Potential conflict of interest: Advisory Board: Merck; Clinical Research: Gilead Merck.
Predicting the response to the treatment of hepatitis C virus infection†
Article first published online: 26 APR 2012
Copyright © 2012 the American Association for the Study of Liver Diseases
Clinical Liver Disease
Special Issue: Hepatitis C Infection – Treatment Part 2
Volume 1, Issue 2, pages 46–48, April 2012
How to Cite
Thomas, D. L. (2012), Predicting the response to the treatment of hepatitis C virus infection. Clinical Liver Disease, 1: 46–48. doi: 10.1002/cld.11
- Issue published online: 26 APR 2012
- Article first published online: 26 APR 2012
- National Institutes of Health. Grant Number: R01013324
Watch the interview with the authors
Watch the video presentation of this article
As long as the treatment of hepatitis C virus (HCV) infection is expensive and is associated with adverse events, patients and providers will need to know the likelihood of a response so that they can decide whether the benefits outweigh the risks. The factors that determine this likelihood are called pretreatment predictors of response, and they can be classified as viral or host determinants (Table 1). Although there are many clinically relevant responses that occur during and after treatment, the primary endpoint is a sustained virological response (SVR), which is defined as the suppression of HCV RNA below the level of detection at the end of treatment and 6 months afterward. SVR is widely regarded as a virological cure, and unless otherwise specified, it is synonymous with a response in this review. Although many interesting viral and host response predictors have been discovered, this review focuses on those widely available to medical providers. Finally, because treatment response predictors may differ as treatments for HCV differ, this review focuses on US Food and Drug Administration–approved treatments as of September 2011.
|HCV viral load|
|Determinant||SVR (%)||Determinant||SVR (%)|
|Viral load||Viral load|
|<800,000 IU/mL||78||70||≤800,000 IU/mL||85||64|
|≥800,000 IU/mL||74||36||>800,000 IU/mL||63||33|
|≤45 years||83||52||≤40 years||69||52|
|>45-65 years||70||38||>40 years||65||34|
|<25 kg/m2||83||44||≤25 kg/m2||67||47|
|≥25-<30 kg/m2||67||45||25-30 kg/m2||65||33|
|≥30 kg/m2||71||41||>30 kg/m2||66||33|
HCV is the most genetically diverse virus known to cause disease in man, and at least 6 distinct genetic groupings (genotypes) have been identified (a seventh genotype has also recently been provisionally characterized). HCV treatment responses vary with the genotype, with the highest SVR rates observed for genotype 2 and the lowest rates observed for genotypes 1 and 4. For example, when peginterferon and ribavirin were used at similar doses for 48 weeks, SVR was achieved by 52% of genotype 1–infected patients and by 80% of genotype 2/3–infected patients.1 An abbreviated treatment duration and a lower ribavirin dose reduced SVR rates for patients with HCV genotype 1 infections but not for patients with genotype 2 or 3 infections. Thus, subsequent clinical trials largely focused on either genotype 1 infections or genotype 2/3 infections. Additional smaller studies have suggested that similarly to genotype 1, genotype 4 is poorly responsive2; genotype 5 is similar to genotype 35; genotype 3 is less responsive than genotype 24; and similarly to genotype 2, genotype 6 is very responsive.5 With HCV protease inhibitors, higher SVR rates have been detected with genotype 1b versus genotype 1a6 (Table 2).
The amount of HCV detected in the blood before treatment (the baseline HCV viral load) may also determine the likelihood of a response to peginterferon and ribavirin.1, 4 The predictive value of the baseline viral load is lower for genotypes with higher SVR rates and is also lower with more potent treatments for the same genotype. In one study of patients with HCV genotype 3 infections who were treated with peginterferon alfa-2a and 800 mg of ribavirin for 24 weeks, SVR was achieved by 81%, 70%, and 59% when the pretreatment viral loads were ≤400,000, >400,000 to 800,000, and >800,000 IU/mL, respectively; the results for patients with genotype 2 (a more responsive genotype) were 82%, 79%, and 73%, respectively.4 For genotype 1–infected patients, the HCV viral load is less predictive of a response if HCV protease inhibitors are taken. For example, in a study of patients with HCV genotype 1 infections, patients treated with peginterferon, ribavirin, and telaprevir had high SVR rates despite their baseline HCV RNA levels: 78% at <800,000 IU/mL and 74% at ≥800,000 IU/mL. For patients who were randomized to peginterferon, ribavirin, and a placebo, the SVR rates were 70% and 36% in the respective viral load groups.7 In another study of patients with HCV genotype 1 infections who were treated with peginterferon, ribavirin, and boceprevir, the SVR rates were 76% to 85% for patients with HCV RNA levels < 800,000 IU/mL and 61% to 63% for patients with HCV RNA levels ≥800,000 IU/mL; for patients randomized to peginterferon, ribavirin, and a placebo, the SVR rates were 64% and 33% for the respective viral load groups.6
Interestingly, the strongest host determinant of a response to peginterferon and ribavirin is the patient's DNA sequence near the gene for interleukin-28B (IL-28B), or lambda interferon 3. According to one commercially available test (the C versus T allele for rs12979860), the response rates of patients with a chronic HCV genotype 1 infection who were treated with peginterferon and ribavirin were 69%, 33%, and 27% for Caucasians with the CC, CT, and TT genotypes, respectively; for black patients, the SVR rates were 48%, 15%, and 13%, respectively.8 For Caucasians and Asian patients, the T versus G allele at position rs8099917 provides similar information.9 The IL-28B genotype is also predictive of treatment outcomes for patients with non-1 HCV genotypes and for human immunodeficiency virus (HIV)/HCV-coinfected patients.10, 11 However, the predictive value of the genetic test is diminished in persons treated with HCV protease inhibitors. Among Caucasian patients with the CC, CT, and TT genotypes who were taking telaprevir, peginterferon, and ribavirin, SVR was achieved by 90%, 71%, and 73%, respectively (Fig. 1).12 Among Caucasian patients with the CC, CT, and TT genotypes who were taking boceprevir, peginterferon, and ribavirin, SVR was achieved by 80%, 71%, and 59%, respectively (Fig. 2).13
Differences in the responses to peginterferon and ribavirin have been reported in patients from different ethnic groups: Asians have the highest response rates, Caucasians have intermediate response rates, and patients of Latino or African ancestry have the lowest response rates.14, 15 These differences are strongly correlated with differences in the frequencies of favorable IL-28B alleles, and they appear to persist with at least the first generation of HCV protease inhibitors.6, 7, 16
The liver disease stage is also a determinant of the response to peginterferon and ribavirin. SVR was achieved by 81%, 75%, and 62% of genotype 1–infected patients taking peginterferon, ribavirin, and telaprevir who had minimal fibrosis, portal fibrosis, or bridging fibrosis/cirrhosis, respectively.7 Likewise, SVR was achieved by 67% and 41% to 52% of genotype 1–infected patients taking peginterferon, ribavirin, and boceprevir who had minimal to moderate fibrosis or bridging fibrosis/cirrhosis, respectively.6 In comparison with patients without cirrhosis, patients with HCV genotype 2 or 3 infections and cirrhosis had 15% and 21% lower SVR rates, respectively.4
Persons with insulin resistance may also have lower SVR rates than younger individuals without insulin resistance. The predictive value of insulin resistance varies, perhaps to some extent because of correlated factors such as a higher body mass index (BMI), steatosis, male sex, and older age. In a recent meta-analysis of 2732 patients treated with peginterferon and ribavirin, the estimated SVR rates were 20% lower for persons with insulin resistance (as determined by the homoeostasis model assessment).17 A high BMI is related to insulin resistance and typically is also associated with a lower response to peginterferon and ribavirin. In the boceprevir registration trial, a low BMI was associated with SVR in the control arm but not in the boceprevir arms.6 However, in the telaprevir registration trial, a higher SVR was seen with a low BMI in the telaprevir arms but not in the control arms.7
An HIV coinfection diminishes the response to peginterferon and ribavirin for all HCV genotypes (and probably IL-28B host genotypes). The lower SVR rates can be largely explained by the higher HCV viral loads of HIV/HCV-coinfected patients because the SVR rates are similarly high for patients with baseline HCV RNA levels < 800,000 IU/mL.1, 18 Studies are ongoing to assess the response of HCV genotype 1/HIV-coinfected persons to HCV protease inhibitors, peginterferon, and ribavirin.
The response to HCV treatment varies according to both viral and host factors. Although the predictive value of each factor differs in various studies and with different HCV treatments, the HCV genotype, baseline viral load, age, IL-28B genotype, race, liver disease stage, and HIV coinfection status remain potentially important determinants of the likelihood of SVR. By using the results for all these markers, providers can provide patients relatively precise estimates of their odds of being cured with today's standard of care.
- 12Telaprevir substantially improves SVR rates across all IL28b genotypes in the advanced trial. J Hepatol 2011; 54: S1369., , , , , , et al.
- 13IL28b polymorphism predicts virological response in patients with chronic hepatitis C genotype 1 treated with boceprevir (BOC) combination therapy. J Hepatol 2011; 54: S12., , , , , , et al.