Pregnancy is a physiologically dynamic period characterized by changes in sex-steroid hormone concentrations, including progesterone, estrogens, and testosterone (Tal et al., 2000). During pregnancy, sex-steroid hormones are derived from maternal and fetal cholesterol via a series of stepwise enzymatic reactions (Tal et al., 2000). Highly regulated, dynamic changes in sex-steroid hormone concentrations, specifically substantial increases in progesterone, testosterone, and estrogens support and regulate numerous pregnancy processes, including implantation, placental angiogenesis, uterine growth, maternal immune response, maternal metabolism, and parturition (O'Leary et al., 1991; Tal et al., 2000). Consequently, deviations from “normal” sex-steroid hormone synthesis and signaling at any point in gestation may have a lasting influence on maternal and child health. Specifically, disruptions in maternal sex-steroid hormone concentrations have been implicated in preeclampsia, gestational diabetes, pre-term birth, inappropriate fetal growth, and breast cancer (Cohn et al., 2017; Makieva et al., 2014; Morisset et al., 2013; Salamalekis et al., 2006). Therefore, to protect maternal and child health, it is critical to identify common modifiable risk factors for maternal sex-steroid hormone disruption.

Pregnant women are ubiquitously exposed to per- and polyfluoroalkyl substances (PFAS), which are a class of chemicals used in food packaging, textiles, firefighting foams, and many other consumer products and industrial processes (Blake and Fenton, 2020; ITRC, 2023). The carbon-fluorine bonds make them resistant to chemical, thermal, and biological degradation – thus, they are highly persistent in the environment (Xiao et al., 2020). Perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS), two well-studied PFAS, were introduced in the United States (U.S.) during the 1940s–50s. PFOA and PFOS were voluntarily phased out by industry through a stewardship program with the U.S. Environmental Protection Agency during the 2000s due to their persistence in the environment and substantial evidence of potential unfavorable health effects, including adverse reproductive and developmental health effects (Blake and Fenton, 2020; Rickard et al., 2022). However, humans are still continuously exposed to PFOA and PFOS because of their long biological half-lives in humans (years) and persistence in the environment, especially in land and aquatic wildlife, soil, and water (Lebeaux et al., 2020). Additionally, there have been other PFAS, (e.g., perfluorobutane sulfonate (PFBS), perfluoroundecanoic acid (PFUdA)) introduced into the U.S. market since the 1950s, but their health consequences, especially during pregnancy, are relatively under-studied (Blake and Fenton, 2020; Rickard et al., 2022).

Gestational PFAS exposure is of particular concern since PFAS can cross the placenta, which results in direct fetal exposure (Kummu et al., 2015). Furthermore, associations of prenatal PFAS exposure, specifically PFOA and PFOS, with low birth weight have been reported in human epidemiologic and animal studies (Blake and Fenton, 2020; Johnson et al., 2014; Koustas et al., 2014), although some of these associations may be subject to confounding from changes in pregnancy hemodynamics (Dzierlenga et al., 2020; Verner et al., 2015). There is also some evidence that maternal exposure to PFOA or PFOS are associated with gestational diabetes (Zhang et al., 2015) or excessive gestational weight gain (Ashley-Martin et al., 2016), while maternal exposure to PFBS, perfluorohexanesulphonic acid (PFHxS), and PFUdA have been associated with gestational hypertensive disorders (Huang et al., 2019). Given that certain PFAS are associated with adverse pregnancy and birth outcomes, studies are needed to identify the biological targets of PFAS that may explain these associations.

PFAS are classified as endocrine disrupting chemicals, primarily based on studies evaluating the impacts of PFOA and PFOS in non-pregnant humans and experimental animal models (Rickard et al., 2022). Many studies have focused on the thyroid-disrupting effects of PFOA and PFOS, as experimental evidence has shown that PFAS target thyroid hormone biosynthesis, transport, metabolism, and thyroid receptors (Chang et al., 2009; Coperchini et al., 2015; Weiss et al., 2009; Yu et al., 2009). PFAS also have structural similarities to fatty acids and have been shown to activate peroxisome proliferator-activated receptors (Vanden Heuvel et al., 2006), which regulate lipid metabolism including cholesterol biosynthesis – the main precursor for sex-steroid hormones (Roth et al., 2020). However, few studies have focused on sex-steroid hormones as potential targets of PFOA, PFOS, and other PFAS, especially during pregnancy. To our knowledge, only two studies have evaluated associations of prenatal exposure to PFOA, PFOS, and several other PFAS with maternal sex-steroid hormone levels (Rivera-Nunez et al., 2023; Yang et al., 2022), whereas others focused on changes in cord blood sex-steroid hormone levels (Itoh et al., 2016; Kobayashi et al., 2021; Liu et al., 2020, 2021; Nian et al., 2020; Yao et al., 2019, 2021). If exposure to certain PFAS alters maternal circulating hormone levels, particularly earlier in pregnancy, that may explain the adverse pregnancy and birth outcomes related to PFAS. Additionally, pregnant women are exposed to PFAS mixtures (Blake and Fenton, 2020; Rickard et al., 2022), but few studies evaluated associations of a PFAS mixture with sex-steroid hormones (Liu et al., 2021; Nian et al., 2020), and none of these assessed maternal circulating hormone levels. Therefore, our main objective was to evaluate associations of PFAS (individually and as a mixture) with maternal progesterone, total testosterone, and estradiol in the early second trimester of pregnancy. Given the potential endocrine disrupting actions of PFAS and because the majority of gestational sex-steroid hormone concentrations are produced by the maternal-fetal-placental unit (Rivera-Nunez et al., 2023; Yang et al., 2022), our secondary objective was to determine if these associations differed in women carrying females versus males.