Doctor: Astrid Coste
Title: The role of environmental exposure to natural ultraviolet radiation and agricultural pesticides in the incidence of haematological malignancies in children
Supervisor: Denis Hemon
Doctoral school: ED 420. Doctoral school of Public health, University Paris Saclay
Date of thesis defense: 10/2017
Jury: Denis Hémon, Alexis Elbaz, Cécile Chevrier, Alain Monnereau, Agnès Lefranc, Stéphanie Goujon
This thesis deals with the role of two environmental exposures not much studied and that could have an impact on the incidence of childhood hematological malignancies (CHM): ultraviolet radiation (UV) and agricultural pesticides.The two major diagnostic groups are acute leukemia (AL) and lymphomas and represent respectively around 470 and 200 new cases per year in France. Despite the progress made in improving therapeutic caring and survival, the etiology of these cancers remains largely unknown.There are very few studies on the association between UV and these cancers. Meta-analyses found a coherent association between childhood AL and parental professional or domestic pesticides exposure during pregnancy. However the association with residential exposure to agricultural pesticides has been less studied and results are heterogeneous.The first, ecological, part of the thesis addressed the associations between residential UV exposure at diagnosis and the incidence of types and subtypes of CHM in France. The 9,082 cases of acute leukemia (AL) and 3,563 cases of lymphoma diagnosed before the age of 15 years from 1990 to 2009 were provided by the French National Registry of Childhood Hematological Malignancies. UV data from 1988 to 2007 were extracted from the EUROSUN database. The annual daily average UV exposure of the children estimated at the municipalities of residence was considered. There was a significant increase in precursor B-cell acute lymphoblastic leukemia (PBC-ALL) in children aged less than 5 years with residential exposure to UV. The results remained stable in analyses stratifying by deprivation index or degree of urbanization of the municipalities.A second, individual, study found no change in the association between UV and PBC-ALL after taking into account the influence of suspected individual risk factors for ALL, and evaluated this association at birth. Individual data were collected by interviews in the ESCALE (2003-2004) and ESTELLE (2010-2011) nationwide case-control studies.The last part of this work focused on the residential exposure to agricultural pesticides. The 10,994 cases of AL and the 4,301 cases of lymphomas diagnosed among children less than 15 years old were obtained from the French National Registry of Childhood Hematological Malignancies over the period 1990-2013. Intensity of agricultural activities by Canton was used as a proxy of residential agricultural pesticides exposure. This proxy was estimated from the 2000 French agricultural census data. At this first step of the analysis, no associations were found between total agricultural area and incidence of CHM. Analyses by types of crops showed, at this first step, a significant association between oilseeds and PBC-ALL and Burkitt lymphoma. Sensitivity analyses evidenced heterogeneous results by time period.