Failure to predict cyclosporine area under the curve using a limited sampling strategy

Abstract

Graft survival in organ transplantation has significantly improved with the introduction of cyclosporin A (CsA) as a part of the immunosuppressive protocols. However, CsA has a variety of side effects, the primary one being renal toxicity, which may manifest as an acute reduction in the glomerular filtration rate (GFR) or as chronic vasculopathy and interstitial fibrosis. CsA nephrotoxicity is normally the consequence of overtreatment and/or drug interaction, but renal dysfunction in transplant patients can also be due to graft rejection often resulting from undertreatment. Major efforts have therefore been made in the last decade to find the ideal dose of CsA that achieves the desired immunosuppression without renal damage. This, unfortunately, is difficult, given the difficulties of drug monitoring. Current methods include measurement of trough plasma or whole blood concentrations. Monitoring trough concentrations of CsA can reveal cases of unusually rapid drug metabolism or poor absorption, but has limited value for assessing adequate immunosuppression or predicting protection against renal toxicity [1, 2]. The limitation of trough level monitoring may be attributed to many factors, including a poor correlation between blood CsA concentration in the circulation and at receptor target sites, and the intensity or duration of drug effect due to interindividual pharmacokinetic differences. Determination of the area under the timeconcentration curve (AUC), calculated from the individual pharmacokinetic profile, is more informative and allows a better index of drug exposure [3]. Thus, monitoring of AUC at clinical steady state has been reported to be more effective than trough level in dosage adjustment to control CsA therapy [3]. However, despite a complete pharmacokinetic profile (AUC) could provide more precise information, it is expensive and time consuming. Single CsA concentrations at five or six hours after dosing have been recently reported as a suitable alternative as they appear to correlate better with AUC values than trough levels [4, 5]. Other studies have suggested abbreviated AUC profiles involving three time points as a reliable alternative to full AUC [6, 7]. The present study was designed with the following aims: (a) to establish the time points of CsA pharmacokinetic profile that correlated best with AUC in renal transplant patients; and (b) to evaluate which of the equations derived from already published models, when applied to the present data, were predictive of AUC.