Given the lack of toxicological data and targeted analytical methods for most PFAS, when conducting an assessment of human health risk from PFAS in drinking water, do you think a screening level risk assessment could be conducted by assuming the total adsorbable/extractable organic fluorine concentration is equal to the concentration of a known toxic PFAS (e.g., PFOA)? What is the scientific merit of such an approach? If the results of TOF suggest a concentration greater than a PFOA-based threshold, then what would be conclusion and next steps? If the results of TOF suggest a concentration less than a PFOA-based threshold, then what would be the conclusion and next steps?

Answer Explanations

• Yes, total adsorbable/extractable organic fluorine concentration can be used for screening level risk assessment

  Expert 3

  I think this is a reasonable screening approach. However, it seems that in Round 1 experts better versed in PFAS chemistry than I raised objections that should be considered.

• No, total adsorbable/extractable organic fluorine concentration cannot be used for screening level risk assessment

  Expert 10

  I don't think TOF-methods can be used for risk assessment (they are not even a measure of total PFAS), but they could be used to identify ("screen") high concentration locations to save resources (if e.g. extractable organic fluorine (EOF) is indeed easier, faster and cheaper than targeted PFAS analysis, which is debatable). For the "hot spot" samples, EOF should be followed up with targeted PFAS analysis, which would form the basis of any risk assessment.

• Yes, total adsorbable/extractable organic fluorine concentration can be used for screening level risk assessment

  Expert 8

  I am not sure what the context is for this question. If one simply wants to know whether there is a hazard from a sample of drinking water then this might be a simple approach. If the concentration is sufficiently below the regulatory level of the target PFAS (i.e. PFOA or some other PFAS), then one could assume there is no risk from the drinking water. This approach does not lead to the development of a MCL, if that is the ultimate goal.
  If the total fluorine turned out to be greater than the regulatory threshold for the target PFAS, then one has to figure out which PFAS are present and which are contributing to the exposure. Then one would have to understand the toxicity of those compounds to determine which are of actual concern.

• No, total adsorbable/extractable organic fluorine concentration cannot be used for screening level risk assessment

  Expert 9

  This question is a little unclear. Current mass balance estimates for PFAS in environmental compartments demonstrate a substantial gap between targeted PFAS and EOF/TOF (larger for TOF than for EOF; see McDonough et al., 2019). Assuming that levels of a known PFAS are equivalent to EOF/TOF will significantly underestimate the number and concentrations of PFAS in a particular environmental media. It's therefore highly unlikely in complex environmental media that EOF/TOF will be less than a PFOA-based threshold. The more likely scenario is that EOF/TOF will exceed a PFOA-based threshold. Additionally, in some environmental media, PFOA is lower than other legacy/long-chain PFAS and even in the same media, such as drinking water, the distribution of PFAS can vary quite a bit depending on the source of contamination.

• Yes, total adsorbable/extractable organic fluorine concentration can be used for screening level risk assessment

  Expert 1

  Although this approach is not perfect (TOF can include chemicals which are not typically considered like PFAS), it is one that can realistically be applied in the near future. A first screening would allow determining whether at least one PFAS exceeds the PFOA/PFOS drinking water guideline. Analytical methods are relatively novel, but the limits of detection seem to be lower than most drinking water guidelines, in the low ng/L or ppt (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7584354/).

  Samples exceeding the PFOA/PFOS guideline level should be analyzed using targeted methods to allow evaluating which known PFAS contribute the most, and the % of unknown PFAS in the mixture. In cases where known chemicals exceed guideline levels, risk should be managed. Where TOF levels exceed the PFOA/PFOS guideline level, but the guidelines are not exceeded by individual PFAS, risk could be considered as uncertain, and futher investigation would be warranted to determine the mixture composition. Where TOF levels do not exceed PFOA/PFOS guideline levels, risk could be considered as negligible.

• Yes, total adsorbable/extractable organic fluorine concentration can be used for screening level risk assessment

  Expert 11

  Yes, a screening test based on TOF could have great value. It would need to be carefully developed to exclude inorganic fluoride, insoluble organic compounds that contain fluoride, and fluoride polymers that are poorly absorbed.

  PFAS substances are not expected to widely occur in public water supplies, thus positives would not be expected to occur frequently. Under these conditions a screening test would be valuable as a cost-effective method of identifying the small number of systems that are affected by PFAS.

  The development of such a test is dependent on the analytical chemistry involved. Organic fluorine pesticide drugs and pesticides occur. If they produce background levels of TOF that exceed the screening value based on PFOA the approach will be of little value.

  Exceeding the TOF criteria would not indicate the need for controls. It would indicate the need for additional testing to determine the actual fluorochemicals in the water. For example, the water system could test to determine if the TOF was in fact due to PFOA. If it was, then actions to control the contamination should be taken. If it is not PFOA, then the water system should test to determine if the source is a known organic fluoride (fluoride containing drugs and pesticides or a known PFAS). If the substance(s) are unknown, then NTA could be used to identify molecular weight and number of carbons. Based on these results, guidance should be provided to the system on how to use the results to decide if the contamination needs to be controlled or merely monitored. This guidance should be in place before the TOF-based screening is performed.

• Yes, total adsorbable/extractable organic fluorine concentration can be used for screening level risk assessment

  Expert 5

  I think a defensible case for such an approach can be made. Nearly two decades past, EPA mixture researchers were struggling with unidentified halogenated disinfection byproducts in drinking water. An approach using total organic carbon, halogen concentrations and ratios of C to halogen to address the potential risk posed by unidentified halogenated disinfection byproducts - a completely analogous problem. For the DBPs, toxicity and dose response data were more robust than for PFAS - so, those researchers had a pretty good idea of which chemical was the most potent. In contrast, there is a much weaker toxicity database for PFAS chemicals, which will translate into uncertainty in "benchmarking" a toxic dose of PFAS, which will be compared to F content of water. So, the approach might be better described and proposed that enacted with an acceptable level of confidence.

• Yes, total adsorbable/extractable organic fluorine concentration can be used for screening level risk assessment

  Expert 4

  The viability of this approach depends upon the defensibility of the screening criterion (e.g. PFOA-based). Provided the screening criterion is sufficiently stringent, including addressing uncertainties the outcomes could be followed up as follows:

  Concentrations are LESS than screening criterion/criteria: no further assessment or management required
  Concentrations are GREATER than screening criterion/criteria: Quantitative RA required, addressinig specific components

• No, total adsorbable/extractable organic fluorine concentration cannot be used for screening level risk assessment

  Expert 2

  The analytical methods for TOF/EOF are still not well developed and not harmonized. Very few labs perform such analysis, but they are not harmonized. First such methods need to be standardized before even we use for regulatory decision making. The quality of data produced by such methods currently is not reliable. They (EOF/TOF values) may provide some worst case scenario, but still those methods need to be standardized. If standardized method are used and if TOF is > PFOA-based threshold, efforts should be made to identify what PFAS is contributing to that higher TOF value.

• No, total adsorbable/extractable organic fluorine concentration cannot be used for screening level risk assessment

  Expert 7

  Method does not generate comparable results.
  The parameter cannot be associated with any toxicity.
  TOF is not a fast nor cheap method and needs a fully functioning chemical laboratory.

• No, total adsorbable/extractable organic fluorine concentration cannot be used for screening level risk assessment

  Expert 6

  While TOF might provide a useful initial number for screening potential EXPOSURE, the likely differences in toxic potency across the range of PFAS means that it has limited usefulness in screening for potential health RISK. If a TOF analysis is to be compared to some toxicity-based threshold as a basis for reaching a risk-based conclusion, where the TOF value appears to be below such a threshold, PFOA would probably be the wrong choice. You would need to select a PFAS with known toxicity at the lowest regulatory level (currently PFOS?).