Development of PBPK model for intra-articular injection in human: methotrexate solution and rheumatoid arthritis case study

Authors: Le Merdy M, Mullin JM, Lukacova V
Publication: J Pharmacokinet Pharmacodyn
Keywords: intra-articular, Methotrexate, PBPK, product development, rheumatoid arthritis
Software: GastroPlus®
Division: Simulations Plus

Abstract

A physiologically based model describing the dissolution, diffusion, and transfer of drug from the intra-articular (IA) space to the plasma, was developed for GastroPlus v9.8. The model is subdivided into compartments representing the synovial fluid, synovium, and cartilage. The synovium is broken up into two sublayers. The intimal layer acts as a diffusion barrier between the synovial fluid and the subintimal layer. The subintimal layer of the synovium has fenestrated capillaries that allow the free drug to be transported into systemic circulation. The articular cartilage is broken up into 10 diffusion sublayers as it is much thicker than the synovium. The cartilage acts as a depot tissue for the drug to diffuse into from synovial fluid. At later times, the drug will diffuse from the cartilage back into synovial fluid once a portion of the dose enters systemic circulation. In this study, a listing of all relevant details and equations for the model is presented. Methotrexate was chosen as a case study to show the application and utility of the model, based on the availability of intravenous (IV), oral (PO) and IA administration data in patients presenting rheumatoid arthritis (RA) symptoms. Systemic disposition of methotrexate in RA patients was described by compartmental pharmacokinetic (PK) model with PK parameters extracted using the PKPlusTM module in GastroPlus. The systemic PK parameters were validated by simulating PO administration of methotrexate before being used for simulation of IA administration. For methotrexate, the concentrations of drug in the synovial fluid and plasma were well described after adjustments of physiological parameters to account for RA disease state, and with certain assumptions about binding and diffusion. The results indicate that the model can correctly describe PK profiles resulting from administration in the IA space, however, additional cases studies will be required to evaluate ability of the model to scale between species and/or doses.

By Maxime Le Merdy, Jim Mullin, Viera Lukacova