Aliment Pharmacol Ther 2012; 35: 37–47
Background The efficacy of individualised antiviral treatment durations for chronic hepatitis C remains unclear.
Aim To evaluate treatment durations based on virological responses at week 4, 8 and 12 of peginterferon alfa-2a plus ribavirin therapy.
Methods Previously untreated patients with HCV genotypes, other than 2 or 3, initiated therapy with peginterferon alfa-2a 180 μg/week plus ribavirin 1000–1400 mg/day. HCV-RNA-negative patients at week 4 rapid virological response (RVR) were randomised to 24 or 48 weeks of treatment; those negative at week 8 were randomised to 36 or 48 weeks; and those who were negative or had a ≥2-log drop at week 12 were randomised to 72 or 48 weeks. Sustained virological response (SVR) was defined as undetectable HCV-RNA after 24 weeks of follow-up.
Results The study was terminated prematurely due to lagging enrolment. Of 236 patients who started treatment, 195 were randomised at week 4 (n = 50), 8 (n = 61) or 12 (n = 84). Ninety-five per cent of patients had genotype 1. SVR rates were not significantly different between patients randomised to 24 (84%) or 48 weeks (84%) at week 4, to 36 (73%) or 48 weeks (74%) at week 8, or to 48 (49%) or 72 weeks (40%) at week 12.
Conclusions In this predominantly genotype 1 cohort, shortening therapy to 24 weeks in patients with a week-4 response and 36 weeks in those with a week-8 response produced SVR rates that were similar to a 48-week regimen. Lengthening treatment to 72 weeks did not improve SVR rates. Genotype 1 patients with RVR can be treated for 24 weeks (clinicaltrials.gov NCT00483938).
The combination of peginterferon plus ribavirin is the standard of care for chronic hepatitis C, with the goal of achieving a sustained virological response (SVR).1 Patients who achieve rapid viral clearance have the highest rates of SVR. Among patients with HCV genotype 1 infection and a rapid virological response (RVR; undetectable HCV RNA at week 4), shortening treatment duration to 24 weeks may result in SVR rates that are similar to those achieved with the standard 48-week regimen.2, 3 In contrast, SVR rates are lower in patients who clear HCV RNA later during a course of treatment and few patients achieve an SVR if they do not achieve an early virological response (EVR; ≥2 log reduction in HCV RNA at week 12).1 An extended 72-week regimen has been evaluated in patients who are slow responders, with conflicting results as to the efficacy of this strategy.4–8
The randomised, multicentre, IMPROVE study was undertaken to evaluate an individualised treatment strategy based on the virological response to treatment with peginterferon alfa-2a plus ribavirin. The primary objective was to compare 72 and 48 weeks of treatment in patients who were HCV RNA positive at both treatment weeks 4 and 8 but who achieved an EVR at week 12 of treatment. Secondary objectives were to compare 24 and 48 weeks of treatment in those with an RVR at week 4, and 36 weeks compared to 48 weeks of treatment in patients who became HCV RNA negative between weeks 4 and 8 of treatment.
Patients and Methods
The IMPROVE study was an investigator-initiated, open-label, randomised, phase IIIb study conducted at 24 centres in Canada. The study was conducted in accordance with the Declaration of Helsinki and the tenets of good clinical practice. The protocol and all amendments were approved by the research ethics board at each institution and all patients provided written informed consent. The study was registered: clinicaltrials.gov NCT00483938.
Adults aged at least 18 years with chronic hepatitis C as evidenced by a positive anti-HCV antibody test and detectable HCV RNA in serum were eligible. Only patients infected with HCV genotypes other than 2 or 3 (genotype ‘non-2–3’) were included. A liver biopsy was not obligatory for study inclusion; however, patients with cirrhosis were required to have compensated liver disease defined as Child-Pugh class A disease. Patients with cirrhosis or transition to cirrhosis (F3) on a previous biopsy were required to undergo an imaging procedure within 2 months of baseline (abdominal ultrasound, CT scan or MRI scan) and to have a serum alfa-fetoprotein level <100 ng/mL to rule out hepatocellular carcinoma (HCC).
Patients were excluded if they had any of the following: infection with HCV genotypes 2 or 3; coinfection with HIV or hepatitis A or B virus; previous antiviral treatment for chronic hepatitis C including interferon-based therapy and treatment with an investigational direct acting antiviral agent; history or evidence of decompensated cirrhosis (Child-Pugh class B or C); evidence of HCC or chronic liver disease attributed to a cause other than HCV; history of a severe psychiatric disorder, uncontrolled thyroid disorder, autoimmune disease, or haemoglobinopathy; haemoglobin concentration <100 g/L in women or <120 g/L in men or patients at increased risk of anaemia or for whom anaemia would be medically problematic, an absolute neutrophil count <1.2 × 109/L, a platelet count <80 × 109/L or a serum creatinine level >1.8 times the upper limit of normal. Pregnant and breastfeeding women and male partners of women who were pregnant were not eligible for the study. Women were required to have a negative pregnancy test within 24 h of receiving the first dose of study medication. Women and male partners of fertile women were required to use of two forms of effective contraception during treatment and for 6 months after treatment. Patients were not allowed to receive systemic antiviral, antineoplastic or immunomodulatory treatment including radiation treatment or systemic corticosteroids (unless prescribed as replacement therapy).
Treatment and study design
All patients received subcutaneous peginterferon alfa-2a 180 μg/week plus oral ribavirin twice daily (Pegasys RBV, Roche, Mississauga, Canada). The total daily dose of oral ribavirin was based on body weight: 1000 mg/day for patients with a body weight <75 kg, 1200 mg/day for patients weighing ≥75 kg and <85 kg, and 1400 mg/day for patients weighing ≥85 kg. Treatment was administered in accordance with the approved Canadian product monograph and established treatment practices.
The duration of treatment was determined based on the virological response during therapy as determined by the Roche Cobas TaqMan HCV test (limit of detection 15 IU/mL) (Indianapolis, IN, USA). Patients were randomised to either a response-guided treatment duration or to a standard 48-week treatment duration (Figure 1) as follows: patients with a RVR at week 4 (HCV RNA <15 IU/mL) were randomised to 24 or 48 weeks of treatment; patients with a week-8 virological response (HCV RNA <15 IU/mL) were randomised to 36 or 48 weeks of treatment; patients with an EVR, defined as HCV RNA <15 IU/mL or, if ≥15 IU/mL, at least a 2-log decrease in HCV RNA from baseline, were randomised to 72 or 48 weeks of treatment.
Three independent sets of randomisation numbers were generated by computer, one for each randomisation point by Syreon Corporation, Vancouver. The sponsor and investigators did not have access to the randomisation sequence or assignments. Investigators were informed of treatment group assignments via a secure website (Clinstream electronic data capture system) and fax.
Nonresponders at week 12 (detectable HCV RNA and <2-log drop from baseline) stopped treatment and exited the study.
In the event of adverse events (AEs) or laboratory abnormalities, the dose of peginterferon alfa-2a and or ribavirin could be modified at the investigator’s discretion and was to reflect current standards of practice. The dose of ribavirin could be modified in 200 mg decrements. At the investigator’s discretion, the dose of either study drug could be reinstituted or increased towards the original dose after resolution of the precipitating event. Monotherapy with peginterferon alfa-2a, but not ribavirin, was allowed. Concomitant use of erythropoiesis-stimulating agents was allowed at the discretion of the investigator according to standard of care practices at the study site.
Assessments and outcomes
Serum HCV RNA levels were determined by Cobas Taqman assay (limit of detection 15 IU/mL) at baseline and treatment weeks 1, 2, 4, 8, 12, 24 and 48. Patients randomised to 72 weeks also had HCV RNA determinations at week 72. The primary efficacy end-point was SVR defined as undetectable serum HCV RNA at the end of the 24 week untreated follow-up phase (study week 48, 60, 72 or 96 depending on the assigned treatment duration (Figure 1).
Virological relapse was defined as detection of HCV RNA at the end of untreated follow-up in a patient that was HCV RNA-negative at the end of treatment. End-of-treatment virological response was defined as undetectable HCV RNA at the end of treatment.
Safety assessments included all serious AEs (SAEs), AEs of special interest, AEs leading to dose reductions or discontinuations of study medication and AE clusters. AEs of special interest included neutropenia defined as an absolute neutrophil count of <0.75 × 109/L; thrombocytopenia defined as a platelet count <50 × 109/L; anaemia defined as haemoglobin <100 g/L, and ALT elevations that led to dosage modifications. AE clusters included flu-like symptoms (one or more of fever, chills, headache, aches, pains); gastrointestinal symptoms (one or more of diarrhoea, gas, bloating), skin (one or more of injection site reaction, rash, pruritus) and emotional/mental symptoms (one or more of anger, anxiety, confusion, depression, forgetfulness, irritability).
As patients were not randomised until they had received at least 4 weeks of treatment, the intention-to-treat (ITT) population comprised all patients who were randomised to a treatment group at or after week 4 of treatment and who received at least one dose of study medication post-randomisation. Patients who were included in the ITT population and who had missing end-of-follow-up virological results were considered not to have achieved an SVR. All patients who received at least one dose of study medication were included in the safety analysis.
The primary objective was to compare SVR rates in patients with an EVR at week 12 who were randomised to 72 or 48 weeks of treatment. Secondary objectives were to compare SVR rates in patients with an RVR who were randomised to 24 or 48 weeks of treatment and in patients with a week-8 virological response who were randomised to 36 or 48 weeks of treatment. Between-group comparisons were performed with a two-sided Pearson Chi-Square test with a type I (α) error rate of 0.05. The planned enrolment of 387 patients was based on the actual virological response rates of the first 100 patients. Based on these data as well as a study that was published after our study had begun (Mangia et al.6), it was assumed that the SVR rate in patients randomised at week 12 would be 18% and 38%, respectively, in those assigned to 48 and 72 weeks of treatment. Assuming a two-sided alpha level of 0.05 and a beta level of 0.80, an enrolment of 62 patients in each of these two arms was required. In July 2008, after several months of lagging enrolment, the principal investigator and study sponsor chose to terminate enrolment short of the target. A survey of the sites revealed that enrolment was very slow because of competing clinical trials evaluating direct-acting antiviral agents and the preference of patients to wait for approval of these agents in the future. Given this situation, continuing enrolment was deemed unfeasible.
The first patient was enrolled on 9 June 2007 and the last patient visit on 7 May 2010.
A total of 236 patients were enrolled of whom 196 (83%) were randomised to one of the six treatment groups and 156 (66%) completed treatment and follow-up (Figure 2). Among the 80 patients who did not complete treatment and follow-up, 40 were withdrawn for nonresponse, 13 withdrew consent, 10 were lost to follow-up, three violated the protocol, two died and 12 withdrew for other reasons. Of the 13 patients who withdrew consent, seven had been randomised to the week-72 group.
Of 1951 patients who were included in the ITT population, 50 (26%) achieved an RVR and were randomised to 24 or 48 weeks of treatment; 62 (32%) had a week-8 response and were randomised to 36 or 48 weeks of treatment; and 84 (43%) patients achieved an EVR and were randomised to 72 or 48 weeks of treatment (Figure 2).
The baseline characteristics of the patients and treatment exposure according to treatment assignments are presented in Tables 1 and 2 respectively. Reflecting the paucity of HCV genotypes 4, 5 and 6 in the Canadian population, the vast majority (95%) were genotype 1. The mean duration of treatment was close to the planned duration in all groups with the exception of the 72-week group, in which the mean duration of treatment was 60.6 weeks. Whereas at least 84% of patients in all other treatment groups received >80% of the planned dose of peginterferon alfa-2a, only two-thirds (67%) of patients in the 72 week arm received >80% of the planned dose of peginterferon alfa-2a.
|HCV RNA-negative at week 4 (RVR)||HCV RNA-negative at week 8||HCV RNA-negative or ≥2-log drop at week 12 (EVR)||No EVR N = 40|
|24 weeks (N = 25)||48 weeks (N = 26)||36 weeks (N = 30)||48 weeks (N = 31)||72 weeks (N = 43)||48 weeks (N = 41)|
|Age, years ± s.d.||47.0 ± 10.4||47.0 ± 10.0||48.7 ± 11.5||47.8 ± 9.2||51.0 ± 7.4||50.9 ± 8.6||51.7 ± 7.5|
|Male gender, n (%)||19 (76.0)||17 (65.4)||20 (66.7)||19 (61.3)||24 (55.8)||25 (61.0)||27 (67.5)|
|Caucasian||18 (72)||21 (80.8)||24 (80)||24 (77.4)||43 (100)||35 (85.4)||34 (85)|
|Black||1 (4)||0||0||0||0||2 (4.9)||3 (7.5)|
|Asian||5 (20)||4 (15.4)||6 (20)||3 (9.7)||0||2 (4.9||0|
|Aboriginal||0||0||0||3 (9.7)||0||1 (2.4)||1 (2.5)|
|Other||1 (4)||1 (3.8)||0||1 (3.2)||0||1 (2.4)||2 (5)|
|Weight ± s.d.||77.4 ± 16.1||79.2 ± 17.7||78.4 ± 13.0||81.7 ± 18.5||81.5 ± 16.6||80.0 ± 15.7||85.5 ± 18.9|
|BMI ± s.d.||26.7 ± 4.5||27.7 ± 4.9||27.4 ± 4.3||28.5 ± 5.1||28.3 ± 5.2||28.0 ± 4.4||29.0 ± 4.7|
|HCV genotype, n (%):|
|1||20 (80)||23 (92)||29 (97)||31 (100)||43 (100)||39 (95)||38 (95)|
|4||2 (8)||2 (8)||1 (3)||2 (5)||2 (5)|
|HCV RNA level, ×106 IU/mL ± s.d.||1.72 ± 3.36||0.68 ± 1.54||3.19 ± 4.30||1.83 ± 1.40||2.97 ± 2.86||3.80 ± 5.02||2.77 ± 3.86|
|HCV RNA >400 000 IU/mL, n (%)||12 (48)||6 (24)||23 (76.7)||28 (90.3)||37 (86.0)||35 (85.4)||33 (84.6)|
|Pre-treatment liver biopsy performed, n (%)||17 (68.0)||17 (68.0)||14 (46.7)||11 (35.5)||21 (48.8)||27 (65.9)||19 (47.5)|
|Histological diagnosis of cirrhosis, n (%)*||2 (11.8)||1 (5.9)||0||0||1 (4.8)||5 (18.5)||2 (10.5)|
|Assigned ribavirin dose, n (%)|
|<1000 mg/day||3 (12.0)||2 (7.7)||4 (13.3)||4 (12.9)||9 (20.9)||9 (22.0)||5 (12.5)|
|1000 mg/day||11 (44.0)||9 (34.6)||12 (40.0)||8 (25.8)||14 (32.6)||12 (29.3)||10 (25.0)|
|1200 mg/day||5 (20.0)||7 (26.9)||8 (26.7)||7 (22.6)||6 (14.0)||8 (19.5)||6 (15.0)|
|1400 mg/day||6 (24.0)||8 (30.8)||6 (20.0)||12 (38.7)||14 (32.6)||12 (29.3)||19 (47.5)|
|HCV RNA-negative at week 4 (RVR)||HCV RNA–negative at week 8||HCV RNA-negative or ≥2-log drop at week 12 (EVR)|
|24 weeks (N = 25)||48 weeks (N = 25)||36 weeks (N = 30)||48 weeks (N = 31)||72 weeks (N = 43)||48 weeks (N = 41)|
|Treatment with peginterferon alfa-2a|
|Mean duration, weeks||24.2 ± 0.5||43.7 ± 10.2||35.2 ± 2.3||47.6 ± 1.3||60.6 ± 17.1||46.4 ± 5.6|
|Median duration (interquartile range), weeks||24 (24–24)||48 (48–48)||36 (32–36)||48 (48–48)||72 (52–72)||48 (48–48)|
|Received >80% of the planned dose of peginterferon alfa-2a, n (%)||25 (100)||21 (84)||30 (100)||31 (100)||29 (67.4)||38 (92.7)|
|Received >80% of the planned dose of ribavirin, n (%)||25 (100)||20 (80)||29 (96.7%)||31 (100)||27 (62.8)||37 (90.2)|
There was no statistically significant difference between the SVR rates in patients with an EVR who were randomised to 72 or 48 weeks of treatment (40% vs. 49%, respectively, P = 0.51, Figure 3a). The SVR rates were identical (84%) in patients with an RVR who were randomised at week 4–24 or 48 weeks of treatment, and there was no statistically significant difference between the SVR rates in patients with a week 8 response who were randomised to 36 or 48 weeks of treatment (73% vs. 74%, respectively, P = 1.000, Figure 3a).
SVR rates among patients randomised to an individualised treatment regimen (61%) were similar to those in patients randomised to the standard 48-week treatment regimen (66%; P = 0.55). The overall SVR rate among all patients who received at least one dose of study medication was 53% (124/236). Relapse rates were not statistically different between any of the response-guided and standard treatment groups, nor the overall response-guided vs. control group (Figure 3b).
Among 50 patients with an RVR, 13, 8 and 29 individuals were HCV RNA negative by week 1, 2 and 4 respectively. Only 1 of 20 (5.0%) patients who became HCV RNA negative by week one or two and who completed treatment relapsed; this patient was randomised to 48 weeks of treatment. In contrast, 5 of 28 (17.9%) individuals who became HCV RNA negative between weeks 2 and 4 of treatment and who completed treatment relapsed.
Week-12 EVR is now subclassified as complete EVR (HCV RNA undetectable at week 12) or partial EVR (detectable HCV RNA but at least a 2-log decrease from baseline). Half the 84 patients randomised at week-12 showed each type of EVR. Among patients randomised to 72 or 48 weeks of treatment, 23 (54%) and 19 (46%) patients, respectively, had a complete EVR. Of the other 42 patients, 20 (47%) and 22 (54%) patients had a partial EVR). Of patients randomised to 72 and 48 weeks, respectively, SVR rates were 52% (12/23) and 68% (13/19) in those with a complete EVR, and 32% (7/22) and 25% (5/20) in those with a partial EVR.
A total of 77 patients were assigned to an initial ribavirin dose of 1400 mg/day of whom 58 were randomised to one of the six treatment groups. Overall, 66% (38/58) of these randomised patients achieved an SVR including 29% (4/14) and 58% (7/12) assigned to 72 and 48 weeks of treatment at week 12, 67% (4/6) and 83% (10/12) assigned to 36 and 48 weeks of treatment at week 8, and 100% (6/6) and 88% (7/8) assigned to 24 and 48 weeks of treatment at week 4 respectively.
At least one dose of peginterferon alfa-2a or ribavirin was reduced or withheld in 13% and 17% of patients, respectively, because of AEs or laboratory abnormalities (Table 3). The highest incidence of dose reductions or withdrawal for AEs or laboratory abnormalities was in the 72-week treatment group. Twenty-six patients experienced a serious AE and 80 (34%) patients experienced an AE of special interest. Across the six randomised groups, the incidence of serious AEs ranged from 10% to 17% and the incidence of AEs of special interest ranged from 26% to 46% (Table 3), with no apparent relationship with treatment duration. The incidence of the AE clusters did not appear to be related to treatment duration.
|HCV RNA-negative at week 4 (RVR)||HCV RNA–negative at week 8||HCV RNA-negative or ≥2-log drop at week 12 (EVR)||No EVR (N = 40)||All patients (N = 236)|
|24 weeks (N = 25)||48 weeks (N = 26)||36 weeks (N = 30)||48 weeks (N = 31)||72 weeks (N = 43)||48 weeks (N = 41)|
|Deaths||0||0||0||0||2 (4.6%)||0||0||2 (0.8)|
|Patients with at least one SAE, n (%)||0||4 (15)||5 (17)||4 (13)||6 (14)||4 (10)||3 (7)||26 (11)|
|Patients with SAEs leading to dose reductions or drug discontinuations||0||0||0||0||2 (5)||1 (2)||1 (2)||4 (2)|
|Dose reductions or discontinuations for AEs or laboratory abnormalities, n (%)|
|Peginterferon alfa-2a||3 (12)||4 (15)||1 (3)||3 (10)||10 (23)||4 (10)||5 (12)||30 (13)|
|Ribavirin||1 (4)||7 (27)||4 (13)||5 (16)||12 (28)||7 (17)||5 (12)||41 (17)|
|Patients with at least one AE of special interest, n (%)||9 (36)||12 (46)||12 (40)||8 (26)||18 (42)||16 (39)||1 (2)||80 (34)|
|Haemoglobin <100 g/L||4 (16)||4 (15)||5 (17)||7 (23)||16 (37)||12 (29)||1 (2)||49 (21)|
|Neutrophils <0.75 × 109/L||6 (24)||9 (35)||8 (27)||3 (10)||9 (21)||9 (22)||1 (2)||45 (19)|
|Platelets <50 × 109/L||1 (4)||1 (4)||2 (7)||1 (3)||3 (7)||2 (5)||2 (5)||12 (5)|
|ALT elevations leading to dose modifications||0||0||0||0||1 (2)||0||1 (2)||2 (1)|
|Patients with at least one event included in an AE cluster, n (%)|
|Emotional/mental*||11 (44)||15 (58)||15 (50)||17 (55)||25 (58)||21 (51)||14 (35)||118 (50)|
|Flu-like symptoms†||16 (64)||21 (81)||23 (77)||24 (77)||32 (74)||37 (90)||28 (70)||181 (77)|
|Gastrointestinal‡||1 (4)||6 (23)||12 (40)||11 (35)||9 (21)||14 (34)||4 (10)||57 (24)|
|Skin§||12 (48)||18 (69)||20 (67)||16 (52)||24 (56)||24 (58)||16 (40)||130 (55)|
The lowest rate of anaemia was observed among patients with an RVR and the highest rate in those with an EVR. Among patients randomised to 48 weeks of treatment, the incidence of anaemia increased with the time of randomisation and was 15%, 23% and 29% among those randomised at week 4, 8 and 12, respectively. In contrast, the highest rate of neutropenia was in patients with an RVR (29% vs. 18% and 21% of those randomised at week 8 and 12).
Two patients randomised to 72 weeks of treatment died. One death, attributed to intracranial haemorrhage, occurred approximately 2 months after the last dose of medication. The other death was attributed to pulmonary sepsis and occurred 4 months after the last dose of study medication. Neither death was considered to be related to study drug treatment in the opinion of the investigator.
The primary comparison of SVR showed no significant difference between 72 and 48 weeks of treatment in non-genotype 2 or 3 patients who achieved an EVR at week 12 after having detectable HCV RNA in serum at both treatments weeks 4 and 8. The trends in both the SVR rates and the relapse rates in these two groups (higher SVR rates and lower relapse rates in patients randomised to 48 weeks of treatment) were contrary to what was expected. Previous studies that have compared 48 and 72 weeks of treatment in patients without an RVR have reported higher SVR rates and lower relapse rates (not all statistically significant) in patients randomised to the longer treatment duration.5–8 It remains unclear why our data conflict with other reports; however, we believe that this may be due to the relatively large number of patients who withdrew prematurely or were lost to follow-up, both of which were disproportionately higher in the longer treatment group. According to the intention-to-treat principle, these patients were counted as treatment failures. Consistent with our observations, all other studies of extended treatment have reported higher withdrawal rates in patients randomised to 72 weeks of treatment.4–8
Our trial also showed that SVR rates were similar among patients with an RVR who were randomised to 24 or 48 weeks of treatment and those with a week-8 virological response randomised to 36 or 48 weeks. To our knowledge, no prior study has randomised patients with a week-8 response to a shortened duration of treatment. Overall SVR rates in patients with a week-8 virological response were very good (approximately 74% in either 36 or 48 week arm) and considerably higher than those achieved in patients without a week-8 response who were randomised at week 12 (<50%). Other studies that have assigned HCV genotype 1 patients to treatment on the basis of the week-8 response have also reported high SVR rates. For example, Mangia et al., reported SVR rates of 70% and 72% in patients with a week-8 virological response assigned to standard and individualised treatment durations.6 Unfortunately, all patients with a week-8 response were assigned to 48 weeks of treatment in that trial6; therefore, their results do not shed any light on the feasibility of abbreviated therapy in week-8 responders. Although the relatively small sample size of this subgroup in our study raises the possibility of a type II error, we are encouraged by the close concordance between SVR rates in our week-8 cohort randomised to 48 weeks and the much larger cohort in the study by Mangia et al.6
The overall SVR rates were numerically similar in patients randomised to individualised treatment durations and those randomised to 48 weeks of treatment, in agreement with other studies of response-guided treatment.6, 9 Thus, if abbreviated 24 and 36 week regimens were incorporated into clinical practice, the individualised treatment strategy would be cost-saving.
There is increasing interest in very early viral kinetics, even before week 4 of treatment, as indicators of therapeutic success. Our study agrees with other studies showing that the faster HCV RNA becomes undetectable during the first month of treatment, the greater the likelihood of achieving an SVR.10, 11 The week 1 and 2 data showed much lower relapse rates amongst those patients who first became virus-negative by week 1 or 2 compared with those who needed the full 4 weeks to achieve undetectable HCV RNA.
Finally, what is the relevance of our findings at the onset of the direct-acting antiviral (DAA) era? The approval of the first class of DAAs (protease inhibitors) has occurred or is imminent in most global regions. Nevertheless, despite these advances, the combination of peginterferon plus ribavirin will continue to play an important role in our therapeutic arsenal because protease inhibitors must be combined with these agents to prevent resistance. Moreover, triple combination regimens have a higher side effect burden than dual combination therapy.12–14 Therefore, some patients who cannot tolerate protease inhibitors will need to be treated with dual therapy. Finally, the vast majority of the world’s 180 million HCV-infected people live in economically disadvantaged countries such as Egypt, Pakistan and Mongolia.15 In these regions, the high cost of the DAAs will probably preclude use of triple-combination therapies except in interferon-ribavirin treatment failures. Therefore, the results of our study may be relevant in many global regions of high HCV endemicity where the current dual-combination therapy will almost certainly remain the standard of care for many years to come.
The high cost of the first DAAs has been justified by their manufacturers in part because they allow shorter total duration of treatment. However, in their cost projections, they continue to insist that the dual-therapy regimens should be a standard 48 weeks. Our results showing similar efficacy of the 24-week regimen, along with other previous studies 2, 3, 6, 9 clearly indicate that genotype 1 patients who achieve RVR can shorten treatment duration to 24 weeks without sacrificing efficacy. In this regard, although the percentage of genotype 1 patients who achieve RVR with dual therapy is relatively low in predominantly American cohorts, approximately 8–11%,12–14 21% of our Canadian patients achieved RVR, and 27% achieved RVR in the Italian multicenter study of Mangia et al.6 In a global database of 1383 genotype 1 patients treated in the registration trials of peginterferon alfa-2a + ribavirin, 16% achieved RVR 16 and in Taiwan, 44% achieved RVR.9 Thus, the cost/efficacy analyses of expensive DAA triple-therapy must account for these significant regional variations in dual-therapy RVR and use the shortened 24-week duration in such patients for cost calculations. Indeed, in some regions with relatively high RVR rates, it may make sense to start treatment with a 4-week lead-in of only dual-therapy and only add the expensive DAA in those who fail to achieve RVR. Those who achieve RVR in the lead-in period can be treated for 24 weeks with only dual therapy, thus significantly saving costs and reducing the risk of the additive AEs with triple-therapy.
Our study has several limitations. Importantly, the number of participants included in the trial was lower than the target enrolment, which limits the statistical power of the study. Moreover, as the importance of IL-28B genotype as a determinant of treatment response has only recently been described, we lack information on this important predictor of SVR.17, 18 As an imbalance in the distribution of this genetic polymorphism between treatment groups can significantly impact treatment responses, particularly in small randomised trials such as ours, future studies should stratify patients according to this host factor.
In conclusion, the major new finding of this randomised trial in non-genotype 2 or 3 patients is that shortening therapy to 36 weeks in those with a week-8 virological response resulted in SVR rates that are similar to the standard 48-week regimen. Week 8 may be a useful intermediate time ‘checkpoint’ for additional virological testing to better individualise response-guided treatment. Moreover, this study confirms that genotype 1 patients who achieve RVR can be treated for 24 weeks – a paradigm that remains relevant and applicable even in the DAA era.
One patient randomised at week 4 to 24 weeks of treatment was subsequently excluded from the ITT analysis because of a violation in entry criteria.
IMPROVE study investigators: Nav Anand, Robert Bailey, Vince Bain, Brian Chai, Curtis Cooper, William Depew, Magdy Elkashab, Siegried Erb, Victor Feinman, Peter Ghali, Wayne Ghesquiere, Susan Greenbloom, Jenny Heathcote, Nir Hilzenrat, Samuel S. Lee, Paul Marotta, Roger Mousseau, Kevork Peltekian, Henryk Pluta, Jeff Powis, Alnoor Ramji, Morris Sherman, Helga Witt-Sullivan and Lawrence Worobetz. Declaration of personal interests: SS Lee declares: consulting for Bristol Myers Squibb (BMS), Gilead, Janssen, Merck, Novartis, Pharmasset, Roche, Tibotec. Research support: BMS, Gilead, Merck, Roche, Novartis, Pharmasset, Roche, Tibotec, Vertex. Speaker for: BMS, Gilead, Merck, Roche; M Sherman declares: consulting for Bayer, Celsion, BMS, Gilead, Merck, Roche, Tibotec, Vertex. Speaker for: Gilead, Roche, BMS; A Ramji declares: consulting for Astra Zeneca, Roche, Merck, Gilead, Vertex. Research support: Roche, Merck, Gilead; M Elkashab declares: consulting and speaking for Roche; R Balshaw is Director, Biometrics of Syreon Corporation, the contract research organisation hired by the sponsor to conduct the trial and perform the analyses; C Usaty is an employee of Roche Canada, Mississauga; RP Myers declares: consulting and speaking for Roche and Merck; S Greenbloom, H Pluka and N Hilzenrat: no personal interests to disclose in relation to this manuscript. Declaration of funding interests: The study was funded by Roche Canada. The statistical analysis was performed by Dr. Robert Balshaw, with the assistance of Nadia Lesnikova of Syreon Corporation. Writing assistance was provided by Blair Jarvis, funded by Roche. Dr Lee made all final decisions regarding the manuscript. Roche Canada did not have veto power over the content of the manuscript or the decision to publish.