Maša Roganović (1), Marija Jovanović (1), Ana Homšek (1), Đorđe Kralj (2), Olga Odanović (2), Tamara Knežević Ivanovski (2), Petar Svorcan (2,3), Srđan Marković (2,3), Katarina Vučićević (1)
(1) University of Belgrade – Faculty of Pharmacy, Department of Pharmacokinetics and Clinical Pharmacy, Belgrade, Republic of Serbia, (2) Clinical Hospital Center “Zvezdara” – Department of Gastroenterology and Hepatology, Belgrade, Republic of Serbia, (3) School of Medicine - University of Belgrade, Belgrade, Republic of Serbia
Introduction: Adalimumab is a fully human monoclonal antibody that belongs to the group of tumor necrosis factor (TNF)-alpha inhibitors. It is approved for the treatment of patients with ulcerative colitis and Crohn’s disease and other immune-related diseases characterized by inflammation [1]. However, a significant number of patients either do not respond to initial therapy with adalimumab or experience a loss of clinical response over time, leading to inadequate disease control [2]. Therefore, proactive therapeutic drug monitoring (TDM) and model-informed precision dosing techniques are essential to prevent therapy failure [3,4].
Objectives: The aim of this study is to evaluate the predictive performance and reliability of available adalimumab pharmacometric models using selected evaluation techniques. This assessment aims to determine their suitability for potential application in the clinical setting.
Methods: Data for the evaluation of models was retrospectively collected from the medical records of adult inflammatory bowel disease (IBD) patients at the Clinical Hospital Center “Zvezdara”. The data included demographic characteristics (such as age and sex), adalimumab dosing information (every week or every other week), at least one trough concentration of adalimumab, level of anti-adalimumab antibodies (ADA) categorized as present or absent, albumin (ALB) levels, and other pertinent laboratory findings. Concentrations below limit of quantification (BLQ) were excluded from data analysis. PubMed was searched for full-text adalimumab pharmacokinetic population models in January 2024. Exclusion criteria included pediatric population, pregnancy and diseases other than IBD. Selected adalimumab pharmacokinetic population models in IBD patients were then transcribed into control-stream file format, and their predictive performance was externally evaluated through visual inspection of goodness-of-fit plots, including visual predictive checks (VPCs) and normalized prediction distribution errors (NPDEs). Mean prediction error (MPE) and root mean square prediction error (RMSPE) were calculated for individual predicted (IPRED) and population predicted concentrations (PRED). Analysis was performed using NONMEM® (Icon Development Solutions, version 7.3) and R software.
Results: Our PubMed search yielded a total of 9 models, with 3 developed in Monolix® and 6 in NONMEM®. After considering the covariates available in our dataset and the compatibility of published models with our data format, we selected the following models for external evaluation: Berends et al. [5], Marquez-Megias et al. [6], Ternant et al. [7], de Klaver et al. [8]. Our dataset for external model evaluation comprised 48 patients, with a mean age of 37.70 years (range: 21 – 61) including 28 females (58.33%). The majority of patients were diagnosed with Crohn’s disease (44,91.66%). A total of 190 concentrations were included in the evaluation. For the model developed by de Klaver et al., the calculated validation parameters in our validation group were as follows: MPE (95% CI) for PRED and IPRED, respectively, were -0.67 (-1.66 – 0.32), -0.32 (-0.55 – -0.099); RMSPE for PRED and IPRED, respectively, were 6.96 and 1.6. Evaluation of Berends et al., Marquez-Megias et al., and Ternant et al. models revealed bias in both PRED and IPRED predictions, although precision in prediction was observed as RMSPE values were also small for these models. The VPC and NPDE plots showed that Klaver et al. model performed better than the other three models.
Conclusion: Based on the analysis of goodness-of-fit plots, VPCs, NPDEs and calculated MPE and RMSPE, the model developed by de Klaver et al. demonstrates the most accurate prediction of PRED of adalimumab in our patient population. These findings underscore the need for additional population pharmacokinetic models of adalimumab. Additional models are essential to comprehensively identify all potential sources of variability and, consequently, enhance the accuracy of predictions.
Acknowledgement: This research was supported by the Science Fund of the Republic of Serbia, grant no. 6777, project: Improving Clinical Outcomes with Precision Dosing in Patients with Inflammatory Bowel Disease Through Investigating Variability of Monoclonal Antibodies Based on Population Pharmacokinetic-Pharmacodynamic Modeling – optYmAb.
[1] Summary Product Characteristics of Humira. Available on: https://www.ema.europa.eu/en/documents/product-information/humira-epar-product-information_en.pdf. Last accessed: 14.3.2024.
[2] Roda G, Jharap B, Neeraj N, Colombel JF. Loss of Response to Anti-TNFs: Definition, Epidemiology, and Management. Clin Transl Gastroenterol. 2016;7(1):e135.
[3] Hinojosa J, Muñoz F, Martínez-Romero GJ. Relationship between Serum Adalimumab Levels and Clinical Outcome in the Treatment of Inflammatory Bowel Disease. Dig Dis. 2019;37(6):444-450.
[4] Kantasiripitak W, Van Daele R, Gijsen M, Ferrante M, Spriet I, Dreesen E. Software Tools for Model-Informed Precision Dosing: How Well Do They Satisfy the Needs? Front Pharmacol. 2020;11:620.
[5] Berends SE, Strik AS, Van Selm JC, Löwenberg M, Ponsioen CY, DʼHaens GR, Mathôt RA. Explaining Interpatient Variability in Adalimumab Pharmacokinetics in Patients with Crohn’s Disease. Ther Drug Monit. 2018;40(2):202-211.
[6] Marquez-Megias S, Nalda-Molina R, Más-Serrano P, Ramon-Lopez A. Population Pharmacokinetic Model of Adalimumab Based on Prior Information Using Real World Data. Biomedicines. 2023;11(10):2822.
[7] Ternant D, Karmiris K, Vermeire S, Desvignes C, Azzopardi N, Bejan-Angoulvant T, van Assche G, Paintaud G. Pharmacokinetics of adalimumab in Crohn’s disease. Eur J Clin Pharmacol. 2015;71(9):1155-7.
[8] de Klaver PAG, Keizer RJ, Ter Heine R, Smits L, Boekema PJ, Kuntzel I, Schaap T, de Vries A, Bloem K, Rispens T, Hoentjen F, Derijks LJJ. Early At-Home Measurement of Adalimumab Concentrations to Guide Anti-TNF Precision Dosing: A Pilot Study. Eur J Drug Metab Pharmacokinet. 2023;48(4):377-385.
Reference: PAGE 32 (2024) Abstr 11175 [www.page-meeting.org/?abstract=11175]
Poster: Methodology - Model Evaluation