Abstract
Hepatitis B virus (HBV) reactivation is well documented in previously resolved or inactive HBV carriers who receive cancer chemotherapy. The consequences of HBV reactivation range from self-limited conditions to fulminant hepatic failure and death. HBV reactivation also leads to premature termination of chemotherapy or delay in treatment schedules. This review summarizes current knowledge of management of HBV reactivation in patients receiving cancer chemotherapy. HBV surface antigen (HBsAg) testing should be performed in patients who require cancer chemotherapy. Four meta-analyses support lamivudine prophylaxis for HBV reactivation during chemotherapy in HBsAg-positive patients. Randomized controlled trials to compare different HBV antiviral agents are needed to define optimal regimens for the prevention and treatment of HBV reactivation in patients receiving cancer chemotherapy.
Introduction
There are 400 million chronic carriers of hepatitis B virus (HBV) worldwide and 800,000 to 1.4 million in the US [Sorrell et al. 2009; Weinbaum et al. 2009]. Approximately half to two thirds of HBV carriers in US were born in other countries [Weinbaum et al. 2009]. HBV reactivation has been well documented in previously resolved or inactive HBV carriers who receive cancer chemotherapy. The consequences of HBV reactivation range from a self-limited hepatitis to fulminant hepatic failure and death [Liang et al. 1990; Galbraith et al. 1975; Pinto et al. 1990; Yeo et al. 2009]. HBV reactivation also necessitates premature termination of chemotherapy or delay in treatment schedules [Yeo et al. 2003, 2005]. Several factors have served to further accentuate the clinical importance of HBV reactivation: (1) advances in potent chemotherapy for cancer; (2) the increasing utilization of immunosuppressive agents in transplantation or autoimmune diseases; (3) the increasing influx of immigrants from HBV high endemic to low endemic areas; and (4) a potentially fatal outcome that is readily preventable with prophylactic antiviral treatment. This work reviews current concepts regarding management of HBV reactivation in patients receiving cancer chemotherapy. A PubMed search with the keywords ‘hepatitis B’ AND ‘chemotherapy’ was conducted to retrieve published articles focused on natural history, pathogenesis, prevention and treatment of HBV reactivation.
Natural history of HBV reactivation
Definition of HBV reactivation
HBV reactivation is defined as an abrupt increase in serum HBV DNA and alanine transaminase (ALT) levels in a patient with resolved or inactive HBV infection [Hoofnagle, 2009; Lok and McMahon, 2009]. In patients with negative HBV surface antigen (HBsAg) and positive antibody to HBV core antigen (anti-HBc), HBV reactivation, termed de novo hepatitis, is defined as HBsAg reappearance with threefold elevated ALT on two consecutive tests 5 days apart and accompanied by an increase in HBV DNA to more than 105 copies/ml [Hui et al. 2006].
Epidemiology
The incidence of HBV reactivation in HBsAg positive patients receiving cancer chemotherapy has been reported to range from 20% to 57% [Kim et al. 2007; Kumagai et al. 1997; Yeo et al. 2000a, 2000b, 2003]. HBV reactivation was observed in 50% of HBsAg-positive patients receiving rituximab-based therapy [Mendez-Navarro et al. 2011]. Reactivation does not necessarily require receipt of systemic chemotherapy. For instance, the reactivation rate in those receiving transarterial chemoembolization (TACE) for hepatocellular carcinoma (HCC) has been reported to be as high as 34% [Jang et al. 2004].
Reactivation can be seen in persons with past exposure who do not harbor active HBV. The HBV reactivation rate in subjects with negative HBsAg and positive anti-HBc was 3% [Hui et al. 2006], and ranges from 2% to 25% in those receiving rituximab-based therapy [Ji et al. 2010; Koo et al. 2011; Matsue et al. 2010; Yeo et al. 2009].
Clinical features of HBV reactivation
In hematologic malignancies, early reports with serial follow up of HBV serologic markers observed reduction of antibody titers to HBV with subsequent reappearance of HBV antigen in 29% of patients under cancer chemotherapy. In addition, HBsAg was persistent in 40% of them and an increase in antigen titer was associated with elevated ALT [Wands et al. 1975]. Elevation in HBsAg titer was noted in 83% of patients with hematologic malignancies who received glucocorticoids [Ohtsu et al. 1991]. During follow up for up to 1 year after cancer chemotherapy, loss of antibody to HBV surface antigen (anti-HBs) was observed in 1% [Alexopoulos et al. 1999].
In breast cancer patients receiving doxorubicin and cyclophosphamide who were premedicated with dexamethasone, serum HBV DNA peaked prior to ALT by nearly 2 weeks [Yeo et al. 2001]. In contrast, serum HBV DNA rose 5–8 weeks prior to the ALT in chronic HBV patients with spontaneous reactivation [Maruyama et al. 1993] and in those treated with corticosteroid alone, 4 weeks in advance of peak ALT [Nair et al. 1986].
In de novo HBV-related hepatitis, a hundredfold rise in serum HBV DNA preceded threefold elevated ALT by 12 to 28 weeks [Hui et al. 2006]. HBsAg seroreversion developed after rise in serum HBV DNA and before ALT elevation [Hui et al. 2006]. Rituximab plus steroid is an independent factor associated with de novo HBV-related hepatitis [Hui et al. 2006]. Thus, patients with negative HBsAg and positive anti-HBc need a much longer interval to develop clinical hepatitis as compared with those with positive HBsAg. This late onset of clinical hepatitis suggests a delayed immune recovery because of the prolonged suppressive effects of rituximab [Dai et al. 2004a]. Rituximab-induced deficiency in antigen-presenting cells which led the HBV to escape the control of cytotoxic T-lymphocyte [Hui et al. 2006].
Serial serum HBV DNA and ALT monitoring before and during chemotherapy is important for a timely diagnosis of HBV reactivation. Since HBV DNA elevation precedes ALT elevation in both HBsAg-positive and HBsAg-negative/anti-HBc-positive patients (Figure 1), vigilance for HBV DNA during chemotherapy is essential so that prompt preemptive antiviral therapy may be instituted.

Typical clinical course of HBV reactivation in HBsAg-positive and HBsAg-negative/anti-HBc-positive individuals receiving chemotherapy. In HBsAg-positive patients, serum HBV DNA was undetectable at the time of peak ALT, instead, it peaked prior to ALT by around 2 weeks [Yeo et al. 2001]. In HBsAg-negative/anti-HBc-positive subjects, serum HBV DNA preceded elevated ALT by 12–28 weeks [Hui et al. 2006]. HBsAg seroreversion developed after rise in serum HBV DNA and before ALT elevation [Hui et al. 2006].
Risk factors for HBV reactivation
HBV reactivation has been observed in patients receiving chemotherapy for hematologic malignancy [Galbraith et al. 1975; Yeo et al. 2009; Hwang et al. 2011], breast cancer [Kim et al. 2007; Yeo et al. 2003], hepatocellular carcinoma (HCC) [Nagamatsu et al. 2003; Yeo et al. 2004c], nasopharyngeal carcinoma [Yeo et al. 2005], and chemoradiation in postgastrectomy for gastric/esophageal adenocarcinoma [Cheng et al. 2004].
The risks of HBV reactivation in patients with positive HBsAg and negative HBsAg/positive anti-HBc are shown in Table 1. The cutoff level of serum HBV DNA between HBsAg-positive patients with versus without reactivation was 3 × 105 copies/ml [Zhong et al. 2004]. One of the modifiable risks among the risk factors for HBsAg-positive patients may be glucocorticoid use. This may be due to steroid increased HBV replication through glucocorticoid responsive element in HBV genome which enhances HBV replication greatly. In this regard, steroid-free regimens in HBsAg-positive patients with lymphoma were associated with reduced incidence and severity of HBV reactivation [Cheng et al. 2003].
Risks for HBV reactivation in patients with positive HBsAg and negative HBsAg/positive anti-HBc.
HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; HBeAg, hepatitis B extracellular antigen; ALT, alanine transaminase; TACE, transarterial chemoembolization; HCC, hepatocellular carcinoma; OR, odds ratio; ARR, adjusted relative risk; CI, confidence interval.
Consequences of HBV reactivation
HBV reactivation has been associated with jaundice in 10% of patients [Liang et al. 1990], fulminant hepatic failure in 6% [Liang et al. 1990] and death [Galbraith et al. 1975; Pinto et al. 1990; Yeo et al. 2009]. In fulminant hepatic failure or severe hepatitis, patients harboring precore mutations had increased risk compared with those without these mutants [Dai et al. 2001; Steinberg et al. 2000]. The overall liver related mortality associated with reactivation has been estimated to be 5% [Liang et al. 1990], with high mortality rates (30%) in patients with HCC [Yeo et al. 2004c]. Around 11% of HCC patients receiving TACE died of hepatic decompensation due to HBV reactivation [Jang et al. 2004]. In patients with de novo HBV-related hepatitis, fulminant hepatic failure developed in 38% [Hui et al. 2006]. In addition, patients with HBV reactivation had delays or interruption of chemotherapy schedule or premature termination in 68–71% versus 19–33% of those without [Yeo et al. 2003, 2005], which may itself be associated with decreased cancer survival.
Pathogenesis of HBV reactivation
The underlying mechanism of HBV reactivation during cancer chemotherapy remains unclear. The reactivation is consistent with replication induced from latent forms of HBV even after serologic evidence of viral clearance [Hoofnagle, 2009]. It has been generally hypothesized that cancer chemotherapy leads to widespread HBV infection in hepatocytes by enhancing replication and suppressing the cellular immune response to the virus. Subsequent resurgence of immune function following chemotherapy discontinuation then results in rapid destruction of all infected hepatocytes [Galbraith et al. 1975; Xunrong et al. 2001].
In support of this premise are data demonstrating that direct exposure of HBV-expressing human hepatoma 1.3ES2 cells to doxorubicin or etoposide increased HBV DNA replication and HBV pregenomic transcription [Chung and Tsai, 2009]. Furthermore, HBV replication during chemotherapy is associated with activation of DNA repair pathways. Activation of chemotherapy-induced DNA repair signaling upregulated promyelocytic leukemia protein in its nuclear body (PML-NB) and the upregulated PML-NB initiated HBV replication [Chung and Tsai, 2009].
Prevention and treatment
Screening
The National Institutes of Health Consensus Development Conference Statement recommends routine screening for hepatitis B of newly arrived immigrants to the US from countries where the HBV prevalence rate is intermediate or high (i.e. greater than 2%) [Sorrell et al. 2009]. The US Centers for Disease Control and Prevention (CDC) also recommend testing of persons born in geographic regions with HBsAg prevalence of greater than 2% [Weinbaum et al. 2009]. The CDC recommends testing for HBsAg, anti-HBc and anti-HBs for all patients before the receipt of chemotherapy [Weinbaum et al. 2009]. The American Society of Clinical Oncology provisional clinical opinion suggested testing for HBsAg on high risk of previous HBV exposure or reactivation and testing for anti-HBc in some but not all patients [Artz et al. 2010]. However, testing for anti-HBs was not supported by evidence [Artz et al. 2010]. In patients with solid tumors who require chemotherapy, testing for HBsAg alone was the most economical strategy [Day et al. 2011]. For patients with reported HBV reactivation rates of more than or equal to 41% as in HBsAg-positive patients receiving rituximab-based therapy and in populations with expected HBsAg prevalence of 2.7%, testing for HBsAg and anti-HBc maximized cost-effectiveness [Day et al. 2011].
In clinical practice, a retrospective study in US showed that only 17% of cancer patients were screened for HBsAg and anti-HBc prior to chemotherapy [Hwang et al. 2011]. Testing for HBV has not been adequately performed in patients at risk for HBV in US.
Prevention and treatment
Lamivudine is a nucleoside analogue that effectively suppresses HBV replication. Four meta-analyses have supported lamivudine prophylaxis for HBV reactivation during chemotherapy in HBsAg-positive patients (Table 2). All studies included in these four meta-analyses showed a lower rate of HBV reactivation in patients receiving lamivudine prophylaxis (Table 2). The earliest analysis revealed prophylactic lamivudine decreased HBV reactivation by fourfold to sevenfold [Kohrt et al. 2006]. Two other meta-analyses, with much overlap in the studies included in both analyses, showed that lamivudine reduced the risk for HBV reactivation by 80–87% and reactivation-related mortality by 70% over no or deferred lamivudine use [Loomba et al. 2008; Martyak et al. 2008]. Preventive lamivudine also reduced treatment delays and premature termination of chemotherapy by 59% [Martyak et al. 2008]. A recent meta-analysis focused on lymphoma confirmed the efficacy of prophylactic lamivudine in reducing HBV reactivation [Ziakas et al. 2009]. Prophylactic lamivudine also reduced the severity of chemotherapy-related HBV reactivation in patients with solid tumors as well as lymphoma in a randomized trial and a retrospective study [Hsu et al. 2008; Yun et al. 2011]. A more recent retrospective study did not support the use of prophylactic lamivudine in HBsAg-negative anti-HBc-positive individuals receiving chemotherapy or rituximab [Koo et al. 2010] given the low frequency of this event [Hui et al. 2006].
Meta-analyses on lamivudine prophylaxis for HBV reactivation during chemotherapy in HBsAg-positive patients.
RCT, randomized controlled trial; Lmv, lamivudine prophylaxis; No, no prophylaxis; E/N, events/total number; RR, relative risk; HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen.
Lymphoma patients.
Timing and duration of therapy
The optimal duration of preventive lamivudine remains unclear. Discontinuation of preventive lamivudine at 3 months after completion of chemotherapy resulted in a 24% rate of HBV reactivation [Hui et al. 2005]. Prechemotherapy serum HBV DNA of ≥104 copies/ml predicted reactivation after discontinuation [Hui et al. 2005]. In a decision analysis, extended lamivudine prophylaxis for 2 years throughout rituximab maintenance therapy improved 3-year overall survival rates by 2.4% in HBsAg-positive patients [Ziakas et al. 2009]. A cost-effectiveness study supported prophylactic lamivudine in HBsAg-positive patients with lymphoma 1 week before chemotherapy and for 6 months after the discontinuation of chemotherapy [Saab et al. 2007].
The use of lamivudine is limited by the development of resistance mutations that may contribute to hepatic decompensation and liver failure. The frequency of lamivudine resistance ranges from 3% to 8% in HBsAg-positive patients with hematologic malignancies under prophylactic use [Hsu et al. 2008; Pelizzari et al. 2004]. Adefovir has lower primary resistance rates and has been shown to be effective against lamivudine-resistant HBV strains. Two case reports of successful adefovir use either alone or in combination with lamivudine in patients with HBV reactivation under cancer chemotherapy have been reported [Cortelezzi et al. 2006; Perez-Roldan et al. 2005]. Entecavir is a more potent inhibitor of HBV replication with a very low long-term rate of primary resistance [Chang et al. 2006]. A retrospective multicenter study indicated that entecavir was superior to lamivudine in preventing HBV reactivation in lymphoma patients [Li et al. 2011]. Based on the anticipated duration of treatment, AASLD Practice Guidelines currently recommend lamivudine or telbivudine for a planned duration of ≤12 months, while tenofovir or entecavir are recommended for longer planned duration [Lok and McMahon, 2009]. Randomized controlled trials to compare different HBV antiviral agents are needed to establish optimal drug regimens for prevention and treatment of HBV reactivation in patients receiving cancer chemotherapy. Case reports described liver transplantation as a life-saving option for patients with liver failure due to HBV reactivation when the prognosis of the coexistent malignancy is good [Kim et al. 2010; King and Chung, 2010; Noterdaeme et al. 2011].
Discussion
The time to discontinue prophylactic lamivudine has been suggested as 6 months after completion of chemotherapy in a cost-effectiveness study [Saab et al. 2007]. A shorter duration of 3 months resulted in high HBV reactivation rate in a longitudinal follow-up study [Hui et al. 2005], whereas an extended duration for 2 years improved 3-year overall survival rates in a decision analysis by only 2.4% [Ziakas et al. 2009]. Therefore, prophylactic lamivudine should be continued for 6 months after completion of chemotherapy (Table 3), however, in patients with prechemotherapy serum HBV DNA of ≥2000 IU/ml, extended lamivudine use is recommended (Table 3). Further study is needed to determine the timing of treatment cessation in: (1) prophylactic use of antiviral agents with higher potency; and (2) HBsAg-negative patients receiving antiviral agents for serum HBV DNA elevation during chemotherapy. Data on monitoring following treatment discontinuation as well as management of reactivation in hepatitis B and delta co-infection have also been limited.
Recommendations for management of HBV reactivation in patients receiving cancer chemotherapy.
HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; HBeAg, hepatitis B extracellular antigen; ALT, alanine transaminase.
Footnotes
Funding
Yi-Wen Huang is a research fellow at Massachusetts General Hospital and Harvard Medical School sponsored by National Science Council of Taiwan, Republic of China and Cathay General Hospital Medical Center, Taipei, Taiwan. Dr. Chung was supported by NIH DK078772 and DK082919.
Conflict of interest statement
The authors declare no conflicts of interest in preparing this article.
