- Email: [email protected]
3 infection
Accepted Manuscript IFNL3 polymorphisms predict response to therapy in chronic hepatitis C genotype 2/3 infection Mohammed Eslam, Reynold Leung, Manuel Romero-Gomez, Alessandra Mangia, William L. Irving, David Sheridan, Ulrich Spengler, Lindsay Mollison, Wendy Cheng, Elisabetta Bugianesi, Duncan McLeod, Abed M. Zaitoun, Vito Attino, Diane Goeltz, Jacob Nattermann, Mark Douglas, David R. Booth, Jacob George, Golo Ahlenstiel PII: DOI: Reference:
S0168-8278(14)00269-4 http://dx.doi.org/10.1016/j.jhep.2014.03.039 JHEPAT 5118
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
13 August 2013 14 March 2014 30 March 2014
Please cite this article as: Eslam, M., Leung, R., Romero-Gomez, M., Mangia, A., Irving, W.L., Sheridan, D., Spengler, U., Mollison, L., Cheng, W., Bugianesi, E., McLeod, D., Zaitoun, A.M., Attino, V., Goeltz, D., Nattermann, J., Douglas, M., Booth, D.R., George, J., Ahlenstiel, G., IFNL3 polymorphisms predict response to therapy in chronic hepatitis C genotype 2/3 infection, Journal of Hepatology (2014), doi: http://dx.doi.org/10.1016/ j.jhep.2014.03.039
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
IFNL3 polymorphisms predict response to therapy in chronic hepatitis C genotype 2/3 infection Mohammed Eslam1, Reynold Leung1,2, Manuel Romero-Gomez3, Alessandra Mangia4, William L. Irving5, David Sheridan6, Ulrich Spengler7, Lindsay Mollison8, Wendy Cheng9, Elisabetta Bugianesi 10, Duncan McLeod11, Abed M. Zaitoun12, Vito Attino13, Diane Goeltz14, Jacob Nattermann7, Mark Douglas1, David R. Booth2, Jacob George1, Golo Ahlenstiel1 1
Storr Liver Unit, Westmead Millennium Institute and Westmead Hospital, University of Sydney, NSW, Australia 2 Institute of Immunology and Allergy Research, Westmead Hospital and Westmead Millennium Institute, University of Sydney, NSW, Australia. 3 Unit for The Clinical Management of Digestive Diseases and CIBERehd, Hospital Universitario de Valme, Sevilla, Spain. 4 Division of Hepatology, Ospedale Casa Sollievo della Sofferenza, IRCCS, San Giovanni Rotondo, Italy 5 NIHR Biomedical Research Unit in Gastroenterology and the Liver, University of Nottingham, Nottingham, UK 6 Liver Research Group, Institute of Cellular Medicine, Medical School, Newcastle University, Newcastle upon Tyne, UK 7 Department of Internal Medicine I, University of Bonn, Sigmund-Freud- Strasse, Bonn, Germany 8 Fremantle Hepatitis Services, Fremantle, Australia 9 Department of Gastroenterology and Hepatology, Royal Perth Hospital, Western Australia 10 Division of Gastro-Hepatology, S. Giovanni Battista Hospital, Turin, Italy. 11
Department of Anatomical Pathology, Institute of Clinical Pathology and Medical Research
(ICPMR), Westmead Hospital, Sydney, Australia. 12
Department of Histopathology, University Hospital Queens Medical Centre, Nottingham, UK Department is Anatoma Patologica IRCCS, Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy 14 Pathologisches Institute, Universitaetsklinikum Bonn, Germany 13
Corresponding Author Jacob George Department of Medicine, Westmead Hospital Westmead, NSW 2145 Ph: 61-2-98457705; Fx 61-2-96357582 Email: [email protected] 1
Running title: IFNL3 in HCV genotypes 2 and 3 Key words: Chronic hepatitis C, IFNL3 (IL28B), SVR, genotype 2, 3, response to therapy Electronic word count: 4951 words [including the abstract, references, tables, and figure legends] Number of figures and tables: 3 tables, 7 figures. List of abbreviations in the order of appearance: SNPs: Single nucleotide polymorphisms, IFNL3: interferon lambda 3, SVR: sustained virologic response, HCV-2/3: Hepatitis C virus genotype 2 and 3, RVR: rapid virological response, OR: odds ratio, Peg-IFNα/RBV: pegylated interferon-alpha and ribavirin, CHC: chronic hepatitis C, BMI: body mass index, ALT: Alanine aminotransferase, AST: aspartate aminotransferase, γGT: gamma-glutamyl-transferase, SD: standard deviation,
Conflict of interest: None. Financial support: GA, MD and JG are supported by National Health and Medical Research Council (NHMRC) Program and Project grants (1053206 and 1006759) and the Robert W. Storr bequest to the Sydney Medical Foundation of the University of Sydney. ME is supported by an International Postgraduate Research Scholarships (IPRS) and an Australian Postgraduate Award (APA) of the University of Sydney.
2
ABSTRACT (word count: 245) Background & aim: Single nucleotide polymorphisms (SNPs) near the interferon lambda 3 (IFNL3, previously known as IL28B) region are the strongest baseline predictors of sustained virologic response (SVR) to pegylated interferon and ribavirin therapy in hepatitis C virus (HCV) genotype 1 infection. Whether IFNL3 SNPs influence treatment response in genotype 2 and 3 (HCV-2/3) infection remains controversial. This study sought to clarify in a large cohort, whether SNPs in the IFNL3 region are associated with treatment response in HCV-2/3 patients. Methods: The cohort comprised 1002 HCV-2/3 Caucasians patients treated with pegylated interferon-alpha and ribavirin who underwent genotyping for the SNPs rs12979860 and rs8099917. Results: Overall, 736 (73.5%) patients achieved SVR (81.9%, 67.9%, and 57.8% for rs12979860 CC, CT, and TT [P=0.0001]; 78%, 68.7%, and 46.3% for rs8099917 TT, TG, and GG [P=0.0001]). By logistic regression, both rs12979860 CC and rs8099917 TT were independent predictors of SVR with an odds ratio (OR) of 2.39 (1.19-3.81) P= 0.0001 and OR 1.85 (1.15-2.23) P=0.0001, respectively. IFNL3 responder genotypes were more frequent in relapsers than null-responders (P=0.0001 for both SNPs). On-treatment rapid virological response (RVR) was predictive of SVR only in those individuals with IFNL3 non-responder genotypes (rs12979860 CT/TT and rs8099917 TG/GG). Conclusion: This adequately powered study in patients with HCV genotypes 2 or 3 infection clearly demonstrates that IFNL3 genotypes are the strongest baseline predictor of SVR, in keeping with the known association for genotype 1 infection. IFNL3 genotyping can aid in therapeutic decision making for these patients.
3
INTRODUCTION Hepatitis C virus (HCV) infects 170 million people worldwide [1] and is a leading cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma [2]. Treatment with pegylated interferon-alpha and ribavirin (Peg-IFNα/RBV) results in a sustained virologic response (SVR) in approximately 50% of people infected with HCV genotype 1 (HCV-1) and in ~75% of those infected with genotypes 2 or 3 (HCV-2/3) [3,4]. Although the addition of protease inhibitors such as telaprevir and boceprevir leads to a substantial improvement in the SVR rate for HCV-1 infection [5], Peg-IFNα/RBV remains the standard of care for non-1 genotypes [5]. In this context, there is growing interest in stratifying patients based on pre-treatment predictors of response, in order to have abbreviated treatment regimens both to minimize cost and more importantly, to reduce the adverse effects of interferon-based therapies. The discovery of polymorphisms near the IFNL3 (formerly known as IL28B) gene, which codes for interferon lambda 3, represented a milestone in hepatitis C research [6,7,8]. Four years later, the value of polymorphisms in this region for predicting response to Peg-IFNα/RBV treatment in HCV-1 is well established [6-8]. In contrast, similar data for HCV-2/3 infection remains contentious [9,10,11], with results of two recent meta-analyses contradicting each other [12,13]. This controversy quite likely reflects the relatively small sample sizes of the previous studies and differences in ethnic stratification, which critically determines the frequencies of observed IFNL3 genotypes [9-11]. The aims of this study were 1) to clarify the role of pre-treatment IFNL3 polymorphisms for predicting SVR in HCV-2/3 patients, 2) to explore the predictors of response in patients with the IFNL3 responder genotype and 3) to clarify the distribution of IFNL3 in non-SVR patients. To address these questions, we analysed a large cohort of patients with genotype 2 or 3 chronic hepatitis C (CHC) infection and known treatment outcomes. 4
PATIENTS AND METHODS Patient cohort The cohort comprised 1002 Caucasians patients with CHC from Australia (N=251), Spain (N=261), Italy (N=235) , the United Kingdom (N=199), and Germany (N=56), fulfilling the following inclusion criteria: (a) adults aged 18 years or older with CHC based on the presence of anti-HCV and detectable serum HCV-RNA for >6 months, (b) infection with HCV-2 or 3 and (c) known outcome following Peg-IFNα/RBV based therapy. Patients were excluded if they were co-infected with either hepatitis B virus or HIV or had other liver diseases as assessed by standard tests. 89 of the patients from this study have been included in previous reports [14]. All patients were treated with Peg-IFNα/RBV; the duration of therapy and stopping rules employed were according to standard guidelines [5]. SVR was defined as undetectable HCV RNA 24 weeks after completion of therapy. Rapid virologic response (RVR) was defined as undetectable HCV RNA levels after 4 weeks of therapy. The non-SVR cohort included patients with either non-response or relapse. Virologic relapse was defined as HCV RNA undetectable at the end-of-treatment but positive thereafter, whereas virologic non-response was defined as HCV RNA detectable throughout the entire therapy of at least 24 weeks. Ethical approval was obtained from the Human Research Ethics Committees of the Sydney West Area Health Service and the University of Sydney. All other sites had ethical approval from their respective ethics committees. Written informed consent for genetic testing was obtained from all participants. The study was conducted in accordance with the international ethical guidelines of the International Conference on Harmonization Guidelines for Good Clinical Practice [15].
5
Clinical and laboratory assessment The following data were collected at baseline: gender, age, ethnicity (Caucasian Vs. nonCaucasian), recruitment center, body mass index (BMI), and routine laboratory tests. BMI was calculated as weight divided by the square of the height (kg/m2). Alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl-transferase (γGT), bilirubin, hemoglobin, leukocyte and platelet count were determined by routine laboratory techniques. Liver biopsy Data on liver fibrosis at biopsy was available in 605 patients. Biopsies were interpreted according to the scoring schema developed by the METAVIR group [16] by a single expert liver pathologists in each center who was blinded to patient clinical characteristics and serum measurements. Thirty-five biopsies were scored independently by pathologists (DM, AZ, DG and VA) from the various centers, and inter observer agreement was calculated by using the κ statistic. Fibrosis was scored on a 5-point scale: F0, no fibrosis; F1, portal fibrosis alone; F2, portal fibrosis with rare septae; F3, portal fibrosis with many septae; F4, cirrhosis. The presence of stage F2, F3, or F4 was termed “significant fibrosis”, whereas the term “advanced fibrosis” was reserved for stage F3 or F4. Necro-inflammatory activity, based on assessment of piecemeal and lobular necrosis, was graded on a 4-point scale: A0, no activity; A1, mild; A2, moderate; A3, severe. IFNL3 Genotyping Genotyping for IFNL3 SNPs was undertaken using the TaqMan SNP genotyping allelic discrimination method (Applied Biosystems, Foster City, CA, USA). The rs8099917 genotyping kit was supplied by Applied Biosystems and rs12979860 genotyping was performed using a custom designed genotyping assay by Applied Biosystems. Detailed procedures have been described previously [17]. Genotyping was performed using the StepOne RT system and 6
analyzed with StepOne software v.2.3.0 (Applied Biosystems, Foster City, CA, USA). All genotyping was blinded to treatment outcome. Statistical Analysis Quantitative data are expressed as mean ± SD (standard deviation), and categorical data as number (percentage) of patients. Skewed variables are reported as median and range. The Student t-test or non-parametric, i.e. Wilcoxon-Mann-Whitney U-test or Kruskal-Wallis tests were used to compare quantitative data, while appropriate. χ2 and Fisher-exact tests were used for comparison of frequency data and to evaluate the relationships between groups. All tests were two-tailed and p values <0.05 were considered significant. Multiple regression models were used to assess for factors independently associated with RVR and SVR. IFNL3 SNP comparisons were made using a dominant model, in which patients carrying one allele; or a recessive model (two copies of the minor allele), were compared with others unless otherwise indicated. Necroinflammation score was dichotomized as (absent/ mild necroinflammation or minimal changes (Metavir score A0-A1) or (presence of moderate/severe necroinflammation (Metavir score A2-A3). Fibrosis stage was dichotomized into two groups; absent or mild fibrosis (Metavir score F0-1) and significant fibrosis (Metavir score F2-4). To assess the cohort size required to detect significant differences, retrospective power analysis was performed, a power curve was plotted for power, sample size and odds ratio. All analyses were carried out using the statistical software package SPSS for Windows, version 14 (SPSS, Chicago, IL) and SAS version 9.1 and SAS Enterprise 9.4
7
RESULTS The characteristics of the study cohort are shown in Table 1. The median age was 43 (range: 19 - 69) years, with 560 (55.9%) patients being male. Of the cohort, 587 (58.6%) were infected with HCV-3, while the remaining had infection with HCV-2. The interobserver agreement between pathologists was good (κ = 77.5) for METAVIR staging. An SVR was achieved in 736 (73.5%) patients; all others were classified as having a non-SVR. The latter included those with non-response (N=139) and those with relapse (N=127).
IFNL3 genotype distribution Genotype distribution of the two IFNL3 SNPs was in Hardy–Weinberg equilibrium (data not shown). The overall genotype distribution of IFNL3 rs12979860 CC, CT, and TT was 46.9%, 42.9%, and 10.2%, and the distribution of rs8099917 TT, TG, and GG was 63.1%, 32.4%, and 4.5%, respectively. 86 patients had no rs8099917 genotype due to a lack of sufficient DNA for genotyping. The characteristics of those patients were matched to that of the whole cohort. SVR rates were 81.9%, 67.9%, and 57.8% for rs12979860 CC, CT, and TT (P=0.0001) and 78%, 68.7%, and 46.3% for rs8099917 TT, TG, and GG (P=0.0001) (Figure 1).
Baseline factors associated with a sustained virological response By univariate analysis, apart from IFNL3 rs12979860 and rs8099917, SVR was associated with previously reported baseline factors including younger age, female gender, lower HCV RNA level and none/mild fibrosis. Of the on treatment factors, as expected, RVR was a strong predictor of subsequent SVR (Table 2).
8
Adjusting for the covariates of age, sex, recruitment center, HCV RNA level, RVR, and the stage of histological fibrosis, two multivariate logistic regression models were developed. Each included only one of the IFNL3 SNPs as the 2 SNPs are in linkage disequilibrium [18]. In the first model, all significant covariates from the univarite analysis were included except for rs8099917. rs12979860 CC showed an odds ratio (OR) for an association with SVR (CC versus CT/TT) of 2.39 (1.19 - 3.81), P= 0.0001). The second regression model contained all covariates, except for rs12979860. Here, rs8099917 TT, was significantly associated with SVR (TT versus TG/GG: OR 1.85 (1.15 - 2.23) P=0.0001) (Table 2). Of note, all models were also calculated for pretreatment predictors alone, i.e. without RVR, and the results were similar (data not shown). To explore if the conflicting results reported in the literature could be due to the small sample sizes of previous studies, we undertook a retrospective power analysis. Notably, cohorts of >300 patients (for rs12979860) and >400 patients (for rs8099917) were required in order to have 90% power to detect an odds ratio of 2 and 1.5, respectively with a P-value (< 0.05) for the differences in SVR rates of patients with and without the responder genotype (Figure 2). This indicates that our cohort was sufficiently powered to detect associations in this patient group.
IFNL3 and treatment failure With respect to the correlation of IFNL3 with treatment failure, non-SVR rates were 22%, 31.3%, and 53.7% for rs8099917 TT, TG, and GG (P=0.001), and 18.1%, 32.1%, and 44.2% for rs12979860 CC, CT, and TT (P=0.001). Based on multiple logistic analysis to adjust for possible cofounders, the minor allele homozygotes for rs8099917 had the highest predictive value, GG: OR 3.45 (95%CI: 1.83-6.48), P=0.0001, which was higher than for the rs12979860 TT homozygote (OR: 2.21 (95%CI: 1.45-3.7), P=0.0002).
9
The predictive value for
heterozygotes rs8099917 GT and rs12979860 CT was similar (OR 1.43 (95%CI: 1.05-1.95), and (OR 1.63 (95%CI: 1.23-2.17), respectively, P< 0.01, for both). IFNL3 and rapid virologic response RVR rates for the rs12979860 CC, CT, and TT genotypes were 74.4%, 68.9%, and 60.7% (p=0.09), while for rs8099917, TT and TG were more likely to achieve RVR than GG (74.3% and 70.1% vs. 51.7%, p=0.03; Figure 3). In multiple regression analysis corrected for other variables significantly associated with RVR in univariate analysis, i.e. age, HCV-genotype and baseline viral load, rs8099917 GG remained significantly associated with failure to achieve RVR (OR: 1.98 (95% CI: 1.27-3.52), P=0.001). The negative correlation of GG with RVR translated to lower SVR rates in this group as compared to rs12979860 TT (46.3% vs. 57.8%, p=0.001).
Differential impact of IFNL3 genotypes in HCV genotype 2 versus 3 infection The baseline characteristics of patients infected with HCV-2 (n=415) or 3 (n=587) are summarized separately in Table 1. The responder genotype distribution of both SNPs of IFNL3 was higher in HCV-3 as compared to HCV-2 patients (Table 1). RVR, but not SVR, was more frequent in HCV-2 than HCV-3 (RVR: 79.5% vs. 63.2%, P=0.0001; SVR: 71.6% vs. 74.8%, P=0.2). As for the relationship between the IFNL3 SNPs and treatment response, both rs12979860 and rs8099917 showed significant associations with SVR in both HCV-3 and HCV2 patients (Figure 4). Both IFNL3 SNPs also remained independently correlated with SVR in multivariate analysis of HCV-3 patients (OR 2.71 (95% CI: 1.83-4.2), P=0.0001 and OR 1.97 (95% CI: 1.31-2.96), P=0.001, for rs12979860 and rs8099917, respectively). In HCV-2 patients, the results were similarly significant (OR 1.88 (95% CI: 1.19-2.93), P=0.005 and OR 1.63 (95% CI: 1.05-2.54), P=0.02 for rs12979860 and rs8099917, respectively). 10
Predictors of SVR in patients with IFNL3 responder genotypes In patients with rs12979860 CC, i.e. the responder genotype, younger age, none/mild fibrosis, lower baseline HCV-RNA log10, higher platelets counts and RVR were associated with SVR by univariate analysis, while only lower HCV-RNA log10 remained independently correlated with SVR in multivariate analysis (OR 2.66 (95% CI: 1.49-8.48), P=0.001). Separate analysis using the median of HCV-RNA log10 (5.89 IU/mL) yielded similar results. In patients with rs12979860 non-responder genotypes (CT/TT), the independent predictors of SVR were RVR (OR 4.8 (95% CI: 2.1-8.79), P=0.0001) and HCV-RNA log10 (OR 2.34 (95% CI: 1.374.45), P=0.001). In patients with rs8099917 TT, SVR was also associated with younger age, female gender, lower basal HCV-RNA log10 level, none or mild fibrosis and RVR. Only lower baseline HCV-RNA log10 level (OR 2.42 (95% CI: 1.86-5.12), P=0.001) was independently correlated with SVR in multivariate analysis.
Predictive value of IFNL3 in non-sustained virologic response group. Finally, we studied the distribution of IFNL3 genotypes among the non-SVR group (n=266), comparing non-responders (N=139) and relapsers (N=127). The responder genotypes rs12979860 CC and rs8099917 TT were more frequent in relapsers than in non-responders (P=0.0001 for both) (Figure 5).
11
DISCUSSION Despite new direct acting antiviral drugs for HCV-1 infection, Peg-IFNα/RBV remains the standard of care for patients with HCV-2/3, and will remain so for some time in less privileged parts of the world [5,19]. In this context, reliable pre-treatment predictors associated with a favorable outcome can improve cost-effectiveness and support clinical decision making [20]. In the present study, we clarify the debate regarding the impact of IFNL3 SNPs rs12979860 and rs8099917 on outcomes in patients with HCV genotype 2 or 3 infection. Using the largest patient cohort to date (N=1002), we demonstrate that IFNL3 SNPs are the strongest baseline predictors of treatment response in this group, though only rs8099917 GG was associated with lower RVR rates. Our retrospective power analysis provides insights into the controversy arising from previously published reports: Moghaddam et al. for example reported that rs12979860 and rs8099917 were not associated with SVR to Peg-IFNα/RBV therapy in 281 patients infected with HCV-3 [9] and Rauch A, et al. similarly reported that rs8099917 was not associated with response in 303 HCV-2/3 [21].
In contrast, Mangia et al. reported that in their cohort of 268
patients, IFNL3 rs12979860 genotype was only associated with SVR in HCV-2/3 patients who did not achieve RVR [11]. Sarrazin et al. reported that only rs12979860 but not rs8099917 was associated with SVR in a cohort of 267 HCV-2/3 patients [22]. Recently, Stenkvist et al. showed in a cohort of 100 HCV-2/3 patients a significantly higher median 1st phase decline in HCVRNA levels in patients with rs12979860 CC than non-CC genotype patients [23]. Here we demonstrate that an impact of rs12979860 or rs8099917 on SVR rates is only reliably observed in cohorts of >300 and 400 patients, respectively. This suggests that previous reports were not adequately powered and that the divergent results principally reflect sample size differences between cohorts. Consistent with this observation, in a large cohort from Japan (n=719 patients, 12
with either genotype 2a or 2b), in the sub-cohort of patients treated with Peg-IFNα/RBV (n=160), IFNL3 polymorphisms were not associated with SVR. However, in the subset treated with interferon monotherapy (n=559), rs8099917 independently predict SVR [24]. The current larger cohort study assessed Caucasian patients with both genotype 2/3 infection and all were treated with Peg-IFNα/RBV. In keeping with data from HCV-1, rs8099917 has better correlation with treatment failure than rs12979860 [18]. This is due to rs12979860 C tagging two distinct haplotypes (43% and 10% of alleles in CHC with HCV-1), which are both responsive (odds ratios of 0.7 and 0.6 for treatment failure), whereas rs12979860 T includes a mixed group of haplotypes (24%, 10%, 2% and 1% of alleles; ORs of 2.2, 1.0, 0.8 and 1.5, respectively for treatment failure) [18]. Thus, rs12979860 tagged by a C allele will always improve the chances of response, whereas with rs12979860 tagged by a T, it depends on the particular haplotype. In contrast, only one distinct IFNL3 haplotype, haplotype 2, is tagged by rs8099917 G, which is strongly associated with treatment failure. rs8099917 G’s have an OR of 2.2 in this data for treatment failure [18]. In contrast, only rs12979860 T’s, which are also rs8099917 G, have an OR of 2.2 for treatment failure, the remaining haplotypes have neutral odds ratios. By the same token, rs8099917 T’s shows a mixed picture with some responding and others not, as unlike rs12979860 C’s, not all rs8099917 T-tagged haplotypes are responders, i.e. 20% of rs8099917 T’s are neutral responders (compared with 0% for rs12979860 C) [18]. Thus, from clinical point of view, both SNPs play a complimentary role with regard to prediction of response. A fourth type 3 interferon (IFNL4) has recently been described from primary human hepatocytes [25] that better predicts response/non response to dual therapy in genotype 1 CHC. However, it is important to note that SS4694115590 likewise, does not explain the difference in haplotype-based responses, as
13
acknowledged by the authors of that report [25]. Further studies are required to better understand the mechanisms for the IFNL3 haplotype-based responses. Given that SVR rates in patients carrying the responder genotype was 82% for rs12979860 CC and 78% for rs8099917 TT, we analyzed the potential predictors for treatment response in these patients. For rs12979860 CC genotypes (~44% of Caucasian subjects in this cohort), only baseline HCV-RNA log10 was independently correlated with SVR, whereas for non-responder genotypes (CT/TT), RVR and baseline HCV-RNA log10 independently predicted SVR. IFNL3 responder genotypes were more frequent in relapsers than non-responders. This is consistent with the established better SVR rate following retreatment of relapsers as compared to non-responders in HCV genotype 1 infection [26,27] and underlines the fact that IFNL3 genotype is an important marker of IFN sensitivity. Hence, IFNL3 genotyping may have a particular role in identifying patients with HCV-2/3 for re-treatment, especially in scenarios where interferon containing regimens remain the standard of care in cost-constrained economies. In conclusion, we show that IFNL3 genotypes are associated with treatment response to Peg-IFNα/RBV in HCV-2/3 patients, similar to the effect observed in HCV-1. Importantly, the data indicates that confounders including cohort size explain to a large degree the controversy from previous reports. The predictive value of RVR is more obvious in those with IFNL3 nonresponder, than in those with the responder genotype. Although new treatments may become available in the future [28], IFNL3 genotyping can aid in clinical decision making for patients with HCV-2/3 infection, and for patient stratification to abbreviated therapy trials or for retreatment.
14
References
[1] World Health Organization . Hepatitis C. Fact Sheet Number 164, Revised October 2000. (http://www.who.int/mediacentre/factssheets/fs164len) Accessed 6 March 2012. [2] Seeff LB. The history of the "natural history" of hepatitis C (1968-2009). Liver Int. 2009;29 Suppl 1:89-99 [3] Manns MP, McHutchison JG, Gordon SC, Rustgi VK, Shiffman M, Reindollar R, et al. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 2001;358:958–965 [4] Hadziyannis SJ, Sette Jr H, Morgan TR, Balan V, Diago M, Marcellin P, et al. Peginterferonalpha2a and ribavirin combination therapy in chronic hepatitis C: a randomized study of treatment duration and ribavirin dose. Ann Intern Med 2004;140:346–355. [5] Ghany MG, Nelson DR, Strader DB, Thomas DL, Seeff LB; American Association for Study of Liver Diseases. An update on treatment of genotype 1 chronic hepatitis C virus infection: 2011 practice guideline by the American Association for the Study of Liver Diseases. Hepatology. 2011;54(4):1433-44. [6] Ge D, Fellay J, Thompson AJ, Simon JS, Shianna KV, Urban TJ, et al. Genetic variation in IL28B predicts hepatitis C treatment-induced viral clearance. Nature 2009; 461:399–401. [7] Suppiah V, Moldovan M, Ahlenstiel G, Berg T, Weltman M, Abate ML, et al. IL28B is associated with response to chronic hepatitis C interferon-alpha and ribavirin therapy. Nat Genet 2009; 41:1100–1104. [8] Tanaka Y, Nishida N, Sugiyama M, Kurosaki M, Matsuura K, Sakamoto N, et al. Genomewide association of IL28B with response to pegylated interferon-alpha and ribavirin therapy for chronic hepatitis C. Nat Genet 2009;41:1105–1109. [9] Moghaddam A, Melum E, Reinton N, Ring-Larsen H, Verbaan H, Bjøro K, et al. IL28B genetic variation and treatment response in patients with hepatitis C virus genotype 3 infection. Hepatology 2011;53 (3):746–754. [10] Yu ML, Huang CF, Huang JF, Chang NC, Yang JF, Lin ZY, et al. Role of interleukin-28B polymorphisms in the treatment of hepatitis C virus genotype 2 infection in Asian patients. Hepatology 2011;53 (1):7–13. 15
[11] Mangia A, Thompson AJ, Santoro R, Piazzolla V, Tillmann HL, Patel K, et al. An IL28B polymorphism determines treatment response of hepatitis C virus genotype 2 or 3 patients who do not achieve a rapid virologic response. Gastroenterology 2010; 139: 821–7. [12] Jia Z, Ding Y, Tian S, Niu J, Jiang J. Test of IL28B polymorphisms in chronic hepatitis C patients treated with PegIFN and ribavirin depends on HCV genotypes: results from a metaanalysis. PLoS One. 2012;7(9):e45698. doi: 10.1371/journal.pone.0045698. [13] Schreiber J, Moreno C, Garcia BG, Louvet A, Trepo E, Henrion J, et al. Meta-analysis: the impact of IL28B polymorphisms on rapid and sustained virological response in HCV-2 and -3 patients. Aliment Pharmacol Ther. 2012; 36(4):353-62. [14] Bucci C, von Delft A, Christian A, Flemming VM, Harrison A, Halliday J, et al. 'Favourable' IL28B polymorphisms are associated with a marked increase in baseline viral load in hepatitis C virus subtype 3a infection and do not predict a sustained virological response after 24 weeks of therapy. J Gen Virol. 2013;94(Pt 6):1259-65. [15] Dixon JR Jr. The International Conference on Harmonization Good Clinical Practice guideline. Qual Assur. 1998;6(2):65-74. [16] Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. The
METAVIR Cooperative Study Group. Hepatology 1996;24:289-293. [17] Suppiah V, Gaudieri S, Armstrong NJ, O'Connor KS, Berg T, Weltman M, et al. IL28B, HLA-C, and KIR variants additively predict response to therapy in chronic hepatitis C virus infection in a European Cohort: a cross-sectional study. PLoS Med. 2011;8(9):e1001092. doi: 10.1371/journal.pmed [18] Smith KR, Suppiah V, O'Connor K, Berg T, Weltman M, Abate ML, et al. Identification of improved IL28B SNPs and haplotypes for prediction of drug response in treatment of hepatitis C using massively parallel sequencing in a cross-sectional European cohort. Genome Med. 2011;3(8):57. [19] Foster GR, Hézode C, Bronowicki JP, Carosi G, Weiland O, Verlinden L, et al. Telaprevir alone or with peginterferon and ribavirin reduces HCV RNA in patients with chronic genotype 2 but not genotype 3 infections. Gastroenterology. 2011; 141:881-889. [20] Shiffman ML. Retreatment of patients with chronic hepatitis C. Hepatology 2002;36:S128– S134. 16
[21] Rauch A, Kutalik Z, Descombes P, Cai T, Di Iulio J, Mueller T, et al. Genetic variation in IL28B is associated with chronic hepatitis C and treatment failure: a genome-wide association study. Gastroenterology. 2010; 138(4):1338-45, 1345.e1-7. [22] Sarrazin C, Susser S, Doehring A, Lange CM, Müller T, Schlecker C, et al. Importance of IL28B gene polymorphisms in hepatitis C virus genotype 2 and 3 infected patients. J Hepatol. 2011;54(3):415-21. [23] Stenkvist J, Sönnerborg A, Weiland O. HCV RNA decline in chronic HCV genotype 2 and 3 during standard of care treatment according to IL28B polymorphism. J Viral Hepat. 2013; 20(3):193-9. [24] Kawaoka T, Hayes CN, Ohishi W, Ochi H, Maekawa T, Abe H, et al. Predictive value of the IL28B polymorphism on the effect of interferon therapy in chronic hepatitis C patients with genotypes 2a and 2b. J Hepatol. 2011; 54(3):408-14. [25] Prokunina-Olsson L, Muchmore B, Tang W, Pfeiffer RM, Park H, Dickensheets H, et al. A variant upstream of IFNL3 (IL28B) creating a new interferon gene IFNL4 is associated with impaired clearance of hepatitis C virus. Nat Genet. 2013;45(2):164-71. [26] Shiffman ML, Di Bisceglie AM, Lindsay KL, Morishima C, Wright EC, Everson GT, et al. Peginterferon alfa-2a and ribavirin in patients with chronic hepatitis C who have failed prior treatment. Gastroenterology 2004, 126:1015–1023. [27] Jacobson IM, Gonzales SA, Ahmed F, Lebovics E, Min AD, Bodenheimer HC, et al. A randomized trial of pegylated interferon a-2b plus ribavirin in the retreatment of chronic hepatitis C. Am J Gastroenterol 2005; 100:2453–2462. [28] Gane EJ, Stedman CA, Hyland RH, Ding X, Svarovskaia E, Symonds WT, et al. Nucleotide polymerase inhibitor sofosbuvir plus ribavirin for hepatitis C. N Engl J Med. 2013; 368(1):34-44.
17
FIGURE LEGENDS
Figure 1: Association between IFNL3 rs12979860 and rs8099917 genotypes and SVR (P=0.0001 for both).
Figure 2: Retrospective power analysis. The power curve represents the correlation between statistical power, sample size and odds ratio. Cohorts of >300 patients (for rs12979860) and >400 patients (for rs8099917) are required in order to have 90% power to detect an odds ratio of 2 and 1.5, respectively with P-value < 0.05 for the difference between SVR rates of patients with and without the responder genotypes
Figure 3: Association between IFNL3 rs12979860 and rs8099917 genotypes and RVR (P=0.09 and P=0.03, respectively).
Figure 4: Association between IFNL3 rs12979860 and rs8099917 genotypes and SVR stratified by HCV genotypes (HCV-3 and 2).
Figure 5: Distribution of IFNL3 rs12979860 and rs8099917 genotypes in non-SVR patients (Relapser vs. Null-responder) (P=0.0001 for both).
Table 1. Baseline demographic characteristics and treatment response
Variables
All patients (n=1002)
HCV-2
HCV-3
P-value
(n=415)
(n=587)
(HCV-2 vs. HCV3)
Age
43 (19-69)
46 (19-69)
41 (18-69)
0.0001
Male (%)
560(55.9)
243(58.6)
317(54)
0.1
BMI(Kg/m2)
25 (18-34)
25 (18-34)
25 (18-34)
0.1
HCV-RNA log10 (n=863)
5.89 (1.11-7.95)
5.9 (2.75-7.95)
5.88 (1.11-7.48)
0.7
SVR(%)
736 (73.5)
297 (71.6)
439(74.8)
0.2
Non-SVR(%)
266 (26.5)
118(28.4)
148(25.2)
RVR
346 (70.5)
174(79.5)
172(63.2)
NON-RVR
145 (29.5)
45(20.5)
100(36.8)
Absent/mild fibrosis
281 (46.4)
106(48.2)
175(45.2)
Moderate/severe fibrosis
324 (53.6)
114(51.8)
210 (54.5)
Absent/mild Steatosis
425(70.2)
167 (39.3)
258 (60.7)
Moderate/severe Steatosis
180(29.8)
53 (29.4)
127 (70.5)
470 (46.9), 430 (42.9), 102 (10.2)
178 (42.9), 200 (48.2), 37 (8.9)
292 (49.7), 230 (39.2), 65 (11.1)
0.01
578(63.1), 297 (32.4), 41(4.5)
234 (59.8), 142 (36.3), 15 (3.8)
344 (65.5), 155 (29.5), 26 (5)
0.08
Response rate
RVR (n=491) 0.0001
Liver fibrosis (n=605) 0.5
Steatosis (n=605) 0.02
IFNL3 rs12979860 (%) CC, CT, TT IFNL3 rs8099917 (%) TT, TG, GG
Data are given as median (range) or as %.
Table 2. Univariate analysis of factors associated with sustained virologic response.
Variable
SVR
Non- SVR
(N=736)
(n=266)
Univariate Analysis
Unadjusted odds ratio
p value
Age (yrs)
41.77(18-69)
46 (21-69)
0.0001
1.25 (1.02-1.98) 0.01
Male Gender (%)
395(53.7)
165(62)
0.02
1.28 (1.23-2.24) 0.01
HCV genotype 3 (%)
439(59.6)
148(55.6)
0.2
1.17 (0.88-1.56) 0.2
Body Mass Index (Kg/m2)
25 (18-32)
25.39 (19-34)
0.7
0.98 (0.93-1.03) 0.4
ALT (U/L)*
74 (12-389)
80 (11-358)
0.7
1.001 (0.997-1.005) 0.7
AST (U/L)*
54 (11-267)
58 (16-279)
0.3
1.002 (0.995-1.008) 0.7
Platelet (x109/L)
207.02± 69.77
202.74± 69.91
0.4
1.15 (0.24-1.67) 0.3
HCV-RNA log10 (n=863)*
5.74 (1.11-7.95)
5.92 (2.93-7.48)
0.01
1.96 (1.24-3.12) 0.004
None/mild
216 (57.1)
138 (60.7)
0.3
Moderate/ severe
162 (42.9)
89 (39.3)
None/mild
225 (52.6)
56 (31.6)
Moderate/ severe
203(47.4)
121 (68.4)
Abcent/mild
200 (73.2)
225 (67.7)
Moderate/ severe
73 (26.8)
107 (32.3)
CC
385 (52.3)
85 (32)
CT-TT
351 (47.7)
181 (68)
Inflamamtion score (%) (n=605) 0.94 (0.68-1.31)
0.7
Fibrosis stage (%) (n= 605) 0.0001
2.19 (1.65-4.46) 0.006
0.1
1.3 (0.91-1.85) 0.1
0.0001
2.4(1.73-3.13) 0.0001
Steatosis (%) (n=605)
IFNL3 rs12979860 (%)
IFNL3 rs8099917 (%) TT
451 (66.9)
127 (52.5)
GT/GG
223 (33.1)
115 (47.5)
RVR
295 (77.8)
51(45.5)
NON-RVR
84 (22.5)
61 (54.5)
0.0001
1.85 (1.35-2.46) 0.0001
0.0001
4.2 (2.69-6.54) 0.0001
RVR (%) (n=491)
Data are given as mean ± standard deviation, median and range or as %.
Table 3. Multivariate analysis of factors associated with sustained virologic response.
Model 1
Model 2
Multivariate analysis
Multivariate analysis
Variable
OR (95% CI) p value
OR (95% CI) p value
Age (yrs)
1.06 (1.002-1.46) 0.04
1.05 (1.003-1.22) 0.04
Male Gender (%)
1.23 (0.62-2.46) 0.5
1.27 (0.51-2.25) 0.8
HCV-RNA log10
1.45(1.05-2.22) 0.01
1.58 (1.09-2.52) 0.01
1.65 (1.13-3.27) 0.01
1.71 (1.18-3.6) 0.01
2.39(1.19-3.81) 0.0001
-
Fibrosis stage (%) None/mild Moderate/ severe
IFNL3 rs12979860 (%) CC /CT-TT
IFNL3 1.85 (1.15-2.23) 0.0001
rs8099917 (%) TT/GT/GG
-
RVR (%) RVR/ NON-RVR
3.25 (1.67-6.32) 0.0001
3.27 (1.56-6.78) 0.0001
Two multivariate logistic regression models were developed. Each included only one of the IFNL3 SNPs as the 2 SNPs are in linkage disequilibrium. Recruitment center was included as a variable in all the univariate and multivariate analyses.
S0168-8278(14)00269-4 http://dx.doi.org/10.1016/j.jhep.2014.03.039 JHEPAT 5118
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
13 August 2013 14 March 2014 30 March 2014
Please cite this article as: Eslam, M., Leung, R., Romero-Gomez, M., Mangia, A., Irving, W.L., Sheridan, D., Spengler, U., Mollison, L., Cheng, W., Bugianesi, E., McLeod, D., Zaitoun, A.M., Attino, V., Goeltz, D., Nattermann, J., Douglas, M., Booth, D.R., George, J., Ahlenstiel, G., IFNL3 polymorphisms predict response to therapy in chronic hepatitis C genotype 2/3 infection, Journal of Hepatology (2014), doi: http://dx.doi.org/10.1016/ j.jhep.2014.03.039
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
IFNL3 polymorphisms predict response to therapy in chronic hepatitis C genotype 2/3 infection Mohammed Eslam1, Reynold Leung1,2, Manuel Romero-Gomez3, Alessandra Mangia4, William L. Irving5, David Sheridan6, Ulrich Spengler7, Lindsay Mollison8, Wendy Cheng9, Elisabetta Bugianesi 10, Duncan McLeod11, Abed M. Zaitoun12, Vito Attino13, Diane Goeltz14, Jacob Nattermann7, Mark Douglas1, David R. Booth2, Jacob George1, Golo Ahlenstiel1 1
Storr Liver Unit, Westmead Millennium Institute and Westmead Hospital, University of Sydney, NSW, Australia 2 Institute of Immunology and Allergy Research, Westmead Hospital and Westmead Millennium Institute, University of Sydney, NSW, Australia. 3 Unit for The Clinical Management of Digestive Diseases and CIBERehd, Hospital Universitario de Valme, Sevilla, Spain. 4 Division of Hepatology, Ospedale Casa Sollievo della Sofferenza, IRCCS, San Giovanni Rotondo, Italy 5 NIHR Biomedical Research Unit in Gastroenterology and the Liver, University of Nottingham, Nottingham, UK 6 Liver Research Group, Institute of Cellular Medicine, Medical School, Newcastle University, Newcastle upon Tyne, UK 7 Department of Internal Medicine I, University of Bonn, Sigmund-Freud- Strasse, Bonn, Germany 8 Fremantle Hepatitis Services, Fremantle, Australia 9 Department of Gastroenterology and Hepatology, Royal Perth Hospital, Western Australia 10 Division of Gastro-Hepatology, S. Giovanni Battista Hospital, Turin, Italy. 11
Department of Anatomical Pathology, Institute of Clinical Pathology and Medical Research
(ICPMR), Westmead Hospital, Sydney, Australia. 12
Department of Histopathology, University Hospital Queens Medical Centre, Nottingham, UK Department is Anatoma Patologica IRCCS, Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy 14 Pathologisches Institute, Universitaetsklinikum Bonn, Germany 13
Corresponding Author Jacob George Department of Medicine, Westmead Hospital Westmead, NSW 2145 Ph: 61-2-98457705; Fx 61-2-96357582 Email: [email protected] 1
Running title: IFNL3 in HCV genotypes 2 and 3 Key words: Chronic hepatitis C, IFNL3 (IL28B), SVR, genotype 2, 3, response to therapy Electronic word count: 4951 words [including the abstract, references, tables, and figure legends] Number of figures and tables: 3 tables, 7 figures. List of abbreviations in the order of appearance: SNPs: Single nucleotide polymorphisms, IFNL3: interferon lambda 3, SVR: sustained virologic response, HCV-2/3: Hepatitis C virus genotype 2 and 3, RVR: rapid virological response, OR: odds ratio, Peg-IFNα/RBV: pegylated interferon-alpha and ribavirin, CHC: chronic hepatitis C, BMI: body mass index, ALT: Alanine aminotransferase, AST: aspartate aminotransferase, γGT: gamma-glutamyl-transferase, SD: standard deviation,
Conflict of interest: None. Financial support: GA, MD and JG are supported by National Health and Medical Research Council (NHMRC) Program and Project grants (1053206 and 1006759) and the Robert W. Storr bequest to the Sydney Medical Foundation of the University of Sydney. ME is supported by an International Postgraduate Research Scholarships (IPRS) and an Australian Postgraduate Award (APA) of the University of Sydney.
2
ABSTRACT (word count: 245) Background & aim: Single nucleotide polymorphisms (SNPs) near the interferon lambda 3 (IFNL3, previously known as IL28B) region are the strongest baseline predictors of sustained virologic response (SVR) to pegylated interferon and ribavirin therapy in hepatitis C virus (HCV) genotype 1 infection. Whether IFNL3 SNPs influence treatment response in genotype 2 and 3 (HCV-2/3) infection remains controversial. This study sought to clarify in a large cohort, whether SNPs in the IFNL3 region are associated with treatment response in HCV-2/3 patients. Methods: The cohort comprised 1002 HCV-2/3 Caucasians patients treated with pegylated interferon-alpha and ribavirin who underwent genotyping for the SNPs rs12979860 and rs8099917. Results: Overall, 736 (73.5%) patients achieved SVR (81.9%, 67.9%, and 57.8% for rs12979860 CC, CT, and TT [P=0.0001]; 78%, 68.7%, and 46.3% for rs8099917 TT, TG, and GG [P=0.0001]). By logistic regression, both rs12979860 CC and rs8099917 TT were independent predictors of SVR with an odds ratio (OR) of 2.39 (1.19-3.81) P= 0.0001 and OR 1.85 (1.15-2.23) P=0.0001, respectively. IFNL3 responder genotypes were more frequent in relapsers than null-responders (P=0.0001 for both SNPs). On-treatment rapid virological response (RVR) was predictive of SVR only in those individuals with IFNL3 non-responder genotypes (rs12979860 CT/TT and rs8099917 TG/GG). Conclusion: This adequately powered study in patients with HCV genotypes 2 or 3 infection clearly demonstrates that IFNL3 genotypes are the strongest baseline predictor of SVR, in keeping with the known association for genotype 1 infection. IFNL3 genotyping can aid in therapeutic decision making for these patients.
3
INTRODUCTION Hepatitis C virus (HCV) infects 170 million people worldwide [1] and is a leading cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma [2]. Treatment with pegylated interferon-alpha and ribavirin (Peg-IFNα/RBV) results in a sustained virologic response (SVR) in approximately 50% of people infected with HCV genotype 1 (HCV-1) and in ~75% of those infected with genotypes 2 or 3 (HCV-2/3) [3,4]. Although the addition of protease inhibitors such as telaprevir and boceprevir leads to a substantial improvement in the SVR rate for HCV-1 infection [5], Peg-IFNα/RBV remains the standard of care for non-1 genotypes [5]. In this context, there is growing interest in stratifying patients based on pre-treatment predictors of response, in order to have abbreviated treatment regimens both to minimize cost and more importantly, to reduce the adverse effects of interferon-based therapies. The discovery of polymorphisms near the IFNL3 (formerly known as IL28B) gene, which codes for interferon lambda 3, represented a milestone in hepatitis C research [6,7,8]. Four years later, the value of polymorphisms in this region for predicting response to Peg-IFNα/RBV treatment in HCV-1 is well established [6-8]. In contrast, similar data for HCV-2/3 infection remains contentious [9,10,11], with results of two recent meta-analyses contradicting each other [12,13]. This controversy quite likely reflects the relatively small sample sizes of the previous studies and differences in ethnic stratification, which critically determines the frequencies of observed IFNL3 genotypes [9-11]. The aims of this study were 1) to clarify the role of pre-treatment IFNL3 polymorphisms for predicting SVR in HCV-2/3 patients, 2) to explore the predictors of response in patients with the IFNL3 responder genotype and 3) to clarify the distribution of IFNL3 in non-SVR patients. To address these questions, we analysed a large cohort of patients with genotype 2 or 3 chronic hepatitis C (CHC) infection and known treatment outcomes. 4
PATIENTS AND METHODS Patient cohort The cohort comprised 1002 Caucasians patients with CHC from Australia (N=251), Spain (N=261), Italy (N=235) , the United Kingdom (N=199), and Germany (N=56), fulfilling the following inclusion criteria: (a) adults aged 18 years or older with CHC based on the presence of anti-HCV and detectable serum HCV-RNA for >6 months, (b) infection with HCV-2 or 3 and (c) known outcome following Peg-IFNα/RBV based therapy. Patients were excluded if they were co-infected with either hepatitis B virus or HIV or had other liver diseases as assessed by standard tests. 89 of the patients from this study have been included in previous reports [14]. All patients were treated with Peg-IFNα/RBV; the duration of therapy and stopping rules employed were according to standard guidelines [5]. SVR was defined as undetectable HCV RNA 24 weeks after completion of therapy. Rapid virologic response (RVR) was defined as undetectable HCV RNA levels after 4 weeks of therapy. The non-SVR cohort included patients with either non-response or relapse. Virologic relapse was defined as HCV RNA undetectable at the end-of-treatment but positive thereafter, whereas virologic non-response was defined as HCV RNA detectable throughout the entire therapy of at least 24 weeks. Ethical approval was obtained from the Human Research Ethics Committees of the Sydney West Area Health Service and the University of Sydney. All other sites had ethical approval from their respective ethics committees. Written informed consent for genetic testing was obtained from all participants. The study was conducted in accordance with the international ethical guidelines of the International Conference on Harmonization Guidelines for Good Clinical Practice [15].
5
Clinical and laboratory assessment The following data were collected at baseline: gender, age, ethnicity (Caucasian Vs. nonCaucasian), recruitment center, body mass index (BMI), and routine laboratory tests. BMI was calculated as weight divided by the square of the height (kg/m2). Alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl-transferase (γGT), bilirubin, hemoglobin, leukocyte and platelet count were determined by routine laboratory techniques. Liver biopsy Data on liver fibrosis at biopsy was available in 605 patients. Biopsies were interpreted according to the scoring schema developed by the METAVIR group [16] by a single expert liver pathologists in each center who was blinded to patient clinical characteristics and serum measurements. Thirty-five biopsies were scored independently by pathologists (DM, AZ, DG and VA) from the various centers, and inter observer agreement was calculated by using the κ statistic. Fibrosis was scored on a 5-point scale: F0, no fibrosis; F1, portal fibrosis alone; F2, portal fibrosis with rare septae; F3, portal fibrosis with many septae; F4, cirrhosis. The presence of stage F2, F3, or F4 was termed “significant fibrosis”, whereas the term “advanced fibrosis” was reserved for stage F3 or F4. Necro-inflammatory activity, based on assessment of piecemeal and lobular necrosis, was graded on a 4-point scale: A0, no activity; A1, mild; A2, moderate; A3, severe. IFNL3 Genotyping Genotyping for IFNL3 SNPs was undertaken using the TaqMan SNP genotyping allelic discrimination method (Applied Biosystems, Foster City, CA, USA). The rs8099917 genotyping kit was supplied by Applied Biosystems and rs12979860 genotyping was performed using a custom designed genotyping assay by Applied Biosystems. Detailed procedures have been described previously [17]. Genotyping was performed using the StepOne RT system and 6
analyzed with StepOne software v.2.3.0 (Applied Biosystems, Foster City, CA, USA). All genotyping was blinded to treatment outcome. Statistical Analysis Quantitative data are expressed as mean ± SD (standard deviation), and categorical data as number (percentage) of patients. Skewed variables are reported as median and range. The Student t-test or non-parametric, i.e. Wilcoxon-Mann-Whitney U-test or Kruskal-Wallis tests were used to compare quantitative data, while appropriate. χ2 and Fisher-exact tests were used for comparison of frequency data and to evaluate the relationships between groups. All tests were two-tailed and p values <0.05 were considered significant. Multiple regression models were used to assess for factors independently associated with RVR and SVR. IFNL3 SNP comparisons were made using a dominant model, in which patients carrying one allele; or a recessive model (two copies of the minor allele), were compared with others unless otherwise indicated. Necroinflammation score was dichotomized as (absent/ mild necroinflammation or minimal changes (Metavir score A0-A1) or (presence of moderate/severe necroinflammation (Metavir score A2-A3). Fibrosis stage was dichotomized into two groups; absent or mild fibrosis (Metavir score F0-1) and significant fibrosis (Metavir score F2-4). To assess the cohort size required to detect significant differences, retrospective power analysis was performed, a power curve was plotted for power, sample size and odds ratio. All analyses were carried out using the statistical software package SPSS for Windows, version 14 (SPSS, Chicago, IL) and SAS version 9.1 and SAS Enterprise 9.4
7
RESULTS The characteristics of the study cohort are shown in Table 1. The median age was 43 (range: 19 - 69) years, with 560 (55.9%) patients being male. Of the cohort, 587 (58.6%) were infected with HCV-3, while the remaining had infection with HCV-2. The interobserver agreement between pathologists was good (κ = 77.5) for METAVIR staging. An SVR was achieved in 736 (73.5%) patients; all others were classified as having a non-SVR. The latter included those with non-response (N=139) and those with relapse (N=127).
IFNL3 genotype distribution Genotype distribution of the two IFNL3 SNPs was in Hardy–Weinberg equilibrium (data not shown). The overall genotype distribution of IFNL3 rs12979860 CC, CT, and TT was 46.9%, 42.9%, and 10.2%, and the distribution of rs8099917 TT, TG, and GG was 63.1%, 32.4%, and 4.5%, respectively. 86 patients had no rs8099917 genotype due to a lack of sufficient DNA for genotyping. The characteristics of those patients were matched to that of the whole cohort. SVR rates were 81.9%, 67.9%, and 57.8% for rs12979860 CC, CT, and TT (P=0.0001) and 78%, 68.7%, and 46.3% for rs8099917 TT, TG, and GG (P=0.0001) (Figure 1).
Baseline factors associated with a sustained virological response By univariate analysis, apart from IFNL3 rs12979860 and rs8099917, SVR was associated with previously reported baseline factors including younger age, female gender, lower HCV RNA level and none/mild fibrosis. Of the on treatment factors, as expected, RVR was a strong predictor of subsequent SVR (Table 2).
8
Adjusting for the covariates of age, sex, recruitment center, HCV RNA level, RVR, and the stage of histological fibrosis, two multivariate logistic regression models were developed. Each included only one of the IFNL3 SNPs as the 2 SNPs are in linkage disequilibrium [18]. In the first model, all significant covariates from the univarite analysis were included except for rs8099917. rs12979860 CC showed an odds ratio (OR) for an association with SVR (CC versus CT/TT) of 2.39 (1.19 - 3.81), P= 0.0001). The second regression model contained all covariates, except for rs12979860. Here, rs8099917 TT, was significantly associated with SVR (TT versus TG/GG: OR 1.85 (1.15 - 2.23) P=0.0001) (Table 2). Of note, all models were also calculated for pretreatment predictors alone, i.e. without RVR, and the results were similar (data not shown). To explore if the conflicting results reported in the literature could be due to the small sample sizes of previous studies, we undertook a retrospective power analysis. Notably, cohorts of >300 patients (for rs12979860) and >400 patients (for rs8099917) were required in order to have 90% power to detect an odds ratio of 2 and 1.5, respectively with a P-value (< 0.05) for the differences in SVR rates of patients with and without the responder genotype (Figure 2). This indicates that our cohort was sufficiently powered to detect associations in this patient group.
IFNL3 and treatment failure With respect to the correlation of IFNL3 with treatment failure, non-SVR rates were 22%, 31.3%, and 53.7% for rs8099917 TT, TG, and GG (P=0.001), and 18.1%, 32.1%, and 44.2% for rs12979860 CC, CT, and TT (P=0.001). Based on multiple logistic analysis to adjust for possible cofounders, the minor allele homozygotes for rs8099917 had the highest predictive value, GG: OR 3.45 (95%CI: 1.83-6.48), P=0.0001, which was higher than for the rs12979860 TT homozygote (OR: 2.21 (95%CI: 1.45-3.7), P=0.0002).
9
The predictive value for
heterozygotes rs8099917 GT and rs12979860 CT was similar (OR 1.43 (95%CI: 1.05-1.95), and (OR 1.63 (95%CI: 1.23-2.17), respectively, P< 0.01, for both). IFNL3 and rapid virologic response RVR rates for the rs12979860 CC, CT, and TT genotypes were 74.4%, 68.9%, and 60.7% (p=0.09), while for rs8099917, TT and TG were more likely to achieve RVR than GG (74.3% and 70.1% vs. 51.7%, p=0.03; Figure 3). In multiple regression analysis corrected for other variables significantly associated with RVR in univariate analysis, i.e. age, HCV-genotype and baseline viral load, rs8099917 GG remained significantly associated with failure to achieve RVR (OR: 1.98 (95% CI: 1.27-3.52), P=0.001). The negative correlation of GG with RVR translated to lower SVR rates in this group as compared to rs12979860 TT (46.3% vs. 57.8%, p=0.001).
Differential impact of IFNL3 genotypes in HCV genotype 2 versus 3 infection The baseline characteristics of patients infected with HCV-2 (n=415) or 3 (n=587) are summarized separately in Table 1. The responder genotype distribution of both SNPs of IFNL3 was higher in HCV-3 as compared to HCV-2 patients (Table 1). RVR, but not SVR, was more frequent in HCV-2 than HCV-3 (RVR: 79.5% vs. 63.2%, P=0.0001; SVR: 71.6% vs. 74.8%, P=0.2). As for the relationship between the IFNL3 SNPs and treatment response, both rs12979860 and rs8099917 showed significant associations with SVR in both HCV-3 and HCV2 patients (Figure 4). Both IFNL3 SNPs also remained independently correlated with SVR in multivariate analysis of HCV-3 patients (OR 2.71 (95% CI: 1.83-4.2), P=0.0001 and OR 1.97 (95% CI: 1.31-2.96), P=0.001, for rs12979860 and rs8099917, respectively). In HCV-2 patients, the results were similarly significant (OR 1.88 (95% CI: 1.19-2.93), P=0.005 and OR 1.63 (95% CI: 1.05-2.54), P=0.02 for rs12979860 and rs8099917, respectively). 10
Predictors of SVR in patients with IFNL3 responder genotypes In patients with rs12979860 CC, i.e. the responder genotype, younger age, none/mild fibrosis, lower baseline HCV-RNA log10, higher platelets counts and RVR were associated with SVR by univariate analysis, while only lower HCV-RNA log10 remained independently correlated with SVR in multivariate analysis (OR 2.66 (95% CI: 1.49-8.48), P=0.001). Separate analysis using the median of HCV-RNA log10 (5.89 IU/mL) yielded similar results. In patients with rs12979860 non-responder genotypes (CT/TT), the independent predictors of SVR were RVR (OR 4.8 (95% CI: 2.1-8.79), P=0.0001) and HCV-RNA log10 (OR 2.34 (95% CI: 1.374.45), P=0.001). In patients with rs8099917 TT, SVR was also associated with younger age, female gender, lower basal HCV-RNA log10 level, none or mild fibrosis and RVR. Only lower baseline HCV-RNA log10 level (OR 2.42 (95% CI: 1.86-5.12), P=0.001) was independently correlated with SVR in multivariate analysis.
Predictive value of IFNL3 in non-sustained virologic response group. Finally, we studied the distribution of IFNL3 genotypes among the non-SVR group (n=266), comparing non-responders (N=139) and relapsers (N=127). The responder genotypes rs12979860 CC and rs8099917 TT were more frequent in relapsers than in non-responders (P=0.0001 for both) (Figure 5).
11
DISCUSSION Despite new direct acting antiviral drugs for HCV-1 infection, Peg-IFNα/RBV remains the standard of care for patients with HCV-2/3, and will remain so for some time in less privileged parts of the world [5,19]. In this context, reliable pre-treatment predictors associated with a favorable outcome can improve cost-effectiveness and support clinical decision making [20]. In the present study, we clarify the debate regarding the impact of IFNL3 SNPs rs12979860 and rs8099917 on outcomes in patients with HCV genotype 2 or 3 infection. Using the largest patient cohort to date (N=1002), we demonstrate that IFNL3 SNPs are the strongest baseline predictors of treatment response in this group, though only rs8099917 GG was associated with lower RVR rates. Our retrospective power analysis provides insights into the controversy arising from previously published reports: Moghaddam et al. for example reported that rs12979860 and rs8099917 were not associated with SVR to Peg-IFNα/RBV therapy in 281 patients infected with HCV-3 [9] and Rauch A, et al. similarly reported that rs8099917 was not associated with response in 303 HCV-2/3 [21].
In contrast, Mangia et al. reported that in their cohort of 268
patients, IFNL3 rs12979860 genotype was only associated with SVR in HCV-2/3 patients who did not achieve RVR [11]. Sarrazin et al. reported that only rs12979860 but not rs8099917 was associated with SVR in a cohort of 267 HCV-2/3 patients [22]. Recently, Stenkvist et al. showed in a cohort of 100 HCV-2/3 patients a significantly higher median 1st phase decline in HCVRNA levels in patients with rs12979860 CC than non-CC genotype patients [23]. Here we demonstrate that an impact of rs12979860 or rs8099917 on SVR rates is only reliably observed in cohorts of >300 and 400 patients, respectively. This suggests that previous reports were not adequately powered and that the divergent results principally reflect sample size differences between cohorts. Consistent with this observation, in a large cohort from Japan (n=719 patients, 12
with either genotype 2a or 2b), in the sub-cohort of patients treated with Peg-IFNα/RBV (n=160), IFNL3 polymorphisms were not associated with SVR. However, in the subset treated with interferon monotherapy (n=559), rs8099917 independently predict SVR [24]. The current larger cohort study assessed Caucasian patients with both genotype 2/3 infection and all were treated with Peg-IFNα/RBV. In keeping with data from HCV-1, rs8099917 has better correlation with treatment failure than rs12979860 [18]. This is due to rs12979860 C tagging two distinct haplotypes (43% and 10% of alleles in CHC with HCV-1), which are both responsive (odds ratios of 0.7 and 0.6 for treatment failure), whereas rs12979860 T includes a mixed group of haplotypes (24%, 10%, 2% and 1% of alleles; ORs of 2.2, 1.0, 0.8 and 1.5, respectively for treatment failure) [18]. Thus, rs12979860 tagged by a C allele will always improve the chances of response, whereas with rs12979860 tagged by a T, it depends on the particular haplotype. In contrast, only one distinct IFNL3 haplotype, haplotype 2, is tagged by rs8099917 G, which is strongly associated with treatment failure. rs8099917 G’s have an OR of 2.2 in this data for treatment failure [18]. In contrast, only rs12979860 T’s, which are also rs8099917 G, have an OR of 2.2 for treatment failure, the remaining haplotypes have neutral odds ratios. By the same token, rs8099917 T’s shows a mixed picture with some responding and others not, as unlike rs12979860 C’s, not all rs8099917 T-tagged haplotypes are responders, i.e. 20% of rs8099917 T’s are neutral responders (compared with 0% for rs12979860 C) [18]. Thus, from clinical point of view, both SNPs play a complimentary role with regard to prediction of response. A fourth type 3 interferon (IFNL4) has recently been described from primary human hepatocytes [25] that better predicts response/non response to dual therapy in genotype 1 CHC. However, it is important to note that SS4694115590 likewise, does not explain the difference in haplotype-based responses, as
13
acknowledged by the authors of that report [25]. Further studies are required to better understand the mechanisms for the IFNL3 haplotype-based responses. Given that SVR rates in patients carrying the responder genotype was 82% for rs12979860 CC and 78% for rs8099917 TT, we analyzed the potential predictors for treatment response in these patients. For rs12979860 CC genotypes (~44% of Caucasian subjects in this cohort), only baseline HCV-RNA log10 was independently correlated with SVR, whereas for non-responder genotypes (CT/TT), RVR and baseline HCV-RNA log10 independently predicted SVR. IFNL3 responder genotypes were more frequent in relapsers than non-responders. This is consistent with the established better SVR rate following retreatment of relapsers as compared to non-responders in HCV genotype 1 infection [26,27] and underlines the fact that IFNL3 genotype is an important marker of IFN sensitivity. Hence, IFNL3 genotyping may have a particular role in identifying patients with HCV-2/3 for re-treatment, especially in scenarios where interferon containing regimens remain the standard of care in cost-constrained economies. In conclusion, we show that IFNL3 genotypes are associated with treatment response to Peg-IFNα/RBV in HCV-2/3 patients, similar to the effect observed in HCV-1. Importantly, the data indicates that confounders including cohort size explain to a large degree the controversy from previous reports. The predictive value of RVR is more obvious in those with IFNL3 nonresponder, than in those with the responder genotype. Although new treatments may become available in the future [28], IFNL3 genotyping can aid in clinical decision making for patients with HCV-2/3 infection, and for patient stratification to abbreviated therapy trials or for retreatment.
14
References
[1] World Health Organization . Hepatitis C. Fact Sheet Number 164, Revised October 2000. (http://www.who.int/mediacentre/factssheets/fs164len) Accessed 6 March 2012. [2] Seeff LB. The history of the "natural history" of hepatitis C (1968-2009). Liver Int. 2009;29 Suppl 1:89-99 [3] Manns MP, McHutchison JG, Gordon SC, Rustgi VK, Shiffman M, Reindollar R, et al. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 2001;358:958–965 [4] Hadziyannis SJ, Sette Jr H, Morgan TR, Balan V, Diago M, Marcellin P, et al. Peginterferonalpha2a and ribavirin combination therapy in chronic hepatitis C: a randomized study of treatment duration and ribavirin dose. Ann Intern Med 2004;140:346–355. [5] Ghany MG, Nelson DR, Strader DB, Thomas DL, Seeff LB; American Association for Study of Liver Diseases. An update on treatment of genotype 1 chronic hepatitis C virus infection: 2011 practice guideline by the American Association for the Study of Liver Diseases. Hepatology. 2011;54(4):1433-44. [6] Ge D, Fellay J, Thompson AJ, Simon JS, Shianna KV, Urban TJ, et al. Genetic variation in IL28B predicts hepatitis C treatment-induced viral clearance. Nature 2009; 461:399–401. [7] Suppiah V, Moldovan M, Ahlenstiel G, Berg T, Weltman M, Abate ML, et al. IL28B is associated with response to chronic hepatitis C interferon-alpha and ribavirin therapy. Nat Genet 2009; 41:1100–1104. [8] Tanaka Y, Nishida N, Sugiyama M, Kurosaki M, Matsuura K, Sakamoto N, et al. Genomewide association of IL28B with response to pegylated interferon-alpha and ribavirin therapy for chronic hepatitis C. Nat Genet 2009;41:1105–1109. [9] Moghaddam A, Melum E, Reinton N, Ring-Larsen H, Verbaan H, Bjøro K, et al. IL28B genetic variation and treatment response in patients with hepatitis C virus genotype 3 infection. Hepatology 2011;53 (3):746–754. [10] Yu ML, Huang CF, Huang JF, Chang NC, Yang JF, Lin ZY, et al. Role of interleukin-28B polymorphisms in the treatment of hepatitis C virus genotype 2 infection in Asian patients. Hepatology 2011;53 (1):7–13. 15
[11] Mangia A, Thompson AJ, Santoro R, Piazzolla V, Tillmann HL, Patel K, et al. An IL28B polymorphism determines treatment response of hepatitis C virus genotype 2 or 3 patients who do not achieve a rapid virologic response. Gastroenterology 2010; 139: 821–7. [12] Jia Z, Ding Y, Tian S, Niu J, Jiang J. Test of IL28B polymorphisms in chronic hepatitis C patients treated with PegIFN and ribavirin depends on HCV genotypes: results from a metaanalysis. PLoS One. 2012;7(9):e45698. doi: 10.1371/journal.pone.0045698. [13] Schreiber J, Moreno C, Garcia BG, Louvet A, Trepo E, Henrion J, et al. Meta-analysis: the impact of IL28B polymorphisms on rapid and sustained virological response in HCV-2 and -3 patients. Aliment Pharmacol Ther. 2012; 36(4):353-62. [14] Bucci C, von Delft A, Christian A, Flemming VM, Harrison A, Halliday J, et al. 'Favourable' IL28B polymorphisms are associated with a marked increase in baseline viral load in hepatitis C virus subtype 3a infection and do not predict a sustained virological response after 24 weeks of therapy. J Gen Virol. 2013;94(Pt 6):1259-65. [15] Dixon JR Jr. The International Conference on Harmonization Good Clinical Practice guideline. Qual Assur. 1998;6(2):65-74. [16] Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. The
METAVIR Cooperative Study Group. Hepatology 1996;24:289-293. [17] Suppiah V, Gaudieri S, Armstrong NJ, O'Connor KS, Berg T, Weltman M, et al. IL28B, HLA-C, and KIR variants additively predict response to therapy in chronic hepatitis C virus infection in a European Cohort: a cross-sectional study. PLoS Med. 2011;8(9):e1001092. doi: 10.1371/journal.pmed [18] Smith KR, Suppiah V, O'Connor K, Berg T, Weltman M, Abate ML, et al. Identification of improved IL28B SNPs and haplotypes for prediction of drug response in treatment of hepatitis C using massively parallel sequencing in a cross-sectional European cohort. Genome Med. 2011;3(8):57. [19] Foster GR, Hézode C, Bronowicki JP, Carosi G, Weiland O, Verlinden L, et al. Telaprevir alone or with peginterferon and ribavirin reduces HCV RNA in patients with chronic genotype 2 but not genotype 3 infections. Gastroenterology. 2011; 141:881-889. [20] Shiffman ML. Retreatment of patients with chronic hepatitis C. Hepatology 2002;36:S128– S134. 16
[21] Rauch A, Kutalik Z, Descombes P, Cai T, Di Iulio J, Mueller T, et al. Genetic variation in IL28B is associated with chronic hepatitis C and treatment failure: a genome-wide association study. Gastroenterology. 2010; 138(4):1338-45, 1345.e1-7. [22] Sarrazin C, Susser S, Doehring A, Lange CM, Müller T, Schlecker C, et al. Importance of IL28B gene polymorphisms in hepatitis C virus genotype 2 and 3 infected patients. J Hepatol. 2011;54(3):415-21. [23] Stenkvist J, Sönnerborg A, Weiland O. HCV RNA decline in chronic HCV genotype 2 and 3 during standard of care treatment according to IL28B polymorphism. J Viral Hepat. 2013; 20(3):193-9. [24] Kawaoka T, Hayes CN, Ohishi W, Ochi H, Maekawa T, Abe H, et al. Predictive value of the IL28B polymorphism on the effect of interferon therapy in chronic hepatitis C patients with genotypes 2a and 2b. J Hepatol. 2011; 54(3):408-14. [25] Prokunina-Olsson L, Muchmore B, Tang W, Pfeiffer RM, Park H, Dickensheets H, et al. A variant upstream of IFNL3 (IL28B) creating a new interferon gene IFNL4 is associated with impaired clearance of hepatitis C virus. Nat Genet. 2013;45(2):164-71. [26] Shiffman ML, Di Bisceglie AM, Lindsay KL, Morishima C, Wright EC, Everson GT, et al. Peginterferon alfa-2a and ribavirin in patients with chronic hepatitis C who have failed prior treatment. Gastroenterology 2004, 126:1015–1023. [27] Jacobson IM, Gonzales SA, Ahmed F, Lebovics E, Min AD, Bodenheimer HC, et al. A randomized trial of pegylated interferon a-2b plus ribavirin in the retreatment of chronic hepatitis C. Am J Gastroenterol 2005; 100:2453–2462. [28] Gane EJ, Stedman CA, Hyland RH, Ding X, Svarovskaia E, Symonds WT, et al. Nucleotide polymerase inhibitor sofosbuvir plus ribavirin for hepatitis C. N Engl J Med. 2013; 368(1):34-44.
17
FIGURE LEGENDS
Figure 1: Association between IFNL3 rs12979860 and rs8099917 genotypes and SVR (P=0.0001 for both).
Figure 2: Retrospective power analysis. The power curve represents the correlation between statistical power, sample size and odds ratio. Cohorts of >300 patients (for rs12979860) and >400 patients (for rs8099917) are required in order to have 90% power to detect an odds ratio of 2 and 1.5, respectively with P-value < 0.05 for the difference between SVR rates of patients with and without the responder genotypes
Figure 3: Association between IFNL3 rs12979860 and rs8099917 genotypes and RVR (P=0.09 and P=0.03, respectively).
Figure 4: Association between IFNL3 rs12979860 and rs8099917 genotypes and SVR stratified by HCV genotypes (HCV-3 and 2).
Figure 5: Distribution of IFNL3 rs12979860 and rs8099917 genotypes in non-SVR patients (Relapser vs. Null-responder) (P=0.0001 for both).
Table 1. Baseline demographic characteristics and treatment response
Variables
All patients (n=1002)
HCV-2
HCV-3
P-value
(n=415)
(n=587)
(HCV-2 vs. HCV3)
Age
43 (19-69)
46 (19-69)
41 (18-69)
0.0001
Male (%)
560(55.9)
243(58.6)
317(54)
0.1
BMI(Kg/m2)
25 (18-34)
25 (18-34)
25 (18-34)
0.1
HCV-RNA log10 (n=863)
5.89 (1.11-7.95)
5.9 (2.75-7.95)
5.88 (1.11-7.48)
0.7
SVR(%)
736 (73.5)
297 (71.6)
439(74.8)
0.2
Non-SVR(%)
266 (26.5)
118(28.4)
148(25.2)
RVR
346 (70.5)
174(79.5)
172(63.2)
NON-RVR
145 (29.5)
45(20.5)
100(36.8)
Absent/mild fibrosis
281 (46.4)
106(48.2)
175(45.2)
Moderate/severe fibrosis
324 (53.6)
114(51.8)
210 (54.5)
Absent/mild Steatosis
425(70.2)
167 (39.3)
258 (60.7)
Moderate/severe Steatosis
180(29.8)
53 (29.4)
127 (70.5)
470 (46.9), 430 (42.9), 102 (10.2)
178 (42.9), 200 (48.2), 37 (8.9)
292 (49.7), 230 (39.2), 65 (11.1)
0.01
578(63.1), 297 (32.4), 41(4.5)
234 (59.8), 142 (36.3), 15 (3.8)
344 (65.5), 155 (29.5), 26 (5)
0.08
Response rate
RVR (n=491) 0.0001
Liver fibrosis (n=605) 0.5
Steatosis (n=605) 0.02
IFNL3 rs12979860 (%) CC, CT, TT IFNL3 rs8099917 (%) TT, TG, GG
Data are given as median (range) or as %.
Table 2. Univariate analysis of factors associated with sustained virologic response.
Variable
SVR
Non- SVR
(N=736)
(n=266)
Univariate Analysis
Unadjusted odds ratio
p value
Age (yrs)
41.77(18-69)
46 (21-69)
0.0001
1.25 (1.02-1.98) 0.01
Male Gender (%)
395(53.7)
165(62)
0.02
1.28 (1.23-2.24) 0.01
HCV genotype 3 (%)
439(59.6)
148(55.6)
0.2
1.17 (0.88-1.56) 0.2
Body Mass Index (Kg/m2)
25 (18-32)
25.39 (19-34)
0.7
0.98 (0.93-1.03) 0.4
ALT (U/L)*
74 (12-389)
80 (11-358)
0.7
1.001 (0.997-1.005) 0.7
AST (U/L)*
54 (11-267)
58 (16-279)
0.3
1.002 (0.995-1.008) 0.7
Platelet (x109/L)
207.02± 69.77
202.74± 69.91
0.4
1.15 (0.24-1.67) 0.3
HCV-RNA log10 (n=863)*
5.74 (1.11-7.95)
5.92 (2.93-7.48)
0.01
1.96 (1.24-3.12) 0.004
None/mild
216 (57.1)
138 (60.7)
0.3
Moderate/ severe
162 (42.9)
89 (39.3)
None/mild
225 (52.6)
56 (31.6)
Moderate/ severe
203(47.4)
121 (68.4)
Abcent/mild
200 (73.2)
225 (67.7)
Moderate/ severe
73 (26.8)
107 (32.3)
CC
385 (52.3)
85 (32)
CT-TT
351 (47.7)
181 (68)
Inflamamtion score (%) (n=605) 0.94 (0.68-1.31)
0.7
Fibrosis stage (%) (n= 605) 0.0001
2.19 (1.65-4.46) 0.006
0.1
1.3 (0.91-1.85) 0.1
0.0001
2.4(1.73-3.13) 0.0001
Steatosis (%) (n=605)
IFNL3 rs12979860 (%)
IFNL3 rs8099917 (%) TT
451 (66.9)
127 (52.5)
GT/GG
223 (33.1)
115 (47.5)
RVR
295 (77.8)
51(45.5)
NON-RVR
84 (22.5)
61 (54.5)
0.0001
1.85 (1.35-2.46) 0.0001
0.0001
4.2 (2.69-6.54) 0.0001
RVR (%) (n=491)
Data are given as mean ± standard deviation, median and range or as %.
Table 3. Multivariate analysis of factors associated with sustained virologic response.
Model 1
Model 2
Multivariate analysis
Multivariate analysis
Variable
OR (95% CI) p value
OR (95% CI) p value
Age (yrs)
1.06 (1.002-1.46) 0.04
1.05 (1.003-1.22) 0.04
Male Gender (%)
1.23 (0.62-2.46) 0.5
1.27 (0.51-2.25) 0.8
HCV-RNA log10
1.45(1.05-2.22) 0.01
1.58 (1.09-2.52) 0.01
1.65 (1.13-3.27) 0.01
1.71 (1.18-3.6) 0.01
2.39(1.19-3.81) 0.0001
-
Fibrosis stage (%) None/mild Moderate/ severe
IFNL3 rs12979860 (%) CC /CT-TT
IFNL3 1.85 (1.15-2.23) 0.0001
rs8099917 (%) TT/GT/GG
-
RVR (%) RVR/ NON-RVR
3.25 (1.67-6.32) 0.0001
3.27 (1.56-6.78) 0.0001
Two multivariate logistic regression models were developed. Each included only one of the IFNL3 SNPs as the 2 SNPs are in linkage disequilibrium. Recruitment center was included as a variable in all the univariate and multivariate analyses.