Journal of Hepatology 44 (2006)

Transcription

1 Journal of Hepatology 44 (2006) Treatment with daily consensus interferon (CIFN) plus ribavirin in non-responder patients with chronic hepatitis C: A randomized open-label pilot study * Markus Cornberg 1, Johannes Hadem 1, Eva Herrmann 2, Frank Schuppert 3, Hartmut H.-J. Schmidt 4,5, Markus Reiser 6, Oliver Marschal 7, Martin Steffen 8, Michael P. Manns 1, *, Heiner Wedemeyer 1 1 Abt. Gastroenterologie, Hepatologie und Endokrinologie, Medizinische Hochschule Hannover, Hannover, Germany 2 Klinik für Innere Medizin II, Universität des Saarlandes, Homburg, Saar, Germany 3 Medizinische Klinik II, Krankenhaus Bad Oeynhausen, Bad Oeynhausen, Germany 4 Medizinische Klinik m. S. Hepatologie und Gastroenterologie, Charité-Campus Mitte, Berlin, Germany 5 Transplantationshepatologie, Universitätsklinikum Münster, Münster, Germany 6 Ruhr-Universität Bochum, Abt. Gastroenterologie und Hepatologie, Berufsgenossenschaftliche Kliniken Bergmannsheil, Bochum, Germany 7 Praxis Braunschweig, Braunschweig, Germany 8 Medizinischen Klinik I, Franziskus Hospital Bielefeld, Bielefeld, Germany Background/Aims: Therapeutic options for hepatitis C non-responder patients are limited. Methods: We initiated an open-label pilot study to investigate the efficacy of CIFN plus ribavirin on viral kinetics, sustained virological response (SVR), and histological response in hepatitis C non-responder patients. Seventy-seven patients were enrolled to receive CIFN given daily in combination with 1000/1200 mg ribavirin. An 8-week inductiondosing regimen of 18 mg CIFN, followed by 9 mg for 40 weeks was compared to 9 mg CIFN for 48 weeks. 90% of patients were infected with HCV-genotype 1. Results: Overall, 82% of the patients demonstrated an early virological response, 65% had an end-of-treatment response, and the SVR was 30%. Interferon/ribavirin non-responders demonstrated a SVR of 22%. Induction-dosing resulted in a greater first-phase HCV-RNA decay that, however, did not translate to better SVRs, presumably due to more dose modifications. High ALT, younger age, and second-phase viral kinetics were associated with SVR. Only sustained responders and relapse patients showed an improved liver histology. Conclusions: Daily dosing of CIFN plus ribavirin may be a promising concept for selected non-responder patients before considering therapies which are anti-viral but not curative. However, motivation and compliance are requisites and a CIFN induction is not required. q 2005 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Hepatitis C virus; Chronic hepatitis C; Consensus interferon; Ribavirin; Non-responder; Induction-dosing; Daily dosing Received 27 July 2005; received in revised form 6 October 2005; accepted 24 October 2005; available online 28 November 2005 * The authors have declared that they received funding from the drug companies involved to carry out their research. * Corresponding author. Address: Department of Gastroenterology, Hepatology, and Endocrinology, Medizinische Hochschule Hannover, Carl-Neuberg-Str. 1, D Hannover, Germany. Tel.: C ; fax: C address: manns.michael@mh-hannover.de (M.P. Manns) /$30.00 q 2005 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi: /j.jhep

2 Introduction M. Cornberg et al. / Journal of Hepatology 44 (2006) Methods Despite the enormous advances that have been achieved in the treatment of chronic hepatitis C over the last decade [1], there is still a need for improved therapies, especially for the difficult-to-treat patients such as HCV-genotype 1 infected individuals, patients with liver cirrhosis, or patients who did not respond to a previous interferon alfa (IFN)- based therapy [2]. Even the new standard therapy of pegylated interferon alfa (PEG-IFN) in combination with ribavirin is not very effective for the so called non-responder patients. Relapse patients may benefit from re-treatment but patients infected with HCV-genotype 1 who were true nonresponders to IFN and ribavirin demonstrated only 12% sustained virological response (SVR) with PEG-IFN and ribavirin [3,4]. Thus, there is currently no generally accepted therapeutic strategy for this patient population. One approach, currently evaluated in clinical trials, is the long-term maintenance therapy with PEG-IFN in nonresponders with advanced fibrosis or cirrhosis to prevent hepatocellular carcinoma and/or decompensation (EPIC3, HALT-C, COPILOT trials [3 5]). However, viral eradication should be still the first achievable goal whenever possible. Some study results suggest that consensus interferon (CIFN), which is a consensus molecule of the type-1 interferons with a higher biological activity in vitro [6,7], may be more effective than standard IFN for the difficult-to-treat HCV-genotype 1 patients [8 10]. CIFN monotherapy demonstrated a substantial SVR in nonresponder patients [11]. To mimic the half-life of PEG- IFN, CIFN has to be administered on a daily basis. Recently, one study showed that daily dosing of 9 mg CIFN significantly increased the SVR compared to a 9 mg tiw regimen [12]. Preliminary data from a single center study presented in abstract form, suggested that daily dosing of CIFN in combination with ribavirin can achieve SVRs of 38 45% in non-responder patients to standard IFN and ribavirin depending on the CIFN dose [13]. High-doseinduction, although not effective in studies with standard IFN [14 16], seemed to have further improved the SVR in this study [13]. Cotler and colleagues [17] assessed the firstphase viral kinetics in 20 previous non-responders after a single dose of 15 or 30 mg CIFN and demonstrated a significantly sharper decline of the HCV-RNA with the higher dose after 24 h (0.8 and 1.5, respectively). Whether high-dose-induction with CIFN may serve as an option to further increase the SVR in difficult-to-treat patients remains unknown. Here, we initiated an open-label pilot study to assess the efficacy of daily dosing of CIFN in combination with ribavirin on the first- and second-phase of viral kinetics, sustained virological response, and histological response in patients with chronic hepatitis C who did not respond to a prior IFN based therapy. Additionally, we tested if a shortterm induction with a double dose of CIFN would further enhance the SVR Selection of patients Adult patients (18 65 years) with chronic hepatitis C infection were eligible for the study if they were virological non-responders to a previous therapy with IFN or IFN plus ribavirin (HCV-RNA positive after at least 3 MU IFN alfa tiw for 24 weeks). Non-response was further classified as flat non-response (!1 log 10 reduction of HCV-RNA), flat partial response (O 1 log 10!2 log 10 reduction of HCV-RNA), and partial response with a significant reduction of viral load (O2 log 10 reduction of HCV-RNA). However, in a number of patients quantitative HCV-RNA levels of the prior therapy were not available, as this was not standard during former therapies. All patients were seropositive for HCV-RNA by testing with polymerase chain reaction by Roche Amplicor w and had elevated serum alanine aminotransferase activity (ALT). Patients were excluded if they had decompensated liver disease, liver diseases unrelated to HCV infection, anemia (hemoglobin concentration less than 12 g/dl for women and less than 13 g/dl for men), leukocytopenia (less than 3000/ml), thrombocytopenia (less than 100,000/ml), decompensated renal disease (serum creatinine above 130 mmol/l), decompensated thyroid disease, HIV or hepatitis B infection, psychiatric conditions, history of seizures, poorly controlled autoimmune diseases, previous organ transplantation. Ongoing intravenous drug (IVDA) abuse and lack of abstinence for at least 12 months were also exclusion criteria. There were two protocol deviations: two patients were included despite platelets lower than 100,000/ml and one of them was also 70 years old. Both patients did not respond to therapy (one breakthrough, one non-response). 3. Study design The study was an open-label trial with a central randomization procedure performed in Hannover. Seventy-seven patients were randomly assigned into two groups without further stratification using sequentially numbered cards in sealed envelopes. Patients in Group A were treated with 9 mg CIFN daily for 48 weeks whereas patients in Group B received an induction therapy of 18 mg CIFN daily for the first 8 weeks followed by 9 mg CIFN daily for the remaining 40 weeks. In both groups, ribavirin was given at the standard dose of 1 g (!75 kg) or 1.2 g daily (O75 kg). The patients were evaluated as outpatients at the following visits: screening visit, entry visit, day 1, day 3, week 1, 2, 4, 12, 18, 24, 30, 36, 42, 48, 52, 60, and 72. The laboratory of each center performed biochemical and hematological testing. Serum HCV-RNA levels (IU/ml) were determined with the Cobas Amplicor Hepatitis C Monitor Test (v2.0, Roche Diagnostics). Viral genotypes were determined with the INNO-LiPA HCV II Kit (Innogenetics, Gent, Belgium). This protocol was approved by the local ethics committees, and was conducted according to principles of the Helsinki Declaration. Written informed consent was obtained from all participants prior to enrollment. 4. Assessment of efficacy The primary endpoint was sustained virological response (SVR), defined as undetectable HCV-RNA in serum 24 weeks after the end-of-treatment (EOT). Secondary

3 M. Cornberg et al. / Journal of Hepatology 44 (2006) endpoints were early virological response (EVR: HCV- RNA declineo2 log 10 before week 12) and absence of serum HCV-RNA at the EOT and normalization of serum ALT at the end of follow-up. Therapy was discontinued once the HCV-RNA was detectable at week 24 of treatment. The first-phase of early viral kinetics was assessed by the log 10 decay during the first 24 h as by Cotler et al. [17].The second-phase slope of the initial viral kinetics was calculated by log-linear regression on measurements of HCV-RNA at 1, 2, and 4 weeks and rapid virological response (RVR) was defined as a second-phase slope faster than K0.10/day as previously [18]. In 37 patients, a second liver biopsy was performed at the end of the follow-up (24 weeks after end of therapy) and the histology (Ishak score [19]) was compared to the histology before the initiation of therapy. 5. Statistical analysis Data were described by rates, means with standard deviation, medians, and ranges. Furthermore, groups were compared by c 2 test, Fishers exact test, Man Whitney and Kruskal Wallis test as appropriate. In addition, multivariate step-wise logistic regression was used to identify independent predictors from baseline characteristics which were associated with SVR in univariate analysis. All p values reported were two-sided and p values below 5% were considered significant. 6. Results 6.1. Patients Between March 2000 and September 2001, 77 patients were randomized (Group AZ38, Group BZ39 and treated at eight different sites (Hannover, Bad Oeynhausen, Berlin, Bochum, Braunschweig, Bielefeld, Kassel, Koblenz). The baseline characteristics are shown in Table 1. The only significant difference between the treatment groups was found in baseline ALT as Group B tended to have lower ALT levels (pz0.048) Safety The spectrum of side effects of daily CIFN therapy was comparable to previous trials with IFN and ribavirin combination therapy. However, the high-dose-inductionphase had a high side effect profile. There were four serious adverse events (suspected drug related) during therapy (Table 2). A case of menigococcus sepsis occurred during the last 2 weeks of therapy in Group A. Despite discontinuation of therapy, the patient remained HCV-RNA negative. Table 1 Baseline characteristics of the 77 patients Group A (nz38) Group B (nz39) Male patients 27 (71%) 31 (79%) Mean age [yearsgsd] 45.2G G10.5 Mean body weight [kggsd] 80.7G G12.9 HCV-genotype distribution HCV-Genotype 1a/b 34 (89%) 35 (90%) HCV-Genotype 3a 2 (5%) 2 (5%) HCV-Genotype 4 or 5 2 (5%) 2 (5%) HCV-virus loado850!10 3 unit/ml 20 (53%) 24 (61%) Median ALT(Min, Max) [units/ml] 64 (20,198) 40 (18,222) (pz0.048) Mode of transmission Blood products/transfusion 19 (50%) 21 (54%) I.V. drug abuse 2 (5%) 3 (8%) Surgery, needle stick, tattoo 8 (21%) 7 (18%) Unknown 9 (24%) 8 (21%) Liver histology before therapy No/mild fibrosis (Ishak scora F0-2) 15 (41%) 20 (51%) Liver fibrosis (Ishak score F3-4) 13 (34%) 11 (28%) Liver cirrhosis (Ishak score F5-6) 8 (21%) 8 (21%) Not available 2 (5%) 0 (0%) Previous therapy IFN monotherapy 17 (45%) 19 (49%) IFNCribravin combination therapy 21 (55%) 20 (51%) Response to prior therapy Flat non-response(!1 log red. of HCV-RNA) 6 (16%) 9 (23%) Flat partial response(o1 log!2 red. of HCV-RNA) 5 (13%) 3 (8%) Significant HCV-RNA reduction(o2 log red. of HCV-RNA) 6 (16%) 7 (18%) Only qualitative HCV-RNA available 21 (55%) 20 (51%)

4 294 M. Cornberg et al. / Journal of Hepatology 44 (2006) Table 2 Laboratory abnormalities and adverse events during the therapy Laboratory abnormalities and adverse events Group A (nz38) Group B (nz39) Leucopenia (!2000 leucocytes) 6 (16%) 8 (21%) Thrombopenia (!50,000 platelets) 4 (11%) 3 (8%) Anemia (Hb!10 g/dl) 9 (24%) 3 (8%) Flu-like symptoms (myalgia, arthralgia, fever) 25 (66%) 33 (85%) Gastrointestinal disorder (nausea) 10 (26%) 17 (44%) Dyspnoe, vertigo 19 (50%) 12 (31%) Skin (dry skin, hair loss, itching, exacerbation of psoriasis [nz3, all in B]) 12 (32%) 15 (38%) Psychiatric disorders (depression, anger, anxiety) 10 (26%) 7 (18%) Serious adverse events (suspected drug related) 1 (3%) 3 (10%) Dose modification/discontinuation CIFN before week 8 due to laboratory 5 (13%) 19 (49%) abnormalities or adverse events Dose modification/discontinuation CIFN during the complete study period due to 10 (26%) 19 (49%) laboratory abnormalities or adverse events Dose modification/discontinuation Ribavirin during the complete study period due to laboratory abnormalities or adverse events a 15 (39%) 7 (18%) The noted adverse events were moderate to severe (subjective reports from patients) and occurred at least once during the treatment period. Serious adverse events: (A) meningococcus sepsis; (B) severe weight loss and poor physical condition resulted in drop out, two patients with injection site abscesses resulting in surgical procedures. a Ribavirin dose reductions did not significantly affect the relapse rate in this study. All other patients with serious adverse events discontinued treatment and relapsed or remained HCV-RNA positive. There was a trend towards more severe flu-like symptoms and nausea in the induction group, which resulted in more frequent dose modifications early during treatment. Reduction of leucocytes and platelets were significantly higher in the induction group during the first 8 weeks of therapy (leucocytes TW1, TW2, TW4; platelets TW1, Fig. 1). Due to laboratory abnormalities and the side effects, almost every other patient required a reduction of the 18 mg CIFN dose before week 8, while this was the case in only 13% of patients receiving 9 mg CIFN during the first 8 weeks (pz0.001, Table 2). Patients in Group A developed anemia and associated side effects (dyspnoe, vertigo) more frequently than patients in Group B, resulting in ribavirin dose modifications in twice as many patients in Group A than B (pz0.04). However, the overall percent reduction of the hemoglobin was not different between both groups. The reason for this finding may have been the lower hemoglobin concentration in Group A at TW0, which was almost significant. No signs of diminished liver function (as measured clinically and by coagulation function and serum albumin) were noted during therapy. We had no evidence for hypocalcemia as previously reported for daily dosing of CIFN [20] Efficacy Overall, 59 patients (77%) completed more than 42 weeks of therapy (O80% of treatment duration), whereas 4 patients dropped out before TW24. In 12 patients, therapy was stopped at TW24 due to non-response, and 2 patients discontinued the study between TW24 and TW42 (Fig. 2). Leucocytes Platelets Hemoglobin % of TW0 value 100% 80% 60% 40% * * * Group A Group B % of TW0 value 100% 80% 60% 40% * Group A Group B % of TW0 value 100% 80% 60% 40% Group A Group B 20% 20% 20% Treatment weeks Treatment weeks Treatment weeks * p<0.05, values from at least patients for each time-point. Fig. 1. Laboratory changes of leucocytes, platelets and hemoglobin during the first 24 weeks of therapy. Percent of the value at TW0 (Z100%) is plotted (*p!0.05).

5 M. Cornberg et al. / Journal of Hepatology 44 (2006) patients (65%) had an EOT response. There was no difference between the two treatment groups (Fig. 3A). Unfortunately, 26 of the EOT patients (52%) had a virological relapse during the follow-up period. Twentythree patients (30%) had a SVR and one patient was lost to follow-up (Fig. 2). Again, there was no difference between both treatment regimens (Fig. 3A). Comparing the HCV-RNA levels during therapy, we observed a more pronounced decline in HCV-RNA in patients treated with 18 mg CIFN during the first 8 weeks of therapy. Especially the median log 10 drop of HCV- RNA after 24 and 72 h was 0.29 and 0.57 in Group A, and 0.30 and 1.03 in Group B (pz0.16 and 0.002, respectively). The stronger first-phase decline in HCV- RNA resulted indeed in a higher virological response at TW8, as 56% of patients treated with the high dose were already HCV-RNA negative compared to only 42% in Group A (Fig. 4A). However, the number of patients with an EVR defined aso2 log 10 reduction at TW12 was not different between both groups (Fig. 3A). The RVR as defined by Layden [18], that is a secondphase slope during TW1 and TW4 faster than K0.1/day, was not statistically different between the treatment groups (28% in Group A, 38% in Group B, po0.2). N=77 Discontinuation before week 8 N=3 (all group B) 1 patient had > 2 log decline n= 63 EVR (> 2 log decline before week 12) 3 non-responder n= 10 no EVR (< 2 log decline before week 12) 8 non-responder 1 stopped N=1 HCV-RNA n.a. 1 non-responder N=60 HCV-RNA negative during therapy N=1 HCV-RNA negative during therapy = 1 relapse) N=2 Discontinuation before week 40 (<80% of therapy) N=50 End of treatment Response (65%) 1 patient stopped TW42 (SAE) N=9 Breakthrough N=1 Lost during follow-up N=23 Sustained Virological Response (30%) N=26 Relapse Fig. 2. Response to therapy of all patients.

6 296 M. Cornberg et al. / Journal of Hepatology 44 (2006) Fig. 3. (A) Rapid virological response (RVR: second-phase slope faster than K0.1/day), early virological response (EVR:O2 log 10 decline of HCV- RNA before week 12), end of treatment response, sustained virological response, sustained virological response in patients who were treated according to the protocol, and sustained virological response rates of patients without CIFN modifications in the first 8 weeks are shown. (B) Sutained virological response rates of patients with different baseline characteristics are shown. Non-responders to IFN monotherapy had the best sustained response and patients with liver cirrhosis had the poorest sustained response rates Factors influencing sustained virological response It is known that patients who are adherent to therapy respond much better to therapy than those who are not compliant [21]. Here, patients who were treated according to the protocol and did not drop out showed a SVR of 33% (Fig. 3A). The SVR was even higher in the group of patients that did not require CIFN dose modifications during the induction phase. When comparing baseline characteristics in patients with and without SVR, only high baseline ALT and low age were significantly associated in univariate analysis (Table 3) and both variables remained significant in multivariate logistic regression (pz0.001, 0.046, respectively). Interestingly, SVR was significantly but not exclusively associated with RVR as 13 of 23 (57%) patients with RVR but also still 10 of 47 patients (21%) without RVR achieved a SVR (Table 3). Nevertheless, patients who were non-responder to an IFN monotherapy showed a higher chance to achieve a SVR (39%) than IFN/ribavirin non-responder (22%, pz0.1, Fig. 3B). This trend was also seen for patients with a faster HCV-RNA decay during the previous therapy as the median log 10 decay after 8 weeks of treatment was 3.38 in patients with a prior HCV-RNA decayo1 log 10 compared to 2.35 in patients with a flat response during the previous therapy (pz0.07, Fig. 4B). For six patients we were able to compare the HCV-RNA kinetics from the prior therapy to the current CIFN

7 M. Cornberg et al. / Journal of Hepatology 44 (2006) Fig. 4. Log 10 decay of HCV-RNA during and after treatment. Panel A shows all patients differentiated by the treatment groups, Panel B shows all patients differentiated by virological response of previous therapy. treatment (Fig. 5). All six patients were previously treated in well-documented studies and were compliant according to the patient records. Interestingly, different patterns of HCV- RNA kinetics during earlier and current treatment were observed. Three out of four patients who were treated with 3 MU IFN alfa-2b tiw during the prior therapy had a rapid viral decline of HCV-RNA and two of these had received induction dosing of IFN alfa-2b during their prior treatment. Both were able to clear HCV-RNA during treatment with CIFN. However, this was not true for patient 67 who seemed to be resistant also to CIFN therapy (Fig. 5) Histological response All patients underwent liver biopsy before the start of therapy. Thirty-seven patients had a second liver biopsy at the end of the follow-up. Patients with SVR had the best histological response, and relapse patients had a better response than did non-responders or breakthrough patients (Table 4). Only three of the non-responder patients had an improvement in the grading but none of them improved in the fibrosis score. In contrast, 60% of the patients with a SVR improved their fibrosis stage. 7. Discussion Currently, there are no recommendations how to manage patients with chronic hepatitis C who did not respond to a prior IFN/ribavirin therapy. Here, we investigated a daily dosing regimen with CIFN in combination with ribavirin. As expected for non-responders, our 77 patients showed baseline characteristics that are associated with a poor treatment response (Table 1). The treatment with daily dosing of CIFN plus ribavirin resulted in surprisingly high EVR of more than 80% but after the end of therapy a high relapse rate was observed. The overall SVR was 30% for this difficult-to-treat patient group. The most important group of patients in our study population, the IFN/ribavirin non-responder group achieved a SVR of 22%, which is about 5 10% higher than reported in prior non-responder studies [3,4,22]. However, it is always a problem to compare studies, especially non-responder trials. Non-responder patients are a highly heterogeneous group (Table 1). It is sometimes difficult to determine if the patients were true nonresponders to the prior therapy or if these patients were just non-compliant. Another factor contributing to the heterogeneity of the non-responders is the response kinetic

8 298 M. Cornberg et al. / Journal of Hepatology 44 (2006) Table 3 Patient characteristics explored in univariate analysis Patient characteristics SVR (nz23) No SVR (nz54) Treatment group A(9 mg) /B(18 mg) 12/11 26/28 O0.2 Male/female patients 17/6 41/13 O0.2 Mean age [yearsgsd] 41.5G G Mean body weight [kggsd] 77.7G G13.1 O0.2 HCV-genotype distribution O0.2 HCV-Genotype 1a/b HCV-Genotype 3a 1 1 HCV-Genotype 4 or Mean HCV-virus load [Log 10 units/mlgsd] 6.1G G0.5 O0.2 Median ALT(Max) 96 (222) 40 (136) Mode of transmission O0.2 Blood products/transfusion I.V. drug abuse 1 4 Surgery, needle stick, tattoo 4 11 Unknown 5 12 Liver histology before therapy 0.18 No/mild fibrosis (Ishak Score F0-2) Liver fibrosis (Ishak Score F3-4) 5 19 Liver cirrhosis (Ishak Score F5-6) 3 13 Not available 2 0 Previous therapy 0.14 IFN monotherapy/ifncribavirin combination therapy 14/9 22/32 Response to prior therapy O0.2 Flat non-response(!1 log red. of HCV-RNA) 3 12 Flat partial response(o1 log!2 log red. of HCV-RNA) 2 6 Significant HCV-RNA reduction(o2 log red. of HCV-RNA) 5 8 Only qualitative HCV-RNA available First-phase decay of HCV RNA after 24 h [Log 10 units/mlgsd] 1.0G G1.1 O0.2 Rapid virologic response(rvr)/no RVR 13/10 10/ Median Second-phase slope (Min, Max) [day K1 ] 0.17 (0.09,0.69) 0.07 (0.0,0.31) p during the previous therapy. The different response kinetics during prior therapies may influence the response during the re-treatment. Patients who achieved a O1 log 10 decay of HCV-RNA in the earlier therapy had a sharper viral decline during the re-treatment (Fig. 4B). One limitation in this study and in many other studies is that the response kinetic to the prior therapy is unknown in many patients (Table 1). Comparing the response to the prior therapy with the response to the daily CIFN/ribavirin treatment in individual patients, it seems that CIFN can indeed have a higher antiviral potency in some patients (Fig. 5). The high EVR in our study but on the other hand the high relapse rate may raise new questions and ideas for further studies. A longer treatment with CIFN plus ribavirin might be an option for these patients. In this specific difficult-to-treat patient group, age and baseline ALT were independent predictors of SVR whereas other well-known predictive factors for SVR as, e.g. HCVgenotype, were not significantly associated with SVR here (Table 3). In Layden et al. [18], RVR defined as a secondphase slope during TW1 and TW4 faster than K0.1/day was highly predictive for SVR in previously untreated patients receiving CIFN monotherapy. They reported that no patient without RVR achieved SVR. Here, RVR remained to be associated with SVR but even 23% of the patients without RVR achieved SVR. This may be due to combination therapy with ribavirin, mainly. Adherence to therapy is another important factor for the success of the treatment [21]. Therefore, therapies that induce severe side effects are in the end less effective despite higher antiviral efficacy. The adherence to therapy here was comparable to previous tiw CIFN regimens [10]. However, in contrast to a report from the USA in treatment naïve patients [20], 9 mg daily CIFN in combination with ribavirin were well tolerated in our study. The selection of non-responders in our study who are motivated and already tolerated a prior IFN therapy may account for this discrepancy. Another German study also reported good tolerability of a daily 9 mg CIFN plus ribavirin regimen in non-responders (P. Buggisch personal communication). However, the induction-dosing of 18 mg CIFN in combination with ribavirin required early dose modification due to side effects and laboratory abnormalities in 49% of the patients such as patient 32 and 70 (Fig. 5). This is in line with a previous non-responder study from Canada [23]. Three of our patients treated with the high dose even dropped out due to adverse events before TW8. This might help to explain why the overall response rates were not higher in Group B (Fig. 3A) despite the initially higher response rates (Fig. 4A). The stronger viral decline in the

9 M. Cornberg et al. / Journal of Hepatology 44 (2006) Fig. 5. HCV-RNA levels of six individual patients during the prior therapy with standard IFN alfa-2b plus ribavirin and during the current therapy with CIFN plus ribavirin. first-phase with the induction-dosing might reflect the higher antiviral efficacy of a higher dose but due to dose reductions and the comparable RVR, this effect could not be transferred into SVR. Layden et al. discussed another possibility that might help to explain why an induction regimen is in the end less effective as they speculate that induction would select for a therapy resistant virus strain [18]. Overall, our data suggest that the high-daily-doseinduction is not superior. The patients with liver cirrhosis had as expected very low response rates. Patient 67 is an example with almost no response to therapy (Fig. 5). This male patient had HCV-genotype-1, was older than 45 years, had relative low baseline ALT levels, had liver cirrhosis, and had a flat non-response to the prior IFN/ribavirin therapy. Furthermore, a dose modification of CIFN was required due to thrombopenia. Unfortunately, these patients who would benefit the most from a curative antiviral treatment have the worst outcome. These patients may be better treated with low dose interferon maintenance therapy to prevent complications of liver cirrhosis, although only one study to date has demonstrated any benefit in this approach [5]. We were also interested in the effect of CIFN on the histology as we compared the liver biopsies from 37 patients before and after therapy (Table 4). As expected,

10 300 M. Cornberg et al. / Journal of Hepatology 44 (2006) Table 4 Histological response of 37 patients with paired liver biopsies Patients with paired biopsies (before therapy, week 72) SVR (nz15) Relapse (nz14) Breakthrough and NR (nz8) All patients (nz37) Staging: Fibrosis (Mean Ishak F) Before/week /2.0 (n.s.) 3.2/2.9 (n.s.) 3.1/3.6 (n.s.) 2.9/2.7 (n.s.) Fibrosis improved 9 (60%) 5 (36%) 0 (0%) 14 (38%) Fibrosis not changed 4 (27%) 6 (43%) 4 (50%) 14 (38%) Fibrosis worsened 2 (13%) 3 (21%) 4 (50%) 9 (24%) Grading: inflammation (Mean Ishak A D) before/week /3.5 (pz0.007) 6.1/4.0 (pz0.006) 5.3/5.2 (pz0.9) 5.6/4.0 (pz0.0001) Periportal necroinflammation (Mean Ishak A) 1.7/1.1 (pz0.09) 1.9/1.3 (pz0.04) 1.8/2.1 (pz0.3) 1.8/1.4 (pz0.07) Bridging and confluent necrosis (Mean Ishak B) 0.1/0.1 (n.s.) 0.1/0.1 (n.s.) 0.0/0.0 (n.s.) 0.1/0.1 (n.s.) Focal lobular necrosis/apoptosis (Mean Ishak C) 1.3/0.8 (pz0.009) 1.6/0.9 (pz0.02) 1.3/1.1 (pz0.6) 1.4/0.9 (pz0.0005) Portal inflammation (Mean Ishak D) 2.3/1.5 (pz0.01) 2.5/1.7 (pz0.03) 2.3/1.9 (pz0.3) 2.4/1.6 (pz0.0006) Inflammation improved 10 (67%) 9 (64%) 3 (38%) 22 (59%) Inflammation not changed 5 (33%) 4 (29%) 2 (25%) 11 (30%) Inflammation worsened 0 (0%) 1 (27%) 3 (38%) 4 (11%) the patients who demonstrated a SVR had the best improvement in staging and grading (Table 4). Only 2 of the 15 sustained responder analyzed had a worsening in the fibrosis score. Both patients had elevated transaminases at the end of follow-up and high gamma-gt values, possibly due to alcohol consumption. Also the relapse patients showed a histological benefit after the end of follow-up (Table 4). In contrast, patients with only short-term virus suppression (breakthrough) or a complete non-response demonstrated no histological benefit. This data suggest that only a long-term virus suppression or clearance can lead to a histological response. In conclusion, before starting long-term maintenance therapies with IFN in all non-responder patients, one may consider alternative therapies such as daily dosing of CIFN plus ribavirin to achieve viral clearance. Selected, highly motivated and compliant patients may benefit from this type of therapy. 8. Study group Medizinische Hochschule Hannover (M.P. Manns, M. Cornberg, H. Wedemeyer, J. Hadem, M. Bastürk, A. Schüler, H.L. Tillmann, C. Trautwein), Krankenhaus Bad Oeynhausen (F. Schuppert, O. Becker, W. Steuber), Charité Berlin (E. Gentz, A. Roske, J. Jetschmann, H.H.-J. Schmidt), Ruhr-Universität Bochum (W. Schmiegel, M. Reiser), Praxis Braunschweig (O. Marschal), Franziskus- Hospital Bielefeld (M. Steffen, R. Müller), Klinikum Kassel (J. Pausch), Praxis Koblenz (G. Hermesdorf). Acknowledgements The study was supported by an unrestricted research grant from Yamanouchi Pharma GmbH (now Astellas Pharma GmbH), Germany, the German competence network for viral hepatitis (Kompetenznetz Hepatitis), and the clinical research unit (KFO 129, supported by the Deutsche Forschungsgemeinschaft). M.P. Manns is a consultant to companies marketing alfa-interferons. We thank N. Kothe, P. Magerstedt, Hannover, and K.H. Schweickert for technical assistance, and L.A. Wilkinson and F.S. Saccoccio for reading the manuscript and helpful discussion. References [1] Cornberg M, Wedemeyer H, Manns MP. Treatment of chronic hepatitis C with PEGylated interferon and ribavirin. Curr Gastroenterol Rep 2002;4: [2] Shiffman ML. Management of patients with chronic hepatitis C virus infection and previous nonresponse. Rev Gastroenterol Disord 2004; 4(Suppl. 1):S22 S30. [3] Shiffman ML, Di Bisceglie AM, Lindsay KL, Morishima C, Wright EC, Everson GT, et al. Peginterferon alfa-2a and ribavirin in patients with chronic hepatitis C who have failed prior treatment. Gastroenterology 2004;126: [4] Poynard T, Schiff ER, Terg R, Goncales F, Diago M, Reichen J, et al. Sustained virological response (SVR) in the EPIC3 trial: week 12 virology predicts SVR in previous interferon/ribavirin treatment failures receiving PEG-intron/rebetol (PR) weight based dosing (WBD). J Hepatol 2005;42(Suppl. 2): [5] Curry M, Cardenas A, Afdhal NH. Effect of maintenance PEG-Intron therapy on portal hypertension and its complications: results from the COPILOT study. J Hepatol 2005;42(Suppl. 2):40. [6] Blatt LM, Davis JM, Klein SB, Taylor MW. The biologic activity and molecular characterization of a novel synthetic interferon-alpha species, consensus interferon. J Interferon Cytokine Res 1996;16: [7] Ozes ON, Reiter Z, Klein S, Blatt LM, Taylor MW. A comparison of interferon-con1 with natural recombinant interferons-alpha: antiviral, antiproliferative, and natural killer-inducing activities. J Interferon Res 1992;12: [8] Tong MJ, Reddy KR, Lee WM, Pockros PJ, Hoefs JC, Keeffe EB, et al. Treatment of chronic hepatitis C with consensus interferon: a multicenter, randomized, controlled trial. Consensus interferon study group. Hepatology 1997;26:

11 M. Cornberg et al. / Journal of Hepatology 44 (2006) [9] Sjogren MH, Sjogren R, Holtzmuller K, Winston B, Butterfield B, Drake S, et al. Interferon alfacon-1 and ribavirin versus interferon alpha-2b and ribavirin in the treatment of chronic hepatitis C. Dig Dis Sci 2005;50: [10] Fattovich G, Zagni I, Minola E, Felder M, Rovere P, Carlotto A, et al. A randomized trial of consensus interferon in combination with ribavirin as initial treatment for chronic hepatitis C. J Hepatol 2003; 39: [11] Heathcote EJ, Keeffe EB, Lee SS, Feinman SV, Tong MJ, Reddy KR, et al. Re-treatment of chronic hepatitis C with consensus interferon. Hepatology 1998;27: [12] Rustgi VK, Bonkovsky HL, Arora S, Zeuzem S, Manns M, Blatt LM. Daily versus three-times-weekly IFN alfacon-1 in previously untreated HCV patients results in a significant greater rate of SVR: final results of an international phase 4 study. Gastroenterology 2005; 128(Supp. 2):A-716. [13] Kaiser S, Hass HG, Gregor M. Successful retreatment of interferon/ribavirin nonresponders with daily dosing of consensus interferon. J Hepatol 2005;42(Suppl. 2): [14] Carithers RL, Zeuzem S, Manns MP, McHutchison JG, Perrillo RP, Bailey R, et al. Multicenter, randomized, controlled trail comparing high dose daily induction interferon plus ribavirin versus standard interferon alfa-2b plus ribavirin. Hepatology 2000;32:317A. [15] Fried MW, Shiffman M, Sterling RK, Weinstein J, Crippin J, Garcia G, et al. A multicenter, randomized trial of daily high-dose interferon-alfa 2b for the treatment of chronic hepatitis c: pretreatment stratification by viral burden and genotype. Am J Gastroenterol 2000; 95: [16] Hadziyannis AS, Papaioannou C, Spanou F, Manesis EK, Hadziyannis SJ. Induction interferon therapy in naive patients with chronic hepatitis C: increased end-of-treatment virological responses but absence of long-term benefit. Aliment Pharmacol Ther 2001;15: [17] Cotler SJ, Layden JE, Neumann AU, Jensen DM. First phase hepatitis c viral kinetics in previous nonresponders patients. J Viral Hepat 2003;10: [18] Layden TJ, Layden JE, Reddy KR, Levy-Drummer RS, Poulakos J, Neumann AU. Induction therapy with consensus interferon (CIFN) does not improve sustained virologic response in chronic hepatitis C. J Viral Hepat 2002;9: [19] Ishak K, Baptista A, Bianchi L, Callea F, De Groote J, Gudat F, et al. Histological grading and staging of chronic hepatitis. J Hepatol 1995; 22: [20] Pockros PJ, Reindollar R, McHutchinson J, Reddy R, Wright T, Boyd DG, et al. The safety and tolerability of daily infergen plus ribavirin in the treatment of naiive chronic hepatitis C patients. J Viral Hepat 2003;10: [21] McHutchison JG, Manns M, Patel K, Poynard T, Lindsay KL, Trepo C, et al. Adherence to combination therapy enhances sustained response in genotype-1-infected patients with chronic hepatitis C. Gastroenterology 2002;123: [22] Krawitt EL, Ashikaga T, Gordon SC, Ferrentino N, Ray MA, Lidofsky SD. Peginterferon alfa-2b and ribavirin for treatmentrefractory chronic hepatitis C. J Hepatol 2005;43: [23] Moskovitz DN, Manoharan P, Heathcote EJ. High dose consensus interferon in nonresponders to interferon alpha-2b and ribavirin with chronic hepatitis C. Can J Gastroenterol 2003;17: