Some individuals can spontaneously clear the hepatitis C virus (HCV) after infection, whereas others develop a chronic infection. The exact mechanism of this phenomenon is unknown. We aimed to evaluate the association of plasma levels of MBL, L-ficolin, and cytokines with outcome of HCV infections in two groups of patients who cleared HCV spontaneously (CHS), and who developed chronic HCV infections (CHC). Altogether, 86 patients and 183 healthy controls were included. Of 86 patients, 36 had CHS and 50 had CHC. Concentrations of plasma MBL and L-ficolin were measured in patients and controls. Twenty plasma cytokines and adhesion molecules, including GM-CSF, ICAM-1, IFN-γ, IFN-α, IL-1α, IL-1β, IL-10, IL-12p70, IL-13, IL-17A, IL-4, IL-8, IP-10, MCP-1, IL-6, MIP-1α, MIP-1β, sE-Selectin, sP-Selectin, and TNF-α, were determined in all patients and randomly selected 45 controls. The level of MBL was significantly lower in subjects with CHS than in healthy controls (median: 293.10 vs. 482.64 ng/ml, p = 0.008), whereas the level of MBL was significantly higher in patients with CHC than in controls (median: 681.32 vs. 482.64 ng/ml, p = 0.001). No such differences in plasma L-ficolin were observed. Plasma levels of all cytokines and adhesion molecules, except ICAM-1, were significantly higher in patients than in controls. Moreover, patients with CHC had significantly higher levels of IFN-γ, IFN-α, IL-1α, IL-10, IL-13, IL-4, IL-6, and TNF-α than those with CHS. These findings implicate that lower levels of plasma MBL, together with lower levels of above mentioned cytokines may play a part in virus clearance of HCV infection.

Hepatitis C patients with advanced fibrosis or cirrhosis are at high risk of developing hepatocellular carcinoma (HCC) even after sustained virological response (SVR). Several HCC risk scores have been developed but which one is most suitable for this population is unclear. In this study, we compared the prediction ability of the aMAP model, THRI model, PAGE‐B model and Models of HCV in a prospective hepatitis C cohort in order to propose better model(s) to clinical practice. Adult hepatitis C patients with baseline advanced fibrosis (141 cases), compensated cirrhosis (330 cases) and decompensated cirrhosis (80 cases) were included and followed up every 6 months for about 7 years or until HCC development. Demographic data, medical history and laboratory results were recorded. HCCs were diagnosed by radiography, AFP or liver histology. The median follow‐up period was 69.93(60.99–74.93) months, during which 53 (9.62%) patients developed HCC. The areas under the receiver operating characteristic curve of aMAP, THRI, PAGE‐B and Models of HCV scores were 0.74, 0.72, 0.70 and 0.63 respectively. The predictive power of the aMAP model score was comparable to that of THRI, PAGE‐Band higher than that of Models of HCV (p < 0.05). Dividing patients into non‐high‐risk and high‐risk groups, the cumulative incidence rates of HCC based on aMAP, THRI, PAGE‐B and Models of HCV was 5.57% vs. 24.17%, 1.10% vs. 13.90%, 5.80% vs. 15.90% and 6.41% vs. 13.81% (all p < 0.05). The AUC of the four models were all below 0.7 in male while all were higher than 0.7 in female. The performance of all the models was not influenced by fibrosis stage. aMAP, THRI model and PAGE‐B model were all performed well while THRI model and PAGE‐B model were easier to calculate. There was no need to select score according to fibrosis stage but should be caution when explain the results in male patients.