Hepatitis C Recurrence After Liver Transplantation: Viral and Histologic Response to Full-Dose Peg-Interferon and Ribavirin

Authors


*Corresponding author: R. Barcena, rbarcena.hrc@salud.madrid.org

Abstract

Hepatitis C recurrence after liver transplantation (LT) is universal, and frequently leads to cirrhosis and death. The aim of our study was to assess the efficacy and safety of 48-weeks of full-dose peg-interferon-α-2a (n = 4) or α-2b (n = 51) plus ribavirin (>11 mg/kg/day) in a multicentric cohort of 55 patients ≥12 months after LT. All subjects had histologically proven HCV recurrence, excluding severe cholestatic recurrence. Mean age was 54.3 ± 9.7, 77% male, 90.9% genotype 1, 32.7% cirrhotics. All but 5 patients received monotherapy with tacrolimus (54.5%), cyclosporine (30.7%) or mycophenolate mofetil (5.5%). The rates of end-of-treatment response and sustained virological response (SVR) were 66.7% and 43.6%, respectively. Low baseline HCV-RNA (p = 0.005) and a length from LT to therapy between 2–4 years (p = 0.011) were predictors of SVR. The lack of achieving a viral load decrease ≥1-log10 at week 4 and/or 2-log10 at week 12 was 100% predictive of failure. The most frequent side effects were neutropenia (76,4%), anemia (60%) and infectious complications (30.9%). Toxicity led to peg-interferon withdrawal in 16 (29%) subjects. In 15 patients with post-treatment biopsy, the histological activity index was significantly improved (p = 0.006), whereas fibrosis did not change (p = 0.14). Three patients died (cholangitis, hepatic artery thrombosis and lung cancer). In conclusion, HCV therapy after LT was very effective, although it led to a significant rate of toxicity.

Introduction

Hepatitis C virus (HCV) is the most common etiology of liver transplantation (LT) due to cirrhosis and hepatocellular carcinoma. In western countries, it represents up to 50% of LT. Most patients have evidence of viral replication at the time of surgery and, therefore, hepatitis recurrence is almost universal, leading to a worse prognosis when compared to other indications of LT. In a large cohort of 11 036 LT patients, survival was significantly decreased among those with chronic hepatitis C: 86.4%, 77.8% and 69.9% at 1, 3 and 5 years versus 87.5%, 81.8% and 76.6%, respectively (p < 0.0001) (1).

HCV recurrence after LT shows an accelerated evolution to cirrhosis, present in 20–40% of patients after 5 years (2–7). When compared to immunocompetent patients, the rate of fibrosis progression per year is much higher (4), and progression to cirrhosis is significantly shortened (8–10). Therefore, hepatitis C recurrence and its management seem critical for its impact on the outcome of HCV infected patients after transplantation.

Treatment of CHC with interferon (IFN) monotherapy after LT yielded poor results, and acute rejection was reported in more than 30% of patients (11). The addition of ribavirin increased the rate of sustained viral response (SVR) with a lower incidence of rejection (12). However, most reports have been based on low number of patients and low doses of both drugs. The aim of the present study was to describe the clinical and virological outcome after combined therapy with full-dose pegylated interferon (peg-IFN) and ribavirin in a cohort of 55 LT patients attended at two tertiary centers in Madrid, Spain.

Patients and Methods

Patients

Patients were included from November 2001 (first availability of peg-IFN-α-2b) to April 2004, from specialized Liver Transplantation Outpatient Clinics at two tertiary centers in Madrid (Spain): Ramon y Cajal Hospital and Puerta de Hierro Clinic.

The following inclusion criteria were considered: LT subjects from 18 to 70 years, with histological evidence of HCV recurrence and after at least 12 months of surgery. Patients should have maintained the same immunosuppressive regimen in the 3 months before HCV treatment, and steroids withdrawal for at least 6 months was required. Diagnosis of HCV recurrence was based on elevated serum alanine aminotransferase levels for at least 6 months, detectable serum HCV-RNA by PCR (COBAS HCV Amplification and Detection version 2.0, Roche Diagnostic, sensitivity 60 IU/mL), and a liver biopsy showing fibrosis >1 using a modified Ishak score (13) or lobular hepatitis. Patients with early acute severe cholestatic recurrence, human immunodeficiency virus coinfection, creatinin levels above 2.2 mg/dL, severe cardio-respiratory disease, other allograft disfunctions, hemoglobin levels below 10 g/dL, neutrophil counts below 1200 cells/μL, and platelet counts below 50 000/μL were excluded. All subjects were informed about the study and signed an informed consent. The study was approved by the ethical committee. In the present study, we show the results of the first 55 treated patients under those inclusion criteria, until April 2004 to achieve 18 months of follow-up in all of them. Of note, all the subjects treated in both centers with those criteria have been included in this study.

Treatment and follow-up

Patients were initially treated with peg-IFN-α-2a (180 μg/week) or peg-IFN-α-2b (1.5 μg/kg/week) plus weight-adjusted ribavirin (800–1200 mg/day, with at least 11 mg/kg/day). Therapy was planned for 48 weeks in genotypes 1 or 4, and 24 weeks in genotypes 2 or 3. All subjects were followed during therapy and for 18 months since therapy initiation, in order to obtain the percentage of SVR and the overall safety. The early stopping rules according to viral decrease at 12 and 24 weeks accepted for immunocompetent patients (14) were not considered in our study due to the assumed potential different behavior of HCV-RNA among LT patients. Therapy was discontinued in case of significant side effects or in patients with HCV-RNA decreases below 0.5 log10 at week 12 or 1.5 log10 at week 24, because it was considered non-ethical to continue therapy in patients with evidence of non-response. Those criteria have been used in this study to define therapy withdrawal due to non-response.

The doses of peg-IFN and ribavirin were adjusted during therapy in case of significant cytopenias, and hematopoietic growth factors were used according to physician's criteria. Severe anemia was considered for hemoglobin levels below 9 g/dL or when erythropoietin, transfusion or dose reduction of ribavirin were needed. Ribavirin doses were not initially adjusted to renal function as we only considered creatinine below 2.2 mg/dL. Neutropenia was considered for neutrophil counts below 750 cells/μL; filgrastim was used for neutrophil counts below 500 cells/μL.

Viral assays

HCV infection was diagnosed by positive HCV antibodies and positive serum HCV-RNA by the qualitative PCR assay reported above. HCV-RNA was determined at baseline, and at weeks 4, 12, 24, 48 and 72. Quantitative HCV-RNA (COBAS Amplicor MONITOR™ Roche Diagnostic, limit of detection 600 IU/mL) was performed during the first 24 weeks and whenever the qualitative test was positive. The logarithmic decrease from baseline HCV-RNA was calculated at weeks 4, 12 and 24. HCV genotypes were determined by reverse hybridization using the Inno-Lipa HCV (Innogenetics, Ghent, Belgium).

Liver allograft histology

All patients had a first liver biopsy before treatment assessed by a single experienced liver pathologist as defined previously (13). Cirrhosis was defined as fibrosis 5–6 out of 6. As this was a clinical study, post-treatment liver biopsies were not mandatory, and were performed when indicated by the physician and accepted by the patient. Post-treatment biopsies were performed only at one of the centers (Ramón y Cajal Hospital). All paired assessments were blindly performed by the same experienced pathologist who was not aware of patient's name or the status pre- or post-treatment.

Statistical analysis

A database was created including age, sex, length from LT to therapy initiation, previous IFN-based therapies, genotype, body mass index, histological features (and cirrhosis or not), immunosuppressive therapy, HCV-RNA evolution during therapy, adverse events, dose adjustments, or use of erythropoietin and/or filgrastim. SVR was considered for patients with a negative qualitative HCV-RNA assay 24 weeks after the end of therapy. End-of-treatment response was considered when the qualitative HCV-RNA PCR assay was negative at the end of therapy.

Statistical analyses were performed with the SPSS (Statistical Package of Services Solutions, SPSS Inc., Chicago, IL) software, version 12.0. To assess baseline predictors of SVR, the following variables were considered: sex, age, body mass index, baseline HCV-RNA, immunosuppressive regimen, fibrosis score, cirrhosis or not, and length in months from LT to treatment. This latter variable was also considered as a categoric variable divided into ‘less than 2 years’, ‘from 2 to 4 years’, and ‘more than 4 years’. Genotype was not included due to the low population with genotypes 2 or 3 (n = 2). Univariate analyses were performed using the Mann-Whitney-U-test for continuous variables whereas categorical variables were evaluated by the Pearson chi-square test. A p-value <0.05 was considered significant. A multivariate logistic regression analysis was performed including those variables statistically significant by univariate analyses and those considered clinically significant by the investigators.

To assess the ‘on therapy’ predictors of SVR, there were also calculated the negative predictive value (NPV) and positive predictive value (PPV) of early HCV-RNA decreases on SVR: decrease of at least 1 log10 at week 4 (D1L4W) and decrease of at least 2 log10 at week 12 (D2L12W).

In the group with paired biopsies, histological activity index and fibrosis scores prior to and after treatment were compared by non-parametric test (Wilcoxon test for paired samples).

Results

Patients features

From 171 alive patients transplanted due to HCV in both centers, 55 subjects (42 men and 13 women) until April 2004, fulfilled inclusion criteria and accepted the therapy, after being informed about the possible risks of treatment post-transplantation.

Baseline features are summarized in Table 1. No patient had previously received peg-IFN and ribavirin, but 6 patients were non-responders to standard IFN monotherapy pre-transplantation. Most patients (90.9%) were genotype 1. The majority (n = 51) received peg-IFN-α-2b for its earlier availability during the enrolment period.

Table 1.  Baseline features (N = 55)
  1. M = male; F = female; SD = standard deviation; IFN = Interferon; peg-IFN = pegylated interferon; HCV = hepatitis C virus; ALT = alanine aminotransferase; LT = liver transplantation.

Sex (M:F)42/13
Age (years, mean ± SD)54.3 ± 9.7
Body mass index (mean ± SD)25.9 ± 3.4
Type of immunosuppression
 Tacrolimus (monotherapy)30 (54.5%)
 Cyclosporine (monotherapy)17 (30.9%)
 Mycophenolate mofetil (monotherapy)3 (5.5%)
 Tacrolimus and mycophenolate4 (7.2%)
 Cyclosporine and mycophenolate1(1.8%)
Type of patient
 Naïve49(89.1%)
 Non-responder to IFN-α monotherapy6 (10.9%)
Type of peg-IFN α2a/2b4/51
Mean HCV RNA (IU/mL)997 780 ± 758 579
 Low viral load (<800 000 IU/mL)26 (48.3%)
 High viral load (≥800 000 IU/mL)29 (52.7%)
Genotype
 150 (90.9%)
  1a5
  1b44
  1a + 1b1
 32 (3.6%)
 42 (3.6%)
 Non-available1 (1.8%)
ALT (mean ± SD) IU/L185.9 ± 115.8
Months from LT to treatment (mean ± SD)63.3 ± 45.5
Fibrosis Ishak score (mean ± SD)3.6 ± 2.0
Cirrhosis (n, %)18 (32.7%)

Immunosuppressive monotherapy was used in 50 subjects (91%); the remaining 5 subjects were on tacrolimus plus mycophenolate mofetil (n = 4), and cyclosporine plus mycophenolate (n = 1). Eighteen patients were cirrhotics, most of them (n = 16) in class A of the Child-Pugh score, with only two in class B. Two patients developed ascites after LT, one of them also variceal bleeding.

Two patients with hypersplenism-related severe thrombocytopenia underwent successful partial splenic embolization to improve platelet count enough prior to HCV treatment (15).

Viral response

Negative HCV-RNA at weeks 4, 12, 24 and 48 was observed in 16 (29.1%), 31 (56.4%), 34 (63.0%) and 36 (66.7%) of the patients, respectively. Among the 36 subjects with end-of-treatment response, 11 had HCV-RNA recurrence at week 72, and one died due to arterial thrombosis. Therefore, the overall rate of SVR was 43.6% (n = 24). When considering only naïve subjects (n = 49), SVR was achieved in 22 (44.9%). Two out of 6 (33%) non-responders to standard IFN monotherapy before LT, obtained SVR. Among genotype 1 subjects, 20 (40%) achieved SVR. The two patients with genotype 3 reached SVR, one of them after only 3 months of treatment that was prematurely stopped due to major depression. The two genotype 4 patients achieved end-of-treatment response, but one of them relapsed 24 weeks after finishing treatment.

Histological benefit

There were 15 cases with paired liver biopsies before treatment and after follow-up. In those cases (Table 2), it was observed a statistically significant improvement in the histological activity index, from 7.3 ± 1.6 before therapy to 4.2 ± 3.1 after treatment (p = 0.006). There were no differences in the mean score of fibrosis (from 2.5 ± 1.8 to 3.1 ± 1.5, p = 0.14). The improvement in the histological activity index was statistically significant only among sustained responders (p = 0.02), whereas the fibrosis score did not significantly change according to viral response.

Table 2.  Histological evolution before treatment and after follow-up
 HAI-1 Mean ± SDHAI-2 Mean ± SDFibrosis-1 Mean ± SDFibrosis-2 Mean ± SD
  1. HAI-1 and -2 = histological activity index before and after treatment, respectively; SD = standard deviation; Fibrosis-1 and -2 = Ishak fibrosis score before and after treatment, respectively; SVR = sustained viral response.

  2. 1p < 0.05.

SVR7.5 ± 2.13.3 ± 2.82.4 ± 1.92.6 ± 1.3
(n = 8) p = 0.021 p = 0.7
Non-SVR7.1 ± 1.15.3 ± 3.22.7 ± 1.73.7 ± 1.6
(n = 7) p = 0.2 p = 0.1
All7.3 ± 1.64.2 ± 3.12.5 ± 1.83.1 ± 1.5
(n = 15) p = 0.0061 p = 0.14

Side effects

Side effects are shown in Table 3 and withdrawal moment and reasons are reported in Figure 1. The most frequent clinical adverse event was fatigue (89.1%), followed by weight loss (n = 27, 49.1%). Seventeen patients (30.9%) developed infectious complications: urinary tract (n = 3), cholangitis (n = 3), skin (n = 3), gastroenteritis (n = 2), ear-nose-throat infections (n = 3), and bacteremia, pneumonia and liver abscesses due to arterial thrombosis in one case each. The mean time from LT to treatment in patients with infectious complications was 61.2 months. Six of these patients had to stop treatment due to the infectious event and two of them achieved SVR with only 12 and 24 weeks of therapy, respectively. The rate of infection was higher in cirrhotics than in patients with lower degrees of histological severity, but it did not reach statistical significance (44% vs. 24%, p = 0.13). Similarly, the rate of infectious complications was not affected by the presence or not of severe neutropenia (35% vs. 28%, p = 0.62), and we did not found a temporal relationship between neutropenia and infection.

Table 3.  Adverse events during treatment
Side effectsN (%)
Asthenia49 (89.1%)
Leukopenia42 (76.4%)
Anaemia (Hb < 9 g/dL)33 (60%)
Thrombocytopenia31 (56.4%)
Weight loss27 (49.1%)
Infection17 (30.9%)
Depressive symptoms15 (27.3%)
Hyperthyroidism1 (1.8%)
Hypothyroidism1 (1.8%)
Chronic rejection1 (1.8%)
Hepatic artery thrombosis1 (1.8%)
Figure 1.

Therapy withdrawal. From a total of 55 treated patients, we show the number of patients who discontinued treatment at each time, the cause and the ones who showed sustained viral response (SVR) after therapy withdrawal at each moment.

There were three deaths during follow-up. One patient developed severe cholangitis at week 32. Therapy was immediately stopped, but bacteremia led to death 1 week later. This patient did not have a bilio-digestive anastomosis, but had developed biliary stenosis after LT treated with a plastic stent. He had not presented any other biliary complication in the previous 7 years. Another patient developed hepatic arterial thrombosis at week 44. This subject had no previous evidence of arterial stenosis or thrombosis by Doppler ultrasound. Despite therapy withdrawal (with end-of-treatment response) and inclusion in the waiting list for re-transplantation, she died 2 months later. She had not presented any vascular complication in the 3 years following LT. The third patient, non-responder, was diagnosed of bronchogenic lung cancer probably related to long-time smoking, and died at month 17 of follow-up.

There were no cases of acute rejection. One subject developed chronic rejection at week 40 of treatment; despite liver biopsy did not confirm diagnosis, he responded to the addition of rapamicin and peg-IFN withdrawal. Other graft disfunctions were excluded and diagnosis was established based on clinical data. He reached SVR and chronic rejection was successfully controlled. Of note, this patient had to stop ribavirin 4 months before the episode of rejection due to anemia, and was only on peg-IFN at the time of the event.

Thirty-three (60%) patients developed severe anemia, and 21 (38.2%) received erythropoietin. Only 2 patients had basal creatinine levels above 1.7 mg/dL, and both developed severe anemia. One of them needed erythropoietin and he obtained SVR. Forty-two (76.4%) presented neutropenia, and 8 (14.5%) needed filgrastim.

In 6 cases, therapy was stopped due to ‘non-response’ as defined above. In 16 patients (29%) peg-IFN was prematurely discontinued due to side effects: infection (n = 6), poor tolerability (n = 6), depression (n = 2), chronic rejection (n = 1), and cytopenias (n = 1). The moment of treatment discontinuation has been shown in Figure 1.

Seven subjects required peg-IFN dose adjustments, due to neutropenia (n = 5), or gastrointestinal intolerance (n = 2). Isolated ribavirin withdrawal was needed only in one patient, due to severe anemia at week 24. In 17 patients, ribavirin dose was adjusted within the first 12 weeks, and later in 9 subjects.

Baseline predictors of SVR

By univariate analyses, a low baseline HCV-RNA and a length between LT and therapy between 2–4 years were statistically associated to SVR (Table 4). The type of immunosuppressive regimen—tacrolimus vs. cyclosporine—was also included in the multivariate analysis (Table 4), for univariate analysis showed a trend to a better outcome in patients on tacrolimus (p = 0.053). None of the rest of baseline features had a statistically significant association with SVR.

Table 4.  Statistical analyses of baseline predictive factors of SVR
VariablesSVRNon-SVRUnivariate analyses p valueMultivariate model (logistic regression)
p valueOdds ratioConfidence interval (95%)
  1. LT = liver transplantation; SVR = sustained viral response; CsA = cyclosporine.

  2. 1p < 0.05.

Sex (male/female)19/523/80.67 
Age (years)55.3 ± 8.953.5 ± 10.30.57 
Body mass index26.3 ± 3.725.8 ± 3.30.73 
Baseline HCV-RNA (log 10 IU/mL)5.6 ± 0.56.0 ± 0.30.00310.00510.0520.007–0.401
Fibrosis (1–6)3.7 ± 2.03.5 ± 2.10.75 
Cirrhosis (No vs. Yes)16/821/100.93 
Months from LT52.7 ± 31.971.5 ± 52.80.46 
<2 years after LT n = 10 patients370.48 
2–4 years after LT n = 17 patients1160.03510.01116.6241.53–28.70
>4 years after LT n = 28 patients10180.22 
Immunosuppression (Tacrolimus vs. CsA)19/515/130.0530.120.290.06–1.40

Logistic regression multivariate analysis included baseline HCV-RNA, a length between LT and therapy between 2 and 4 years, and the use of tacrolimus vs. cyclosporine. Again, only a lower baseline HCV-RNA and a length between LT and therapy from 2–4 years remained statistically associated to SVR. Patients starting therapy after 2–4 years of LT responded more frequently than the overall group starting before 2 years or more than 4 years after surgery. SVR was achieved in 30%, 64.7% and 35.7% in <2 years, 2–4 years and >4 years, respectively.

‘On therapy’ predictive factors

Table 5 shows the PPV and NPV of HCV-RNA decreases at week 4 (1 log10) and week 12 (2 log10) on SVR. None of 13 patients with HCV-RNA decreases <1 log10 at week 4, and none of 9 with less than 2 log10 decrease at week 12 achieved SVR, yielding a NPV of 100% in both instances. On the other hand, all patients who reached negative HCV-RNA at week 4 (n = 16) obtained SVR (PPV = 100%). The PPV of negative HCV RNA at week 12 on SVR was 70.9%.

Table 5.  PPV and NPV of early HCV-RNA decreases on SVR Thumbnail image of

Among 20 subjects with positive HCV-RNA at week 24, only one achieved SVR, but his HCV-RNA decrease at week 24 was >3 log10.

Discussion

To our knowledge, this is the largest series on the outcomes of peg-IFN plus ribavirin in LT patients after at least 12 months of surgery. The main finding of this study is that the use of full-dose combined therapy in these patients yielded a rate of SVR similar to that observed among immunocompetent patients (16). This more than acceptable outcome might just be related to the length between LT and therapy (mean 5.25 years), with its associated better clinical situation, lower levels of immunosuppression and less significant presence of cytopenias. It should be remarked the histological severity observed in the cohort, with a mean fibrosis score of 3.6 + 2 and cirrhosis in 32.7%, as well as the high prevalence of genotype 1 (90.9%). We think that this population should be considered a representative sample of the wide population of HCV patients who have received a liver graft in the previous years. This ‘delayed approach’ and good baseline situation allowed the use of full doses therapy during the first 12 weeks, which has been shown to be an important predictor of SVR in immunocompetent patients (17). The observed rate of SVR over 40% is encouraging, even taking into account the higher risk of adverse effects and the possibility of rejection.

Previous studies with initially low and increasing dose regimen of standard IFN and ribavirin (18–24) produced SVR rates ranging from 8 to 60%, but the trials were heterogeneous and with low number of patients. We have previously reported the preliminary results of our strategy including the first 21 patients of this series (25).

Three different approaches to HCV recurrence therapy have been proposed in LT patients. ‘Preemptive therapy’ (22,26–28) aimed to prevent HCV recurrence early after LT is not currently being considered, due to low effectivity and the low proportion of ‘eligible’ population. The other two strategies are therapy during the acute phase of hepatitis C recurrence (29) or, as we propose, ‘a delayed approach’ waiting until chronic hepatitis C is established.

Early after transplantation viral load is high, related to more intensive immunosuppression and the concomitant use of steroids. In that situation, Castells et al. (29) reported very good results, with a SVR rate of 34.7%. However, the rate of eligibility (56%) to start therapy in the initial months after surgery makes difficult to generalize treatment for many patients. In the literature there are also some studies (18,25,30) which do not consider separately patients receiving therapy in the early vs. tardive phase of HCV recurrence. Although we have also treated patients with minimal fibrosis, if therapy is delayed until established chronic hepatitis, the prognostic criteria of hepatitis severity itself could be applied, and the decision to treat would not be based only on clinical eligibility. In our study, which only included patients after at least 12 months of LT, the optimal time frame with the best outcome was the period between 2 and 4 years after LT, perhaps due to a good balance between a lower immunosuppression with lower fibrosis.

The behavior of HCV-RNA during peg-IFN plus ribavirin in LT patients is not well known, but ours and some previous data suggest a kinetic pattern similar to immunocompetent patients, in contrast to observed slower clearance observed among other immunocompromised patients, such as human immunodeficiency virus coinfected patients (31).

As previously reported (21,25,30,32), combined therapy in LT patients is associated with a higher frequency and severity of adverse events, with the frequent need of erythropoietin and/or filgrastim. Our rate of treatment discontinuation due to adverse events was 29%, compared to about 10% among immunocompetent subjects (15). There were three episodes of cholangitis, which led to therapy withdrawal, and 1 patient died. The rate of infectious complications has been very high reaching 44% in cirrhotic patients. One adverse event, which might have also been related to IFN was hepatic arterial thrombosis, that led to death before completing therapy, in a patient with no previous evidence of vascular complications. Vascular disease has rarely been reported as an IFN-related adverse event (33–35), but in these patients, hepatic arterial thrombosis is of concern, for the frequent evolution to graft failure and death or re-transplantation. To our knowledge, only hepatic arterial stenosis without thrombosis has been reported before (36).

There were no cases of acute rejection, but 1 patient developed chronic rejection that could be successfully managed. This complication may also evolve to graft loss, and has been well documented among renal transplant patients on IFN (37,38). It is worth noting that our patient had stopped ribavirin before the episode, suggesting a potential immunomodulatory beneficial effect of ribavirin on IFN-induced rejection (39). In a recent report (40), Stravitz et al. observed a 35% of acute or chronic rejection. Only four of their patients tolerated ribavirin, so the use of IFN monotherapy in most subjects may have led to the high rate of rejection. Therefore, some data from literature and this study seem to support the strategy of maintaining ribavirin for the whole course of therapy, to minimize the risk of rejection.

In the frequent case of severe anemia, ribavirin doses may be decreased and/or erythropoietin should be used.

The high rate of adverse events in this population makes mandatory to find early markers of non-response, both before and during treatment, to avoid unnecessary treatment in as many non-responders as possible while preventing premature discontinuation in those who might achieve SVR. As among immunocompetent patients (14), our study also confirms for the first time the usefulness of early viral kinetics in LT patients. The lack of D1L4W or D2L12W had a NPV of 100% on SVR, and might have allowed the avoidance of severe complications and the prolongation of unsuccessful treatment. The lack of D1L4W was observed in 13 (23.6%) patients who therefore could have stopped therapy as early as at week 4, and might have prevented 3 episodes of infection, 4 of severe depressive symptoms, 1 of hypothyroidism, and 9 each of neutropenia and anemia, developed by these patients during treatment. The lack of D2L12W was observed in 9 patients, all of them lacking also D1L4W, so perhaps it could not be as useful as D1L4W in this group of patients.

On the other hand, viral clearance at week 4 was associated to SVR in all subjects, so in this instance it is critical to maintain full-dose therapy using hematopoietic growth factors if needed. Only 1 patient with positive HCV-RNA at week 24 but a 3 log10 decrease from baseline achieved SVR. Therefore, the decision of IFN discontinuation in this rare situation should be individualized.

There were 15 patients with paired biopsies, and 8 reached SVR. Overall, there was a significant improvement in the histological activity index, but not in the fibrosis degree, that might be related to the relatively short period of follow-up (18 months), suggesting that fibrosis improvement could be a delayed phenomenon. In post-treatment biopsies, the evolution of chronic hepatitis is 18 months longer than in pre-treatment biopsies, and that is a long time for a post-transplantation hepatitis C. This could be the reason of a non-significant tendency to increase fibrosis immediately after follow-up. A recent study with a mean follow-up of 52 months (41) showed that among LT patients, SVR led to a statistically significant improvement in fibrosis score when compared to non-SVR. In 38% of their patients, fibrosis improved in relation to pre-treatment biopsy. One study (42) with IFN-α-2b and ribavirin detected a significant amelioration from baseline fibrosis, effect which was not observed by other authors (43).

Another remarkable finding was the almost significantly better effect of tacrolimus on SVR when compared to cyclosporine. In several articles (44–46) cyclosporine was reported to produce in vitro suppression of viral replication driven by blockade of cyclophilins, effect not proved for tacrolimus. The type of immunosuppression, however, does not seem to clinically affect the evolution of CHC after transplantation (5,47–52). A better response to IFN has been suggested recently with cyclosporine compared to tacrolimus (46). In our study the group of tacrolimus had a significantly shorter period since LT, so both samples were not comparable, as the length from LT to therapy was statistically significant both by univariate and multivariate analyses. Further studies are needed to analyze this finding.

In conclusion, full-dose peg-IFN plus ribavirin seems to be quite effective in LT patients, with rates of SVR similar to immunocompetent subjects. However, the high risk of hematological toxicity and infectious complications makes essential a strict monitoring of these patients. A HCV-RNA decrease at week 4 below 1 log10 seems to be a useful tool to detect non-responders, allowing early withdrawal and the avoidance of serious adverse events. In our study, the optimal period to start therapy was between 2 and 4 years after LT.

Acknowledgments

We would like to thank Miguel Angel Rodriguez Sagredo and Teresa Bermejo Vicedo from Pharmacy Department in Ramon y Cajal Hospital for their collaboration in this study. This study was supported in part by Red Nacional de Investigación en Gastroenterología y Hepatología and by Grant from the Investigation Department of Ramon y Cajal Hospital.

Ancillary