Correction to: A pharmacogenomic study on the pharmacokinetics of tacrolimus in healthy subjects using the DMETTM Plus platform (The Pharmacogenomics Journal, (2017), 17, 2, (174-179), 10.1038/tpj.2015.99)

Correction to: The Pharmacogenomics Journalhttps://doi.org/10.1038/tpj.2015.99, published online 16 February 2016 The authors identified errors in the published article. Corrections are in bold. These corrections do not affect the results, discussion, or conclusions. The authors regret the errors and sincerely apologize to the Journal and readers of the Journal for any inconvenience. 1. There were errors in Tables 1 and 2. Corrections to the tables have been made, as well as their corresponding entries in the INTRODUCTION section. The “Reference allele” and “Mutant allele” columns in Tables 1 and 2 previously contained the information based on the PharmGKB database and PubMed database. But the authors found that the allele information from different databases can cause discrepancies. Therefore, a correction has been made based on the dbSNP database which is the largest public repository for genetic variation to clarify any confusion or misunderstanding. The corrected INTRODUCTION section, Table 1 and 2 appear below. Although rs776746 T > C (also known as CYP3A5*3), which is a nonfunctioning allele of the CYP3A5 gene, is associated with decreased tacrolimus metabolism⁴, the role of other genes, including the ABCB1, CYP2C19, POR, UGT1A8, NOD2, and PPARA, in the pharmacokinetics of tacrolimus was either inconsistent or insignificant³. (Table presented.) (Table presented.) Generic variants associated with tacrolimus AUC_last based on the FDR‐adjusted multiple testing analysis. The four most significant associations are shown here. SNP Gene Reference allele^‡ Mutant allele^‡ P-value FDR adjusted P-value rs776746 T C 0.00001 0.00466 rs2257401 G C 0.00001 0.00466 rs2242480 0.00029 0.06444 rs3814055 C T 0.00073 0.12125 ^‡ Reference and mutant alleles designated in this study were based on the dbSNP database in National Center for Biotechnology Information (NCBI). AUC_last: area under the concentration curve from time zero to the last quantifiable time point; SNP: single nucleotide polymorphism; FDR: false discovery rate. Genetic variants having a coefficient greater than zero in the LASSO models for tacrolimus AUC_last and C_max. SNP Gene Reference allele^‡ Mutant allele^‡ Coefficient AUC_last rs4986949 0.28485 rs776746 0.18146 rs3814055 C T 0.10679 rs2257401 G C 0.05650 rs16947 0.05085 rs7496 C T 0.03105 rs1736565 C T 0.02985 rs2020861 A G* 0.02571 rs6068816 0.01274 rs3803390 C T* 0.00302 rs1783811 A G* 0.00050 rs1080983 0.00008 C_max rs776746 T C 0.10999 rs3814055 0.00039 ^‡ Reference and mutant alleles designated in this study were based on the dbSNP database in National Center for Biotechnology Information (NCBI). ^* Multiple alleles exist. AUC_last, area under the concentration curve from time zero to the last quantifiable time point; C_max, maximum whole-blood concentration; SNP: single nucleotide polymorphism. 2. There was an error in the SUBJECTS and METHODS section, subsection “Determination of plasma concentrations of tacrolimus”, subsection “Pharmacokinetic analysis”, where “plasma concentration” should be “whole blood concentration”. A correction has been made, as well as their corresponding entries in Figure 1 and the caption on Table 3. Whole-blood concentrations of tacrolimus were determined using a previously published LC/MS/MS method¹⁸ with some modifications. The blood sample preparation involved a liquid/liquid extraction with methyl tert-butyl ether. Tacrolimus concentrations from the reference formulation were used for the pharmacokinetic analysis in the present study. The maximum whole-blood concentration (C_max) of tacrolimus was determined directly from the observed whole-blood concentration data. (Table presented.) (Figure presented.) P-values from a general linear model of the pharmacokinetic parameters for tacrolimus, where the CYP3A5 (rs776746) and NR1I2 (rs3814055) genotypes and their interaction term were the independent variables. P-value Adjusted r^{2 a} Interaction AUC_last <0.01 <0.05 0.16 0.54 C_max 0.20 <0.05 0.34 0.24 AUC_last, area under the concentration curve from time zero to the last quantifiable timepoint; C_max, maximum whole-blood concentration. ^a Proportion of variability that can be explained by the model consisting of the CYP3A5 and NR1I2 genotypes. Figure 1. Mean concentration-time profiles of tacrolimus (a) by different CYP3A5 and NR1I2 genotypes (n = 42) and (b) by two different combined CYP3A5 and NR1I2 genotypes (n = 9) where the genotypes represented the highest (CYP3A5 *3/*3 and NR1I2 T/T) and the lowest (CYP3A5 *1/*1 and NR1I2 C/C) exposure to tacrolimus. The error bars represent the standard deviations. 3. There was an error in the RESULTS section, subsection “Genetic effects of CYP3A5 and NR1I2 on tacrolimus pharmacokinetics”, where “CYP2A5*3/*3” should be “CYP3A5*3/*3”. The greater the number of nonfunctioning *3 alleles in the CYP3A5 gene, the greater the mean exposure to tacrolimus (Figure 1a). Consequently, the geometric mean AUC_last and C_max of tacrolimus was 2.78 (95% CI: 1.66–4.66) and 1.64 (95% CI: 1.04–2.60) times greater, respectively, in the CYP3A5*3/*3 homozygote than in the *1/*1 wild-type (P < 0.05; Figure 1a). 4. There was a typo in the DISCUSSION section. The corrected sentence appears below. Recently, a clinical trial in 32 kidney transplant patients showed that subjects with the rs3814055 C/C genotype had 1.2 and 1.5 times greater clearance of tacrolimus than the rs3814055 T carriers, C/T and T/T genotypes, respectively,²⁷ which supports the findings in our study.