The clinical characteristics of the study populations(31 CHB patients, 17 HBV carriers and 45 healthy controls)are described in Table 1. At the beginning of the observation period, we found that serum ficolin-2 concentrations of CHB patients were significantly higher than that of the HBV carriers(*p = 0.018) and healthy donors(**p < 0.0001), but there were no differences in ficolin-2 concentrations between HBV carriers and healthy donors(p = 0.7443, Figure 1). The mean concentration of ficolin-2 was 5.41 μg/mL in the CHB patients, 4.53 μg/mL in the HBV carriers, and 4.58 μg/mL in the healthy controls. Age and gender produced no significant differences between the healthy control group and the HBV group(CHB and HBV carriers)(p > 0.05).
Figure 1. Increased initial ficolin-2 protein concentrations in CHB patients compared with those of the HBV carriers and healthy donors.
Table 1. Characteristic of subjects involved in this study
Thirty-one CHB patients were divided into a "favorable outcome group" and an "unfavorable outcome group" based on their ALT, HBV DNA or HBeAg outcome responses according to the clinical practice guidelines published(European Association for the Study of the Liver, 2012)(Table 2). Next, we compared the initial serum ficolin-2 concentrations between the favorable outcome group and the unfavorable outcome group. We found that CHB patients with favorable viral responses, whose HBV DNA copies( > 500 copies/ml)declined after 48 weeks of amelioration liver function treatment(n = 23), had significantly higher levels of ficolin-2 than those in the unfavorable outcome group(n = 8)(Figure 2A, *p = 0.0474). The mean initial ficolin-2 concentration was 5.65 μg/mL in the DNA-decreased favorable group, while the mean ficolin-2 concentration was 4.76 μg/mL in the HBV DNA unfavorable outcome group.
Figure 2. Comparison of initial ficolin-2 concentrations between the favorable group and unfavorable group, according to the outcome responses of their HBV DNA levels (A), HBeAg conversion (B) and ALT levels (C).
Table 2. Clinical parameters of study participants by classification group
At the beginning of our observation, 20 of the 31 CHB patients were HBeAg positive. HBeAg seroconversion and HBeAg loss were defined as described in the Materials and Methods. We found that serum ficolin-2 concentrations in the HBeAg conversion group(n = 10) were significantly higher than that in the HBeAg loss and non-conversion(HBeAg positive)group(Figure 2B; Ⅰ vs Ⅱ, *p = 0.05; Ⅰ vs Ⅲ, *p = 0.036); however, there were no significant differences in ficolin-2 concentrations between those in the HBeAg loss group(n = 4) and HBeAg non-conversion group(n = 6)(Figure 2B, Ⅱ vs Ⅲ, p = 0.437).
However, we did not observe significant differences in the initial serum ficolin-2 concentrations between the favorable outcome ALT group(ALT values > 40 IU/L)where levels returned to normal ALT(≤ 40 IU/L, n = 23) and the unfavorable outcome ALT group(ALT values abnormal, > 40 IU/L, n = 8)(Figure 2C, p = 0.8257).
We further compared the correlation between ficolin-2 concentration changes and disease outcome. Ficolin-2 concentrations significantly decreased in the favorable outcome ALT group(ALT values returned to normal, ≤ 40 IU/L)(Figure 3A, *p = 0.0129), but not in the unfavorable outcome ALT group(ALT values abnormal, > 40 IU/L)(Figure 3A, p = 0.4027). Similarly, ficolin-2 concentrations decreased significantly in the HBV DNA favorable outcome group(DNA decreased group, n = 23)(Figure 3B, **p = 0.004), but not in the DNA unfavorable outcome group(no declined group, n = 8)(Figure 3B, p = 0.248).
Figure 3. Serum ficolin-2 concentrations were decreased and positively correlated with the favorable ALT, HBV DNA and HBeAg-seroconversion outcome. Analysis of ficolin-2 concentrations between Week 0 and Week 48 of treatment with the favorable and unfavorable ALT (A), HBV DNA (B) and HBeAg-seroconversion outcome (C).
We also found that serum ficolin-2 concentrations decreased significantly in the favorable HBeAg(HBeAg conversion)group(n = 10) after 48 weeks(Figure 3C, *p = 0.043), while levels did not decrease in the HBeAg loss group(n = 4)(Figure 3C, p = 0.115) or HBeAg non-conversion group(n = 6)(Figure 3C, p = 0.2326).
We observed that early increased serum ficolin-2 concentrations were dynamically decreased(Figure 4A, Wk 48 vs Wk 0, *p = 0.0312) and were accompanied by decreased ALT values(Figure 4B, Wk 48 vs Wk 0, **p < 0.0001) and HBV DNA levels(Figure 4C, Wk 48 vs Wk 0, **p < 0.0001) in the 31 CHB patients during 48 weeks of treatment. We also observed that hepatic inflammation(by hematoxylin and eosin(H & E)staining)at Week 0 was more serious than that at Week 48(Figure 4D). Liver histological inflammation grades were assessed using a standard Knodell scoring system by experienced pathologists according to previous reports(Desmet et al., 1994; Mannan et al., 2014). Inflammation scores of the Knodell histological activity index(HAI)were displayed by the sum of portal inflammation(from 0 to 4), periportal piecemeal necrosis(from 0 to 10) and intralobular inflammation(from 0 to 4) and total scores were 18(Desmet et al., 1994). Our data showed that inflammation scores of HAI were 4 and 2 at week 0 and week 48, respectively(Figure 4D). This result suggests that the patients' inflammation was alleviated after 48 weeks and was accompanied by decreased ficolin-2 expression and HBV DNA levels.
Figure 4. Dynamically decreased serum ficolin-2 concentrations were observed, accompanied by decreased ALT values/HBV DNA levels. Dynamically decreased serum ficolin-2 concentrations (A), ALT values (B) and HBV DNA levels (C) were analyzed during the 48 weeks of treatment in the 31 CHB patients. Data in A–C were expressed as the mean values at each time point. (D) The representative data from liver biopsy sections (n= 4) at Week 0 and Week 48 of treatment were analyzed with H & E reagent and evaluated by light microscope. Arrows represent the infiltration of inflammatory cells.
The dynamic changes in ficolin-2 concentrations in HBeAg-positive patients were further analyzed. Ten patients(Patients 1 to 10) with HBeAg seroconversion displayed a significant decrease in serum ficolin-2 concentrations that were accompanied by decreased ALT values and HBV DNA levels(Figure 5A); 4 patients(Patients 11 to 14) with HBeAg loss(HBeAg-, HBeAb-)displayed a slight decrease in ficolin-2 concentrations(Figure 5B); 6 patients(Patients 15 to 20) with HBeAg non-conversion(HBeAg+, HBeAb-)displayed no change in serum ficolin-2 concentrations, HBV DNA levels or ALT values(Figure 5C).
Figure 5. Representative data of ficolin-2, ALT and HBV-DNA after 48 weeks of treatment among 20 positive HBeAg CHB patients. (A) Decreased concentrations of serum ficolin-2 in the 10 individual patients with HBeAg conversion were observed and correlated with the ALT and/or HBV DNA levels during 48 weeks of treatment. (B) Serum ficolin-2 concentrations were slightly decreased in the 4 HBeAg loss patients (Patients 11 to 14) associated with decreased HBV DNA levels and/or ALT values. (C) Six patients (Patients 15 to 20) with non-conversion of HBeAg (HBeAg+, HBeAb-), and unfavorable outcome of both HBV DNA and ALT, displayed no decrease in serum ficolin-2 concentrations.
Previous reports have shown that ficolin-2 is mainly produced by liver cells. Little is known about serum and intrahepatic ficolin-2 expression in HCC tissue cells. We examined serum ficolin-2 concentrations in 45 positive HBV(+) and 15 negative HBV(-)HCC patients(Table 3) and found significantly lower serum ficolin-2 concentrations in HCC patients compared to those in 45 healthy donors(Figure 6A). HBV(+)HCC patients had slightly higher mean serum ficolin-2 concentrations than HBV(-)HCC patients(Figure 6A). We also examined intrahepatic ficolin-2 expression in liver biopsy specimens from 26 HCC and 10 cirrhosis patients(Table 3). By using immunohistochemistry, we found that much lower intrahepatic ficolin-2 expression occurred in HCC tissue cells and different stages of HCC tissue cells(from tumor node metastasis(TNM)-Ⅰ to TNM-Ⅳ)compared to adjacent normal tissue cells(Figures 6B–6E). Hepatic cirrhosis also led to less ficolin-2 expression compared to adjacent normal hepatic cells(Figure 6B). These data suggest that liver cell damage in HCC and cirrhosis tissue could cause decreased production of intrahepatic ficolin-2 because, as previously noted, ficolin-2 is mainly produced by liver cells. We also observed that HBV(+)HCC tissue cells showed slightly higher ficolin-2 expression than HBV(-)HCC tissue cells(Figure 6D), which is consistent with the observation of increased serum ficolin-2 levels in CHB patients(Figure 1).
Figure 6. Much lower protein expressions of serum and intrahepatic ficolin-2 in HCC and cirrhosis compared with those of healthy controls. (A) Decreased serum ficolin-2 protein concentrations in HBV (+) HCC and HBV (-) HCC patients compared with those of healthy donors. (B) Statistical analysis of intrahepatic ficolin-2 expression among adjacent normal tissue cells, cirrhosis liver cells and HCC tissue cells. The data shown are means ± SEM of the indicated independent experiments. (C) Representative data of intrahepatic ficolin-2 expression in the HCC tumor tissue cells and adjacent normal tissue cells from HCC patients by immunohistochemistry analysis. (D) Representative data of intrahepatic ficolin-2 expression by immunohistochemistry analysis in the adjacent normal hepatic cells, HBV (+) and HBV (-) tissue cells from HCC patients, and the negative control (without addition of the primary antibody anti-ficolin-2 mAb for the normal hepatic tissue cells). (E) Representative data of intrahepatic ficolin-2 expression by immunohistochemistry analysis in the different stages of HCC tumor tissue cells.
Table 3. Characteristic of HCC and cirrhosis subjects involved in this study