Hepatitis B after transplantation: Competition between the recipient virus and the donor virus


  • David Mutimer

    Corresponding author
    1. Liver and Hepatobiliary Unit, Queen Elizabeth Hospital, Edgbaston, Birmingham, United Kingdom
    • Liver and Hepatobiliary Unit, Queen Elizabeth Hospital, Nuffield House, 3rd Floor, Edgbaston, Birmingham, United Kingdom B15 2TH
    Search for more papers by this author
    • Telephone: 011-44-121-627-2337; FAX: 011-44-121-627-2412

  • See Article on Page 1314

Hepatitis B virus (HBV) infection of the liver graft can be a result of reinfection (after transplantation for HBV-related liver disease) or the acquisition of HBV from a liver donor with a past resolved HBV infection (so-called de novo infection), and occasionally transmission involves a third party (eg, sexual transmission post-transplant). A clinically important infection of recipient origin, donor origin, or both is effectively prevented by the use of appropriate antiviral prophylaxis before and/or after transplantation. For the prevention of recurrent infection, the combination of hepatitis B immunoglobulin (HBIG) and nucleos(t)ide analogues (NAs) is almost always effective.1 For the prevention of de novo infection, lamivudine prophylaxis is effective and usually adequate.2 However, despite the application of protocols that maintain posttransplant serum hepatitis B surface antigen (HBsAg) negativity and prevent clinically relevant graft damage, sensitive polymerase chain reaction (PCR) assays have been able to detect HBV DNA sequences in the serum and/or liver in a high proportion of patients.3–6 These studies examined posttransplant specimens from patients who underwent transplantation for HBV-related liver disease. A single study also included donor HBV details (Hussain et al.4 noted that 3 of 25 donors were HBV core antibody–positive), so detectable HBV DNA in these studies was presumed to be of recipient (ie, recurrent) origin. That assumption was probably correct in most cases because the frequency of core antibody positivity in donors in the countries in which these studies were undertaken (the United States and Western Europe) is relatively low. In contrast, Chinese donors have a much higher likelihood of past HBV infection. Therefore, Chinese transplant centers have an obvious opportunity to examine posttransplant specimens for HBV sequences of both recipient and donor origin.


cccDNA, covalently closed circular DNA; HBIG, hepatitis B immunoglobulin; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; NA, nucleos(t)ide analogue; PCR, polymerase chain reaction.

Cheung and colleagues7 looked for HBV DNA, including HBV covalently closed circular DNA (cccDNA), in posttransplant specimens collected from 31 patients who underwent transplantation for HBsAg-positive liver disease. All patients were serum HBsAg–negative at the time of sampling. Nineteen of the 31 liver donors were core antibody–positive, and this reflected past resolved infection. The prophylaxis involved NAs, lamivudine prophylaxis for the majority, and none received HBIG. All posttransplant specimens were sampled more than 1 year after transplantation, and some were sampled as long as 10 years after transplantation. Sensitive PCR assays were employed to examine both serum and tissues. In addition to the detection of total HBV DNA (ie, any HBV DNA sequence), PCR primers and conditions were chosen specifically to detect and amplify cccDNA if it was present.

Serum and blood leukocytes were HBV DNA–negative, but HBV DNA was detected in liver biopsy specimens from the majority of the patients. HBV DNA was detected in 26 of 31 liver biopsy specimens, and cccDNA was detected in 16 of 31. Pretransplant recipient characteristics did not predict the presence or absence of HBV DNA in posttransplant biopsy samples. However, in those with detectable DNA (ie, the majority examined), there was a weak correlation between pretransplant serum HBV DNA and the amount of DNA detected in posttransplant biopsy samples. HBV DNA, including cccDNA, was more likely to be detected in the posttransplant liver when the donor was core antibody–positive. Indeed, cccDNA was detected in the posttransplant liver biopsy samples of 12 of 19 recipients when the donor was core antibody–positive and in only 4 of 12 biopsy samples when the donor was antibody-negative. This observation implies that a donor HBV infection might be a relevant determinant of persisting HBV DNA in the recipient. Therefore, viral genetic sequencing of donor and recipient specimens was undertaken for those patients for whom informative specimens were available; they included 8 patients with core antibody–positive donations. A phylogenetic comparison indicated that the detectable DNA in the posttransplant biopsy samples was a mixture of donor and recipient HBV DNA in 4 patients, only recipient HBV DNA in 2 patients, and only donor HBV DNA in 2 patients.

Overall, the findings of this study indicate long-term occult HBV infection despite successful NA prophylaxis and convincingly show that the donor strain, recipient strain, or both may persist.

This study extends investigations undertaken by others who have looked for evidence of persistent (presumed recurrent) occult HBV infection in HBV transplant patients. In addition to looking for evidence of HBV infection, the Hong Kong cohort and study design addressed the possibility that an HBV infection might represent a recurrent recipient HBV infection, a de novo donor HBV infection, or both. Observations made by these investigators convincingly show that recipient species, donor species, or both may persist despite long-term NA prophylaxis. The resilience of donor-derived infections has been clearly demonstrated by the numerous examples of de novo infection observed in liver recipients who were naturally HBV-immune and were receiving NA prophylaxis. Observations made in this study support the argument for sustained and indefinite-duration prophylaxis for recipients of core antibody–positive donor livers.

Unfortunately, the context of the study cohort is not explained in the article. It appears that routine prophylaxis for HBsAg-positive recipients at this center comprised NAs without the use of HBIG. Published studies, including a report from the University of Hong Kong,8 have shown a high failure rate for lamivudine monoprophylaxis, with HBsAg recurrence and significant graft HBV infection experienced during posttransplant follow-up. The failure of lamivudine monoprophylaxis (given to the majority of the patients in this study) can be associated with recurrent or persistent serum HBsAg positivity, and the emergent species is typically lamivudine-resistant (YMDD mutant).8–10 It would have been interesting, if informative donor/recipient pairs and specimens had been available, to examine the HBV species in the serum and livers of patients who experienced overt HBV recurrence despite prophylaxis. This might impart additional clinical relevance to the observations made for the reported cohort with covert infection. For instance, the HBV genotype might determine the long-term outcome of NA treatment given to those with overt HBV recurrence. Recent reports of the nonimmunosuppressed population indicate that HBsAg seroconversion may be observed during long-term NA treatment, but the likelihood of seroconversion appears to be HBV genotype–dependent.11 One could speculate that for any given donor/recipient genotype discordance, it might be preferable for a patient to have a de novo infection (eg, genotype A) instead of experiencing HBV recurrence (genotype B or C). This may have greater relevance for clinical practice in Western transplant centers, in which favorable discordance (A versus B/C) is more likely to be observed.

Relevant observations concerning the competition between donor and recipient viral strains have also been made in the context of liver transplantation for hepatitis C virus infection.12–14 Although the total number of published informative donor/recipient genotype pairs is relatively low, it appears that hepatitis C virus genotype 1 may be more resilient than other genotypes and may emerge as the long-term dominant species when it is competing with the more interferon-sensitive genotypes. Under this circumstance, the outcome of the battle for survival between competing genotypes may have significant long-term clinical consequences.

Organ donor cytomegalovirus may also compete with recipient cytomegalovirus and may be responsible for a clinically significant infection.15 The concept of competing donor and recipient viral species in the solid organ transplant recipient is not new!

The observations made in this study differ in detail from those made by other investigators. For instance, Cheung et al.7 were unable to detect circulating HBV DNA, but others have found serum HBV DNA in a high proportion of patients.3, 4 Cheung et al. detected liver DNA in the majority (including those with core antibody–negative donors), but Lenci et al.6 detected liver DNA in only a few. This discordance probably reflects differences in specimen collection, handling, and storage, differences in assay conditions and sensitivities, and differences in antiviral prophylaxis. Nevertheless, the consistent message from all studies is that HBV DNA persists in the recipient for many years despite effective antiviral prophylaxis. Cheung et al. show us that the donor HBV strain may compete with the recipient HBV strain and become dominant during antiviral prophylaxis. The clinical implications of this observation are uncertain and need to be determined by further studies.