Abstract
BACKGROUND:
Tacrolimus has large inter-patient variability in pharmacokinetics (PK); however the variability has not been fully explained by predictive covariates commonly identified in routine clinical settings. To explore other contributing factors to the variability, we performed physiologically-based pharmacokinetic (PBPK) modeling using patient information in addition to pharmacogenetics analysis.
METHODS:
A base tacrolimus PBPK model was developed with the Simcyp simulator (version 16) using published data: physicochemical parameters; in vitro kinetic data for CYP3A4/5; and in vivo renal clearance. The predictive performance of the PBPK model was evaluated using published clinical tacrolimus PK data, including our original bioequivalence study with extensive sampling in renal transplant patients (n=35; Alloway et al., PLOS Medicine, 2017; NCT01889758). As part of the bioequivalence study, patients were genotyped for SNPs in CYP3A4/5 and PXR genes (14 total individual SNPs). The relationship between the genotypes and trough concentration (Ctrough) of tacrolimus was assessed by association analysis and examined through sensitivity analyses of CYP3A4/5 abundances using the developed model. Sensitivity analyses were also conducted to assess the impact of hematocrit, serum albumin and creatinine levels on tacrolimus PK.
RESULTS:
The PBPK model, with physiological parameters from renal transplant patients, described tacrolimus concentration-time profiles in CYP3A5 expressors and non-expressors which were comparable to clinical PK observations. The predicted PK profiles were sensitive to changes in CYP3A4 abundance, hematocrit and serum albumin levels, while the impact of serum creatinine level was negligible in CYP3A5 non-expressors. Association analysis revealed that the T/T genotype for rs2472677 SNP in PXR showed considerably higher Ctrough/dose ratios compared to the C/T genotypes, followed by C/C genotype (P = 0.006) in CYP3A5 non-expressors. The PBPK simulations captured PK variability observed in CYP3A5 non-expressors well, which is mainly due to the individual CYP3A4 contribution to tacrolimus PK.
CONCLUSION:
This study indicates that CYP3A4 abundance, regulated by PXR, appears to be an important determinant of PK variability of tacrolimus, when combined with CYP3A5 expression in renal transplant patients. The approach helps provide better understanding of factors contributing to PK variability, which will improve the outcomes for patients receiving tacrolimus therapy, although further confirmation using clinical data is needed.
