Retreating chronic hepatitis C with daily interferon alfacon-1/ribavirin after nonresponse to pegylated interferon/ribavirin: DIRECT results


  • fax: 314-577-8125


Up to 50% of patients with chronic hepatitis C fail to respond to initial therapy with pegylated interferon (PEG-IFN) and ribavirin (RBV). With unsuccessful viral eradication, these patients remain at risk for developing progression of their liver disease. Retreatment with PEG-IFN/RBV yields sustained virologic response (SVR) rates that are under 10%. A wholly synthetic interferon, interferon alfacon-1 or consensus interferon (CIFN) given with RBV, was evaluated in patients who failed initial PEG-IFN/RBV therapy. The intent-to-treat analysis included 487 patients; 245 received CIFN 9 μg/day and RBV, and 242 received CIFN 15 μg/day and RBV. Within this group of patients, 59.3% had documented advanced fibrosis at baseline liver biopsy (stage F3 or F4). SVR rates were 6.9% (17/245 patients) in the 9 μg group and 10.7% (26/242) in the 15 μg group. In the intent-to-treat analysis, SVR rates were higher among patients with a >2-log10 decrease in hepatitis C virus RNA during prior PEG-IFN/RBV therapy: 11% (4/38) in the 9 μg group and 23% (7/31) in the 15 μg group. Among patients with lower baseline fibrosis scores (F0-F3), SVR rates were 7.8% (15/192) in the 9 μg group and 13.1% (23/175) in the 15 μg group. In this same group of patients (F0-F3), if a >2-log10 decrease in hepatitis C virus RNA with previous PEG-IFN/RBV treatment was achieved, SVR rates improved to 10.7% and 31.6% in the 9 μg and 15 μg groups, respectively. CIFN/RBV combination retreatment was safe and well tolerated. Conclusion: Retreatment of PEG-IFN and RBV nonresponders with CIFN and RBV is safe and efficacious and can be considered a retreatment strategy for patients failing previous therapy with PEG-IFN/RBV, especially in interferon-sensitive patients with lower baseline fibrosis scores. (HEPATOLOGY 2009.)

Since 2001, the standard of care for patients with chronic hepatitis C has been the combination of pegylated interferon (PEG-IFN) and ribavirin (RBV).1, 2 This combination has produced sustained virologic response (SVR) rates of 50%-60% in patients infected with hepatitis C virus (HCV) genotype 1 who adhere to their therapeutic regimens and 40% in intention-to-treat populations.1, 2 However, because only about 65% of patients become HCV RNA–undetectable when treated with this regimen, more than one-third of all patients are classified as nonresponders. Some of these patients have relatively mild liver disease but may have symptoms of HCV viremia, while other patients have advanced fibrosis and are at risk for developing complications of chronic liver disease, including decompensated cirrhosis and hepatocellular carcinoma, and may require liver transplantation.3–5

The optimal approach to PEG-IFN/RBV nonresponders has not been well defined. Some clinicians have used the “watchful waiting” approach6 and are anticipating new antiviral therapies with either protease inhibitors or polymerase inhibitors. However, it remains to be determined just how effective these new agents will be when combined with PEG-IFN and RBV in the retreatment of PEG-IFN/RBV nonresponders.7

Alternative therapies have included retreatment with the alternative brand of PEG-IFN not used in the initial therapy, although most results with this approach have been disappointing. Other approaches have included prolonged treatment with PEG-IFN, maintenance therapy, or the use of higher dosages of either PEG-IFN and/or RBV.8–12 The strategy studied in the current investigation included high doses of daily consensus interferon (CIFN) (Infergen; interferon alfacon-1) 9 or 15 μg/day given with RBV.


AE, adverse event; bDNA, branched DNA; CIFN, consensus interferon; DIRECT, Daily-Dose Consensus Interferon and Ribavirin: Efficacy of Combined Therapy; Hb, hemoglobin; HCV, hepatitis C virus; ITT, intention to treat; PEG-IFN, pegylated interferon; RBV, ribavirin; REPEAT, REtreatment with PEgasys in PATients Not Responding to Peg-Intron Therapy; SVR, sustained virologic response; TMA, transcription-mediated amplification.

Patients and Methods

This study, referred to as the DIRECT (Daily-Dose Consensus Interferon and Ribavirin: Efficacy of Combined Therapy) trial, was designed by the sponsor and by several of the academic investigators. The data were managed by the sponsor and the academic investigators. The sponsor performed the statistical analysis. The academic investigators were responsible for the development of the final manuscript and had unrestricted access to the data. An author involved with the design or execution of this study either wrote or edited every section of the manuscript. Both an academic author (B. R. B.) and an industry representative (Michael Beckloff, Three Rivers Pharmaceuticals, Cranberry Township, PA) attest to the completeness and accuracy of the data.

Study Design.

This was a phase 3, randomized, open-label, multicenter, U.S.-based registration trial conducted to investigate the efficacy, tolerability, and safety of daily CIFN at dosages of 9 and 15 μg/day (interferon alfacon-1, Infergen; Three Rivers Pharmaceuticals, LLC, Cranberry Township, PA) administered with daily RBV (Ribasphere, Three Rivers Pharmaceuticals, LLC) compared with no treatment in patients who did not respond to prior therapy with either PEG-IFN alfa-2a or alfa-2b and RBV. The trial was divided into 2 sections: DIRECT-001 and DIRECT-002 (Fig. 1).

Figure 1.

DIRECT study design: a randomized, open-label study of CIFN and RBV in patients who did not respond to previous combination therapy with PEG-IFN and ribavirin.

Patients were randomized at a 1:1:1 ratio into three study groups: CIFN 9 μg/day (group 1), 15 μg/day (group 2) plus oral RBV 1,000-1,200 mg/day (based on body weight), or a control, no-treatment group (group 3). The no-treatment group was mandated by the U.S. Food and Drug Administration in order to provide a comparison of safety for the two treatment groups. It was not anticipated that any of the patients randomized to the control group would have a spontaneous response. After 24 weeks of observation, all patients in the control group of DIRECT-001 were offered randomization into DIRECT-002 to receive CIFN 9 or 15 μg/day plus RBV.

At week 24, patients who had undetectable plasma HCV RNA by branched DNA (bDNA) assay, confirmed by transcription-mediated amplification (TMA) assay, or who had a ≥2-log10 decrease from baseline in HCV RNA were assigned to continue therapy to week 48. Patients with a <2-log10 decrease from baseline in plasma HCV RNA (bDNA assay) were considered nonresponders and were withdrawn from treatment. At week 48, patients with undetectable plasma HCV RNA (by bDNA and TMA assays) were assigned to return for regular visits in the follow-up period (weeks 52, 60, 68, and 72) until 24 weeks after their last dose of study drug (week 72). Patients with detectable plasma HCV RNA (bDNA or TMA assay) at any time between weeks 48 and 72 were classified as relapsers.

All patients who discontinued therapy early at any time were instructed to return for a single follow-up visit 30 days after their last dose of study drug to complete early termination/discontinuation assessments. Patients who had undetectable plasma HCV RNA by bDNA and TMA assays at the time of stopping therapy or at the early termination/discontinuation visit were to return for follow-up plasma HCV RNA assessments through week 72, as long as their plasma HCV RNA levels remained undetectable by both bDNA and TMA assays. Patients who discontinued for nonresponse at weeks 12 or 24 did not undergo a follow-up HCV RNA measurement.

Patients in the no-treatment group in DIRECT-001 were given the option to enroll in DIRECT-002 under the following conditions: if they achieved a <2-log10 decrease in plasma HCV RNA at week 24 compared with baseline or if they had detectable plasma HCV RNA by bDNA or by TMA at week 48. These patients were then treated according to the same protocol followed in DIRECT-001. The results of DIRECT-001 and DIRECT-002 were pooled for the purpose of this analysis.


Men and women were eligible for enrollment if they were chronically infected with HCV of any genotype. Chronic infection was identified based on a history of being positive for serum anti-HCV and/or HCV RNA. A liver biopsy performed within 3 years of screening must have demonstrated evidence of chronic HCV infection. Hepatic fibrosis was interpreted by local pathologists based upon the Metavir scoring system. Patients with advanced liver disease, including bridging fibrosis (F3) and cirrhosis (F4), were eligible for the study as long as they had normal liver function as evidenced by serum albumin >3.5 mg/dL, platelet count >75,000/mm3, and no prior episode of hepatic decompensation (variceal hemorrhage, hepatic encephalopathy, ascites, or hepatocellular carcinoma).

Prior nonresponse and adherence after initial therapy with PEG-IFN alfa-2a (180 μg/week) or PEG-IFN alfa-2b (1.5 μg/kg/week) plus RBV was determined via careful chart review completed by the study site principal investigator and confirmed by an external study monitor. Nonresponders had to have had a <2-log10 decrease in HCV RNA between weeks 12 and 24 or detectable HCV RNA at weeks 24 or 48. Patients were required to have had completed a minimum of 90 days between discontinuation of their prior regimen and the start of the current study medication. All patients had to have received at least 80% of the cumulative standard dosages of PEG-IFN and RBV for at least 38 weeks (80% of the planned treatment duration). Patients were excluded if this previous treatment was prematurely discontinued, dosing was interrupted, or if the dose of PEG-IFN was reduced because of noncompliance, safety, or tolerability issues (including hematologic or psychiatric side effects).

Patients were also excluded if they were pregnant or lactating women or male partners of pregnant women, or if they were not suitable candidates for enrollment or unlikely to comply with the requirements of the study in the opinion of the investigator or sponsor.

Treatment and Assessments.

Screening took place between 8 weeks and 1 day before the first day of treatment. After providing informed consent, patients were screened for inclusion criteria, underwent a physical examination, provided a baseline medical history, and had blood drawn for laboratory testing. After screening, eligible patients were randomized in a 1:1:1 ratio to receive CIFN 9 or 15 μg/day plus RBV 1,000 mg/day (body weight <75 kg) or 1,200 mg/day (body weight >75 kg) or no treatment (Fig. 1). RBV was provided as capsules containing 200 mg of active drug. An independent data monitoring committee conducted regular interim safety assessments throughout the study.

Plasma HCV RNA levels were determined first using the bDNA quantitative assay, which has a sensitivity of detection of 615 IU/mL and a reportable range of 615 to 6,920,000 IU/mL. The Bayer TMA assay, with a sensitivity of detection of 5 IU/mL, was used whenever HCV RNA levels were undetectable via bDNA assay.

Patients developing anemia, defined as hemoglobin (Hb) <10 g/dL, were managed by reducing the dose of RBV to 600 mg/day. The use of growth factors was not permitted. If the Hb increased to >10 g/dL, the RBV dose could be increased in 200-mg/day increments as tolerated according to the discretion of the site principal investigator. RBV dose was not increased after being reduced to 600 mg/day for patients with a history of cardiovascular disease whose Hb decreased by 2 g/dL or more during any 4-week period. RBV was permanently discontinued in patients whose Hb dropped below 8.5 g/dL. In those patients with a history of cardiac or cerebrovascular disease, Hb remaining below 12 g/dL after 4 weeks on a reduced dose required permanent discontinuation of RBV. Neutropenia was managed by CIFN dose reduction; in patients whose absolute neutrophil count fell to <0.75 × 109/L, starting doses of 15 μg were lowered to 9 μg and then to 6 μg, and starting doses of 9 μg were lowered to 6 μg.

Efficacy Variables.

The primary efficacy variable was the proportion of patients with SVR, defined as undetectable plasma HCV RNA by both bDNA and TMA assays at 24 weeks after the last dose of study drug. In addition, SVR was further explored for the effect of race, genotype, sex, age, baseline HCV RNA, presence/absence of cirrhosis, body weight, and previous response to PEG-IFN/RBV.

Safety and Tolerability.

All adverse events (AEs) and serious AEs were recorded for patients who received at least 1 dose of study medication (active-treatment groups) or who completed baseline assessments (no-treatment group). AEs were recorded until either 30 days after the last dose of study medication (active-treatment groups) or until the last study visit (no-treatment group). AEs were graded from 1 to 5 (1, mild; 2, moderate; 3, severe; 4, life-threatening or disabling; 5, death) based on the Common Toxicity Criteria for Adverse Events v3.0. An AE was considered a serious AE if it resulted in death, was life-threatening, required inpatient hospitalization, or resulted in persistent or significant disability or incapacity.

Statistical Methods.

At least 170 patients were needed in each of the three study groups (for a total of 510 patients) to provide an approximately 91% power to detect a difference in SVR between each of the active treatment groups and the no-treatment group. This analysis was performed using a two-sided Fisher's exact test at α = 0.05 significance level, with adjustment for multiple comparisons, and assumed an SVR rate of 10% for either of the active treatment groups and 1.2% for the no-treatment group. The study was not powered to detect differences between the 9 μg and the 15 μg arms.

Data were summarized and analyzed for two patient populations: the intention-to-treat (ITT) population and patients who did not receive any dose modifications. The ITT group consisted of all patients who were randomized to receive CIFN in DIRECT-001, as well as all patients from the no-treatment group in DIRECT-001 who went on to receive at least 1 dose of CIFN in DIRECT-002. Data from the ITT population were used in all efficacy and safety analyses.

Descriptive statistics were determined for continuous variables (patient counts, mean, standard deviation, median, minimum, and maximum) and categorical variables (number and percentage of patients for each category). Percentages were calculated using the number of patients without missing data as the denominator unless otherwise indicated. Calculations of virologic response (both sustained and at specific visits) used the number of ITT patients as the denominator. All statistical testing was conducted at the 0.05 level of significance using SAS software, version 8.2.


Patients and Disposition.

Five hundred fifteen patients were randomized at 44 sites in the United States and Puerto Rico to receive CIFN 9 μg/day plus RBV 1,000 or 1,200 mg/day (n = 171), CIFN 15 μg/day plus RBV (n = 172), or no treatment (n = 172). Of the 172 patients in the no-treatment group in DIRECT-001, 144 continued on to DIRECT-002. Of these, 74 received CIFN 9 μg/day plus RBV 1,000 or 1,200 mg/day, and 70 received CIFN 15 μg/day plus RBV 1,000 or 1,200 mg/day. The final ITT population included 487 patients (245 who received CIFN 9 μg/day and 242 who received CIFN 15 μg/day).

Baseline demographic and clinical characteristics of the two CIFN treatment groups are presented in Table 1. The majority of patients were male (70%) and Caucasian (64%). Of the enrolled patients, 59.3% had advanced liver disease on biopsy, including bridging fibrosis (F3; 35%) or cirrhosis (F4; 25%). In addition, 52% of patients had hepatic steatosis. The average time between biopsy sampling and study day 1 was 1.6 years. Patients included in the DIRECT trial were required to be off PEG-IFN/RBV therapy for at least 3 months prior to starting CIFN therapy. The median washout period between previous treatment and day 1 of CIFN therapy was 448 days (15 months) and 506 days (16.8 months) for the 9 μg and 15 μg groups, respectively. Sixty-eight percent of the patients had high baseline HCV RNA levels of >850,000 IU/mL. The majority of patients (79%) failed to achieve an early virologic response (at least a 2-log10 drop in HCV RNA from the pretreatment baseline) to previous PEG-IFN therapy.

Table 1. DIRECT-001 and DIRECT-002: Baseline Characteristics
 CIFN 9 μg/day + RBV (n = 245)CIFN 15 μg/day + RBV (n = 242)
  • Abbreviation: SD, standard deviation.

  • *

    Steatosis defined as present or absent.

Mean age ± SD, years51 ± 6.6550 ± 6.59
HCV genotype 195%96%
High viral load  
 ≥850,000 IU/mL68%68%
Mean weight ± SD, kg89.1 ± 18.6489.4 ± 17.59
Mean body mass index ± SD, kg/m229.3 ± 5.2029.6 ± 5.02
 African American21%17%
Liver biopsy results  
 Cirrhosis (F4)22%28%
 Bridging fibrosis (F3)36%34%
Response to prior therapy  
 <2-log10 drop78%80%
 >2-log10 drop15%12%
Washout interval, days  
 Mean ± SD453 ± 345594 ± 372

Antiviral Efficacy.

By ITT analysis, pooled end-of-treatment response via TMA assay in the pooled 9 μg arm was 14.7% (36/245), with a subsequent SVR of 6.9% (17/245). In the 15 μg arm, pooled end-of-treatment response via TMA assay was 18.5% (45/242), with an SVR rate of 10.7% (26/242). Relapse rates pooled for both arms were 52% (19/36) and 42% (19/45) for the 9 μg and 15 μg groups, respectively. Post hoc analysis revealed steatosis and time to viral negativity had the most impact on relapse rates.13 As expected, in DIRECT 001, patients in the no-treatment arm achieved a 0% SVR. In patients who did not have dose modifications, overall SVR rates were 7% in the CIFN 9 μg group and 17% in the 15 μg group (Fig. 2A,B). The SVR rates were not significantly different between the 001 and 002 arms (P = 0.818). Although the study was not powered to detect differences between the 9 μg and 15 μg groups, a post hoc analysis revealed no difference in SVR rates between the two (P = 0.141, 95% CI −8.8%-1.2%).

Figure 2.

Rates of SVR, defined as undetectable viral levels at least 24 weeks after the end of treatment with CIFN and RBV, in (A) the ITT population (n = 487) and (B) patients who did not receive dose modifications (n = 281). Active versus no treatment (n = 172).

Patients who achieved a complete early virologic response (defined as viral negativity at week 12 via TMA assay) were more likely to demonstrate an SVR than the general study population. In the 9 μg group, 81.3% (13/16) of patients with complete early virologic response achieved SVR, whereas in the 15 μg group, 63.6% (14/22) of patients with complete early virologic response demonstrated SVR. In patients deemed slow responders (>2-log drop at week 12, viral-negative at week 24), SVR rates were 11.7% (2/17) and 35.4% (11/31) in the 9 μg and 15 μg groups, respectively. Two patients in the 9 μg arm and one patient in the 15 μg arm achieved SVR despite being viral-positive at week 24.

Patients achieving the greatest log reduction in terms of viral response to initial PEG-IFN/RBV therapy had the best likelihood of responding to retreatment with CIFN and RBV (Fig. 3A,B). Among F0-F2 patients with >2-log10 decreases in HCV RNA during their prior PEG-IFN/RBV therapy, SVR rates were 13.3% (2/15) and 30.0% (3/10) in the 9 μg and 15 μg groups, respectively. A similar trend was seen in the 15 μg arm in patients with bridging fibrosis (F3). SVR for the F0-F2 group of patients was 8.7% (9/104) in the 9μg group and 14.9% (14/94) in the 15 μg group. With patients displaying bridging fibrosis only on biopsy, overall SVR was 6.8% (6/88) and 11.1% (9/81) in the 9 and 15 μg groups, respectively. Complete SVR for noncirrhotics (F0-F3) was 7.8% (15/192) in the 9 μg group versus 13.1% (23/175) in the 15 μg group, whereas cirrhotics achieved SVR rates of 3.8% (2/53) and 4.5% (3/67) in the 9 μg and 15 μg groups, respectively. In the cirrhotic cohort, patients required at least a 1-log drop on prior therapy to benefit from retreatment with CIFN and RBV. African American patients achieved lower SVR rates than Caucasians (4.2% versus 11%, respectively). In this population, pooled analysis between the two dosage arms revealed that 35.4% of the patients had failed at least two or more prior treatment regimens, with 52.7% having obtained a <1-log drop on prior PEG-IFN/RBV therapy. In addition to this, 60.2% had dose reductions while on CIFN/RBV therapy. Finally, non–genotype 1 patients (genotype 2/3) achieved an overall SVR rate of 23.1% (3/13) and 88.8% (8/9) in the 9 μg and 15 μg groups, respectively. Further univariate predictors of response are discussed in Table 2.

Figure 3.

Rates of SVR, defined as undetectable viral levels at 24 weeks after the end of treatment, in the ITT analysis by known previous response to therapy among patients without F0-F2 fibrosis, F3, and with cirrhosis (F4) in the (A) 9 μg and (B) 15 μg treatment groups. One patient in the 9 μg group and two patients in the 15 μg group achieving SVR had unknown prior response to PEG-IFN/RBV therapy.

Table 2. Univariate Predictors of SVR
 n (% SVR)P Value
  1. Abbreviation: NS, not significant.

Treatment group  
 9 μg/day (n = 245)17 (6.9)NS
 15 μg/day (n = 242)26 (10.7) 
 Male (n = 342)30 (8.8)NS
 Female (n = 145)13 (9.0) 
Fibrosis scores  
 F0-F2 (n = 198)23 (11.6)NS
 F3-F4 (n = 289)20 (6.9) 
 1 (n = 464)32 (6.9)<0.001
 Non–genotype 1 (n = 23)11 (47.8) 
Viral load  
 <850,000 IU/mL (n = 155)25 (16.1)<0.001
 >850,000 IU/mL (n = 331)18 (5.4) 

Safety and Tolerability.

A total of 83.6% of patients in the 9 μg group and 71.7% of patients in the 15 μg group received at least 80% of their cumulative CIFN dose. The most common reason for early termination was treatment failure. Discontinuation due to not achieving a >2-log drop at week 24 was similar between the pooled dosage arms of 001 and 002 (32.3% versus 28.4%, P value not significant). Other reasons for treatment discontinuation are listed in Table 3.

Table 3. Early Treatment Discontinuation: ITT Population
 Patients, n (%)
CIFN 9 μg/day + RBV (n = 245)CIFN 15 μg/day + RBV (n = 242)Total (N = 487)
Treatment failure127 (51.8)107 (44.2)234 (48.0)
 <2-log10 reduction in HCV RNA at week 2496 (39.2)64 (26.4)160 (32.9)
 Detectable HCV RNA at week 4831 (12.7)43 (17.8)74 (15.2)
Adverse events35 (14.3)51 (21.1)86 (17.7)
Withdrawal of consent20 (8.2)18 (7.4)38 (7.8)
Decision by principal investigator or sponsor3 (1.2)9 (3.7)12 (2.5)
Lost to follow-up9 (3.7)9 (3.7)18 (3.7)
Other15 (6.1)7 (2.9)2 (4.5)

Table 4 summarizes the most common AEs experienced by patients in the pooled 001 and 002 arms of the DIRECT trial. Most AEs were grade 2 or 3 and were more commonly related to administration of both CIFN and RBV than either drug alone. Most patients experienced at least one AE in the study. Individual AEs resulting from treatment with CIFN were typical of those reported with IFN-based therapy. All AEs were more common in the CIFN 9 and 15 μg groups than in the no-treatment group. RBV-induced hemolytic anemia occurred in 6.4% of patients. In general, most AEs were either not drug-related based on the opinion of the study site principal investigator or were thought to be related to the combination of study drugs rather than to either CIFN or RBV alone. The most common AEs leading to dose modifications in both CIFN treatment groups included neutropenia, fatigue, leukopenia, depression, nausea, myalgia, lymphopenia, and anemia. Overall, discontinuations for AEs occurred in 14% of the 9 μg group and 21% of the 15 μg group in the pooled ITT analysis.

Table 4. Adverse Events Reactions Occurring Regardless of Attribution
Body System/Preferred Term (MedDRA)Retreatment*
CIFN 9 μg/day for 48 weeks (n = 244)CIFN 15 μg/day for 48 weeks (n = 242)
  • Only events that occurred at a frequency of ≥5% in any treatment group of both IRHC-001 and IRHC-002 studies combined are included. Patients can appear more than once in Table 5. All values are percentages.

  • Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; WBC, white blood cell.

  • *

    Adverse events reported in patients during treatment or posttreatment observation are listed regardless of attribution to treatment.

Blood and lymphatic disorders  
Hemolytic anemia1619
Eye disorders  
Vision blurred87
Gastrointestinal disorders  
Abdominal pain77
Abdominal pain upper87
Dry mouth (saliva decreased)35
General disorders and administration site conditions (or body as a whole)  
Influenza-like illness (or symptoms)4042
Injection site erythema1616
Injection site pain35
Injection site reaction1512
Injection site rash25
Pain (or body pain)56
Pyrexia (or fever)1317
Upper respiratory tract infection56
Blood ALT increased46
Blood AST increased610
Blood phosphorus decreased75
Blood uric acid increased35
Lymphocyte count decreased35
Neutrophil count decreased54
WBC count decreased65
Weight decrease1622
Metabolism and nutrition disorders  
Decreased appetite1718
Musculoskeletal and connective tissue disorders  
Back pain129
Muscle cramp56
Nervous system disorders  
Memory impairment56
Psychiatric disorder  
State of confusion45
Respiratory, thoracic, and mediastinal disorders  
Dyspnea exertional109
Pharyngolaryngeal pain56
Skin and subcutaneous tissue disorders  
Dry skin98
Hyperhidrosis (or sweating increased)25


Retreatment of PEG-IFN/RBV nonresponders with daily CIFN/RBV resulted in an SVR rate of 6.9% with 9 μg/day CIFN and 10.7% with 15 μg/day CIFN. Patients whose doses were not reduced achieved SVR rates of 7% in the 9 μg group and 17% in the 15 μg group. These findings are consistent with a previous clinical trial demonstrating encouraging SVR rates with this higher dose of CIFN, 15 μg/day.14 The best response rate, 31.6%, was observed in noncirrhotic patients (F0-F3) who had a partial virologic response with a >2-log10 decline in HCV RNA during their previous course of PEG-IFN treatment.

These results were achieved even though the patients in the DIRECT trial had numerous poor prognostic factors for a successful response. Approximately 95% had HCV genotype 1, about 20% were African American, 68% had a high baseline HCV RNA level of >850,000 IU/mL, and almost 90% had a baseline HCV RNA level of >400,000 IU/mL. Approximately 80% had a <2-log10 decline in HCV RNA during prior treatment. Sixty percent of patients had advanced liver disease, including cirrhosis (25%) and bridging fibrosis (35%), and 52% had steatosis on biopsy. All of these factors have been shown to significantly reduce rates of SVR.

Overall, the alternative strategies for improving SVR in PEG-IFN/RBV nonresponders have not met with success. Two trials of maintenance IFN therapy were evaluated in PEG-IFN/RBV nonresponders with advanced fibrosis or cirrhosis to determine if this strategy can reduce progression to cirrhosis, complications of cirrhosis, hepatocellular carcinoma, the need for liver transplantation, and death.8, 15 The Hepatitis C Antiviral Long-Term Treatment Against Cirrhosis trial demonstrated that maintenance PEG-IFN alfa-2a therapy at a dose of 90 μg/week over 3.5 years provided no overall benefit compared with no treatment.8 Similar results were observed in the Colchicine versus PEG-Intron Long Term study, which compared PEG-IFN alfa-2b 0.5 μg/kg/week to colchicine over 3.5 years.15

In the recently completed REtreatment with PEgasys in PATients Not Responding to Peg-Intron Therapy (REPEAT) trial, nonresponders and relapsers to previous PEG-IFN alfa-2b and RBV were retreated with either a standard dose of PEG-IFN alfa-2a 180 μg/week or a higher dose of 360 μg/week for 12 weeks, after which the dose was reduced to the standard dose.16 Patients who became HCV RNA undetectable by week 24 were treated for either 48 or 72 weeks. In a protocol-defined primary analysis, SVR rates after retreatment with PEG-IFN alfa-2a and RBV were only 7% to 9% with 48 weeks of treatment but increased to 14% to 16% in those patients treated for 72 weeks. This increase in SVR resulted from a decline in relapse with the prolonged course of treatment. The use of the higher induction dose of PEG-IFN alfa-2a, 360 μg/week, did not impact SVR rates. The SVR results of this study, using 72 weeks of PEG-IFN alfa-2a are comparable with those achieved with 48 weeks of treatment with CIFN/RBV in the DIRECT trial. Several differences exist between the REPEAT and DIRECT trials that confound direct comparison of the results. It is not known what proportion of patients in the REPEAT trial were treatment-compliant, and the number of relapse patients included in the trial is not clear.17 Furthermore, the patients enrolled in the DIRECT trial had more advanced liver disease than those in REPEAT (60% versus 27% with stage F3-F4) and contained a higher percentage of African American patients (20% versus 10%).

Recent studies of CIFN and RBV have demonstrated a favorable response in the retreatment of PEG-IFN/RBV nonresponders.14, 18, 19 Two open-label trials demonstrated SVR rates ranging from 10% to 37%, with varying CIFN regimens.17, 19 In a third study, 137 consecutive patients who did not become HCV RNA–undetectable during treatment with PEG-IFN alfa-2b with RBV were switched to CIFN 15 μg/day for 12 weeks, followed by CIFN three times weekly for an additional 36 weeks with weight-based doses of RBV.14 SVR rates were noted in 37% of patients who remained on their full doses of therapy. The SVR rate was 27% in African Americans and 41% in Caucasian patients.

Several other new and promising therapies are under development for the treatment of chronic hepatitis C. These include RBV-like molecules, polymerase and protease inhibitors, and novel IFN formulations.7, 20 Two protease inhibitors, Boceprevir and Telaprevir, are the furthest along in development. Phase III studies are in the midst of enrollment, and these agents may gain U.S. Food and Drug Administration approval in 2 to 3 years. In addition, it has already been demonstrated that both protease and polymerase inhibitors will require the use of both IFN and RBV to achieve an SVR in treatment-naïve or treatment-experienced patients.

In conclusion, the current study shows the benefit CIFN holds for difficult-to-treat patients with chronic hepatitis C who have failed to respond to previous treatment with PEG-IFN and RBV. The present study demonstrated that some patients with chronic hepatitis C who have failed to respond to treatment with PEG-IFN and RBV can be successfully retreated with daily CIFN and RBV. The greatest SVR rate during retreatment in the present study was observed in F0-F3 patients who had a partial virologic response during their prior course of treatment. Therefore, once-daily CIFN in combination with RBV can be considered for select patients with chronic HCV who have failed to respond to prior treatment with PEG-IFN and RBV.


The authors would like to thank the DIRECT-001 and DIRECT-002 study investigators (Bashar Aqel, Sanjeev Arora, Carl Berg, David Bernstein, Edmund Bini, Scott Cotler, Michael Fallon, Kenneth Flora, Steven-Huy Bui Han, Joanne Imperial, Mark Jonas, Gerond Lake-Bakaar, Anne Larson, Eric Lawitz, Edward Lebovics, Michael Lyons, Gary Matusow, Douglas Meyer, Abdul Nadir, Robert Reindollar, Adrian Reuben, Maribel Rodriguez-Torres, Vinod Rustgi, Michael Ryan, John Santoro, Muhammad Sheikh, Robert Sjogren, Mark Swaim, Helen Te, Tram Tran, Andrzej Triebling, Harlan Wright, and Nizar Zein) for their participation in the studies. Dr. John Kincaid is also acknowledged for his assistance in the preparation of the first draft of the manuscript. All subsequent revisions of this manuscript were made by Bruce R. Bacon and Mitchell L. Shiffman with some input from the authors on the masthead.