This work was supported in part by the Spanish Ministry of Science through the Health Research Fund of the Carlos III Institute of Health (PS09/01707 and Network Center for Biomedical Research in Hepatic and Digestive Diseases).
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
After liver transplantation (LT), a response to interferon (IFN)-based therapies resulting in the permanent eradication of hepatitis C virus (HCV) is associated with improved histology, reduced disease progression, a decreased risk of clinical decompensation, and improved overall graft and patient survival.1-8 The standard current therapy in the setting of LT is pegylated interferon (PEG-IFN) with ribavirin (RBV), and a sustained virological response (SVR) is achieved by approximately one-third of patients.1-11 As for those who do not achieve SVR, there are 2 major categories: nonresponders and responders who relapse. Because of the accelerated course of recurrent HCV disease in the LT population and the limited efficacy of current antiviral therapy, it is expected that the number of prior failures with a first course of antiviral therapy will increase in the future.
Among immunocompetent patients, the retreatment of prior relapsers with PEG-IFN and RBV results in higher rates of SVR than the retreatment of prior nonresponders; this is particularly true for patients who received suboptimal therapy during the first course (ie, monotherapy with IFN, therapy for a suboptimal duration, or suboptimal doses of RBV or IFN). In contrast, the retreatment of prior nonresponders (particularly those for whom a prior course with PEG-IFN and RBV has failed) results in a significantly smaller proportion of patients achieving SVR, even when the patients are retreated with aggressive approaches such as an induction phase with high PEG-IFN doses together with the prolongation of therapy.12, 13
Triple therapy with protease inhibitors in combination with PEG-IFN and RBV has become the standard of care in immunocompetent patients with genotype 114-16 and most specifically in prior relapsers because of the high rates of SVR achieved (on the order of 85%). There are no data yet about the efficacy of these compounds in the transplant population. In order to determine adequately the benefits of these new therapies versus PEG-IFN and RBV in transplant patients, we first need to define the efficacy and tolerability of PEG-IFN and RBV in both naive and retreated recipients. Although many studies have now been published about antiviral therapy in naive recipients after transplantation,1-11 there are almost no data on retreatment in this setting9, 17, 18 or studies evaluating factors that might predict retreatment outcomes.
In this study, we aimed to determine in patients with recurrent HCV (relapsers or prior nonresponders to a previous posttransplant IFN-based course) the efficacy, safety, and benefits of retreatment with PEG-IFN and RBV. In particular, our aims were (1) to determine whether retreatment with PEG-IFN with or without RBV results in virological or clinical benefits and (2) to assess predictive factors associated with a virological response, including rapid virological response (RVR), treatment adherence, donor age, baseline immunosuppressive therapy (cyclosporine A versus tacrolimus), and baseline fibrosis.
PATIENTS AND METHODS
Data on patients retreated after transplantation were retrieved from prospectively collected databases of 4 European centers. The study protocol received a priori approval by the appropriate institutional review committees. Patients were relapsers or nonresponders to a prior course of standard or PEG-IFN–based therapy administered after transplantation. Only patients treated for a second or third time after LT with information available about HCV RNA 6 months after treatment were included in this analysis.
PEG-IFN (Pegasys from Roche, Inc., and PegIntron from Schering-Plough, Inc.) and RBV (Rebetol from Schering-Plough and Copegus from Roche) were started at full or reduced doses according to the hemoglobin, neutrophil, or platelet count. Growth factors were used whenever they were considered appropriate by the hepatologist in charge. The intended duration of therapy was 48 weeks whenever possible.
HCV RNA was quantified in serum at the baseline, during therapy (4, 12, 24, and 48 weeks), and 24 weeks after treatment. A biochemical response was defined as the normalization of liver enzyme levels. In addition, a virological response was defined as HCV RNA negativity in serum according to a qualitative polymerase chain reaction. The response was considered complete if it occurred at the end of the completion of therapy (end-of-treatment virological response) and was considered sustained if it was maintained for 6 months after the completion of therapy (SVR). An early virological response (EVR) was defined as a >2 log10 IU/mL reduction in HCV RNA levels from the baseline 3 months after the beginning of therapy. In turn, RVR was defined by the undetectability of HCV RNA 1 month after the beginning of therapy.
Variables that were analyzed as potential predictors of SVR included the following: donor and recipient demographics, history of antiviral therapy before transplantation, baseline immunosuppression, baseline viremia and infecting genotype, baseline disease severity, laboratory parameters (including transaminases, bilirubin, hemoglobin, leukocytes, platelets, glycemia, and creatinine), type of PEG-IFN (α2a versus α2b), treatment adherence, time from transplantation to treatment initiation, time from the first course of therapy to the second, and variables from the first course of posttransplant antiviral therapy [type of IFN (PEG-IFN versus standard IFN), type of response (relapse versus nonresponse), and adherence-related factors (initial RBV dose, cumulative RBV dose, and treatment duration)].
The primary endpoint of the analysis was SVR. The secondary endpoints were the end-of-treatment virological response and the tolerability of treatment (dose reductions and treatment discontinuation rate).
The baseline characteristics and measures of the tolerability and efficacy of treatment are described as proportions or as medians and ranges. Comparisons of sustained responders and nonresponders were made with Pearson chi-square and Wilcoxon tests.
The McNemar test was used to compare proportions between the first and second courses of therapy. A P value < 0.05 was considered statistically significant. A multivariate analysis using logistic regression analysis to identify independent predictors of SVR was performed for the variables that showed a level of significance of P < 0.1 in a univariate analysis. In addition, a Cox regression analysis was performed to identify independent predictors associated with survival.
From 1997 to 2009, 489 patients were treated with a first course of IFN-based therapy at the 4 centers. One hundred eighty-eight of these patients (38.5%) achieved SVR, so 301 were potential candidates for retreatment. Only 79 of these 301 patients (26%) were finally retreated from 2000 to 2011. The reasons for not retreating the remaining 222 patients included the following: advanced liver disease, death, or retransplantation (n = 80); advanced age or comorbidities (n = 42); tumor recurrence or a de novo tumor (n = 9); a combination of negative predictive factors for a viral response or a lack of patient acceptance (n = 33); waiting for triple therapy (n = 36); and a lack of data (n = 22).
The 79 patients, who had a median age of 59 years (range = 35-77 years), were mostly male (72%) and infected with genotype 1 (87%), and mostly had a high baseline viral load (85% with viremia > 600,000 IU/mL), were retreated at a median of 6.9 years (range = 175 days to 21.6 years) after transplantation. The time elapsing from the end of the first course of therapy to retreatment was 1.9 years (range = 45 days to 8.2 years). The most commonly chosen PEG-IFN was α2a, which was used in 60% of the retreated patients. In turn, the baseline immunosuppression included tacrolimus (51%), cyclosporine A (49%), mycophenolate mofetil (19%), and steroids (22%). Most patients (63%) were retreated at advanced stages of fibrosis (F3-F4), with cirrhosis present in 37%. Retreatment was started at a median of 257 days after the previous biopsy. The median body mass index was 27 kg/m2 (range = 20-38 kg/m2). Thirty-three percent were diabetics at retreatment. The baseline laboratory values were as follows: alanine aminotransferase, 89 IU/L (range = 20-485 IU/L); total bilirubin, 1.06 mg/dL (range = 0.41-13.10 mg/dL); creatinine, 1.05 mg/dL (range = 0.6-4 mg/dL); hemoglobin, 14 g/dL (range = 10-17.8 g/dL); leukocyte count, 4800/mm3 (range = 1510-9600/mm3); and platelet count, 124,000/mm3 (range = 36,000-412,000/mm3).
Forty-eight patients achieved an end-of-treatment response (61%), with relapse occurring in 20 (25%); therefore, SVR was achieved by 28 patients (35%). The SVR rate was 32% for genotype 1–infected patients [28.5% for genotype 1a (n = 7) and 34.5% for genotype 1b (n = 55), P = not significant] and 60% for non–genotype 1 patients [genotype 2 (2/2), genotype 5 (1/1), genotype 3 (3/6), and genotype 4 (0/1), P = not significant]. The median duration of therapy was 357 days (range = 44-763 days), and there was a trend for a longer period of therapy in genotype 1 patients versus non–genotype 1 patients [362 days (range = 119-763 days) for genotype 1 versus 321 days (44-378 days) for genotype 2, P = 0.06]. Only 49 patients (62%) were started at full doses of RBV. Adverse events were frequent and resulted in dose reductions in 44% and premature treatment discontinuation in 20%. Only 57% of the patients fully adhered to an optimal course of therapy and received more than 80% of both drugs for more than 80% of the estimated duration (Table 1). Five cases of rejection occurred among these patients, but treatment discontinuation was required for only 1 of these patients. No case of plasma cell hepatitis was reported in this series.
Table 1. Comparison of the First and Second Courses of Antiviral Therapy (n = 79)
NOTE: Bolded P values are significant.
The data are presented as medians and ranges.
Data were available for 69 patients.
At least 80% of the doses of antiviral agents for at least 80% of the expected duration of therapy.
An assessment of fibrosis after retreatment was available for only 48 patients: 15 SVR patients and 33 nonresponders. An improvement in fibrosis greater than or equal to 1 fibrosis unit, stabilization with similar fibrosis, and a worsening of fibrosis greater than or equal to 1 fibrosis unit were observed in 32.5%, 41.5%, and 26%, respectively. Among patients with SVR, the vast majority had biopsy samples after retreatment that showed either an improvement in the fibrosis score or stabilization (14/15 or 93%). In contrast, among nonresponders, an improvement or stabilization was observed in a lower proportion of cases (21/33 or 64%, P = 0.07).
SVR was associated with significant improvements in survival after therapy (Fig. 1) and hence after transplantation (Fig. 2). The 1-, 5-, and 10-year survival rates after retreatment were 100%, 93%, and 93%, respectively, for patients achieving SVR and 96%, 75%, and 61%, respectively, for nonresponders to antiviral retreatment (P = 0.04, log-rank test). There were 2 deaths among the 28 SVR patients (7%) and 13 deaths among the 51 nonresponders (25%, P = 0.047). One of the 2 deaths among the SVR patients was HCV-related, and the other was unrelated. Eight of the 13 deaths (61.5%) among the nonresponders were HCV-related, and the remainder were unrelated to HCV. When additional variables associated with survival in the univariate analysis [age at retreatment (P = 0.04) and baseline fibrosis (P = 0.04)] were entered into a Cox regression multivariate analysis, the only independent variable predicting survival was baseline fibrosis (odds ratio = 4.618, 95% confidence interval = 1.036-20.57, P = 0.04), and SVR was no longer an independent variable predicting survival (P = 0.08).
Comparison of the First and Second Courses of Therapy
The 2 courses of therapy differed in many variables and particularly in variables related to disease severity at the baseline (with an increase in patients with cirrhosis from 10% to 37%, P < 0.001), in the age at initiation (with the median age increasing from 55 to 59 years, P = 0.02), and in adherence-related variables. This suggested a more aggressive approach for retreated patients: a higher proportion of patients received full initial RBV doses (45% versus 62%, P = 0.03), a higher proportion used growth factors (erythropoietin, 32% versus 72%, P < 0.001; granulocyte colony-stimulating factor, 6% versus 16%, P = 0.048) or required transfusions (9% versus 22%, P = 0.03), and the treatment duration was longer (P = 0.03; Table 1). In addition, the proportion of patients using steroids decreased from 52% to 22% (P < 0.001).
Other variables, such as tacrolimus usage (49% and 51% during the first and second courses, respectively), mycophenolate mofetil usage (18% and 19% during the first and second courses, respectively), diabetes, baseline renal function, body mass index, and viral load, remained unchanged (Table 1).
Variables Associated With SVR in Retreated Patients
Both RVR and EVR were more common in retreated patients versus naive patients [24% versus 7% for RVR (P = 0.008) and 75% versus 50% for EVR (P = 0.002); Table 1]. SVR was achieved by 28 patients (35%).
In the univariate analysis, the baseline variables associated with SVR were disease severity (baseline fibrosis, platelet counts, and bilirubin levels) and age at therapy. More specifically, older age, advanced fibrosis, higher bilirubin levels, and lower platelet counts were associated with poorer SVR rates (Table 2). In addition, a trend between baseline viremia and SVR was observed (P = 0.08). The on-treatment variables associated with SVR were viral kinetics and treatment adherence (Table 3). EVR was significantly associated with SVR: SVR was achieved by 47% of EVR patients and by 0% of non-EVR patients (P < 0.001). RVR was also associated with SVR: 61% of the 18 RVR patients achieved SVR, whereas 26% of the non-RVR patients did (P = 0.006). Dose reductions and premature terminations of therapy that resulted in patients not reaching 80% of the recommended doses or not maintaining the treatment for at least 80% of the recommended duration resulted in lower response rates.
Table 2. Baseline Factors Associated With SVR in Retreated Patients
Interestingly, neither the type of IFN used in the first course of therapy (standard versus pegylated) nor the type of response achieved during the first course (relapse versus nonresponse) were significantly associated with SVR during the retreatment course. There was, however, a trend between the type of response to the first course and SVR after retreatment, with prior relapsers achieving SVR more frequently than nonresponders (44% versus 27.5%, P = 0.13).
Recurrent progressive disease is one of the major challenges in transplant centers.1 Treatment with antivirals is the only means for improving the outcome. Persistent viral eradication has indeed been shown to improve histology, decrease the risk of decompensation, and thereby increase graft and patient survival.1-10 Unfortunately, only one-third of treated patients achieve SVR when they are treated after LT, whereas another third are considered nonresponders, and one-quarter to one-third relapse after they achieve a viral response at the end of therapy.1, 9-11, 18-24 Although significant data have accumulated for the immunocompetent population with respect to treatment options for these nonresponders,12-16 there is almost no information for the transplant population.
The aim of our study was, therefore, to determine the benefits, limitations, and predictive factors of retreatment with both PEG-IFN and RBV in a relatively large number of HCV-infected LT recipients recruited at 4 transplant centers who had not achieved SVR after the first course of antiviral therapy. Although we acknowledge that PEG-IFN and RBV will soon be history and that the transplant community should be and is already focusing on testing direct antiviral agents with or without PEG-IFN and RBV to improve tolerability and SVR rates, this article sets the bar that those studies will need to exceed in the future.
The main findings of the study can be summarized as follows: (1) patients who are selected for or agree to retreatment represent a heavily biased subset of the pool of potential candidates for retreatment; (2) in this difficult-to-treat population of prior nonresponders to IFN-based regimens who are mainly infected with genotype 1 virus with a high viral load and include a high proportion of patients with an advanced stage of disease, SVR can be achieved by a substantial number of patients (30% on average); (2) SVR is associated with improved outcomes; (3) advanced disease is associated with impaired SVR rates; (4) an aggressive approach to treatment adherence results in enhanced viral eradication; and (5) viral kinetics are highly predictive of SVR.
The first important consideration is that a failure to respond to an initial course of antiviral therapy should not preclude a second course, even in the presence of negative factors for a treatment response. However, an effort should be made to retreat patients before the disease is too advanced because treatment at advanced stages of fibrosis is associated with poorer SVR rates as well as lower treatment benefits in terms of survival. Baseline severity is a variable that has been shown in previous studies to be associated with SVR in both immunocompetent patients and LT recipients. In particular, cirrhosis has been found to be associated not only with decreased treatment efficacy but also with worse tolerability after a first course of antiviral therapy.2, 4, 7, 8, 18, 22, 24 Importantly, just as with first courses of therapy,1-6 SVR was significantly associated with improved survival in the univariate analysis. Although in the multivariate analysis survival was no longer associated with SVR but mostly depended on baseline disease severity, the P value for SVR as an independent predictor of survival approached statistical significance (P = 0.08), and this emphasized the concept that treatment and/or retreatment should not be delayed until advanced stages of fibrosis.
In addition, it is also important to ensure a full course with full doses of antivirals in order to improve the results. An aggressive approach to treatment adherence (particularly RBV doses) has also been shown to improve treatment efficacy in both immunocompetent patients and LT patients.8, 22-24
Finally, just as in the nontransplant population and in transplant patients treated for the first time,1, 6, 7-9, 18-24 viral kinetics were highly predictive of SVR, with 61% of RVR patients finally achieving SVR; in contrast, none of the non-EVR patients achieved SVR. In the new era of direct antiviral agents, data on RVR and its predictability for SVR can be relevant.
Surprisingly, neither the type of response to antiviral therapy during the first course nor the type of therapy (monotherapy versus combination therapy, standard IFN versus PEG-IFN, or relapse versus nonresponse)—factors known to significantly affect treatment outcomes in immunocompetent patients12-16—had an effect on SVR in our cohort of retreated LT patients. There was a trend, however, in the type of response to prior therapy: prior nonresponders had lower probabilities of achieving SVR in comparison with prior relapsers. This might be a reflection of the sample size; alternatively, it is likely that other cofactors extremely common in the LT setting (eg, degree of immunosuppression, treatment adherence, and patients' and donors' genetic backgrounds) play a more important role in this setting. Indeed, a favorable interleukin-28B genotype in both the donor and the recipient is significantly associated with SVR.22, 25-27 Unfortunately, we had no data on interleukin-28B in our study. However, the distribution of these genotypes could not have changed between the 2 treatment courses. In immunocompetent patients treated with triple therapy (PEG-IFN, RBV, and telaprevir or boceprevir), the type of response to a prior course of therapy is the strongest predictor of retreatment outcome, so that an SVR rate of 85% can be achieved when relapsers are retreated, regardless of baseline fibrosis; however, these rates are substantially lower for prior nonresponders and range from approximately 30% for prior null responders to 50% for prior partial responders.14-16 If the same results are obtained for LT recipients, retreatment with double therapy (with which SVR is achieved in approximately 25% of prior nonresponders and in 40%-45% of prior relapsers) should no longer be used. Future studies with triple therapy in LT recipients will answer this question.
In conclusion, SVR can be achieved by approximately one-third of LT recipients retreated with a course of PEG-IFN and RBV. The treatment of patients before they reach an advanced stage of fibrosis is associated with better results. However, relatively good results can be achieved even in the presence of advanced fibrosis, with an SVR rate of approximately 25%. An aggressive approach to treatment adherence, however, is required to obtain these results. Viral kinetics are extremely helpful in this setting and have predictive values similar to those for the treatment of naive LT recipients.