Perfluorinated alkylated substances serum concentration and breast cancer risk: Evidence from a nested case‐control study in the French E3N cohort

Abstract

Endocrine‐disrupting chemicals are proposed to increase breast cancer (BC) incidence. Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA), two perfluorinated alkylated substances (PFASs), are suspected to be ubiquitously present in the blood of human population worldwide. We investigated the associations between serum concentrations of these substances and BC risk. Etude Epidémiologique auprès de femmes de l'Education Nationale is a cohort of 98,995 French women born in 1925–1950 and followed up since 1990. We sampled 194 BC cases and 194 controls from women with available blood samples. Serum concentrations of PFASs were measured by liquid chromatography coupled to tandem mass spectrometry (LC–MS/MS). Adjusted conditional logistic regression models were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs). All statistical tests were two sided. While PFASs concentrations were not associated with BC risk overall, we found positively linear associations between PFOS concentrations and the risk of ER+ (3rd quartile: OR = 2.22 [CI = 1.05–4.69]; 4th quartile: OR = 2.33 [CI = 1.11–4.90]); Ptrend = 0.04) and PR+ tumors (3rd quartile: OR = 2.47 [CI = 1.07–5.65]; 4th quartile: OR = 2.76 [CI = 1.21–6.30]; Ptrend = 0.02). When considering receptor‐negative tumors, only the 2nd quartile of PFOS was associated with risk (ER−: OR = 15.40 [CI = 1.84–129.19]; PR−: OR = 3.47 [CI = 1.29–9.15]). While there was no association between PFOA and receptor‐positive BC risk, the 2nd quartile of PFOA was positively associated with the risk of receptor‐negative tumors (ER−: OR = 7.73 [CI = 1.46–41.08]; PR−: OR = 3.44 [CI = 1.30–9.10]). PFAS circulating levels were differentially associated with BC risk. While PFOS concentration was linearly associated with receptor‐positive tumors, only low concentrations of PFOS and PFOA were associated with receptor‐negative tumors. Our findings highlight the importance of considering exposure to PFASs as a potential risk factor for BC.

Serum Levels of Commonly Detected Persistent Organic Pollutants and Per- and Polyfluoroalkyl Substances (PFASs) and Mammographic Density in Postmenopausal Women

“…This may have contributed to the conflicting findings. In particular, the median level of PFOS in Inuits was 18.2 ng/mL, similar to the level in the E3N study (17.3 ng/mL) [21]. Effect modification by polymorphisms in genes involved in estrogen metabolism such as CYP17A1 -34T/C (rs743572) and COMT Val158Met (rs4680) has been proposed [42,43]; however, the modifying effects of these polymorphisms were not consistent across PFAS compounds or across studies, and frequencies of alleles potentially associated with increased susceptibility to PFASs among Inuits [42] were only slightly higher (rs4680 Met allele) or even lower (rs743572 T allele) in Inuits than in European women [42,44].…”

“…Another prospective study nested in the French E3N (Etude Epidémiologique auprès de femmes de la Mutuelle Générale de l'Education Nationale) cohort investigated two compounds, PFOS and PFOA, and did not observe significant associations with overall breast cancer; however, there was a significant positive linear association between one of the compounds (PFOS) and the risk of estrogen receptor-positive subtype of breast cancer [21]. PFOS levels in the E3N study were much higher than the levels in the CTS (median levels among controls: 17.3 ng/mL and 6.95 ng/mL, respectively) [21]. PFOA was also investigated in two studies in a contaminated area in North Carolina using an ecological study design [19] or mathematically estimated PFOA levels [18], both reporting no associations [18,19].…”

“…The only published prospective study of diverse classes of PFASs (16 PFASs) used serum samples collected during pregnancy in Danish women and also reported no associations or non-significant inverse associations between those PFASs and premenopausal breast cancer risk, except for a weak positive association observed for perfluorooctane-sulfonamide (PFOSA) [17]. Another prospective study nested in the French E3N (Etude Epidémiologique auprès de femmes de la Mutuelle Générale de l'Education Nationale) cohort investigated two compounds, PFOS and PFOA, and did not observe significant associations with overall breast cancer; however, there was a significant positive linear association between one of the compounds (PFOS) and the risk of estrogen receptor-positive subtype of breast cancer [21]. PFOS levels in the E3N study were much higher than the levels in the CTS (median levels among controls: 17.3 ng/mL and 6.95 ng/mL, respectively) [21].…”

“…Data on other classes of POPs such as polybrominated diphenyl ethers (PBDEs) and per-and poly-fluoroalkyl substances (PFASs) remain scarce and inconsistent [12][13][14][15][16][17][18][19][20][21][22]. Five studies using post-diagnostic samples have investigated serum or adipose-level PBDEs in relation to breast cancer risk [12,[14][15][16]22].…”

“…Whereas positive associations were reported for a few PBDE congeners such as BDE-47, BDE-100, and BDE-153 in an area in China highly contaminated from electronic waste recycling [12], and suggestive positive associations were reported for BDE-47 in Alaska Native women [15] and young Canadian women [22], two studies conducted among general populations in California [14,16] reported no associations. Results on PFASs have also been mixed, with studies reporting positive associations for five PFAS compounds among Greenland Inuits [20], weak positive association for a PFAS compound perfluorooctane-sulfonamide (PFOSA) and non-significant inverse association for another PFAS compound among Danish women [17], and a positive association with estrogen receptor-positive breast cancer for perfluorooctane sulfonate (PFOS) in French women [21]; in addition, no associations were found for perfluorooctanoic acid (PFOA) in two studies conducted in an area near a chemical plant [18,19], nor for six PFAS compounds commonly detected in the general population in California [13]. Little is known about the role of these chemicals with respect to pathological changes along the cancer development pathway or cancer treatment.…”