Significant improvements have been made in recent years in the treatment of hepatitis C in the immune-competent population.1 More than half of currently treatment-naïve patients can be cured with the combination of long-acting pegylated interferon and ribavirin. Demonstration that these advances improve survival is however lacking. In the HCV-infected transplant population, antiviral therapy has historically yielded poor results.2 Fortunately, results have also improved in this setting, such that sustained viral response can currently be achieved in more than 35% of treated patients.2, 3 However, toxicity remains a substantial limiting factor, and about one-third of patients discontinue therapy due to severe side effects, limiting the population that might benefit from peginterferon plus ribavirin. In light of the reduced response and increased side effects, one strategy would be to treat only patients at high risk of severe HCV recurrence. Such a strategy requires a detailed understanding not only of the natural history of HCV after transplantation, but also of the variables associated with disease progression as well as of treatment response.
Despite early reports that recurrent HCV infection did not compromise patient or graft survival,2 numerous studies show that histological progression of HCV is accelerated following liver transplantation4 resulting in the rapid development (within 5 years) of cirrhosis in 30% of transplant patients with HCV.2 This rapidly progressing disease ultimately results in reduced graft and patient survival.2, 5–7 At 5 and 10 years after transplantation, survival reaches 61%-75% and 68% compared with 76%-85% and 78%, respectively, in other indications.5–7 In a recent U.S. study of 11,036 patients, transplantation in HCV-positive recipients was associated with an increased rate of death of 1.23 (95% confidence interval, 1.12-1.35) as compared with transplantation in recipients without HCV.5
With the clear understanding that HCV infection of the graft impairs both graft and patient survival, there has been intensive investigation into variables associated with these poor outcomes, particularly those variables that may be modified in order to improve outcome.
An important variable that was investigated in the studies of Belli et al.8 and Thuluvath et al.9 in this issue of Liver Transplantation is “year of transplantation”. Recent observations suggest that the speed at which fibrosis progresses has increased in patients undergoing transplantation in recent years,4 so that the proportion of patients developing advanced disease in the first 3-5 years after transplantation is higher among patients who underwent transplantation in the late 1990s compared with those who underwent transplantation in the late 1980s to early 1990s.6 In 2000, using data from La Fe Hospital in Spain and the University of California San Francisco in the United States, we showed for the first time an accelerated rate of posttransplant fibrosis progression among HCV-positive patients who had more recently undergone liver transplantation.4 The cumulative probability of developing stage 4 fibrosis at 5 years increased from 0% in patients who underwent transplantation in 1988-1989, to 7% and 18% in those who underwent transplantation in 1990-1991 and 1992-1993, respectively. In addition, we showed that Spanish patients had a higher rate of posttransplant fibrosis progression than U.S. patients despite similar progression rates prior to transplantation. Two years later, we provided further insight into these results through the analysis of 522 patients with cirrhosis who underwent transplantation between 1991-2000, 283 (54%) of whom were infected with HCV.6 We not only confirmed the more aggressive course of recurrent hepatitis C in patients who underwent transplantation recently but also the negative effect it had on survival. Thus, whereas patient and graft survival increased in non–HCV infected patients in recent years, both have decreased in HCV-infected recipients.
The finding of a worsening natural history has been challenged in the study by Belli et al. in this issue of Liver Transplantation,8 as well as by others.10, 11 In the multicenter, retrospective study by Belli et al.,8 a survival outcome analysis was performed on 502 HCV-positive recipients who underwent transplantation between 1990 and 2002 in 3 transplant centers (2 in Italy, 1 in the United Kingdom). In addition, histological data based on yearly protocol liver biopsies were available for 354 patients who survived at least 1 year. The main results can be summarized as follows: (1) an improvement in survival rates over the years; and (2) a significant association between older donor age, female sex, and lack of antibody induction therapy with the development of severe recurrent disease.
How can we reconcile these discrepant results from centers all over the world? An understanding of the mechanisms of disease injury and assessment of variables associated with disease severity and/or progression may be the key.12 Unfortunately, the number of factors which have been linked with progressive disease are numerous and frequently controversial.12 For instance, antibody induction therapy, found in the Belli study8 to be associated with improved outcome has, in contrast, been associated with aggressive recurrence in previous studies.13 In addition, whereas female sex seems to be protective in immune-competent patients, an opposite effect has been described by Belli et al.8 and others5, 6 in the transplant recipient population. Of all described potential associations, advanced donor age together with steroid pulse therapy and overimmunosuppression, have emerged as the most convincing factors that explain differences in outcome. Differences in donor age between centers and the changing donor profile with the increasing use of older donors may account for the discrepancies regarding HCV-related disease progression after transplantation and the observed worsening in outcomes in recent years in some but not all centers. In our own study, the median age of the donor rose from 23 (range = 15-58) years in 1991-1992 to 52 (range = 16-77) years in 1999-2000.6 The overall age of the donor in the Belli study ranged from 37 to 42 years depending on the center.8 The authors though did not provide information on the change in donor age over time. In fact, in a recent study from one of the participating centers, the age of the donor was not found to affect outcome in HCV-infected patients.14 The authors themselves acknowledged in that study that the age of the donor was relatively stable over time. The same may apply to immunosuppression.12 Regimens used in liver transplantation vary substantially from center to center (in the study by Belli et al.,8 extremely different protocols were used in the 3 participating centers) and, most importantly, have profoundly changed over the years. This not only concerns the immunosuppressive agents used but also in the way we have combined these agents over time. In fact, data from one of the contributing centers recently confirmed the protective effect of maintenance steroids against severe recurrent HCV disease.14 Unfortunately, the data on immunosuppression were scarce in the Belli study.8
In summary, the association with worse outcomes in recent years, and the controversy that seems to surround this observation, is likely due to differences in distribution of these variables from study to study, from center to center, and from year to year. The year of transplantation is likely merely a surrogate for these other variables. Given the difficulty of quantifying all these variables and the fact that there may well be unmeasured variables that also contribute, it is hardly surprising that the literature is conflicting. In the absence of animal models to understand the pathogenesis and given the shortage of donors and hence need to use “compromised donors” that in turn seem to be particularly problematic in patients with HCV, there are 2 potential approaches to clarify prior inconsistencies. The first is the use of large databases that, although not the best way to identify factors associated with outcome, can be very helpful to fully appreciate the trends in liver transplantation. Mutimer et al. examined a large European liver transplant database to define the impact of transplantation date and donor age on graft and patient survival for patients with HCV (n = 4736) compared to a group who underwent transplantation for alcoholic liver disease (ALD, n = 5406).15 In the entire cohort, graft and patient survival of HCV-positive recipients was inferior to recipients with ALD. Interestingly, the authors showed that since 1987, there has been a steady and ongoing improvement in the outcome of recipients with ALD, but no such improvement was observed for recipients with HCV. In this issue of Liver Transplantation, Thuluvath et al.9 use the United Network for Organ Sharing (UNOS) database, a large U.S. data set (n = 28,193 patients [HCV-positive = 7459 and HCV-negative = 20,734]), and show very similar results to those reported by Mutimer et al. in European centers.15 In this study, after adjusting for other known confounding variables that may influence outcome, graft (72.8%, 71.0%, and 69.8%) and patient (77.4%, 79.6%, and 78.5%) survival of HCV-positive patients were shown to remain unchanged during 3 study periods (1991-1993, 1994-1997, 1998-2001), respectively. In contrast, the graft and patient survival rates of HCV-negative recipients improved significantly during study periods 2 and 3 compared to period 1. Finally, Futagawa et al. recently published similar results using the Scientific Registry of Transplant Recipients (SRTR), another large U.S. database in the United States.16 Futagawa et al. analyzed the change in outcomes during 2 successive 5-year periods (period I = 1992-1996 versus period II = 1997-2002) among 35,186 deceased adult liver transplant recipients. Despite the assumed improvement in immunosuppression and the clear improvement in 1-year graft survival from 81% to 85% in period II, the analysis of SRTR data showed no improvement in the 5-year survival rate (67%). This lack of improvement could be attributed to the increase in the proportion of HCV-positive patients undergoing transplantation, who were also the group of patients who had the worst survival. HCV recurrence was shown to be the cause of graft failure in 55.2% of graft losses in period I and 61.3% of those in period II. Altogether these studies show worse results, or at least lack of improvement, in recent years in the population infected with HCV. The second alternative is to perform studies (preferentially prospective, randomized, controlled trials) that compare different management strategies. Unfortunately, adequately powered studies are lacking for a variety of reasons, particularly the difficulty in controlling for many confounding variables and the existence of center-to-center differences in management of immune suppression that are hard to standardize. In a recent single-center prospective, though underpowered, study where a historical control group was used, we showed that simple changes in immunosuppression, including avoidance of excessive immunosuppression such as triple and quadruple therapy and pulse steroids, together with a slow and gentle weaning off prednisone, resulted in a significant reduction in the rate of severe recurrent disease at 1 year after transplantation, despite the increasing age of the donor.17
In conclusion, although effective antivirals will likely change disease management, such effective antivirals are not yet close at hand for patients who underwent liver transplantation because of HCV. In the absence of such therapies, we are obliged to make sure through a better understanding of factors associated with outcome12 that we are minimizing harm to patients with our current management strategies. Minimizing the warm ischemia time, using prophylaxis against cytomegalovirus, favoring a nondiabetogenic environment,18 and promoting antifibrotic agents such as angiotensin-blocking drugs19 may have a positive effect on recurrent HCV. Most importantly, establishing which is the less deleterious immunosuppressive regimen will eventually lead to improved results. In that sense, current large, multicenter, randomized, controlled studies are finally underway to determine the potential role of each immunosuppressive agent.