Y. Choi, F. Jiang, H. Lee; Department of Clinical Pharmacology and Therapeutics, Seoul National University Hospital, Seoul, Korea, Republic of

BACKGROUND: Understanding the associations between the pharmacogenomic (PGx) variants and pharmacokinetics (PK) of tacrolimus, an immunosuppressant with a narrow therapeutic index and a wide PK inter-individual variability is important for personalized treatment. The aims of our study were to investigate the associations between genetic polymorphisms with tacrolimus PK variations using a high throughput genotyping method and to compare two data analysis methods frequently used in gene association studies: ANOVA and LASSO.
METHODS: We conducted a 96-hour full PK study in 42 healthy volunteers after administration of a single oral dose of tacrolimus. The DNA samples were genotyped for PGx markers using an Affymetrix DMET plus array. The associations between PGx markers and the PK(AUC0-48, Cmax) of tacrolimus were analyzed using LASSO and ANOVA.
RESULTS: An ANOVA analysis showed that three SNPs in CYP3A5, CYP3A7 and CYP3A4 genes (rs776746, rs2257401, rs2242480, respectively) were significantly related to the tacrolimus AUC0-48 (P < 0.05); a SNP in NR1I2 gene (rs3814055) also showed a marginal significance (P = 0.12). A LASSO analysis identified two SNPs in NR1I2 and CYP3A5 genes (rs3814055, rs776746, respectively) with the largest impacts on both the tacrolimus AUC0-48 and Cmax.
CONCLUSION: Our results agree with previous studies that CYP3A5*3 (rs776746) has a significant impact on the tacrolimus PK. The association identified for the first time between a SNP (rs3814055) in NR1I2 gene and the tacrolimus PK is interesting because NR1I2 gene encodes PXR, a transcriptional regulator of CYP3A enzymes; however, this association warrants further in vitro and in vivo studies. Using more than just one data analysis method may improve the interpretation of the results of gene association studies.