Document VjYbmO8aXKwBwYRnmgdaNzvJq

Malignant Lymphomas Research Paper Polymorphisms in the oxidative stress genes, superoxide dismutase, glutathione peroxidase and catalase and risk of non-Hodgkin's lymphoma Ferrata Storti Foundation Tracy J. Lightfoot Christine F. Skibola Alex G. Smith Matthew S. Forrest Peter J. Adamson Gareth J. Morgan Paige M. Bracci Eve Roman Martyn T. Smith Elizabeth A. Holly TJL and CFS contributed equally to the work. Background and Objectives. The risk of non-Hodgkin's lymphoma (NHL) has been associated with inflammation. One possible mechanism may involve oxidative stress as reactive oxygen species (ROS) can generate pro-inflammatory signals. Anti-oxidant enzymes including superoxide dismutase, glutathione peroxidase and catalase protect against the harmful effects of ROS. Genetic variation in the genes coding for these enzymes (SOD2, GPX1, and CAT, respectively) alters ROS production and therefore may provide a mechanism for the relationship between inflammation and NHL. Design and Methods. Data from two population-based, case-control studies of lymphoma in the UK (700 cases and 915 controls) and USA (1593 cases and 2517 controls) were pooled to analyze polymorphisms in genes involved in the oxidative stress response (SOD2 Val16Ala, CAT C-262T and GPX1 Pro197Leu). Results. No associations were observed between SOD2 Val16Ala and CAT C-262T and total NHL, diffuse large-B cell lymphoma or follicular lymphoma. However, when we looked at marginal zone lymphoma, a specific subtype of lymphoma characterised by inflammation, we found that homozygosity for the SOD2 16Ala allele was associated with a decreased risk among UK study participants. The GPX1 197Leu allele was weakly associated with NHL and follicular lymphoma. Interpretation and Conclusion. Analysis of genetic variation in oxidative stress genes in two lymphoma case-control studies suggests a possible role for oxidative stress in the risk of NHL. The risk modification is seen predominantly for marginal zone lymphomas which frequently arise in the context of chronic inflammation. However, in order to clarify the role of oxidative stress in the etiology of NHL analyses of additional polymorphisms and haplotypes in these and other genes involved in the oxidative stress response are needed. Key words: non-Hodgkin's lymphoma, oxidative stress, antioxidants, single nucleotide polymorphisms. Haematologica 2006; 91:1222-1227 2006 Ferrata Storti Foundation From the Epidemiology & Genetics Unit, University of York, York, YO10 5DD, UK (TJL, AGS, PJA, ER); Division of Environmental Health Sciences, School of Public Health, University of California, Berkeley, CA 94720 USA (CFS, MSF, MTS); Institute of Cancer Research & Royal Marsden NHS Trust Department of HaematoOncology, Downs Road, Surrey, SM2 5PT UK (GJM); Department of Epidemiology and Biostatistics, University of California, San Francisco, CA 94143 USA (PMB, EAH); Department of Health Research and Policy, Stanford University School of Medicine, Stanford, CA, USA 94305 (EAH). Correspondence: Tracy Lightfoot, Epidemiology & Genetics Unit, Area 3, Seebohm Rowntree Building, Department of Health Sciences, University of York YO10 5DD, United Kingdom. E-mail: tracy.lightfoot@egu.york.ac.uk Chronic inflammation has been associated with a risk of non-Hodgkin's lymphoma (NHL).1-3 Although the mechanisms underlying this association are uncertain, it is known that inflammation is associated with increased oxidative stress caused by the generation of reactive oxygen species (ROS). In addition to inducing oxidative DNA damage that can lead to neoplasia,4-6 ROS propagate pro-inflammatory cytokines including interleukin-1 which stimulates B cells to produce antibodies that initiate signaling for B-cell activation. Furthermore, a number of diseases characterized by a chronic B-cell activated phenotype, including arthritis, systemic lupus erythematosus, and Sjgren's syndrome have all been associated with an increased risk of NHL.3,7-9 Cells are protected from the harmful effects of ROS by anti-oxidant enzymes such as superoxide dismutase, glutathione peroxidase and catalase. Superoxide dismu- tase catalyses the dismutation of superoxide radicals (a major form of ROS) into hydrogen peroxide that glutathione peroxidase and catalase break down into water.10-13 Thus, an imbalance in anti-oxidant mechanisms may influence cellular sensitivity to free radical damage and alter susceptibility to disease. This may be particularly important in lymphocytes and their precursors in which chronic inflammation increases the risk of neoplastic changes, as is typified by the marginal-zone group of lymphomas which have been associated with chronic inflammation. Associations between oxidative stress and cancer have generally been investigated by measuring DNA damage, free radicals and anti-oxidants. Both increased superoxidase dismutase activity14,15 and decreased glutathione peroxidase activity16 have previously been reported in conjunction with lymphoproliferative disorders. However, greater | 1222 | haematologica/the hematology journal | 2006; 91(9) Oxidative stress genes and non-Hodgkin's lymphoma Ferrata Storti Foundation insight into the relationship between oxidative stress and disease susceptibility may be gained by studying functional polymorphisms in genes that control levels of cellular oxidative damage. Candidate pathways include anti-oxidant and DNA repair mechanisms which are modulated by individual genetic variation. A number of polymorphisms in the genes coding for superoxide dismutase, catalase and glutathione peroxidase (SOD2, CAT and GPX1, respectively) have been identified and associated with cancer risk.17-22 In the mitochondrial targeting sequence for SOD2 a T-C transition in the -9 position occurs resulting in a valine to alanine substitution (Val16Ala).23 This substitution alters protein structure which affects localization and transport of the superoxide dismutase enzyme into mitochondria where it exerts its antioxidant effects; the valine allele has been associated with decreased enzyme transport, reduced defense against superoxide radicals and increased risk of cancer.17-19 Several polymorphisms in GPX1 have been identified, including Pro197Leu, which has been associated with lung cancer,20 breast cancer21 and cardiovascular disease.24 The CAT C-262T polymorphism alters transcription factor binding and basal expression of catalase in red blood cells25 and has been associated with an increased risk of breast22 and pancreatic cancer. These and other polymorphisms that impair anti-oxidative capacity may influence NHL risk by increasing levels of pro-inflammatory cytokines leading to increased B-cell activation. To test this hypothesis we pooled data from two population-based studies, one conducted in the UK and the other in the USA, to examine the association between single nucleotide polymorphisms (SNP) in genes involved in anti-oxidative mechanisms (SOD2 V16A, rS4880; GPX1 P197L, rS1050450; CAT C-262T, rS1001179) and NHL subtypes. Design and Methods Study population Full details of both lymphoma case-control studies have been previously described.26-29 Briefly, in the UK study, patients aged 18-64 years diagnosed with NHL between 1998 and 2001 in certain areas of the country were identified. For each case, one individually sex-, age- and ethnicity- matched control was randomly selected from the same primary care practice list as the case. In the US, NHL patients aged 21-74 years who were residents of one of six San-Francisco Bay Area counties when newly diagnosed from 1988 to 1993 were identified through the Northern California Cancer Center's rapid case ascertainment. Controls were identified by random-digit dial and supplemented by random sampling of Health Care Financing Administration lists for participants aged 65 years or older and were matched to patients by sex, in 5-year age groups and county of residence. Participants from both studies provided informed consent to interview, collection and analysis of biological samples and, in the UK, access to their medical records. NHL histological subtype and grade were coded to the Working Formulation (US study) or the International Classification for Diseases for Oncology, Third Edition (ICD-O3) (UK study) (Fritz et al., 2000)30 with diagnoses grouped to reflect comparable ICD-O3 categories for analysis. Participation rates were over 70% for cases and controls in both the UK and USA studies. The present analysis includes 2293 cases (700 UK; 1593 US) with a confirmed diagnosis of NHL and 3432 controls (915 UK; 2517 US). DNA was available from 928 and 1446 HIVnegative, white non-Hispanic cases and controls, respectively. There were no significant differences in age, sex or diagnostic sub-group between those individuals who did and did not have DNA available for genotyping. The distributions by histological subtype, sex, age and study site for genotyped NHL cases are shown in Table 1. The majority of NHL cases with available genotyping data had either diffuse large B-cell lymphomas (41%) or follicular lymphomas (35%). However, DNA was also available from a number of patients with other histological NHL subtypes including mantle cell lymphoma (n=23), marginal-zone lymphoma (MZL) (n=54) and T-cell lymphoma (n=26) in the UK study and Burkitt-like lymphomas (n=2), diffuse mixed lymphomas (n=20), diffuse small lymphomas (n=18), lymphoblastic lymphoma (n=1), mycosis fungoides (n=10) and small lymphocytic lymphoma (n=47) in the US study. In addition 18 not otherwise specified diagnoses of NHL were included in the analyses. DNA extraction and genotyping DNA was extracted and quantified as previously described.26-29 Genotyping was performed using Taqman-based assays supplied by Applied Biosystems (ABI) (Applied Biosystems, Foster City, CA, USA) on a GeneAmp PCR System 7700 ABI Sequence Detection System using the following protocol: 95C for 10 minutes, then 40 cycles of 95C for 15 seconds and 60C for 1 minute. Primer and probe sets used were 5'TCGAAGCCCTGCTGTCTCA'-3 (forward primer), 5'CGAGACAGCAGCACTGCAA-3' (reverse primer), VIC- AGGGCCCAGCTGT (C-specific allele probe), 6FAM-AGGGCTCAGCTGTG (T-specific allele probe) for rs1050450 (GPX1 P197L); 5'-GGCCTGAAGGATGCTGATAACC-3' (forward primer), 5'-CTCTGGCCCAGCAATTGGA-3' (reverse primer), VICCCCGGGATAGCCG (C-allele specific probe) and 6FAM-CCCGGAATAGCCG (T-allele specific probe) for rs1001179 (CAT C-262T); and 5'-GGCTGTGCTTTCTCGTCTTCA-3' (forward primer), 5'-TCTGCCTGGAGCCCAGATAC-3' (reverse primer), VIC- haematologica/the hematology journal | 2006; 91(9) | 1223 | T. J. Lightfoot et al. Table 1. Distribution by histological subtype, sex and age for genotyped HIV-negative non-Hodgkin's lymphoma (NHL) cases and controls by study site. Variable USA San Francisco NHL Study Cases Controls Cases n (%) n (%) n (%) UK NHL study Controls n (%) Cases n (%) Both sites combined Controls n (%) Total 308 (100) 684 (100) 620 (100) 762 (100) 928 (100) 1446 (100) Histological subtypea DLBCL Follicular lymphoma Other 98 (32) 112 (36) 98 (32) - 283 (46) - 214 (34) - 123 (20) - 381 (41) - 326 (35) - 221 (24) - - - Sex Males Females 174 (62) 134 (38) 463 (73) 221 (27) 319 (51) 301 (49) 409 (54) 353 (46) 493 (53) 435 (47) 872 (60) 574 (40) Age <45 45-<55 55-<65 65+ Mean ageSD 48 (16) 60 (19) 93 (30) 107 (35) 57.511.7 231 (34) 125 (18) 102 (15) 226 (33) 52.914.7 87 (14) 222 (36) 311 (50) 0 (0) 53.68.4 159 (21) 255 (33) 348 (46) 0 (0) 52.09.6 135 (15) 282 (30) 404 (43) 107 (12) 52.49.8 390 (27) 380 (26) 450 (31) 226 (16) 54.912.3 aDLBCL, diffuse large B-cell lymphoma; other NHL cases with genotyping data from the UK lymphoma study include 23 mantle cell lymphomas, 54 marginal-zone lymphomas, 26 T-cell lymphomas and 20 NHL not otherwise specified and from the US San Francisco NHL study include 2 Burkitt-like lymphomas, 20 diffuse mixed lymphomas, 18 diffuse small lymphomas, a lymphoblastic lymphoma, 10 mycosis fungoides, 47 small lymphocytic and 9 NHL not otherwise specified. Ferrata Storti Foundation CCAAAACCGGAGCCA (T-specific allele probe) and 6FAM-CCCAAAGCCGGAGC (C-specific allele probe) for rs4880 (SOD2 V16A). Statistical analysis Hardy-Weinberg equilibrium for genotype frequencies was assessed in controls using standard 2 procedures. Odds ratios (OR) and 95% confidence intervals (CI) were estimated from pooled data using unconditional logistic regression and adjusted for age, sex and study region. The likelihood ratio test was used to test for trend and for interaction between SNP. Results were considered statistically significant for two-sided p-values of p<0.05. All analyses were conducted using STATA version 8.2 (College Station, TX, USA).31 Results Pooled genotype distributions for cases and controls are shown in Table 2. Genotype frequencies for controls were in Hardy-Weinberg equilibrium and, for SOD2 and CAT, were similar to those previously reported in Caucasian populations.25,32 The frequencies for GPX1 P197L in the two control groups differed between the US (Pro/Pro 50%, Pro/Leu 40%, Leu/Leu 10%) and UK populations (Pro/Pro 58%, Pro/Leu 35%, Leu/Leu 7%) (p=0.007), although the study-specific odds ratios for NHL risk were comparable (data not shown). No significant differences in the distribution of SOD2 V16A or CAT C-262T polymorphisms were observed between cases and controls in pooled analyses. However, modestly increased risk estimates were observed with GPX1 heterozygotes and homozygote variants, respectively, for all NHL (OR 1.24, 95% CI 1.04-1.48; OR 1.28, 95% 0.94-1.75), follicular lymphoma (OR 1.34; 95% CI 1.04-1.73; OR 1.29 95% CI 0.82-2.03) and diffuse large B-cell lymphoma (OR 1.24; 95% CI 0.97-1.58; OR 1.30 95% CI 0.85-1.99). In the UK study, data were further analysed to examine the risk associated with the well-defined NHL subtype MZL (n=54 cases) which is known to have an inflammatory origin. No associations were observed with respect to the CAT or GPX1 polymorphisms. However, for the SOD2 V16A polymorphism, differences in the distribution of the genotype groups between cases (25.5% Val/Val; 60.8% Val/Ala; 13.7% Ala/Ala) and controls (23.9% Val/Val; 49.1% Val/Ala; 27.0% Ala/Ala) were observed such that homozygosity for the alanine allele was associated with a non-significant decreased risk of MZL (OR 0.45, 95% CI 0.601.12). Stratification by sex for both the pooled and single study analyses revealed no differences in risk estimates. There was no evidence of gene-gene interactions for all NHL (data not shown). Discussion We examined associations between polymorphisms in genes involved in the oxidative stress response and risk of lymphoma in two population-based, case-control studies. Whereas our findings do not support a major | 1224 | haematologica/the hematology journal | 2006; 91(9) Oxidative stress genes and non-Hodgkin's lymphoma Table 2. Number of cases and controls, adjusted odds ratiosa and 95% confidence intervals by subtype of non-Hodgkin's lymphoma for SOD2 16VA, GPX1 P197L and CAT C-262T - UK and US pooled study data. Ferrata Storti Foundation Controls n (%) NHLb OR (CI) Cases n (%) DLBCL OR (CI) FL OR (CI) Total 1446 (100) 928 (100) - 381 (100) - 326 (100) - SOD2 16VA VV VA AA VV v VA/AA sna 358 (25) 713 (49) 371 (26) 1071 (74) 4 211 (24) 463 (51) 229 (25) 674 (75) 25 1.00 1.07 (0.87-1.33) 1.01 (0.79-1.29) 1.05 (0.86-1.28) 97 (26) 179 (49) 91 (25) 276 (75) 14 1.00 0.90 (0.68-1.19) 0.85 (0.61-1.18) 0.88 (0.68-1.15) 65 (20) 175 (55) 81 (25) 240 (75) 5 1.00 1.33 (0.97-1.82) 1.17 (0.81-1.67) 1.27 (0.94-1.72) CAT C-262T CC CT TT CC v CT/TT sna 867 (60) 498 (35) 72 (5) 570 (40) 9 554 (61) 298 (33) 57 (6) 355 (39) 19 1 0.93 (0.77-1.11) 1.22 (0.84-1.77) 0.97 (0.81-1.15) 218 (59) 128 (34) 25 (7) 153 (41) 10 1 1.02 (0.79-1.30) 1.35 (0.83-2.19) 1.06 (0.84-1.34) 205 (63) 104 (32) 16 (5) 120 (37) 3 1 0.86 (0.66-1.12) 0.82 (0.45-1.49) 0.86 (0.67-1.10) GPX1 P197L PP PL LL PP v PL/LL sna 773 (54) 551 (38) 120 (8) 671 (46) 2 453 (49) 387 (42) 81 (9) 468 (51) 7 1 1.24 (1.04-1.48)d 1.28 (0.94-1.75) 1.25 (1.05-1.48)e 186 (49) 158 (42) 33 (9) 191 (51) 4 1 1.24 (0.97-1.58) 1.30 (0.85-1.99) 1.25 (0.99-1.58) 156 (48) 142 (44) 28 (8) 170 (52) 0 1 1.34 (1.04-1.73)f 1.29 (0.82-2.03) 1.33 (1.04-1.70)g GPX1 P197L-UK PP PL LL PP v PL/LL sna 762 (100) 438 (58) 268 (35) 55 (7) 323 (42) 1 620 (100) 311 (51) 259 (42) 43 (7) 302 (49) 7 1 1.35 (1.07-1.69)h 1.12 (0.74-1.72) 1.31 (1.06-1.62)i 283 (100) 137 122 20 142 4 1 1.44 (1.08-1.92)j 1.17 (0.68-2.03) 1.40 (1.06-1.84)k 214 (100) 106 (50) 91 (42) 17 (8) 108 (50) 1 1.40 (1.02-1.93)l 1.35 (0.75-2.43) 1.39 (1.02-1.89)m GPX1 P197L-USA PP PL LL PP v PL/LL sna 684 (100) 335 (49) 283 (41) 65 (10) 348 (51) 1 308 (100) 142 (46) 128 (43) 38 (12) 166 (54) 0 1 1.08 (0.81-1.45) 1.42 (0.90-2.24) 1.14 (0.87-1.51) 98 (100) 49 (50.0) 36 (36.7) 13 (13.3) 49 (50.0) 0 1 0.87 (0.55-1.38) 1.42 (0.73-2.79) 0.97 (0.63-1.49) 112 (100) 50 (45) 51 (45) 11 (10) 62 (55) 0 1 1.21 (0.79-1.85) 1.16 (0.57-2.36) 1.20 (0.80-1.80) aOdds ratios (OR) and 95% confidence intervals (CI) were estimated using unconditional logistic regression and were adjusted for age and sex; bNHL: non-Hodgkin's lymphoma, DLBCL: diffuse large B-cell lymphoma, FL: follicular lymphoma, MZL: marginal zone lymphoma. sna: sample not amplified; dp=0.018; ep=0.010; fp=0.024; g p=0.020; hp=0.009; ip=0.014; jp=0.013; kp=0.017; lp=0.040; mp=0.035. role for GPX1, SOD2 or CAT in the etiology of NHL a modest increase in the risk for all NHL was observed with the GPX1 197Leu allele. Reduction in glutathione peroxidase activity has previously been reported in patients with lymphopoliferative disorders16 and more recently has been associated with poor prognosis in diffuse large B-cell lymphomas33 suggesting a possible role for this enzyme in the etiology and outcome of lymphoma. Manganese superoxide (MnSOD) expression has also been linked with a poor prognosis of diffuse large B-cell lymphoma.33 Although no association was observed between the SOD2 polymorphism and diffuse large Bcell lymphoma in the present study, Wang et al.32 recently reported an increased risk of this type of lymphoma in carriers of the SOD2 16Ala allele in a large US multicenter case control study. However, neither study found evidence of an association between SOD2 and all cases of NHL.32 In the UK population, fewer than expected cases of MZL were homozygous for the SOD2 16Ala allele such that the ala/ala genotype was associated with a decreased risk of MZL. The precise mechanism by which homozygosity for the alanine allele might result in a decreased risk for MZL is not clear. However, it may be a consequence of the improved defense against ROS conferred by the presence of alanine in the mitochondrial targeting sequence of SOD2. Indeed, it has been shown that the alanine form of superoxide dismutase is up to 40% more efficiently localized to the mitochondrial matrix, the major site for cellular metabolism and production of ROS, compared to the valine form.34 Furthermore, four-fold higher levels of superoxide dismutase protein expression and activity have been observed for the alanine variant.35 Although the NHL classification method used in the UK did not take into account of tumor site, it is likely that our population included a high proportion of gastric MZL given that gastric tissue is a more common site of MZL36 than the haematologica/the hematology journal | 2006; 91(9) | 1225 | T. J. Lightfoot et al. thyroid or salivary gland. Gastric MZL is associated with Helicobacter pylori infection which induces a chronic inflammatory response resulting in ROS and subsequent DNA damage. The lower than expected frequency of the ala/ala genotype observed in MZL is consistent with this variant having an impact on the response to infection due to its predicted increased ability to break down ROS. However, further clarification of the functional significance of the polymorphism is required since the effect on MnSOD expression is not clear.23 Although these findings were based on a limited number of cases (n=54), they are consistent with a previous observation of an association between inflammatory response genes and risk of MZL.36 In this earlier study polymorphisms in interleukin-1 (IL-1) and glutathione S-transferase (GST) genes, which regulate the inflammatory response to infection and exhibit anti-oxidant properties were analyzed in a series of patients with gastric MZL. The authors concluded that risk of gastric MZL was influenced by interindividual variation in inflammatory response and anti-oxidative capacity, particularly through combined effects of the IL-1 RN and GST T1 genotypes.37 The differential associations observed in this study between polymorphisms and NHL subtypes reflect the complexity of the disease. NHL is often referred to as a single entity when in fact it represents a group of independent lymphomas distinguished by histological subtypes that are likely to have differing etiologies. Thus, by analyzing data according to histological subtype, we may gain a greater understanding of the pathways involved in disease initiation and progression. We recently observed a similar effect by subtype for genetic polymorphisms in the folate metabolic pathway and risk of NHL subtypes.38,39 Our data provide some support for the hypothesis that polymorphisms that impair anti-oxidative capacity may influence NHL risk, with some aspects clearly warranting further investigation. In summary, further analyses of these and additional polymorphisms and haplotypes in GPX1 and SOD2 and other oxidative stress genes in large pooled studies that include well defined NHL subtypes will help to clarify their role in lymphomagenesis, and may provide a mechanism for the relationship between inflammation and NHL. In addition, evaluation of polymorphisms in oxidative stress response genes may provide useful screening and/or prognostic markers for lymphoma. TJL and CFS: this individual should be designated as joint author as they have contributed equally to the work. They were involved in the design of the experimental work, data acquisition, interpretation of data, drafting the article and critical revision of the article; AS: contributed to data acquisition, analysis and interpretation of the data, drafting the article and critically reviewing the article. MSF: contributed to data acquisition, interpretation of data, article content and revision; PJA: contributed to analysis of pooled data and UK data, data interpretation and article preparation and revision; GJM: contributed to the initiation of the UK study, experimental design, data interpretation, manuscript preparation and revision; PMB: contributed to the conception and design of the US study, acquisition of data, analysis of US data, interpretation of results, drafting article, critical revision of article. RE: contributed to initiation, design and conduct of the UK study, acquisition of data, interpretation of results, critical revision of article; MTS: contributed to conception and design of the US study, experimental design acquisition of data, interpretation of results, critical revision of article; EAH: ontributed to the conception and design of the US study, acquisition of data, interpretation of results, critical revision of article. This work was supported by the Leukaemia Research Fund of Great Britain, NIH grant numbers CA104862 (MTS), CA45614, CA89745, CA87014 (EAH), and by the National Foundation for Cancer Research (MT). We thank all consultants, hospital staff, general practitioners, and interviewees who participated in the study. The authors also declare that they have no potential conflict of interest. Manuscript received March 8, 2006. Accepted July 6, 2006. Ferrata Storti Foundation References 1. Tavani A, La Vecchia C, Franceschi S, Serraino D, Carbone A. Medical history and risk of Hodgkin's and nonHodgkin's lymphomas. Eur J Cancer Prev 2000;9:59-64. 2. Smedby KE, Hjalgrim H, Askling J, Chang ET, Gregersen H, PorwitMacDonald A, et al. Autoimmune and chronic inflammatory disorders and risk of non-Hodgkin lymphoma by subtype. J Natl Cancer Inst 2006;98:5160. 3. Baecklund E, Iliadou A, Askling J, Ekbom A, Backlin C, Granath F, et al. Association of chronic inflammation, not its treatment, with increased lymphoma risk in rheumatoid arthritis. Arthritis Rheum 2006;54:692-701. 4. Halliwell B, Gutteridge JM. Role of free radicals and catalytic metal ions in human disease: an overview. Methods Enzymol 1990;186:1-85. 5. Cerutti PA, Trump BF. Inflammation and oxidative stress in carcinogenesis. Cancer Cells 1991;3:1-7. 6. Klaunig JE, Kamendulis LM. The role of oxidative stress in carcinogenesis. Annu Rev Pharmacol Toxicol 2004; 44: 239-67. 7. Callan MF. Epstein-Barr virus, arthritis, and the development of lymphoma in arthritis patients. Curr Opin Rheumatol 2004;16:399-405. 8. Xu Y, Wiernik PH. Systemic lupus erythematosus and B-cell hematologic neoplasm. Lupus 2001;10:841-50. 9. Ramos-Casals M, De Vita S, Tzioufas AG. Hepatitis C virus, Sjogren's syndrome and B-cell lymphoma: linking infection, autoimmunity and cancer. Autoimmun Rev 2005;4:8-15. 10. Fridovich I. The biology of oxygen radicals. Science 1978;201:875-80. 11. Halliwell B. Free radicals, antioxidants, and human disease: curiosity, cause, or consequence? Lancet 1994;344:721-4. 12. Robinson BH. The role of manganese superoxide dismutase in health and disease. J Inherit Metab Dis 1998;21:598603. 13. Betteridge DJ. What is oxidative stress? Metabolism 2000;49 Suppl 1:3-8. 14. Gonzales R, Auclair C, Voisin E, Gau- tero H, Dhermy D, Boivin P. Superoxide dismutase, catalase, and glutathione peroxidase in red blood cells from patients with malignant diseases. Cancer Res 1984;44:4137-9. 15. Abdel-Aziz AF, El Naggar MM. Superoxide dismutase activities in serum and white blood cells of patients with some malignancies. Cancer Lett 1997;113:614. 16. Bewick M, Coutie W, Tudhope GR. Superoxide dismutase, glutathione peroxidase and catalase in the red cells of patients with malignant lymphoma. Br J Haematol 1987;65:347-50. 17. Wang LI, Miller DP, Sai Y, Liu G, Su L, Wain JC, et al. Manganese superoxide dismutase alanine-to-valine polymorphism at codon 16 and lung cancer risk. J Natl Cancer Inst 2001;93:1818-21. 18. Liu G, Zhou W, Wang LI, Park S, Miller DP, Xu LL, et al. MPO and SOD2 polymorphisms, gender, and the risk of non-small cell lung carcinoma. Cancer Lett 2004;214:69-79. | 1226 | haematologica/the hematology journal | 2006; 91(9) Oxidative stress genes and non-Hodgkin's lymphoma 19. Olson SH, Carlson MD, Ostrer H, Harlap S, Stone A, Winters M, et al. Genetic variants in SOD2, MPO, and NQO1, and risk of ovarian cancer. Gynecol Oncol 2004;93:615-20. 20. Ratnasinghe D, Tangrea JA, Andersen MR, Barrett MJ, Virtamo J, Taylor PR, et al. Glutathione peroxidase codon 198 polymorphism variant increases lung cancer risk. Cancer Res 2000; 60: 6381-3. 21. Ravn-Haren G, Olsen A, Tjonneland A, Dragsted LO, Nexo BA, Wallin H, et al. Associations between GPX1 Pro198Leu polymorphism, erythrocyte GPX activity, alcohol consumption and breast cancer risk in a prospective cohort study. Carcinogenesis 2006; 27:820-5. 22. Ahn J, Gammon MD, Santella RM, Gaudet MM, Britton JA, Teitelbaum SL, et al. Associations between breast cancer risk and the catalase genotype, fruit and vegetable consumption, and supplement use. Am J Epidemiol 2005; 162:943-52. 23. Rosenblum JS, Gilula NB, Lerner RA. On signal sequence polymorphisms and diseases of distribution. Proc Natl Acad Sci USA 1996;93:4471-3. 24. Hamanishi T, Furuta H, Kato H, Doi A, Tamai M, Shimomura H, et al. Functional variants in the glutathione peroxidase-1 (GPx-1) gene are associated with increased intima-media thickness of carotid arteries and risk of macrovascular diseases in Japanese type 2 diabetic patients. Diabetes 2004;53:2455-60. 25. Forsberg L, Lyrenas L, de Faire U, Morgenstern R. A common functional C-T substitution polymorphism in the promoter region of the human catalase gene influences transcription factor binding, reporter gene transcription and is correlated to blood catalase levels. Free Radic Biol Med 2001;30:5005. 26. Holly EA, Lele C, Bracci PM, McGrath MS. Case-control study of nonHodgkin's lymphoma among women and heterosexual men in the San Francisco Bay Area, California. Am J Epidemiol 1999;150:375-89. 27. Holly EA, Bracci PM. Population-based study of non-Hodgkin lymphoma, histology, and medical history among human immunodeficiency virus-negative participants in San Francisco. Am J Epidemiol 2003;158:316-27. 28. Skibola CF, Holly EA, Forrest MS, Hubbard A, Bracci PM, Skibola DR, et al. Body mass index, leptin and leptin receptor polymorphisms, and nonHodgkin's lymphoma. Cancer Epidemiol Biomarkers Prev 2004; 13:77986. 29. Willett EV, Smith AG, Dovey GJ, Morgan GJ, Parker J, Roman E. Tobacco and alcohol consumption and the risk of non-Hodgkin lymphoma. Cancer Causes Control 2004;15:77180. 30. Fritz A, Percy C, Jack A, Shanmugaratnan K, Sobin L, Parkin D, Whelan S. International Classification of Diseases for Oncology. 3rd ed. WHO Geneva, 2001. 31. StataCorp. Stata Statistical Software: Release 8.2. College Station, Texas: Stata Corporation, 2003. 32. Wang SS, Davis S, Cerhan JR, Hartge P, Severson RK, Cozen W, et al. Polymorphisms in oxidative stress genes and risk for non-Hodgkin lymphoma. Carcinogenesis 2006 [Epub ahead of print]. 33. Tome ME, Johnson DB, Rimsza LM, Roberts RA, Grogan TM, Miller TP, et al. A redox signature score identifies diffuse large B-cell lymphoma patients with a poor prognosis. Blood 2005; 106:3594-601. 34. Sutton A, Khoury H, Prip-Buus C, Cepanec C, Pessayre D, Degoul F. The Ala16Val genetic dimorphism modulates the import of human manganese superoxide dismutase into rat liver mitochondria. Pharmacogenetics 2003; 13:145-57. 35. Sutton A, Imbert A, Igoudjil A, Descatoire V, Cazanave S, Pessayre D et al. The manganese superoxide dismutase Ala16Val dimorphism modulates both mitochondrial import and mRNA stability. Pharmacogenet Genomics 2005;15:311-9. 36. Zucca E, Conconi A, Pedrinis E, Cortelazzo S, Motta T, Gospodarowicz MK et al. Nongastric marginal zone B-cell lymphoma of mucosaassociated lymphoid tissue. Blood 2003;101:2489-95. 37. Rollinson S, Levene AP, Mensah FK, Roddam PL, Allan JM, Diss TC et al. Gastric marginal zone lymphoma is associated with polymorphisms in genes involved in inflammatory response and anti-oxidative capacity. Blood 2003;102:1007-11. 38. Lightfoot TJ, Skibola CF, Willett EV, Skibola DR, Allan JM, Coppede F, et al. Risk of non-Hodgkin lymphoma associated with polymorphisms in folate metabolising genes. Cancer Epidemiol Biomarkers Prev 2005;14:2999-3003. 39. Skibola CF, Forrest MS, Coppede F, Agana L, Hubbard A, Smith MT, et al. Polymorphisms and haplotypes in folate-metabolizing genes and risk of non-Hodgkin lymphoma. Blood 2004; 104:2155-62. Ferrata Storti Foundation haematologica/the hematology journal | 2006; 91(9) | 1227 |