IL28B polymorphism determines treatment response of patients with hepatitis C

IL28B polymorphism determines treatment response of patients with hepatitis C genotypes 2 or 3 who do not achieve a rapid virologic response


Alessandra Mangia, Liver Unit, IRCCS Casa Sollievo della Sofferenza Hospital, San Giovanni Rotondo, Italy

Alexander J. Thompson, Duke Clinical Research Institute and Duke University Medical Center, Durham, NC, USA

Rosanna Santoro, Liver Unit, IRCCS Casa Sollievo della Sofferenza Hospital, San Giovanni Rotondo, Italy

Valeria Piazzolla, Liver Unit, IRCCS Casa Sollievo della Sofferenza Hospital, San Giovanni Rotondo, Italy

Hans L. Tillmann, Duke Clinical Research Institute and Duke University Medical Center, Durham, NC, USA

Keyur Patel, Duke Clinical Research Institute and Duke University Medical Center, Durham, NC, USA

Kevin V. Shianna, Institute for Genome Sciences & Policy, Center for Human Genome Variation, Duke University, Durham, NC, USA

Leonardo Mottola, Liver Unit, IRCCS Casa Sollievo della Sofferenza Hospital, San Giovanni Rotondo, Italy

Daniela Petruzzellis, Liver Unit, IRCCS Casa Sollievo della Sofferenza Hospital, San Giovanni Rotondo, Italy

Donato Bacca, Department of Internal Medicine, Hospital Castrano, Italy

Vito Carretta, Department of Internal Medicine Hospital Venosa, Italy

Nicola Minerva, Department of Internal Medicine, Hospital Canosa, Italy

David B. Goldstein, Institute for Genome Sciences & Policy, Center for Human Genome Variation, Duke University, Durham, NC, USA

John G. McHutchison, Duke Clinical Research Institute and Duke University Medical Center, Durham, NC, USA


Received 17 February 2010; received in revised form 23 April 2010; accepted 25 May 2010. published online 04 June 2010.

Accepted Manuscript


Abstract

Background & Aims


Polymorphisms in the region of the interleukin (IL)28B gene on chromosome 19 have been associated with peginterferon-a (pegIFN)-induced clearance of genotype 1 hepatitis C virus (HCV); there are no data for patients with genotypes 2 or 3 (G2/3) HCV. We evaluated the effects of IL28B polymorphisms on response to treatment with pegIFN and ribavirin in a well-characterized cohort of G2/3 patients.

Methods


DNA was analyzed from 268 patients (Caucasian, G2=213, G3=55). Patients were randomly assigned to groups that received standard (24 weeks, SD24, n=68) or variable durations of therapy. Patients that received variable durations and had a rapid virological response (RVR) were treated for 12 weeks (VD12, n = 122); those without a RVR were treated for 24 weeks (VD24, n=78). IL28B genotypes (rs12979860) were analyzed for association with treatment response.

Results


The frequencies of the IL28B genotypes were: CC=37%, CT=48%, TT=15% ; 82% of patients with the CC genotype achieved a SVR, compared to 75% with the CT and 58% with the TT genotypes (P = 0.0046). Differences between IL28B genotypes were greatest among patients who failed to attain RVR (VD24 SVR rates: CC=87%, CT=67%, and TT=29%, P=0.0002). Among patients with RVRs (61%), the IL28B genotype was not associated with SVR (>70% for all IL28B genotypes). In a multi-variable logistic regression model, IL28B genotype predicted SVR (odds ratio=1.76; 95% confidence interval=1.16 - 2.7).

Conclusions


An IL28B polymorphism was associated with a SVR in patients infected with genotypes 2 or 3 HCV who did not achieve a RVR. Analysis of IL28B genotype might be used to guide treatment for these patients.

INTRODUCTION

Several viral and host-related factors determine the outcome of peginterferon-␣ (PegIFN) and ribavirin (RBV) treatment in patients with chronic hepatitis C virus (HCV) infection. In addition to viral genotype, serum HCV RNA level and stage of hepatic fibrosis, host genetic factors have recently been identified to influence treatment responsiveness in patients with chronic hepatitis C. Single nucleotide polymorphisms (SNPs) on a linkage disequilibrium block encompassing two interferon lambda genes on chromosome 19 have been strongly associated with response to pegIFN plus RBV combination treatment in three genome wide association studies of HCV genotype 1 infected patients 1-3. The strongest association signal to date has been for the SNP rs12979860 1. Adherent patients with the CC genotype of rs12979860 were more than twice as likely to respond to 48 weeks of treatment as compared to the non-CC genotypes1. Although the mechanism by which this genetic variant acts remains unclear, it is likely to influence innate antiviral immune responses. The influence of this genetic variation on response to shorter treatment courses in patients chronically infected with genotype 2 or 3 HCV is not known.

The current standard therapy for HCV genotype 2 and 3 infection is 24 weeks of pegIFN plus RBV, which cures approximately 80% of patients 4. In recent years, refinements to this regimen have been developed to minimize side effects and costs. In patients with low viral load at baseline who attain week 4 rapid virological response (RVR), 12 – 16 weeks of therapy is associated with rates of SVR that are comparable to those attained after a standard 24 weeks of treatment 5-7 . Although relapse rates may be somewhat higher with shorter therapy 8, the cost-effectiveness of this approach has been demonstrated 6, 9. Individualized therapy for patients with genotype 2/3 HCV infection according to RVR has recently been included in a number of European treatment guidelines 10, 11.

We have now investigated the relevance of IL28B genetic variation to treatment response in a well characterized cohort of European patients with genotype 2 or 3 HCV infection who were treated with pegIFN-alpha-2b and weight-based RBV for 12 or 24 weeks duration according to week 4 virological response, as previously published 5.

DISCUSSION

To our knowledge this is the first report of the role and relevance of the recently described IL28B genetic polymorphism to PegIFN and RBV treatment outcome in the setting of chronic infection with genotype 2 or 3 HCV. This unique cohort, from a prospectively performed landmark clinical trial 5, has allowed us to carefully evaluate the interaction between this genetic determinant and treatment response in patients who received a variable duration of therapy based upon week 4 virologic response.

The data indicate that the IL28B polymorphism is relevant to treatment outcome, although the effect size was attenuated in these genotype 2/ 3 infected European patients, compared to the original observation in North American patients with genotype 1 HCV infection (genotype 2/3: 1.76 [95% CI 1.16 – 2.66], Table 4A, vs genotype 1: OR 5.2 [95%CI 4.1 – 6.7] 14). The attenuation of overall effect size was largely due to the much higher rates of RVR that occurred in the genotype 2/3 patients; IL28B-type did not influence outcome in these rapid responders. The major effect of this genetic variant was in patients who did not achieve a rapid response at week 4, where IL28B-type had a strong influence on the rate of SVR. A second point of contrast between this cohort and the genotype 1 North American cohort previously reported 14 was that the presence of a single C allele was noted to confer significant clinical benefit. In contrast, in genotype 1 HCV infection, the clinical benefit was more restricted to IL28B CC homozygotes 14.

Despite the association between IL28B-type and overall SVR, there was no significant association between the IL28B polymorphism and week 4 viral clearance, although RVR was numerically higher in CC and CT patients compared to TT patients. However, by week 8, viral clearance was clearly more common in CC and CT patients. We interpret this to mean that the IL28B polymorphism is associated with differential viral kinetics, similar to that observed in genotype 1, but that the different rate of decline between IL28B-types is attenuated, such that rates of viral clearance were not demonstrably different by week 4, but were significantly different at week 8. It is likely that quantitative measures of viral load may have been more sensitive for differentiating kinetics between patients with different IL28B-types, but unfortunately such measures were not available. We therefore acknowledge that the lack of HCV RNA decline evaluation represents a possible limitation of this study and that HCV RNA viral decline could have shown a better correlation with the IL28B genetic variant.

The study was not powered to formally evaluate the utility of IL28B-type for making decisions about selecting RVR patients for 12 weeks of therapy. Current guidelines recommend this approach for RVR patients with genotype 2/3 HCV infection and low viral load 10, 11. It is tempting to speculate that such treatment would be most suitable for CC patients. Although our data would be consistent with this being a subgroup of RVR patients most likely to achieve equivalent response rates, no significant difference was noted between 12 and 24 weeks of therapy in this cohort and we cannot draw firm conclusions. What appears more impressive is the potential role for IL28B genotyping in selecting non-RVR patients for standard 24 week therapy or more prolonged therapy. Prospective studies randomizing patients according to IL28B-type will be necessary to address both of these issues.

Of interest and as previously reported by Ge et al. in a genotype 1 HCV cohort 1, we also noted that the prevalence of the favourable C allele was lower in genotype 2/3 HCV-infected patients than in the non-HCV control population. This suggests that the T allele is associated with persistence, and is consistent with the recent observation that the IL28B CC-type is associated with spontaneous clearance following acute HCV infection 13, 15. The data therefore suggest that the role of IL28B variants in spontaneous clearance may be independent of HCV genotype.

Finally, the identification that genetic variation in the IL28B gene region is associated with pegIFN treatment outcome raises the possibility that IL28B might have therapeutic potential. IL28B (or IFN-lambda (␣)-3) is one of three members of the IFN-␣ family, first identified in 2003 16, 17. All 3 members signal via a common IFN- ␣ receptor, to trigger an IFN-stimulated gene response that overlaps considerably with the downstream signaling pathway of type 1 IFNs. Antiviral activity against both HCV and HBV has been demonstrated in vitro 18. Phase 1 studies investigating IL29 (IFN-␣-1) have recently confirmed an anti-HCV effect in genotype 1 non-responders 19. Phase II development in a genotype 1-naïve population is planned. The efficacy of IL29 in the setting of genotype 2/3 HCV is yet to be studied.

In conclusion, these findings highlight the importance of the interaction between HCV genotype and host genetic determinants in influencing treatment response. We have demonstrated that genetic variation in IL28B is associated with treatment response in patients infected with genotype 2/3 HCV. This effect is attenuated compared to previous reports in genotype 1 HCV, consistent with the recognized IFN-sensitivity of these genotypes, and also suggesting that key differences in the ability to evade innate antiviral immune responses are likely to exist between genotype 1 and genotype 2/3 HCV. IL28B- type was particularly important in patients who did not achieve RVR. Further prospective randomized studies are needed to investigate whether patients without RVR carrying the IL28B non-CC genotypes benefit from extended treatment duration.