Annals of Hematology

, Volume 87, Issue 6, pp 475–480

High prevalence of occult hepatitis B virus infection in patients with B cell non-Hodgkin’s lymphoma

Authors

  • Ming-Huang Chen
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
  • Tzeon-Jye Chiou
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
  • Jin-Hwang Liu
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
  • Jyh-Pyng Gau
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
  • Hao-Wei Teng
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
  • Wei-Shu Wang
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
  • Ta-Chung Chao
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
  • Chueh-chuan Yen
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
  • Po-Min Chen
    • Division of Hematology & Oncology, Department of MedicineTaipei Veterans General Hospital
    • National Yang-Ming University School of Medicine
Original Article

DOI: 10.1007/s00277-008-0469-9

Cite this article as:
Chen, M., Hsiao, L., Chiou, T. et al. Ann Hematol (2008) 87: 475. doi:10.1007/s00277-008-0469-9

Abstract

Several reports recently found that patients with B cell non-Hodgkin’s lymphoma (NHL) had a higher carrier rate of hepatitis B surface antigen (HBsAg). The current study aimed to examine the hepatitis B virus (HBV) infection status of NHL patients in Taiwan, an HBV-endemic area. Serum HBV and serum hepatitis C virus were measured in 471 NHL patients and 1,013 non-lymphoma cancer patients enrolled between February 2000 and March 2007. Furthermore, nested polymerase chain reaction of HBV-DNA was used to examine the sera from selected patients in these two populations and healthy volunteers for the presence of occult HBV infection. The infection rates (as indicated by the rates of HBsAg and occult HBV) were compared between different groups. There was a higher incidence of HBV infection in B cell NHL patients (23.5%), especially patients with diffuse large B lymphoma, than solid tumor patients (15.6%, P = 0.001). Among HbsAg-negative patients, those with B cell NHL had a higher prevalence of occult HBV infection (6%) than those with non-lymphoma solid tumors and healthy volunteers, 0% and 0.9%, respectively (P = 0.005). B cell NHL patients, even HBsAg-negative B cell NHL patients, but not T cell NHL patients, have a higher incidence of HBV infection than patients with solid tumors. Our findings support the etiologic role of HBV infection in B cell NHL.

Keywords

Hepatitis B virusNon-Hodgkin’s lymphomaOccult hepatitis B

Introduction

Patients with chronic hepatitis B and C virus (HBV and HCV) infections have an increased risk of developing hepatocellular carcinoma [13]. Evidence indicates that both viruses are lymphotropic [4, 5]. Several studies conducted in different parts of the world have shown that patients with non-Hodgkin’s lymphoma have a higher prevalence of HCV infection [6, 7].

Similarly, several recent reports from different countries also found that patients with B cell non-Hodgkin’s lymphoma (NHL) had a higher carrier rate of hepatitis B, i.e., were positive for hepatitis B surface antigen (HBsAg) [812].

With the advance of polymerase chain reaction (PCR) technology, occult hepatitis B virus infection has been redefined in terms of persistence of hepatitis B virus genomes in the liver or serum of HbsAg-negative individuals [1318]. Several studies showed that rate of occult HBV infection was increased in patients infected with human immunodeficiency virus (HIV) [1921] and HCV [16, 18, 22] and in individuals at high risk of parenterally transmitted infection including intravenous drug addicts [23], hemophilia patients [24], and hemodialysis patients [25, 26]. Moreover, a higher rate of occult HBV infection has been observed in most HBsAg-negative populations in HBV-endemic areas (as defined by the carrier rate of HBsAg) [18].

As Taiwan is an HBV-endemic area, analyzing the status of HBV infection in Taiwanese patients with non-Hodgkin’s lymphoma would be informative. The current study examined and compared the HBV infection status (i.e., the HBsAg carrier rate and occult HBV infection rate) of NHL patients with that of different control populations. We provide additional evidence supporting the possible role of HBV infection in the etiology of NHL.

Materials and methods

Patients

Totally, 471 patients with the diagnosis of NHL at Taipei Veterans General Hospital were enrolled between February 2000 and March 2007. The type of lymphoma was diagnosed according to the World Health Organization (WHO) classification criteria [27]. Blood samples were routinely drawn from all patients, and serological tests for HBsAg, HCV, and HIV were routinely performed at diagnosis. The available sera (stored at diagnosis) from HBsAg-negative patients were screened for occult HBV infection (i.e., HBV-DNA by nested PCR). The study used three different control groups: (1) patients with non-lymphoma cancers (except hepatocellular carcinoma) enrolled by our department during the period of NHL patient enrollment (n = 1013) and used to determine the prevalence of HBsAg carrier status; (2) HBsAg-negative patients in control group 1 with sera stored at diagnosis (n = 124) and chosen at random to determine occult HBV infection status; and (3) HBsAg-negative healthy volunteers (n = 108) enrolled at the time of the study also to determine occult status. All participants provided written informed consent. The evaluations were based on the data available on March 31, 2007.

Methods

Serological tests for viral infection

HBsAg was detected by the microparticle enzyme immunoassay (IMx-Abbott Laboratories, Abbott Park, IL, USA), and the reactivity of HBsAg was calculated from the ratio of the sample rate to the MODE 1 calibrator rate. Antibodies to hepatitis C virus were also detected using a microparticle enzyme immunoassay (MEIA, Abbott IMx HCV version 3.0, USA).

Nested PCR for occult HBV infection

Viral DNA was extracted from 200 μl of each serum and lymphoma biopsy tissue using the High PureTM viral nucleic acid kit (Roche Molecular Biochemicals, Indianapolis, IN, USA) according to the manufacturer’s recommendations. DNA was dissolved in 50 μl of elution buffer. The presence of occult HBV infection was then examined using nested PCR with three independent sets of primers to detect different HBV proteins, including pre-S/S (S), pre-core/core (Core), and X viral regions. The sequences of these primers are shown in Table 1 [28]. DNA (20 ng per sample) was analyzed for the presence of HBV genomes by performing three different nested PCR amplification assays. The first nested PCR amplification was performed for 35 cycles, each consisting of denaturing for 40 s at 94°C, annealing for 30 s at the annealing temperature (pre-S/S region 55°C, X region 56°C, precore/core region 58°C), and extension for 40 s at 72°C. The second nested PCR amplification was performed for 30 cycles, each consisting of denaturing for 40 s at 94°C, annealing for 30 s at the annealing temperature (pre-S/S region 53°C, X region 60°C, precore/core region 58°C), and extension for 40 s at 72°C. In patients seronegative for HBsAg, the presence of occult HBV infection was considered detected if at least two of the three nested PCR amplifications was HBV DNA-positive when serum was HBsAg-negative [29].
Table 1

PCR primers for detecting HBV genome

 

Sense primer

Anti-sense primer

 

Primer set designation

Nucleotide sequence

Nucleotide sequence

PCR reaction product size (bp)

Pre-S/S

5′-CAT CAG GAC TCC TAG GAC CC-3

5′-TGC TCG TGT TAC AGG CG-3′

239

5′-GAG GAC AAA CGG GCA ACA-3′

5′-GAG GCA TAG CAG CAG GAT G-3′

X

5′-CCA TAC TGC GGA ACT CCT AGC-3′

5′-GCT AGG CTG TGC TGC CAA CT-3¢

159

5′-CGT TCA CGG TGG TCT CCA T-3′

5′-CGT AAA GAG AGG TGC GCC CCG-3′

Pre-core/core

5′-TCG CAT GGA GAC CAC CGT GA-3′

5′-CAT AAG AGG ACT CTT GGA CT-3′

358

5′-ATA GCT TGC CTG AGT GC-3′

5′-GGA AAG AAG TCA GAA GGC-3′

bp base pair

Statistical analysis

The values are expressed as number (%) for categorical variables and median (range) for continuous variables. Patient demographics and baseline characteristics were compared using an independent-sample t test for continuous variables and chi-square test or Fisher’s exact test for categorical variables. All analyses were performed using the SPSS 13.0 statistical package (SPSS Inc, Chicago IL, USA) and P values of <0.05 were considered statistically significant.

Results

Carrier rates of HBsAg in patients with non-Hodgkin’s lymphoma and other solid tumors

The carrier rate of HBsAg was significantly higher in NHL patients (107 of 471, 22.7%) than non-lymphoma cancer patients (control group 1; 158 of 1013, 15.6%, P = 0.001), but there was no between-group difference in prevalence of HCV infection (Table 2).
Table 2

Serological status of hepatitis B surface antigen in patients with B cell and T cell NHL and non-lymphoma cancer patients

 

NHL (%)

Non-lymphoma cancer (%)

P value

No. of patients

 

471 (100)

 

1,013 (100)

 

HBsAg-positive

 

107 (22.7)

 

158 (15.6)

0.001

Anti HCV-positive

 

23 (4.8)

 

54 (5.3)

0.913

Immunophenotype

B cell

 

T cell

  

No. of patients

400 (100)

 

71 (100)

1013

 

Gender

M/F

258/142

 

40/31

630/383

0.419/0.456a

Age (years)

Median

63

 

46

53

0.001/0.001a

Range

16–91

 

14–83

18–86

 

HBsAg-positive

94 (23.5)b

 

13 (18.3)b

158 (15.6)

0.001/0.613a

NHL non-Hodgkin’s lymphoma, HBsAg hepatitis B surface antigen

aP value for the comparisons between patients with B cell and T cell NHL and patients with non-lymphoma cancers (except hepatocellular carcinoma), respectively.

bThe proportion of B cell or T cell NHL patients, respectively

Carrier rates of HBsAg in NHL patients of different types

The higher HBsAg carrier rate was analyzed in immunophenotypes of NHL patients. The difference was present in patients with B cell NHL (P = 0.001), not in those with T cell NHL (P = 0.613; Table 2). With respect to B cell lymphoma subtypes, only diffuse large B cell lymphoma (DLBCL) patients had a significantly higher carrier rate (P < 0.001; Tables 3 and 4).
Table 3

Hepatitis B surface antigen status in patients classified according to WHO criteria for B cell and T cell NHL

 

Total

HBsAg-positive

Percentage

B cell lymphoma

Total

400

94

23

 Precursor B lymphoblastic leukemia/lymphoma

6

1

16

Chronic lymphoblastic leukemia

9

1

11

Lymphoplasmacytic lymphoma

1

0

0

Marginal zone B cell lymphoma

10

3

30

Maltoma

6

0

0

Follicular lymphoma

50

10

20

Mantle cell lymphoma

11

2

18

DLBCL

271

68

25a

Burkitt’s lymphoma/leukemia

16

3

18

Unclassified

20

6

30

T cell lymphoma

Total

71

14

19

Precursor T lymphoblastic leukemia

4

1

25

NK/T cell lymphoma

10

1

10

Subcutaneous panniculitis-like T cell lymphoma

4

1

25

Primary cutaneous anaplastic large cell lymphoma

5

0

0

Peripheral T cell lymphoma-unclassified

25

4

16

Angioimmunoblastic T cell lymphoma

11

4

36

Anaplastic large cell lymphoma

4

2

50

Unclassified

8

1

12

NHL non-Hodgkin’s lymphoma, HBsAg hepatitis B surface antigen, Maltoma mucosa-associated lymphatic tissue lymphoma, DLBCL diffuse large B cell lymphoma

aStatistically significant compared with non-lymphoma cancer patients (p < 0.001, chi-square test)

Table 4

Hepatitis B surface antigen status in patients with non-lymphoma cancers

 

Total

HBsAg-positive

Percentage

Total

1,013

158

16

Gastrointestinal tumor

Esophageal cancer

16

5

31

Gastric cancer

47

6

13

Cholangiocarcinoma

18

5

28

Pancreatic cancer

28

4

14

Colorectal cancer

277

42

15

Breast cancer

148

25

17

Lung cancer

108

12

11

Melanoma

8

1

13

Head and neck cancer

70

13

19

Genitourinary tumor

Prostate cancer

20

4

20

RCC/bladder cancer/ureteral cancer

79

7

9

Germ cell tumor

25

5

20

Cervical cancer

8

1

13

Soft tissue tumor

Sarcoma

69

8

12

Gastrointestinal stromal tumor

11

4

36

Neuroendocrine tumor

10

2

20

Adrenal gland tumor

5

1

20

Unknown primary cancer

66

13

20

The difference in rate of HBV infection between patients with different solid tumors is not statistically significant.

HBsAg hepatitis B surface antigen, RCC renal cell carcinoma

Occult HBV infection in HBsAg-negative NHL patients and non-lymphoma cancer patients

The sera from 83 HBsAg-negative B cell NHL patients were examined, but none from HBsAg-negative T cell NHL patients was examined because of the relatively small number of cases. As shown in Table 5, the rate of occult HBV infection was significantly higher in NHL patients (5 of 83, 6%) than in healthy volunteers (1 of 108, 0.9%) and patients with solid tumors (0 of 124, P = 0.005). For those five infected with occult hepatitis B, the lymphoma subtypes included diffuse large B cell lymphoma (two patients), follicular lymphoma (one), mucosa-associated lymphoid tissue (one), and mantle cell lymphoma (one). Nested PCR of DNA extracted from tumor samples of these five patients failed to detect occult HBV infection (i.e., the HBV genome was undetectable).
Table 5

Characteristics of HBsAg-negative B cell NHL patients and different control populations

 

B cell NHL

Non-lymphoma cancer

Healthy volunteers

P value

No. of patients

83

124

108

 

Gender

M/F

52/31

76/48

60/48

0.423

Age (years)

Median

56

54.3

54.5

0.328

Range

19–86

18–82

19–93

 

Occult HBV

Positive (%)

5 (6)

0 (0)

1 (0.9)

0.005

Occult HBV is indicated by sera negative for hepatitis B surface antigen and positive for nested PCR.

NHL non-Hodgkin’s lymphoma

Discussion

The rates of both serum HBsAg and occult HBV infection in our present study showed that HBV infection was more prevalent in NHL patients than in patients in different control populations. This finding provides additional support for the relationship between HBV infection and B cell NHL. In current study, the infection rate (i.e., HBsAg seropositivity) in patients with solid tumors (15.6%) was similar to the previously reported rates in Taiwanese adults (12.8–15.7%) [30, 31]. The rate of occult hepatitis B infection was relatively lower in the HBsAg-negative population including patients with B cell NHL (6%), those with solid tumors (0%), and healthy volunteers (0.9%) than in those reported from other HBV-endemic areas (0–18% of subjects without liver disease) [32, 33]. In addition to differences in geographic distribution and in population, there were differences in samples (e.g., sera or liver tissues), PCR methods (simple vs. nested), and definition of occult HBV infection between the current and previous studies [1416, 18].

Furthermore, the finding that the higher HBV infection rate was mainly in patients with diffuse large B cell lymphoma (Table 3) has not been previously reported [812]. Although possibly biased by differences in B cell lymphoma subtype frequency between our study populations and those of past studies, our finding may provide information that could help decide the population makeup needed to study the relationship between NHL and HBV infection.

Although HBV infection occurred at a higher rate in NHL patients’ sera, we failed to provide direct evidence linking HBV infection to lymphoma, i.e., to detect the viral genome in lymphoma tissues. This pattern of evidence was similar to that for HCV infection in NHL patients in that detection of HCV infection (HCV antibodies and RNA) was also almost entirely restricted to patients’ sera [4, 7, 34]. Possibly, HBV infection does not directly cause lymphoma but rather acts indirectly on B lymphocytes in hosts with persistent HBV infection to transform them after an incubation period [12].

Although our findings did not provide direct evidence supporting the etiologic role of HBV in the development of lymphoma, the impact of a higher HBV infection rate in patients with B cell NHL (especially those with occult HBV infection) cannot be overlooked. During chemotherapy, occult hepatitis B in NHL patients may reactivate and lead to a typical course of chronic HBV infection [35, 36]. Early detection of occult hepatitis B infection and preemptive use of anti-HBV drugs (e.g., lamivudine) may be effective in decreasing these HBV-related events [36].

In conclusion, the current study demonstrated a higher prevalence rate of HBV infection in patients with B cell NHL, including those negative for surface antigen. In addition to its etiologic role in lymphoma, occult HBV infection affects the therapy of lymphoma.

Acknowledgments

We thank Ms. Shu-Chou Chaou and Ms. Chia-Min Shih for performing the laboratory analyses of HBV and HCV markers and Ms. Pui-Ching Lee for statistical analysis. The study was supported by the grants from Taiwan Cancer Clinic Foundation, Taipei Veterans General Hospital (VGH93-192 and 94-236 for PMC, and 96B2-012 for LTH) and National Science Council (NSC), Taiwan (NSC94-2314-B075-049 for PMC, and NSC95-2745-B-075-008 & 96-2321-B-075-008 for LTH).

Copyright information

© Springer-Verlag 2008