Introduction Even though some risk factors for breast cancer are well-established (reproductive history or lifestyle), they explain no more than 50% of cases. Occupational exposures, including farming, have been seldom studied and were based on small-scale studies. Our aim was to assess the association between farming activities and breast cancer in the AGRICAN cohort.

Methods AGRICAN consisted of 181,842 participants affiliated to the French agricultural health insurance. Data on lifetime agricultural exposures (18 farming activities, up to 5 tasks) and living in a farm during first year of life were collected from the enrollment questionnaire (2005–2007). Hazard ratios were estimated using Cox regression analysis with age as time scale.

Results From enrolment to 2011, 743 incident breast cancers were identified among 55,558 postmenopausal women (555 ductal and 99 lobular carcinoma) through linkage with cancer registries. We found an overall lower breast cancer risk among farmers (HR 0.81, 95% CI: 0.65–1.01) and cattle breeders, with a linear inverse relationship with duration (≥40 years HR 0.71, 95% CI: 0.52–0.95; p-trend < 0.05), mainly for those performing care, milking or insecticide treatment. Peas growers had a significant increased risk (HR 1.52), mainly those involved for at least 30 years and those performing harvest. Elevated risks were also suggested for women involved on vegetables, beets and greenhouses. No association was observed with pesticide use but elevated risks were suggested for re-entry tasks or harvesting in potato, tobacco or fruit growing. Living in a farm during first year of life when the main activity was market gardening also increased risk (HR 1.44). Results remained unchanged with adjustment for reproductive history and lifestyle.

Conclusion This work provides new results on associations between farming activities and breast cancer among postmenopausal women. It also emphasises the need to consider pesticide exposures during other tasks than application.