Accurate staging of liver fibrosis is important for clinical decision making and personalized management. Liver fibrosis is influenced by patients' genomics, including IFNL3 genotype and microRNA expression. However, incorporating microRNAs into fibrosis prediction algorithms has not been investigated. We examined the potential of eight selected serum microRNAs; miR-122, miR-126, miR-129, miR-199a, miR-155, miR-203a, miR-221, and miR-223 as non-invasive biomarkers to stage liver fibrosis in HCV-associated chronic liver disease (HCV-CLD). 145 Egyptian HCV-CLD patients were divided according to Metavir fibrosis scores. MicroRNAs and IFNL3 rs12979860 genotype were assayed by RT-qPCR and allelic discrimination techniques, respectively. Serum miR-122 was downregulated, whereas miR-203a and miR-223 were upregulated in significant fibrosis (≥F2) compared with no/mild fibrosis (F0-F1). Serum miR-126, miR-129, miR-203a, and miR-223 were upregulated in severe fibrosis (≥F3) and cirrhosis (F4) compared with F0-F2 and F0-F3, respectively. miR-221 was upregulated in ≥F3, but unchanged in F4. miR-155, miR-199a, and IFNL3 rs12979860 genotype were not significantly different in all comparisons. Differentially expressed serum microRNAs discriminated ≥F2, ≥F3, and F4 by receiver-operating-characteristic analysis. Multivariate logistic analysis revealed a model combining miR-129, miR-223, AST, and platelet count with high diagnostic accuracy for ≥F3 (AUC=0.91). The model also discriminated F4 (AUC=0.96) and ≥F2 (AUC=0.783), and was superior to APRI and FIB-4 in discriminating ≥F3 and F4, but not ≥F2. In conclusion, combining serum microRNAs with baseline predictors could serve as a new non-invasive algorithm for staging HCV-associated liver fibrosis. Additional studies are required to confirm this model and test its significance in liver fibrosis of other etiologies.
Keywords: HCV; IFNL3; fibrosis; microRNAs.
© 2017 John Wiley & Sons Ltd.