Development of a gene signature for predicting cirrhosis risk score of chronic liver disease associated with HCV infection in Egyptians.

ABSTRACT

BACKGROUND: Complex diseases such as fibrosis are likely polygenic. Lately, cirrhosis risk score (CRS) clearly discriminated Chronic HCV patients with high-risk versus those with low-risk for cirrhosis better than clinical factors. METHODS: Herein, the CRS was assessed via genotyping by allelic discrimination assays in 243 HCV Egyptian patients categorized into 164 patients didn't develop HCC (93 mild, 71 advanced fibrosis); and 79 patients developed HCC. APRI and FIB-4 scores were calculated, compared with CRS and correlated with degree of fibrosis progression. RESULTS: Median of the three CRS, APRI and FIB-4 scores were significantly elevated in late fibrotic and HCC patients (p < 0.001); however CRS displayed proper discrimination (mild fibrosis (0.59; 0.4-0.75), advanced fibrosis (0.75; 0.7-0.86) and HCC (0.73; 0.57-0.77); (p < 0.001)). The ROC analysis of CRS score displayed modest accuracy to discriminate between mild and advanced fibrotic patient; AUC was 0.73; p < 0.0001), while AUC was only 0.57 (p = 0.05) for the discrimination between HCC and no HCC. Moreover, the combination of CRS, APRI and FIB4 lessened the power of correlation (AUC, 0.63 (p < 0.0001)) in fibrosis prognosis. In HCC prognosis, the combination of CRS, APRI and FIB4 in HCC patients showed modest accuracy with AUC, 0.59 (p = 0.0001). CONCLUSION: The diagnostic accuracy of FIB-4 for predicting liver fibrosis was nearly identical to that of CRS, however the strength of CRS score stemmed from that it is built on 7 SNPs host genetic factor. Our study validates non invasive algorithms for fibrosis prognosis purposes which may aid in decision making for therapeutic intervention.