Intrahepatic interferon-stimulated gene responses: Can they predict treatment responses in chronic hepatitis C infection?


  • Nicholas A. Shackel,

    1. A.W. Morrow Gastroenterology and Liver Centre, Centenary Institute, Royal Prince Alfred Hospital, Sydney, Australia
    2. The University of Sydney, Sydney, Australia
    Search for more papers by this author
  • Geoffrey W. McCaughan

    Corresponding author
    1. A.W. Morrow Gastroenterology and Liver Centre, Centenary Institute, Royal Prince Alfred Hospital, Sydney, Australia
    • Centenary Institute, Locked Bag No. 6, Newtown, NSW, 2042, Australia===

    Search for more papers by this author
    • fax: 61-2-9565-6101

  • Potential conflict of interest: Nothing to report.

  • See Article on Page 1548

Clearance of hepatitis C virus (HCV) infection either spontaneously or following interferon (IFN) and ribavirin treatment is characterized by specific innate and subsequent adaptive immune responses.1 Effective treatment of chronic infection requires 24-48 weeks of therapy with an overall sustained virological response rate (SVR) of 50%. Although many factors such as viral genotype, pretreatment viral load, alcohol use, insulin resistance, and ethnicity are associated with variable treatment responses, the molecular mechanisms underlying viral clearance following therapy are poorly characterized.


HCV, hepatitis C virus; IFN, interferon; ISG, interferon-stimulated gene; PBMC, peripheral blood mononuclear cell; RR, rapid virological response; SR, slow virological response.

It could be argued that molecular markers used on an individual basis might improve prediction of treatment responses prior to commencement of therapy. Previously, Chen and colleagues examined liver biopsies prior to α-IFN and ribavirin therapy from 16 responders, 15 nonresponders, and 20 normal individuals by gene array analysis and determined that 18 genes were predictive of an SVR.2 These investigators identified a gene expression signature of 8 genes that could predict the likelihood of achieving an SVR in 30 of 31 individuals (GIP2/IFI15/ISG15, ATF5, IFIT1, MX1, USP18/UBP43, DUSP1, CEB1, and RPS28).2 The striking outcome from this study was that these genes, known to be involved in interferon responsiveness, were overexpressed in nonresponders and formed part of the predictive gene signature profile. Additionally, 2 genes (ISG15/IFI15 and USP18/UBP43) were identified as part of a previously unrecognized novel IFN regulatory pathway.2, 3 In a further study of peripheral blood mononuclear cells (PBMCs), a relative lack of interferon-stimulated gene (ISG) expression was associated with a poor response to antiviral therapy with pegylated IFN.4

In an article in this issue of HEPATOLOGY, Feld and colleagues present work demonstrating unique patterns of liver gene expression that correlate with IFN and ribavirin treatment responses in HCV genotype 1 infection.5 The authors used gene arrays to study 11 individuals in whom 6 had an early rapid virological response (RR) at 4 weeks of treatment, defined as >2 × log10 decrease from pretreatment levels of HCV viral load. These individuals were compared to 5 who had a slow virological response (SR) and 19 individuals with chronic HCV genotype 1 infection who were treatment-naïve. Their results are intriguing. Individuals with an RR were more likely to have a significant induction of IFN-stimulated genes after 72-96 hours of pegylated IFN compared to SR or treatment-naïve individuals. Further, the expression of genes known to inhibit the IFN response was increased in SR compared to individuals with an RR. Additionally, ribavirin appeared to diminish gene expression that could inhibit IFN responsiveness. Importantly, these investigators confirmed key findings by real-time reverse-transcription polymerase chain reaction. Therefore, these authors have identified molecular pathways associated with IFN responsiveness within the liver associated with HCV infection.

However, there are limitations to this study. Feld and colleagues were not able to demonstrate a significant difference in IFN-inducible gene expression in individuals with RR compared to SR. To demonstrate a difference, the investigators had to examine the changes relative to a separate group of treatment-naïve individuals who went on to have an RR or SR. Hence, this study did not profile serial samples from the same individuals before and after treatment. Further, gene expression in treatment-naïve individuals showed that those individuals with a future RR had low levels of IFN-responsive gene expression compared to individuals with an SR. This is consistent with the study of Chen et al. in which pretreatment levels of IFN-responsive gene expression were greater in individuals who subsequently responded poorly to treatment.2

Therefore, to determine if an individual is likely to respond to exogenous IFN, 2 key questions need to be answered: (1) Is the endogenous intrahepatic IFN-associated gene response already significantly induced prior to exogenous IFN administration? If the IFN response is already significantly activated, it then appears that exogenous IFN is likely to be less successful in achieving further activation of this innate immune system effector arm and optimal viral suppression (Fig. 1). (2) Is there a significant induction of gene expression associated with an IFN treatment response? To achieve suppression of viral replication and achieve a treatment response following IFN administration, there needs to be a significant induction of ISG expression. This means that exogenous IFN appears to only be effective in enhancing the innate immune response if there is a significant increase in ISG expression and that this requires an immune response that is not already near maximally stimulated.

Figure 1.

Interferon stimulated gene (ISG) responses in HCV infection. (A) Individuals who have an SVR seem to have relatively low levels of ISG expression before therapy. The ISG expression is then significantly increased with interferon therapy and this is associated with viral clearance. (B) Individuals who do not respond to interferon seem to have relatively higher ISG expression before therapy but fail to mount a further significant induction following treatment and this is associated with viral persistence.

The ribavirin-associated gene expression demonstrated by Feld et al. is novel and may for the first time demonstrate a previously unrecognized mode of action of this antiviral agent. It has, for a long time, been established that ribavirin alone is an ineffective anti-HCV agent but in combination with IFN is responsible for enhancing the likelihood of achieving an SVR. Feld and colleagues demonstrate that ribavirin was associated with suppression of a set of genes responsible for an inhibition of IFN responsiveness. This mechanism may explain the means by which ribavirin helps to achieve an SVR. However, this effect was seen in combination therapy with ribavirin for 72 hours and 24 hours of pegylated IFN and it is unclear if this represents a synergistic interaction with exogenous interferon or an effect solely due to ribavirin. Moreover, an increase in RR and further suppression of viral load has not been demonstrated with ribavirin. Therefore, the exact mechanism of ribavirin action still remains unclear but is likely to be multifactorial.

Although the results of Feld et al. and others are intriguing, a number of fundamental questions remain unanswered: Does the IFN response gene expression reflect leukocyte-mediated clearance of infected hepatocytes? Do hepatocytes express IFN response genes? The answer to both questions is yes, but the balance between these 2 pathways of viral clearance is unknown. Liver leukocytes express IFN and it is well established that intrahepatic natural killer, CD8, and CD4 activity is enhanced by IFN.1, 6, 7 Hepatocytes are directly IFN responsive and express a number of genes such as the chemokine (C-X-C motif) ligand (CXCL)-10, interferon-inducible protein p78 (MxA), and protein kinase RNA.8–10 Further, hepatocyte expression of MxA prior to IFN therapy is predictive of a nonresponse to treatment.10 Additionally, IFN can directly suppress HCV replication within hepatocytes.11 Therefore, there appears to be a number of synergistic actions of IFN in HCV infection including an enhanced clearance of infected hepatocytes, inhibition of viral replication within infected hepatocytes, as well as active recruitment of leukocytes to infected hepatocytes. To what extent exogenous IFN alters these synergistic actions is unknown.

Finally, it is unclear why individuals with chronic HCV infection who already have a high level of ISG expression do not achieve viral clearance. Perhaps these results are indicative of a high level of ISG expression in leukocytes but not in hepatocytes where gene expression is inhibited by the presence of HCV.12 Furthermore, the factors that are responsible for the comparatively greater ISG expression in certain individuals compared to others with chronic HCV infection are unknown. These apparent contradictions raise many questions, which (like most questions in hepatitis C pathogenesis) are likely to have complex answers.13–15

The study by Feld et al. highlights the important role of ISG expression in determining IFN treatment responses in HCV infection. The future promises optimization of IFN responses, identification of likely responders, and the development of individualized tailored treatment regimes. The age of individualized and personalized treatment therapies in HCV using molecular host responses may rapidly be approaching. 1

Table 1. Summary of ISG Responses in HCV Infection
Feld et al.5LiverGene expression signature containing multiple ISG predictive of SR and RR following IFN treatment
Taylor et al.4PBMCLack of response to IFN therapy associated with blunted ISG induction in PBMCs
Chen et al.2LiverGene expression signature containing multiple ISG predictive of SVR following IFN treatment
Asahina et al.16PMBCHigher levels of IFN-induced PKR and MxA expression associated with chronic HCV eradication
Lanford et al.20Liver and PBMCHigh level of ISG expression in chronic HCV infection and lack of further induction following IFN. In PMBCs, ISG induction equivalent to uninfected. (Chimpanzee study)
Bigger et al.17LiverOngoing ISG expression in chronic HCV infection (Chimpanzee study)
Bigger et al.18LiverBiphasic ISG expression associated with acute HCV clearance (Chimpanzee study)
He et al.19PBMCISG expression predictive of racial differences in IFN responsiveness in chronic HCV infection