Diagnosis, prevention and management of hepatitis B virus reactivation during anticancer therapy

Authors

  • Winnie Yeo,

    Corresponding author
    1. Department of Clinical Oncology, Sir Y. K. Pao Centre for Cancer, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong
    • Department of Clinical Oncology, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong
    Search for more papers by this author
    • fax: (852) 2649-7426

  • Philip J. Johnson

    1. Cancer Research UK Institute for Cancer Studies, University of Birmingham, Birmingham, United Kingdom
    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

It is estimated conservatively that more than one third of the world's population has been infected with the hepatitis B virus (HBV) and that there are 350 million people with chronic infection, 75% of whom live in Southeast Asia and the Western Pacific regions.1–3 Reactivation of HBV infection is now a well-recognized complication in infected patients who undergo cytotoxic chemotherapy for cancer. The condition ranges from asymptomatic self-limiting anicteric hepatitis to severe, potentially fatal progressive decompensated hepatitis. With the increasing use of potent cytotoxic chemotherapy, reactivation of hepatitis B in endemic regions is becoming a common clinical problem. This adversely affects advances made in various forms of cancer therapy. In this review, we consider the diagnosis, prevention, and management of HBV reactivation in association with chemotherapy and hematopoietic stem cell transplantation, particularly in light of the availability of effective prophylactic antiviral therapy.

Reactivation of HBV was first described by Wands et al.,4 who in 1975 reported the condition in 20 patients with lympho- and myeloproliferative disorders. Most of the cases reported since were similar in patients with hematological malignancies—in particular, lymphomas.4–16 The skewed observations may be due to the fact that these patients are often subjected to intense myelosuppressive treatment regimens, the malignancies per se are often associated with an immunocompromised state, and the patients have been consistently reported to have a higher rate of hepatitis B surface antigen (HBsAg) seropositivity than the normal population.9, 17–19

Over the past decade, HBV reactivation has been increasingly observed in patients with solid tumors.18–22 In patients with hepatocellular carcinoma (85% of whom have chronic HBV infection in Southern China3, 23), hepatitis following systemic chemotherapy has been reported in 60%20, 24; this is mostly attributed to HBV reactivation, which has a 30% mortality rate.20 In other cancer subpopulations, the incidence of HBV reactivation ranges between 25% and 40%.10, 15, 18, 21, 25

In the setting of hematopoietic stem cell transplantation (HSCT) for various hematological and oncological conditions, HBV reactivation has been reported in over 50% of patients.26–28 This is related to the intense chemotherapy with or without total body irradiation, and the coexistence of acute graft-versus-host disease.28

Although the frequency of viral reactivation among HBsAg-positive patients with cancer would be expected to be the same irrespective of geographical area, the prevalence of HBV infection varies between different populations from 10% to 25% in endemic areas to less than 1% in others.3, 10, 29 As a result, there would be proportional difference in the incidence of viral reactivation in a given population.

Abbreviations

HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; HSCT, hematopoietic stem cell transplantation; anti-HBs, hepatitis B surface antibody; ALT, alanine aminotransferase; anti-HBc, hepatitis B core antibody; HBeAg, hepatitis B e antigen.

Definitions and Diagnosis

In early reports, the diagnosis of HBV reactivation was based on HBsAg and hepatitis B surface antibody (anti-HBs) antibody titers. Wands et al.4 described 2 separate clinical scenarios during immunosuppressive therapy: (1) HBsAg-positive patients experiencing an increase in serum HBsAg titer and (2) HBsAg-negative/anti-HBs–positive patients showing anti-HBs decline associated with the reappearance of HBsAg (seroreversion).

With the availability of quantitative HBV DNA assays, HBV reactivation could be tracked by the temporal relationship of the rise in HBV DNA titers with hepatitis and chemotherapy administration.10, 18 Other definitions have been used, such as “the occurrence of hepatitis during or immediately after cytotoxic chemotherapy, accompanied either by an increase in HBV-DNA levels of ≥10-fold, or an absolute increase that exceeds 9log10copies/mL, in the absence of other systemic infections.”10, 18 Thus, other than close monitoring before, during, and after chemotherapy, the sensitivity of the HBV DNA assays is crucial. The earlier branched DNA hybridization assay (Quantiplex HBV DNA assay; Chiron, Berkeley, CA) has a detection limit of 0.7 × 106 copies/mL, whereas recent polymerase chain reaction assays have a lower detection limit of 300 copies/mL.30

In earlier studies, alanine aminotransferase (ALT) has been the mainstay modality of monitoring.18 Whereas a proportion of patients were detected to have a simultaneous rise in ALT and viral DNA, others had undetectable HBV DNA by the time ALT rise was noted. Serial ALT and HBV DNA monitoring for individuals during chemotherapy showed that the rise in HBV DNA preceded that of ALT by a median of 2 to 3 weeks (Fig. 1), so that by the time hepatitis became evident, the HBV DNA level may have decreased to an undetectable level. Close monitoring using ALT and HBV DNA will enable early diagnosis and appropriate management of HBV reactivation.21

Figure 1.

HBV DNA and ALT profiles of a typical patient who had increased HBV DNA preceding a rise in ALT during chemotherapy. Triangles (▴) represent ALT activity measured in international units per milliliter. Circles (●) represent HBV DNA measured in (loge) genome equivalent per milliliter. Days are number of days from the first day of the first cycle of chemotherapy. ALT, alanine aminotransferase; HBV, hepatitis B virus.

Clinical Picture of HBV Reactivation

Reactivation may occur during or after completion of the full course of chemotherapy.17 Although some patients will recover from the condition spontaneously, anticancer treatment has to be interrupted due to hepatitis, and thus the patient's prognosis is jeopardized. In a study on breast cancer patients, over 70% of patients who developed HBV reactivation required premature termination of chemotherapy or disruption in treatment schedules compared with 30% in those without reactivation.21 The importance of chemotherapeutic dose intensity as a determinant of treatment response and outcome in patients with non-Hodgkin's lymphoma31, 32 and breast cancer33, 34 has been well documented.

Some patients experience icteric hepatitis with HBV reactivation,15 with fatality rates ranging between 5% and 40%.10, 15, 16, 20, 28 Patients with cirrhosis are more likely to develop hepatic decompensation. This in part explains the particularly poor clinical outcome of patients who have hepatocellular carcinoma with HBV reactivation, as the majority of them have coexisting cirrhosis.20, 35 Hepatic failure in patients who have cirrhosis may be complicated by sepsis, which occurs more commonly during immunosuppression. Patients with cirrhosis require more stringent monitoring. Diagnosis of cirrhosis is ideally determined via histology; however, clinical diagnosis with radiological assessment of the liver and evidence of thrombocytopenia, hypoalbuminemia, or coagulopathy may suffice when histology is not available.36 Recent advances using noninvasive Fibrotest, Fibroscan, or constellation of biochemical markers may be useful additions to evaluation systems for assessing hepatic fibrosis and cirrhosis.

Reactivation in HBsAg-Negative and Anti-HBs–Positive and/or Hepatitis B Core Antibody–Positive Patients

In individuals with resolved HBV infection (i.e., HBsAg-seronegative, anti-HBs–postive and/or hepatitis B core antibody [anti-HBc]–positive), HBV replication has been shown to persist in the liver and in peripheral blood mononuclear cells.37, 38 Two reports described patients receiving conventional dose chemotherapy. In 1 report,4 12 of 17 patients with hematological malignancies had dramatic reduction of anti-HBs titer, and 5 (30%) had evidence of seroreversion with reappearance of HBsAg—2 of whom had persistent HBsAg despite cessation of chemotherapy. The second report described 51 Chinese patients with lymphoma—9 of whom developed hepatitis, with 2 having raised HBV DNA during chemotherapy, resulting in a 4% HBV reactivation rate.10 In both reports, hepatic impairment had been mild and no direct HBV-related mortality was observed.

For anti-HBs–positive hematological patients undergoing allogeneic HSCT, isolated cases of HBV reactivation have been described.39–50 Based on reported series,47, 51–54 the frequency of seroreversion ranges between 14% and 50% (Table 1A). The wide range reported could be explained by risk estimation based on small series, missed cases of seroreversion due to absence of symptoms, and variable duration of posttransplant follow-up. Seroreversion occurs relatively late after transplant (6-52 mo; median 19 mo) compared with that in HBsAg-positive patients (median 2-3 mo).18 This is probably caused by prolonged existence of recipient-type memory B cell immunity in the former. HBV reactivation in this setting has been considered to result from immunosuppression following decline in recipient-derived immunoglobulin G over 1-2 years after allo-HSCT. Seroreversion hepatitis is thought to be caused by naïve donor immunity activity against reactivated HBV replication.53 Although no HBV-related mortality had been observed in these series, 2 isolated cases of fulminant hepatic failure have been reported after allo-HSCT among a total of 39 patients in the literature.48–49 The suggested factors for reversion hepatitis include the use of corticosteroids,50, 54 lack of anti-HBs in donor,50, 51, 54 and graft-versus-host disease.52–53

Table 1A. HBV Reactivation in HBsAg-Negative Patients in the Absence of Prophylactic Antiviral Therapy
ReferenceNo. of Patients StudiedDiseaseAnticancer TherapyPretreatment HBV MarkersEvidence of HBV ReactivationNo. of Patients With HepatitisPeak ALTTreatment for HBV-Related HepatitisOutcome of HBV-Related Hepatitis
  1. Abbreviations: AML, acute myeloid leukemia; ALL, acute lymphoid leukemia; CT, chemotherapy; RS, seroreversion; NHL, non-Hodgkin's lymphoma; CML, chronic myeloid leukemia; SAA, severe aplastic anemia; HD, Hodgkin's disease; MM, multiple myeloma; MDS, myelodysplastic syndrome.

Wands et al.4179 AML 5 ALL 3 MMChemotherapy17 anti-HBs+5 RS52- to 5-fold of normal5 supportive3 resolved, 2 persistent RS
Lok et al.105151 NHLChemotherapy51 anti-HBs+, 33 of whom were anti-HBc+2 HBV DNA+2Not specified (but 1 developed hyperbilirubinemia and the other developed hepatic failure; both recovered)2 supportive9 resolved
Chen et al.4799 hematological disordersChemotherapy + HSCT9 anti-HBs+1 RS1Not specified1 supportive1 persistent RS
Dhedin et al.513713 AML 8 ALL 7 CML 3 NHL 2 SAA 4 othersChemotherapy + HSCT37 anti-HBs+ and anti-HBc+4 RS and HBV DNA+428- to 80-fold of normal4 supportive3 resolved, 1 persistent RS
Seth et al.524218 ALL 15 CML 4 MDS 3 SAA 2 NHLChemotherapy + HSCT42 anti-HBc+6 RS ± HBV DNA+6167-2,250 U/L1 supportive, 5 lamivudine5 resolved, 1 had refractory lymphoma and died of multiorgan failure
Onozawa et al.53145 CML 4 ALL 3 MDS 2 SAAChemotherapy + HSCT13 anti-HBs+ and anti-HBc+7 RS7196-1,460 U/L6 supportive, 1 lamivudine5 resolved, 2 persistent RS
Knöll et al.5461 ALL 1 AML 2 CML 1 NHL 1 HDChemotherapy + HSCT6 anti-HBs+ and anti-HBc+3 RS and HBV DNA+17-fold of normal2 supportive, 1 lamivudine1 resolved, 2 persistent RS

For patients receiving autologous HSCT, Lau et al.53 reported that only 1 of 37 anti-HBs–positive patients (23 of whom were also anti-HBc–positive) developed viral reactivation after transplant, a finding that was associated with nonfulminant hepatitis. There have been 5 other cases of seroreversion reported in the literature, 2 of which were fulminant.50, 56, 57

In addition to the above-mentioned findings, data based on 38 HBsAg-negative/anti-HBc–positive patients receiving solid organ transplantation revealed that while 44% had posttransplant hepatitis B viremia, only 5% had detectable HBsAg and none had clinical evidence of hepatitis.58

The recent introduction of therapeutic monoclonal antibodies against B and T lymphocytes such as rituximab (a chimeric mouse human monoclonal antibody against CD20+ malignant lymphoid cells) and alemtuzumab (a humanized monoclonal antibody against anti-CD52+ malignant lymphoid cells), used alone or in combination with cytotoxic therapy, has been associated with HBV reactivation.59–68 In October 2004, the U.S. Food and Drug Administration reported a possible relationship between fulminant hepatitis and rituximab use. These agents have been found to induce profound and durable B and T cell depletion.69–70 Although lysis of HBV-infected hepatocytes is mainly mediated by CD8+ cytotoxic T cell immunity, B cells may also act as antigen-presenting cells and prime cytotoxic T lymphocyte–specific responses in HBV infection.71 This is supported by reports of rituximab- and alemtuzumab-induced severe or fatal cytomegalovirus reactivation, parvovirus B19 infection, adenovirus infection, and pneumocystis carinii pneumonia.70, 72–75 The progressive B and T cell depletion may also account for the increasing incidence of HBV reactivation in anti-HBs– and/or anti-HBc–positive patients undergoing chemotherapy with these agents (Table 1B). Among the 7 cases reported, 3 developed fatal hepatic failure despite lamivudine therapy, possibly related to the delay in antiviral administration, and another had persistent HBsAg positivity.63–68

Table 1B. HBV Reactivation in HBsAg-Negative Patients in Association With Rituximab or Alemtuzumab in the Absence of Prophylactic Antiviral Therapy
ReferencesAge/SexDiseaseAnticancer TherapyOther Pretreatment HBV MarkersPeak ALT (U/L)Treatment of HBV ReactivationOutcome
  1. Abbreviations: DLC, diffuse large cell lymphoma; R, rituximab; O, vincristine; C, cyclophosphamide; Ara-C, high-dose arabinoside-C; VP16, etoposide; Dex, dexamethasone; CHOP, cyclophosphamide/doxorubicin/vincristine/prednisolone; IFN, interferon; S mutant, mutation detected in region L110R, R122K, Y/F134S, P142L, or D144A; NR, not reported; CLL, chronic lymphocytic leukemia; A, alemtuzumab; NHL, non-Hodgkin's lymphoma type not specified.

Tsutsumi et al.6380/MDLCR + OAnti-HBs+, anti-HBe+101SupportiveResolved
 55/MDLCR + C + Ara-C + VP16 + DexAnti-HBs+, anti-HBe+84SupportiveResolved
Dervite et al.6473/MFLCHOP + IFN + Ara-C + RAnti-HBs+1230SupportivePersistent HBsAg+, anti-HBs−
Westhoff et al.6573/MDLCRAnti-HBs+ (S mutant)NRLamivudineDied of hepatic failure
Iannitto et al.66NRCLLAAnti-HBs+932LamivudineResolved
Sarrecchia et al.6753/MCLLRAnti-HBs+, anti-HBc+2120LamivudineDied of hepatic failure
Law et al.6867/MNHLR + CHOPAnti-HBs+2240LamivudineDied of hepatic failure

Risk Factors: Which Patients Develop HBV Reactivation?

Several studies have provided data on risk factors associated with HBV reactivation (Table 2). Those that have undertaken multivariate, as opposed to univariate, analyses should bear a greater weight and provide more definitive evidence. Male sex,10, 18 younger age,10, 18 and pre-chemotherapy ALT20 have been reported to be associated factors.

Table 2. Incidence and Associated Risk Factors of HBV Reactivation in Cancer Patients Undergoing Chemotherapy
Patient PopulationNo. of HBsAg Carriers StudiedRate of HBV ReactivationStatistical AnalysisRisk Factors IdentifiedReference
  • Abbreviations: PCR, polymerase chain reaction; cccDNA, covalently closed circular DNA.

  • *

    Of 137 patients undergoing HSCT, 23 were HBsAg-positive.

  • Of the 83 patients who received transarterial chemo-lipidization, 21.6% developed HBV reactivation; 1.6% of the 63 patients who received other forms of non-immunosuppressive therapy (e.g., percutaneous ethanol injection, surgery, or supportive management) developed viral reactivation.

Lymphoma2748%UnivariateMale; HBeAg seropositivityLok et al.10
Lymphoma4538%UnivariateSecond- or third-line chemotherapy (as compared to first line)Kumagai et al.16
Various malignancies7819%UnivariateMale; younger age; HBeAg seropositivity; lymphoma; use of anthracyclinesYeo et al.18
Lymphoma23*45%MultivariateDetectable pre-chemotherapy HBV DNA using Digene II assay (cutoff limit 0.142 × 106 copies/mL)Lau et al.55
Breast cancer4141%UnivariateHigh pre-chemotherapy HBV DNA using real-time PCR assay (optimal cutoff 3 × 105 copies/mL)Zhong et al.82
Various malignancies13826%MultivariateLymphoma or breast cancer; use of anthracyclines and/or steroids; detectable pre-chemotherapy HBV DNA using real-time PCR assay (cutoff limit 2.9 × 103 copies/mL)Yeo90
Hepatocellular carcinoma3324%MultivariateHBeAg seropositivityNagamatsu et al.24
Hepatocellular carcinoma1461.6%-21.7%MultivariateHBeAg seropositivity; use of chemotherapy (as compared to non-immunosuppressive therapy)Jang et al.76
Hepatocellular carcinoma10236%MultivariatePre-chemotherapy ALTYeo et al.20
Lymphoma2241%MultivariatePre-chemotherapy intrahepatic cccDNAHui et al.84

Virological Aspects.

Although hepatitis B e antigen (HBeAg) positivity in patients with cancer appears to be a risk factor,18, 24, 76 this has not been found to be universally the case.10, 77–79 An increased risk has also been observed in the absence of HBeAg seropositivity, and this has been attributed to the presence of the precore/core promoter HBV mutants (i.e., HBeAg-negative/hepatitis B e antigen–positive chronic hepatitis B infection).80 This finding has also been associated with severe fulminant hepatitis.77–81 Cytotoxic chemotherapy does not appear to put wild-type or mutant HBV under preferential selection pressure.77

Early reports assessing pre-chemotherapy viral load failed to demonstrate it to be a predictor for HBV reactivation.18, 21, 76 However, with sensitive assays such as real-time polymerase chain reaction, recent studies in patients undergoing conventional-dose and high-dose chemotherapy with HSCT have shown that high pre-chemotherapy HBV DNA load, defined as above 3 × 105 copies/mL, is associated with increased likelihood of developing reactivation.58, 82 It is of note that different studies have applied different assays in measuring viral load, and the lack of standardization with variable range of detection may have limited the understanding of the disease.

Intrahepatic covalently closed circular DNA, a key intermediate in HBV replication,83 has been reported to have clinical significance in HBV reactivation.84 Intrahepatic covalently closed circular DNA will facilitate the understanding of HBV reactivation. However, this requires liver biopsy with inherent sampling error and is likely to limit its clinical applicability.

Cancer Type and Treatment Factors.

Several anticancer immunosuppressive agents have been associated with HBV reactivation (Table 3); these could be broadly categorized into conventional cytotoxic agents5–14, 18, 20 and biologic response modifiers with anti-B and -T monoclonal antibody therapies as discussed in the previous section. Among the former, corticosteroids and anthracyclines have been most frequently associated with the condition.10, 18, 85–87 HBV DNA contains a glucocorticoid responsive element that has been reported to facilitate HBV replication,88, 89 while anthracycline has been demonstrated in vitro to stimulate HBV DNA secretion from HepG2-derived 2.2.15 cells in a dose-dependent manner.90

Table 3. Chemotherapeutic Agents That Have Been Associated With Development of HBV Reactivation
ClassAgentsReferences
Conventional cytotoxics  
CorticosteroidsPrednisolone4-6,8-11,13-15,18,84
 Dexamethasone8-11,14,15,18
 Methylprednisolone15,85
Antitumor antibioticsDoxorubicin8-11,13,14,18,84
 Epirubicin18,87
 Daunorubicin4,14,85
 Bleomycin6,8-11,13,14,84
 Mitomycin-C18
 Actinomycin D6
Plant alkaloidsVincristine4-6,8-11,13,14,18,84
 Vinblastine18,84,85
Alkylating agentsCyclophosphamide4-6,8-11,13,14,18
 Ifosfamide84
 Chlorambucil5,11
 Carboplatin18
 Cisplatin6,11,18
AntimetabolitesCtyarabine4-5,11,14,85
 Azauridine5
 Fluouracil18
 Gemcitabine18
 Mercaptopurine5,85
 Methotrexate4,5,8-11,14,18,84
 Thioguanine5
OthersColaspase5
 Folinic acid5,9-11,18
 Procarbazine4,6,9,10,84
 Docetaxel18
 Etoposide11,14,18,84
 Fludarabine78
Biologic response modifiers  
Monoclonal antibodiesRituximab63-65,67,68
 Alemtuzumab66
OthersInterferon15,20,35

Both anthracyclines and corticosteroids are commonly used as part of the anticancer treatment and antiemetic premedication for patients who have hematological and breast malignancies. This explains the observation of increased HBV reactivation in these patients.9, 15, 16, 18, 21, 25 Although individual agents may be associated with HBV reactivation through specific mechanisms, the degree of immunosuppression as a consequence of combining these agents with others could also contribute to the development of the condition. First, patients who undergo intensive chemotherapy with bone marrow transplantation have been reported to have a higher incidence of HBV reactivation when compared with the more commonly used, standard dose of chemotherapy. Second, patients with gastrointestinal malignancies who undergo cytotoxic chemotherapy, mainly consisting of less-immunosuppressive agents (fluorouracil and folinic acid), have a lower risk of developing viral reactivation.89 Third, in HBV-related patients with hepatocellular carcinoma, the incidence of HBV reactivation appears to correlate with the level of immunosuppression of the anticancer therapy administered; viral reactivation was reported in 40%, 25%, and 2% of patients who underwent systemic chemotherapy, transarterial chemotherapy, and percutaneous ethanol injection or surgical resection, respectively, in descending order of immunosuppressive effects.20, 24, 76

Although the duration of chemotherapy has not been proven to be an associating factor,18 Kumagai et al.16 have reported that patients who were on second or third line of chemotherapy had a higher incidence of HBV reactivation.

HBV reactivation during chemotherapy is the result of a multifactorial interaction. A mathematical model for risk calculation has been suggested by incorporating the host, viral, and therapy factors: detectable pre-chemotherapy HBV DNA level (as measured via sensitive polymerase chain reaction assay, with a lower detection limit of 2.9 × 103 copies/mL), the use of steroids, and a diagnosis of lymphoma or breast cancer.89

Management of Established HBV Reactivation

Until recently, aggressive supportive therapy and discontinuation of cytotoxic chemotherapy has been the mainstay of treatment. Interferon, which has both antiviral and immunomodulatory functions, has been shown to control hepatitis during chemotherapy.16 However, the use of conventional interferon may be limited by the possibility of fatal hepatitic flare via augmentation of the immunomediated destruction of hepatocytes. Corticosteroids had initially been suggested to be effective in chronic HBV infection and HBV reactivation.91, 92 However, this has not been confirmed by other studies93, 94; on the contrary, and as discussed previously, corticosteroids have been frequently associated with the development of HBV reactivation.

Lamivudine, a nucleoside analogue, has been shown to have substantial activity in chronic HBV infection.95–100 Lamivudine has also been claimed to be effective in controlling viral replication during HBV reactivation,79, 101, 102 thereby allowing individual patients to continue cytotoxic chemotherapy upon maintenance of a sufficient hepatic function.102, 103 Sustained HBeAg seroconversion and undetectable serum HBV DNA level for at least 3 months after the discontinuation of lamivudine therapy have also been observed in some patients who were initially HBeAg-seropositive,103 and the postulated mechanism has been a high immune response to HBV reactivation that allows elimination of covalently closed circular DNA in hepatocytes in conjunction with suppression of viral replication. Lamivudine has also been reported to be effective in cases of hepatic decompensation during HBV reactivation94, 104 and in cases that involve a precore HBV mutant.102, 103

Such claims should be treated with caution, however. It is clear that even before the widespread availability of lamivudine, spontaneous resolution of viral reactivation was a common outcome. Thus, the reported efficacy during HBV reactivation in recent reports could well have been a coincidental finding that parallels the spontaneous decrease in viral replication that occurs upon immune recovery of the host after chemotherapy.105 Figure 1 shows a breast cancer patient who was started on lamivudine upon biochemical evidence of hepatitis while undergoing chemotherapy. With the prospectively collected sera during the course of her chemotherapy, it was shown that the rise in HBV DNA preceded that of ALT, and that by the time lamivudine was administered, the viral DNA had become undetectable. In the absence of serial HBV DNA measurement, the subsequent clinical and biochemical improvement might well have been attributed to the antiviral administration.

On the other hand, despite lamivudine, HBV-associated mortality has been reported in up to 20% of the HBsAg-positive patients treated. This has been postulated to be due to a delay in antiviral administration at a time when severe hepatic impairment with massive hepatic damage had already occurred.18, 48–49, 79, 103 Thus, it is possible that only with prompt administration of the antiviral (i.e., the first instance in which HBV DNA first starts to rise and before severe irreversible hepatic damage) will it be effective. However, the potential intense monitoring modality may not be cost-effective and may be difficult to conduct in clinics that lack adequate laboratory support. Thus, the efficacy of the antiviral therapy administered after symptoms have developed has yet to be demonstrated.

Prevention of HBV Reactivation

Even if lamivudine is an effective therapeutic measure, the development of HBV reactivation invariably means that anticancer therapy is disrupted, with delay at the least and premature termination in the more severe circumstances. Therefore, preventing the occurrence of this condition may provide a more practical approach in managing patients who require chemotherapy.

One potential means of minimizing the risk of HBV reactivation is the avoidance of corticosteroid therapy as part of chemotherapeutic/antiemetic regimes in HBsAg carriers.14–15, 106, 107 However, HBV reactivation may still occur in association with the use of other cytotoxic/immunosuppressive agents.108 Furthermore, this may lead to suboptimal therapy and may even jeopardize the patient's chance of cure. In a prospective study of 50 patients with non-Hodgkin's lymphoma who were randomized to receive either the standard steroid-containing regimen (prednisolone/epirubicin/cyclophosphamide/etoposide) or a steroid-free regimen (epirubicin/cyclophosphamide/etoposide), whereas the cumulative incidence of HBV reactivation was significantly higher in the steroid-containing study arm (73% vs. 38%; P = .03), patients in the steroid-free arm had a significantly lower rate of complete remission and shorter overall survival.86

Continuous low-dose steroids and a gradual tailing off of immunosuppressive cytotoxics have been suggested,6, 8, 10 but others have been unable to confirm any advantage of such a regimen.14, 15

Leaw et al.109 reported that interferon administered from the start of chemotherapy prevented HBV reactivation in 13 lymphoma patients. However, using interferon-containing combination chemotherapy for patients with inoperable HBV-related hepatocellular carcinoma has not reduced the incidence of HBV reactivation.20

The Role and Rationale of Prophylactic Lamivudine to Prevent Reactivation.

Because viral replication occurs before clinical evidence of hepatitis,105 it raises the possibility of using lamivudine in a prophylactic manner before the administration of chemotherapy.

Earlier studies assessing the prophylactic role of lamivudine have been based on retrospective series.110–118 Table 4 lists reports that compared patients who did and did not receive prophylactic lamivudine; all of these reports demonstrated that the prophylactic antiviral reduced HBV reactivation.

Table 4. Studies of Prophylactic Lamivudine for HBV Reactivation
Patient PopulationType of ChemotherapyType of StudyNo. of Patients in the Study ArmDuration of Prophylactic Lamivudine Treatment*No. of Controls (No Prophylactic Lamivudine)Incidence of HBV Reactivation (Studied Population vs. Controls)Reference
  • *

    Time of commencing prophylactic lamivudine treatment (from start of chemotherapy).

  • Systemic chemotherapy with HSCT.

  • Systemic chemotherapy with/without HSCT.

Various malignanciesSystemicRetrospective series13Variable; 0.5-24 months after completion of chemotherapy50% vs. 40%Shibolet et al.113
LymphomaSystemicRetrospective series3Until 2 mo after completion of chemotherapy210% vs. 57%Persico et al.114
LymphomaSystemicRetrospective series20Until 12 mo after completion of chemotherapy205% vs. 45%Lau et al.115
Various malignanciesSystemicRetrospective series16Details not reported190% vs. 37%Lim et al.116
LymphomaSystemicRetrospective series11Details not reported209% vs. 85%Lee et al.110
Hematological malignanciesSystemicRetrospective series8Until 12 mo after completion of chemotherapy100% vs. 50%Idilman et al.111
Hepatocellular carcinomaTranshepatic intra-arterial chemotherapyRetrospective series8Variable; up to 1 mo after completion of chemotherapy90% vs. 67%Nagamatsu et al.117
LymphomaSystemicRetrospective series4Details not reported80% vs. 67%Ozguroglu et al.118
LymphomaSystemicProspective randomized series15Until 6 wk after completion of chemotherapy or when white cell count is >4.0 × 109/L150% vs. 54%Lau et al.119
Various malignanciesSystemicProspective phase II study with historical controls65Until 2 mo after completion of chemotherapy1934% vs. 25%Yeo et al.120
LymphomaSystemicProspective series with historical controls11Until 1 mo after completion of chemotherapy530% vs. 32%Leaw et al.109
Breast cancerSystemicProspective series with historical controls31Until 2 mo after completion of chemotherapy616% vs. 31%Yeo et al.121
Breast cancerSystemicProspective series with historical controls11Until 1 mo after completion of chemotherapy90% vs. 55%Dai et al.25

More recent studies have been conducted prospectively.109, 119–121 In 1 of these studies,119 30 lymphoma patients undergoing chemotherapy with or without HSCT were randomized to receive lamivudine either as prophylactic therapy or deferred therapy initiated upon serological evidence of HBV reactivation. Prophylactic lamivudine significantly reduced reactivation (0% vs. 53%) and led to a significant improvement in the survival-free-from-hepatitis due to HBV reactivation.

In another study, 65 patients treated with prophylactic lamivudine were compared with 193 controls without prophylactic lamivudine before conventional-dose chemotherapy.120 Prophylactic lamivudine significantly reduced the incidence of HBV reactivation (5% vs 24%) and the associated morbidity, as evidenced by a significant reduction in incidences of hepatitis that were less severe and a significant decrease in disruption of chemotherapy. Furthermore, although not statistically significant, it was suggested that reactivation-associated mortality could be avoided.

In the setting of HSCT, 2 interesting issues are noteworthy. For HBsAg-positive HSCT recipients who received hematopoietic cells from an anti-HBs–positive donor, adoptive transfer of immunity with clearance of HBV infection has been reported,3, 122, 123 and the coadministration of lamivudine has been suggested to confer advantage in HBV clearance.124 On the other hand, for HSCT recipients who received HSCT from HBsAg donors, a high incidence of HBV-related hepatitis after HSCT—with up to one sixth of patients dying from HBV-related hepatic failure—has been reported.49, 125, 126 Recently, pre-emptive antiviral therapy before HSCT in all positive donors and all recipients as well as HBV vaccination in all HBsAg-negative recipients has been shown to significantly reduce HBV-related hepatitis and hepatic failure.127

Limitations of Lamivudine.

Careful clinical monitoring is still required despite the potential benefits of the prophylactic lamivudine approach. In the case of chronic HBV infection, prolonged therapy with lamivudine has been associated with an increased likelihood of treatment-resistant HBV variants with YMDD mutations,128, 129 from 24% at 1 year, to 38% at 2 years, to 50% at 3 years, to 67% at 4 years.130, 131 The emergence of lamivudine-resistant mutants is usually associated with a breakthrough with a moderate increase of serum HBV DNA and ALT levels, which may remain lower than at baseline for several months. In patients with decompensated cirrhosis, lamivudine therapy can be associated with flares that result in liver failure and death.

Among patients receiving the standard dose of chemotherapy, a resistant mutant has been found sporadically after 6 to 8 months of prophylactic lamivudine.132, 133 Data on the possible consequences of such a mutant in an immunocompromised patient remain very limited, and its potential link with the development of HBV reactivation is unknown.

Another concern with the use of lamivudine has been the occurrence of withdrawal hepatitic flares upon cessation of the antiviral.134 With patients being followed for up to 3 months after lamivudine treatment, withdrawal hepatitis flares have been observed in up to 13% of patients112, 118–125 and have been associated with high pre-chemotherapy HBV DNA load.135

Currently, apart from lamivudine, there are 2 other approved nucleoside analogues for the treatment of chronic HBV infection: adefovir and entecavir. Adefovir has been used as a first-line therapy for established HBV reactivation.136 It has also been used as a second-line antiviral therapy in immunocompromised transplant recipients who developed lamivudine resistance.137, 138 Although presently there are no data on adefovir and entecavir as a prophylactic treatment in preventing HBV reactivation during chemotherapy, there are potential advantages to using these agents. In chronic HBV infection, resistance to either agent appears to be lower than lamivudine: 12% for adefovir at 3-year follow-up,139 and sporadically for entecavir for patients with prior lamivudine resistance at 1-year follow-up.140, 141 Longer follow-up upon drug discontinuation is required to determine the true incidences of resistant mutants,142 the possibility of virological relapses, and the associated side effects—especially in the setting of oncological practices, where concurrent administration of anticancer therapies could exacerbate treatment toxicities.

Conclusions

HBV reactivation is a common complication in HBsAg-positive patients undergoing immunosuppressive anticancer therapy. For patients with cancer who are from HBV-endemic areas, routine screening for HBsAg before cytotoxic chemotherapy should be performed. Prophylactic therapy with nucleoside analogues has been shown to significantly decrease the incidence and overall morbidity of HBV reactivation. However, considering that (1) not all patients receiving immunosuppressive therapy will develop reactivation and (2) lamivudine therapy has its potential shortcomings, one might suggest that lamivudine be offered only to patients at high risk of reactivation. However, until such risk factors can be clearly identified, HBsAg-positive patients should begin prophylactic antiviral treatment before chemotherapy.

The optimal duration of such therapy remains to be determined. Most of the reported studies on prophylactic lamivudine continued antiviral therapy for another 1 to 2 months after completion of chemotherapy (Table 3), as has been recommended by the Asian-Pacific Consensus Update Working Party on Chronic Hepatitis B, who advised that in such circumstances, lamivudine should be continued until at least 6 weeks after the end of immunosuppression or chemotherapy.143 Treatment may have to be prolonged for patients receiving HSCT or anti–B or –T cell therapy, and those who have high pre-chemotherapy viral load.119, 135 The exact duration may be guided by the monitoring of immune recovery and, in the case of rituximab therapy, measurement of CD20 counts would provide a more direct monitoring modality. Furthermore, all patients treated should be monitored for postlamivudine hepatic flare, the occurrence of which may prompt retreatment with lamivudine.

For HBsAg-negative patients who have evidence of previous HBV infection, the data are currently insufficient to provide information on the incidence of HBV reactivation. In highly endemic areas, 20% of cancer patients have been reported to be HBsAg-negative and anti-HBc positive9; therefore, a universal adoption of the prophylactic antiviral approach is currently not justified and is unlikely to be cost-effective. Patients planning for anticancer therapies that include anti–B or –T cell therapies or HSCT should be screened for previous HBV infection. Those patients found to be anti-HBs/anti-HBc–positive should be monitored during immunosuppressive treatment—preferably with up to weekly liver function tests and HBV DNA titers—and antivirals should be administered as a deferred therapy upon evidence of HBV reactivation.

With further research leading to a better understanding of the mechanism of HBV reactivation and the availability of established and potential new antiviral agents such as adefovir and entecavir, the prospects are good that HBV-seropositive patients with cancer may be offered appropriate cytotoxic chemotherapeutic treatment without jeopardizing their prognosis.

Acknowledgements

We thank Dr. Nancy Leung for her expert advice and critical review of the manuscript.

Ancillary