Liver fibrosis is the net result of dynamic changes between fibrogenesis and fibrolysis. Evidence has shown that antiviral therapy can reverse liver fibrosis or even early cirrhosis caused by hepatitis B virus. However, current evaluation systems mainly focus on the severity of, but not the dynamic changes in, fibrosis. Here, we propose a new classification to evaluate the dynamic changes in the quality of fibrosis, namely: predominantly progressive (thick/broad/loose/pale septa with inflammation); predominately regressive (delicate/thin/dense/splitting septa); and indeterminate, which displayed an overall balance between progressive and regressive scarring. Then, we used this classification to evaluate 71 paired liver biopsies of chronic hepatitis B patients before and after entecavir‐based therapy for 78 weeks. Progressive, indeterminate, and regressive were observed in 58%, 29%, and 13% of patients before treatment versus in 11%, 11%, and 78% after treatment. Of the 55 patients who showed predominantly regressive changes on posttreatment liver biopsy, 29 cases (53%) had fibrosis improvement of at least one Ishak stage, and, more interestingly, 25 cases (45%) had significant improvement in terms of Laennec substage, collagen percentage area, and liver stiffness despite remaining in the same Ishak stage. Conclusion: This new classification highlights the importance of assessing and identifying the dynamic changes in the quality of fibrosis, especially relevant in the era of antiviral therapy.(Hepatology 2017;65:1438‐1450)

Activation of TGFB (transforming growth factor β) promotes liver fibrosis by activating hepatic stellate cells (HSCs), but the mechanisms of TGFB activation are not clear. We investigated the role of ECM1 (extracellular matrix protein 1), which interacts with extracellular and structural proteins, in TGFB activation in mouse livers. We performed studies with C57BL/6J mice (controls), ECM1-knockout (ECM1-KO) mice, and mice with hepatocyte-specific knockout of EMC1 (ECM1Δhep). ECM1 or soluble TGFBR2 (TGFB receptor 2) were expressed in livers of mice after injection of an adeno-associated virus vector. Liver fibrosis was induced by carbon tetrachloride (CCl4) administration. Livers were collected from mice and analyzed by histology, immunohistochemistry, in situ hybridization, and immunofluorescence analyses. Hepatocytes and HSCs were isolated from livers of mice and incubated with ECM1; production of cytokines and activation of reporter genes were quantified. Liver tissues from patients with viral or alcohol-induced hepatitis (with different stages of fibrosis) and individuals with healthy livers were analyzed by immunohistochemistry and in situ hybridization. ECM1-KO mice spontaneously developed liver fibrosis and died by 2 months of age without significant hepatocyte damage or inflammation. In liver tissues of mice, we found that ECM1 stabilized extracellular matrix-deposited TGFB in its inactive form by interacting with αv integrins to prevent activation of HSCs. In liver tissues from patients and in mice with CCl4-induced liver fibrosis, we found an inverse correlation between level of ECM1 and severity of fibrosis. CCl4-induced liver fibrosis was accelerated in ECM1Δhep mice compared with control mice. Hepatocytes produced the highest levels of ECM1 in livers of mice. Ectopic expression of ECM1 or soluble TGFBR2 in liver prevented fibrogenesis in ECM1-KO mice and prolonged their survival. Ectopic expression of ECM1 in liver also reduced the severity of CCl4-induced fibrosis in mice. ECM1, produced by hepatocytes, inhibits activation of TGFB and its activation of HSCs to prevent fibrogenesis in mouse liver. Strategies to increase levels of ECM1 in liver might be developed for treatment of fibrosis.

Progression of liver fibrosis still occurs in some patients with chronic hepatitis B virus (HBV) infection despite antiviral therapy. We aimed to identify risk factors for fibrosis progression in patients who received antiviral therapy. We conducted a longitudinal study of patients with chronic HBV infection and liver biopsies collected before and after 78 weeks of anti-HBV therapy. Fibrosis progression was defined as Ishak stage increase ≥ 1 or as predominantly progressive classified by P-I-R system (Beijing Classification). Levels of HBV DNA and HBV RNA in blood samples were measured by real-time quantitative PCR. HBV RNA in liver tissue was detected by in situ hybridization. A total of 239 patients with chronic HBV infection with paired liver biopsies were included. Among the 163 patients with significant fibrosis at baseline (Ishak ≥ stage 3), fibrosis progressed in 22 patients (13%), was indeterminate in 24 patients (15%), and regressed in 117 patients (72%). Univariate and multivariate analyses revealed that independent risk factors for fibrosis progression were higher rate of detected HBV DNA at week 78 (odds ratio, 4.84; 95% CI, 1.30-17.98; P = .019) and alcohol intake (odds ratio, 23.84; 95% CI, 2.68-212.50; P = .004). HBV DNA was detected in blood samples from a significantly higher proportion of patients with fibrosis progression (50%) at week 78 than patients with fibrosis regression (19%) or indeterminate fibrosis (26%) (P = .015), despite low viremia (20-200 IU/mL) in all groups. The decrease of serum HBV RNA from baseline in the fibrosis regression group was larger than that in the fibrosis progression group. In a longitudinal study of patients with chronic HBV infection, we associated liver fibrosis progression at week 78 of treatment with higher rates of detected HBV DNA. We propose that a low level of residual HBV may still promote fibrosis progression, and that patients' levels of HBV DNA should be carefully monitored.