To the Editor:
It has been reported a high incidence of de novo hepatitis B virus (HBV) infection after liver transplantation (LT) from hepatitis B surface antigen (HBsAg) negative but anti-core antibody (anti-HBc) positive donors (1). The frequency of the HBV transmission depends on the HBV serological status of the recipient: the presence of anti-HBc and anti-HBs (past HBV infection) or anti-HBs levels > 10 IU/mL (HBV vaccine prior to LT) in organ recipients usually confers resistance to HBV infection while their absence results in de novo HBV infection (2). Then, liver grafts from anti-HBc positive donors should be directed to patients with anti-HBs as a result of past HBV infection or HBV vaccine prior LT.
It is known that the “a” determinant, located within the HBV HBsAg, is the target of the immune response providing immunity against HBV infection (3). A change of amino acids in the “a” determinant could lead to a structural variation in the epitope of the surface antigen recognized by the anti-HBs with a loss of immunoreactivity and then, affecting virus biding and entry into the cell (4). HBV envelope mutants associated with the “a” determinant after HBV vaccination have been identified (5). In patients with a compromised immune system, such as transplant recipients, HBV variants with mutations in the surface gene could proliferate and, subsequently, these patients could be at high risk of de novo HBV infection (6). We described a liver transplant recipient who developed a de novo HBV infection despite she was vaccinated and had protective serum anti-HBs titers (>10 IU/mL) before LT.
A 50-year-old woman underwent LT for end-stage cryptogenic cirrhosis receiving a liver from an anti-HBc and anti-HBs positive donor. The serological HBV profile of the liver recipient prior LT (anti-HBc negative, anti-HBe negative, HBsAg negative, HBeAg negative, anti-HBs negative) indicated that the patient did not have any HBV exposure. She received recombinant HBV vaccine in the pre-LT period (40 μg/dose intramuscularly administered at 0, 15 and 30 days starting 10 months before LT; Engerix B, SmithKline Beechman, Belgium) and developed anti-HBs levels >10 IU. Post-transplant immunosuppression regimen consisted of tacrolimus and steroids. At 6 days after LT, the patient had an episode of acute graft rejection, as confirmed by a liver biopsy, so three steroid pulses (1 g/pulse) were administered. The patient was discharged 18 days after LT without any other complication. Recipient was routinely screened for anti-HBs and serum HBV-DNA during the post-LT period (Table 1). Seven months post LT, liver ultrasound was normal, but her routine biochemical tests revealed high level of liver enzymes (AST 97 U/L, ALT 151 U/L, GGT 136 U/L) (Table 1). In addition, virological analysis showed that HBsAg and serum HBV-DNA were positive, even if her serum anti-HBs titer was determined to be 17 IU/L (Table 1). Specific virological tests did not reveal any other viral infections (CMV, CEV, HCV, HAV). Based on these results, a de novo HBV infection occurred in this patient despite the presence of anti-HBs at a protective level. Consequently, steroid therapy was suspended although tacrolimus level was still maintained. Treatment with lamivudine was started (100 mg/day), but serum HBV-DNA still remained positive, so a combination therapy of lamivudine and adefovir was administered at 32 months after LT. At this time, a second liver biopsy revealed marked steatosis with positive immunostaining for HBcAg (nucleus and cytoplasm) and HBsAg (cytoplasm and membrane). Currently, 40 months postLT, the patient still maintains serum HBV-DNA, HBsAg and HBeAg positive (Table 1).
|Months after LT|
|HBV-DNA (copies/mL)||–||–||+||+||>2 × 105||>2 × 105||1.1 × 106||>2 × 105||7.2 × 104||1.5 × 105|
The post-LT follow-up of this patient clearly indicated that a de novo HBV infection occurred, even if protective anti-HBs levels were present when serum HBV-DNA became positive. It has been described the emergence of HBV with surface gene mutations, which are able to escape immune response against HBV vaccine and cause infection (5). Based on this fact and in the post-LT follow-up of the recipient, we thought about the possibility that the patient could be infected with one of these HBV mutants. To prove this hypothesis, a serum sample taken 36 months after LT was used for HBV-DNA extraction, PCR amplification of the 'a' determinant of the S gene and sequencing. Viral sequence revealed that the patient harbor an HBV variant with 2-point mutations at amino acid positions 127 and 145 of HBsAg (Figure 1). The first mutation resulted in a substitution of either proline or lysine for threonine (Pro or Lys 127 Thr); the second mutation was a substitution of glycine for alanine (Gly 145 Ala). Several HBV mutants with amino acid changes in the “a” determinant have been reported in the post-transplant situation (6). Among these mutations, the particular and almost invariably change is Gly 145 Arg, which has been described to cause persistent infections (7). Our patient also presented this change at position 145; however, the sequence analysis revealed the presence of an Ala instead of an Arg. This amino acid change has never been previously found in the literature. The change Pro 127 Thr has already been described in liver transplant patients (8).
Our patient has been treated with lamivudine followed by therapy with lamivudine in combination with adefovir. Treatment with lamivudine is initially efficient in HBV infections but it induces, over time, mutations in the conserved catalytic YMDD domain of the polymerase gene. Then, adefovir therapy appears to be very promising in patients with HBV infection and lamivudine-resistant HBV. Therefore, the reverse transcriptase domain of the polymerase was also sequenced in order to detect both lamivudine and adefovir mutations. Sequence analysis showed that no changes were presented in this domain. Because of the frame shift between HBV envelope and polymerase genes, HBsAg mutations can also generate mutations in the HBV polymerase. However, the amino acid change Gly 145 Ala in the surface gene did not cause an amino acid exchange in HBV polymerase, so polymerase activity was not affected. We do not have a clear answer to the mechanism underlying the lack of response to therapy with lamivudine and adefovir, even in the absence of drug resistant mutants. It is known that pretreatment ALT levels are the best predictor of treatment response to both, lamivudine and adefovir (9,10). Patients with pretreatment ALT levels below two times the upper normal limit is unlikely to respond. In our patient, only a moderate rise in ALT level occurred, which remained unchanged during the follow-up period (Table 1). Alternatively, another explanation could result from an altered host metabolism; possible polymorphisms in cellular enzymes could affect the metabolic steps needed for the proper function of the drug into the HBV-infected cells, leading to the lack of response to therapy. However, more extensive molecular studies will be needed in order to find an answer.
It is known that liver failure decreases vaccine response rates as well as that HBV vaccine is poorly effective in immunocompromized patients. The seroconversion rate is low after HBV vaccination of liver transplant patients with non-HBV-related cirrhosis. It has been suggested that in vaccinated anti-HBs-positive recipients, hepatitis B immune globulin (HBIG) administration or revaccination after LT should be necessary in order to prevent de novo HBV infection by maintaining anti-HBs levels >100 IU/mL (11,12). In addition, it is also known that these patients are at higher risk of de novo HBV infection compared with patients having humoral and cellular immunity against HBV as a consequence of childhood HBV vaccination. In our patient both situations occurred: she received HBV vaccine once her liver failure was established and, consequently, her response to HBV vaccine was low but still, reaching protective serum anti-HBs titers. Furthermore, she did not receive any post-LT prophylaxis trying to avoid de novo HBV infection.
In conclusion, we described a new vaccine HBV mutant in a liver transplant recipient causing a de novo HBV infection, even if the patient still had protective anti-HBs levels. Although the efficacy of HBV vaccine in organ recipients has been proved in the literature, hereafter this fact could not be considered as universal due to the development of immune escape HBV mutants associated with the “a” determinant, which can evade the anti-HBs protection.