Effect of Nonviral Factors on Hepatitis C Recurrence After Liver Transplantation

Abstract

Objective:

Hepatitis C (HCV) is now the most common indication for orthotopic liver transplantation (OLT). While graft reinfection remains universal, progression to graft cirrhosis is highly variable. This study examined donor, recipient, and operative variables to identify factors that affect recurrence of HCV post-OLT to facilitate graft-recipient matching.

Methods:

Retrospective review of 307 patients who underwent OLT for HCV over a 10-year period at our center. Recurrence of HCV was identified by the presence of biochemical graft dysfunction and concurrent liver biopsy showing diagnostic pathologic features. Time to recurrence was the endpoint for statistical analysis. Five donor, 6 recipient, and 2 operative variables that may affect recurrence were analyzed by univariate comparison and Cox proportional hazard regression models.

Results:

Recurrence-free survival in the 307 study patients was 69% and 34% at 1 and 5 years, respectively. Four predictive variables related to either donor or recipient characteristics were identified. Advanced donor age, prolonged donor hospitalization, increasing recipient age, and elevated recipient MELD scores were found to increase the relative risk of HCV recurrence. Examination of HLA disparity between donors and recipients demonstrated no correlation between class I or class II mismatches and recurrence-free survival.

Conclusions:

We have identified donor and recipient characteristics that significantly predict hepatitis C recurrence following liver transplantation. These factors are identifiable before transplant and, if considered when matching donors to HCV recipients, may decrease the incidence of HCV recurrence after OLT. A change in the current national liver allocation system would be needed to realize the full value of this benefit.

We retrospectively evaluated recurrence of hepatitis C in 307 patients who underwent orthotopic liver transplant over an 11-year period at our center. Univariate and multivariate analyses were used to examine donor, recipient, and operative variables that might affect recurrence. The risk factors that proved significant in this cohort are identifiable preoperatively and should be considered for better matching of donor grafts to hepatitis C recipients. This strategy would require modification of the current national organ distribution algorithm.

Four million Americans are infected with hepatitis C virus (HCV), a prevalence double that of the human immunodeficiency virus (HIV) and hepatitis B (HBV) combined.¹ Of patients with chronic infection, 20% will progress to cirrhosis over 10 to 20 years, making HCV the most common cause of chronic liver disease in the United States, the major cause of hepatocellular cancer, and the leading indication for orthotopic liver transplantation (OLT).^2–4 Risk factors for disease progression include age, obesity, alcohol intake, and coinfection with HBV and/or HIV.⁵ HCV virologic factors, including recognition of 6 genotypes with over 100 subtypes, do not appear to contribute to progression but do affect response to therapy.⁶

Unfortunately, recurrence of HCV infection in the allograft is immediate and universal after OLT.⁷ However, the natural history of recurrent HCV disease after OLT is variable, with approximately 20% of recipients progressing to graft cirrhosis within 5 years of transplant.⁸ Overall, HCV disease is more aggressive in OLT recipients than in patients whose immunity is intact.⁹ Enhanced disease progression is multifactorial and probably depends upon interactions among host, donor, viral, and external factors, although these relationships remain controversial and poorly defined. Recent reports suggesting that grafts from living related donors fare worse than those from unrelated cadaver donors in HCV recipients have implied an immunologic role as well.¹⁰

The diagnosis of recurrent HCV disease, as opposed to reinfection, is based on biochemical graft dysfunction and histologic findings on liver biopsy. Recent data suggest that the frequency and severity of HCV disease recurrence after OLT are accelerating.^11–13 This is an ominous finding, given that retransplantation for recurrent HCV has significantly worse outcome than primary OLT in most recipients.⁹ Identifying recipients at risk for rapid recurrence would be especially helpful when considering treatment with the currently available antiviral agents either as prophylaxis or therapy.

Most studies examining HCV recurrence have focused on viral characteristics, while little is published on the impact of host, donor, or perioperative factors. Interestingly, in the much larger immunocompetent population with HCV infection, the most powerful predictors of disease severity are those related to the host, while viral factors such as genotype and HCV RNA level have been shown to play a lesser role in determining outcome. We have previously examined nonviral factors that contribute to overall patient and graft survival in OLT for HCV.¹⁴ The present study was undertaken to determine whether these factors might contribute specifically to rapid HCV recurrence following OLT.

MATERIALS AND METHODS

Patient Inclusion Criteria

We performed a retrospective analysis of primary adult liver transplants for HCV cirrhosis at the Dumont-UCLA Transplant Center between January 1, 1990 and December 31, 2000. We identified 510 such patients (approximately 25% of all transplants performed at our center during this time) and obtained demographic and clinical data from both the United Network for Organ Sharing (UNOS) and the UCLA Hepatitis C Transplant Registry. Ninety-seven patients were excluded due to incomplete information regarding HCV recurrence, as were an additional 106 patients who had not developed HCV recurrence but had accumulated fewer than 36 months of follow-up time. The remaining 307 patients with accurate HCV recurrence information and more than 36 months of follow-up were used for subsequent analysis.

Candidates for OLT were considered as urgent or nonurgent recipients according to their medical condition before transplantation, as defined by UNOS categories. From 1990 to 1994, urgent recipients were categorized as status 4, and nonurgent recipients included status 3, 2, and 1. After November 1994, the status designation was modified by UNOS on 2 occasions. Urgent recipients included either status 1 or 2A, and nonurgent recipients included status 2, 2B, 3, or 4, according to the designated UNOS criteria at the time of transplantation. Current acuity status is determined by either status 1 designation or the MELD formula, which is based on serum creatinine, total bilirubin, and INR; these values were included for analysis as well.

Diagnosis of Recurrent HCV Disease

Hepatitis C was diagnosed before OLT by anti-HCV seropositivity using enzyme-linked immunosorbent assay (ELISA 2.0) and/or PCR for detection of HCV RNA. The preoperative diagnosis was confirmed by pathologic examination of the explanted livers after operation. Recurrence of HCV disease was diagnosed by the presence of biochemical graft dysfunction with concomitant liver biopsy showing features consistent with recurrent HCV, including portal or lobular infiltration by mononuclear cells with piecemeal necrosis, hepatocellular fatty change, and/or fibrosis in the absence of other specific causes such as cellular rejection.

Immunosuppression

Maintenance immunosuppression consisted of either a triple cyclosporine-based drug regimen (Sandimmune or Neoral, azathioprine, and prednisone) or a dual tacrolimus-based regimen (tacrolimus and prednisone). In 1996, Neoral (Novartis, East Hanover, NJ) was routinely substituted for Sandimmune (Novartis, East Hanover). Routine use of tacrolimus was begun at our institution in 1994. On the day of transplantation, patients were started on a rapid steroid taper according to our standard protocol. Oral prednisone (20 mg/day) was started on day 8 and tapered over 2 months to 5 mg/day. Beginning in 1995, steroids were discontinued at 6 months in HCV patients who did not exhibit rejection episodes.

Variables Affecting Recurrence

Prior work from our center has examined factors that influence patient and graft survival after OLT for HCV¹⁴ and is summarized in Table 1. Based on these studies, we selected factors to examine that may impact on HCV recurrence. Donor factors considered included donor age greater or less than 50 years, donor serum sodium level greater or less than 150 mEq/mL, donor hospital stay greater or less than 5 days, donor HCV status, and donor sex. Six recipient factors were analyzed: recipient age greater or less than 40 years, total bilirubin level greater or less than 2.4, serum creatinine greater or less than 1, prothrombin time greater or less than 20, MELD score over or under 27, and UNOS status of urgent or nonurgent. Finally, 2 perioperative variables were considered: cold ischemia time greater or less than 10 hours and warm ischemia time greater than or less than 45 minutes. The thresholds for donor, recipient, and operative variables were selected based on values which were predictive of patient and graft survival in prior studies.

TABLE 1. Nonviral Factors Analyzed for Impact on HCV Recurrence

Statistical Analysis of Variables Affecting HCV Recurrence

Time to recurrence was considered as the endpoint for statistical analysis. For each variable, patient subgroups were evaluated for cumulative person-months of follow-up and number or recurrences within the subgroup. In this way, recurrence rates per subgroup were generated and compared with calculated relative risk of HCV disease recurrence for each variable threshold considered. Both univariate and multivariate analyses were performed. Univariate comparison used log-rank methods. Cox proportional hazard regression model was used for multivariate analyses.

Examination of HLA Matching on HCV Recurrence

To examine the effect of HLA matching on HCV disease recurrence, we stratified our 307 donor-recipient pairs for the degree of immunologic similarity among HLA-A, HLA-B, and HLA-DR loci. Each subgroup of HLA disparity was evaluated in terms of number of cumulative person-years follow-up and total number of HCV disease recurrences for that subgroup. Recurrence rate per 100 person-years was thus calculated for each degree of mismatch.

RESULTS

Overall HCV-Free Survival and Univariate Predictors of HCV Disease Recurrence

Of 307 recipients in the study group, HCV disease recurrence occurred within the first 12 months in 95 patients (31%); from months 12 to 24 in 54 patients (18%); from months 24 to 36 in 26 patients (8%); and more than 3 years after OLT in 29 patients (9%). The remaining 103 patients (34%) had not developed recurrence in more than 36 months of follow-up. Overall HCV disease-free recurrence was 69% and 34% at 1 and 5 years, respectively (Fig. 1). In univariate analysis, donor age greater than 50 years and donor hospital stay greater than 5 days were significant predictors for higher risk of HCV recurrence post-OLT (Table 2).

FIGURE 1. Hepatitis C virus disease-free overall survival in the 307-patient study group.

TABLE 2. Univariate Predictors of HCV-Free Survival

Examination of HLA Matching on HCV Recurrence

There were no donor-recipient pairs with no or only one HLA mismatch; 3% of our recipients had only 2 mismatches, while the remainder had 3 to 6 mismatches. Recurrence rate per 100 person-years are shown in Table 3. There was no statistically significant difference between recurrence rates for different degrees of class I or class II HLA matching.

TABLE 3. HCV Recurrence Rates per HLA Mismatch

HCV-Positive Donors

Grafts from donors with HCV were used in 37 of the 307 recipients in this series was (12%). Use of such grafts reduced mean time of recurrence from 35.7 months to 22.9 months (relative risk [RR] = 1.4; P = 0.08). However, neither graft survival (RR = 1.0, P = 0.293) nor patient survival (RR = 1.02, P = 0.907) was reduced significantly with utilization of these donors.

Recipients With Hepatocellular Cancer

This series included 35 recipients (11%) with hepatocellular cancer (HCC). The presence of HCC did not significantly affect graft survival (RR = 1.26; P = 0.248) or patient survival (RR = 1.5, P = 0.093) in this study. In a related, larger analysis of patients with HCC and HCV who underwent OLT in our institution,¹⁵ 1-, 3-, and 5-year survivals were reduced in the recipients with cancer when compared with the group with HCV cirrhosis alone.

Multivariate Analysis of Predictors of HCV Recurrence

We next performed multivariate analysis using a Cox proportional hazard regression model on our donor, recipient, and operative variables of interest. Four of these factors were shown to be statistically significant in predicting risk of HCV disease recurrence after OLT (Fig. 2). Donor age greater than 50 years imparted a 1.5-fold risk of recurrence (P = 0.017). Donor hospital stay greater than 5 days imparted a 1.7-fold greater risk of recurrence (P = 0.02). Elevated recipient MELD gave a 1.6-fold greater risk of HCV recurrence (P = 0.036) and older recipient age increased disease recurrence by a factor of 1.5-fold (P = 0.08). Cold and warm ischemia times and donor HCV status were not independent factors with impact on HCV recurrence in this analysis. These results are summarized in Table 4. Further analysis indicates that these risk factors for HCV recurrence are additive in nature and that rates of disease recurrence rise sharply as they accumulate in donor-recipient pairs (data not shown).

FIGURE 2. Summary of multivariate analysis of donor, recipient, and operative factors affecting risk of HCV recurrence after OLT.

TABLE 4. Summary of Multivariate Predictors of HCV Recurrence After OLT

DISCUSSION

Hepatitis C is now the most common indication for liver transplantation, and disease recurrence after OLT has become a central problem in the field. Early recurrence of HCV disease is recognized as a poor prognostic indicator with lower subsequent graft and patient survivals.¹⁴ Identification of recipients who are predisposed to rapidly recurrent HCV disease would be useful when considering organ allocation and prophylactic antiviral treatment. Defining HCV recurrence has varied in the literature. For this study, we have used pathologic criteria in the context of biochemical graft dysfunction as our end point, as previously.¹⁴ Others have successfully applied quantitative scoring to protocol biopsies,^16,17 although this was not possible in this retrospective study in which biopsies were performed only when graft dysfunction was present. Likewise, the 10-year scope of this work prevented any routine surveillance of hepatitis C viral loads as quantitative assays changed several times during this period.

In this study, we identified older donor age, extended donor hospital stay, elevated recipient MELD, and advancing recipient age as significant predictors for HCV recurrence. These results suggest that grafts from extended criteria donors are particularly poorly suited for sicker recipients with HCV and may help illuminate recent national trends showing accelerating rates of HCV recurrence after OLT. Donor livers with greater than 30% macrosteatosis were not routinely in this series; thus, no attempt was made to correlate highly fatty livers or donor BMI with recurrence.

Work from Baylor University shows a similar decline in outcomes for HCV recipients with MELD scores of 25 and greater.¹⁷ Like that report, our study used the biologic MELD computed only from laboratory values and was unaffected by the presence of HCC exception points. Other variables examined in the present study, including warm and cold ischemia times as well as degree of immunologic mismatch, were found to be nonpredictive of HCV recurrence. Previous analyses of our HCV positive patients undergoing OLT have detected a small but statistically significant deleterious effect (RR = 1.02, P = 0.001) of prolonged warm ischemia time¹⁴ on overall graft survival. Such an effect was not seen in this analysis, which focused only on HCV disease recurrence, but we continue to attempt to minimize cold and warm ischemic times in all of our OLTs, and especially in cases involving the use of extended criteria donor grafts. Interestingly, no effect on graft or patient survival was seen in the study group due to the presence of recipient HCC. If there was a liberalizing impact on donor selection for these recipients, it did not statistically impact outcomes in this study.

The mechanism by which the identified factors exert their undesirable effect on HCV recurrence presumably involves host-viral interactions similar to those seen in the immunocompetent HCV-infected population. HCV damage in this group is mediated by the host immune response, as HCV is not directly cytopathic. Both CD4 and CD8 positive T cells participate in recognition of HCV peptide displayed by infected hepatocytes¹⁸ and resultant elevated gamma-interferon levels correlate with observed liver injury. HCV infection initiates a specific host recognition resulting in a host response that is ineffective in clearing virus and generates hepatic cellular damage in a nonspecific fashion.¹⁹ HCV recurrence is accelerated after OLT due to high viral loads and an exaggeration of this host response, which occurs even in the presence of exogenous immunosuppression. The alloimmune response and ischemia-reperfusion injury may contribute as well.^20,21

Grafts from extended criteria donors (older donors and/or longer hospital stays) have suffered cellular damage to some extent. The relationship between hepatocellular injury, hepatic regeneration, viral replicative activity, HCV antigen expression, and pathologic host response has been invoked previously but remains unproven. Increased allograft damage is related to enhanced levels of known immune modulators, including IL-6 and IL-10. These cytokines are released in the milieu of injured or proliferating cells and are known to participate in the pathogenesis of hepatitis C via increased viral activity, exaggerated host response, or both.²² Recipient acuity or advancing age may reflect host inability to mount effective antiviral response, which in turn leads to a final common pathway of heavier viral load and increased nonspecific host immune response. Cytokine levels were not analyzed in this study cohort but further studies examining cellular and molecular changes in patients with early recurrence are expected to be of great value, especially in guiding antiviral therapy. As in the immunocompetent population, specific viral characteristics are less important in this process.^23–27 The effect of immunosuppression regimens on HCV recurrence was not examined in this study, as all recipients received a standardized steroid and calcineurin inhibitor protocol; however, recent studies suggest that immunosuppression withdrawal protocols may lead to accelerated HCV recurrence consistent with the mechanism of exaggerated nonspecific host response described herein.²⁸ This study focused on nonviral perioperative risk factors for HCV recurrence and identified factors that elevated risk of recurrence on the order of 1.5- to 2-fold. How this degree of risk elevation compares to that from other risk factors such as steroid pulses or preoperative viral load was not quantified in this study.

Previous reports have suggested that donor-recipient matching at the HLA-B locus reduces the incidence of acute rejection after OLT but promotes the recurrence of viral hepatitis.^29,30 Additionally, a recent report suggested that HCV infection pursues a more aggressive course following living donor as compared with cadaveric liver transplantation.¹⁰ However, a recent U.S. report identified no histologic differences in living versus deceased donor recipients with HCV, with no occurrence of cirrhosis in either group.³¹ Similarly, UNOS data from 1999 to 2003 for transplants performed for HCV cirrhosis with either living or deceased donor grafts found no difference in survival between recipient groups.³² In the current study, we observed no statistically significant effect of donor-recipient HLA disparity on the risk of developing recurrent HCV disease after OLT.

The use of HCV positive grafts was 12% in this series, and recipients demonstrated more rapid HCV recurrence. However, ultimately neither graft nor patient survival was significantly reduced with utilization of these donors, a finding that has been observed by other investigators.³³ Earlier reports from our institution and from Baylor University^14,34 have shown that early HCV recurrence can be associated with significantly increased rates of patient death and graft loss, such that use of HCV positive grafts must be handled judiciously.

Antiviral therapy with pegylated interferon and ribaviron has been studied prior to OLT, after OLT but before HCV recurrence (“prophylactic therapy”), and after OLT with documented recurrent disease. As some patients with recurrent HCV have a benign course, and tolerability of antiviral therapy is severely limited by adverse effects, the decision is difficult. Our policy at UCLA is not to treat HCV cirrhosis with antiviral therapy in patients awaiting OLT, as the benefit of low viral load pre-OLT has been controversial, and drug tolerability in these patients is poor. However, other centers have reported improved results with treatment before OLT and decreased HCV recurrence after transplant.³⁵ Prophylactic treatment has been shown effective in delaying recurrence^36–41 but has limited efficacy (9%–39% sustained viral response), and an effect on patient and graft survival has been more difficult to demonstrate. Adverse effects may outweigh benefits in a population in which not all patients will develop recurrent disease. Treatment of recurrent HCV disease post-OLT shows promising early results in several small series but often is poorly tolerated.^42,43

Retransplantation for graft failure due to recurrent HCV disease is controversial. Overall worse outcomes are expected in cases of retransplantation for any indication,⁴⁴ and most series report results inferior to first time OLT for HCV.^45,46 A preferable strategy would be to identify patients at greatest risk for disease recurrence and offer prophylactic treatment to this group. One-year protocol biopsies may be a helpful aid in identifying rapidly recurring disease.^16,17,47

Recent data have shown an acceleration of HCV disease progression after OLT.^11–13 The causes are probably multifactorial and may be related to use of antiviral therapy before transplantation, with selection of more virulent viral strains in patients who come to OLT. Increased rates of HCV recurrence are also predicted by our findings, which suggest that greater use of extended criteria donors for HCV recipients will have adverse effects on recurrence.^9,48 These observations suggest that certain modifications in donor allocation and recipient management might lead to better results for HCV patients.

The current system of allocating liver grafts in the United States is governed by the MELD formula and directs donor organs of all quality indiscriminately to the most critically ill recipients regardless of HCV status. This policy may be contributory to high rates of HCV recurrence after OLT. Under a proposed modification, grafts from optimal donors would be directed toward recipients with HCV, while grafts from extended criteria donors should be reserved for non-HCV patients. Also, centers would attempt to minimize cold and warm ischemic times in these situations, and the use of HCV-positive donors would be limited.

For high MELD recipients with HCV, we recommend consideration of rapid steroid withdrawal and prophylactic antiviral therapy if tolerated. Retransplantation for graft loss from recurrent HCV disease after OLT remains controversial and undesirable, with recent studies showing extremely high overall mortality rates.¹⁶ Understanding and modifying variables that impact HCV recurrence after OLT remain a preferable alternative, both on an individual and national level.

Discussions

Dr. Goran B. Klintmalm (Dallas, Texas): Hepatitis C has emerged as the overwhelming indication for adult liver transplantation in the United States. It is an epidemic affecting, as we heard, 4 million Americans, some of whom constitute the 50% of patients receiving liver transplantation. In the 1980s and early 90s, hepatitis B was the viral hepatitis that caused the biggest concern to the transplant community. However, we soon developed remedies for these as HBIG and Lamivudine. We are not as fortunate with hepatitis C. No immunoglobulin exists and the current medical therapy in the form of Ribavirin and pegulated Interferon has a poor response rate in these patients as they are poorly tolerated, especially posttransplant.

Consequently, the question of how and what affects are of major importance to all in liver transplantation. Today's presentation by the UCLA group is the latest important contribution to our understanding of this topic.

The current study of 307 patients transplanted between 1990 and 2000 contains much information. This period is special since it was during this time period we began to recognize the importance and devastation of recurrent hepatitis C. We also made changes in our immunosuppressive regimens and donor selections.

I have had the privilege to review the manuscript and study its findings and conclusions. It is an excellent manuscript. As do all good manuscripts, it raises additional questions. One question is put to rest, however. That is HLA matching does not seem to influence recurrence.

My first question is for donor age 50 to 55. It has been reported by several to be of importance. But in the study that we presented last year on 478 patients, we did not find donor age to affect the incidence of recurrence. It did show, however, that the time to recurrence was shorter in the higher age donors. We wonder if you have any comments on these different findings.

On a side note, your comment that advancing age is a significant predictor of the recipient for recurrence, the audience probably paused to reflect that you define advanced age as being over 40.

Second, in a previous publication from UCLA on the recurrence of hepatitis C, you noted that the incidence was unaffected, but the time to recurrence was significantly shorter in patients treated with Tacrolimus instead of Cyclosporine. The same finding was reported by the group from Berlin and also from Baylor in a publication in 2000 as well as last year, and still you decided not to use immunosuppression as one of your risk factors. I would like to hear your arguments for this.

Thirdly, it is known that some patients develop cirrhosis in their transplant faster than others do. We just don't know who these patients are and why. Whether the recurrence caused the cirrhosis in 1 or 6 years seems possible and even more important than time to recurrence. The risk factors you used in this study may give significant new insight on this vexing problem. Did you consider using time to cirrhosis as an endpoint?

Fourth, one of the major questions that I have had over the last several years is how we use our donors. The pristine donors of yore comprise a small percentage of the donor population today. The thought of systematically taking the best livers and putting them into patients with poorer outcomes due to recurrent hepatitis C and giving the less perfect grafts to patients with potentially better long-term outcomes troubles me. To me it would be like robbing Peter to pay Paul. I have less problems using older donors for older recipients and younger donors for younger recipients, but I am getting very reluctant to ask UNOS for new rules since you may get more than you ask for. For those of you who not in transplant, to deal with UNOS is almost like dealing with the IRS. I would like to hear more of your thoughts about the allocation scheme that you mentioned briefly at the end.

Lastly, the use of hepatitis C positive donors for hepatitis C has been shown by many authors and also from the SRTR data not to negatively affect the outcome or survival rates. Yet, you find a negative effect in the UCLA data. Do you suggest we stop using hepatitis C positive donors for hepatitis C positive recipients?

The presented study represents a huge effort with an analysis of a large patient material with hepatitis C. Single center studies can provide detailed data impossible to obtain from national registries. The study from UCLA presented here today represents another milestone that has been given to the transplant community.

I sincerely enjoyed this excellent presentation on a vitally important topic and I thank Dr. Khuri and the Association for the privilege to discuss this paper.

Dr. Andrew M. Cameron (Los Angeles, California): Dr. Klintmalm, thank you for your comments and kind words. It is a privilege to have you discuss this work.

Your first question concerns the relative risk of advancing donor age. Our data is in agreement with yours. We do see an increased risk with each subgroup of donors at successive stratified age groups. We chose the donor age of under or over 50 years as a statistical cutoff for 2 reasons. Number 1, this variable has been shown to be predictive of overall outcome in previous studies. Second, that cutoff, while not a median stratification of our donor population, did divide the cohort into 2 large groups with around 70 donors over age 50. And that breakdown certainly demonstrates your observation that the donor pool is now much older.

Your second question regarding advancing recipient age requires some diplomacy. Despite our selection of age 40 as “older” recipients, I can report that in Southern California 40 years old is still considered young. In fact, many of our recipients are now in their 60s and 70s, and a report regarding the very good outcomes that can be on obtained in those groups of selected older patients is being prepared at UCLA. Our oldest liver transplant recipient at UCLA was 82 years old at the time of her liver transplant and is still alive and well today. We have clearly entered not only the era of extended criteria donors but also extended criteria recipients.

Your third question concerned Prograf versus Cyclosporine in terms of an effect on hepatitis C recurrence. We were not able to analyze that difference in this study as all patients at UCLA are maintained on Prograf unless they are unable to tolerate that drug. Our experience is identical to that published from your group at Baylor which shows that the dominant important negative prognostic risk factor regarding immunosuppression is rejection episodes requiring steroid pulse. Our rates of rejection are on the order of 20% for the first year after transplant, and any beneficial Prograf effect is probably mediated via fewer rejection episodes.

Your fourth question concerned progression to cirrhosis in this series. Our progression to graft cirrhosis is about 20% at 5 years. We would imagine that the risk factors identified here that predict HCV disease recurrence would also participate in progression to cirrhosis, but that has not been definitively shown.

Your fifth and most challenging question concerned changes in organ allocation systems. The modification we would cautiously suggest would be creation of an independent extended criteria donor list similar to that currently used with renal transplantation in our country. Such marginal grafts, to use an older and currently unpopular term, would be offered not to a specific recipient but to a transplant center, and its use would be at their discretion, identical to the system currently in place in Europe. In this way, extended criteria grafts could be matched to appropriate recipients, in this case lower MELD, hepatitis C negative recipients.

Finally, your last question was regarding hepatitis C positive donors. We do use them at UCLA. The rate was 12% in this series. As you correctly point out HCV recurrence is faster in this subgroup but fell just short of statistical significance in our study. As you also clarified, overall patient and graft survival is not reduced with the use of these donors. They are often optimal donors in all other respects, and are usually younger donors in our experience. Therefore, hepatitis C positive donors are used judiciously at UCLA.

Dr. Nancy L. Ascher (San Francisco, California): My first question relates to emerging information mainly from Hugo Rosen from University of Colorado indicating that it is possible to actually identify specific a host responses to the HCV viral antigens. Do you have information regarding this approach? Perhaps host response to the HCV can, in part, explain that subset of patients that really seem to live with their virus in the long term?

The second question relates to genotyping the recipients. The audience may not know that genotype 3 is quite responsive to therapy, unlike genotype 1-A and B. Nonetheless, genotype 1-A and B are the most common genotypes in the United States. What were the specific genotypes in your patients with HCV?

Dr. Andrew M. Cameron (Los Angeles, California): I don't have specific molecular evidence regarding mechanism in our rapid progressors. As you point out, there is better and better lab data and now clinical data about the role of host NK cells, T Cell receptor Hepatitis C antigen specificity, and IL-6 and IL-10 levels in the role of HCV progression. We have not yet looked at our rapidly progressing recipients for evidence of cellular or molecular differences.

In regards to your question about genotype. Not in this study, but in a previous study we genotyped our Hepatitis C patients. As you point out, 90% are genotype 1-B. It is our experience that in the transplant recipient population, genotype predicts response to anti-viral therapy better than it predicts disease progression, a finding similar to that in the immunocompetent population. Our hepatologists still struggle with who should be put on toxic anti-viral therapy. Certainly, if you could identify a recipient likely to recur and progress rapidly, likewise if you knew that recipient was infected with genotype 1-B, you would be tempted to start treatment, but unfortunately you would also know that this recipient would be expected to respond less well. So it is a difficult problem.

Dr. John P. Roberts (San Francisco, California): To expand upon what Dr. Klintmalm was asking, the biopsy criteria for the diagnosis of hepatitis C recurrence was pretty broad, with a lot of non-specific criteria. How about just looking at people that reach stage 2 fibrosis? The second question is, did you do protocol biopsies so that we have a fairly good idea about how well these patients were sampled?

Dr. Andrew M. Cameron (Los Angeles, California): To answer the last question first, we did not perform protocol biopsies, a potential weakness of this study. A relatively recent study from the Klintmalm group with 300 patients studied protocol biopsies at 1 week, a year, 2 years, and 3 years found that there were some patients with histopathologic evidence of HCV recurrence that didn't have biochemical graft function abnormalities. We used a clinical definition of graft dysfunction and then confirmed the diagnosis of recurrence with a biopsy. So all patients in this study did have a liver biopsy, but they were not protocol biopsies, rather they were initiated by graft dysfunction.

We did not subdivide for grade of fibrosis on our biopsy reads. Our endpoint for analysis was a firm diagnosis of HCV recurrence in the absence of cellular rejection. Distinguishing between hepatitis and rejection was often the main dilemma for the pathologist.

Dr. Susan L. Orloff (Portland, Oregon): I have 2 questions. One is, in this growing incidence of obesity in our population, did you look at the percentage of macrovesicular steatosis in the donors and how that factored into risk of hepatitis C recurrence?

The second question I have is, did you look at a subset of patients with fibrosing cholestatic hepatitis, which is a very severe incidence of recurrent hepatitis C, one that requires either retransplantation with bad outcome or death? And I am just wondering if you looked at a subset of risk factors for that particular entity.

Dr. Andrew M. Cameron (Los Angeles, California): Regarding fatty livers, we did not subdivide our donor grafts for varying percentages of macrosteatosis. We routinely use grafts only with fatty content of less than 30%, so we did not have a large number of moderate or highly macrosteatotic grafts to look at. I think the presupposition is that those grafts would be considered extended criteria grafts and would therefore contribute to higher rates of viral recurrence.

Your second question was regarding fibrosing cholestatic hepatitis, which is indeed a horrible problem. In our series the incidence of fibrosing cholestatic hepatitis after transplant for hepatitis C was on the order of 2%. These are cases where recurrence is hyperaggressive with very high viral loads and very poor outcomes. Those patients represent a difficult problem. We treat them by lowering their level of immunosuppression and consider earlier retransplant.