O1A.2 Exposure to benzimidazole fungicides in agriculture and non-hodgkin lymphomas, overall and by subtypes, in the AGRIculture and CANcer (AGRICAN) cohort

PurposeIncreased risks of Non-Hodgkin’s Lymphomas (NHLs) have been reported among farmers in several meta-analyses, with exposure to pesticides as the main explanation. Few studies investigated associations by NHL subtypes and considering specific pesticides. From the French agricultural cohort AGRICAN, we assessed the role of benzimidazole fungicides use on NHL risk, overall and for multiple myeloma (MM), Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma (CLL-SLL) and Diffuse Large B-cell Lymphoma (DLBCL).MethodsNearly 1 82 000 participants affiliated in a health assurance system reported in the enrollment questionnaire (2005–2007) lifetime pesticide use with start and end year on 13 crops and were followed until 2013. Using the crop-exposure matrix PESTIMAT, associations between NHLs and benzimidazole fungicides exposure (overall and for benomyl, carbendazim, fenzaflore, thiabendazole and thiophanate methyl) were estimated using Cox models.ResultsAfter exclusion of prevalent cases, individuals with incomplete agricultural profession history data or a zero tracking period, 1,133 NHL incident cases were identified from cancer registries (269 MM, 244 CLL-SLL, 190 DLBCL). Nearly 20% of participants were considered exposed (median duration from 7 to 20 years according to active ingredient). Increased NHL risk was observed with exposure to benzimidazole, overall, on any crops (NHL: HR=1.13, 95%CI=0.94–1.37, 150 cases, no duration relationship). Moreover, borderline positive associations were reported on specific crops: wheat/barley (NHL: HR=1.23, CLL-SLL: HR=1.42), beets (DLBCL, HR=2.19) and rape (DLBCL, 2.32). Significant increased risks were reported with MM for use of thiophanate-methyl on wheat/barley (HR=3.46, 23 exposed cases, no duration relationship) and with DLBCL for all 4 benzimidazole used on beets (HR from 2.34 to 2.57 and for the 2 used on rape (HR=2.39 and 2.46), although based on respectively 7 and 6 cases.ConclusionThese findings suggest positive associations between incidence of specific NHL subtypes, and exposure to benzimidazole as a chemical family or specific ingredients in this family.