Long-term impact of per- and polyfluoroalkyl substances on kidney health: a large-scale genetic study

The widespread use and bioaccumulation of per- and polyfluoroalkyl substances (PFASs) have raised significant concerns regarding their potential impact on kidney health. However, previous studies on PFASs exposure and kidney health have produced inconsistent results. In this study, we performed a system-wide bidirectional Mendelian randomization (MR) design to explore the comprehensive causal relationship between PFASs exposure (PFOA, PFOS) and kidney-related traits. Genetic instruments for PFASs were sourced from the GWAS Catalog, while kidney-related traits (including 47 renal function indicators and 34 kidney diseases) were obtained from the UK Biobank, FinnGen, and meta-GWAS consortia. Statistical analyses were performed using multiple MR approaches to evaluate associations between genetically predicted PFASs levels and kidney traits, with additional sensitivity analyses and reverse MR analyses conducted to assess robustness. The results showed that higher genetically predicted PFOA levels were associated with lower urea (OR = 0.93, P = 0.024) and creatinine levels (OR = 0.98, P = 0.013), as well as a reduced long-term risk of tubulo-interstitial nephritis (OR = 0.84, P = 0.046). In contrast, higher genetically predicted PFOS levels were linked to increased estimated glomerular filtration rate (OR = 3.01, P = 0.002) and cystatin M levels (OR = 1.62, P = 0.046), but also a greater long-term risk of renal tubulo-interstitial diseases (OR = 1.18, P = 0.044) and unspecified kidney failure (OR = 1.16, P = 0.019). Both reverse MR and sensitivity analysis support these results. In conclusion, our MR study provides strong evidence for the causal impact of PFASs on renal function indicators and kidney disease, highlighting the complexity of PFASs toxicology, which is essential for refining risk assessments and informing regulatory decisions on these persistent environmental contaminants.