Population pharmacokinetic model to predict steady-state exposure to once-daily cyclosporin microemulsion in renal transplant recipients.

backgroundThe microemulsion formulation of cyclosporin (CsA-ME) has a less variable absorption profile than the standard formulation (CsA-S), but only limited information is available about once-daily administration of CsA-ME.

objectiveTo develop a population pharmacokinetic model for once-daily CsA-ME that enables the prediction of individual steady-state area under the concentration-time curve (AUC) on the basis of blood concentration measurements and patient covariates. PATIENTS AND

methodsThe steady-state pharmacokinetics of once-daily cyclosporin were studied in 60 stable renal transplant recipients before and after conversion from CsA-S to CsA-ME. For each formulation, 7 blood samples were collected from 50 patients (group A) at sparse timepoints over 2 weeks, and 10 blood samples were collected from 10 patients (group B) at fixed timepoints over 24 hours. A 2-compartment population model assuming time-lagged first-order oral absorption was fitted to the data from group A, using nonlinear mixed effects modelling (NONMEM). The data from group B were used to evaluate the predictive performance of the model.

resultsMean [+/- SD; coefficient of variation (%CV)] CsA-S doses of 245mg (+/- 92) resulted in cyclosporin blood concentrations of 214 microg/L (+/- 70) after 12 hours and 108 microg/L (+/- 23) after 24 hours; the mean estimated AUC to 24 hours was 7658 microg x h/L (30%). With mean CsA-ME doses of 206mg (+/- 59), cyclosporin blood concentrations were 212 microg/L (+/- 33) and 132 microg/L (25%) after 12 and 24 hours, respectively, and the mean estimated AUC(24) was 9357 microg x h/L (23%). A strong correlation between 12-hour concentrations and AUC(24) was observed for CsA-ME (r = 0.95, p < 0.001), but not for CsA-S (r = 0.59, nonsignificant); the correlation between 24-hour trough concentrations and AUC(24) was weaker for both formulations (r = 0.64, p < 0.05 and r = 0.37, nonsignificant, respectively). On the basis of the population model derived from group A, the single best timepoint to predict AUC(24) from blood cyclosporin concentration was at 8 hours [AUC(24) (microg x h/L) = 19.6 x cyclosporin concentration at 8 hours (microg/L) + 3035], resulting in a prediction error of 8.3 +/- 6.6% when applied to the measured AUC(24) of group B. Adverse events were observed after conversion in 18 patients; these events generally resolved spontaneously or after dosage reduction, but twice-daily administration was required in some patients.

conclusionsSwitching from once-daily CsA-S to CsA-ME results in more consistent and predictable cyclosporin pharmacokinetics. Adjustment of dosage or regimen may be required in some patients.