Association between residential proximity to crop areas and childhood acute leukemia risk in mainland France - GEOCAP case-control study.

[BACKGROUND] Pesticide exposures are suspected to be a risk factor for several childhood cancers, particularly acute leukemia (AL) and the most common type, acute lymphoblastic leukemia (ALL). The role of environmental exposures due to agricultural uses of pesticides remains debated.

[OBJECTIVE] The study aimed to investigate whether residential proximity to different crops, used as a proxy for pesticide exposures, is associated with an increased risk of AL in France.

[METHODS] We included 2731 AL cases diagnosed in 2008-2013 and 28,303 controls representative of the contemporary French pediatric population (1-14 years old), drawn from the national registry-based GEOCAP study. National annual maps of agricultural land use were developed to assess proximity to 13 crop types. Multivariate logistic regression models adjusted for age were used to evaluate the association with the densities of crops within 1000 m of children's addresses. A hierarchical classification was also used to categorize children according to their agricultural neighborhood profiles.

[RESULTS] The majority of children (69 %) lived less than 1000 m from crops at time of diagnosis/inclusion. Barley and viticulture densities were positively associated with ALL: Odds ratio (OR) = 1.05 (1.00-1.10) and 1.06 (0.99-1.13) for a density increase of 3 % and 10 %, respectively. Two agricultural neighborhood profiles were also associated with an increased risk of ALL: 'Vines (large areas)', OR = 1.35 (1.03-1.78) and 'Wheat with diverse crops', OR = 1.28 (1.10-1.49). The results remained stable across sensitivity analyses.

[CONCLUSION] Our results support the hypothesis of a role of agricultural pesticide exposures in childhood ALL risk. Future research should focus on the chemical substances applied to crops.