A Model-Based Tool for Guiding Infliximab Induction Dosing to Maximize Long-term Deep Remission in Children with Inflammatory Bowel Diseases.

Background And Aims: Adequate infliximab concentrations during induction treatment are predictive for deep remission [corticosteroid-free clinical and endoscopic remission] at 6 months in children with inflammatory bowel diseases [IBD]. Under standard infliximab induction dosing, children often have low infliximab trough concentrations. Model-informed precision dosing [MIPD; i.e. model-based therapeutic drug monitoring] is advocated as a promising infliximab dosing strategy. We aimed to develop and validate an MIPD framework for guiding paediatric infliximab induction treatment.

Methods: Data from 31 children with IBD [4-18 years] receiving standard infliximab induction dosing (5 mg/kg at week [w]0, w2 and w6) were repurposed. Eight paediatric population pharmacokinetic models were evaluated. Modelling and simulation were used to identify exposure targets, identify an optimal sampling strategy, and develop a multi-model prediction algorithm for implementation into an MIPD software tool. A role for infliximab clearance monitoring was evaluated.

Results: A 7.5 mg/L infliximab concentration target at w12 was associated with 64% probability of deep remission at 6 months. With standard dosing, less than 80% of simulated children <40 kg attained this target. The w12 target was most accurately and precisely achieved by implementing MIPD at w6 using the w6 infliximab concentration [rapid assay required]. The multi-model algorithm outperformed single models when optimizing the w6 dose based on both w2 and w4 concentrations. MIPD using only the w2 concentration resulted in biased and imprecise predictions. Infliximab clearances at w6 and w12 were predictive for deep remission.

Conclusions: A freely available, multi-model MIPD tool facilitates infliximab induction dosing and improves deep remission rates in children with IBD.