Institutional members access full text with Ovid®

Share this article on:

Spatial variations of childhood acute leukaemia in France, 1990–2006: global spatial heterogeneity and cluster detection at ‘living-zone’ level

Demoury, Clairea,b; Goujon-Bellec, Stéphaniea,b,c; Guyot-Goubin, Auréliea,b,c; Hémon, Denisa,b; Clavel, Jacquelinea,b,c

European Journal of Cancer Prevention: July 2012 - Volume 21 - Issue 4 - p 367–374
doi: 10.1097/CEJ.0b013e32834e31d8
Research Papers: Leukemia

Childhood acute leukaemia (AL) accounts for a third of childhood cancers. Analysing the spatial distribution of the incidence of AL and its lymphoblastic and myeloblastic subtypes may contribute to the identification of risk factors. This national registry-based study aimed to evaluate global spatial heterogeneity in the incidence rates of AL and to detect clusters in France over the period 1990–2006 on the ‘living-zone’ scale. Between 1990 and 2006, 7675 cases of AL were registered in the National Registry of Childhood Haematopoietic malignancies. Their spatial distribution in the 1895 ‘living zone’ was first evaluated with two tests for global spatial heterogeneity (Potthoff–Witthinghill and Rogerson's tests) and then with the SaTScan and FleXScan methods, which aim to locate spatial and space–time clusters. Over 1990–2006, no spatial heterogeneity of AL or its subtypes was evidenced. In addition, none of the most likely clusters identified with SaTScan and FleXScan over the whole period was significant, and the systematic search for space–time clusters yielded nonsignificant results. However, when three subperiods were considered, five statistically significant nonoverlapping spatial clusters were identified. This study did not find evidence of any global spatial heterogeneity of AL incidence rates in France over the period 1990–2006. Although no significant spatial cluster was detected over the whole period, the study identified a few significant spatial clusters in specific periods. Even though the significance levels of those clusters do not strongly support the existence of local risk factors, the clusters may still reflect a slight impact of shared risk factors, including background environmental exposures, which require further investigation.

aINSERM, CESP Environmental Epidemiology of Cancer

bUniversity of Paris-Sud 11, UMRS

cFrench National Registry of Childhood Haematopoietic malignancies (NRCH), Villejuif, France

Correspondence to Jacqueline Clavel, PhD, Research Center in Epidemiology and Population Health (CESP), Environmental Epidemiology of Cancer UMRS-1018, Group 6 INSERM, 16 Avenue Paul Vaillant-Couturier, F-94807 Villejuif Cedex, France Tel: +33 145 595 037; fax: +33 145 595 151; e-mail:

Received June 28, 2011

Accepted October 14, 2011

© 2012 Lippincott Williams & Wilkins, Inc.