Accelerating tacrolimus model-informed precision dosing in kidney transplant recipients: Model evaluation and...

Area of Science:

• Pharmacokinetics and Pharmacodynamics
• Transplantation Medicine
• Computational Biology

Background:

• Tacrolimus is a crucial immunosuppressant post-kidney transplant.
• Model-Informed Precision Dosing (MIPD) aims to optimize tacrolimus therapy.
• Evaluating existing population pharmacokinetic (popPK) models is key for efficient MIPD implementation.

Purpose of the Study:

• Compare the predictive performance of published tacrolimus popPK models against an in-house model.
• Assess model performance in Uruguayan kidney transplant recipients.
• Guide efficient MIPD implementation through external model evaluation.

Main Methods:

• Systematic review identified 25 adult tacrolimus popPK models.
• External evaluation used retrospective data from 39 patients (396 concentrations).
• An in-house model was developed using 49 patients (727 concentrations); predictive performance assessed via a priori and Bayesian forecasting.

Main Results:

• Both published and in-house models showed poor a priori predictive performance (MAPE ~50%).
• Bayesian forecasting significantly improved prediction accuracy with data integration.
• High precision (bias <10%, MAPE <30%) was achieved with at least 3 post-dose concentrations via Bayesian updating.

Conclusions:

• Published tacrolimus popPK models, with Bayesian updating, achieve clinically reliable predictions.
• These models perform comparably to in-house developed models.
• Efficient MIPD implementation is feasible without local model development using existing popPK models and early pharmacokinetic curve data.