Research Highlight

Modeling Human Exposure to Poly- and Perfluoroalkyl Substances (PFAS) from Source to Dose

By Xindi Cindy Hu, Harvard University

Poly- and perfluoroalkyl substances (PFAS) are a class of synthetic organic chemicals that have been in production since the 1950s. They are detectable in virtually all Americans. Exposure to some PFAS have been linked to a suite of health problems including developmental delay, metabolic diseases such as diabetes and high cholesterol, and immune suppression. Elucidating the origin of contamination and the relative importance of exposure pathways is critical for designing effective public health interventions to reduce exposure and prevent adverse health outcomes. Through a partnership of the University of Rhode Island (URI), Harvard T.H. Chan School of Public Health, and Silent Spring Institute, scientists and engineers are studying the Sources, Transport, Exposure and Effects of PFAS (STEEP).

Pathways for human exposure to these compounds include marine foods, drinking water, and consumer goods. Communities impacted by PFAS contaminated drinking water have drawn much public attention nationwide. There is known PFAS pollution at 94 sites in 22 states[1], including industrial manufacturers, military bases, and civilian firefighting sites. As long-chain PFAS are being phased out, concentrations of shorter-chain PFAS and alternative fluorochemicals in the environment increase dramatically. Public health and environmental agencies nationwide have grappled with decisions aimed at reducing human exposures to the entire class of PFAS under many uncertainties. Major questions of relevance include: What are the relative contributions of multiple exposure pathways? Where are the potential contamination “hot spots”? What is the future trajectory of PFAS exposures?

Under the STEEP research program, two recent PhD graduates Xindi (Cindy) Hu and Clifton Dassuncao in Dr. Elsie Sunderland’s Lab at Harvard University are combining statistical and mechanistic models to better understand the relationships among environmental sources, occurrence of PFAS contamination in drinking water and marine foods, and human exposure and health impacts. Traditional approaches in exposure assessment rely on bottom-up estimates and data such as contact frequency, PFAS concentrations in the environment, and toxicokinetic parameters. This approach is data intensive, which can result in large uncertainties. An alternative approach proposed by our research uses correlations among multiple chemical homologues in environmental samples that contain information on their origin. This process is referred to as “chemical fingerprinting” and has been applied to polychlorinated biphenyls (PCBs), and polycyclic aromatic hydrocarbons (PAHs). Hu and co-authors extended this source tracking approach to human serum and discussed caveats associated with applying such methods to PFAS in their recent publication in Environmental Health (Hu et al. 2018). Using principal components analysis (PCA) confirmed by hierarchical clustering and questionnaire data, the research contrasted unique serum PFAS profiles in whaling men, children and pregnant women, and associated dominating exposure source with elevated compositions of a couple of PFAS tracers (Figure 1).

The team considers this method to be a qualitative tool that can be enhanced with the expansion the list of PFAS analytes in the future research. Regular epidemiological studies usually report six legacy PFAS and future investigations would benefit from including additional alternative PFAS as the current human exposure changes with global production shift. In future STEEP research, more data on source profiles (such as food, drinking water, indoor air, and dust), will be generated. Comprehensive data will help impacted communities to understand their major exposure source and to make well-informed decisions on the best strategy to reduce their exposure. These data can also help policy makers to design the most effective public health intervention to prevent harmful health impacts of PFAS.

Figure 1: Profiles of PFAS in human serum provide information on major exposure sources. Principal component analysis of serum PFAS profiles show clustering of three groups, confirmed by questionnaire data. C9-C12 perfluoroalkyl carboxylic acids (PFCAs) are associated with seafood exposure; while perfluorohexane sulfonic acid (PFHxS) and N-ethyl perfluorooctanesulfonamido acetic acid (N-EtFOSAA) are associated with exposure from consumer products.


Hu XC, Dassuncao C, Zhang X, Grandjean P, Weihe P, Webster GM, et al. 2018. Can profiles of poly-and perfluoroalkyl substances (PFASs) in human serum provide information on major exposure sources? Environmental Health 17:11.

[1]  PFAS Contamination Site Tracker. Social Science Environmental Health Research Institute (SSEHRI) at Northeastern University Accessed June 25th 2018.

More News