Model-Based Tacrolimus Follow-up Dosing in Adult Renal Transplant Recipients: A Simulation Trial

Background: Initial algorithm-based dosing appears to be effective in predicting tacrolimus dose requirement. However, achieving and maintaining the target concentrations is challenging. Model-based follow-up dosing, which considers patient characteristics and pharmacological data, may further personalize treatment. This study investigated whether model-based follow-up dosing could lead to more accurate tacrolimus exposure than standard therapeutic drug monitoring (TDM) in kidney transplant recipients after an initial algorithm-based dose.

Methods: This simulation trial included patients from a prospective trial that received an algorithm-based tacrolimus starting dose followed by TDM. For every measured tacrolimus predose concentration (C 0,obs ), model-based dosing advice was simulated using the InsightRX software. Based on previous tacrolimus doses and C 0 , age, body surface area, CYP3A4 and CYP3A5 genotypes, hematocrit, albumin, and creatinine, the optimal next dose, and corresponding tacrolimus concentration (C 0,pred ) were predicted.

Results: Of 190 tacrolimus C 0 values measured in 59 patients, 121 (63.7%; 95% CI 56.8-70.5) C 0,obs were within the therapeutic range (7.5-12.5 ng/mL) versus 126 (66.3%, 95% CI 59.6-73.0) for C 0,pred ( P = 0.89). The median absolute difference between the tacrolimus C 0 and the target tacrolimus concentration (10.0 ng/mL) was 1.9 ng/mL for C 0,obs versus 1.6 ng/mL for C 0,pred . In a historical cohort of 114 kidney transplant recipients who received a body weight-based starting dose followed by TDM, 172 of 335 tacrolimus C 0 (51.3%) were within the therapeutic range (10.0-15.0 ng/mL).

Conclusions: The combination of an algorithm-based tacrolimus starting dose with model-based follow-up dosing has the potential to minimize under- and overexposure to tacrolimus in the early posttransplant phase, although the additional effect of model-based follow-up dosing on initial algorithm-based dosing seems small.