Transplantation of hepatitis B surface antigen–positive livers into hepatitis B virus–positive recipients and the role of hepatitis delta coinfection†
See Editorial on Page 875
The scarcity of liver donors requires consideration of grafts from sources not previously used. Allografts from hepatitis B surface antigen (HBsAg)-carriers without a significant liver disease have been proposed for liver transplant recipients with hepatitis B virus (HBV)-related cirrhosis and hepatocellular carcinoma (HCC). Combination prophylaxis schemes against HBV post-liver transplantation (LT) recurrence are currently available; the efficacy of those schemes in HBV-related cirrhosis and HCC must be assessed. This report describes the allocation of HBsAg-positive grafts in three HBsAg-positive recipients, with HBV-related cirrhosis and evolving HCC lesions, two of them with hepatitis Delta virus (HDV) coinfection. Patients were administered anti-hepatitis B immunoglobulins (HBIGs) and lamivudine in order to prevent HBV recurrence. In spite of anti-HBV prophylaxis, HBV infection did persist after LT in all patients (no serum clearance of HBsAg). HBV replication assessed by serum HBV deoxyribonucleic acid (DNA) presence was detected in the first month after LT in the 3 recipients. A prompt HDV reinfection with a clinical and histological pattern of hepatitis was observed in the 2 HBV / HDV coinfected recipients. In 1 of them, an evolving chronic hepatitis required a second LT. The non-HDV–infected patient showed an uneventful follow-up, but the lack of the neutralizing effect of HBIGs and the high risk of escape mutants forced the addition of adefovir-dipivoxil to lamivudine, in order to prevent viral variants and hepatitis recurrence. In conclusion, allografts from HBsAg-positive donors in HBsAg-positive recipients are associated with the persistence of the HBsAg after LT due to the failure of HBIG prophylaxis, even if lamivudine does inhibit virion production. This condition favors HDV replication and HDV hepatitis recurrence in coinfected patients. The allocation of HBsAg-positive grafts in HBsAg-positive recipients could be justified only in recipients without HDV coinfection and a combined prophylaxis with lamivudine and adefovir-dipivoxil is currently the best way to manage escape mutants in these recipients. (Liver Transpl 2005;11:922–928.)
The worldwide disparity between patients requiring liver transplantation (LT) and the availability of organs has induced many LT centers to accept marginal grafts like the ones from donors with evidence of past exposure to the hepatitis B virus (HBV), that is with serologic markers of past-HBV infection such as the antibody to hepatitis B core antigen (antiHBc) in absence of hepatitis B surface antigen (HBsAg). Transmission of hepatitis B infection has been reported in liver transplant recipients whose liver donors were negative for HBsAg and positive for antiHBc.1–4 This is explained by the presence of intrahepatic HBV deoxyribonucleic acid (DNA) (seldom in serum) by polymerase chain reaction (PCR) in the HBsAg-negative, antiHBc-positive population.1, 5
HBV posttransplantation recurrence, defined by the appearance of HBsAg in serum samples anytime after LT, has been reported in studies on HBsAg-negative recipients of donors with serologic signs of previous HBV exposure. In these recipients, the incidence of HBV post-LT reactivation increases from 16 to 95% in the absence of HBV prophylaxis.6 Many effective prophylactic schemes are currently available in LT centers to prevent HBV recurrence or de novo hepatitis.6–8 Selective protocols based on donor and recipient risk factors for HBV post-LT recurrence have been proposed to prevent hepatitis B reinfection of the graft.8, 9 Due to successful anti-HBV prophylaxis schemes, it is nowadays very well established that the allocation of antiHBc-positive livers does not affect graft or patient survival.1–6
It is generally recommended that antiHBc-positive livers should be preferentially offered to HBsAg-positive recipients who must undergo anti-HBV prophylaxis. Recipients other than HBsAg-positive ones can receive these grafts as well, such as those with signs of previous exposure to HBV (anti-HBsAg-positive and / or antiHBc-positive) or even naïve patients provided that an effective prophylaxis is administered.
Due to the organ shortage, grafts from HBsAg-carriers without a significant liver disease, as confirmed by histology on liver biopsy, have been occasionally proposed.6, 10 However, many doubts remain about the protective effect of hepatitis B immunoglobulins (HBIGs) and lamivudine prophylaxis in recipients of HBsAg-positive grafts; clinical reports on this effect are not available at this time. Data from the literature show that, in settings other than LT, in HBsAg-positive carriers who do respond to lamivudine therapy, HBV prophylaxis with HBIG offers a very low neutralizing effect in spite of high dosages of intravenous HBIG and the inhibition of virion production due to lamivudine.7
In Italy, the huge disparity between potential liver recipients (more than 20,000 deaths/year due to cirrhosis or hepatocarcinoma)11 and transplanted liver grafts (about 1,000 every year),12 has forced the Italian Health Authority (Ministry of Health) to propose and officially establish (in December 2003) an experimental protocol concerning the allocation of HBsAg-positive grafts without HDV coinfection in HBsAg-positive recipients.12, 13
In this report, we describe the first 3 cases of HBsAg-positive liver recipients of HBsAg-positive grafts performed at the Liver Transplantation Centre, Molinette Hospital, Turin, Italy. The 3 patients underwent LT after signing a protocol fulfilling the ethical guidelines of the Helsinki Declaration, 1975. Informed consensus was obtained from each patient before transplantation.
Materials and Methods
A 57-year-old man, coinfected with HBV and HDV, had a histological diagnosis of cirrhosis in 1999 during a laparoscopic cholecystectomy for asymptomatic cholelithiasis. Serum HBV markers were negative for hepatitis B “e” antigen (HBeAg) and positive for the antibody to HBeAg (antiHBe). In 2002, he suffered from 2 episodes of variceal bleeding that were successfully treated with endoscopic eradication. In March 2003, the patient was admitted to the hospital for the first ascitic decompensation; 3 hepatic nodular lesions (diameter <3 cm) compatible with hepatocellular carcinoma (HCC) were discovered by ultrasound and computerized tomography. After percutaneous alcoholizations of the lesions, the patient was enlisted for LT and a strict ultrasound follow-up (every 3 months) was performed. Lamivudine was started in September 2003 (HBV DNA serum level before lamivudine: 42,200 copies/mL) and a complete virologic response was achieved in 71 days of treatment. No evidence of HDV replication was found pre-LT, or development of HBV resistant strains due to mutations in the tyrosine, methionine, aspartate, aspartate (YMDD) domain of the viral polymerase gene. The patient underwent transplantation on January 13, 2004.
In June 2003, a 43-year-old man suffering from a HBV / HDV chronic hepatitis since 1983 was admitted to the Gastroenterology Department with 3 HCC nodules (diameter <3 cm) and ascitic decompensation. Six months later, an episode of hepatic encephalopathy occurred. Serum HBV markers were negative for HBeAg and positive for antiHBe. The patient underwent LT on February 26, 2004. No evidence of HDV replication was found before surgery. Before LT, the patient was not treated with lamivudine because serum HBV DNA was persistently negative.
A 56-year-old man with a diagnosis of chronic hepatitis B (without HDV infection) since 1980 was admitted to the hospital for ascitic decompensation and encephalopathy. Esophageal varices without bleeding and 2 HCC nodules (1 of these with a diameter of 3 cm) were diagnosed. Serum HBV markers were negative for HBeAg and positive for antiHBe. While he was enlisted for LT, lamivudine was started (HBV DNA serum level before lamivudine: 545,000 copies/mL). A complete virologic response was achieved in 4 months without the development of HBV lamivudine-resistant strains. He underwent LT in November 17, 2003.
The graft was from a 43-year-old man with a mild elevation of liver enzymes (alanine amino transferase [ALT] = 47 U/L, aspartate aminotransferase = 107 U/L) and of the prothrombin time (international normalized ratio = 1.24); thrombocytopenia (101,000 × 109/L) was present. Serologic HBV markers were positive for HBsAg, antiHBc and HBV-DNA (22,000 copies/mL), negative for HBeAg and antiHBe. HDV serologic markers (total anti-HDV antibody and Immunoglobulin M (IgM) to HDV) were negative. A liver biopsy showed steatosis, minimal portal fibrotic enlargement, and a sporadic presence of HBsAg (<1%), in absence of hepatitis B core antigen by immunohistochemistry (DAKO, Carpenteria, CA).
The liver donor was a 48-year-old woman with normal levels of liver enzymes, mild coagulopathy (international normalized ratio = 1.35), and mild thrombocytopenia (135,000 × 109/L); she was positive in serum for HBsAg, antiHBc, antiHBe, and HBV-DNA (412 copies/mL). HDV serologic markers were negative. A liver biopsy showed aspecific degeneration of the hepatocytes together with a sporadic presence of HBsAg (<1%), in absence of reactivity to the viral core antigen by immunohistochemistry.
The liver donor was a 20-year-old man with high levels of liver enzymes (aspartate aminotransferase = 565 U/L, ALT = 207 U/L), an important coagulopathy (international normalized ratio = 1.65), and mild thrombocytopenia (133,000 × 109/L); the HBV serologic features were positive for HBsAg, antiHBc, antiHBe, and HBV-DNA (7,700 copies/mL); HDV serologic markers were negative. A liver biopsy showed mild chronic inflammation and mild portal enlargement; HBsAg was positive (30%), in absence of reactivity to hepatitis B core antigen by immunohistochemistry.
The antirejection regimen consisted of cyclosporine and steroids; cyclosporine (Sandimmun [Neoral]; Novartis, Basel, Switzerland, 100 mg/mL oral solution) was started at 5 mg/kg every 12 hours, adjusting the dosage to the blood concentration 2 hours after the given dosage (C2). In order to avoid overimmunosuppression, steroids were gradually discontinued in patient 2 and 3 within 6 months after LT. In patient 1, steroids were briefly discontinued, then reintroduced due to an aggressive episode of acute hepatitis.
Prophylaxis of HBV Recurrence After LT
Recipients were administered the Centre's standard schedule of anti-HBIGs: 10,000 IU of HBIG intravenously in the first 2 days, then 6,000 IU for 4 days and 2,000 IU for 2 days more, followed by 3,000 IU every 7–15 days. Lamivudine treatment was always associated with the administration of HBIG. Immunoglobulins were suspended after 7, 12, and 20 weeks from LT in patients 1, 2, and 3, respectively, due to the lack of neutralizing effect.
Serologic Assays for HBV and HDV Markers of Infection
Serologic markers of HBV infection were tested with the Chemiluminescent Microparticle Immunoassay technology (Abbot Laboratories, North Chicago, IL) either in liver recipients or donors. HBsAg was tested in a qualitative format whose detection limit was 0.05 IU/mL, according to World Health Organization standards. Serum markers of HDV infection were studied by enzyme-linked immunosorbent assay (Sorin, Saluggia, Italy). Total antibody to HDV was detected with a competitive enzyme-linked immunosorbent assay technique in a microwell format with a recombinant HDV antigen. IgM antibody to HDV was detected with a μ-capture enzyme-linked immunosorbent assay technique using a genetically engineered HDV antigen and an anti-HDV antibody fragments (mouse monoclonal).
Serologic and virologic data were collected in recipients at the time of the enrollment in the LT waiting list, at the time of transplantation, weekly for the first month after surgery, monthly for other 5 months, and then every 3 months.
Molecular assays for HBV and Delta Viruses
HBV DNA Detection
HBV DNA was quantified in serum samples with the commercially available quantitative PCR system COBAS Amplicor HBV Monitor (Roche Diagnostics, Basel, Switzerland). DNA was extracted from 100-μL serum specimens according to the manufacturer's procedure. The sensitivity of the system was 200 copies/mL with a linear range up to 200,000 copies/mL.
HBV Polymerase Gene Mutants
HBV lamivudine-resistant strains were detected with a reverse hybridization line probe assay (INNO-LIPA; Innogenetics, Ghent, Belgium) after amplifying domain B and C of the viral polymerase gene with specific primers.
HDV Ribonucleic Acid Detection
Ribonucleic acid (RNA) extraction was performed from 140 μL of serum samples with a procedure based on Qiagen silica-gel membranes adsorbed onto spin columns (Qiagen, Milan, Italy). RNA was reverse-transcribed with random examers; a nested-PCR was performed for the amplification of a 406-bp fragment corresponding to position 883–1288 of the HDV genome.14 PCR products were visualized by electrophoresis on a 2% agarose gel and ethidium bromide staining. The PCR detection limit was 10 copies/reaction.
The HBsAg-positive grafts were allocated to the above described LT candidates due to the evidence of a rapid evolution of HCC while in the waiting list: signs of increasing nodule growth in patient 1 and 2 and a new HCC lesion in patient 3. Recipient and related-donor characteristics pre-LT and after-LT are shown in Tables 1 and 2, respectively.
Table 1. Pre- and Post-Liver Transplantation Characteristics of Recipients
|Before LT|| || || |
| HBV DNA (copies/ml)†||Negative||Negative||Negative|
| HDV RNAo||Negative||Negative||Nd§|
| Lamivudine therapy||Yes||No||Yes|
|After LT|| || || |
| Follow-up (months)||12||10||13|
| HBV prophylaxis‖||HBIG + Lam||HBIG + Lam||HBIG + Lam + Adv|
| HBV hepatitis||Yes||Yes||No|
| Liver enzymes||Elevated (++)||Elevated (+)||Normal|
| HBV DNA†|| || || |
| Month 1||1180||54100||470|
| Month 3||3900||Negative||Negative|
| Month 6||Negative||Negative||Negative|
| Month 9||Negative||Negative||Negative|
| HDV RNA‡|| || || |
| Month 1||Positive||Positive||—|
| Month 3||Positive||Positive||—|
| Month 6||Positive||Positive||—|
| Month 9||Negative||Positive||—|
| IgM-HBc/IgM-HDV|| || || |
| Month 1||−/+||−/Not tested||−/Nd**|
| Month 3||−/+||−/+||−/Nd**|
| Liver biopsy||At week 9: acute hepatitis (HBsAg-, HDV: 30%)||At week 16: chronic hepatitis (HBsAg: 15%, HDV: 90%)||—|
| Outcome||Alive||Alive (re-LT)||Alive|
Table 2. Characteristics of Related-Donors
|Platelets (× 109/L)||110,000||135,000||133,000|
|HBV DNA (copies/mL)†||22,000||412||7,700|
|Liver biopsy||Steatosis with minimal portal enlargement (HBsAg <1%, HBcAg absent)||Hepatocytes aspecific degeneration (HBsAg <1%, HBcAg absent)||Mild chronic inflammation with mild portal enlargement (HBsAg: 30%, HBcAg absent)|
Clinical and Virologic Outcome After LT
The patient showed an uneventful postsurgical outcome and was discharged 2 weeks after LT. Despite the administration of HBIG and lamivudine, serum HBsAg never became negative after surgery. Liver enzymes were normal for the first 4 weeks; very low levels of HBV DNA (up to 1,180 copies/mL) were detected in serum from the third week after LT, without evidence of circulating HBeAg or mutations at the YMDD polymerase motif. HDV RNA appeared in serum in the first months after LT (Table 1). Four hepatitis flares occurred between weeks 5 and 12 post-LT. Three liver biopsies confirmed the presence of acute hepatitis (at week 9: immunohistochemistry showed HDV antigen in 30% of the hepatocytes; Table 1). In order to reduce overimmunosuppression, steroids were discontinued, but a new aggressive episode of acute hepatitis with coagulopathy and jaundice occurred. A mild improvement of liver function tests was observed when steroids were reintroduced (at week 14 post-LT) with bilirubin and transaminase normalization at week 18. At month 9 and 1 year after LT, transaminases and bilirubin were normal, serum HBV DNA and HDV RNA were negative, but the HBsAg was always positive while on lamivudine treatment.
Liver enzymes normalized 2 weeks after LT and remained normal during the following 3 months. The serum HBsAg never became negative in spite of the administration of HBIG and lamivudine. Serum HBV DNA and HDV RNA were detected 2–4 weeks and 4–9 weeks from LT, respectively. Steroids were discontinued at month 6 from LT. A rise of aspartate aminotransferase and ALT at month 4 forced a liver biopsy that showed a pattern of mild inflammation and the presence of HBsAg (15%) and hepatitis Delta surface antigens (90%) in absence of hepatitis B core antigen by immunohistochemistry. One month later, a second liver biopsy was performed due to ascitic decompensation refractory to therapy and a new ALT and aspartate aminotransferase rise. Histology showed a chronic hepatitis with significant fibrosis (3 / 6 fibrosis according to the Ishak score). The patient was enlisted for retransplantation due to the histological findings, recurrent ascitic decompensation, and increasing renal dysfunction. Ten months from the first LT, the patient underwent a second LT. At the time of retransplantation, the patient was treated with lamivudine and high HBIG dosage (10,000 IU intravenous from the anhepatic phase, then daily during the first postoperative week, and weekly thereafter in order to maintain an anti-HBsAg titer higher than 500 IU/L).
Hepatic enzymes normalized in 4 weeks from LT and were always normal during the follow-up (385 days). The patient never became HBsAg-negative, in spite of the administration of HBIG and lamivudine. Very low levels of HBV replication (HBV DNA: 470 copies/mL) were detected in serum 5 weeks from surgery without evidence of mutations at the YMDD polymerase motif. The double immunosuppressive scheme of therapy was maintained for the fist 6 months after LT; no biopsies were performed because the patient was clinically stable. Adefovir-dipivoxil (Hepsera, Gilead Sciences, Foster City, CA) was added to lamivudine at week 24 in order to prevent the development of lamivudine-resistant strains and hepatitis B recurrence. After 13 months of adefovir and lamivudine, aspartate aminotransferase and ALT were normal; serum HBV DNA was negative, with a persistent positivity of the serum HBsAg.
The present scarcity of organ donors requires consideration of grafts from sources not previously used. Allografts from HBsAg-carriers without a significant liver disease, as confirmed by histology, have been proposed for recipients with HBV-related cirrhosis and hepatocellular carcinoma.6, 10 Combination schemes for prophylaxis of HBV post-LT recurrence are currently available whose efficacy in those particular conditions must be assessed.
The first Italian experience concerning HBsAg-positive recipients of HBsAg-positive grafts has been undertaken in 3 HBV-infected recipients, 2 of them with HDV coinfection. The patients were administered anti-HBV prophylaxis with HBIG and lamivudine in order to prevent HBV recurrence.
The major issue we faced in HBsAg-positive recipients of HBsAg-positive grafts was that in spite of anti-HBV prophylaxis, HBV infection did persist after LT in all patients. In fact, no serum HBsAg clearance was observed and serum signs of HBV replication (HBV DNA positive) were detected in all the 3 recipients, as previously reported.10 The main cause of persistent HBV infection was the ineffective neutralization of HBsAg levels that are normally produced by HBsAg-carriers, with the administration of HBIG, as previously observed in patients with chronic hepatitis B on lamivudine treatment.7 The relapse of HBV infection was also correlated with the immunosuppression regimen and particularly with the effect of steroids due to the activation of the steroid-sensitive transcriptional enhancer component of the HBV genome.
The second major issue we faced was HDV reinfection in the 2 HBV / HDV coinfected patients with a clinical and histological pattern of hepatitis, whereas in the non-HDV coinfected patient, liver enzymes were persistently normal during an uneventful follow-up. Histology was not available for this patient because the policy of our Centre requires a post-LT liver biopsy only if patients show clinical and functional symptoms and signs of hepatitis.
Patients 1 and 2 developed a prompt Delta reinfection within the first 3 months from LT (HDV RNA positive in serum), with low levels of HBV DNA in serum (Table 1). Biopsies from both of these patients showed the presence of Delta antigens in the liver. This data is in agreement with previously published works showing a fast Delta virus reinfection associated with liver HBV replication in the absence of HBIG prophylaxis.15, 16 In our patients, the failure of HBIGs in their neutralizing effect and the presence of a HBsAg-positive graft, which is the natural substrate for HDV replication, were the leading causes of HDV recurrence and hepatitis. In patient 2, a rapid evolving chronic hepatitis with 3/6 fibrosis (Ishak histology score) at month 4 after surgery, required a second LT performed 10 months after the first one.
Due to HBIG failure and discontinuation, our patients could be considered on lamivudine monoprophylaxis and consequently HBV replication could be interpreted as a predictor of polymerase mutant selection. In our experience, even when HBV DNA was positive in serum, no lamivudine-resistant strains were found during the follow-up of the patients.
In the 2 HDV-infected patients, a residual HBV replication at months 1 and 3 after LT, respectively (Table 1), allowed the penetration of HDV16 in the hepatocytes from natural reservoirs,15 thus inhibiting HBV replication and activating a specific immune response and a clinical and histological HDV. This hypothesis was confirmed in patient 1 by the evidence of hepatitis flares after the immune-reconstitution due to steroid suspension and a prompt clinical response when steroids were reintroduced, as reported in hepatitis C virus reinfection after LT.17
In patient 3, the absence of HDV coinfection and the residual HBV replication, together with the inefficacy of HBIG and the high risk of escape mutants, forced the addition of adefovir-dipivoxil to lamivudine, in order to prevent viral variants.
Even if no definitive conclusions can be withdrawn from our experience on HBsAg-positive grafts, due to the lack of long-term follow-up and histologic graft assessment, this report suggests that HBsAg-positive grafts in HBsAg-positive recipients could be justified only in recipients without HDV coinfection. In fact, allografts from HBsAg-positive donors in HBsAg-positive recipients are associated with the persistence of the HBsAg due to the failure of HBIG prophylaxis, in spite of the inhibition of virion production obtained with the administration of lamivudine. This condition favors HDV replication and HDV hepatitis recurrence. Further studies are required to address the issue of the importance of the level of HBV DNA in donor serum since its absence by standard molecular methods cannot rule out the risk related to HBV replication in liver.
Moreover, the allocation of HBsAg-positive grafts in HBsAg-positive recipients without HDV coinfection requires long-term monotherapy maintenance with lamivudine. This situation is at high risk for the development of polymerase mutants and hepatitis recurrence due to resistant strains, especially if patients have already undergone pre-LT HBV therapy, as previously reported.18 Therefore, in these patients a combined prophylaxis with lamivudine and adefovir-dipivoxil could be proposed as the best way to manage escape mutants and hepatitis recurrence.