Publications : 2024

Thompson CM, Heintz MM, Rogers SI, Fitch SE, Rivera BN, Klaren WD, Vincent MJ, Wikoff DS, Haws LC. Evidence identification and appraisal supporting development of an updated toxicity value for HFPO-DA. Abstract 3654, Society of Toxicology Annual Meeting, Salt Lake City, UT, March 2024.

Abstract

Background and Purpose: HFPO-DA (ammonium 2,3,3,3‐tetrafluoro‐2‐ (heptafluoropropoxy)‐propanoate) is a short-chain polyfluoroether carboxylic acid used as a polymerization aid in the manufacture of some types of fluorinated polymers. Existing oral reference dose (RfD) values for HFPO-DA are based on liver effects in rodents. Recent studies indicate that the mode of action for some liver effects may lack human relevance and thus there may be need for development of an alternative RfD value that is relevant to human health. Moreover, there are new HFPO-DA studies, including chronic animal bioassays and epidemiological studies, that may warrant evaluation. Herein, an evidence-based risk assessment was conducted to identify studies with oral exposures to HFPO-DA suitable for toxicity criteria development for use in human health risk assessment. Methods: Systematic review was used to facilitate the risk assessment process. A protocol was developed describing the approach to systematically identify and select experimental animal and epidemiological studies assessing apical outcomes for use in toxicity value development. Evidence was identified from the peerreviewed literature and by targeted searching of authoritative websites and reports for unpublished studies. Studies identified from these searches were reviewed against inclusion/exclusion criteria defined by the PECO (Population, Exposure, Comparator and Outcome) statement as follows, P: in vivo experimental animal (mammalian) or epidemiological studies, E: controlled or measured oral exposure to HFPO-DA, C: untreated or vehicle-exposed negative control or a comparison or reference population exposed to lower levels of HFPO-DA, O: non-cancer and cancer apical outcomes. Studies using in vitro, ex vivo or non-mammalian models, and those investigating mechanistic effects of HFPO-DA exposure were excluded as candidates for toxicity value development. However, these study types were categorized during evidence identification for downstream targeted assessments of apical outcome biological plausibility and human relevance. Included studies were critically appraised for internal validity and risk of bias using ToxRTool (Toxicological data Reliability Assessment Tool) for the assessment of animal studies and the approaches detailed in Money et al. (2013) for the assessment of epidemiological studies. Results: Following de-duplication of search results, titles and abstracts of 417 articles were screened for inclusion or exclusion based on the defined PECO statement. A total of 47 articles met the inclusion criteria at the title and abstract level. At full text, 19 studies met the inclusion criteria, including 3 epidemiological studies and 16 experimental animal studies. Each of the 3 epidemiological studies assessed health effects associated with per- or polyfluorinated alkyl substances (PFAS), including HFPO-DA, measured in participant serum or plasma. Health effects evaluated in these studies included serum lipid levels (e.g., cholesterol, triglycerides), unexplained recurrent spontaneous abortion, and polycystic ovarian syndrome. The epidemiological study evaluating lipid outcomes did not detect HFPO-DA in the serum of any participant. Assessment of study reliability found that all three epidemiological studies had critical limitations that precluded their utility for assessing causal relationships and use in dose-response asessment. Specifically, in each study, serum or plasma PFAS measurements were taken as a single measurement during study enrollment, and therefore may not be reflective of PFAS concentrations at the time of disease onset. In addition, risks reported by study authors did not adequately account for co-exposures to other correlated PFAS compounds. Out of the 16 animal studies, 6 were unpublished laboratory reports and 10 were published studies from the peer-reviewed literature. Common outcomes evaluated in these studies were related to liver or developmental toxicity. Assessment of study reliability using ToxRTool found that 12 studies were reliable without restrictions (i.e., reliability score = 1), 3 studies were reliable with restrictions (i.e., reliability score = 2), and 1 study was determined to be not reliable (i.e., reliability score = 3). Based on study reliability assessment results, a total of 15 animal studies met both inclusion and reliability criteria for consideration in toxicity value development. Conclusions: Following a systematic evaluation of relevant data for HFPO-DA, no epidemiological studies were determined to be acceptable for use in toxicity value development. Fifteen experimental animal studies were determined to be relevant and reliable for consideration in the development of an updated oral toxicity value for HFPO-DA. Several of these animal studies investigated hepatic or developmental effects in rodents following HFPO-DA exposure. Additional data investigating the mode of action and human relevance of these effects will be considered prior to carrying specific endpoints forward for toxicity value derivation. New toxicity criteria based on these appraisals will be developed and compared to existing criteria for HFPO-DA to determine if the existing criteria are appropriate for protection of human health.