As it would for any chronic disease, an understanding of the natural history of HCV provides a basis for rational decisions about management of patients. Early studies of transfusion-associated hepatitis identified chronicity and histologic progression as prominent features of what was subsequently identified as HCV infection [1]. Before HCV was identified, prospective studies of hepatitis in transfusion recipients found that at least half of these recipients had elevated serum aminotransferase levels that persisted for more than 6 months after the first evidence of hepatic dysfunction, implying that the hepatitis had become chronic. Liver biopsy specimens in many of these affected persons subsequently documented histologically severe injury, including chronic active hepatitis and cirrhosis [1]. There has been some debate, however, about whether such findings actually translate into increased morbidity and mortality for the individual patient [2, 3].

The frequency of HCV infection in patients presenting to liver transplantation programs and the increasingly obvious association between cirrhosis and hepatocellular carcinoma [4] indicate that HCV can indeed be linked to clinically significant disease. The failure of some studies to detect adverse consequences in patients with chronic HCV infection probably reflects the indolent natural history of HCV infection rather than the absence of such complications. A retrospective review of patients with post-transfusion hepatitis documented the development of cirrhosis after a mean of 21 years and of hepatocellular carcinoma after a mean of 29 years [5]. Thus, the longer the follow-up, the more likely that clinically overt liver disease will be recognized in patients with chronic HCV infection. Because the most serious consequences of HCV may not become apparent for many years, it is important to identify those patients infected with HCV who are at greatest risk for progressive liver disease and who may benefit most from early therapeutic intervention with antiviral agents. These persons may include those who consume excessive amounts of alcohol, as has been indicated by studies of patients with decompensated liver disease and HCV infection. Patients with alcoholic cirrhosis have a higher prevalence of HCV markers than do alcoholics with less severe forms of liver damage [6]. Other host factors leading to more severe liver injury may include an older age at the time of initial HCV infection or an immunocompromised state [7]. It is likely, however, that the intrinsic properties of HCV are also important determinants of disease severity.

Additionally, HCV is characterized by a high mutation rate and considerable genomic heterogeneity, factors that may mitigate against effective vaccine development [8]. It is now possible to distinguish several different HCV genotypes by sequencing viral isolates. The hepatitis C virus is a single-stranded RNA virus, divided into a structural and a nonstructural region. The structural region contains genes for the nucleocapsid protein and the envelope glycoproteins. The genes in the nonstructural region code for various functional proteins. In addition, a nucleotide sequence at the 5' portion in the structural region does not appear to have a gene product. This 5' "noncoding" or "untranslated" sequence is the most highly conserved region; the greatest heterogeneity occurs in the envelope proteins. Numerous techniques to genotype HCV isolates have been described; these methods generally use the polymerase chain reaction (PCR) to amplify subgenomic HCV RNA. Genotype identification has been done using type-specific primers for PCR [9], restriction enzyme digestion of the amplified PCR product [10], or type-specific probes to hybridize with HCV RNA after PCR [11]. The nomenclature used to describe the different genotypes has been confusing, leading Simmonds and colleagues to propose a standardized classification system [12]. This new system designates six genotypes numbered 1 through 6. Within these types, "subtypes" designated by lower-case letters identify more closely related isolates. The clinicopathologic correlation of specific genotypes is currently the subject of active investigation. The geographic distribution of genotypes varies substantially [9-14]. In Western Europe and the United States, HCV infection in patients with liver disease and in blood donors is predominantly caused by genotypes 1a, 1b, 2b, and 3a, with some variation in frequency. In Japan and Taiwan, genotypes 1b, 2a, and 2b are seen most frequently; elsewhere in Asia, type 3 is the most common. Type 4 is found frequently in the Middle East.

Several investigators have reported on the significance of specific genotypes as determinants of disease severity, responsiveness to interferon- therapy, and durability of response to interferon after therapy is stopped [15, 16]. The study reported by Nousbaum and colleagues [17] in this issue expands our knowledge about the relation between specific genotypes and disease severity. Nousbaum and colleagues use Okamoto and colleagues' classification system, which recognizes five major genotypes and identifies them using Roman numerals [9]; in this system, type II corresponds to Simmonds and colleagues' type 1b. They provide data on 220 French and Italian patients whose biopsy specimens showed evidence of chronic hepatitis in association with HCV viremia detected by PCR. Serum isolates were typed, and the branched-chain DNA (bDNA) technique of signal amplification was used to quantify serum HCV RNA [18].

In this study, type II (Simmonds and colleagues' type 1b) was the most prevalent HCV genotype overall [17]. This genotype, cirrhosis, and hepatocellular carcinoma were significantly associated. Thus, type II (1b) was found in 75% of patients with cirrhosis and 84% of those with hepatocellular carcinoma, but in only 54% of those with less severe biopsy changes of chronic hepatitis. Type II (1b) was also associated with a longer disease duration and an age greater than 40 years. Thus, 36 of 57 (63%) patients with type II (1b) had had disease for 10 years or more, whereas only 19 of 50 (38%) patients with nontype II (1b) had had known disease activity for more than 10 years. More patients in the nontype II (1b) group had a history of intravenous drug use, which is typically associated with younger age. Therefore, the increased severity of disease in patients with type II (1b) may reflect the cumulative effects of longer disease duration. A multivariate analysis, however, confirmed the association between cirrhosis and type II (1b). This genotype was also associated with a decreased likelihood of success with interferon therapy. Thirty of 41 (73.2%) nonresponders to interferon had the type II (1b) genotype, whereas only 28 of 59 (47.5%) responders had this genotype. Previous studies have suggested that longer disease duration, evidence of cirrhosis shown on biopsy specimens, increasing age, and higher viral load may predict diminished responsiveness to interferon therapy; again, multivariate analysis confirmed that type II (1b) was a predictor of nonresponse. The viral load as quantified by bDNA was higher overall in the nonresponders. Type II (1b) nonresponders had a higher level of viremia than did nonresponders with other genotypes, but this genotype was also a predictor of nonresponsiveness independent of viral load.

Among the important conclusions to be drawn from this article is that HCV genotype II (1b) is associated with increased histologic severity of disease and with nonresponsiveness to interferon therapy. Larger, population-based studies are needed to confirm that HCV genotypes vary in their propensity to produce clinically significant liver disease. Development of less cumbersome methods of HCV genotyping will facilitate such studies [19, 20]. This article also suggests both a changing prevalence of HCV genotypes in the European population and an increased likelihood of severe recurrence of HCV after liver transplantation in patients with genotype II (1b). These observations merit further investigation. In the future, routine evaluation of patients with HCV infection will probably include viral genotyping, which will contribute to the prognostic profile for long-term patient management. For now, recognition of HCV genotypes provides insight into the differing outcomes of HCV infection and responsiveness to interferon therapy.