A model integrating donor gene polymorphisms predicts fibrosis after liver transplantation

Abstract

Post-transplant liver fibrosis (PTLF) is a common and severe complication in liver recipients. In this study, we assessed the impact of donor liver genetics on the development of PTLF. A total of 232 patients undergoing liver transplantation were included. Twenty-two single nucleotide polymorphisms (SNPs) associated with liver fibrosis were analyzed. Univariate analysis revealed seven donor SNPs to be associated with PTLF. In a multivariate analysis, independent risk factors of PTLF were genetic variation of donor GRP78 rs430397 (OR = 8.99, p = 0.003), GSTP1 rs1695 (OR = 0.13, p = 0.021), miRNA-196a rs12304647 (OR = 16.01, p =0.001), and TNF-α rs1800630 (OR = 79.78, p = 0.001); blood tacrolimus levels at maintenance > 7 ng/ml (OR =7.48, p <0.001); and post-transplant diabetes mellitus (OR = 7.50, p = 0.001). A predictive model that included donor SNPs showed better prognostic ability for PTLF than a model with only clinical parameters (AUROC: 0.863 vs 0.707, P < 0.001). Given that donor gene SNPs are associated with an increased risk of PTLF, this model integrated with donor gene polymorphisms may help clinicians predict PTLF.

Keywords: liver fibrosis; liver transplantation; novel integrated model; single nucleotide polymorphisms.

Figure 1

ROC curves of the model to predict LF after LT. LF: liver fibrosis; LT: liver transplantation; ROC: receiver operating characteristic curve.