Poly- and perfluoroalkyl substances (PFAS) are a class of synthetic chemicals widely used in industrial and consumer products due to their strong water- and oil-repellent properties (Cousins et al. 2020; Houde et al. 2006; Trudel et al. 2008). Among them, perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) are two of the most commonly applied compounds. These substances have been detected in various environmental and biological media, including drinking water, seafood, soil, and even human tissues, largely due to their high persistence and bioaccumulative potential (Houde et al. 2006).

Accumulating evidence has linked PFAS exposure to a broad spectrum of adverse health outcomes, including endocrine disruption, immune dysfunction, neurotoxicity, reproductive toxicity, and cardiovascular diseases (Fenton et al. 2021; Morales-Grahl et al. 2024; Schlezinger and Gokce 2024; Zhang et al. 2023). In recent years, growing attention has been directed toward the potential impact of PFAS on skeletal health. Several epidemiological studies have reported associations between elevated PFAS levels and reduced bone mineral density (BMD), altered bone turnover, increased risk of osteoporosis, and a higher incidence of fractures (Carwile et al. 2022; Fan et al. 2023; Kirk et al. 2023; Xu et al. 2023). For instance, a large-scale study in southern China observed inverse associations between PFAS exposure and BMD, particularly among women and younger individuals (Fan et al. 2023). In vitro experiments further support these findings, showing that PFOS and PFOA impair osteogenic differentiation, enhance adipogenesis, and disrupt bone remodeling processes, possibly through mechanisms involving peroxisome proliferator-activated receptor (PPAR) signaling and hormonal imbalances (Evans et al. 2022; Kirk et al. 2021; Szilagyi et al. 2020).

Despite these findings, current evidence remains largely observational, making it difficult to establish causality due to potential confounding and reverse causation. Moreover, inconsistencies across studies, variations in PFAS types, population heterogeneity, and differing outcome definitions further obscure the nature of the PFAS-bone relationship. Given the increasing ubiquity of PFAS exposure and the burden of bone-related diseases, there is an urgent need for rigorous, hypothesis-driven approaches to clarify whether PFAS exposure has a causal role in skeletal disorders.

Leveraging advances in human genetics and large-scale biobank resources, genetically anchored analyses now offer an opportunity to explore environmental exposures through a more mechanistically informed lens. By tracing inherited variants linked to internal PFAS levels, these methods allow us to approximate long-term exposure effects while mitigating common biases in observational epidemiology. The growing availability of genome-wide association summary statistics further enhances the feasibility of conducting robust, high-throughput investigations across a wide range of skeletal phenotypes. As genetic variants are randomly allocated at conception and generally unaffected by environmental factors or reverse causation, Mendelian randomization (MR) can overcome key limitations of traditional observational studies (Larsson et al. 2023). Additionally, the availability of large-scale genome-wide association study (GWAS) datasets enables two-sample MR analyses across diverse exposures and outcomes, enhancing statistical power and generalizability (Larsson et al. 2023).

To address current knowledge gaps, we systematically examined the relationship between genetically proxied PFOA/PFOS burden and a comprehensive set of bone-related traits. In the first stage, we examined physiological indicators related to bone metabolism, including BMD at various anatomical sites, bone turnover markers (e.g., osteocalcin, collagen-related proteins, matrix metalloproteinases), mineral levels (e.g., calcium and phosphate), and regulatory hormones (e.g., parathyroid hormone and FGF23). These biomarkers provide insight into subclinical changes that may precede overt skeletal disease. In the second stage, we explored a range of clinically relevant pathological outcomes, such as osteoporosis, pathological fractures, osteonecrosis, osteomyelitis, and bone neoplasms. This study offers the first genome-informed, system-wide evaluation of PFAS-associated bone alterations across both molecular and disease-level phenotypes. By integrating genetic data with clinical phenotypes, this study aims to provide robust evidence on the causal effects of PFASs on bone health, contributing to the understanding of their potential role in skeletal disease etiology and informing prevention and intervention strategies.

PFASs, particularly PFOA and PFOS, are persistent environmental pollutants with well-established endocrine-disrupting and immunotoxic effects (Cao and Ng 2021). However, their potential impact on skeletal health remains underexplored and controversial. In this study, we used a two-sample Mendelian randomization approach to investigate the causal associations between genetically predicted PFOA/PFOS exposure and both physiological and pathological bone-related traits.

We found that PFOA exposure was causally associated with increased levels of parathyroid hormone-related protein and collagen alpha-1(XX) chain, while PFOS was associated with circulating parathyroid hormone and multiple heel BMD measurements. These findings suggest that PFASs may influence bone metabolism through hormonal regulation and extracellular matrix remodeling. While some studies have reported associations between PFAS and parathyroid hormone concentration (PTH), others have found no such relationship (Averina et al. 2024). However, no significant association between PFOA and collagen alpha-1(XX) chain has been established to date.

On the pathological level, PFOA exposure was found to be significantly associated with a spectrum of skeletal disorders, including benign neoplasms of the skull and face, osteonecrosis due to drugs, osteoarthritis, osteoporosis, and spinal fractures. While most previous studies have focused on PFAS-related carcinogenic risk in organs such as the breast, thyroid, liver, and kidney, limited attention has been given to their potential role in benign skeletal conditions. Our findings expand this field by identifying possible causal links between PFOA and non-malignant bone pathology. Moreover, the observed associations between PFOA and osteoarthritis or osteoporosis are consistent with prior observational studies, which reported higher PFAS exposure levels among individuals with reduced bone mineral density or musculoskeletal disorders. By using a genetic epidemiological approach, our study provides complementary evidence supporting these associations, while minimizing confounding and reverse causation.

Mechanistically, several pathways may underlie the observed associations between PFAS exposure and compromised bone health. One of the most well-characterized mechanisms involves activation of peroxisome proliferator-activated receptors (PPARs), particularly PPARγ. PFAS compounds are known to function as potent PPARγ agonists, which can skew mesenchymal stem cell differentiation toward adipogenesis at the expense of osteogenesis. This imbalance contributes to impaired bone formation and increased marrow fat accumulation, ultimately reducing bone mass and strength (Kirk et al. 2021). Almeida et al. further revealed that PFAS can directly bind to the ligand- and DNA-binding domains of PPARγ/RXRα complexes, altering gene transcription in a manner that mimics endogenous ligands, leading to sustained suppression of osteogenic programs (Almeida et al. 2023). Another potential mechanism involves disruption of the vitamin D endocrine axis. Vitamin D is a critical regulator of calcium homeostasis and osteoblast differentiation. Epidemiological studies have shown that PFAS exposure is associated with decreased circulating levels of vitamin D (Zhao et al. 2024). Computational docking analyses and experimental studies have demonstrated that PFAS molecules may antagonize vitamin D receptor (VDR) function, impairing the expression of downstream osteogenic genes and weakening bone mineralization (Azhagiya Singam et al. 2023). Di Nisio et al. provided mechanistic evidence that PFOA interferes with vitamin D-induced transcriptional activity in bone cells, thereby promoting osteotoxicity through endocrine disruption (Di Nisio et al. 2020).

This study has several notable strengths. By applying a bidirectional two-sample Mendelian randomization design, we reduced the influence of residual confounding and reverse causation, which often limit the interpretation of observational studies. We incorporated a broad range of bone-related phenotypes, from molecular and hormonal markers to clinically diagnosed outcomes, providing a comprehensive assessment of the skeletal effects of PFAS exposure. The use of multiple MR methods and extensive sensitivity analyses, including MR-Egger, MR-PRESSO, and leave-one-out procedures, further validated the robustness of our findings. Nevertheless, several limitations should be considered. Due to the limited sample size of existing GWAS on PFAS exposure, we adopted a relaxed significance threshold (P < 1×10⁻⁵) to identify suitable instruments, which may introduce weak instrument bias. However, all SNPs showed sufficient strength (F > 10), and the consistency across MR methods supports the reliability of our estimates. In addition, MR cannot directly reveal biological mechanisms, and our analysis was restricted to European populations, potentially limiting generalizability. Finally, as MR reflects lifelong genetic predisposition, it cannot distinguish between acute and chronic exposure or capture dose-response effects.

In summary, our study provides the first genetic evidence supporting potential causal links between PFOA/PFOS exposure and bone-related outcomes. These findings highlight the need for mechanistic studies and public health interventions targeting PFAS-induced skeletal toxicity.