Abstract
Objective: To explore the predictive role of dynamic changes in serum Golgi protein 73 (GP73) and its inflammatory influencing factors on the reversal of hepatitis B virus-related liver fibrosis. Methods: Two hundred and seventy-eight patients with hepatitis B virus-related liver fibrosis who received entecavir or combined Fuzheng Huayu tablets treatment and completed two liver biopsies (biopsy) in Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine from September 2014 to July 2019 were selected. The correlation between serum GP73 level and fibrosis stage (Ishak) and inflammation grade (HAI) was analyzed. The patients were divided into a fibrosis reversal group (Ishak decreased≥1 point) and a non-reversal group (Ishak score remained unchanged or increased), and an inflammation improvement group (ΔHAI≤-2) and a non-improvement group (ΔHAI>-2) according to the pathological changes of liver tissue before and after treatment. The cross-sectional value of GP73, its change value (ΔGP73), and the role of inflammatory influencing factors on the liver before and after treatment were evaluated for their predictive efficacy regarding liver fibrosis regression. The receiver operating characteristic curve was used to explore the predictive value of serum ΔGP73 combined with liver stiffness change value (ΔLSM) for the reversal of hepatitis B virus-related liver fibrosis. One-way analysis of variance was used to compare the data between the groups of quantitative data, and a paired t-test or rank sum test was used for the data before and after treatment. The χ(2) test was used to compare the differences between the groups of enumeration data. Spearman and Pearson correlation methods were used for correlation analysis. Results: The serum GP73 level was higher in the cirrhosis group than that in the group without significant fibrosis (P<0.01). The GP73 level was higher in patients with moderate and severe inflammation than that in the mild group (P<0.05). Pre-treatment serum GP73 was positively correlated with fibrosis stage (r=0.248), inflammation grade (r=0.318), and alanine aminotransferase level (r=0.203) (P<0.01). The area under the receiver operating characteristic curve (AUROC) for the predictive ability of post-treatment GP73 levels in the fibrosis reversal was 0.633 (95%CI: 0.573-0.689, sensitivity 62.68%, and specificity 59.56%). The decrease in ΔGP73 was significantly higher in the liver fibrosis reversal group (n=142) than that in the non-reversal group (n=136) [-39.22(-85.08,-14.31) ng/mL vs. -30.06(-61.29,-5.84) ng/mL, P<0.01]. ΔGP73 was also associated with liver inflammation changes (AUROC=0.634, 95%CI: 0.574-0.690, sensitivity of 51.64%, specificity of 69.87%). Additionally, the predictive effectiveness of GP73 for fibrosis reversal improved after normalization of serum ALT (AUROC: 0.651 vs. 0.522 at baseline). ΔGP73 combined with ΔLSM had improved the AUROC predictive effectiveness from single indicators of 0.609 (ΔGP73) and 0.656 (ΔLSM) to 0.800 (95%CI: 0.662-0.899), with specificity increasing from 72.22% to 86.11%. Conclusion: Serum GP73 level is positively correlated with the degree of liver fibrosis and inflammation. Serum GP73 levels and ΔGP73 can predict the reversal of fibrosis, with liver inflammation being an important influencing factor following treatment. ΔGP73 combined with ΔLSM can significantly optimize the evaluation efficiency of liver fibrosis reversal.