IL28B gene polymorphism rs12979860, but not rs8099917, contributes to the occurrence of chronic HCV infection in Uruguayan patients

A cross-sectional and observational study was conducted in Uruguayan individuals with and without HCV infection, recruited between 2014 and 2015 at the Gastroenterology Clinic from Hospital de Clínicas (Control group, n = 92; HCV-infected group, n = 78). Chronic HCV-infected patients were treatment naïve and over 18 years old. All samples analysed in this study were negative for hBsAg and HIV and in the case of the control group, all individuals showed absence of reactive serology for HCV. The information obtained from the study populations included sex and age. For the HCV-infected patients virus genotype and stage of liver disease were included, if previously determined (See Table 1). Additionally, 48 of the 78 enrolled HCV-infected patients underwent dual therapy (peg-IFN-α plus RBV) during the study, 42 of which had completed it by the time of writing. SVR was defined as absence HCV RNA in serum 24 weeks after the end of treatment. Non-response (NR) to therapy was defined as HCV viral load decline less than 2 logs at week 12 during therapy or detectable serum HCV RNA at any other time during therapy. Relapse (R) was defined as undetectable level of HCV RNA by the end of treatment which becomes detectable after discontinuation of therapy. For analyses, we divided patients into two groups: patients who achieved SVR and those who did not (NR and R).

The study was conducted according to national and international ethical guidelines (good clinical practice, Nuremberg statements, Declaration of Helsinki) and local regulatory rules (Mercosur Standards/GMC/RES No. 129/96). The protocol was approved by the Ethics Committee of the Hospital de Clínicas on October 24th, 2014 and all patients gave written informed consent. Access to personal information was restricted to the medical doctors. The genetic information extracted from the samples was used exclusively for the purposes of this study.

Genomic DNA was extracted from peripheral white blood cells by QIAamp DNA Kit (QIAGEN). All samples from HCV-infected individuals were SNP genotyped by end-point PCR amplification following restriction fragment length polymorphism analysis as previously described [33]. For rs12979860 primers used were: rs12-F 5’-GCGGAAGGAGCAGTTGCGCT-3′ (sense) and bst-R 5’-GGGGCTTTGCTGGGGGAGTG-3′ (antisense). For rs8099917 primers used were: rs80-F 5’-CCCACTTCTGGAACAAATCGTCCC-3′ (sense) and rs80-R 5’-TCTCCTCCCCAAGTCAGGCAACC-3′ (antisense) [33]. All samples from the control group were genotyped by real-time PCR using TaqMan SNP allelic discrimination assays carried out using StepOne Real-Time PCR System (Applied Biosystems). The primers, MGB probes and conditions used to amplify rs12979860 have been previously described [27]. The rs8099917 SNP genotyping was determined by a TaqMan® Pre-designed SNP Assay (Applied Biosystems) (AB) reference: C_11710096_10. Genotyping of each sample was attributed automatically by the StepOne Software v2.2.2. Positive and negative controls (previously verified by direct Sanger sequencing) were used in each genotyping assay.

Data are presented as percentages (categorical variables), means and standard deviations (continuous variables). Comparisons between the two groups were made using the chi-square (χ²) test or Fisher’s exact test (when cell sample sizes were less than five) for categorical variables and Student’s t-test for continuous variables. The existence of differences in genotypic frequencies between groups was assessed by Chi-square test under the three main inheritance models: codominant, dominant and recessive. A Monte Carlo permutation method was used for multiple test correction. A p value less than 0.05 was considered statistically significant. Raw p values are shown in the tables. Possible deviations from Hardy-Weinberg equilibrium were studied in the control population (both exact and Chi-squared tests were performed) using PLINK Package version 1.9 (http://​zzz.​bwh.​harvard.​edu/​plink/​) [34]. We performed univariable and multivariable logistic regression analyses in order to analyse the independent contribution of each polymorphism to the occurrence of chronic HCV infection (dependent or outcome variable). The categorical independent variables included in the full multivariable model were rs12979860 and rs8099917 SNP genotypes (CC, CT, TT and TT, TG, GG, respectively) adjusted by age and gender. An interaction multiplicative term between both SNPs was also tested in order to consider their joint effect, if any. Odds Ratios (OR) and 95% confidence intervals were calculated. The analyses were performed using the IBM® SPSS® Statistics version 23 Software and R software version 3.4.0 [35]. Linkage disequilibrium analysis was performed using Haploview Software [36].

Seventy-eight treatment-naïve patients with chronic HCV infection were recruited. 61.5% were male with an average age of 46 years. In all cases, chronic infection was confirmed by a viral load above the limit of detection or by qualitative PCR detection. HCV genotypic analysis confirmed the prevalence of genotype 1 (57.7%) in the cohort. The control group included 92 individuals, all of which were HCV, HIV and HBV negative by serologic methods. 72.8% were male with an average age of 40 years. All demographic characteristics are presented in Table 1, including the histological stages for the infected population and the viral genotypes, if previously determined. No significant difference was found in terms of gender distribution (p ≥ 0.05). On the contrary, the difference in age between groups was statistically significant (p < 0.05). The stage of liver disease according to Metavir score was available in 49 of the 78 infected patients. Cirrhosis (stage 3 and 4) was observed in 29 (59.2%) of them (Table 1).

Several reports establish a strong link between SNPs rs12979860 and rs8099917 with the response to peg-IFN-α/RBV treatment for HCV genotype 1 infected patients [7, 9, 13, 25, 28, 29, 31, 37‐39]. To date no study exists that shows the distribution of these polymorphisms within the Uruguayan population despite the fact that HCV genotype 1 is confirmed as the most prevalent viral genotype in the country (Table 1) [40]. To gain information regarding the distribution of IL28B SNPs among the Uruguayan population the genomic DNA from 92 non-infected, non-related individuals were analysed. The distribution of rs12979860 genotypes within this control group were 45.7% CC, 42.4% CT and 11.9% TT, while for the rs8099917 they were 60.9% TT, 33.7% GT and 5.4% GG (Fig. 1 and Table 2). SNP rs12979860 and rs8099917 showed to be in Hardy-Weinberg equilibrium (χ² tests, p ≥ 0.05) in the control group, which allowed us to proceed with the genotype distribution comparison in both populations. Next we sought to establish the distribution of IL28B polymorphisms within a Uruguayan cohort of 78 treated naïve HCV-infected patients. SNP analysis in the HCV-infected cohort revealed that the frequencies for the rs12979860 genotypes were 29.5% CC, 47.4% CT and 23.1% TT, while frequencies for the rs8099917 TT, GT and GG genotypes were 57.7%, 28.2% and 14.1%, respectively (Fig. 1 and Table 2). Under both a codominant as well as a dominant model, the genotype distribution corresponding to rs12979860 (CC, CT, TT) between the non-infected control group and the HCV-infected cohort (Fig. 1a) evidenced to be statistically significant (p˂0.05) even after multiple test correction, with a higher prevalence of the favourable genotype (CC) within the control group (Table 2). A similar analysis was done for rs8099917 genotype distributions; however no statistically significant differences between both cohorts were found (Fig. 1b and Table 2). The allelic frequencies between both populations also showed to be statistically different only for rs12979860 (Fig. 1c and 1d). No association was observed between IL28B genotype and other variables such as gender, HCV genotype or liver fibrosis stage (Additional files 1 and 2).

Of the 42 HCV-infected patients who underwent and finished IFN-α/RBV treatment, 17 achieved SVR, whereas 25 did not (3 NR and 21 relapsers). Among those who responded favourably, 23.5% carried the rs12979860 good-response genotype (CC) while 58.8% carried the rs8099917 favourable genotype (TT). Among non-responder and relapser patients, these frequencies were lower (20.0% - CC and 48.0% - TT, respectively). Nevertheless, chi-squared and Fischer’s exact tests did not evidence statistically significant differences in distributions of IL28B genotypes between both groups.

Next we sought to determine the contribution of IL28B SNPs to the occurrence of chronic HCV-infection by performing logistic regression analyses (Table 3). The results of univariable and multivariable (adjusted by age and sex) models suggest that when rs8099917 genotype is constant, individuals carrying rs12979860 unfavourable genotypes TT and CT on average have a 5.783 (TT) and 3.086 (CT) fold higher likelihood (adjusted OR_TT = 5.783, p = 0.023; adjusted OR_CT = 3.086, p = 0.015) of developing a chronic infection upon infection with HCV when compared to individuals hosting the favourable genotype (CC). Noteworthy, all individuals homozygous for the favourable allele of rs12979860 (CC) also carried the favourable genotype of rs8099917 (TT). It should be noted that no interaction between the SNPs was evidenced, as revealed by a non-converging model. In addition, rs12979860 – TT/CT as well as age were found to be associated with a higher chance of chronic infection occurrence in both univariable and multivariable models.

The results of the genotype distribution comparisons as well as the logistic regression are supported by Haploview analyses showing that both SNPs are in weak Linkage Disequilibrium in both of the studied populations (controls: r² = 0.57; infected-patients: r² = 0.37; both: r² = 0.48). This observation also indicates that for the studied population of HCV-infected patients the association between IL28B SNPs and development of HCV chronicity is primarily driven by only one of the evaluated polymorphisms.