                  	Office of Chemical Safety and United States	Pollution Prevention
                  Environmental Protection Agency	April 2019

                                       
                                       
                                       
Summary of External Peer Review and Public Comments and Disposition for the Environmental and Human Health Hazards for Persistent, Bioaccumulative and Toxic (PBT) Chemicals







                 Draft to Support Proposed Rule: RIN 2070-AK34
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                  April 2019 

This document summarizes the public and external peer review comments that the EPA's Office of Pollution Prevention and Toxics (OPPT) received for the environmental and human health hazards for the persistent, bioaccumulative, and toxic (PBT) chemicals. It also provides EPA/OPPT's response to the comments received from the public and the peer review panel.
EPA/OPPT appreciates the valuable input provided by the public and peer review panel. The input resulted in numerous revisions to the hazard summary.
Peer review charge questions are used to categorize the peer review and public comments into specific issues related to the five main themes.  
 Organization and clarity of the document
 Methodology for collecting and summarizing information
 Representativeness and completeness of presented information
 Appropriateness and sufficiency of information
 Reliability and relevance of information
All peer review comments for the nine charge questions are presented first, organized by charge question in the following section. These are followed by the public comments. For each theme, general comments pertaining to all chemicals are presented first, and then additional comments pertaining to only one or several chemicals follows.

Organization and Clarity of the Document - Peer Review Comments
Charge question 1: This information, when finalized will inform a technical support document to support EPA's regulatory activities on PBTs. Please comment on the organization and structure of this document to inform this use. Do you have specific recommendations to improve clarity and presentation of this information?
                                       #
Summary of Peer Review Comments for Specific Issues Related to Charge Question 1
                               EPA/OPPT Response
                                       1
Summarization and Conclusions: 
 There is no discussion of the data presented. For example, there is no discussion of the weight of the evidence for each endpoint, the possible relevance to humans, or the adequacy of the database. 
 The executive summary of this document is just a summary of the purpose of the document and is not a summary of the contents of the document. The executive summary should be expanded to include a short summary of the hazard findings for each of the chemicals.
 A summary that compares the margin of exposure in the animal studies or environmental studies to human exposure (from the Exposure and Use Assessment, if an estimated exposure has been identified) would be useful, particularly because that aspect of the risk assessment is not being completed in this expedited process.
 The addition of a summary table of physical-chemical properties for each chemical would enable the reader to more easily compare exposure dose to solubility, route of exposure and distribution within the organism to log Kow, etc.  Addition of any available persistence information and metabolic pathways would also be important. 
 The environmental hazard assessment should be structured similarly to the human health summary, by including explicit NOAELs and LOAELs.
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.
 Margins of exposure are not necessary to characterize hazards of the five PBT chemicals.
 Physical-chemical properties, as well as exposure and persistence information are available in the Exposure and Use Assessment.  Including this information in the hazards document would be redundant.   
 The executive summary now includes a short summary of hazard findings for each of the chemicals.
                                       2
Editorial:
 The document should be carefully reviewed for errors (e.g., the environmental hazard summary table has some cells that use "per cent" instead of %). Fixing errors and ensuring consistency will make the document easier to understand for the reader.
 EPA has performed a review of the document to remove errors and ensure consistency in the text.
                                       
DecaBDE-specific comments:
 Acute toxicity is reported at levels above the aqueous solubility value listed in Table 4.1.  
 Summary text should reflect the key results of Table 4.1
 Many Deca-BDE studies use commercial products that do not specify the percent purity, nor the other components of the product, which may enhance solubility of Deca-BDE. Thus, the commercial product toxicity data may be confounded by the additional chemical ingredients. Section 4.1 discusses instances where the data are from commercial mixtures.
 Summary text was updated to reflect key results of Table 4.1.
                                       
HCBD-specific comments:
 No aqueous solubility data is given in the table as was done for decabromodiphenyl ether.
 Aqueous solubility and other physical-chemical properties are available in the exposure and use document.
                                       
PIP (3:1)-specific comments:
 The first paragraph on page 20 provides useful information on the chemical identify of PIP. Since the information applies to both the environmental and human health sections, this paragraph should appear before section 6.1.
 The document tried to evaluate a multi-chemical product rather the constituents of the product, and thus the purported hazard was unclear.  Indeed the CASRN is actually isopropyl phenyl phosphate. This is a single compound and not a mixture as reported in the assessment.  Hazard assessments for the specific constituents would provide a better mechanism for regulators to assess the risks of the mixture. Perhaps a strategy for pesticide registration, which selects an "active ingredient" within the product, may improve clarity.
 In the human health hazard summary of PIP (3:1), the first sentence refers to hazard data for "Isopropylated, phosphate (3:1)". If this is a typo, please fix it. If it is intentional, please note whether this is a synonym (although it isn't labeled as such in the exposure and use assessment), or if a different chemical was assessed for hazard data. 
 In the environmental hazard summary, the % of PIP (3:1) contained in the test mixture is included in the summary table. This information should also be included for the human health hazard summary. If it is not available, this should be stated in the text or in the table.
 In Table 6-2, there is no information about the dose timing or number of days of dosing for the OECD 422 oral gavage study in Sprague-Dawley rats. This information should be added to the table.
 Comment has been addressed in the editing of the document.
 Hazard data for PIP (3:1) and related compounds is generally scarce. EPA provided all hazard data it located in order to provide as much information as possible regarding the hazards associated with the family of PIP (3:1) structures. EPA agrees that a component-based or `active ingredient' type hazard assessment would be ideal; however, EPA did not locate data that would support such an analysis. 
 A review of the document was performed to remove errors and ensure accuracy and consistency in the text and tables.
 OECD 422 is a guideline with a specific dosing protocol. Restating this information is unnecessary for the summary table and has not been done for other guideline studies.
 The percentage of Isopropylated, phosphate (3:1) is generally unavailable because most of the studies used commercial formulations.
 Table 6-2 was edited to include number of days of dosing duration as part of the OECD 422 oral gavage study in Sprague-Dawley rats.
                                       
PCTP-specific comments: 
 The title of Table 8-2 is misleading. It should indicate that only PCTP metabolites or breakdown products are included. 
 Clarifying text was added to Table 8-2.
                                       
2,4,6 TTBP-specific comments:
 In the human health hazard summary, the second paragraph contains a sentence that states "One unpublished OECD 422 guideline study report observed reduced body weights in the offspring and increased postnatal." Please finish this sentence to describe what was increased postnatally.
 Clarifying text was added.

Methodology for Collecting and Summarizing Information - Peer Review Comments
Charge question 2: Please comment on the methodology for collecting and summarizing information from secondary sources. Was the approach appropriate? Is the description for how and where the literature was obtained clearly described?
                                       #
Summary of Peer Review Comments for Specific Issues Related to Charge Question 2
                               EPA/OPPT Response
                                      11
Specific details needed in description of search:
 Did the search strategy supplement the reviews by searching forward from the date of the review (allowing for suitable overlap)?
 Which chemical identifiers and chemical names and synonyms were used as search terms?
 Did the search strategy include PubMed, ToxNet, or Toxline? What was the rationale for database selection and how were data from different databases weighted? For example, was data from ECOTOX considered as higher value than other sources?
 Did the search strategy include the National Toxicology Program, Report on Carcinogens, or National Academy of Sciences?
 What were the dates of the searches?
 Information on the search strategy is outlined in Section 3: Approach for Surveying the Chemical-Specific Hazard Data. No forward searching was conducted. No restrictions were placed on the search results.
                                      12
Recommendations for Scope of Search:
 The assessment should have been a systematic review.
 Specific searches for vulnerable populations should have been included.
 The compilation of information from secondary sources leaves much room for error in the interpretation of the primary literature and in the aggregation of data.  For example an average or median LC or EC value loses ALL of the structure of the underlying data set(s).  Also, the ECHA 2018 references are basically lengthy fact sheets and as such are useless for review purposes.  Actual study data are needed for hazard or risk assessments.  If these are proprietary studies from manufacturers, they still must be made available to US regulators and appropriate reviewers before serious consideration can be given to them. 
 It was stated in the review document that Web of Science was utilized as a secondary literature review source. An additional resource should be Scifinder Scholar produced by Chemical Abstract Services (CAS).  Conducting a search using the terms BPE toxicity and refining that to "review," a number of documents were identified that were not included in this search.  
 A risk evaluation to support a 6(h) action for PBTs was not conducted. The presented data provide an overview of the range of hazard information available for each category of chemicals. The hazard summary is not a full hazard assessment that would be used for a risk evaluation. Systematic review and additional full searches were not conducted for the purpose of the hazard summary in support of 6(h). 
 The literature searches were broad enough to identify vulnerable populations.
 Aggregate data are adequate for a literature survey intended to characterize types and magnitude of effects and not to identify specific points of departure or dose-response analyses.
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.
       
                                       
Recommendations for Presentation of Search:
 The search strategy should explain the general procedure for the collection and organization of the literature data for the first chemical and mention only differences or unique cases for all the other compounds.
 Information on the search strategy is outlined in Section 3: Approach for Surveying the Chemical-Specific Hazard Data.
                                       
PIP (3:1)-specific comments:
 Many of the references, especially the human health summary, refer to dossiers prepared by the manufacturer. These references adequately support the conclusion that PIP is toxic. However, it is preferable to review and cite original references for key studies if they are readily available, such as data reported to EPA (TSCATS).
 For this particular agent(s), it was very unclear which compound was being evaluated.  Although an alkylated phosphate, triphenyl phosphate is structurally different than 68937-41-7 and has a host of adverse developmental effects already reported (van der Veen and de Boer 2012).  It also would likely undergo significantly different biotransformation and disposition. In the exposure and use assessment the CASRN is 2502-15-0. The CASRN being used for this assessment of PIP (3:1) should be clarified.
 The search strategy collected publicly available hazard information and did not apply quality or source limitations on the search results. Whenever possible, original references were cited, however, many of the studies were accessed from databases containing only summaries of the data.
 Hazard data on PIP (3:1) as a discrete substance is largely unavailable. Most studies tested PIP (3:1) that also contains triphenyl phosphate (TPP) in commercial product.  TPP is mentioned because of its presence in a majority of the publicly available studies with PIP (3:1).  Comments for the exposure and use assessment are not applicable here.
 Hazard data for PIP (3:1) and related compounds are generally scarce. EPA presented all hazard data it located to provide as much information as possible regarding the hazards associated with the family of PIP (3:1) structures.
                                       
PCTP-specific comments: 
 Using studies of pentachloronitrobenzene as a surrogate for 133-49-3 is unacceptable.  Renner and Nguyen 1982 clearly show that 133-49-3 is 16 times higher in toxicity than Pentachloronitrobenze. Additional software Scifinder Scholar should be an additional source for publically available literature.
 The Renner and Nguyen data were obtained from acute intraperitoneal exposures. Intraperitoneal exposure is not an appropriate route of administration for toxicity testing. 
 Per TOXNET, pentachlorothiophenol (CAS 133-49-3) has a rat oral LD50 of 11,900 mg/kg/bw and a mouse oral LD50 of 1,000-4000 mg/kg/bw. According to Extoxnet, oral LD50 values for pentachloronitrobenzene (the surrogate cited in the human health hazard summary) varies in rats and range from 1500 mg/kg/bw when administered in oil to-12,000 mg/kg/bw when administered in water. These data suggest both chemicals are category 4 acute toxicity hazards and are appropriate analogues. Thus, looking at the relevant oral exposure pathway, the two chemicals have similar acute toxicity. 
 Information on the search strategy is outlined in Section 3: Approach for Surveying the Chemical-Specific Hazard Data; Scifinder would not identify additional information from the abovementioned hazard endpoints.


Representativeness and Completeness of Presented Information - Peer Review Comments
Charge question 3: Please comment on the representativeness and adequacy of the selected literature used in the evidence tables and described in the text to summarize the hazard concerns. Are there additional primary peer-reviewed and publicly available literature that warrant further consideration? If so, please identify or provide these sources and comment on whether acquisition and summarization of this additional information would change the hazard characterization already provided in the existing summaries? Does the information presented clearly and accurately represent the hazards associated with each chemical? Are there any studies that should be removed from the summary and evidence tables due to study quality or data transparency concerns?
                                       #
Summary of Peer Review Comments for Specific Issues Related to Charge Question 3
                               EPA/OPPT Response
                                      19
Overall comments about representativeness of data:
 There are more studies available for some of the chemicals (e.g., decaBDE) than are provided in the summary tables. This is consistent with the stated purpose of providing a summary, not a comprehensive literature review. However, it is not clear how the studies that were put into the summary tables were chosen. The reason for study choice should be explicitly stated. 
 A better description should be provided explaining why only repeat-dose studies are relevant to these chemicals ("given the PBT nature of the chemicals of interest"  -  page 1 of the charge question document). Not all the studies listed for the human health assessment are repeat-dose studies  -  e.g., for decaBDE Johansson et al. (2008) or Viberg et al. (2007).
 The nature of this document as a hazard summary, instead of a comprehensive hazard assessment, makes it very likely that there will be additional studies that were not identified in the summary. The most important consideration is to ask whether enough studies have been identified in order to perform an adequate weight-of-evidence evaluation to determine the hazards. Looking at a single study can lead to erroneous conclusions about hazard. It is important to collect all key studies for a chemical and evaluate all the findings from those studies to reach a balanced conclusion about hazard. 
 A full systematic review process was not used or required under 6(h) for the development of the hazard summaries. The studies identified by panelists were incorporated in the document.  
 The sentence containing the phrase "given the PBT nature..." was amended to clarify that acute studies were used to document hazard concerns for the PBTs.
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA believes that an adequate number of key studies across hazard endpoints have been identified and summarized to characterize the hazards for DecaBDE and HCDB.
 Other chemicals were so data poor that EPA has presented all the information it identified to gauge reproducibility, concordance and/or qualitative weight of evidence, if possible, in characterizing hazards.
                                       
DecaBDE-specific comments:
 The current human health hazard summary is not representative of the literature for decaBDE hazard identification, because available summary documents were not used, and a primary literature search was not conducted. Since this decaBDE hazard summary does not complete a full review of the primary literature, there are many studies not cited here (e.g., ATSDR (2017) cites 91 studies on decaBDE for human health-related toxicity, whereas this hazard summary cites 14). As noted above, the reason for EPA's choice of these 14 particular studies is unclear and should be explained.
 Additional studies: 
             The National Research Council assessed the health risks of flame retardant chemicals including Decabrom (NRC 2000). This assessment includes a review of the health effects of Decabrom.
             Reviews by Hardy et al. 2009, Xu et al. 2018, and Lu et al. 2018.
             Missing key studies identified by Health Canada (2012) and ATSDR (2017) that pass study quality criteria.
 Acquisition of this information and assessment of study quality and weight-of-evidence does change the hazard characterization. In this hazard summary EPA concludes that "Oral repeated dose animal data for DecaBDE indicate developmental neurological effects, developmental immunological effects, general developmental toxicity, and liver effects." And "Dose-related brain effects were reported in adult rats as well, which was demonstrated by a decrease in brain weight following 28-days of oral gavage (Van der Ven et al., 2008)."  The reviewer provides an alternate interpretation of the evidence based on their own search and review.
 Based on study quality review Liang et al. (2010) and Viberg et al. (2003) should be removed, because their study quality is unacceptable. Hsu et al. (2006) and raw results from van der Ven et al. (2008) were not available, so these should be removed. Several of the key studies identified in the Health Canada (2012) and ATSDR (2017) assessments should also be removed because of unacceptable study quality (Zhang et al. (2013), Xing et al. (2009)).
 The additional studies cited in this comment have been added to the document.
 A study quality analysis was not conducted for this hazard summary. 
 The fact that well recognized, authoritative peer-reviewed assessments such as those by Health Canada and ATSDR have relied on certain studies that the commenter finds unacceptable highlights that differences in interpretation may occur depending on the context or specific purpose of an assessment in which they are used. Because the commenter does not provide information on what purpose nor the approach or criteria by which the cited studies were deemed unacceptable, EPA cannot comment on the basis for these assertions. The purpose of EPA's Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. The inclusion of the studies cited aid in accomplishing this goal.
   
                                       
HCBD-specific comments:
   ::	There is no Health Canada (2012) assessment of HCBD  -  the study cited in the hazard summary is the decaBDE assessment. Presumably the Health Canada (2000) study was intended to be cited  -  "Priority Substances List Assessment Report: Hexachlorobutadiene".
   ::	Additional reviews/assessments to consider (including their key studies):
             US EPA Office of Water 2003 Health Effects Support Document on Hexachlorobutadiene.
             Taylor et al. 2003
  ::	Acquisition of this information and assessment of study quality and weight-of-evidence does change the hazard characterization. The reviewer provides an alternate interpretation of the evidence based on their own search and review. 
  ::	Based on study quality review, Gage (1970) (identified in US EPA 2007) should be removed because the study quality is unacceptable, and Field et al. (1990) should be removed, because this study does not appear to be available.
 The additional studies cited in this comment have been added to the document.
 A study quality analysis was not conducted for this hazard summary.
                                       
PIP (3:1)-specific comments:
 The database on PIP is relatively data poor. Read-across methods were used to assess PIP exposure. A similar approach might be considered for the human health or ecological effects document, as well.  At a minimum, the toxicity of the class of aryl phosphates should be acknowledged.
 Because TPP is generally included in the mixture with PIP (3:1) and TPP has its own known toxicological effects, data should be presented that describes hazard assessments of TPP so that the hazards from each can be compared to determine what may be a PIP (3:1)-specific effect. 
 The hazard summary references both ECHA 2018a and 2018b for PIP (3:1). However, only 2018b should be referenced  -  2018a is for 2,4,6 TTBP.
 Additional reviews/assessments to consider (including their key studies):
             UK  2009: Environmental risk evaluation report: Isopropylated triphenyl phosphate (CAS nos. 28108-99-8, 26967-76-0 & 68937-41-7)
 Acquisition of this information and assessment of study quality and weight-of-evidence does change the hazard characterization. The reviewer provides an alternate interpretation of the evidence based on their own search and review. 
 Based on study quality review, Patisaul et al. (2013) (identified in US EPA 2015) should be removed because the compound tested (FireMaster 550) seems to contain mostly non-PIP (3:1) chemicals. 
 The chemicals identified under 6(h) are identified by CAS nos. and the hazard summary focuses specifically on those CAS nos. with other information used as needed.
 The hazard summary references were updated so that only ECHA 2018b is referenced for PIP (3:1).
 UK  2009: Environmental risk evaluation report: Isopropylated triphenyl phosphate (CAS nos. 28108-99-8, 26967-76-0 & 68937-41-7) were incorporated.
 Assessing the hazard of TPP is beyond the scope of this assessment as EPA is not using a `read-across' approach for characterizing hazards for PIP (3:1), however it is mentioned because it is contained in many studies with PIP (3:1), i.e., it provides transparency in interpreting results
 Patisaul et al. (2013) was removed.

PCTP-specific comments: 
 The exposure and use assessment Section 8.2 stated that PCTP has been banned in most parts of the world because it forms several teratogenic decomposition products. The information about PCTP's decomposition products and their potential teratogenic properties is very important to include in this hazard assessment.  
 The ToxNet Hazardous Substances Database (referenced in the Exposure and Use Assessment) provides some toxicological information about PCTP. Many of the references are for studies in the ECHA IUCLID database.
 Acquisition of this information and assessment of study quality and weight-of-evidence could completely change the hazard characterization of PCTP, because the current characterization is not based on PCTP data. 
 No relevant repeated-dose animal toxicity studies or human data were available for PCTP, thus analogous chemicals that are known to metabolize or degrade into PCTP were included in the hazard summary.  Characterization of the hazards of decomposition products is outside the scope of this assessment since the 6(h) action is specific to PCTP. However, any actions to reduce exposure to PCTP would commensurately reduce the possibility of formation of decomposition products. The studies identified by panelists were incorporated in the document.  
 The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking. A hazard summary was developed under 6(h) while a full hazard assessment and risk evaluation were not completed. 
                                       
2,4,6 TTBP-specific comments:
 Acquisition of additional primary and summary information and assessment of study quality and weight-of-evidence does not change the hazard characterization.
 The aquatic studies indicate that TTBP is quite toxic. The document states: "No toxicity data for terrestrial species were identified." The paucity suggests further investigation of exposure and toxicity are warranted.  Perhaps some structure activity relationship work or investigation of the effects of other butylated or alkylated phenols is merited.  
       
 The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking. 
 The Hazard Summary focuses on the specific PBT chemicals such as 2,4,6 TTBP and is fit for purpose for the type of analysis performed in support of 6(h) rulemaking. A full hazard assessment and risk evaluation were not conducted for this rulemaking.


Appropriateness and Sufficiency of Information - Peer Review Comments
Charge question 4: Has appropriate and sufficient information been obtained from the literature to support the identification of hazards?
                                       #
Summary of Peer Review Comments for Specific Issues Related to Charge Question 4
                               EPA/OPPT Response
                                      25
Overall comments about the sufficiency of the data:
 Appropriate and sufficient information was not obtained from the literature to support a weight-of-evidence evaluation of the hazards. Many of the studies use compounds that have similar structures but do not use the chemical being considered.  Some of the studies compiled are from databases where the primary literature or reports are not identified, which should disqualify them from consideration.  
 The current summary does not present the hazards with enough detail to accurately portray the nuances of the data. Different levels of confidence could be expressed in the endpoints, for example strong evidence of hepatoxicity from adult or developmental exposure; moderate evidence for thyroid impacts; limited evidence of immunotoxicity. That would help provide more information to the audience about the confidence of the hazard identification. 
 The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking. hazard assessment for a risk evaluation A full hazard assessment and risk evaluation were not conducted for this rulemaking. When available, EPA obtained data on the chemical itself. However, as noted by the reviewer, the available dataset for some of the chemicals lacked chemical-specific data, so EPA presented data for analogs and metabolites to fill data gaps.
 While EPA acknowledges that the literature summary presented is not exhaustive, for data rich chemicals EPA has presented an adequate number and characterization of toxicity studies to characterize types and magnitude of toxicity endpoints. For data poor chemicals, EPA has presented all the information it identified to gauge reproducibility, concordance and/or qualitative weight of evidence, if possible in characterizing hazards.
       
                                       
DecaBDE-specific comments:
 The studies summarized by the Agency end at 2014. While there have been additional studies on DecaBDE since 2014, in general, most of the studies and endpoints summarized by the Agency are relevant to the hazards under consideration and there are a sufficient number to draw conclusions regarding DecaBDE-induced hazards subject to the PBT assessment.
 EPA incorporated additional studies published after 2014 for DecaBDE.
                                       
HCBD-specific comments:
 The review reports that EPA considers HCBD a "possible" human carcinogen based on kidney adenomas and carcinomas. What exactly does this mean? Is "possible carcinogen" sufficient to support a regulation? 
 This language is a weight of evidence determination using the Guideline for Carcinogen Risk Assessment (1986), and can specifically address chemical agents that have limited animal evidence and human data (web accessible at: http://ofmpub.epa.gov/eims/eimscomm. getfile?p_download_id=439779). 
                                       
PIP (3:1)-specific comments:
 Although not all of the available data are cited, the toxicity of PIP, especially neurotoxicity, is well established.
 As a class, triaryl phosphates are generally neurotoxic, and structure-activity relationships have been identified. This raises the question as to whether other aryl phosphates, including triphenyl phosphate, should be included in this rulemaking process.
 TPP is not one of the identified PBTs under 6(h), so a hazard summary was not specifically developed, however it is mentioned because it is contained in many studies with PIP (3:1).
 This is consistent with the hazard summary EPA presents in Section 6.2: Human Health Hazard Summary (Phenol, Isopropylated, Phosphate (3:1)), of this document. 
 The scope of the TSCA section 6(h) rulemaking is determined according to specific criteria outlined in the statute and summarized in the Background section of this document. Evidence for neurotoxic effects for PIP (3:1) are presented in section 6.2 of this document. 
                                       
PCTP-specific comments: 
 In the human health hazards summary, it is noted that PCTP is a metabolite and/or breakdown product of other chemicals. Only data on the parent compounds were identified. Is there any evidence that PCTP is more toxic than the parent compounds or that it is the active metabolite responsible for toxicity? Otherwise, the toxicity of PCTP in humans has not been established.
 The studies and endpoints summarized by the Agency are relevant to the hazards under consideration. However, given the limitations in the database of studies, it will be especially important for the Agency to provide a thorough uncertainty analysis in its risk characterization for PCTP. 
 The use of surrogate compounds is highly uncertain for human health assessment and few ecological endpoints for 133-49-3 suggest uncertain information for hazard assessments.
 No relevant repeated-dose animal toxicity studies or human data were available for PCTP, thus analogous chemicals that are known to metabolize or degrade into PCTP were included in the hazard summary.  Decomposition products of the PBT chemicals are not being evaluated in the hazard document.  Likewise, surrogate compounds are not being used to assess the human health or ecological hazard for the PBT chemicals.
 The hazard summary is not supporting a risk evaluation for a which an uncertainty analysis would be conducted. 

                                       
2,4,6 TTBP-specific comments:
 GLP studies support the identification of hazardous chemical to ecological receptors.  Bioaccumulation data was not present.   For human health assessment, one peer-reviewed study and few ECHA documents suggested hazard, but confirmation with peer-reviewed literature is uncertain.
 Identification of hazards associated with 2,4,6-TTBP is consistent with the information presented in Section 7.2: Human Health Hazard Summary (2,4,6-Tris(tert-butyl) Phenol) of this document.


Reliability and Relevance of Information - Peer Review Comments
Charge question 5: Please comment on the reliability and relevance of the identified data sets for each chemical and on the strengths and weaknesses of individual data sets that will help to better inform future regulatory actions.
                                       #
Summary of Peer Review Comments for Specific Issues Related to Charge Question 5
                               EPA/OPPT Response
                                      26
Overall comments about the reliability of the data:
 Additional information on the selection process for these chemicals is needed. Why are data on persistence and bioaccumulation not included in the reviews? Why were these selected over other chemicals?
 There is no assessment provided of the study quality and methods that might impact interpretation, such as test material purity, solubility and stability, and study design, e.g., guideline study, GLP, and other methodological considerations affecting study quality. 
 The health effects summary should be strengthened to include a discussion of weight of the evidence with some conclusions.
 The scope of the TSCA section 6(h) rulemaking is determined according to specific criteria outlined in the statute and summarized in the Background section of this document and the NPRM. Persistence and bioaccumulation characteristics of the chemicals are such criteria and were considered in identifying the chemicals subject to the 6(h) action, as outlined in the Background section of this document.  Further details regarding how the PBT chemicals were evaluated to be listed in the TSCA Work Plan can be found in EPA's TSCA Work Plan for Chemical Assessments: 2014 Update (web accessible at: https://www.epa.gov/sites/production/files/2015-01/documents/tsca_work_plan_chemicals_2014_update-final.pdf). 
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. EPA did not conduct a systematic review, or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.
                                       
DecaBDE-specific comments:
 The Agency should consider that most of the identified studies are not guideline studies conducted under good laboratory practices and so it is often not possible to confirm in these non-GLP studies key critical considerations for establishing an accurate dose or exposure response relationship for observed effect. Therefore, in its weight-of-evidence assessment, the Agency should place more consideration on those studies conducted following established guidelines and conducted under good laboratory practices.
 Many of the decaBDE studies treated animals with oral gavage using a lipophilic medium. This both increased the acute dose (a gavage dose is administered all at once so the body has more difficulty recovering from it), and it increases bioavailability in a manner that may not be relevant to exposure in the general public (with the possible exception of breast feeding babies). This should be addressed in decisions about hazards from decaBDE.
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Because TSCA does not require a risk evaluation be conducted to support a 6(h) action for PBTs, EPA does not deem it necessary to conduct systematic review, weight of evidence analyses to characterize hazard (toxicity) for the five PBT chemicals.  
 Analyzing the pharmacokinetic differences between exposure methods and study quality analysis is outside the scope the hazard summary. Because the hazard data presented herein will not be used to conduct dose-response analyses or identify points-of-departure, PBPK differences introduced by dosing regimens is not critical to hazard identification.
       

                                       
HCBD-specific comments:
 The dataset for HCBD is sufficiently comprehensive for the desired review. The Agency should consider that most of the identified studies are not guideline studies conducted under good laboratory practices and so it is often not possible to confirm in these non-GLP studies key critical considerations for establishing an accurate dose or exposure response relationship for observed effects. Therefore, in its weight-of-evidence assessment, the Agency should place more consideration on those studies and technical reports conducted following established guidelines and conducted under good laboratory practices (GLP) or principles of GLP for older studies and reports conducted prior to implementation of GLP.
 Peer-reviewed, government and manufacturer-based data were provided and used to provide necessary NOEC, NOAEL and other thresholds that could be used to derive hazard.  Strengths of the data sets were the numerous sources of data that have presumably been vetted by the Agency. However, a noticeable weakness would be the lack of description of inclusion/exclusion criteria for non-governmental documents or the peer-reviewed literature. 
 Bioaccumulation description was not present.
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.
       Persistence and bioaccumulation characteristics of the chemicals are criteria that were considered previously in identifying the chemicals subject to the 6(h) action, as outlined in the Background section of this document.  Further details regarding how the PBT chemicals were evaluated to be listed in the TSCA Work Plan can be found in EPA's TSCA Work Plan for Chemical Assessments: 2014 Update (web accessible at: https://www.epa.gov/sites/production /files/201501/documents/tsca_work_plan_ chemicals_2014_update-final.pdf). EPA agrees that the dataset for HCBD is sufficient for the desired purpose. 

                                       
PIP (3:1)-specific comments:
 The dataset for environmental and human health hazards for PIP includes a large number of guideline studies conducted under GLP and is sufficiently comprehensive for the desired review.  However, the Agency should obtain copies of the original study reports to review the accuracy of the data and information reported in the ECHA database.
 Limited strengths are present for this assessment.  There is only one peer-reviewed study provided on a product that was 28% triphenylphosphate.  Thus, the literature is extremely limited.
 The data used is what was publicly available during the initial data extraction; original study reports, especially those from other jurisdictions, are not always readily attainable.
 EPA agrees that the dataset for PIP (3:1) is sufficient for the desired purpose. The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.
 EPA agrees the data are limited for PIP (3:1); however, as noted by another commenter, neurotoxicity of PIP (3:1) is well established, consistent with the reports summarized in this document.
                                       
PCTP-specific comments: 
 The dataset for environmental and human health hazards for PCTP is minimal and appears limited. The Agency should obtain original study reports to assess the quality and validity of the information in order to determine the most reliable data for hazard and risk assessment. In its weight-of-evidence assessment, the Agency should place more consideration on those studies and reports conducted following most closely established guidelines and principles of GLP.  
 EPA agrees that the dataset for PCTP is limited, but believes it is sufficient for the desired purpose. The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.
                                       
2,4,6 TTBP-specific comments:
 The dataset for environmental and human health hazards for 2,4,6 TTBP includes both guideline GLP studies and non-guideline studies.  In its weight-of-evidence assessment, the Agency should place more consideration on those studies and reports conducted following established guidelines and conducted under GLP or following GLP principles with adequate information and data provided to confirm such.  In addition, the Agency should obtain copies of the original study reports to review the accuracy of the data and information reported in the ECHA database.
 GLP studies indicate hazard, but because few peer-reviewed studies have been performed with this specific compound, there is significant uncertainty with regard to hazard.
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking. EPA agrees that the dataset for 2,4,6 TTBP somewhat limited, but believes it is sufficient for the desired purpose. 



Organization and Clarity of the Document - Public Comments
Charge question 1: This information, when finalized will inform a technical support document to support EPA's regulatory activities on PBTs. Please comment on the organization and structure of this document to inform this use. Do you have specific recommendations to improve clarity and presentation of this information?
                                       #
  Summary of Public Comments for Specific Issues Related to Charge Question 1
                               EPA/OPPT Response
                                       1
Editorial:
 ACC Recommendation: EPA should establish the human health and ecological health thresholds for each chemical in order to be able to determine what level of exposure represents an unreasonable risk in order to ascertain what exposure reduction is practicable.
 Establishing human health and ecological health thresholds as to demonstrate unreasonable risk for each chemical would be conducted in a risk evaluation. Risk evaluations are not required under 6(h). The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.
                                       
B. EPA should leverage information generated from the first 10 chemicals for risk evaluation where it can.
 ACC Recommendation: EPA should leverage the information generated from the first 10 chemicals for risk evaluation where it can. For example, it was noted that hexachlorobutadiene (HCBD) is primarily generated as a by-product of the manufacture of chlorinated hydrocarbons, particularly perchloroethylene, trichloroethylene and carbon tetrachloride -- each of which is the subject of one of the first ten risk evaluations. EPA should consider information associated with relevant risk evaluations where appropriate. 
 EPA did not conduct risk evaluations of these chemicals for this rulemaking. 
 EPA did leverage information generated from the first 10 chemicals where it was possible.

Methodology for Collecting and Summarizing Information - Public Comments
Charge question 2: Please comment on the methodology for collecting and summarizing information from secondary sources. Was the approach appropriate? Is the description for how and where the literature was obtained clearly described?
                                       #
  Summary of Public Comments for Specific Issues Related to Charge Question 2
                               EPA/OPPT Response
                                      12
Recommendations for Scope of Search:
 EPA should apply a similar structural and analytical approach to the Section 6(h) PBT chemicals even though EPA is not required to conduct a risk evaluation on those chemicals.
 ACC Recommendation: Many useful lessons may be learned from the draft risk evaluations prepared by EPA on the initial 10 chemicals for risk evaluation under TSCA. It would be useful for EPA to apply a similar structural and analytical approach to the Section 6(h) chemicals regardless of the fact that EPA is not required to conduct a risk evaluation on those chemicals.
 EPA is required to address the risks of injury to health or the environment that may be presented by these chemicals. Consequently, many aspects of the TSCA risk evaluation process may be valuable in determining those risks. For example, before EPA can identify appropriate risk management for Section 6(h) PBT chemicals, EPA should consider controls already in place under other EPA regulatory programs implementing other EPA-administered statutes. To do so, EPA should review exposure pathways of these chemicals and the risks associated with those.
 EPA should apply a similar structural and analytical approach to the Section 6(h) PBT chemicals even though EPA is not required to conduct a risk evaluation on those chemicals.
 ACC Recommendation: EPA should ensure that it understands how other federal laws not administered by EPA may already be managing the risks of these PBT chemicals for certain exposure pathways. Again, EPA should assess the risks associated with these chemicals under certain exposure pathways in order to do so.
 Risk evaluations were not conducted for the PBTs. A hazard summary was developed to address the hazards of each chemical identified under 6(h).
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking.







Representativeness and Completeness of Presented Information - Public Comments
Charge question 3: Please comment on the representativeness and adequacy of the selected literature used in the evidence tables and described in the text to summarize the hazard concerns. Are there additional primary peer-reviewed and publicly available literature that warrant further consideration? If so, please identify or provide these sources and comment on whether acquisition and summarization of this additional information would change the hazard characterization already provided in the existing summaries? Does the information presented clearly and accurately represent the hazards associated with each chemical? Are there any studies that should be removed from the summary and evidence tables due to study quality or data transparency concerns?
                                       #
  Summary of Public Comments for Specific Issues Related to Charge Question 3
                               EPA/OPPT Response
                                      19
Overall comments about representativeness of data:
 As was noted in the Environmental Hazard Summary for PIP (3:1), much of the environmental hazard data was collected using commercial mixtures with significant fractions (>5%) of triphenyl phosphate (TPP) which appears to drive toxicity. When the PIP (3:1) commercial mixtures have less than 5% TPP, aquatic toxicity appears to be much lower.
 ACC Recommendation: Greater clarity regarding the environmental toxicity of PIP (3:1) is needed.
 The purpose of the hazard document is not to evaluate the hazard of chemicals typically present in mixtures containing the PBT chemicals. The percentage of TPP was provided (as reported) so the reader is aware that a mixture may have been used in the study.
       


Appropriateness and Sufficiency of Information - Public Comments
Charge question 4: Has appropriate and sufficient information been obtained from the literature to support the identification of hazards?
                                       #
  Summary of Public Comments for Specific Issues Related to Charge Question 4
                               EPA/OPPT Response
                                      25
Overall comments about the sufficiency of the data:
 ACC Recommendation: EPA should consider the application of New Approach Methodologies (NAMs) particularly in those cases where the available data is limited. For example, at least two of the PBTs had limited hazard data. The use of NAMs may help establish a threshold for risk management that could be appropriate.
 PCTP is both a metabolite and biodegradation product of pentachloronitrobenzene (PCNB) (Khan et al., 2011) and a metabolite of hexachlorobenzene (WHO, 1997). EPA used data on these compounds because there was little publicly available data on PCTP.
 The Environmental and Human Health Hazard Summary is intended to be a survey of existing hazard data to provide a general characterization of the toxicity profile of the five PBT chemicals, e.g. species, types of effects and the range of toxicity. Hence, the hazard data are reported from the literature with no additional analysis or assessment. EPA did not conduct a systematic review or calculate margins of exposure to characterize hazard (toxicity) for the five PBT chemicals. The Hazard Summary is fit for purpose for the type of analysis performed in support of 6(h) rulemaking. Because TSCA does not require a risk evaluation be conducted to support a 6(h) action for PBTs, EPA does not deem it necessary to establish a threshold for the five PBT chemicals.


Reliability and Relevance of Information - Public Comments
Charge question 5: Please comment on the reliability and relevance of the identified data sets for each chemical and on the strengths and weaknesses of individual data sets that will help to better inform future regulatory actions.
                                       #
  Summary of Public Comments for Specific Issues Related to Charge Question 5
                               EPA/OPPT Response
                                      26
Overall comments about the reliability of the data:
 ACC Recommendation: EPA is not required to conduct a risk evaluation for Section 6(h) PBT chemicals, nor is it required to apply systematic review to the data associated with the Section 6(h) PBT chemicals. EPA should apply, however, the data quality criteria from its Application of Systematic Review in TSCA Risk Evaluations to the environmental and human health hazard data to enable the public to understand which data may be of sufficient reliability and relevance for regulatory decision-making.
 A number of the acute and chronic environmental hazard endpoint values for the PBT chemicals exceed the water solubility limit which raises questions about the reliability of those data. For example, it is reported for DecaBDE that "[a]cute toxicity to fish varies among species, with acute effects reported in the range of 0.01 to >500 mg/L" (p.8) when water solubility is reported in the Use and Exposure Assessment as 0.02 mg/L (p. 21).
 ACC Recommendation: EPA should re-evaluate the reliability of the acute and chronic environmental hazard data for the PBT chemicals. As part of this, EPA should include the details of the test article, materials, and methods of each acute fish study and water solubility study cited by EPA since these parameters may explain the variation of toxicity in different fish species. ACC also urges EPA to apply its Systematic Review principles to the data that will be used as part of the risk evaluation and risk management decisions on the five PBTs.
  Study details are listed in the hazard document as reported to provide further insight on potential study limitations.
 Publicly available information on the PBT chemicals were provided in the hazard document, and the study details were provided as reported.  

