Ribavirin with either standard or pegylated interferon to treat recurrent hepatitis C after liver transplantation

Authors

  • V. R. CICINNATI,

    1. Department of Gastroenterology and Hepatology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    2. Department of General, Visceral and Transplantation Surgery, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author
  • S. IACOB,

    1. Department of General, Visceral and Transplantation Surgery, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    2. Gastroenterology and Hepatology Center, Fundeni Clinical Institute, Bucharest, Romania
    Search for more papers by this author
  • C. G. KLEIN,

    1. Department of Gastroenterology and Hepatology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    2. Department of General, Visceral and Transplantation Surgery, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author
  • H. A. BABA,

    1. Institute of Pathology and Neuropathology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author
  • G. C. SOTIROPOULOS,

    1. Department of General, Visceral and Transplantation Surgery, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author
  • P. HILGARD,

    1. Department of Gastroenterology and Hepatology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author
  • Y. ERIM,

    1. Department of Psychosomatic Medicine and Psychotherapy, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author
  • C. E. BROELSCH,

    1. Department of General, Visceral and Transplantation Surgery, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author
  • G. GERKEN,

    1. Department of Gastroenterology and Hepatology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author
  • S. BECKEBAUM

    1. Department of Gastroenterology and Hepatology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    2. Department of General, Visceral and Transplantation Surgery, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
    Search for more papers by this author

PD Dr S. Beckebaum, Interdisciplinary Liver Transplant Unit, University Hospital Essen, OPZ 2, Ebene A1, Hufelandstr. 55, 45122 Essen, Germany.
E-mail: susanne.beckebaum@uni-due.de

Summary

Aim  To investigate the efficacy of two anti-viral protocols in hepatitis C virus-reinfected liver transplant recipients.

Methods  In this prospective study, 26 liver transplant patients were treated with standard interferon-α2b for 12 months or standard interferon-α2b for 3 months followed by pegylated interferon-α2b for 9 months. Interferon was combined with ribavirin in all patients. The histological course of the study population was compared with an untreated historic control group (n = 38) with similar baseline characteristics.

Results  The sustained virological response rates in the standard interferon group and in the pegylated interferon group were 27.3% and 26.7%, respectively. Only 29% of patients with sustained virological response had end of treatment histological response, whereas 47% of viral non-responders showed end of treatment histological response. The percentage of patients with histological improvement was significantly higher in the study population when compared to the controls. Univariate analysis indicated that hepatitis C virus genotype non-1, high baseline alanine aminotransferase, the time interval between liver transplant and interferon therapy and the body mass index predicted sustained virological response. In the multivariate model, baseline alanine aminotransferase and the body mass index remained a significant predictor of sustained virological response.

Conclusions  Both treatment regimens offer similar efficacy profiles. Failure to eradicate hepatitis C virus should not lead to treatment discontinuation if serial liver biopsies demonstrate histological response.

Introduction

Hepatitis C virus (HCV)-related cirrhosis represents the leading indication for liver transplantation (LT) worldwide. Recurrence of HCV infection in replicative patients occurs during reperfusion of the allograft and the natural course of the disease is accelerated compared to the non-transplant setting.1 At least 50% of recipients develop histological evidence of recurrent disease within 1 year post-transplant2 and progression to allograft cirrhosis occurs in 10–30% at 5 years of follow-up,3 leading to increasing incidence of retransplantation in HCV recipients.4 In the context of a growing demand for LT services and shortage of cadaveric donor organs, there are continuous efforts to improve the outcome in HCV recipients.

Currently, available anti-viral regimens are effective in immunocompetent patients with overall sustained virological response (SVR) rates of approximately 55%.5–7 However, in the setting of LT, sustained response rates are markedly lower and highly variable.8 Potential factors responsible for reduced virological response to anti-viral therapy are non-response to interferon (IFN) and ribavirin (RBV) therapy prior to LT, influence of immunosuppression on drug efficacy and tolerance, predominance of genotype 1 infection, frequent dose reduction because of cytopenia or renal dysfunction and high rates of discontinuation. Thus, treatment of established HCV infection in the LT setting is much more challenging as in non-immune compromised patients, and several questions such as the onset, the modality and the duration of treatment have yet to be answered.9 In addition, the criteria for selecting patients for anti-viral treatment have not been evaluated yet.

Anti-viral therapy has been applied prior to LT, pre-emptively in the early post-operative period or for established recurrent hepatitis C. However, pretransplant IFN treatment has not shown to be a convincing approach because of low therapeutic benefit, high rate of premature discontinuation and severe side effects including the risk of infectious complications, hepatic decompensation and death.10, 11 Moreover, patients with pretransplant response to anti-viral therapy may relapse after surgery.10 Only a few studies have been published reporting experiences with pre-emptive anti-viral therapy within the first weeks after LT.12–15 This ‘prophylactic’ approach though does not seem feasible in a considerable proportion of patients due to initial intense immunosuppression, post-operative infections, cytopenia and insufficient recovery from surgery. Another approach, focusing on the treatment of patients with established HCV reinfection of the allograft, seems to be the most applicable strategy.16–18

The aim of this prospective study was to determine the efficacy and the tolerability of two anti-viral treatment strategies in LT recipients with established HCV reinfection and to compare their histological course with an untreated historic control group. In addition, potential predictive factors for virological, histological and biochemical response were assessed to achieve a better selection of patients for effective anti-viral therapy.

Materials and methods

Study population

Since 2001, we prospectively evaluated transplant recipients with recurrent HCV infection in our Liver Transplant Unit.

Patients were eligible for anti-viral therapy if they fulfilled the following inclusion criteria: (i) positive HCV-RNA [serum HCV-RNA levels determined by the branched DNA signal amplification assay (Bayer Diagnostics, Emeryville, CA, USA; limit of detection 615 IU/mL) and qualitative testing utilizing the TMA assay (Bayer Diagnostics, limit of detection 9.6 IU/mL)]; (ii) histological evidence of recurrent HCV infection, scored according to the Ishak-modified HAI System using a seven-point scale for fibrosis stage (0–6) and an 18-point scale for inflammatory activity. Patients were excluded from the study if they presented with any of the following manifestations: white blood counts <3000/mm3, neutrophils <1500/mm3, platelet counts <80 000/mm3, haemoglobin levels <11 g/dL, liver cirrhosis, hepatitis B virus or human immunodeficiency virus coinfection, other concurrent causes of liver disease, alcohol abuse or drug dependence within the previous 6 months, severe psychiatric illness, thyroid dysfunction, severe pulmonary or cardiovascular disease, renal impairment defined as serum creatinine >1.5 mg/dL and/or creatinine clearance <50 mL/min, current pregnancy or breastfeeding of infants, unable/unwilling to use contraception during the treatment and 6 months after discontinuation. For exclusion of severe psychiatric illness, patients were evaluated by a specialist from the Department of Psychosomatic Medicine and Psychotherapy at the University Hospital Essen. None of our patients included in this study met the criteria for major depression prior to administration of IFN. Patients were educated about potential coping strategies that are beneficial for IFN-related neuropsychiatric adverse events. Patients were extensively informed about potential IFN-induced neuropsychiatric side effects and were educated about the possibility of antidepressant therapy to allow them to play an active role in the decision whether to start therapy. Patients were also educated to receive, if necessary, psychotherapist consultations.

Historic control group

Potential control individuals were obtained from a historic database including 102 patients who were transplanted for HCV-induced cirrhosis in our department and had data from regular follow-up visits. Among these recipients, 38 patients fulfilled the selection criteria for inclusion in the historic control group: positive HCV-RNA levels in the serum and histological evidence of recurrent HCV infection, available paired graft biopsies (defined as two consecutive liver biopsies at an interval of 12 months), no IFN therapy after LT and lacking evidence of hepatitis B virus or human immunodeficiency virus coinfection or other concurrent causes of liver disease. None of the patients in the historic control group had a history of alcohol abuse.

Therapeutic protocol and follow-up

Patients were treated with standard IFN-α2b administered at a dose of 2 million units (MU) daily for 12 months. After pegylated (PEG)-IFN became available, the remaining patients were treated with standard IFN-α2b (2 MU/day) during the first 3 months followed by PEG-IFN-α2b at a dose of 1.5 μg/kg once weekly for the following 9 months. RBV was administered concomitantly with IFN in both groups at a dose of 10–12 mg/kg/days.

All patients on treatment were scheduled to have complete blood count and liver function tests that were assessed weekly for the first 4 weeks, then for every 2–4 weeks and whenever required while on treatment. RBV dose was reduced by 25–50% for haemoglobin levels below 9.5 g/dL and was discontinued if haemoglobin levels were <8 g/dL despite erythropoietin (EPO) therapy. The IFN dose was reduced by 25–50% for leucocyte counts between 1000/mm3 and 1500/mm3 and/or neutrophil counts between 500/mm3 and 750/mm3 despite concomitant granulocyte colony-stimulating factor (G-CSF) therapy (administered if leucocyte count was <1700/mL and/or neutrophil count was <750/mm3), and for platelet counts <50 000/mm3. However, IFN was discontinued in case of platelet counts <25 000/mm3 or white blood cell counts <1000/mm3 and/or neutrophils <500/mm3.

Quantitative HCV-RNA was measured monthly within the first 12 weeks after the onset of therapy and thereafter every 3 months until 6 months after the completion of therapy. All patients underwent protocol liver biopsies 1–3 months before the onset of therapy, 6 months after the beginning of therapy, at the end of therapy and thereafter at yearly intervals. All patients had follow-up investigations in the post-treatment period for at least 24 weeks.

Study end points and response assessment

The primary end point was the achievement of SVR (negative qualitative serum HCV-RNA at week 72). Secondary end points included end of treatment virological response (EOTVR) as defined by negative qualitative serum HCV-RNA assay at the end of anti-viral therapy at week 48, end of treatment biochemical response (EOTBR) as defined by normalization of serum alanine aminotransferase (ALT) activity at week 48 and end of treatment histological response (EOTHR) which was considered in the case of a decrease of two or more points in the grading score in association with unchanged or improved fibrosis score according to the Ishak-modified HAI system.

Very early virological response (VEVR) and early virological response (EVR) were defined as at least a 2 log10 drop or undetectable HCV-RNA at week 4 and at week 12 after onset of therapy.

To identify factors associated with virological, histological and biochemical response, we merged the two treatment groups and included the following parameters in the univariate analysis.

Related to the recipient prior to anti-viral therapy

Presence of diabetes mellitus type II, gender, administration of bolus corticosteroids for acute rejection episodes, mycophenolate mofetil or azathioprine induction and prior prednisone therapy.

Related to the recipient at the onset of anti-viral therapy

Age, body mass index (BMI)/BMI ≥ 25 kg/m2, inflammatory activity, fibrosis score and presence of steatosis prior to therapy, type of immunosuppression [ciclosporin (CSA) or tacrolimus (TAC) monotherapy], time period between LT and start of IFN therapy, ALT, ALT/aspartate aminotransferase (AST)/gamma glutamyl transferase (γGT)/alkaline phosphatase (AP) values more than two times the upper normal limit (UNL) and AST/ALT ratio >2.

Related to the virus

Hepatitis C virus genotype, viral load at baseline, high level of viraemia before the onset of therapy (>1 500 000 IU/mL).

Related to the donor

Living or deceased donor, age and gender.

Related to the anti-viral therapy

Administration of pretransplant anti-viral therapy, type of IFN (standard IFN/PEG-IFN), a 2 log10 drop of viraemia after 1 month/after 3 months of therapy, the need to decrease the IFN or RBV doses and completion of anti-viral regimen.

Safety and tolerability were assessed through monitoring of adverse events (e.g. influenza-like symptoms, gastrointestinal, respiratory, neuropsychiatric side effects) and laboratory abnormalities (anaemia, leucopenia/neutropenia and thrombocytopenia) at scheduled visits or when they were reported during the treatment. Adverse events were graded as mild, moderate or severe requiring withdrawal of therapy.

Statistical analysis

The analysis of the study population was conducted on an intent-to-treat basis. Continuous data were expressed as mean ± s.d. (if not otherwise indicated) and Mann–Whitney U-test was used to compare continous variables between the IFN group and the PEG-IFN group, and between the overall study population and the historic control group. Kruskal–Wallis test was used to compare between the IFN group, the PEG-IFN group and the historic control group. Wilcoxon’s rank test was carried out to compare continuous variables at the beginning and at the end of therapy. Fisher’s exact test was used to compare categorical data between the groups. To identify single variables associated with EOTVR, EOTHR, EOTBR and SVR, univariate analyses were conducted by logistic regression. Stepwise regression analyses with a foreward variable selection were carried out to detect factor combinations for the prediction of response to anti-viral therapy.

All factors with a P-value of <0.05 were entered in a stepwise logistic regression model, whereas all factors with a P-value of >0.10 were excluded from the model. The significance of the logistic regression models was tested by using likelihood ratio of chi-squared tests. To test the significance of each independent variable in the stepwise logistic regressions, Wald tests (IzI) were conducted. A two-tailed P-value of <0.05 was required for measuring statistical significance. Statistics were performed using Intercooled stata 9.1 (StataCorp LP, College Station, TX, USA).

Results

Baseline features

From November 2001 to March 2005, 79 LT recipients with recurrent hepatitis C infection in the allograft were screened for eligibility criteria. A total of 46 patients did not qualify for anti-viral therapy; 17 patients were discarded due to anaemia and/or thrombocytopenia and/or leucopoenia; three patients showed histologically proven ductopenic rejection; seven patients had biliary complications; one patient revealed severe cardiovascular disease, two patients had HBV coinfection, five were transplanted due to concurrent causes of liver disease, four had renal failure and seven patients did not give their consent to receive anti-viral therapy.

Furthermore, seven patients are still under anti-viral therapy or have a follow-up period of <24 weeks and were therefore excluded from the analysis. Thus, 26 patients were included in the present analysis.

Eleven patients (standard IFN group) were treated with standard IFN-α2b plus RBV, the remaining 15 patients (PEG-IFN group) were treated with standard IFN-α2b plus RBV during the first 3 months followed by PEG-IFN-α2b plus RBV for the following 9 months.

Comparative analysis of the patients’ baseline characteristics did not show statistically significant differences between both groups (Table 1). The standard IFN and PEG-IFN groups consisted of patients with a mean age of 56.4 ± 5.7 years (range: 48–66) and 52.1 ± 6.9 years (range: 38–67). Eight of 11 patients (standard IFN group) and 14 of 15 patients (PEG-IFN group) were diagnosed with diabetes mellitus. Mycophenolate mofetil induction therapy directly after LT was administered in the standard IFN group and PEG-IFN group in 55% and 67%, whereas azathioprine induction therapy was administered in four and three patients, respectively. All patients received initial prednisone treatment for a mean time period of 47.9 ± 50.7 weeks and 40.8 ± 22.4 weeks, respectively. Four vs. five patients in the PEG-IFN group underwent previous corticosteroid bolus therapy because of acute rejection. None of the study patients had a history of chronic rejection. At the onset of anti-viral therapy, immunosuppression consisted of CSA monotherapy in the majority of patients, while the remainders received TAC monotherapy. Liver values are depicted in Table 1. Serum ALT values were averaged more than two times than the UNL in 55% vs. 60% in the PEG-IFN group; γGT and AP were more than two times than the UNL in three vs. two patients and in none vs. one patient. In the standard IFN group, eight patients were infected with genotype 1 and 3 with genotype 3, whereas 11 patients were infected with genotype 1, two with genotype 2, one with genotype 3 and one with genotype 4 in the PEG-IFN group. Anti-viral therapy after LT was started after a mean time period of 142 ± 145 weeks (standard IFN group) and 151 ± 100 weeks (PEG-IFN group), respectively (= 0.516).

Table 1.   Comparative analysis of the patients’ baseline characteristics
Variable IFN group (n = 11)PEG-IFN group (n = 15)P-valueOverall study population (n = 26)Historic control group (n = 38)P-valueP-value**
  1. Values are expressed as mean ± s.d. or percentages.

  2. IFN, interferon; PEG-IFN, pegylated IFN; OLT, orthotopic liver transplantation; BMI, body mass index; ALT, alanine aminotransferase.

  3. * According to the Ishak-modified HAI system; ** For comparison between the IFN group, PEG-IFN group and historic control group.

Age at OLT (years)56.4 ± 5.752.1 ± 6.90.1053.9 ± 6.754.8 ± 10.80.490.35
Recipient gender (male)73%93%0.2885%73%0.370.25
Donor gender (male)64%73%0.6869%61%0.600.72
Donor age (years)37.4 ± 17.133.7 ± 14.20.5635.3 ± 15.337.0 ± 14.4 0.590.73
Pre-OLT anti-viral therapy46%40%1.042%34%0.800.94
HCV genotype 173%73%1.073%76%0.781.0
BMI before therapy (kg/m2)26.8 ± 1.426.7 ± 3.10.8926.8 ± 3.826.4 ± 8.60.260.53
HCV-RNA >1.5 × 106 IU/mL before therapy64%73%0.4969%58%0.440.59
Necroinflammatory score*4.1 ± 1.43.9 ± 1.20.764.0 ± 1.33.7 ± 1.90.310.56
Fibrosis score*1.4 ± 0.511.6 ± 1.00.471.5 ± 0.81.7 ± 1.00.280.42
Steatosis46%40%1.042%39%1.00.94
Baseline ALT >2 times the UNL55%60%1.058%50%0.620.83
Baseline bilirubin (mg/dL)1.1 ± 0.420.96 ± 0.410.321.0 ± 0.41.2 ± 0.60.390.30
Baseline creatinine (mg/dL)1.2 ± 0.251.2 ± 0.190.681.2 ± 0.21.1 ± 0.50.430.14
Ciclosporin/tacrolimus monotherapy9/211/41.020/627/110.780.92
Deceased/living donor 8/313/20.6221/531/71.00.74

Characteristics of the historic control group are depicted in Table 1. There were no statistically significant differences between the overall study population (IFN group plus PEG-IFN group) and the historic control group as well as between the three distinct groups. In the historic control group, the mean time period between LT and ‘baseline’ biopsy (defined as the first of the two consecutive graft biopsies) was comparable with the mean time period between LT and the baseline biopsy before onset of anti-viral therapy in the study group (159 ± 178 weeks vs. 147 ± 119 weeks, P = 0.67).

Efficacy

The virological and biochemical responses achieved in the two treatment groups are illustrated in Figure 1. Overall, there was no statistically significant difference between the two treatment groups regarding the anti-viral efficacy. The VEVR rates were observed as 45% and 27% (= 0.42) and the EVR rates were 73% and 53% (= 0.43) in the standard IFN and the PEG-IFN group, respectively. Likewise, the EOTVR rates (36% vs. 40%) were not statistically significantly different between the two groups. Three of four patients (standard IFN group) and four of six patients (PEG-IFN group) who achieved EOTVR maintained viral response. All patients with SVR were men with a mean age of 55.4 ± 6.0 years (range: 49–67) and all had completed anti-viral therapy. The comparison between the onset and end of anti-viral therapy revealed a marked decrease of ALT values with normalization in about 50% of patients in both groups. Two of three sustained responders vs. two of four sustained responders (PEG-IFN group) had EOTBR. Interestingly, only one of three patients and one of four patients with SVR had EOTHR in the standard IFN group and PEG-IFN group, respectively. For both, the standard IFN group and the PEG-IFN group, there was a slight but not a significant improvement in the necroinflammatory activity (4.1 ± 1.4 at onset of therapy vs. 3.2 ± 1.6 at 12 months after start of therapy, = 0.17 and 3.9 ± 1.2 vs. 3.4 ± 2.0, P = 0.37), and an increase in the fibrosis score (1.4 ± 0.51 vs. 2.1 ± 0.83 and 1.6 ± 1.0 vs. 2.7 ± 1.5, < 0.005 and < 0.05). All patients except one of each group who had EOTHR had concomitant EOTBR. Figure 2 illustrates grading and staging scores and histological response after 12 months of therapy.

Figure 1.

 Biochemical and virological response rates were not statistically significantly different in the standard interferon (IFN) group (n = 11) and the PEG-IFN group (n = 15), respectively (intention-to-treat analysis).

Figure 2.

 Improvement of necroinflammatory activity and staging score stabilization (unchanged or improved fibrosis score according to the Ishak-modified HAI system) in the standard interferon (IFN) group and the PEG-IFN group, respectively. End of treatment histological response was detectable in 27% of patients in both groups.

Analysis of the histological course in the historic control group vs. the treatment group

For comparison of the historic control group with the study population we merged the two treatment groups.

In the historic control group, patients showed an increase in the necroinflammatory score (3.7 ± 1.9 vs. 4.8 ± 3.4 after 12 months, P = 0.081), and a significant increase in the fibrosis score (1.7 ± 1.0 vs. 3.4 ± 1.8 after 12 months, < 0.001). To the contrary, in the overall study population, the necroinflammatory activity tended to improve upon anti-viral treatment (4.0 ± 1.3 vs. 3.3 ± 1.8, = 0.119). Improvement of necroinflammatory activity occurred in 42% of study patients vs. 11% of patients in the control group (= 0.006, Figure 3). Fibrosis was unchanged or improved in 42% of treated patients vs. 21% of patients in the control group (= 0.096). Histological improvement [grading score decrease (≥2 points) in association with unchanged or improved fibrosis score, defined as EOTHR in the study group] occurred in 27% of treated patients vs. 5% of controls (= 0.025).

Figure 3.

 Improvement of necroinflammatory activity and staging score stabilization in the overall study population and the historic control group, respectively. Histological improvement was considered in those with improved necroinflammatory activity (decrease of ≥2 points) and concomitant stabilization of fibrosis (unchanged or improved staging score).

Safety: side effects, dose reduction and premature withdrawal

No patient developed acute rejection during IFN therapy and no patient began the anti-viral therapy within the first 3 months post-LT. The most common side effects during the first 12 weeks of therapy were influenza-like symptoms, sleep disturbances, nausea, fatigue, myalgia and arthralgia. These adverse events were not statistically significantly different between both groups. None of our patients had to stop therapy because of severe neuropsychiatric side effects and none of them needed or required psychotropic medication.

The side effects between months 4 and 12 of therapy are illustrated in Figure 4. Myalgia/arthralgia, pyrexia and shortness of breath occurred more often in the PEG-IFN group, whereas loss of appetite and nausea were more often reported in the standard IFN group; these differences, however, were not statistically significant.

Figure 4.

 The incidence and the profile of side effects between 4 and 12 months of anti-viral therapy [standard interferon (IFN)/RBV vs. PEG-IFN/RBV] were comparable between both study groups.

During the first 3 months after onset of therapy, anaemia (haemoglobin <13.5 g/dL) occurred in nine and 13 patients, leucopoenia (<3000/mm3) in eight and 11 patients and thrombocytopenia (<80 000/mm3) in four and three patients in the standard IFN group and the PEG-IFN group, respectively. In addition, during this early study period, severe anaemia with haemoglobin values ≤10 g/dL was only detectable in two patients (standard IFN group) and three patients (PEG-IFN group), respectively, whereas severe leucopoenia (<1500/mm3) or thrombocytopenia (<50 000/mm3) was not detectable.

The haematological side effects between 4 and 12 months of anti-viral therapy encountered in both study groups are depicted in Table 2. There was a significant difference between the baseline values of blood counts and values at the end of therapy in the standard IFN group and the PEG-IFN group, respectively [haemoglobin: 13.7 ± 1.4 g/dL vs. 12.1 ± 1.7 g/dL (< 0.05) and 14.4 ± 1.3 g/dL vs. 11.6 ± 1.2 g/dL (< 0.001); leucocytes: 4.7 ± 1.2/nL vs. 3.7 ± 1.8/nL (< 0.05) and 3.5 ± 0.73/nL vs. 2.9 ± 0.17/nL (< 0.05), thrombocytes: 146 ± 78.4/nL vs. 126 ± 76.8/nL (< 0.01) and 168 ± 52.1/nL vs. 131 ± 43.8/nL (< 0.01)]. The incidence of IFN and RBV dose reduction was not statistically significantly different between both groups. During the first 3 months, IFN was reduced to 25–50% in five (standard IFN group) and four (PEG-IFN group) patients, respectively (= 0.42). During the following 9 months, reduced standard IFN and PEG-IFN were administered in five and six patients (= 1.00) and RBV dose reduction (by 25–50%) was necessary in four and five patients, respectively (= 1.00). Discontinuation of therapy was necessary in eight of 26 patients (31%). The causes of treatment discontinuation were: severe side effects [nausea and extensive weight loss (n = 2), fatigue and shortness of breath (n = 2), depression (n = 1), confusion (n = 1), severe accentuation of the psoriasis cutaneous lesions (n = 1)] and emigration to the US (n = 1).

Table 2.   Haematological side effects between 4 and 12 months of anti-viral therapy
VariableStandard IFN [n = 11; % (n)]PEG-IFN [n = 15; % (n)]P-value
  1. IFN, interferon; PEG-IFN, pegylated IFN; G-CSF, granulocyte colony-stimulating factor.

Anaemia (haemoglobin <13.5 g/dL)81.8 (9)100 (15)0.17
Haemoglobin 8–10 g/dL36.4 (4)33.3 (5)1.0
Repeated (>3×) erythropoietin use 36.4 (4)33.3 (5)1.0
Leucocytes 1500–3000/mm381.8 (9)86.7 (13)1.0
Leucocytes <1500/mm39.1 (1)0.0 (0)0.42
Neutrophils <750/mm39.1 (1)20 (3)0.61
Repeated (>3×) G-CSF use 9.1 (1)20 (3)0.61
Thrombocytes 50 000–80 000/mm354.5 (6)26.7 (4)0.23
Thrombocytes <50 000/mm318.2 (2)6.7 (1)0.56

Predictive factors for anti-viral response

Univariate analysis showed a positive correlation between SVR and HCV genotype non-1 and the presence of high ALT value (Table 3). The time period between LT and the onset of anti-viral therapy was negatively correlated with SVR. Analysing predictive factors for response at the end of treatment, we found a positive correlation between EOTVR and high baseline ALT value, decrease of viral load ≥2 log at 1 month and presence of genotype non-1. Fifteen patients had a BMI ≥25 kg/m2; the highest BMI value was 34.7 kg/m2. The BMI value at baseline was a negative predictor for EOTHR. ALT >2 times the UNL was associated with EOTBR, whereas previous bolus corticosteroid therapy and BMI ≥25 kg/m2 were negatively correlated with biochemical response.

Table 3.   Relation between EOTVR, EOTBR, EOTHR, SVR and patients’ characteristics by univariate analysis (only significant variables shown)
Variablesχ2> χ2Odds ratio95% confidence limits
  1. EOTVR, end of treatment virological response; ALT, alanine aminotransferase; EOTBR, end of treatment biochemical response; BMI, body mass index; UNL, upper normal limit; EOTHR, end of treatment histological response; SVR, sustained virological response; IFN, interferon.

EOTVR
 ALT >2× UNL7.80.0054151.5–149
 ALT7.60.00571.71.1–2.8
 Decrease of viral load  ≥2 log at month 14.60.03136.51.1–38.6
 Genotype non-14.40.03667.01.0–48.3
EOTBR
 BMI ≥25 kg/m28.20.00420.080.01–0.55
 ALT >2× UNL4.10.04405.30.97–29.3
 Previous bolus  corticosteroid therapy3.90.04700.190.04–1.0
EOTHR
 BMI7.20.00750.620.39–1.0
SVR
 ALT8.50.00371.81.1–3.0
 Genotype non-14.20.04157.11.0–49.5
 Time period between  OLT and start of IFN  therapy5.60.01770.990.99–1.0

By the multivariate logistic regression analysis, a high ALT value was identified as predictive of SVR (Table 4). Furthermore, overweight (BMI ≥25 kg/m2) was negatively associated with EOTBR, whereas an elevated ALT value (more than two times the UNL) positively predicted EOTVR. Histological response upon the end of treatment was negatively associated with BMI.

Table 4.   Relation between EOTVR, EOTBR, EOTHR, SVR and baseline characteristics by multivariate analysis (only significant variables shown)
VariablesOdds ratio> IzI95% confidence limits> χ2
  1. EOTVR, end of treatment virological response; ALT, alanine aminotransferase; EOTBR, end of treatment biochemical response; BMI, body mass index; EOTHR, end of treatment histological response; SVR, sustained virological response.

EOTVR
 ALT >2× UNL150.0211.5–1490.0054
EOTBR
 BMI ≥25 kg/m20.080.0100.01–0.550.0042
EOTHR
 BMI0.60.0480.4–1.00.0075
SVR
 ALT1.80.0281.1–3.00.0037

From those patients who had anti-viral therapy prior to LT (n = 11), only one had completed pretransplant anti-viral therapy with EOTVR but early relapse. In our study, this patient had VEVR and EVR under IFN/RBV treatment, but decided to stop therapy because of adverse events. All the other patients were withdrawn from preoperative anti-viral drugs because of non-response. By univariate analysis, we did not find any correlation between SVR and anti-viral therapy prior to LT.

Discussion

Hepatitis C virus patients show accelerated rates of cirrhosis and graft loss from recurrent hepatitis C in the medium- and long-term after LT.1, 19, 20 Although an increasing number of transplant centres use anti-viral regimens, treatment is not standardized and is often associated with low response rates because of insufficient dosing and patient’s intolerance. Quantitative analysis of published data suggest that approximately 24% of HCV transplant recipients treated with standard IFN/RBV and 27% of patients with PEG-IFN/RBV obtain SVR.21 In a recently published study by Fernández et al. 23% of patients administered PEG-IFN/RBV achieved SVR.22 So far, the therapeutic advantage of PEG-IFN is uncertain in the transplant setting, because studies comparing standard IFN head-to-head to PEG-IFN are limited.16, 23

This study has focused on the efficacy, the safety and potential predictive factors of response to anti-viral therapy in patients with HCV recurence after LT. In contrast to previous studies referring to a thrice weekly dosing of standard IFN, we chose to administer standard IFN daily to obtain a more favourable pharmacokinetic profile. Moreover, in the PEG-IFN group, the initial treatment during the first 3 months was performed with standard IFN for safety reasons and eventually faster dose adjustment in case of adverse effects. However, this approach limits the direct comparison of the efficacy of standard IFN vs. PEG-IFN therapy.

During the early treatment period, we found that the percentage of patients with at least a 2 log10 decrease of viral load was similar in both groups. Likewise, the EOTVR (36% vs. 40%) and SVR (27% vs. 27%) were not statistically significantly different in the standard IFN group and the PEG-IFN group, respectively, and sustained response rates are similar to those reported in the literature.8, 23

The most important determinant of prognosis is the impact of anti-viral treatment on fibrosis. We compared the histological course of the study population with an untreated historic control group. We found that the percentage of patients with histological improvement within 12 months was significantly higher in the study population when compared to the untreated controls.

As in the non-transplant setting, clinical and histological improvement may occur even when viral clearance is not achieved. On the other hand, even in the presence of SVR, necroinflammation may persist and fibrosis may remain unchanged or worsen in a subgroup of patients. A recently published study including 29 patients achieving SVR reported progression of fibrosis in one-third of patients at 2 years.24 In another study, sustained responders had a significant improvement in their liver inflammatory activity score, but not in their fibrosis score.25 Furthermore, several studies reported that inflammation and fibrosis improved in viral non-responders.26, 27

In our study, four of seven (57%) patients with SVR had improvement in their inflammatory activity, but only two (29%) of them had an improved or at least unchanged fibrosis score 12 months after the onset of therapy. The latter had EVR with no detectable HCV-RNA in the quantitative HCV-RNA assay after 3 months and concomitant EOTBR. Patients without SVR had declined inflammation grade in only 37% (seven of 19) but an improved/maintained fibrosis score in 47% (nine of 19). Thus, our data and the results from previous studies suggest that the antifibrotic effects of IFN may be independent of its anti-viral activity.27 The lacking association between sustained viral clearance and histological response might also be explained either by (i) other mechanisms of hepatic injury (e.g. calcineurin inhibitor (CNI)- or autoimmune-related toxicity and biliary complications) or (ii) a longer time period until fibrosis regression occurs in the allograft. Patients with failure to eradicate HCV but histological response in serial biopsies may though benefit from IFN therapy. Therefore, we offer this subgroup of patients a low-dose maintenance PEG-IFN monotherapy upon completion of this study. To date, little information has been published about the histological outcome and viral clearance in the long-term after IFN therapy. In our study, we followed up five of seven SVR patients. All these patients remained HCV-RNA-negative and none developed cirrhosis within a median post-treatment follow-up period of 37 months. Serial liver biopsies revealed an unchanged fibrosis and inflammatory activity score in all patients except one who had progression of fibrosis by 1 point.

Treatment of recurrent HCV infection is complicated by the lower patient tolerability necessitating frequent dose adjustment, adjuvant growth factors or treatment discontinuation.28 A remarkable part of our patients developed constellations of depression-related symptoms (irritability, sleep disturbances, fatigue, loss of appetite and anxiety) without also reporting significant sadness, hopelessness, guilt or loss of pleasure in life. The reasons for these missing findings can only be speculative and may include the initial dosing regimen for standard IFN (daily low-dose instead of thrice weekly), the exclusion of vulnerable psychiatric patients and meticulous provision of potential coping strategies and support systems. During therapy, IFN dose reduction was mandatory in 45% and 40% of the patients in the standard IFN and the PEG-IFN groups, respectively. In our study, withdrawal from therapy became essential in 31% of patients due to severe side effects. Published standard IFN/RBV or PEG-IFN/RBV combination studies report discontinuation rates between 18% and 63% compared to about 10% in immunocompetent patients.8, 29 Toniutto et al.16 found that treatment with PEG-IFN was associated more frequently with dosage reduction and withdrawal of therapy when compared to unmodified IFN (92% vs. 50%). The differences were mainly due to haematological side effects. However, in our study, as in most non-transplant studies, the adverse event profile of unmodified IFN was similar to that of PEG-IFN.30, 31 None of our patients discontinued prematurely because of haematological side effects as cytopenia could be successfully treated with EPO and/or G-CSF. Treatment with growth factors increases the costs and complexicity of therapy but seems to be inevitable in the LT setting to improve treatment adherence rates, and thus an increase in the anti-viral therapeutic efficacy.

The identification of pretreatment variables for the prediction of anti-viral response may help to develop patient selection algorithms to (i) identify patients who may have benefit from therapy and (ii) avoid inefficient treatment. Thus, possible recipient-, host- and virus-related predictive factors were analysed in this study. In non-immune compromised patients, increasing evidence has shown that HCV genotype, low viral load, adherence to therapy and combined IFN/RBV therapy are the most important factors influencing the success of anti-viral therapy.32–35 In our study, the decrease of viral load ≥2 log at 1 month but not the baseline HCV-RNA titre was associated with EOTVR. Conflicting results do exist in the literature about the predictive value of baseline HCV on viral response in transplant recipients.17, 18, 27, 36 Therefore, larger trials need to be conducted to further elucidate the influence of baseline HCV-RNA on viral response.

In our study, HCV genotype non-1 was identified as a predictive factor for EOTVR as well as SVR to IFN treatment. This was reflected by a sustained response rate of only 21% in patients with genotype 1 vs. 71% in those with other genotypes. Several studies in the non-transplant setting have demonstrated that HCV genotype 1 correlated with high viral titres in serum.32, 37 In our study, the baseline viral load was higher in genotype 1 hepatitis C-infected patients when compared to those with other genotypes; this difference, however, was not statistically significant.

There is ongoing discussion if the type of calcineurin inhibitor impacts the outcome of therapy. Interestingly, a recently published Spanish group found an almost significantly better effect of TAC on SVR when compared to CSA.36 Other studies have reported that CSA suppresses HCV in cell cultures;38, 39 however, it has been shown that the results from in vitro experiments can not to be necessarily transferred into the clinical setting.40, 41 In a previously published report from our transplant centre42 and in this present study, no significant differences between TAC and CSA monotherapy with regard to post-transplant viral load and to viral response upon IFN treatment became evident. Our findings are concordant with results from recent studies by Berenguer et al. and Murkherjee et al.27, 43

The optimal time point of initiation of anti-viral therapy after LT has not been determined yet. A recently published study including patients after at least 12 months of LT reported the best outcome upon 2–4 years after LT.36 Our study consisted of patients who received anti-viral therapy after a median time of 2.3 years (range: 4.3 months to 9.3 years) after LT. We found a negative correlation between the length of time between LT and the sustained response to IFN therapy.

In this study, half of the patients experienced a biochemical response to anti-viral therapy. Baseline ALT values were observed to be more than two times the UNL in 58% of patients (range: normal to 12 times the UNL) and, as among immunocompetent patients,6, 44 the presence of high ALT values was predictive of SVR by univariate and multivariate regression analysis.

In immune competent patients, it has been shown that a high BMI may be associated with lower biological response rates to IFN.45–47 This is in line with our data demonstrating that patients with normal BMI (<25 kg/m2) had a greater likelihood of EOTBR and EOTHR. Pharmacological effect of IFN on the induction of 2,5′-oligoadenylatesynthetase, a key enzyme in the IFN-elicited anti-viral response, has been shown to be reduced in obese patients;48 moreover, initial absorption of subcutaneously administered IFN may be impaired in high abdominal fat mass thus contributing to low response rates in this patient group.

In conclusion, based on the results of this prospective trial, daily standard IFN/RBV as well as daily standard IFN/RBV followed by PEG-IFN/RBV therapy are comparable in their safety and efficacy profiles. Furthermore, this study shows a histological benefit in patients with anti-viral treatment when compared to untreated controls. Growth factors should be used more amply to ensure adequate anti-viral drug dosing. Histological response rather than SVR may be considered as primary end point of IFN treatment. Serial biopsies under therapy and prolonged histological follow-up investigations are necessary to determine the progression of fibrosis following post-transplant anti-viral therapy. Recommendations regarding appropriate patient selection in the LT setting remain problematic. Several pretreatment variables have been associated with response or non-response to anti-viral treatment. These parameters may facilitate in providing advice to patients and in the estimation of the risk-benefit of anti-viral therapy. However, variables associated with viral response still vary between studies and yet, none of these factors has proven to be sufficiently reliable to allow exclusion of a transplant patient from IFN therapy.

Acknowledgements

Authors’ declaration of personal interests: None. Declaration of funding interests: This study was funded by Essex Pharma GmbH, Munich, Germany (PO4203), ID number 00-164-1519Z.

Ancillary