[Zurück]

C. Urach, G. Zauner, F. Miksch, W. Reinisch, A. Eser, F. Breitenecker:
"Simulating pharmacokinetics using compartment models to assess the best time for followup medication with infliximab";
Vortrag: Medical Decision Making, Oslo; 10.06.2012 - 12.06.2012.

Purpose: Treatment of inflammatory bowel disease with the intravenously administered biologic infliximab is highly efficacious, but expensive. Application is currently performed according to a standardized regimen irrespective of the individual pharmacokinetic profile of a patient. Ideally, timing of administration could be adapted to the actual serum concentration of the drug which should not fall below a certain threshold level. Thus, calculating the optimal time for the next infusion may further improve the outcome of the treatment.
Methods: A clearance model for infliximab already exists. We aimed to adapt and extend this model by use of a multi-dimensional optimization strategy and inclusion of further patient characteristics. The model consists of a central and a peripheral compartment. The fluid kinetics are determined by the amount of infliximab in the two compartments, infusion strategy, exchange transports and elimination process. The flows are weighted by additional parameters like weight, sex, baseline albumin and ATI status of the patient. The applicability of the standard model was analysed for 100 patients with Crohn´s disease. Afterwards, we performed parameter identification to optimize the model for a specific test sample. Due to the lack of enough measure points the corresponding equations for parameter identification for single patients are underdetermined, that is why multi-dimensional optimization algorithms have to be used.
Results: Optimizing the model with data from many patients leads to a significantly reduction of variance of calculated to actual concentration of the substance in the bloodstream without increasing overestimation of the real concentration, which is good provided that declining under a threshold makes the treatment less likely efficacious.
Conclusion: Dynamic compartment models can be used to simulate blood kinetics in humans. Simulating elimination of infliximab in human blood helps deciding when the next dose should be administered and therefore may lead to more beneficial outcomes both safety- and efficacy-wise.