Case Report of Lamivudine-Resistant Hepatitis B Virus Infection Post Liver Transplantation from a Hepatitis B Core Antibody Donor
The use of allografts from donors with hepatitis B core antibody in liver transplantation (LT) is associated with the risk of de novo hepatitis B virus (HBV) infection. Prophylaxis using hepatitis B Immune globulin (HBIg) and lamivudine alone or in combination has been reported. Yet, there are no standardized regimens and long-term efficacy is not known. We report a case of a patient who underwent LT for alcoholic liver disease who received an allograft from a donor with Hepatitis B core antibody. The patient had no previous exposure to HBV, was vaccinated against HBV, and had demonstrated Hepatitis B surface antibody present in serum before and 6 months after transplantation. Prophylaxis with short-term HBIg (1 week) and indefinite lamivudine was given. De novo HBV infection developed more than 3 years after LT with a lamivudine-resistant polymerase mutant containing the rtM204I and rtl180L/M mutations. We reviewed the risk of de novo post-LT HBV infection in recipients of livers from hepatitis B core antibody positive donors. High risk were HBV naïve recipients, moderate risk recipients had isolated hepatitis B surface antibody (anti-HBs) or hepatitis B core antibody (anti-HBc), while low-risk recipients had both anti-HBs and anti-HBc. We reviewed prophylaxis protocols reported in the literature and made recommendations for management.
The use of allografts from donors with hepatitis B core antibody (anti-HBc) has expanded the donor pool in liver transplantation (LT). These donors represented 3.8% of 1109 hepatic allografts used in the National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database from 1989 to 1994 (1). A recent series from our center revealed that 28% of the 402 recipients and 4.3% of all donors were anti-HBc (+) (2). The primary concern with the use of these grafts is the risk of de novo post-LT hepatitis B virus infection. In those patients requiring transplantation that are naïve to hepatitis B virus (HBV) exposure, the risk for de novo hepatitis has been previously reported to range from 37% to 100% in the absence of prophylaxis. Livers from anti-HBc donors have been shown to be the major risk factor for post-LT HBV infection in patients transplanted for non-HBV indications (1,3–12).
Unlike anti-HBc, the presence of donor hepatitis B surface antibody (anti-HBs) alone or in association with anti-HBc was not associated with an increased risk of infection (1,11). Thus, donor anti-HBs status does not need to be considered when assessing risk for de novo post-LT HBV infection. In addition, the presence of anti HBc (in the absence of hepatitis B surface antigen (HBsAg)) in the recipient prior to transplantation has negligible risk for post-LT HBV infection (13) even in the presence of hepatitis B virus deoxyribonucleic acid (HBV DNA) in the recipient serum or liver tissue (14).
We present a case of de novo HBV infection post-LT resulting from a liver from an anti-HBc (+) donor into an anti-HBs (+) recipient despite long-term lamivudine and short-term hepatitis B immune globulin (HBIg) treatment. We review the literature regarding de novo post-LT HBV infection associated with the use of grafts from anti-HBc donors in LT.
The patient was a 61-year-old Caucasian male with a 30-year history of significant alcohol consumption. He discontinued alcohol in June 1998, but presented to our institution in January 1999 with symptoms of muscle wasting, intermittent confusion, jaundice, edema and ascites requiring large-volume paracentesis. His past medical history was notable for an appendectomy as a child and an umbilical hernia repair in 1994. He had no known risk factors for viral hepatitis and had been vaccinated for HBV in the past.
Physical examination was significant only for tense ascites, lower extremity edema and asterixis. Initial laboratory testing showed WBC 1.9, Hemoglobin 7.8 g/dL, Platelets 42, Na 132 mEq/L, creatinine 0.8 mg/dL, alkaline phosphatase 273 U/L, total bilirubin 11.3 mg/dL, albumin 2.7 g/dL, ALT 38 U/L, and AST 77 U/L. Viral serologies revealed anti-HBs (+), HBsAg (−), anti-HBc (−), HCV Ab (−), HCV RNA (−), and HIV (−). A CT of the abdomen/pelvis demonstrated moderately large ascites, a small irregular liver, and a 1.6-cm low-density rounded lesion in the anterior segment of right lobe of the liver. Biopsy of the liver mass revealed regenerative nodular hyperplasia without malignancy. The patient was subsequently hospitalized for management of refractory ascites, encephalopathy and poor functional status. Due to deteriorating mental status, progressive refractory volume overload and synthetic dysfunction, the patient was transferred to the ICU and listed for LT at Status 2A (MELD 26). Laboratory parameters prior to LT revealed a total bilirubin 20.1 mg/dL, INR 1.9, albumin 2.8 g/dL, and creatinine 1.1 mg/dL.
On January 21, 1999, a liver from a 50-year-old donor was available. The donor tested anti-HBc (+), HBsAg (−) and hepatitis B e antigen (HBeAg) (−). Anti-HBs titer was not performed. Donor status for CMV, RPR, HCV, HTLV III and HIV were all negative. Repeat patient serologies revealed anti-HBs (+), HBsAg (−), HBcAb IgM (−) and total anti-HBc (−). Our clinical protocol at that time was to consider the use of these grafts in HBsAg (−) recipients with concomitant use of prophylactic intravenous HBIg and lamivudine. The risks associated with use of this graft were discussed at length with the patient and his family, and they agreed to accept the organ. LT was performed with full-sized graft and duct-to-duct anastamosis by piggyback technique on January 21, 1999. There were 8.25 h of cold ischemia time and 41 min of warm ischemia time. The patient received i.v. HBIg 10 000 IU during the anhepatic phase as well as postperfusion. He then received 10 000 IU i.v. daily on postoperative days 1 through 7 for a total of 90 000 IU. Lamivudine (Epivir-HB GlaxoSmithKline, Research Park, NC) 100 mg by mouth daily was started on the day of LT and was continued indefinitely as outlined per protocol. Lamivudine was not dose adjusted for renal insufficiency.
Postoperatively the patient received tacrolimus-based immunosuppression. The course was complicated by renal insufficiency and persistent confusion, but he was discharged in good condition on February 19, 1999. On May 27, 1999, tacrolimus was changed to cyclosporine and mycophenolate due to confusion and weight loss, which resulted in a return to baseline preillness mental status. In August 1999, the patient developed acute cellular rejection associated with an increase in alkaline phosphatase from 166 to 1724 U/L, ALT from 14 to 502 U/L, AST from 22 to 327 U/L, total bilirubin from 0.7 to 10.6 mg/dL. A liver biopsy revealed severe acute rejection with no evidence of significant necrosis or hepatitis. Treatment with pulse i.v. solumedrol (total 3 g) led to biochemical and histological resolution.
The patient had no further clinical complications and had uneventful annual evaluations including protocol liver biopsies in January 2000 and May 2001 (all with negative staining for HBV core and surface antigens). Lamivudine had been continued during the entire posttransplant period.
Despite feeling well in April 2002, he was found to have positive serologies for HBsAg, HBeAg and anti-HBc. The HBV DNA level was 3.64 × 108 copies/mL. Anti-HBs was negative and last demonstrated to be present on August 26, 1999. HBV resistance testing based on sequencing of the polymerase gene on April 24, 2002 demonstrated the presence of polymerase, specifically rtM204I and rtL180L/M mutations, associated with a diminished response to Lamivudine. Liver biopsy on April 12, 2002 revealed no significant histopathologic abnormality, but immunostaining demonstrated patchy positive staining for HBsAg, and showed rare positive cells for hepatitis B core antigen (HBcAg). The patient's liver function tests in April 2002 showed total bilirubin 0.8 mg/dL, alkaline phosphatase 237 U/L, albumin 4.4 g/dL, ALT 20 U/L, AST 25 U/L and GGT 45 U/L.
Adefovir 10 mg daily (Hepsera, Gilead Sciences, Inc, Foster City, CA) was added to his regimen of lamivudine 100 mg daily at this time. The patient had serial testing for HBV DNA quantification with a decrease from 1.02 × 109 copies/mL in May 2002 to 2.64 × 107 copies/mL in July 2002 to 7.56 × 105 copies/mL in April 2003. A liver biopsy on April 15, 2003 revealed mild portal lymphocytic infiltrate with no increased fibrosis and minimal activity (grade 1, stage 0). Immunostaining demonstrated diffuse HBcAg (moderate positive) but was negative for HBsAg. There was no finding of acute or chronic rejection. In October 2003, the patient had worsening of his renal function with an increase in baseline serum creatinine from 1.7 to 2.5 by April 2003. This led to initiation of rapamycin with discontinuation of cyclosporine and adjustment of the adefovir dose for renal insufficiency.
In August 2004, the patient was feeling clinically well and had no further changes in his immunosuppression or antiviral regimen. Laboratory testing revealed alkaline phosphatase 127 U/L, ALT 31 U/L, AST 34 U/L, albumin 4.4 g/dL, total bilirubin 0.4 mg/dL, creatinine 2.0 mg/dL, HBeAg (+), anti-HBe (−), HBsAg (+), anti-HBs (−) and anti-HBc (+) with HBV DNA now 39 735 IU/mL (2.22 × 105 copies/mL).
In June 2005, the patient was diagnosed with metastatic gastric carcinoma and died on July 12, 2005. Last liver tests were in May 2005 with ALT 25 U/L (45 normal), AST 26 U/L (31 normal), total Bilirubin 0.7 mg/dL and HBV DNA 5 914 IU/mL.
We have presented a case of lamivudine-resistant post-LT HBV infection from an anti-HBc (+) donor graft which developed over 3 years following LT. The patient had anti-HBs from vaccination prior to transplant and received prophylaxis with lamivudine (indefinite) and short-term HBIg (1 week) with anti-HBs titers present for at least 6 months. Jain et al., have also recently described the development of a YMDD mutant in 2 of 28 patients receiving anti-HBc (+) grafts despite the use of lamivudine indefinitely and HBIg (10 000 units daily for 4 days). This data emphasize the need to better define risk and to develop a more standard approach to using these grafts (15).
To assess the risk of post-LT HBV infection in recipients of anti-HBc (+) donor grafts, we reviewed 11 published studies with a total of 194 patients who received an anti-HBc (+) donor graft without HBIg or lamivudine prophylaxis. We grouped patients according to the recipient's status for hepatitis B markers (see Table 1). Naïve recipients [anti-HBc (−), anti-HBs (−)] had the highest incidence at 77%. Recipients with isolated anti-HBs (+) [anti-HBc (−)] had an incidence of post-LT HBV of 19%, while recipients with only anti-HBc (+) [anti-HBs (−)] had an incidence risk of 13%. However, there were no cases of de novo hepatitis in recipients with both anti-HBs (+) and anti-HBc (+).
Table 1. Cases de novo HBV per recipient status using no prophylaxis
|Wachs, 1995||6|| ||0/1 (0%)|| ||3/5 (60%)||(9–52)||3, 3, 24|
|Roche, 1997||7|| ||7/7 (100%)||36 (14–57)||2, 4, 4, 5, 8, 9, 12|
|Dodson, 1997||47||0/7 (0%)|| ||2/15 (13%)||18/25 (72%)||(6–84)||Not given|
|Dickson, 1997||20||1/2 (50%)||0/1 (0%)||0/1 (0%)||14/16 (87.5%)||29 (1–62)||Median 12 (2–37)|
|Douglas, 1997||9||0/1 (0%)|| ||3/8 (37%)||63 (<1–124)||6, 6, 24|
|Uemoto, 1998||16||1/1 (100%)|| ||14/15 (93%)||42 (16–76)||Mean 12 (5–26)|
|Prieto, 2001||30||0/2 (0%)||0/2 (0%)||0/3 (0%)||15/23 (65%)||29 (0.2–58)||Median 12 (3–24)|
|Roque-Afonso, 2002||6||0/2 (0%)|| ||4/4 (100%)||36 (32–45)||8, 9, 11, 15|
|Manzarbeitia, 2002||27||1/1 (100%)||0/13 (0%)||2/11 (18%)||2/2 (100%)||25.6 (3–65)||6, 9, 17, 18, 24|
|Chen, 2002||8||1/4 (25%)||0/1 (0%)||0/1 (0%)||2/2 (100%)||63 (45–86)||14, 35, 39|
|Nery, 2003||18||0/2 (0%)||0/16(0%)|| ||28 (1–75)||Not given|
|Overall||194||4/22(18%)||0/34 (0%)||4/31 (13%)||82/107 (77%)||(<1–124)|| |
These data suggest that the presence of an isolated anti-HBs in recipients of anti-HBc donor livers decreases but does not eliminate the risk of post-LT HBV. The lack of universal protection could be due to the presence of a HBsAg escape mutation, the loss of anti-HBs or lack of complete immunologic protection from humoral immunity alone. Mutations in the antigenic ‘a’ determinant of the HBsAg can lead to the loss of protection from anti-HBs. This mutant could be preexisting in the donor graft or potentially induced by HBIg prophylaxis or lamivudine therapy as has been previously described (16,17). The loss of protective anti-HBs serologies after known response to vaccination has been reported to range from 27% to 34% in studies with 3 years of follow-up (18–21). In addition, loss of anti-HBs is not uncommon following LT, even in patients with previous HBV exposure. One study reported the loss of anti-HBs post LT in 31% patients with pretransplant anti-HBs and anti-HBc (14). Our patient experienced loss of anti-HBs after LT despite short-term HBIg treatment and the presence of pre-LT anti-HBs. There was no attempt to revaccinate the patient after LT. Whether prolonged anti-HBs either through indefinite HBIg therapy or successful vaccination would have prevented post-LT HBV is unknown, but retrospectively may have been of benefit in our patient.
The presence of an isolated anti-HBc in recipients of anti-HBc donor liver also is only partially protective in preventing post-LT HBV infection. The protective effect in the recipient may be due to anti-HBc being a surrogate marker of a cytotoxic T-cell reaction to the HBV. A specific cytotoxic T-cell response has been shown to be important in controlling acute and chronic HBV infection (22–25). It is unclear why all patients would not be protected. One possibility is that it may be necessary to have a good HLA match between donor and recipient as donor antigen presenting hepatocytes must find their recipient lymphocyte counterpart, as immune reaction against intracellular viruses such as CMV, EBV and HBV is HLA restricted (26–28).
The presence of both anti-HBs and anti-HBc in the recipient appears to provide almost complete protection from post-LT HBV infection from an anti-HBc donor liver. This perhaps represents protection from both the cellular and humoral components of the immune system.
Some authors have proposed the use of donor HBV DNA as a determinant of risk for development of post-LT HBV infection. Several studies have looked at this issue in small numbers of patients, primarily using retrospective analyses of HBV DNA in stored serum and liver tissue. Collectively, information from these reports strongly suggests that the absence of serum HBV DNA in anti-HBc donors does not preclude transmission of HBV to liver recipients (1,4,10,11,29).
HBV DNA in the pre-LT donor liver biopsy may be equally unreliable. Presence of HBV DNA has been found in 50% of anti-HBc donors (24 of 48 donors) (5,10,30,31). However, in Loss et al., 1 of 6 patient with a donor biopsy negative for DNA pre-LT was demonstrated to have HBV DNA in the liver biopsy 2 months after LT. In addition, a patient from this series with undetected liver HBV DNA at 1 month post LT developed post-LT HBV infection at 20 months after discontinuation of lamivudine. The patient remained HBsAg (+) despite subsequent treatment with lamivudine and HBIg (30). Nery et al. describe a patient with the absence of HBV DNA in the donor liver biopsy that developed post-LT HBV infection at 14 months after discontinuing lamivudine. Again emphasizing the limitations of using this approach to determining prophylaxis (12).
Newer techniques to detect covalently closed circular DNA, pregenomic RNA or other episomal forms of HBV may be able to better determine the risk of HBV transmission to recipients (32). However, these approaches would likely be more valuable in the posttransplant management of patients than in donor recipient matching due to the inherent delay of such techniques and the time limitations associated with organ procurement (33). Other authors have suggested that Child Pugh score could be used to determine the risk due to decreased levels of innate immunity in sicker individuals. While, this may give some guidance in determining risk, it is not adequate to make decisions regarding prophylaxis strategies (11).
Based on this review of the data, we consider recipients of grafts from anti-HBc donors to have the following risk groups based on recipient serology alone. High risk (70–80%): HBV naïve recipients; Moderate risk (10–20%): recipients with either isolated anti-HBs or anti-HBc; low risk (0–5%): recipients with anti-HBs and anti-HBc. It is important to note that these studies are limited by their retrospective nature and lack of consistent long-term follow-up.
Currently, there is no standardized prophylaxis regimen for recipients of anti-HBc (+) grafts. In a survey of 28 U.S. transplant centers, 33% use HBIg week/month and lamivudine indefinitely, 22% lamivudine alone indefinitely, 19% HBIg and lamivudine indefinitely, 15% HBIg at transplant and lamivudine, and 11% HBIg alone (34). Reviewing the results of the different published regiments, de novo hepatitis was reported only in the studies using HBIg monotherapy (3 of 17). All cases occurred in naïve recipients, and two-thirds occurred after discontinuation of HBIg. This correlates well with the level of recurrence seen in recipients with isolated anti-HBs (Tables 2–4) (5–7,10,12,30,31,35–37).
Table 2. Cases de novo HBV per recipient status using lamivudine monotherapy
|Yu, 2001||6||0/1 (0%)||0/1 (0%)||0/2 (0%)||0/5 (0%)||100 or 150 mg daily|| ||21 (2–36)|
|Chen, 2002||16||0/13 (0%)||0/2 (0%)||0/1 (0%)|| ||3–4 mg/kg/day (children) or 100 mg daily (adolescent)||Yes||22 (14–40)|
|Nery, 20031||14|| ||0/81 (0%)||0/51 (0%)||150 mg daily|| ||19 (1–69)|
Table 3. Cases de novo HBV per recipient status using HBIg monotherapy
|Uemoto, 1998||3|| ||0/3 (0%)||100 IU/kg daily ×7, then 1000 IU prn to keep anti-HBs >100 IU/L||18 (13–24)|
|Dodson, 1999||1|| ||1/1 (100%)||10 000 IU daily ×7, then monthly ×6||11 months months|
|Roque-Afonso, 2002||12||0/3 (0%)|| ||0/4 (0%)||1/51 (20%)||5000 IU daily ×7, then keep anti-HBs >100 IU/mL||18 (6–36)|
|Manzarbeiti, 2002|| ||10 000 IU monthly ×6|| |
|Total||17||0/3|| ||0/4||3/10|| ||(11–63)|
Table 4. Cases de novo HBV per recipient status using combination HBIg and lamivudine
|Dodson, 1999||13|| ||0/8 (0%)||0/5 (0%)||150 mg daily||10 000 IU daily ×7, monthly ×6, 1000 IU IM every 2 weeks ×18 months|| ||16 (6–25)|
|Holt, 2002||12||0/5 (0%)|| ||0/1 (0%)||0/6 (0%)||150 mg bid||10 000 IU daily ×7||Yes||29 (2–38)|
|Fabrega, 2003||7||0/1 (0%)|| ||0/6 (0%)||100 mg daily||10 000 IU daily ×7 Continue if donor is HBV DNA + in liver or serum|| ||21 (9–36)|
|Loss, 20031||13||0/2 (0%)|| ||0/11 (0%)||150 mg daily||10 000 IU ×1 anhepatic||Yes||33 (22–51)|
|Nery, 2003||11|| ||0/6 (0%)||0/3 (0%)||0/2 (0%)||150 mg daily||10 000 IU daily ×7, weekly ×4, monthly ×6|| ||22 (8–56)|
The success of lamivudine while promising must be viewed with caution. Yu et al. demonstrated no recurrence in nine recipients of anti-HBc donors receiving lamivudine prophylaxis. However, 4 out of 9 patients died within the first 6 months, only 2 out of 9 were alive at 24 months, and no patient survived beyond 36 months. Thus, no patient survived into the time period where de novo post-LT HBV developed in our patient (35). Chen et al. had a mean follow-up of 25 months (minimum 14 months) in 16 Chinese patients. However, all recipients had at least 1 anti-HBV anti body including anti-HBs only in 13 patients, anti-HBc only in 1 patient and both anti-HBs and anti-HBc in 2 patients making all patients in this group at lower risk for recurrence (7). Nery et al. described no recurrence in 13 patients although 2 patients who developed de novo post-LT HBV infection were excluded due to ‘noncompliance.’ One of these patients had persistence of serum HBV DNA despite reinitiating lamivudine suggesting the possibility of HBV mutation (12). Lamivudine prophylaxis may be further limited by long-term viral persistence requiring long-term therapy and lamivudine resistance has been shown to increase with time (38).
The success of combination therapy is not surprising given the almost complete lack of recurrence using this regimen for patients undergoing LT for underlying HBV liver disease (39). Patients undergoing LT for HBV have much higher levels of baseline HBV viral load. The results from our case report would suggest that if this strategy is to be undertaken, anti-HBs should be maintained long term as our patient developed post-LT HBV despite the presence of anti-HBs for 6 months. However, the cost and inconvenience to maintain anti-HBs make this a less-than-ideal approach in the long term. Pre-LT vaccination to provide endogenous anti-HBs is unlikely to be successful in the vast majority of patients. A recent North American study demonstrated only 37% response rate to double-dose hepatitis B vaccine in cirrhotics without history of hepatitis B infection (18). This contradicts data from some Asian studies, though the higher seroconversion rates in Asia may correlate with a higher underlying prevalence of HBV exposure (40).
Lamivudine monotherapy should be adequate for those patients in our lower risk group. The higher risk patients may require a combination of anti-HBs (either through vaccination or HBIg), while some of the new antivirals with lower resistance profiles such as adefovir, entecavir or tenofovir may need to be considered in those with intermediate risk or as monotherapy in higher risk patients.
The ramifications of donor-transmitted HBV infection after LT at this point are unclear. Loss of HBsAg once de novo HBV develops is thought to be rare (1,11). Several authors have reported mild histological disease following initial HBV infection, even in the absence of antiviral therapy. Douglas described 1 of 5 with significant graft dysfunction during an average follow-up of 7 years with four deaths unrelated to HBV infection (4). Dickson et al. also reported mild initial histologic recurrence with 85% of biopsies either normal or with only mild portal and lobular inflammation at 1 year (11 of 13). However, follow-up biopsies in this series revealed significant histologic recurrence in 5 out of 8 biopsies from 1.6 to 4.5 years after LT and an overall decreased 4-year survival in recipients of grafts from anti-HBc donors (1). Other authors in addition have reported more severe disease including Uemoto et al. who described 15 cases with one death from fibrosis cholestatic hepatitis and two with chronic active hepatitis (3,10,41). The effect of new antiviral therapies and antiviral-resistant HBV infections on the course of de novo post-LT HBV is yet to be determined. In addition, the risk of long-term serum HBsAg following de novo infection and the risk of hepatocellular carcinoma is not known. In this case recurrent HBV did not lead to graft failure and enzymes normalized with treatment. However, viral load remained greater than 105 copies after 27 months of adefovir therapy and did not fall below 104 until almost 3 years of treatment. HBsAg remained in the serum.
In summary, our case report demonstrates that there is persistent risk of HBV from an anti-HBc donor even more than 3 years following LT. It also highlights the need for long-term follow-up prior to determining the efficacy of a prophylactic strategy and the need to determine long-term prophylaxis strategy. Our review of the literature suggests that the recipient's status for HBV markers should be used to assess risk for de novo hepatitis and these risk groups can be used to design prophylactic strategies. High-risk and intermediate-risk patients may require a long-term combination therapy to prevent de novo HBV, but the actual risk of monotherapy or shorter-term combination prophylaxis can only be determined by longer term follow-up. Patients in the lowest risk group could be considered for monotherapy with lamivudine either with or without perioperative HBIg.
Dr. Bonatti received financial support from the Detiger Fellowship.