In spite of rigorous dose adjustments by way of therapeutic drug monitoring, a large proportion of kidney transplant recipients are unable to achieve the target tacrolimus trough concentrations. This is attributed to the narrow therapeutic window of the drug (10-15 ng/mL) and large inter-individual variability in pharmacokinetic parameter such as clearance. There is a need for development of clinical dosing models that can help prospectively predict the dose for an individual, especially in the critical period immediately post-transplant. Therefore, we established and quantified the effect of clinical and genetic factors on tacrolimus clearance (CL/F) using a large population of adult kidney transplant recipients. Tacrolimus troughs (n=11823) from 681 transplant recipients over the first 6-months post-transplant were analyzed using non-linear mixed effects modeling approach in NONMEM®. The troughs were characterized by a steady state infusion model. Covariates were analyzed using a forward selection (p<0.0.1) backward elimination (p<0.001) approach. We formulated an equation that predicts the CL/F of an individual based on the days post-transplant, presence of the highly influential CYP3A5*1 genotype, transplant at a steroid sparing center, age and concomitant use of a calcium channel blocker at the time of trough collection. The CL/F was seen to decrease with increasing days post transplant, transplant at a steroid sparing center and use of a calcium channel blocker. Transplant recipients with the CYP3A5*1/*3 and *1/*1 genotypes had a CL/F that was 70% and 100% higher, respectively, than those with the CYP3A5*3/*3 genotype. The dose required in order to achieve a particular target trough can be prospectively determined from this equation. The above equation was validated in a separate cohort of adult kidney transplant recipients. The equation was assessed by predictive performance in 795 transplant recipients (n=13,968 troughs) receiving tacrolimus using bias and precision. Assessment was done for the initial troughs as well as for all troughs over the entire 6 months. The equation has low bias (0.2 ng/ml) and good precision (within ± 20% for a typical trough of 10 ng/mL) in predicting initial troughs and could be safely used to predict initial doses. This is critical as an accurate initial dose will help the recipient to get to therapeutic range faster and reduce the number of out-of-range troughs. For all the troughs, over the 6 months post-transplant, the equation did better than a basic model with no covariates but had higher bias and imprecision than the prediction of initial troughs. We were presented with 119 single nucleotide polymorphisms (SNP) in this study. Due to software limitations and impracticalities associated with such a large number of covariates, we developed and validated a novel "winnowing method" of covariate selection that is able to test and select SNPs in combination. This method uses random selection, repetitions of generalized additive modeling in the R statistical package and post-hoc estimates from NONMEM®. The salient feature of this method is the creation of an index, ranging from 0-1, that defines the relative importance of the SNP when tested in a combination. With this method, we were able to select 26 SNPs out of the 119 SNPs, which included the well-established CYP3A5*1 SNP. We validated this method using a simulated dataset. In the validation dataset, the winnowing method was able to select all the important SNPs. The type I and type II error rates were 9% and 0% respectively. Although NONMEM® is the oldest and most widely used population pharmacokinetics software, several other software packages are now becoming available such as the Phoenix® NLME<sup>TM</sup>. One desirable feature in this new software package is a graphical user interface and menu-driven covariate selection options. Therefore, we compared these two software packages in terms of covariates selected and predictive performance using both clinical and simulated data. For the tacrolimus data, NONMEM® predictions had lower bias and imprecision as compared to Phoenix® NLME<sup>TM</sup>. For the clinical data, NONMEM® predictions had higher bias but were more precise than the Phoenix® NLME<sup>TM</sup> predictions.>
University of Minnesota Ph.D. dissertation. December 2012. Major: Experimental & Clinical Pharmacology. Advisor: Dr. Angela K. Birnbaum. 1 computer file (PDF); xiii, 130 pages, appendices p. 124-130.
Application of pharmacometrics for covariate selection and dose optimization of tacrolimus in adult kidney transplant recipients.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.