| |
Importance of IL28B gene polymorphisms in hepatitis C virus genotype 2 and 3 infected patients - pdf attached
|
| |
| |
Download the PDF here
Jnl of Hepatology March 2011
Christoph Sarrazin1Corresponding Author Informationemail address, Simone Susser1, Alexandra Doehring2, Christian Markus Lange1, Tobias Muller3, Christina Schlecker2, Eva Herrmann4, Jorn Lotsch2, Thomas Berg5
1Medizinische Klinik 1, J. W. Goethe-University Hospital, Frankfurt am Main, Germany; 2pharmazentrum frankfurt, Institute of Clinical Pharmacology, J. W. Goethe-University, Frankfurt am Main, Germany; 3Medizinische Klinik m. S. Hepatologie und Gastroenterologie, Campus Virchow Klinikum, Charite Universitatsmedizin Berlin, Germany; 4Institute of Biostatistics and Mathematical Modeling, J. W. Goethe-University, Frankfurt am Main, Germany; 5Klinik fur Gastroenterologie, Sektion Hepatologie, Universitatsklinikum Leipzig, Germany
"Multivariate analysis revealed that lower age, HCV genotype 2, and the rs12979860 CC genotype were significantly associated with sustained virologic response.......The rapid virologic response (RVR) rates for the rs12979860 CC, CT, and TT genotypes were 87%, 77%, and 64%, respectively, but this failed to be statistically significant (p=0.17). For SVR in patients who achieved an RVR, a significant correlation with the rs12979860 genotype was observed (p=0.05), while for SVR in non-RVR patients no association was found (p=0.48) (Fig. 4).....A higher HCV RNA concentration was significantly associated with the TT genotype of rs8099917, the CC genotype of rs12979860, and the AA genotype of rs12980275 for HCV genotype 1 and genotype 2/3 infected patients (Table 3). Finally, the rs8099917 TT genotype was significantly associated with a higher fibrosis stage in HCV genotype 1 infected patients (Table 3)."
"Ge et al. reported a twofold greater rate of SVR for the rs12979860 CC genotype in comparison with the TT genotype [6], while in the present study 87.4% of patients with the rs12979860 CC genotype vs. 73.1% of patients with the TT genotype achieved an SVR. Because of the high natural SVR rates of genotype 2/3 infected patients, differences between the rs12979860 CC vs. the non-CC genotype patients became more evident for virologic non-response. Only 47.5% of all patients with SVR had the rs12979860 CC genotype but 85% of patients with virologic non-response had CT or TT genotypes. In addition, SVR in patients who achieved an RVR was significantly associated with the rs12979860 CC genotype."
"In addition to a correlation with virologic response and viral load, IL28B genotypes may also contribute to the grade of inflammation and the stage of liver fibrosis. Finally, a correlation between the frequencies of the rs12979860 CC genotype with SVR rates in genotype 1-, 2- and 3-infected patients was observed which needs further investigation."
ABSTRACT
Background & Aims
Genetic variation in the interleukin 28B (IL28B) gene has been associated with the response to interferon-alfa/ribavirin therapy in hepatitis C virus (HCV) genotype 1-infected patients. The importance of three IL28B single nucleotide polymorphisms (rs8099917, rs12980275 and rs12979860) for HCV genotype 2/3-infected patients is unknown.
Methods
In patients with chronic hepatitis C genotype 2/3 (n=267), IL28B host genotypes (rs8099917, rs12980275 and rs12979860) were analyzed for associations with sustained virologic response (SVR) to antiviral therapy with (pegylated) interferon-alfa and ribavirin and with respect to epidemiological, biochemical, and virological parameters. For comparison, hepatitis C genotype 1 patients (n=378) and healthy controls (n=200) were included.
Results
The rs12979860 CC genotype, lower age, and genotype 2 were significantly associated with SVR in HCV genotype 2/3-infected patients (p=0.01, p=0.03 and p=0.03, respectively). No association was observed for rs8099917 and rs12980275. In addition, an SVR in patients with rapid virologic response (RVR) was associated with the rs12979860 CC genotype (p=0.05), while for non-RVR no association was found. Furthermore, a significant association with a higher baseline viral load was observed for all three IL28B genotypes in genotype 1/2/3-infected patients. Finally, increasing frequencies of the rs12979860 CC genotypes were observed in genotype 1- (33.9%), genotype 3- (38.9%), and genotype 2-infected (51.9%) patients in comparison with healthy controls (49.0%) (p<0.01).
Conclusions
In genotype 2/3-infected patients, rs12979860 was significantly associated with SVR. The frequency of the rs12979860 CC genotype is lower in HCV genotype 1 vs. genotype 2/3 patients. All major IL28B genotypes are associated with HCV-RNA concentration.
Abbreviations: IL28B, interleukin 28B gene, HCV, hepatitis C virus, SNP, single nucleotide polymorphism, WHO, World Health Organisation, SOC, standard of care, SVR, sustained virologic response, BMI, body mass index, IFN, interferon, RNA, ribonucleic acid, IU, international unit, μg, microgram, kg, kilogram, mg, milligram, mL, milliliter, HBV, hepatitis B virus, HIV, human immunodeficiency virus, EDTA, ethylendiamin-tetraacetat, DNA, desoxyribonucleicacid, OD, optical density, GWAS, genome-wide association studies, ALT, alanine-aminotransferase, PPV, positive predictive value, NPV, negative predictive value, GT, genotype, n.a., not available
Introduction
Chronic hepatitis C virus infection is still a major cause for developing cirrhosis and hepatocellular carcinoma which often results in liver failure and thus in liver transplantation. According to the World Health Organisation, 180 million people are infected worldwide and 3-4 million new infections per year were estimated [1]. Up to now, the standard of care (SOC) treatment consists of (pegylated) interferon-alfa and ribavirin. However, depending on the viral genotype, treatment response rates differ significantly among infected patients. While up to 80% of the genotype 2 and 3 infected patients can be cured, the response rate is only 40-50% in genotype 1 infections [2], [3]. Virus-specific characteristics (viral load, genotype, viral variants for example within the interferon sensitivity determining region, ISDR) may be responsible for these differences but also clinical parameters (age, gender, BMI, fibrosis stage, liver enzymes) have been shown to be associated with virologic response [4], [5]. The impact of genetic variation near the interleukin 28B (IL28B) gene for response in HCV genotype 1 infected patients was shown recently [6], [7], [8], [9]. IL28B encodes interferon λ-3 (IFN-λ3) a cytokine distantly related to type I interferons and the IL-10 family. Together with interleukin 28A (IFN-λ2) and interleukin 29 (IFN-λ1), IL28B forms a cytokine gene cluster on a chromosomal region mapped to 19q13. Expression of the cytokines encoded by these three genes can be induced by RNA virus infection [10].
While the association of the virologic response to (pegylated) interferon-alfa/ribavirin combination therapy with IL28B variants in HCV genotype 1 infected patients was shown and confirmed in several independent studies, little is known about the importance of these IL28B polymorphisms for genotype 2 or 3 infected patients [11], [12].
In the present study, the three main IL28B single nucleotide polymorphisms (SNPs), which had so far shown the strongest association with virologic response (rs12979860, rs8099917, and rs12980275) in genotype 1 infected patients, were investigated in HCV genotype 2/3 infected patients (n=267) in correlation with epidemiological, biochemical, and virological parameters as well as in response to antiviral therapy with (pegylated) interferon-alfa and ribavirin. For comparison, epidemiological IL28B genotype frequencies in patients with chronic HCV genotype 1 infection (n=378) and healthy controls (n=200) were investigated.
Results
Association of IL28B genotypes with virologic response to PEG-IFNα and ribavirin therapy in HCV genotype 2/3 infected patients
Clinical characteristics of HCV genotype 2/3 infected patients are shown in Table 1. In 205/267 patients antiviral combination therapy with (pegylated) interferon and ribavirin was completed and the virologic treatment outcome was known. From the known parameters associated with sustained virologic response, in the present study only age (p=0.03), genotype 2 (p=0.03), and fibrosis stage (p=0.04) were significant, while sex, ALT, and HCV RNA concentration at baseline were not associated with sustained virologic response (Table 2). Fig. 1A-C show the IL28B genotyping results of patients with chronic hepatitis C genotype 2/3 infection for the three SNPs related to virologic response.
Fig. 1. Association of the IL28B genotypes with (A) sustained virologic response (SVR), (B) relapse, (C) non-response (NR) to (pegylated) interferon-alfa and ribavirin in HCV genotype 2/3 infected patients.
Statistical analysis identified rs12979860 as being significantly associated with virologic response to (PEG)-IFNα/ribavirin treatment in HCV genotype 2 and 3 infected patients (p=0.01). Rates of sustained virologic response, relapse, and non-response were 78%, 12%, and 10%, respectively in the entire cohort of genotype 2/3 infected patients. For patients with the rs12979860 CC genotype a higher SVR rate (87.4%) was observed in comparison with CT and TT genotypes (70.7% and 73.1%) (Fig. 1A). For virologic relapse, a variable association with the different rs12979860 genotypes was observed (Fig. 1B). The most pronounced difference was observed for virologic non-response. Only 3.4% of patients with the CC genotype vs. 13.0% and 19.2% with the CT and TT genotypes, respectively, were virologic non-responders (Fig. 1C).
Due to the high SVR rates obtained in genotype 2/3 infected patients, the sensitivity for prediction of a sustained virologic response in patients with the rs12979860 CC genotype vs. the CT/TT genotype was only 47.5% (47.5% of all SVR patients had the rs12979860 CC genotype, Fig. 2), while a high specificity for the exclusion of non-response in patients with the rs12979860 CC genotype with 85.0% (85.0% of non-responders had rs12979860 CT/TT genotypes) was observed (Fig. 2).
The rapid virologic response (RVR) rates for the rs12979860 CC, CT, and TT genotypes were 87%, 77%, and 64%, respectively, but this failed to be statistically significant (p=0.17). For SVR in patients who achieved an RVR, a significant correlation with the rs12979860 genotype was observed (p=0.05), while for SVR in non-RVR patients no association was found (p=0.48) (Fig. 4).
Genome-wide association studies also described the rs12980275 and rs8099917 genotypes as predictive for treatment outcome in HCV genotype 1 patients. In the present study, in genotype 2/3 infected patients with the favorable genotypes (rs8099917 TT and rs12980275 AA) high SVR rates of 80.5% and 83.9%, but also in the unfavorable genotypes (rs8099917 GG and rs12980275 GG) high SVR rates of 87.5% and 80.0% in comparison with 78.0% in the overall cohort, were observed (Fig. 1A). The unfavorable genotype (rs8099917 GG and rs12980275 GG) was more rarely associated with virologic relapse. Furthermore, a trend towards lower frequency of the genotypes rs8099917 TT and rs12980275 AA was observed in non-responder patients (Fig. 1B and C). However, in the present study, for genotype 2/3 infected patients, no statistically significant association between the IL28B rs8099917 and rs12980275 genotypes with virologic response, was observed.
Multivariate analysis revealed that lower age, HCV genotype 2, and the rs12979860 CC genotype were significantly associated with sustained virologic response (Table 2).
Association of the IL28B genotype with ALT, HCV RNA concentration and fibrosis
In addition to an association with virologic response, the IL28B genotypes could also be related to biochemical, virological, and histological parameters. ALT levels, HCV RNA concentration at baseline and the fibrosis stage were correlated with the IL28B SNPs rs12979860, rs8099917, and rs12980275. For higher ALT values, a significant association was observed for the rs12979860 CC genotype in HCV genotype 2/3 infected patients and for the rs12980275 AA genotype in HCV genotype 1 infected patients. In addition, for the remaining major IL28B genotypes, we observed a trend towards increased association with higher ALT levels (Table 3). A higher HCV RNA concentration was significantly associated with the TT genotype of rs8099917, the CC genotype of rs12979860, and the AA genotype of rs12980275 for HCV genotype 1 and genotype 2/3 infected patients (Table 3). Finally, the rs8099917 TT genotype was significantly associated with a higher fibrosis stage in HCV genotype 1 infected patients (Table 3).
Frequencies of the IL28B genotypes in HCV genotype 2, 3 compared with HCV genotype 1 and healthy controls
Frequencies of the IL28B genotypes in HCV genotype 1, 2, and 3 infected patients as well as healthy controls are shown in Table 4. There were no significant deviations from the Hardy-Weinberg equilibrium (p>0.10). In the present study, only the rs12979860 CC genotype was associated with virologic response in genotype 2/3 infected patients. For genotype 1 infected patients, a lower frequency of the rs12979860 CC genotype in comparison with the healthy control was reported previously [6] and this could be confirmed in the present study (genotype 1, 33.9% vs. healthy control, 49%). Interestingly, for the HCV genotype 2/3 patients, an intermediate frequency of the rs12979860 CC genotype (42.7%) was obtained (Table 4, Fig. 3). The difference between HCV genotype 1 infected patients and healthy controls had a high level of statistical significance (p<0.001), while the difference between control subjects and HCV genotype 2/3 patients was not significant (p=0.116). Separate analysis of genotype 2 and 3 infected patients showed that a significant difference is present between genotype 1 and 2 (p=0.045), but not for genotype 1 and 3 (p=0.43) infected patients, for the frequency of the rs12979860 CC genotype (33.9%, 51.9%, 38.9%) (Fig. 3).
Discussion
Parameters for the prediction of sustained virologic response in patients with chronic hepatitis C before initiation of antiviral therapy are important in order to be able to estimate the potential for treatment success. They can help clinicians in the decision on whether or not to start antiviral therapy and this information can also motivate patients who might have a high chance for virologic response. Different studies have shown that HCV genotype, HCV RNA concentration, age, gender, BMI, fibrosis stage, alanine aminotransferase (ALT), and gamma glutamyltranspeptidase (GGT) levels, insulin resistance as well as host genetic polymorphisms of several genes (HLA, chemokines, interleukins and IFN-stimulated genes) are associated with sustained virologic response [4], [5], [15], [16]. However, in clinical practice guidelines only the HCV genotype and HCV RNA concentration at baseline are currently recommended to be used in order to determine treatment duration in response to guided therapy approaches [17], [18], [19]. Most recently, three genome wide association studies reported associations of different SNPs in IL28B (interferon lambda gene region) with response to antiviral therapy [6], [7], [8]. Here, in HCV genotype 1 infected patients a highly significant correlation of sustained virologic response to interferon-alfa/ribavirin combination treatment was observed with the genotypes rs12979860 CC, rs8099917 TT, and rs12980275 AA. In the present study, the importance of these three major IL28B SNPs for European patients with HCV genotype 2/3 infection was investigated.
The main result of this study is that in genotype 2/3 infected patients only a significant association of the rs12979860 CC genotype with SVR was observed. This is in line with a recent study on a relative small cohort of genotype 2/3 infected patients (n=45) by McCarthy et al., in which only rs12979860 was investigated and also found to be associated with SVR [12]. Because of the high SVR rates of genotype 2/3 infected patients to interferon-alfa/ribavirin combination therapy, differences among patients with and without the rs12979860 CC genotype were much smaller than in genotype 1 patients. Ge et al. reported a twofold greater rate of SVR for the rs12979860 CC genotype in comparison with the TT genotype [6], while in the present study 87.4% of patients with the rs12979860 CC genotype vs. 73.1% of patients with the TT genotype achieved an SVR. Because of the high natural SVR rates of genotype 2/3 infected patients, differences between the rs12979860 CC vs. the non-CC genotype patients became more evident for virologic non-response. Only 47.5% of all patients with SVR had the rs12979860 CC genotype but 85% of patients with virologic non-response had CT or TT genotypes. In addition, SVR in patients who achieved an RVR was significantly associated with the rs12979860 CC genotype.
Generally, the separate analysis of genotype 2 and 3 infected patients in the present study showed a significant association of the IL28B rs12979860 genotype only in genotype 3 infected patients. However, this might be due to the larger number of genotype 3 patients together with the rare event of virologic relapse or non-response in genotype 2 patients. For all single parameters (SVR, RVR, and HCV RNA concentration), and also for genotype 2 infected patients with the rs12979860 CC vs. CT/TT genotype, a trend towards higher SVR and RVR rates as well as higher HCV RNA concentrations at baseline was observed and p-values increased in the combined analysis together with HCV genotype 3 infected patients (data not shown).
For the other two IL28B SNPs we studied, rs8099917 and rs12980275, no significant association with virologic treatment response was observed. Investigation of a comparable number of genotype 2/3 infected patients in a study by Rauch et al. also showed no correlation between the rs8099917 genotype and virologic response to pegylated interferon/ribavirin combination therapy [11]. In the present study, only a slightly reduced frequency of the genotypes rs8099917 TT and rs12980275 AA in comparison with carriers of the risk alleles (rs8099917 TG/GG and rs12980275 AG/GG) in patients with non-response was detected and it remains unclear whether in a larger patient cohort this association might become significant.
In addition, an association of the major IL28B genotypes with higher baseline HCV RNA concentration in genotype 1 infected patients is known from previous studies. Interestingly, in the present study for all three IL28B SNPs, a highly significant association with higher baseline viral loads was observed in genotype 2/3 as well as in genotype 1 infected patients carrying the genotypes rs12979860 CC, rs8099917 TT, and rs12980275 AA. Usually a lower baseline viral load is associated with a higher chance of SVR and it remains unclear why here the same IL28B genotype is associated with SVR and higher viral load. Furthermore, a trend towards higher ALT levels was observed in carriers of the rs12979860 CC genotype in HCV genotype 2/3 patients and in carriers of the rs12980275 AA genotype in HCV genotype 1 infected patients. In addition, a trend towards higher fibrosis stages in HCV genotype 1 infected patients was seen with the rs8099917 TT genotype. Generally, it is possible that the IL28B genotypes are also associated with inflammatory activity and fibrosis stage. However, genotype 2 and 3 patients in the present study were not completely balanced for the frequency of different fibrosis stages and larger patient populations are required to prove this potential association.
An interesting finding from a previous study describes a strong association between the rs12979860 CC genotype with the resolution of acute HCV infection in comparison to those who developed chronic hepatitis C [9]. This implies that an unfavorable IL28B rs12979860 genotype may predispose patients to chronic HCV infection [8]. The favorable rs12979860 CC genotype seems to be protective against the development of chronic hepatitis C and thus in patients who did not clear the virus spontaneously, non-favorable rs12979860 genotypes are enriched. However, this may be different according to different HCV genotypes. In the present study, the frequency of the rs12979860 CC genotype in healthy controls was 49.0%. However, in patients with chronic hepatitis C, different rs12979860 CC genotype frequencies according to different HCV genotypes were observed. The lowest frequency of the favorable rs12979860 CC genotype for SVR was detected in genotype 1 infected patients (33.9%), followed by genotype 3 patients (38.9%) and genotype 2 patients (51.9%). For chronic hepatitis C, the lowest SVR rates are observed in genotype 1 infected patients while for genotype 3 patients, higher SVR rates are observed and the highest SVR rates can be seen in genotype 2 infected patients [20], [21], [22]. Thus, the different SVR rates seen in the different HCV genotypes might be in part explained by different rs12979860 CC genotype frequencies. In a recent study by McCarthy et al., lower rs12979860 CC genotype frequencies in genotype 1 vs. genotype 2/3 infected patients were also observed. However, in this study, the rs12979860 CC genotype frequencies were higher in genotype 3 vs. genotype 2 infected patients [12]. Thus, additional studies are required to estimate the differences between IL28B genotype frequencies in patients with different HCV genotypes. Furthermore, the underlying functional mechanisms for the development of chronic hepatitis C with different HCV geno- and subtypes in the presence of a specific IL28B genotype needs to be elucidated.
In conclusion, the rs12979860 CC genotype but not the rs8099917 and rs12980275 genotypes was significantly associated with sustained virologic response to (pegylated) interferon-alfa/ribavirin combination therapy in genotype 2/3 infected patients. All three major IL28B genotypes previously observed to be associated with sustained virologic response and HCV RNA concentration in genotype 1 infected patients are also associated with HCV RNA concentration in genotype 2/3 infected patients. In addition to a correlation with virologic response and viral load, IL28B genotypes may also contribute to the grade of inflammation and the stage of liver fibrosis. Finally, a correlation between the frequencies of the rs12979860 CC genotype with SVR rates in genotype 1-, 2- and 3-infected patients was observed which needs further investigation.
|
|
| |
| |
|
|
|
