Predictors of Response to 24-Week Telaprevir-Based Triple Therapy for Treatment-Naïve Genotype 1b Chronic Hepatitis C Patients

We evaluated the genetic variation in rs8099917, substitutions in core amino acid (aa) 70, and the number of aa substitutions in the interferon sensitivity-determining region (ISDR) on the prediction of sustained virological response (SVR) in treatment-naïve hepatitis C virus (HCV) genotype 1b (G1b) patients. This multicenter study involved 150 Asian treatment-naïve patients infected with HCV G1b who received 12 weeks of telaprevir in combination with 24 weeks of peginterferon-α-2b and ribavirin. The baseline and treatment-related factors potentially associated with SVR were determined by multivariate logistic regression analysis. Virological response was analyzed on an intent-to-treat basis. Cessation of the therapy due to adverse effects occurred in only 2 patients, who discontinued the trial at 10 weeks and at 2 weeks due to cerebral infarction and renal impairment, respectively. Among the 150 patients in whom the final virological response was determined, only genotype TT in rs8099917 was identified as a pretreatment predictor (P = 7.38 × 10−4). Achievement of a rapid virological response (RVR), defined as undetectable HCV RNA at week 4 of treatment, was identified as an after-starting-treatment predictor (P = 2.47 × 10−5). However, neither a substitution in core aa 70 nor the number of substitutions in the ISDR affected treatment outcome.


Introduction
Chronic hepatitis C virus (HCV) infection was generally treated with pegylated interferon (PEG-IFN) and ribavirin combination therapy. This treatment provides rates of sustained virological response (SVR, undetectable serum levels of HCV RNA at least 6 months after completion of therapy) of 40-50% among patients with HCV genotype 1 (G1) who have not received previous treatment (treatment-naïve patients) [1][2][3].
Telaprevir, a first generation orally bioavailable inhibitor of the nonstructural 3/4A HCV protease, shows significantly higher rates of SVR than standard care of PEG-IFN and ribavirin in patients with HCV G1 disease when given for 12 weeks in combination with regimens of PEG-IFN plus ribavirin lasting 12, 24, or 48 weeks [4][5][6]. Although the SVR rates in patients receiving telaprevir with PEG-IFN and ribavirin (triple therapy) were higher than those in PEG-IFN plus ribavirin (combination therapy) patients, drop-out rates due to increased side effects were also higher in the triple therapy patients [4][5][6]. Triple therapy is associated with increased risks of anemia, skin lesions, digestive symptoms, hyperuricemia, renal dysfunction, and other side effects compared to the combination therapy, especially among women and older patients [4,7,8]. The majority of patients in Japan who are infected with HCV genotype 1b (G1b) are older than the patients in the United States and/or Europe and the frequency of patients who discontinued due to adverse events was reported to be as high as 11.1-16.7% [7][8][9]. In most previous reports [4][5][6][7][8][9][10], the age inclusion criterion limited participants to those who were aged <65 years. Therefore, the safety and efficacy of the triple therapy for patients aged >65 has not been established.
The outcome of the triple therapy in patients who had previously received the combination therapy was largely dependent on the response to the previous therapy [10,11]. In treatment-naïve patients, the genetic variation near the interleukin-28B (IL-28B) gene and amino acid (aa) substitution in the core 70 (core aa 70) of HCV may be the candidate predictors of the virological response to triple therapy. Although the significance of substitution in the core aa 70 on the outcome of the triple therapy has been reported [8,9], a considerable number of the patients with a history of the combination therapy were included in these reports. Therefore, the factors predicting SVR in treatment-naïve patients have not been clearly determined.
In the present study, the SVR rate was examined in treatment-naïve patients with HCV G1b who were treated with a regimen consisting of 12 weeks of telaprevir, combined with PEG-IFN and ribavirin, followed by 12 weeks of the PEG-IFN and ribavirin. The significance of genetic variation in rs8099917 near the IL28B gene (rs8099917 genotype), core aa 70, and the number of amino acid substitutions in the interferon sensitivity-determining region (ISDR) on the outcome of triple therapy were then evaluated. In addition, we evaluated the significance of rapid virological response (RVR, serum HCV RNA not detectable within 4 weeks of therapy) [12] in the prediction of SVR and tried to define the patients who were suitable for triple therapy.  , there were no patients who changed the  treatment policy by the results of the genetic variation near  the IL-28B gene, aa substitution in the core 70, and ISDR of  HCV. Of the 359 patients, 150 had not received previous combination therapy (i.e., they were treatment-naïve) and were included in this analysis ( Table 1). The age range of the enrolled patients was 18-75 years. All of the patients satisfied the following criteria. The patients were persistently seropositive for HCV RNA for >6 months and the amount of serum HCV RNA, as determined by a qualitative realtime polymerase chain reaction (RT-PCR) method (COBAS TaqMan HCV test, Roche Diagnostics, Tokyo, Japan), was ≥5 log 10 IU/mL, which has been defined as a "high viral load, " according to the Japanese criteria [13]. In addition, they had white blood cell counts ≥2000 per cubic millimeter, neutrophil counts ≥1500 per cubic millimeter, hemoglobin levels ≥11 g/dL, and platelet counts ≥70,000 per cubic millimeter, were aged 18-75 years, and had body weights >35 kg at the time of study entry. The presence or absence of cirrhosis was determined by the liver biopsy METAVIR scores [14] within 12 months of enrollment or by an aspartate aminotransferase (AST) to platelet ratio index (APRI) > 2.0, as proposed by Wai et al. [15], at the time of the enrollment. Patients who were positive for the hepatitis B surface antigen or anti-human immunodeficiency virus antibody or who had hepatocellular carcinoma, other liver diseases, psychiatric conditions, or current alcohol consumption levels >20 g per day were excluded from this study.

Materials and Methods
All patients were scheduled to receive telaprevir (1500-2250 mg per day) combined with weekly subcutaneous injections of PEG-IFN -2b (1.5 g/kg) and ribavirin (600-1000 mg per day, according to body weight: <60 kg: 600 mg per day; 60-80 kg: 800 mg per day; >80 kg: 1000 mg per day; if the patient's hemoglobin was <13 g/dL at the start of therapy, ribavirin was reduced by 200 mg) for 12 weeks. This was followed by 12 weeks of PEG-IFN -2b plus ribavirin combination therapy. Telaprevir was administered every 8 hours after meals at 500 mg or 750 mg, or every 12 hours after meals at 750 mg or 1125 mg. Ribavirin was orally administered every 12 hours after meals. The patient's attending physicians determined the initial dose of telaprevir (1500 mg per day or 2250 mg per day), based on patient age, gender, body weight, and hemoglobin level. Telaprevir adherence was calculated on the basis of 2250 mg per day. Each drug was appropriately reduced or withdrawn when a serious adverse event developed during the treatment course. In patients who had HCV RNA >3 log 10 IU/mL at week 4, detectable HCV RNA at 12 week, or a 2 log 10 IU/mL increase in HCV RNA from the lowest level during therapy, therapy was discontinued because of the extremely low likelihood of achieving SVR and the high risk of developing antiviral resistance.
Virological response was analyzed on an intent-to-treat basis. Virological "relapse" was defined as HCV RNA levels that became undetectable by the end of treatment but became positive again during the follow-up period; viral "breakthrough" was defined as HCV RNA levels that became undetectable during the treatment but became detectable again before the end of the treatment, and "nonvirological PEG-IFN: peginterferon; RBV: ribavirin; TVR: telaprevir; HCV: hepatitis C virus; ISDR: interferon-sensitivity determining region; ND: not detected. * Determined by the liver biopsy METAVIR scores within 12 months of enrollment or by an aspartate aminotransferase to platelet ratio index (APRI) >1.5. * * "Wild-type" (arginine) or "mutant-type" (glutamine or histidine).
Data expressed as number of patients or median (range).
response" (NVR) reflected HCV RNA levels that never dropped below the detection level during therapy. We also defined rapid virological response as the absence of detectable HCV RNA 4 weeks after starting treatment.
The study protocol was conducted in accordance with the provisions of the Declaration of Helsinki and Good Clinical Practice guidelines and was approved by the Institutional Review Boards of all participating sites. Written informed consent was acquired from each individual.

Genotyping and Quantification of HCV RNA. The HCV
G1b genotype was defined according to the method previously reported by Ohno et al. [16]. The serum HCV RNA concentration was measured at the beginning of therapy and every 4 weeks until 24 weeks after the end of therapy, using the previously described RT-PCR method. The linear dynamic range of the assay was 1.2-7.8 Log 10 IU/mL, and samples below the level of detection were considered HCV-negative.

Detection of aa Substitutions in
Core aa 70 and the ISDR of HCV Genotype 1b. The substitutions in core aa 70 and in the ISDR were determined using a direct sequencing method. Briefly, RNA was extracted from the serum and, after reverse transcription, the substitution in aa 70 was determined according to the method previously reported by Akuta et al. [17]. The "wild-type" aa 70 in the core region is arginine and a "mutant-type" involved a change to glutamine or histidine. Also, aa substitutions in the range of 2209-2248 in the NS5A (the ISDR) were determined using the method of Enomoto et al. [18]. The number of aa substitutions in ISDR was classified into "0-1" and 2 or more.

Genetic Variation
Near the IL-28B Gene. Genomic DNA was extracted from whole blood using MagNA Pure LC and the DNA Isolation Kit (Roche Diagnostics). The rs8099917 single nucleotide polymorphism (SNP) near the IL-28B gene [19] was genotyped by RT-PCR using the TaqMan SNP Genotyping Assay and the 7500 Fast RT-PCR System (Applied Biosystems, Foster City, CA, USA). The rs8099917 genotype was classified into 2 categories: TT (major genotype) and non-TT genotype (minor genotype, TG or GG).

Statistical Analysis.
Pearson or Mantel-Haenszel chisquare test, Fisher's exact test, or Mann-Whitney test was used to compare frequencies in categorical data or differences in continuous data between groups, respectively. Possible variables contributing to RVR and SVR included baseline and on-treatment features. Variables that reached statistical significance ( < 0.05) or marginal significance ( < 0.10) in bivariate comparisons were subsequently entered into multivariate logistic regression analysis using forward and backward stepwise selection method to identify significantly independent factors associated with RVR and SVR. values of <0.05 denoted the presence of a statistically significant difference. All statistical analyses were carried out using STATISTICA for Windows version 6 (StatSoft, Tulsa, OK, USA).

Virological Responses during Triple Therapy.
Of the 150 patients, 116 (77.3%) had a RVR. The frequency of undetectable serum HCV RNA was 94.4% at the end of treatment. Of the 150 patients for whom a final virological response was evaluated (either SVR or non-SVR), 131 patients (87.3%) achieved SVR.

Factors
Affecting SVR with the Triple Therapy. The frequency of rs8099917 genotype TT was higher in patients with SVR than non-SVR (85.5% versus 15.8%; = 1.33 × 10 −10 ). Serum gamma-glutamyl-transpeptidase concentration was significantly lower in SVR patients than in non-SVR individuals ( = 0.0459). Serum albumin ( = 0.0968) and fasting plasma glucose concentrations ( = 0.0549) demonstrated only marginal differences. The substitutions in core aa 70 and the number of substitutions in the ISDR did not predict SVR. In the multivariate logistic regression analysis for elucidating pretreatment independent predictive factors for SVR, rs8099917 genotype TT (OR, 0.071; 95% CI, 0.015-0.337; = 0.0007) alone was identified. In addition, achievement of RVR (83.2% versus 36.8%; = 2.47 × 10 −5 ) after starting treatment was a significant predictor of SVR (Table 3).

Influence of the Substitution in Core aa 70 and the Substitution Number in ISDR on SVR in Patients with rs8099917
Genotype TT or Non-TT. Among the 150 patients, 115 had an rs8099917 TT genotype and 35 had a non-TT (TG or GG) genotype. Of the 115 patients who had genotype TT, 74 had "wild-type" amino acids, whereas 41 had "non-wild-type" aa in the core aa 70 of the HCV isolate. Of the 35 patients who had non-TT (TG or GG) genotype, 23 were "wild-type, " whereas 12 had a "non-wild-type" core aa 70 genotype.
The rates of SVR in these patients are illustrated in Figure 1. Neither the substitution in aa 70 in the HCV core region nor the number of aa substitutions in the ISDR impacted the prediction of SVR, regardless of the rs8099917 genotype.

Significance of RVR on Achieving SVR in Patients with an rs8099917
Non-TT (TG or GG) Genotype. In the 94 patients carrying an rs8099917 TT genotype and achieving RVR, the SVR rate was 98.9% (93/94), whereas it was 90.5% (19/21) in patients carrying genotype TT, but not achieving RVR. In contrast, the SVR rate was only 23.1% (3/13) in patients who had the rs8099917 genotype non-TT (TG or GG) without achieving RVR, whereas it was 72.7% (16/22) in patients achieving RVR ( = 0.0125) (Figure 1).

Safety and Characteristics of the Patients Who Did Not
Achieve SVR. Of the 150 patients who had a defined final virological response, only 19 (12.7%) were classified as non-SVR: 3 (15.8%) "NVR"; 10 (52.6%) "relapse"; and 6 (31.6%) "breakthrough. " Three patients were carrying genotype TT, whereas 16 had a non-TT genotype. Cessation of the therapy due to adverse effects occurred in only 2 patients who were       Figure 1: According to the genetic variation in rs8099917, near the IL28B gene, a significantly higher proportion of patients with the TT genotype showed a sustained virological response (SVR) than did patients with the TG or GG genotype. In contrast, based on the amino acid substitutions in the core region (amino acid 70) and interferon sensitivity-determining region (ISDR), there was no significant association between the SVR rate and these substitutions, irrespective of the rs8099917 genotype. Furthermore, the SVR rate was 98.9% in patients with the rs8099917 TT genotype who achieved rapid virological response (RVR), whereas the SVR rate was 23.1% among patients with the TG or GG rs8099917 genotype who did not achieve RVR.
carrying genotype TT; these patients discontinued the therapy at 2 weeks and at 10 weeks due to cerebral infarction and renal impairment, respectively. The remaining 1 patient who was carrying genotype TT finished the scheduled treatment, but the HCV RNA reappeared 4 weeks after the end of the therapy. These 3 genotype TT patients were classified as having relapses.

Comparison of SVR Rates, according to rs8099917 Genotype and the Existence of Cirrhosis, between Patients Aged ≤65
Years and Those Aged >65 Years. Among the 150 patients, 29 (19.3%) were aged >65 years. The frequency of the non-TT genotype (TG or GG) rs8099917 in the patients in both age groups was similar. However, the frequency of cirrhosis in patients aged over 65 years (9/29; 31.0%) was higher than that in the younger patients (20/121; 16.5%). The SVR rate in the older (aged >65 years) patients was similar to that in those aged ≤65 years, regardless of the rs8099917 genotype or the existence of cirrhosis ( Figure 2). In 29 patients aged >65 years, the frequencies of the major side effect were as follows: anemia occurred in 28 (96.6%), elevation of serum uric acid in 20 (69.0%), skin rashes in 18 (62.1%), headache in 15 (51.7%), nausea in 11 (37.9%), and elevation of serum creatinine in 10 (34.4%). The frequencies were similar to those of the ≤65 years patients. Cessation of the therapy due to adverse effects occurred in only 1 (3.4%) patient; this patient discontinued the therapy at 10 weeks due to renal impairment (data not shown).

Discussion
Genetic variation near the IL28B gene and the substitutions in core aa 70 of HCV G1b have been suggested to be predictive of virological outcomes for triple therapy. However, in the present study of treatment-naïve patients, none of the viral factors (substitution of core aa70 or number of substitutions in the ISDR) or the host factors [12,20], except for the rs8099917 genotype, were observed as predictors of 24-week triple therapy efficacy.
In the previous study, IL28 SNP genotype had a limited impact on SVR rates with triple therapy in treatmentexperienced patients [21], and the strength of association between IL28B genotype and treatment outcome was attenuated in the triple therapy arms compared to the combination therapy arm [22].
In this study, the rs8099917 genotype displayed a striking influence on the outcome of triple therapy, along with "achievement of RVR. " In the patients carrying the rs8099917 genotype TT, even if RVR was not achieved, the SVR rate was as high as 90%. In contrast, the SVR rate was only 23.1% for patients with the TG or GG genotype who did not achieve RVR but reached 72.7% when RVR was achieved. Therefore,  Figure 2: There was no significant relation between sustained virological response rate and age. The patients older than 65 years had hemoglobin levels similar to those aged ≤65 years.
"achievement of RVR" was particularly important for SVR in patients with a minor (non-TT) rs8099917 genotype. From these findings, rs8099917 genotyping should be examined in treatment-naïve patients and the patients carrying the TT genotype should be considered for triple therapy. On the other hand, in patients with a TG or GG genotype, the introduction of the triple therapy may be recommended, but extension of treatment period or cessation of the therapy should be considered if HCV RNA remains detectable following 4 weeks of therapy. Substitutions in the core aa 70 and the number of substitutions in the ISDR of HCV G1b were reported to be important predictors of combination therapy efficacy, and these findings are gaining consensus [12,23]. In previous reports on the triple therapy, the rs8099917 genotype and the presence of a substitution in the core aa 70 were repeatedly identified as predictors of SVR in patients with HCV G1b infection [8,9]. However, our study on treatment-naïve patients revealed that core aa 70 substitutions do not predict the achievement of SVR. This discrepancy may be explained due to the differences in the study populations and the inclusion criteria of drop-out patients. Akuta et al. [9] reported that their study population contained both treatment-naïve patients and patients with a history of interferon-based therapy. In addition, 25% of their study population was treated with triple drugs for 12 weeks alone, without the follow-up PEG-IFN plus ribavirin combination therapy. Similarly, Chayama et al. [8] reported that their study population contained not only treatment-naïve patients but also combination therapy experienced patients. In addition, their study included patients who received a very short duration of the triple therapy. Therefore, therapy duration variability and the difference of study population from our study may have had an impact on their study outcomes.
All of our patients were in the intent-to-treat with the 24week triple therapy and only 6 patients (4.0%) were disrupted. This may lead to our conclusion that substitutions in the core aa 70 did not impact treatment outcomes in treatment-naïve, HCV G1b-infected patients. However, the relatively small number of patients in our study may limit these conclusions; the results should be verified by a larger-scale study.
In our study, the RVR and SVR rates were 80.2% and 87.3%, respectively; these rates tend to be higher than those reported previously [4,5]. One of the reasons for the higher SVR rate may be the comparatively high frequency (76.7%) of the TT genotype in our cohort. Another reason for higher SVR rate was the extremely low prevalence of dropouts, owing to the aggressive management of adverse effects and careful adjustment of the dosage of triple drugs. The drop-out rate in the present study was only 1.3% (2 of 150 patients), as compared to 11.1-16.7% in other Japanese studies [7,9,24]. Although it is the small number of cases (only 9 of 150 patients), detailed search for aa substitutions in the HCV NS3 protease domain may be required for the care of patients who are classified as NVR or breakthrough, because the relationship between aa substitutions in this domain and resistance to NS3-4A protease inhibitors has been documented [25,26].
In previous studies [4][5][6][7][8][9][10], the treatment of elderly patients with triple therapy is approached with caution and the inclusion criteria for clinical trials are usually set to maximum age of 65 years. Although differences in efficacy and in the frequency and severity of side effects between the patients aged >60 and those aged ≤60 years have not been observed [24], the safety and the efficacy of the treatment in patients aged >65 years has not been adequately demonstrated. However, there is a need for an effective antiviral therapy for older patients in Japan because of the large numbers of older HCV G1 patients. Thus, we included a small number of patients aged >65 years in the present study. The SVR rate of older (>65 years) patients seemed to be similar to that in those aged ≤65 years and there was not a notable increase in the number of drop-outs. Our finding that adherence to the telaprevir or ribavirin dose schedule did not clearly affect the achievement of SVR suggests that continuation of the scheduled length of therapy may be far more important for achieving SVR than maintaining a particular drug dosage. Moreover, the efficacy of triple therapy for the minor rs8099917 genotype or the presence of cirrhosis was similar between patients aged >65 years and those aged ≤65 years. From these observations, we propose that patients aged >65 years can be treated with triple therapy if the drug dosages are adequately regulated.

Conclusions
In conclusion, for treatment-naïve, chronic HCV G1binfected patients, the rs8099917 genotype was a significant factor predicting a successful virological effect of the triple therapy; a substitution in core aa 70 and the number of substitutions in the ISDR did not impact treatment outcomes. In addition, achievement of RVR after starting treatment was exceedingly important to accomplish SVR in patients carrying the rs8099917 non-TT genotype. Thus, the patient's rs8099917 genotype should be determined before starting triple therapy.