(M1330-10-67) Population Pharmacokinetic Model of Tacrolimus in Brazilian Bone Marrow Transplantation Patients: A Tool for Dosing Regimen Optimization

Purpose: Tacrolimus (TAC) is an immunosuppressive drug commonly administered to prevent graft-versus-host disease (GVHD) following hematopoietic stem cell transplantation (HSCT)¹. Despite its widespread clinical use, TAC exhibits significant interindividual variability in its pharmacokinetic (PK) parameters, such as clearance (CL) and volume of distribution (Vd), making it challenging to achieve the desired therapeutic range². The aim of this study was to develop a POPPK model of TAC in Brazilian adults and children after bone marrow transplantation at Porto Alegre Clinical Hospital (HCPA), a public hospital.

Methods: Patients undergoing HSCT were administered an initial dose of 0.3 mg/kg via continuous infusion. Therapeutic drug monitoring (TDM) was employed to maintain plasma TAC concentrations within the target range of 10-20 ng/mL. Dose adjustments were made using empiric protocols established by the hospital. A total of 455 observations from retrospective TDM data were utilized for model development. The popPK model was constructed using NONMEN® (Icon), and interindividual variability (IIV) was examined among subjects. Several demographic and biochemical covariates were evaluated to explain the observed IIV. External validation was performed by calculating the absolute prediction error (APE%) for population and individual predictions using additional plasma samples from different patients.

Results: A one-compartment model was obtained based on samples from 50 patients (ages 5 to 60 years old) treated with TAC. IIV in CL was accounted for by incorporating age and hematocrit (HCT) as covariates. The estimated typical values for CL and Vd were 9.37 L/h and 155 L, respectively. The IIV for CL and Vd were determined to be 30% and 101%, respectively. Model validation was conducted through visual predictive checks and non-parametric bootstrapping. The APE% for individual and population predictions were found to be 22.96 % and 17.19 %, respectively, which is consistent with the acceptance criteria of < 30%. The algorithm for individualized TAC therapy was described using individual CL equation and the target plasma concentration in steady-state resulting in the algorithm: Dose (mg)=Cpss_target*9.37*(AGE/26.84)^0.353)*(-0.0072*HCT)*e^(0.09).

Conclusion: This study successfully developed a popPK model for TAC tailored to the Brazilian population of HSCT patients. By incorporating patient-specific factors and covariates, the model provides a valuable tool for individualized dosing strategies, aiming to optimize TAC therapy outcomes. This tailored popPK model is the first of its kind in Brazil and offers more accurate and precise information for achieving the desired therapeutic target range in this specific patient population.

References: [1] Sharma N, et al. Cancers (Basel). v.13(4), p.613, 2021.
[2] Cheung J, et al. Can J Hosp Pharm. v.73(1), p.37-44, 2020.