|Home | About | Journals | Submit | Contact Us | Français|
Early identification of factors contributing to successful treatment of hepatitis C infection is important for researchers and clinicians. Studies conducted on the role of ultra rapid viral response (URVR) for prediction of sustained viral response (SVR) have shown its high positive predictive value (PPV). However, data on the combined effect of URVR with IL28B genotypes for prediction of SVR are lacking. Our aim was to study the role of URVR and IL28B genotypes for prediction of SVR among patients in Georgia infected with genotype 1.
Of a total of 156 patients enrolled in the study, 143 were included in the final analyses. Viral load testing for monitoring viral response was done at 3, 24, and 48, 72 hours and at 1, 2, and 4 weeks after treatment initiation. IL28B single nucleotide polymorphisms in rs12979860 were genotyped by real time PCR methods.
Our study revealed URVR as the earliest treatment predictor among genotype 1 patients harboring IL28B C/C genotype (PPV-100%). Moreover, C/C genotype was found have a high PPV among genotype 1 patients without URVR or RVR unlike patients infected with genotype 2 or 3. URVR and IL28B C/C genotype were not as predictive of an SVR among genotype 2 and 3 patients; however RVR were highly predictive of an SVR in these patients.
Our results suggest that testing for IL28B genotypes and viral load at week one and two may improve the ability to predict an SVR among HCV genotype 1 patients; this information can be useful to encourage patients to remain on treatment.
Current estimates suggest that approximately 180 million people worldwide are infected with hepatitis C virus (HCV) with the highest prevalence rates reported in Africa and Asia . According to the World Health Organization (WHO), as many as 4 million new infections occur annually and more than 350, 000 people die from HCV related liver diseases each year . In western countries, HCV is the leading cause of end-stage liver disease and hepatocellular carcinoma, as well as the main indication for liver transplantation . HCV infection is currently curable using increasingly effective antivirals. Pegylated interferon alpha and ribavirin combination therapy has proven to be an effective treatment for individuals with chronic HCV infection [3–6].
Racial disparities have been reported for this recommended combination therapy with respect to the sustained viral response (SVR) rate, defined as the absence of HCV RNA in serum 6 months after completion of treatment. An SVR of 20% to 28% has been reported among black patients compared with 40% to 52% among white patients with genotype 1 infection and 57% versus 82% for genotype 2 or 3 infection respectively [7–10].
Sustained response rates of 61–79% were found among Asian patients, infected with HCV genotype 1 and 80–95% for Asian patients infected with genotypes 2 or 3 [11–15]. These racial differences in the responses to treatment suggested a possible genetic influence on HCV treatment outcome.
Various other host characteristics have been identified to be associated with favorable treatment outcome including female sex, younger age, absence of liver steatosis and insulin resistance [3, 5, 6]. Being infected with HCV genotype 2 or 3 and having a low viral load are key viral determinants associated with a successful treatment outcome. In addition, rapid virological response (RVR) to treatment, i.e. undetectable HCV RNA in serum, at week 4 and early viral response (EVR) at week 12 are important therapeutic milestones which predict SVR [16, 17].
Studies of HCV viral kinetics during treatment revealed that interferon alpha-2b causes a rapid dose-dependent reduction in HCV RNA levels in the serum within 24 to 48 hours. Mathematical calculations revealed that HCV has a serum half-life of 3 hours and a viral production rate of 1.0 × 1012 virions per day [4, 18]. Therefore a rapid viral decline after the initiation of treatment is a strong predictor of an SVR response to treatment [19, 20].
The rapid initial decline is followed by a slower phase that varies widely among patients and is attributed to the death rate of infected hepatocytes. The second phase of viral decline, mediated by immune clearance of infected and dead hepatocytes, is also related to viral clearance and SVR .
The recent discovery of host genetic factors that influenced treatment outcome has changed the approach to HCV treatment. Three independent genome-wide association studies have identified single nucleotide polymorphisms (SNPs) near the gene coding for interferon lambda-3 (or IL28B) that are associated with favorable response to antiviral treatment and spontaneous resolution of infection in patients infected with HCV genotype 1 [22, 23, 24].
The identification of genetic variation in the IL28B gene region raised the possibility that determination of the IL28B allele might have the potential to predict the response to therapy in order to alter the duration of treatment. Among the SNPs studied, rs12979860 located near the IL28B gene was reported to have the highest SVR predictive potential . The rates of SVR among patients infected with genotype 1 and harboring an IL28B C/C genotype reached 80%. A high SVR rate was observed among genotype 2 and 3 patients harboring IL28B genotype C/C who did not achieve a RVR during the treatment course .
The ability to predict either a positive or negative therapeutic response is of obvious benefit to clinicians and patients. The ability to predict the likelihood of a favorable treatment response as early as possible in the course of antiviral treatment could be beneficial to the treating physician and the patient. The reported studies of an ultra rapid viral response (URVR), a viral load reduction by >2 log at week one or two, to therapy have found it to be predictive of an SVR. However, to our knowledge data on the combined effect of URVR and IL28B genotypes for prediction of SVR have not been widely reported.
The prevalence of HCV infection in Georgia, in the general population as well as in high risk groups, is among the highest of the Eastern European countries. The most comprehensive surveys of hepatitis in Georgia indicate that hepatitis C is widespread in the country . Screening of 2,000 randomly selected adults in Tbilisi, Georgia in 2008, found a HCV seroprevalence of 6.7% . These data indicate that the prevalence of HCV in Georgia is more than twice the global HCV prevalence estimate of 3% and is higher than other countries in the region .
We have conducted a prospective cohort study to investigate the role of viral and host characteristics in predicting SVR during interferon treatment among hepatitis C infected individuals from Georgia. We have studied the utility of URVR and IL28B genotypes for prediction of SVR among patients on antiviral treatment with interferon and ribavirin. This study is the first prospective cohort study to examine the HCV viral kinetics and host genetic markers in patients undergoing antiviral treatment in Georgia.
Of 179 patients with chronic HCV infection seeking care at the Infectious Diseases, AIDS and Clinical Immunology Research Center (IDACIRC) in 2009–2010, 156 adult patients were eligible for the study based on the following inclusion criteria: (1) detectable HCV viral load and HCV genotype 1; (2) no history of previous interferon treatment and (3) liver fibrosis METAVIR scores of F2/F3. Patients who were co-infected with hepatitis B (HBV) or human immunodeficiency virus (HIV) or with diagnosed autoimmune or alcoholic liver diseases were excluded. The study was approved by the IDACIRC’s Institutional Review Board and all subjects provided written consent to participate.
Antiviral treatment included pegIFNa-2a or pegIFNa-2b at standard doses (180mg/week or 1.5mg/kg per week, respectively) and weight-adjusted ribavirin (1000 mg/day for genotype 1 patients weighing <75 kg and 1200 mg/day for patients weighing >75 kg and 800 mg/day for genotype 2 and 3 respectively) following current HCV treatment guidelines . Patients infected with HCV genotype 1 were treated for 48 weeks, whereas genotype 2 and 3 patients received 24 weeks of combination therapy according to the current treatment recommendations for HCV mono-infected adults . Treatment stopping rules were applied for patients with suboptimal virologic response (<2 log reduction of HCV RNA) at week 12 or detectable virus at week 24 after achieving EVR .
Plasma HCV RNA was determined using the COBAS® TaqMan® HCV Test, v.2 (Roche, Switzerland) with the detection limit of 25 IU/ml. Plasma samples for HCV viral load testing were collected at baseline, 3, 24, 48 and 72 hours after initial interferon injection and at 1, 2, 4, 12, 24 and 48 weeks. SVR was measured by HCV viral load testing at 24 weeks after treatment completion.
HCV genotyping was conducted by amplifying the 5′ untranslated region of HCV (COBAS AMPLICOR™ Hepatitis C Virus Test, version 2.0 (Roche, Basel, Switzerland) and analyzing the amplification products using the Versant HCV Genotype v2.0 LiPA strips (Innogenetics, Belgium). Results were interpreted using the manufacturer’s protocol.
Human genomic DNA extracted from a blood specimen was used for IL28B genotype determination. SNP rs12979860 was genotyped by allele-specific TaqMan minor groove binding (MGB) probes and TaqMan Genotyping kit (Applied Biosystems, Foster City, California, USA) on ABI 7500 Real Time PCR System as previously described . Allele-specific genotyping calls were automatically applied on the allelic discrimination plot upon the real time PCR amplification and detection completion. Three possible allelic calls were presented, distinguishing C/C, C/T and T/T genotypes.
Liver fibrosis was measured using transient elastography by FibroScan (Echosens, Paris, France). The median value per patient was expressed in kilopascal (kPa) units. Moderate or significant liver fibrosis and its corresponding METAVIR scores F2 and F3 was defined for liver stiffness values of 6 – 9.5 kPa, based on results from studies conducted in both HCV mono-infected and HIV/HCV co-infected patients [31, 32].
Univariate analysis was used to evaluate variable distributions. Comparisons were tested using Pearson’s chi-square or Fisher’s exact test as appropriate. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) with respective 95% confidence intervals (95% CI) were calculated by IL28B genotype status to assess predictive value of early viral decline for achieving SVR. All statistical analyses were performed using SAS 9.2 (SAS Institute Inc, Cary, NC, USA). P values less than 0.05 were considered statistically significant.
Of the total of 156 patients enrolled in the study, 13 (8.3%) discontinued treatment due to interferon/ribaivirin side effects or other reasons and were not included in the final analyses. Of the remaining 143 patients, 103 (72.0%) completed the course of treatment; 40 subjects (28.0%) terminated the treatment due to a suboptimal virologic response at week 12.
The majority of the patients were male (82.5 %) and older than 23 years of age (mean age 38) (Table 1). All were of white race and European ancestry. Of the 143 patients, 83 (58%) reported previous exposure to intravenous drugs, which was the likely source for their HCV acquisition. Eighty-four patients (58.7%) of 143 had a high viral load (>600 000 IU/ml) at baseline and all were diagnosed with either moderate or significant liver fibrosis by METAVIR score (6–9.5 Kpa) as per study enrollment criteria.
IL28B genotype analysis indicated that 76 patients (53.1%) were harboring C/T and T/T variants of SNP rs12979860, while C/C genotype was found among 67 individuals (46.8%). A significantly greater proportion of patients with the C/C genotype had a high viral load (>600 000 IU/ml) than those with the non C/C group. Since the predictive values for SVR among patients with unfavorable genotypes of C/T and T/T as well as among those with favorable HCV genotype 2 and 3 were reported to be similar, we have grouped them together in the final analyses.
The HCV genotype distribution among the cohort was as follows: 50 patients (34.9%) were infected with HCV genotype 1 and 93 patients (65.1%) with HCV genotypes 2 and 3. An SVR was reached in 23 (46 %) patients infected with HCV genotype 1 and among 67 (72 %) patients infected with HCV genotype 2 and 3.
The rate of decline in the HCV viral load at each time point was calculated and compared between C/C vs. non C/C groups. The decline in HCV viral load of >2log10 at each time point compared to the baseline was considered to be a favorable virological response. The first phase viral decline was defined as the decline of HCV RNA between baseline and day three. No significant associations were found between first phase decline (at hours 3, 24, 48 and 72) and an SVR (data not shown). Data on the second phase decline rate, namely, URVR at weeks one and two, RVR, EVR, ETR, and IL28B and HCV genotypes are summarized in Table 2.
When the on-treatment response rates among HCV genotype 1 patients were stratified according to the IL28B genotypes, the group harboring C/C genotype showed a higher rate of viral decline at all second phase time points, compared to the group harboring C/T and T/T genotypes (p values at all time points were <0.05 (Table 2). Therefore, the T allele had a negative impact on the second phase viral decline and was associated with lower rates of URVR at weeks one and two, RVR and SVR (Table 2).
Of 17 genotype 1 patients from the C/C group who achieved SVR, 8 patients attained URVR at week one and week two with further RVR, EVR and ETR. None of these 8 patients had a breakthrough or relapsed after treatment completion. Conversely, URVR was not predictive of SVR among non C/C genotype 1 group (Table 2).
All second phase viral load decline rates among genotype 2 and 3 patients’ harboring the C/C genotype were somewhat higher than the non C/C group (Table 2). The effect of IL28B C/C genotypes was only marginally significant on an URVR at week one and two and did not contribute to the later viral road decline rates (Table 2). When the HCV genotype 2 and 3 group was stratified by HCV genotype, the C/C genotype was a slightly better predictor for URVR at week one and two and SVR among genotype 3 patients than it was for those infected with genotype 2 (data not shown).
The mean HCV viral load decline between baseline and URVR week one and two was calculated among all IL28B and HCV genotype groups. A greater HCV viral load decline was observed among the HCV genotype 1 C/C group which was not associated with baseline viral load. However, the decline among the non C/C group was significantly associated with a low baseline viral load (p<0.04). There were no differences identified among the genotype 2 and 3 group in this regard (p>0.05).
An SVR was achieved among 77.2% of patients with HCV genotype 1 C/C versus 22.4% of patients with non-C/C. However, SVR occurred among 80% and 64.5% patients whose genotypes were C/C and non C/C in those infected with HCV genotypes 2 and 3 respectively (Table 2). The rate of SVR among genotype 1 C/C patients not achieving URVR and RVR was greater compared to the non C/C group (p<0.05). However, C/C genotype was not predictive of SVR among genotype 2 and 3 patients without an URVR and RVR (p>0.05).
The positive predictive value (PPV) and negative predictive value (NPV) for SVR at each second phase time point are summarized in Table 3. Combining the IL28B C/C genotype with either URVR at week two or RVR responses yielded a PPV of 100%, and NPV of 43%–45%, while among the subjects with non C/C genotype the PPV was only 50% for URVR at week two and 75% for RVR and the NPV was 81% for URVR and 87% for RVR respectively. The PPV and NPV values for each time point among-the genotype 2 and 3 group were similar between the two IL28B groups (data not shown).
In addition to HCV genotype, viral load and liver fibrosis stage, recent studies have demonstrated an association between certain host genetic factors and successful treatment outcome [23, 24]. Among the on-treatment predictors, viral load decline of more than 2log10 at week 4 and week 12 has been shown to have a high predictive value for achieving an SVR [17, 18].
Because of the high cost of treatment and relatively low effectiveness of current treatment regimens among genotype 1 patients, it is important to identify markers that can reliably predict the likelihood of treatment success. Early prediction of treatment outcome is essential for encouraging the continuation of therapy among patients with a high likelihood of cure and deferring treatment when the chances of achieving an SVR are minimal.
Because of the high cost of antiviral treatment, neither the health system nor the insurance schemes in Georgia will cover the costs of hepatitis C treatment services. Therefore, less than 10% of the patients diagnosed with HCV infection who are in urgent need of antiviral treatment undergo antiviral therapy. Moreover, nearly all of those patients have to pay for their treatment out of pocket, an option that is not possible for most patients. Thus, the early identification of reliable markers of treatment success is of critical importance.
We have performed a detailed evaluation of the first and second phase viral declines in conjunction with identifying IL28B genotypes among patients with chronic HCV infections in Georgia who were receiving treatment with pegylated interferon and ribavirin. The main findings of our study were that among patients infected with HCV genotype 1 with homozygous C/C alleles, the viral load decline (>2log 10) at week one and two was highly predictive of SVR (PPV-100%) and was as useful as RVR in predicting a successful outcome of therapy. Therefore, the probability of reaching an SVR can be assessed as accurately at week one and two as at week 4 or 12. Additionally, the chance of achieving SVR among genotype 1 C/C patients without URVR at week one and two or RVR, is greater than that among the non C/C group. Therefore, according to our data, it is useful both to measure the presence of virological response to therapy at weeks one and two and also to determine the IL28B genotype of HCV genotype 1 patients in order to predict the eventual probability of an SVR.
Our finding was consistent with the finding of Thompson et al  in which they reported that patients with C/C genotype had the strongest viral reduction levels at week two compared to other IL28B genotypes and therefore were associated with an SVR. In addition, our study showed the effect of URVR at week one to be the earliest effective prediction time point for an SVR. This observation differed from the finding of Bochud et al  and Arends et al . They reported the effect of viral load decline at 24 and 48 hours as the earliest markers for the prediction of an SVR among C/C carriers. As the first phase viral load decline is largely dependent on the fibrosis stage, as well as insulinemia and gamma-glutamyl transpeptidase levels, the differences observed in our cohort may be due to demographic differences between study populations. Despite the effect of C/C genotype for achieving higher URVR and SVR rates among genotype 2 and 3 patients, there was not a significant association between C/C genotypes and viral reduction rates at week 4, although the RVR was numerically higher among C/C carriers than those with C/T and T/T genotypes.
Our results were slightly different from Mangia et al  and Rallon et al . Similarly to the Mangia results, the rate of an SVR was high among patients who attained URVR and RVR regardless of the IL28B genotype. However, unlike their reports, the C/C genotype was not predictive of SVR among patients not achieving URVR and RVR. An unexplained observation was found among this group as C/C genotype was a predictor for achieving URVR and SVR but not attaining RVR, EVR and ETR. The numbers in IL28B genotype groups not achieving URVR and RVR were small; therefore we cannot draw firm conclusions. However, we can speculate that a favorable IL28B genotype influences early viral kinetics and does not have a substantial effect in the later treatment course among genotype 2 and 3 patients. Thus, RVR and not IL28B genotype can be regarded as the better predictor for SVR among genotype 2 and 3 patients unlike genotype1 patients. Altogether these observations suggest that different mechanisms of viral eradication may be operative during the early and late treatment course among patients infected with different HCV genotypes.
Our study has several important implications. First, it is the first study of HCV viral kinetics and IL28B genotypes among the Georgian population, who represent a highly homogeneous ethnic group. Second, we have identified URVR at week one and two to be the earliest effective treatment predictor for SVR among C/C genotype carriers. Third, we found the influence of C/C genotype for SVR among patients in Georgia who were infected with different HCV genotypes. The effect of the C/C genotype among HCV genotype 1 patients is not only mediated by its effect on URVR and RVR, but still is predictive of a favorable outcome of therapy among patients who do not achieve a URVR or RVR. Finally, our data suggest that, a favorable IL28B genotype might only influence the viral eradication early in the treatment course among genotype 2 and 3 patients and does not have a beneficial effect later during therapy among patients without URVR and RVR.
There were several important limitations to our study that should be considered. First, we did not have a large number of female participants in the study, which might allow us to evaluate the gender related responses to interferon therapy among women with different IL28B genotypes. Second, due to the limited number of patients, we did not evaluate HCV treatment responses among genotype 1 patients with different subtypes. Third, the liver fibroses score was evaluated by transient elastography and not liver biopsy, which some consider to be the gold standard for liver fibrosis evaluation. Nevertheless, transient elastography has been shown to accurately diagnose patients with advanced fibrosis and, i.e. METAVIR scores of F2 or 3. Fourth, we do not have information on the liver enzymes or hepatic steatosis and insulinemia levels; these factors have been reported to be related to SVR responses to interferon treatment. Finally, we did not study the effect of other SNPs on interferon responses.
Our study demonstrates importance of defining IL28B genotypes and URVR as the earliest prediction marker for SVR among HCV genotype 1 mono-infected Georgian patients undergoing antiviral therapy. In addition, genotype 1 patients who had an unfavorable IL28B genotype and did not achieve URVR or RVR had a minimal chance of eventually achieving an SVR; however, treatment discontinuation cannot be considered at this point. The predictive power of the IL28B genotype for an SVR continues to be significant beyond the early therapeutic course; however, an RVR is the most important predictor of an SVR among genotype 2 and 3 patients regardless of the IL28B genotype. More research is needed to fully understand the predictive role of URVR and IL28B genotypes among larger patients groups consisting of females and subjects who are co-infected with HIV or HBV.
The project was supported by Infectious Diseases, AIDS and Clinical Immunology Research Center.
The project was partially supported by Award Number D43TW000233 and D43TW007384 from the Fogarty International Center. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Fogarty International Center or the National Institutes of Health.
The authors acknowledge the assistance for Monica Parker, Renee Hallack, and Tea Iobashvili for valuable comments on the article.
Conflict of interest: none declared
Author’s contributionConceived and designed the experiment: Marine Karchava, Lali Sharvadze and Tengiz Tsertsvadze.
Performed experiments: Marine Karchava, Natia Dvali, Lana Gatserelia, Ekaterine Dolmazashvili, Lela Dzigua
Analyzed the data: Nikoloz Chkhartishvili
Wrote the paper: Marine Karchava
Revised the manuscript for important intellectual content: Lali Sharvadze, Kenrad Nelson, Nino Gochitashivli, Natia Dvali, Maia Zhamutashvili, Ekaterine Dolmazashvili, Lela Dzigua, Tengiz Tsertsvadze