A PHARMACOKINETIC POPULATION MODEL FOR CYCLOSPORIN A IN RENAL TRANSPLANT RECIPIENTS : Development of a model for whole blood- and intracellular concentrations

Abstract

ABSTRACT

Background: Cyclosporine A (CsA) has since its introduction in the 1980’s played a substantial part of the success in solid organ transplantation. Like many other immunosuppressive drugs, CsA has a narrow therapeutic window and a large inter-individual variability. Drug exposure above the therapeutic window is associated with adverse events like nephrotoxicity, infection and cancer while drug exposure below will yield a lack of effect and increased risk for acute rejection episodes. Obtaining an optimal drug concentration will prevent acute organ rejections and optimize survival of the grafts and ultimately the patients.

The primary aim of this study is to implement a T-cell compartment to an already existing whole blood model. Another goal was to further develop the basic whole blood model after the inclusion of 20 new patients followed for at least 8 weeks, by re-evaluating for relevant covariates and include estimation of interoccasional variation in the model.

Methods: Data was gathered from four separate clinical trials, performed by the department of Pharmaceutical Biosciences at the University of Oslo in co-operation with the Medical Department at Rikshospitalet, Oslo University Hospital, for the whole blood model. In all 70 patients and a total of 1276 whole blood samples were included in the whole blood model.

Of the 70 patients, 20 patients also had intracellular concentrations measured. These 430 intracellular samples were included in the development of the extended model.

By using the nonlinear mixed-effect modelling program NONMEM two population pharmacokinetic models were developed. Results: When re-analyzing for significant covariates, many similar results as earlier tested for the whole blood model was found. Age was a significant co-factor on the parameters: clearance (CL), absorption (Ka) and compartmental volumes (V1), while cytochrom P450 3A5 (CYP3A5) genotype had a significant impact on clearance. The steroid dose and weight influenced the inter-compartmental clearance (Q), while BMI had an effect on volume (V1) and absorption (Ka). Interoccasional variability was found significant on the parameter V2, and included in the final model.

An intracellular population pharmacokinetic kinetic (PPK) model was developed for CsA. The concentrations were homogenized to the same unit (ng/ml), by estimating the T-lymphocyte volume, and LN-transformed because of the large concentration range difference. The developed model predicts whole blood and intracellular concentrations, but does not predict accurately or stable enough in its current state.

Conclusion: Two models were developed, one for whole blood concentrations and one extended model also including intracellular concentrations of CsA. There is no unambiguous answer if the whole blood model gave a significant improvement on the already existing model, but the model showed somewhat improvement in the visual plots and also included prednisolone and CYP3A5 as a covariate. Interoccasional variability was found significant and further included for the whole blood model. The whole blood and intracellular model is still in an early stage and needs to be further developed, tested for covariates and interoccasional variability, and finally validated