P
Range of the PFOA/S Safe Dose

 

A coalition of interested parties is forming to develop a range in the safe dose(s) for perfluorooctanote (PFOA) and its sulfonic acid (PFOS), which is reflected in differences in safe doses world-wide of over 100,000-fold (see attached presentation).  This collaboration started in September of 2022 with an open call for nominations to an advisory committee, now selected,* who will lead the coalition through a series of virtual and face-to-face and possibly in-person meetings.  Results are anticipated in the Spring of 2023. 
As with all ARA projects, all interested groups are welcome to be involved.  For example, 

  • Your group can endorse the effort.  Here endorsement simply means that your group agrees with the collaborative effort being done under the auspices of the Alliance for Risk Assessment (ARA) (see: https://www.tera.org/Alliance%20for%20Risk/index.htm).    If you wish to endorse, please send us your group’s logo.
  • Your group could participate directly.  Here participation is contribution to the actual technical work by one or more of your group’s scientists.  If you wish to participate, please send us bio sketches of selected scientists.
  • Your group could donating money or other resources, like a meeting place, to the cause.  Toxicology Excellence for Risk Assessment has been designated as the nonprofit organization to support this collaboration.  Because TERA is a 501c3 nonprofit organization, all donations to this effort are tax deductible.  

 

For interest in joining the coalition, please contact Michael Dourson at dourson@tera.org

 

*Advisory Committee:

  • Lyle Burgoon with Raptor Pharm & Tox, Ltd, USA
  • Harvey Clewell with Ramboll, Global
  • Tony Cox with Cox Associates, USA
  • Michael Dourson, Toxicology Excellence for Risk Assessment, USA
  • Tamara House-Knight with GHD, Global
  • Ravi Naidu with CRC CARE, Australia
  • Paul Nathanail with LQM, United Kingdom
  • James S. Smith with US DoD, USA
  • Nitin Verma, Chitkara University, India

 

Keynote Presentation to the CleanUp Conference in Adelaide, Australia, and the Australian Government, Department of Health and Aged Care: Click Here

 

 

Workshop Report published by Regulatory Toxicology and Pharmacology: Click Here

 

 

Range of the PFOA/S Safe Dose References:

ARA PFOA RfD Range

 

-----

Present at the May 1, 2024 conference call (in dark print)

 

Group 1

  • Paul Nathanail, LQM, UK, consultant
  • Linda Dell, Ramboll, USA, consultant
  • Bernard Gadagbui, Toxicology Excellence for Risk Assessment, USA, NGO
  • Helmut Greim, University of Munich University(retired), University, Germany, university,
  • Vijay Kannappan, Cook Medical, USA, industry
  • Mike Luster, retired NIOSH, USA, government 
  • Therese Manning, Environmental Risk Sciences Pty Ltd, Australia, consultant
  • Tiago Severo-Peixi, State University of Londrina, Brazil, university
  • Andrea Wojtyniak, Geosyntec Consultants International, Canada, consultant

 

Group 2

  • Harvey Clewell, Rambol, USA, consultant
  • Tamara House-Knight, GHD, USA, consultant
  • Thomas Colnot, CiS Toxicology, Chile, consultant
  • Edmund Crouch, Green Toxicology, USA, consultant
  • James Deyo, Environmental Protection Authority, New Zealand, government
  • Mahesh Gupta, University of Saskatchewan, Canada, university
  • Travis Kline, Geosyntec Consultants, USA, consultant
  • Anurag Sharma, Nitte University, India, university
  • Katie Richardson, Senversa, Australia, consultant

 

Group 3

  • James Smith, NMCPHC, USA, government
  • Nitin Verma, Chitkara University, India, university
  • Wolfgang Dekant, University of Würzburg (retired), Germany, university
  • Philip Goodrum, GSI, USA, consultant
  • Laura Green, Green Toxicology, USA, consultant
  • Andrew Pawlisz. Trihydro, USA, industry
  • Frank Pagone, RHP Risk Management, USA, consultant
  • Jackie Wright, Environmental Risk Sciences Pty Ltd, Australia, consultant

 

 

“Secretariat”

 

  • Tony Cox with Cox Associates, USA, consultant
  • Michael Dourson with TERA, USA, NGO
  • Ashish Jachak from RHP Risk Management Inc., USA, consultant
  • Ravi Naidu with CRC CARE, Australia

 

After a brief introduction by Michael Dourson, Group 2 was asked to offer some insights on the second task of determining the critical studies for one or more of PFOS’s critical effect(s). 

Speaking for Group 2, Harvey Clewell noted that human observational studies were judged not to be sufficiently reliable to form the basis of the safe PFOS dose.  In contrast, experimental animal studies in monkeys for liver effects (Seacat et al. 2002, Butenhoff et al. 2012) and in rodents for developmental toxicity (Lau et al. 2003, Luebker et al. 2005, Thibodeaux et al. 2003) were judged to be sufficiently reliable.  Moreover, these same studies were used differently by various authorities (MAK, 2015, Health Canada, 2018, FSANZ, 2017, ATSDR, 2021) for developing their safe PFOS dose.  Group 2 also suggested two studies from which a human clearance estimate might be developed (Li et al., 2022 and Zhang et al., 2013). Such a parameter is necessary to estimate the safe PFOS dose from the serum concentrations of PFOS that are available in several of the experimental animal studies.

Discussion on Group 2’s included information on the UK Committee on Toxicology (Paul Nathanail), ways to access additional information that would allow a PFOS human clearance based urinary and biliary excretion (Edmund Crouch), and clarity on differences between PFOA and PFOS, where the latter is a much stronger acid (Laura Green).  It was also mentioned that Tony Fletcher might have information on the amount of PFOS cleared in the bile.  Paul Nathanail volunteered to contact this gentleman.  Michael Dourson volunteered to contact Li et al. (2022) to obtain their raw data on clearance values or alternatively their geometric mean clearance values.  The latter information would be important based on the study of Zhang et al. (2013) which indicated that the geometric mean was a better descriptor of clearance than the arithmetic mean.  Late that night, Edmund Crouch emailed around half-life estimates derived from data presented in Li et al. (2022).

Wolfgang Dekant and Laura Green then discussed the findings of Group 3, which mirrored those described by Group 2.  Human data were not only considered unreliable to form the basis of the safe dose, but were even less reliable than similar data for PFOA.  Liver effects in monkeys were considered to be the better choice for safe dose assessment because of similarity between monkeys and humans with respect to hepatic effects.  Developmental toxicity in rodents was also considered to be a usable endpoint, although one with more overall uncertainty than the monkey data.  One comment made was that PFAS in general, and PFOS in particular, is not an emerging issue, as evidenced by dramatically decreasing serum concentrations in human populations throughout the world.

Group 1 then described its findings for various endpoints.  Like the other two groups, Group 1 did not consider the observational studies in humans to be a reliable basis for deriving the PFOS safe dose, and found that immune system effects cannot be used since the ones in particular used by EFSA and US EPA for the estimation of their safe doses are not of clinical relevance.  Afterwards, Linda Dell described human studies on reproduction and birth weight with their attendant difficulties but firmer observations in rodents; Bernard Gadagbui described the cancer findings stating that while these were considered relevant by IARC and US EPA, other authorities disagreed; Vijay Kannappan briefly discussed thyroid effects, comments including differences in rodent and human thyroid homeostasis; and Tiago Severo-Peixi described liver effects, which while mild nevertheless seemed consistent in several experimental animal species.  The general feeling of Group 1 was that liver and developmental effects in experimental animals were more reliable than other effects, but a clear consensus on this was not yet reached by the group.  Laura Green added that if we rely on developmental effects, it should perhaps be made clear that PFOS is not a teratogen.

The overall conclusions of all three groups were that:

  • Epidemiological studies cannot form a reliable basis for deriving the PFOS safe dose;
  • One or more experimental animal studies focusing on liver and perhaps developmental toxicity are reliable bases for deriving the PFOS safe dose;
  • The studies of Li et al. (2022) and Zhang et al. (2013) should be further investigated to see whether a credible estimate of human clearance can be obtained focusing on both biliary and urinary excretion.

 

The meeting ended with Michael Dourson reiterating the next assignments for the 3 groups.  Specifically, groups are to independently:

  • Continue to contemplate the PFOS’s mode(s) of action for its critical effect(s) based on discussions in the first consensus call;
  • Continue to contemplate the critical studies for one or more of its critical effect(s) based on discussions in this second consensus call;
  • Select the choice of extrapolation method including the choice of uncertainty factors [deadline May 24]; and
  • Identify additional data gaps [deadline June 21].  
As before, this sequence will be periodically punctuated by zoom conference calls for the purpose of team presentations and attempted consensus around the

 

-----

Notes of March 26/27, 2024 International Zoom-Call:

Group 1

  • Paul Nathanail, LQM, UK, consultant
  • Linda Dell, Ramboll, USA, consultant
  • Bernard Gadagbui, Toxicology Excellence for Risk Assessment, USA, NGO
  • Helmut Greim, University of Munich University(retired), University, Germany, university,
  • Vijay Kannappan, Cook Medical, USA, industry
  • Mike Luster, retired NIOSH, USA, government 
  • Therese Manning, Environmental Risk Sciences Pty Ltd, Australia, consultant
  • Tiago Severo-Peixi, State University of Londrina, Brazil, university
  • Andrea Wojtyniak, Geosyntec Consultants International, Canada, consultant

 

Group 2

  • Harvey Clewell, Rambol, USA, consultant
  • Tamara House-Knight, GHD, USA, consultant
  • Thomas Colnot, CiS Toxicology, Chile, consultant
  • Edmund Couch, Green Toxicology, USA, consultant
  • James Deyo, Environmental Protection Authority, New Zealand, government
  • Mahesh Gupta, University of Saskatchewan, Canada, university
  • Travis Kline, Geosyntec Consultants, USA, consultant
  • Anurag Sharma, Nitte University, India, university
  • Katie Richardson, Senversa, Australia, consultant

 

Group 3

  • James Smith, NMCPHC, USA, government
  • Nitin Verma, Chitkara University, India, university
  • Wolfgang Dekant, University of Würzburg (retired), Germany, university
  • Philip Goodrum, GSI, USA, consultant
  • Laura Green, Green Toxicology, USA, consultant
  • Andrew Pawlisz. Trihydro, USA, industry
  • Frank Pagone, RHP Risk Management, USA, consultant
  • Jackie Wright, Environmental Risk Sciences Pty Ltd, Australia, consultant

 

 

“Secretariat”

 

  • Tony Cox with Cox Associates, USA, consultant
  • Michael Dourson with TERA, USA, NGO
  • Ashish Jachak from RHP Risk Management Inc., USA, consultant
  • Ravi Naidu with CRC CARE, Australia

 

After a brief introduction by Tony Cox, Group 1 was asked to offer some insights on the first task of determining PFOS’s mode of action for its critical effect(s).  Speaking for Group 1 Helmut Greim noted that while immunotoxicity was the critical effect in whole or in part by EFSA and USEPA, other authorities have judged that these effects are primarily based on associations from human studies.  Furthermore, the stated effects have little clinical relevance and are at doses much lower than those evoking immunotoxic effects in experimental animals.  Moreover, the US NAS (2016) and several more recent papers studying this response in humans to PFOS and other PFAS exposures do not support immunotoxicity as a critical effect.  Nor was a MOA for this potential critical effect readily discernable.  Other effects such as liver and/or the reproductive system should be considered.

Harvey Clewell then spoke for Group 2 and stated that little difference appeared to exist between PFOA and PFOS regarding their MOAs and arrays of adverse effects.  Therefore, it would not be surprising if the range of the safe dose for PFOA and PFOS were likewise similar.  Future research focus should be on collecting data from mechanistic studies.  Similar to Group 1, Group 2 went on to state that the causality of the epidemiology studies has not been adequately demonstrated, as amply described by WHO (2022).  Harvey also mentioned that the urinary clearance of PFOS was smaller than the urinary clearance of PFOA in that more PFOS is excreted in the feces via the bile.  Finally, disturbance of lipid metabolism might be the underlying MOA for the host of adverse effects seen.

Jackie Wright and Nitin Verma then spoke for Group 3 and stated that, like both Groups 1 & 2, Group felt that the immunotoxicity of PFOS was not conclusive since the majority of findings in humans were associations of little clinical relevance, and newer studies, such as on COVID patients, showed no effects.  Immunotoxicity results in experimental animals were also inconclusive and generally at much higher doses.  Group 3 went on to state that more work was needed on liver effects where studies in monkeys are important.  Also to be considered are reproductive toxicity in rodents, cholesterol effects in humans focusing on effects of clinical relevance, and bladder, lung, and cerebral-vascular tumors.

Another point brought up referred to the UK Committee on Toxicology (COT) who concluded that "Whilst the COT is unable to suggest an alternative to the [EFSA] TWI [tolerable weekly intake] at this time, there are strong caveats when comparing the exposure estimates with the TWI established by EFSA. There is considerable uncertainty as to the appropriateness of the derivation of the TWI, and of the biological significance of the response on which it is based, which complicates interpretation of the possible toxicological significance of exceedances."

 

It was further mentioned that the TWI is a baseline below which there is no appreciable risk to human health however exceeding the TWI did not necessarily warrant remediation.

 

The meeting ended with Michael Dourson reiterating the next assignments for the 3 groups.  Specifically, groups are to independently:

  • Continue to contemplate the PFOS’s mode of action for its critical effect(s);
  • Focus on determination of the critical studies for one or more of its critical effect(s) [deadline April 12];
  • Select the choice of extrapolation method including the choice of uncertainty factors [deadline May 10]; and
  • Identify additional data gaps [deadline June 7].  

This sequence will be periodically punctuated by zoom conference calls for the purpose of team presentations and attempted consensus around the various focus topics.  

 

-----

Notes from Zoom-Call:

3/29/2023

Third International Conference Call on Safe Dose Range of PFOA & PFOS

Present (P)/Excused (E):

Chair: Tony Cox with Cox Associates, USA, consultant (E)
Co-Chair: Ravi Naidu with CRC CARE, Australia, NGO (E)
Rapporteur: Michael Dourson with TERA, USA, NGO (P)

Group 1
Lyle Burgoon, RaptorPharmTox, USA, consultant (P)
Paul Nathanail, LQM, UK, consultant (E)
Shanon Ethridge, International Association for Plumbing and Mechanical Officials Research and Testing, USA, NGO (E)
Vijay Kannappan, Cook Medical, USA, industry (P)
Michael Luster, NIOSH (retired), government, USA (E)
Therese Manning, Environmental Risk Sciences Pty Ltd, Australia, consultant (P)
Tiago Severo-Peixi, State University of Londrina, Brazil, University (E)
Andrea Wojtyniak, Geosyntec, Canada, consultant (P)

Group 2
Harvey Clewell, Rambol, Global, consultant (P)
Tamara House-Knight, GHD, Global, consultant (e)
Linda Dell, Ramboll, Global, consultant (P)
James Deyo, Environmental Protection Authority, New Zealand, government (E)
Bernard Gadagbui, Toxicology Excellence for Risk Assessment, USA, NGO (P)
Travis Kline, Geosyntec Consultants, USA, consultant (E)
Katie Richardson, Senversa, Australia, consultant (E)
Anurag Sharma, Nitte University, India, university (E)

Group 3
James Smith, NMCPHC, USA, government (E)
Nitin Verma, Chitkara University, India, university (P)
Wolfgang Dekant, University of Würzburg (retired), Germany, university (E)
Philip Goodrum, GSI, consultant (E)
Laura Green, Green Toxicology LLC, USA, consultant (E)
Tom Jonaitis, RegTox Solutions Inc., Canada, consultant (E)
Frank Pagone, RHP Risk Management, USA, consultant (E)
Jackie Wright, Environmental Risk Sciences Pty Ltd, Australia, consultant (P)

 

 

Michael initiated the meeting by restating the charge for this phase of the effort, a determination of a provisional safe dose range for PFOA.  He then outlined the process of the call starting with presentations and clarifying questions, and then continuing with discussion and consensus statements.  Afterwards, he invited Group 2 to present.

Harvey went over the findings of Group 2, who collectively decided to build a range based on several studies.  The first study was the Butenhoff et al. (2002) where liver weight increases in monkeys were observed and from which Laura Green and Edmond Crouch developed a benchmark concentration of 19 ug/ml.  Dividing this concentration by an uncertainty factor of 1-fold for experimental animal to human variability (since the effect seen was in primates), and a human clearance factor by Lorber and Egeghy (2011) as reported by Post (2021), resulted in an RfD of 0.27 ug/kg-day. 

The second study considered by Group 2 was Lau et al. (2006) with a NOAEL of 23 ug/ml for dose dependent growth deficits in offspring.  An uncertainty factor for experimental animals to humans of 10 was used as well as a human clearance factor of Lorber and Egeghy (2011) as reported by Post (2021).  The resulting RfD was 0.3 ug/kg-day. 

Other studies considered were Onishchenko et al. (2011), Koskela et al. (2015), Loveless et al. (2006), and Macon et al. (2011).  An uncertainty factor for experimental animals to humans of 10 was used with each study as well as a human clearance factor by Lorber and Egeghy (2011) as reported by Post (2021).  The resulting RfDs varied from 0.011 to ~0.6 ug/kg-day. 

Clarify questions and discussion included:

  • Would not the use of a clearance value from human study Zhang et al. (2013), as describe by Campbell et al. (2022), be a better choice than clearance values from human observational studies described by Lorber and Egeghy (2011)?  Harvey responded that this would be the case and a revision of these tentative RfD values would be appropriate.
  • Would not the use of a database uncertainty factor be reasonable given the large uncertainty in the overall database?  Harvey thought that the data for PFOA was extensive, but perhaps a factor of 3 might be appropriate since a 2-generation study was not available. 
  • Some concern was expressed over the use of the Onishchenko et al. (2011), Koskela et al. (2015) due to the small number of experimental animals.
  • It was also argued that a UF of 0.1 for experimental animal to human toxicodynamics should be applied for differences in response to PPAR activation between rodents and humans.  While appropriate, no one seemed willing to invoke this science-based factor.

Lyle then summarized the findings of Group 1, members of whom were still of the general opinion that the overall database was insufficient at this time to make a reliable judgment of critical effect.  Nevertheless, in order to develop a provisional range, Group 1 focused on two mouse studies, specifically the developmental/reproduction study of Abbott et al. (2007) and the immunotoxicity study of DeWitt et al. (2016), with a range in the NOAELs from 0.3 to 0.94 mg/kg-day. 

Group 1 then developed an RfD range of 3 to 9.4 ug/kg-day from these two values by dividing this range by the classic 100-fold uncertainty factor.  Group 1 also developed a separate range by adjusting the kinetic comparison between mice and humans based on the work of Zhang et al. (2013) to develop a range of 0.3 to 515 ug/kg-day.

Clarify questions and discussion included:

  • Would not the use of a database uncertainty factor be reasonable given the large uncertainty in the overall database?  The general consensus was that such a factor might be needed, but if so, a factor of 3-fold might be more appropriate than a value of 10-foldsince a 2-generation study was not available.
  • The large range in the second RfD calculation appeared to be due to conflating the mouse to human uncertainty factor for toxicokinetic variability with the within human uncertainty factor for toxicokinetic variability.  Perhaps separate these two? Everyone seemed to agree with this.

Michael then went over initial findings of Group 3 based on a conversation with James S.  The group had previously considered dose response, known or suspected Mode of Action (MOA), consistency, coherence between experimental animal and epidemiology data, and robustness of the overall response in their evaluation of potential critical effects.  In general, Group 3 considered liver effect as best meeting these criteria and that the results in monkey were most relevant due to comparability of PPAR-alpha activation for potential liver effects and general physiology with humans, despite the few numbers of animals and some inconsistency with the reported observations.  Group 3 did not state a range in the provisional RfD, but it would be similar to what Group 2 had proposed. 

After these presentations, clarifying questions and discussion, the following consensus positions were developed:

  • The various positions of the three sciences teams appear to overlap, so that developing a range in the PFOA safe dose, based on differing experimental animal studies, seemed reasonable.  The use of human data for this exercise was not entertained, consistent with the consensus of all three science teams from the second conference call.
  • PFOA has an enormous database, but still has some uncertainty, especially in choosing the critical effect.  A factor of 3-fold for this area of uncertainty should be considered.
  • The use of the average clearance value (either mean, median, mode or geometric versions of these) from the Zhang et al. (2013) human study should be used with any of the experimental animal points of departure if in ug/ml of serum, or by comparison with kinetic information from the relevant species if the points of departure are in units of dose.  Moreover, the Zhang et al. (2013) also shows human variability that can be used to develop a data-derived value for within human toxicokinetics.  A preliminary analysis gives this a value of ~9-fold. 
  • The development of a specific range in the PFOA safe dose based on information from this meeting will be tasked to a smaller group of volunteers who will then send around a draft for consensus review.

 

References

Abbott, BD; Wolf, CJ; Schmid, JE; Das, KP; Zehr, RD; Helfant, L; Nakayama, S; Lindstrom, AB; Strynar, MJ; Lau, C. (2007). Perfluorooctanoic acid induced developmental toxicity in the mouse is dependent on expression of peroxisome proliferator activated receptor-alpha. Toxicol Sci 98: 571-581.

Butenhoff, J; Costa, G; Elcombe, C; Farrar, D; Hansen, K; Iwai, H; Jung, R; Kennedy, G; Lieder, P; Olsen, G; Thomford, P. (2002). Toxicity of ammonium perfluorooctanoate in male cynomolgus monkeys after oral dosing for 6 months. Toxicol Sci 69: 244-257.

Butenhoff, J.L., G.L. Kennedy, S.R. Frame, J.C. O’Conner, and R.G. York. (2004). The reproductive toxicology of ammonium perfluorooctanoate (APFO) in the rat. Toxicology 196:95–116.”

Campbell, Jerry, Harvey Clewell, Tony Cox, Michael Dourson, Shannon Ethridge, Norman Forsberg, Bernard Gadagbui, Ali Hamade, Ravi Naidu, Nathan Pechacek, Tiago Severo Peixe, Robyn Prueitt, Mahesh Rachamalla, Lorenz Rhomberg, James Smith, Nitin Verma.  (2022).  The Conundrum of the PFOA human half-life, an international collaboration.  Regulatory Toxicology and Pharmacology 132 (2022) 105185.

Dewitt, JC; Williams, WC; Creech, NJ; Luebke, RW. (2016). Suppression of antigen-specific antibody responses in mice exposed to perfluorooctanoic acid: Role of PPARα and T- and B-cell targeting.  J Immunotoxicol 13: 38-45.

Koskela, A; Koponen, J; Lehenkari, P; Viluksela, M; Korkalainen, M; Tuukkanen, J. (2017).
Perfluoroalkyl substances in human bone: concentrations in bones and effects on bone cell
differentiation. Sci Rep 7: 6841.

Lau, C; Thibodeaux, JR; Hanson, RG; Narotsky, MG; Rogers, JM; Lindstrom, AB; Strynar, MJ. (2006).  Effects of perfluorooctanoic acid exposure during pregnancy in the mouse. Toxicol Sci 90: 510-518.

Lorber, M; Egeghy, PP. (2011). Simple intake and pharmacokinetic modeling to characterize exposure of Americans to perfluoroctanoic acid, PFOA. Environ Sci Technol 45: 8006-8014.

Macon, MB; Villanueva, LR; Tatum-Gibbs, K; Zehr, RD; Strynar, MJ; Stanko, JP; White, SS; Helfant, L; Fenton, SE. (2011). Prenatal perfluorooctanoic acid exposure in CD-1 mice: low-dose developmental effects and internal dosimetry. Toxicol Sci 122: 134-145.

Post, Gloria.  (2021).  Recent US State and Federal Drinking Water Guidelines for Per- and Polyfluoroalkyl Substances.  Environmental Toxicology and Chemistry.  26 August 2020.  https://doi.org/10.1002/etc.4863.

Onishchenko, N; Fischer, C; Wan Ibrahim, WN; Negri, S; Spulber, S; Cottica, D; Ceccatelli, S. (2011). Prenatal exposure to PFOS or PFOA alters motor function in mice in a sex-related manner.  Neurotox Res 19: 452-461.

Zhang, Y; Beesoon, S; Zhu, L; Martin, JW. (2013). Biomonitoring of perfluoroalkyl acids in human urine and estimates of biological half-life. Environ Sci Technol 47: 10619-10627.


After the meeting several members pointed out that a comprehensive two-generation reproductive toxicity study was conducted in Sprague-Dawley Rats by Butenhoff et al. (2004). EPA used this study to help justify a database UF of 1.
  

 

 

2/ 8-9 /2023

Second International Conference Call on Safe Dose Range of PFOA & PFOS


Present (P)/Excused (E):

Chair: Tony Cox with Cox Associates, USA, consultant (P)
Co-Chair: Ravi Naidu with CRC CARE, Australia, NGO (E)
Rapporteur: Michael Dourson with TERA, USA, NGO (P)

Group 1
Lyle Burgoon, RaptorPharmTox, USA, consultant (P)
Paul Nathanail, LQM, UK, consultant (E)
Shanon Ethridge, International Association for Plumbing and Mechanical Officials Research and Testing, USA, NGO (P)
Vijay Kannappan, Cook Medical, USA, industry (E)
Michael Luster, NIOSH (retired), government, USA (P)
Therese Manning, Environmental Risk Sciences Pty Ltd, Australia, consultant (P)
Tiago Severo-Peixi, State University of Londrina, Brazil, University (E)
Andrea Wojtyniak, Geosyntec, Canada, consultant (P)

Group 2
Harvey Clewell, Rambol, Global, consultant (P)
Tamara House-Knight, GHD, Global, consultant (P)
Linda Dell, Ramboll, Global, consultant (P)
James Deyo, Environmental Protection Authority, New Zealand, government (E)
Bernard Gadagbui, Toxicology Excellence for Risk Assessment, USA, NGO (E)
Travis Kline, Geosyntec Consultants, USA, consultant (E)
Katie Richardson, Senversa, Australia, consultant (P)
Anurag Sharma, Nitte University, India, university (E)

Group 3
James Smith, NMCPHC, USA, government (E)
Nitin Verma, Chitkara University, India, university (E)
Wolfgang Dekant, University of Würzburg (retired), Germany, university (E)
Philip Goodrum, GSI, consultant (P)
Laura Green, Green Toxicology LLC, USA, consultant (P)
Tom Jonaitis, RegTox Solutions Inc., Canada, consultant (E)
Frank Pagone, RHP Risk Management, USA, consultant (P)
Jackie Wright, Environmental Risk Sciences Pty Ltd, Australia, consultant (E)

 

Michael initiated the meeting by restating the charge for this phase of the effort, a determination of the critical studies for one or more of PFOA's critical effect(s). Tony then outlined the process of the call starting with presentations and clarifying questions, and then continuing with discussion and consensus statements.  Afterwards, he invited Group 3 to present.

Laura went over the findings of Group 3.  The group considered the following criteria in their evaluation of potential critical effects:

  • Dose response,
  • Known or suspected Mode of Action (MOA),
  • Consistency,
  • Coherence between experimental animal and epidemiology data, and
  • Robustness of the overall response. 

After reviewing the plethora of relevant information, Group 3 did not consider the epidemiology data, composed primarily of observational studies, to be sufficient to determine a critical effect and instead focused on experimental animal work. 

Group 3 considered monkey studies as most relevant due to the closeness to humans with PPAR-alpha activation for potential liver effects and general physiology, and the difficulty in interpretation of rodent developmental effects: 

  • Non-adverse liver effects were seen at all of the doses tested in monkeys (3, 10, 20, and 30 mg/kg-day).  These effects correlated roughly with some non-adverse liver effects seen in the human observational studies and was consistent with the sole human clinical study that, while short term, showed no adverse liver effects. 
  • Although these liver effects were not considered to be adverse in monkeys, mortality was also observed in monkeys at the higher doses leading to a clear NOAEL/LOAEL boundary.  
  • One member of Group 3 reached out to the investigators of the monkey studies to ask for any additional data.  No additional data were available.

Harvey then went over the findings of Group 2.  After reviewing the relevant plethora of information, Group 2 also did not consider the epidemiology data, composed primarily of observational studies, to be sufficient to determine a critical effect. These studies were considered to be:

  • Confounded, and confounding was not readily quantified, which created a hurdle with the use of human data, and
  • Exposures were not significantly different from background in most studies to assign an association, much less causation

Because of these concerns, Group 2 also focused on experimental animals for consideration of the critical effect.

In contrast to Group 3, Group 2 selected rodent developmental studies rather than liver changes, and specifically Lau et al, 2006, as most relevant due to the consistency in response of several rodent species and the fact that the likely MOA was fatty acid mimicry.  This selection was based on:

  • PFOA access to mid-chain fatty acid transport, and biliary and renal excretion and resorption. 
  • While such mimicry might be readily handled by organs, such as the liver, it might more readily disturb fatty acid homeostasis in the developing organism, thus supporting its selection as the critical, or perhaps co-critical effect. 
  • Ppar-alpha induced liver effects occurred in rodents at about a 10-fold higher dose than those evoking developmental toxicity.

Other issues/questions raised included:

  • Laura raised the idea that a developmental study in rabbits might also be worth considering since rodents are known not to representative human development as well.  Group 2 agreed to review the available rabbit developmental study. 
  • Lyle raised the issue of statistical problems with the Lau et al. (2006) study in particular, and for most, if not all of the experimental animal work in general.  
  • Several, but not all, human observational studies show a decrease in birth weight when PFOA is sampled in the 3rd trimester of pregnancy, but not at earlier sampling times.
  • Monkey and rodent NOAEL/LOAEL interfaces are approximately the same at around 1 to 3 mg/kg-day.

Lyle then summarized the findings of Group 1, which were based on a very nice summary of studies by Jackie and Therese (attached).  Like Groups 2 and 3, Group 1 did not feel that any of the human studies were sufficiently reliable to be used to determine the critical effect, and for the various reasons already indicated.  Nor did Group 1 feel that the liver effects seen in monkeys, or perhaps other species, were appropriate, since the effects seen were not adverse.  Nor did Group 1 consider the developmental effects appropriate due to statistical issues mentioned above.

Group 1 was of the general opinion that the overall database was insufficient at this time to make a reliable judgment of critical effect, and supported this position with the observation that different health agencies around the world have come to very different decisions.  While these differences may not be direct evidence for the overall weakness in the database, the WHO (2022) came to the same conclusion.  Specifically, the overall database was too uncertain to determine a scientifically based judgment of critical effect.  Instead WHO (2022) made a risk management recommendation.

After these presentations, clarifying questions and discussion, the following consensus positions were developed:

  • Should human studies be used for the development of the critical effect? No, existing human observational studies cannot be used reliably for this purpose.  For example, changes in cholesterol have only a small effect and are not dose responsive.  These studies may support the choice of critical effect with some of the experimental animal work, however.
  • Should vaccine responses be used for the development of the critical effect?  No, existing human observational vaccine findings are not primary immune responses and not of clinical relevance.  Moreover, higher dose worker exposures do not suggest immune responses.  
  • The overall uncertainty in the database, both epidemiology and experimental animal, is sufficient to give pause to the development of a credible critical effect for PFOA.  This conclusion is similar to what WHO (2022) found and for the same or similar reasons.  However, in recognition of the importance of managing PFOA potential health risk, and despite the overall difficulties in the experimental animal studies, a provisional approach will be explored along the following lines:
    • Frank toxicity in both monkeys and rats has been observed in a dose related manner.  We might be able to tie these effects into other liver and or developmental endpoints.  TERA will conduct a Benchmark Dose (BMD) approach on the relevant monkey and rodent studies, and send this to Groups 1, 2, and 3 for consideration.  Laura asked participants to critique and improve upon her and Edmund Crouch’s work in this regard from 2019 (as submitted to MassDEP, and available at https://greentoxicology.com/Reports/PFAS_comments_to_MADEP.pdf ).
    • PFOA is the fluorinated version of the naturally occurring caprylic acid.  A big difference between these is ½ life in the human body.  TERA will conduct a limited literature review on the toxicity of caprylic acid, and Groups 1, 2, and 3 will conduct a thought experiment to probe whether potential long term toxicity from caprylic acid matches any of the findings with PFOA.
Group 2 will look at the rabbit developmental toxicity study.

The human clinical study of Elcombe et al. (2013) is in the same range and showed no overt effects (50 – 1200 mg/week ÷ 7 days ÷ 70 kg ~ 0.1 – 2.4 mg/kg-day).

Experimental animal work indicates some immune toxicity but only at doses higher than those suggested in human observational studies. 

 

 

1/4/2023

International Conference Call On Safe Dose Range Of Perfluorooctanoate (PFOA) & Its Sulfonic Acid (PFOS)

Present (P)/Excused (E):

Chair: Tony Cox with Cox Associates, USA, consultant (P)
Co-Chair: Ravi Naidu with CRC CARE, Australia, NGO (P)
Rapporteur: Michael Dourson with TERA, USA, NGO (P)

Group 1
Lyle Burgoon, RaptorPharmTox, USA, consultant (P)
Paul Nathanail, LQM, UK, consultant (P)
Shanon Ethridge, International Association for Plumbing and Mechanical Officials Research and Testing, USA, NGO (P)
Vijay Kannappan, Cook Medical, USA, industry (P)
Michael Luster, NIOSH (retired), government, USA (P)
Therese Manning, Environmental Risk Sciences Pty Ltd, Australia, consultant (P)
Tiago Severo-Peixi, State University of Londrina, Brazil, University (E)

Group 2
Harvey Clewell, Rambol, USA, consultant (E)
Tamara House-Knight, GHD, Global, consultant (P)
Linda Dell, Ramboll, Global, consultant (P)
James Deyo, Environmental Protection Authority, New Zealand, government (E)
Bernard Gadagbui, Toxicology Excellence for Risk Assessment, USA, NGO (P)
Travis Kline, Geosyntec Consultants, USA, consultant (E)
Anurag Sharma, Nitte University, India, university (E)

Group 3
James Smith, NMCPHC, USA, government (P)
Nitin Verma, Chitkara University, India, university (P)
Wolfgang Dekant, University of Würzburg (retired), Germany, university (P)
Philip Goodrum, GSI, consultant (P)
Tom Jonaitis, RegTox Solutions Inc., Canada, consultant (E)
Frank Pagone, RHP Risk Management, USA, consultant (P)
Jackie Wright, Environmental Risk Sciences Pty Ltd, Australia, consultant (P)

Observers
Laura Green, Green Toxicology LLC, USA, consultant (P)
Katie Richardson, Senversa, Australia, consultant (P)
Andrea Wojtyniak, Geosyntec, Canada, consultant (P)
 

 

Michael and Tony opened the meeting by restating the charge for this phase of the effort, a focus on PFOA’s mode of action for its critical effect(s), and then arranging for presentations, clarifying questions, and consensus statements.  Afterwards, Group 1 was invited to present.

Lyle went over the findings of group 1 summarized in an excel spreadsheet (to be loaded at website). 

  • Findings included that widely different choices of critical effect and their tentative mode of action (MOA) are evident among national authorities, not all of these critical effects may be relevant to risk assessment, and in particular for the development of a PFOA safe dose range. 
  • For example, Mike opined that while vaccine data might suggest additional testing, the current assortment of vaccine effects is not suitable for developing a Reference Dose (RfD).  This is because small decreases in vaccine responses are not clinically relevant due to the greater variability in these responses amongst the population.  Several folks agreed with this statement. 
  • Paul added that comments on the level of uncertainty for differing choices of critical effect by each government position might be helpful for the next phase of our work (to select a critical effect among the many suggested candidates). 
  • Ravi added that most sample sizes in rodents are too small to be readily helpful for risk assessment. 
  • Michael asked whether membrane fluid dynamics was due to lodging of PFOA into plasma membranes, which might be expected due to its chemical similarity to plasma lipids and limited volume of distribution from the sole clinical study in humans.  Such insertions without associated hydrogen bonding might make such membranes less efficient.

Linda then went over the findings of group 2 summarized in a word document ((to be loaded at website). 

  • Findings included that clinical effects in many of the human observational studies, such as increases in cholesterol and decreases in vaccine titer and birth weight, are of small magnitude or are imprecise. Investigators generally report differences within normal laboratory reference ranges in relation to PFOA blood concentrations.  These findings might reflect pharmacological bias or reverse causality due to the fatty acid mimicry evident in PFOA’s chemical structure.
  • Several of the critical effects found in experimental animals need pPAR activation; since humans and rodents have strikingly different pPAR activations, this has direct relevance to development of a safe dose range for PFOA based on data from experimental animals.
  • A vast inter- and intra-individual human variability in natural vaccine response exists that precludes any definitive statement in regards to the use of this endpoint for it choice as the critical effect.
  • Reverse causality may apply to more than one effect.
  • Answers to questions regarding relevance of these findings and their associated MOAs in humans will not likely come from human studies, but at the same time we need experimental models that more closely resemble humans.

Tony then asked if any conflicts were evident between the presentations and discussion of topics between groups 1 and 2.  None were identified.  Paul then asked whether we could go through Group 1’s list of government positions and opine whether each group was likely to have under- or over-predicted the likely range in the PFOA safe dose.  No one disagreed with this suggestion.

James then went over the findings of group 3 summarized in a word document (to be loaded at website). 

  • Findings included that it was difficult to discuss any particular MOA as the critical effect had not yet been selected. 
  • Overall, we have little information on MOA other than perhaps for the liver effects found in rodents due to pPAR activation.
  • Do we have areas where the dose response information is good enough to develop a safe dose range?  This question led to discussion of inflection points or potentially hormetic responses that might yield useful information, such as human observational studies showing an increase in cholesterol but the sole human clinical study on PFOA showing decreases (Convertino et al., 2018). 
  • Although cholesterol changes did not appear to be definitive and not likely to be the critical effect, studying other inflection points or hormetic responses seemed like a good idea.
  • Humans are much less sensitive than experimental animals to pPAR related events.
  • Membrane fluid dynamics was again raised as a possible clue to the inculcation of PFOA into plasma membranes, which might be expected, if given sufficient dose, to cause a host of effects.  While this is a plausible hypothesis, it was not known how much PFOA would be needed per cell membrane to cause the expected leakage or fluidity.

Tony then led the group in a general discussion.  Items noted included:

  • A SciPinion panel on immunotoxicity of PFOA found that the vaccine threshold of 0.1  IU/ml was not helpful for risk assessment since it is just a guideline.  Additional discussed mentioned that this value is not an appropriate biomarker nor should it be considered as a threshold.  It use in the development of a safe dose is not credible.  No one disagreed with these findings.
  • Could the varying COVID responses world-wide be studied in relation to differing PFOA serum concentrations?  Wolfgang thought that this was likely already happening.
  • Several MOAs could be envisioned but not enough evidence exists to establish any one of these MOAs with certainty. 
  • Certain effects appear to be irrelevant for the determination of a safe dose, specifically cholesterol changes and vaccine status.   No one disagreed with this statement.
  • Studying inflection points or perhaps hormesis might help resolve why we have 100,000-fold differences in the PFOA safe dose internationally.  While differences among such groups can often be a factor of 3 due to differing times of analysis and methods, this difference in PFOA clearly not acceptable, nor can all groups be correct.

Critical effect is defined here as the first adverse effect, or its known and immediate precursor, that occurs as dose is increased. It is recognized that multiple effects may be critical (occurring at or around the same dose), and that critical effects in experimental animals may not reflect these same effects found or expected in humans.  However, if the critical effect is prevented, then it is assumed that all subsequent adverse effects are prevented.  This assumption is well established in the risk assessment community.

 

 

10/31/2022

Notes of the Advisory Committee for International Collaboration on the PFOA/S Safe Doses
Conference Call of October 31/November 1

Present

Lyle Burgoon, Raptor Pharm & Tox, Ltd, USA
Harvey Clewell, Ramboll, Global
Michael Dourson, TERA, USA
Laura Green, Green Toxicology, LLC, USA 
Tamara House-Knight, GHD, Global
Ravi Naidu, CRC CARE, Australia
James S. Smith, US DoD, USA
Nitin Verma, Chitkara University, India

Excused:

Tony Cox, Cox Associates, USA
Eric Saunders, Pfizer, Global

Summary of Discussions:

After introductions by members individually, Michael Dourson went over the ground rules for Alliance for Risk Assessment (ARA) projects.  Briefly stated, ARA projects are to be focused on collaborative work where no party is excluded and all viewpoints are considered.  Additionally, ARA projects are not to be used as a pretext for criticizing a single existing or developing organization’s position.  With this particular project, several agency positions are found to be widely disparate (PFOA safe dose estimates are more than 100,000-fold apart), and so an international collaboration to attempt a narrowing of this range seems reasonable and well within the ARA framework. 

At this point several members of the advisory group described findings of a previous international collaboration on PFOA human 1/2 life (see: https://tera.org/Alliance%20for%20Risk/Projects/pfoahumanhalflife.html). The prior effort was successful in reducing the credible range of the PFOA human ½ life 1 and has been cited in the recent WHO (2022)2 findings. 
Discussion then centered on whether the safe dose ranges of PFOA and PFOS would be addressed together or separately.  The second approach was selected with discussions of PFOA occurring first, in part because of a recent publication on long term follow up with findings from the original PFOA C8 study.3   However, it was recognized by all that many of the insights gained with PFOA would also apply to PFOS. 

Next discussed was the process for the evaluation.  The creation of three or perhaps four separate, and independent, science groups was proposed, based on the success of this approach with the previous international collaboration.  Each group would be given identical tasks and instructions to work without discussion with other groups.  Everyone agreed with this approach and also thought it best to address specific issues sequentially.  Specific issues were determined to be mode(s) of action, choice of critical effect(s), method of extrapolation such as uncertainty factor, and data gaps.  Periodic discussions among all groups was envision to share findings on specific issues and then to attempt a consensus.

Next discussed was the invitation for folks to assist.  It was decided to send an invitation to already identified interested folks, and to make a broader announcement along these lines on social media.  This general invitation would be followed by specific invitations to scientists known to be erudite in certain of the issues. 

The time line for all of this work was recognized to vary depending on the issue under discussion.  For example, discussion on the mode of action might be more easily done then that done on the critical effect.  However, giving the science groups specific deadlines was considered important and would be determined at a subsequent advisory committee meeting.  Interactions with outside parties was also considered to be important but premature at this call. 

Finally, the committee thought that the listing of important references was needed and that the developing website should contain a section devoted to this, but perhaps organized by specific issue.

 

(1.) Campbell, Jerry, Harvey Clewell, Tony Cox, Michael Dourson, Shannon Ethridge, Norman Forsberg, Bernard Gadagbui, Ali Hamade, Ravi Naidu, Nathan Pechacek, Tiago Severo Peixe, Robyn Prueitt, Mahesh Rachamalla, Lorenz Rhomberg, James Smith, Nitin Verma.  2022.  The Conundrum of the PFOA human half-life, an international collaboration.  Regulatory Toxicology and Pharmacology 132 (2022) 105185.

(2.) WHO, 2022. PFOS and PFOA in Drinking-water.  Background document for development of  WHO Guidelines for Drinking-water Quality 29 September.  Version for public review.

(3.) Steenland K, Fletcher T, Stein CR, Bartell SM, Darrow L, Lopez-Espinosa MJ, Barry Ryan P, Savitz 50 DA (2020). Review: Evolution of evidence on PFOA and health following the assessments of the C8 51 Science Panel. Environ Int. 145:106125

 


 

 

Project Contact:
Dr. Michael Doursoni
dourson@tera.org

Interested Partners:

ACS

CH2MHill
Gradient

Hull and Associates

TERA