Worse Recent Efficacy of Antiviral Therapy in Liver Transplant Recipients with Recurrent Hepatitis C: Impact of Donor Age and Baseline Cirrhosis

Transcription

1 LIVER TRANSPLANTATION 15: , 2009 ORIGINAL ARTICLE Worse Recent Efficacy of Antiviral Therapy in Liver Transplant Recipients with Recurrent Hepatitis C: Impact of Donor Age and Baseline Cirrhosis Marina Berenguer, 1,3,4 Victoria Aguilera, 1,3 Martín Prieto, 1,3 Cecilia Ortiz, 3 Maria Rodríguez, Federica Gentili, 1 Blas Risalde, 3 Angel Rubin, 1 Raquel Cañada, 3 Antonio Palau, 1,3 and Jose-Miguel Rayón 2,3 1 Hepatogastroenterology Service and 2 Pathology Service, Hospital Universitari La Fe, Valencia, Spain; and 3 Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas, 4 Facultad de Medicina, Universidad de Valencia, Valencia, Spain We hypothesized that antiviral efficacy [sustained virologic response (SVR)] has improved in recent years in the transplant setting. Our aim was to assess whether the efficacy of pegylated interferon (PegIFN) ribavirin (Rbv) has improved over time. One hundred seven liver transplant patients [74% men, 55.5 years old (range: ), 86% genotype 1a or 1b] were treated with PegIFN-Rbv for 355 (16-623) days at 20.1 ( ) months after transplantation. Tacrolimus was used in 61%. Sixty-seven percent had baseline F3-F4 (cirrhosis: 20.5%). Donor age was 49 (12-78) years. SVR was achieved in 39 (36.5%) patients, with worse results achieved in recent years ( : n 27, 46.5%; 2004: n 23, 43.5%; 2005: n 21, 35%; 2006 to January 2007: n 36, 24%; P 0.043). Variables associated with SVR in the univariate analysis included donor age, baseline viremia and cirrhosis, bilirubin levels, rapid virologic response and early virologic response (EVR), premature discontinuation of PegIFN or Rbv, and accumulated Rbv dose. In the multivariate analysis, the variables in the model were EVR [odds ratio (OR): 0.08, 95% confidence interval (CI): , P 0.002] and donor age (OR: 1.039, 95% CI: , P 0.01). Variables that had changed over time included donor age, baseline viremia, disease severity (cirrhosis, baseline bilirubin, and leukocyte and platelet counts), interval between transplantation and therapy, and use of growth factors. In the multivariate analysis, variables independently changing were donor age (OR: 1.041, 95% CI: , P 0.004), duration from transplantation to antiviral therapy (OR: 1.001, 95% CI: , P 0.013), and baseline leukocyte count (OR: 1.000, 95% CI: , P 0.034). In conclusion, the efficacy of antiviral therapy with PegIFN-Rbv has worsened over time, at least in our center. The increase in donor age and greater proportion of patients treated at advanced stages of disease are potential causes. Liver Transpl 15: , AASLD. Received September 3, 2008; accepted November 16, See Editorial on Page 677 Hepatitis C related end-stage liver disease is the main indication for liver transplantation in most transplant centers. Largely as a result of recurrent hepatitis C, the 5- and 10-year survival rates are inferior to those reported in uninfected recipients. 1,2 Strategies to improve the outcome include the modification of factors associated with progressive disease and the effective management of established hepatitis C through the use of antiviral therapy. 1-3 Both strategies result in improved Abbreviations: CI, confidence interval; EPO, erythropoietin; EVR, early virologic response; GCSF, granulocyte colony stimulating factor; GSF, granulocyte stimulating factor; HAI, histologic activity index; HCV, hepatitis C virus; IFN, interferon; NR, nonresponse; NS, not significant; OR, odds ratio; Peg 2a, pegylated interferon alfa 2a; Peg 2b, pegylated interferon alfa 2b; PegIFN, pegylated interferon; Rbv, ribavirin; RVR, rapid virologic response; SVR, sustained virologic response. Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas is funded by the Instituto de Salud Carlos III (PI and CB06/04/0065). The study also received funding from the Colegio de Médicos de Valencia. Address reprint requests to Marina Berenguer, Servicio de Hepatogastroenterología, Hospital Universitario La Fe, Avenida Campanar 21, Valencia, Spain. Telephone: ; FAX: ; mbhaym@teleline.es DOI /lt Published online in Wiley InterScience ( American Association for the Study of Liver Diseases.

2 PEGIFN AND RIBAVIRIN IN HCV LIVER TRANSPLANT RECIPIENTS 739 survival. 4-6 In a recent study from our group, 5 patient survival was found to be significantly greater among patients treated with interferon (IFN)-based therapies compared to disease-matched recipients who were left untreated. As in the nontransplant population, pegylated interferon alfa (PegIFN ) in combination with ribavirin (Rbv) yields the best results in terms of sustained virologic response (SVR) However, a number of questions remain unanswered regarding doses, duration of therapy, necessity of adjuvant therapy (eg, with growth factors), risk for rejection, and mostly factors associated with SVR, particularly those that might be modified at an early stage so that toxic therapy is not continued in the subgroup of patients with a low likelihood of response. We report our experience with PegIFN in combination with Rbv in a large cohort of hepatitis C liver transplant recipients managed in a university hospital based hepatology clinic. We hypothesized that SVR as a marker of effectiveness would be (1) greater in patients treated in more recent years because of a theoretical learning curve, (2) similar with PegIFN 2a-Rbv and PegIFN 2b-Rbv, (3) similar under cyclosporine-based regimes and under tacrolimus, and (4) worse in previously treated patients (before transplantation) in comparison with treatment-naïve patients. We also hypothesized that the same predictors of response described in the immune-competent population, namely the genotype and viral load before therapy and at 1 and 3 months of therapy, would apply to the liver transplant patients. The aims of this study were therefore (1) to determine whether the efficacy, defined by the rate of SVR, and the tolerance, defined by the proportion of patients who required dose reduction/premature discontinuation of therapy, had improved over time and (2) to identify baseline and on-treatment factors associated with SVR. PATIENTS AND METHODS A retrospective chart review was performed on all hepatitis C virus (HCV) infected liver transplant patients who received combination therapy with PegIFN and Rbv at the hepatology clinic between December 2001 and January Only patients for whom data on viral response 6 months following the end of antiviral therapy were available were included in this analysis. Patient Selection Since the inception of the transplant program, HCVinfected transplant recipients have been followed up by yearly protocol liver biopsies. In the last 3 to 5 years, a new protocol has been put into place so that patients with mild recurrence (defined as fibrosis 0 and mild necroinflammation) undergo subsequent protocol biopsies at longer intervals (1, 3, 5, 7, 10, and 15 years). Additional liver biopsies are performed if clinically indicated, generally within the first months post-transplantation to exclude coexistent diseases or complications such as rejection. In fact, all patients had an initial liver biopsy to assess the severity of liver disease. The decision to initiate treatment was individualized and made jointly by the patient and attending hepatologist. The absolute number of patients who were potentially candidates for antiviral therapy is unknown. However, in general, therapy was initiated when there were changes compatible with progressive disease (ie, progression to fibrosis 1 and/or moderate to severe necroinflammation within the first year post-transplantation, changes in the stage of fibrosis of more than 1 unit between yearly biopsies, or a change in the pattern of recurrent hepatitis to a more cholestatic form during follow-up). Patient acceptance to undergo antiviral therapy was always taken into account. The viral load did not have an impact on the decision process. No patient without established recurrence (ie, preemptive therapy) was included. In addition, patients treated with standard IFN or PegIFN in monotherapy or those treated with standard IFN in combination with Rbv were excluded. Other causes of graft dysfunction, including rejection, biliary or vascular complications, and other viral infections, were carefully excluded by Doppler ultrasound, cholangiograms, cultures, and histological examination of the graft (multiple if necessary). Patients with ongoing rejection, evidence of autoimmune hepatitis, and a history of heart disease were not treated. In addition, therapy was stopped in patients who developed a severe episode of acute rejection or chronic rejection under IFN therapy. Antiviral Treatment Regimen PegIFN (Pegasys, Roche, Inc., and Pegintron, Schering- Plough, Inc.) and Rbv (Rebetol, Schering-Plough, and Copegus, Roche) were started at full or reduced doses according to the hemoglobin, total white blood cell count, absolute neutrophil count, or platelet count. Doses were modified according to standard criteria, the known effects of both drugs being taken into consideration. Adjuvant therapies including erythropoietin (EPO) and granulocyte colony stimulating factor injections were used whenever they were considered appropriate by the hepatologist in charge. The intended duration of therapy was 48 weeks whenever possible. A biochemical response was defined by normalization of liver enzyme levels (serum alanine aminotransferase and aspartate aminotransferase). In addition, a virological response was defined by the negativity of HCV-RNA in serum by qualitative polymerase chain reaction. The response was considered complete if it occurred at the end of completion of therapy. A sustained response was achieved when both biochemical and virologic responses were observed 6 months after the completion of therapy. In addition, histological improvement was defined by the reduction of at least 2 points in the activity grade or 1 point in the fibrosis stage in posttreatment biopsies. Virological Assays HCV-RNA was detected in serum at baseline, at 4, 12, 24, and 48 weeks of therapy, and at 24 weeks post-

3 740 BERENGUER ET AL. treatment. HCV-RNA (qualitative) was detected with the Cobas Amplicor HCV test, version 2.0 (Roche Diagnostics, Branchburg, NJ). HCV-RNA quantitation was performed either with the Cobas Amplicor HCV Monitor test, version 2.0 (Roche Diagnostics), or with the branched DNA assay, version 3.0 (Bayer Diagnostics, Tarrytown, NY). The viral load values (IU/mL) obtained with the 2 techniques within their respective dynamic ranges were highly correlated. HCV genotyping was performed with a commercial reverse-hybridization genotyping assay (Inno-LIPA HCV II, Innogenetics, Zwijndrecht, Belgium). Histological Examination Liver biopsies were available before therapy and subsequently at the end of completion of therapy if possible. All biopsy specimens were read by a single pathologist (J.M.R.) who was blinded to biochemical and virological responses of the patient. Histological recurrence of HCV and disease severity were scored according to a slight modification of the histologic activity index (HAI) proposed by Knodell, and the histological grade (activity) and stage (fibrosis) were evaluated separately. The grade was determined by the combination of the HAI scores for periportal necrosis (0-6: 0, none; 1, mild piecemeal necrosis; 3, moderate piecemeal necrosis; 4, marked piecemeal necrosis; 5, moderate piecemeal necrosis plus bridging necrosis; and 6, marked piecemeal necrosis plus bridging necrosis), lobular degeneration and necrosis (0-4), and portal inflammation (0-4), and it was defined as follows: 1 to 2, minimal; 3 to 6, mild; 7 to 10, moderate; and 11 to 14, severe. In addition, liver biopsy samples were staged according to the original HAI fibrosis score: 0, none; 1, fibrous portal expansion; 3, bridging fibrosis; and 4, cirrhosis. Graft biopsy specimens were also examined for features of acute and chronic rejection. Cellular rejection was always based on histological findings, including mixed portal infiltrate, venous endothelitis, and bile duct injury. A formal Banff score was not given because of the overlap with recurrent hepatitis C. Chronic rejection was defined by the presence of bile duct atrophy/ pyknosis, bile duct paucity, and foam cell obliterative arteriopathy. Statistical Analysis The primary endpoint of the analysis is SVR. Secondary endpoints are end-of-treatment biochemical and virological responses, histological improvement, and tolerability of treatment (treatment discontinuation rate). Baseline characteristics and measures of tolerability and efficacy of treatment are described as proportions or medians and range. Comparisons between sustained responders and nonresponders are made with 2-sided Fisher s exact tests and Wilcoxon s tests. Multivariate analysis using logistic regression analysis was performed for the variables that showed a level of significance of P 0.1 to identify independent predictors of TABLE 1. Baseline Features (n 107) Male gender 77 (72%) Age at therapy (years) 55.5 ( ) Genotype 1 92 (86%) Viral load pre-therapy (IU/ ml) ( to ) Pretransplantation antiviral 32 (30%) therapy Posttransplantation antiviral therapy Naïve 90 (84%) Retreatment 17 (16%) Treatment cohort (47%) (53%) Baseline immunosuppression Cyclosporine 42 (39%) Tacrolimus 65 (61%) Steroids at initiation of 15 (14%) therapy Type of therapy Peg 2a-Rbv 66 (62%) Peg 2b-Rbv 41 (38%) Histology at initiation* Acute hepatitis 4 (3.5%) Cholestatic hepatitis 11 (10%) Chronic hepatitis F0-F1 35 (33%) F3-F4 70 (67%) Body mass index at initiation 26 ( ) Donor age (years) 49 (12 78) Donor age 50 years 53 (49%) Time to therapy since 20.1 ( ) transplantation (months) Abbreviations: Peg 2a, pegylated interferon alfa 2a; Peg 2b, pegylated interferon alfa 2b; Rbv, ribavirin. *Patients might be assigned to more than 1 category. SVR. A P value of 0.05 was considered statistically significant. RESULTS Patient Population This study included 107 hepatitis C infected liver transplant patients treated with PegIFN in combination with Rbv for recurrent HCV-related liver disease with a minimum of 6 months since the end of antiviral therapy (Table 1). Of these, 72% were men, and the median age was 55.5 years (range: ). The median patient body mass index was 26 ( ). Most patients were infected with HCV genotype 1 (n 92, 86%). Except for 2 patients, the remainder had elevated baseline alanine aminotransferase levels (median: 151 IU/L; range: ). The median time to treatment after liver transplantation was 603 days ( days). The median donor age was 49 (range: 12-78) years. Tacrolimusbased immunosuppression was used in 61%. Most patients were IFN-naïve post-transplant. Baseline disease severity was considered advanced in 33 patients (cir-

4 PEGIFN AND RIBAVIRIN IN HCV LIVER TRANSPLANT RECIPIENTS 741 rhosis, n 22; cholestatic hepatitis, n 11). Although none of the patients had clinical evidence of hepatic decompensation, 17 (16%) had bilirubin levels higher than 3 g/dl. In addition, 32% of the patients had a platelet count 100,000/mm 3, 14% had a leukocyte count 3000/mm 3, 10% had hemoglobin levels 12 g/dl, and 8.5% had creatinine levels 1.5 mg/dl. End-of-Treatment and Sustained Biochemical and Virologic Responses End-of-treatment biochemical responses and end-oftreatment virologic responses were seen in 40 (37.5%) and 73 (68%) patients, respectively. Sustained biochemical and virologic responses were seen in 36 (34%) and 39 (36.5%) patients, respectively. Dose Reductions and Premature Termination of Therapy Nearly half of the patients had premature discontinuation of IFN (n 37, 35%), Rbv (n 43, 40%), or both agents secondary to adverse events, including anemia (n 14, 32%), rejection episodes (n 9, 25%), intolerance (n 6, 16%), and miscellaneous [n 10, 27%; a lack of early virologic response (EVR) was the cause of early treatment discontinuation in only 2 of these patients]. In some instances, 2 or more causes led to the decision to stop therapy. In addition, 50 patients had adverse events, mainly anemia, neutropenia, or thrombocytopenia, necessitating dose reductions of IFN (n 23, 21.5%) or Rbv (n 50, 47%). The premature discontinuation of either drug was not affected by the use of either EPO or granulocyte stimulating factor (GSF). In addition, there was no association between the number of patients who had received more than 80% of Rbv or PegIFN and the use of EPO and/or GSF (data not shown). Histological Data Paired biopsies at baseline and at the end of therapy/ end of follow-up were available in 47 patients: 26 had completed therapy and 21 had prematurely discontinued treatment because of side effects. A comparison of activity scores between M0 and M12 to M18 showed a stabilization and/or improvement in activity in 61.5% of virological responders; in turn, fibrosis remained unchanged and/or improved in 46% of SVR patients. The results were not statistically different from those observed in nonresponders (59% and 44%, respectively). However, it is difficult to draw meaningful conclusions from these data because a posttreatment biopsy was available in only 13 SVR patients; indeed, these biopsies were mostly performed in nonresponders (n 34) or in SVR patients who did not achieve a complete normalization of liver enzymes. Predictors of SVR (Table 2) Variables evaluated as potential predictors of SVR were donor and recipient demographics, baseline immunosuppression, baseline viremia and infecting genotype, baseline disease severity, laboratory parameters (including levels of transaminases, bilirubin, hemoglobin, leukocytes, platelets, glycemia, and creatinine), type of PegIFN, treatment adherence, time interval between transplantation and treatment initiation, rapid virologic response (RVR; evaluated at 1 month of therapy) and EVR (evaluated at 3 months), and year of therapy. Univariate Analysis Univariate analysis demonstrated that there was no significant difference in SVR rates between those treated with PegIFN 2b and those treated with PegIFN 2a (32% versus 39%). The response was also similar regardless of the calcineurin inhibitor used, tacrolimus or Neoral cyclosporine (35% versus 38%). Finally, prior antiviral therapy did not affect the treatment response (Table 2). Pretreatment variables associated with SVR were donor age, baseline viremia, bilirubin levels, and fibrosis. More specifically, older donor age, presence of cirrhosis, high bilirubin levels, and high viremia were associated with poorer SVR rates. Although patients with genotype 2/3 were more likely to achieve SVR than those infected with genotype 1 (53% versus 34%), the difference did not reach statistical significance. In addition, on-treatment variables associated with SVR were viral load at 1 and 3 months and treatment adherence. EVR (defined as a decline in the viral load higher than 2 logs from baseline) was significantly associated with SVR, so that SVR was achieved in 49% of those with EVR versus 9% of those without EVR (P 0.001). RVR (defined as negative viremia 1 month after therapy was started) was also associated with SVR. This analysis was, however, based on only 70 patients for whom a serum sample at this time point was available. SVR rates were not significantly affected by transient dose reductions of either drug alone (Table 3). However, dose reductions or premature terminations of therapy that resulted in the patient not reaching 80% of the recommended dose of both drugs during 80% of the recommended duration resulted in lower response rates. The SVR rate among patients achieving at least 80% of the recommended dose of Rbv was 49% versus 27% for patients who were unable to achieve this criterion (P 0.02). The remaining variables that were analyzed, including gender, age, pretransplantation antiviral therapy or history of significant alcohol consumption, Child-Pugh classification and hepatocellular carcinoma at transplantation, laboratory tests and body mass index at initiation of therapy, diabetes, time from transplantation to therapy, use of colony growth factors, and past rejection episodes, had no significant effect on SVR rates (Tables 2 and 3). When only genotype 1 infected patients were analyzed (n 92), variables associated with SVR were donor age (P 0.01), baseline viremia (P 0.06), baseline bilirubin levels (P 0.45), RVR (P 0.01) and EVR (P 0.03), and treatment adherence (accumulated dose ofrbv greater than 80% of full expected treatment, P 0.03; premature discontinuation of PegIFN, P 0.09).Although the response tended to be worse for

5 742 BERENGUER ET AL. TABLE 2. Baseline and On-Treatment Predictive Factors of SVR SVR (n 39) NR (n 68) P Value Type of therapy Peg 2a (n 66) 26 (39%) 27 (61%) NS Peg 2b (n 41) 13 (32%) 18 (68%) Genotype Genotype 1 (n 92) 31 (34%) 61 (66%) NS Not genotype 1 (n 15) 8 (53%) 7 (47%) Viral load at baseline (IU/L) ( to ( to ) ) Cholestatic hepatitis pre-therapy Yes (n 11) 4 (36%) 7 (64%) NS No (n 96) 35 (36.5%) 61 (63.5%) Gender Male (n 77) 28 (36%) 49 (64%) NS Female (n 30) 11 (37%) 19 (63%) Age at therapy (years) 54 (37 66) 56 (38 69) NS Body mass index at therapy (kg/m 2 ) 25.5 (17 32) 26 (18 38) NS Donor age (years) 46 (12 69) 53 (17 78) 0.03 Fibrosis pre-therapy F0-F1 (n 35) 16 (46%) 19 (54%) NS F3-F4 (n 70) 22 (31%) 48 (69%) Cirrhosis pre-therapy Yes (n 22) 4 (18%) 18 (82%) No (n 83) 34 (41%) 49 (59%) Severe necroinflammatory grade pre-therapy 15 (40%) 28 (44%) NS (n 43) Antiviral therapy in the past (before transplantation) Yes (n 32) 11 (34%) 21 (66%) NS No (n 75) 28 (37%) 47 (63%) Antiviral therapy post-lt Naïve (n 90) 33 (37%) 57 (63%) NS Retreatment (n 17) 6 (35%) 11 (65%) Immunosuppression Cyclosporine (n 42) 16 (38%) 26 (62%) NS Tacrolimus (n 65) 23 (35%) 42 (65%) Time from transplantation to therapy (years) 608 ( ) 595 ( ) NS RVR at 1 month* Yes (n 19) 13 (68%) 6 (32%) No (n 51) 10 (20%) 41 (80%) EVR at 3 months Yes (n 70) 34 (49%) 36 (51%) No (n 23) 2 (9%) 21 (91%) Erythropoietin use Yes (n 56) 20 (36%) 36 (64%) NS No (n 51) 19 (37%) 32 (63%) GCSF use Yes (n 44) 15 (34%) 29 (66%) NS No (n 63) 24 (38%) 39 (62%) Bilirubin pre-therapy (mg/dl) 1.1 ( ) 1.4 (0.5 24) Pretreatment glycemia (mg/dl) 101 (74 183) 105 (71 380) NS Abbreviations: EVR, early virologic response; GCSF, granulocyte colony stimulating factor; LT, liver transplantation; NR, nonresponse; NS, not significant; Peg 2a, pegylated interferon alfa 2a; Peg 2b, pegylated interferon alfa 2b; RVR, rapid virologic response; SVR, sustained virologic response. *RVR was determined in only 70 cases. patients with advanced disease (cirrhosis and cholestatic hepatitis) compared to those with mild recurrence (fibrosis 0 or 1), the difference did not reach statistical significance (data not shown). Multivariate Analysis RVR was not included in the multivariate analysis because 35% of patients had this value missing. The only

6 PEGIFN AND RIBAVIRIN IN HCV LIVER TRANSPLANT RECIPIENTS 743 TABLE 3. Impact of the Dose and Duration of Combination Antiviral Therapy on SVR SVR NR P Value Duration of therapy (days) 365 (81 623) 338 (16 618) NS Early discontinuation Yes (n 40) 9 (22.5%) 31 (77.5%) No (n 67) 30 (45%) 37 (55%) Premature IFN discontinuation Yes (n 37) 8 (22%) 29 (78%) 0.01 No (n 69) 31 (45%) 38 (55%) Premature Rbv discontinuation Yes (n 43) 10 (23%) 33 (77%) 0.01 No (n 63) 29 (46%) 34 (54%) Dose reductions Yes (n 58) 19 (33%) 39 (67%) NS No (n 49) 20 (41%) 29 (59%) IFN dose reduction Yes (n 23) 6 (26%) 17 (64%) NS No (n 84) 33 (39%) 51 (61%) Rbv dose reduction Yes (n 50) 17 (34%) 33 (66%) NS No (n 57) 22 (39%) 35 (61%) Initial Rbv dose Full (n 65) 24 (37%) 41 (63%) NS Reduced (n 42) 15 (36%) 27 (64%) Rbv dose 80% recommended (n 47) 23 (49%) 24 (51%) % recommended (n 59) 16 (27%) 43 (73%) Abbreviations: IFN, interferon; NR, nonresponse; NS, not significant; Rbv, ribavirin; SVR, sustained virologic response. variables that remained in the model were EVR at 3 months [odds ratio (OR): 0.08, 95% confidence interval (CI): , P 0.002] and donor age (OR: 1.039, 95% CI: , P 0.01). The same results were found when only patients with genotype 1 were analyzed (data not shown). Cohort effect In order to assess whether a potential learning curve had resulted in better SVR rates, we used 4 periods: (n 27), 2004 (n 23), 2005 (n 21), and 2006 to January 2007 (n 36). Surprisingly, the SVR rates had decreased over time (P 0.043), and the relapse rates had increased (P 0.052). The SVR rates achieved in , 2004, 2005, and were 48%, 43.5%, 33.5%, and 25%, respectively. In turn, the rates of viral relapse were 26%, 44%, 40%, and 60%, respectively. Interestingly, EVR was similar between the cohorts. In order to understand what might have negatively affected the efficacy of antiviral therapy, we compared treatment-related variables between 2 major periods ( and 2005 to January 2007). The SVR rate was 46% in those treated in the early cohort versus 28% in those treated more recently (P 0.05). Variables that significantly differed between these 2 periods were donor age, baseline disease severity, baseline viremia, and use of EPO. More specifically, donors of patients treated more recently were significantly older than those treated in the first period [53 (12-78) versus 45 (17-76) years, P 0.03]; only 36% of donors were older than 50 years in the first period as opposed to 61% in the second period (P 0.009). Baseline disease severity was more advanced among those treated more recently; indeed, the percentage of patients with baseline cirrhosis was greater (28% versus 12.5%, P 0.05), baseline bilirubin levels were higher [1.6 ( ) versus 1.1 (0.4-24) mg/dl, P 0.08], and leukocyte and platelet counts were lower [4500 ( ,900) versus 5400 ( ,000)/mm 3, P 0.05, and 108,000 (30, ,000) versus 122,000 (65, ,000)/mm 3, P 0.05]. In addition, both the use of EPO (34% versus 68%, P ) and baseline viremia [ ( to ) versus ( to ) IU, P 0.05] had significantly increased over time. A trend was observed for other variables without statistical significance being reached. For instance, the Child-Pugh score at the time of transplantation was higher in those treated more recently (Child score C: 38% versus 56%, P 0.09). In addition, the use of GSF had also increased over time (32% versus 49%, P 0.07). Finally, the duration between liver transplantation and treatment initiation was shorter in those treated in the first cohort versus the last cohort [504 ( ) versus 715 ( ) days, P 0.09]. Interestingly, besides the use of growth factors, all the variables related to treatment adherence, including early IFN and/or Rbv discontinuation rate, treatment reductions, full IFN and Rbv doses versus reduced

7 744 BERENGUER ET AL. doses at treatment initiation, and treatment duration, were similar between cohorts. There were no differences between cohorts in the remaining variables, including demographics, baseline immunosuppression, type of PegIFN, and genotype distribution (data not shown). In the multivariate analysis, we introduced all variables except for EPO and GSF use. Variables that had independently changed over time were donor age (OR: 1.041, 95% CI: , P 0.004), duration from transplantation to antiviral therapy (OR, 1.001, 95% CI: , P 0.013), and baseline leukocyte count (OR: 1.000, 95% CI: , P 0.034). DISCUSSION The need to optimize outcomes for hepatitis C infected recipients is one of the most pressing issues facing transplant physicians. 1,2 Antivirals have been used for this goal in patients with recurrent hepatitis C. 1-3,5,8-12 Results with PegIFN and Rbv have been shown to be better than those obtained with standard IFN-Rbv. Most studies have suggested that sustained viral eradication leads in the long term to histologic improvement, 15,16 reduced risk of developing decompensated allograft cirrhosis, and improved survival. 4,5 The response, however, appears to be lower in the liver transplant setting compared to that achieved in the immunecompetent population. Indeed, in a recent systematic review, SVR was achieved in only 29% to 31% of treated transplant recipients; this percentage was significantly lower than that reported in the immune-competent population treated with the same regimen. 7,17,18 Reasons for this lower response include a high prevalence of factors known to be associated with a lack of response in the nontransplant population (high viral load, high prevalence of HCV genotype 1, low tolerability with difficulties in achieving full-dose treatment, and high prevalence of prior nonresponders) 18 and, presumably, a lower response to HCV therapy in patients with impaired immune function. As in the immune-competent and human immunodeficiency virus HCV-coinfected populations, 17 there might also be a learning curve, particularly with respect to the management of adverse effects and the maintenance of optimal doses of Rbv. In addition, determining pretreatment and early on-treatment factors associated with viral clearance are extremely relevant aspects in the management of transplant recipients, so a potentially useful therapy is offered to a maximum of candidates but is not used or stopped at an early stage in those with low chances of success. Our experience covers several years of antiviral therapy and is based on a large number of treated patients at different stages of disease severity. The main conclusions from this large study based on a treatment-on-recurrence strategy may be summarized as follows: 1. HCV clearance is achieved in a third of HCV-infected liver transplant recipients treated with PegIFN-Rbv. 2. Recipients with advanced disease, particularly allograft cirrhosis, have a significantly lower chance of achieving SVR than those treated at earlier stages of disease. 3. A lack of EVR at 3 months of therapy is a very useful tool for predicting failure of therapy. 4. The type of calcineurin inhibitor used does not influence the outcome of antiviral therapy. 5. The type of PegIFN used does not influence the outcome of antiviral therapy. 6. A history of prior nonresponse to IFN therapy before transplantation does not influence the outcome following posttransplantation antiviral therapy. 7. The age of the donor is highly related to the outcome of antiviral therapy, so treatment failures are significantly more frequent in recipients of grafts from older donors. 8. The efficacy of antiviral therapy has decreased in recent years in our center; this is likely due to the increased donor age and the increased number of patients treated at advanced stages of disease. In our study, SVR was achieved in 34% of HCV genotype 1 patients treated with the PegIFN-Rbv combination and in 53% of those infected with genotype 2 or 3. Although the difference did not reach statistical significance, these results are in accordance with those reported in the literature for other populations, including human immunodeficiency virus HCV-coinfected patients. 7,17,18 It is interesting to note that the variables that have been shown to determine treatment outcome in the immune-competent population are the same in the liver transplant setting. In particular, RVR and EVR have been shown to have highly positive and negative predictive values, respectively, in a way similar to what is reported in the general population. 18 In addition, the observation that donor age is an independent predictor of response is analogous to the observations from many large studies in nontransplant patients with chronic hepatitis C infection, in which age at the start of therapy is an independent predictor of treatment outcome. 19 Potentially modifiable baseline variables such as immunosuppression, type of PegIFN, body mass index, and glycemia have not been found to have a significant effect on treatment outcome. Recently, some studies have suggested that cyclosporine may have an antiviral effect, 20 which may, in turn, increase the rate of viral clearance. We were, however, unable to show differences in treatment response among those immunosuppressed with cyclosporine and those who used tacrolimus. Likewise, a history of prior treatment with antivirals before liver transplantation did not appear to affect treatment success following antiviral therapy in the posttransplantation period. Although this is an interesting observation, it must be interpreted with caution because, in most cases, we were unable to assess whether the dose and duration of therapy in the past had been adequate. Finally, as recently shown in the nontransplant population, 21 the 2 approved PegIFNs achieved similar results in patients with recurrent hepatitis C.

8 PEGIFN AND RIBAVIRIN IN HCV LIVER TRANSPLANT RECIPIENTS 745 Also, similarly to what is reported in the immunecompetent population, adherence to therapy has been found to be a major factor that determines outcome. 18,19 Premature discontinuation of both IFN and Rbv was associated with reduced SVR rates in the univariate analysis. In addition, achieving more than 80% of Rbv adherence was significantly associated with SVR. Unfortunately, as reported previously in several studies assessing antiviral therapy in the liver transplant population, 1-3,5-14 the tolerance was problematic, with half of the patients requiring dose reductions, mostly of Rbv; in addition, therapy was prematurely discontinued in 37.5%. These results emphasize the importance of support to maximize treatment compliance with an IFN-based regimen. Strict monitoring and management of side effects, particularly hematologic cytopenias and psychiatric complications, might be useful in optimizing treatment outcomes. In our study, however, neither the use of EPO nor the use of GSF was associated with SVR or with treatment discontinuation. In fact, although the use of growth factors increased significantly with time, treatment response rates declined. This lack of effect of EPO on SVR has also been recently shown in the immune-competent population. In a 3-arm study by Shiffman and colleagues, 22 the use of EPO in all subjects at the initiation of antiviral therapy did not enhance the SVR rate when the same starting dose of Rbv was given. Only those who used a higher starting dose of Rbv achieved higher rates of SVR with lower relapse rates. 22 It is likely that the same will apply to the transplant patient, but this needs to be confirmed in prospective studies. The worse results achieved in recent years, not reported previously by any other group, are likely due to a combination of factors, particularly the increasing donor age and greater baseline disease severity. Indeed, both of these factors were significantly associated with treatment outcome, and both had substantially changed over time. In the immune-competent population, SVR rates are lower in patients with cirrhosis versus patients treated at earlier stages of disease. 18,19 In the transplant population, 2 studies 4,11 have also shown worse results in patients with advanced disease. On the basis of these findings, we would recommend treating patients at earlier stages of disease before they reach the stage of allograft cirrhosis. Donor age is one of the strongest factors influencing HCV-related disease progression in the transplant setting. 1,2 This is the first study to report that it is also a very strong factor determining the chances of antiviral treatment success. In fact, none of the variables analyzed except for the rate of SVR were significantly different between recipients of grafts from older ( 50 years) and younger donors (data not shown). In conclusion, our results indicate that response to PegIFN-Rbv is achieved in about one-third of treated hepatitis C liver transplant patients. The success of antiviral therapy, however, is strongly related to baseline disease severity and donor age, so poor results are obtained in recipients of grafts from old donors treated when they have established cirrhosis. On the basis of these findings, we strongly recommend that disease monitoring be performed at more frequent intervals and that antiviral therapy be started at early stages of disease in recipients of grafts from old donors, in whom the risk of progressive recurrent disease is higher and the chances of antiviral success are lower. REFERENCES 1. Berenguer M. Hepatitis C after liver transplantation: risk factors, outcomes, and treatment. Curr Opin Org Transplant 2005;10: Samuel D, Forns X, Berenguer M, Trautwein C, Burroughs A, Rizzetto M, et al. Report of the monothematic EASL conference on liver transplantation for viral hepatitis (Paris, France, January 12-14, 2006). J Hepatol 2006;45: Berenguer M. Treatment of hepatitis C after liver transplantation. Clin Liver Dis 2005;9: Picciotto FP, Tritto G, Lanza AG, Addario L, De Luca M, Di Costanzo GG, et al. Sustained virological response to antiviral therapy reduces mortality in HCV reinfection after liver transplantation. J Hepatol 2007;46: Berenguer M, Palau A, Aguilera V, Rayón JM, San Juan F, Prieto M. Clinical benefits of antiviral therapy in patients with recurrent hepatitis C following liver transplantation. Am J Transpl 2008;8: Berenguer M, Aguilera V, Prieto M, San Juan F, Rayón JM, Benlloch S, Berenguer J. Significant improvement in the outcome of HCV-infected transplant recipients by avoiding rapid steroid tapering and potent induction immunosuppression. J Hepatol 2006;44: Berenguer M. Systematic review of the treatment of established recurrent hepatitis C with pegylated interferon in combination with ribavirin. J Hepatol 2008;49: Samuel D, Bizollon T, Feray C, Roche B, Ahmed SNS, Lemonnier C, et al. Interferon- 2b plus ribavirin in patients with chronic hepatitis C after liver transplantation: a randomized study. Gastroenterology 2003;124: Toniutto P, Fabris C, Fumo E, Apollonio L, Caldato M, Avellini C, et al. Pegylated versus standard interferon- in antiviral regimens for post-transplant recurrent hepatitis C: comparison of tolerability and efficacy. J Gastroenterol Hepatol 2005;20: Berenguer M, Palau A, Fernandez A, Benlloch S, Aguilera V, Prieto M, et al. Efficacy, predictors of response, and potential risks associated with antiviral therapy in liver transplant recipients with recurrent hepatitis C. Liver Transpl 2006;12: Carrion JA, Navasa M, Garcia-Retortillo M, Garcia-Pagan JC, Crespo G, Bruguera M, et al. Efficacy of antiviral therapy on hepatitis C recurrence after liver transplantation: a randomized controlled study. Gastroenterology 2007;132: Oton E, Barcena R, Moreno-Planas JM, Cuervas-Mons V, Moreno-Zamora A, Barrios C, et al. Hepatitis C recurrence after liver transplantation: viral and histologic response to full-dose peg-interferon and ribavirin. Am J Transplant 2006;6: Neumann U, Puhl G, Bahra M, Berg T, Langrehr JM, Neuhaus R, et al. Treatment of patients with recurrent hepatitis C after liver transplantation with peginterferon alfa-2b plus ribavirin. Transplantation 2006;82: Sharma P, Marrero JA, Fontana RJ, Greenson JK, Conjeevaram H, Su GL, et al. Sustained virologic response to therapy of recurrent hepatitis C after liver transplantation is related to early virologic response and dose adherence. Liver Transpl 2007;13: Abdelmaleck MF, Firpi RJ, Soldevila-Pico C, Reed AI,

9 746 BERENGUER ET AL. Hemming AW, Liu C, et al. Sustained viral response to interferon and ribavirin in liver transplant recipients with recurrent hepatitis C. Liver Transpl 2004;10: Bizollon T, Admed SNS, Radenne S, Chevallier M, Chevallier P, Paraz P, et al. Long-term histologic improvement and clearance of intrahepatic hepatitis C virus RNA following sustained response to interferon-ribavirin combination therapy in liver transplant patients with hepatitis C recurrence. Gut 2003;52: Sulkowski MS. Management of hepatic complications in HIV-infectedpersons. JInfectDis2008;197(suppl3):S279 S Heathcote J. Antiviral therapy: chronic hepatitis C. J Viral Hepat 2007;14(suppl 1): Cainelli F. Hepatitis C virus infection in the elderly: epidemiology, natural history and management. Drugs Aging 2008;25: Watashi K, Hijikata M, Hosaka M, Yamagi M, Shimotohno K. Cyclosporin A suppresses replication of hepatitis C virus genome in cultured hepatocytes. Hepatology 2003; 38: Sulkowski M, Lawitz E, Shiffman ML, Muir AJ, Galler G, McCone J, et al. Final results of the ideal (individualized dosing efficacy vs flat dosing to assess optimal pegylated interferon therapy) phase IIIB study. J Hepatol 2008; 48(suppl 2):S Shiffman ML, Salvatore J, Hubbard S, Price A, Sterling RK, Stravitz RT, et al. Treatment of chronic hepatitis C virus genotype 1 with peginterferon, ribavirin, and epoetin alpha. Hepatology 2007;46: