Interleukin 28B polymorphism predicts pegylated interferon plus ribavirin treatment outcome in chronic hepatitis C genotype 4†
Article first published online: 3 JAN 2012
Copyright © 2012 American Association for the Study of Liver Diseases
Volume 55, Issue 2, pages 336–342, February 2012
How to Cite
De Nicola, S., Aghemo, A., Grazia Rumi, M., Galmozzi, E., Valenti, L., Soffredini, R., De Francesco, R., Prati, G. M., D'Ambrosio, R., Cheroni, C., Donato, M. F. and Colombo, M. (2012), Interleukin 28B polymorphism predicts pegylated interferon plus ribavirin treatment outcome in chronic hepatitis C genotype 4. Hepatology, 55: 336–342. doi: 10.1002/hep.24683
Potential conflicts of interest: Massimo Colombo, MD, Grant and research support: Merck, Roche, BMS, Gilead Science; Advisory committees: Merck, Roche, Novartis, Bayer, BMS, Gilead Science, Tibotec, Vertex, Achillion; Speaking and teaching: Tibotec, Roche, Novartis, Bayer, BMS, Gilead Science, Vertex. Maria Grazia Rumi, MD, Speaking and Teaching: Roche; Advisory committees: Jannsen; Travel support: Roche. Alessio Aghemo, MD, Advisory committees: Roche; Travel support: BMS, Glaxo Smith-Kline, Bayer. The other authors have no financial disclosures to declare.
- Issue published online: 27 JAN 2012
- Article first published online: 3 JAN 2012
- Accepted manuscript online: 19 SEP 2011 09:08AM EST
- Manuscript Accepted: 3 SEP 2011
- Manuscript Received: 18 JUN 2011
Single nucleotide polymorphisms (SNPs) near the interleukin 28B (IL28B) region are the strongest baseline predictors of a sustained virologic response (SVR) to peg-interferon (PegIFN) and ribavirin (Rbv) in patients with hepatitis C virus (HCV) genotype 1 infection. Whether this holds true for HCV-4 patients too is unknown. The aim was to investigate the predictive power of the rs12979860 IL28B SNP for a response to Peg-IFN and Rbv in HCV-4 patients. All HCV-4 patients consecutively treated between September 2004 and June 2010 with PegIFN and Rbv at two liver centers at the Maggiore Hospital Milan (Italy) underwent TaqMan SNP Genotyping assays for testing rs12979860 genotype. Of 112 treated patients (98 males, 75 of Egyptian descent, 26 with cirrhosis) 103 were included in the final analysis; five discontinued treatment for nonvirologic reasons and four did not consent to genetic testing. Twenty-four (23%) were genotype CC, 65 (63%) CT, and 14 (14%) TT. Overall, 50 (49%) achieved an SVR: 21 (88%) CC patients versus 29 (37%) CT/TT (P < 0.0001). CC patients more often had a rapid virologic response (RVR) than CT/TT patients (12, 50% versus 23, 29%; P = 0.08), while also showing lower relapse rates (0% [0/21] versus 36% [16/45] P = 0.0013). In non-RVR patients, SVR rates were higher in CC than CT/TT patients (9 [75%] versus 13 [23%] P = 0.001). By logistic regression, the IL28B rs12979860 CC genotype was an independent predictor of SVR with an odds ratio of 8.0 (95% confidence interval 2.00-32.01; P = 0.003). Conclusion: The IL28B rs12979860 SNP may have an added value in the treatment algorithm of HCV-4 patients because it is the strongest predictor of an SVR to PegIFN/Rbv therapy. (HEPATOLOGY 2012)
Chronic infection with hepatitis C virus (HCV) is a global health problem, affecting ≈3% of the world's population, and a major cause of anticipated liver-related death worldwide.1, 2 Pegylated interferon (PegIFN) and ribavirin (Rbv) therapy is the consolidated standard of care (SOC), with achievable rates of viral clearance that are largely influenced by virus and host-related factors.3, 4 Recently, single nucleotide polymorphisms (SNP) on chromosome 19 were found to predict treatment outcome in the more difficult to cure genotype 1 patients and spontaneous clearance in acutely infected individuals.5, 6 Although the SNPs are located 3 kb upstream of the interleukin 28B (IL28B) gene that encodes for interferon lambda-3, the exact biological mechanisms underlying this association are unknown. Still, the predictive power of IL28B SNPs in genotype 1 patients has been validated in many studies conducted in different geographical areas, to the point that this test has become part of the therapeutic algorithm.7-11 However, the role of IL28B as a moderator of treatment outcome in other HCV genotypes is less clear. Indeed, in HCV-2 and 3 patients the impact of IL28B polymorphisms on sustained virological response (SVR) rates are less pronounced, because acceptable rates of viral clearance have been documented even in the unfavorable IL28B genotypes, CT and TT.12 Relatively unknown is the role of rs12979860 in HCV-4 patients, as only two subgroup analyses of human immunodeficiency virus (HIV)-HCV coinfected and Egyptian patients addressed this topic, finding the IL28B CC genotype to be an independent predictor of an SVR to PegIFN plus Rbv.13, 14 Understanding of the predictive role of IL28B SNPs in HCV-4 monoinfected patients has important clinical implications, as HCV-4 is not only the most prevalent agent of chronic hepatitis C in the Middle East and North Africa, but also because, as a consequence of migratory flows and spread among the community of drug users, infection rates with HCV-4 are on the rise in Europe and other Western countries.15-17 Moreover, even if HCV-4-infected patients have historically been considered a difficult to cure population, like HCV-1 patients, thus receiving similarly high dosing and prolonged treatment duration, unfortunately they will not benefit from the arrival of the first-generation NS3 protease inhibitors of HCV, which have a weak antiviral activity against HCV-4.18, 19 Because individualization of SOC therapy will remain for years the only optimization option for HCV-4-infected patients, we investigated whether a relationship exists between the rs12979860 IL28B CC genotype and the outcome of PegIFN plus Rbv therapy in HCV-4 naïve patients in the Milan area.
Patients and Methods
Between September 2004 and June 2010, 128 previously untreated adult patients with chronic hepatitis C genotype 4 were consecutively evaluated for antiviral treatment at the Liver Center and the Metabolic Center for Liver Disease of the Ospedale Maggiore Policlinico in Milan. Six patients refused treatment and eight were not eligible because of decompensated cirrhosis or contraindications to IFN. Thus, 112 (88%) patients gave written consent to receive Peg-IFN alpha-2b or 2a plus Rbv therapy.
Chronic hepatitis C was defined as at least 1 year serum positivity for serum HCV-RNA, alanine aminotransferase (ALT) levels >1.5 times the upper limit of normal, and by a liver biopsy, performed in the year preceding treatment, consistent with chronic hepatitis C. Patients were excluded if they had coinfection with hepatitis B virus (HBV), human immunodeficiency virus (HIV), or Schistosoma, decompensated liver disease, drug dependence, or >20 g/day alcohol intake. Patients with poorly controlled diabetes, severe depression, autoimmune diseases, or concomitant malignant neoplastic diseases were also excluded. Patients with high titers (>1:80) of nonorgan-specific autoantibodies were not excluded if their liver biopsy did not show the histological hallmarks of autoimmune liver disease and was consistent with chronic hepatitis C. Patients gave written informed consent to receive therapy and to make their medical records available for this study, which was approved by the Institutional Review Board of the Department of Internal Medicine. Patients were recalled in 2010 to provide written informed consent to rs12979860 genotype testing.
Genomic DNA was obtained from peripheral blood mononuclear cells (PBMC) using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany). Genotyping of rs12979860 was performed using a 5′ nuclease assay with allele-specific TaqMan probes; assays were run on a 7900HT real-time polymerase chain reaction (PCR) instrument (Applied Biosystems, Carlsbad, CA), following the manufacturer's instructions, or by using the tetra primer ARMS method.20 Serum HCV-RNA was assessed by qualitative reverse-transcription (RT)-PCR assay (COBAS Amplicor HCV test v. 2.0, Roche Diagnostics) with a detection limit of 50 IU/mL, during treatment at weeks 4, 12, 24, and 48, and after therapy at weeks 4, 12, 24. Serum HCV-RNA was quantified at baseline and week 12 with Versant HCV-RNA 3.0 assay (bDNA 3.0, Bayer, Emeryville, CA), with a sensitivity limit of 615 IU/mL and a dynamic range from 615 to 7,700,000 IU/mL. HCV was genotyped by Line Probe Assay (INNO-LIPA HCV 2, Innogenetics, Zwijndrecht, Belgium). Liver biopsies were performed with a 16G Tru-Cut needle (Uro-Cut 16G, TSK, Tokyo, Japan) and read by a single pathologist (M.F.D.), who was unaware of patient identity and treatment regimen. The severity of hepatic inflammation was evaluated by the Ishak score in separate reports for grading and staging.21 The exclusion of schistosomiasis was made by microscopic examination of the stool and urine following concentration methods.22
Patients received Rbv (Rebetol, Schering Plough, Kenilworth, NJ) combined with either PegIFNα2a (Pegasys, Roche, Basel, Switzerland) 180 μg/week or PegIFNα2b 1.5 μg/kg/week (PegIntron, Schering Plough) for 48 weeks. PegIFNα2a was associated with Rbv 1,000-1,200 mg day (<75 kg; ≥75 kg) and PegIFNα2b with Rbv 800 mg for patients of less than 65 kg body weight, 1,000 mg for 65-85 kg, and 1,200 mg for ≥85 kg. Therapy was discontinued if quantitative HCV-RNA testing at week 12 dropped by less than 2 logs compared with baseline values, and at week 24 if HCV-RNA was still detectable in those patients in whom HCV-RNA dropped >2 log at week 12. All patients were evaluated for safety and tolerance of treatment every 4 weeks during the treatment period. PegIFNα2a was reduced to 135 μg and PegIFNα2b to 1.0 μg/kg per week in patients with <0.75 × 109/L neutrophils at two consecutive tests, whereas it was withdrawn in patients with <0.50 × 109/L. The same dose reductions were applied if platelets fell under 50,000 cells/mm3, with PegIFN being discontinued when reaching the 25,000 cells/mm3 threshold. In both treatment arms Rbv dose was tapered by 200 mg/day in patients with hemoglobin <10 g/dL, whereas it was discontinued in patients with <8.5 g/dL hemoglobin. Growth factors were allowed for the management of grade 2 anemia (erythropoietin alpha 40,000 IU/week) and grade 3 neutropenia (GCSF 300 μg/week) only in patients with advanced fibrosis starting from January 2006 as recommended by the Italian National Health System.
Definition of Response.
Clearance of serum HCV-RNA by RT-PCR was assessed at week 4 (rapid virological response, RVR), at week 12 (complete early virological response, cEVR), at week 24, and at week 48 of treatment (end-of-treatment response, ETR). An SVR was undetectable HCV-RNA by RT-PCR at week 24 of posttreatment. Patients with an ETR who tested HCV-RNA-positive during follow-up were classified as relapsers. Patients who had any other virologic response were considered nonresponders.
Comparisons between groups were made by using the Mann-Whitney U test or Student's t test for continuous variables and the χ2 or Fisher exact probability test for categorical data. A probability value of P < 0.05 was considered statistically significant. Logistic regression analysis was performed to identify the variables associated with PegIFN/Rbv treatment SVR. All variables with statistical significance at the univariate analysis were included in the final model and odds ratios (OR) and corresponding 95% confidence intervals (95% CI) were computed. Calculations were done with Stata 10.0 statistical package (Stata 1944-2007, College Station, TX).
The demographic and clinical characteristics of the 112 patients enrolled in the study are shown in Table 1. There were 102 males (91%) and 76 were of Egyptian ethnicity (68%). The median age was 46 years (range 27-68) and 26 (23%) patients had a diagnosis of cirrhosis before starting therapy. The subtype of HCV-4 was a in 4 patients, a/c/d in 6, c/d in 26, e in 6, h in 4, and not classifiable by INNO-LIPA in the remaining 66 (59%). Of the 112 patients included, four were excluded from the analysis who did not consent to genetic testing for the IL28B SNP and five who discontinued treatment prematurely for nonvirological reasons. The final analysis included 103 naïve HCV-4 patients, 68 treated with PegIFNα2a (66%), and 35 treated with PegIFNα2b (34%). The rs12979860 genotype was CC in 24 (23%), CT in 65 (63%), TT in 14 (14%) patients, with a minor allele frequency (MAF) of 0.45. The MAF was similar among Egyptian and Italian patients (0.41 versus 0.53); CC patients presented a trend toward higher median baseline viral load compared with CT/TT patients (1,128,142 IU/mL versus 353,243 IU/mL; P = 0.06).
|Egyptian ethnicity||76 (68%)|
|Age, years (mean)||46 (range 27-63)|
|Weight, kg (mean)||78 (55-120)|
|BMI (mean)||26 (19-33)|
|Source of infection|
|Blood transfusion||3 (3%)|
|Other risks||2 (4%)|
|Liver fibrosis stage (Ishak)|
|S 0-2||40 (36%)|
|S 3,4||46 (40%)|
|S 5,6||26 (24%)|
|Baseline serum HCV-RNA|
|<0.6 x 10 6 IU/ml||64 (65%)|
|Baseline ALT IU/l > 2 fold increase||62 (61%)|
In all, 35 patients (34%) achieved an RVR, 66 (64%) an ETR, and 50 an SVR (49%). A posttreatment relapse occurred in 16 ETR patients (24%). By stratifying patients on the basis of the IL28B genotype (CC versus CT/TT), the CC genotype patients achieved significantly higher SVR rates (21/24, 88%) compared with CT/TT patients (29/79, 38%) (P < 0.0001). CC patients also showed higher rates of ETR than CT/TT patients (21/24 versus 45/79, P = 0.04), and significantly lower rates of posttreatment relapse (0% versus 16%, P = 0.001) (Fig. 1). By univariate analysis, the baseline variables associated with an SVR were female gender (P = 0.007), absence of cirrhosis (P = 0.02), Egyptian ethnicity (P = 0.02), and the IL28 genotype CC (P < 0.0001) (Supporting Table 1). By evaluating the impact of the IL28B genotype on Egyptian and Italian patients separately, a significant association between IL28B and SVR remained in each ethnic group. In the 69 Egyptians an SVR was achieved in 17 (89%) of the CC and in 23 (45%) of the CT/TT patients (P < 0.001). The corresponding figures for the 34 Italian patients were: SVR in four (80%) CC versus seven (24%) in CT/TT (P = 0.02). By logistic regression analysis the IL28B CC genotype was confirmed as an independent predictor of SVR, with an OR of 8.0 (95% CI: 2.00-32.01, P = 0.003) (Table 2).
|Variables||Odds Ratio||95% CIs||P|
|IL28B CC genotype||8.00||2.00-32.01||0.003|
|Fibrosis stage 0-4||4.24||1.21-14.88||0.02|
IL28B rs12979860 and Early On-Treatment HCV RNA Kinetics.
The achievement of an RVR was a strong predictor of an SVR, obtained in 80% of the 35 RVR patients, as compared with 33% in the 68 who were still viremic at week 4 (P < 0.0001). CC patients showed higher rates of RVR than CT/TT patients (50% versus 29%), both overall and analyzed on the basis of ethnic group (Fig. 2). However, once an RVR was achieved the SVR rates were independent from the host genotype, being 100% (12/12) in CC patients and 70% (16/23) in CT/TT (P = 0.06), once again demonstrating the relevant predictive power of early HCV RNA decline on treatment outcome. In contrast, among patients lacking an RVR, the SVR rates were higher in the CC patients than in CT/TT, 75% (9/12) versus 23% (13/56) (P = 0.001). By a logistic regression model including RVR as an independent variable, the achievement of an RVR (OR 9.16; 95% CI: 2.90-28.84, P <0.0001), IL28B CC genotype (OR 7.57; 95% CI: 1.78-32.12, P = 0.006), Egyptian ethnicity (OR 4.03; 95% CI: 1.25-13.00, P = 0.02), and absence of cirrhosis (OR 4.50; 95% CI: 1.03-19.70, P = 0.045) were independent predictors of an SVR.
To our knowledge this is the first study designed to investigate the role of the IL28B rs12979860 genotype as a predictor of SVR to PegIFN plus Rbv in HCV-4 monoinfected patients of European and North African ethnicity. Our finding of SVR rates being significantly higher in CC patients compared with CT/TT patients (88% versus 38%) are similar to those reported for HCV-1 patients where the CC genotype was first discovered and consequently validated as the strongest pretreatment predictor of an SVR to Peg-IFN plus Rbv in Caucasian, Afro-American, and Asian patients.5-10 Although the exact mechanisms behind this association are still only partially understood, with the most accepted theory supporting a correlation between IL28B genotypes and IFN-stimulated gene expression,23 there is unanimous agreement that on-treatment viral kinetics differ substantially on the basis of IL28B genotypes.24 CC patients with HCV-1 exhibit a faster viral decline once PegIFN plus Rbv is started that results into more patients achieving an RVR and eventually achieving an SVR once treatment is completed. On the other hand, the CT/TT genotype is characterized by a slower viral decline that translates into an increased rate of posttreatment relapse of hepatitis. Our study confirms these findings also in HCV-4 patients, as the CC patients showed a trend toward higher RVR rates (50% versus 29%) but having significantly lower rates of posttreatment relapse compared with CT/TT patients (0% versus 36%). Although the RVR rates seen in our study (34%) are slightly higher than expected, this difference is likely caused by the sensitivity of the method used for week 4 HCV-RNA testing.24 Indeed, our figures fit well with other studies from our group as well as with another European study on HCV-4 patients where HCV RNA was tested with the same commercial method.14, 25
In analogy with studies in patients with HCV-1, in HCV-4 patients IL28B was not the only driving force behind the achievement of an RVR, because once this key endpoint is reached response rates were high regardless of background genetic factors.26 This finding once again reaffirms the dominant role of RVR on any other baseline predictor and effectively suggests that the treatment decision-making process should not rely on the pretreatment chances of achieving an SVR, because, even in the worst pretreatment case scenario, an SVR is still possible in RVR patients.27 Although such an assumption would de facto limit the interest for any pretreatment predictor of treatment outcome, IL28B included, our study shows that IL28B genotype testing has a place in the treatment algorithm in HCV-4 patients. The reasons for this rely on the peculiar ability of the CC genotype to partially overshadow the negative predictive power of the lack of an RVR. Indeed, in our study, even in the absence of an RVR, CC patients still achieved high SVR rates (75%), thus confirming the results of a previous study in HCV-2 and 3 patients.12 The interplay between IL28B genotype and week 4 HCV-RNA testing is also clinically relevant in HCV-4 CT/TT patients, where the lack of an RVR, which translated into an SVR rate of 32% independently from IL28B genotype, was associated with extremely disappointing SVR rates (23%). Although these figures do not support the introduction of an anticipated stopping rule, the addition of IL28B genotype to HCV-RNA at week 4 testing has the added benefit of informing the patient more precisely about the SVR chances and eventually suggesting treatment discontinuation in CT/TT patients without an RVR in the presence of severe treatment-related side effects, suboptimal patient motivation, or severe comorbidities.
The finding that SVR rates were higher in Egyptians compared with Italians confirms a previous observational study where HCV-4 patients from North African descent obtained higher SVR rates compared with HCV-4 patients from France.17 Based on our data we can reasonably exclude IL28B as the reason behind these findings, as not only the distribution of the CC IL28B genotype did not differ between the two populations but also because its role as an independent pretreatment predictor of SVR was confirmed in both ethnic groups.
Although we are still searching for the exact explanation behind the different SVR rates between Egyptians and Italians, it is important to understand that the HCV spread occurred at different times in the two countries. Indeed, in Europe HCV-4 was mainly spread in the 1970s-1980s through intravenous drug use, whereas in Egypt the mass antischistosoma campaign of tartar emetic in the 1950-1980s is the main culprit.28, 29 This is well reflected in our study by the significant differences in terms of baseline clinical features seen in the two ethnic groups. Italian patients more often acquired the infection through drug use (29% versus 0%), and were also older at the start of therapy (mean age: 46.2 versus 41.7) compared with Egyptians. Moreover, whereas no patient had an ongoing alcohol intake of more than 20 g/day, Egyptian patients were more likely to be teetotalers than Italians (87% versus 29%) due to religious convictions. It seems therefore likely that the presence of many negative moderators of treatment outcome in HCV-4 Italian patients might be the more plausible explanation for the reduced SVR rates in this ethnic group30; still, we cannot confidently exclude that the relatively small sample size of our study might have played a role in biasing these results. Also, we cannot completely exclude that these findings might be explained by differences in other unknown genetic factors between the two ethnic groups or by different virological features, as the HCV-4 subtype was either not available or undetermined in a large number of patients and our study unfortunately did not include any viral sequencing analysis.
We also acknowledge that our findings are mainly applicable to male patients only, as a striking imbalance in the male-to-female ratio (9:1) was seen in our cohort. The main reason for this lies in the epidemiology of HCV-4 infection in Italy, which has either been acquired through drug use in the 1980s, a practice historically more common among males than females, or is currently found in immigrants from North Africa. In the latter group of patients the higher prevalence of males is explained by socioeconomic reasons, as immigrants often enter the country alone, leaving their wives and families in their original country.
Still, all these caveats notwithstanding, we think that our study provides solid evidence that the IL28B rs12979860 genotype is a strong predictor of treatment outcome to PegIFN plus Rbv in HCV-4 monoinfected patients, independently from ethnicity. This supports the incorporation of the IL28B genotype in the treatment algorithm of HCV-4 patients, while highlighting the need to consider this factor in the design and randomization of drug discovery studies, as an imbalanced distribution might artificially affect the trial outcome.31, 32
Additional Supporting Information may be found in the online version of this article.
|HEP_24683_sm_SuppTab1.doc||55K||Supplementary Table 1: Epidemiological and clinical characteristics stratified by treatment outcome.|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.