It is estimated that hepatitis C (HCV), a single-stranded RNA virus, infects over 100 million people worldwide. In 1991, when screening for HCV infection first became widely available, 17.8% of liver transplants were carried out for end-stage liver disease associated with HCV infection. This proportion has increased every year since 1991 and is currently over 30%, making end-stage liver disease associated with HCV infection the most common indication for liver transplantation (LT) in the United States and Europe.
Five-year patient and graft survival for HCV-infected recipients is similar to that of other indications (Figure 1). However, in contrast to the other leading causes of decompensated liver disease, such as alcohol, cholestatic liver diseases and cryptogenic cirrhosis, post-transplant recurrence of HCV infection, as defined by detection of HCV RNA, is nearly universal (1–6). With follow-up periods of up to 10 years, eventual histological recurrence of hepatitis C, depending upon the necroinflammatory activity and fibrosis criteria for a particular study, has been reported to occur in 42–88% of HCV-infected recipients. The great majority of studies reporting the frequency of histological recurrence of HCV have relied upon biopsies obtained in response to abnormal liver biochemistries. A recent study reporting the histological progression of post-transplant HCV infection, in which biopsies were obtained regardless of biochemical profile, found that almost half of HCV-infected liver transplant recipients develop histological evidence of recurrence (defined by histology activity index ≥ 3 and/or fibrosis stage ≥ 2) within the first postoperative year (7). One in five recipients who developed histological evidence of recurrent hepatitis C in the National Institutes of Health-sponsored Liver Transplant Database either died or required retransplantation due to hepatitis C-induced allograft failure within the first 5 postoperative years (8). In other words, ≈ 10% of patients transplanted for hepatitis C will die or lose their allograft secondary to recurrent disease in the medium term. Reports of an increasing prevalence of cirrhosis among HCV-infected recipients transplanted more recently (post 1994) are of particular concern.
Whether a subset of HCV-infected recipients exists who are at increased risk for poor patient or graft survival continues to be a subject of substantial debate. Several analyses have identified viral and host variables associated with more severe post-transplant HCV infection. A summary of variables known to be associated with more post-transplant mortality with or without graft loss is presented in Table 1.
Table 1. : Recipient variables associated with increased relative risk of mortality or graft loss
a Relative risk (RR) of 1.05 is per recipient year of age (8,16).
Reports of the impact of HCV genotype on outcomes in HCV-infected recipients have been varied. A recent report by the European Concerted Action on Hepatitis (EUROHEP), based upon combined results of liver transplantation for hepatitis C from 15 European centers between 1984 and 1995, reported that HCV genotype 1b is associated with a higher rate of recurrent hepatitis and that the development of cirrhosis portends a very poor prognosis (fully three-quarters of recipients with histologically demonstrated cirrhosis died during the follow-up period) (9). Conversely, a large prospective multicenter North American study found no association with graft survival or histological recurrence and genotype (8). It is likely that differences in case definitions, biopsy protocols (or the lack thereof), genotyping methodology (e.g. whether genotype is assigned through comparison with the 5′ untranslated vs. the NS5 region), length of follow-up and selection bias contribute to the heterogeneity in the results regarding the impact of HCV genotype on histological recurrence of HCV infection. Even in the studies that found HCV genotype 1b to be associated with more severe chronic allograft injury (10,11), no association was found with overall graft loss or patient mortality. If HCV genotype is associated with more severe allograft injury, this should become manifest by an increased frequency of graft failure in 1b-infected recipients. To date, such an association has not been demonstrated. The impact of HCV quasi-species emergence has not been studied sufficiently to determine the importance of viral mutations on disease recurrence.
Race or ethnicity has not been widely considered to be an important factor in patient or graft survival after liver transplantation. However, in analyses of the national registry on liver transplantation conducted by the United Network for Organ Sharing, race has been found to correlate with outcome, in both children and adult recipients. Thus, among adults, overall 6-year survival in UNOS was reported to be 64% in whites, 67% in Hispanics, 53% in blacks, and 47% in Asians. In the National Institutes of Digestive Diseases and Kidney (NIDDK) Liver Transplantation Database, race was not associated with survival in the overall cohort, but was associated with survival among patients with hepatitis C. Multivariate analyses showed that factors independently associated with outcome were age, recipient race, Child–Pugh score and location before transplantation.
Higher pre- and post-transplant viral loads are associated with more severe histological recurrence of hepatitis C in both the transplant and nontransplant settings. Forms of immunosuppression that enhance viral replication or attenuate viral clearance are thus likely to be associated with more severe recurrence of hepatitis C. The relative impact of commonly utilized immunosuppressants upon HCV viremia and post-transplant outcomes, including rejection, are discussed below.
Calcineurin inhibitors. Cyclosporine A has no impact on viral levels when administered to HCV-infected patients in the nontransplant setting. Although no studies of the independent impact of tacrolimus on HCV viremia have been carried out, levels of post-transplant HCV viremia are similar among patients receiving tacrolimus and steroids and those receiving cyclosporine A and steroids. Increased allograft rejection in patients infected with HCV has been alternately associated with the use of tacrolimus and cyclosporine. Most studies have found no difference. On balance, there are no compelling data to suggest an advantage in using either tacrolimus or cyclosporine A in patients undergoing liver transplantation for hepatitis C.
Corticosteroids. In the nontransplant setting, corticosteroids are known to increase levels of hepatitis C viremia (12) and to be associated with more severe patterns of histological injury (13). A similar impact has been demonstrated among liver transplant recipients. Gane et al. (14), in a comprehensive study, reported the effects of corticosteroids on post-transplant levels of viremia and graft injury. The principal findings were:
1 pulsed intravenous methylprednisolone therapy is associated with transient four- to 100-fold increases in HCV RNA levels;
2 methylprednisolone therapy for acute cellular rejection is associated with an increased frequency of the subsequent development of acute hepatitis;
3 HCV RNA levels increase steeply (between two and 20 times) during episodes of acute lobular hepatitis and decrease with improved graft function;
4 higher HCV RNA levels are associated with increased histological severity of graft injury.
Similarly, Sheiner et al. (15) found a strong correlation between the number of rejection episodes, and thus exposure to corticosteroids, and histological recurrence of hepatitis C. Among patients studied in the NIDDK Liver Transplant Database, higher average daily steroid doses were independently associated with increased mortality and graft loss. In the NIDDK study, recipients with a mean daily steroid dose in the first 42 days post-transplantation of ≥ 100 mg had a significantly increased risk of death (relative risk = 2.7, p = 0.04) when compared with recipients with an average daily dose of < 50 mg. The relationship between rejection treatment and patient survival among HCV-infected recipients was also reported in the NIDDK Liver transplant Database study. Compared with patients not treated for rejection, being treated for acute cellular rejection was independently associated with increased mortality (relative risk = 2.9, p = 0.03) (16). This is in contrast to non-HCV infected recipients, for whom a single treated episode of acute cellular rejection is associated with diminished mortality (relative risk = 0.58, 95% CI = 0.39–0.86, p = 0.007). While the basis of the increased risk of patient mortality associated with acute cellular rejection among HCV-infected recipients is not clear, the two most common causes of death among HCV-infected recipients are non-HCV infections and allograft failure secondary to HCV recurrence. Although graft failure secondary to HCV occurred more commonly in the NIDDK study among recipients with higher pretransplant HCV RNA levels, this accounted for less than half of the excess mortality and graft loss experienced by these recipients, in whom death and graft loss due to non-HCV infections were also more common (19.8 vs. 3.7%, p = 0.003). It is possible that HCV recurrence either mimics or causes allograft rejection, resulting in greater exposure to immunosuppressive agents and, thereby, increased susceptibility to serious infections and/or HCV-induced allograft injury.
Azathioprine. Although azathioprine, in combination with corticosteroids, is known to be associated with increased HCV RNA levels in the nontransplant setting, the impact on post-transplant HCV RNA levels is not known. The NIDDK Liver Transplant Database found no variation in histological grade of recurrence, mortality or graft loss among patients who received tacrolimus or cyclosporine and steroids alone or with concomitant azathioprine.
OKT3. Because of the inherent overlap between OKT3, rejection and cumulative steroid exposure, the independent impact of OKT3 on recurrence of hepatitis C is difficult to discern. Rosen et al. (17) reported that OKT3 administration is a significant risk factor for both the time to development and the severity of histological recurrence of hepatitis C. In contrast, Sheiner et al. (15) reported that OKT3 induction was associated with less frequent early recurrence of hepatitis C. Useful insight might be gleaned from a study by Gane et al. (14) who, in a program that does not use OKT3, found no relationship between the number of episodes of rejection and the histological severity of recurrence of hepatitis C. This would suggest that OKT3, rather than the number of episodes or severity of rejection, is associated with more severe histological recurrence of hepatitis C. Regardless of the cause, the NIDDK Liver Transplant Database found that steroid-resistant rejection is associated with a greater than fivefold increased risk of mortality in HCV-infected recipients.
Mycophenolate mofetil. This is a nonbiological immunosuppressive antimetabolite. By releasing mycophenolic acid, it selectively inhibits inosine monophosphate dehydrogenase, reducing proliferation of lymphocytes and cytotoxic T-cell differentiation. Mycophenolate mofetil markedly potentiates the antiherpesvirus activities of acyclovir, ganciclovir, lobucavir and penciclovir in vitro and in vivo. In addition, another inosine monophosphate dehydrogenase inhibitor, ribavirin, has limited efficacy in the treatment of HCV when used in combination with interferon-α. Unfortunately, there are no meaningful data to suggest a benefit of mycophenolate mofetil in HCV-infected liver transplant recipients. An increase in the prevalence of more severe histological recurrence of HCV has been reported recently (18). The increased frequency of cirrhotic stage recurrence was linked by the authors to the use of mycophenolate mofetil. The study, however, lacked prospective histological data upon which to draw a firm conclusion. In contrast, in a phase III multicenter randomized, controlled study of mycophenolate mofetil, there was a decreased frequency of histological recurrence of HCV at 6 months and a trend for decreased histological recurrence at 12 months among participants receiving mycophenolate mofetil when compared with subjects randomized to azathioprine. On aggregate, the reported data provide no compelling basis for either administering or avoiding mycophenolate mofetil in HCV-infected liver transplant recipients.
Il-2 receptor antibodies. Data concerning the impact of Il-2 receptor antibodies on recurrence of HCV following liver transplantation appear only in abstract form. None of the data presented to date, including the results of a phase III study, point to a meaningful benefit or detriment to adjunctive Il-2 receptor antibody therapy in HCV-infected liver transplant recipients.
TOR inhibitors. Although TOR inhibitors have been shown to have antifibrotic effects in vivo in animal models of liver disease and to decrease levels of mRNA of TGF-β, procollagen and transglutaminase, there are no data concerning the effects of TOR inhibition in HCV-infected liver transplant recipients.
Treatment of HCV Infection Following Liver Transplantation (Table 2)
Following the observation that recipients with higher pre-LT HCV RNA titers experience mortality and graft loss rates ≈ 30% greater than recipients with lower pre-LT HCV RNA titers, a pilot study of the tolerability and efficacy of pretransplant antiviral therapy was conducted. Pretransplant antiviral therapy, with the agents and doses used in this pilot study, was poorly tolerated in the minority of UNOS status 2b HCV-infected patients who met treatment initiation criteria. Although the frequency of viral response was comparable to that reported in other cirrhotic cohorts, most potential recipients do not meet treatment initiation criteria, and serious adverse events, including systemic infections, occurred commonly. Thus, while interferon-α alone or in combination with ribavirin may be effective in reducing the viral load in hepatitis C-infected patients awaiting liver transplantation, the adverse events associated with therapy are frequent and often severe in the minority of potential recipients who meet treatment initiation criteria. The routine use of immediate pretransplant antiviral therapy is thus not recommended.
The limited published experience of nonpegylated interferon monotherapy therapy in the treatment of post-transplant HCV infection has also been disappointing. In a small study of interferon monotherapy at UCSF, Wright et al. (19) achieved transient decreases in HCV RNA but no impact on histology. Sheiner et al. (20), in a study of 86 liver transplant recipients utilizing interferon monotherapy prophylactically, found that the frequency of histological recurrence of HCV was decreased in patients receiving interferon monotherapy. None of the recipients who received interferon cleared virus in this study. While the number and size of studies are both small, reports of nonpegylated interferon-α and ribavirin combination therapy in the treatment of post-transplant HCV infection suggest that the overall efficacy of these agents is broadly comparable to that seen in the nontransplant setting (21–24). Interferon and ribavirin combination therapy in the nontransplant setting has produced a greater than twofold increase in both end of treatment and sustained virological response rates when compared with nonpegylated interferon monotherapy. Unfortunately, the plethora of robust data in the nontransplant setting has not been reproduced in transplant recipients. The data that do exist are relatively encouraging, however. Three European studies found end of treatment virological response rates of 35–50% following 6 months of combination therapy, with ≈ 15% of recipients failing to tolerate therapy (21,25,26). In the largest report to date, the virological relapse rate was > 50% at 6 months after cessation of combination therapy. All three of the European studies found histological improvement in virological responders. The North American experience has been less impressive. The University of Miami reported, in abstract form, an end of treatment virological response rate of 12%, with histological improvement in a further 12% (24). Another North American center found a ≈ 25% end of treatment virological response to combination therapy (27). The differences in efficacy in the European and North American studies may reflect a fundamental difference in the study designs: treatment was initiated on a prophylactic basis in two of the three European studies (regardless of biochemical or histological profile) compared with initiation of treatment following documentation of histological/biochemical recurrence in the North American studies. Furthermore, the efficacy of combination therapy with nonpegylated interferon and ribavirin in the treatment of post-transplant HCV infection appears to be reduced in recipients with more advanced recurrence (25). The high prevalences of anemia, thrombocytopenia, leucopenia and renal insufficiency contribute to the high reported rates of dose reduction and treatment cessation of both interferon and ribavirin post-transplantation. Dose reduction/cessation of ribavirin alone can be expected in ≈ 50% of recipients receiving this agent. Early initiation of erythropoietin has been reported in small studies to facilitate increased toloerability of ribavirin. Reversal of fibrosing cholestatic post-transplant hepatitis C infection, although reported, appears to be unusual (24,28). A major limitation of post-transplant combination therapy is the relatively poor tolerability of ribavirin in transplant recipients. Most studies report a ≈ 50% dose reduction/cessation rate for ribavirin.
The advent of pegylated preparations of interferon-α has altered the therapeutic landscape for hepatitis C. Studies in the nontransplant setting indicate a greater than twofold increase in efficacy of pegylated interferon-α2a monotherapy when compared with monotherapy with nonpegylated interferon. Recently published large multicenter studies of pegylated interferon-α2a monotherapy demonstrate similar efficacy to nonpegylated interferon and ribavirin combination therapy in cirrhotic and noncirrhotic patients (29,30). Studies of the efficacy of pegylated interferon in the treatment of post-transplant HCV infection are still underway, and preliminary data suggest that virological response rates of pegylated interferon and ribavirin in the nontransplant setting are superior to therapy with combination therapy with nonpegylated interferon and ribavirin. A large (n = 692) randomized controlled multicenter study of combination therapy with pegylated interferon-α2a (40 kDa) and ribavirin in the nontransplant setting reported a sustained virological response rate of 56% on an intention-to-treat basis (Fried et al., Oral presentation at the annual meeting of the American Gastroenterology Association, Atlanta GA, May 2001). Future protocols for liver transplant recipients are likely to incorporate combination therapy with pegylated interferon and ribavirin. The optimal timing for the initiation of such therapy is not known. The data suggest that post-transplant HCV infection may be most amenable to therapy when initiated on a prophylactic basis early in the post-transplant course. However, a direct comparison of the safety and efficacy of antiviral therapy initiated prophylactically compared with when recurrence of HCV is evident histologically has not been performed.
Interferon and the risk of rejection in liver transplant recipients
Of the handful of studies describing the outcome of treating HCV with interferon in liver transplant recipients, two have suggested that interferon treatment may be associated with an increased risk of rejection. In a study by Feray et al. (31) chronic rejection occurred in five of the 14 treated patients, leading to retransplantation in three patients, compared with one of the 32 untreated recipients. In a study by Vargas et al. (32) one of seven treated cases developed severe acute rejection and another chronic rejection. An acute vanishing bile duct syndrome has been also described after interferon therapy for recurrent hepatitis C. Interferon monotherapy was not associated with increased risk of rejection in three other studies. Only one case of possible rejection was reported among the 18 patients treated by Wright et al. (19). None of the larger, more recent studies of combination therapy with interferon and ribavirin have reported an increased risk of frequency or severity of rejection (24,33,34). Some have reported remarkably low rates of rejection during combination therapy [e.g. 0/44 patients (22), 1/33 patients (35)]. Preliminary data from a randomized study of PEG-IFNα-2a (Pegasys) versus no treatment in patients with established recurrent HCV show no episodes of late acute cellular rejection among patients receiving PEG-IFNα-2a to date (n = 28).
On balance, there is no compelling evidence that interferon therapy is associated with increased frequency or severity of rejection in liver transplant recipients.
Retransplantation of HCV-infected Recipients
While, intuitively, retransplanting recipients with HCV-associated allograft failure may seem likely to result in poor graft survival, the preponderance of the data suggest otherwise. Two relatively large studies [one based on the UNOS database (36) and the other a report of the combined results of three large North American transplant centers (37)] have observed that both patient and graft survival following retransplantation in HCV-infected recipients are similar to those of recipients without HCV infection. Rosen et al. (36), in the UNOS analysis, noted, however, that survival following retransplantation was poor in HCV-infected recipients who were cholestatic and had poor renal function. This was true for early and late retransplantation. Few transplant centers will perform retransplantation in HCV-infected recipients who meet UNOS status 2a criteria.
Two reports of > 10 cases of outcomes following transplantation of HCV-infected donor livers into HCV-infected recipients have been published. Both suggest that medium-term patient and graft survival are similar to those seen following transplantation of non-HCV-infected organs. In lieu of larger cohorts of recipients receiving liver allografts from HCV-infected donors, this sporadic practice seems reasonable in the context of full consent of the recipient and normal donor histology at the time of transplantation. The advent of more efficacious therapies for HCV may extend this practice.
Hepatitis C Immunoglobulin
Hepatitis B immunoglobulin has moved hepatitis B-infected patients from the ranks of the untransplantable to ideal candidates for liver transplantation. The hope has long been that HCV immunoglobulin (HCIg) will similarly ameliorate the impact of recurrence of HCV infection. Feray et al. (38) demonstrated that anti-HCV containing HBIg reduced HCV reinfection in HBV/HCV coinfected liver transplant recipients and conferred limited protection against de novo HCV infection. It was subsequently speculated that polyclonal immunoglobulins from anti-HCV-positive donors could prevent HCV infection. A trial of an HCIg preparation (Civacir™, NABI Corporation, Jacksonville, FL) in chimpanzees demonstrated that HCIg could decrease levels of viremia and decrease histological expression of HCV antigens. An NIAID-sponsored phase II study is currently planned. The availability, cost and concerns about potential infectivity of pooled HCIg are likely to limit the impact of this therapy.
Hepatitis C-associated liver failure is the most common indication for liver transplantation and recurs nearly universally following transplantation. Histological evidence of recurrence is apparent in ≈ 50% of HCV-infected recipients in the first postoperative year. Approximately 10% of HCV-infected recipients will die or lose their allograft secondary to hepatitis C-associated allograft failure in the medium term. While the choice of calcineurin inhibitor and/or the use of azathioprine have not been clearly shown to affect histological recurrence of hepatitis C or the frequency of rejection in hepatitis C-infected recipients, cumulative exposure to corticosteroids is associated with increased mortality, higher levels of HCV viremia and more severe histological recurrence. In contrast to nonhepatitis C-infected recipients, treatment for acute cellular rejection is associated with attenuated patient survival among recipients with hepatitis C. The development of steroid-resistant rejection is associated with a greater than fivefold increased risk of mortality in HCV-infected liver transplant recipients. In lieu of large studies in a post-transplant population, therapy with pegylated interferon (± ribavirin) should be considered in recipients with histologically apparent recurrence of hepatitis C before total bilirubin exceeds 3 mg/dL. The role of hepatitis C immunoglobulin and new immunosuppression agents in the management of post-transplant hepatitis C infection is still evolving. Overall, HCV-infected recipients who undergo retransplantation experience 5-year patient and graft survival rates that are similar to recipients undergoing retransplantation who are not HCV-infected.
This work has been supported by Public Health Service grant GCRC RR00585.