Predicting patients' exposure to cyclosporin

Abstract

The introduction of a new formulation of cyclosporin, Neoral, has reduced pharmacokinetic variability and it may be possible to simplify area‐under‐curve (AUC) measurements using a limited sampling strategy. We have examined the timing of blood samples necessary to obtain accurate AUC predictions for cyclosporin using limited data from stable renal transplant patients dosed twice daily with Neoral. Best subset regression of blood concentration profile data obtained from ten patients at steady state indicated that two samples, timed at 2 and 8 h post‐dose, accounted for 97 % of the variance in AUC. The accuracy of this prediction was tested using profile data collected in a further 36 patients on three occasions separated by 4 and 44 weeks. Using the regression, AUC = 1.96 × [2 h] + 11.5 × [8 h] + 355.2, the mean (95 % CI) prediction errors of the three occasions were 1.7 % (‐ 2.1–5.4 %), 3.3 % (‐ 2.6–9.2 %) and 0.4 % (‐ 3.4–4.2 %). Data are presented that suggest AUC monitoring with a single blood sample could be feasible in a clinical setting.