Recurrent HCV infection after liver transplantation is universal and sustained clearance of HCV-RNA rarely occurs. The aim of this study was to characterize cell-mediated immunity and cytokine production in HCV-infected patients after liver transplant. The study included 6 pretransplantation patients (PT) and 15 liver transplanted patients, including 5 with spontaneous HCV-RNA clearance (SC group), 5 with sustained virological response after antiviral treatment (SVR group), and 5 no response (NR group). The control group included 5 HCV-RNA negative, anti-HCV negative healthy individuals. This study examines proliferative T-cell response and cytokine production (gamma-interferon and IL-10) after HCV specific and phytohemagglutinin (PHA) stimulation in cultured peripheral blood mononuclear cells (PBMCs) from each group. Multispecific proliferative responses to HCV antigens (mean Stimulation Index; SI) were higher in the SVR group (mean SI 7.4 ± 2) and SC group, as compared with the NR group (P < .05, vs SVR) and PT group (P < .05, vs SVR and SC). After PHA stimulation, gamma-interferon levels were similar to controls (4330 ± 640 pg/ml) in the SC (4474 ± 300 pg/mL) and SVR groups (3647 ± 300 pg/mL), but were significantly lower than controls in the PT (401 ± 331 pg/mL; P < .02) and NR groups (546 ± 360 pg/mL; P < .01). IL-10 production after PHA stimulation was similar in SC, SVR, and controls (647 ± 279 pg/mL, 674 ± 310 pg/mL and 841 ± 294 pg/mL, respectively), but was lower in PT patients (232 ± 94 pg/mL). The NR group showed high basal IL-10 production with little increase after stimulation. In conclusion, liver post-transplantation patients with spontaneous clearance of HCV-RNA and those with sustained viral response after therapy showed an immune response despite immunosuppression that might have contributed to their favorable outcome. (Liver Transpl 2004;10:584–594.)
End-stage liver disease associated with hepatitis C virus (HCV) infection is the most common indication for liver transplantation and accounts for 50% of these procedures in Spain.1 Unfortunately, HCV recurrence is universal and is associated with substantial morbidity, mortality, and graft loss.2, 3 In contrast to immunocompetent individuals, HCV infection in immunosuppressed transplant recipients usually has an accelerated course. Medium-term studies indicate that most recipients have significant fibrosis and about 30% have cirrhosis by 5 years after liver transplantation.4
Studies in the general population have shown that liver damage due to HCV infection occurs in the context of an immune response.5, 6 Spontaneous clearance of HCV, with HCV-RNA testing negative after years of follow-up, has been reported sporadically.7 The host immune response plays a critical role in controlling HCV replication and liver damage.5 CD4+ and CD8+ T-cell responses may be involved in HCV-RNA clearance either by a direct cytolytic effect on infected cells or by induction of cytokines. Moreover, some studies have shown that the intrahepatic cytokine milieu tends toward a Th1 subtype profile.8, 9 In the posttransplantation setting, immunosuppression can modify HCV pathogenic pathways, changing the efficacy of immune response. The temporal evolution of the different immune mechanisms is another important factor to take into account. Mechanisms that initially predominate may become less important over time, and new immune responses may emerge that modify the host-virus interaction.3 Therapeutic clinical trials in chronic hepatitis C in the general population have shown that alpha-interferon (α-IFN) given with ribavirin (combination therapy) results in a sustained viral response (SVR) in 31 to 64% of patients, depending on viral genotype, viral load, and histological fibrosis stage.10 The mechanisms of action of these drugs, and particularly their effects on T-cell response, are not completely understood; however, their immunomodulatory properties and direct antiviral effects, especially those of α-IFN, are well recognized.11
The observation at our institution of a subgroup of transplanted patients spontaneously clearing virus after transplant or achievement of SVR in patients treated with interferon-based therapies prompted us to analyze the relevant characteristics of the immune response in these groups of patients. For this purpose, proliferative T-cell response and cytokine production in response to HCV-specific and mitogenic stimuli were studied in these patients and compared with those of other patients from the same cohort who did not respond to treatment. In addition, results for these factors were compared with those from chronic HCV-infected pretransplanted patients and healthy HCV-negative controls.
HCV, hepatitis C virus; anti-HCV, antibody to hepatitis C virus; HBV, hepatitis B virus; SC, spontaneous clearance; SVR, sustained virological response; NR, no response to antiviral treatment; PBMCs, peripheral blood mononuclear cells; PHA, phytohemagglutinin; α-IFN, alpha-interferon; γ-IFN, gamma-interferon.
Patients and Methods
Between 1990 and 2001 in our center, 278 liver transplantations were performed in HCV-RNA+ patients. Among them, 10 patients had spontaneously cleared HCV infection, with persistently negative serum HCV-RNA after a mean follow-up of eight years. Out of the 14 treated patients, 6 had long-term SVR.12
Six pretransplant patients (PT group) (Table 1) and 15 liver-transplanted patients (Table 2), including 5 with spontaneous clearance (SC group), 5 with sustained viral response after antiviral treatment (SVR group), and 5 with no response after treatment (NR group), were studied. The control group included 5 healthy individuals.
Table 1. Characteristics of Pretransplant Chronic HCV-infected Patients
According to clinical criteria and transaminase levels, several nonprotocolized liver biopsies were performed, in the liver-transplanted patients, during follow-up monitoring. Recurrent hepatitis C in the absence of rejection was defined as the time of histological diagnosis. Other causes of hepatitis or biliary tract pathology were ruled out with appropriate studies. Liver biopsy was mandatory before antiviral treatment in order to evaluate liver damage (grade and stage) and to detect other transplant complications. Control liver biopsies were performed after antiviral treatment had ended.
Biopsy specimens were stained with hematoxylin-eosin and Masson's trichrome. The degree of inflammatory activity (grade) and fibrosis (stage) were assigned following the method described by Scheuer et al.13
Characteristics of these groups
The PT group consisted of 6 HCV-RNA positive (5 genotype 1b and 1 genotype 2), anti-HCV-positive chronically infected patients on the waiting list for a liver transplant. There were 4 men and 2 women, aged 38–62 years (mean, 54 years).
The SC group was composed of 5 patients randomly selected from a larger group that had cleared HCV-RNA without antiviral treatment. All patients had been genotyped as 1b before transplantation. There were 2 men and 3 women, aged 39–60 years old (mean, 52 years). Three of them had concomitant events that might have contributed to viral clearance: de novo HBV infection (Fig. 1), hydroxyurea treatment (Fig. 2), and CMV infection and recovery from severe acute cholestatic hepatitis (Fig. 3). No related factors have been documented in the other two patients.
The SVR group consisted of 5 transplanted patients, 1 woman and 4 men, aged 39–61 years old (mean, 49), who showed virus clearance after antiviral treatment. The percentage of response in the total group was 42.8%.
The NR group included 5 patients with persistent infection after antiviral therapy. One woman and four men with an age range of 40–52 years (mean, 47) were included.
The 5 control subjects were anti-HCV-negative, HCV-RNA-negative, healthy individuals, 2 men and 3 women, aged 28–58 years old (mean, 41), with normal serum ALT levels and no biochemical evidence of liver disease.
The liver transplant donors were anti-HCV-negative. In the absence of clinical indication, liver biopsy was not performed. All the subjects gave written informed consent to participate in the study, which was approved by the local ethics committee.
Hepatitis C antibody was determined by enzyme immune assay (AxSYM® HCV, v 3.0 Abbott, Division Diagnostics, Wiesbaden, Germany). Antibody specificity was identified by RIBA SIA 3.0 (Chiron Corp., Emeryville, CA). Serum qualitative HCV-RNA was determined by the COBAS Amplicore HCV Monitor test 2.0 (Roche Diagnostics, Basel, Switzerland), which detects approximately 50 copies of HCV per milliliter. Genotyping was done by in-house methodology based on restriction fragment length polymorphism, as described by Davidson.14 HCV genotypes were classified according to the nomenclature of Simmonds et al.15
After informed consent was given, patients meeting the following criteria were invited to receive antiviral treatment: age range 18–65 years, a minimum of 3 months since transplantation, leukocyte count higher than 3 × 109/L, platelets higher than 100 × 109/L, hemoglobin concentration greater than 12 gr/mL (men) or 11 gr/mL (women). Exclusion criteria were as follows: cirrhosis, recent rejection episode, alcohol abuse, tumor, serum creatinine higher than 200 mmol/L, HBV carrier status or autoimmune disease. Monotherapy immunosuppression with one calcineurinic was mandatory. Only 14 patients met all the criteria. Antiviral treatment has been available in our institution only since 1999, so the patients were treated at a mean of 31 ± 27 months after transplant.
Patients in the SVR and NR groups (Table 2) were given a combination of subcutaneous α-IFN 2a (Roferon A, Productos Roche SA, Madrid, Spain) at a dose of 3 million units three times weekly and oral ribavirin (Roche 20-9963, Productos Roche SA, Madrid, Spain) at 1g per day (2 doses of 400mg and 600mg) for 12 months.
Virological Response to Treatment
Sustained viral response to therapy was defined as serum clearance of HCV-RNA by 24 weeks after the end of therapy. Nonresponders had detectable serum HCV-RNA levels throughout treatment.
Recombinant HCV proteins covering the core, NS3, NS4, and NS5 regions were purchased from Biodesign (Saco, ME). These proteins were derived from HCV genotype 1a, coded for HCV (core amino acid 2–192), NS3 (amino acid 1450–1643), NS4 (amino acid 1658–1863), and NS5 (amino acid 2322–2423), and expressed as carboxyl terminal fusion proteins with beta-galactosidase in Escherichia coli. E. coli extracts and beta-galactosidase were tested as controls for nonspecific stimulation in each proliferation assay. Purity of the antigens was higher than 90%.
Peripheral blood mononuclear cells (PBMCs) were isolated on Ficoll-Isopaque gradients (Pharmacia-Biotech, Upsala, Sweden). Cells were washed and then cultured at 106 cells/mL in assay media, consisting of RPMI-1640 supplemented with 10% fetal bovine serum and L-glutamine (all supplied by Gibco, Grand Island, NY). PBMCs were resuspended in RPMI 1640, plated at 2 × 105/200 μL/well, and stimulated with 1 μg/mL of recombinant HCV and control proteins. All experiments were done in 4 replicate wells of 96-well U-bottoned plates (Nunc, Roskilde, Denmark). Concurrently, to assess overall T-cell responsiveness in each subject, PBMCs were stimulated in the same conditions with phytohemagglutinin (PHA) at 10 μg/mL (Sigma Chemical Co., St. Louis, MO).
The proliferative response was evaluated by [3H] thymidine uptake during the last 20 hours of culture (Amersham, Buckinghamshire, UK) and measured in counts per minute (cpm) by a beta-counter (Betaplate, Wallac LKB Instrument, Sweden). The Stimulation Index (SI) was calculated by dividing the cpm value found in the presence of antigen by that found without antigen. We established a cutoff value of 1.3, which was 3 standard deviations higher than the mean SI after HCV-specific stimulation in negative controls.
After 24 hours, cell-free supernatants from nonstimulated (basal) and stimulated cells were collected, aliquoted, and stored at −80°C until cytokine analysis. Previous kinetic studies showed that the best time to detect maximum cytokine release was between 18 and 40 hours.
Interleukin measurements were performed using the Cytokine Bead Array kit (Becton Dickinson, San Jose, CA) following the manufacturer's instructions. Briefly, a mixture of 50 μL of each antibody-bead reagent and detector antibody-PE reagent was added to 50 μL of cell supernatants, activated with different recombinant HCV proteins or mitogens, and calibrators, and incubated for 3 hours at room temperature prior to washing and data acquisition using flow cytometry.
Flow cytometric analysis was performed using a FCASCalibur system (Becton Dickinson; Madrid, Spain). Data were acquired and analyzed using Becton Dickinson Cytometric Bead Array software. Cytokine concentrations were determined by comparison with the standard curves from known concentrations run simultaneously. Specific cytokine production was calculated as the difference between cytokine production after stimulation and the basal production (corrected value).
Results of proliferative responses were expressed as mean ± SEM and analyzed by the Student's t test. The nonparametric Mann-Whitney test was used to compare levels of interleukin production. A P value < .05 was considered statistically significant. All statistical procedures were performed using the GraphPad Prism 3.0 software package (San Diego, CA).
Clinical Outcome of Transplanted Patients
Spontaneous clearance patients
All patients from the SC group were alive at the time of writing, at a range of 133 to 32 months (mean, 76.2) after transplantation. Three patients have concomitant events that may have contributed to viral clearance and immunological enhancement (Fig. 1–3). The time between transplantation and the diagnosis of HCV reinfection was documented only in Patient 3, and occurred at three months. In the early phase after liver transplantation, these patients were treated with a quadruple immunosuppressive regimen: antilymphocytic immunoglobulin (ATGAM, Lab Pasteur-Merieux, Lyon, France), cyclosporine A (CsA) plus corticosteroids, and azathioprine (stepwise reduction). None of them have received tacrolimus (FK) immunosuppression at any time (Table 2). There were no differences in blood levels of the individual immunosuppressive agents among these patients during the period of study.
Antiviral Treated Patients (SVR and NR groups)
The 10 patients with histological diagnoses of recurrent HCV were treated 31 ± 27 months after transplantation, with α-IFN 2a and ribavirin. Five patients experienced SVR and 5 patients did not respond to treatment (Table 2). During antiviral therapy, immunosuppressor treatment consisted of a single calcineurin inhibitor, including CsA in all SVR patients, and CsA in 2 and FK in 3 of the NR patients.
Response to HCV antigens
HCV-specific CD4+ T-cell responses were studied in all the groups and controls by measuring the level of PBMC proliferation in response to HCV core, NS3, NS4, and NS5 proteins relative to the control beta-galactosidase protein. The patient treated with hydroxyurea (SC group) was excluded from this analysis because of the antiproliferative treatment.
Despite the patients' immunosuppressed status, proliferative responses to all specific HCV proteins were detectable. The SVR group had the strongest mean response (SI, 7.4 ± 2), which was significantly higher than that of the NR group (SI, 1.6 ± 0.3; P = .02) and the PT group (SI, 1.5 ± 0.3; P = .03) (Fig. 4). In the SC group, mean proliferative response was also higher than that of both the NR and PT groups (SI, 3.7 ± 1.2), but differences were only statistically significant as compared with the latter group (P = .01) (Fig. 4). The mean SI value in the control group was 1.0 ± 0.1.
Response to phytohemagglutinin
PBMCs from all patient groups responded similarly to PHA challenge (mean SI values for PT, SC, SVR, and NR were 4.5 ± 2, 3.9 ± 0.7, 9.9 ± 7, and 4.9 ± 1.6, respectively). However, as was expected in the context of immunosuppressive therapy, SI values were much lower in all these patient groups than those in the control group (mean SI, 31 ± 17) (Table 3).
Table 3. Proliferation Assay and Cytokine Production Results
Mean SI After HCV-Antigen Stimulation
Mean SI After PHA Stimulation
Mean γ-IFN Corrected Value After PHA Stimulation (pg/mL)
To gain further insight into T-cell responsiveness against HCV recurrence after liver transplantation, we assessed cytokine production by HCV antigen-stimulated and PHA-stimulated PBMCs. gamma-interferon and IL-10 production was measured in the supernatants of cultured cells from each patient group and the controls.
Response to HCV antigens
Only two patients, both from the SC group, showed production of gamma-interferon (γ-IFN) and IL-10 after stimulation with HCV recombinant antigens. As occurred with the proliferation assay, the response was multispecific, because high cytokines values were obtained with all the HCV antigens. No other measurable responses were found in any of the other groups.
Response to phytohemagglutinin
After PHA stimulation γ-IFN production increased to a similar degree in the SC group (4474 ± 300 pg/mL), in the SVR group (3647 ± 300 pg/mL) and in the controls (4330 ± 640 pg/mL). However, γ-IFN production was significantly lower in the PT patients (401 ± 331 pg/mL; P < .02 vs controls) and in the NR group (546 ± 360 pg/mL; P < .01 vs controls) (Table 3 and Fig. 5, Panel A).
In contrast to γ-IFN findings, we observed significantly higher basal IL-10 production in the NR patients (802 ± 410 pg/mL) than in the controls (183 ± 31 pg/mL; P = .04). This fact could be an indication that PBMCs were pre-activated in vivo in these patients. Their response to PHA stimulation (891 ± 428) was, therefore, difficult to evaluate because it was likely that maximum secretion capacity had already been reached in the culture medium. No significant differences were found in basal IL-10 levels in any of the other groups as compared with controls (SVR 112 ± 14 pg/mL, SC 268 ± 104 pg/mL and PT 37 ± 7 pg/mL) (Table 3 and Fig. 5, Panel B), allowing us to analyze the corrected IL-10 value (the difference between basal and PHA-stimulated values). After PBMCs activation, the IL-10 corrected value in the SC (647 ± 279 pg/mL) and SVR groups (674 ± 310 pg/mL) was similar to that of the control group (841 ± 294 pg/mL). The PT group, however, showed much lower IL-10 production after PHA-stimulation (232 ± 94 pg/mL; P = .12) (Table 3 and Fig. 5, Panel B).
Recurrent hepatitis C after liver transplantation is a major problem, occurring in 95–100% of patients.2, 3 HCV-RNA clearance has been sporadically reported in this setting,16 suggesting that a small number of patients are able to develop an effective immune response against recurrent HCV infection in the graft. Several favorable prognostic factors, such as donor age,17 the absence of rejection episodes,18 and geographical differences19, 20 have been described to explain this good outcome. The balance between immunosuppression and virological immune response is crucial with regard to viral persistence and liver damage.
Because therapeutic studies involving antiviral agents to treat recurrent HCV have reported generally poor results,21 studies focusing on the infrequent patients showing favorable outcome could provide worthwhile information.
Techniques to measure the proliferative response of PBMC to challenge with HCV antigens or PHA and assays to determine cytokine production are frequently used as barometers of mononuclear functional integrity. In this study we analyze T-cell response in various groups of HCV-infected recipients of liver transplants to investigate the role of the immune response in long-term outcome. Our results show some important differences between the clinical groups.
First, patients with sustained viral clearance after antiviral treatment had a significantly higher proliferative response to all HCV antigens than treatment nonresponders and chronically infected pretransplant patients. Second, there was a significant increase in γ-IFN production after PHA stimulation, both in patients who cleared spontaneously and those who cleared after antiviral treatment. Lastly, PBMCs from patients who did not respond to treatment produced high levels of IL-10 even before in vitro stimulation, suggesting the presence of pre-activated cells in vivo.
Studies in nontransplanted patients have documented a strong, multispecific immune response in HCV antibody-positive patients without detectable viremia and in patients who cleared the virus spontaneously,22, 23 but not in those with chronic HCV infection and persistent viremia.24
γ-IFN has been related with a noncytopathic antiviral pathway, a mechanism that may represent an important host survival strategy to control viral infections, especially when vital organs are massively infected.25, 26 The active induction of cytokine within the liver by agents (viral or otherwise) may contribute to clearing the HCV.
Among the five patients with spontaneous clearance, one had HBV superinfection. Concomitant infection with HBV can induce an immune response able to cross-clear HCV, as has been described in murine models27 and in humans.28–31 In another patient from the SC group, cholestasis associated with a rejection episode may have induced local production of Th1 cytokines in the liver, leading to viral clearance. However, it cannot be ruled out that the presence of CMV superinfection in this case might have contributed to the immune response observed. CMV infection has been described as a poor prognostic factor in chronic HCV carriers,32 whereas HCV-RNA clearance in cholestatic hepatitis was reported in a liver-transplanted patient without antiviral treatment.33 Another SC patient received hydroxyurea treatment. Besides its antiproliferative action, hydroxyurea has well-recognized antiviral activity, as demonstrated in anti-HIV infection protocols.34 Although no reference to this effect has been made in HCV infection, activity of hydroxyurea in HCV cannot be ruled out. No other related intercurrent factors were recorded in the other 2 SC patients included in the series.
IL-10 was originally described as a cytokine with an inhibitory action by virtue of its ability to inhibit the production of several cytokines through Th1 clones.35, 36 IL-10-producing T cells have been detected in a higher proportion of patients with chronic infection than in those who had cleared the virus.37 We found that PBMCs from patients who did not respond to treatment displayed high basal IL-10 production, suggesting the presence of pre-activated cells in vivo. High IL-10 levels may have led to a low proliferative response that might have prevented viral clearance in these patients.
γ-IFN and IL-10 production after stimulation with HCV recombinant antigens was observed in only two patients, both belonging to the SC group, suggesting that these patients have developed a high number of memory HCV-specific T-cells. Thus, there could be a relationship between persistence of memory cells and sustained virological clearance after liver transplantation. The infrequent occurrence of this response may be explained by the extremely small number of HCV-specific lymphocytes compared to the total pool.
Because the PT patients were not immunossuppresed, high response would be expected in both the proliferation and cytokine production assays. Instead, immune response in this group was similar to that of the NR patients. This may indicate that chronic HCV infection in advanced liver disease affects immune response to a greater degree than immunosuppressive treatment does. Together with findings from other authors,38 our study performed long-term posttransplantation suggests that some transplanted patients are able to mount an immune response that might help to clear HCV-RNA even while under immunosuppressive monotherapy. Furthermore, our data indicate that persistence of T-cell reactivity is important for the control of HCV replication.
We thank Ms. Celine L. Cavallo for English language assistance.