BACKGROUND AND AIM: Per- and polyfluoroalkyl substances (PFASs) are persistent chemicals with a wide range of industrial applications, measured in human serum samples. Experimental studies have identified PFASs as endocrine disrupting chemicals and epidemiological studies suggest that PFASs can affect normal thyroid function in humans. We investigated the association between exposure to mixtures of serum PFASs and changes in serum thyroid hormone levels among 12 to 20 year olds. METHODS: Using data from the NHANES 2011-2012 cycle, we employed principal component analysis (PCA) and identified three factors, describing mixtures of PFASs. Using multivariable linear regression models, we estimated associations between exposure to PFAS mixtures and changes in serum thyroid hormone concentrations. RESULTS:Factor 1 was characterized by carboxylates, comprising of PFUnA, PFDA, and PFNA, Factor 2 was described by sulfonates which included PFHxS, PFOS, PFA, and MeFOSAA and Factor 3 was described PFHpA alone. Greater exposure to Factor 1 PFASs was associated with an overall increase in free T4, TSH, and thyroid peroxidase antibodies levels, without any statistical significance. Exposure to Factor 2 PFASs was associated with an overall decrease in free T3, total T3, free T4, and total T4 levels. Factor 3 resulted in an increase in most of the serum thyroid hormone levels with the exception of thyroglobulin and thyroglobulin antibodies levels. CONCLUSIONS:Our results suggest the importance of modeling the effects of exposure to a mixture of PFASs simultaneously. The findings of this analysis can better inform researchers and policy makes on which PFASs behave analogously or inversely in an effort to better assess the effect PFASs may have on health effects. KEYWORDS: biomarkers, chemical mixtures, endocrine disruptors, polyfluoroalkyl substances, thyroid function