Per- and polyfluoroalkyl substances (PFAS), thyroid hormones, sexual hormones and pubertal development in adolescents residing in the neighborhood of a 3M factory.

Abstract

Background: Near Antwerp a 3M factory has been active since 1971 emitting PFAS, mainly PFOS, in the local environment. Production of C8 compounds was stopped in 2002, production of other PFAS continued until 2024. This study aimed to examine the association between internal PFAS concentrations and thyroid hormones, sexual hormones, and pubertal development in adolescents living in the neighborhood of the factory.

Methods: We measured PFAS in serum of 146 female and 139 male adolescents. For males sex hormones (LH, testosterone, estradiol, progesterone, inhibin B, FSH) and SHBG were measured in serum. For males and females we assessed serum thyroid hormone levels (TSH, T3, T4 and T3/T4) and pubertal development parameters self-assessed through a standardized questionnaire. Associations between PFAS concentrations and effect biomarkers/health effects were assessed through Generalized Estimating Equations (GEE), using linear models for continuous outcomes, logistic models for binary outcomes, and proportional odds models for ordinal outcomes.

Results: For males LH, total and bioavailable testosterone showed significant negative associations with PFHxS and PFOA. LH and bioavailable testosterone also showed significant negative associations with other PFAS compounds. SHBG showed significant positive associations with PFDA, PFNA, PFHxS, PFOS and the sum of the linear forms of PFOS, PFOA, PFNA and PFHxS. Males' length and growth spurt showed significant negative associations with PFOS, PFOA and PFAS sum parameters and length and growth spurt separately also with other PFAS compounds. For females growth spurt showed significant negative association with PFOA and a significant positive association with PFOS(branched). For both males and females body hair development showed significant negative associations with PFHxS, and, for males and females separately also with other PFAS compounds. For females, breast development showed significant negative associations with PFOA, pubertal development scale showed significant negative associations with PFOA, PFHxS, PFOS(linear) and the sum of 4 PFAS. For males, TSH showed a significant negative association with PFDA and FT3 showed significant positive associations with PFOA, PFOA and PFNA. For females, FT3 showed a significant negative association with PFOS(branched).

Conclusion: We observed significant, consistent and biologically relevant associations of PFAS serum concentrations with sex hormone and SHBG levels in male adolescents. Moreover, a significant delay in physiological processes occurring in puberty was observed in females and males. Associations with thyroid hormones differed significantly by sex.