Case-Control Study on Concentrations of Organohalogen
Compounds and Titers of Antibodies to Epstein-Barr
Virus Antigens in the Etiology of Non-Hodgkin
Lymphoma

Leukemia and Lymphoma. v.00 Jan01

Leukemia and Lymphoma. 2001, Vol. 00. pp. 1-11

Lennart Hardell ^a*, Mikael Eriksson ^c, Gunilla Lindstrom ^d, Bert van Bavel ^d,  Annika Linde ^c, Michael Carlberg ^a,  Göran Liljegren ^b 

^a Department of Oncology and ^b Department of Surgery, Örebro Medical Center, S-701 85 Örebro, Sweden, ^c Department of Oncology, University Hospital, S-221 85 Lund, ^d Department of Environmental Chemistry, Umeå University, S-901 87 Umeå and ^e Department of Virology, Swedish Institute for Infectious Disease Control and Microbiology Tumour Biology Center, Karolinska Institute, S-171 82 Stockholm

(In final form December 15, 2000)

* Correspondence to: Department of Oncology, Örebro Medical Centre, S-701 85 Örebro, Sweden. Fax: +46 19 101768 E-mail: lennart.hardell@orebroll.se 

A rapid increase in incidence of non-Hodgkin lymphoma (NHL) has been reported from many countries. Exposure to certain pesticides and organochlorines has been shown to be risk factors. Epstein-Barr virus (EBV) is a human herpesvirus that has been associated with some subgroups of NHL, such as Burkitt lymphoma and lymphomas related to severe immunosuppression. In this study, we measured lipid adjusted blood concentrations of 36 congeners of polychlorinated biphenyls (PCBs), p,p'-dichlorodiphenyl-dichloroethylene (p,p'-DDE), hexachlorobenzene (HCB), four different subgroups of chlordanes (trans-nonachlordane, cis-nonachlordane, MC6 and oxychlordane) and 2,2',4,4'-tetrabrominated diphenyl ether (TBDE) in incident cases of NHL and controls from the general population. Titers of antibodies to the Epstein-Barr early antigen (EA) were correlated to concentrations of organochlorines. We found a significant difference in lipid adjusted blood concentrations of total PCBs and T13DE between cases and controls. Titers of antibodies to EA IgG > 80 were correlated to an increased risk for NHL with odds ratio (OR) = 1.9, 95% confidence interval (CI) =0.943.8. This risk was further increased in those with a level above the median value of "sum of PCBs" (OR=4.0, CI=1.2-14), HCB (OR=5.3, CI=1.6-19), sum of chlordanes (OR=4.0, CI=1.2-14) and TBDE (OR=21, CI=4.6-124), suggesting an interaction between EBV and a higher concentration of these chemicals. Also for the "sum of immunotoxic PCBs" increased risk was found in that group (OR=6.4, CI=1.9-24). Subdivision of NHL in histological types yielded highest risks for low-grade B-cell NHL.

Keywords: non-Hodgkin lymphoma, PCB, DDE, hexachlorobenzene, chlordane, tetrabrominated diphenyl ether, Epstein-Barr virus, etiology

Contract grant sponsors: Cancer-och Allergifonden, Örebro County Council Research Committee, Örebro Medical Centre Foundation, Gunnar Nilsson Foundation, IngaBritt och Arne Lundbergs forskningsstiftelse, and Lions Research Foundation

A rapid increase in incidence of non-Hodgkin lymphoma (NHL) during the last 20-30 years has been reported in many Western countries.(1-3) Several studies have shown pesticide exposure to be a risk factor for NHL.(4,5) Conditions characterized by immunosuppression such as immunosuppressive treatment after organ transplantation, severe inborn immunodeficiencies and HIV-infection have been reported to increase the risk for NHL.(4,6,7) Some factors evaluated as possible risk factors for NHL, such as exposure to organochlorines, have been reported to induce immunological changes.(8-10)10)

Epstein-Barr virus (EBV) is a human herpesvirus with a tropism for B-lymphocytes, and the virus is found worldwide. The majority of the world's adult population has antibodies to EBV antigens. The primary infection usually occurs during childhood and is often subclinical. After the primary infection, a latent infection is established, which is balanced by the immune response by the host, and among others antibodies to the Epstein-Barr Early antigen (EA), Viral Capsid antigen (VCA) and Epstein-Barr nuclear antigen (EBNA) may be detected.(11) EBV has been associated with certain types of NHL such as Burkitt lymphoma and lymphomas occurring in immunologically compromised or HIV infected subjects.(12)

Polychlorinated biphenyls (PCBs) are aromatic chemicals that are very stable in the environment. They do not occur naturally but have been widely used in electrical equipment and building constructions because of their physical properties. Theoretically, 209 congeners are possible, but only about 130 of these are likely to occur in commercial products. Humans are exposed to PCB mainly through the food chain by consumption of contaminated fish, meat and dairy products. In humans, PCB is stored mainly in the adipose tissue.(13) PCB was used in Sweden until 1973 when use was prohibited by law.

Exposure to PCB has been suggested to induce measurable changes in the immune system,(8,14) although in doses higher than background exposure. Increased concentrations of some specific PCBs in

patients with NHL compared with controls has been shown.('5) In another study, a dose-response relation between lipid-corrected concentrations of total PCB (quartiles) and risk of NHL overall was found.(16) Flame-retardants such as polybrominated diphenyl ethers (PBDEs) have been increasingly used since the 1970's and are environmental hazards identified more recently.(17,18) They are halogenated aromatic compounds just as PCBs and may exert similar toxicological effects.(19,20) In our previous study we reported increased concentrations of 2,2',4,4'-tetrabrominated diphenyl ether (TBDE), one congener of PBDE, in patients with NHL.(21)

DDT is an insecticide that has been widely used because of its chemical stability. DDT and its metabolites p,p'-dichlorodiphenyl-dichloroethyelene (p,p'-DDE) and DDD are lipid soluble and bioaccumlate. No association has been found between exposure to DDT or concentrations of p,p'-DDE and NHL in epidemiological studies. (5,15-17,22,23)

Hexachlorobenzene (HCB) has been used in agriculture as a fungicide and occurs also in industry as an intermediate in chemical processes. HCB has been found as a contaminant in pesticides. (244n a previous study, no association between HCB arid NHL was found.(15)

Chlordanes are substances originating from synthesis of cyklodienes and have been described to be distributed throughout the biosphere. Technical chlordane has been used both in agriculture and in the control of termites. Production of chlordane started in USA in 1947 and chlordanes were registered in Sweden until 1969. Also these compounds are stored mainly in adipose tissue. Exposure to chlordane has been suggested to induce immunological changes measured in lymphocyte functions in vitro.(9) An association between chlordanes and NHL has also been suggested(25)

The purpose of this investigation was to determine concentrations of these organohalogen compounds in a larger sample of patients with NHL and population based controls. We also wanted to correlate concentrations of these substances to titers of antibodies to EBV Ethical committee approved the study.

TABLE I Concentrations of organohalogen compounds (ng/g lipid) in cases and controls. Wilcoxon P-value is given

a. n=81 cases, 83 controls
b. n=81 cases, 82 controls
c. n=53 cases, 65 controls

In the first phase of this study adipose tissue samples were obtained from the abdominal wall of incident NHL patients. Approximately 2-10 g adipose tissue was taken in local anesthesia. As controls, patients operated for a benign lesion at the Department of Surgery at the Örebro Medical Center Hospital were used. During that procedure an adipose tissue sample was taken from the abdominal wall. They were in the same age group (within 5 year interval) and of the same sex as the NHL patients and without a cancer diagnosis. This part of the study consisted of 50 cases and 47 controls. All these cases and controls were recruited during the same time period, 1994-97. They were all living in the catchment area of the Örebro-Uppsala medical region. The first results of this part of the study have been previously reported.(15,21,25)

In the second part of the study, blood samples were taken from 32 incident patients with NHL during the time period 1997-99. Of them 16 were derived from the Department of Oncology at the Örebro Medical Center and 16 from the Department of Oncology at the University Hospital in Lund. After informed consent approximately 50 ml of blood was drawn in glass tubes with heparin. The blood was centrifuged and plasma was frozen in glass bottles for later analysis. One control in the same age group as the case (5year interval) and same sex was recruited to each case from the Population Register. They were sent a letter asking for participation in the study. Glass tubes were then sent for blood sampling at the nearest medical department. In total 21 controls from the Örebro region and 15 from the Lund region were included.

Altogether, both parts of this study encompassed 82 cases and 83 controls. For technical reasons results were not available for included persons. From the majority of these subjects, blood was also drawn for analysis of antibodies to EBV-antigens. In total, this was done for 145 subjects, 67 cases and 78 controls.

All samples were assigned a unique id-number, which did not show if it was from a case or a control. Case or control identity was disclosed during the statistical analysis of the results.

The participants were sent a questionnaire, asking detailed questions about previous occupations and exposure to potential risk factors, such as pesticides.

Also length and weight at the time of sampling of adipose tissue or blood was assessed. Additionally the weight one year before the time when blood was drawn was asked for since weight might be changed during the development of NHL. The questionnaire was answered by all cases and all but one control. Body Mass Index (BMI) was calculated as the weight in kilograms divided by the square of the height in meters.

All cases were recruited by two of us (LH and ME) from our clinics. No case with NHL refused to participate. Controls for adipose tissue sampling at the surgical department were included on a consecutive basis by one of us (GL) when a control patient fulfilling the inclusion criteria was identified. No control subject refused to participate.

Antibodies to different EBV antigens were analyzed as earlier described. (26) In brief, indirect immunofluorescence (IF) was used for IgG and IgM antibodies to the virus capsid antigen (VCA) and the combined restricted and diffuse components of early antigens (EA R+D). The VCA IgG antibodies were endpoint titrated in four-fold dilutions from 1/20, while EA IgG was analyzed in one dilution, allowing for detection of antibodies in serum dilution >1/40 (titre >1/40). Positive samples were end-titrated. EBNA-1 IgG antibodies were screened in a peptide ELISA with the alanine-glycine repeat as antigen. The EA and VCA IgG titres found in healthy Swedes with the used methods have been published previously.(26) The samples were coded at examination, and blinded for if they were derived from a case or a control person. The completed results were uncoded by the clinicians, and titers and seroprevalences in cases and controls were compared statistically.

Approximately 20 ml of blood plasma was used for analyses of 36 PCB congeners, p,p'-DDE, HCB, four chlordane congeners and TBDE. The plasma samples and blank samples were fortified with ¹³C-labelled internal standards. The lipid fraction, including the organochlorines, was first removed from the plasma by use of a Hydromatrix column.(27) The lipid content was then determined gravimetrically and further cleaned up by multi-layer silica chromatography. Congener specific analyses and quantification of the organochlorines were done by high-resolution gas chromatography and mass spectrometry, HRGC-MS, running in El and SIM mode. The methods detection level, MDL, was in the range of 0.3-1 pg/g for the various analytes and samples. All results are expressed in ng/g lipid.

TABLE II Odds ratio (OR) and 95% confidence interval (CI) for different organohalogen compounds and NHL. As cut off the median concentration of the chemicals in the controls was used, numbers > median are shown for cases and controls. Adjustment was made for age, sex, BMI and specimen analysed (blood or adipose tissue)

TABLE III Multivariate analysis of organohalogen compounds and odds ratio (OR) with 95% confidence interval (CI) for NHL. As cut off the median concentration of the chemicals in the controls was used, numbers > median are shown for cases and controls. Adjustment was made for age, sex, BMI and specimen analysed (blood or adipose tissue)

TABLE IV Titers of antibodies to EBV-antigens and Wilcoxon p-value are shown for cases and controls

Unconditional logistic regression was performed using the SAS system (SAS Institute, Cary, NC) for calculation of odds ratio (OR) and 95% confidence interval (CI). In the analyses adjustments were made for age, sex, Body Mass Index (BMI) at the time of sampling, and specimen analyzed (adipose tissue sample or blood). Antibody variables and organohalogen variables were dichotomized using the median concentration of the controls. The SAS System was also used for descriptive statistics and Wilcoxon rank sum tests. In one of the analyses, PCBs were grouped according to immunotoxic properties for the detected PCBs (PCBs #66, 110, 105, 118, 74, 128/167, 156, 138, 170/190) as suggested by Moysich et al.(27)

Results for the sum of PCBs (36 congeners), sum of chlordanes (oxychlordane, MC6, trans-and cis-nonachlordane), p,p'-DDE, HCB, TBDE and antibody titers to EBV antigens are presented for 165 subjects, 82 cases and 83 controls. Of the cases 45 were men and 37 women and of the controls 48 were men and 35 were women. The mean age for the cases was 62.5 years (range 29-90 years) and for the controls 62.3 years (range 30-81 years). The age for one case was high, 90 years old, the next highest age for a case was 83 years. Exclusion of the 90-year-old case from the analyses did not change the results. Furthermore, no titers to EBV antigens were available for that case.

TABLE V Odds ratio (OR) and 95% confidence interval (CI) for NHL analysed with median values of antibodies to EBV antigens for controls as cut-off. Numbers of cases and controls > median are shown. Adjustment was made for age, sex, BMI and specimen analysed (blood or adipose tissue)

TABLE VI Odds ratio (OR) and 95% confidence interval (CI) for different organohalogen compounds and NHL in relation to titer to Epstein-Barr virus early antigen (EA) IgG. As cut off the median concentration of the chemicals in the controls was used Numbers > median (expressed in ng/g lipid) are shown for cases and controls. Adjustment was made for age, sex, BMI and specimen analysed (blood or adipose tissue)

The following results are based on all 82 cases and 83 controls. Analyses of chlordanes was missing for one case for technical reasons. For TBDE, results were obtained for 53 cases and 65 controls since these chemical analyses were not available at the start of this study, c.f. Hardell et al.(15,21) Regarding interaction between organohalogen compounds and titers to EBV antigens the results were based on 67 cases and 78 controls, however for TBDE on 44 cases and 59 controls.

TABLE VII Results presented for low-grade B-cell NHL. For details, see Table VI

Concentrations of organohalogens for cases and controls are shown in Table I. Significantly higher concentrations of the sum of PCBs (36 congeners), immunotoxic PCBs (11 congeners), and TBDE were found among the cases. The sum of chlordanes did not differ significantly between cases and controls, whereas the concentrations of MC6 and cis-nonachlordane were significantly higher in the cases.

HCB and TBDE did not correlate with the other study compounds. Some correlation was found between PCB and pp-DDE (cases rr=0.58, controls rr=-0.61), PCB and chlordane (cases rr=0.69, controls rr=0.49), and pp-DDE and chlordane (cases rr=0.74, controls rr=0.33).

TABLE VIII Results presented for high-grade B-cell NHL. For details, see Table VI