SHORT REPORT Atopy, exposure to pesticides and risk of non-Hodgkin lymphoma Claire M. Vajdic 1 * , Lin Fritschi 2 , Andrew E. Grulich 1 , John M. Kaldor 1 , Geza Benke 3 , Anne Kricker 4 , Ann Maree Hughes 5 , Jennifer J. Turner 6 , Sam Milliken 7 , Chris Goumas 4 and Bruce K. Armstrong 4,8 1 National Centre in HIV Epidemiology and Clinical Research, University of NSW, Darlinghurst, NSW, Australia 2 Laboratory for Cancer Epidemiology, Western Australian Institute for Medical Research, Perth, WA, Australia 3 Department of Epidemiology and Preventative Medicine, Monash University, Melbourne, VIC, Australia 4 School of Public Health, The University of Sydney, Sydney, NSW, Australia 5 Epidemiology and Genetics Unit, University of York, York, United Kingdom 6 Department of Anatomical Pathology, St. Vincent’s Hospital, Sydney, NSW, Australia 7 Department of Haematology, St. Vincent’s Hospital, Sydney, NSW, Australia 8 Sydney Cancer Centre, Royal Prince Alfred Hospital, Sydney, NSW, Australia Pesticide exposure has been associated with non-Hodgkin lym- phoma (NHL) risk in a number of studies, and two recent studies suggest that the increased risk may be confined to those with a history of asthma. We examined the interaction between occupa- tional pesticide exposure and atopy on risk of NHL in an Austra- lian population-based case–control study. Incident cases (n 5 694) were diagnosed in New South Wales or the Australian Capital Territory between 2000 and 2001 and controls (n 5 694) were randomly selected from electoral rolls and frequency-matched to cases by age, sex and State of residence. Occupational pesticide exposure was determined by an expert occupational hygienist’s assessment of job-specific questionnaires administered by tele- phone. History of atopy (asthma, hay fever, eczema and food allergy) was self-reported. Logistic regression models included the three matching variables, ethnicity and sun exposure. The OR for NHL with substantial pesticide exposure and any history of asthma was 3.07 (95% CI 0.55–17.10) and with substantial pesti- cide exposure and no asthma history it was 4.23 (95% CI 1.76– 10.16). The p-value for interaction was 0.29. A similar pattern of risk was observed for each of the pesticide subtypes; for asthma at various times of life; for hay fever, eczema, food allergy and any atopy, in men only and for follicular lymphomas only. Although this study had limited power, the findings do not suggest modifica- tion of the association between pesticide exposure and NHL risk by asthma or atopic disease more generally. ' 2007 Wiley-Liss, Inc. Key words: non-Hodgkin lymphoma; risk; case–control; pesticides; asthma; atopy Occupational exposure to pesticides and a personal history of atopy have been widely examined as risk factors for non-Hodg- kin lymphoma (NHL), a neoplasm arising from cells of the immune system. These studies have typically found that expo- sure to pesticides increases risk of NHL, 1 while asthma, or atopy more generally, has been inconsistently protective. 2 Since both pesticide exposure and asthma may influence NHL risk by mod- ulating the immune system, 2 US studies of NHL examined the interaction between the 2 exposures; they found that risk of NHL with pesticide exposure was higher in asthmatics than in non-asthmatics. 3,4 Although there is no biological evidence that asthma modi- fies the effect of pesticide exposure on NHL risk, given the potential importance of such a relationship, we sought to repli- cate it in our case–control study of NHL in Australia. We have previously reported an odds ratio of 3.09 (95% CI 1.42–6.70) for NHL with substantial occupational pesticide exposure 5 and apparently protective effects of having asthma (OR 0.88, 95% CI 0.67–1.17), eczema (OR 0.72, 95% CI 0.57–1.10) or hay fever (OR 0.65, 95% CI 0.52–0.82) as an adult, or for being al- lergic to one or more foods at any time of life (OR 0.29, 95% CI 0.20–0.42). 6 Material and methods Detailed methods for this study have been described previously. 5–7 Ethical approval was obtained from each participating institution. Participants Cases were patients aged 20–74 years with incident NHL diag- nosed between 2000 and 2001 and resident in New South Wales or the Australian Capital Territory. They were considered ineli- gible if they had poor English language skills, if they were unable to complete a telephone interview or if they had a history of trans- plantation or HIV infection or a diagnosis of chronic lymphocytic leukaemia, plasma cell myeloma or B- or T-lymphoblastic leukae- mia. An anatomical pathologist reviewed pathology reports for all consenting cases and diagnostic histopathology sections for all uncertain cases to provide assurance of the correct diagnosis. Con- trols were randomly selected from electoral rolls to approximately match the expected distributions of cases with respect to age, sex and State or Territory of residence. Electoral registration is com- pulsory for adult Australian citizens. Controls were considered ineligible if they had poor English language skills or if they were unable to complete a telephone interview. HIV status was not ascertained for controls; few Australians are HIV positive. Of the 1,230 cases ascertained, the study enrolled a total of 694, or 78% of the 842 apparently eligible and contactable. Of the 1,687 controls ascertained, a total of 694 were enrolled, 61% of the 1,134 apparently eligible and contactable. 5,7 Data collection Cases and controls received an introductory letter and an infor- mation leaflet, and on consenting, completed a self-administered questionnaire that included the following information for each job held for 1 year or more: job title, employer, industry, first and last years, working hours per day and days per week. Those who reported a job with potential exposure to pesticides were asked for details in a subsequent computer-assisted telephone interview (CATI). The CATI questionnaire used job-specific modules adapted from the US National Cancer Institute (see Ref. 5). If her- bicides, insecticides or fungicides were mixed or applied, the Grant sponsors: National Health and Medical Research Council of Aus- tralia (NHMRC), Grant number 990920; The Cancer Council NSW; The University of Sydney Medical Foundation. *Correspondence to: National Centre in HIV Epidemiology and Clini- cal Research, University of New South Wales, Level 2/376 Victoria Street, Darlinghurst, NSW 2010, Australia. Fax: 1612-9385-0920. E-mail: cvajdic@nchecr.unsw.edu.au Received 22 October 2006; Accepted after revision 7 December 2006 DOI 10.1002/ijc.22602 Published online 8 February 2007 in Wiley InterScience (www.interscience. wiley.com). Int. J. Cancer: 120, 2271–2274 (2007) ' 2007 Wiley-Liss, Inc. Publication of the International Union Against Cancer