Approximately 180 million individuals (i.e. 2.5% of the world population) are estimated to have chronic hepatitis C virus (HCV) infection, with the highest prevalence reported in Egypt and the lowest in Finland (1, 2). HCV belongs to the Flaviviridae family and is a small 55 nm virus with a lipid envelope and a single-stranded RNA viral genome with approximately 9600 nucleotides (3, 4). The positive-strand RNA genome includes a 5′-non-coding region with an internal ribosome entry site, an open reading frame that encodes structural (core, envelope 1, 2, p7) and non-structural (NS2, NS3, NS4A, NS4B, NS5A, NS5B) proteins and a 3′-non-coding region. The internal ribosome entry site causes the translation of a polyprotein precursor, which is processed into mature structural and non-structural proteins (5). HCV is mainly transmitted via various parenteral routes such as blood and blood product transfusions, (intravenous) drug abuse, contaminated medical equipment, tattoos, as well as sexually or perinatally. Acute HCV infection usually occurs 2–12 weeks after exposure. Patients with symptomatic acute hepatitis C develop chronic infection in 48–75% of the cases while asymptomatic courses result in chronic infection in most cases (85–90%) (6–8). Chronic hepatitis C is strongly associated with the development of cirrhosis, end-stage liver disease and hepatocellular carcinoma (HCC). Antiviral therapy can prevent these complications (9, 10). The current standard of care includes the administration of PEG-IFN-α/ribavirin combination therapy. In HCV genotype 2- or 3-infected patients, sustained virological response (SVR) rates of ∼80% (11) are achieved while in genotype 1-infected patients, viral eradication is achieved in only 40–50% (12–14).
Approximately 180 million individuals are chronically infected with hepatitis C, which is strongly associated with the development of cirrhosis, end-stage liver disease and hepatocellular carcinoma. Several virological tools (anti-HCV antibody assays, measurement of HCV-RNA, HCV-genotyping) are useful in management of hepatitis C infected patients. The primary goal of antiviral therapy in chronic hepatitis C is a sustained virological response (SVR). The HCV genotype should be determined in every patient considered for antiviral therapy because the currently recommended treatment duration and ribavirin doses differ among HCV genotypes. Exact subtyping might gain increased importance for future therapies with direct-acting antiviral agents (DAA) because of differences of antiviral activities and barriers to resistance among HCV subtypes. Monitoring HCV RNA by a highly sensitive assay (LOD≤15 IU/ml) is the basis for management of response guided therapy of chronic hepatitis C with pegylated IFN plus ribavirin. Rules for early discontinuation of antiviral therapy in non-responders and determination of optimal treatment durations in virologic responders have been developed for application of individualized treatment strategies.
COBAS® Amplicor HCV Monitor 2.0;
hepatitis C virus;
limit of detection;
negative predictive value;
positive predictive value;
reverse transcription-polymerase chain reaction;
rapid virological response;
sustained virological response;
Anti-hepatitis C virus antibody detection
The most common screening test for HCV is an immunoassay [enzyme immunoassay (EIA), microparticle EIA, chemiluminescence immunoassay (CIA)] that detects anti-HCV antibodies. These assays have many advantages for diagnosis: they are easy to use, automation is simple, have a low variability and are inexpensive.
Nowadays, the detection of antibodies directed against various HCV epitopes in plasma or serum is based on the use of third-generation EIAs. These assays detect antibodies to recombinant antigens from core (c22) and non-structural proteins 3 (c33), 4 (c100, c200) and 5. The specificity and sensitivity of third-generation immunoassays in patients with chronic liver disease were found to be >98 and >97% respectively (15–17). The mean time to seroconversion is shortened by 2–3 weeks with third-generation compared with second-generation assays with the detection of HCV-specific antibodies approximately 4–6 weeks after infection (18). Anti-HCV-IgM detection cannot discriminate between acute and chronic hepatitis C because some chronically infected patients produce anti-HCV IgM intermittently and not all patients respond to acute HCV infection by producing anti-HCV IgM. Anti-HCV antibodies may become undetectable via commercial assays in some patients many years after successful treatment.
Detection and quantification of hepatitis C virus RNA
There are a number of different commercial assays approved by the FDA and EMEA for the determination of HCV RNA (19–24).
Cobas® Amplicor HCV version 2.0 (Roche Molecular Diagnostics, Pleasonton, CA, USA) based on a standard reverse transcription-polymerase chain reaction (RT-PCR) is available for the qualitative measurement of HCV RNA. The lowest detection limit is 50 IU/ml whatever the HCV genotype (19).
The versant HCV qualitative assay (Siemens Healthcare Diagnostics, Deerfield, IL, USA) based on transcription-mediated amplification is also a highly sensitive test for a qualitative HCV RNA measurement with the lower detection limit of 5–10 IU/ml whatever the HCV genotype (25).
The versant HCV quantitative Test (Siemens Healthcare Diagnostics) is a quantitative HCV RNA assay based on signal amplification by branched DNA (bDNA). The bDNA assay version 3.0 is standardized for IU, and the assay has been reported to be linear over its entire dynamic range from the lower detection limit of 615 IU/ml to 8 million IU/ml whatever the HCV genotype (22).
The COBAS® Amplicor HCV Monitor 2.0 (CAM, Roche Molecular Diagnostics) is a standard RT-PCR-based assay with a linear detection range of 500–500 000 IU/ml, whatever the HCV genotype. For higher HCV RNA concentrations, predilution of the original sample is required (21).
Currently, two real-time PCR assays are available: the COBAS® Ampliprep/Cobas® TaqMan® assay (CAP/CTM, Roche Molecular Diagnostics) and the real-time HCV assay (also named AccuGene® HCV, Abbott Molecular Inc., Des Plaines, IL, USA). These assays have the advantage of having a broad dynamic range of amplification, thus improving the limits of detection (LOD) to ≤10 IU/ml, and linear quantification up to 107–108 IU/ml (26, 27). Four different modes of results are possible with the Abbott HCV assay: (a) undetectable (below the LOD of the assay), (b) possible detection but below 12 IU/ml, (c) positive above 12 IU/ml with an exact HCV RNA concentration or (d) positive above 8.0 log10 IU/ml (which represents the upper limit of quantification).
The results of the Roche CAP/CTM are reported in four stages: (a) undetectable (below the detection limit), (b) positive but unquantifiable (<15 IU/ml), (c) detectable and quantifiable (>15 IU/ml), reported as an exact HCV RNA concentration in IU/ml above 30–40 IU/ml, or (d) detectable, quantifiable but above the upper limit (>6.9 × 107 IU/ml) (28, 29).
It has also been shown that the results of assays can vary significantly with different HCV genotypes despite IU standardization. Generally, the HCV RNA levels in genotype 1 samples measured with the Roche CAP/CTM are higher than those obtained by the Abbott real-time HCV assay and the Siemens bDNA assay (approximately 0.5 log10 IU/ml), while the HCV RNA concentrations in samples with genotype 4 are lower (26, 29–31).
The HCV genotype can be determined by direct sequencing from subgenomic regions such as core/E1 or NS5B. The most conserved regions of the HCV genome are the 5′-untranslated region and a nearly invariant 98 nucleotide RNA element (X-tail) within the 3′-untranslated region (32, 33). The most variable region of the HCV genome is the hypervariable region of E2 (34, 35). For everyday use, there are few commercial kits that use direct sequence analysis of the 5′-untranslated region (Trugene® 5′NC HCV Genotyping Kit, Siemens Healthcare Diagnostics) or reverse hybridization with genotype-specific probes complementary to the 5′-untranslated and the core region (INNO-LiPA HCV II, Siemens Healthcare Diagnostics) (36–39).
The determination of HCV subtypes has no clinical relevance for the current standard of care with pegylated IFN/ribavirin therapy, while different treatment durations based on viral kinetics are recommended based on different HCV genotypes.
Novel, direct-acting antiviral agents (DAA), also called ‘specifically targeted antiviral therapy for hepatitis C’, are currently under clinical development and are substantially improving the SVR rates in genotype 1 patients (40). Subtype determination may become important in future clinical practice mainly because of resistance profiles for these DAA agents according to HCV genotype and subtype. The correct identification of HCV subtypes 1a and 1b has been reported in >96% of cases for second-generation INNO-LiPA assays (41).
Clinical application of virological tools
Acute hepatitis C
The diagnosis of acute HCV infection can be difficult. Although the most reliable approach is proof of seroconversion to HCV antibodies in a previously seronegative individual (42), this is rarely possible in clinical practice. The absence of detectable HCV antibodies in the acute phase does not exclude acute hepatitis C, because the appearance of antibodies can be delayed in as many as 30% of patients at the onset of symptoms (43). An impaired ability to develop antibodies is especially seen in immunocompromised patients. HCV RNA is typically detected in a seronegative patient, followed by the development of HCV antibodies several days or weeks later. When patients are positive for both anti-HCV antibodies and HCV RNA at the initial presentation, it can be difficult to discriminate between acute and acute exacerbated chronic hepatitis C. In these cases, it may be helpful to monitor liver enzymes and to evaluate the risks of HCV transmission. The value of IgM antibodies in the diagnosis of acute infection is considered to be low, because they are detected in both acute and chronic infection (44). Although measurable HCV RNA serum concentrations emerge within the first days after infection, HCV RNA can fluctuate during acute hepatitis C. Therefore, HCV RNA tests must be performed again several weeks later in all negatively tested patients suspected of acute hepatitis C.
Chronic hepatitis C
The primary goal of antiviral therapy in chronic hepatitis C is a SVR, defined as undetectable serum HCV RNA by a sensitive molecular assay 24 weeks after the end of therapy. Recently, it has been suggested that 12 weeks post-treatment follow-up is as relevant as 24 weeks to determine the SVR in patients with HCV receiving PEG-IFN-α and ribavirin (45).
With the current standard therapy of 24–48 weeks of treatment with PEG-IFN-α and ribavirin, the SVR rates are still unsatisfactory and only reach about 40–50% (12–14). In recent years, treatment regimens have been individualized in an attempt to improve the treatment response with the identification of several viral- and host-related factors that affect response to antiviral therapy. Monitoring HCV RNA was found to be a key parameter in the management of response-guided therapy of chronic hepatitis C with PEG-IFN plus ribavirin.
The treatment recommendations of the current German/Austrian/Swiss guidelines are described below (46).
If HCV RNA has decreased by <2 log10 by week 12 compared with the concentrations before the initiation of therapy, the probability of an SVR is minimal (0–3%) and discontinuation of therapy is recommended (47–49). An absolute HCV RNA concentration of >30 000 IU/ml at week 12 has been suggested (50, 51) as an alternative to the <2 log decline stopping rule (Fig. 1).
Treatment can also be discontinued in patients with detectable HCV RNA (≥50 IU/ml) after 24 weeks of therapy. Once again, the chances of an SVR in these patients are minimal (1–3%) (12, 48, 52).
Because of the higher sensitivity of the current most extensively used real-time PCR-based HCV RNA assays (LOD≤10 IU/ml) and with the extension of therapy to up to 72 weeks, the negative predictive value (NPV) of the 2 log rule and the stopping rule at week 24 based on HCV RNA detectability must be re-evaluated. In the INDIV-2 study, for example, patients with initial HCV RNA negativity by a highly sensitive assay at week 30 were treated for 72 weeks and achieved SVR rates of 50% (53).
Low-dose monotherapy with PEG-IFN-α in patients who fail to respond to a full course of antiviral therapy cannot be recommended. Three independent studies did not show a significant improvement in histological and/or clinical courses in these patients. Whether certain sub-groups (i.e. patients with portal hypertension) can benefit from low-dose PEG-IFN monotherapy remains to be determined (54–56).
Treatment duration for genotype 1(4)-infected patients
Twenty-four weeks of treatment is recommended for genotype 1 patients with a low baseline viral load (VL) (<600 000–800 000 IU/ml) who achieve a rapid virological response (RVR; HCV RNA<50 IU/ml at week 4 of treatment). There was no significant difference in the SVR rates when patients with RVR and low VL were treated for 24 weeks or 48 weeks (57–59) (Fig. 1). There are few data on a reduced treatment duration in patients with advanced fibrosis, steatosis and insulin resistance, and thus shortened treatment should not be considered for these groups (60–62).
Various studies have investigated the benefit of extending treatment in genotype 1-infected slow responders [patients with detectable (≥50 IU/ml) HCV RNA levels at week 12, but undetectable HCV RNA at week 24] (50, 52, 63–66) (Table 1). Analysis of three different European studies showed significantly higher SVR and lower relapse rates in the 72-week compared with the 48-week treatment group (52) (Fig. 1). A trend towards lower relapse rates in slow responders treated for 72 weeks compared with 48 weeks was reported in a recent prospective study, which investigated extending treatment (SUCCESS). However, because of a uniform trend in all studies and the significant improvement found in the meta-analysis of three European studies, 72 weeks of treatment can still be considered in slow responders who tolerate therapy (66).
|Definition of a slow response|
(HCV RNA concentration)
|SVR (relapse) in a slow responder (%)||Treatment discontinuation (%)|
|48 weeks||72 weeks||48 weeks||72 weeks|
|Sánchez-Tapias et al. (52)||291||180||800||>50 IU/ml, week4||28 (53%)||44 (17%)||18||36|
|≥ 2 log10 week12||16||44|
|Berg et al. (50)||455||180||800||>50 IU/ml, week 12||17 (NA)||29 (NA)||24||41|
|>50 IU/ml, week 4 and <50 IU/ml, week 24||NA (64%)||NA (40%)|
|≥2 log10 week 12||33||46|
|Pearlman et al. (63)||361||1.5/kg b.w.||800–1400||≥2 log10 week 12 and <10 IU/ml, week 24||18 (59%)||38 (20%)||14||15|
|Ferenci et al. (59)||289||180||1000–1200||≥ 2log10 week 12||31 (56%)||37 (35%)||9||8|
|Mangia et al. (64)||696||180 or 1.5/kg b.w.||1000–1200||<50 IU/ml, week 12||38 (43%)||64 (15%)||10||13|
|≥2 log10, week 12||0 (100%)||8 (60%)|
|Buti et al. (66)||1428||1.5/kg b.w.||800–1400||≥2 log10, week 12||43 (47%)||48 (33%)||9||23|
Patients who attain a complete early virological response (cEVR; undetectable HCV RNA at week 12 using an assay with a lower limit of quantification cut-off of 50 IU/ml) should be treated for 48 weeks (50, 52, 64, 65) (Fig. 1). Currently, baseline VL is only used to shorten the treatment duration in rapid responders. However, the determination of low and high baseline VL may also be useful in patients with cEVR. Higher SVR rates were reported in patients with high baseline viraemia and cEVR if therapy was prolonged to 60–72 weeks (53, 64).
Treatment duration for genotype 2/3-infected patients
Various studies have investigated the reduction of treatment to 12–16 weeks. Overall, reducing the treatment to <24 weeks increases the number of relapsers. However, many HCV genotype 2/3 patients may be treated for 12–16 weeks if certain preconditions are fulfilled, especially an RVR because only patients with RVR at week 4 had high SVR rates after 16, 14 or even 12 weeks of treatment (67–72) (Table 2). Nevertheless, in the ACCELERATE study, patients who received PEG-IFN-α2a (180 μg/week) plus a fixed ribavirin dose of 800 mg per day achieved significantly lower SVR rates with a shorter treatment (79 vs 85%) (67).
at week 4)
|SVR (relapse) in a rapid responder (%)||SVR (relapse) in a rapid responder and low viraemia|
(<4–8 × 105 IU/ml) (%)
|12–16 weeks||24 weeks||12–16 weeks||24 weeks|
|Dalgard et al. (68)||122||1.5/kg b.w.||800–1400||14||90 (10%)||NA||96 (4%)||NA|
|Mangia et al. (70)||283||1.0/kg b.w.||1000–1200||12||85 (10%)||91 (2%)||NA (7%)||NA|
|von Wagner et al. (69)||153||180||800–1200||16||82 (16%)||80 (3%)||94 (NA)||87 (NA)|
|Yu et al. (71)||150||180||1000–1200||16||100 (0%)||98 (2%)||NA||NA|
|Shiffman et al. (67)||1469||180||800||16||79 (NA)||85 (NA)||NA||NA|
|Dalgard et al. (72)||428||1.5/kg b.w.||800–1400||14||81 (11%)||91 (5%)||83 (17%)||100 (0%)|
In addition to RVR, the specific HCV genotype and the baseline VL are associated with a virological response. HCV genotype 2 patients respond better to pegylated IFN and ribavirin than those infected with genotype 3 (61). Furthermore, patients with a low baseline VL (<400 000 IU/ml) had significantly higher SVR rates than those with high HCV RNA concentrations at baseline (>400 000–800 000 IU/ml) (67, 69). Genotype 2/3-infected patients with a baseline VL≤400 000–800 000 IU/ml and RVR can be considered for a shorter treatment. Ribavirin dosing appears to be an important factor in defining the treatment outcome: shorter treatment durations have mainly been effective in studies with a weight-based ribavirin regimen, whereas studies with a fixed ribavirin dose of 800 mg per day have generally resulted in a significant decrease in SVR with a shorter treatment. However, reducing the duration of treatment is not recommended in patients with advanced liver fibrosis or those with low alanine transaminase values at baseline (67, 70, 74). Genotype 2/3-infected patients who do not achieve RVR showed low SVR rates (45–55%)(67, 72, 75). Whether patients without an RVR should be treated for longer than 24 weeks is based on retrospective studies. These data show that genotype 2/3-infected patients without an RVR who receive 48 weeks of PEG-IFN-α2a plus ribavirin 1000/1200 mg/day have higher SVR rates than those receiving 24 weeks of PEG-IFN-α2a plus ribavirin 800 mg/day (76 vs 67%; relapse: 4 vs 26%) (76). Prospective studies have begun to investigate extending the treatment to 36 or 48 weeks in non-RVR patients (Fig. 2).
Does the use of highly sensitive assays affect recent recommendations for a response-guided therapy?
Recommendations for the duration of treatment and early discontinuation were established using HCV RNA assays with a detection limit of ≤50 IU/ml. Numerous studies have shown that patients with a low baseline VL (<400 000–800 000 IU/ml) and a RVR, defined by undetectable HCV RNA at week 4, are appropriate candidates for shorter treatment regimens (12–16 and 24 weeks in genotype 2/3- and 1-infected patients respectively).
As mentioned above, the currently used real-time PCR-based CAP–CTM test has a detection limit of ≤15 IU/ml. Sarrazin and colleagues re-analysed frozen serum samples with CAP–CTM from patients with chronic hepatitis C enrolled in two large, randomized studies. The RVR rates were highly concordant for the CAM, with a LOD of 50 IU/ml, and the CAP–CTM. Although a significantly smaller number of samples had undetectable HCV RNA with the CAP–CTM, there was no difference in the SVR rates after shorter therapy in patients with an RVR<50 IU/ml, an RVR<15 IU/ml and undetectable RVR (82, 83 and 83% for 24 weeks for genotype 1 and 95, 95 and 94% for 16 weeks genotype 2/3) (77). Treatment regimens can therefore be shortened to 16/24 weeks on the basis of an RVR with HCV RNA concentrations <15 IU/ml by the CAP–CTM.
Genotype 1-infected patients with residual viraemia at week 12 (<15 IU/ml, but detectable by CAP–CTM) have also been shown to have a high relapse rate (55%). This group may benefit from prolonged treatment (72 weeks). Furthermore, low viraemia (between 15 and 50 IU/ml) at week 12 in genotype 1-infected patients was associated with even higher relapse rates (75%) (77).
Is the time point a relevant predictor of sustained virological response?
Up-to-date virological response profiles at weeks 4 and 12 provide a robust framework for predicting SVR in patients with genotype 1 infection. Neumann and colleagues investigated the positive predictive value (PPV) and NPV of an EVR at weeks 2 and 4 during treatment with IFN-α2b or pegylated IFN-α2a in treatment-naïve patients infected with genotype 1. All patients with undetectable HCV RNA at week 2 achieved an SVR (PPV: 100%). Patients with a rapid initial virological response at week 2, defined as an HCV RNA decline >2 log10IU/ml, had a high PPV for SVR of 88–97%. A VL>6 log10 IU/ml at week 2 has a high NPV (82–100%) for achieving an SVR. The combination of VL and viral decline (VD) at week 4 had the best NPV. A VL>5.5 log10 IU/ml and VD<2 log10 IU/ml had a 100% NPV in all treatment arms (4–13% of all patients; specificity: 12–29%) (78).
Despite an improvement in anti-HCV therapy in the last few years, the treatment of chronic hepatitis C is still challenging and must be improved. Many viral-related factors have been evaluated in association with the virological response to PEG-IFN-α and ribavirin-based therapy. It is currently agreed that the algorithms for treatment duration and early discontinuation can be applied based on highly sensitive HCV RNA assays (LOD≤15 IU/ml) (77, 79).
One of the most important patient predictors of a reduced SVR rate is advanced fibrosis and cirrhosis. Other patient-related factors negatively influencing the treatment outcome are ethnicity, male gender, older age, higher body weight, liver steatosis, elevated pretreatment serum γ-glutamyltransferase levels and the recently identified polymorphism upstream from IL-28B (rs12979860) (80–82).
Several independent genome-wide association studies have reported the presence of single nucleotide polymorphisms in the IL28B region to be associated with response to treatment (83, 84). The exact mechanisms underlying this association between IL28B polymorphism and response to treatment are unknown (85).
The gene IL-28B on chromosome 19 codes for IFN-λ-3. The protein product is one of the three members of the recently described type 3 IFN family (86). In Caucasians infected with genotype 1, the CC IL-28B type was associated with improved early viral kinetics and a greater likelihood of RVR (28 vs 5 vs 5% for CC, CT, TT respectively) and cEVR (87 vs 38 vs 28%) and SVR (69 vs 33 vs 37%) (82). Recently, it was shown that the IL28B polymorphism (rs12979860) also determines the treatment response in genotype 2- or 3-infected patients who do not achieve RVR (87).
In the future, DAA such as the protease inhibitors telaprevir and boceprevir in addition to PEG-IFN-α and ribavirin will improve the SVR rates in treatment-naïve genotype 1 patients as well as in genotype 1-infected non-responders and relapsers to standard therapy. Clinical trials with the NS3/4A protease inhibitor telaprevir show that the treatment duration can be shortened to 24 weeks in more than 60% of patients (40). The results from on-treatment HCV RNA measurements must be analysed and response-guided algorithms must be established based on the correlation of baseline VL with RVR and SVR. Telaprevir will be approved for a 24-week course in patients with undetectable HCV RNA at weeks 4 and 12 (extended RVR). For all other patients, 48 weeks of treatment is recommended. Treatment algorithms for boceprevir will be different. After a 4-week lead-in phase of PEG-IFN-α plus ribavirin, boceprevir will be added for either 24 or 48 weeks. Individuals achieving an RVR at week 8 will be treated for a total of 28 weeks, while patients showing a clearance of the virus between weeks 8 and 12 of treatment will continue for a total of 48 weeks.
Conflicts of interest
C. Sarrazin has served as a clinical investigator, consultant and/or member of speakers' bureau for Abbott, Roche and Siemens. S. Zeuzem has served as a clinical investigator, consultant and/or member of speakers' bureau for Abbott, Achillion, Anadys, BMS, Gilead, Merck, Novartis, Pfizer, Roche, Tibotec and Vertex. P. de Leuw has no conflicts to declare.