Article Text

Is the hepatocyte a Trojan horse for hepatitis C virus?
  1. 111B, GI Unit,
  2. Veterans Administration Medical Center,
  3. 4150 Clement Street,
  4. San Francisco, CA 94121, USA
  1. Dr Wright.

Statistics from

See article on page 570

Hepatitis C virus (HCV) is a major public health problem around the world, with a prevalence ranging from 0.5 to 2% in most developed countries. After acute infection, spontaneous remission is a rare event. Although long term consequences of HCV infection are controversial, there is a consensus emerging that HCV infection is a progressive disease resulting in liver failure in some but probably in a minority of infected individuals. Patients who lack biochemical evidence of liver injury may be an important subset with a “more benign” natural history.1

In the rare situation where spontaneous cure occurs after acute infection, aminotransferase activities normalise and HCV RNA becomes undetectable in serum. By contrast, the serological profile shows minor modifications with time. With chronicity, serum HCV RNA remains detectable in the vast majority of individuals. Undetectable HCV RNA in serum in patients with abnormal liver enzymes may be related to low viraemia, be under the limit of detection of virological assays, or be due to a fluctuating course. In anti-HCV positive individuals with normal enzymes, however, the lack of serum HCV RNA has been interpreted as complete cure of the infection. However, the study by Haydonet al in this issue (see page 570) challenges this interpretation. These investigators have documented HCV RNA in liver in the majority of patients with exposure to HCV (confirmed by RIBA 3) whether or not virus was detectable in serum, suggesting that residual HCV infection occurs not only among those individuals with abnormal liver enzymes but also among those considered to have “resolved infection”.

Although a high correlation exists between third generation assays and HCV RNA in serum,2 a small subgroup of patients remains, focused on in this study, which are RIBA 3 positive but HCV RNA negative in serum, regardless of liver enzymes. The question arises whether this represents a past cured infection, mainly in those with normal enzymes, or whether it represents an actual infection with undetectable serum HCV RNA. The authors sought to resolve this question by assessing intrahepatic HCV RNA levels in two different groups of subjects: (i) 12 patients with the above mentioned characteristics, seven of whom had normal aminotransferase activities; and (ii) 86 patients with the classic picture of chronic HCV infection (RIBA 3 positive and detectable serum HCV RNA by reverse transcription polymerase chain reaction (RT-PCR)). The main findings can be summarised as follows: (i) 87.5% of patients without detectable HCV RNA in serum were HCV RNA positive in liver, suggesting that these patients had a true infection; (ii) serum RT-PCR negative and positive groups were similar with regard to demographic, histological and virological data, suggesting that RT-PCR negative patients may indeed have clinically significant liver disease; and (iii) intrahepatic viral levels were not associated with any host or viral factor, including age, sex, duration of disease, mode of HCV acquisition, alcohol consumption, disease severity, and viral genotype.

This study represents a further step in the understanding of the mechanisms of viral induced liver disease and supports the hypothesis that it is immune mediated rather than cytopathic.3Previous studies done in “apparently healthy blood donors” have reported similar results with high rates of liver damage in anti-HCV positive individuals with or without viraemia and normal aminotransferase activities.1 When analysing these data, however, several factors should be clearly defined in order to compare the results. The first is the definition of persistently normal aminotransferase activities. Due to the fluctuating nature of chronic HCV infection, repeat tests for alanine aminotransferase (ALT) activities are needed in order to differentiate those individuals with intermittently raised ALT from those who have truly persistently normal activities. The second issue is RT-PCR variability. Undetectability of serum HCV RNA by RT-PCR may be a false negative result from inadequate specimen handling and storage, type of sample used, incorrect design of amplification primers or serum contamination with interfering substances.4 Indeed, sensitivity of RT-PCR assays has been shown to vary from one centre to another, especially when in-house methods are applied.5 Furthermore, although PCR is a highly sensitive technique, there is a limit of detection below which a very low rate of replication may not be detected. Finally, absence of viraemia may result from time dependent fluctuations in serum HCV RNA levels. The methods used by Haydon et al were meticulously applied state of the art techniques, and HCV RNA was assessed at three different time points in order to eliminate the bias of time dependent variations in viraemia. Moreover, intrahepatic HCV RNA was detected in the majority of patients with histologically confirmed liver damage, suggesting that HCV RNA was not detected in serum as a result of a low viraemic status rather than methodological difficulties.

Despite improvements in the diagnosis of HCV infection, we have little information about factors which determine progression of disease.3 An intriguing question which is under investigation is the relation between genotype, HCV replication and histological severity. Conflicting results exist as to the association between genotype 1 (subtype 1b) infection and severe liver disease, independent of viral load. In some of the studies where an association has been found, patients with HCV genotype 1b had higher levels of viraemia, suggesting that this particular genotype was capable of inducing more severe liver disease through its greater replicative capacity. Indeed, other studies have found an association between genotype 1 and higher HCV RNA levels. Studies relating genotype to HCV RNA levels or assessing the independent effect of both factors must be interpreted with caution however, as some primers and/or probes used to quantify HCV RNA hybridise with different efficiency to different genotypes.6 More recent studies, using appropriate correction factors for the bDNA version 1.0 assays or using different means of quantitation, have not found differences in HCV RNA levels among genotypes.7 Haydon et al, using limiting dilution PCR with primers from 5′NCR provide further support for this observation.

The association between HCV RNA levels and disease severity, independent of genotype, is also controversial. One possible reason is the existence of inaccuracies in viral quantitation. In order to avoid them meticulous attention must be given to processing of serum samples,3 yet methods of processing have varied from study to study. An inherent assumption in all the studies is that HCV RNA quantified in serum reflects HCV RNA in liver, and that HCV RNA detected in a needle biopsy sample is representative of HCV RNA levels in the liver as a whole, both of which have been shown in recent studies.8 Haydon et al support this assumption as serum RT-PCR positive patients had higher levels of intrahepatic HCV RNA than serum negative ones.

Thus it seems that quantitation of HCV RNA in serum is a reasonable reflection of viral quantitation in hepatocytes with non-viraemic anti-HCV positive patients simply representing the lower end of the spectrum of patients with HCV infection—that is, those with very low replication rate below the limit of detection of the PCR. A comparable group of patients are those who respond biochemically and virologically to interferon. Indeed, relapse has been shown to occur, even several years after patients seem to have achieved a treatment induced “cure”, emphasising the concept that negative serum HCV RNA does not necessarily imply complete cure.9 In some of these cases, liver HCV RNA is still present, suggesting that the rate of replication is too low to be detected by current assays. In contrast, negative HCV RNA both in liver and serum has been associated with sustained response to treatment.10

In conclusion, (i) the detection of HCV RNA in liver demonstrates residual HCV infection in patients who have up until now be considered to have resolved infection. These data explain the presence of liver injury in the vast majority of so called “healthy HCV carriers”, and raises serious questions as to whether a true “HCV healthy carrier state” really exists. Only careful prospective natural history studies of patients with confirmed seropositivity for HCV yet persistently normal liver enzymes or undetectable serum HCV RNA will be able to answer these questions. (ii) The virological, epidemiological and histological similarities between viraemic and non-viraemic anti-HCV individuals and the absence of a correlation between intrahepatic HCV RNA levels and disease severity represent additional data that favour an immune mediated mechanism of liver injury in HCV infection. (iii) Antiviral therapy should be considered for patients with HCV related liver disease whether or not they have delectable virus in serum as the majority will have virus in liver and histological evidence of hepatic disease. (iv) Similar precautions and advice regarding transmission should be applied to anti-HCV (RIBA) positive patients whether or not virus can be detected in serum.

See article on page 570


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles