Infection with hepatitis C virus (HCV) is a major cause of chronic liver disease, with an estimated 170 million people infected worldwide or ~3% of the world population.[1] Currently there is no prophylactic vaccine. HCV is transmitted mainly through contact with blood or blood products. Acute infection is usually asymptomatic or shows nonspecific mild symptoms, but most patients (70–80%) with HCV infection fail to clear the virus from their body and go on to develop chronic hepatitis.[2] The most significant consequence of chronic infection is the slow evolution of hepatic fibrosis over many years, culminating in cirrhosis and an increased risk of developing hepatocellular carcinoma.[2,3] The rate of development of fibrosis varies substantially among patients and is influenced by a number of demographic and environmental factors. However, these factors account for only a small proportion of the variability[4] and there are no clinical markers or tests that predict the rate of fibrosis progression in an individual subject. Thus, there has been increasing interest in the influence of host genetic factors on the rate of disease progression, and whether a genetic signature can be developed to reliably identify subjects at risk of severe disease.

1. The Virus and Antiviral Therapy

Discovered in 1989,[5] the HCV is a positive-strand RNA virus that belongs to the Flaviviridae family. The 9.6-kilobase HCV genome comprises untranslated regions that flank an uninterrupted open-reading frame encoding a single polyprotein of ~3000 amino acids that is processed into structural and nonstructural subunits by host and viral proteases.[6] HCV is highly heterogeneous and has been classified into six major genotypes (HCV 1–6) that can be further divided into numerous subtypes.[7] The six HCV genotypes show marked differences in geographic distribution. Genotypes 1–3 have a worldwide distribution, while genotype 4 is most common in the Middle East and North Africa; genotypes 5 and 6 are rarely seen outside of South Africa and Southeast Asia, respectively.[1,8]

Although the different genotypes are not considered to differ dramatically in their virulence or pathogenicity, viral genotype does influence responsiveness to antiviral therapy. The principal goal of therapy for chronic HCV infection is permanent viral eradication, termed a sustained virologic response (SVR). However, treatment of HCV is suboptimal and many patients do not tolerate or respond to therapy. Currently, the best available treatment is the combination of pegylated interferon (Peg-IFN) and ribavirin. Combination therapy is expensive and may have debilitating adverse effects, including fatigue, influenza-like illness, neuropsychiatric symptoms, gastrointestinal disturbances, thyroid dysfunction, dermatologic effects, and hematologic abnormalities.[9–11] Adverse effects impact compliance as they may require dose reductions in up to 40% of patients and drug discontinuation in up to 14% of patients.[10]

HCV genotype is the single most important predictor of treatment response. Peg-IFN/ribavirin therapy results in viral eradication in approximately 40–50% of individuals with genotype 1 or 4 infections and 70–80% of those with non-1/4 type infections.[12,13] Other factors that contribute to nonresponse to treatment include advanced fibrosis, high viral load, older age, obesity,[12–16] and possibly host genetic factors.[17]

2. Factors Associated with Increased Rate of Fibrosis Progression

Percutaneous liver biopsy is the gold standard for assessing disease activity and severity in patients with chronic HCV infection. Liver disease is characterized by hepatocellular injury and inflammation that is thought to lead to fibrosis, which is an increase in the extracellular matrix constituents such as collagen.[18] The staging of the liver disease by evaluation of fibrosis is a key component of liver biopsy assessment and the stages are commonly assigned a numerical value to indicate increasing severity (e.g. F0–F4). The stages are determined on the basis of the quantity and location of fibrosis. Chronic HCV infection leads to fibrous expansion of portal tracts that may develop short septa (F1). As the disease progresses, fibrous septa extend to form bridges between portal tracts (F2) and between portal tracts and central veins (F3). With further progression, parenchymal nodules surrounded by fibrosis indicate the development of cirrhosis (F4).[19] Steatosis (fatty liver) is a common histologic feature in chronic HCV infection with around 55% of biopsies showing some steatosis.[20]

Although chronic HCV infection is generally characterized by slowly progressive fibrosis, large cross-sectional studies have shown that the rate at which liver injury progresses varies substantially.[21–23] The rate of fibrosis progression can be estimated using a ratio of the stage of fibrosis to the duration of infection (fibrosis units per year). Approximately one-third of patients with chronic HCV infection may progress to cirrhosis within 20 years, one-third within 50 years and one-third may not progress at all.[21,22] A number of demographic and environmental factors have been identified that influence the rate of fibrosis development. An increased rate of fibrosis progression has been associated with older age at infection, male gender and excessive alcohol consumption,[21–23] co-infection with hepatitis B virus (HBV),[24,25] or immunosuppression related to organ transplantation,[26,27] or co-infection with human immunodeficiency virus (HIV).[26,28,29] Inflammation on biopsy contributes to fibrosis progression.[20] It is also now recognized that obesity-related steatosis[30,31] and insulin resistance[32,33] impact fibrosis progression in chronic HCV infection. While numerous studies have examined the effects of viral related factors (HCV genotype, viral load, viral quasispecies) on fibrosis progression, the weight of evidence indicates that these factors play no role in progression once chronic hepatitis has developed.[34–36]

3. Evidence for Host Genetic Factors Determining Fibrosis Progression

While it is now accepted that the rate of development of fibrosis in HCV infection is influenced by demographic and environmental factors, these account for only a small proportion of the variability. Wright and colleagues[4] examined the influence of age at infection, age at biopsy, duration of infection, gender, viral genotype, alcohol consumption, ethnicity, and mode of acquisition in HIV/HBV-negative HCV-infected patients. Using bivariate and multivariate analyses to construct a regression model to predict rate of fibrosis progression, the model accounted for only 30% of the variability in fibrosis rate. In an earlier study, demographic and environmental factors explained only 17% of fibrosis variability.[21] Thus, there has been increasing interest in the influence of host genetic factors on liver fibrosis.

Fibrogenesis is a complex process (figure 1), mediated by necroinflammation and activated hepatic stellate cells (HSCs) and driven by a number of concurrent pathways involving oxidative stress, inflammation, and steatosis.[36–38] More than 100 reports have examined the relationship between host single nucleotide polymorphisms (SNPs) or other genetic mutations and fibrosis in chronic HCV infection. These have mainly been case-control, candidate gene, allele-association studies where the candidate gene has been selected on the basis of its putative role in fibrogenic pathways and disease pathogenesis. These studies have been extensively reviewed.[39–42] Only genetic mutations that have been examined in at least two independent studies in which fibrosis was assessed by biopsy will be discussed further in this paper (see table I).

Fig. 1
figure 1

Fibrogenesis is a complex process, mediated by necroinflammation and activated hepatic stellate cells (HSC) and driven by a number of concurrent cell types and pathways involving cytokines, chemokines, hormones, oxidative stress, and steatosis.[36–38] Polymorphisms in genes within any of these pathways may potentially influence fibrogenesis. CTGF = connective tissue growth factor; EDN1 = endothelin-1 (also known as ET1); EGF = epidermal growth factor; FGF = fibroblast growth factor; HCV = hepatitis C virus; HFE = hemochromatosis gene; HSC = hepatic stellate cell; IGF = insulin-like growth factor; IL-6 = interleukin-6; MTTP = microsomal triglyceride transfer protein; PDGF = platelet-derived growth factor; RAS = renin-angiotensin system; ROS = reactive oxygen species; TGF-β1 = transforming growth factor-β1; TLR = toll-like receptor; TNFα = tumor necrosis factor-α VEGF = vascular endothelial growth factor.

Full size image
Table I
figure Tab1figure Tab1

Selected candidate gene association studies that have examined the role of host genetic factors on fibrosis progression in patients chronically infected with hepatitis C virus (HCV)

Full size image

3.1 Cytokine and Immune Response Genes

Several studies have focused on polymorphisms in cytokine and chemokine genes as these may influence inflammation and fibrosis.

Tumor necrosis factor-α (TNFα) has a major role as a mediator of the inflammatory response, and may promote apoptosis of inflammatory and fibrogenic cells but may also downregulate collagen synthesis.[75,76] Polymorphisms in the promoter of the TNF gene that influence the level of transcription have been associated with advanced liver disease,[43,44] but other studies have not confirmed these findings.[45–49] Interleukin 10 (IL-10) is an anti-inflammatory cytokine that inhibits activation or function of T cells, macrophages, antigen presenting cells, and HSC.[77,78] The promoter region of the IL10 gene contains three biallelic polymorphisms (−1082G>A, −819C>T, and −592C>A) that produce three different haplotypes with differing effects on IL-10 production: GCC (high), ACC (intermediate), and ATA (low).[79] The low IL-10-producing genotypes have been associated with more[50] or less[51] fibrosis in some studies, but not in others.[46,47,52,53]

Chemokine (C-C motif) receptor 5 (CCR5) is a receptor for proinflammatory chemokines that have key roles in host responses to viruses. A common 32-bp deletion mutation in the CCR5 gene (CCR5Δ32) causes truncation and loss of CCR5 receptors on lymphoid cells of homozygotes.[80] Interest in the Δ32 mutation was greatly increased when it was found to be of paramount importance for protection against HIV infection.[80,81] Studies of CCR5Δ32 in HCV have led to discrepant results. Hellier et al.[54] found that possession of the mutant allele was associated with more advanced fibrosis. However, other studies have found the opposite[55] or no association.[56–61]

The dominant fibrogenic cytokine in hepatic fibrosis is transforming growth factor-β1 (TGF-β1), which contributes to the activation of HSCs and their production of extracellular matrix proteins.[82] Its action is exerted through two complementary pathways, one that stimulates matrix accumulation and the other that reduces matrix degradation.[76] Several polymorphic sites have been described within the TGFB1 gene, including two in the promoter region (−800G>A and −509C>T), one at position +72 (C insertion) in a nontranslated region, and two in the signal sequence at codons 10 (Leu/Pro) and 25 (Arg/Pro).[83] Studies of TGFB1 polymorphisms in HCV have led to discrepant results. One study has reported that individuals with the high TGF-β1-producing genotype (homozygosity for the Arg25 allele) are more likely to have increased fibrosis.[47] Another study has suggested that the presence of proline at this position leads to more rapid fibrosis progression,[62] while a third study found no association between codon 25 genotype and fibrosis.[46]

Chemokine (C-C motif) ligand 2 (CCL2; also known as monocyte chemotactic protein type 1 [MCP1], acts on multiple leukocyte populations to promote recruitment.[84] CCL2 production by HSCs regulates leukocyte trafficking and may have a direct profibrogenic action via HSC chemotaxis.[85] A G>A SNP in the CCL2 gene at position −2518 results in increased production of CCL2.[86] Muhlbauer et al.[63] found that HCV-infected patients carrying the −2518G allele were more likely to have advanced fibrosis and severe inflammation on liver biopsy. Other studies have not been able to replicate this finding.[54,64,65]

3.2 Genes that Influence Steatosis, Oxidative Stress, or Fibrogenic Pathways

Recognition of the role of steatosis in increased fibrosis progression in patients with HCV has led to the investigation of genes that influence its development and severity.

Microsomal triglyceride transfer protein (MTTP) plays a role in the synthesis and secretion of very low-density lipoprotein (VLDL) in the liver. A SNP in the promoter of the MTTP gene (−493G>T) has been associated with lower levels of transcription, resulting in lower levels of MTTP and decreased secretion of triacylglycerol from the liver. Association studies for this polymorphism have produced conflicting results.[55,66]

Heavy iron overload may cause significant liver injury, including progressive fibrosis, cirrhosis, and hepatocellular carcinoma. This is well described in patients with hereditary forms of hemochromatosis, where the most common known genetic factor is homozygosity for the 282 Cys>Tyr (C282Y) variant in the HFE gene.[87,88] Heterozygosity for C282Y or other mutations in the HFE gene (e.g. His63Asp [H63D]) may result in relatively modest cellular iron accumulation[89] that may nevertheless increase the production of reactive oxygen species leading to activation of HSCs.[90] A number of studies have examined the association of mutations in the HFE gene with fibrosis in patients who are chronically infected with HCV, but are not homozygous for C282Y (i.e. nonhemochromatosis). Some studies,[55,67–70,91] but not others,[71–73] have reported an association between HFE mutations and advanced fibrosis. The heterogeneity of clinical penetrance of HFE mutations[88] may contribute to the discrepancies in these studies.

In cardiac and renal fibrosis, TGF-β1 production may be enhanced by angiotensin II, the principal effector molecule of the renin-angiotensin system (RAS). Angiotensin II may also augment the accumulation of extracellular matrix.[92] Functional polymorphisms of genes of the RAS have been described, including −6G>A in the promoter of the AGT gene, encoding the precursor peptide, angiotensinogen.[93] Powell et al.,[47] but not Forrest et al.,[74] found an association between the high-producing angiotensinogen avariant allele AGT-6A and increased fibrosis.

4. Evaluation of Study Design

4.1 Patient Categorization

The interpretation of studies that have examined the role of genetic factors in fibrosis progression in HCV is confounded by the method used to categorize subjects as having slow or rapid disease progression. Most studies have been cross-sectional in design, and the associations were assessed using the stage of fibrosis from a single liver biopsy, without taking into account the duration of HCV infection.[17] Within a single center, the majority of patients with chronic HCV infection have minimal or mild fibrosis, but it may not be valid to assign them to a slow fibrosis progression category if they have a relatively short duration of disease. A stronger, more relevant approach is to use strict criteria to categorize subjects based on both the histologic stage of fibrosis and duration of infection. Richardson et al.[55] categorized patients as ‘slow’ progressors if they had no or minimal fibrosis ≥20 years after acquisition of HCV (figure 2). Patients with rapidly progressive disease had stage 3 or 4 fibrosis on liver biopsy or stage 2 fibrosis ≤10 years after acquisition of HCV. Although this limited the number of subjects in the analysis (149 from an initial cohort of 326 patients), it provided a more robust assessment of identifying high- and low-risk patients based on the rate of disease progression.

Fig. 2
figure 2

Strategy for selection of a training cohort from patients (n = 326) with chronic hepatitis C virus (HCV) infection. Patients with slowly progressive disease had no or minimal fibrosis (stage 0 or 1) on liver biopsy ≥20 years after acquisition of HCV. Patients with rapidly progressive disease had stage 3 or 4 fibrosis on liver biopsy or stage 2 fibrosis ≤10 years after acquisition of HCV (reproduced from Richardson et al.,[55] with permission).

Full size image

4.2 Multiple Gene Testing

For complex diseases, genetic associations are usually of small magnitude with odds ratios of 1.1–1.5, and any single polymorphism accounts for only 1–8% of the overall disease risk in the population.[94] Although the risk attributed to an individual polymorphism is very small, the additive effect of several genetic variants from different loci may account for a greater proportion of the disease risk.[94] However, few studies in chronic HCV infection have examined the combined effect of multiple predisposing alleles.

Richardson et al.[55] analyzed associations between polymorphisms in six genes and more rapidly progressing fibrosis. Individual adjusted odds ratios ranged from 2.1 to 4.5. However, for possession of ≥3, ≥4, or ≥5 pro-fibrotic alleles, the adjusted odds ratios were 9.1, 15.5, and 24.1, respectively. Using results from the logistic regression analysis (that adjusted for potential confounding by gender, age at infection, age at biopsy, viral genotype, and BMI), a predictive equation and a corresponding receiver-operating characteristic (ROC) curve was constructed to assess the ability to predict fibrosis progression. The area under the ROC curve was 0.868 and the predictive equation correctly classified 80% of patients in a validation cohort.[55]

Agnostic genome-wide association studies (GWAS) represent a major advance in the study of complex diseases.[95] It is now possible, using high throughput DNA microarrays, to simultaneously examine variations among thousands of genes. SNP chips to examine 500 000 to 1 million SNPs are now routinely analyzed. Huang and colleagues[96] used a method similar to that described above for categorizing patients with chronic HCV infection. In conjunction with a GWAS of 24 823 gene-centric SNPs and sophisticated statistical analysis, a cirrhosis risk score was developed from a 7-gene signature. The cirrhosis risk score was a better predictor than clinical factors for differentiating high- versus low-risk for cirrhosis, although steatosis, BMI, and insulin resistance were not taken into account. There are advantages and disadvantages with using a gene-centric approach. While this methodology should be more complete with regard to the coverage of genes, many putative causal variants lie outside genes and are in relatively low linkage disequilibrium with genic-tag SNPs.[97]

It may be suggested simplistically that the identification of a gene with an odds ratio of less than 1.5 may not usefully contribute to a genetic test to identify patients at risk of advanced fibrosis. However, the identification of a number of such genes, with additive or synergistic effects, may account for up to 70% of the overall disease risk.[94] Indeed, in the studies by Richardson et al.[55] and Huang et al.[96] that used a combination of SNPs in just 6 or 7 genes, 80% of patients in validation cohorts were correctly classified.

While the functionality of many of the SNPs may be unclear, it has been proposed that data from functional studies cannot substitute for compelling statistical genetic evidence of the existence and location of a susceptibility locus.[95] At present, experimental systems may not be sophisticated enough to fully determine the function of the very many SNPs that have been identified. Indeed, a tagged SNP may just be linked to the true functional polymorphism.

5. Genetic Tests and the Potential for Personalized Therapy

Despite enormous effort, to date studies that have investigated the contribution of host genetic factors to fibrosis progression in chronic HCV have been irreproducible and disappointing. This outcome is not, however, restricted to studies involving HCV.[94,98] As seen with genetic studies for other diseases, small study cohorts and poor study design (including mis-phenotyping and heterogeneous study populations) have resulted in limited meaningful findings.[99]

Investigation of factors contributing to fibrosis progression in chronic HCV is particularly challenging for a number of reasons. The restricted ability to collect incident cases and natural history of a disease that spans >20 years limits prospective long-term longitudinal studies. Because of the nature of the mode of acquisition, for most patients it is not possible to obtain a precise duration of infection. The lack of accurate non-invasive tests to assess disease progression necessitates liver biopsy, which is not without risk. The factors that influence the progression of established cirrhosis are also unknown. Early cirrhosis progresses to advanced cirrhosis in a minority of patients[100] and there is a need to uncover the genetic determinants.

Because of the high cost (both financial and in terms of adverse effects) of antiviral therapy, it has been suggested that a genetic test predicting the likelihood of rapid fibrosis progression could be used to decide which patients should be selected to undergo treatment. Recent improvements in antiviral therapy have resulted in 80% of patients with HCV genotype 2 or 3, and 50% of patients with genotype 1 or 4, achieving a SVR.[12,13] With recent reports that obesity and insulin resistance impair response to antiviral therapy,[14,101,102] the possibility is raised that weight loss prior to treatment will further improve the SVR rate. Is there still a need for a genetic test to aid in selection of patients to treat? Until recently, histology-based strategies were in widespread use, with treatment being determined by the degree of fibrosis and inflammation on presenting liver biopsy. However, an increasing number of health systems no longer require a liver biopsy for access to treatment, and the decision to treat is a more consultative process between patient and physician that balances the prospects of enduring treatment morbidity in the hope of preventing advanced fibrosis. Knowledge of a patient’s risk of developing severe disease will remain an important adjunct to the decision-making process.

The primary goal of treatment for patients with HCV is the eradication of the virus. However, for those patients who are unable to undertake or do not respond to antiviral therapy, the aim is to limit fibrosis progression. There are currently no approved therapeutic options designed to delay or reverse the progression of fibrosis. Many of the genes identified through association studies have provided mechanistic clues that may lead to potential targets for antifibrotic drug development. For example, following the association of the RAS with increased hepatic fibrosis,[47] animal models,[103,104] and, more recently, preliminary human studies,[105,106] have demonstrated that RAS inhibitors can attenuate development of hepatic fibrosis. As newer treatment options become available, with attendant clinical trials to evaluate efficacy, it will be vital that equal numbers of patients with a given risk of fibrosis progression are distributed in both control and treatment arms.[107] Since demographic and environmental factors account for only a small proportion of the variability in fibrosis progres sion,[4] the development of a genetic profile or other test that accurately predicts risk will be central to the establishment of drug efficacy.

Will there be a ‘one size fits all’ genetic signature that can predict rapid fibrosis progression? Racial, ethnic, and geographic variations in polymorphism frequency and function are already well documented for a number of genes[108,109] and it is recognized that study populations for genetic studies should be ethnically homogenous[110] and without residual confounding from population stratification. While viral-related factors apparently do not contribute to the rate of disease progression,[34–36] it is not known whether the same predisposing host genetic factors contribute to rapid fibrosis progression for the different viral genotypes. For example, it is now generally accepted that steatosis increases the rate of fibrosis progression in patients with chronic HCV infection (as well as in a number of other liver diseases).[111] However, for patients infected with HCV genotype 3, steatosis is thought to principally be a viral effect, while obesity related factors are considered to be the predominant influence on development of steatosis for patients infected with genotype 1.[35,111]

Similarly, with such a dramatic difference between males and females in the rate of fibrosis progression (independent of alcohol consumption), is it possible that genetic predisposition in causal pathways may differ? It must also be established whether histologic[112] and demographic information can increase the accuracy of a genetic signature.

6. The Way Forward

Successful determination of genetic signatures for fibrosis progression in patients with chronic HCV infection will require multicenter collaborations using GWAS, with large, phenotypically well-defined sample sets. Large sample sets (n > 1000 for the smallest subgroup) are needed to detect even moderate effects in order to eliminate the possibility of false-positive associations.[99] The approach will also need to be multidisciplinary, as the skill sets required are now beyond the scope of an individual specialist physician, scientist, geneticist, or epidemiologist.

As GWAS technology advances, it is becoming more affordable, but currently remains extremely expensive, and it will be necessary to convince funding agencies that the significant financial commitment required will provide ‘value for money.’ In this regard it will be prudent to take advantage of experience gained in other disease settings when designing and analyzing these studies.[99]

Although the strategy shift from hypothesis-driven to hypothesis-generating research[113] may not appeal to all, the outcome offers the potential for personalized therapy and better patient management.