Systematic Evidence Map for the Per- and Polyfluoroalkyl Substances (PFAS) Universe

Background: Per and polyfluoroalkyl substances (PFAS) are a research priority for the US EPA. Because PFAS include thousands of structurally diverse chemicals, there is a pressing need for identifying what data are available to assess human health hazard of these compounds.

Objectives: We used systematic evidence map (SEM) methods to summarize the available epidemiological and mammalian bioassay evidence for ~14,735 chemicals identified as PFAS by EPA’s Center for Computational Toxicology and Exposure (CCTE). This work is a continuation of our previous 2022 and 2024 SEMs that inventoried evidence on a separate set of ~500 PFAS. The comprehensive PFAS dashboard includes evidence identified from our past SEMs and completed US EPA assessments.  

Methods: We conducted literature searches from peer-reviewed and gray literature sources to identify, screen, and inventory mammalian bioassay and epidemiological literature. A combination of manual review and machine learning software were utilized. A diverse array of potentially relevant supplemental content was also tracked, including mechanistic data, exposure-only studies, and studies informing chemical toxicokinetics and clearance.  For each study meeting predefined population, exposure, comparator, and outcome (PECO) criteria, experimental design details and health endpoints evaluated were summarized in interactive web-based literature inventory visuals.  Epidemiology studies and animal bioassay studies with ≥21-d exposure duration or reproductive/developmental study design proceeded to undergo a study evaluation for risk of bias and sensitivity, as well as detailed extraction of health endpoint data.  Underlying data is publicly available and can be downloaded.

Results: Scientific database searches retrieved 152,205 references.  After full text screening, there were 347 mammalian bioassay and 44 epidemiological studies that met PECO criteria. The mammalian bioassay and epidemiological evidence assessed 99 and 30 individual PFAS respectively (n=18 PFAS with both). The epidemiological evidence assessed 15 health systems and the mammalian bioassay evidence assessed 16 health systems.

Discussion:  Results from our 2022 and 2024 SEMs and completed US EPA assessments are compiled into a comprehensive PFAS dashboard. This dashboard is a resource for better understanding the currently available PFAS human health hazard data. It can be used as a tool for researchers and regulators interested in PFAS data gaps and research needs. Across all the data sources compiled into the comprehensive PFAS dashboard, only 1.4% (214/14,735) of PFAS had any mammalian bioassay or epidemiological data available. The vast majority of PFAS lack publicly available information about the potential human health effects of exposure to these chemicals.