Hepatitis C virus 1 b ( II ) Infection and Development of Chronic Hepatitis , Liver Cirrhosis and Hepatocellular Carcinoma : a Case-Control Study in Japan

. Hepatology, 1996; 24: 350A. 36. Tsukuma H, Hiyama T, Tanaka S, et al. Risk factors for hepatocellular carcinoma among patients with chronic liver disease. N Engl J Med, 1993; 328: 1797-1801. 37. Kuzushita N, Hayashi N, Katayama K, et al. Increased frequency of HLA DR 13 in hepatitis C virus carriers with persistently normal ALT levels. J Med Virol, 1996; 48: 17. 38. Tibbs C, Donaldson P, Underhill J, et al. Evidence that the HLA DQAl*03 allele confers protection from chronic HCV-Infection in northern European caucasoids. Hepatology, 1996; 24: 1342-5. 39. Tanaka H, Hiyama T, Tsukuma H, et al. Cumulative risk of hepatocellular carcinoma in Hepatitis C virus carriers: statistical estimations from cross-sectional data. Jpn J Cancer Res, 1994; 85: 485-90. 40. Bisceglie AMD, Goodman ZD, Ishak KG, et al. Longterm clinical and histopathological follow-up of chronic posttransfusion hepatitis. Hepatology, 1991; 14: 969-74. 41. Seeff LB, Buskell BZ, Wright EC, et al. Long-term mortality after transfusion-associated non-A, non-B hepatitis. N Engl J Med, 1992; 327: 1906-11.

Hepatitis C virus (HCV) has been identified as the major causative agent of chronic liver disease and greatly contributes to the etiology of liver cirrhosis (LC) 1.2.3) and hepatocellular carcinoma (HCC) '~.It is widely assumed that HCV, a nonintegrating RNA virus, causes HCC through a pathogenic mechanism inducing chronic liver inflammation and subsequent cirrhosis 6.6), although there are some reports of HCVrelated HCC in the absence of cirrhosis ',).
Previously reported genomic nucleotide sequence analysis of HCV showed that HCV has at least 6 major genotypes and a series of subtypes9-13).Recent studies suggested a possible role for HCV genotype in deterioration of liver chronicity and advanced liver disease outcome.A cohort study showed that blood donors with genotype-1 (mainly lb) tend to have elevated alanine aminotransferase levels more frequently than those with other genotypes 14), that chronic hepatitis of genotype-1 (mainly lb) is associated with more rapid deterioration of liver histology 15), and that patients with LC with type lb have a significantly higher risk of HCC than those infected by other genotypes 16).Case-control studies showed that HCV type lb infection in healthy carriers In and in patients with LC 18,11) had a significantly higher risk of developing HCC than those with HCV type 2a and other genotypes.However, another cohort study has shown no differences in the prognosis and risk of progression to HCC based on the genotype of HCV 2D), and a case-control study has shown that HCV type lb infection was not a risk factor for cirrhosis in patients with NC-CH 21).This report assesses the prevalence of major HCV genotypes in 1,151 individuals with HCV infection in the Osaka area over a period of five years.The relation between the genotype of HCV and liver disease outcome has been evaluated by three case-control studies using ASCs as control subjects to calculate the odds ratios contrasting NC-CH, LC, and HCC, respectively.

SUBJECTS AND METHODS
Control subjects who were found positive by some type of HCV test but had no symptoms of chronic hepatitis were recruited from Osaka Red Cross Blood Center (ORCBC) and Osaka Medical Center for Cancer and Cardiovascular Diseases (OMCC) between 1992 and 1994.Of 17,168 blood donors at the ORCBC, 267 were anti-HCV positive by a second generation passive hemagglutination assay (PHA) (Dainabot Co. Ltd., Tokyo, Japan) Za.Among these, 114 subjects showed a positive result for HCV by the HCV-RNA test, but had serum alanine amino transferase levels (ALT) lower than 1.5 times the upper limit of normal (50 Karmen units).We collected additional data on HCV genotypes among 31 out patients at OMCC, who showed a positive result for HCV-RNA testing but their ALT values were lower than 1.5 times the upper limit of normal, without any clinical symptoms of chronic liver disease.All the 145 subjects for control were negative for hepatitis B surface antigen (HBsAg).
We collected data on HCV genotypes from 742 patients with NC-CH among those who were treated at the First Department of Internal Medicine, Osaka University and its referral hospitals between 1990 and 1994.The inclusion criteria of the NC-CH patients were: 1) the presence of a chronic liver disease of more than 12 months' duration from diagnosis based on the patient's medical history and physical examination, 2) liver histology or ultrasonography compatible with chronic hepatitis without cirrhosis, 3) the presence of both anti-HCV and HCV-RNA, and 4) the absence of HBsAg.Most of these chronic hepatitis patients had been histologically confirmed in preparation for interferon therapy.
Two hundred eighty-five patients with liver cirrhosis who were diagnosed at the Department of Internal Medicine, OMCC, from 1990 to 1994 were enrolled.Of these patients, 182 had HCC.All patients were positive for HCV with anti-HCV and HCV-RNA testing.The diagnosis of LC was histologically confirmed or was based on clinical signs of portal hypertension, such as esophageal or gastric varices, ascites and splenomegaly.The diagnosis of HCC was made by ultrasonography-guided fine-needle biopsy in most of the patients or by liver-selective angiographic and/or computed tomographic imaging, with a -fetoprotein values of more than 200 ng/mL.Tests for hepatitis B surface antigens were negative in all of these patients.
HCV-RNA was detected by nested reverse-transcription polymerase chain reaction with primers derived from the 5 non-coding regions, as described by Okamoto, et al. a> HCV genotyping was performed by reverse-transcription polymerase chain reaction amplification of core region sequences with universal and 5 subtype-specific primers, according to Okamoto, et al. 24.25)Through this method, five genotypes (I , 1 1, III , N and V which correspond to la, l b, 2a, 2b and 3a, respectively, by the classification system of Simmonds, et al. 12)) were identified.The detection of HCV-RNA and the genotyping among NC-CH were performed before the introduction of interferon therapy.
The association between HCV lb infection and the development of NC-CH, LC and HCC was expressed as odds ratios with 95% confidence interval (CI) calculated by unconditional logistic regression using the SAS/PC statistical package (SAS Institute Inc., Cary, NC).Variables included in the analysis were sex, age at diagnosis, and HCV genotype (lb vs. 2a, 2b).

RESULTS
Genotyping was accomplished for all of the 1,172 subjects.Among these, 21 cases were excluded because they had multiple HCV genotypes (2 with ASC had 2a and 2b; 5 with NC-CH had lb and 2a; 2 with NC-CH had lb and 2b, 9 with NC-CH had 2a and 2b, 1 with HCC had la and lb; and 2 with HCC had 2a and 2b) Table 1 shows the demographic factors and distribution of HCV genotypes.There were no subjects with HCV type 3a.HCV type lb was detected in 100 out of 143 subjects with ASC (69.9%), 551 out of 726 with NC-CH (75.9%), 86 out of 103 with LC (83.5%), and 153 out of 179 with HCC (85.5%).The number of patients who had HCV type 2a represented the second largest group.Only three subjects with NC-CH had HCV type la genotype.Age-specific distributions are presented in Table 2.The prevalence of type lb among ASC patients was 60-78%.A similar result was seen in patients with NC-CH (66-77%).The prevalence of type lb in patients with LC and with HCC were 78-100% , which was higher than among ASC and NC-CH in all age groups.There was no significant difference of the prevalence of type lb by sex in each of the four diagnostic categories.(male v.s.female : 67%, 74% in ASC, 77%, 73% in NC-CH, 80%, 87% in LC and 85%, 88% in HCC).
The odds ratios contrasting NC-CH with ASC, LC with ASC and HCC with ASC with respect to HCV genotype lb were calculated.Subjects with type la were excluded from this analysis because the sequence of HCV type la has poor homology to those of type 2a and 2b, in addition to the fact that those patients comprise a small number among the total number of study subjects.Using univariate analysis, the odds ratios were 1.38 (95%CI=0.93-2.05),2.17 (95%CI=1.16-4.09),and 2.51 (95%CI=1.45-4.35)for development of NC-CH, LC and HCC, respectively (Table 3).The age-and sex-adjusted odds ratios contrasting LC and HCC with ASC controls with respect to HCV genotype Ib was significantly higher than unity at the Table 1

. Demographic factors and distribution of HCV genotypes in asymptomatic carriers (ASC), and patients with chronic hepatitis without cirrhosis (NC-CH), liver cirrhosis (LC) and hepatocellular carcinoma (HCC).
There were no subjects with type 3a.

DISCUSSION
There is increasing evidence in epidemiological studies that show that there are differences in the geographic distribution of HCV genotypes.In the United States, HCV genotype la is most prevalent, followed by HCV genotype lb 26).In France and Italy 27, and Germany 28), HCV genotype lb is the most prevalent, although the proportion of HCV genotype lb is lower in younger people.Genotype lb is the most prevalent in Japan with a prevalence of 70-80%'), and genotype la is rare.In Thailand29), genotype 3a is the most prevalent, followed by type lb.When we compare the geographic distribution of HCV genotypes, it is important to clarify the demographic characteristics and stage of liver disease of subjects, which may substantially relate to the distribution of HCV genotypes.Our study in Japan revealed that genotype lb is the most prevalent in all stages of liver disease among NC-CH, LC, and HCC patients as well as among the ASC, the predominance of genotype lb was seen in all age groups, the second most prevalent genotype among the HCV carriers was type 2a, and the prevalence of genotype la was extremely rare.
The present study showed that the risk of LC and HCC among ASC with type lb was significantly higher as compared to ASC with type 2a or 2b at the same level (OR=2.3),but there was no evidence to show that ASC with type lb have a higher risk of developing NC-CH as compared to ASC with type 2a or 2b.It has been shown that patients with chronic type-C hepatitis with genotype lb had a lower rate of response to interferon therapy than did patients with HCV genotype 2a or 2b 30.31).Kobayashi et al. reported in their follow-up studies that chronic hepatitis patients with HCV genotype lb develop a more severe hepatitis as compared with those with other genotypes 14).Our results did not contradict these reports, suggesting that infection with HCV genotype lb causes liver diseases of greater severity and persistence, which probably induces LC and subsequent HCC more frequently than do the other major HCV genotypes.
However, limitations of the present study are as follows.First, the control group were diagnosed as "asymptomatic" by lower level of transaminases but did not have liver biopsies.Therefore, the fact would not rule out a few cases of undiagnosed liver cirrhosis, which biased the data toward underestimation of significance of genotype lb.Second, we did not take into consideration factors other than age, sex and the genotype, which possibly influence the progression of liver diseases.These other factors include the viral dose 32, 33), the role of HCV quasispecies 34.35), alcohol intake, smoking 38), and the immunoresponse of the host 37.3a).However, we do not have any plausible explanation that would indicate that these factors would be associated with the distribution of HCV genotypes, and appeared as confounding factors.Third, we could not get information of the length of HCV infection which is associated with the increased risk of LC and HCC.In contrast to the results of studies conducted in France and Italy 2n, and Germany 28), our study showed that there was no significant difference in the age-dependent distribution of HCV genotypes.This finding supports the notion that there was no significant difference among epidemics of HCV in the Osaka area on a time-dependent distribution of HCV genotypes, suggesting that the time period after infection with HCV does not differ much among different genotypes of HCV.
In conclusion, our results demonstrate that HCV type lb is predominant both in healthy carriers and patients with chronic liver disease in Osaka, and indicate that type lb infection is more closely associated with the development of LC and HCC through its role in accelerating the progression of chronic liver inflammation to a cirrhotic stage than are other HCV genotypes.These findings would partially explain why the cumulative risk of Japanese HCV carriers of developing HCC is considered to be higher than that in the United States 11.40.41).Further studies are necessary to clarify the pathogenic mechanism by which HCV type lb induces chronic liver diseases and to develop means to stop the progression to LC after infection with HCV.

Table 2 .
Distribution of HCV genotypes by age group.

Table 3 .
The odds ratios (OR) for the development of CH, LC and HCC among ASC with HCV type lb infection against those with type 2a or 2b infection.Adjustment for sex and age .Age was used as a consecutive variable. *