PT-004 - PROSPECTIVE VALIDATION OF MAXIMUM A POSTERIORI-BAYESIAN ESTIMATION OF TACROLIMUS EXPOSURE IN STABLE KIDNEY TRANSPLANT RECIPIENTS.

N. Smith, T. Nguyen, K. Tornatore-Morse; University at Buffalo, Buffalo, NY, USA.

Background: Tacrolimus (TAC) exhibits interpatient pharmacokinetic (PK) variability, requiring therapeutic drug monitoring (TDM) of troughs. TAC troughs may not consistently predict drug exposure using area under the concentration-time curve 0-12 h (AUC0-12). This report is a prospective validation comparing TAC AUC0-12 estimation by maximum a posteriori-Bayesian (MAP-B) approach using robust (n=9) and sparse (n=2) timed samples relative to noncompartmental analysis (NCA) in stable kidney transplant recipients (KTR).

Methods: A prospective, open-label 12-hr PK study was conducted in stable KTR >12-months post-transplant. Nine serial samples determined steady-state TAC AUC0-12 using NCA. Robust and Sparse AUC0-12 estimates by MAP-B were obtained using the Advanced Dosing Solutions freeware. Differences between MAP-B and NCA estimates of AUC0-12 were tested using a paired sign test with intra-class correlation. Bias and precision were assessed using relative root mean squared error (rRMSE) and relative bias (rBias).

Results: 5 Black and 10 White stable KTRs (eGFR = 56.4 + 16.8 ml/min/1.73m2) received TAC dose:1.9±1.2 mg/study with troughs of 6.0±1.7 ng/ml. The NCA-AUC0-12 was 88.4 + 33.1 ng.h.ml-1 compared to MAP-B estimates for Robust-AUC0-12 = 85.1 + 33.8 ng.h.ml-1 (ICC=0.934; P=0.12) and Sparse-AUC0-12 of 86.7 + 33.9 ng.h.ml-1 (ICC=0.908; P=0.61) based on C(t0h) and C(t4h). NCA-AUC0-12 compared to Robust-AUC0-12 had R2 = 0.942; rBias = 1.83%; and rRMSE = 13.5%. The NCA-AUC0-12 comparison to Sparse-AUC0-12 had R2 = 0.897; rBias = 2.61%; and rRMSE = 16.1%.

Conclusion: MAP-Bayesian estimates using sparse 2-sample approach produced TAC AUC0-12 comparable to nine-sample NCA- AUC0-12. This sparse TAC AUC0-12 estimation with troughs may improve dosing individualization.

Campagne O, et al. Tacrolimus Population Pharmacokinetics and Multiple CYP3A5 Genotypes in Black and White Renal Transplant Recipients. Journal of clinical pharmacology. Sep 2018;58(9):1184-1195.