Long-term histological effects of preemptive antiviral therapy in liver transplant recipients with hepatitis C virus infection

Authors


Abstract

The long-term effects of preemptive antiviral therapy on fibrosis progression in liver transplant recipients with hepatitis C virus (HCV) were examined in a cohort of consecutive liver transplant recipients who received preemptive antiviral therapy for 48 weeks (95% were virologic nonresponders). Control patients were transplanted during this same period but did not receive preemptive therapy. Patients were followed to the date of last biopsy and censored at the time of subsequent HCV treatment. Eighty-six patients surviving ≥90 days were included. Treated and control patients were similar, except that treated patients had longer histological follow-up (60 versus 50 months), a lower median Model for End-Stage Liver Disease score at liver transplant (17 versus 23), and a shorter median length of hospital stay (6 versus 9.5 days). In the uncensored analysis, the cumulative probability of a Batts-Ludwig fibrosis score ≥ 2 at 48 months post–liver transplant was 22% in the preemptive therapy group and 49% in the nonpreemptive therapy group (P = 0.08). In multivariate analysis, preemptive therapy was associated with a 48% reduced risk of a fibrosis score ≥ 2 (hazard ratio = 0.52, 95% confidence interval = 0.24-1.12, P = 0.09), but this failed to achieve statistical significance. Receipt of preemptive therapy was associated with a delay in subsequent HCV therapy for moderate to severe disease (fibrosis score ≥ 2 or moderate necroinflammatory activity) with a median time of 36.3 months versus 20.3 months in the preemptive and nonpreemptive groups (P = 0.004). We conclude that preemptive antiviral therapy in virologic nonresponders delays the time to subsequent HCV treatment and may confer a reduced risk of fibrosis progression. Further study of preemptive antiviral therapy is warranted. Liver Transpl 14:1491–1497, 2008. © 2008 AASLD.

Recurrent hepatitis C virus (HCV) infection after liver transplantation is a significant cause of graft loss in the United States and Europe, where HCV-associated cirrhosis in the most common indication for liver transplantation.1 In an effort to prevent graft loss from recurrent disease, treatment with interferon (IFN) and ribavirin is typically undertaken in those with progressive or severe disease. Achievement of a sustained virological response (SVR) is the goal of therapy as this is associated with histological benefits, including a reduction in necroinflammation and stabilization of fibrosis in most patients.2, 3 Unfortunately, SVR rates are less than desired, with only 23% to 45% of treated patients achieving this endpoint.4-8

Initiation of treatment in the early posttransplant period has been proposed as a potentially more effective means of preventing or ameliorating recurrent HCV disease. With preemptive therapy, as typically defined, antiviral therapy is usually begun within the first 2 to 8 weeks post-transplantation before there is histological evidence of recurrent disease. Our center evaluated this strategy in a randomized trial comparing IFN (nonpegylated or pegylated) monotherapy and IFN (nonpegylated or pegylated) combined with ribavirin.9 Treatment was initiated 2 to 6 weeks post-transplantation and continued for a total of 48 weeks. We found that only 41% of patients were eligible for antiviral therapy in the early posttransplant period, mainly because of persistent anemia or posttransplantation complications. Treatment was also poorly tolerated, with only 23% receiving full-dose IFN and only 9% receiving full-dose IFN with ribavirin. Moreover, efficacy was poor, with SVR achieved in only 9% of patients. On the basis of the results of this study, we recommended against the use of preemptive antiviral therapy. Other preemptive studies have shown similarly low rates of viral clearance.10, 11

Although preemptive antiviral therapy does not enhance rates of SVR, the question of whether such treatment may provide histological benefits has not been previously addressed. Certainly, treatment regimens that attenuate fibrosis progression, even in the absence of viral eradication, would be a valuable weapon in the fight against graft losses due to recurrent HCV. In this study, we examined the long-term histological outcomes in transplant patients who received preemptive antiviral therapy.

Abbreviations

ALT, alanine aminotransferase; CMV, cytomegalovirus; HCV, hepatitis C virus; IFN, interferon; LOS, length of hospital stay; LT, liver transplantation; MELD, Model for End-Stage Liver Disease; SVR, sustained virological response.

PATIENTS AND METHODS

Study Population

This is a retrospective analysis of a prospective cohort of adult patients who underwent liver transplantation for HCV-associated cirrhosis at the University of California–San Francisco from December 1998 through June 2002. Patients were included if they met all of the following criteria: ≥18 years of age; pretransplant diagnosis of HCV-associated cirrhosis; and clinical stability with white blood cell count > 3.0/mm3, hemoglobin > 10.0 g/dL, platelets > 45,000/mm3, creatinine < 1.5 mg/dL, international normalized ratio < 2.0, aspartate aminotransferase < 200 IU/L, alanine aminotransferase (ALT) < 200 IU/L, and total bilirubin < 2.5 mg/dL. Exclusions included serum anti–human immunodeficiency virus or hepatitis B surface antigen positivity prior to transplantation; acute rejection (by clinical and histological criteria) within 14 days of consent; vascular and biliary complications post-transplantation resulting in the need for interventions; current untreated infection; abnormal thyroid-stimulating hormone; medically uncontrolled psychosis or depression; a history of hemoglobinopathies or any cause of chronic hemolysis; clinically significant retinopathy; uncontrolled diabetes mellitus; a medical history of concomitant autoimmune disease; a continued need for intensive care unit support; and unstable cardiopulmonary status, including myocardial infarction within the preceding 4 weeks. The most frequent cause for exclusion from the randomized study was failure to meet the laboratory inclusion criteria.9

For the current study, the preemptive therapy group, which comprised all patients who received at least 1 dose of preemptive antiviral therapy, was compared with a control group termed the nonpreemptive group, which comprised all patients undergoing liver transplantation during this same period who did not meet the inclusion and exclusion criteria for preemptive therapy or who declined participation in the preemptive study. In addition, patients were excluded if they died within the first 3 months post-transplantation or did not have a liver biopsy available for examination during the follow-up period.

Use of Antiviral Therapy

In the preemptive treatment group, patients were randomized to receive either IFN monotherapy (IFN alfa-2b, 3 MU 3 times weekly, or peg-IFN alfa-2b, 1.5 μg/kg/week) or combination therapy with IFN and ribavirin (IFN alfa-2b or peg-IFN alfa-2b and ribavirin, 600 mg/day, which was increased to 1-1.2 g/day if tolerated) initiated between weeks 2 and 6 post-transplantation and continued for 48 weeks. Patients in the control group did not receive preemptive antiviral therapy.

In both groups, subsequent antiviral therapy was given for either severe or progressive recurrent HCV disease. The decision to treat was at the discretion of the transplant hepatologist and based on the presence of at least moderate fibrosis or necroinflammatory activity on biopsy.

Histological Evaluation

Liver biopsies were obtained annually or as clinically indicated. Refusals and missed biopsy appointments contributed to missing biopsy data. Liver biopsies were obtained from the patients in the preemptive study at the end of treatment and annually thereafter.

The fibrosis stage was assessed with the Batts-Ludwig scoring system (F0-F4).12 The grade of necroinflammation was assessed on a scale from 0 to 4. Biopsies that were scored as being between grades, or stages were rounded up (that is, a fibrosis score of 1-2 was assigned a value of 2 for this analysis). All biopsies were reviewed by a single transplant pathologist who was blinded to each patient's treatment group.

Immunosuppression Protocol

Patients were treated with a 3-drug regimen of tacrolimus, mycophenolate mofetil, and prednisone. Tacrolimus levels were kept between 8 and 12 ng/dL during the first 3 to 6 months and then at 6 to 8 ng/dL subsequently. Mycophenolate mofetil was given at a dose of 1 to 1.5 g/day initially and was tapered off within the first 6 to 12 months. Prednisone was started at 200 mg daily and tapered by 40 mg daily until 20 mg daily was reached. Prednisone was then tapered by 2.5 mg weekly until 5 mg daily was reached by week 6 post-transplantation. In some patients, prednisone was gradually tapered off by 12 months post-transplantation, but this was not a uniform practice. Sirolimus was used as an alternative to tacrolimus in patients with renal dysfunction. Antibody induction regimens were not routinely used.

Treatment of Acute Rejection

Patients with biopsy-proven mild or moderate acute cellular rejection were treated with 1 g of methylprednisolone intravenously on 2 consecutive days followed by 200 mg of prednisone daily, which was tapered by 40 mg daily down to 20 mg daily. Prednisone was then tapered by 2.5 mg weekly until 5 mg daily was reached. Severe acute rejection or acute rejection unresponsive to corticosteroids was treated daily with OKT3 (5 mg intravenously) for 7 to 10 days or Thymoglobulin® (0.5-1.0 mg/kg) for 5 to 7 days.

Statistical Analysis

Descriptive statistics included the median, mean, standard error, and range, as appropriate. Comparisons between groups were performed with the chi-square and Fisher exact tests for categorical variables and the Mann-Whitney test for continuous variables. Kaplan-Meier methods were used to estimate the time to a fibrosis score ≥ 2 in the preemptive group versus the nonpreemptive group, with comparison between the groups performed with the log-rank test. Patients in the preemptive group were censored at the start of retreatment with IFN-based therapy. Patients in the nonpreemptive group were censored at the start of IFN-based therapy. Patients achieving SVR were not excluded from the analysis. Cox regression was used to assess the relationship between receipt of preemptive antiviral therapy and a fibrosis score ≥ 2; adjustments were made for potential confounders, including donor age, treatment of acute cellular rejection, and cytomegalovirus infection. All statistical analyses were performed with Stata 8.0 (Stata Corp., College Station, TX).

RESULTS

Patient Characteristics

A total of 124 patients underwent liver transplantation for HCV-associated cirrhosis at the University of California–San Francisco during the study period. Eighteen patients were not considered candidates for preemptive therapy for the following reasons: hepatitis B surface antigen positivity, repeat liver transplantation, or lack of follow-up due to the patient not living in the region after transplantation. Of the remaining 106 patients, 11 patients were excluded from the current analysis because of a lack of a liver biopsy during the follow-up period, and 9 patients were excluded from analysis because of death within the first 3 months post-transplantation. The causes of death were neurologic complications (n = 2), recurrent hepatocellular carcinoma (n = 1), cardiac complications (n = 2), and sepsis/multiorgan failure (n = 4); this left a total of 86 patients. Of these 86 patients, 41 patients were in the preemptive group, and 45 patients were in the nonpreemptive group. Study groups were similar except for the following: the median duration of histological follow-up in the preemptive group was 10 months longer than that in the nonpreemptive group (60 versus 50 months respectively, P = 0.05); the median biological Model for End-Stage Liver Disease (MELD) score at transplantation in the preemptive group was lower than that in the nonpreemptive group (17 versus 23, P = 0.02); and the median length of hospital stay after transplantation in the preemptive group was lower than that in the nonpreemptive group (6 versus 9.5 days, P < 0.0001). The patient characteristics are detailed in Table 1.

Table 1. Patient Characteristics
 Preemptive Group (n = 41)Nonpreemptive Group (n = 45)P Value
  • Abbreviations: CMV, cytomegalovirus; LOS, length of hospital stay; MELD, Model for End-Stage Liver Disease.

  • *

    Includes 2 domino liver transplants.

Age (years), median, range50 (39–70)51 (36–70)0.25
Gender: male (%)33 (80%)30 (67%)0.15
Donor age (years), median, range41 (9–70)39 (14–67)0.98
MELD at transplant, median, range17 (9–40)23 (11–49)0.02
LOS after transplantation, median, range6 (4–11)9.5 (4–36)<0.0001
Type of transplant   
 Deceased donor (%)30 (73%)38 (84%)0.42
 Living donor (%)*11 (27%)7 (16%)0.20
Ischemia times, median, range   
 Cold, hours8.3 (1–17)9.1 (1–20)0.19
 Warm, minutes45.5 (37–88)46.0 (31–99)0.70
Treated acute rejection (%)14 (34%)21 (51%)0.28
Treated CMV infection (%)3 (7%)2 (4%)0.57
Histological follow-up, median, range60 (4–96)50 (4–91)0.05

Use of Antiviral Therapy

Figure 1 outlines the use of antiviral therapy in the study groups. The median duration of preemptive therapy was 46.3 weeks (range, 1.0-78.1), with a premature treatment discontinuation rate of 41%. The most common causes for discontinuation were persistent cytopenias (growth factors were not routinely used in more than half of the patients) and acute rejection. Of the 41 patients in the preemptive group, 17 received IFN monotherapy, 5 received peg-IFN monotherapy, 15 received IFN plus ribavirin, and 4 received peg-IFN plus ribavirin. Twelve of the 41 patients in the preemptive group (29%) were retreated during the follow-up period. Nine of these 12 patients (75%) were begun on retreatment after reaching a fibrosis score ≥ 2. At the start of retreatment, the median necroinflammation and fibrosis scores were 2.0 and 2.0 (mean fibrosis score = 2.1), respectively.

Figure 1.

Use of subsequent antiviral therapy in the study groups. Eighteen patients were not candidates for preemptive therapy for the following reasons: hepatitis B surface antigen positivity, repeat LT, or lack of adequate follow-up due to the patient not living in the region after transplantation. Nine patients were excluded from the analysis because of death within the first 3 months after LT. Eleven patients were excluded from the analysis because of a lack of a liver biopsy during the follow-up period. Abbreviations: HCV, hepatitis C virus; LT, liver transplantation.

Of the 45 patients who did not receive preemptive therapy, 26 (58%) were treated for recurrent HCV during the follow-up period, and all received combination therapy. Of the 26 patients treated, 15 patients (58%) were begun on treatment within the first 6 months from the date of transplantation for severe but not cholestatic recurrent HCV, as evidenced by the median ALT at the time of treatment initiation of 205 IU/L (range, 16-831) and median total bilirubin of 1.0 mg/dL (range, 0.5-5.2). The median time to treatment initiation in these 15 patients with early, severe recurrent HCV was 3.4 months (range, 2.0-5.5 months). In the remaining 11 patients who received therapy, the median necroinflammation and fibrosis scores at the start of treatment were 2.75 and 2.0 (mean fibrosis score = 1.9), respectively. Among all 26 patients who received therapy, the median duration of antiviral treatment was 51.1 weeks (range, 3.0-239.4), with a premature treatment discontinuation rate of 38%. Persistent cytopenias and acute rejection were the most common reasons for treatment discontinuation.

The median time from transplantation to retreatment for moderate fibrosis or necroinflammation was 36.3 months in the preemptive group versus 20.3 months in the nonpreemptive group (P = 0.004).

Cholestatic HCV

Four patients in the nonpreemptive group had a total bilirubin level > 2.5 mg/dL at the start of antiviral therapy versus no patients in the preemptive therapy group.

Cumulative Probability of Moderate to Severe Fibrosis

The cumulative probability of moderate to severe fibrosis defined as a fibrosis score ≥ 2 was examined in 2 different ways: with and without censoring for subsequent antiviral therapy. In the uncensored analysis, the cumulative probability of a fibrosis score ≥ 2 at 48 months post-transplantation was 22% and 49% in the preemptive and nonpreemptive groups, respectively (P = 0.08), with divergence of the curves after 36 months (Fig. 2).

Figure 2.

Cumulative probability of progressing to a Batts-Ludwig fibrosis score ≥ 2 unadjusted for subsequent interferon use. The cumulative probability of progressing to a fibrosis score ≥ 2 was 16% and 22% at 36 and 48 months in the preemptive group and 31% and 49% at 36 and 48 months in the nonpreemptive group (P = 0.08, log-rank test).

In order to adjust for the potential effects of subsequent IFN use on fibrosis progression, the analysis was performed with patients censored at the start of antiviral therapy (retreatment in the preemptive group and first treatment in the nonpreemptive group). In this analysis, the cumulative probability of a fibrosis score ≥ 2 was significantly lower in the preemptive group compared to the nonpreemptive group (P = 0.006; Fig. 3). However, a total of 15 patients in the nonpreemptive group were started on treatment within the first 6 months of transplantation prior to reaching a fibrosis score ≥ 2. When these 15 patients were excluded and the censored analysis was performed, there was no longer a statistically significant difference (P = 0.13; Fig. 4).

Figure 3.

Cumulative probability of progressing to a Batts-Ludwig fibrosis score ≥ 2 with patients censored at the time of subsequent interferon use. The cumulative probability of progressing to a fibrosis score ≥ 2 was significantly different between the 2 groups (P = 0.006, log-rank test).

Figure 4.

Cumulative probability of progressing to a Batts-Ludwig fibrosis score ≥ 2 with patients censored for subsequent antiviral therapy and with the exclusion of 15 patients in the nonpreemptive group in which treatment was initiated within the first 6 months after transplantation before they reached a fibrosis score ≥ 2. The cumulative probability of progressing to a fibrosis score ≥ 2 was not significantly different between the 2 groups (P = 0.13, log-rank test).

Predictors of Moderate to Severe Fibrosis

In univariate analysis, preemptive therapy was associated with a 46% lower likelihood of progressing to a fibrosis score ≥ 2 with a hazard ratio of 0.54 (95% confidence interval = 0.28-1.05), but the association did not achieve statistical significance (P = 0.07). The association was preserved after adjustments for covariates known to be associated with fibrosis progression, including donor age, treated acute cellular rejection (rejection treated with steroid boluses or lymphocyte-depleting antibody therapies), and cytomegalovirus infection, as well as covariates not known to be associated with fibrosis progression, including biological MELD score at transplantation and length of hospital stay after transplantation. In this multivariate model (Table 2), preemptive antiviral therapy was associated with a hazard ratio of 0.52 (95% confidence interval = 0.24-1.12, P = 0.09) for developing a fibrosis score ≥ 2.

Table 2. Multivariate Analysis of Predictors of a Batts-Ludwig Fibrosis Score ≥ 2
 Hazard Ratio95% Confidence IntervalP Value
  1. Abbreviations: CMV, cytomegalovirus; LOS, length of hospital stay; MELD, Model for End-Stage Liver Disease.

Preemptive antiviral therapy0.520.24–1.120.09
Donor age (per year)1.010.98–1.030.52
Treated acute rejection1.800.87–3.730.11
CMV infection2.550.72–8.990.15
MELD at transplant1.010.97–1.050.64
LOS after transplantation0.990.95–1.040.80

Analysis of Severity of Hepatitis C Recurrence

In order to determine whether preemptive therapy had a beneficial effect of ameliorating the course of recurrent HCV, we analyzed the peak ALT and peak total bilirubin within the first 3 months after liver transplantation. In the nonpreemptive group, the median peak ALT was 86 IU/L (mean, 167; range, 16-1120), and the median peak total bilirubin was 1.1 mg/dL (mean, 1.5; range, 0.5-9.4). By comparison, in the preemptive group, the median peak ALT was 91 IU/L (mean, 157; range, 16-1250), and the median peak total bilirubin was 1.0 mg/dL (mean, 1.5; range, 0.4-4.3). There was no significant difference between the peak ALT (P = 0.89, Mann-Whitney test) or the peak total bilirubin (P = 0.06, Mann-Whitney test) between the 2 groups.

DISCUSSION

Preemptive antiviral therapy currently has no established role in the management of transplant recipients with HCV infection. Previous studies have shown a low rate of SVR and no clear advantage over delaying treatment until recurrent histological disease has been documented. Indeed, our own prior results found an SVR rate of only 9%.9 However, the current follow-up study highlights other benefits of preemptive therapy. We found preemptive antiviral therapy to be associated with a 48% reduced risk of progressing to a Batts-Ludwig fibrosis score ≥ 2 with a median histological follow-up of nearly 5 years. Although this difference did not achieve statistical significance, the magnitude of the potential benefit is striking. At 48 months post-transplantation, 22% of patients who had received preemptive therapy had progressed to a fibrosis score ≥ 2 versus 49% of those not receiving preemptive therapy. Preemptive therapy also delayed the need for retreatment of moderate to severe disease, with the median time to retreatment in the preemptive group being significantly longer than the median time to eventual treatment in the nonpreemptive group (36.3 months and 20.3 months respectively, P = 0.004). Although it is commonly accepted that achieving SVR leads to histological benefits, it has not been determined whether the histologic benefit of antiviral therapy extends to virologic nonresponders.2, 3 Our cohort is largely one of nonresponders because only 1 patient in each group achieved SVR during the period of study.

Preemptive therapy, as used in our study, must be distinguished from early treatment, which has been used in other studies and is currently under evaluation in a large multicenter US trial.13 Reinfection is an early event, with HCV RNA detectable in serum with rising titers during the first weeks following transplantation and with biochemical and histological evidence of hepatitis seen within months (median, 3-6 months). Thus, for treatment to be preemptive, it must be initiated early, prior to the development of clinically evident infection. In our preemptive cohort, treatment was initiated within the first 6 weeks after transplantation. This is contrasted against studies of early antiviral therapy, which begin treatment 6 months after transplantation or later. Whether earlier exposure to IFN and ribavirin is more likely to result in a modulation in the rate of histological progression remains to be proven, and longitudinal results of randomized studies of early treatment may also shed light on this issue.

The major advantage of our study is its relatively long duration of histological follow-up (median, 55 months). This allowed us to report on fibrosis progression as the main histological outcome of preemptive therapy rather than on changes in necroinflammation as previous studies have done. The long duration of follow-up also allowed us to observe an apparent late benefit of preemptive antiviral therapy, with the cumulative risk of progressing to a fibrosis score ≥ 2 beginning to diverge at 36 months after transplantation between the preemptive and nonpreemptive groups. A second strength of the study is that we performed an intention-to-treat analysis in which all patients who received even a single dose of preemptive therapy were included in the preemptive group. This may have resulted in a more conservative estimate of the beneficial effect of preemptive therapy in comparison with a per-protocol analysis.

It must be noted that this study has several limitations. First, the patients were not randomized to receive preemptive antiviral therapy versus no treatment, and this may have introduced bias by indication. Patients in the nonpreemptive group were enriched with patients who were often too sick in the early posttransplant period to qualify for preemptive therapy, and this is highlighted by the higher biological MELD score at transplantation and the longer hospital length of stay after transplantation in the nonpreemptive group. However, these differences are unlikely to affect the results of our study, as neither MELD nor length of stay has been shown to correlate with severity of HCV disease following transplantation. In addition, the mean rate of fibrosis progression seen in our nonpreemptive group (0.6 stage units/year) was similar to those reported in other natural history studies in transplant recipients, which range from 0.3 to 0.6 stage units per year (scale 0-4).14-16

In addition, a large percentage of patients (58%) received antiviral therapy at some point during the follow-up period for progression of disease, and this made it difficult to report the natural history of each group in the absence of IFN therapy. When patients receiving antiviral therapy in the nonpreemptive group and retreatment in the preemptive group were censored at the time of subsequent treatment, a statistically significant difference in the cumulative probability of moderate to severe fibrosis (≥F2) was present. Divergence of the probability curves was most apparent 36 months post-transplantation, nearly 2 years after completion of the preemptive therapy. In order to minimize a bias toward early censoring and thus worse outcomes in the nonpreemptive group, we repeated the censored analysis, excluding patients in the nonpreemptive group who developed early severe recurrence and started treatment within the first 6 months, and we found that the cumulative probability of ≥F2 was no longer significantly different, although the direction of the association was preserved. These subanalyses suggest that the histological benefit of preemptive therapy may be due to prevention of early, severe recurrent HCV, which has been associated with more aggressive fibrosis progression.17 Although we did not observe a significant difference in the peak ALT or total bilirubin within the first 3 months after transplantation between the 2 groups, perhaps it is because we were not capturing all cases of severe HCV recurrence in the time frame of three months. Preemptive therapy may be modifying the immune response in new HCV infection in the graft and resulting in a lower risk of severe or progressive fibrosis. Additional studies of virologic and immunologic events during the early posttransplant period would be needed to elucidate the mechanism.

In conclusion, our results show that preemptive therapy was associated with a 48% reduced risk of progressing to a fibrosis score ≥ 2, although the result did not meet statistical significance. Preemptive therapy was, however, associated with a significant delay in the need for subsequent HCV therapy for moderate to severe disease. In the absence of therapies that uniformly achieve viral eradication, there remains a need for alternative management strategies that can prolong graft survival. For this reason and on the basis of the results of our study, we recommend additional prospective studies to confirm our results and to determine the optimal timing of preemptive antiviral therapy to maximize both histological and virological outcomes.

Ancillary

Advertisement