Recurrent Primary Disease After Liver Transplantation

Natural History After Liver Transplantation

The early experience with LT in HBV-infected patients revealed a high rate of reinfection and accelerated progression of disease after LT, which resulted in 5-year survival rates of less than 50%. The availability of prophylactic therapies—first, hepatitis B immunoglobulin (HBIG), and later, oral nucleoside/nucleotide analogs such as lamivudine, adefovir, entecavir, and tenofovir—transformed the outcomes of patients with HBV infection undergoing LT. In a retrospective study of HBV-infected adults undergoing primary LT in the United States between 1987 and 2002, the 1-year survival probability improved from 71% in 1987 to 1991 to 87% in 1997 to 2002, and the 5-year survival rate increased from 53% in 1987 to 1991 to 76% in 1997 to 2002 ( Fig. 53-3 ). Similarly, the European experience shows temporal improvement in patient survival among HBV patients regardless of transplant indication ( Table 53-2 ). The progressive improvement in survival rates paralleled the availability of therapeutic interventions to prevent and treat HBV infection in transplant recipients. A more contemporary U.S. multicenter experience with LT in 170 HBV-infected patients who underwent LT between 2001 and 2007 reported a 5-year survival rate of 85% or greater, and from 2002 to 2011 the U.S. Scientific Registry of Transplant Recipients found that among HBV-infected adults who underwent LT for HCC, the 5-year graft survival rate was 93%. Thus the survival of HBV-infected LT recipients is among the highest for adults with cirrhosis of all causes undergoing LT.

Natural history studies from the era before the use of prophylactic therapies showed that the level of HBV DNA at the time of LT was the principal factor affecting the posttransplant risk for recurrent infection. In the landmark study by Samuel et al. of 372 European hepatitis B surface antigen (HBsAg)-positive patients who underwent LT from 1977 to 1990 (with and without HBIG prophylaxis), the 3-year actuarial risk of recurrence of HBV infection was highest (83%) in those with HBV-related cirrhosis with a serum HBV DNA level of 10 ⁵ copies per milliliter or greater at time of LT, intermediate in those without detectable HBV DNA or hepatitis B e antigen (HBeAg) (58%), and lowest in those with hepatitis D virus (HDV) coinfection (32%) or fulminant HBV infection (17%) ( Fig. 53-4 ).

In the current era of routine use of prophylactic therapies to prevent HBV infection, recurrence of HBV infection after LT is infrequent but is still most consistently associated with the levels of HBV DNA before LT. The presence of drug-resistant HBV before LT and recurrence of HCC with HBV replication in tumor cells possibly serving as a source for recurrence of HBV infection have also been associated with higher rates of failure of prophylactic therapy. In a recent large single-center study of 185 consecutive LT recipients for HBV-related liver disease, pretransplant HBV DNA level, presence of HCC, antiviral treatment status, and posttransplant viral resistance were identified as the major risk factors associated with recurrence of HBV infection after LT.

Pathology of Hepatitis B After Liver Transplantation

In general, the histopathologic findings of recurrent HBV infection in a liver allograft are similar to those in a nontransplanted liver with HBV, with acute and chronic pathologic features seen. In addition, a unique variant form of severe recurrent HBV infection termed the fibrosing cholestatic variant can rarely occur. This histologic variant is generally unique to HBV in the transplant setting, with case reports in renal and bone marrow transplant recipients as well as LT recipients, with only one case report of a nontransplant patient treated with immunosuppressive therapy.

In the absence of prophylactic therapy, the earliest manifestations of recurrent HBV infection include cytoplasmic and nuclear expression of hepatitis B core antigen (HBcAg) in hepatocytes seen by immunostaining 2 weeks to 5 weeks after LT, typically in the absence of hepatocyte necrosis or inflammation. Acute hepatitis, which develops on average 8 weeks to 10 weeks after LT, is manifested histologically as diffuse hepatocyte ballooning and lobular disarray with spotty hepatocyte necrosis. A portal-based mononuclear infiltrate of varying severity is seen. There is a gradual increase in the percentage of hepatocytes containing both nuclear and cytoplasmic HBcAg, especially during the evolution of acute disease to chronic disease. Ground-glass cells are infrequently seen during the acute and early phase of infection. There is no significant bile duct damage or endothelial injury, thus making the distinction between early recurrent HBV infection and acute cellular rejection relatively straightforward. In a small percentage of patients, a severer hepatitis with bridging and submassive necrosis is seen and, before the availability of antiviral therapy, was typically associated with rapid progression to cirrhosis or graft loss.

Over time the changes associated with acute hepatitis evolve into those of chronic hepatitis, with portal-based inflammation and fibrosis and the presence of piecemeal necrosis and ground-glass cells, typically in inverse proportion to the necroinflammatory activity. The inflammatory infiltrates are predominantly of the plasma lymphocytic type ( Fig. 53-5, A ). Immunohistochemical studies reveal HBcAg in the cytoplasm and nucleus of a large number of hepatocytes (see Fig. 53-5, B ). Overall, the histologic features in the chronic phase of infection are the same as those in nontransplant patients except for the rapidity of progression over time. Evolution to recurrent cirrhosis can occur within 2 years in the absence of antiviral therapy.

As previously mentioned, fibrosing cholestatic HBV is an infrequent form of recurrent HBV infection that is seen only in immunosuppressed patient populations. Unique histologic features include extensive hepatocyte ballooning, marked ductular reaction and cholestasis, pericellular and portal fibrosis, and a paucity of inflammatory infiltrates ( Fig. 53-6, A ). High levels of viral replication are present as evidenced by high levels of expression of HBcAg and HBsAg in the liver (see Fig. 53-6, B ), thus lending support to the concept of a direct cytopathic form of liver injury. Biochemical features include a markedly abnormal serum bilirubin level and prothrombin time but only modest increases in serum aminotransferase levels and signs of progressive liver failure. In the absence of antiviral therapy, the outcome is rapidly fatal.

Prevention of Recurrent Hepatitis B Virus Disease

Historically, recurrent HBV infection after LT led to rapidly progressive disease with graft losses in 50% of patients within 5 years. Before the availability of nucleoside/nucleotide analogs, the focus of HBV infection management was on preventing reinfection with use of HBIG. Now, although effective antivirals are available to treat recurrent HBV infection, the primary management strategy is still to prevent recurrence of HBV infection. A combination of HBIG and nucleoside/nucleotide analogs is most frequently used. In the past 2 decades, multiple single-center and multicenter studies have established the high efficacy of this combined approach. However, HBIG is costly and inconvenient for long-term use, so protocols that use lower doses, a shorter duration, and nonintravenous routes of administration have evolved. Additionally, the availability of antivirals with high potency and a very low resistance risk, such as entecavir and tenofovir, has led to use of nucleoside/nucleotide analogs alone for prophylaxis.

Posttransplant recurrence of HBV infection was historically defined by reappearance of HBsAg in serum. In the early HBIG prophylaxis era, this signaled a failure of prophylaxis and was accompanied by quantifiable HBV DNA and later by clinical and histologic evidence of recurrent HBV disease. However, in the current era of long-term use of nucleoside/nucleotide analogs, recurrence of HBsAg in serum may occur without quantifiable HBV DNA. The former indicates the presence of recurrent infection in the transplanted liver, but the lack of quantifiable HBV DNA is consistent with adequate suppressive therapy. Additionally, HBsAg that is detectable in serum without HBV DNA has been described in patients receiving prophylaxis who develop recurrent HCC. In this scenario the tumor rather than an infected allograft is hypothesized to be the source of HBsAg. Thus although prevention of HBV reinfection has always been the goal of prophylaxis, the high efficacy of nucleoside/nucleotide analogs in suppressing HBV DNA after transplant and preventing graft loss has led to less concern about recurrence per se (as defined by the presence of HBsAg in serum), because recurrent disease is generally controllable with antivirals such as entecavir and tenofovir. However, once recurrence occurs, the risk of drug resistance and progressive hepatitis in the graft is always a concern, particularly in patients with adherence issues, or those with multidrug-resistant HBV for whom long-term suppressive therapies may be less available. For these reasons the optimal prophylactic therapy is one that prevents rather than suppresses graft reinfection (i.e., achieves negative HBsAg and undetectable HBV DNA in serum after LT).

Pretransplant Antiviral Treatment

The most consistent predictor of post-LT prophylaxis failure is the level of HBV DNA at the time of LT. Thus before LT, antiviral therapy should be started in all patients with indications for LT and detectable HBV DNA. The goals of pre-LT antiviral treatment are to achieve an undetectable HBV DNA level before LT and to minimize the risk of emergence of resistant viral strains; the preferred drugs are tenofovir and entecavir. Lamivudine, telbivudine, and adefovir are not ideal first-line agents in this setting, given the high rates of resistance and, in the case of adefovir, the risk of nephrotoxicity. However, in resource-constrained situations, lamivudine—when given for a limited duration of treatment before transplant to minimize the risk of resistance—may be an acceptable low-cost option.

The collective experience with entecavir and tenofovir indicates high efficacy and safety in patients with decompensated cirrhosis. In a study that included 120 patients with advanced fibrosis treated for 48 weeks, entecavir resulted in undetectable HBV DNA levels in 91% of HBeAg-positive patients and 96% of HBeAg-negative patients versus 57% of HBeAg-positive patients and 61% of HBeAg-negative patients treated with lamivudine. In a study of 70 treatment-naïve patients with decompensated cirrhosis, 75% of HBeAg-positive patients and 98% of HBeAg-negative patients achieved HBV DNA suppression after 1 year of therapy. Tenofovir disoproxil has a favorable safety profile and high antiviral potency with the absence of resistance with treatment periods of up to 6 years. Use of tenofovir in patients with decompensated cirrhosis was initially reported via case reports only until 45 patients with decompensated cirrhosis were treated as part of a phase 2, double-blind randomized study. Treatment discontinuation due to side effects was infrequent (6.7%), and at week 48 of treatment, the level of HBV DNA was less than 400 copies per milliliter in 71% of patients, there was normalization of the alanine aminotransferase level in 57% of patients, and HBeAg loss or seroconversion occurred in 21% of patients. Some experts also advocate the use of de novo combination therapy, such as tenofovir-emtricitabine, to further minimize the risk of resistance; however, data to support this approach are lacking. In the only study to compare these agents directly, 112 patients with HBV and decompensated liver disease were randomized to receive entecavir, tenofovir, or tenofovir-emtricitabine. The rates of viral suppression to less than 400 copies per milliliter by 48 weeks were 73%, 71%, and 88% respectively and a decrease in the Child-Turcotte-Pugh score of at least 2 units was achieved in 42%, 26%, and 48% of patients respectively. Importantly, serious adverse events and frequency of death (9%, 4%, and 4% respectively) were comparable between all three treatment groups (entecavir, tenofovir, and tenofovir-emtricitabine), with no reports of lactic acidosis in any study patients.

For patients with lamivudine-resistant HBV infection, the use of tenofovir or tenofovir-emtricitabine is recommended. Before the approval of these agents, adefovir was the preferred drug, and in a multicenter, open-label study of 128 patients awaiting LT—98% of whom had lamivudine-resistant HBV infection—81% achieved an undetectable HBV DNA level after 48 weeks of treatment with adefovir, with a mean change in serum HBV DNA of −3.5 log ₁₀ copies per milliliter. Although there are no studies of tenofovir or tenofovir-emtricitabine in lamivudine-resistant HBV infection in patients with decompensated cirrhosis, their use in this population is supported by data in patients with compensated cirrhosis.

Safety is a major concern with any drug in patients with decompensated cirrhosis, many of whom have concurrent renal dysfunction. In a report of 16 patients with cirrhosis and chronic hepatitis B treated with entecavir, five patients, all of whom had a baseline Model for End-Stage Liver Disease (MELD) score of 20 or higher, developed lactic acidosis between 4 days and 240 days after treatment initiation. Lactic acidosis is a listed potential side effect of all nucleoside/nucleotide analogs, and patients with decompensated cirrhosis may be a higher-risk group. In a small follow-up study, development of lactic acidosis was again noted after initiation of entecavir therapy in a patient with decompensated cirrhosis. The frequency and magnitude of lactic acidosis were not different from those in similar untreated controls, suggesting that the perceived safety signal may be related to the natural history of decompensated cirrhosis as opposed to entecavir. Regardless, awareness of this potential complication in patients with decompensated cirrhosis is necessary. Additionally, all approved antiviral drugs for HBV infection require dose adjustment in the setting of renal dysfunction (creatinine clearance less than 60 mL/min).

Prophylactic Therapy

Historical Perspective

Early studies using HBIG monotherapy for prevention of graft reinfection after LT demonstrated that 81% to 86% were free of recurrence of HBV infection at 2 years when high-dose HBIG was used to maintain hepatitis B surface antibody (anti-HBs) titers higher than 500 IU/L. With the subsequent approval of lamivudine and later adefovir, the combination of HBIG and a nucleoside/nucleotide analog became established at the standard of care for HBV-infected LT recipients in the mid-1990s. Three meta-analyses have clearly demonstrated the superiority of HBIG plus lamivudine over HBIG alone, with a reduction in posttransplant recurrence of HBV infection by 62% to 81% with the combination regimen. For the past 2 decades the use of the combination of HBIG and nucleoside/nucleotide analogs has prevailed as the primary management strategy, with a more individualized approach to prophylaxis used in recent years ( Fig. 53-7 ).

The need for lifelong prophylactic therapy is supported by findings of HBV DNA and covalently closed circular DNA in serum, peripheral blood mononuclear cells, and/or liver of patients without serologic evidence of recurrence of HBV infection. Of 25 patients who received HBIG and lamivudine combination therapy continuously after LT and remained HBsAg negative, HBcAg was found in 4 (16%) by immunochemical staining of liver biopsy specimens. In two studies that included 66 patients without evidence of recurrence of HBV infection, total HBV DNA and HBV covalently closed circular DNA was detected in the posttransplant liver biopsy specimens of 23 (35%) and 11 (17%) of patients, respectively. Although there may be a subset of patients in whom HBV infection has been completely eradicated after LT and/or whose own immune system can control the infection without prophylaxis, thus far the diagnostic tools to identify this group of patients are limited. Evaluation of total and covalently closed circular DNA detection in liver tissue may be helpful in identifying patients who may be able to undergo prophylaxis withdrawal. However, further study is required to fully understand how to incorporate such tests into posttransplant decision making.

Treatment of Recurrent Hepatitis B Virus Disease

Given the safety, tolerability, and efficacy of current prophylactic therapies for the prevention of HBV reinfection, recurrence of HBV infection after LT is uncommon. In patients in whom recurrence develops despite optimal prophylaxis, long-term suppressive therapy is required to prevent progressive fibrosis and graft loss. Lifelong antiviral therapy prolongs graft survival, and with current antiviral options graft loss from recurrent HBV disease is very infrequent. Given the need for long-term therapy, a drug or drug combination with a low likelihood of treatment failure due to HBV resistance is desirable, and monitoring of HBV DNA levels at regular intervals is needed.

The limitations of suboptimal antiviral therapy after LT are highlighted by the previous experience with lamivudine. High rates of lamivudine resistance, approaching 40% after 1 year to 5 years of follow-up, were seen with posttransplant lamivudine monotherapy. Several reports have demonstrated the emergence of multidrug-resistant HBV strains. Preexisting viral variants, as well as the selection of mutations during exposure to sequential antiviral therapies and HBIG, contribute to the emergence of complex multidrug-resistant HBV. Because the complexity of management of recurrent HBV infection is greater in the setting of multidrug-resistant HBV, choosing drugs with a high barrier to resistance as first-line drugs in the transplant setting is the best strategy.

For transplant recipients with lamivudine resistance, there is cross-resistance with the other nucleoside analogs telbivudine and emtricitabine and reduced efficacy of entecavir. Thus for patients with resistance to lamivudine or any of the nucleoside analogs, the only treatment options are the nucleotide analogs adefovir and tenofovir ( Table 53-5 ). Combination therapy is recommended once drug resistance has been documented to minimize the risk of subsequent treatment failure and the development of multidrug-resistant HBV. The largest post-LT study of combination antiviral therapy for patients with recurrent HBV and lamivudine resistance involved 241 patients treated with adefovir plus lamivudine; 78% achieved undetectable serum HBV DNA by week 144. Adverse events of any kind resulted in discontinuation of treatment in only 4% of patients. Published experience with the use of tenofovir is limited but one study that included eight lamivudine-resistant patients who received tenofovir reported that seven of the patients (88%) had undetectable HBV DNA levels after a median follow-up of 19.3 months (range 14 months to 26 months). Given its high potency against HBV in the nontransplant setting, tenofovir is preferred for treatment of patients with recurrent HBV infection and nucleoside-resistant HBV (see Table 53-5 ).

All five oral antiviral treatments for HBV—the two nucleotide analogs (adefovir and tenofovir) and the three nucleoside analogs (lamivudine, entecavir, and telbivudine)—are primarily renally eliminated and therefore dose reductions and/or increased dose intervals are required in patients with renal insufficiency. A decrease in renal function has been observed in patients treated with tenofovir or entecavir, an outcome LT recipients are already at higher risk of because of exposure to calcineurin inhibitors for immunosuppression. It is therefore of interest that two recent studies have shown that use of telbivudine in either the pretransplant or the posttransplant setting is associated with improvement in renal function. However, telbivudine is not a preferred antiviral because of the risk of resistance and associated side effects of myopathy and polyneuropathy.

Management of Hepatitis B in Special Populations

Hepatitis D Virus–Coinfected Patients

The 10-year graft and patient survival rates among HBsAg-positive European patients who underwent LT for decompensated cirrhosis were 86% and 80% for HDV-infected LT recipients compared with only 68% and 64% for HBV-monoinfected LT recipients ( Fig. 53-8 ). LT recipients coinfected with HBV/HDV and HBV-monoinfected LT recipients who underwent LT for HCC had similar patient and graft survival rates. In the absence of HBV prophylaxis, recurrent HDV infection occurs in most patients (≥80%) and disease manifestations coincide with recurrence of HBV infection. The universal use of prophylactic HBV therapy has greatly reduced the burden of recurrence of HDV infection. Long-term administration of HBIG monotherapy in 68 HDV-coinfected patients resulted in a 5-year actuarial survival rate of 88%. Although liver HDV antigen or serum HDV RNA was detected in 88% of patients within the first year, active HBV and HDV replication and clinical hepatitis developed in only 7 of 68 patients (10%). Because HDV requires HBsAg to produce virions, prevention of HBsAg recurrence may be of particular importance in preventing recurrence of HDV infection after LT. For this reason prophylaxis using a combination of HBIG and nucleoside/nucleotide analogs may be the preferred approach (see Table 53-3 ). A more recent study that included 25 patients with HDV infection who received HBIG and lamivudine combination immunoprophylaxis reported no recurrences of HDV or HBV infections after a mean follow-up of 40 months (range 13 months to 74 months).

Natural History After Liver Transplantation

Viral recurrence after LT is universal in patients who are viremic at the time of transplantation. Alanine or aspartate aminotransferase levels are elevated persistently or intermittently in most transplant recipients, but up to 30% of patients have persistently normal levels despite the presence of histologic damage in biopsy specimens. Approximately 60% to 80% of recipients show evidence of recurrent histologic disease on liver biopsy by the end of the first year after LT. Liver stiffness measurement via transient elastography is useful in identifying patients with significant fibrosis. Delayed, spontaneous clearance of HCV infection has been reported after LT but is rare.

Progression of HCV-related disease is accelerated in LT recipients in comparison with immunocompetent patients with HCV infection. The rate of HCV-associated progression of fibrosis is nonlinear and highly variable ( Fig. 53-9 ). The estimated median time to cirrhosis is 9 years but in up to 30% of untreated patients cirrhosis develops within the first 5 years. Once cirrhosis has become established, patients have a 30% to 42% annual risk of decompensation—a rate that is markedly elevated relative to that of immunocompetent patients with HCV-associated cirrhosis.

Overall survival is reduced in HCV-infected patients in comparison with HCV-negative patients, with 5-year patient and graft survival rates of 64% to 70% and 57% to 76% respectively, and 10-year patient and graft survival rates of 51% to 69% and 57% to 63%, respectively ( Fig. 53-10 ; see also Fig. 53-8 ). In a large retrospective study using data from 11,000 transplant recipients (4400 were HCV positive) available through the United Network for Organ Sharing (UNOS) registry, patients who received transplants for HCV-related liver disease had a 23% increased risk of death and a 30% increased risk of graft loss at 5 years when compared with patients who received transplants for non–HCV-related causes (see Fig. 53-10 ). The major causes of death and graft loss in HCV-infected transplant recipients are complications related to recurrent infection. These results reflect the natural history of HCV largely in the absence of effective HCV therapy. Outcomes are expected to improve significantly related to the availability of safe and highly effective direct antiviral drug therapies for LT recipients (see “ Prevention and Treatment of Recurrent Disease ”).

Assessment of Hepatic Fibrosis in Liver Transplant Recipients

Although liver biopsy has been the standard method for assessing disease severity, this procedure is cumbersome as a repeated measure, is associated with some risk, and may understage the severity of fibrosis, especially with smaller specimens and those stained with hematoxylin and eosin alone (without trichrome). Given these limitations of liver biopsy, alternative methods for staging disease have been examined.

Transient elastography measures the velocity of a low-frequency shear wave across the hepatic parenchyma and correlates impedance with the severity of fibrosis. In a meta-analysis of six studies of transient elastography in LT recipients with HCV infection, the pooled sensitivity was 83% [95% confidence interval (CI) 77% to 88%] and the specificity was 83% (95% CI 77% to 88%) for the diagnosis of significant fibrosis (fibrosis stage ≥2 for METAVIR, Scheuer or ≥3 for Ishak). For the diagnosis of cirrhosis by transient elastography, the pooled estimate for sensitivity was 98% (95% CI 90% to 100%) and for specificity was 84% (95% CI 80% to 88%). However, there was significant heterogeneity among the studies examined, with variation of elastography score cutoffs for significant and advanced fibrosis. Transient elastography may be useful prognostically, with one study showing that a transient elastography score of 8.7 kPa or more at 1 year after LT was associated with liver decompensation at 5 years after LT in 47% of patients. Measurement of the hepatic venous pressure gradient has also been evaluated as a predictor of disease severity and graft survival. The disadvantage of hepatic venous pressure gradient measurement as a predictive test is its invasive nature and the technical expertise needed. Serum fibrosis markers have also been studied as an alternative method of staging recurrent disease but have not been widely adopted. Thus transient elastography and liver biopsy are the usual means of assessing fibrosis severity in HCV-infected transplant recipients.

Pathology of Hepatitis C After Liver Transplant

Although findings may differ, early histopathologic features of recurrent HCV infection include lobular inflammation and focal apoptotic hepatocyte necrosis ( Fig. 53-11, A ). Steatosis is a nonspecific finding of early HCV infection, and portal inflammation—mainly with mononuclear cells often as lymphoid aggregates—may be found as the disease progresses. Any bile duct injury, if present, is typically mild. Severe necroinflammatory lesions, including focal necrosis, interface hepatitis, and confluent necrosis, may be seen and are highly associated with the early development of cirrhosis. Over time, the histologic appearance of recurrent chronic HCV infection is indistinguishable from that seen in the nontransplant setting (see Fig. 53-11, B ).

An aggressive variant of recurrent HCV infection has been recognized and called severe cholestatic hepatitis C ; it occurs in 2% to 8% of patients who receive a transplant for HCV-related liver disease. Initially labeled fibrosing cholestatic hepatitis on the basis of similarities to fibrosing cholestatic hepatitis in transplant recipients with recurrent hepatitis B, cholestatic hepatitis C is characterized clinically by severe hyperbilirubinemia (mean rise in serum bilirubin of 24.7 mg/dL) in the setting of high HCV RNA levels, typically occurring within the first 2 years (often within the first 6 months) after LT. Examination of biopsy specimens reveals lobular inflammation, bile duct proliferation, and cholestasis; areas of bridging and confluent necrosis can be rapidly replaced by fibrosis (see Fig. 53-11, C ). In the absence of treatment, cholestatic hepatitis can lead to early graft loss.

Another variant of recurrent hepatitis C is plasma cell or autoimmune-like hepatitis. This has been described primarily in the context of antiviral therapy with interferon and ribavirin and has unique histologic findings. Some cases are associated with elevated levels of autoantibodies (antinuclear antibody, anti–smooth muscle antibody, and anti–liver-kidney microsomal antibody) and elevated immunoglobulin levels. The key histologic features include severe interface inflammatory activity consisting predominantly of plasma cells and perivenular necroinflammation (see Fig. 53-11, D ). Biopsy-proven acute cellular rejection before treatment initiation (hazard ratio [HR] 4.87, p = 0.009) and the type of immunosuppression at the time of initiation of treatment (tacrolimus has a lower risk than cyclosporine, HR 0.25, p = 0.02) have been associated with development of plasma cell hepatitis in the context of interferon therapy. Patients with plasma cell hepatitis have a significantly higher rate of graft failure compared with patients without plasma cell hepatitis. Clinical and histologic responses to treatment with corticosteroids and amplification of baseline immunosuppression have been associated with good outcomes.

Acute Rejection in Patients With Hepatitis C

Histopathologically, acute rejection (AR) may be difficult to differentiate from recurrent HCV infection, with some but not all studies reporting low interobserver and intraobserver agreement in the diagnosis. Because recurrent HCV infection occurs universally after LT in the absence of treatment, most biopsy samples will have evidence of HCV disease, and the features of acute cellular rejection are superimposed on that background. Features that are characteristically associated with acute cellular rejection—and not recurrent HCV infection—include bile duct injury and necrosis with overlapping nuclei, endotheliitis, and inflammatory infiltrates around the portal tracts consisting of eosinophils, lymphocytes, and occasional neutrophils.

New tools to improve the accuracy of diagnosis of AR in the setting of HCV infection have been sought. The protein MxA, a marker for type I interferon production that is strongly expressed by hepatocytes in the presence of HCV, has yielded mixed results. In a retrospective study of 54 HCV-infected patients with or without acute cellular rejection, the Cylex immune function assay, which measures levels of adenosine triphosphate released from CD4 ⁺ T cells, found that a cutoff of 220 ng/mL had 88.5% sensitivity and 90.9% specificity for identifying acute cellular rejection. Various immunohistochemical stains, including those for lymphocyte expression of minichromosome maintenance protein 2, C4d, and IG222 monoclonal antibody against HCV E2 glycoprotein, in addition to CD28 expression on peripheral blood mononuclear cells, have shown promise in small single-center studies. Finally, in a recent study of 54 liver allograft samples from unique HCV-infected recipients, microarray analysis was done on a training set ( n = 32) and a validation set ( n = 19). One hundred seventy-nine probe sets were differentially expressed among groups, with 71 exclusive genes between recipients with recurrence of HCV infection alone and HCV with acute cellular rejection. The best-fitting model included 15 genes and had an accuracy of 100% in the training set and a sensitivity of 50%, specificity of 91%, positive predictive value of 71% and negative predictive value of 80%. None of these biomarkers are ready for routine clinical use, and liver biopsy remains the gold standard for diagnosis of AR.

Factors Associated With Disease Progression and Graft Loss

Several viral-, recipient-, donor-, and transplant-related factors influence the rate of progression of HCV-related disease and the risk of graft loss ( Table 53-6 ). Before the availability of highly effective HCV therapies, the identification of factors that could be modified to reduce the risk of disease progression was the mainstay of management. The recipient-related factors most consistently associated with worse posttransplant outcomes include older age, African American race, and HIV coinfection. The most important donor-related factor associated with the risk of recurrent cirrhosis is donor age. Posttransplant factors of importance include a history of treated AR, diabetes, and cytomegalovirus (CMV) infection. Immunosuppression is of some importance but no specific immunosuppressive regimen has established itself to be superior to others. Viral factors have not been consistently linked with the risk of recurrent disease or survival.

Recipient-Related Factors

Older recipient age has been shown to be associated with patient death and graft loss but not disease progression. A UNOS registry–based study found a modest but statistically significant effect, with patients older than 60 years experiencing a 5-year survival rate of 66.5% versus 70.3% for recipients aged 60 years or younger ( p < 0.01). Some but not all studies have shown recipient female sex to be associated with increased graft loss and recurrent HCV-related disease progression.

African American race has been associated with a higher risk of death and graft loss in HCV-infected recipients. Among nearly 3500 HCV-related transplant recipients from the Scientific Registry of Transplant Recipients, African Americans had an approximately 30% higher risk of graft loss and death compared with Caucasians. Studies linking race with recurrent HCV infection severity have also shown an association between African American race and increased rate of progression of fibrosis, with a statistically significant HR of 1.47 in multivariate analysis in one study. A separate donor risk index (DRI) for African Americans with HCV was proposed including donor age, race, and cold ischemia time and resulted in the correct reclassification of 27% of patients compared with the original DRI.

LT recipients with HIV-HCV coinfection have decreased graft survival rates and higher rates of recurrent HCV cirrhosis compared with HCV-infected LT recipients without HIV ( Fig. 53-12 ). Graft survival rates differ among studies but range from 72% to 94% at 1 year, to 59% to 80% at 2 years and 29% to 51% at 5 years. Fibrosis scores were significantly higher in the coinfected group than in the monoinfected group at 2 years after LT (1.4 ± 1.1 vs. 2.4 ± 1.3, p = 0.01), and death was more common in the patients in whom cholestatic hepatitis developed. Additionally, the 3-year incidence of treated AR was 1.6-fold higher for the patients coinfected with HCV and HIV than in monoinfected recipients (39% vs. 24%, respectively), and this may have contributed to the higher risk of HCV progression after LT.