This article is mentioned in:

Introduction

Reactivation of hepatitis B virus (HBV) replication with an increase in serum HBV DNA and alanine aminotransferase (ALT) level has been reported in 20–50% of chronic hepatitis B (CHB) carriers undergoing immunosuppressive or cancer chemotherapy (1–4). A study on lymphoma patients (1) demonstrated that hepatitis caused by HBV reactivation usually occurred after two or three courses of chemotherapy; however, it may develop at any time during or after immunosuppressive therapy. The clinical presentation of HBV reactivation may vary, ranging from symptomatic, mild hepatitis to hepatic failure or death (4–6). As previously described, the risk of HBV reactivation and prognosis are determined by host factors, underlying disease, baseline HBV status and the intensity of immunosuppression (1,2,7–12).

Regarding the significance of baseline HBV replication status, as previously described (6), chronic HBV-infected patients with positive hepatitis B surface antigen (HBsAg) have a higher risk of reactivation following immunosuppressive or cancer chemotherapy compared with patients with resolved HBV infection (i.e., anti-HBc-positive, HBsAg-negative). For patients with positive HBsAg, a higher HBV DNA level prior to therapy (>2,000 IU/ml) has been demonstrated to be an important risk factor for HBV reactivation compared with those having lower HBV DNA levels (11,12). The majority of patients with resolved HBV infection have undetectable serum HBV DNA levels; however, the risk of HBV reactivation remains higher for the minority with detectable HBV DNA at baseline (13,14).

To prevent HBV reactivation during cancer chemotherapy, clinical studies, including randomized controlled trials, demonstrated that prophylactic use of an oral antiviral agent, mainly lamivudine, may reduce the rate of HBV reactivation and the severity of associated hepatitis flares and mortality (15–19). According to the 2009 American Association for the Study of Liver Diseases (AASLD) CHB practice guideline update (20), patients with a high baseline HBV DNA (>2,000 IU/ml) level must continue treatment until they reach treatment endpoints as in immunocompetent patients. In addition, oral antiviral agents with high potency and low resistance profile, such as entecavir or tenofovir, were preferred if a longer duration of treatment is anticipated, or in patients with baseline HBV DNA >2,000 IU/ml to avoid selection of drug-resistant mutations. Therefore, measurement of serum quantitative HBV DNA level was mandatory prior to initiation of chemotherapy.

To date, the data on baseline HBV DNA level prior to the initiation of cancer chemotherapy, particularly in CHB patients with non-hematological malignancies, are limited. Therefore, this retrospective study was conducted to address this issue. In addition, factors associated with virological activities were investigated.

Patients and methods

Eligible candidates

This study was a retrospective, single-center, data collection study. From November, 2011 to December, 2014, a total of 105 consecutive CHB carriers with non-hematological malignancies who received prophylactic oral antiviral agents prior to the initiation of chemotherapy were enrolled by the investigators for analysis. CHB carrier status was defined as follows: Positive HBsAg for >6 months preceding entry into the study, with either positive or negative hepatitis B e antigen (HBeAg). Patients were excluded if they were already on antiviral therapy for CHB, or had evidence of coinfection with hepatitis C virus, hepatitis D virus or human immunodeficiency virus. After November, 2009, the National Health Insurance Administration, Ministry of Health and Welfare of the Taiwanese government reimbursed the cost of prophylactic oral antiviral agent for CHB carriers to prevent HBV reactivation during cancer chemotherapy. In the present study, 6 patients received prophylactic therapy with lamivudine (100 mg/day) and the remaining 99 patients were treated with entecavir (0.5 mg/day). Prophylactic antiviral agent was started within 1 week prior to the initiation of chemotherapy and was maintained for ≥6 months after completion of cancer chemotherapy. The present study was performed in accordance with the principles of Good Clinical Practice, the principles of the Declaration of Helsinki and its appendices, and local and national laws. Written informed consent was obtained from all participants.

Study design and laboratory examinations

Demographic data and clinical information on the enrolled patients, including cancer diagnosis, staging and chemotherapy regimen, were collected. Biochemical and hematological testing were performed by commercially available assays. Hepatitis B serological markers (HBeAg and anti-HBe) were tested using commercially available enzyme-linked immunosorbent assay kits. Quantitative HBsAg levels were determined by the ARCHITECT HBsAg assay (Abbott Laboratories, Abbott Park, IL, USA), with a sensitivity of 0.05 IU/ml (21). Serum quantitative HBV DNA levels prior to the initiation of chemotherapy was measured by commercially available quantitative polymerase chain reaction (PCR) assay (COBAS® AmpliPrep/COBAS® TaqMan® HBV Test, Roche Molecular Diagnostics, Basel, Switzerland), with a lower detection limit of 20 IU/ml (22).

Statistical analyses

The results are expressed as mean ± standard error of the mean. Data were analyzed using the SPSS software package, version 12.0 (SPSS Inc., Chicago, IL, USA). Statistical analyses were performed using Chi-square and Fisher's exact tests for categorical variables. Independent t-tests were used for continuous variables. The association between HBV DNA level (in log10 IU/ml) and serum quantitative HBsAg level (in log10 IU/ml) were determined by the Pearson's correlation model. The results were considered statistically significant at P<0.05.

Results

Baseline characteristics of enrolled patients (Table I)

The mean age of the enrolled patients was 55.2±1.1 years; 48 of the patients were male; 3 of the patients were HBeAg-positive, and 27.6% had abnormal ALT levels at baseline (>40 U/l). The patient composition by diagnosis was as follows: Breast cancer (37.1%), lung cancer (18.1%), colon cancer (17.1%), head and neck cancer (10.5%), other gastrointestinal tract malignancies (8.6%), gynecological cancer (4.8%) and others (malignant thymoma, melanoma and osteosarcoma) (3.8%).

Distribution of baseline HBV DNA levels prior to the initiation of chemotherapy

The median HBV DNA level prior to chemotherapy was 3.30 log10 IU/ml. The distribution of HBV DNA levels were as follows: Undetectable by PCR, i.e., <20 IU/ml (15.2%); 20≤DNA<2,000 IU/ml (35.3%); 2,000≤DNA<20,000 IU/ml (26.6%); 20,000≤DNA<106 IU/ml (17.2%); and ≥106 IU/ml (5.7%) (Table II). A total of 49.5% of the enrolled patients had baseline HBV DNA levels of >2,000 IU/ml.

Table II. Distribution of baseline HBV DNA levels prior to the initiation of chemotherapy.

Table II.

Distribution of baseline HBV DNA levels prior to the initiation of chemotherapy.

HBV DNA level (IU/ml)	No. (%)
Undetectable by quantitative PCR (<20)	16 (15.2)
20≤DNA<2,000	37 (35.3)
2,000≤DNA<20,000	28 (26.6)
20,000≤DNA<106	18 (17.2)
≥106	6 (5.7)

[i] HBV, hepatitis B virus; PCR, polymerase chain reaction.

Comparisons between patients with baseline HBV DNA levels ≤2,000 IU/ml and >2,000 IU/ml (Table III)

Age, gender and hematological data were not significantly associated with virological activity. Patients with a baseline HBV DNA level of >2,000 IU/ml tended to have a higher mean level of ALT and aspartate aminotransferase; however, the difference did not reach significance. As compared with patients with baseline normal ALT, a higher percentage of patients with abnormal baseline ALT had HBV DNA level >2,000 IU/ml (65.5 vs. 44.7%, respectively) but the difference was not statistically significant (P=0.057). The mean serum quantitative HBV DNA level was comparable between patients diagnosed with breast cancer (3.16±0.24 log10 IU/ml), colon cancer (3.14±0.45 log10 IU/ml), lung cancer (3.20±0.33 log10 IU/ml) and other types of camcer (3.68±0.28 log10 IU/ml). The percentage of patients with HBV DNA level >2,000 IU/ml were similar among these groups. All 3 patients with positive HBeAg had HBV DNA levels >2,000 IU/ml. The quantitative HBsAg level was a major determinant of baseline HBV DNA level: 38/51 (74.5%) patients with HBsAg ≥250 IU/ml had a HBV DNA level >2,000 IU/ml, as compared with patients with HBsAg <250 IU/ml (27.8%) (P<0.001). In addition, 20/23 (87.0%) patients with HBsAg >1,000 IU/ml had a HBV DNA level >2,000 IU/ml, whereas only 20% (8/40) with HBsAg <100 IU/ml had HBV DNA >2,000 IU/ml.

Correlation between quantitative HBsAg and HBV DNA levels

To investigate the correlation between quantitative HBsAg and HBV DNA levels, these data were transformed into log10 IU/ml and a significant correlation was observed between log10 HBsAg and log10 HBV DNA levels (correlation coefficient γ=0.641, P<0.001).

Discussion

HBV reactivation caused by cancer chemotherapy has been extensively investigated in patients with lymphoma. The earliest prospective study from Hong Kong (1) enrolled 100 patients for analysis and the serum HBV DNA levels were measured regularly prior to and during chemotherapy. At baseline, 27 of the patients were chronic HBV carriers (i.e., HBsAg-positive), 51 belonged to the resolved HBV infection group (i.e., anti-HBc-positive and HBsAg-negative) and 22 patients did not exhibit evidence of previous HBV infection. The results demonstrated that hepatitis associated with HBV reactivation developed in 48, 4 and 0% of the patients, respectively. In addition, hepatic failure due to HBV reactivation occurred in 7, 2 and 0% of the patients, respectively (1). The increased risk of HBV carriers may be explained by the necessity of intensive chemotherapy for lymphoma and the addition of corticosteroids, which exert a potent immunosuppressive effect, and the direct stimulation of viral replication via glucocorticoid-responsive elements in the HBV genome (23).

HBV reactivation may also occur in patients receiving chemotherapy for solid tumors, the targeted population of the present study. Initially, the majority of the reports were based on the findings from patients treated for breast cancer (7,24). However, later reports suggested that HBV reactivation by chemotherapy developed in a wide range of solid tumors, including colon, lung and head and neck cancers (2,8). In a study that included 63 HBsAg-positive patients with solid tumors (2), the incidence of HBV reactivation was reported to be ~10–38% among different diagnoses. A higher incidence of reactivation has been reported in patients with breast cancer (41%) compared with that in patients with other solid tumors (7–29%) (8); the concomitant use of anthracyclines has been suggested to be one of the possible underlying mechanisms (25,26).

To the best of our knowledge, this is the first study to evaluate the baseline HBV replication status prior to the initiation of chemotherapy for consecutive CHB patients with non-hematological malignancies. Interestingly, a wide variation of baseline HBV DNA distribution was observed. Approximately 15% of the enrolled CHB-infected patients had undetectable HBV DNA by quantitative PCR prior to the initiation of cancer chemotherapy. By contrast, a total of 49.5% of the enrolled patients with a baseline HBV DNA level of >2,000 IU/ml, were at risk for HBV reactivation induced by immunosuppression (11,12). Our results supported the significance of universal screening prior to initiation of cancer chemotherapy recommended by the US Centers for Disease Control and Prevention (27), the European Association for the Study of the Liver (28) and the Asian-Pacific Association for the Study of the Liver (29), particularly in highly endemic area of HBV infection, such as Taiwan. As several patients are unaware of their HBV infection status, the purpose of HBV screening is to identify patients who are infected and who may benefit from preventative antiviral therapy or close monitoring.

Regarding the selection of oral antiviral agents, the majority of published studies investigating the efficacy of prophylactic use of antivirals in preventing HBV reactivation by chemotherapy, were based on the use of lamivudine, the first nucleos(t)ide analogue approved for the treatment of hepatitis B. However, the emergence of resistance to lamivudine may lead to hepatitis flares, liver failure and death, negating the clinical benefit (20). Based on previous report, the incidence of lamivudine resistance was ~14–32% after 1 year and ~60–70% after 5 years of therapy for patients with CHB (20). Despite its prophylactic use, drug-resistant mutations and hepatitis flares have been reported in patients treated with lamivudine during immunosuppressive therapy (18,30). Recently, a newer generation of oral antivirals with higher antiviral potency, such as entecavir and tenofovir, have become available. Studies on immunocompetent patients with CHB demonstrated that the cumulative incidence of antiviral resistance for naïve patients was 1.2% to entecavir and 0% to tenofovir up to 5 years (31,32). A non-randomized study from China (33) demonstrated that prophylactic therapy with entecavir (n=34) was associated with significantly lower rates of hepatitis (5.9 vs. 27.0%, P=0.007), hepatitis B reactivation (0 vs. 12.4%, P=0.024) and disruption of chemotherapy (5.9 vs. 20.2%, P=0.042) as compared with the lamivudine group (n=89). Based on the results obtained from the present study, lamivudine may not be the optimal option for all patients and the aim of ‘preventive use’ cannot be applied to all enrolled subjects. In agreement with the 2009 AASLD CHB practice guideline (20), we suggest that individuals with positive HBsAg and high HBV DNA levels (≥2,000 IU/ml), and those who require long-term prophylaxis (i.e., >12 months), should be treated with more potent antiviral agents possessing a higher genetic barrier to resistance, such as entecavir or tenofovir. Tenofovir may be preferable in patients who have had prior exposure to lamivudine, whereas entecavir may be preferred in those with a risk of renal function deterioration.

Strong evidence currently supports that the prophylactic use of antivirals to prevent HBV reactivation after cancer chemotherapy is more effective compared with rescue or on-demand use (6,14). However, data regarding the timepoint for antiviral agent discontinuation are very limited. A previous study demonstrated that high levels of serum HBV DNA (≥2,000 IU/ml) prior to chemotherapy was the most important predictor of HBV relapse after withdrawal of prophylactic antiviral therapy (34). Baseline HBV DNA level may be used as an important guide to determine the duration of antiviral therapy. According to the 2009 AASLD CHB practice guideline (20), patients with a baseline HBV DNA level <2,000 IU/ml should continue treatment for 6 months after completion of chemotherapy. Prophylactic antiviral therapy should be continued until the therapeutic endpoint for CHB is achieved in patients who are HBsAg-positive with a high viral load (≥2,000 IU/ml) prior to the initiation of chemotherapy to reduce the risk of post-treatment relapse.

Since the identification of HBsAg by Blumberg in 1965, HBsAg has been used as a diagnostic marker for overt HBV infection (35). Automated quantitative assays for serum HBsAg have recently become available, facilitating standardized quantification of serum HBsAg level. This has led to increased interest in the clinical application of quantitative serum HBsAg levels for predicting therapeutic response (36,37). Our present study demonstrated that quantitative HBsAg level was the only baseline factor associated with HBV DNA levels. A total of 74.5% of the patients with HBsAg ≥250 IU/ml have HBV DNA levels >2,000 IU/ml, compared with 27.8% patients with HBsAg <250 IU/ml. Other clinical and demographic variables are not helpful in discriminating baseline virological activities prior to the initiation of cancer chemotherapy. In concordance with previous reports (38,39), we observed a significant correlation between HBsAg and HBV DNA levels. Serum quantitative HBsAg levels at baseline may provide complementary information on virological activity if HBV DNA data are not available.

In summary, the key to the prevention of HBV reactivation is the identification of patients with HBV infection prior to initiation of cancer chemotherapy. We recommend universal screening in highly endemic areas, such as Taiwan, as no validated risk tools to identify HBV infection may be easily applied in clinical practice. We observed a wide distribution of baseline HBV DNA levels in CHB patients diagnosed with non-hematological malignancies. Of note, approximately half of the patients (i.e., those with HBV DNA >2,000 IU/ml) had a higher risk of HBV reactivation if no proper antiviral prophylaxis was undertaken. The selection of antiviral agents and the duration of therapy should be adjusted according to the baseline HBV replication status.

Acknowledgements

The present study was supported in part by grants from the Taipei Veterans General Hospital (V105C-068), Szu-Yuan Research Foundation of Internal Medicine (104-011) and Research Foundation of Digestive Medicine (104-10), Taiwan, R.O.C.

Glossary

Abbreviations

Abbreviations:

References