.88–0.97), and Patient-Reported Outcome-2 (PRO2) (OR 0.93; 95% CI 0.87–0.99) were associated with Week 26 remission (p ≤ 0.033 for all predictors). Conclusions: Patients who are predicted to have accelerated baseline CZP clearance are at risk of subtherapeutic CZP concentrations. Patient-level probabilities for a composite remission outcome can be predicted for patients with CD by entering commonly available patient- and disease-related factors into an online CDST (https://premedibd.com) incorporating predicted CZP clearance.

Original languageEnglish (US)
Pages (from-to)85-93
Number of pages9
Issue number1
StatePublished - Jan 2022
Externally publishedYes

ASJC Scopus subject areas

  • Biotechnology
  • Pharmacology
  • Pharmacology (medical)


Dive into the research topics of 'A Clinical Prediction Model to Determine Probability of Response to Certolizumab Pegol for Crohn’s Disease'. Together they form a unique fingerprint.

Cite this