Breast cancer, the first female cancer, affects one in eight women in her lifetime. The increase and unequal distribution of its prevalence throughout the world, regardless of age and genetic factors (< 10% of cases), observations in migrant women and the increased risk in women who have been exposed in utero to diethylstilbestrol, suggest the involvement of environmental factors that can act very early in development, such as persistent chemical pollutants (POPs) that are endocrine disrupting (EP) acting via nuclear receptors. But the demonstration of the deleterious role of such exposure to chemical pollutants is confronted epidemiologically with methodological difficulties: the correlation is most often sought at the time of cancer discovery when the critical windows of exposure are for the breast, fetus, perinatal, peri-advertising or pregnant and a single pollutant is usually measured, whereas they can be potentiated ("cocktail effect").

The objective of this project is therefore to assess the risk related to pre- and per-gravidic exposure to a range of POPs families (dioxins, dioxin-like, PolyChlorinated Bysphenyls PCBs, flame-retardant polybrominated compounds, waterproofing perfluorinated compounds and several organochlorine pesticides), to develop breast cancer within 15 years of delivery, taking into account the conventional risk factors for breast cancer.

This project benefits from a historical bio-bank of 6242 cord blood, an indirect reflection of the pre and per-gravidic maternal exposure, a bank set up between 2002 and 2005, during a PHRC at the Nice University Hospital, frozen and stored under strict and regulatory conditions, declared to the CNIL, and the Cancer Observatory / CRISAP of the PACA-Corsica region, with an exhaustive register of over 92% since 2005, and containing the main characteristics of cancer. The cross-referencing of these two registers (biobanks and CRISAP), supported by the preliminary feasibility study (155 cases of cancer expected, 35% of controls lost to follow-up or opposed to participation), makes it possible to consider a prospective case-control study nested in the cohort of mothers included in the bio-bank.

It will therefore be investigated whether women parturient women of this period who had the highest levels of POPs cord blood between 2002 and 2005 had a higher risk of developing breast cancer than those who had the lowest levels, taking into account other known risk factors.

The assays will be carried out by LABERCA in Nantes, the national reference laboratory, using gas chromatography/high-resolution mass spectrometry coupling. Blood lipid concentrations will be described in quartile and analyzed in continuous values, alone or in combination with a cumulative score, in N=140 cases and 2N=280 controls, matched for age and parity at birth by random drawing, sample required for 80% power, risk 5% OR at 2, frequency of exposure>70%.

The investigators propose to develop a predictive model of breast cancer occurrence based on exposure to endocrine disrupters measured during pregnancy by adjusting for known breast cancer risk factors.

These conditions should provide information on the possible association of exposure to POPs present in the domestic environment during critical pre- and per-gravidic periods and the risk of breast cancer and reinforce the relevance of preventive measures recently recommended during pre-conceptional and/or pre-natal consultations.

Inclusion Criteria:

• women who have given birth 2002/2005
• collection of cord blood during childbirth
• woman who developed invasive or non-invasive breast cancer in the following years (2005- O6/2018)

Exclusion Criteria:

• 0

women who have given birth 2002/2005