Neoral monitoring by simplified sparse sampling area under the concentration-time curve: Its relationship to acute rejection and cyclosporine nephrotoxicity early after kidney transplantation

Background. Cyclosporine (CsA) dosing is traditionally based on trough blood levels (C₀) rather than area under the concentration-time curve (AUC), although AUC correlates better with posttransplantation clinical events. For Neoral, AUC based on limited sampling correlates closely with full 12-hr AUC. The purpose of our study was to correlate C₀ with AUC based on CsA levels at 0, 1, 2, 3, and 4 hr after dose (PK_0-4) and to compare this AUC with C₀ in predicting acute rejection (AR) and acute cyclosporine nephrotoxicity (CsANT) in de novo first kidney transplant patients. Methods. PK_0-4 was done 2-4 days after starting Neoral for 156 patients. All received CsA-based triple-drug immunosuppression without antibody induction. AUC was calculated as projected 12-hr (AUC_0-12) and actual 4-hr (AUC_0-4) from the PK_{0- 4} using the parallel trapezoid rule. Neoral dosing was based on C₀ not AUC. AUC was retrospectively compared with C₀ as a predictor of AR and CsANT during the first 90 days. Results. C₀ correlated poorly with AUC_0-12 and AUC_0-4 (r=0.61 and r=0.42). C₀ (mean ± SEM) levels were not significantly different in 34 patients with and 109 without AR (293±21 vs. 294±11 μg/L, P=0.95). AUC_0-12 and AUC_0-4 were significantly lower in patients with than without AR (AUC_0-12 9090±598 vs. 10608±336 μg·h/L, P=0.01; AUC_0-4 3934±306 vs. 4802±166 μg.h/L, P=0.006). In stepwise regression analysis only AUC_0-12 or AUC_0-4 (P=0.03[P=0.02) and delayed graft function (P=0.007) predicted AR. AUC_0-12, AUC_0-4, and C₀ were all significantly higher in patients with CsANT than without CsANT (AUC_{0- 12} 11746±650 vs. 10023±301 μg·h/L, P=0.01; AUC_0-4 5270±358 vs. 4474±150 μg·h/L, P=0.01; C₀ 343±18 vs. 287± 10 μg/L, P=0.01), but in stepwise regression analysis C₀ was not an independent predictor of CsANT. Patients with AUC_0-12 in the range of 9500 to 11500 μg·h/L or AUC_{0- 4} between 4400 and 5500 μg·h/L had the lowest incidence of AR (13% and 7%, respectively) without significantly higher risk for CsANT. Conclusion. C₀ correlates poorly with AUC based on PK_0-4. Early AUC based on PK_0-4 is more closely associated with AR and CsANT than is C₀. Our data suggest that a target AUC_0-12 of 9500-11500 or AUC_0-4 of 4400-5500 μg·h/L may provide optimal Neoral immunosuppression.