Dietary intake of polycyclic aromatic hydrocarbons (PAHs) and breast cancer risk: Evidence from the French E3N-Generations prospective cohort
Amina Amadou, Delphine Praud, Chloé Marques, Hwayoung Noh, Pauline Frénoy, Arnaud Vigneron, Thomas Coudon, Floriane Deygas, Gianluca Severi, Béatrice Fervers, Francesca Romana Mancini
Abstract
• This is the largest study of the effect of PAH dietary intake on breast cancer risk. • PAH dietary intake was assessed through a validated food questionnaire. • Findings support that PAH dietary intake was associated with breast cancer risk. • PAH dietary intake was positively associated with risk of ER-PR- breast cancer. • Breast cancer risk was positively associated with BaP dietary intake. While there is compelling evidence of the association between occupational exposure to polycyclic aromatic hydrocarbons (PAHs) and risk of breast cancer (BC), findings on PAH dietary exposure are less consistent. The present study aims to evaluate the association between PAH dietary intake and BC risk. The study included 67,879 women who completed a validated semi-quantitative dietary questionnaire (208 food items) from the E3N-Generations cohort study. PAH dietary intake was estimated by combining E3N food consumption data with food contamination levels obtained from the second French total diet study (TDS2). Cox regression was used to estimate adjusted hazard ratios (HRs) and 95 % confidence intervals (CIs) for the association between PAH dietary intake (sum of four PAHs (PAH 4 ) namely benzo[a]pyrene (BaP), chrysene (CHR), benzo[a]anthracene (BaA) and benzo[b]fluoranthene (BbF)) and BC risk. Additionally, BaP, a surrogate for total PAHs, was investigated as the second exposure variable. After an average follow-up of 17.6 years, 5,686 incident BC were diagnosed. Overall, the estimated HRs for the associations between each quintile of PAH 4 and BC risk, taking the first quintile as reference, were all greater than 1, but were statistically significant only for the third quintile (HR Q3 vs Q1 = 1.10; CI: 1.01–1.20). By estrogen (ER) and progesterone (PR) hormone receptor status, we observed a positive association between PAH 4 dietary intake and ER-PR- BC (HR Q4 vs Q1 = 1.34; CI: 1.01–1.76). Moreover, there was a borderline positive association with BaP, for the second (HR Q2 vs Q1 = 1.08; CI: 0.99–1.17) and third (HR Q3 vs Q1 = 1.07; CI: 0.98–1.16) quintiles. This study supports a relationship between PAH 4 dietary intake and BC risk, notably with a non-linear trend. A positive but marginal association was observed between BaP dietary intake and BC risk. Abbreviations: BC, breast cancer; PAHs, polycyclic aromatic hydrocarbons; BaA, benzo[a]anthracene; BaP, benzo[a]pyrene; BbF, benzo[b]fluoranthene; CHR, chrysene; TDS2, second French total diet study; IARC, International Agency for Research on Cancer; EPA, Environmental Protection Agency; CrI, credible intervals; CIs, confidence intervals; ORs, odds ratios; HR, hazard ratios; SD, standard deviation; E3N, Etude Epidémiologique auprès des femmes de la Mutuelle Générale de l’Education Nationale; MGEN, Mutuelle Générale de l’Education Nationale; ER, estrogen receptor; PR, progesterone receptor; TNM, tumor-node-metastasis; CIQUAL, French Information Centre on Food Quality; LB, lower bound; MB, middle bound; LOD, limit of detection; FHBC, family history of BC history; AFFP, age at first full-term pregnancy; BMI, body mass index; MHT, menopausal hormone replacement therapy; CNIL, Commission for data protection and privacy.