Role of interleukin-28B polymorphisms in the treatment of hepatitis C virus genotype 2 infection in Asian patients


  • Potential conflict of interest: Nothing to report.


Genome-wide association studies have linked single nucleotide polymorphisms (SNPs) near the interleukin-28B gene to the hepatitis C virus genotype 1 (HCV-1) response to peginterferon/ribavirin treatment. We aimed to explore the impact on the treatment outcomes of Asian HCV-2 patients. We determined rs8105790, rs8099917, rs4803219, and rs10853728 to be candidate SNPs in 482 Asian HCV-2 patients treated with the standard of care. Because the first three SNPs were in very strong linkage disequilibrium with one another (r2 = 0.94-0.96), rs8099917 and rs10853728 were selected for an analysis of their influence on the achievement of rapid virological response [RVR; seronegativity for hepatitis C virus (HCV) RNA in treatment week 4] and sustained virological response (SVR; seronegativity for HCV RNA throughout 24 weeks of posttreatment follow-up). The rs10853728 genotype did not predict RVR or SVR in HCV-2 patients. However, patients with the rs8099917 TT genotype, in comparison with patients with GT/GG genotypes, had a significantly higher rate of achieving RVR (85.2% versus 72.0%, P = 0.017) but did have not a significantly higher rate of achieving SVR (89.4% versus 86.0%). Multivariate analysis revealed that a baseline HCV viral load <400,000 IU/mL was the strongest predictor of RVR [odds ratio (OR) = 4.27, 95% confidence interval (CI) = 2.31-7.87, P < 0.001], and this was followed by advanced liver fibrosis (OR = 0.28, 95% CI = 0.15-0.53, P < 0.001), the carriage of the rs8099917 TT genotype (OR = 3.10, 95% CI = 1.34-7.21, P = 0.008), and the pretreatment level of aspartate aminotransferase (OR = 0.996, 95% CI = 0.99-1.00, P = 0.04). Nevertheless, the achievement of RVR was the single predictor of SVR with an OR of 19.37 (95% CI = 8.89-42.23, P < 0.001), whereas the rs8099917 genotypes played no role in achieving SVR with or without RVR. Conclusion: The rs8099917 TT genotype is significantly independently predictive of RVR, which is the single best predictor of SVR, in Asian HCV-2 patients. (Hepatology 2011)

Peginterferon/ribavirin combination therapy is recommended for patients with hepatitis C virus (HCV) infection. Among the pretreatment virological variables, the presence of a hepatitis C virus genotype 2 (HCV-2) or HCV-3 infection is the most powerful predictor of a sustained virological response (SVR).1 Patients infected with HCV-2/HCV-3 have significantly better virological responses after antiviral therapy in comparison with patients infected with HCV-1/HCV-4. Although a rapid virological response (RVR) rate of 62% to 87% and an SVR rate of 80% to 93% can be achieved after 24 weeks of peginterferon/ribavirin combination therapy in patients with HCV-2/HCV-3 infection,2-7 up to 30% and 20% of patients still fail to attain RVR and SVR, respectively, with the current standard-of-care regimen.8-10 There clearly exists a genotype-specific difference in the viral kinetics.11 Beyond the virological elements, the diversity of host genetic factors among different races partially accounts for variations in treatment responses. Studies based on genome-wide association studies have shown that single nucleotide polymorphisms (SNPs) at and/or near the interleukin-28B (IL-28B) gene, which encodes interferon-λ, play a critical role in the treatment of HCV-1 infection.12-15 The linkage of the genetic variants to the on-treatment and posttreatment responses of HCV-2 patients has not been well investigated; there have been discordant results regarding its role in treatment outcomes in recent studies using Caucasian populations.14, 16 In the current study, therefore, we aimed to elucidate the role of IL-28B polymorphisms in the treatment response with respect to viral kinetics in a large Chinese cohort residing in Taiwan with HCV-2 infection.


ALT, alanine aminotransferase; AST, aspartate aminotransferase; CI, confidence interval; EOTVR, end-of-treatment virological response; EVR,early virological response; HCV, hepatitis C virus; HCV-#, hepatitis C virus genotype #; IL-28B, interleukin-28B; OR, odds ratio; PCR, polymerase chain reaction; RVR, rapid virological response; SD, standard deviation; SNP, single nucleotide polymorphism; SVR, sustained virological response.

Patients and Methods

Selection of Patients.

We retrospectively recruited 497 consecutive therapy-naive patients from Taiwan with chronic hepatitis C and HCV-2 infection who underwent the current standard-of-care regimens.8-10 Patients received peginterferon alfa-2a (180 μg/week) or peginterferon alfa-2b (1.5 μg/kg/week) subcutaneously and oral ribavirin according to body weight (<60 kg, 800 mg/day; 60-75 kg, 1000 mg/day; and >75 kg, 1200 mg/day). Eligible subjects were seropositive for HCV antibodies (third-generation enzyme immunoassay; Abbott Laboratories, North Chicago, IL) and for HCV RNA according to real-time polymerase chain reaction (PCR) for more than 6 months. The exclusion criteria were as follows: human immunodeficiency virus infection, autoimmune hepatitis, primary biliary cirrhosis, sclerosing cholangitis, Wilson's disease, alpha-1-antitrypsin deficiency, decompensated cirrhosis, overt hepatic failure, a current or past history of alcohol abuse (≥20 g daily), a psychiatric condition, previous liver transplantation, and evidence of hepatocellular carcinoma. Serum levels of HCV RNA at the baseline, in treatment weeks 4 and 12, at the end of treatment, and 24 weeks after therapy were determined by qualitative PCR. Serum levels of HCV RNA at the baseline and in week 12 were measured with a branched DNA assay (Versant HCV RNA 3.0, Bayer, Tarrytown, NJ; quantification limit = 615 IU/mL) if qualitative HCV RNA seropositivity was found. HCV genotypes were determined by the method described by Okamoto et al.17 The study was approved by the ethics committees at the participating hospitals and was carried out according to the guidelines of the International Conference on Harmonization for Good Clinical Practice. All patients gave written informed consent before enrollment. Four hundred eighty-two patients (97%) who continued treatment for at least 80% of the assigned duration were included in the analysis; 349 of the 482 patients (72.4%) had undergone liver biopsy within 1 year of antiviral therapy and had available histological data, which were graded and staged according to the scoring system described by Knodell and Scheuer.18 The distributions of IL-28B genotypes were not different between the patients who were included in the analysis and those who were excluded from the analysis because they continued treatment for less than 80% of the assigned duration (Supporting Information Table 1).

Table 1. Basic Demographic, Virological, and Clinical Features of the Patients
  • *

    Data were available for 349 patients.

Age (years), mean (SD)52.8 (11.1)
Male sex, n (%)264 (54.8)
Body weight (kg), mean (SD)65.7 (10.9)
Baseline HCV RNA (log IU/mL), mean (SD)5.35 (0.92)
Baseline HCV RNA > 400,000 IU/mL, n (%)182 (37.8)
AST (IU/L), mean (SD)106 (62)
ALT (IU/L), mean (SD)162 (102)
Creatinine clearance rate (mL/minute), mean (SD)91.8 (28.5)
Fasting blood sugar (mg/dL), mean (SD)105.7 (41.8)
rs8099917 TT/GT + GG, n (%)432/50 (89.6/10.4)
rs10853728 CC/CG + GG, n (%)318/164 (66.0/34.0)
Fibrosis score of 3-4, n (%)*99 (28.4)

Assessment of Efficacy.

The endpoint of the study was the achievement of SVR, which was defined as seronegativity for HCV RNA throughout the 24 weeks of the posttreatment follow-up period. RVR was defined as seronegativity for HCV RNA at 4 weeks of therapy. Early virological response (EVR) was defined as seronegativity or at least a 2-log10 decrease from the baseline for serum HCV RNA at 12 weeks of treatment. Complete EVR was defined as PCR positivity for HCV RNA in week 4 but PCR negativity in week 12 of treatment. End-of-treatment virological response (EOTVR) was defined as seronegativity for HCV RNA at the end of treatment. Relapse was defined as the reappearance of HCV RNA during the follow-up period in patients who achieved EOTVR.

IL-28B Genotyping.

Previous genome-wide association studies have indicated that SNPs rs12979860 and rs8099917 are related to treatment outcomes for Caucasians and African Americans with HCV-1 infection.13-16 On the other hand, rs12980275, rs11881222, rs7248668, rs8105790, rs8099917, rs4803219, and rs10853728 have been noted to be associated with antiviral treatment responses on the basis of genome-wide association studies and a replication study in patients of Asian ethnicity.12 Because the minor allele frequencies of rs12980275, rs11881222, and rs7248668 are all less than 1% in Taiwanese,19 rs8105790, rs8099917, rs4803219, and rs10853728 were selected as candidate SNPs in the present study. The genotypes of the patients were determined with the ABI TaqMan SNP genotyping assays (Applied Biosystems, Foster City, CA) and with predesigned commercial genotyping assays (ABI assay C__11710096_10). Briefly, PCR primers and two allelic-specific probes were designed to detect a specific SNP target. The PCR reactions were performed in 96-well microplates with ABI 7500 real-time PCR (Applied Biosystems). Allele discrimination was achieved by the detection of fluorescence with System SDS 1.2.3. In the initial analysis, rs8105790, rs8099917, and rs4803219 were noted to be in very strong linkage disequilibrium with one another (r2 = 0.94-0.96). Therefore, rs8099917 and rs10853728 were selected for the final analysis with respect to the other variables in the current study (Fig. 1 and Supporting Information Table 2).

Figure 1.

Linkage disequilibrium plot of the candidate SNPs. Pairwise r2 was used for analyzing the correlation.

Table 2. Univariate Analysis of Factors Associated With RVR and SVR
VariableRVR-Positive (n = 404)RVR-Negative (n = 78)P ValueSVR-Positive (n = 429)SVR-Negative (n = 53)P Value
  • *

    Data were available for 349 patients.

rs8099917 TT/GT + GG, n (%)368/36 (91.1/8.9)64/14 (82.1/17.9)0.02386/43 (90.0/10.0)46/7 (86.8/13.2)0.50
rs10853728 CC/CG + GG, n (%)262/142 (64.9/35.1)56/22 (71.8/28.2)0.24281/148 (65.5/34.5)37/16 (69.8/30.2)0.53
Male sex, n (%)226 (55.9)38 (48.7)0.25242 (56.4)22 (40.7)0.03
Age (years), mean (SD)52.4 (10.9)55.0 (11.5)0.0652.4 (11.1)56.3 (10.5)0.01
Body weight (kg), mean (SD)65.5 (11.0)66.8 (10.7)0.3465.7 (10.8)66.2 (11.7)0.74
Baseline HCV RNA (log IU/mL), mean (SD)5.32 (0.93)5.50 (0.86)0.115.35 (0.92)5.35 (0.91)0.99
Baseline HCV RNA > 400,000 IU/mL, n (%)131 (32.5)51 (65.4)<0.001152 (35.5)30 (55.6)0.004
AST (IU/L), mean (SD)102 (58)126 (73)0.002104 (62)123 (58)0.04
ALT (IU/L), mean (SD)158 (98)185 (118)0.03160 (100)178 (119)0.23
Fasting blood sugar (mg/dL), mean (SD)104.2 (40.7)113.3 (46.3)0.11104.6 (41.1)114.3 (45.8)0.11
RVR-positive, n (%)382 (89.0)22 (41.5)<0.001
EVR-positive, n (%)423 (98.6)43 (79.6)<0.001
Fibrosis score of 3-4, n (%)*66 (23.2)33 (51.6)<0.00183 (26.8)16 (41.0)0.06

Statistical Analyses.

The Hardy-Weinberg disequilibrium test was performed for each SNP. The linkage disequilibrium index (Lewontin's D′ and r2) was calculated with Haploview 4.2.20 The frequencies were compared between groups with the χ2 test with the Yates correction or Fisher's exact test. Group means, presented as means and standard deviations (SDs), were compared with analysis of variance and the Student t test. Serum HCV RNA levels were expressed after the logarithmic transformation of the original values. Creatinine clearance was estimated with the Cockcroft-Gault equation, which includes the sex, age, body weight, and serum creatinine level as values. The frequencies of the rare alleles of rs8099917 and rs10853728 genotypes were too low, and we combined the rare homozygote and heterozygote together when we analyzed these two SNPs. To assess the relative contributions of predictors of RVR and SVR, we applied stepwise logistic regression analysis and used age, sex, baseline HCV RNA levels, the degree of liver fibrosis, IL-28B genotypes, and pretreatment aminotransferase levels as covariants. The statistical analyses were performed with the SPSS 12.0 statistical package (SPSS, Chicago, IL). All statistical analyses were based on two-sided hypothesis tests with a significance level of P < 0.05.


Patient Profiles and Virological Responses.

The basic demographic, virological, and clinical features of the patients are shown in Table 1. The rates of RVR, EVR, EOTVR, SVR, and relapse were 83.8%, 96.7%, 96.9%, 89.0%, and 8.1%, respectively.

Factors Associated With RVR and SVR.

In the univariate analysis, the genotypes of rs10853728 were not associated with RVR or SVR (Table 2). The TT genotype of rs8099917, low baseline HCV RNA levels (<400,000 IU/mL), low pretreatment levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT), and less advanced liver fibrosis were significantly associated with a higher RVR rate. The stepwise logistic regression analysis revealed that a baseline HCV viral load <400,000 IU/mL was the strongest predictor of RVR, and this was followed by less advanced liver fibrosis, the carriage of the rs8099917 TT genotype, and a lower pretreatment level of AST (Table 3). Younger patients, male sex, baseline HCV RNA levels <400,000 IU/mL, low pretreatment levels of AST, and RVR and EVR achievement were factors predictive of SVR in the univariate analysis (Table 2). According to the multivariate analysis, RVR achievement was the single predictor of SVR (Table 3), whereas neither rs8099917 nor rs10853728 offered significant predictive value for SVR in HCV-2 patients.

Table 3. Logistic Regression Analysis of Factors Associated With RVR and SVR
VariableOR95% CIP Value
  • *

    Data were available for 349 patients.

 Baseline HCV RNA levels
  >400,000 IU/mL1  
  <400,000 IU/mL4.272.31-7.87<0.001
Liver fibrosis score (F0-F2 = 0, F3-F4 = 1)*0.280.15-0.53<0.001
  GT/GG genotype1  
  TT genotype3.101.34-7.210.008
 Pretreatment AST level (per 1 IU/L increase)0.9960.99-1.000.04
 Achievement of RVR
  Yes19.378.89-42.23P < 0.001

Characteristics of Patients With SNP rs8099917 and Its Impact on Treatment Responses.

The basic demographic, virological, and clinical features were similar between patients with the major homozygote (TT) or GT/GG genotypes of rs8099917, except that those with the TT genotype had significantly lower baseline levels of HCV RNA (5.32 ± 0.94 versus 5.59 ± 0.66 log IU/mL, P = 0.02; Table 4).

Table 4. Basic Demographic, Virological, and Clinical Features of the Patients With Different rs8099917 Genotypes
Variablers8099917 TT Genotype (n = 432)rs8099917 GT/GG Genotype (n = 50)P Value
  • *

    Data were available for 349 patients.

Age (years), mean (SD)52.6 (11.3)54.6 (9.0)0.23
Male sex, n (%)233 (53.9)31 (62.0)0.28
Body weight (kg), mean (SD)65.5 (10.7)67.7 (12.6)0.19
Baseline HCV RNA (log IU/mL), mean (SD)5.32 (0.94)5.59 (0.66)0.02
Baseline HCV RNA >400 kIU/mL, n (%)162 (37.5)20 (40.0)0.74
AST (IU/L), mean (SD)107 (62)97 (54)0.28
ALT (IU/L), mean (SD)165 (104)144 (83)0.17
Creatinine clearance rate (mL/minute), mean (SD)91.8 (29.1)91.5 (23.0)0.94
Fasting blood sugar (mg/dL), mean (SD)105 (41)114 (45)0.13
Fibrosis score of 3-4, n (%)*86 (27.3)13 (38.2)0.18

Patients with the homozygous TT genotype had a significantly higher rate of RVR than G allele carriers (GT/GG; 85.2% versus 72.0%, P = 0.017). The other treatment responses, including the rates of EVR (99.1% versus 98.0%, P = 0.48), EOTVR (97.2% versus 94.0%, P = 0.2), SVR (89.4% versus 86.0%, P = 0.47), and relapse (8.1% versus 8.5%, P = 0.78), were not different between patients with the TT or GT/GG genotypes. Between-groups analysis by stratification of RVR achievement demonstrated that the rates of EOTVR, relapse, and SVR were similar between patients with the TT or GT/GG genotypes of rs8099917, regardless of RVR achievement (Table 5). Further analysis by stepwise logistic regression revealed that factors associated with SVR in patients with RVR were HCV RNA levels < 400,000 IU/mL [odds ratio (OR) = 2.91, 95% confidence interval (CI) = 1.18-7.19, P = 0.02] and age (OR = 0.94, 95% CI = 0.90-0.99, P = 0.01), whereas the achievement of complete EVR was the sole factor predictive of SVR in non-RVR patients (OR = ∞, 95% CI = 0.00-∞, P < 0.0001).

Table 5. Impact of rs8099917 on the On-Treatment Response in HCV-2 Patients With or Without RVR
Variablers8099917 TT Genotypers8099917 GT + GG GenotypeP Value
 EOTVR, n/N (%)367/368 (99.7)36/36 (100.0)1.00
 SVR, n/N (%)348/368 (94.6)34/36 (94.4)1.00
 Relapse, n/N (%)19/367 (5.2)2/36 (5.6)1.00
 EVR, n/N (%)57/64 (89.1)13/14 (92.9)1.00
 EOTVR, n/N (%)53/64 (82.8)11/14 (78.6)0.71
 SVR, n/N (%)38/64 (59.4)9/14 (64.3)0.73
 Relapse, n/N (%)15/53 (28.3)2/11 (18.2)0.71


Apart from environmental and viral factors, genetic variations are probably involved in the efficacy of interferon-based therapies for chronic hepatitis C.21 Interferon-λ induces antiviral, antiproliferative, and immune responses.22 It has been mentioned in the context of HCV infection (i.e., suppression of its replication in vitro)23, 24 and has been applied in clinical HCV treatment.25 IL-28B, located on chromosome 19, encodes interferon-λ3 and has been reported to be involved in the suppression of HCV replication. The present study, to the best of our knowledge, presents the largest cohort for elucidating the influence of genetic polymorphisms near the IL-28B gene on the treatment of HCV-2 patients in a Chinese population residing in Taiwan. We have demonstrated that carriers with the TT genotype of rs8099917, located 8 kb upstream of IL-28B, are more likely to achieve RVR with HCV-2 infection. In addition, a baseline HCV RNA level <400,000 IU/mL, a lower pretreatment level of AST, and a low degree of fibrosis also have predictive value for RVR. By taking into account the attainment of RVR or failure to achieve RVR, we have eliminated the impact of IL-28B genetic polymorphisms on the final treatment outcome of HCV-2 patients.

RVR achievement is by far the most important predictive factor for SVR attainment. Patients with RVR have an approximately 90% chance of treatment success after standard of care with peginterferon/ribavirin combination therapy, regardless of the viral genotypes.2, 4, 26, 27 The achievement of this early goal provides greater flexibility for tailoring the treatment duration on an individual basis and also enhances the cost-effectiveness of treatment.28 An awareness of what baseline factors can predict RVR before treatment is therefore pivotal. Several factors, including age, pretreatment HCV RNA levels, higher pretreatment AST levels, liver fibrosis status, insulin resistance, and higher on-treatment ribavirin doses, have been reported to be associated with RVR in HCV-1 patients.29, 30 Insulin resistance and baseline HCV viral loads have also been noted to predict RVR in patients with HCV-4 infection.31 Mangia et al.32 showed that a high dose of ribavirin and low pretreatment HCV viral loads with an HCV-2 infection were independent factors predicting RVR in patients with HCV-2/HCV-3 infection.32 However, little is known about the general relevance of host genetics to the early phases of viral kinetics in the treatment of HCV infection. Thompson et al.15 recently noted that rs12979860, located close to IL-28B, has a tremendous influence on the attainment of RVR in HCV-1 patients in Western populations. In the current study, we replicated previous findings showing that a low degree of fibrosis and low pretreatment HCV RNA levels were associated with RVR. It is noteworthy that Chinese patients with the TT genotype of rs8099917 were prone to attaining RVR; this was independent of the baseline viral loads, pretreatment AST levels, and degree of fibrosis. Patients with the TT genotype of rs8099917 were associated with improved early viral suppression. The determination of whether our findings can be applied to other ethnicities and other HCV genotypes needs further investigation.

Recent studies focused on SNPs surrounding the IL-28B gene have demonstrated their determinant role in SVR.12-16, 33 In addition, SNPs have also been shown to be associated with the spontaneous clearance of HCV infection.14, 34 Subjects enrolled in genetic association studies were mostly infected with HCV-1, however. Two studies that were conducted in Western populations showed its relevance to Caucasians with HCV-2/HCV-3 infection, but the sample size was relatively small, and the results were inconsistent.14, 16 McCarthy et al.16 noted that there was no difference in effect of rs12979860, a novel SNP that determines treatment efficacy, on SVR comparing subjects with HCV-1 and -2/3 infection among Caucasians, whereas Rauch et al.14 showed that the discriminatory power of rs8099917 to identify likely responders to treatment was restricted to HCV-1 patients and did not apply to HCV-2 patients. Our results have demonstrated that the effect of SNP rs8099917 in the context of other variables is confined to early viral kinetics and does not apply to antiviral therapy outcomes in HCV-2 patients. The real cause is not clear, but it is plausible that the unique character of rapid virological decline after interferon-based therapy might offset a host genetic predisposition in patients with RVR. It is noteworthy that emerging evidence suggests a potential role for genetic polymorphisms of IL-28B in HCV-1 patients without RVR.15 However, host genetic diversity did not show predictive value for final treatment outcomes in non-RVR Chinese patients with HCV-2 infection in the current study. Instead, the results echo our previous finding that the achievement of complete EVR is the most important factor predictive of treatment success in patients who fail to attain RVR.6 Because only approximately 60% of non-RVR patients can achieve SVR, a prolonged course of treatment or a therapy adding other potent antivirals such as protease inhibitors35 might be anticipated in those patients with HCV-2 refractory to current standard regimens.

Intriguingly, patients carrying the favorable TT genotype had significantly lower levels of HCV RNA among our HCV-2 patients. This finding contrasts with the findings of two previous studies. Ge et al.33 reported that among Caucasian patients with HCV-1, those with the rs12979860 wild CC genotype, an independent predictor favoring SVR, had higher baseline HCV viral loads. McCarthy et al.16 demonstrated a similar finding with respect to off-treatment viral loads in Caucasian patients with HCV-1. The exact mechanism underlying this genetic association with viral loads remains unclear. However, the polymorphism has no association with the categorization of individuals' baseline viral loads (which might influence the treatment response) as higher or lower, and this implies that the association of the polymorphism with viral clearance and viral loads may be unrelated.

In conclusion, treatment decisions for patients with chronic hepatitis C infection currently are based mainly on their virological clinical characteristics. Host genetic polymorphisms in the vicinity of IL-28B might determine the RVR rate, the most important predictor of treatment outcome, for Asian patients with HCV-2 infection. Further studies of different populations and other HCV genotypes are warranted to validate these findings.