Multiple xenoestrogen air pollutants and breast cancer risk: Statistical approaches to investigate combined exposures effect

Amina Amadou, Camille Giampiccolo, Fabiola Bibi Ngaleu, Delphine Praud, Thomas Coudon, Lény Grassot, Elodie Faure, Florian Couvidat, Pauline Frenoy, Gianluca Severi, Francesca Romana Mancini, Pascal Roy, Béatrice Fervers

Research output: Contribution to journalArticlepeer-review

Abstract

Studies suggested that exposure to air pollutants, with endocrine disrupting (ED) properties, have a key role in breast cancer (BC) development. Although the population is exposed simultaneously to a mixture of multiple pollutants and ED pollutants may act via common biological mechanisms leading to synergic effects, epidemiological studies generally evaluate the effect of each pollutant separately. We aimed to assess the complex effect of exposure to a mixture of four xenoestrogen air pollutants (benzo-[a]-pyrene (BaP), cadmium, dioxin (2,3,7,8-Tétrachlorodibenzo-p-dioxin TCDD)), and polychlorinated biphenyl 153 (PCB153)) on the risk of BC, using three recent statistical methods, namely weighted quantile sum (WQS), quantile g-computation (QGC) and Bayesian kernel machine regression (BKMR). The study was conducted on 5222 cases and 5222 matched controls nested within the French prospective E3N cohort initiated in 1990. Annual average exposure estimates to the pollutants were assessed using a chemistry transport model, at the participants’ residence address between 1990 and 2011. We found a positive association between the WQS index of the joint effect and the risk of overall BC (adjusted odds ratio (OR) = 1.10, 95% confidence intervals (CI): 1.03–1.19). Similar results were found for QGC (OR = 1.11, 95%CI: 1.03–1.19). Despite the association did not reach statistical significance in the BKMR model, we observed an increasing trend between the joint effect of the four pollutants and the risk of BC, when fixing other chemicals at their median concentrations. BaP, cadmium and PCB153 also showed positive trends in the multi-pollutant mixture, while dioxin showed a modest inverse trend. Despite we found a clear evidence of a positive association between the joint exposure to pollutants and BC risk only from WQS and QGC regression, we observed a similar suggestive trend using BKMR. This study makes a major contribution to the understanding of the joint effects of air pollution.

Original languageEnglish
Article number124043
JournalEnvironmental Pollution
Volume351
DOIs
Publication statusPublished - 15 Jun 2024
Externally publishedYes

Keywords

  • Air pollution
  • Bayesian kernel machine regression (BKMR)
  • Breast cancer
  • Multiple exposure
  • Weighted quantile sum (WQS) regression
  • Xenoestrogen

Cite this