AUTHOR=Galvez Carla , Boza Pía , González Mariluz , Hormazabal Catalina , Encina Marlene , Azócar Manuel , Castañeda Luis E. , Rojo Angélica , Ceballos María Luisa , Krall Paola 

TITLE=Evaluation of limited-sampling strategies to calculate AUC(0–24) and the role of CYP3A5 in Chilean pediatric kidney recipients using extended-release tacrolimus

JOURNAL=Frontiers in Pharmacology

VOLUME=14

YEAR=2023

URL=https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2023.1044050

DOI=10.3389/fphar.2023.1044050

ISSN=1663-9812

ABSTRACT=<p><bold>Background:</bold> Kidney transplantation (KTx) requires immunosuppressive drugs such as Tacrolimus (TAC) which is mainly metabolized by CYP3A5. TAC is routinely monitored by trough levels (C<sub>0</sub>) although it has not shown to be a reliable marker. The area-under-curve (AUC) is a more realistic measure of drug exposure, but sampling is challenging in pediatric patients. Limited-sampling strategies (LSS) have been developed to estimate AUC. Herein, we aimed to determine AUC<sub>(0–24)</sub> and <italic>CYP3A5</italic> genotype in Chilean pediatric kidney recipients using extended-release TAC, to evaluate different LSS-AUC<sub>(0–24)</sub> formulas and dose requirements.</p><p><bold>Patients and methods:</bold> We analyzed pediatric kidney recipients using different extended-release TAC brands to determine their trapezoidal AUC<sub>(0–24)</sub> and <italic>CYP3A5</italic> genotypes (SNP rs776746). Daily TAC dose (TAC-D mg/kg) and AUC<sub>(0–24)</sub> normalized by dose were compared between CYP3A5 expressors (*1/*1 and *1/*3) and non-expressors (*3/*3). We evaluated the single and combined time-points to identify the best LSS-AUC<sub>(0–24)</sub> model. We compared the performance of this model with two pediatric LSS-AUC<sub>(0–24)</sub> equations for clinical validation.</p><p><bold>Results:</bold> Fifty-one pharmacokinetic profiles were obtained from kidney recipients (age 13.1 ± 2.9 years). When normalizing AUC<sub>(0–24)</sub> by TAC-D significant differences were found between CYP3A5 expressors and non-expressors (1701.9 vs. 2718.1 ng*h/mL/mg/kg, <italic>p</italic> &lt; 0.05). C<sub>0</sub> had a poor fit with AUC<sub>(0–24)</sub> (<italic>r</italic><sup>2</sup> = 0.5011). The model which included C<sub>0</sub>, C<sub>1</sub> and C<sub>4</sub>, showed the best performance to predict LSS-AUC<sub>(0–24)</sub> (<italic>r</italic><sup>2</sup> = 0.8765) and yielded the lowest precision error (7.1% ± 6.4%) with the lowest fraction (9.8%) of deviated AUC<sub>(0–24)</sub>, in comparison to other LSS equations.</p><p><bold>Conclusion:</bold> Estimation of LSS-AUC<sub>(0–24)</sub> with 3 time-points is an advisable and clinically useful option for pediatric kidney recipients using extended-release TAC to provide better guidance of decisions if toxicity or drug inefficacy is suspected. The different <italic>CYP3A5</italic> genotypes associated with variable dose requirements reinforce considering genotyping before KTx. Further multi-centric studies with admixed cohorts are needed to determine the short- and long-term clinical benefits.</p>