Patients and methods

Results

Table 1 summarises the clinical features of the enrolled patients. A total of 2014 patients were considered to be eligible for the study. Of these, 973 had a diagnosis of CH, 430 had LC, and 611 had HCC. The mean age of the patients with HCC was significantly higher than that of those with CH or LC (65.8 v 59.1 and 60.3 respectively). The proportion of women patients was much smaller for HCC than CH or LC (22.3% v 46.4 and 51.2% respectively). Both the smoking and alcohol indices were significantly higher in patients with HCC than in those with either CH or LC in accordance with many previous reports.8 ,22-24 At the time of enrollment, all patients were interviewed about their previous history; of the 1775 patients whose information was considered to be reliable, 629 (35.4%) had received a blood transfusion. The percentage of patients with a history of blood transfusion was not significantly different among patients with CH, LC, and HCC (38.0, 36.1, and 30.6% respectively). Information on the history of intravenous drug use and previous HBV vaccination was also obtained at enrollment, and a total of 12 and eight patients respectively had such histories. However, no association was found between these factors and the clinical stage of liver disease.

Table 2 shows serological status with regard to anti-HBV antibodies. A total of 52 anti-HCV positive patients with HBV marker (12 CH, 15 LC, and 25 HCC) were selected randomly and examined for the presence of circulating HBV DNA sequences, as detected by amplification of the S region.10 Serum HBV DNA was not present in any of the patients. Of those with HCV related chronic liver disease, 29.1% were positive for anti-HBs, a value that is clearly higher than that observed in the general population.25 More significantly, a total of 1004 patients (49.9%) were positive for the anti-HBc antibody; again, this rate is higher than that observed in the general population.25 The proportion of patients who were positive for the anti-HBs and anti-HBc increased with progression of the liver disease. In particular, the percentage of patients testing positive for anti-HBs and the percentage positive for anti-HBc were both significantly higher in cases of HCC than in cases of either CH or LC. It is widely recognised that the prevalence of anti-HBV serological markers increases with age, and the mean age of patients with HCV related chronic liver disease is generally high.23 ,24Thus it may be that the high prevalence of anti-HBV serological markers in these patients is merely a reflection of their advanced age. In order to clarify this, subjects without HCV infection were also tested for anti-HBc antibody, and compared with anti-HCV positive patients (table 3). Anti-HBc was detected in 44 of 352 (12.2%) controls and was distributed around 10% in any age group. This result was consistent with a previous report on blood donors in Japan.25 Even after matching for age, the rate of seropositivity for anti-HBc in HCV related chronic liver disease was significantly higher than that in the control of any age group. Furthermore, in order to assess the influence of the difference in mean age between HCC and non-HCC groups on the rate of anti-HBc positivity, the anti-HCV positive patients were split into two age groups (table 2). The positive rate for anti-HBc in the HCC group was found to be significantly higher than that in the CH and LC groups not only in patients under the age of 60 years, but also in those over the age of 60 years. The difference was more pronounced in the younger age group.

To exclude the influence of other factors on the increased positivity of anti-HBc in patients with HCC, the subgroup analyses by sex, alcohol intake, cigarette smoking, and history of blood transfusion were performed between the CH and HCC groups (table 4). We confirmed that the odds ratios for each factor were similar in each subgroup and these factors have no effect on seropositivity for anti-HBc in these patients. Furthermore, there were also no interactive effects of the history of intravenous drug abuse on the high prevalence of anti-HBc in the HCC group (data not shown).

According to a previous report, most blood donors with anti-HBc are usually also anti-HBs positive, while only 2.8% of patients are positive only for anti-HBc.25 In contrast with the pattern seen in healthy blood donors, there were many patients with HCV infection who were positive only for anti-HBc but negative for anti-HBs. In fact, 24.1% of the total number of patients with HCV related chronic liver disease had anti-HBc in the absence of anti-HBs. In contrast, the positivity of anti-HBs alone was 4.6% among anti-HCV positive patients, and this group included all eight individuals who had received an HBV vaccination.

It may be speculated that the high prevalence of anti-HBV serological markers in patients with HCV infection is due to the fact that many of them have had a blood transfusion in the past. Therefore, we evaluated the possible relation between a history of blood transfusion and the positive rate for anti-HBc in our study population. The positive rate for the anti-HBc was 46.9% and 50.9% in patients with and without a history of blood transfusion respectively. Thus, a history of blood transfusion cannot account for the higher rate of anti-HBV serological markers in patients with HCV infection. Similarly, other risk factors predisposing to viral transmission such as history of intravenous drug use or chance of sexual transmission were not different between anti-HBc positive and negative patients.

Discussion

In this study, we found that the prevalence of anti-HBc is almost 50% in patients with HCV related chronic liver disease, which is significantly higher than that reported in healthy blood donors at a blood centre in Japan.25 It is also higher than that observed in our control group without liver disease (12.2%). Several previous reports have already shown a high prevalence of anti-HBV serological markers in HCV infected patients.5-9 However, most of the previous reports have mainly focused on patients with HCC, and moreover have not assessed the anti-HBV serological status in relation to age. It is known that the prevalence of anti-HBV serological markers increases with age. Moreover, the mean age of patients with HCV related chronic liver disease is generally high,23 ,24 raising the possibility that the high prevalence of anti-HBV markers in HCV related chronic liver disease found in this study merely reflects their advanced age. Indeed, our patients with anti-HCV are older than the blood donors in the previous report. Therefore we selected subjects without HCV infection as a control group in order to compare the rates of seropositivity in age matched groups. This showed that the prevalence of anti-HBV serological markers was higher in patients with HCV related chronic liver disease than in control subjects irrespective of age.

It is also known that many patients with anti-HCV have a history of blood transfusion.23 ,24 Thus it could be postulated that the high prevalence of anti-HBV markers in patients with HCV infection is merely attributable to having received a blood transfusion in the past. In this study, however, we found that the percentage of patients with a history of blood transfusion was not different in those with and without anti-HBc. Moreover, although the prevalence of anti-HBc increases as the liver disease progresses, the percentage of patients with a history of blood transfusion was not different among patients with CH, LC, and HCC. Taking the results together, we could not find any relation between a previous history of blood transfusion and being positive for anti-HBc. In addition, we confirmed that other factors, such as sex, smoking, and drinking, had no influence on the seropositivity for anti-HBc. No remarkable differences were observed between the anti-HBc positive and negative group in the prevalence of histories predisposing to viral infection such as intravenous drug use. Thus, by analysing a large number of diverse patients including those with HCC, CH, and LC in Japan, we have clearly shown that the prevalence of anti-HBV serological markers is high in HCV related chronic liver disease even when other factors such as a past history of blood transfusion and differences in age are excluded.

It has been generally accepted that patients who are negative for HBsAg, but positive for anti-HBs and/or anti-HBc (low titre), have previously had transient HBV infection. However, it has been recently shown that traces of HBV are often detectable in the blood for many years after clinical recovery from acute hepatitis despite the presence of serum antibodies against the HBV and HBV specific cytotoxic T lymphocytes.26 ,27 Furthermore, we have recently shown the transmission of HBV from healthy donors who were negative for HBsAg but positive for anti-HBc (low titre) to recipients in living related liver transplantation.10 In addition, Koikeet al 6 ,21 have reported that clonally integrated HBV DNA sequences were identified at a high rate in patients with anti-HCV positive HCC who were seronegative for HBsAg. Taking these results together, it is strongly suggested that seronegativity for HBsAg does not exclude the existence of HBV DNA, and that HBV is often present in the liver tissue of patients who are positive for anti-HBc (low titre) even in the absence of HBsAg.11-21 ,28 Thus it is conceivable that the patients with HCV related chronic liver disease who are positive for anti-HBc may have dual sustained infection with both HBV and HCV in the liver.

In this study, we found a strong correlation between the clinical stage of liver disease and the prevalence of anti-HBV antibodies. Indeed, the prevalence of the anti-HBc in patients with HCC was about 60%, which was significantly higher than that observed in CH and LC. These results are in accordance with several previous reports,5-9 and suggest that HBV co-infection may be involved in hepatocarcinogenesis in patients with HCV related chronic liver disease. In contrast with our data, however, there is one study which showed that the rate of positivity for anti-HBV markers in patients with anti-HCV positive HCC was not high.29 The reasons for these discrepant findings are unclear. They may be due to geographical difference, even though both studies were carried out in Japanese patients. However, it is more likely to be attributable to the different methods of evaluation for HBV infection. HBV serostatus was evaluated mainly by measuring anti-HBs status in the previous study. Our data on HBV transmission in liver transplantation, however, suggest that it is more important to test for anti-HBc rather than anti-HBs to predict latent HBV infection.10

In relation to the importance of anti-HBc in evaluating latent HBV infection, an interesting observation in this study is the characteristic pattern of anti-HBV serological status in patients with HCV infection. Generally, healthy blood donors who are positive for anti-HBc also tend to be positive for anti-HBs. In this study, however, we found that as many as 24.1% of the patients with HCV related chronic liver disease were positive for anti-HBc but negative for anti-HBs. The reason for the difference in anti-HBV serological status between healthy blood donors and patients with HCV related chronic liver disease is unclear at present.30 It may be that HBV infection occurs in anti-HCV positive patients at a younger age than the healthy population, because anti-HBc remains detectable long after anti-HBs is no longer detectable.31 However, an alternative explanation may be that co-existing HCV protein has a negative effect on the synthesis of HBsAg, with a resulting decrease in anti-HBs production. Indeed, a previous study has shown an inverse relation between HBV and HCV replication.32 ,33 In another report, HCV core protein has been shown to dramatically suppress HBsAg synthesis.34

In conclusion, this study has clearly shown a high prevalence of anti-HBc in HCV related chronic liver disease, including not only HCC but also CH and LC. Considering the accumulating evidence showing that the seropositivity of anti-HBc is often associated with sustained hepatic infection with HBV, our data suggest the possibility that HBV infection, including latent infection, may be important in the development of HCC in patients with HCV related chronic liver disease.