Genetic and clinical determinants of early, acute calcineurin inhibitor-related nephrotoxicity: Results from a kidney transplant consortium

Background: Calcineurin inhibitor (CNI)-related acute nephrotoxicity is a common complication of transplantation. Clinical factors and elevated CNI levels are associated with nephrotoxicity; however, they do not fully explain the risk. Genetic factors may also predispose individuals to nephrotoxicity. Methods: We enrolled 945 kidney recipients into a multicenter, prospective study. DNA was genotyped for 2724 single-nucleotide polymorphisms (SNPs) using a customized chip. Cox models, unadjusted and adjusted for clinical factors, examined the association between SNPs and time to early CNI-related acute nephrotoxicity in the first 6 months posttransplant. Results: Cyclosporine was associated with a 1.49 hazard (95% confidence interval, 1.04-2.14) of acute nephrotoxicity relative to tacrolimus. Acute nephrotoxicity occurred in 22.6% of cyclosporine and 19.8% of tacrolimus recipients. The median (interquartile range) daily dose and trough concentration at time of nephrotoxicity were 400 mg (400-500 mg) and 228 ng/mL (190-272 ng/mL) in the cyclosporine group, and 6 mg (4-8 mg) and 12.6 ng/mL (10.2-15.9 ng/mL) in the tacrolimus group, respectively. In single-SNP adjusted analysis, nine SNPs in the XPC, CYP2C9, PAX4, MTRR, and GAN genes were associated with cyclosporine nephrotoxicity. In a multi-SNP analysis, SNPs from the same genes remained significant after adjusting for the clinical factors, showing that the SNPs are jointly and independently predictive of cyclosporine nephrotoxicity. No SNPs were associated with tacrolimus nephrotoxicity. Conclusion: We identified SNPs that were potentially associated with early, acute cyclosporine-related nephrotoxicity. Identifying risk SNPs before transplantation provides an opportunity for personalization of immunosuppression by identifying those who may benefit from CNI-avoidance or minimization, or assist in selecting CNI type. These SNPs require independent validation.