Tacrolimus pharmacokinetics in lung transplantation: new strategies for monitoring.

BACKGROUND: Tacrolimus (TAC) dosing in lung transplantation is traditionally based on blood trough levels (C0). The best sampling strategy for the estimation of total drug exposure (area-under-the-curve [AUC]) has not been determined.
METHODS: Thirty-one 12-hour pharmacokinetic profiles were studied in 15 patients (8 men and 7 women, 42.0 +/- 13 years) post-bilateral lung transplantation (7.3 +/- 3.7 months; range, 3-18 months). Twelve-hour AUC (AUC0-12) was calculated by trapezoidal rule. Relationships between individual concentration points or abbreviated kinetics (2-4 concentration points) and AUC0-12 were determined by linear regression analysis (R2; absolute prediction error [APE]).
RESULTS: Pharmacokinetic profiles showed high variability, particularly in the absorption phase. AUC was 221 +/- 47.2 ng/ml (range, 156-329.3 ng/ml) at C0 10 to 15 ng/ml and was independent of TAC dose (R2 = 0.002). C0 was poorly predictive of AUC0-12 (R2 = 0.64; APE, 16.1% +/- 10.9%; range, 1.4%-37.8%). The predictive performance for AUC0-12 was highest with abbreviated kinetics using 4 (C0/C2/C3/C4: R(2) = 0.99; APE, 2.6% +/- 2.0%; range, 0.1%-7%) or 3 concentration points (C0/C2/C4: R2 = 0.98; APE, 2.6% +/- 2.1%; range, 0.1%-9.1%). Of the 2-point kinetics C2/C6 (R2 = 0.96; APE, 5.3% +/- 3.7%; range, 0.1%-12.7%), C2/C4 (R2 = 0.94, APE 6.7% +/- 4.8%; range 0.1%-14.6%) and C0/C4 (R2 = 0.94; APE 4.1% +/- 2.9%; range, 0.5%-11.4%) performed best. Single point strategies (best was C4: R2 = 0.94; APE 7.1% +/- 5.5%, range, 0.2%-24.1%) all had unacceptably high APE (range > 15%).
CONCLUSION: True TAC exposure shows high variability in stable lung transplant patients and is poorly predicted by C0. Abbreviated kinetics covering at least 2 concentration points between 0 and 4 hours post-drug intake are required for an accurate estimation of AUC.