Ana Homšek 1, Srđan Marković 2, Petar Svorcan 2, Marija Jovanović 1, Katarina Vučićević 1
1 Faculty of Pharmacy – University of Belgrade, Department of Pharmacokinetics and Clinical Pharmacy (Belgrade, Serbia), 2 University Hospital Medical Center "Zvezdara", Gastroenterology and Hepatology Department (Belgrade, Serbia)
Introduction: Patients with inflammatory bowel disease (IBD) often face challenges in maintaining stable disease without complications. Mucosal healing (MH), assessed endoscopically, remains the most reliable indicator of remission [1]. However, the endoscopic procedures are time and money consuming, as well as uncomfortable for patients. This highlights the need for easily accessible and routinely measurable biomarkers that reliably reflect MH. While faecal calprotectin is widely used as a non-invasive predictor of response, stool samples are not always available or provided by patients. Recent studies suggest that platelet counts may also serve as useful predictors of treatment response [2, 3].
Objectives: In this study, we aim to investigate how platelet levels intertwined with infliximab (IFX) pharmacokinetics influence the achievement of MH.
Methods: Data for this study were obtained retrospectively, from medical charts of IBD patients on IFX at the University Hospital Medical Center “Zvezdara”. Patients were included in the study if they had undergone endoscopic evaluation during treatment, up to 104 weeks. A total of 248 patients were included (male 58.5%, age at diagnosis 36.5 (17-76) years, baseline weight 73.5 (43-110) kg). IFX was administered as a two-hour infusion after appropriate premedication with doses calculated according to body weight (200-800 mg). The dosing regimen was individualized to the patient’s needs as standard (week 0, 2 and 6) or accelerated induction (every or every other week), followed by maintenance therapy (every 4, 6 or 8 weeks).
For each patient, individual pharmacokinetic parameters were obtained in a previously developed model using prior analysis, utilizing the population modelling approach using NONMEM 7.5 software (ICON Development Solutions Inc., Dublin, Ireland) [4]. Data were processed using R software by implementing different packages (dplyr, pROC, magrittr, ggplot2, patchwork, psych, tidyr) in RStudio (version 2025.09.1+401). Normality of the tested variables was checked with a Shapiro-Wilk test. To compare groups with and without MH either Student’s t-test or Mann-Whitney U test were performed. Univariate binary logistic regression was implemented to evaluate the diagnostic performance of variables. Multiple regression analysis was used to assess combined influence of the two variables on the probability of MH achievement.
Results: Out of 248 patients, 77 (31%) did not achieve MH during IFX treatment. Patients that achieved MH had lower median clearance (0.412 L/day vs 0.494 L/day, p < 0.05) indicating lower drug exposure in unstable cohort. Median platelet levels were higher in patients without MH achievement (307•10⁹/L vs 260•10⁹/L, p < 0.05) indicating higher inflammation burden which contributes to lower IFX concentrations and unstable disease status.
Univariate logistic regression suggested IFX clearance could be a good predictor of MH (AUCROC 70.8%, 95%CI: 63.75-77.87%), while platelet levels had slightly less promising results (AUCROC 66.56%, 95%CI: 56.48-76.64%). In multivariate analysis, patients were stratified by platelet levels: low <150•10⁹/L, normal 150-300•10⁹/L, high normal 300-400•10⁹/L and high >400•10⁹/L. Statistically significant difference was observed between subcategories (p = 0.0288). Patients with high platelet levels had lower MH probability (<60%) and higher drug clearance. Patients in low platelet level cohort had lower drug clearance and a high probability of MH achievement (>90%). Normal platelet level patients had various clearance values, and any probability to achieve MH, while the high normal ones had slightly higher clearance and slightly lower probability (<80%).
Conclusion: In conclusion, both IFX clearance and platelet levels are important determinants of MH in IBD patients. The achievement of MH is predicted well by IFX clearance. Platelet levels should be considered as a promising and routinely available biomarker together with IFX pharmacokinetics to predict response in IBD patients.
Acknowledgement: This research was supported by the Science Fund of the Republic of Serbia, through grant agreement No. 6777, Improving Clinical Outcomes with Precision Dosing in Patients with Inflammatory Bowel Disease Through Investigating Variability of Monoclonal Antibodies Based on Population Pharmacokinetic-Pharmacodynamic Modelling —optYmAb.
References:
[1] Vatn MH. Mucosal healing: impact on the natural course or therapeutic strategies. Digestive diseases. 2009 Nov 4;27(4):470-5.
[2] Liu X, Pan LX, Pei JX, Pu T, Wen HT, Zhao Y. Role of serological biomarkers in evaluating and predicting endoscopic activity in inflammatory bowel disease. World Journal of Gastroenterology. 2025 May 14;31(18):104206.
[3] Voudoukis E, Karmiris K, Koutroubakis IE. Multipotent role of platelets in inflammatory bowel diseases: a clinical approach. World Journal of Gastroenterology: WJG. 2014 Mar 28;20(12):3180.
[4] Owen J.S., Fiedler-Kelly J. Introduction to population pharmacokinetic/pharmacodynamic analysis with nonlinear mixed effects models. John Wiley & Sons; 2014.
Reference: PAGE 34 (2026) Abstr 12100 [www.page-meeting.org/?abstract=12100]
Poster: Drug/Disease Modelling - Other Topics