What is the scientific merit of using environmental persistence as a means to group PFAS for regulations regarding manufacture, import and use? How could environmental persistence be defined? For any PFAS (based on the broadest OECD definition) that is not persistent itself and does not degrade into a persistent PFAS, would those then be excluded from further evaluation?

Environmental persistence needs to be defined and will be calculated for most PFAS. Persistence is of importance only after the function of the chemical for which purpose it was synthesized and used. It is not realistic to assume that a PFAS used as a processing aid will remain unchanged in the process but will degrade rapidly when out of the system.
For in-process uses of PFAS, technology such as closed systems and release controls will be more important than parameters such as persistence or bioaccumulation or half-life in certain environmental media under environmental conditions.
The same applies to agrochemicals, pharmaceuticals, etc. which first have to fulfill their function.
The OECD definition is far too broad and does not warrant to investigate all substances; better move to other groups of chemicals.

Persistence is in itself not that important in predicting the risk posed by a chemical (certain artificial sweeteners are persistent). Persistence eliminates the option of relying on natural degradation as a clean-up option. Persistence in conjunction with high levels of release over extended periods of time are more of a concern (DDT, dioxins, PCBs, etc.). My impression is that for PFAS the levels of release were not as high or as widespread since they are largely limited to releases during manufacture and not during use (except for fire fighting foams). Persistence in combination with bioaccumulation is also more of a concern because of the potential for higher doses for animals at the top of the food chain. Thus, persistence alone may not be that useful a criterion for defining a PFAS substances versus not a PFAS substance. I would not exclude non-persistent compounds from the definition of PFAS.

Such an approach is based on green chemistry principles (principle #10: design for degradation) (Anastas and Warner, 1998). The vast majority of PFAS are synthetic and without mankind they would not be present in the environment. Once released, PFAS will stay in the environment indefinitely, and if continually released, the amount in the environment will increase. The EU REACH regulation has a suitable definition of persistence, but maybe there should be specific higher thresholds for extremely persistent substances such as PFAS? If a substance is defined as PFAS (e.g. under the OECD definition), but is mineralizable (there are a few substances e.g. a CF3 attached to a heteroatom (O, S, N), see e.g. Cousins et al., 2020), then it is not persistent. If one of the rare non-persistent PFAS is shown to be highly toxic or bioaccumulative or used in high quantities, it cannot be excluded from further evaluation. There can, however, be substances that meet the PFAS structural definition, that can be judged to be safe for human health and the environment.

Anastas, P.T.; Warner, J.C. (1998) Green Chemistry: Theory and Practice, Oxford University Press: New York, p.30.

Cousins, I.T.; DeWitt, J.C.; Glüge, J.; Goldenman, G.; Herzke, D.; Lohmann, R.; Ng, C.A.; Scheringer, M.; Wang, Z. (2020) The High Persistence of PFAS is Sufficient for their Management as a Chemical Class. Environ Sci-Process Impacts, 22, 2307-2312.

Environmental persistence is an important parameter than defines exposure potential, but I do not believe it is the right parameter to use for grouping PFAS for risk assessment (the main focus of this exercise?). If the question is limited to regulations regarding manufacture, import and use, environmental persistence assumes greater importance. I am not well across definitions of "environmental persistence" but it should be possible to select one from the literature. For the example where a PFAS is not persistent itself, nor degrades to a persistent metabolite or degradation product, I would be comfortable for it to be excluded from such regulatory action.

On the issue of "exclusion from further evaluation" (also raised at question 1.5 in round 1), I was tempted to note that the concept of Threshold of Toxicity Concern (TTC) is sometimes used in risk assessments involving exposures to small amounts of multiple chemicals as a means of excluding some from a list of Chemicals of Concern (CoC) that are taken forward in the risk assessment. Where exposure estimates for individual chemicals are below the TTC levels developed for three Kramer classes or carcinogenic substances (Kroes et al Tox Sci 86: 226-230, 2005; see also EFSA guidance on the TTC concept; EFSA J 17(6): 5708, 2019) the contribution to risk from that component is likely to be small enough to be neglected. However, there is a warning in both the Kroes et al (2005) and EFSA (2019) papers that the application of the TTC to mixtures analysis needs to be addressed with caution. Furthermore, the TTC values in these references have been developed before the latest very low RfD values for PFOS, PFOA and some other PFAS were developed, and since these newer values are well below any of the published TTC cut-offs, application of the TTC concept would actually result in exclusion of most PFAS from further risk assessment. Therefore, application of TTC principles in this case would be counter-productive.

Criteria for persistence needs to be identified. There are some publications with regard to the criteria for persistence in air, water, soil and biota; half-lives or residence times are indicators, but this may be complicated by parent versus transformation products (i.e., parent chemical may not be persistent, but the transformation product could be). Klecka et al. 2000 (SETAC press) have reported some criteria for persistence. For a PFAS that is not persistent itself and does not degrade into a persistent PFAS, I would suggest that it be excluded from persistence criterion. However, it is important to make sure that there is a strong evidence to support the criteria. The notion of strong C-F bond makes everyone believe that if there is a C-F bond, a chemical with that moiety would persist in the environment. Therefore, a strong evidence is needed to convince PFAS or its transformation product is not persistent.

Doesn't this consideration depart somewhat from the Problem Formulation as written? It specifies PFAS in drinking water and existing data. Presumably then, the world of PFAS is limited to those found or likely to be found in surface water. If so, excluding non-persistent PFAS would not be appropriate, as these could be directly discharged to surface water. Exclusion would be supported only if there are data indicating that no such purposeful or inadvertent discharges are likely.

Persistence has been defined by several regulatory entities. The USEPA defines persistent substances as those that have a half-life in water, soil, and sediment of greater than 60 days and very persistent substances as those that have a half-life in water, soil, and sediment as greater than 180 days. The vast majority of PFAS or PFAS precursor degradation products that also are PFAS fall into the very persistent category. These definitions also are more or less in agreement with EU REACH Annex XIII definitions as well. Most PFAS can't really be grouped differently by persistence as they or their degradation products meet the definition of very persistent. PFAS that do not meet the definition of persistent or very persistent would likely need to be grouped or categorized or evaluated with other criteria, such as mobility, bioaccumulation potential, and/or toxicity. Persistence therefore is sufficient as a grouping/regulatory criteria for many, if not most, PFAS. Cousins et al. (2019 - Why is high persistence alone a major cause of concern) notes that persistence leads to continuously increasing contamination and increasing probabilities of known and unknown effects and that once adverse effects are identified, it will take decades, centuries or even longer to reverse contamination and therefore effects. This paper very clearly lays out the scientific merits of using environmental persistence or a "P-sufficient approach" for chemical management/regulation.

The toxicological outcome, the adverse effect is the optimal means to conduct a grouping activity for mixtures or cumulative risk assessment. It is agnostic to physical chemical properties, environmental fate and transport, biological longevity and chemical structural attributes. A specific investigation into the correlation between environmental persistence and the types of health effects (and the relative potencies of studies components) would be informative. Given the relative state of knowledge regarding toxicities and potencies of chemicals to be grouped, environmental persistence may by a lesser evil. The most relevant "persistence" is biological - the accumulative issue. The use of environmental persistence may be approached as a surrogate for bioaccumulative, taking advantage of what is known about the distribution of molecules with potentially-groupable molecular attributes into bins organized according to environmental persistence and bioaccumulative potential. But, it will have to be recognized that chemicals in these potential bins may have different toxicological properties including types of effects and relative potencies. The application of environmental persistence, alone, should be viewed as an insufficient means of grouping. No, those chemicals should not be excluded, they would have a characterized level of persistence, which would be on the lower end of the scale. But, they will have measurable persistence - nothing instantly vanishes from the environment.

In theory, substances with relatively low persistence can achieve relatively high steady-state concentrations in the environment if input exceeds the clearance rate. I'd rather see these lower-persistence substances included at the outset of a risk assessment, and then excluded if appropriate through a screening process based on both rate of entry to and depletion from the receiving environment. In a practical sense, this screening could support grouping for regulatory purposes, but not based solely on persistence.

A non-persistent PFAS could represent a human health risk if exposure is frequent/sustained and potency is high. In addition, a non-persistent PFAS could have a mechanism of action that is similar to that of persistent PFAS. If both are present in drinking water, they should be considered together in a dose additivity model. For this reason, I do not believe environmental/biological persistence should be the basis for grouping PFAS in the context of risk assessment.

Environmental persistence alone is not adequate. One would also need to show the potential for bioavailability and/or degradation into something that is bioavailable and toxic. Persistence is important in that the longer a compound is present, the longer it has to do damage. This is evident from the difference in half-life in long and short chain PFAAs. However, if something is not bioavailable or does not degrade (such as some polymers) then this would not have scientific merit.