Orwa Albitar, Sabariah Noor Harun, Dzul Azri Mohamed Noor, Siti Maisharah Sheikh Ghadzi.
School of Pharmaceutical Sciences, Universiti Sains Malaysia, 11800 Pulau Pinang, Malaysia.
Objectives: Cyclosporine is a calcineurin inhibitor that is still one of the main drugs of choice in renal transplant immunosuppression therapy.1 It is characterized by large pharmacokinetic variability, a narrow therapeutic window, and numerous covariates affecting its pharmacokinetic parameters.2 Despite its importance, a population pharmacokinetic model of cyclosporine hasn’t been developed yet in the Malaysian population. This study aimed to develop a population pharmacokinetic model of cyclosporine in Malaysian renal transplant recipients.
Methods: A total of 2807 concentration points at pre-dose and two hours after dose were collected retrospectively from 113 renal transplant patients on cyclosporine. The model structure and pharmacokinetic parameters were estimated using non-linear mixed-effect modeling NONMEM software.3 A combined proportional and additive model was used to describe the residual error. Missing doses were estimated using the compartment initialization method.4 Multiple imputations were used to handle missing covariates in order to perform covariate analysis using a stepwise forward addition (P < 0.05) followed by a backward elimination (P < 0.001). The final model was chosen based on the lowest objective function value, prediction corrected visual predictive check (pcVPC), relative standard errors (RSE) produced from the sampling-importance resampling (SIR) method,5 and scientific plausibility.
Results: The majority of patients were males 61.1%, Chinese 87.6% followed by Malay 7.1%, and Indians 5.3% with a mean (range) age of 44 years (17-77), and weight of 63 Kg (32-106). One-compartment model with first-order absorption and elimination was the best to fit the data. The value (RSE) of cyclosporine clearance (CL/F) was estimated as 15.2 L/h (6%), the volume of distribution (V/F) was 107 L (4%), and the rate of absorption (Ka) was 1.28 h-1 (fixed to the published value). Post-operative days, sex, creatinine clearance, and calcium channels blockers were identified as significant covariates on CL/F, while sex and cholesterol levels were identified as significant covariates on V/F. The model’s estimated inter-individual variability was 20.9% and 23.6% for CL/F and V/F, respectively, while additive and proportional error were estimated to be 0.39 ng/ml and 11%, respectively. Based on the pcVPC, the 95% confidence interval of the median, 10th and 90th percentiles of the simulated data have overlaid the median, 10th and 90th percentile of the observed data, implying the good predictive performance of the final model and all the models’ parameters have shown an RSE of not more than 30% based on the SIR results.
Conclusions: This is the first study involving Malaysian renal transplant recipients using a large number of drug monitoring data in a single-center and advanced methods to handle missing data that allows the development of a population pharmacokinetic model with the quantification of the covariate effects. The model developed in the current study provides useful information on the factors affecting cyclosporine pharmacokinetic to be considered when making the appropriate dosing adjustment in the future.
References:
[1] Kidney Disease: Improving Global Outcomes (KDIGO) Transplant Work Group. Special Issue: KDIGO Clinical Practice Guideline for the Care of Kidney Transplant Recipients. Am J Transplant. 2009;9:S1–155.
[2] Han K, Pillai VC, Venkataramanan R. Population Pharmacokinetics of Cyclosporine in Transplant Recipients. AAPS J. 2013;15:901–12.
[3] Beal, S. L., Sheiner, L. B. & Boeckman, A. J. NONMEM Users Guides. (1989–2006) (Icon Development Solutions, Ellicott City, MD, 2008).
[4] Wang Y, Liu X. Handling Missing Dosing History in Population Pharmacokinetic Modeling: An Extension to MDM Method. CPT Pharmacometrics Syst Pharmacol 2019;8(1):39-49.
[5] Dosne AG, Bergstrand M, Harling K, Karlsson MO. Improving the estimation of parameter uncertainty distributions in nonlinear mixed effects models using sampling importance resampling. Journal of pharmacokinetics and pharmacodynamics. 2016;43(6):583-96.
Reference: PAGE () Abstr 9348 [www.page-meeting.org/?abstract=9348]
Poster: Drug/Disease Modelling - Other Topics