Impact of high-dose peginterferon alfa-2A on virological response rates in patients with hepatitis C genotype 1: A randomized controlled trial

Authors


  • Potential conflict of interest: The study was sponsored by Roche. No financial support was received by the Australian-based Protocol Steering Committee who governed the conduct of the study. Dr. Roberts is a consultant for, advises, is on the speakers' bureau of, and received grants from Roche. Dr. Crawford received grants from Roche. Dr. McCaughan is a consultant for, is on the speakers' bureau of, and received grants from Roche and Schering-Plough. Dr. Sievert advises and received grants from Roche. Dr. Desmond advises, is on the speakers' bureau of, and received grants from Roche. He is also on the speakers' bureau of Schering-Plough and Gilead. Dr. Marks received grants from Roche. Drs. Yoshihara and DePamphilis own stocks in Roche. Dr. Dore advises, is on the speakers' bureau of, and received grants from Roche.

  • The study was governed by an independent protocol steering committee comprising authors S.K.R., M.D.W., D.H.C., G.W.M., W.S., W.S.C., W.R., and G.J.D., with Roche employee M.Y. performing the statistical analysis for the study.

Abstract

This study tested the hypothesis that high-dose peginterferon alfa-2a (PEG-IFNα-2a) for the first 12 weeks would increase early and sustained virological response (SVR) rates in patients with chronic hepatitis C genotype 1. Eight hundred ninety-six patients were randomized 1:1 to 360 μg (n = 448) or 180 μg (n = 448) PEG-IFNα-2a weekly plus ribavirin at 1000-1200 mg/day for 12 weeks, followed by 36 weeks of 180 μg PEG-IFNα-2a weekly plus ribavirin at 1000-1200 mg/day with 871 patients evaluable for the intention-to-treat analysis. Virological responses were assessed by TaqMan (limit of detection 15 IU/mL) at week 4, 8, 12, 24, 48 (end of therapy), and 24 weeks following therapy (SVR). Undetectable hepatitis C virus RNA rates were significantly higher among patients receiving high-dose induction therapy at week 4 (36% versus 26%, P < 0.005), week 8 (61% versus 50%, P < 0.005), and week 12 (74% versus 62%, P < 0.005). However, SVR was not significantly different between patients receiving high-dose (53%) and standard (50%) therapy. Significant baseline prognostic factors for SVR included age, sex, race, histological stage, and viral load. SVR was considerably higher among patients with no or minimal fibrosis (64% and 60%, respectively) compared to those with severe fibrosis/cirrhosis (28% and 24%, respectively). The frequency of serious adverse events and drug discontinuations were similar in both groups, whereas PEG-IFN dose modification, weight and appetite reduction, and grade IV neutropenia were significantly higher in the induction arm. Conclusion: Induction dosing with 360 μg/week PEG-IFNα-2a for 12 weeks was well tolerated and enhanced early virological response but not SVR rates. The high SVR rates in patients with minimal fibrosis highlight the benefit of early treatment in patients with hepatitis C virus genotype 1. (HEPATOLOGY 2009.)

Combination pegylated interferon (peginterferon, PEG-IFN) plus ribavirin therapy is the current standard of care for patients with chronic hepatitis C (CHC) achieving a sustained virological response (SVR) rate of 55%-65%.1, 2 Response rates are lower, however, among patients with hepatitis C virus (HCV) genotype 1 infection in whom only 40%-50% achieve an SVR.1–3 Therefore, more-effective treatment regimens are needed for these patients. Although novel direct-acting antiviral strategies look promising in the longer term, peginterferon and ribavirin will likely remain the cornerstone of therapy in the immediate future.

Intensified therapy with higher and/or more frequent doses of (peg)interferon for up to 12 weeks initially has been evaluated in several studies.4–11 The rationale for this so-called induction strategy is that high-dose interferon may improve SVR by inducing a more rapid initial decline in HCV RNA. The relationships between interferon dose and early phase HCV kinetics, including phase I and II slope and time to undetectable HCV RNA,12, 13 and early virological responses and SVR support such a rationale.1, 14–16 Studies utilizing induction therapy with (peg)interferon in patients with genotype 1 CHC have yielded mixed results, with some patients demonstrating an enhanced SVR4, 5 and others not.6–11 However, there has been no published randomized trial of high-dose induction therapy with peginterferon alfa-2a/b (PEG-IFNα-2a/b) in a large population of treatment-naïve patients with genotype 1 CHC.

Thus, we tested the hypothesis that high-dose PEG-IFNα-2a) induction therapy will increase the SVR rate via an increase in the early virological response (EVR) rate compared to standard dosing in HCV genotype 1 patients. In this large, international, multicenter, randomized study, we compared the efficacy and safety of 360 μg of PEG-IFNα-2a weekly for the first 12 weeks versus standard 180 μg of PEG-IFNα-2a weekly both in combination with standard-dose ribavirin in HCV genotype 1 treatment-naïve patients with CHC.

Abbreviations

ALT, alanine aminotransferase; BMI, body mass index; cEVR, complete early virological response; CHC, chronic hepatitis C; CI, confidence interval; EVR, early virological response; HCV, hepatitis C virus; PEG-IFNα-2a, peginterferon alfa-2a; pEVR, partial early virological response; RVR, rapid virological response; SVR, sustained virological response.

Patients and Methods

Patient Population.

Study recruitment commenced through a network of Australian sites in August 2004. Eligible subjects included treatment-naïve adults aged 18-75 years with serological evidence of HCV genotype 1 infection (repeatedly anti-HCV positive and/or HCV RNA–positive), quantifiable serum HCV RNA > 600 IU/mL (Roche Cobas TaqMan HCV Test; Roche Diagnostics, Indianapolis, IN; limit of quantification, 15 IU/mL), elevated serum alanine aminotransferase (ALT) level, compensated liver disease (Child-Pugh score < 7), and histological findings consistent with CHC on a liver biopsy performed within the previous 36 months.

Patient exclusion criteria included: infection with HCV genotype other than genotype 1 (genotype non-1), hepatitis B virus, and/or human immunodeficiency virus; history of decompensated liver disease; evidence of hepatocellular carcinoma; causes of chronic liver disease other than HCV; therapy with any systemic antiviral, antineoplastic, or immunomodulatory agent within 6 months prior to first receiving study drug; pregnancy or breast feeding, or male partners of women who were pregnant; neutrophil count <1500 cells/mm3; platelet count <90,000 cells/mm3; hemoglobin concentration <120 g/L in women or <130 g/L in men; serum creatinine level >1.5 times the upper limit of normal at screening; active severe psychiatric disease; history of immunologically mediated disease or any severe chronic or uncontrolled disease; current or recent drug or alcohol abuse; and unwillingness to provide informed consent. Patients with cirrhosis were also excluded due to the potential enhanced toxicity of induction therapy.

In order for patients in Australia to receive subsidized commercial study drugs, the patients had to meet the disease eligibility criteria for CHC treatment through the Australian Highly Specialized Drugs Program, including elevated ALT and CHC proven by pretreatment liver biopsy. However, during the course of the study, these criteria were altered with removal of elevated ALT (December 2005) and pretreatment liver biopsy (April 2006) requirements. To enhance recruitment, the protocol was changed in accordance with the revised criteria, to allow normal ALT and patients without baseline liver biopsy. In addition, patients with compensated cirrhosis were able to enroll. To further enhance recruitment, the study was expanded in mid-2006 to include patients in New Zealand, Thailand, Argentina, Mexico, and Canada. Patients from these countries except New Zealand required baseline liver biopsy. Study recruitment was through 51 sites including 33 in Australia.

Study Design and Treatment Regimens.

This was a multicenter, multinational, randomized, open-label, active controlled study. Patients meeting screening eligibility criteria were randomly assigned 1:1 to receive PEG-IFNα-2a either in an induction dose or standard-dose regimen. The induction regimen consisted of 360 μg of PEG-IFNα-2a weekly for the first 12 weeks followed by 180 μg of PEG-IFNα-2a weekly for 36 weeks and 1000/1200 mg of ribavirin daily for 48 weeks. The standard-dose regimen involved 180 μg of PEG-IFNα-2a weekly and 1000/1200 mg of ribavirin daily for 48 weeks. Randomization was performed centrally via a computer program with stratification according to country and screening HCV viral load (<800,000 or ≥800,000 IU/mL). All patients whose study treatment was prematurely discontinued underwent safety assessments at 4 and 12 weeks after their last dose of study medication. Subjects who prematurely discontinued treatment and had undetectable HCV RNA at 4 weeks after their last dose of medication were asked to return for HCV RNA determination at study weeks 48 and 72. Patients with detectable HCV RNA and less than a 2-log drop in HCV RNA level at week 12 were encouraged to continue therapy up until week 24. Subjects were required to discontinue study therapy if they had detectable HCV RNA at week 24.

Dose Modifications.

Dose modifications of PEG-IFNα-2a and ribavirin for adverse events and laboratory abnormalities were performed in accordance with protocol-defined guidelines. In patients receiving standard dose PEG-IFNα-2a, the decremental adjustment in the weekly dose was from 180 μg to 135 μg, 90 μg, and 45 μg; in patients who were receiving 360 μg of PEG-IFNα-2a, the decremental adjustment was from 360 μg to 270 μg, 180 μg, 135 μg, 90 μg, and 45 μg, depending on the severity of adverse events. Specific guidelines for PEG-IFNα-2a dose adjustment for neutropenia included dose reduction to 135 μg for patients with fewer than 750 cells/mm3, and for those with fewer than 500 cells/mm3 to suspend therapy until counts rose above 1000 cells/mm3. When required, adjustments to the daily dose of ribavirin were performed in gradual stepwise decrements of 200 mg. Ribavirin was withheld in patients who had hemoglobin below 85 g/L or below 120 g/L in patients with clinically significant cardiovascular disease. The use of growth factors was discouraged but permitted if deemed necessary on clinical safety grounds.

Patient adherence to therapy was assessed via recording the injections and doses of PEG-IFNα-2a and ribavirin at each visit according to the patient's detailed statements, and via documentation of drugs dispensed.

Assessments and Efficacy Endpoints.

Clinical and laboratory safety and efficacy assessments were performed during the treatment period every 4 weeks during the first 24 weeks, then every 6 weeks through to week 48; and after 4 weeks (week 52), 12 weeks (week 60), and 24 weeks (week 72) of follow-up. Quantitative serum HCV RNA levels were measured at baseline and at weeks 4, 8, 12, 24, 48, and 72. HCV RNA assessments were performed by two central laboratories (Australia, United States) using the Roche Ampliprep/Cobas TaqMan HCV Test with a detection limit of 15 IU/mL. The TaqMan assessment platforms in the two central laboratories were harmonized before initiating study assessments. HCV virological efficacy endpoints based on undetectable HCV RNA, including the primary endpoint at week 72, were based on TaqMan levels <15 IU/mL (both undetectable and detectable <15 IU/mL).

When this study commenced, HCV RNA samples at week 48 (end-of-treatment) were being assessed by local laboratories using the Roche Amplicor assay (detection limit, 50 IU/mL). The protocol was subsequently amended to implement central laboratory assessment for HCV RNA at week 48. Thus, a proportion of patients had only local laboratory results available for week 48 HCV RNA assessment. To determine the end-of-treatment virological response results, central laboratory assessments were used whenever available. If results from central laboratory assessments were not available, the results of local laboratory assessments were used.

The primary efficacy endpoint was SVR defined as undetectable HCV RNA in serum at the completion of 24 weeks follow-up after the 48-week treatment period (a single last HCV RNA measured at or following week 68). Patients without measurements during the required assessment time window were considered nonresponders. Secondary endpoints included rapid virological response (RVR) defined as undetectable HCV RNA in serum at the end of 4 weeks of treatment, and complete early virological response (cEVR) defined as undetectable HCV RNA in serum at the end of 12 weeks of treatment.

Statistical Analysis.

The Cochran-Mantel-Haenszel test stratified by viral load (<800,000 and ≥800,000 IU/mL) and country was used to compare SVR rates between the two treatment groups of the intent-to-treat population, which was defined as all patients randomized who received at least one dose of study medication. Odds ratios and corresponding 95% confidence intervals (CIs) were calculated. The Breslow-Day test was performed to assess the homogeneity of the odds ratios across the strata. Exact 95% (two-sided) CIs from the binomial distribution were provided for virological response in individual treatment groups. For the sample size calculation in this study, SVR in the standard dose PEG-IFNα-2a group was set to 50% with the SVR in the PEG-IFNα-2a induction dose set to 60%. With the level of significance set to 0.05 and the power to 80%, a two-sided test required a sample size of 408 in each treatment group.

Study Conduct.

Study conduct was overseen by an Australian-based Protocol Steering Committee. Approval from the institutional review board or ethics committee was obtained before commencement of the study at all sites and before modifications were made to the conduct of the trial. An independent data safety review board closely monitored safety throughout the study. The study was sponsored by Roche. The clinical trial was registered with both the U.S. National Institutes of Health (NCT00192647) and the Australian Therapeutic Goods Administration (ACTRN12605000488606).

Results

A total of 896 patients were enrolled into the study between September 2004 and February 2007 and randomly assigned 1:1 to the induction group (448 patients) and the standard group (448 patients). The disposition of patients in the study is shown in Fig. 1. There were 25 randomized patients who did not receive study drug, including 15 in the induction group and 10 in the standard group, for reasons that included refusal of treatment, failure to return, violation of the protocol inclusion or exclusion criteria, and administration/other reasons. Baseline characteristics for the 871 patients forming the intention-to-treat analysis population are shown in Table 1. The two groups were well balanced with respect to baseline characteristics.

Figure 1.

Study flow and disposition of patients. Patients who discontinued prematurely and whose HCV RNA was undetectable at the end of treatment were expected to return for HCV RNA assessments 12 weeks after actual end of treatment and at end of follow-up. Thus, the number of patients who completed follow-up may be higher than the number of patients who completed treatment.

Table 1. Demographic and Baseline Disease Characteristics of Patients
CharacteristicInduction Group Peg-IFN alfa-2a 360/180 μg Plus Ribavirin (n = 433)Standard Group Peg-IFN alfa-2a 180 μg Plus Ribavirin (n = 438)
  • *

    There were two additional patients with mixed HCV infection with genotype 1/2b (standard group) and 1/3a (induction group), and a further patient in the standard group who had genotype 3a.

Sex (M/F)298/135285/153
Age (yr), mean ± SD43.6 ± 9.643.3 ± 9.2
Race group, n (%)  
 Caucasian355 (82)365 (83)
 Asian61 (14)55 (13)
 Black2 (<1)1 (<1)
 Other15 (3)17 (4)
Weight (kg), mean ± SD77.3 ± 16.978.7 ± 16.1
Weight > 85 kg, n (%)139 (32)141 (32)
Body mass index (kg/m2), mean ± SD26.2 ± 4.626.9 ± 4.8
Alanine aminotransferase (U/L), mean ± SD64.6 ± 49.170.5 ± 62.9
Alanine aminotransferase quotient, n (%)  
 0–192 (21)89 (20)
 1–3248 (57)255 (58)
 >392 (21)91 (21)
 Missing1 (<1)3 (<1)
HCV genotype subtype*, n (%)  
 1152 (35)135 (31)
 1a131 (30)135 (31)
 1a/b24 (6)33 (8)
 1b125 (29)133 (30)
HCV RNA titer (IU/mL)  
 Mean ± SD (×106)4.0 ± 5.63.9 ± 5.2
 > 400,000 n (%)356 (82)343 (78)
 ≥ 800,000 n (%)302 (70)298 (67)
Histologic stage, n (%)  
 F028 (6)25 (6)
 F1107 (25)104 (24)
 F2121 (28)113 (26)
 F346 (11)51 (12)
 F414 (3)16 (4)
 Missing117 (27)129 (29)
Cirrhosis/transition to cirrhosis, n (%)60 (14)67 (15)

Virological Response.

Virological responses over the initial 12 weeks of therapy were higher in the induction dose PEG-IFNα-2a arm compared to standard dose PEG-IFN α-2a arm, including undetectable HCV RNA at week 4 (RVR) (36% versus 26%; P = 0.0007), week 8 (61% versus 50%; P = 0.0005), and week 12 (cEVR) (74% versus 62%; P < 0.0001; Fig. 2A). At week 24, virological response in the induction dose arm remained higher (75% versus 68%; P < 0.05); however, the absolute difference between the two groups had reduced. At the end of treatment at week 48, there was no significant difference in virological response between the induction dose and standard dose arms (70% versus 66%; P = 0.18). SVR rates at week 72 were not significantly different between the induction dose and standard dose arms (53% versus 50%; P = 0.29; Fig. 2B). Similar SVR results were obtained when analysis was performed according to the more restricted per-protocol population (53% versus 51%) or when actual rather than scheduled treatment period was used (data not shown). The lack of improvement in SVR with induction dose PEG-IFNα-2a was not due to a higher rate of virological relapse in follow-up, because this was similar to that following standard dose therapy (24% versus 22%).

Figure 2.

Rates of early viral responses in the initial 12 weeks (A) and end-of-treatment virological response and sustained virological response (B) in both the induction and standard-dose treatment groups according to intention-to-treat analysis. Virologic response was defined as undetectable serum level HCV RNA (<15 IU/mL). Differences between the two treatment groups were assessed by the Cochran-Mantel-Haenszel test stratified by viral load (<800,000 and ≥800,000 IU/mL) and country.

Predictive Value of Early Viral Response.

Virological response at week 12 was evaluated as (1) early virological response (EVR, undetectable HCV RNA or a ≥ 2-log10 decrease from baseline in HCV RNA); (2) cEVR (undetectable HCV RNA); and (3) partial EVR (pEVR, detectable HCV RNA >15 IU/mL but a ≥ 2-log10 decrease from baseline in HCV RNA) (Table 2). The proportion of patients in the induction arm who achieved an EVR, cEVR, and pEVR was 89%, 74%, and 14%, respectively, whereas the corresponding proportions in the standard arm were 80%, 62%, and 18%, respectively. The positive predictive values of EVR for SVR were similar for induction and standard arm being 59%-60% for EVR, 66%-72% for cEVR, and 19%-21% for pEVR. Negative predictive value of EVR for SVR was 100% and 98% in the induction regimen and standard arms, respectively (Table 2).

Table 2. Predictive Value of Early Viral Responses
  Induction Group Peg-IFN alfa-2a 360/180 μg Plus Ribavirin (n = 433)Standard Group Peg-IFN alfa-2a 180 μg Plus Ribavirin (n = 438)
NSVRNSVR
YesNoYesNo
N (%)N (%)N (%)N (%)
  1. Undetectable HCV RNA = < 15 IU/mL. Patients with missing HC RNA samples at week 12 are excluded. †Patients with missing HCV result at week 12 or nonassessable HCV RNA values are excluded. EVR, early viral response; cEVR, complete EVR; pEVR, partial EVR; PPV, positive predictive value; NPV, negative predictive value.

Undetectable HCV RNA or ≥ 2-log drop in HCV RNA at week 12 (EVR)*Yes384226 (59)158 (41)350211 (60)139 (40)
No280 (0)28 (100)631 (2)62 (98)
 % with EVR 89%  80% 
 PPV 59%  60% 
 NPV 100%  98% 
Undetectable HCV RNA at week 12 (complete EVR)*Yes322214 (66)108 (34)270194 (72)76 (28)
No9012 (13)78 (87)14318 (13)125 (87)
 % with cEVR 74%  62% 
 PPV 66%  72% 
 NPV 87%  87% 
Detectable HCV RNA with ≥ 2-log drop in HCV RNA at week 12 (partial EVR)†Yes6212 (19)50 (81)8017 (21)63 (79)
No250 (0)25 (100)611 (2)60 (98)
 % with pEVR 14%  18% 
 PPV 19%  21% 
 NPV 100%  98% 

Predictors of Sustained Virological Response.

Stepwise logistic regression analyses were used to identify baseline factors predictive of an SVR among the pooled intent-to-treat population. In the multiple logistic-regression model, the following factors were entered as variables in the final regression model: sex, age (≤40 years versus >40 years), race (Caucasian versus Asian; Caucasian versus other), body weight (<85 kg versus ≥85 kg), ALT quotient (≤1 versus >3; 1-≤3 versus >3), screening viral load (<400 kilo IU [KIU]/mL versus ≥800 KIU/mL; 400 to <800 KIU/mL versus ≥800 KIU/mL), HCV genotype subtype (non-1B versus 1B), cirrhotic status (cirrhosis versus noncirrhosis; missing versus noncirrhosis), and treatment group (induction versus standard). In multivariate analyses, age (odds ratio ≤40 years versus >40 years, 2.59; 95% confidence interval [CI], 1.87-3.58; P < 0.001), sex (odds ratio male versus female, 0.70; 95% CI, 0.50-0.97; P = 0.03), race (odds ratio Asian versus Caucasian, 2.33; 95% CI, 1.46-3.72; P = 0.0004), cirrhotic status (odds ratio cirrhosis versus noncirrhosis, 0.34; 95% CI, 0.21-0.54; P < 0.0001), and HCV viral load (odds ratio <400,000 IU/mL versus ≥800,000 IU/mL, 2.20; 95% CI, 1.47-3.28; P = 0.0001) were independent predictors of SVR. Notably, body weight was not significant. Additional analysis was performed using the same model but replacing body weight with body mass index (BMI) as an independent variable. However, BMI was not significant in predicting SVR.

Subgroup Analyses.

Subgroup analyses of various baseline factors in the intent-to-treat population indicated that there were no significant or consistent differences or trends in SVR between induction and standard arms (Table 3). In both treatment groups, SVR rates in general decreased with increasing age, body weight and BMI, higher baseline HCV RNA titer, presence of cirrhosis, and increasing Metavir stage of liver fibrosis. This trend is consistent with the results from the multivariate logistic regression analysis as detailed above. In particular, in both the standard and induction arms, SVR was higher among patients with no or minimal fibrosis (64% and 60%, respectively) compared to those with moderate (51% and 50%) or severe fibrosis/cirrhosis (28% and 24%; Table 3). Of note, high-dose PEG-IFNα-2a given for 12 weeks did not confer an additional SVR advantage in patients with high body weight (>85 kg).

Table 3. Effect of Pretreatment Patient Characteristics on Sustained Virological Response (SVR)
GroupInduction Group Peg-IFN alfa-2a 360/180 μg Plus Ribavirin (n = 433)Standard Group Peg-IFN alfa-2a 180 μg Plus Ribavirin (n = 438)
 n/N (%)n/N (%)
  1. SVR defined as HCV RNA < 15 IU/mL at week 72.

All patients230/433 (53)219/438 (50)
Age  
 ≤ 40 years104/146 (71)97/141 (69)
 > 40 years126/287 (44)122/297 (41)
Sex  
 Male149/298 (50)134/285 (47)
 Female81/135 (60)85/153 (56)
Race  
 Caucasian183/355 (52)167/365 (46)
 Asian40/61 (66)40/55 (73)
 Other7/17 (41)12/18 (67)
Weight (kg)  
 < 6565/108 (60)51/88 (58)
 65 to < 7565/108 (60)55/103 (53)
 75 to < 8537/78 (47)50/106 (47)
 ≥ 8563/139 (45)63/141 (45)
Body mass index (kg/m2)  
 < 27143/260 (55)122/236 (52)
 ≥ 2787/173 (50)97/202 (48)
Pretreatment HCV RNA level (IU/mL)  
 Low (< 400,000)49/71 (69)59/88 (67)
 Middle (400,000 to < 800,000)32/54 (59)25/50 (50)
 High (≥ 800,000)147/302 (49)132/293 (45)
Metavir stage of liver fibrosis  
 F020/28 (71)17/25 (68)
 F166/107 (62)60/104 (58)
 F262/121 (51)57/113 (50)
 F315/46 (33)15/51 (29)
 F42/14 (14)1/16 (6)
Cirrhosis/transition to cirrhosis  
 Yes17/60 (28)16/67 (24)
 No148/256 (58)134/242 (55)

Safety.

The overall safety profile was similar in the two groups with no new safety concerns observed with the induction dose regimen. Although the types of adverse events reported in both groups were similar, some common adverse events were reported with slightly increased frequency in the induction arm including loss of appetite, weight decrease, and chills (Table 4). The frequency of severe adverse events was comparable in both groups (24% versus 26%). Similarly, serious adverse events occurred at a similar frequency in the two groups (11% versus 9%). The most frequent type of serious adverse event in both groups was serious infections (Table 4). One death occurred in each group; both were caused by aortic dissection.

Table 4. Incidence of Adverse Events, Dose Modifications, and Premature Discontinuation
VariableInduction Group Peg-IFN alfa-2a 360/180 μg Plus Ribavirin (n = 433)Standard Group Peg-IFN alfa-2a 180 μg Plus Ribavirin (n = 438)
  1. AE, adverse event. *Serious AE considered by the investigator to be remotely, possibly or probably related to treatment.

Adverse events (AEs), n (%)  
Overall431 (100)435 (99)
Most frequent (≥ 15% in at least one group)  
 Headache227 (52)208 (47)
 Insomnia195 (45)207 (47)
 Influenza-like illness180 (42)183 (42)
 Nausea179 (41)169 (39)
 Fatigue159 (37)174 (40)
 Decreased appetite160 (37)113 (26)
 Lethargy134 (31)123 (28)
 Alopecia122 (28)99 (23)
 Weight decreased120 (28)87 (20)
 Myalgia114 (26)97 (22)
 Rash110 (25)116 (26)
 Irritability109 (25)111 (25)
 Diarrhoea101 (23)76 (17)
 Depression84 (19)85 (19)
 Arthralgia82 (19)76 (17)
 Dizziness84 (19)79 (18)
 Dry skin76 (18)86 (20)
 Pruritus76 (18)88 (20)
 Cough70 (16)61 (14)
 Pyrexia66 (15)47 (11)
 Chills64 (15)34 (8)
Severe (of all adverse events)105 (24)112 (26)
Serious AEs46 (11)45 (10)
 Treatment related*17 (4)20 (5)
 Serious infection14 (3)8 (3)
Dose modification for adverse event or laboratory abnormality, n (%)  
Peginterferon alfa-2a119 (27)78 (18)
 Adverse event34 (8)21 (5)
 Laboratory abnormality93 (21)62 (14)
  Neutropaenia76 (18)55 (13)
  Thrombocytopaenia17 (4)6 (1)
  Anemia5 (1)3 (1)
Ribavirin108 (25)90 (21)
 Adverse event59 (14)37 (8)
 Laboratory abnormality64 (15)63 (14)
  Anemia59 (14)61 (14)
Grade 4 laboratory abnormality  
 Neutrophil count < 0.50 × 109/L34 (8)16 (4)
 Platelet count < 20 × 109/L2 (<1)0 (0)
 Hemoglobin <8.5 g/dL10 (2)9 (2)
Premature treatment discontinuation for adverse event or laboratory abnormality, n (%)  
Peginterferon alfa-2a41 (9)32 (7)
Ribavirin44 (10)36 (8)

Dose modification of PEG-IFNα-2a was higher overall in the induction group (27%) than in the standard group (18%) due to a higher incidence of laboratory abnormalities and adverse events in the induction group (Table 4). During the first 12 weeks, PEG-IFNα-2a dose modifications for laboratory abnormalities (mainly neutropenia) and adverse events occurred more frequently in the induction group (19%) than in the standard group (9%). Rates of premature withdrawal from PEG-IFNα-2a treatment for safety reasons were similar in the two groups (9% versus 7%). Ribavirin dose modifications (mainly for anemia) were higher in the induction arm (14% versus 8%) during the first 12 weeks and remained slightly higher for the total treatment period (25% versus 21%).

Severe neutropenia (neutrophils <0.50 × 109/L) occurred more often in the induction arm (8% versus 4%) especially during the first 12 weeks of treatment (6% versus 2%). The use of growth factors for the treatment of neutropenia was low in both groups: 2% in the induction group and 1% in the standard group. Thrombocytopenia (platelets <50 × 109/L) occurred at a similar frequency in the two arms (4% and 3%), whereas severe thrombocytopenia (platelets <20× 109/L) occurred in two patients in the induction arm. Severe anemia (hemoglobin <8.5 g/dL) occurred at 2% in both arms. The use of hematopoietic stimulants for anemia was infrequent (2% in each group; Table 4).

Discussion

This large randomized study of treatment-naïve CHC patients infected with HCV genotype 1 demonstrates that treatment with a 360 μg induction dose of PEG-IFNα-2a for the first 12 weeks followed by a standard 180 μg dose for 36 weeks in combination ribavirin fails to significantly improve SVR rates compared to treatment with standard dose PEG-IFNα-2a plus ribavirin for 48 weeks. The lack of benefit from the induction strategy we evaluated is supported by near identical SVR rates (53% versus 50%) within a large study population, similar findings from both the intent-to-treat and per-protocol analysis, and lack of trends toward benefit in subgroups, including the high viral load group and high body weight group.

Several smaller studies have evaluated induction therapy with high-dose and/or frequent dose (peg)interferon, yielding generally disappointing results.4–11 Most of these studies investigated standard interferon and included both HCV genotype 1 and non-1 patients, with variable periods of induction (4-14 weeks). Induction with standard interferon has improved initial HCV RNA decay11 and initial virological responses,8 but has in general failed to improve SVR.6–11 The exception is the study by Ferenci et al. that demonstrated significantly higher SVR rates with high-dose daily interferon among the subgroup of patients with HCV genotype 1, although not in the overall study population.4 Limited data exists on the efficacy of induction therapy with peginterferon alfa in patients infected with HCV genotype 1.5, 10, 17 Of these, only one previous published study has reported on high-dose induction therapy with peginterferon alfa in patients with HCV genotype 1.10 This Romanian study evaluated the efficacy of an initial 12 weeks of induction therapy with high-dose PEG-IFNα-2b (3.0 μg/kg/week) compared to standard dose PEG-IFNα-2b (1.5 μg/kg/week) in a small, randomized controlled trial of CHC patients including 75 with HCV genotype 1. Virological responses through the first 12 weeks and SVR rates were similar between the two study arms in both genotype 1 and non-1 patients, suggesting there was no early or late benefit from high-dose PEG-IFNα-2b therapy.10 A larger randomized study comparing the same high-dose induction PEG-IFNα-2b regimen (3.0 μg/kg/week for 12 weeks), recently reported in abstract form, showed similar outcomes in patients with HCV genotype 1.17 SVR rates were low in both study groups with no differences seen between those treated with induction dosing versus standard dosing (34% versus 32%). Our study is the largest randomized controlled trial of induction dosing of PEG-IFNα-2a/2b conducted to date in CHC genotype 1 patients, and the results clearly demonstrate lack of benefit of induction dosing for enhancing SVR in treatment-naïve CHC genotype 1 patients.

However, the 360-μg induction dose of PEG-IFNα-2a given for 12 weeks was associated with improvements in EVR; rates of undetectable HCV RNA (weeks 4, 8, and 12) were 10%-12% higher in the induction arm and initial mean HCV RNA reductions were greater (data not shown), demonstrating that enhanced efficacy occurred early and was maintained throughout the induction period. These results are in contrast to data by Brady et al., who found no significant difference in week-12 virological response (undetectable HCV RNA) between HCV genotype 1 patients treated with induction dose compared to standard dose PEG-IFNα-2b (63% versus 58%).17 The benefit of PEG-IFNα-2a induction therapy in enhancing EVR, and in particular RVR rates, may have potential relevance in the development of future treatment strategies involving the combination of PEG-IFNα-2a with direct antiviral agents. Further evaluation of this paradigm is needed to determine whether such an approach improves treatment response and/or shortens treatment duration.

A key observation from our study is that the superior RVR rates achieved with PEG-IFNα-2a induction therapy failed to translate into improved SVR rates. This contrasts with studies involving direct-acting antiviral agents such as telaprevir, that enhance both early and SVR rates when combined with standard PEG-IFNα-2a and ribavirin.18 Why then does induction therapy fail to improve treatment outcomes despite higher EVR? Analysis of viral response rates between the two treatment groups over the first 24 weeks provides some possible insights. Throughout the 12-week induction period, there was a 10%-12% differential in virological response rates between the groups in favor of induction dosing. However, this gap reduced to 7% by treatment week 24, after 12 weeks of standard therapy. Notably, this was not due to the failure to maintain viral suppression in the induction group but rather continued improvement in the virological response rate in the standard group. This suggests that the main effect of high-dose PEG-IFNα-2a therapy is to augment the first phase decay of HCV RNA related to viral replication inhibition rather than to achieve a significant and/or sustainable increase in the rate of second-phase viral decline related to intrahepatic HCV clearance that is an important driver of treatment outcome. Comparative studies of early viral kinetics on high-dose and standard-dose PEG-IFNα-2a therapy during the initial 12 weeks of therapy and beyond are underway to test this concept.

As one of the largest clinical trials conducted in this patient population, our study also provided important additional information relevant to the treatment of patients with HCV genotype 1. The 50% SVR rate achieved was similar to that reported in previous studies while the independent baseline prognostic factors identified in multivariate analyses including age, sex, histological stage, and viral load were consistent with prognostic factors identified previously in patients infected with genotype 1. An RVR, achieved in around one-quarter of patients in the standard arm, had a positive predictive value for SVR of 80%, similar to that recently reported in patients treated with 24 weeks of standard therapy19 although somewhat lower than the 89% reported from a retrospective analysis of a large international phase III registration study.15 Consistent with other studies, EVR was the most reliable predictor of nonresponse with a negative predictive value of 98% in the standard therapy group.1, 20, 21 Interestingly, the data also suggested that induction dosing, although associated with higher EVR, did not significantly change the predictive value of virological response parameters at week 4 or 12 for achieving an SVR.

The relationship between stage of fibrosis and SVR was particularly striking. In both the standard and induction arms, SVR was considerably higher among patients with no or minimal fibrosis (64% and 60%, respectively) compared to those with severe fibrosis or cirrhosis (28% and 24%). In addition, significantly higher SVR rates were noted in younger (≤ 40 years) subjects in both treatment groups compared to those >40 years (69%-71% versus 41%-44%). Together, these results highlight the benefit of early treatment in maximizing SVR rates in patients with HCV genotype 1. Importantly, body weight and BMI did not influence SVR. Moreover, no increase in SVR was observed in high body weight patients treated with induction dosing compared to standard dosing (45% versus 45%).

The safety profile of PEG-IFNα-2a and ribavirin combination therapy in this study was similar to that previously reported in treatment-naïve patients with CHC, with no new significant safety concerns identified either in the overall population or in those treated with PEG-IFNα-2a induction therapy. Indeed, the safety profile and tolerability of the induction treatment regimen was similar to the standard treatment regimen with the frequency of serious adverse events and rates of premature withdrawal from PEG-IFNα-2a for safety reasons similar between the two groups (9% versus 7%). However, some safety-related events likely associated with high-dose PEG-IFNα-2a were observed with increased frequency in the induction group, including decreased appetite and weight, diarrhea, pyrexia, and chills. In addition, the induction group had a higher frequency of dose modifications of PEG-IFNα-2a for safety reasons predominantly related to laboratory abnormalities and in particular neutropenia, with grade 4 neutropenia (neutrophil count <0.50 × 109/L) occurring slightly more frequently with induction therapy. Nevertheless, these differences in clinical and laboratory adverse events did not lead to increased discontinuation rates of PEG-IFNα-2a therapy, which suggests that such events were clinically manageable, often by dose modification.

In treatment-naïve CHC patients infected with HCV genotype 1, high-dose PEG-IFNα-2a during the first 12 weeks of treatment in combination with ribavirin was safe and well tolerated and resulted in significant improvements in EVR but not SVR compared with standard dose PEG-IFNα-2a. Although the results conclusively demonstrate that a high-dose PEG-IFNα-2a induction strategy does not achieve additional efficacy benefits as compared with the currently approved treatment regimen, the enhanced EVR observed with high-dose PEG-IFNα-2a could have potential relevance to future treatment strategies that combine direct antiviral agents with PEG-IFNα-2a and ribavirin.

Acknowledgements

The authors thank Deborah Richards, John Miller, Arthur Alston, and Jill Schaefer from Roche Australia for their managerial assistance, the late Amy Lin and Phillip McCloud from Roche for their statistical consultant advice, Monica Eason for acting as global study leader, James Thommes for acting as global science leader, and Helen Cattell for acting as study data manager. In addition, we thank all the study coordinators and study management team members who worked on this study.

Appendix 1

List of Chariot Study Group Investigators by Country.

Argentina: José Curciarello, Jorge Daruich, Hugo Fainboim, Hugo Tanno.

Australia: Peter Angus, David Badov, Wendy Cheng, Mark Cornwell, Darrell Crawford, Paul Desmond, Greg Dore, Damian Dowling, John Freiman, Jacob George, Peter Gibson, Hugh Harley, Brian Hughes, Gary Jeffrey, Graham Kaye, Ian Kronborg, Alice Lee, Barbara Leggett, Miriam Levy, George Marinos, John Masson, Geoff McCaughan, Graeme McDonald, Jenny McDonald, Christopher Merridith, Lindsay Mollison, John Quin, Stuart Roberts, Alexander Rodgers, Joe Sasadeusz, William Sievert, Don Walker, Martin Weltman.

Canada: Robert Bailey, Nir Hilzenrat, Krishna Menon, Mark Swain, Eric Yoshida.

Mexico: Teresita de Jesus Beltran, Pedro Gomez-Quiroz, Linda Munoz.

New Zealand: Edward Gane, Catherine Stedman, Frank Weilert.

Thailand: Anuchit Chutaputti, Tawesak Tanwandee, Satawat Thongsawat.

Ancillary