Childhood leukaemia and infectious exposure: A report from the United Kingdom Childhood Cancer Study (UKCCS) Jill Simpson*, Alex Smith, Pat Ansell, Eve Roman, on behalf of the United Kingdom Childhood, Cancer Study Investigators a Epidemiology and Genetics Unit, Department of Health Sciences, University of York, YO10 5DD, UK ARTICLE INFO Article history: Received 26 June 2007 Received in revised form 24 July 2007 Accepted 31 July 2007 Available online 7 September 2007 Keywords: Childhood leukaemia Infection Case-control Birth order Social activity Deprivation Maternal recall Primary care Clinical records ABSTRACT Data from a national case-control study are used to explore the relationships between childhood leukaemia, infant infection and three markers of infectious exposure – birth order, infant-activity group attendance and area-based deprivation. Amongst controls, clin- ically diagnosed infection in the first year varied little with birth order and infant-activity group attendance – with 4 in 5 children having at least one infection, and each child aver- aging around 2.9 (2.8–3.0). Amongst cases of acute lymphoblastic leukaemia (ALL), the lev- els of infection increased as the indices of infectious exposure increased – for example, odds ratios associated with at least one infection in the first year being 0.9 (95% confidence interval (CI): 0.6–1.4) for birth order one and 1.6 (95% CI: 1.1–2.2) for birth order two or more. By contrast, interview data were misleading, with mothers – particularly case mothers – consistently under-reporting. We conclude that the findings based on clinical data, combined with the markers of infectious exposure, confirm the observation that immune dysregulation among children who develop ALL is detectable from an early age. Ó 2007 Elsevier Ltd. All rights reserved. 1. Introduction The potential aetiological role of infection in the development of childhood leukaemia has been the subject of many epide- miological studies and reports. To date, however, no specific agents have been identified and the mechanism by which infectious exposures might influence subsequent leukaemia risk remains a much debated topic. 1,2 A key reason for this uncertainty is the lack of consistency between study findings, and the present paper explores some of the reasons why this may occur. Unravelling the relationship between disease risk and pre- vious infectious exposure is not straightforward. With a view to quantifying children’s likely exposure to infectious agents at various time-points, a wide range of proxies have been em- ployed including family measures of socio-economic status and residential location 3,4 ; parental indicators of social con- tact outside the home 1,5 ; markers of the child’s social activity such as birth order 6–8 and pre-school group attendance 9–11 ; as well as infectious illness histories of both the child 8,12–17 and their mothers. 17–22 In order to investigate the relationship between leukaemia and infection as comprehensively as possible, the United Kingdom Childhood Cancer Study (UKCCS) collected informa- tion on a number of markers of infectious exposure from multiple sources (www.ukccs.org). A unique facet of the UKCCS is that, in addition to asking mothers about their child’s health, systematic abstractions of primary-health care 0959-8049/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.ejca.2007.07.027 * Corresponding author: Tel.: +44 (0) 1904 321891; fax: +44 (0) 1904 321899. E-mail address: jill.simpson@egu.york.ac.uk (J. Simpson). a See Acknowledgements for Regional Investigators, and Management Committee. EUROPEAN JOURNAL OF CANCER 43 (2007) 2396 – 2403 available at www.sciencedirect.com journal homepage: www.ejconline.com