Scientific Report on consumer health‐based guidance values for trifluoroacetic acid Authors: EFSA (European Food Safety Authority), Adeline Cavelier, Tamara Coja, Andrea Gall, Antonio Hernández‐Jerez, Christiane Vleminckx, Elodie Bergsma, Marco Binaglia, Federica Crivellente, Mathilde Colas, Anna Lanzoni, Jochem Louisse, Rafaela Rodrigues De Jesus, Annetta Grillo, Iris Mangas, Juan Manuel Parra Morte Published: 2026 ABSTRACT AbstractThe European Commission asked EFSA to provide a scientific report according to Article 31 of Regulation (EC) No 178/2002, in conjunction with Regulation (EC) No 1107/2009, to revise consumer health‐based guidance values, i.e. acceptable daily intake (ADI) and acute reference dose (ARfD), for trifluoroacetic acid (TFA). EFSA launched a targeted call for data from 6 August to 7 October 2024 and established an EFSA Working Group. Based on the weight of evidence, the WG concluded that TFA and its salt, TFA Na, are unlikely to be genotoxic. The WG established an ADI of 0.014 mg/kg body weight (bw) per day (expressed as trifluoroacetic acid) derived from the parental benchmark dose lower confidence limit (BMDL20) of 8.6 mg/kg bw per day. This value is based on changes in thyroid hormones, specifically decreased thyroxine (T4) levels, observed in the extended one‐generation reproductive toxicity study in rats. A standard uncertainty factor of 100 was applied to account for inter‐ and intra‐species differences. An additional factor of 5 was applied to address remaining uncertainties, namely the absence of a long‐term toxicity/carcinogenicity study and the potential for developmental immunotoxicity, given the lack of functional immunotoxicity testing for TFA during the developmental phase. In addition, an ARfD of 0.07 mg/kg bw (expressed as trifluoroacetic acid) was derived based on the same BMDL20 of 8.6 mg/kg bw per day, by applying a standard uncertainty factor of 100.Keywords: ADI, ARfD, common metabolite, health‐based guidance values, mammalian toxicology, PFAS, trifluoroacetic acid INTRODUCTION Trifluoroacetic acid (TFA) is formed as a degradation product of various chemicals such as hydrochlorofluorocarbons (HCFCs) and hydrofluorocarbons (HFCs), as well as pharmaceuticals. Additionally, TFA can result from the metabolism and breakdown of certain per‐ and polyfluoroalkyl substances (PFAS) 1 and some active substances used in plant protection products (Solomon et al., 2016). The existing consumer health‐based guidance values (HBGVs) for TFA were set during the peer review of the pesticide risk assessment of the active substance flurtamone (EFSA, 2017): an acceptable daily intake (ADI) of 0.05 mg/kg body weight (bw) per day (expressed as sodium trifluoroacetate (TFA Na)) was established based on a 90‐day rat study. This value incorporates an uncertainty factor (UF) of 200, which includes a twofold factor to account for the extrapolation from sub‐chronic to chronic exposure. No acute reference dose (ARfD) was considered necessary based on the available toxicological studies. On 7 January 2021, a notification according to Article 56 of Regulation (EC) No 1107/2009 2 was submitted to EFSA, the European Commission and Member States Competent Authorities by the TFA task force, which consists of companies which market chemicals forming TFA in the European Union. The notification raised concerns regarding potentially harmful or unacceptable effects of TFA, as adverse developmental effects were observed in rabbits after exposure to TFA in a newly conducted prenatal developmental toxicity study (OECD Test Guideline (TG) 414). In 2023, this study was submitted in the context of the REACH registration procedure for TFA. 3 Furthermore, in the recent assessment of TFA as part of the evaluation of tritosulfuron (EFSA, 2023a) and flufenacet (EFSA, 2024), a data gap was identified by EFSA for aneugenicity potential. The same HBGVs for consumer risk assessment as previously peer reviewed in the context of the EFSA conclusion of flurtamone (EFSA, 2017) were used. According to the latest scientific state of the art (EFSA Scientific Committee, 2021), the derivation of HBGVs could be affected by the conclusion on aneugenicity. However, this was not discussed during the peer review since the EFSA Scientific Opinion on the guidance on aneugenicity assessment was not available at the time of the dossier submission. On 11 June 2024, in accordance with Regulation (EC) No 1272/2008, 4 Germany submitted a harmonised classification and labelling (CLH) dossier on TFA and its salt, TFA Na, 5 to the European Chemicals Agency (ECHA), proposing new classifications as a reproductive toxicant (Category 1B, H360fD 6 ); persistent, mobile and toxic (PMT); and very persistent and very mobile (vPvM). The CLH proposal for reproductive toxicity was based on the prenatal developmental toxicity study in rabbits3 and fertility effects in an extended one‐generation reproductive toxicity study (EOGRTS) in rats, having an impact on the relevance assessment as a groundwater metabolite. In addition to the ADI, the effects observed in the developmental toxicity may trigger the need to set an ARfD for TFA. From 26 May 2025 to 25 July 2025, ECHA launched a public consultation on the CLH dossier 7 submitted by Germany, inviting stakeholders to provide comments on the new CLH proposals. The CLH proposals were discussed at the ECHA Risk Assessment Committee (RAC) 77 WG meeting on 27 April 2026 and at the RAC‐77 Plenary meeting on 2 June 2026. The ECHA RAC recommended a classification for TFA as Acute Tox. 4 (oral, H302 8 ) with an acute toxicity estimate (ATE) of 500 mg/kg bw; for TFA and TFA Na as Acute Tox. 3 (inhalation, H331 9 ) with an ATE of 3 mg/L (vapour), as Reproductive toxicity category 1B for effects on development (H360Df 10 ) and as PMT and vPvM. 11 Background and Terms of Reference as provided by the requestor In accordance with Article 31 of Regulation (EC) No 178/2002, 12 in conjunction with Regulation (EC) No 1107/2009, the European Commission requested EFSA to issue an EFSA output and to review the recommended consumer HBGVs, i.e. ADI and ARfD, for TFA. 13 The review should take into account the available studies: those previously submitted and evaluated as part of active substance assessments, new studies submitted by the TFA task force related to notification under Article 56 of Regulation (EC) No 1107/2009 14 received on 7 January 2021, data used by Germany for the CLH dossier on TFA and TFA Na, and, where available, studies submitted to Member States for the assessment of plant protection products. More specifically, EFSA was requested to conduct the following tasks: Term of reference (ToR) 1: to collect evidence in addition to data on TFA already available from EFSA peer review processes, including: (i) additional data on TFA and TFA Na generated by the TFA task force, (ii) data used by Germany for the CLH dossier on TFA and TFA Na, and (iii) data submitted to Member States as part of the process for authorisation of plant protection products. Term of reference (ToR) 1: to collect evidence in addition to data on TFA already available from EFSA peer review processes, including: (i) additional data on TFA and TFA Na generated by the TFA task force, (ii) data used by Germany for the CLH dossier on TFA and TFA Na, and (iii) data submitted to Member States as part of the process for authorisation of plant protection products. Term of Reference (ToR) 2: to assess the available evidence in relation to the toxicological properties of the metabolite TFA and TFA Na and, if possible, to derive HBGVs (ADI and ARfD) for TFA to be used in risk assessments. Term of Reference (ToR) 2: to assess the available evidence in relation to the toxicological properties of the metabolite TFA and TFA Na and, if possible, to derive HBGVs (ADI and ARfD) for TFA to be used in risk assessments. As part of this work, EFSA was invited to consult Member States and, where relevant, experts from EFSA's Panels and their working groups, as well as ECHA's Committee on Risk Assessment (RAC). Interpretation of the Terms of Reference Term of Reference (ToR) 1 To address ToR 1, EFSA launched a targeted call for data, inviting Member States, the TFA task force and the Competent Authority Germany, 15 which prepared the CLH dossier on TFA and TFA Na, to submit toxicological and metabolism studies on TFA and TFA Na. Specifically, the following data was requested: For Member States, data from the PPP authorisation dossiers for pesticide active substances identified as PFAS 16 that either form or potentially form TFA, according to the outcome of EFSA peer review procedures. In this context, Member States were invited to submit any available data related to the toxicological hazard properties of TFA or TFA Na covering the endpoints of Section 5 – Subsections 5.1. to 5.9.7 in Commission Regulation (EU) No 283/2013. 17 For Member States, data from the PPP authorisation dossiers for pesticide active substances identified as PFAS 16 that either form or potentially form TFA, according to the outcome of EFSA peer review procedures. In this context, Member States were invited to submit any available data related to the toxicological hazard properties of TFA or TFA Na covering the endpoints of Section 5 – Subsections 5.1. to 5.9.7 in Commission Regulation (EU) No 283/2013. 17 For the Competent Authority Germany, studies used in the context of the CLH report and not available at EFSA level and the confidential version of the CLH report, for the sake of ECHA/EFSA alignment on data package and assessment, and in line with the one substance one assessment concept (BAuA, 2025). As part of this procedure, the list of references used in the context of the CLH dossier was beforehand compared with the data available at EFSA level. For the Competent Authority Germany, studies used in the context of the CLH report and not available at EFSA level and the confidential version of the CLH report, for the sake of ECHA/EFSA alignment on data package and assessment, and in line with the one substance one assessment concept (BAuA, 2025). As part of this procedure, the list of references used in the context of the CLH dossier was beforehand compared with the data available at EFSA level. For the TFA task force, the list of studies generated by the TFA task force and related original study reports or position papers. For the TFA task force, the list of studies generated by the TFA task force and related original study reports or position papers. In parallel to the targeted call for data, EFSA gathered all studies relevant for this Mandate from the dossiers submitted within the EFSA peer review processes. The targeted call was launched from 6 August to 7 October 2024. A total of 170 studies and position papers were collected. The list of studies can be found in Annex 2 and at: https://open.efsa.europa.eu/questions/EFSA‐Q‐2024‐00502. Term of Reference (ToR) 2 An EFSA Working Group (WG) was established in order to address ToR 2. The WG interpreted the ToR 2 by considering the following tasks in the first tier: To conduct the assessment, including relevance and reliability, of studies available for genotoxicity hazard identification; To conduct the assessment, including relevance and reliability, of studies available for genotoxicity hazard identification; To conduct the assessment, including relevance and reliability of studies available for assessment of toxicokinetic and toxicodynamic properties of TFA; To conduct the assessment, including relevance and reliability of studies available for assessment of toxicokinetic and toxicodynamic properties of TFA; To identify the most suitable basis for setting the ADI and ARfD, if applicable; To identify the most suitable basis for setting the ADI and ARfD, if applicable; To derive HBGVs (ADI and ARfD, if applicable) for TFA to be used in consumer risk assessments. To derive HBGVs (ADI and ARfD, if applicable) for TFA to be used in consumer risk assessments. The WG defined the assessment questions as follows: Assessment question 01: Based on the results of the targeted call for data, do the available data raise a concern for genotoxicity? Assessment question 01: Based on the results of the targeted call for data, do the available data raise a concern for genotoxicity? Assessment question 02: Based on the results of the targeted call for data, what is the most suitable basis for setting the ADI? Assessment question 02: Based on the results of the targeted call for data, what is the most suitable basis for setting the ADI? Assessment question 03: Based on the results of the targeted call for data, is there a need for setting an ARfD? If so, what is the most suitable basis? Assessment question 03: Based on the results of the targeted call for data, is there a need for setting an ARfD? If so, what is the most suitable basis? Evaluations carried out under other regulatory contexts TFA and TFA Na have been evaluated by several organisations including UBA (2020), DTU (2021), RIVM (2023), Department Zorg (2024) and JMPR (2024). TFA is registered under REACH Regulation and is manufactured in and/or imported to the European Economic Area. At the time EFSA was drafting this report, ECHA CLH procedure for TFA and TFA Na was underway for proposed classification as a reproductive toxicant (Category 1B, H360fD), PMT and vPvM. There are ongoing activities to assess several PFAS including TFA under different regulatory frameworks. On 13 January 2023, five national authorities (Denmark, Germany, the Netherlands, Norway and Sweden) submitted a joint proposal to ECHA to introduce a restriction under the REACH Regulation covering approximately 10,000 PFAS (ECHA, 2023). The proposal sought to establish an EU‐wide framework to address risks associated with the manufacture, placing on the market and use of this broad group of substances. From 22 March 2023 to 22 September 2023, ECHA conducted a 6‐month public consultation on the proposed restriction, enabling stakeholders to provide scientific, technical and socio‐economic information. ECHA aims to complete its scientific evaluation of the proposed restriction by the end of 2026. In the area of drinking water regulation, TFA has been proposed for inclusion in the most recent revision of the Drinking Water Directive, 18 with the objective of ensuring that TFA is considered among the PFAS substances subject to regular monitoring in drinking water. As of 12 January 2026, 19 PFAS analysis in drinking water (including TFA) has become mandatory across Member States. Member States are required to monitor compliance using one of the following parametric values: –Total PFAS (including TFA), with a limit value of 0.5 μg/L; or Total PFAS (including TFA), with a limit value of 0.5 μg/L; or –The sum of 20 specified PFAS, with a limit value of 0.1 μg/L. The sum of 20 specified PFAS, with a limit value of 0.1 μg/L. Prior to this date, PFAS monitoring in drinking water was conducted on a voluntary basis, except where more stringent national or regional requirements applied. The transitional period has now ended, and mandatory monitoring requirements are in force. As regards the assessment of TFA within the PFAS group, the WG did not assess substances other than TFA and TFA Na since this is considered out of the scope of the current mandate. In 2020, the EFSA Panel on Contaminants in the Food Chain (CONTAM) assessed the risks related to the presence of PFAS in food, focusing mainly on four PFAS: perfluorooctane sulfonic acid (PFOS), perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA) and perfluorohexane sulfonic acid (PFHxS), which were the most frequent PFAS detected in human blood and for which sufficient occurrence and toxicological data were available. Based on toxicological evidence, particularly effects on the immune system and measured human exposure, a tolerable weekly intake of 4.4 ng/kg bw per week was established for the sum of these four PFAS (EFSA, 2020). In that EFSA opinion, no data on TFA or its sodium salt were included, and no explicit reference was made to TFA, TFA Na or related synonyms (e.g. trifluoroacetate) in the main text, tables or annexes. This absence reflects that TFA was outside the defined scope of the EFSA's assessment from 2020, which focused on selected perfluoroalkyl carboxylic acids (PFCAs), perfluoroalkyl sulfonic acids (PFSAs) and related substances, mainly those with carbon chain ranging from C4 to C18, and for which adequate occurrence and toxicological data were available at the time. DATA AND METHODOLOGIES Data collection Collection of evidence from targeted call for data To address ToR 1, EFSA collected evidence through a targeted call for data launched from 6 August 2024 to 7 October 2024. This includes data on TFA and TFA Na already available from EFSA peer review processes, additional data on TFA and TFA Na generated by the TFA task force, data used by the dossier submitter (Germany) for the drafting of the CLH dossier on TFA and data submitted to Member States as part of the process for authorisation of plant protection products. EFSA noted that a literature search 20 had been conducted by the dossier submitter for the drafting of the CLH dossier and additional literature studies were part of the data package submitted to EFSA by different stakeholders. Data generated by the TFA task force notably included REACH registration data. These data have also been considered by the WG. The complete list of 170 studies and position papers can be found in Annex 2. In the interests of transparency, data owners were invited to submit redacted versions of original study reports or position papers that were not previously made publicly available. The studies are available on Open EFSA. 21 The studies are identified and referenced throughout the present scientific report using the numbering #1 to #170, as provided in Column A of Annex 2. Consolidated list of studies outside the window of call for data During the development of this report, the WG was also made aware of new studies out of time frame of targeted call of data (6 August 2024 to 7 October 2024). Formally, there is no legal obligation to consider newly available data submitted outside of the dedicated targeted call for data, unless such information qualifies as adverse data. A list of the collected newly available public literature or studies is reported in the format of an Excel spreadsheet and made publicly available as part of the background documentation to the EFSA Scientific report (Annex 9). EFSA screened the newly submitted public literature or studies to determine their potential impact on the risk assessment and whether they would have altered the weight of evidence (WoE) in the current risk assessment. Overall, this exercise aimed to ensure that information from published literature or new studies, even if they became available at a later stage and outside the legislative frame, was screened to ensure that available evidence that could have led to a different outcome was considered. Methodologies Based on the collected data (ToR 1), a structured, systematic approach was used to assess all submitted data and to set consumer HGBVs, if appropriate (ToR 2). To address ToR 2, a weight of evidence (WoE) was applied, following EFSA's established principles for data handling and uncertainty analysis, with methods planned through a detailed protocol (see Annex 1). All the evidence retrieved followed the same process, i.e. extraction of the data, appraisal of the evidence and uncertainty analysis. Furthermore, specific eligibility criteria for relevance of studies were defined by the WG and applied to all the submitted studies. For instance, only studies on TFA or TFA Na were included, whereas studies on mixtures containing TFA or TFA salts (e.g. TFSK: KTMS (potassium trifluoromethanesulfinate) TFAK (potassium trifluoroacetate) reaction mass), were excluded as it was not possible to distinguish toxic effects of TFAK from those of KTMS. In addition, for the in vivo studies, only the oral route was considered, since it is the relevant route of exposure for setting the consumer reference values. The detailed protocol is available in Annex 1. The reasons for excluding 84 studies and position papers from the body of evidence according to the eligibility criteria can be found in Appendix H. The WG divided studies providing relevant information for genotoxicity hazard identification, for assessment of toxicokinetic and toxicodynamic properties of TFA, and studies providing mechanistic information on mode of action (MoA). All the studies were classified as either OECD TG (regulatory) or non‐OECD TG studies and by streams of evidence (i.e. in vitro and in vivo animal, human observational studies). The WG extracted the data from selected studies and appraised the evidence. The methodology for appraising evidence on the relationship between TFA exposure and: Genotoxicity endpoints were based on the approach described by the EFSA cross‐cutting Working Group on Genotoxicity (EFSA, 2023b). Genotoxicity endpoints were based on the approach described by the EFSA cross‐cutting Working Group on Genotoxicity (EFSA, 2023b). Non‐genotoxicity endpoints, coming from OECD TG studies, were based on the Klimisch et al. (1997) scoring system. Non‐genotoxicity endpoints, coming from OECD TG studies, were based on the Klimisch et al. (1997) scoring system. Non‐genotoxicity endpoints, coming from non‐OECD TG studies, were based on the risk of bias (RoB). The RoB methodology was tailored to three streams of evidence: in vitro, in vivo and human studies. The internal validity, or RoB, of each study was appraised using a customised version of the OHAT/NTP RoB assessment tool (NTP, 2015, 2016). Each study was appraised by two independent reviewers from the WG and EFSA. Possible discrepancies not resolved by discussion were taken to the attention of the whole WG. Non‐genotoxicity endpoints, coming from non‐OECD TG studies, were based on the risk of bias (RoB). The RoB methodology was tailored to three streams of evidence: in vitro, in vivo and human studies. The internal validity, or RoB, of each study was appraised using a customised version of the OHAT/NTP RoB assessment tool (NTP, 2015, 2016). Each study was appraised by two independent reviewers from the WG and EFSA. Possible discrepancies not resolved by discussion were taken to the attention of the whole WG. All studies, including already peer‐reviewed studies, were subject to evaluation by the WG. For further details on the methodology, please refer to the protocol in Annex 1. The WG also considered the evaluations carried out under other regulatory contexts described under Section 1.3. The draft output and annexes of the current Scientific Report on TFA were open for public consultation before its finalisation from 22 July 2025 to 22 September 2025. The outcome of the public consultation is available in Annex 8. ASSESSMENT Identity and method of chemical analysis Most toxicological studies have been conducted using the salt of TFA, e.g. TFA Na. The purity of the test material in the toxicological studies ranged from 98% to 99.1%. Trifluoroacetic acid and sodium trifluoroacetate are, respectively, the IUPAC chemical names for compound codified or named in this Scientific Report as TFA and TFA Na (see Table 1). TABLE 1Used compound codes.Code/trivial name a IUPAC name/SMILES notation/InChiKey b Structural formula c Trifluoroacetic acid, TFA Trifluoroacetic acid FC(F)(F)C(=O)O DTQVDTLACAAQTR‐UHFFFAOYSA‐N Sodium trifluoroacetate, TFA Na, perfluoroacetate Sodium trifluoroacetate [Na+].FC(F)(F)C([O‐])=O UYCAUPASBSROMS‐UHFFFAOYSA‐M Abbreviations: InChiKey, International Chemical Identifier; IUPAC, International Union of Pure and Applied Chemistry; SMILES, Simplified Molecular Input Line Entry System. a The name in bold is the name used in this scientific report. b ACD/Name 2021.1.3 ACD/Labs 2021.1.3 (File Version N15E41, Build 123232, 7 July 2021). c ACD/ChemSketch 2021.1.3 ACD/Labs 2021.1.3 (File Version C25H41, Build 123835, 28 August 2021). Used compound codes. Trifluoroacetic acid FC(F)(F)C(=O)O DTQVDTLACAAQTR‐UHFFFAOYSA‐N Sodium trifluoroacetate [Na+].FC(F)(F)C([O‐])=O UYCAUPASBSROMS‐UHFFFAOYSA‐M Abbreviations: InChiKey, International Chemical Identifier; IUPAC, International Union of Pure and Applied Chemistry; SMILES, Simplified Molecular Input Line Entry System. The name in bold is the name used in this scientific report. ACD/Name 2021.1.3 ACD/Labs 2021.1.3 (File Version N15E41, Build 123232, 7 July 2021). ACD/ChemSketch 2021.1.3 ACD/Labs 2021.1.3 (File Version C25H41, Build 123835, 28 August 2021). The method of analysis of TFA in the diet (#150) used in the key study to set HBGVs was considered reliable (i.e. fit for purpose). 22 Hazard identification and characterisation Toxicokinetics No test guideline‐compliant absorption, distribution, metabolism and excretion (ADME) studies were submitted to EFSA. It should be noted that under physiological pH, TFA and TFA salts dissociate, resulting in systemic exposure to the TFA anion. Toxicokinetic data are available from plasma measurements in the newly submitted rat studies (#99, #109) and rabbit study (#102). The rabbit developmental toxicity study #103 lacked the kinetic analyses needed to assess linearity but provided information on tissue distribution. In addition, a publicly available in vivo study (#97) was reviewed but excluded due to a non‐oral route of administration, although it provided contextual information regarding TFA kinetics. Collectively, the toxicokinetic data derived from these oral toxicity and developmental studies (#99, #102, #103 and #109) were considered sufficient to characterise systemic exposure, distribution and general kinetic behaviour under the tested conditions. In a 52‐week rat drinking water study (#109), apparent steady‐state plasma concentrations were achieved by week 52. Systemic exposure (Concentration maximum (Cmax), Area under the Curve (AUC)) increased approximately proportionally over the concentration range of 30–600 ppm, with no relevant sex‐related differences. In contrast, a preliminary dietary reproductive study in rats (#99) showed non‐linear, sub‐proportional kinetics in dams, suggesting saturation of absorption and/or elimination processes at higher doses. F1 offspring exhibited age‐dependent toxicokinetic differences, with low and dose‐independent plasma concentrations shortly after birth (postnatal day (PND) 4) and markedly higher systemic exposure by weaning (PND 21), consistent with increasing internal exposure as pups matured and because F1 offspring generally consumed diet from mid‐lactation. In rabbits administered TFA by oral gavage (#102), absorption was evident, with median time to maximum concentration (Tmax) values of approximately 3–8 h on gestation day (GD) 6, decreasing to about 2–3 h by GD 28, indicating faster absorption later in gestation. Systemic exposure increased in a sub‐proportional manner with dose, consistent with non‐linear kinetics, likely reflecting saturation of absorption processes and/or a relative increase in elimination efficiency at higher doses. The rabbit developmental toxicity study (#103) confirmed systemic exposure and demonstrated efficient placental transfer, with fetal plasma levels comparable to maternal levels, as well as distribution to the ocular aqueous and vitreous humour, but did not allow assessment of kinetic linearity. Overall, TFA is absorbed after oral exposure and distributes systemically, including to the fetus and the eye. Its toxicokinetic behaviour is context‐dependent, with approximately linear kinetics under long‐term low‐dose exposure in rats and non‐linear kinetics observed at higher doses in developmental studies in both species. A more detailed assessment is provided in Appendix B. Integrating these toxicokinetic data is critical for interpreting both hazard and risk, particularly in contextualising systemic exposures in animal studies relative to human biomonitoring data. Serum concentrations at the No Observed Adverse Effect Level (NOAEL) and Lowest Observed Adverse Effect Level (LOAEL) in the animal studies are several orders of magnitude higher than TFA levels reported in human populations (typically in the μM range, see Table 17). This indicates a substantial margin of exposure (MOE) between experimental exposures and environmental or background human exposure levels. Based on its physicochemical properties (low log Kow, low bioconcentration factor and high polarity) and the absence of progressive systemic accumulation upon repeated dosing, TFA is expected to have a low potential for bioaccumulation or undergo significant metabolism. Although specific data on absorption, mass balance and elimination are limited, qualitative evidence supports high oral absorption, systemic distribution and rapid excretion, aligning with expectations for small, hydrophilic acids. TABLE 17Summary of the biomonitoring studies measuring TFA levels.Study referencePopulation groupBiological matrixMethodMedian levels and detection rateStudy design#91Adults (n = 252)SerumHPLC‐MS/MS Median: 8.46 ng/mL Detection rate: 97.0% Cross‐sectional#94Neonates (n = 66)Umbilical cord serumUPLC‐MS/MS Median: 0.229 ng/mL Detection rate: 55% Cross‐sectionalAbbreviations: HPLC‐MS/MS, High Performance Liquid Chromatography – Mass Spectrometry; UPLC‐MS/MS, Ultra‐high Performance Liquid Chromatography – Mass Spectrometry. Summary of the biomonitoring studies measuring TFA levels. Median: 8.46 ng/mL Detection rate: 97.0% Median: 0.229 ng/mL Detection rate: 55% Abbreviations: HPLC‐MS/MS, High Performance Liquid Chromatography – Mass Spectrometry; UPLC‐MS/MS, Ultra‐high Performance Liquid Chromatography – Mass Spectrometry. Genotoxicity The genotoxic potential of TFA and TFA Na was investigated in vitro in several tests covering different endpoints (see Appendix A). Induction of reverse mutations in bacteria was investigated in a total of five studies. Several strains of Salmonella typhimurium were tested in the presence or absence of metabolic activation. Three tests were performed in accordance with OECD TG 471, and the study results were considered highly relevant. One additional test was not fully in accordance with OECD TG 471, and the study results were considered of low relevance. A spot test was also reported in the literature, with a reliability score of 4 (‘not assignable’) and the results were considered of low relevance. All tests gave negative results with or without S9 (metabolic activation). Induction of gene mutations in mammalian cells was investigated in five studies. All tests were performed in mouse lymphoma cells (L5178Y) in the presence or absence of metabolic activation. Four tests investigating mutations at the thymidine kinase locus and one at the hypoxanthine‐guanine phosphoribosyltransferase (HPRT) locus were conducted in accordance with OECD TG 490 and OECD TG 476, respectively. All the study results with or without S9 were negative and considered of high relevance. During the public consultation on this Scientific Report, comments were provided on the interpretation of the results from an in vitro clastogenicity assay (#141). The EFSA cross‐cutting WG on genotoxicity was consulted on the clastogenic and aneugenic potential of TFA and TFA Na (see Appendix J). The TFA WG concurred with the advice of the EFSA Genotoxicity WG on chromosomal damage. The potential of TFA and TFA Na to induce chromosomal damage was evaluated with the chromosome aberration test performed in human peripheral blood lymphocytes and CHL cell line in the presence and absence of metabolic activation. Negative results were obtained in three studies of limited relevance carried out using TFA Na, while a highly relevant study conducted with TFA was positive. The potential of TFA to induce chromosomal damage was also investigated in two in vitro micronucleus tests, conducted in the presence and absence of metabolic activation using the human lymphoblastoid TK6 and CHO‐K1 cell lines. Negative results were obtained in both studies, considered of high relevance. The WG noted that the former data gap on aneugenicity (EFSA, 2023a) is now fulfilled by applying the in vitro micronucleus test. Overall, the studies carried out with TFA Na report no induction of chromosomal damage, while inconsistencies were observed among the studies performed with TFA showing a positive response in one chromosome aberration test and negative results in two in vitro micronucleus tests. In addition, the WG noted that studies with TFA showed considerable variability in cytotoxicity levels, both within individual studies (e.g. between range‐finding and main experiments) and across different studies. The reasons for these inconsistencies remain unclear but could be related to the physico‐chemical characteristics of TFA (a strong acid) (Morita et al., 1989) and technical factors (e.g. pH changes) modulating the effects of TFA. In contrast, no such variability was observed with TFA Na. Considering: (i) the substantial uncertainty surrounding the mechanisms underlying these findings with TFA compared to TFA Na and (ii) that exposure to TFA in its acid form is unlikely under physiological conditions, the WG assigned greater weight to the consistently negative results of chromosomal aberration studies conducted with the TFA salt than to those observed with the acid. Overall, the weight of evidence does not raise a concern for chromosomal damage of TFA and its Na salt. Based on the in vitro studies, the weight of evidence suggests that genotoxicity is unlikely. There is no need to perform in vivo tests, in line with EFSA Scientific Opinion on genotoxicity testing strategies applicable to food and feed safety assessment (EFSA Scientific Committee, 2011). In conclusion, based on available data, TFA or TFA Na are unlikely to be genotoxic. Toxicity in experimental animals Single oral toxicity Based on the available information, TFA Na is of low acute oral toxicity to rats (#1, #122) with the lethal dose (LD) 50 value of TFA Na above 2000 mg/kg bw in female rats (#1). The WG noted that, at the ECHA RAC 77 Plenary meeting (2 June 2026), the ECHA RAC recommended a classification for TFA as Acute Tox. 4 (oral, H302) with an acute toxicity estimate (ATE) of 500 mg/kg bw and for TFA and TFA Na as Acute Tox. 3 (inhalation, H331 with an ATE of 3 mg/L (vapour)). 23 Repeated oral toxicity Repeated oral toxicity after sub‐chronic (# 8, #15, #16, #124) and chronic (52‐week, #109) exposure to TFA Na indicated the liver as the main target of toxicity in rats, as evidenced by increased organ weight and hepatocellular hypertrophy. Rats treated with TFA Na showed changes in some clinical chemistry parameters across all repeated dose studies (e.g. decrease in cholesterol, decrease in glucose and bilirubin levels in both sexes) (see Appendix C and Section 3.2.7 ). Nevertheless, in the absence of associated histopathological findings or significant increase in liver weight (higher than 15%), the WG did not consider the effects on clinical chemistry parameters relevant for setting the NOAEL. Changes in haematological parameters observed across studies were mild and not considered to reach the relevant level of adversity. There were no TFA Na treatment‐related effects noted by the end of the 52‐week rat study (#109) or following the 6‐week recovery phase (i.e. the maximum tolerable dose (MTD) was not reached, as the highest dose tested (37.8 mg/kg bw per day) was not properly selected). Ophthalmological investigation in rats in 52‐week study (#109) revealed no treatment‐related effects, as most of the effects were observed in animals also during pretreatment. The ophthalmological observations were macroscopic and not related to folded retina observed in rabbit developmental studies. Across all repeated‐dose oral toxicity studies, the lowest NOAEL of 9.9 mg/kg bw per day was identified in the 90‐day study in Wistar rats (#8) with the corresponding LOAEL of 98 mg/kg bw per day. At the LOAEL, effects included increased relative liver weight in both sexes and hepatocellular hypertrophy in males and a decrease in total bilirubin and glucose in both sexes (see Table 2). In Sprague Dawley rats, the NOAEL in the 90‐day study (#124) was 89.2 mg/kg bw per day, whereas in the 52‐week study (#109), the NOAEL was 37.8 mg/kg bw per day (highest dose level tested). TABLE 2A 90‐day rat study (#8).ParameterSex N Vehicle 0 ppmLow dose – 160 ppm (9.9 mg/kg bw per day in males/12.2 mg/kg bw per day in females)Mid dose – 1600 ppm (98 mg/kg bw per day in males/123 mg/kg bw per day in females)High dose – 16,000 ppm (1043 mg/kg bw per day in males/1216 mg/kg bw per day in females)Mean/IncidenceSD ±Mean/IncidenceSD ±%Mean/IncidenceSD ±%Mean/IncidenceSD ±% Absolute liver weight (g) M1012.150.8311.610.92−413.25* 1.01914.48** 0.9519F105.960.486.250.6756.71* 0.46137.36** 0.6823 Relative liver weight % M102.3270.1192.2580.083−32.657** 0.148143.102** 0.15733F102.2430.0892.3430.16542.520** 0.221122.880** 0.14928 Centrilobular to panlobular Hepatocellular hypertrophy, diffuse –minimal M1/100/103/91/10F0/100/100/105/10 Centrilobular to panlobular Hepatocellular hypertrophy, diffuse – slight M0/100/102/96/10F0/100/100/104/10 Centrilobular to panlobular Hepatocellular hypertrophy, diffuse – moderate M0/100/100/93/10F0/100/100/100/10 Total bilirubin μmol/L M10 a 1.60.41.10.2−310.5** 0.1−690.3** 0.2−81F102.10.51.80.4−141** 0.6−520.5** 0.3−76 Glucose, mmol/L M10 a 5.870.535.40.64−84.21** 0.44−284.14** 0.84−29F105.570.865.130.56−84.19** 0.45−254.62* 1.11−17 Alanine AminoTransferase (ALT), IU/L M10 a 47254720087848565** 1938F103894010548172645518Abbreviations: % Effect size/N, Number of animals examined; SD, standard deviation. a Nine instead of 10 males was included in group 3 (1600 ppm).**The group mean is significantly different from the control at p < 0.01.*The group mean is significantly different from the control at p < 0.05.– If the Bartlett test was not significant (p > 0.05), means were compared using the analysis of variance (ANOVA). If the ANOVA was not significant (p > 0.05), the group means are considered to be homogeneous, and no further analysis were performed. If the ANOVA was significant (p ≤ 0.05), the group means are considered to be homogeneous, and no further analysis were performed. If the ANOVA was significant (p < 0.05), means of the exposed groups were compared to the mean of the control group using the Dunnett test (2‐sided).– If the Bartlett test was significant (p < 0.05), group means were compared using the non‐parametric Kruskal–Wallis test. If the Kruskal–Wallis test was not significant (p > 0.05), the group means are considered to be homogeneous, and no further analysis were performed. If the Kruskal–Wallis test was significant (p < 0.05), means of the exposed groups were compared to the mean of the control group using the Dunn test (2‐sided). A 90‐day rat study (#8). Abbreviations: % Effect size/N, Number of animals examined; SD, standard deviation. Nine instead of 10 males was included in group 3 (1600 ppm). The group mean is significantly different from the control at p < 0.01. The group mean is significantly different from the control at p < 0.05. – If the Bartlett test was not significant (p > 0.05), means were compared using the analysis of variance (ANOVA). If the ANOVA was not significant (p > 0.05), the group means are considered to be homogeneous, and no further analysis were performed. If the ANOVA was significant (p ≤ 0.05), the group means are considered to be homogeneous, and no further analysis were performed. If the ANOVA was significant (p < 0.05), means of the exposed groups were compared to the mean of the control group using the Dunnett test (2‐sided). – If the Bartlett test was significant (p < 0.05), group means were compared using the non‐parametric Kruskal–Wallis test. If the Kruskal–Wallis test was not significant (p > 0.05), the group means are considered to be homogeneous, and no further analysis were performed. If the Kruskal–Wallis test was significant (p < 0.05), means of the exposed groups were compared to the mean of the control group using the Dunn test (2‐sided). Reproductive toxicity after oral exposure Multigeneration reproductive toxicity A preliminary reproductive toxicity study in rats (#99) and an extended one‐generation reproductive toxicity study (EOGRTS) in rats (#100) were conducted on TFA Na. In the EOGRTS in rats (#100) conducted on TFA Na (see Appendix D), thyroid hormone (thyroxine, T4) concentrations were determined in F0 and F1 adults at termination, as well as in F1 offspring on PND22 (see Table 3). The offspring NOAEL was set at 8.65 mg/kg bw per day based on decreased T4 levels at 44.3 mg/kg bw per day and higher. Dose–response relationship and biological significance (benchmark of 20%, see Appendix G) were considered the main criteria for determining the NOAEL for T4 decrease. It is noted that, although an increased relative thyroid weight was observed in some groups, no dose relationship was observed; the absolute thyroid weight was not changed, and no treatment‐related histopathological findings were reported in adults of both generations (see Table 4). TABLE 3Thyroid hormone (T4) levels, EOGRTS (#100).Thyroxine (T4) concentrations (pg/mL)ppm0120/60600/3003000/1500mg/kg bw per day08.6544.3223 N 10101010F0 male adult terminal (Week 18) Mean ± SD CV % % 41,700 ± 7690 18.4 43,400 ± 9630 22.2 +4% 37,100 ± 7710 20.8 −11% 24,700 ± 7610** 30.8 −41% F0 female adult terminal (Day 28 of lactation) Mean ± SD CV % % difference 32,200 ± 7390 23.0 31,400 ± 7970 25.4 −3% 31,800 ± 3630 11.4 −1% 22,000 ± 5090** 23.1 −32% F1 male offspring (Day 22 of age) Mean ± SD CV % % 56,900 ± 7370 13.0 60,200 ± 6510 10.8 +6% 47,300 ± 6590* 13.9 −17% 40,700 (36600) a ± 9680 (15800) a , ** 23.8 (43.2) a −36% F1 female offspring (Day 22 of age) Mean ± SD CV % % difference 53,800 ± 11,000 20.4 50,800 ± 8340 16.4 −6% 50,800 ± 8020 15.8 −6% 34,100 ± 8340** 24.5 −37% F1 male adult terminal (Week 13) Mean ± SD CV % % 50,400 ± 7130 14.1 42,700 ± 8980* 21.0 −15% 34,400 ± 10100** 29.4 −32% 24,400 ± 3690** 15.1 −52% F1 female adult terminal (Week 13) Mean ± SD CV% % 33,800 ± 13,700 40.5 34,000 ± 8480 24.9 +1% 35,900 ± 10,900 30.4 +6% 26,100 ± 5750 22.0 −23% Abbreviations: % Effect size/N, Number of animals examined; CV, coefficient of variation; SD, standard deviation.* p < 0.05 (Williams test).** p < 0.01 (Williams test). a Data for animal 289‐1 (below the limit of quantification) excluded from statistical calculations. Data in parenthesis include all data in statistical calculations. Thyroid hormone (T4) levels, EOGRTS (#100). Mean ± SD CV % % 41,700 ± 7690 18.4 43,400 ± 9630 22.2 +4% 37,100 ± 7710 20.8 −11% 24,700 ± 7610** 30.8 −41% Mean ± SD CV % % difference 32,200 ± 7390 23.0 31,400 ± 7970 25.4 −3% 31,800 ± 3630 11.4 −1% 22,000 ± 5090** 23.1 −32% Mean ± SD CV % % 56,900 ± 7370 13.0 60,200 ± 6510 10.8 +6% 47,300 ± 6590* 13.9 −17% 40,700 (36600) a ± 9680 (15800) a , ** 23.8 (43.2) a −36% Mean ± SD CV % % difference 53,800 ± 11,000 20.4 50,800 ± 8340 16.4 −6% 50,800 ± 8020 15.8 −6% 34,100 ± 8340** 24.5 −37% Mean ± SD CV % % 50,400 ± 7130 14.1 42,700 ± 8980* 21.0 −15% 34,400 ± 10100** 29.4 −32% 24,400 ± 3690** 15.1 −52% Mean ± SD CV% % 33,800 ± 13,700 40.5 34,000 ± 8480 24.9 +1% 35,900 ± 10,900 30.4 +6% 26,100 ± 5750 22.0 −23% Abbreviations: % Effect size/N, Number of animals examined; CV, coefficient of variation; SD, standard deviation. p < 0.05 (Williams test). p < 0.01 (Williams test). Data for animal 289‐1 (below the limit of quantification) excluded from statistical calculations. Data in parenthesis include all data in statistical calculations. TABLE 4Thyroid weights, EOGRTS (#100).Thyroid weightsppm0120/60600/3003000/1500mg/kg bw per day08.6544.3223 F0 generation males (week 17) N 25252525Thyroids and parathyroids weight a – absolute (g)Mean ± SD0.019 ± 0.0030.020 ± 0.0030.107 ± 0.437 a 0.019 ± 0.003Thyroids and parathyroids weight – relative to bw (%)Mean ± SD0.0042 ± 0.00080.0044 ± 0.00060.0215 ± 0.08590.0044 ± 0.0007 F0 generation females (Day 28 post‐partum) N 23232423Thyroids and parathyroids weight – absolute (g)Mean ± SD0.016 ± 0.0030.017 ± 0.0030.017 ± 0.0040.019 ± 0.005* Thyroids and parathyroids weight – relative to bw (%)Mean ± SD0.0068 ± 0.00120.0071 ± 0.00110.0070 ± 0.00130.0082 ± 0.0026** F1 generation males – cohort 1A (Week 13) N 20202020Thyroids and parathyroids weight – absolute (g)Mean ± SD0.017 ± 0.0020.019 ± 0.0030.020 ± 0.0040.018 ± 0.003Thyroids and parathyroids weight – relative to bw (%)Mean ± SD0.0045 ± 0.00070.0053 ± 0.0008* 0.0052 ± 0.0011* 0.0052 ± 0.0009* F1 generation females – cohort 1A (Week 13) N 20202020Thyroids and parathyroids weight – absolute (g)Mean ± SD0.015 ± 0.0020.015 ± 0.0010.015 ± 0.0020.015 ± 0.003Thyroids and parathyroids weight – relative to bw (%)Mean ± SD0.0068 ± 0.00120.0068 ± 0.00090.0068 ± 0.00080.0076 ± 0.0011* Abbreviations: N, Number of animals examined; SD, standard deviation.* p < 0.05 (Williams' test).** p < 0.01 (Williams' test). a One single animal showed a very high increased thyroid and parathyroids' weight (2.206 g). Thyroid weights, EOGRTS (#100). Abbreviations: N, Number of animals examined; SD, standard deviation. p < 0.05 (Williams' test). p < 0.01 (Williams' test). One single animal showed a very high increased thyroid and parathyroids' weight (2.206 g). The parental NOAEL was determined to be 8.65 mg/kg bw per day based on decreased T4 levels at 44.3 mg/kg bw per day and higher. This NOAEL is also supported by decreased body weight gain and minor histopathological findings in the stomach (mainly glandular dilatation). Similar to the findings in repeated dose toxicity studies following sub‐chronic and chronic exposure, parental generations treated with TFA Na showed changes in clinical chemistry parameters (i.e. decreased levels of glucose, non‐esterified fatty acids, triglycerides and cholesterol). Nevertheless, in the absence of associated histopathological findings or significant increase liver weight (higher than 15%), the WG did not consider the effects on clinical chemistry parameters at the dose level of 8.65 mg/kg bw per day as relevant for setting the NOAEL. The WG considered whether the decrease in thyroid hormone levels observed in the EOGRTS (#100) lead to a concern for developmental neurotoxicity (DNT). In this study, no effect on brain weight was observed in pups on PND22. In young adults (13 weeks old, exposed in utero and from PND22), a decrease in body weight and a slight decrease in the absolute brain weight were reported in both sexes in the high dose group only; but no histopathological findings were noted in the brain. This finding on brain weight was considered of low biological relevance due to the slight magnitude of the finding. The sensitive populations, i.e. dams, fetuses and newborns, were not investigated for thyroid toxicity (effects on T4, TSH and thyroid histopathology) as would be the case in a specific study (i.e. Comparative Thyroid Assay). Nevertheless, the WG considered that the NOAEL (or BMDL20) based on decreased T4 level in F1 offspring on PND22 is considered sufficiently conservative and protective of potential DNT effects, and that remaining uncertainties (lack of data in the target population) would be covered by the application of the standard UF of 100 (see Appendix G). Reduced sperm/spermatid counts, as well as increased abnormal sperm, were reported at the highest dose of 223 mg/kg bw per day in both generations (see Table 5). Decreased absolute testes and epididymis weights were also noted in the F1 generation (see Table 6). It is noted that the effects observed in this study were not associated with marked systemic toxicity and may be consistent with the decreased testosterone production by human chorionic gonadotropin (hCG)‐stimulated Leydig cells reported in the mechanistic in vitro rat study (#169, see Section 3.2.5 ). The changes in sperm motion parameters (average path velocity (VAP), curvilinear velocity (VCL), beat cross frequency (BCF)) observed in F1 generation only were of small magnitude, showed weak dose–response consistency and had no impact on overall sperm motility (see Table 5). However, when considered together with reduced sperm/spermatid counts, increased abnormalities and decreased testes and epididymis weights, these findings indicate a mild but coherent deterioration in male reproductive parameters at the highest dose (223 mg/kg bw per day). The WG noted that the assessment of sperm parameters (except sperm motility in both generations and sperm morphology in F1 generation) was conducted only at the highest dose. However, the WG considered the lack of data on sperm morphology at the low‐ and mid‐dose levels in F0 generation as a low uncertainty since this parameter was examined at the low‐ and mid‐dose levels in F1 generation and no treatment‐related effects were reported. Regarding the lack of data on sperm/spermatid counts at the low‐ and mid‐dose levels in both generations, the WG considered that, since treatment‐related decreased sperm (F0)/spermatid (F1) counts were observed at the high dose, there are uncertainties in setting a reproductive NOAEL for this study. The WG concluded that the reproductive LOAEL of this study is the highest dose level of 223 mg/kg bw per day. Usually, when a LOAEL is used for setting a health‐based guidance value, an additional uncertainty factor of 3 is applied. This would give a value of 74 mg/kg bw per day, i.e. higher than the parental/offspring NOAEL identified for this study (8.65 mg/kg bw per day). Therefore, the WG considered that the parental/offspring NOAEL would be sufficiently protective of reproductive effects. TABLE 5Sperm parameters, EOGRTS (#100).Sperm parametersppm0120/60600/3003000/1500mg/kg bw per day08.6544.3223 F0 generation males N 24 25 24 25 Motile sperm (%)Mean ± SD94 ± 493 ± 591 ± 592 ± 6Progressively motile sperm (%)Mean ± SD42 ± 940 ± 1038 ± 637 ± 10Cauda epididymis (left) – Weight (g)Mean ± SD0.253 ± 0.035––0.231 ± 0.027* Cauda epididymis (left) – Sperm count (millions/g)Mean ± SD484 ± 138––437 ± 116Cauda epididymis (left) – Total sperm count (million)Mean ± SD123 ± 41––101 ± 29* Testis (left) – Weight (g)Mean ± SD1.98 ± 0.15––1.95 ± 0.18Testis (left) – Spermatid count (millions/g)Mean ± SD91 ± 19––90 ± 15Testis (left) – Total spermatid count (million)Mean ± SD180 ± 36––175 ± 33Normal sperm (%)Mean ± SD97.1 ± 1.5––95.8 ± 2.3* Total abnormal sperm (%)Mean ± SD2.9 ± 1.5––4.2 ± 2.3* Head abnormal (%)Mean ± SD0.8 ± 0.8––1.0 ± 0.8Head flat (%)Mean ± SD0.7 ± 0.7––1.0 ± 0.8 F1 generation males N 20 20 20 20 Motile sperm (%)Mean ± SD93 ± 592 ± 594 ± 394 ± 4Progressively motile sperm (%)Mean ± SD44 ± 941 ± 1043 ± 1243 ± 12Cauda epididymis (left) – Weight (g)Mean ± SD0.204 ± 0.019––0.191 ± 0.025Cauda epididymis (left) – Sperm count (millions/g)Mean ± SD453 ± 93––478 ± 71Cauda epididymis (left) – Total sperm count (million)Mean ± SD92 ± 18––92 ± 21Testis (left) – Weight (g)Mean ± SD1.83 ± 0.14––1.70 ± 0.16** Testis (left) – Spermatid count (millions/g)Mean ± SD97 ± 19––81 ± 18** Testis (left) – Total spermatid count (million)Mean ± SD178 ± 39––136 ± 30** Normal sperm (%)Mean ± SD96.4 ± 2.296.6 ± 2.496.3 ± 2.694.3 ± 5.4Total abnormal sperm (%)Mean ± SD3.6 ± 2.23.4 ± 2.43.7 ± 2.65.7 ± 5.4Head abnormal (%)Mean ± SD1.1 ± 1.01.0 ± 0.71.0 ± 1.02.3 ± 2.1* Head flat (%)Mean ± SD1.0 ± 0.90.7 ± 0.60.9 ± 1.02.1 ± 1.9* Average path velocity (VAP) (μm/s)Mean ± SD146 ± 12137 ± 12* 138 ± 14* 140 ± 12* Curvilinear velocity (VCL) (μm/s)Mean ± SD357 ± 28337 ± 37328 ± 7* 331 ± 36* Beat cross frequency (BCF) (Hz)Mean ± SD36 ± 235 ± 235 ± 335 ± 3* Abbreviations: N, Number of animals examined; –, not measured; SD, standard deviation.* p < 0.05 (t‐test).** p < 0.01 (t‐test). Sperm parameters, EOGRTS (#100). Abbreviations: N, Number of animals examined; –, not measured; SD, standard deviation. p < 0.05 (t‐test). p < 0.01 (t‐test). TABLE 6Testes and epididymis weights, EOGRTS (#100).Testes and epididymis weights # ppm0120/60600/3003000/1500mg/kg bw per day08.6544.3223 F0 generation males N 25 25 25 25 Testes weight – absolute (g)Mean ± SD3.810 ± 0.5734.105 ± 0.295* 3.751 ± 0.7313.864 ± 0.368Testes weight – relative to body weight (%)Mean ± SD0.842 ± 0.1450.894 ± 0.0760.845 ± 0.1780.903 ± 0.105Epididymis weight – absolute (g)Mean ± SD1.459 ± 0.1721.522 ± 0.1151.389 ± 0.2341.425 ± 0.117Epididymis weight – relative to body weight (%)Mean ± SD0.322 ± 0.0450.331 ± 0.0310.314 ± 0.0550.333 ± 0.032 F1 generation males – cohort 1A N 20 20 20 20 Testes weight – absolute (g)Mean ± SD3.635 ± 0.2633.572 ± 0.2573.732 ± 0.2453.352 ± 0.343** Testes weight – relative to body weight (%)Mean ± SD0.967 ± 0.0860.968 ± 0.0761.000 ± 0.1000.986 ± 0.091Epididymis weight – absolute (g)Mean ± SD1.252 ± 0.1011.234 ± 0.1081.246 ± 0.1191.156 ± 0.136*Epididymis weight – relative to body weight (%)Mean ± SD0.333 ± 0.0300.335 ± 0.0300.333 ± 0.0360.340 ± 0.035 F1 generation males – cohort 1B N 20 20 20 20 Testes weight – absolute (g)Mean ± SD3.724 ± 0.4233.740 ± 0.2873.828 ± 0.2993.473 ± 0.276*Testes weight – relative to body weight (%)Mean ± SD0.956 ± 0.1260.963 ± 0.1080.979 ± 0.1000.934 ± 0.069Epididymis weight – absolute (g)Mean ± SD1.299 ± 0.1471.330 ± 0.1451.316 ± 0.1211.217 ± 0.105Epididymis weight – relative to body weight (%)Mean ± SD0.334 ± 0.0480.342 ± 0.0410.336 ± 0.0350.328 ± 0.031Abbreviations: N, Number of animals examined; SD, standard deviation. # Left and right.* p < 0.05 (Williams' test).** p < 0.01 (Williams' test). Testes and epididymis weights, EOGRTS (#100). Abbreviations: N, Number of animals examined; SD, standard deviation. Left and right. p < 0.05 (Williams' test). p < 0.01 (Williams' test). F1 cohort 1A showed changes in immunophenotyping. Results are discussed under immunotoxicity (see Section 3.2.3.4.1 ). Developmental toxicity Developmental toxicity was investigated in rats (#6, #7, #50) and rabbits (#87, #88, #101, #102, #103, #130, #162) (see Appendix E). The most sensitive species to the developmental toxicity of TFA Na is the rabbit as evidenced by different eye malformations, e.g. multiple folded retina, absent aqueous/vitreous humour in several studies. The relevant developmental toxicity NOAEL is 60 mg/kg bw per day in an OECD TG 414 (#103) based on different eye malformations at 250 mg/kg bw per day (see Table 7). Some skeletal findings (i.e. fused and partially fused ribs and sternebrae), reduced fetal weight and increased post‐implantation loss were observed at the dose level of 750 mg/kg bw per day (Tables 8, 9, 10, respectively). TABLE 7Fetal examinations – major abnormality findings – group incidences/group mean percentage of litters (developmental toxicity study in rabbits, #103).Dose (mg/kg bw per day)03060250750Number examined (Litters)2829282828 N/% N/% N/% N/% N/%Multiple folded retina0/00/00/02/7.19/32.1** Absent aqueous/vitreous0/00/00/01/3.67/25.0** Misshapen lens(es)0/00/00/00/02/7.1Haemorrhage aq./vit.0/00/00/00/05/17.9* Abbreviations: N, number of affected animals.* p < 0.05.** p < 0.01.– The proportion of litters with at least one fetus affected were analysed using Fisher's (Thakur, 1985) exact tests (upper tail). The percentage of affected fetuses for each animal/litter were analysed using a Kruskal–Wallis nonparametric ANOVA (Lehmann, 1975).– If the Kruskal–Wallis test was significant (p ≤ 0.05), pairwise comparisons of each treated group with the control group were made using the Wilcoxon rank sum test (Lehmann, 1975).– If the Kruskal–Wallis test was not significant (p > 0.05), no further analyses were conducted. Where only two groups were available for analysis a Wilcoxon rank sum test was performed. Fetal examinations – major abnormality findings – group incidences/group mean percentage of litters (developmental toxicity study in rabbits, #103). Abbreviations: N, number of affected animals. p < 0.05. p < 0.01. – The proportion of litters with at least one fetus affected were analysed using Fisher's (Thakur, 1985) exact tests (upper tail). The percentage of affected fetuses for each animal/litter were analysed using a Kruskal–Wallis nonparametric ANOVA (Lehmann, 1975). – If the Kruskal–Wallis test was significant (p ≤ 0.05), pairwise comparisons of each treated group with the control group were made using the Wilcoxon rank sum test (Lehmann, 1975). – If the Kruskal–Wallis test was not significant (p > 0.05), no further analyses were conducted. Where only two groups were available for analysis a Wilcoxon rank sum test was performed. TABLE 8Fetal examinations – skeletal abnormality findings (prenatal developmental toxicity study in rabbits, #103).Dose (mg/kg bw per day)FetusesLitters0306025075003060250750Group 1Group 2Group 3Group 4Group 5Group 1Group 2Group 3Group 4Group 5 Number examined 209 244 219 193 178 28 29 28 28 28 N/% N/% N/% N/% N/% N/% N/% N/% N/% N/%Ribs Fused/partially fused0/01/0.40/01/0.46/2.80/01/3.40/01/3.63/10.7Sternebrae Fused/partially fused1/0.50/03/2.03/1.53/1.61/3.60/02/7.13/10.73/10.7Abbreviations: N, number of affected animals.* p < 0.05.** p < 0.01.– The proportion of litters with at least one fetus affected were analysed using Fisher's (Thakur, 1985) exact tests (upper tail). The percentage of affected fetuses for each animal/litter was analysed using a Kruskal–Wallis non‐parametric ANOVA (Lehmann, 1975).– If the Kruskal–Wallis test was significant (p ≤ 0.05), pairwise comparisons of each treated group with the control group were made using the Wilcoxon rank sum test (Lehmann, 1975).– If the Kruskal–Wallis test was not significant (p > 0.05), no further analyses were conducted. Where only two groups were available for analysis a Wilcoxon rank sum test was performed. Fetal examinations – skeletal abnormality findings (prenatal developmental toxicity study in rabbits, #103). Abbreviations: N, number of affected animals. p < 0.05. p < 0.01. – The proportion of litters with at least one fetus affected were analysed using Fisher's (Thakur, 1985) exact tests (upper tail). The percentage of affected fetuses for each animal/litter was analysed using a Kruskal–Wallis non‐parametric ANOVA (Lehmann, 1975). – If the Kruskal–Wallis test was significant (p ≤ 0.05), pairwise comparisons of each treated group with the control group were made using the Wilcoxon rank sum test (Lehmann, 1975). – If the Kruskal–Wallis test was not significant (p > 0.05), no further analyses were conducted. Where only two groups were available for analysis a Wilcoxon rank sum test was performed. TABLE 9Overall fetal and litter weight in rabbits on Day 29 of gestation (prenatal developmental toxicity study in rabbits, #103).Dose (mg/kg bw per day)Litters03060250750Mean/SDGroup 1Group 2Group 3Group 4Group 5 N 2829282828Total litter weight (g)299.9/87.6335.8/71.22309.8/94.26265.0/89.59232.7**/84.36Overall fetal weight (g)41.1/5.3940.0/3.5940.4/4.9740.3/5.7637.4**/4.81Abbreviations: N, Number of animals examined; SD, standard deviation.**p < 0.01 – *p < 0.05: treated groups compared with control using Williams' test. Overall fetal and litter weight in rabbits on Day 29 of gestation (prenatal developmental toxicity study in rabbits, #103). Abbreviations: N, Number of animals examined; SD, standard deviation. **p < 0.01 – *p < 0.05: treated groups compared with control using Williams' test. TABLE 10Post‐implantation loss (%) in rabbits (prenatal developmental toxicity study in rabbits, #103).Dose (mg/kg bw per day)Litters03060250750Mean/SDGroup 1Group 2Group 3Group 4Group 5 N 2829282828Post‐implantation loss (%)8.1/13.137.7/9.414.6/7.7416.4/21.8919.4*/22.16Abbreviations: N: Number of animals examined; SD, standard deviation. Note: Data were square root transformed for the statistical analysis. **p < 0.01 – *p < 0.05‐treated groups compared with control using Williams' test. Post‐implantation loss (%) in rabbits (prenatal developmental toxicity study in rabbits, #103). Abbreviations: N: Number of animals examined; SD, standard deviation. Note: Data were square root transformed for the statistical analysis. **p < 0.01 – *p < 0.05‐treated groups compared with control using Williams' test. The maternal NOAEL is 30 mg/kg bw per day based on increased absolute liver and kidney weight with histopathological correlation in the liver at 60 mg/kg bw per day. In addition, satellite phase dams were investigated for toxicokinetics, blood chemistry, haematology and both maternal and fetal aqueous/vitreous humour. An additional developmental toxicity study in rabbits was submitted to ECHA RAC to characterise the changes of the visual pathway of rabbits. 24 In this study, two abortions were found at the dose level of 375 mg/kg bw per day (only one dose level was tested). Study results are not impacting the lowest NOAEL/LOAEL setting for developmental toxicity. In a supplementary study (#88), eye malformations were observed in rabbit fetuses at the lowest dose level of 180 mg/kg bw per day, whereas in dams, only minimal liver effects were noted. Regarding the moderate severity of maternal toxicity in different studies, the findings in rabbit fetuses were not considered to be secondary to toxic effects observed in dams. TFA did not induce maternal and developmental toxicity in rats up to 150 mg/kg bw per day (highest dose level tested, #7). The WG noted that the multiple anomalies observed were found in six individual fetuses from one single litter (1 out of 21) at 150 mg/kg bw per day (see Table 11). In the rat reproductive toxicity study (EOGRTS, #100), rosettes/folds in the retina were observed in three females of the F1 generation at histopathological examination; however, it is noted that rosettes were already observed in the male control group of the F0 generation (see Table 12). TABLE 11Abnormalities observed in fetal and litters examinations – (developmental toxicity study in rats, #7).Dose level (mg/kg per day)Dam No.‐fetus No.ExternalVisceralSkeletal0––––37.52504‐3– Subclavian artery: retroesophageal–75––––1504501‐1 Umbilicus: omphalocele Hindlimbs: flexure, malrotated – Ribs: partially fused4501‐2 Umbilicus: omphalocele Abdomen: omphalocele Eye: retina, bilateral, folded –4501‐3 Umbilicus: omphalocele– Ribs: partially fused Sternum: partially cleft/split Thoracic vertebral arches: partially fused 4501‐4 Abdomen: gastroschisis Hindlimbs: flexure, malrotated – Ribs: partially fused Thoracic vertebral arches: partially fused 4501‐6 Umbilicus: omphalocele Abdomen: omphalocele–4501‐9 Umbilicus: omphalocele Ribs: partially fused Abnormalities observed in fetal and litters examinations – (developmental toxicity study in rats, #7). Umbilicus: omphalocele Hindlimbs: flexure, malrotated Abdomen: omphalocele Eye: retina, bilateral, folded Ribs: partially fused Sternum: partially cleft/split Thoracic vertebral arches: partially fused Abdomen: gastroschisis Hindlimbs: flexure, malrotated Ribs: partially fused Thoracic vertebral arches: partially fused TABLE 12Histopathology of the eyes, EOGRTS in the rat (#100).MalesFemalesDose (ppm)0120/60600/3003000/15000120/60600/3003000/1500 Dose (mg/kg bw per day) 0 8.65 44.3 223 0 8.65 44.3 223 F0 generation N 25 0 0 25 23 0 0 23 Eyes – Rosettes, Retina (minimal)3––00––0 F1 generation – Cohort 1A N 20 0 0 20 20 0 0 20 Eyes – Rosettes/Folds, Retina (minimal)0––00––3Abbreviation: N, Number of animals examined. Histopathology of the eyes, EOGRTS in the rat (#100). Abbreviation: N, Number of animals examined. Other toxicological effects Immunotoxicity The WG assessed whether TFA affected immunological endpoints across available repeated‐dose toxicity and reproductive toxicity studies. The assessment was based on relevant immunological endpoints as described by Chemicals Regulation Directorate (CRD, 2015). Within the available toxicity data package on TFA no consistent or dose‐dependent effects on immunological endpoints were observed across studies (Annex 10). For example, a decrease in lymphocyte count was observed at the top dose in male rats in F0 and F1 (#100) and top dose in female rats (#15). However, these findings were not observed in other studies up to the high dose level tested (more than 1000 mg/kg bw per day, #8) (see Tables 13, 14, 15). TABLE 13Female rat lymphocytes (14‐day toxicity study in rat; #15).GroupControl234Dose in mg/kg bw per day045.4191.08189.59 N 5555Mean ± SDMean ± SDMean ± SDMean ± SDFemale lymphocyte count9.3 ± 1.38.3 ± 1.77.1 ± 1.35.8 ± 1.5* Abbreviations: N, Number of animals examined; SD, standard deviation.* p ≤ 0.05 (Dunnett). Female rat lymphocytes (14‐day toxicity study in rat; #15). Abbreviations: N, Number of animals examined; SD, standard deviation. p ≤ 0.05 (Dunnett). TABLE 14Male and female rat lymphocytes (90‐day toxicity study in rat; #8).MaleFemaleGroup12341234Dose in mg/kg bw per day09.9981043012.21231216 N 101091010101010Mean ± SDMean ± SDMean ± SDMean ± SDMean ± SDMean ± SDMean ± SDMean ± SDLymphocyte count (10E9/L)12.0 ± 3.210.0 ± 2.211.3 ± 2.110.6 ± 3.57.5 ± 2.27.1 ± 1.66.4 ± 1.56.1 ± 1.4Lymphocyte count (%)81 ± 378 ± 480 ± 678 ± 481 ± 580 ± 476 ± 679 ± 5 Note: Bart; NSg‐05/Anova: NSg‐05/No unplanned test performed.Abbreviations: N, Number of animals examined; SD, standard deviation. Male and female rat lymphocytes (90‐day toxicity study in rat; #8). Note: Bart; NSg‐05/Anova: NSg‐05/No unplanned test performed. Abbreviations: N, Number of animals examined; SD, standard deviation. TABLE 15Haematology – monocytes and lymphocytes changes in male rats (EORGTS, #100).Dose (ppm)01206003000Group 1MGroup 2MGroup 3MGroup 4M F0 generation N 10 10 10 10 Dose (mg/kg bw per day) 0 9.71 49.1 248 Lymphocytes (× 109/L) (mean/SD)2.74/0.4093.62/2.7313.48/0.9362.80/0.557Monocytes (× 109/L) (mean/SD)0.12/0.0310.14/0.0910.10/0.0340.08*/0.031 F1 generation N 10 10 10 10 Dose (mg/kg bw per day) 0 9.37 47.3 242 Lymphocytes (× 109/L) (mean/SD)4.08/3.1163.27/0.8602.97/0.5532.59*/0.558Monocytes (× 109/L) (mean/SD)0.12/0.0940.12/0.0330.11/0.0570.07*/0.026Abbreviations: N, Number of animals examined; SD, standard deviation. Haematology – monocytes and lymphocytes changes in male rats (EORGTS, #100). Abbreviations: N, Number of animals examined; SD, standard deviation. Evidence from retrospective analysis on pesticide active substances (CRD, 2015; Gehen et al., 2014; US EPA, 2013) indicates that immunotoxicity, as assessed by the T‐cell‐dependent antibody response (TDAR) assays in adult animals, or by analysis of standard immunotoxicity endpoints, is rarely a major driver for deriving the reference point for risk assessment (see Appendix I). Immunophenotyping of T cells (including CD4+ and CD8+ T subsets), B cells, natural killer cells, monocytes and neutrophils was performed by flow cytometry on leucocytes isolated from the spleen of rats (F1 cohort 1A, week 13) in the EOGRTS (#100); however, a cohort for immunotoxicity was not included in this study. A decrease in absolute cell counts in the spleen was observed in both sexes at all dose levels for all previously analysed cell populations. The relative composition of lymphocyte subsets remained unchanged, which indicates no shift in CD4+/CD8+ ratio or other cellular distributions (see Table 16). TABLE 16Male and female rat spleen leucocytes count, white blood cells/spleen (F1 cohort 1A, week 13; EOGRTS (#100)).Dose in ppm0120/60600/3003000/1500Dose in mg/kg bw per day08.6544.3223 N 10101010Mean ± SDMean ± SD (%)Mean ± SD (%)Mean ± SD (%)Male – spleen leucocytes (count)12 E+07 ± 4 E+07 10 E+07 ± 4 E+07 (−12) 8 E+07 ± 2 E+07 (−27) 1 E+07 ± 4 E+07 (−33) Female – spleen leucocytes (count)12 E+07 ± 3 E+07 10 E+07 ± 2 E+07 (−14) 9 E+07 ± 3 E+07 (−22) 8 E+07 ± 3 E+07 (−33) Male – CD4+/CD8+ ratio2.4 ± 0.62.5 ± 0.52.8 ± 0.52.7 ± 05Female – CD4+/CD8+ ratio2.7 ± 0.42.4 ± 0.72.4 ± 0.62.5 ± 0.7Abbreviations: N, Number of animals examined; SD, standard deviation. Note: No statistical analysis conducted. % Effect size. Male and female rat spleen leucocytes count, white blood cells/spleen (F1 cohort 1A, week 13; EOGRTS (#100)). 10 E+07 ± 4 E+07 (−12) 8 E+07 ± 2 E+07 (−27) 1 E+07 ± 4 E+07 (−33) 10 E+07 ± 2 E+07 (−14) 9 E+07 ± 3 E+07 (−22) 8 E+07 ± 3 E+07 (−33) Abbreviations: N, Number of animals examined; SD, standard deviation. Note: No statistical analysis conducted. Effect size. The observed decreases in absolute splenic immune cell numbers across doses suggest a potential immunomodulatory effect of TFA, although their biological significance remains uncertain since immunophenotyping provides mechanistic insight into the steady‐state lymphocyte subset distributions and absolute cell numbers in the spleen, but does not directly assess the immune function, such as increased susceptibility to infection or the ability to generate an effective immune response. While reductions in the number of splenic immune cells occurred in both sexes, lymphocyte subset distribution and the CD4+/CD8+ ratio were unchanged. In addition, no corroborating histopathological alterations were reported. However, the functional immune competence (e.g. the capacity to generate antibody response or resist infection) was not evaluated. In the absence of functional immunotoxicity testing, these findings introduced uncertainty regarding possible developmental immunotoxicity since effects were observed in animals where exposure has already happened in utero. This uncertainty was addressed through an additional database‐related uncertainty factor, as the available data are insufficient to establish a NOAEL specifically for immune effects in the developing organism. Neurotoxicity No specific oral neurotoxicity studies with TFA have been submitted. Animals administered TFA in repeated dose studies (#8, #15, #16, #109, #124) and in the EOGRTS (#100) did not show clinical signs of neurotoxicity nor were the effects observed for exploratory motor activity, in open field observations, in sensory reactivity, or in grip strength (#8). There were also no treatment‐related effects on brain weight (#8, #15, #16, #100, #109, #124) and histopathology (#8, #16, #100, #109, #124) in adults, where these were subject to investigation. Endocrine disruption As described above, evidence of endocrine disruption was observed in the EOGRTS (#100) and discussed in Section 3.2.3.3.1 . Observations in humans Biomonitoring Three biomonitoring studies (#91, #94 and #96) measured TFA in several human matrices, with one of them reporting only qualitative results. All three were evaluated using the dedicated Risk of Bias (RoB) tool (see Appendix E). Based on this assessment, study #91 was assigned to Tier 1 and study #94 to Tier 2. Consequently, only these two studies (both measuring serum concentrations in non‐European populations) were retained for further assessment (see Table 17). Study #96, which analysed urine samples from a Belgian population, was classified as Tier 3 and excluded from further consideration due to a high RoB. This study relied on a non‐targeted suspected‐screening analytical approach, reported only presence/absence data without quantitative measurements, and did not employ a TFA reference standard. Compound identification was based on indirect MS/MS spectral confirmation, resulting in only tentative identification of TFA and low analytical confidence. The two biomonitoring studies appraised as Tier 1 or Tier 2 following RoB analysis are summarised in Table 17. It should be noted that both cross‐sectional studies, which measured TFA in serum, were conducted outside Europe. Overall, the available human biomonitoring evidence provides qualitative support for internal exposure to TFA in humans, indicating systemic absorption and background population levels, but is limited by methodological constraints and RoB considerations, precluding its use for quantitative hazard characterisation. Epidemiological studies Among the identified biomonitoring studies, only study #91 (summarised in Annex 5) included outcome data in addition to exposure measurements. For this reason, it was additionally evaluated as an epidemiological study using the National Toxicology Program (NTP) Office of Health Assessment and Translation (OHAT) tool (NTP, 2015, 2016) (see Annex 5), which includes criteria relevant to outcome assessment, confounding, attrition and statistical analysis. This study received an overall Tier 3 rating due to a ‘probably high risk of bias’ judgement in a key confounding domain (confounding bias) and was therefore not retained for further consideration. Mode of action of TFA effects Six in vitro mechanistic studies were identified as Tier 1 or Tier 2 following RoB analysis ( Annexes 6 and 7 ) and summarised below: TFA did not show a concentration–response relationship on the differentiation of stem cells in the in vitro embryonic stem cell test (#10); TFA did not show a concentration–response relationship on the differentiation of stem cells in the in vitro embryonic stem cell test (#10); TFA when exposed to DLD‐1 human colorectal adenocarcinoma cells is transported by the human monocarboxylate transporter 1 (MCT1), assessed by quantifying intracellular uptake based on acidification of these cells with an EC50 of 12 mM. MCT1‐mediated cellular uptake was evidenced by the inhibition of its uptake in the presence of the specific and potent MCT1 inhibitor AR‐C155858 (#114); TFA when exposed to DLD‐1 human colorectal adenocarcinoma cells is transported by the human monocarboxylate transporter 1 (MCT1), assessed by quantifying intracellular uptake based on acidification of these cells with an EC50 of 12 mM. MCT1‐mediated cellular uptake was evidenced by the inhibition of its uptake in the presence of the specific and potent MCT1 inhibitor AR‐C155858 (#114); TFA Na did not activate any of the stress response pathways measured in seven human reporter HepG2 cell lines (#131); TFA Na did not activate any of the stress response pathways measured in seven human reporter HepG2 cell lines (#131); Exposure to TFA Na had no significant impact on gene expression profiles or cellular morphology/function in human induced pluripotent stem cells differentiating into cardiomyocyte‐, hepatocyte‐ and neuronal‐like cells (#132); Exposure to TFA Na had no significant impact on gene expression profiles or cellular morphology/function in human induced pluripotent stem cells differentiating into cardiomyocyte‐, hepatocyte‐ and neuronal‐like cells (#132); TFA was found to significantly decrease testosterone production in hormonally stimulated isolated rat Leydig cells following 5‐h exposure (whereas no clear and consistent effect on testosterone production was observed following 24‐h exposure). TFA was also found to enhance lactate production by Sertoli cell‐only cultures hormonally stimulated; the relevance of this single finding, in the absence of effect on the morphology of the cells and on effects on pyruvate production, is unknown according to the authors. Overall, TFA showed a slight effect on Leydig cell function and no clear effect on the viability and function of Sertoli cell‐only cultures, and Sertoli‐germ cell co‐cultures derived from Sprague Dawley rats (#169). TFA was found to significantly decrease testosterone production in hormonally stimulated isolated rat Leydig cells following 5‐h exposure (whereas no clear and consistent effect on testosterone production was observed following 24‐h exposure). TFA was also found to enhance lactate production by Sertoli cell‐only cultures hormonally stimulated; the relevance of this single finding, in the absence of effect on the morphology of the cells and on effects on pyruvate production, is unknown according to the authors. Overall, TFA showed a slight effect on Leydig cell function and no clear effect on the viability and function of Sertoli cell‐only cultures, and Sertoli‐germ cell co‐cultures derived from Sprague Dawley rats (#169). TFA suppressed antibody production (immunoglobulins IgG and IgM) in human peripheral blood mononuclear cells (PBMCs) following 7‐day exposure (24‐h pre‐exposure, followed by 6‐day stimulation of exposed cells) (Iulini et al., 2025). The WG noted that the concentration‐related decrease was clearer for IgM release in female PBMCs compared to the decrease of IgG in female PBMCs or IgG or IgM releases in male PBMCs. The study showed some limitations in reporting: non‐normalised data not shown, results for TFA are presented in a figure only (non‐tabulated data) and it is not clear whether the statistical analysis has been performed in normalised or non‐normalised data. The results pointed to a potential immunomodulatory mode of action for TFA. TFA suppressed antibody production (immunoglobulins IgG and IgM) in human peripheral blood mononuclear cells (PBMCs) following 7‐day exposure (24‐h pre‐exposure, followed by 6‐day stimulation of exposed cells) (Iulini et al., 2025). The WG noted that the concentration‐related decrease was clearer for IgM release in female PBMCs compared to the decrease of IgG in female PBMCs or IgG or IgM releases in male PBMCs. The study showed some limitations in reporting: non‐normalised data not shown, results for TFA are presented in a figure only (non‐tabulated data) and it is not clear whether the statistical analysis has been performed in normalised or non‐normalised data. The results pointed to a potential immunomodulatory mode of action for TFA. Findings from a 14‐day in vivo oral toxicity study (#15) suggest that TFA may act as a weak peroxisome proliferator in male rats, as evidenced by increased palmitoyl‐CoA oxidation activity, a biomarker of peroxisomal β‐oxidation and peroxisome proliferator‐activated receptors (PPARs) activation. PPARs are nuclear receptors that play a central role in regulating lipid and glucose metabolism (Kim & Ahn, 2004; Kim et al., 2003), and their activation can influence energy homeostasis and metabolic adaptation. In addition, PPARs (particularly PPARγ and PPARα) exert important immunomodulatory functions by linking cellular metabolism to immune cell differentiation and activity; for example, PPARγ promotes anti‐inflammatory macrophage polarisation and regulates T‐cell responses, while PPARα can suppress pro‐inflammatory signalling pathways such as NF‐κB. Through these mechanisms, PPAR activation generally shifts immune responses towards a less inflammatory state in a context‐ and ligand‐dependent manner (Christofides et al., 2021). In regulatory studies, TFA exposure was also associated with decreases in total bilirubin, cholesterol, triglycerides and glucose. In the absence of histopathological evidence of hepatocellular degeneration, this pattern is consistent with an adaptive hepatic response characterised by enhanced metabolic and clearance capacity rather than direct toxicity (Dekant & Dekant, 2023; Felter et al., 2018; Hall et al., 2012; Sugatani et al., 2001). Evidence for PPARα involvement is limited, and TFA appears to act, at most, as a weak peroxisome proliferator in rodents and rodent‐specific PPARα‐mediated proliferative responses are considered of low human relevance (Felter et al., 2018; Dekant & Dekant, 2023;). Reductions in circulating bilirubin are more plausibly attributed to induction of hepatic conjugation and transport processes, including UDP‐glucuronosyltransferase 1A1 (UGT1A1), which is primarily regulated via constitutive androstane receptor (CAR)‐mediated pathways (Sugatani et al., 2001). Taken together, these receptor‐mediated responses offer the most plausible explanation for the biochemical alterations reported, although they do not constitute definitive proof of a specific molecular target. Consideration of non‐critical effects During the public consultation on this Scientific Report, numerous comments were received concerning the outputs of UBA (2020) and RIVM (2023), particularly with regard to the consideration of increased alanine aminotransferase (ALT) levels and decreased bilirubin levels observed in the available rat studies with TFA as relevant parameters for derivation of health‐based guidance values. In addition, the EFSA WG on TFA organised a Technical Hearing with various stakeholders (including NGOs, industry representatives, UBA and RIVM) to explicitly gather their views on the selection of relevant parameters and reference points used by them and those to be applied in the current assessment of TFA. Following these activities, the WG carefully examined all arguments submitted. A comprehensive summary of the already available and newly submitted evidence, relevant to decide upon the relevance and benchmark for adversity of the following parameters has been prepared: Increase in alanine aminotransferase (ALT) Increase in alanine aminotransferase (ALT) International guidelines for human safety assessment in evaluation of drugs consistently identify ALT elevation as a biomarker of liver toxicity in both non‐clinical and clinical settings. Increases of two to three times the upper limit of normal (ULN) are generally considered indicative of hepatocellular damage in humans (EMA, 2010; FDA, 2009; HED, 2002), while in assessment of preclinical studies, increases of twofold to fourfold or higher compared with concurrent controls should raise concern unless an alternative explanation exists (Boone et al., 2005). A WoE approach recommends assessing ALT changes alongside other biomarkers of liver toxicity such as aspartate aminotransferase (AST), alkaline phosphatase (ALP), γ‐glutamyltransferase (GGT) and glutamate dehydrogenase (GLDH), as well as clinical chemistry indicators like increase in bilirubin (Hall et al., 2012). In standard toxicity testing, mean ALT increases of twofold to fourfold in dogs or rats may raise concern and increases greater than threefold to fivefold are considered adverse even without histological changes (Boone et al., 2005; FDA, 2009). According to FDA guidance (2009), treatment discontinuation should be considered if ALT or AST exceeds eight times the ULN, remains above five times the ULN for more than 2 weeks or is greater than three times the ULN when accompanied by increased bilirubin or clinical symptoms. Health Canada (2012) applies similar thresholds, defining as one of the criteria for liver toxicity ALT values greater than two to three times ULN. JMPR (2015) notes that minor enzyme fluctuations may be adaptive, with changes greater than 50 % in animal studies serving as a starting point for concern. In Corton et al. (2018), analysis of data indicates that increase in ALT above 141% was related to tumorigenic outcome in long‐term rat studies; however, change in ALT was not observed as an isolated effect and histopathological information was not available in the primary Toxicogenomics Project–Genomics Assisted Toxicity Evaluation System (TG‐GATES) data set used for deriving the thresholds. Beyond its role as a marker of hepatocellular damage, increased intracellular ALT expression has also been linked to the administration of PPAR agonist drugs such as fibrates, suggesting that this mechanism may contribute to elevated serum ALT activity (Thulin et al., 2008). Overall, biological relevance and consistency across parameters are more important than statistical significance alone and although ALT is a relevant clinical pathology parameter for detecting chemically induced hepatocellular injury, the best approach remains an integrated evaluation of clinical pathology, histology and other study data (e.g. liver weight) (Boone et al., 2005; Hall et al., 2012; Ramaiah et al., 2017; Siska et al., 2022). Consequently, in the absence of histopathological evidence of liver injury or sustained functional impairment, elevated ALT level is not considered a critical endpoint for the derivation of health‐based guidance values. Decrease in bilirubin Decrease in bilirubin The vast majority of references on bilirubin changes indicate that increase, rather than decrease, is generally associated with hepatic toxicity. Hall et al. (2012) notes that elevated bilirubin, especially when accompanied by increased bile acids, is a reliable indicator of impaired hepatic function, while decreases may be adaptive if linked to enhanced conjugation and excretion. ECETOC (2002) suggests that isolated changes beyond historical control values without histopathological correlation are typically non‐adverse. EMA (2010) and FDA (2009) guidelines emphasise that concurrent increases in ALT and bilirubin should be given particular attention, as this pattern is associated with risk of liver failure in humans. Health Canada (2012) defines as one of criteria for liver toxicity conjugated bilirubin levels greater than two times ULN. In Corton et al. (2020), analysis of data indicates that increase in bilirubin above 115% was related to tumorigenic outcome in rat long‐term studies; however, change in bilirubin was not observed as an isolated effect and histopathological information was not available in the primary TG‐GATES data set used for deriving the thresholds. As regards decrease in bilirubin, JMPR (2015) states that low bilirubin levels are generally not of concern and often reflect microsomal enzyme induction. Mechanistically, this effect can be attributed to the upregulation of hepatic conjugation pathways, primarily by UGT1A1, and elimination via canalicular transporters such as MRP2 into the bile (Köck & Brouwer, 2012). There is some indication that PPARα agonists activate bilirubin metabolism in liver cells and its elimination by induction of PPARα‐dependent UGT1A1 (Bigo et al., 2014). In addition to the well‐known origin of bilirubin as a natural yellow pigment produced during the breakdown of haemoglobin, there is evidence that bilirubin is a direct natural PPARα ligand that enhances fatty acid oxidation and alters lipoprotein metabolism (Kipp et al., 2025; Lee et al., 2025). Recent publications (Creeden et al., 2021; Kaur et al., 2025; Kipp et al., 2025; Lee et al., 2025) postulate bilirubin protective function stating that lower bilirubin level in humans can be related to higher risk of metabolic and cardiovascular diseases, obesity, type 2 diabetes, chronic kidney disease, systemic inflammation, ulcerative colitis, cognitive decline and sleep apnoea. However, the ‘normal’ reference ranges for bilirubin are debatable and need to be established across different subpopulations by taking into account potential confounders like sex, ethnicity, age, diet, smoking status, alcohol consumption, underlying liver disease, hemolytic disease, circadian rhythms, medication, physical activity, fasting status and other relevant variables (Vítek, 2019; Vítek et al., 2023). Because high variability in measured bilirubin levels is observed across analytical methods and automated clinical chemistry instruments, and bilirubin assays lack full methodological standardisation (Apperloo et al., 2005), these are considered among the least reliable clinical chemistry assays, resulting in poor comparability across epidemiological studies. Large, well‐designed population studies are strongly recommended to overcome these limitations (Vítek, 2019; Vítek et al., 2023). No regulatory thresholds for decrease in bilirubin could be identified in regulatory guidance or relevant literature, either for observations in animals' toxicity studies or for clinical settings. In context of cumulative risk assessment, clinical chemistry parameters and liver enzymes are ancillary and insufficiently specific to identify a liver specific toxicological effect, in contrast to liver‐specific effects like histopathological lesions and increases in relative liver weight of ≥ 15% (EFSA, 2025). Consequently, in the absence of histopathological evidence of liver injury or sustained functional impairment, elevated bilirubin level is not considered a critical endpoint for the derivation of health‐based guidance values. Changes in cholesterol and glucose levels Changes in cholesterol and glucose levels In addition to ALT and bilirubin, the WG considered that observed changes in cholesterol and glucose levels in animal studies may be mechanistically linked to a mild nuclear receptor (e.g. PPARα) activation and subsequent peroxisome proliferation, processes known to modulate lipid and carbohydrate homeostasis in rodents. While these alterations were observed even at low doses tested, the absence of concurrent histopathological changes in key target tissues (such as the liver or endocrine pancreas) is considered indicative of an adaptive physiological response rather than an adverse effect. Although high‐dose hepatocellular hypertrophy further supports a possible PPAR‐mediated mode of action, the weight of evidence remains limited by the absence of robust, mechanistically targeted studies directly confirming PPAR involvement in TFA's biological activity. Overall, these clinical chemistry parameters are not considered critical endpoints for the derivation of health‐based guidance values. Consideration of critical effects and reference values Critical effects As described above (see Section 3.2.3 ), studies in rats have shown that the target organs of TFA are thyroid (reproductive toxicity), liver (repeated dose toxicity) and fetal development (developmental studies). The effects seen at the lowest doses involve the thyroid (decrease in T4) and liver (increase weight and histopathology) (see Tables 2 and 5). Effects on development (eye malformations) have been reported at doses above 60 mg/kg bw per day (see Table 7). Dose–response analysis The dose response analysis of the critical effects identified in experimental animal studies was performed by means of the benchmark dose modelling (BMD). The EFSA guidance on the use of the benchmark dose modelling in risk assessment (EFSA Scientific Committee, 2022) was used. The full details of the BMD modelling are reported in Annex 11. The BMD modelling was considered and reported for all critical effects identified in Section 3.2.7.1 . For the changes in thyroid, decreased T4 levels observed in adults and pups in the EOGRTS were modelled, considering a benchmark response (BMR) of 20% (see Appendix G). Independent analyses were performed for levels measured in adult rats (from parental and F1 generations) and pups (F1 generation, PND22). The BMD analyses resulted in the lowest benchmark dose lower confidence limit (BMDL20) of 8.6 mg TFA Na/kg bw per day (adult male rats, F1), a value well in line with the NOAEL discussed in Section 3.2.3.3.1. Increased relative liver weights and hepatocyte hypertrophy, observed upon subchronic exposure, were considered for BMD modelling of hepatic effects. For the analysis of the increased relative liver weight, a benchmark response (BMR) of 15% was selected as a relevant effect size, as recommended by JMPR (2015) and applied as relevant threshold in the European peer review for approval of pesticide active substances. The default BMR of 10% for quantal data was selected for the modelling of hepatocyte hypertrophy. Overall, for the selected hepatic effects, a lower BMDL10 of 18.6 mg TFA Na/kg bw per day was calculated for increased incidence of hepatocyte hypertrophy (male rats, study #8). The combined incidences of eye malformations (absence of aqueous/vitreous humour and multiple folding of retina) in two developmental toxicity studies in rabbits (studies #88 and #103) were considered for BMD modelling of developmental effects. In view of the severity and low incidence of the effects, a protective BMR of 0.5% was selected. Separate BMD analyses of the individual data from the two studies indicated that in neither of them litter effects substantially affected the dose–response analysis (data not shown); therefore, aggregated analysis was performed, allowing the combined modelling of the two studies using the study as covariate. From this latter analysis, a lower BMDL05 of 41.1 mg TFA Na/mg kg per day was calculated (study #88). Health‐based guidance values The existing HBGVs for TFA were set during the peer review of the pesticide risk assessment of the active substance flurtamone (EFSA, 2017). In particular, an ADI of 0.05 mg/kg bw per day (expressed as TFA Na) was derived from the 90‐day oral toxicity study in Wistar rats (# 8) based on a NOAEL of 9.9 mg/kg bw per day and applying the standard uncertainty factor of 100 plus an additional UF of 2 for the extrapolation from sub‐chronic to long‐term exposure. An ARfD was deemed unnecessary. Since that evaluation, new studies were made available. The BMDL values identified for the critical effects ranged from 8.6 to 41.1 mg TFA Na/kg bw per day. The WG agreed that, based on the new evidence, the most appropriate reference point for setting the ADI is the BMDL20 of 8.6 mg/kg bw per day, derived from changes in thyroid hormones (decreased T4 levels; adult male rats, F1) in the EOGRTS (#100). The WG also agreed to apply the standard uncertainty factor of 100 to account for inter‐ and intra‐species differences (EFSA Scientific Committee, 2012), together with an overall additional factor of 5 to address the uncertainties related to the absence of a long‐term toxicity/carcinogenicity study and the possible developmental immunotoxicity, given the lack of functional immunotoxicity testing for TFA during the developmental phase. The overall additional uncertainty factor of 5 is based on expert's judgement, following the uncertainty analysis reported in Section 4 . The resulting ADI is 0.014 mg/kg bw per day (expressed as TFA 25 ). For compounds having an effect on the hypothalamic–pituitary–thyroid axis, transitory decreased T4 levels during a certain period of pregnancy may have an impact on the development of the brain.Therefore, the effects on decreased circulating T4 levels are considered suitable for setting an ARfD. In the EORGTS (#100), BMDL20 equal to 8.6 mg/kg bw per day is set based on T4 decrease in F1 males and is considered protective of all population in the EOGRTS in the absence of thyroid hormone analysis in the pregnant dams. Based on this BMDL20, an ARfD of 0.07 mg/kg bw (expressed as TFA16) was established by applying a standard uncertainty factor of 100 to allow for inter‐ and intra‐species differences. UNCERTAINTY ANALYSIS In the context of establishing consumer HBGVs, the WG discussed the several areas of assessment with the aim to identify potential sources of uncertainties that may underestimate the hazard and impact derivation of the HBGV (see Table 18). TABLE 18Identification and prioritisation of sources of uncertainty for the derivation of the HBGV.Area of assessmentSources of uncertaintyDescriptionImpact on the derivation of the HBGVsToxicokinetics (TK)Information on absorption No guideline‐compliant ADME studies were submitted for TFA. The absence of mass‐balance ADME studies precludes calculation of absolute oral bioavailability. Based on available TK investigations in rats and rabbits, TFA is absorbed after oral exposure and distributes systemically, including to the fetus and the eye. LowTK parameters in humans No guideline‐compliant ADME studies were submitted for TFA. No direct human toxicokinetic data (such as plasma protein binding or elimination half‐life) are available, limiting interspecies extrapolation and the translation of animal data to potential human risk. LowInformation of accumulation No guideline‐compliant ADME studies were submitted for TFA. Based on its physicochemical properties (low log Kow, low bioconcentration factor and high polarity), TFA is expected to have a low potential for bioaccumulation. LowInformation on elimination No guideline‐compliant ADME studies were submitted for TFA. Based on its physicochemical properties (low log Kow, low bioconcentration factor and high polarity), TFA is expected to have a low potential for bioaccumulation. LowInformation on the extent of metabolism No guideline‐compliant ADME studies were submitted for TFA. Under physiological pH, TFA and TFA salts dissociate, resulting in systemic exposure to the TFA anion. LowDesign of the TK investigations in experimental animals.Analytical issues, including the possibility of contamination, introduce uncertainty around low‐level measurements in control animals, which may affect baseline comparisonsLowGenotoxicityInformation on in vivo genotoxicity Lack of in vivo genotoxicity studies. Considering that TFA is unlikely to be genotoxic based on the in vitro tests, there is no need to perform in vivo tests, in line with EFSA Scientific Opinion on genotoxicity testing strategies applicable to food and feed safety assessment (EFSA Scientific Committee, 2011). LowDesign of chromosomal damage studies in vitro Studies carried out with TFA Na report no induction of chromosomal damage, while inconsistency was observed among the studies performed with TFA showing a positive response in one chromosome aberration test and negative results in two in vitro micronucleus tests. Studies with TFA reported high variability in the level of cytotoxicity. No such variability was observed with TFA Na. The variability in cytotoxicity could be related to the physico‐chemical characteristics of TFA (a strong acid). Under physiological pH, TFA and TFA salts dissociate, resulting in systemic exposure to the TFA anion. Greater weight to the consistently negative results of chromosomal aberration studies conducted with the TFA salt than to those observed with the acid (see Section 3.2.2 ). LowToxicity studies in experimental animals: critical endpoints and critical study designDesign of studies in experimental animals Lack of investigation of sperm/spermatid count at the low dose and mid dose in F0 and F1. Parental/offspring NOAEL considered protective by dividing the high dose (LOAEL) by a factor of 3 to extrapolate to a NOAEL (see Section 3.2.3.3.1 ). LowDesign of studies in experimental animals Lack of investigation of thyroid toxicity (T4 levels and histopathology) in fetuses, in offspring before PND 22 and in pregnant dams. Uncertainty is covered by the standard uncertainty factor of 100 (see Appendix G). LowInformation on other effects that could be considered as critical Lack of a developmental neurotoxicity study. Parental/offspring NOAEL considered protective based on T4 decrease. LowInformation on other effects that could be considered as critical Lack of specific neurotoxicity studies. There are no indications of clinical signs of neurotoxicity in available toxicity studies (see Section 3.2.3.4.2 ). LowInformation on other effects that could be considered as critical Lack of long‐term and carcinogenicity studies. This uncertainty was taken into account by applying an additional uncertainty factor for the derivation of the ADI. MediumInformation on other effects that could be considered as critical Lack of a developmental immunotoxicity cohort or an immunotoxicity testing using functional tests. This uncertainty was taken into account by applying an additional uncertainty factor for the derivation of the ADI. Medium to HighInformation on non‐rodent species. Lack of dog studies The retrospective analysis (EFSA PPR Panel, 2026, in publishing) showed that, although dog studies were sometimes considered for HBGVs setting, genuine dog‐specific sensitivity was rarely observed, as most differences in NOAELs in rodent and dog studies were due to experimental variability, study design or scaling rather than greater sensitivity of dogs. LowObservations in humansInformation in human biomonitoring studiesEvidence for human exposure is limited but indicates that humans are internally exposed to TFA.LowInformation in human epidemiological studies Minimal information from epidemiological studies. However, it is already covered by the application of the standard uncertainty factor of 10. LowMode of actionUncertainties in the strength, consistency and specificity of the association of the key events and the critical effect.Evidence for PPARα involvement is limited but tentatively postulated as the main mode of action.LowIdentification of non‐critical effectsDefinition of adversity of some clinical chemistry findings.Rats treated with TFA Na showed changes in some clinical chemistry parameters across all repeated dose studies with unclear adversity threshold and without clear dose–response (e.g. decrease in cholesterol, decrease in glucose and bilirubin levels in both sexes).LowIdentification of target organ and critical effectsNo uncertainty identified.Dose response analysis of critical endpointsNo uncertainty identified.Abbreviations: ADI, acceptable daily intake; ADME, absorption, distribution, metabolism and excretion; HBGVs, health‐based guidance values; LOAEL, lowest observed adverse effect Level; NOAEL, no observed adverse effect level; TK, toxicokinetics. Identification and prioritisation of sources of uncertainty for the derivation of the HBGV. No guideline‐compliant ADME studies were submitted for TFA. The absence of mass‐balance ADME studies precludes calculation of absolute oral bioavailability. Based on available TK investigations in rats and rabbits, TFA is absorbed after oral exposure and distributes systemically, including to the fetus and the eye. No guideline‐compliant ADME studies were submitted for TFA. No direct human toxicokinetic data (such as plasma protein binding or elimination half‐life) are available, limiting interspecies extrapolation and the translation of animal data to potential human risk. No guideline‐compliant ADME studies were submitted for TFA. Based on its physicochemical properties (low log Kow, low bioconcentration factor and high polarity), TFA is expected to have a low potential for bioaccumulation. No guideline‐compliant ADME studies were submitted for TFA. Based on its physicochemical properties (low log Kow, low bioconcentration factor and high polarity), TFA is expected to have a low potential for bioaccumulation. No guideline‐compliant ADME studies were submitted for TFA. Under physiological pH, TFA and TFA salts dissociate, resulting in systemic exposure to the TFA anion. Lack of in vivo genotoxicity studies. Considering that TFA is unlikely to be genotoxic based on the in vitro tests, there is no need to perform in vivo tests, in line with EFSA Scientific Opinion on genotoxicity testing strategies applicable to food and feed safety assessment (EFSA Scientific Committee, 2011). Studies carried out with TFA Na report no induction of chromosomal damage, while inconsistency was observed among the studies performed with TFA showing a positive response in one chromosome aberration test and negative results in two in vitro micronucleus tests. Studies with TFA reported high variability in the level of cytotoxicity. No such variability was observed with TFA Na. The variability in cytotoxicity could be related to the physico‐chemical characteristics of TFA (a strong acid). Under physiological pH, TFA and TFA salts dissociate, resulting in systemic exposure to the TFA anion. Greater weight to the consistently negative results of chromosomal aberration studies conducted with the TFA salt than to those observed with the acid (see Section 3.2.2 ). Lack of investigation of sperm/spermatid count at the low dose and mid dose in F0 and F1. Parental/offspring NOAEL considered protective by dividing the high dose (LOAEL) by a factor of 3 to extrapolate to a NOAEL (see Section 3.2.3.3.1 ). Lack of investigation of thyroid toxicity (T4 levels and histopathology) in fetuses, in offspring before PND 22 and in pregnant dams. Uncertainty is covered by the standard uncertainty factor of 100 (see Appendix G). Lack of a developmental neurotoxicity study. Parental/offspring NOAEL considered protective based on T4 decrease. Lack of specific neurotoxicity studies. There are no indications of clinical signs of neurotoxicity in available toxicity studies (see Section 3.2.3.4.2 ). Lack of long‐term and carcinogenicity studies. This uncertainty was taken into account by applying an additional uncertainty factor for the derivation of the ADI. Lack of a developmental immunotoxicity cohort or an immunotoxicity testing using functional tests. This uncertainty was taken into account by applying an additional uncertainty factor for the derivation of the ADI. Lack of dog studies The retrospective analysis (EFSA PPR Panel, 2026, in publishing) showed that, although dog studies were sometimes considered for HBGVs setting, genuine dog‐specific sensitivity was rarely observed, as most differences in NOAELs in rodent and dog studies were due to experimental variability, study design or scaling rather than greater sensitivity of dogs. Minimal information from epidemiological studies. However, it is already covered by the application of the standard uncertainty factor of 10. Abbreviations: ADI, acceptable daily intake; ADME, absorption, distribution, metabolism and excretion; HBGVs, health‐based guidance values; LOAEL, lowest observed adverse effect Level; NOAEL, no observed adverse effect level; TK, toxicokinetics. Ordinal scales describe uncertainty using ordered categories like high, medium or low, so the analysis is qualitative for priority ranking (EFSA Scientific Committee, 2018). The WG considered that all uncertainties classified as of low impact would not underestimate the hazard and would not change the setting of HBGV (to a lower value) if they were resolved. 26 Based on the uncertainty analysis done (see Table 18), two main areas of uncertainty were identified: With respect to the lack of a long term (2‐year)/carcinogenicity study on TFA, the WG considered that: In both the short‐term (90‐day) and chronic (52‐week) toxicity studies in rats, there was no histopathological evidence of proliferative lesions, such as hyperplasia or other preneoplastic findings;TFA is unlikely to be genotoxic based on available data;A peroxisome proliferator mode of action has been postulated; however, based on current scientific understanding, the human relevance to humans of this mode of action in relation to liver tumour formation is considered low; however, a robust mode of action study on TFA is not available.Liver hypertrophy was reported in repeated dose toxicity studies (#15, #8, #124); it occurred at dose levels above 85 mg/kg bw per day.In the chronic (52‐week) rat study (#109), the maximum tolerated dose (MTD) was not achieved as the highest dose tested (37.8 mg/kg bw per day) was not properly selected. As a result, this study may have limited sensitivity to detect preneoplastic findings and does not fulfil the requirements of a full carcinogenicity assessment (see Section 3.2.3.2 ); With respect to the lack of a long term (2‐year)/carcinogenicity study on TFA, the WG considered that: In both the short‐term (90‐day) and chronic (52‐week) toxicity studies in rats, there was no histopathological evidence of proliferative lesions, such as hyperplasia or other preneoplastic findings;TFA is unlikely to be genotoxic based on available data;A peroxisome proliferator mode of action has been postulated; however, based on current scientific understanding, the human relevance to humans of this mode of action in relation to liver tumour formation is considered low; however, a robust mode of action study on TFA is not available.Liver hypertrophy was reported in repeated dose toxicity studies (#15, #8, #124); it occurred at dose levels above 85 mg/kg bw per day.In the chronic (52‐week) rat study (#109), the maximum tolerated dose (MTD) was not achieved as the highest dose tested (37.8 mg/kg bw per day) was not properly selected. As a result, this study may have limited sensitivity to detect preneoplastic findings and does not fulfil the requirements of a full carcinogenicity assessment (see Section 3.2.3.2 ); In both the short‐term (90‐day) and chronic (52‐week) toxicity studies in rats, there was no histopathological evidence of proliferative lesions, such as hyperplasia or other preneoplastic findings; In both the short‐term (90‐day) and chronic (52‐week) toxicity studies in rats, there was no histopathological evidence of proliferative lesions, such as hyperplasia or other preneoplastic findings; TFA is unlikely to be genotoxic based on available data; TFA is unlikely to be genotoxic based on available data; A peroxisome proliferator mode of action has been postulated; however, based on current scientific understanding, the human relevance to humans of this mode of action in relation to liver tumour formation is considered low; however, a robust mode of action study on TFA is not available. A peroxisome proliferator mode of action has been postulated; however, based on current scientific understanding, the human relevance to humans of this mode of action in relation to liver tumour formation is considered low; however, a robust mode of action study on TFA is not available. Liver hypertrophy was reported in repeated dose toxicity studies (#15, #8, #124); it occurred at dose levels above 85 mg/kg bw per day. Liver hypertrophy was reported in repeated dose toxicity studies (#15, #8, #124); it occurred at dose levels above 85 mg/kg bw per day. In the chronic (52‐week) rat study (#109), the maximum tolerated dose (MTD) was not achieved as the highest dose tested (37.8 mg/kg bw per day) was not properly selected. As a result, this study may have limited sensitivity to detect preneoplastic findings and does not fulfil the requirements of a full carcinogenicity assessment (see Section 3.2.3.2 ); In the chronic (52‐week) rat study (#109), the maximum tolerated dose (MTD) was not achieved as the highest dose tested (37.8 mg/kg bw per day) was not properly selected. As a result, this study may have limited sensitivity to detect preneoplastic findings and does not fulfil the requirements of a full carcinogenicity assessment (see Section 3.2.3.2 ); Overall, the WG concluded that an uncertainty remains due to the lack of a long‐term (2‐year)/carcinogenicity study and the limitations of the available chronic study. The WG categorised this uncertainty as medium. 2With respect to immunophenotyping findings, the WG considered that: A dose‐related decrease in the total number of splenic immune cells was observed in both sexes at all dose levels in the EOGRTS in rats (F1, #100). For each individual cell type, the dose–response relationship was more evident in males than in females;In the absence of functional immunotoxicity testing, these findings introduced uncertainty regarding possible developmental immunotoxicity;Immunophenotyping results were not subjected to statistical analysis in the respective studies, limiting the interpretation of the findings;A developmental immunotoxicity cohort or a functional test is not available for TFA, which could have elucidated potential immunotoxic consequences of the observed cellular changes;Within the available in vivo toxicity data package on TFA or TFA Na, no consistent or dose‐dependent effects on immunological endpoints were observed across studies in adult animals.TFA suppressed antibody production (IgG and IgM) in human peripheral blood mononuclear cells (PBMCs) following 7‐day exposure (Iulini et al., 2025). With respect to immunophenotyping findings, the WG considered that: A dose‐related decrease in the total number of splenic immune cells was observed in both sexes at all dose levels in the EOGRTS in rats (F1, #100). For each individual cell type, the dose–response relationship was more evident in males than in females;In the absence of functional immunotoxicity testing, these findings introduced uncertainty regarding possible developmental immunotoxicity;Immunophenotyping results were not subjected to statistical analysis in the respective studies, limiting the interpretation of the findings;A developmental immunotoxicity cohort or a functional test is not available for TFA, which could have elucidated potential immunotoxic consequences of the observed cellular changes;Within the available in vivo toxicity data package on TFA or TFA Na, no consistent or dose‐dependent effects on immunological endpoints were observed across studies in adult animals.TFA suppressed antibody production (IgG and IgM) in human peripheral blood mononuclear cells (PBMCs) following 7‐day exposure (Iulini et al., 2025). A dose‐related decrease in the total number of splenic immune cells was observed in both sexes at all dose levels in the EOGRTS in rats (F1, #100). For each individual cell type, the dose–response relationship was more evident in males than in females; A dose‐related decrease in the total number of splenic immune cells was observed in both sexes at all dose levels in the EOGRTS in rats (F1, #100). For each individual cell type, the dose–response relationship was more evident in males than in females; In the absence of functional immunotoxicity testing, these findings introduced uncertainty regarding possible developmental immunotoxicity; In the absence of functional immunotoxicity testing, these findings introduced uncertainty regarding possible developmental immunotoxicity; Immunophenotyping results were not subjected to statistical analysis in the respective studies, limiting the interpretation of the findings; Immunophenotyping results were not subjected to statistical analysis in the respective studies, limiting the interpretation of the findings; A developmental immunotoxicity cohort or a functional test is not available for TFA, which could have elucidated potential immunotoxic consequences of the observed cellular changes; A developmental immunotoxicity cohort or a functional test is not available for TFA, which could have elucidated potential immunotoxic consequences of the observed cellular changes; Within the available in vivo toxicity data package on TFA or TFA Na, no consistent or dose‐dependent effects on immunological endpoints were observed across studies in adult animals. Within the available in vivo toxicity data package on TFA or TFA Na, no consistent or dose‐dependent effects on immunological endpoints were observed across studies in adult animals. TFA suppressed antibody production (IgG and IgM) in human peripheral blood mononuclear cells (PBMCs) following 7‐day exposure (Iulini et al., 2025). TFA suppressed antibody production (IgG and IgM) in human peripheral blood mononuclear cells (PBMCs) following 7‐day exposure (Iulini et al., 2025). Overall, the WG concluded that an uncertainty remains due to the lack of functional immunotoxicity testing for TFA following pre‐ and post‐natal developmental exposure to follow up immunophenotyping findings and the results from the in vitro mechanistic study. The WG categorised this uncertainty as medium to high. These uncertainties have been addressed by applying an additional overall uncertainty factor for the derivation of the ADI (see Section 3.2.7.3 ). CONCLUSIONS Based on a WoE of the available and reliable evidence, the WG concluded that TFA and its salt TFA Na are unlikely to be genotoxic. The WG agreed to set an ADI of 0.014 mg/kg bw per day and an ARfD of 0.07 mg/kg bw (both expressed as TFA). ABBREVIATIONS 4‐Nitroquinoline 1‐oxide Albumin Globulin ratio Acceptable Daily Intake Absorption, Distribution, Metabolism and Excretion Alkaline Phosphatase Alanine Transaminase Activated Partial Thromboplastin Time Acute Reference Dose Aspartate Transaminase Acute Toxicity Estimate Area Under the Curve Benzo(a)pyrene Beat Cross Frequency Benchmark Mose Modelling Benchmark Dose Lower Confidence Limit Body Mass Index Benchmark Response 7‐benzyloxyresorufin O‐dealkylase Body Weight Body Weight Gain Chromosomal Aberration Constitutive Androstane Receptor Chemical Abstract Service Cytokinesis‐Block Proliferation Index Chinese Hamster Lung cells Chinese Hamster Ovary Harmonised Classification and Labelling. Concentration Maximum Coenzyme‐A EFSA Panel on Contaminants in the Food Chain. Cyclophosphamide Comparative Thyroid Assessment Coefficient of Variation Cytochrome Iodothyronine Deiodinase Dimethyl sulfoxide Developmental Neurotoxicity European Chemicals Agency Endocrine Disruptor Ethyl methanesulfonate Extended One‐Generation Reproductive Toxicity Study US Environmental Protection Agency Ethoxyresorufin O‐deethylase Federal Insecticide, Fungicide and Rodenticide Act Gestation Day Global Evaluation Factor γ‐glutamyltransferase Gap Junctional Intercellular Communication Glutamate dehydrogenase Good Laboratory Practice Health‐based guidance values Historical Control Data Hydrochlorofluorocarbons Human Chorionic Gonadotropin Hydrofluorocarbons human‐induced Pluripotent Stem Cells High Performance Liquid Chromatography – Mass Spectrometry human hypoxanthine‐guanine phosphoribosyltransferase Hypothalamic – Pituitary – Thyroid Induced Mutant Frequency International Chemical Identifier International Union of Pure and Applied Chemistry Liquid Chromatography ‐ High Resolution Mass Spectrometry Lactation Day Median lethal dose Lowest Observed Adverse Effect Level Limit of Detection Limit of Quantification Mean Corpuscular Haemoglobin Mean Cell Haemoglobin Concentration Human Monocarboxylate Transporter 1 Mean Corpuscular Volume Maximum Detection Limit Mitotic Index Mitotic Inhibition Mitomycin C Methyl methanesulfonate Methynitronitrosoguanidine Mode of Action Margin of Exposure Mean Platelet Volume Maximum Tolerable Dose Not applicable Non‐esterified Fatty Acids Natural killer No Observed Adverse Effect Level National Toxicology Program Organisation for Economic Co‐operation and Development Office of Health Assessment and Translation human Peripheral Blood Mononuclear Cells Per‐ and polyfluoroalkyl substances Perfluoroalkyl Carboxylic Acids Perfluorohexane sulfonic acid Perfluorononanoic acid Perfluorooctanoic acid Perfluorooctane sulfonic acid Perfluoroalkyl Sulfonic Acids Persistent, Mobile and Toxic Postnatal Day Peroxisome Proliferator‐activated Receptors Plant Protection Product Pentoxy‐resorufin O‐deethylase Prothrombin Time. Propylthiouracil Quality Assurance and Quality Control Risk Assessment Committee Red Blood Cells Registration, Evaluation, Authorisation and restriction of Chemicals Replication Index Relative increase in cell count Risk of Bias Relative Population Doubling Relative Survival Relative Suspension Growth Relative Total Growth Sertoli cell‐only cultures Standard Deviation Sertoli‐germ cell co‐cultures Simplified Molecular Input Line Entry System T‐cell Dependent Antibody Response Sodium Trifluoroacetate Trifluoroacetic Acid Potassium Trifluoroacetate Potassium Trifluoromethanesulfinate Test Guideline Toxicogenomics Project–Genomics Assisted Toxicity Evaluation System Toxicokinetic Time to maximum concentration Terms of Reference Thyroid Peroxidase Uncertainty Factor Upper Limit of Normal Ultra‐high Performance Liquid Chromatography ‐ Mass Spectrometry Average Path Velocity Curvilinear Velocity very persistent and very mobile White Blood Cells Working Group Weight of Evidence REQUESTOR European Commission QUESTION NUMBER EFSA‐Q‐2024‐00502 COPYRIGHT FOR NON‐EFSA CONTENT EFSA may include images or other content for which it does not hold copyright. In such cases, EFSA indicates the copyright holder and users should seek permission to reproduce the content from the original source. MAP DISCLAIMER The designations employed and the presentation of material on any maps included in this scientific output do not imply the expression of any opinion whatsoever on the part of the European Food Safety Authority concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Supporting information Annex 1: Protocol of the toxicological assessment. Annex 2: List of studies collected in the framework of the targeted call for data. Annex 3: Outcome of the Risk of Bias for non‐guideline in vitro and ex vivo studies in the body of evidence. Annex 4: Outcome of the Risk of Bias for non‐guideline in vivo studies in the body of evidence. Annex 5: Outcome of the assessment for human studies. Annex 6: Data extraction of non‐guideline in vivo animal oral toxicity studies. Annex 7: Data extraction of non‐guideline in vitro toxicity studies. Annex 8: Outcome of the public consultation. Annex 9: Consolidated list of studies outside the window of call for data. Annex 10: Assessment of immunotoxicity endpoints in available toxicity studies. Annex 11: Benchmark Dose (BMD) Modelling. ACKNOWLEDGEMENTS EFSA wishes to thank the following for the support provided to this scientific output: Wieneke Bil, Bas Bokkers (RIVM); Alexander Eckhardt, Helena Banning (UBA); Angeliki Lysimachou, Peter Clausing, Helmut Burtscher, Pauline Cervan (NGO representatives); Maura Karina Ferreira Emiliano, Anja Hueser, Antoinette Degroot, Lisa Bertomeu, Rich Bartlett, Stephanie Melching‐Kollmuss (TFA Task force); Alexandra Fischer, Verena Haudek‐Prinz, Lorenz Karl, Therese Weil (AT), Dirk Marien, Liesbeth De Smet, Philippe Castelain, Kristel Brys (BE), Pernille Jacobsen, Peter Hammer Sorensen (DK), Tanja Heise, Franziska Kupprat (DE), Dimitra Nikolopoulou, Vasiliki Hatzi (EL), Manuel Sanz Bernal, Clara Aisa Sanchez, Veronica Campos Avedillo (ES), Francesca Metruccio (IT), Jessica Broeders (NL), Tanja Fatur (SI), Daniel Borg, Anneli Widenfalk, Anna Mentor (SE), Marianne Balmer, Christoph Geiser (CH) (Member States representatives). Assessment of relevance and reliability of genotoxicity studies Test systemExposure conditionsInformation on the characteristics of the test substanceResultReliability/commentsRelevance of test system/of the study resultReference Reverse mutation assay Salmonella typhimurium TA98, TA100, TA1535, TA1537 and TA102 OECD TG 471 (1997): in compliance with version 21 July 2020 GLP: Yes With and without 10% S9 mix (Aroclor 1254‐induced rat liver) 1st experiment: +/− S9: plate incorporation method, 0, 1.6, 8, 40, 200, 1000 and 5000 μg/plate. 2nd experiment: –S9: plate incorporation method, 0, 156.25, 312.5, 625, 1250, 2500 and 5000 μg/plate. +S9: preincubation test, 0, 156.25, 312.5, 625, 1250, 2500 and 5000 μg/plate. Negative control: water (solvent) Appropriate positive controls Triplicate plates TFA Na 99.1% pure Negative No evidence of toxicity observed Preliminary cytotoxicity test with TA100: No evidence of toxicity observed. 1 Historical negative and positive control data provided High/high #2 Reverse mutation assay Salmonella typhimurium TA98 and TA100 OECD TG 471: no GLP: no Plate incorporation assay 1.5 mg/plate With and without S9 mix (Aroclor induced rat liver) Negative control: Not reported Positive controls: no TFA Purity: not reported Negative No revertants above the control < 0.000007 revertants/nmole 4 Test not performed according to OECD TG 471 Very short description of the protocol and results Spontaneous mutant colonies and mutant colonies induced by positive controls are missing. Only one concentration tested Only one experiment High/low #71 Reverse mutation assay Salmonella typhimurium TA97a, TA98, TA100, TA1535, and TA102 OECD TG 471 (1997): in compliance with version 26 June 2020 GLP: Yes With and without 10% S9 mix (phenobarbital and β‐naphthoflavone induced rat liver) Plate incorporation assay 1st experiment: 0, 0.5, 1.5, 5, 15, 50, 150, 500, 1500, 5000 μg/plate. 2nd experiment: 0, 78.125, 156. 25, 312.5, 625, 1250, 2500 and 5000 μg/plate. Negative control: water (solvent) Appropriate positive controls Triplicate plates TFA Na 98% pure Negative No evidence of toxicity observed 1 Historical negative and positive control data provided High/high #123 Reverse mutation assay Salmonella typhimurium TA98, TA100, TA1535, TA1537 and TA102 OECD TG 471 (1997): in compliance with version 26 June 2020 GLP: Yes With and without 5% S9 mix (Aroclor 1254‐induced rat liver) 1st experiment: plate incorporation assay 0, 3.16, 10.0, 31.6, 100, 316 or 1000 μg/plate. 2nd experiment: pre‐incubation test 0, 3.16, 10.0, 31.6, 100, 316 or 1000 μg/plate. Negative control: water (solvent) Appropriate positive controls Triplicate plates Two preliminary cytotoxicity tests (plate incorporation test without and with metabolic activation) in TA100. Ten concentrations from 0.316 to 5000 μg/plate were tested. TFA 99.0% pure Negative Main test: cytotoxicity observed at 1000 μg/plate. Preliminary test: Pronounced cytotoxicity (scarce background lawn and reduction of the number of revertants by more than 50%) noted at ≥ 1000 μg/plate. 1 Historical negative and positive control data provided High/high #140 Reverse mutation assay Salmonella typhimurium TA98, TA100 and TA1535 OECD TG 471: no GLP: no Preincubation assay With and without S9 mix (Aroclor 1254‐induced rat liver or testes or combined) 0, 3.6, 8.9, 17.0 and 35.7 mg/mL Negative control: vehicle Positive controls: B(a)P and 4NQO Duplicate plates TFA Na 98% pure Negative 35.7 mg/mL: maximum non‐toxic concentration (as illustrated by revertant ratios less than 1.0 in some instances). 3 Only 3 strains tested Values reported are ratio of the number of His revertant colonies in presence of test compound to the number of His revertant colonies in the absence of test compound No information on cytotoxicity High/low #27 Reverse mutation assay Salmonella typhimurium T98 and TA100 OECD TG 471: no GLP: no Spot test Sterile 6 mm diameter paper disc placed in the centre of each plate 0.1 mL TFA Positive control: MNNG TFA Purity: not reported Negative According to the authors 4 Test not performed according to OECD TG 471 Very short description of the protocol Results not shown Low/low #22 Gene mutation assay locus TK+/− Mouse lymphoma cells (L5178Y) OECD TG 490 (2016) GLP: yes MicrotitreR Fluctuation Technique With and without 5% S9 mix (Aroclor 1254‐induced rat liver) ‐Without S9: 1st experiment + repeat: 3‐h exposure, 0, 360, 560, 760, 960, 1160 and 1360 μg/mL 2nd experiment: 24‐h exposure, 0, 360 to 1360 (10 mM) μg/mL ‐With S9: 3‐h exposure (two experiments), 0, 360 to 1360 μg/mL Duplicate cultures Negative control: water (solvent) Positive controls: ‐S9: 4‐NQO, +S9: B(a)P Preliminary cytotoxicity tests: 3‐h exposure +/− S9, 42.5 to 1360 μg/mL 24‐h exposure ‐S9, 5.313 to 1360 μg/mL TFA Na 99.1%–99.3% pure Negative Cytotoxicity: 1st experiment: 3h, the highest dose tested yielded 112% and 93% RTG in the absence and presence of S9 respectively. 2nd experiment: the highest dose tested yielded 76% and 104% relative total growth (RTG) in the absence and presence of S9, respectively. Preliminary cytotoxicity tests: 3‐h exposure: The highest dose tested yielded 112% Relative total growth (RTG) in the absence of S9 and 99% RTG in the presence of S9; 24‐h exposure: The highest dose tested yielded 49% RTG. 1 No calculation of Induced Mutation Frequency (IMF) and no Global Evaluation Factor (GEF) evaluation Historical negative and positive control data provided High/high #3 Gene mutation assay locus TK+/− Mouse Lymphoma cells (L5178Y) OECD TG 490 (2016) GLP: yes With and without 10% S9 mix (phenobarbital and β‐naphthoflavone induced rat liver) Exposure 3 h (3.5 h) 0, 8.192, 20.48, 51.2, 128, 320, 800, 2000 and 5000 μg/mL Negative control: medium Positives controls: Methyl methanesulfonate (MMS), Cyclophosphamide (CP) Preliminary cytotoxicity test, +/− S9, 312.5, 625, 1250, 2500 and 5000 μg/mL TFA Na 98% pure Negative IMF(s): < GEF (126 × 10−6) Cytotoxicity (RTG): –S9: 94%, 80%, 81%, 81%, 72%, 78%, 75%, 60% +S9: 95%, 84%, 82%, 84%, 77%, 80%, 60%, 44% Preliminary cytotoxicity test: Cell count: conc.‐related toxicity RSG: > 20% ± S9 at all doses 1 Historical negative and positive control data provided High/high #126 Gene mutation assay locus TK+/− Mouse Lymphoma cells (L5178Y) OECD TG 490 (2015) GLP: yes With and without 5% S9 mix (phenobarbital and β‐naphthoflavone induced rat liver) Preliminary cytotoxicity test and 1st. mutagenicity test, –S9 4 h and 24 h + S9: 4 h 10.00; 3.33; 1.11; 0.37; 0.12; 0.04; 0.01; 0.00 mM = 2.00; 0.67; 0.22; 0.07; 0.02; 0.01; 0.003; 0.00 μL/mL Definitive test: Without S9: – 4‐h exposure: 0.75; 0.38; 0.19; 0.09; 0.00 μL/mL and –24‐h exposure: 1.00; 0.50; 0.25; 0.13; 0.00 μL/mL With S9: 4‐h exposure, 1.00; 0.50; 0.25; 0.13; 0.00 μL/mL Negative control: water Positives controls: MMS, CP TFA > 99.9% pure Negative In the absence and presence of metabolic activation induced mutation frequency level (IMF (s)) did not exceed 126 × 10−6 (GEF) for most of the concentrations tested (only + S9 4 h 1 μL/mL: MF: 158.06 × 10−6). Cytotoxicity: +S9, 4 h: High cytotoxicity at 2 μL/mL Preliminary cytotoxicity: ‐S9, 4 h: RTG 42–153% 24 h: RTG 0–148% +S9, 4 h: RTG 0%–125% 1 High/high #134 Gene mutation Assay locus TK± Mouse Lymphoma cells (L5178Y) OECD TG 490 (2015) GLP: yes With and without S9 mix (Aroclor 1254 induced rat liver) Two experiments Exposure w/o S9: 3 and 24 h Exposure w S9: 3 h 0, 125, 250, 500, 1000 and 2000 μg/mL Negative control: water Positives controls: MMS, 3‐Methylcholanthrene Preliminary cytotoxicity test, ±S9, 0, 10.0, 31.6, 100, 316, 1000 and 2000 μg/mL TFA 99.0% pure Negative No change in the ratio of small to large mutant colonies indicating no clastogenic potential Main test cytotoxicity: decreased relative total growth at 1000 and 2000 μg/mL The pronounced to complete cytotoxic concentration of 2000 μg/mL was excluded from evaluation in all experiments, as well as the cytotoxic concentration of 1000 μg/mL in the absence of metabolic activation with a 24‐h exposure (RTG = 13%). Preliminary cytotoxicity without and with S9 mix (3 and 24 h): decreased survival at 1000 and 2000 μg/mL 1 Historical negative and positive control data provided High/high #139 Gene mutation assay locus HPRT Mouse lymphoma cells (L5178Y) OECD TG 476 (2016): Not reported GLP: yes MicrotitreR fluctuation Technique With and without 2% S9 mix (Aroclor 1254‐induced rat liver) Exposure: 3 h 1st experiment: 0, 200, 400, 600, 800, 1000, 1200 and 1360 μg/mL 2nd experiment: 0, 150, 300, 500, 700, 900, 1100 and 1360 μg/mL Negative control: water Positive controls: 4NQO, B(a)P Preliminary cytotoxicity test: 0, 42.5 to 1360 μg/mL (equivalent to 10 mM) TFA Na 99.9% pure Negative Cytotoxicity: 1st experiment: highest concentration tested, 1360 μg/mL, gave 96% and 111% relative survival (RS) in the absence and presence of S9, respectively. 2nd experiment: highest concentration tested, 1360 μg/mL, gave 80% and 112% relative survival (RS) in the absence and presence of S9, respectively. Preliminary cytotoxicity: 96% and 111% RS in the absence and presence of S9, respectively 1 Historical negative and positive control data provided High/high ##163 Chromosomal aberration assay Chinese hamster lung cells (CHL) OECD TG 473 (2016): not fully compliant GLP: yes With and without 10% S9 mix (phenobarbital and β‐naphthoflavone induced rat liver) Exposure: 3h and 21 h recovery period 0, 128, 320, 800, 2000 and 5000 μg/mL 2 experiments Negative control: medium Positive controls: Ethyl methanesulfonate (EMS) and CP 200 metaphases scored/conc. Single culture Preliminary cytotoxicity test: 0, 312.5, 625, 1250, 2500 and 5000 μg/mL TFA Na 98% pure Negative 1st test. Acceptability criteria were not met as the % of cells with CA (w/o gaps) of the negative control fell outside the 95% confidence interval of the historical negative control distribution in this lab (w/o S9: 0.1–1.1; w S9: −0.2 to 1.3) (validity criteria not met). % of cells with CA (w/o gaps) Control 3h expo w/o S9: 2.0; w S9: 1.5. The study's authors decided to repeat the assay. 2nd test: negative Cytotoxicity: relative increase in cell count (RICC): 1st experiment: –S9: 92%, 97%, 92%, 95% and 79%; +S9: 107%, 92%, 98%, 103% and 82%. 2nd experiment: –S9: 98%, 96%, 94% and 87%, 82%; +S9: 99%, 104%, 96%, 93% and 80%. Preliminary cytotoxicity: Relative Increase in Cell Count (RICC) > 79% at all concentrations ±S9 mix 2 No extended treatment Only 200 metaphases scored/conc. instead of 300. Only 1 test acceptable (2nd test) High/limited #125 Chromosomal aberration assay Human peripheral lymphocytes OECD TG 473: not reported, in compliance with TG 473 (2016) GLP: yes With and without 5% S9 mix (Aroclor 1254‐ induced rat liver) 1st experiment: w/o S9 4‐h exposure, 0, 125, 250, 500, 1000 and 2000 μg/mL 21‐h recovery period 2nd experiment: w/o S9 24‐h exposure, 0, 15.63, 31.3, 62.5, 125 and 250 μg/mL 3rd experiment: w S9 4‐h exposure, 0, 125, 250, 500, 1000 and 2000 μg/mL 21‐h recovery period 4th experiment: w S9 4‐h exposure, 0, 500, 750, 1000, 1000*, 2000* μg/mL 21‐h recovery period 5th experiment: w/o S9 24‐h exposure, 0, 125, 175 and 250 μg/mL Negative control: water Positive controls: MMC, CP 300 metaphases scored/conc. Duplicate cultures Preliminary cytotoxicity: *Neutralised with 1 N NaOH TFA 99.0% pure Positive Increase in the % CA noted at 125 μg/mL (1 of 2 experiments) or at 250 μg/mL w/o S9 after 24‐h exposure and at 1000 μg/mL in both experiments with S9 after 4‐h exposure. No increase in % CA w/o S9 after 4‐h exposure. Positive results are also observed at 1000 and 2000 μg/mL in the absence of cytotoxicity when TFA was neutralised by NaOH (not in the buffer). Cytotoxicity: Complete cytotoxicity at 2000 μg/mL 4‐h exposure ±S9 Cytotoxicity at 1000 μg/mL, 4‐h exposure ‐S9 58% (MI: 0.42) and + S9 49% and 45% (MI: 0.51 and 0.55). Cytotoxicity at 250 μg/mL of 57% and 44% (MI: 0.43 or 0.56) after 24‐h exposure –S9. Decrease in pH compared to the control (pH: 7.53) at 1000 μg/mL (pH: 6.03) and 2000 μg/mL (pH: 2.69). The same concentrations of 1000 or 2000 μg/mL medium adjusted by buffering with 1 N NaOH to pH 7.25 or 7.02, respectively, did not lead to any cytotoxicity. Preliminary cytotoxicity: –S9 24‐h exposure: pronounced to complete cytotoxicity starting at 316 μg/mL +S9 4‐h exposure: complete cytotoxicity at 2000 μg/mL 1 High/high #141 Chromosomal aberration assay Human peripheral lymphocytes (pooled blood from 3 female donors) OECD TG 473: not reported, but nearly in compliance with version 2016 GLP: yes With and without 2% S9 mix (Aroclor 1254‐ induced rat liver) 1st experiment: 3‐h exposure ±S9 and harvesting 17 h later 0, 1000, 1200 and 1360 μg/mL (= 10 mM). 2nd experiment: –S9, 20‐h exposure, harvesting at end of treatment, 0, 750, 1200 and 1360 μg/mL +S9, 3‐h exposure and harvesting 17 h later, 0, 1050, 1200 and 1260 μg/mL Negative control: water Positive controls: 4NQO, CP 200 metaphases scored/conc. Preliminary cytotoxicity: 0, 4.934–1360 μg/mL ± S9 Duplicate cultures TFA Na 99.9% pure Negative Only slight cytotoxicity (as measured by Mitotic inhibition MIH) Cytotoxicity: 1st experiment: –S9: no cytotoxicity 1st experiment: +S9: mitotic inhibition of 2% at 1360 μg/mL 2nd experiment: –S9: MIH of 13%, 35% and 35%, 2nd experiment: +S9: MIH of 5%, 13% and 22% 2 Only 200 metaphases scored/conc. instead of 300. High/limited #164 Chromosomal aberration assay Human peripheral lymphocytes (pooled blood from 3 male donors) OECD TG 473 (1997): nearly in compliance with version 2016 GLP: yes With and without 2% S9 mix (Aroclor 1254‐ induced rat liver) 0, 85, 170, 340, 680 and 1360 μg/mL (= 10 mM) 1st experiment: 3‐h exposure ±S9 and harvesting 17 h later 2nd experiment: –S9, 20‐h exposure, harvesting at end of treatment +S9, 3‐h exposure and harvesting 17 h later Negative control: water Positive controls: 4NQO, CP 200 metaphases scored/conc. Preliminary cytotoxicity: 0, 0.006641 to 1360 μg/mL ± S9 Duplicate cultures TFA Na 99.3% pure Negative % of CA mostly similar to the concurrent vehicle controls in the presence and absence of S9. One exception after a 20‐h exposure in the absence of S9 in exp. 2: 2% structural aberrations excluding gaps vs. 0% in controls at 1360 μg/mL (% of cells with aberrations within the historical vehicle control frequencies, 61% mitotic inhibition). Not considered biologically relevant. No increases in the frequency of cells with numerical aberrations either in the absence or presence of S9, considered to be of any biological relevance. Cytotoxicity: 1st experiment: 1360 μg/mL, induced approximately 16% and 0% mitotic inhibition (reduction in mitotic index) in the absence and presence of S9 respectively (4‐h exposure). 2nd experiment: 1360 μg/mL, induced approximately 61% and 19% mitotic inhibition in the absence (24‐h exposure) and presence of S9 (4‐h exposure) respectively. 2 Only 200 metaphases scored/conc. instead of 300. High/limited #4 Micronucleus test TK6 Human Lymphoblastoid Cells Without Cyt B OECD TG 487 (2016) GLP: yes With and without 5% S9 mix (phenobarbital and 5,6‐benzoflavone induced rat liver) 1st experiment: ±S9, 3‐h exposure, 0, 285, 570 and 1140 μg/mL 2nd experiment: ‐S9, 24‐h exposure, 0, 285, 570 and 1140 μg/mL Negative control: water Positive controls: MMC, CP Analysis: 2000 mononuclear cells/conc. Concentration range finding: 0, 17.8, 35.6, 71.3, 143, 285, 570 and 1140 μg/mL (=10 mM) TFA 99.08% pure Negative No statistically significant increases in % of micro nucleated mononuclear cells. Cytotoxicity: measurement of Relative Population Doubling (RPD). 1st experiment: 3‐h exposure: ‐S9: RPD of 88.4.8% at highest conc.; +S9: RPD of 97.1% at highest concentration 2nd experiment: 24‐h exposure: ‐S9: RPD of 77.0% at highest concentration Range finding: absence of cytotoxicity after 3‐h exposure ±S9. After 24‐h exposure ‐S9: RPD of 71.8% at highest concentration 1 Historical negative and positive control data not provided High/high #5 Micronucleus test CHO‐K1 Cells With Cyt B OECD TG 487 (2016) GLP: yes With and without 1% S9 mix (phenobarbital and β‐Naphtoflavone induced rat liver) 1st experiment: +/‐S9, 3‐h exposure +27‐h recovery, 0, 372, 745, 1117, 1490 μg/mL 2nd experiment: –S9, 27‐h exposure, 0, 372, 745, 1117, 1490 μg/mL 1 μL/mL = 1490 μg/mL Negative control: water Positive controls: MMC, CP CBPI (Cytokinesis‐Block Proliferation Index) and RI (Replication Index) was determined to assess cell proliferation (cytotoxicity) using at least 500 cells per culture. Analysis: 2000 binucleated cells/conc. Preliminary cytotoxicity test: 0.0009–2 μL/mL. The highest tested concentration (2 μL/mL) had cytotoxic effect against CHO‐K1 cells (100% cytotoxicity) in all incubation conditions. TFA > 99.9% pure Negative Cytotoxicity: –S9 27‐h exposure: 40% at 1117 μg/mL and 100% at 1490 μg/mL (no scoring of MN). –S9 3‐h exposure: concentrations are not cytotoxic. +S9 3‐h exposure: concentrations are not cytotoxic. 1 High/high #133Abbreviations: 4NQO, 4‐Nitroquinoline 1‐oxide; CBPI, Cytokinesis‐Block Proliferation Index; CHL, Chinese hamster lung cells; CP, Cyclophosphamide; EMS, Ethyl methanesulfonate; IMF, induced mutant frequency; MMS, Methyl methanesulfonate; MNNG, N‐methyl‐N′‐nitro‐N‐nitrosoguanidine; RI, Replication Index; RICC, Relative Increase in Cell Count; RPD, Relative Population Doubling; RTG, relative total growth. Reverse mutation assay Salmonella typhimurium TA98, TA100, TA1535, TA1537 and TA102 OECD TG 471 (1997): in compliance with version 21 July 2020 GLP: Yes With and without 10% S9 mix (Aroclor 1254‐induced rat liver) 1st experiment: +/− S9: plate incorporation method, 0, 1.6, 8, 40, 200, 1000 and 5000 μg/plate. 2nd experiment: –S9: plate incorporation method, 0, 156.25, 312.5, 625, 1250, 2500 and 5000 μg/plate. +S9: preincubation test, 0, 156.25, 312.5, 625, 1250, 2500 and 5000 μg/plate. Negative control: water (solvent) Appropriate positive controls Triplicate plates TFA Na 99.1% pure Negative No evidence of toxicity observed Preliminary cytotoxicity test with TA100: No evidence of toxicity observed. 1 Historical negative and positive control data provided Reverse mutation assay Salmonella typhimurium TA98 and TA100 OECD TG 471: no GLP: no Plate incorporation assay 1.5 mg/plate With and without S9 mix (Aroclor induced rat liver) Negative control: Not reported Positive controls: no TFA Purity: not reported Negative No revertants above the control < 0.000007 revertants/nmole 4 Test not performed according to OECD TG 471 Very short description of the protocol and results Spontaneous mutant colonies and mutant colonies induced by positive controls are missing. Only one concentration tested Only one experiment Reverse mutation assay Salmonella typhimurium TA97a, TA98, TA100, TA1535, and TA102 OECD TG 471 (1997): in compliance with version 26 June 2020 GLP: Yes With and without 10% S9 mix (phenobarbital and β‐naphthoflavone induced rat liver) Plate incorporation assay 1st experiment: 0, 0.5, 1.5, 5, 15, 50, 150, 500, 1500, 5000 μg/plate. 2nd experiment: 0, 78.125, 156. 25, 312.5, 625, 1250, 2500 and 5000 μg/plate. Negative control: water (solvent) Appropriate positive controls Triplicate plates TFA Na 98% pure Negative No evidence of toxicity observed 1 Historical negative and positive control data provided Reverse mutation assay Salmonella typhimurium TA98, TA100, TA1535, TA1537 and TA102 OECD TG 471 (1997): in compliance with version 26 June 2020 GLP: Yes With and without 5% S9 mix (Aroclor 1254‐induced rat liver) 1st experiment: plate incorporation assay 0, 3.16, 10.0, 31.6, 100, 316 or 1000 μg/plate. 2nd experiment: pre‐incubation test 0, 3.16, 10.0, 31.6, 100, 316 or 1000 μg/plate. Negative control: water (solvent) Appropriate positive controls Triplicate plates Two preliminary cytotoxicity tests (plate incorporation test without and with metabolic activation) in TA100. Ten concentrations from 0.316 to 5000 μg/plate were tested. TFA 99.0% pure Negative Main test: cytotoxicity observed at 1000 μg/plate. Preliminary test: Pronounced cytotoxicity (scarce background lawn and reduction of the number of revertants by more than 50%) noted at ≥ 1000 μg/plate. 1 Historical negative and positive control data provided Reverse mutation assay Salmonella typhimurium TA98, TA100 and TA1535 OECD TG 471: no GLP: no Preincubation assay With and without S9 mix (Aroclor 1254‐induced rat liver or testes or combined) 0, 3.6, 8.9, 17.0 and 35.7 mg/mL Negative control: vehicle Positive controls: B(a)P and 4NQO Duplicate plates TFA Na 98% pure Negative 35.7 mg/mL: maximum non‐toxic concentration (as illustrated by revertant ratios less than 1.0 in some instances). 3 Only 3 strains tested Values reported are ratio of the number of His revertant colonies in presence of test compound to the number of His revertant colonies in the absence of test compound No information on cytotoxicity Reverse mutation assay Salmonella typhimurium T98 and TA100 OECD TG 471: no GLP: no Spot test Sterile 6 mm diameter paper disc placed in the centre of each plate 0.1 mL TFA Positive control: MNNG TFA Purity: not reported Negative According to the authors 4 Test not performed according to OECD TG 471 Very short description of the protocol Results not shown Gene mutation assay locus TK+/− Mouse lymphoma cells (L5178Y) OECD TG 490 (2016) GLP: yes MicrotitreR Fluctuation Technique With and without 5% S9 mix (Aroclor 1254‐induced rat liver) ‐Without S9: ‐Without S9: Without S9: 1st experiment + repeat: 3‐h exposure, 0, 360, 560, 760, 960, 1160 and 1360 μg/mL 2nd experiment: 24‐h exposure, 0, 360 to 1360 (10 mM) μg/mL ‐With S9: 3‐h exposure (two experiments), 0, 360 to 1360 μg/mL ‐With S9: 3‐h exposure (two experiments), 0, 360 to 1360 μg/mL With S9: 3‐h exposure (two experiments), 0, 360 to 1360 μg/mL Duplicate cultures Negative control: water (solvent) Positive controls: ‐S9: 4‐NQO, +S9: B(a)P Preliminary cytotoxicity tests: 3‐h exposure +/− S9, 42.5 to 1360 μg/mL 24‐h exposure ‐S9, 5.313 to 1360 μg/mL TFA Na 99.1%–99.3% pure Negative Cytotoxicity: 1st experiment: 3h, the highest dose tested yielded 112% and 93% RTG in the absence and presence of S9 respectively. 2nd experiment: the highest dose tested yielded 76% and 104% relative total growth (RTG) in the absence and presence of S9, respectively. Preliminary cytotoxicity tests: 3‐h exposure: The highest dose tested yielded 112% Relative total growth (RTG) in the absence of S9 and 99% RTG in the presence of S9; 24‐h exposure: The highest dose tested yielded 49% RTG. 1 No calculation of Induced Mutation Frequency (IMF) and no Global Evaluation Factor (GEF) evaluation Historical negative and positive control data provided Gene mutation assay locus TK+/− Mouse Lymphoma cells (L5178Y) OECD TG 490 (2016) GLP: yes With and without 10% S9 mix (phenobarbital and β‐naphthoflavone induced rat liver) Exposure 3 h (3.5 h) 0, 8.192, 20.48, 51.2, 128, 320, 800, 2000 and 5000 μg/mL Negative control: medium Positives controls: Methyl methanesulfonate (MMS), Cyclophosphamide (CP) Preliminary cytotoxicity test, +/− S9, 312.5, 625, 1250, 2500 and 5000 μg/mL TFA Na 98% pure Negative IMF(s): < GEF (126 × 10−6) Cytotoxicity (RTG): –S9: 94%, 80%, 81%, 81%, 72%, 78%, 75%, 60% +S9: 95%, 84%, 82%, 84%, 77%, 80%, 60%, 44% Preliminary cytotoxicity test: Cell count: conc.‐related toxicity RSG: > 20% ± S9 at all doses 1 Historical negative and positive control data provided Gene mutation assay locus TK+/− Mouse Lymphoma cells (L5178Y) OECD TG 490 (2015) GLP: yes With and without 5% S9 mix (phenobarbital and β‐naphthoflavone induced rat liver) Preliminary cytotoxicity test and 1st. mutagenicity test, –S9 4 h and 24 h + S9: 4 h 10.00; 3.33; 1.11; 0.37; 0.12; 0.04; 0.01; 0.00 mM = 2.00; 0.67; 0.22; 0.07; 0.02; 0.01; 0.003; 0.00 μL/mL Definitive test: Without S9: – 4‐h exposure: 0.75; 0.38; 0.19; 0.09; 0.00 μL/mL and –24‐h exposure: 1.00; 0.50; 0.25; 0.13; 0.00 μL/mL With S9: 4‐h exposure, 1.00; 0.50; 0.25; 0.13; 0.00 μL/mL Negative control: water Positives controls: MMS, CP TFA > 99.9% pure Negative In the absence and presence of metabolic activation induced mutation frequency level (IMF (s)) did not exceed 126 × 10−6 (GEF) for most of the concentrations tested (only + S9 4 h 1 μL/mL: MF: 158.06 × 10−6). Cytotoxicity: +S9, 4 h: High cytotoxicity at 2 μL/mL Preliminary cytotoxicity: ‐S9, 4 h: RTG 42–153% 24 h: RTG 0–148% +S9, 4 h: RTG 0%–125% Gene mutation Assay locus TK± Mouse Lymphoma cells (L5178Y) OECD TG 490 (2015) GLP: yes With and without S9 mix (Aroclor 1254 induced rat liver) Two experiments Exposure w/o S9: 3 and 24 h Exposure w S9: 3 h 0, 125, 250, 500, 1000 and 2000 μg/mL Negative control: water Positives controls: MMS, 3‐Methylcholanthrene Preliminary cytotoxicity test, ±S9, 0, 10.0, 31.6, 100, 316, 1000 and 2000 μg/mL TFA 99.0% pure Negative No change in the ratio of small to large mutant colonies indicating no clastogenic potential Main test cytotoxicity: decreased relative total growth at 1000 and 2000 μg/mL The pronounced to complete cytotoxic concentration of 2000 μg/mL was excluded from evaluation in all experiments, as well as the cytotoxic concentration of 1000 μg/mL in the absence of metabolic activation with a 24‐h exposure (RTG = 13%). Preliminary cytotoxicity without and with S9 mix (3 and 24 h): decreased survival at 1000 and 2000 μg/mL 1 Historical negative and positive control data provided Gene mutation assay locus HPRT Mouse lymphoma cells (L5178Y) OECD TG 476 (2016): Not reported GLP: yes MicrotitreR fluctuation Technique With and without 2% S9 mix (Aroclor 1254‐induced rat liver) Exposure: 3 h 1st experiment: 0, 200, 400, 600, 800, 1000, 1200 and 1360 μg/mL 2nd experiment: 0, 150, 300, 500, 700, 900, 1100 and 1360 μg/mL Negative control: water Positive controls: 4NQO, B(a)P Preliminary cytotoxicity test: 0, 42.5 to 1360 μg/mL (equivalent to 10 mM) TFA Na 99.9% pure Negative Cytotoxicity: 1st experiment: highest concentration tested, 1360 μg/mL, gave 96% and 111% relative survival (RS) in the absence and presence of S9, respectively. 2nd experiment: highest concentration tested, 1360 μg/mL, gave 80% and 112% relative survival (RS) in the absence and presence of S9, respectively. Preliminary cytotoxicity: 96% and 111% RS in the absence and presence of S9, respectively 1 Historical negative and positive control data provided Chromosomal aberration assay Chinese hamster lung cells (CHL) OECD TG 473 (2016): not fully compliant GLP: yes With and without 10% S9 mix (phenobarbital and β‐naphthoflavone induced rat liver) Exposure: 3h and 21 h recovery period 0, 128, 320, 800, 2000 and 5000 μg/mL 2 experiments Negative control: medium Positive controls: Ethyl methanesulfonate (EMS) and CP 200 metaphases scored/conc. Single culture Preliminary cytotoxicity test: 0, 312.5, 625, 1250, 2500 and 5000 μg/mL TFA Na 98% pure Negative 1st test. Acceptability criteria were not met as the % of cells with CA (w/o gaps) of the negative control fell outside the 95% confidence interval of the historical negative control distribution in this lab (w/o S9: 0.1–1.1; w S9: −0.2 to 1.3) (validity criteria not met). % of cells with CA (w/o gaps) Control 3h expo w/o S9: 2.0; w S9: 1.5. The study's authors decided to repeat the assay. 2nd test: negative Cytotoxicity: relative increase in cell count (RICC): 1st experiment: –S9: 92%, 97%, 92%, 95% and 79%; +S9: 107%, 92%, 98%, 103% and 82%. 2nd experiment: –S9: 98%, 96%, 94% and 87%, 82%; +S9: 99%, 104%, 96%, 93% and 80%. Preliminary cytotoxicity: Relative Increase in Cell Count (RICC) > 79% at all concentrations ±S9 mix 2 No extended treatment Only 200 metaphases scored/conc. instead of 300. Only 1 test acceptable (2nd test) Chromosomal aberration assay Human peripheral lymphocytes OECD TG 473: not reported, in compliance with TG 473 (2016) GLP: yes With and without 5% S9 mix (Aroclor 1254‐ induced rat liver) 1st experiment: w/o S9 4‐h exposure, 0, 125, 250, 500, 1000 and 2000 μg/mL 21‐h recovery period 2nd experiment: w/o S9 24‐h exposure, 0, 15.63, 31.3, 62.5, 125 and 250 μg/mL 3rd experiment: w S9 4‐h exposure, 0, 125, 250, 500, 1000 and 2000 μg/mL 21‐h recovery period 4th experiment: w S9 4‐h exposure, 0, 500, 750, 1000, 1000*, 2000* μg/mL 21‐h recovery period 5th experiment: w/o S9 24‐h exposure, 0, 125, 175 and 250 μg/mL Negative control: water Positive controls: MMC, CP 300 metaphases scored/conc. Duplicate cultures Preliminary cytotoxicity: *Neutralised with 1 N NaOH TFA 99.0% pure Positive Increase in the % CA noted at 125 μg/mL (1 of 2 experiments) or at 250 μg/mL w/o S9 after 24‐h exposure and at 1000 μg/mL in both experiments with S9 after 4‐h exposure. No increase in % CA w/o S9 after 4‐h exposure. Positive results are also observed at 1000 and 2000 μg/mL in the absence of cytotoxicity when TFA was neutralised by NaOH (not in the buffer). Cytotoxicity: Complete cytotoxicity at 2000 μg/mL 4‐h exposure ±S9 Cytotoxicity at 1000 μg/mL, 4‐h exposure ‐S9 58% (MI: 0.42) and + S9 49% and 45% (MI: 0.51 and 0.55). Cytotoxicity at 250 μg/mL of 57% and 44% (MI: 0.43 or 0.56) after 24‐h exposure –S9. Decrease in pH compared to the control (pH: 7.53) at 1000 μg/mL (pH: 6.03) and 2000 μg/mL (pH: 2.69). The same concentrations of 1000 or 2000 μg/mL medium adjusted by buffering with 1 N NaOH to pH 7.25 or 7.02, respectively, did not lead to any cytotoxicity. Preliminary cytotoxicity: –S9 24‐h exposure: pronounced to complete cytotoxicity starting at 316 μg/mL +S9 4‐h exposure: complete cytotoxicity at 2000 μg/mL Chromosomal aberration assay Human peripheral lymphocytes (pooled blood from 3 female donors) OECD TG 473: not reported, but nearly in compliance with version 2016 GLP: yes With and without 2% S9 mix (Aroclor 1254‐ induced rat liver) 1st experiment: 3‐h exposure ±S9 and harvesting 17 h later 0, 1000, 1200 and 1360 μg/mL (= 10 mM). 2nd experiment: –S9, 20‐h exposure, harvesting at end of treatment, 0, 750, 1200 and 1360 μg/mL +S9, 3‐h exposure and harvesting 17 h later, 0, 1050, 1200 and 1260 μg/mL Negative control: water Positive controls: 4NQO, CP 200 metaphases scored/conc. Preliminary cytotoxicity: 0, 4.934–1360 μg/mL ± S9 Duplicate cultures TFA Na 99.9% pure Negative Only slight cytotoxicity (as measured by Mitotic inhibition MIH) Cytotoxicity: 1st experiment: –S9: no cytotoxicity 1st experiment: +S9: mitotic inhibition of 2% at 1360 μg/mL 2nd experiment: –S9: MIH of 13%, 35% and 35%, 2nd experiment: +S9: MIH of 5%, 13% and 22% 2 Only 200 metaphases scored/conc. instead of 300. Chromosomal aberration assay Human peripheral lymphocytes (pooled blood from 3 male donors) OECD TG 473 (1997): nearly in compliance with version 2016 GLP: yes With and without 2% S9 mix (Aroclor 1254‐ induced rat liver) 0, 85, 170, 340, 680 and 1360 μg/mL (= 10 mM) 1st experiment: 3‐h exposure ±S9 and harvesting 17 h later 2nd experiment: –S9, 20‐h exposure, harvesting at end of treatment +S9, 3‐h exposure and harvesting 17 h later Negative control: water Positive controls: 4NQO, CP 200 metaphases scored/conc. Preliminary cytotoxicity: 0, 0.006641 to 1360 μg/mL ± S9 Duplicate cultures TFA Na 99.3% pure Negative % of CA mostly similar to the concurrent vehicle controls in the presence and absence of S9. One exception after a 20‐h exposure in the absence of S9 in exp. 2: 2% structural aberrations excluding gaps vs. 0% in controls at 1360 μg/mL (% of cells with aberrations within the historical vehicle control frequencies, 61% mitotic inhibition). Not considered biologically relevant. No increases in the frequency of cells with numerical aberrations either in the absence or presence of S9, considered to be of any biological relevance. Cytotoxicity: 1st experiment: 1360 μg/mL, induced approximately 16% and 0% mitotic inhibition (reduction in mitotic index) in the absence and presence of S9 respectively (4‐h exposure). 2nd experiment: 1360 μg/mL, induced approximately 61% and 19% mitotic inhibition in the absence (24‐h exposure) and presence of S9 (4‐h exposure) respectively. 2 Only 200 metaphases scored/conc. instead of 300. Micronucleus test TK6 Human Lymphoblastoid Cells Without Cyt B OECD TG 487 (2016) GLP: yes With and without 5% S9 mix (phenobarbital and 5,6‐benzoflavone induced rat liver) 1st experiment: ±S9, 3‐h exposure, 0, 285, 570 and 1140 μg/mL 2nd experiment: ‐S9, 24‐h exposure, 0, 285, 570 and 1140 μg/mL Negative control: water Positive controls: MMC, CP Analysis: 2000 mononuclear cells/conc. Concentration range finding: 0, 17.8, 35.6, 71.3, 143, 285, 570 and 1140 μg/mL (=10 mM) TFA 99.08% pure Negative No statistically significant increases in % of micro nucleated mononuclear cells. Cytotoxicity: measurement of Relative Population Doubling (RPD). 1st experiment: 3‐h exposure: ‐S9: RPD of 88.4.8% at highest conc.; +S9: RPD of 97.1% at highest concentration 2nd experiment: 24‐h exposure: ‐S9: RPD of 77.0% at highest concentration Range finding: absence of cytotoxicity after 3‐h exposure ±S9. After 24‐h exposure ‐S9: RPD of 71.8% at highest concentration 1 Historical negative and positive control data not provided Micronucleus test CHO‐K1 Cells With Cyt B OECD TG 487 (2016) GLP: yes With and without 1% S9 mix (phenobarbital and β‐Naphtoflavone induced rat liver) 1st experiment: +/‐S9, 3‐h exposure +27‐h recovery, 0, 372, 745, 1117, 1490 μg/mL 2nd experiment: –S9, 27‐h exposure, 0, 372, 745, 1117, 1490 μg/mL 1 μL/mL = 1490 μg/mL Negative control: water Positive controls: MMC, CP CBPI (Cytokinesis‐Block Proliferation Index) and RI (Replication Index) was determined to assess cell proliferation (cytotoxicity) using at least 500 cells per culture. Analysis: 2000 binucleated cells/conc. Preliminary cytotoxicity test: 0.0009–2 μL/mL. The highest tested concentration (2 μL/mL) had cytotoxic effect against CHO‐K1 cells (100% cytotoxicity) in all incubation conditions. TFA > 99.9% pure Negative Cytotoxicity: –S9 27‐h exposure: 40% at 1117 μg/mL and 100% at 1490 μg/mL (no scoring of MN). –S9 3‐h exposure: concentrations are not cytotoxic. +S9 3‐h exposure: concentrations are not cytotoxic. Abbreviations: 4NQO, 4‐Nitroquinoline 1‐oxide; CBPI, Cytokinesis‐Block Proliferation Index; CHL, Chinese hamster lung cells; CP, Cyclophosphamide; EMS, Ethyl methanesulfonate; IMF, induced mutant frequency; MMS, Methyl methanesulfonate; MNNG, N‐methyl‐N′‐nitro‐N‐nitrosoguanidine; RI, Replication Index; RICC, Relative Increase in Cell Count; RPD, Relative Population Doubling; RTG, relative total growth. Assessment of toxicokinetics No specific ADME studies on TFA or TFA Na were submitted. However, toxicokinetic parameters after oral administration to rats and rabbits have been investigated in the newly submitted toxicity studies (#102, #99, #103 and #109). In a 52‐week drinking water study with a 6‐week recovery phase (#109), male and female rats were exposed to TFA at concentrations of 30, 120, and 600 ppm. By week 52, mean plasma concentrations reached steady‐state levels, ranging from: 11.0–14.6 mg/L at 30 ppm; 11.0–14.6 mg/L at 30 ppm; 39.3–61.9 mg/L at 120 ppm; 39.3–61.9 mg/L at 120 ppm; And 121–185 mg/L at 600 ppm. And 121–185 mg/L at 600 ppm. Across all dose groups, plasma concentrations remained consistent over time, indicating that plateau levels were reached. As a result, Tmax values were deemed not meaningful for interpreting the TK profile and were only reported in the full toxicokinetic report. No significant sex differences were observed in systemic exposure, as measured by Cmax and AUC (AUCD7–W59). Exposure increased in a concentration‐proportional manner across the dose range (30–600 ppm) for both sexes by Week 59. The mean ± SD for Cmax and AUCD7‐W59 values of TFA following drinking water of TFA at 30, 120, or 600 ppm to male and female rats for 52 weeks are presented in Table B.1. TABLE B.1Summary of the Mean TFA TK Parameters in Plasma of male and female rats exposed for 52 weeks (# 109).Dose (ppm)SexCmax (mg/L)AUC day 7‐week 59 (week*mg/L)30Male15.3 ± 2.99641 ± 223Female15.8 ± 1.54667 ± 208120Male52.0 ± 4.602340 ± 173Female64.2 ± 24.72750 ± 695600Male194 ± 34.57870 ± 1450Female176 ± 37.76750 ± 1310 Note: Data are presented as mean ± SD for Cmax and AUC day 7‐week 59 values.Abbreviations: AUC, area under the curve; SD, standard deviation. Summary of the Mean TFA TK Parameters in Plasma of male and female rats exposed for 52 weeks (# 109). Note: Data are presented as mean ± SD for Cmax and AUC day 7‐week 59 values. Abbreviations: AUC, area under the curve; SD, standard deviation. In a preliminary dietary reproductive toxicity study in rats (#99), females were fed with TFA Na at varying concentrations during gestation and lactation. Plasma TFA concentrations were measured using HPLC‐MS/MS. In F0 females, mean Cmax values reached up to 485 mg/L on gestation day (GD) 17 and 405 mg/L on lactation day (LD) 11 at the highest dose. Corresponding AUC values (up to 10.285 μg·h/mL) showed a sub‐proportional increase with dose, suggesting nonlinear kinetics due to saturation of absorption or elimination processes (see Tables B.2, B.3, B.4). In F1 offspring, limited sampling precluded AUC calculation. On postnatal day (PND) 4, mean plasma TFA levels were similar across dose groups (~34 mg/L), showing no dose‐dependency. By PND 21, concentrations rose substantially, with mean values reaching 342–394 mg/L at the highest dose. Unexpectedly high TFA levels were detected in control animals and untreated stock, exceeding typical background levels (approximately 0.04 mg/L). In untreated stock, plasma concentrations ranged from 0.167 to 0.211 mg/L, while control animals in the study exhibited even higher values. However, the facility's water supply contained negligible TFA (0.000382 mg/L), effectively ruling out environmental exposure via drinking water. Since the control values remained below those of treated animals, a dosing error is considered unlikely. Instead, the elevated levels are attributed to probable cross‐contamination during sample processing or analytical artefacts (see Table B.2). This finding underscores the importance of implementing rigorous contamination control and analytical quality assurance measures when handling TFA samples in future investigations. TABLE B.2Toxicokinetic data from a preliminary reproductive toxicity study with TFA Na in rats (#99).GroupTime pointDose (ppm)Cmax (mg/L) – IndividualCmax (mg/L) – MeanAUC (μg·h/mL)NotesF0 FemalesGD 1714002672144819Tmax: 0.5–1 h; dose‐related increase, sub‐proportional34003233076892840063348510,285F0 FemalesLD 1114002201714032Tmax: 4.5–5 h (low); 0.5–1 h (mid/high dose); dose‐related increase, sub‐proportional3400329279633484004664059304F1 PupsPND 47007.936.4* —Pooled sample; no dose‐correlation170051.433.7* —420038.234.2* —F1 PupsPND 21700M: 173, F: 178M: 140, F: 142—Dose‐related increase, sub‐proportional1700M: 394, F: 325M: 289, F: 276—4200M: 475, F: 456M: 342, F: 394—Abbreviations: AUC, area under the curve; GD, gestation day; LD, lactation day; PND, postnatal day.‘Cmax individual’ refers to the highest concentration observed in a single animal among the six included in each group, whereas ‘Cmax mean’ refers to the average concentration across those six animals.*Mean Cmax values on PND 4 are for pooled samples from males and females.– means that AUC as not calculated because only one sample was taken for F1 animals on PND4 and PND21. Toxicokinetic data from a preliminary reproductive toxicity study with TFA Na in rats (#99). Abbreviations: AUC, area under the curve; GD, gestation day; LD, lactation day; PND, postnatal day. ‘Cmax individual’ refers to the highest concentration observed in a single animal among the six included in each group, whereas ‘Cmax mean’ refers to the average concentration across those six animals. Mean Cmax values on PND 4 are for pooled samples from males and females. – means that AUC as not calculated because only one sample was taken for F1 animals on PND4 and PND21. TABLE B.3Summary of the mean TFA toxicokinetics parameters in plasma of dams in the preliminary reproductive toxicity study (#99).ppmGD17 Cmax ratioGD17 AUC ratioLD11 Cmax ratioLD11 Cmax ratio1400111134001.431.431.631.5784002.272.132.372.31Abbreviations: AUC, area under the curve; GD, gestation day; LD, lactation day. Summary of the mean TFA toxicokinetics parameters in plasma of dams in the preliminary reproductive toxicity study (#99). Abbreviations: AUC, area under the curve; GD, gestation day; LD, lactation day. TABLE B.4Summary of the mean TFA toxicokinetics parameters in plasma of pups in the preliminary reproductive toxicity study (#99).ppmPND4 Cmax ratioPND21 males Cmax ratioPND21 females Cmax ratio140011134000.972.061.9484000.962.442.77Abbreviations: AUC, area under the curve; PND, postnatal day. Summary of the mean TFA toxicokinetics parameters in plasma of pups in the preliminary reproductive toxicity study (#99). Abbreviations: AUC, area under the curve; PND, postnatal day. In a preliminary developmental toxicokinetic study in rabbits (#102), TFA Na was administered once daily via oral gavage to presumed pregnant dams from GD 6 to 28 at doses of 10, 45, 180, and 750 mg/kg bw per day. Blood samples were collected at various time points post‐dose (1, 2, 3, 5, 8 and 24 h) on GD 6 and GD 28, and from fetuses at necropsy on GD 29. TFA was consistently detected in plasma, including in controls, likely due to background contamination; however, control values were much lower than those in dosed animals (see Table B.5). TABLE B.5Concentrations in mg/L of TFA in plasma of fetal rabbits from F1 control group (GD 29) (#102).F1 animal numberTFA (mg/L)10.57720.60131.074101517.864.76 Concentrations in mg/L of TFA in plasma of fetal rabbits from F1 control group (GD 29) (#102). Elevated TFA levels observed in control animals can be attributed to probable sample contamination or analytical artefacts rather than to true background exposure. Ultra‐short PFAS such as TFA are difficult to quantify and are susceptible to contamination from consumables or reagents, underscoring the need for rigorous blanks and Quality Assurance and Quality Control procedures (Björnsdotter et al., 2020). This conclusion is further supported by the high variability among control values (including a 100‐fold outlier) and the negligible TFA concentrations measured in the facility's water supply. Importantly, control values remained substantially lower than those observed in treated groups and are not considered to compromise the toxicokinetic interpretation of the assessment. TFA was absorbed after oral administration, with median Tmax values ranging from 3 to 8 h on GD 6 and 2–3 h on GD 28 in rabbits. The mean Cmax and AUC₀₋₂₄ increased with dose but less than proportionally, indicating non‐linear kinetics. Accumulation was observed at lower doses (10 and 45 mg/kg bw per day) after repeated dosing. TFA levels in fetuses on GD 29 were comparable to those in maternal plasma at 24 h post‐dose. Maternal‐to‐foetal concentration ratios ranged from 0.983 (45 mg/kg bw per day) to 1.15 (750 mg/kg bw per day), suggesting efficient transplacental transfer and similar systemic exposure between dams and fetuses (see Table B.6). TABLE B.6Summary of the mean trifluoroacetic acid toxicokinetic parameters in plasma of pregnant rabbits (# 102).Interval (GD)Dose groupDose level (mg/kg/day)Cmax (mg/L)Tmax (h)AUC‐24 (h*mg/L)t1/2 (h)621059.08.001240NR3452185.004730NR41806795.0014,000NR575018603.0026,200NR282101702.003410NR3454452.008840NR41808392.0012,600NR575018603.0021,80011.5Abbreviations: AUC, area under the curve; GD, gestation day; NR, Not reported due to an inability to characterise the elimination phase. Summary of the mean trifluoroacetic acid toxicokinetic parameters in plasma of pregnant rabbits (# 102). Abbreviations: AUC, area under the curve; GD, gestation day; NR, Not reported due to an inability to characterise the elimination phase. Increased levels of TFA were detected in both maternal and fetal aqueous/vitreous humour in a dose‐dependent manner compared with the controls. In addition, the fetal levels were higher than maternal levels (see Table B.7). TABLE B.7Back‐calculated concentrations of trifluoroacetic acid (ng/mL) in maternal and fetal rabbit aqueous/humour samples on GD 29 (#102).Group 1Group 2Group 3Group 4Group 5 Maternal samples Mean966362,383155,333202,017252,833 As % Control – 646 1608 2090 2616 SD18,20610,03426,082140,91667,467 N 66666Min23052,800134,00022,100195,000Max46,20080,600197,0004,500,000379,000 Fetal samples Mean14,14796,700215,833268,200292,500 As % Control – 683 1525 1895 2067 As % of Maternal 146 155 139 133 116 SD26,90943,19039,47439,35469,831 N 65654Min36567,100178,000229,000199,000Max68,300173,000275,000314,00036,100Abbreviations: N, number of animals examined; SD, standard deviation. Back‐calculated concentrations of trifluoroacetic acid (ng/mL) in maternal and fetal rabbit aqueous/humour samples on GD 29 (#102). Abbreviations: N, number of animals examined; SD, standard deviation. Both maternal and foetal aqueous/vitreous humour showed lactate levels that were comparable with the controls, however foetal lactate levels were approximately 2‐fold higher than the maternal levels (see Table B.8). TABLE B.8Back‐calculated concentrations of lactate (ng/mL) in maternal and foetal rabbit aqueous/humour samples on GD 29 (#102).Group 1Group 2Group 3Group 4Group 5 Maternal samples Mean14381290145812041203SD225174174204135 N 66666Min1110111013109831010Max17001460174014401330 Fetal samples Mean24332554236326542650SD164276191166287 N 65653Min22802330212022802320Max26803030254026802840Abbreviations: N, number of animals examined; SD, standard deviation. Back‐calculated concentrations of lactate (ng/mL) in maternal and foetal rabbit aqueous/humour samples on GD 29 (#102). Abbreviations: N, number of animals examined; SD, standard deviation. In the main developmental toxicity study in rabbits (#103), both maternal and fetal aqueous/vitreous humour (pool sample, litter basis) showed increased levels of TFA when compared to controls, with evidence of a dose response and higher levels in fetuses. As TFA was quantified in plasma, systemic exposure to TFA was confirmed (see Table B.9). TABLE B.9Summary of trifluoroacetic acid plasma concentrations (mg/L) in pregnant rabbit and fetal plasma (#103).Satellite phaseMain phasePregnant damsFetalGD 6GD 19GD 20GD 28 Group Dose level (mg/kg per day) 2 a 24 a 2 a 24 a Pme b 2 a 100.4100.45515.113.03.680.98723010396.33923161663153601861744453191934454250674397808342234793575011404099783302511120Abbreviations: GD, gestation day; Pme, Post‐maternal euthanasia. a Plasma concentrations taken at 2 or 24 h post dose. b Plasma concentrations taken post maternal euthanasia. Summary of trifluoroacetic acid plasma concentrations (mg/L) in pregnant rabbit and fetal plasma (#103). Abbreviations: GD, gestation day; Pme, Post‐maternal euthanasia. Plasma concentrations taken at 2 or 24 h post dose. Plasma concentrations taken post maternal euthanasia. Increased levels of TFA were detected in both maternal and fetal aqueous/vitreous humour in a dose‐dependent manner compared with the controls (see Table B.10). TABLE B.10Back‐calculated concentrations of trifluoroacetic acid (ng/mL) in maternal and foetal rabbit aqueous/humour samples.Group 1Group 2Group 3Group 4Group 5 Maternal samples GD20 Mean4922186,125193,250199,625273,875SD720144,23135,60832,64566,694 N 88888Min403139,000165,000159,000170,000Max19,900286,000278,000269,000385,000 Fetal samples GD 29 Mean614209,625254,000281,625322,500SD14535,23448,76578,36053,219 N 88888Min425163,000179,000191,000253,000Max881253,000335,000438,000391,000Abbreviations: GD, gestation day; N, number of animals examined; SD, standard deviation. Back‐calculated concentrations of trifluoroacetic acid (ng/mL) in maternal and foetal rabbit aqueous/humour samples. Abbreviations: GD, gestation day; N, number of animals examined; SD, standard deviation. Both maternal and fetal lactate levels in aqueous/humour samples were comparable with the controls. Lactate levels in fetal aqueous/vitreous humour samples on GD29 were approximately twofold the levels observed in maternal samples on GD20 (see Table B.11). TABLE B.11Back‐calculated concentrations of lactate (ng/mL) in maternal and foetal rabbit aqueous/humour samples (#103).Group 1Group 2Group 3Group 4Group 5 Maternal samples GD20 Mean12961241122011651165SD1751331468181 N 88888Min11001100108010801080Max16801510146013001300 Fetal samples GD 29 Mean24382251215023562445SD359345219284418 N 88888Min19901890186019002000Max31702920238027503300Abbreviations: GD, gestation day; N, number of animals examined; SD, standard deviation. Back‐calculated concentrations of lactate (ng/mL) in maternal and foetal rabbit aqueous/humour samples (#103). Abbreviations: GD, gestation day; N, number of animals examined; SD, standard deviation. Assessment of repeated dose toxicity studies Species n: number reliabilityStudy (method/type, length, route of exposure)Test substanceCritical effectsNOAEL (mg/kg bw per day)LOAEL (mg/kg bw per day)ReferenceRat (Wistar) n: 5 Reliability: 2 (reliable with restrictions)Repeated dose toxicity: oral (pilot, 14 days) OECD TG: – GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 98.7 Oral: feed 0, 600, 1200, 2400 (ppm) Males: 0, 43, 85, 170 mg/kg bw per day Females: 0, 45, 91, 190 mg/kg bw per day Male/female toxicity at the LOAEL: No treatment‐related mortality and no clinical signs were observed in the study. Body weight gain and food consumption were not affected by trifluoroacetate administration. Haematology: males: lower mean absolute lymphocyte count (−13% at 1200 and 2400 ppm), increased number of neutrophils (+15% at 1200, +30% at 2400 ppm) compared to controls; females: dose‐related decrease in total white blood cell count (−17% at 1200 ppm, −30% at 2400 ppm, relative to controls), lower mean absolute lymphocyte count (−24% at 1200 and −38% at 2400 ppm), increased number of neutrophils (+53% at 1200, +60% at 2400 ppm) compared to controls. Clinical chemistry: males/females: no effects. Organ weight: males: mean absolute (+18%) and relative (+19%) liver weights were found to be statistically significantly higher in males at 1200 ppm (and above), when compared to controls; females: no effects. Histopathology: males: a slight diffuse centrilobular hepatocellular hypertrophy was observed in 1/5 at 1200 ppm and 2/5 at 2400 ppm; females: no effects. Hepatotoxicity: males/females: not at 1200 ppm. Enzyme induction: males/females: No significant effect was noticed in benzyloxyresorufin O‐dealkylase (BROD), ethoxyresorufin O‐deethylase (EROD) and pentoxyresorufin O‐dealkylas (PROD) activities after trifluoroacetate administration in females, whereas a significant dose‐related increase in lauric acid hydroxylation was observed in males reaching 125% at 1200 ppm and 159% increase at 2400 ppm, when compared to controls. Cell cycling: males/females: no effect of treatment on hepatocellular proliferation. Palmitoyl‐CoA oxidation activity: males: +145% (1200 ppm) and 152% (2400 ppm) compared to control; females: no effects. Effects at doses higher than the LOAEL: Effects observed at 1200 ppm were more pronounced at 2400 ppm. 43 (males) and 45 (females)85 (males) and 91 (females)#15 Remarks: Limited number of parameters investigated. No effects on total bilirubin; glucose and bile acids not investigated. NOAEL in females has been decreased from > 190 mg/kg bw per day to 45 mg/kg bw per day, based upon effects on total WBC, decreased lymphocytes and increased neutrophils. The effect was seen in both sexes, less pronounced in males. Rat (Wistar) N: 5 Reliability: 1 (reliable without restriction) Repeated dose toxicity: oral (main 28‐days) OECD TG 407 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.1 Oral: feed 0, 600, 1800, 5400, 16,000 (ppm) Males: 0, 50, 149, 436, 1315 mg/kg bw per day Females: 0, 52, 157, 457, 1344 mg/kg bw per day Male/female toxicity at the LOAEL: No treatment‐related mortality and no clinical signs were observed in the study. Body weight gain: tendency to decreased body weight gain (approx. 10%) from 1800 ppm (149 mg/kg bw per day) onwards in males. Liver: Increased absolute and relative liver weight (relative: +15%*, in males) Clinical chemistry: decreased glucose levels −36%*, in males; −16%, in females), decreased cholesterol in males (−‐23%, −). Increased albumin/globulin (A/G) ratio, statistically significant. Urinalysis: higher ketone levels in urine in both sexes (semi‐quantitative analysis). Effects at doses higher than the LOAEL: At 5400 ppm: Clinical chemistry: decreased globulin levels in males (−12%*), decreased calcium levels in males (−4.9%*). Liver weight: increased relative liver weight +19%* in males and 13%* in females Pituitary gland: decreased absolute and relative pituitary gland weight in females (relative: −23%*). At 16000 ppm: Liver: increased absolute and relative liver weight (relative: +33%* in males; +18%* in females) Clinical chemistry: increased ALT (+37%* in males, +23%* in females), decreased chlorine levels in females (−0.9%, −0.9%, −2.8%, −3.7%*), decreased glucose levels (−29%* in males, −30%* in females), decreased cholesterol in males (−30%*). Urinalysis: increased urinary volume in males (+65%*). 50149#16 Rat (Wistar) N: 10 Reliability: 1 (reliable without restriction) Repeated dose toxicity: oral (main 90‐days) OECD TG 408 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99 Oral: feed 0, 160, 1600, 16,000 (ppm) Males: 0, 9.9, 98, 1043 mg/kg bw per day Females: 0, 12.2, 123, 1216 mg/kg bw per day Male/female toxicity at the LOAEL: No treatment‐related mortality and no clinical signs were observed in the study. No effects were recorded for exploratory motor activity, open field observations, sensory reactivity and grip strength. Effects at 1600 ppm (98 mg/kg bw per day (males) and 123 mg/kg bw per day (females)): Body weight gain: no effects. Haematology: effect size of changes in haematological parameters (haemoglobin, mean corpuscular haemoglobin (MCH) in females) are not considered to be biologically relevant. Clinical chemistry: lower mean total bilirubin (−69% (males), −52% (females)) and glucose concentrations (−28% (males), −25% (females)). None of the measured liver enzymes was increased > 2‐fold. Urinalysis: higher ketone levels (grade 4) in males. Organ weight: mean absolute and relative liver weights were increased by 8 and 14% in males and by 9 and 13% in females. Histopathology: 5/9 males had minimal to slight diffuse centrilobular to panlobular hepatocellular hypertrophy. In addition, a loss of the periportal hepatocellular vacuolation was observed in males. Effects at doses higher than the LOAEL: Target organ: liver. Effects at 16,000 ppm (1043 mg/kg bw per day (males), 1216 mg/kg bw per day (females)): Body weight gain: −17% (males), −14% (females). Haematology: lower haemoglobin (−8%), statistically significantly lower mean corpuscular volume (−6%), mean corpuscular haemoglobin (−7%) and haematocrit (−6%), all in females. Effect size of no biological relevance. Clinical chemistry: mean total bilirubin (−81% (males), −76% (females)) and glucose concentrations (−29% (males), −17% (females)) were markedly lower. None of the measured liver enzymes was increased > 2‐fold. Urinalysis: higher ketone levels (grade 4) in both sexes. Organ weight: mean absolute and relative liver weights were increased by 19 and 33% in males and by 23 and 28% in females. Histopathology: higher incidence of hepatocellular necrotic foci in males; all males and 9/10 females had minimal to moderate diffuse centrilobular to panlobular hepatocellular hypertrophy with ground‐class appearance of the hepatocellular cytoplasm. Loss of the periportal hepatocellular vacuolation was observed in both sexes. 9.9 (males) and 12.2 (females)98 (males) and 123 (females)#8 Remarks: The LOAEL (1600 ppm) is mostly based upon histopathological findings in liver, associated with some increase in liver weight. Increase in liver enzymes was not biologically relevant, as were some changes in electrolytes and haematological parameters. As regards rodent cardiomyopathy, comparable incidences were observed in controls (4/10), low (5/10) and mid (4/10) dose and were slightly higher in high dose males (9/10), while no effects were observed in females (2/10, 2/10, 1/10, 1/10). The WG considered that the lesion, observed also in #100, was present in concurrent controls, was more prone in males than in females and it was minor in severity and therefore considered as expression of the background condition. Rat (Sprague Dawley) N: 10 Reliability: 1 (reliable without restriction) Repeated dose toxicity: Oral (main 90 days) Other: GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 98 Oral: feed 0, 450, 1500, 15,000 (ppm) Males: 0, 27.2, 89.2, 962.2 mg/kg bw per day Females: 0, 28.7, 101.4, 1054.8 mg/kg bw per day Male/female toxicity at the LOAEL: In high‐dose animals (15,000 ppm), body weight (gain) was decreased > 10% and food utilisation was impaired. Different haematological and clinical chemistry parameters were statistically significantly different to controls in the high dose animals but were mostly of lower magnitude and the change was not clearly dose related. Relative liver weight in males was 40% higher than in controls, and hepatocellular hypertrophy was reported in all high dose males. In addition, majority of high dose males showed bile duct hyperplasia. 4/10 high‐dose males had also follicular cell hypertrophy. High‐dose females had relative liver weight increase of 18% without any histopathological correlates. Effects at doses higher than the LOAEL: Liver is the target organ. The highest dose was LOAEL. 89.2 (males) 101.4 (females) 962.2 (males) 1054.8 (females) #124 Remarks: Mortality: no mortality was observed. Clinical signs: no clinical signs were observed. Body weight (gain): body weight gain (weeks 0–13) was 25% lower in high dose males and 30% in females, compared to control animals. Terminal body weight was 15% lower in high dose males and 11% lower in high dose females, compared to control animals. Effects on mid dose animals were far below 10% difference to controls. Food consumption and utilisation: food consumption was 5% lower in both males and females of the high dose in period 0–13 weeks. Food utilisation was 17% lower in high dose females and 20% lower in high dose males compared to control animals in period 0–13 weeks. Haematology: in high dose males haemoglobin was statistically significantly decreased, while MCHC was statistically significantly decreased in mid and high dose animals, compared to controls. Magnitude of the effects was ca. 5%. In females, MCV and MCH were statistically significantly decreased in all treated groups, while haemoglobin and haematocrit were statistically significantly decreased in mid and high dose females. Magnitude of the effects was up to 10%. Coagulation: no effects observed. Clinical biochemistry: in high dose males, ALP, AST and AP were statistically significantly increased (ca. 1.5‐fold) compared to control. Total protein, globulin, kalium and cholesterol were statistically significantly decreased in high dose males. In high dose females, only ALP was statistically significantly decreased (1.5‐fold). Statistically significant decrease in total bilirubin was observed in all treated males and females, without clear dose–response. Urinalysis: no effects on urinalysis were observed. Macroscopy: no treatment‐related macroscopic findings were observed. Organ weight: in high dose males, relative organ weight was statistically significantly increased for liver, testes, brain, thymus, kidneys and heart. Increase in relative liver weight was 40% compared to controls, while in mid dose males, it was +11%. In females, only relative liver weight was statistically significantly increased, +18% compared to control animals. Thyroid was not weighed. Histopathology: 7**/10 males in high dose had minimal bile duct hyperplasia while no such finding was observed in control animals. All 10**/10 males in high dose group had mild hepatocellular hypertrophy. 4*/10 males had mild follicular cell hypertrophy (thyroid). No histopathological effects were observed in females. Rat (Crl:CD(SD)) N: 30 for control and high does (10 for recovery), 20 for low and mid dose 3 or 4 for toxicokinetic investigations Reliability: 2 (reliable with restriction) Repeated dose toxicity: oral (main 52‐weeks) with 6 weeks recovery phase OECD TG 452 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.85 Oral: drinking water 0, 30, 120, 600 (ppm) Males: 0, 1.8, 7.6, 37.8 mg/kg bw per day Females: 0, 3.3, 12.2, 64.0 mg/kg bw per day Male/female toxicity at the LOAEL: No LOAEL was derived since the effects on some blood and clinical chemistry parameters, observed in animals of high dose, were biologically not relevant and inconsistent between measuring time points (3 and 12 months), though in some cases statistically significant. There were no effects on body and organ weights or on histopathology, and survival was not affected. Ophthalmological observations were not dose related. Macroscopic records of cataracts were made in most animals also during pre‐treatment inspection. Effects at doses higher than the LOAEL: No target organ could be identified in this study. There were no TFA Na related effects noted by the end of the 6‐week recovery phase. The doses applied in this study are considered too low as maximum tolerable dose (MTD) was not reached. 37.8 (males) 64.0 (females) Highest dose tested > 37.8 (males) > 64.0 (females) 109 Remarks: Mortality: no treatment‐related mortality was observed. The unscheduled deaths were typical spontaneous changes commonly observed in Sprague Dawley rats of this age and strain, and were not considered test article‐related as not occurring at higher incidence in treated animals. Clinical signs: no clinical signs were observed. Body weight (gain): there were no effects on body weight or body weight gain. Haematology: Some statistically significant effects on haematological parameters of minor biological relevance observed after 3 months in high dose animals (males: −4% RBC, −4% haemglobin, −4% haematocrit, −7% MPV; females: −3% haemoglobin, −2.5% MCV, −7% MPV) were not confirmed at the end of the study. Repeated statistically significant decrease of minor biological relevance in MCH in females was observed at 90 day (−2.4%) and at the end of the study (−3%). Coagulation: increase in prothrombin time in males at mid (+19%) and high dose (+16%) at 90 days time point was not confirmed at the end of the study. In females, increase in prothrombin time (+6%, +8%) and activated partial thromboplastin time (APTT) (+8%, +15%) at mid and high dose at 90 days was not confirmed at the end of the study. Clinical chemistry: approximately 1.5‐ to 2‐fold statistically significant increase in ALT was seen in males and females at the high dose only at the end of the study, while statistically significantly increased AST in females was observed only at the end of the study. A statistically significant decrease in total protein, and a statistically significant increase in A/G ratio were seen in males only after 3 months but not at the end of the study. Statistically significantly decreased globulin was observed at both time points, 3 months and at the end. A statistically significant decrease in glucose was observed in males after 3 months but not at the end of the study while in females a statistically significant decrease was measured only at the end of the study. Total bilirubin was statistically significantly decreased in high dose males only at 3 months measurement and in females at 3 months and at the end of the study. Urinalysis: no effects on urinalysis were observed. Macroscopy: no treatment‐related macroscopic findings were observed. Organ weights: no treatment‐related effects on organ weights were observed. In several organs, relative organ weight in the high dose animals was rather decreased than increased. Histopathology: no treatment‐related histopathological findings were observed. Sodium trifluoroacetate (TFA Na) Purity (%): 98.7 Oral: feed 0, 600, 1200, 2400 (ppm) Males: 0, 43, 85, 170 mg/kg bw per day Females: 0, 45, 91, 190 mg/kg bw per day Male/female toxicity at the LOAEL: No treatment‐related mortality and no clinical signs were observed in the study. Body weight gain and food consumption were not affected by trifluoroacetate administration. Haematology: males: lower mean absolute lymphocyte count (−13% at 1200 and 2400 ppm), increased number of neutrophils (+15% at 1200, +30% at 2400 ppm) compared to controls; females: dose‐related decrease in total white blood cell count (−17% at 1200 ppm, −30% at 2400 ppm, relative to controls), lower mean absolute lymphocyte count (−24% at 1200 and −38% at 2400 ppm), increased number of neutrophils (+53% at 1200, +60% at 2400 ppm) compared to controls. Clinical chemistry: males/females: no effects. Organ weight: males: mean absolute (+18%) and relative (+19%) liver weights were found to be statistically significantly higher in males at 1200 ppm (and above), when compared to controls; females: no effects. Histopathology: males: a slight diffuse centrilobular hepatocellular hypertrophy was observed in 1/5 at 1200 ppm and 2/5 at 2400 ppm; females: no effects. Hepatotoxicity: males/females: not at 1200 ppm. Enzyme induction: males/females: No significant effect was noticed in benzyloxyresorufin O‐dealkylase (BROD), ethoxyresorufin O‐deethylase (EROD) and pentoxyresorufin O‐dealkylas (PROD) activities after trifluoroacetate administration in females, whereas a significant dose‐related increase in lauric acid hydroxylation was observed in males reaching 125% at 1200 ppm and 159% increase at 2400 ppm, when compared to controls. Cell cycling: males/females: no effect of treatment on hepatocellular proliferation. Palmitoyl‐CoA oxidation activity: males: +145% (1200 ppm) and 152% (2400 ppm) compared to control; females: no effects. Effects at doses higher than the LOAEL: Effects observed at 1200 ppm were more pronounced at 2400 ppm. Remarks: Limited number of parameters investigated. No effects on total bilirubin; glucose and bile acids not investigated. NOAEL in females has been decreased from > 190 mg/kg bw per day to 45 mg/kg bw per day, based upon effects on total WBC, decreased lymphocytes and increased neutrophils. The effect was seen in both sexes, less pronounced in males. Rat (Wistar) N: 5 Reliability: 1 (reliable without restriction) Repeated dose toxicity: oral (main 28‐days) OECD TG 407 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.1 Oral: feed 0, 600, 1800, 5400, 16,000 (ppm) Males: 0, 50, 149, 436, 1315 mg/kg bw per day Females: 0, 52, 157, 457, 1344 mg/kg bw per day Male/female toxicity at the LOAEL: No treatment‐related mortality and no clinical signs were observed in the study. Body weight gain: tendency to decreased body weight gain (approx. 10%) from 1800 ppm (149 mg/kg bw per day) onwards in males. Liver: Increased absolute and relative liver weight (relative: +15%*, in males) Clinical chemistry: decreased glucose levels −36%*, in males; −16%, in females), decreased cholesterol in males (−‐23%, −). Increased albumin/globulin (A/G) ratio, statistically significant. Urinalysis: higher ketone levels in urine in both sexes (semi‐quantitative analysis). Effects at doses higher than the LOAEL: At 5400 ppm: Clinical chemistry: decreased globulin levels in males (−12%*), decreased calcium levels in males (−4.9%*). Liver weight: increased relative liver weight +19%* in males and 13%* in females Pituitary gland: decreased absolute and relative pituitary gland weight in females (relative: −23%*). At 16000 ppm: Liver: increased absolute and relative liver weight (relative: +33%* in males; +18%* in females) Clinical chemistry: increased ALT (+37%* in males, +23%* in females), decreased chlorine levels in females (−0.9%, −0.9%, −2.8%, −3.7%*), decreased glucose levels (−29%* in males, −30%* in females), decreased cholesterol in males (−30%*). Urinalysis: increased urinary volume in males (+65%*). Rat (Wistar) N: 10 Reliability: 1 (reliable without restriction) Repeated dose toxicity: oral (main 90‐days) OECD TG 408 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99 Oral: feed 0, 160, 1600, 16,000 (ppm) Males: 0, 9.9, 98, 1043 mg/kg bw per day Females: 0, 12.2, 123, 1216 mg/kg bw per day Male/female toxicity at the LOAEL: No treatment‐related mortality and no clinical signs were observed in the study. No effects were recorded for exploratory motor activity, open field observations, sensory reactivity and grip strength. Effects at 1600 ppm (98 mg/kg bw per day (males) and 123 mg/kg bw per day (females)): Body weight gain: no effects. Haematology: effect size of changes in haematological parameters (haemoglobin, mean corpuscular haemoglobin (MCH) in females) are not considered to be biologically relevant. Clinical chemistry: lower mean total bilirubin (−69% (males), −52% (females)) and glucose concentrations (−28% (males), −25% (females)). None of the measured liver enzymes was increased > 2‐fold. Urinalysis: higher ketone levels (grade 4) in males. Organ weight: mean absolute and relative liver weights were increased by 8 and 14% in males and by 9 and 13% in females. Histopathology: 5/9 males had minimal to slight diffuse centrilobular to panlobular hepatocellular hypertrophy. In addition, a loss of the periportal hepatocellular vacuolation was observed in males. Effects at doses higher than the LOAEL: Target organ: liver. Effects at 16,000 ppm (1043 mg/kg bw per day (males), 1216 mg/kg bw per day (females)): Body weight gain: −17% (males), −14% (females). Haematology: lower haemoglobin (−8%), statistically significantly lower mean corpuscular volume (−6%), mean corpuscular haemoglobin (−7%) and haematocrit (−6%), all in females. Effect size of no biological relevance. Clinical chemistry: mean total bilirubin (−81% (males), −76% (females)) and glucose concentrations (−29% (males), −17% (females)) were markedly lower. None of the measured liver enzymes was increased > 2‐fold. Urinalysis: higher ketone levels (grade 4) in both sexes. Organ weight: mean absolute and relative liver weights were increased by 19 and 33% in males and by 23 and 28% in females. Histopathology: higher incidence of hepatocellular necrotic foci in males; all males and 9/10 females had minimal to moderate diffuse centrilobular to panlobular hepatocellular hypertrophy with ground‐class appearance of the hepatocellular cytoplasm. Loss of the periportal hepatocellular vacuolation was observed in both sexes. Remarks: The LOAEL (1600 ppm) is mostly based upon histopathological findings in liver, associated with some increase in liver weight. Increase in liver enzymes was not biologically relevant, as were some changes in electrolytes and haematological parameters. As regards rodent cardiomyopathy, comparable incidences were observed in controls (4/10), low (5/10) and mid (4/10) dose and were slightly higher in high dose males (9/10), while no effects were observed in females (2/10, 2/10, 1/10, 1/10). The WG considered that the lesion, observed also in #100, was present in concurrent controls, was more prone in males than in females and it was minor in severity and therefore considered as expression of the background condition. Rat (Sprague Dawley) N: 10 Reliability: 1 (reliable without restriction) Repeated dose toxicity: Oral (main 90 days) Other: GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 98 Oral: feed 0, 450, 1500, 15,000 (ppm) Males: 0, 27.2, 89.2, 962.2 mg/kg bw per day Females: 0, 28.7, 101.4, 1054.8 mg/kg bw per day Male/female toxicity at the LOAEL: In high‐dose animals (15,000 ppm), body weight (gain) was decreased > 10% and food utilisation was impaired. Different haematological and clinical chemistry parameters were statistically significantly different to controls in the high dose animals but were mostly of lower magnitude and the change was not clearly dose related. Relative liver weight in males was 40% higher than in controls, and hepatocellular hypertrophy was reported in all high dose males. In addition, majority of high dose males showed bile duct hyperplasia. 4/10 high‐dose males had also follicular cell hypertrophy. High‐dose females had relative liver weight increase of 18% without any histopathological correlates. Effects at doses higher than the LOAEL: Liver is the target organ. The highest dose was LOAEL. 89.2 (males) 101.4 (females) 962.2 (males) 1054.8 (females) Remarks: Mortality: no mortality was observed. Clinical signs: no clinical signs were observed. Body weight (gain): body weight gain (weeks 0–13) was 25% lower in high dose males and 30% in females, compared to control animals. Terminal body weight was 15% lower in high dose males and 11% lower in high dose females, compared to control animals. Effects on mid dose animals were far below 10% difference to controls. Food consumption and utilisation: food consumption was 5% lower in both males and females of the high dose in period 0–13 weeks. Food utilisation was 17% lower in high dose females and 20% lower in high dose males compared to control animals in period 0–13 weeks. Haematology: in high dose males haemoglobin was statistically significantly decreased, while MCHC was statistically significantly decreased in mid and high dose animals, compared to controls. Magnitude of the effects was ca. 5%. In females, MCV and MCH were statistically significantly decreased in all treated groups, while haemoglobin and haematocrit were statistically significantly decreased in mid and high dose females. Magnitude of the effects was up to 10%. Coagulation: no effects observed. Clinical biochemistry: in high dose males, ALP, AST and AP were statistically significantly increased (ca. 1.5‐fold) compared to control. Total protein, globulin, kalium and cholesterol were statistically significantly decreased in high dose males. In high dose females, only ALP was statistically significantly decreased (1.5‐fold). Statistically significant decrease in total bilirubin was observed in all treated males and females, without clear dose–response. Urinalysis: no effects on urinalysis were observed. Macroscopy: no treatment‐related macroscopic findings were observed. Organ weight: in high dose males, relative organ weight was statistically significantly increased for liver, testes, brain, thymus, kidneys and heart. Increase in relative liver weight was 40% compared to controls, while in mid dose males, it was +11%. In females, only relative liver weight was statistically significantly increased, +18% compared to control animals. Thyroid was not weighed. Histopathology: 7**/10 males in high dose had minimal bile duct hyperplasia while no such finding was observed in control animals. All 10**/10 males in high dose group had mild hepatocellular hypertrophy. 4*/10 males had mild follicular cell hypertrophy (thyroid). No histopathological effects were observed in females. Rat (Crl:CD(SD)) N: 30 for control and high does (10 for recovery), 20 for low and mid dose 3 or 4 for toxicokinetic investigations Reliability: 2 (reliable with restriction) Repeated dose toxicity: oral (main 52‐weeks) with 6 weeks recovery phase OECD TG 452 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.85 Oral: drinking water 0, 30, 120, 600 (ppm) Males: 0, 1.8, 7.6, 37.8 mg/kg bw per day Females: 0, 3.3, 12.2, 64.0 mg/kg bw per day Male/female toxicity at the LOAEL: No LOAEL was derived since the effects on some blood and clinical chemistry parameters, observed in animals of high dose, were biologically not relevant and inconsistent between measuring time points (3 and 12 months), though in some cases statistically significant. There were no effects on body and organ weights or on histopathology, and survival was not affected. Ophthalmological observations were not dose related. Macroscopic records of cataracts were made in most animals also during pre‐treatment inspection. Effects at doses higher than the LOAEL: No target organ could be identified in this study. There were no TFA Na related effects noted by the end of the 6‐week recovery phase. The doses applied in this study are considered too low as maximum tolerable dose (MTD) was not reached. 37.8 (males) 64.0 (females) Highest dose tested > 37.8 (males) > 64.0 (females) Remarks: Mortality: no treatment‐related mortality was observed. The unscheduled deaths were typical spontaneous changes commonly observed in Sprague Dawley rats of this age and strain, and were not considered test article‐related as not occurring at higher incidence in treated animals. Clinical signs: no clinical signs were observed. Body weight (gain): there were no effects on body weight or body weight gain. Haematology: Some statistically significant effects on haematological parameters of minor biological relevance observed after 3 months in high dose animals (males: −4% RBC, −4% haemglobin, −4% haematocrit, −7% MPV; females: −3% haemoglobin, −2.5% MCV, −7% MPV) were not confirmed at the end of the study. Repeated statistically significant decrease of minor biological relevance in MCH in females was observed at 90 day (−2.4%) and at the end of the study (−3%). Coagulation: increase in prothrombin time in males at mid (+19%) and high dose (+16%) at 90 days time point was not confirmed at the end of the study. In females, increase in prothrombin time (+6%, +8%) and activated partial thromboplastin time (APTT) (+8%, +15%) at mid and high dose at 90 days was not confirmed at the end of the study. Clinical chemistry: approximately 1.5‐ to 2‐fold statistically significant increase in ALT was seen in males and females at the high dose only at the end of the study, while statistically significantly increased AST in females was observed only at the end of the study. A statistically significant decrease in total protein, and a statistically significant increase in A/G ratio were seen in males only after 3 months but not at the end of the study. Statistically significantly decreased globulin was observed at both time points, 3 months and at the end. A statistically significant decrease in glucose was observed in males after 3 months but not at the end of the study while in females a statistically significant decrease was measured only at the end of the study. Total bilirubin was statistically significantly decreased in high dose males only at 3 months measurement and in females at 3 months and at the end of the study. Urinalysis: no effects on urinalysis were observed. Macroscopy: no treatment‐related macroscopic findings were observed. Organ weights: no treatment‐related effects on organ weights were observed. In several organs, relative organ weight in the high dose animals was rather decreased than increased. Histopathology: no treatment‐related histopathological findings were observed. Assessment of multigeneration reproductive toxicity studies Species n: number reliabilityStudy (method/type, length, route of exposure)Test substanceCritical effectsNOAEL (mg/kg bw per day)LOAEL (mg/kg bw per day)Reference Rat (Han Wistar) N: F0 generation: 6 females (no F0 males since females are treated from GD6) F1 generation: 6/sex/group Multi‐generation reproductive toxicity (pilot 1‐extended) OECD TG: – GLP: no Sodium trifluoracetate (TFA Na) Purity (%): 99.9 Oral: feed 0, 1400, 3400, 8400 ppm 0, 700, 1700, 4200 ppm during lactation (LD1 to LD21) F0 females during gestation: 97.5, 230, 547 mg/kg bw per day F0 females during lactation: 110.2, 262, 578 mg/kg bw per day F1 females LD21‐LD34: 228, 557, 1369 mg/kg bw per day F1 males LD21‐LD34: 202, 476, 1290 mg/kg bw per day Maternal toxicity at the LOAEL: At 1400/700 ppm: Body weight and body weight gain: decreased body weight gain in F0 dams at the beginning of gestation (GD6‐10: −25%*, −33%*, −67%**) Liver: increased weight in F1 animals LD35 (+26%**, +28%**, +44%** in males and +25%**, +26%**, +33%** in females) Effects at doses higher than the LOAEL: At 3400/1700 ppm: Liver: increased weight in F0 dams (not statistically significant, 11%) Body weight and body weight gain: decreased body weight in F1 females from LD28 and decreased body weight gain in F1 females LD24‐28 (not clearly dose‐related) At 8400/4200 ppm: Body weight and body weight gain: decreased body weight gain in F0 dams during lactation (−9%*) Maternal: n.a.n.a.#99 Offspring toxicity at the LOAEL: At 1400/700 ppm: Body weight and body weight gain: decreased body weight gain of males and female pups at the beginning of lactation (LD4‐7: −19%*, −19%**, −26%* in males and −17%*, −12%**, −22%** in females) (after LD7: statistically significant decrease at the high dose only) Effects at doses higher than the LOAEL: At 8400/4200 ppm: Survival indices: decreased live birth index (87.6% vs. 100% in the controls, not statistically significant) resulting in a slight decreased litter size (10.3 vs. 11.4 in the controls, not statistically significant) Offspring: n.a.n.a. Remarks: Not considered appropriate to set NOAELs due to the study design (limited investigations, limited number of dams in each group, unusual treatment period). Toxicokinetic analysis: blood samples in GD17 dams, PND4 pups, LD11 dams, PND21 pups (in OECD TG 443, blood samples are recommended to be performed in fetuses during late pregnancy, in pups on PND10 (or milk) and rather post‐weaning (PND28) than PND21). Toxicokinetic analysis not complete (only Cmax available, ± AUC for some groups, not detailed, no graphs), results are available in Appendix B. Control samples were all contaminated (lower than values in treated animals but >> than concentrations in a water sample). Increase of systemic exposure (Cmax and/or AUC) is not proportional with increasing dose levels in all groups. Comments on the study design: Treatment period: F0 females from GD6 to LD21; F1 animals from mid lactation to Day 35 of age. Limited investigations: – Dams: clinical observations, body weight, food consumption, parturition observations (number of live and dead offspring, difficulties observed), gestation length – Littering phase: clinical observations, litter size, sex ratio, offspring body weight. – Necropsy: macroscopic examination, number of implantation sites, liver weight for F0 and F1 adults. – Blood sampling: analysis of concentration in plasma samples (proof of absorption) – no toxicokinetic analysis. Rat (Han Wistar) N: F0 generation: 25 F1A cohort: 20 F1B cohort: 20 Reliability: 1 (reliable without restriction) Multi‐generation reproductive toxicity (main 1‐extended) OECD TG 443 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: feed 0, 120, 600, 3000 ppm 0, 60, 300, 1500 ppm during lactation (females and pups from PND21 to PND35) F0 males: 9.71, 49.2, 248 mg/kg bw per day F0 females: 8.65, 44.3, 223 mg/kg bw per day (during gestation, i.e. lowest compared to premating or lactation, used for NOAEL/LOAEL setting) F1 males: 9.37, 47.3, 242 mg/kg bw per day F1 females: 9.83, 49.4, 248 mg/kg bw per day Parental toxicity at the LOAEL: At 600/300 ppm: Thyroid: decreased T4 levels (F1 males week 13, −32%*). Body weight parameters: decreased body weight gain (F0 males during week 1, statistically significant, around −10%). Blood chemistry: increased ALP M F0, increased A/G ratio M/F F0 and F F1. Stomach: gland dilatation F F0 and F1 (this effect was also observed in M F0 and F1 at the high dose, but in males, low and mid doses were not investigated – contrary to females for which all doses were investigated), erosion of glandular region F F0, congestion F F0, infiltrate of multinucleated cells in the mucosa of the glandular region F F0, decreased secretion of glandular region F F1 at the high dose. Effects at doses higher than the LOAEL: At 3000/1500 ppm: Thyroid: decreased T4 levels (F0 and F1, M and F, −23% to −52%*). Body weight parameters: decreased body weight gain (F0 F premating during wk1–12%*, F0 F gestation occasional statistically significant effects, F0 lactation −4.5%*, F1 adults −5/6%*) and body weight (F0 M −6%*, F0 F premating −4%*, F0 F gestation during GD7‐14 ‐17%*, F1 adults from PND21 to Day 71–7/−9%* at weaning on PND21 and −5%* in M on Day 71). Food consumption: decreased in F0 F lactation from LD4 (e.g. –24%* LD4‐7). Haematology: slight anaemia M/F F0 and M/F F1, decreased monocytes M F0 and F1, increased platelets M/F F0 and F F1, decreased lymphocytes M F1. Blood chemistry: increased ALP M F1, increased AST M F1 and increased ALT M/F F1, increased A/G ratio M F1, decreased glucose, decreased cholesterol M, decreased triglycerides M, decreased non‐esterified fatty acid, decreased bilirubin Liver: increased relative liver weight M/F F0 (+16%* in M and +14%* in F) and M/F F1A (+20%** and +16%**). Testes: decreased absolute weight in M F1A (−7.8%**) and F1B (−6.7%*). Epididymides: decreased absolute weight in M F1A (−7.7%*) and F1B (−6.3% non‐statistically significant). Lung: alveolar macrophage aggregation in M/F F1 (6/20 vs. 4/20 in controls in M, 9/20 vs. 5/20 in controls in F, low and mid doses not examined). Spleen: extramedullary haematopoiesis F F1 (3/20 vs. 1/20 in controls). Parental: 8.6544.3100 (and complementary reports #69, 70, 95) Offspring toxicity at the LOAEL: At 600/300 ppm Thyroid: decreased T4 levels (F1 males PND22; −17%*). Effects at doses higher than the LOAEL: At 3000/1500 ppm: Thyroid: decreased T4 levels (F1 males females PND22, −28/−37%*). Bw parameters: decreased bw from LD7 to LD21 M/F F1 (approx. ‐7 to −8%*) and decreased body weight gain from LD4 M/F F1 (LD1‐21: −7%*). Post‐implantation loss and litter size: decreased post‐implantation survival index (90.3%* vs. 95.6% in the controls) and subsequent slight decrease of litter size (10.9% vs. 11.6% in the controls, non‐statistically significant). Eyes: rosettes/folds in the retina F F1 (3/20 vs. 0/20 in controls). Offspring: 8.6544.3100 Reproductive toxicity at 3000/1500 ppm: Sperm analysis in F0: ‐ Decreased weight of cauda epididymis (−9%*), decreased total sperm (−18%*) and decreased sperm count per gram (−10%, non‐statistically significant) (low and mid dose not examined); – Decreased progressively motile sperm (42, 40, 38, 37% in the control, low, mid and high dose; non‐statistically significant), change in some sperm motion data: decreased rapid (non‐statistically significant), increased slow (non‐statistically significant), decreased straightness (statistically significant); ‐ Change in sperm morphology data: decreased normal sperm (95.8%* vs. 97.1% in the controls), increased abnormal sperm (4.2%* vs. 2.9% in the controls), increased some particular abnormalities (non‐statistically significant) (low and mid dose not examined). Sperm analysis in F1: – Decreased weight of testis (−7%*), decreased total sperm (−24%*) and spermatid count (−17%*) (low and mid dose not examined); ‐ Change in sperm morphology data: decreased normal sperm (94.3% (non‐statistically significant) vs. 96.4% in the controls), increased abnormal sperm (5.72% (non‐statistically significant) vs. 3.6% in the controls), increased some particular abnormalities (abnormal head 2.3%* vs. 1.1% in controls; flat head 2.1%* vs. 1.0% in controls; other increases are non‐statistically significant) (all doses examined). Reproductive: n.a.223100 Remarks: Heart: increased rodent progressive cardiomyopathy M F0 and F1 (7/25 vs. 3/25 in controls in F0, 3/20 vs. 1/20 in controls in F1, low and mid doses not examined). The WG noted that rodent progressive cardiomyopathy (PCM) is a well‐known spontaneous and progressive myocardial disease that occurs in rats and mice, with no counterpart in humans. In rodents it shows some variability in onset, incidence and severity depending on the strain as well as experimental conditions (feeding regimen), and concurrent pathologies (chronic progressive nephropathy) (Berridge et al., 2016). The WG considered that the lesion, observed not only in this study but in other studies (#8), it was present in concurrent controls, more prone in males and it was minor in severity and therefore considered as expression of the background condition and no treatment related. Rat (Han Wistar) N: F0 generation: 6 females (no F0 males since females are treated from GD6) F1 generation: 6/sex/group Multi‐generation reproductive toxicity (pilot 1‐extended) OECD TG: – GLP: no Sodium trifluoracetate (TFA Na) Purity (%): 99.9 Oral: feed 0, 1400, 3400, 8400 ppm 0, 700, 1700, 4200 ppm during lactation (LD1 to LD21) F0 females during gestation: 97.5, 230, 547 mg/kg bw per day F0 females during lactation: 110.2, 262, 578 mg/kg bw per day F1 females LD21‐LD34: 228, 557, 1369 mg/kg bw per day F1 males LD21‐LD34: 202, 476, 1290 mg/kg bw per day Maternal toxicity at the LOAEL: At 1400/700 ppm: Body weight and body weight gain: decreased body weight gain in F0 dams at the beginning of gestation (GD6‐10: −25%*, −33%*, −67%**) Liver: increased weight in F1 animals LD35 (+26%**, +28%**, +44%** in males and +25%**, +26%**, +33%** in females) Effects at doses higher than the LOAEL: At 3400/1700 ppm: Liver: increased weight in F0 dams (not statistically significant, 11%) Body weight and body weight gain: decreased body weight in F1 females from LD28 and decreased body weight gain in F1 females LD24‐28 (not clearly dose‐related) At 8400/4200 ppm: Body weight and body weight gain: decreased body weight gain in F0 dams during lactation (−9%*) Offspring toxicity at the LOAEL: At 1400/700 ppm: Body weight and body weight gain: decreased body weight gain of males and female pups at the beginning of lactation (LD4‐7: −19%*, −19%**, −26%* in males and −17%*, −12%**, −22%** in females) (after LD7: statistically significant decrease at the high dose only) Effects at doses higher than the LOAEL: At 8400/4200 ppm: Survival indices: decreased live birth index (87.6% vs. 100% in the controls, not statistically significant) resulting in a slight decreased litter size (10.3 vs. 11.4 in the controls, not statistically significant) Remarks: Not considered appropriate to set NOAELs due to the study design (limited investigations, limited number of dams in each group, unusual treatment period). Toxicokinetic analysis: blood samples in GD17 dams, PND4 pups, LD11 dams, PND21 pups (in OECD TG 443, blood samples are recommended to be performed in fetuses during late pregnancy, in pups on PND10 (or milk) and rather post‐weaning (PND28) than PND21). Toxicokinetic analysis not complete (only Cmax available, ± AUC for some groups, not detailed, no graphs), results are available in Appendix B. Control samples were all contaminated (lower than values in treated animals but >> than concentrations in a water sample). Increase of systemic exposure (Cmax and/or AUC) is not proportional with increasing dose levels in all groups. Comments on the study design: Treatment period: F0 females from GD6 to LD21; F1 animals from mid lactation to Day 35 of age. Limited investigations: – Dams: clinical observations, body weight, food consumption, parturition observations (number of live and dead offspring, difficulties observed), gestation length – Littering phase: clinical observations, litter size, sex ratio, offspring body weight. – Necropsy: macroscopic examination, number of implantation sites, liver weight for F0 and F1 adults. – Blood sampling: analysis of concentration in plasma samples (proof of absorption) – no toxicokinetic analysis. Rat (Han Wistar) N: F0 generation: 25 F1A cohort: 20 F1B cohort: 20 Reliability: 1 (reliable without restriction) Multi‐generation reproductive toxicity (main 1‐extended) OECD TG 443 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: feed 0, 120, 600, 3000 ppm 0, 60, 300, 1500 ppm during lactation (females and pups from PND21 to PND35) F0 males: 9.71, 49.2, 248 mg/kg bw per day F0 females: 8.65, 44.3, 223 mg/kg bw per day (during gestation, i.e. lowest compared to premating or lactation, used for NOAEL/LOAEL setting) F1 males: 9.37, 47.3, 242 mg/kg bw per day F1 females: 9.83, 49.4, 248 mg/kg bw per day Parental toxicity at the LOAEL: At 600/300 ppm: Thyroid: decreased T4 levels (F1 males week 13, −32%*). Body weight parameters: decreased body weight gain (F0 males during week 1, statistically significant, around −10%). Blood chemistry: increased ALP M F0, increased A/G ratio M/F F0 and F F1. Stomach: gland dilatation F F0 and F1 (this effect was also observed in M F0 and F1 at the high dose, but in males, low and mid doses were not investigated – contrary to females for which all doses were investigated), erosion of glandular region F F0, congestion F F0, infiltrate of multinucleated cells in the mucosa of the glandular region F F0, decreased secretion of glandular region F F1 at the high dose. Effects at doses higher than the LOAEL: At 3000/1500 ppm: Thyroid: decreased T4 levels (F0 and F1, M and F, −23% to −52%*). Body weight parameters: decreased body weight gain (F0 F premating during wk1–12%*, F0 F gestation occasional statistically significant effects, F0 lactation −4.5%*, F1 adults −5/6%*) and body weight (F0 M −6%*, F0 F premating −4%*, F0 F gestation during GD7‐14 ‐17%*, F1 adults from PND21 to Day 71–7/−9%* at weaning on PND21 and −5%* in M on Day 71). Food consumption: decreased in F0 F lactation from LD4 (e.g. –24%* LD4‐7). Haematology: slight anaemia M/F F0 and M/F F1, decreased monocytes M F0 and F1, increased platelets M/F F0 and F F1, decreased lymphocytes M F1. Blood chemistry: increased ALP M F1, increased AST M F1 and increased ALT M/F F1, increased A/G ratio M F1, decreased glucose, decreased cholesterol M, decreased triglycerides M, decreased non‐esterified fatty acid, decreased bilirubin Liver: increased relative liver weight M/F F0 (+16%* in M and +14%* in F) and M/F F1A (+20%** and +16%**). Testes: decreased absolute weight in M F1A (−7.8%**) and F1B (−6.7%*). Epididymides: decreased absolute weight in M F1A (−7.7%*) and F1B (−6.3% non‐statistically significant). Lung: alveolar macrophage aggregation in M/F F1 (6/20 vs. 4/20 in controls in M, 9/20 vs. 5/20 in controls in F, low and mid doses not examined). Spleen: extramedullary haematopoiesis F F1 (3/20 vs. 1/20 in controls). Offspring toxicity at the LOAEL: At 600/300 ppm Thyroid: decreased T4 levels (F1 males PND22; −17%*). Effects at doses higher than the LOAEL: At 3000/1500 ppm: Thyroid: decreased T4 levels (F1 males females PND22, −28/−37%*). Bw parameters: decreased bw from LD7 to LD21 M/F F1 (approx. ‐7 to −8%*) and decreased body weight gain from LD4 M/F F1 (LD1‐21: −7%*). Post‐implantation loss and litter size: decreased post‐implantation survival index (90.3%* vs. 95.6% in the controls) and subsequent slight decrease of litter size (10.9% vs. 11.6% in the controls, non‐statistically significant). Eyes: rosettes/folds in the retina F F1 (3/20 vs. 0/20 in controls). Reproductive toxicity at 3000/1500 ppm: Sperm analysis in F0: ‐ Decreased weight of cauda epididymis (−9%*), decreased total sperm (−18%*) and decreased sperm count per gram (−10%, non‐statistically significant) (low and mid dose not examined); – Decreased progressively motile sperm (42, 40, 38, 37% in the control, low, mid and high dose; non‐statistically significant), change in some sperm motion data: decreased rapid (non‐statistically significant), increased slow (non‐statistically significant), decreased straightness (statistically significant); ‐ Change in sperm morphology data: decreased normal sperm (95.8%* vs. 97.1% in the controls), increased abnormal sperm (4.2%* vs. 2.9% in the controls), increased some particular abnormalities (non‐statistically significant) (low and mid dose not examined). Sperm analysis in F1: – Decreased weight of testis (−7%*), decreased total sperm (−24%*) and spermatid count (−17%*) (low and mid dose not examined); ‐ Change in sperm morphology data: decreased normal sperm (94.3% (non‐statistically significant) vs. 96.4% in the controls), increased abnormal sperm (5.72% (non‐statistically significant) vs. 3.6% in the controls), increased some particular abnormalities (abnormal head 2.3%* vs. 1.1% in controls; flat head 2.1%* vs. 1.0% in controls; other increases are non‐statistically significant) (all doses examined). Remarks: Heart: increased rodent progressive cardiomyopathy M F0 and F1 (7/25 vs. 3/25 in controls in F0, 3/20 vs. 1/20 in controls in F1, low and mid doses not examined). The WG noted that rodent progressive cardiomyopathy (PCM) is a well‐known spontaneous and progressive myocardial disease that occurs in rats and mice, with no counterpart in humans. In rodents it shows some variability in onset, incidence and severity depending on the strain as well as experimental conditions (feeding regimen), and concurrent pathologies (chronic progressive nephropathy) (Berridge et al., 2016). The WG considered that the lesion, observed not only in this study but in other studies (#8), it was present in concurrent controls, more prone in males and it was minor in severity and therefore considered as expression of the background condition and no treatment related. Assessment of developmental toxicity studies Species n: number reliabilityStudy (method/type, length, route of exposure)Test substanceCritical effectsNOAEL (mg/kg bw per day)LOAEL (mg/kg bw per day)Reference Rat (Sprague Dawley (CD)) n: 22 Reliability: 1 (reliable without restriction) Developmental toxicity (main 14‐days) OECD TG 414 GLP: yes Trifluoroacetic acid (TFA) Purity (%): 99 Oral: gavage 0, 37.5, 75, 150 mg/kg bw per day 0, 37.5, 75, 150 mg/kg bw per day Maternal toxicity at the LOAEL: none. Effects at doses higher than the LOAEL: none. 150n.a.#7 Developmental toxicity Multiple malformations (e.g. umbilicus – omphalocele; hindlimbs – flexure, malrotated; abdomen – omphalocele; ribs – partially fused; one fetus with bilateral, folded retina) were observed in six fetuses out of one litter at 150 mg/kg bw per day, at lowest dose one fetus retroesophageal subclavian artery. Effects at doses higher than the LOAEL: none. 150n.a. Remarks: No mortality in dams, no clinical signs, no effect on body weight or food; organ weight increased in kidneys 5.6% and liver 9.6% at 150 mg/kg bw per day but considered not adverse. Caesarean section: no effects. The WG noted that higher doses could have been tested. Rat (Crl:CD(SD)) n: 8 Reliability: 2 (reliable with restrictions) Developmental toxicity (main 14‐days) OECD TG 414 GLP: no Trifluoroacetic acid (TFA) Purity (%): 99 Oral: gavage 0, 75, 150 mg/kg bw per day 0, 75, 150 mg/kg bw per day Maternal toxicity at the LOAEL: Decreased body weight gain and increased relative organ weights of the liver and kidneys. Effects at doses higher than the LOAEL: none. 75150#6 Developmental toxicity at the LOAEL: No treatment related fetal malformation was found in any dose groups. Effects at doses higher than the LOAEL: none. 150n.a. Remarks: Limited due to non‐GLP and to major deviations (e.g. number of dams, only 2 dose groups). Acceptable as supplementary only. Rat (Sprague–Dawley) N: unclear RoB: Tier 2/supplementary. Developmental toxicity (GD 10–20 (daily exposure)) No OECD TG. GLP: no Trifluoroacetic acid (TFA) Purity (%): 99.70 Oral: gavage 0, 75, 150 mg/kg bw per day Based on the presented data, no NOAEL could be derived for maternal toxicity because of slightly increased liver weight (4%) even at lowest dose level of 75 mg/kg bw per day. Liver is the known target organ, but histopathological examination is missing. Derivation of the NOAEL for developmental toxicity was not possible in this study (transient clinical chemistry of hepatic and renal effects even at lowest dose level, limited parameter, unclear number of fetuses examined). Maternal: n.a. Developmental: n.a. Maternal: 75 Developmental: n.a. #50 Rabbit (NZW) N: 7 Reliability: 2 (reliable with restrictions) Developmental toxicity (pilot) OECD TG: ‐ GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 250, 500, 1000/750 mg/kg bw per day Maternal toxicity at the LOAEL: Clinical signs even at lowest dose level: unsteady gait and decreased activity, but much lesser incidence. Physical examination revealed no signs at any dose level, no macroscopic findings. Placental weight not affected. Effects at doses higher than the LOAEL: At 1000 mg/kg bw per day: clinical signs: unsteady, underactive in‐home cages, impaired locomotion, rapid breathing, reduced body tone and flattened posture. Therefore, dose level reduced to 750 mg/kg bw per day. Maternal: n.a.250#87 Developmental toxicity at the LOAEL: Litter data: corpora lutea (n 10.0, 9.0, 9.9, 8.9) implantations (n 8.9, 7.3, 0.0, 7.6) total viable fetuses (n 8.1, 5.4, 7.7, 5.9), resorptions (n 0.7, 1.9, 1.3, 1.7), pre−/post‐implantation loss (%11.2, 19.4, 7.3, 13.9/7.1, 28.5, 14.5, 23.8 respectively) even at 250 mg/kg bw per day, litter/fetal weight but without clear dose response. One fetus with eye right dark area(s), both eyes retained. Effects at doses higher than the LOAEL Anomalies: Absence of accessory lung lobe, 2 incidences of cardiovascular defects at 500 mg/kg bw per day and above, at 750 mg/kg bw per day one fetus with eye right opaque retained, a second fetus with eye right haemorrhage(s), eye right small retained. Developmental: n.a.250 Remarks: Study considered supplementary: limited number of dams in each group, limited investigations, transient clinical signs even at lowest dose level. It was therefore not possible to set a maternal NOAEL. Based on findings even at the lowest dose level, no developmental NOAEL was set: limited investigations (e.g. no skeletal examination). Rabbit (NZW) N: 24 Reliability: 1 (reliable without restriction) Developmental toxicity OECD TG 414 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 180, 375, 750 mg/kg bw per day Maternal toxicity at the LOAEL: Minimal (to moderate at higher doses) generalised hepatocellular hypertrophy n = 0, 6, 8, 9 (no mechanistic data), no further changes (e.g. degeneration, necrosis). Minimal to moderate bile duct hyperplasia/fibrosis n = 0, 3, 6, 11 (no blood chemistry conducted), relative liver weight only 9% increased. Effects at doses higher than the LOAEL: At 375 mg/kg bw per day and above: transient clinical signs, slight bile duct hyperplasia/fibrosis, gravid uterine weight and bw changes. Maternal: n.a.180#88 (and complementary reports 111, 116) Developmental toxicity at the LOAEL: Eye abnormalities: at 180 mg/kg bw perday, one fetus had microphthalmia, with multiple folded retina and absent aqueous/vitreous humour (within HCD, but higher incidences were observed at higher doses). Multiple fold retina fetus n = 0, 1, 5, 9; litter n = 0, 1, 4, 8. Absent aqueous/vitreous humour fetus n = 0, 1, 6, 8; litter n = 0, 1, 4, 6. Effects at doses higher than the LOAEL: Eye abnormalities: at 375 and 750 mg/kg bw per day increased multiple fold retina and absent of aqueous/vitreous humour. Retina ruptured into surrounding tissue fetus n = 0, 0, 1, 1 Small/misshapen lens fetus n = 0, 0, 1, 1 Absent lens fetus n = 0, 0, 1, 1 At 375/750 mg/kg bw per day: low incidences (1 or 2) of major/minor abnormalities affecting heart and major blood vessels (transposition of ascending aorta/pulmonary trunk, narrow/dilated ascending aorta and pulmonary trunk, double outlet ventricle(s) and ventricular septal defects); multiple cervical/thoracic/lumbar/caudal vertebral and rib abnormalities; with associated minor abnormalities affecting the vertebrae, ribs, sternum and costal cartilages; kidneys: absent/displaced/fused considered not treatment related. Litter data: Corpora lutea (n 10.4, 9.6, 9.6, 8.3**), implantations (n 9.1, 8.5, 8.0, 7.2*), resorptions (n 0.8, 1.0, 0.8, 1.1), pre−/post‐implantation loss (% 12.6, 11.1, 17.8, 15.7/10.2, 10.1, 10.7, 14.9), live fetuses (8.3, 7.5, 7.2, 6.1**). At 375 mg/kg bw/d both total litter weight and male fetuses weight was significantly decreased, but for female fetuses and overall fetal weight at highest dose level. Developmental: n.a.180 Remarks: In contrast to the study report, the maternal NOAEL of 180 mg/kg bw per day is not supported, because even at lowest dose level histopathological effects in the liver were observed and no further data are available to disregard these effects as not adverse. No NOAEL for both maternal and developmental toxicity was derived due to adverse findings even at lowest dose level. The significant low number of corpora lutea resulted in lower numbers of implantations and live fetuses but were observed in a comparable range to the control group. Rabbit (NZW) N: 6 Reliability: 2 (reliable with restrictions) Other: (pilot 13 days) Other: GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 750 mg/kg bw per day Plasma concentrations of TFA were observed in all animals through 24 h post dose on day 1 and 13, but no accumulation of TFA after multiple doses (Cmax, AUC0‐24 were similar). Food consumption was low compared to pretreatment values and statistically significantly low to control values (p < 0.05/0.01); overall food consumption at 750 mg/kg bw per day was 59% of controls during the treatment period. Accordingly, body weight (gain) was lower at 750 mg/kg bw per day compared to the control females. No macroscopic findings were observed. On day 13, blood pH was slightly higher at 750 mg/kg bw per day. At 750 mg/kg bw per day blood glucose level was comparable to control animals, remained similar on day 13, but lower in control animals at 8 h post dosing. Lactate profile showed some difference between control and treatment animals, however, without any obvious trend. Statistically significant differences in blood chemistry parameters were observed: lower values for ALP, total bile acids and a higher total cholesterol value at 750 mg/kg bw per day. n.a.n.a.#101 Remarks: The objective was to evaluate toxicokinetic and clinical chemistry parameters, glucose, lactate, Vitamin A levels and blood pH in the non‐pregnant rabbit following oral administration of TFA Na. Based on the results, the study authors considered of no concern to use a dose of 750 mg/kg bw per day as high dose for the preliminary developmental toxicity study in rabbits. Fewer samples for Vitamin A investigation are available. Rabbit (NZW) N: 6 Reliability: 2 (reliable with restrictions) Developmental toxicity (pilot 23 days) OECD TG: ‐ GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 10, 45, 180, 750 mg/kg bw per day Maternal toxicity at the LOAEL: Clinical signs: abnormal gait, rapid breathing mainly during first 3 days. Decreased activity intermittently during treatment period. Faecal output was decreased, loose stool observed, bw gain decreased (69% of control, not significant). Lactate values increased on day 6 (2 and 3 h post‐dosing) but not on day 19 at 750 mg/kg bw per day. Blood chemistry revealed low bile acid concentration (−60%) and high plasma urea (+28%), cholesterol (+55%) and triglyceride concentration (+74%) on day 29. Macroscopic examination revealed no treatment related findings in dams. Accumulation of trifluoroacetic acid was observed at the 10 and 45 mg/kg bw per day dose levels after multiple doses of sodium trifluoroacetate in pregnant rabbits. Exposure was similar between pregnant female rabbits at 24 h post dose and fetal rabbits at necropsy on GD 29. Maternal: 180 750#102 Developmental toxicity: At 10 mg/kg bw per day and 180 mg/kg bw per day one female in each group was not pregnant. At 750 mg/kg bw per day two females were not pregnant, and a third female had an atypical unilateral pregnancy with a high pre‐implantation loss. The number of litters available for assessment from each group were 6, 5, 6, 5 and 3 at 0, 10, 45, 180 and 750 mg/kg bw per day, respectively. Developmental: n.a.n.a. Remarks: At 10 mg/kg bw per day and 180 mg/kg bw per day, one female in each group was not pregnant. At 750 mg/kg per day, two females were not pregnant, and a third female had an atypical unilateral pregnancy with a high pre‐implantation loss. The number of litters available for assessment from each group were only 6, 5, 6, 5 and 3 at 0, 10, 45, 180 and 750 mg/kg bw per day, respectively. Only limited number of litters and limited investigation (e.g. fetal: only macro pathology). Corpora lutea, implantations, total resorptions, implantation loss and number of live fetuses of the remaining litters were without any adverse effects. Fetal weight decreased at 10, 180 and 750 mg/kg bw per day, at highest dose < 10%. Rabbit (NZW) N: 30 for main; 8 for satellite Reliability: 1 (reliable without restriction) Developmental toxicity (main 23 days) OECD TG 414 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 30, 60,250, 750 mg/kg bw per day Maternal toxicity at the LOAEL: Absolute liver weight increased, dose response: 12%, 16%, 26%, 28% with histopathological correlation at 60 mg/kg bw per day (hypertrophy, hyperplasia bile duct) and above. Absolute kidney weight increased 17%, 15%, 27%, 31% with no histopathological correlation, transient decreased bw gain, overall, no affected bw gain. Effects at doses higher than the LOAEL: Clinical signs: At 250 mg/kg bw per day smaller faecal pellets. At 750 mg/kg bw per day: 2/30 rapid breathing, transient reduced faecal pellets, one female euthanized due to evidence of abortion, food consumption transient decrease. Clinical chemistry: Different changes in blood chemical parameters revealed some evidence of liver impairment (e.g. bilirubin and bile acids decreased, cholesterol and triglycerides increased). Changes in creatinine, sodium, potassium and chloride levels were considered to be indicative of changes in the kidney. Histopathology: Minimal to moderate fibrosis, portal/bile duct at 250 mg/kg bw per day (2/28) and 750 mg/kg bw per day (4/28). Organ weights: Gravid uterine weight decreased at 750 mg/kg bw per day, approximately by 82% compared to the control. Maternal: 3060#103 Developmental toxicity at the LOAEL: Litter data: Pre‐implantation loss increased without statistical significance (12.5%, 8.2%, 13.8%, 18.2%, 18.7%), post‐implantation loss (8.1%, 7.7%, 4.6%, 16.4%, 19.4%*). Fetal eye malformation (above HCD): Multiple folded retina: fetus n = 0, 0, 0, 2 (1%), 12 (7.4%**) litter n = 0, 0, 0, 2/28 (7.1%), 9/28 (32.1%**) Absent aqueous/vitreous humour: fetus n = 0, 0, 0, 1 (0.4%), 8 (4.9%**) litter n = 0, 0, 0, 1 (3.6%), 7 (25%**) Misshapen lens: fetus n = 0, 0, 0, 0, 2 (0.8) litter n = 0, 0, 0, 0, 2 (7.1) Haemorrhage(s) aqueous/vitreous humour: fetus n = 0, 0, 0, 0, 5 (3.7%*) litter n = 0, 0, 0, 0, 5 (17.9%*) Retroesophageal right subclavian artery: fetus n = 1 (0.4%), 0, 1 (0.5%), 5 (2.9%), 13 (8.5**) litter n = 1 (3.6), 0, 1 (3.6), 4 (14.3), 8 (28.6*) Effects at doses higher than the LOAEL: Fetal weight: at 750 mg/kg bw per day mean litter weight −22%**, mean overall fetal weight − 9%**. Ribs fused/partially fused: fetus n = 0, 1 (0.4), 0, 1 (0.4), 6 (10.7); litter n = 0, 1 (3.4), 0, 1 (3.6), 3 (10.7). Some further minor skeletal findings (e.g. retarded ossification) were observed. Developmental: 60 250 Remarks: Toxicokinetic: blood samples were collected from main phase and satellite phase animals at different time points. Satellite phase animals: lactate profile, vitamin A, blood chemistry, haematology (peripheral blood), blood pH, bile acid assessment.. Determination of lactate in rabbit maternal and fetal aqueous/vitreous humour (non‐GLP): maternal and fetal aqueous/humour revealed lactate levels comparable to controls. Fetal lactate levels on GD29 were approximately twofold the levels observed in maternal samples on GD20. Some further minor skeletal findings (e.g. retarded ossification) were observed at 750 mg/kg bw per day. Rabbit (Japanese White rabbits (Kbl:JW)) N: 22 mated Reliability: 2 (reliable with limitations) Developmental toxicity (main Developmental toxicity study) OECD TG 414 GLP: yes Trifluoroacetic acid (TFA) Purity (%): 100.1 TFA 99.3 TFA Na Oral: gavage 0, 1, 10 mg/kg bw per day TFA and 750 mg/kg bw per day TFA Na 0, 1, 10 mg/kg bw per day TFA and 750 mg/kg bw per day TFA Na Maternal toxicity at the LOAEL: Dams treated with 750 mg/kg bw per day TFA Na had decreased body weight gain of 17% compared to control animals. No effects were observed at lower doses. Effects at doses higher than the LOAEL: 750 mg/kg bw per day TFA Na was the highest dose tested. Maternal: 10 750#130 Developmental toxicity at the LOAEL: At 750 mg/kg bw per day TFA Na the incidence of fetuses/litters with fold retina, small vitreous chamber and small eyes was statistically significantly increased. No effects were observed at lower doses. Effects at doses higher than the LOAEL: 750 mg/kg bw per day TFA Na was the highest dose tested. Selection of doses and dose‐spacing is not considered appropriate. Developmental: 10750 Remarks: Mortality: No mortality was observed. Clinical signs: No clinical signs were observed. Pregnancy status: No treatment‐related effects on pregnancy status. Body weight: No effects were observed on adjusted and non‐adjusted body weight. Body weight gain: Dams at 750 mg/kg bw per day TFA Na had 17% lower body weight gain over the days 6–28 of gestation. This was especially in the first part of dosing (GD6‐21). Food consumption: dams at 750 mg/kg bw per day TFA Na had statistically significantly lower food consumption over the days 6–18 of gestation; at the end of the study, food consumption was comparable to negative control. Caesarean section: no statistically significant differences between the groups as regards number of corpora lutea, number of implantations, pre‐implantation loss, resorptions/foetal deaths and number of live fetuses. Fetal weight: no effects on fetal weight were observed. Sex ratio: no effects were observed. Variations: in controls, 1 TFA and 10 TFA, % of females with variations was 72.2%, 81.0%, 83.3%. There was a not statistically significant increase of thymic remnant in TFA/1 and TFA/10 groups (6.08% (10 fetuses in 4 litters) and 8.59% (11 fetuses in 6 litters)). In TFA Na/750 group, % females with variations were statistically significantly increased (100*%). Only thymic remnant in the neck was statistically significantly increased in TFA Na/750 group (12.42*% (20 fetuses in 10 litters) compared to 2.21% (3 fetuses in 3 litters) in controls). Malformations: in TFA/1 group, % females with malformations (total) was statistically significantly increased (33.3*% (7 females)). In TFA/10, % females with malformations (total) were 22.2%, (4 females) while it was only 5.6% (1 female) in control. In TFA Na/750 group, % females with malformations (total) were statistically significantly increased (85%** (17 females)). In the TFA‐treated groups (1 and 10 mg/kg bw per day), no specific malformation reached statistical significance. There were no incidences (0%) for fold retina and small vitreous chamber; incidences of small eyes were not dose‐related (0%–3.97% (2 fetuses in 2 litters)–1.96 (1 fetus in 1 litter)). Malformations in the TFA Na/750 fetuses were related to visceral observations: fold retina (27.74**% (21 fetuses in 12 litters) compared to 1.85% (1 fetus in 1 litter) in control), small vitreous chamber (26.36**% (21 fetus in 13 litters) compared to 1.85% (1 fetus in 1 litter) in control) and small eyes (22.83*% (16 fetuses in 10 litters) compared to 0% in control). Rabbit (New Zealand White) N: 3 Repeated dose toxicity: oral (pilot 14‐day) OECD TG: – GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 250, 500, 750, 1000 mg/kg bw per day (no control group) 250, 500, 750, 1000 mg/kg bw per day Female toxicity at the LOAEL: At all dose levels: Food consumption: lower during treatment period than during pre‐treatment. Effects at doses higher than the LOAEL: At 500, 750, 1000 mg/kg bw per day: Body weight: slight body weight loss during Days 1–15 (whereas a slight body weight gain was reported at 250 mg/kg bw per day). Female: n.a.n.a.#162 Remarks: Not considered appropriate to set NOAELs due to the study design (limited investigations (clinical signs, body weight, food consumption, macroscopic examination), limited number of females in each group, no control group, treatment started at different dates for each group therefore age and body weight of the animals in each group were not comparable). Rat (Sprague Dawley (CD)) n: 22 Reliability: 1 (reliable without restriction) Developmental toxicity (main 14‐days) OECD TG 414 GLP: yes Trifluoroacetic acid (TFA) Purity (%): 99 Oral: gavage 0, 37.5, 75, 150 mg/kg bw per day 0, 37.5, 75, 150 mg/kg bw per day Maternal toxicity at the LOAEL: none. Effects at doses higher than the LOAEL: none. Developmental toxicity Multiple malformations (e.g. umbilicus – omphalocele; hindlimbs – flexure, malrotated; abdomen – omphalocele; ribs – partially fused; one fetus with bilateral, folded retina) were observed in six fetuses out of one litter at 150 mg/kg bw per day, at lowest dose one fetus retroesophageal subclavian artery. Effects at doses higher than the LOAEL: none. Remarks: No mortality in dams, no clinical signs, no effect on body weight or food; organ weight increased in kidneys 5.6% and liver 9.6% at 150 mg/kg bw per day but considered not adverse. Caesarean section: no effects. The WG noted that higher doses could have been tested. Rat (Crl:CD(SD)) n: 8 Reliability: 2 (reliable with restrictions) Developmental toxicity (main 14‐days) OECD TG 414 GLP: no Trifluoroacetic acid (TFA) Purity (%): 99 Oral: gavage 0, 75, 150 mg/kg bw per day 0, 75, 150 mg/kg bw per day Maternal toxicity at the LOAEL: Decreased body weight gain and increased relative organ weights of the liver and kidneys. Effects at doses higher than the LOAEL: none. Developmental toxicity at the LOAEL: No treatment related fetal malformation was found in any dose groups. Effects at doses higher than the LOAEL: none. Remarks: Limited due to non‐GLP and to major deviations (e.g. number of dams, only 2 dose groups). Acceptable as supplementary only. Rat (Sprague–Dawley) N: unclear RoB: Tier 2/supplementary. Developmental toxicity (GD 10–20 (daily exposure)) No OECD TG. GLP: no Trifluoroacetic acid (TFA) Purity (%): 99.70 Oral: gavage 0, 75, 150 mg/kg bw per day Maternal: n.a. Developmental: n.a. Maternal: 75 Developmental: n.a. Rabbit (NZW) N: 7 Reliability: 2 (reliable with restrictions) Developmental toxicity (pilot) OECD TG: ‐ GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 250, 500, 1000/750 mg/kg bw per day Maternal toxicity at the LOAEL: Clinical signs even at lowest dose level: unsteady gait and decreased activity, but much lesser incidence. Physical examination revealed no signs at any dose level, no macroscopic findings. Placental weight not affected. Effects at doses higher than the LOAEL: At 1000 mg/kg bw per day: clinical signs: unsteady, underactive in‐home cages, impaired locomotion, rapid breathing, reduced body tone and flattened posture. Therefore, dose level reduced to 750 mg/kg bw per day. Developmental toxicity at the LOAEL: Litter data: corpora lutea (n 10.0, 9.0, 9.9, 8.9) implantations (n 8.9, 7.3, 0.0, 7.6) total viable fetuses (n 8.1, 5.4, 7.7, 5.9), resorptions (n 0.7, 1.9, 1.3, 1.7), pre−/post‐implantation loss (%11.2, 19.4, 7.3, 13.9/7.1, 28.5, 14.5, 23.8 respectively) even at 250 mg/kg bw per day, litter/fetal weight but without clear dose response. One fetus with eye right dark area(s), both eyes retained. Effects at doses higher than the LOAEL Anomalies: Absence of accessory lung lobe, 2 incidences of cardiovascular defects at 500 mg/kg bw per day and above, at 750 mg/kg bw per day one fetus with eye right opaque retained, a second fetus with eye right haemorrhage(s), eye right small retained. Remarks: Study considered supplementary: limited number of dams in each group, limited investigations, transient clinical signs even at lowest dose level. It was therefore not possible to set a maternal NOAEL. Based on findings even at the lowest dose level, no developmental NOAEL was set: limited investigations (e.g. no skeletal examination). Rabbit (NZW) N: 24 Reliability: 1 (reliable without restriction) Developmental toxicity OECD TG 414 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 180, 375, 750 mg/kg bw per day Maternal toxicity at the LOAEL: Minimal (to moderate at higher doses) generalised hepatocellular hypertrophy n = 0, 6, 8, 9 (no mechanistic data), no further changes (e.g. degeneration, necrosis). Minimal to moderate bile duct hyperplasia/fibrosis n = 0, 3, 6, 11 (no blood chemistry conducted), relative liver weight only 9% increased. Effects at doses higher than the LOAEL: At 375 mg/kg bw per day and above: transient clinical signs, slight bile duct hyperplasia/fibrosis, gravid uterine weight and bw changes. Developmental toxicity at the LOAEL: Eye abnormalities: at 180 mg/kg bw perday, one fetus had microphthalmia, with multiple folded retina and absent aqueous/vitreous humour (within HCD, but higher incidences were observed at higher doses). Multiple fold retina fetus n = 0, 1, 5, 9; litter n = 0, 1, 4, 8. Absent aqueous/vitreous humour fetus n = 0, 1, 6, 8; litter n = 0, 1, 4, 6. Effects at doses higher than the LOAEL: Eye abnormalities: at 375 and 750 mg/kg bw per day increased multiple fold retina and absent of aqueous/vitreous humour. Retina ruptured into surrounding tissue fetus n = 0, 0, 1, 1 Small/misshapen lens fetus n = 0, 0, 1, 1 Absent lens fetus n = 0, 0, 1, 1 At 375/750 mg/kg bw per day: low incidences (1 or 2) of major/minor abnormalities affecting heart and major blood vessels (transposition of ascending aorta/pulmonary trunk, narrow/dilated ascending aorta and pulmonary trunk, double outlet ventricle(s) and ventricular septal defects); multiple cervical/thoracic/lumbar/caudal vertebral and rib abnormalities; with associated minor abnormalities affecting the vertebrae, ribs, sternum and costal cartilages; kidneys: absent/displaced/fused considered not treatment related. Litter data: Corpora lutea (n 10.4, 9.6, 9.6, 8.3**), implantations (n 9.1, 8.5, 8.0, 7.2*), resorptions (n 0.8, 1.0, 0.8, 1.1), pre−/post‐implantation loss (% 12.6, 11.1, 17.8, 15.7/10.2, 10.1, 10.7, 14.9), live fetuses (8.3, 7.5, 7.2, 6.1**). At 375 mg/kg bw/d both total litter weight and male fetuses weight was significantly decreased, but for female fetuses and overall fetal weight at highest dose level. Remarks: In contrast to the study report, the maternal NOAEL of 180 mg/kg bw per day is not supported, because even at lowest dose level histopathological effects in the liver were observed and no further data are available to disregard these effects as not adverse. No NOAEL for both maternal and developmental toxicity was derived due to adverse findings even at lowest dose level. The significant low number of corpora lutea resulted in lower numbers of implantations and live fetuses but were observed in a comparable range to the control group. Rabbit (NZW) N: 6 Reliability: 2 (reliable with restrictions) Other: (pilot 13 days) Other: GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 750 mg/kg bw per day Plasma concentrations of TFA were observed in all animals through 24 h post dose on day 1 and 13, but no accumulation of TFA after multiple doses (Cmax, AUC0‐24 were similar). Food consumption was low compared to pretreatment values and statistically significantly low to control values (p < 0.05/0.01); overall food consumption at 750 mg/kg bw per day was 59% of controls during the treatment period. Accordingly, body weight (gain) was lower at 750 mg/kg bw per day compared to the control females. No macroscopic findings were observed. On day 13, blood pH was slightly higher at 750 mg/kg bw per day. At 750 mg/kg bw per day blood glucose level was comparable to control animals, remained similar on day 13, but lower in control animals at 8 h post dosing. Lactate profile showed some difference between control and treatment animals, however, without any obvious trend. Statistically significant differences in blood chemistry parameters were observed: lower values for ALP, total bile acids and a higher total cholesterol value at 750 mg/kg bw per day. Remarks: The objective was to evaluate toxicokinetic and clinical chemistry parameters, glucose, lactate, Vitamin A levels and blood pH in the non‐pregnant rabbit following oral administration of TFA Na. Based on the results, the study authors considered of no concern to use a dose of 750 mg/kg bw per day as high dose for the preliminary developmental toxicity study in rabbits. Fewer samples for Vitamin A investigation are available. Rabbit (NZW) N: 6 Reliability: 2 (reliable with restrictions) Developmental toxicity (pilot 23 days) OECD TG: ‐ GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 10, 45, 180, 750 mg/kg bw per day Maternal toxicity at the LOAEL: Clinical signs: abnormal gait, rapid breathing mainly during first 3 days. Decreased activity intermittently during treatment period. Faecal output was decreased, loose stool observed, bw gain decreased (69% of control, not significant). Lactate values increased on day 6 (2 and 3 h post‐dosing) but not on day 19 at 750 mg/kg bw per day. Blood chemistry revealed low bile acid concentration (−60%) and high plasma urea (+28%), cholesterol (+55%) and triglyceride concentration (+74%) on day 29. Macroscopic examination revealed no treatment related findings in dams. Accumulation of trifluoroacetic acid was observed at the 10 and 45 mg/kg bw per day dose levels after multiple doses of sodium trifluoroacetate in pregnant rabbits. Exposure was similar between pregnant female rabbits at 24 h post dose and fetal rabbits at necropsy on GD 29. Maternal: 180 Developmental toxicity: At 10 mg/kg bw per day and 180 mg/kg bw per day one female in each group was not pregnant. At 750 mg/kg bw per day two females were not pregnant, and a third female had an atypical unilateral pregnancy with a high pre‐implantation loss. The number of litters available for assessment from each group were 6, 5, 6, 5 and 3 at 0, 10, 45, 180 and 750 mg/kg bw per day, respectively. Remarks: At 10 mg/kg bw per day and 180 mg/kg bw per day, one female in each group was not pregnant. At 750 mg/kg per day, two females were not pregnant, and a third female had an atypical unilateral pregnancy with a high pre‐implantation loss. The number of litters available for assessment from each group were only 6, 5, 6, 5 and 3 at 0, 10, 45, 180 and 750 mg/kg bw per day, respectively. Only limited number of litters and limited investigation (e.g. fetal: only macro pathology). Corpora lutea, implantations, total resorptions, implantation loss and number of live fetuses of the remaining litters were without any adverse effects. Fetal weight decreased at 10, 180 and 750 mg/kg bw per day, at highest dose < 10%. Rabbit (NZW) N: 30 for main; 8 for satellite Reliability: 1 (reliable without restriction) Developmental toxicity (main 23 days) OECD TG 414 GLP: yes Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 0, 30, 60,250, 750 mg/kg bw per day Maternal toxicity at the LOAEL: Absolute liver weight increased, dose response: 12%, 16%, 26%, 28% with histopathological correlation at 60 mg/kg bw per day (hypertrophy, hyperplasia bile duct) and above. Absolute kidney weight increased 17%, 15%, 27%, 31% with no histopathological correlation, transient decreased bw gain, overall, no affected bw gain. Effects at doses higher than the LOAEL: Clinical signs: At 250 mg/kg bw per day smaller faecal pellets. At 750 mg/kg bw per day: 2/30 rapid breathing, transient reduced faecal pellets, one female euthanized due to evidence of abortion, food consumption transient decrease. Clinical chemistry: Different changes in blood chemical parameters revealed some evidence of liver impairment (e.g. bilirubin and bile acids decreased, cholesterol and triglycerides increased). Changes in creatinine, sodium, potassium and chloride levels were considered to be indicative of changes in the kidney. Histopathology: Minimal to moderate fibrosis, portal/bile duct at 250 mg/kg bw per day (2/28) and 750 mg/kg bw per day (4/28). Organ weights: Gravid uterine weight decreased at 750 mg/kg bw per day, approximately by 82% compared to the control. Developmental toxicity at the LOAEL: Litter data: Pre‐implantation loss increased without statistical significance (12.5%, 8.2%, 13.8%, 18.2%, 18.7%), post‐implantation loss (8.1%, 7.7%, 4.6%, 16.4%, 19.4%*). Fetal eye malformation (above HCD): Multiple folded retina: fetus n = 0, 0, 0, 2 (1%), 12 (7.4%**) litter n = 0, 0, 0, 2/28 (7.1%), 9/28 (32.1%**) Absent aqueous/vitreous humour: fetus n = 0, 0, 0, 1 (0.4%), 8 (4.9%**) litter n = 0, 0, 0, 1 (3.6%), 7 (25%**) Misshapen lens: fetus n = 0, 0, 0, 0, 2 (0.8) litter n = 0, 0, 0, 0, 2 (7.1) Haemorrhage(s) aqueous/vitreous humour: fetus n = 0, 0, 0, 0, 5 (3.7%*) litter n = 0, 0, 0, 0, 5 (17.9%*) Retroesophageal right subclavian artery: fetus n = 1 (0.4%), 0, 1 (0.5%), 5 (2.9%), 13 (8.5**) litter n = 1 (3.6), 0, 1 (3.6), 4 (14.3), 8 (28.6*) Effects at doses higher than the LOAEL: Fetal weight: at 750 mg/kg bw per day mean litter weight −22%**, mean overall fetal weight − 9%**. Ribs fused/partially fused: fetus n = 0, 1 (0.4), 0, 1 (0.4), 6 (10.7); litter n = 0, 1 (3.4), 0, 1 (3.6), 3 (10.7). Some further minor skeletal findings (e.g. retarded ossification) were observed. Developmental: 60 Remarks: Toxicokinetic: blood samples were collected from main phase and satellite phase animals at different time points. Satellite phase animals: lactate profile, vitamin A, blood chemistry, haematology (peripheral blood), blood pH, bile acid assessment.. Determination of lactate in rabbit maternal and fetal aqueous/vitreous humour (non‐GLP): maternal and fetal aqueous/humour revealed lactate levels comparable to controls. Fetal lactate levels on GD29 were approximately twofold the levels observed in maternal samples on GD20. Some further minor skeletal findings (e.g. retarded ossification) were observed at 750 mg/kg bw per day. Rabbit (Japanese White rabbits (Kbl:JW)) N: 22 mated Reliability: 2 (reliable with limitations) Developmental toxicity (main Developmental toxicity study) OECD TG 414 GLP: yes Trifluoroacetic acid (TFA) Purity (%): 100.1 TFA 99.3 TFA Na Oral: gavage 0, 1, 10 mg/kg bw per day TFA and 750 mg/kg bw per day TFA Na 0, 1, 10 mg/kg bw per day TFA and 750 mg/kg bw per day TFA Na Maternal toxicity at the LOAEL: Dams treated with 750 mg/kg bw per day TFA Na had decreased body weight gain of 17% compared to control animals. No effects were observed at lower doses. Effects at doses higher than the LOAEL: 750 mg/kg bw per day TFA Na was the highest dose tested. Maternal: 10 Developmental toxicity at the LOAEL: At 750 mg/kg bw per day TFA Na the incidence of fetuses/litters with fold retina, small vitreous chamber and small eyes was statistically significantly increased. No effects were observed at lower doses. Effects at doses higher than the LOAEL: 750 mg/kg bw per day TFA Na was the highest dose tested. Selection of doses and dose‐spacing is not considered appropriate. Remarks: Mortality: No mortality was observed. Clinical signs: No clinical signs were observed. Pregnancy status: No treatment‐related effects on pregnancy status. Body weight: No effects were observed on adjusted and non‐adjusted body weight. Body weight gain: Dams at 750 mg/kg bw per day TFA Na had 17% lower body weight gain over the days 6–28 of gestation. This was especially in the first part of dosing (GD6‐21). Food consumption: dams at 750 mg/kg bw per day TFA Na had statistically significantly lower food consumption over the days 6–18 of gestation; at the end of the study, food consumption was comparable to negative control. Caesarean section: no statistically significant differences between the groups as regards number of corpora lutea, number of implantations, pre‐implantation loss, resorptions/foetal deaths and number of live fetuses. Fetal weight: no effects on fetal weight were observed. Sex ratio: no effects were observed. Variations: in controls, 1 TFA and 10 TFA, % of females with variations was 72.2%, 81.0%, 83.3%. There was a not statistically significant increase of thymic remnant in TFA/1 and TFA/10 groups (6.08% (10 fetuses in 4 litters) and 8.59% (11 fetuses in 6 litters)). In TFA Na/750 group, % females with variations were statistically significantly increased (100*%). Only thymic remnant in the neck was statistically significantly increased in TFA Na/750 group (12.42*% (20 fetuses in 10 litters) compared to 2.21% (3 fetuses in 3 litters) in controls). Malformations: in TFA/1 group, % females with malformations (total) was statistically significantly increased (33.3*% (7 females)). In TFA/10, % females with malformations (total) were 22.2%, (4 females) while it was only 5.6% (1 female) in control. In TFA Na/750 group, % females with malformations (total) were statistically significantly increased (85%** (17 females)). In the TFA‐treated groups (1 and 10 mg/kg bw per day), no specific malformation reached statistical significance. There were no incidences (0%) for fold retina and small vitreous chamber; incidences of small eyes were not dose‐related (0%–3.97% (2 fetuses in 2 litters)–1.96 (1 fetus in 1 litter)). Malformations in the TFA Na/750 fetuses were related to visceral observations: fold retina (27.74**% (21 fetuses in 12 litters) compared to 1.85% (1 fetus in 1 litter) in control), small vitreous chamber (26.36**% (21 fetus in 13 litters) compared to 1.85% (1 fetus in 1 litter) in control) and small eyes (22.83*% (16 fetuses in 10 litters) compared to 0% in control). Rabbit (New Zealand White) N: 3 Repeated dose toxicity: oral (pilot 14‐day) OECD TG: – GLP: no Sodium trifluoroacetate (TFA Na) Purity (%): 99.9 Oral: gavage 250, 500, 750, 1000 mg/kg bw per day (no control group) 250, 500, 750, 1000 mg/kg bw per day Female toxicity at the LOAEL: At all dose levels: Food consumption: lower during treatment period than during pre‐treatment. Effects at doses higher than the LOAEL: At 500, 750, 1000 mg/kg bw per day: Body weight: slight body weight loss during Days 1–15 (whereas a slight body weight gain was reported at 250 mg/kg bw per day). Remarks: Not considered appropriate to set NOAELs due to the study design (limited investigations (clinical signs, body weight, food consumption, macroscopic examination), limited number of females in each group, no control group, treatment started at different dates for each group therefore age and body weight of the animals in each group were not comparable). List of Endpoints Impact on Human and Animal Health Absorption, distribution, metabolism and excretion (toxicokinetics) Acute toxicity Short‐term toxicity Genotoxicity Reproductive toxicity Reproduction toxicity (EORGTS) Developmental toxicity Summary Assessment of thyroid‐mediated endocrine disruption Advice from the EFSA Working group on Endocrine Disruptors – April 2025 – 27th Working Group meeting A dose‐range finding and an extended one‐generation toxicity study (EOGRT, with no inclusion of a second generation) were made available to the EFSA Working group on endocrine disruptors (ED). The EFSA Working group on TFA sought experts' opinion on the following questions Have thyroid hormones been properly analysed in the EOGRT study? Have thyroid hormones been properly analysed in the EOGRT study? Should adversity be set based on thyroid hormone effects observed in the EOGRT study? If so, at which dose level (NOAEL/LOAEL)? Should adversity be set based on thyroid hormone effects observed in the EOGRT study? If so, at which dose level (NOAEL/LOAEL)? Should the effects observed on thyroid hormones lead to a concern for DNT in the sensitive population or to conduct additional investigations? Should the effects observed on thyroid hormones lead to a concern for DNT in the sensitive population or to conduct additional investigations? The questions were discussed at the 27th Working Group meeting in April 2025. 1. Have thyroid hormones been properly analysed in the EOGRT study? Experts agreed that T4 was properly analysed in the EOGRT. The experts observed that the historical control data (HCD) correctly show the proficiency of the laboratory to perform the analysis of T4 as the values for the concurrent controls are within the min–max range of the HCD provided. 27 T3 was not reported even though it should have been detected in the employed mass spectrometry method (it is, however, acknowledged that T3 measurements are not required in the OECD TG 443). For TSH measurements, the experts highlighted that a number of samples were reported to be below the limit of quantification (LOQ). This occurred in samples for both adult and post‐weaning animals and suggests a problem with the assays, as the employed method has been commonly used by this testing lab, and there is no description of ‘below the LOQ’ in any of the 15 studies reported in the HCD. 2. Should adversity be set based on thyroid hormone effects observed in the EOGRT study? If so, at which dose level (NOAEL/LOAEL)? The experts considered that the effects observed on T4 at mid and top doses are adverse in both offspring (PND 22) and adult ages. The effects were above the threshold of concern (20%; US EPA, 2011, 28 , and EFSA Scientific Committee, 2025, 29 ), were statistically significant and showed a dose–response relationship and consistency in direction. At low doses, a statistically significant decrease of T4 (−15%) in F1 adult male animals was observed. The experts considered that this effect is not adverse because it is below the threshold of concern. Additionally, no change was observed in F0 and in F1 PND22 at the same dose level. A statistically significant increase in thyroid weight, relative to body weight, was observed at low dose in F1 male cohort 1a; however, no effects were reported in absolute thyroid weight and thyroid histopathology. The ED WG concluded that the NOAEL for thyroid effects should be set at low dose for both offspring and adult ages based on T4 changes observed at mid dose. 3. Should the effects observed on thyroid hormones lead to a concern for DNT in the sensitive population or to conduct additional investigations? The experts agreed that a concern for DNT cannot be dismissed. However, the agreed NOAEL based on changes in T4 is considered protective of DNT effects, and thus, additional investigations are not considered necessary. Advice from the EFSA Working group on Endocrine Disruptors – January 2026 – 33rd Working Group meeting After public consultation, the EFSA Working group on TFA sought experts' opinion on the following questions from the EFSA Working group on Endocrine Disruptors Could the WG confirm that the threshold of concern (20%; US EPA, 2011, 30 and EFSA SC, 2025, 31 ) is applicable to both adult and offspring? Could the WG confirm that the threshold of concern (20%; US EPA, 2011, 30 and EFSA SC, 2025, 31 ) is applicable to both adult and offspring? Could the WG confirm that the agreed NOAEL based on changes in T4 in offspring is considered protective of DNT effects and an additional UF should not apply to setting of health‐based guidance values? Could the WG confirm that the agreed NOAEL based on changes in T4 in offspring is considered protective of DNT effects and an additional UF should not apply to setting of health‐based guidance values? Could the WG advice whether changes in T4 in offspring could be of relevance for setting an ARfD having in mind the EFSA SC (2025)2? Could the WG advice whether changes in T4 in offspring could be of relevance for setting an ARfD having in mind the EFSA SC (2025)2? Is there any correlation between a decrease in thyroid hormones (in particular, decrease in T4) and immunotoxicity? Is there any correlation between a decrease in thyroid hormones (in particular, decrease in T4) and immunotoxicity? 1. Could the WG confirm that the threshold of concern (20%; US EPA, 2011 1 , and EFSA SC, 2025 2 ) is applicable to both adult and offspring? Yes, the WG confirms that the 20% threshold of concern is applicable to both adults and offspring. This threshold reflects the current state of scientific knowledge and can be applied across populations (adults and offspring) and chemicals. This threshold is considered biologically relevant and sufficiently conservative. While offspring are generally considered a more sensitive population, for TFA the most sensitive group identified was F1 adult males (in terms of T4 drop). 2. Could the WG confirm that the agreed NOAEL based on changes in T4 in offspring is considered protective of DNT effects and an additional UF should not apply to setting of health‐based guidance values? If a study would be conducted, the WG considered that a comparative thyroid assessment (CTA) study would be more appropriate than a DNT study to provide information on T4, TSH and thyroid histopathology in the relevant target population. DNT endpoints related to HPT axis disruption have been characterised for prototypic thyroid‐disrupting chemicals (e.g. ectopic neurons in corpus callosum, molecular changes in hippocampus and cortex, deficits in learning and memory). Although relevant endpoints, not all are included in the OECD TG 426 and further characterisation of sensitivity is required. In contrast, the CTA study is intended to evaluate standard thyroid‐related endpoints in all target populations, i.e. dams, fetus and newborns (which are also key events in several thyroid adverse outcome pathways) and is therefore fit for purpose in the risk assessment process. In this case for TFA, the NOAEL (or BMDL20) is considered sufficiently conservative, although uncertainties remain regarding earlier developmental time points (PND 4, fetus and dams) not investigated for disruption of the HPT axis. The WG acknowledged that further investigation could have been conducted. However, considering: the limited magnitude of the observed effect (LOAEL), the limited magnitude of the observed effect (LOAEL), the effect observed in only males at the LOAEL, the effect observed in only males at the LOAEL, Thyroid hormone changes in the rat, being qualitatively similar but quantitatively more sensitive than humans, Thyroid hormone changes in the rat, being qualitatively similar but quantitatively more sensitive than humans, the absence of histopathological changes in the thyroid at doses where concomitant statistically significant changes occur in T4, the absence of histopathological changes in the thyroid at doses where concomitant statistically significant changes occur in T4, The fact that earlier developmental time points (PND 4, fetus and dams) investigations are not mandatory according to OECD TG 443, The fact that earlier developmental time points (PND 4, fetus and dams) investigations are not mandatory according to OECD TG 443, the application of the standard uncertainty factor of 100. the application of the standard uncertainty factor of 100. The F1 offspring data at PND22 are considered sufficiently protective of any potential DNT effects. Any remaining uncertainty (i.e. absence of hormone data from early developmental stages, i.e. PND4, fetus and dams) would be covered by the application of the standard UF of 100 as a conservative default approach. The WG concluded that any additional UF (e.g. 10) regarding the lack of DNT study is not necessary. 3. Could the WG advise whether changes in T4 in offspring could be of relevance for setting an ARfD having in mind the EFSA SC, 2025 ? Yes, the WG considered that the observed changes in thyroid hormones (decreased T4 levels) in the EOGRTS can trigger the setting of an ARfD based on the NOAEL of 8.65. mg/kg bw per day, considering the fact that there is not enough knowledge on the critical timing or duration of exposure that would lead to the eventual (irreversible) effect during brain development. Based on existing knowledge of the mechanisms for modulation of circulating T4 levels, setting the ARfD based on the EOGRTS NOAEL is considered conservative. 4. Is there any correlation between a decrease in thyroid hormones (in particular, decrease in T4) and immunotoxicity? According to the current state of science, there is insufficient data to conclude on a correlation between decreased thyroid hormone levels and immunotoxicity. The available evidence for TFA does not allow any conclusions regarding the effects observed on T cells in the spleen and the noted decreases in circulating T4. Rationale supporting the 20% decrease in serum T4 levels for assessing potential perturbation of hypothalamic–pituitary–thyroid (HPT) axis After the public consultation, the EFSA WG on TFA organised a Technical Hearing with experts from EU Member States together with an expert from the EFSA WG Endocrine Disruption (ED). Upon Member States' request, the WG TFA further elaborated on the rationale behind the 20% decreased in total T4. A decrease of approximately 20% in total T4 is used as a pragmatic benchmark in the assessment of substances (including contaminants) as part of the weight‐of‐evidence (WoE) approach to assess potential disruption of the HPT axis. This value is informed by regulatory experience in the EU (EFSA ED Working Group) and the US (US EPA, 2011, 32 ). The approach is underpinned by the adverse outcome pathway for thyroid disruption, using propylthiouracil (PTU) as a reference compound proposed for the quantification of the AOP ‘Inhibition of thyroid peroxidase leading to cortical heterotopia’ (Hassan et al., 2017; see also AOP402 ‘Thyroid peroxidase (TPO) inhibition leads to periventricular heterotopia formation in the developing rat brain’ 33 ). This AOP is using morphological evidence (e.g. heterotopia indicating permanent neuronal damage, characterised by an abnormal cluster of neurons within the normal structure of the corpus callosum) as AO. TH were measured in dams and fetus serum and fetal brain levels of T4 were estimated from published literature, modelled and eventually linked to the increased in size of the heterotopia in the brain of offspring. The empirical modelling is therefore based on kinetic data in dams and fetus which are interlinked as in the rodent, fetal thyroid is not functional until GD 17 and at birth 17% of fetal T4 derives from the mother. The process is similar in humans where this estimate ranges from 30% to 35%. Correlation analyses between serum T4 levels and heterotopia volume indicate that late gestational neuronal development is sensitive to thyroid hormone fluctuations and the developed model incorporates a mathematical relationship between fetal brain T4 and heterotopia volume. Reduction of 35% in fetal serum T4 induces a clear effect on cortical heterotopia and approximately 20% drop in dam serum T4 at GD 20 was associated with a minimal but statistically significant volume of heterotopia in excess of that seen in control. PTU is a very well‐characterised thyroid‐disrupting chemical by thyroid peroxidase inhibition but is also able to inhibit iodothyronine deiodinase (DIO) 1; though the most relevant deiodinase in the brain is DIO 2, which is not affected by PTU. While PTU shows heterotopia at relevant levels of T4 reduction, it is acknowledged that other substances may induce larger decreases (e.g. 35%) without leading to heterotopia. Therefore, the 20% threshold in rats is considered sufficiently conservative approach developed using a well‐characterised compound. The approach is taking into account several uncertainties: it is a simple model for complex biology, placental transfer exists for the chemical under investigation and thyroid hormones, homeostatic feedback mechanisms in both dams and in the fetus, transport and local regulatory control of TH, partitioning/transport of TH from foetal serum to brain, complexity of TH metabolism in the brain and transporter changes over development time, species differences in TH economy and differential timelines over which brain develop. The weight of evidence (WoE), described in G.2, was considered for assessing the regulatory relevance for TFA. The WoE supports using a 20% decrease in T4 as a conservative and pragmatic benchmark for TFA. The derived reference point is considered adequately protective for potential developmental neurotoxicity. Reasons for excluding studies from body of evidence according to the eligibility criteria Specific eligibility criteria for selecting studies and/or relevant information were defined by the WG and available in the detailed protocol in Annex 1. The reasons for excluding studies from the body of evidence according to the eligibility criteria are indicated in Table H.1. In total, 84 studies were excluded from the assessment: position paper or review/third‐party assessment (n = 29), conference proceedings (n = 1), full text not in English (n = 4), test item: no TFA/TFA Na (n = 33), in vivo study conducted through a route of administration other than oral (n = 15) and studies considered to fall outside of the scope (environmental fate and behaviour and lipolysis assessments) (n = 2). In addition, the original reports of some of the studies listed in Annex 2 were not available to EFSA (see column K in ‘List of studied merged’ tab) and no further considered in the scientific report. TABLE H.1Studies excluded from the body of evidence of TFA according to the eligibility criteria.RefIDAuthorsYearTitleData submitters (targeted call for data)Reasons of exclusion 13 Byun J.Y. et al.2014Occupational airborne contact dermatitis caused by trifluoroacetic acid in an organic chemistry laboratory.EFSA, DEIn vivo study conducted through a route of administration other than oral 14 Dahlin, J. et al.2013Chemical burns caused by trifluoroacetic acid.EFSA, DE, TFA task forceIn vivo study conducted through a route of administration other than oral 17 Confidential2014Flufenacet ‐ Toxicological profile and exposure assessment of the plant metabolitesEFSA, DEPosition paper or review/third‐party assessment 18 Confidential2014Trifluoroacetate (TFA) ‐ Waiver of an acute reference dose (ARfD)EFSA, HR, DEPosition paper or review/third‐party assessment 19 Confidential2012Toxicological relevance of trifluoroacetic acid: Assessment with respect to flumetralinEFSA, DEPosition paper or review/third‐party assessment 20 Confidential2013Trifluoroacetic acid (TFAA): Statement on ToxicologyEFSA, DEPosition paper or review/third‐party assessment 21 Confidential1968Toxicity of the Metabolites of HalothaneEFSA, DEIn vivo study conducted through a route of administration other than oral 23 J. M. Baden et al.1979Carcinogenicity of Halothane in Swiss/ICR MiceEFSATest item: no TFA/TFA Na 24 P. Bently et al.1993Hepatic peroxisome proliferation in rodents and its significance for humansEFSAPosition paper or review/third‐party assessment 25 D. A. Blake et al.1969Animal Toxicity of 2,2,2‐TrifluoroethanolEFSAIn vivo study conducted through a route of administration other than oral 29 R. J. Bull2000Mode of action of liver tumor induction by trichloroethylene and its metabolites, trichloroacetate and dichloroacetateEFSAPosition paper or review/third‐party assessment 30 A. B. DeAngelo et al.1989Species and strain sensitivity to the induction of peroxisome proliferation by chloroacetic acidsEFSATest item: no TFA/TFA Na 31 A. B. DeAngelo et al.1997Failure of monochloroacetic acid and trichloroacetic acid administered in the drinking water to produce liver cancer in male F344/N ratsEFSATest item: no TFA/TFA Na 32 US EPA2003PPARα‐mediated hepatocarcinogenesis in rodents and relevance to human risk assessmentsEFSAPosition paper or review/third‐party assessment 33 J.W. Fisher et al.2001Trichloroethylene, trichloroacetic acid, and dichloroacetic acid: Do they affect fetal rat heart development?EFSATest item: no TFA/TFA Na 34 L. Fu, E. Marshall Johnson and L. M. Newman1990Prediction of the developmental toxicity hazard potential of halogenated drinking water disinfection by‐products tested by the in vitro hydra assayEFSATest item: no TFA/TFA Na 35 Health Council of the Netherlands. Committee for Compounds toxic to reproduction2006Trichloroacetic acid; evaluation of the effects on reproduction, recommendation for classificationEFSAPosition paper or review/third‐party assessment 36 S. L. Herren‐Freund et al.1987The carcinogenicity of trichloroethylene and its metabolites, trichloroacetic acid and dichloroacetic acid, in mouse liverEFSATest item: no TFA/TFA Na 38 IARC Working Group on the Evaluation of Carcinogenic Risks to Humans2004IARC Monographs on the Evaluation of Carcinogenic Risks of Humans, Volume 84; Some Drinking‐water Disinfectants and Contaminants, including ArsenicEFSAPosition paper or review/third‐party assessment 39 P. D. Johnson et al.1998Cardiac Teratogenicity of Trichloroethylene MetabolitesEFSATest item: no TFA/TFA Na 41 L. S. KaminskyToxicology of Atmospheric Degradation Products of Selected HydrochlorofluorocarbonsEFSAPosition paper or review//third‐party assessment 42 L. S. Kheilo and S. N. Kremneva1966Comparative assessment of the toxicity of trifluoroacetic and pentafluoropropionic acidEFSAStudy not in English ‐ Russian 44 Confidential1974Prüfung des einflusses von formamid (Charge XXIII/245) auf das trächtige kaninchen und den foetus bei verabreichung per schlundsondeEFSAStudy not in English ‐ German 45 J. M. Manson et al.1984Effects of Oral Exposure to Trichloroethylene on Female Reproductive FunctionEFSATest item: no TFA/TFA Na 46 G. G. Mather, J. H. Exon and L. D. Koller1990Subchronic 90 day Toxicity of Dichloroacetic and Trichloroacetic Acid in RatsEFSATest item: no TFA/TFA Na 47 W. B. Deichmann et al.1962Industrial Hygiene and Toxicology; Trifluoroacetic AcidEFSAPosition paper or review/third‐party assessment 48 M. A. Pereira1996Carcinogenic Activity of Dichloroacetic Acid and Trichloroacetic Acid in the Liver of Female B6C3F1 miceEFSATest item: no TFA/TFA Na 49 P. H. Rosenberg and T. Wahlström1971Hepatotoxicity of Halothane Metabolites in vivo and Inhibition of Fibroblast Growth in vitroEFSA, CLH dossier, DEIn vivo study conducted through a route of administration other than oral. *It is noted that the in vitro assays conducted in this study report have been included. 55 M. K. Smith et al.1989Teratogenic Activity of Trichloroacetic Acid in the RatEFSATest item: no TFA/TFA Na 58 WHO2004Trichloroacetic Acid in Drinking‐water; Background document for development of WHO Guidelines for Drinking‐water Quality EFSAPosition paper or review/third‐party assessment 59 G. Woodard et al.1941The Acute Oral Toxicity of Acetic, Chloroacetic, Dichloroacetic and Trichloroacetic AcidsEFSATest item: no TFA/TFA Na 60 H. Zenick et al.1984Effects of Trichloroethylene Exposure on Male Reproductive Function in RatsEFSATest item: no TFA/TFA Na 61 Confidential1999Flurtamone; Trifluoroacetic Acid: Health Effect Risk AssessmentEFSA, HR, DE, TFA Task forcePosition paper or review/third‐party assessment 62 Confidential1968Toxic Actions of the Metabolites of Halothane: LD50 and Some Metabolic Effects of Trifluoroethanol and Trifluoroacetic Acid in Mice and Guinea PigsEFSA, CLH dossier, DE, TFA task forceIn vivo study conducted through a route of administration other than oral 63 R. C. Lind and A. J. Gandolfi1990Covalent Binding of Oxidative Biotransformation Reactive Intermediates to Protein Influences Halothane‐Associated Hepatotoxicity in Guinea PigsEFSATest item: no TFA/TFA Na 64 J. M. Fraser and L. S. Kaminsky19882,2,2‐Trifluoroethanol Intestinal and Bone Marrow Toxicity: The Role of Its Metabolism to 2,2,2‐Trifluoroacetldehyde and Trifluoroacetic AcidEFSA, CLH dossier, DE, TFA task forceIn vivo study conducted through a route of administration other than oral 65 Confidential1997TRIFLUOROACETIC ACID: A position statement of the herbicidal effects of this compound on a range of crops and weed species when applied as a pre‐ and post‐emergence treatment in the glasshouse.EFSAPosition paper or review/third‐party assessment 73 ECETOC1995 Joint Assessment of Commodity Chemicals No.31; 1,1,1,2‐Tetrafluoroethane (HFC‐134a) Cas No. 811‐97‐2 EFSAPosition paper or review/third‐party assessment 74 ECETOC1996Joint Assessment of Commodity Chemicals No.33; 1,1‐Dichloro‐2,2,2‐Trifluoroethane (HCFC‐123) CAS No.306‐83‐2EFSAPosition paper or review/third‐party assessment 75 D. J. Ford, D. E. Coyle and J. F. Harrington1984Effects of Hypersensitivity to a Halothane Metabolite on Halothane‐induced Liver DamageEFSA, DEIn vivo study conducted through a route of administration other than oral 76 Confidential1995Trifluoroacetic Acid: A Toxicological ProfileDEPosition paper or review 77 H. Ghantous et al.1986Distribution of Halothane and the Metabolites Trifluoroacetic Acid and Bromide in the Conceptus after Halothane Inhalation by Pregnant MiceEFSA, CLH dossier, DE, TFA task forceIn vivo study conducted through a route of administration other than oral 78 Confidential1984Genotoxicity and Carcinogenicity of Fluorocarbons: Assessment by Short‐Term in Vitro Test and Chronic Exposure in RatsEFSATest item: no TFA/TFA Na 79 A. Mathieu et al.1974Experimental Immunity to a Metabolite of Halothane and Fluroxene: Cutaneous Delayed‐type HypersensitivityEFSA, DEIn vivo study conducted through a route of administration other than oral 80 J. G. Reves and L. E. McCracken1976Failure to Induce Hepatic Pathology in Animals Sensitized to a Halothane Metabolite and Subsequently Challenged with HalothaneEFSA, DEIn vivo study conducted through a route of administration other than oral 82 K. R. Solomon et al.2016Sources, fates, toxicity, and risks of trifluoroacetic acid and its salts: Relevance to substances regulated under the Montreal and Kyoto ProtocolsEFSAPosition paper or review/Third‐party assessment 83 M. M. Airaksinen, P. H. Rosenberg and T. Tammisto1970A Possible Mechanism of Toxicity of Trifluoroethanol and Other Halothane MetabolitesEFSAIn vivo study conducted through a route of administration other than oral. *It is noted that the in vitro assays conducted in this study report have been included. 84 A. Bartorelli2003Calcium Salts with Cytotoxicity ActivityCLH dossier, DEPosition paper or review/third‐party assessment 85 Confidential2014Summary of the toxicological and metabolism studies for FlurtamoneCLH dossierPosition paper or review/third‐party assessment 86 J. C. Boutonnet et al.1999Environmental Risk Assessment of Trifluoroacetic AcidCLH dossier, DEPosition paper or review/third‐party assessment 89 B. R. G. Danielsson, H. Ghantous and L. Dencker1985Accumulation in Murine Amniotic Fluid of Halothane and its MetabolitesCLH dossier, DETest item: no TFA/TFA Na 90 P. Divakaran et al.1980Accumulation and Persistent of Halothane in Adult and Fetal Rat Brain as a Result of Subanesthetic ExposureCLH dossier, DETest item: no TFA/TFA Na 93 Confidential1964Comparative assessment of the toxicity of trifluoroacetic and pentafluoropropionic acidCLH dossierStudy not in English – Russian 97 H. Kinoshita1989Enterohepatic Circulation of Trifluoroacetic Acid: A Pharmacokinetics StudyCLH dossier, DEIn vivo study conducted through a route of administration other than oral 105 T. W. Sadler1979Culture of early somite mouse embryos during organogenesisCLH dossier, DETest item: no TFA/TFA Na 112 Confidential2020Prenatal Development Toxicity Study of TFSK TFAK REACTION MASS Following Oral Administration to Pregnant Rats by GavageTFA Task Force, DETest item: no TFA/TFA Na 115 Confidential2024Position Paper to explain the Mode of Action for observed fetal eye findings in rabbitsTFA Task ForcePosition paper or review//third‐party assessment 117 Confidential2024Updated generic Risk Assessment (not specific to a certain precursor substance)TFA Task ForcePosition paper or review/third‐party assessment 118 Confidential2024Proposal of a health‐based drinking water limit value for TFATFA Task ForcePosition paper or review/third/‐party assessment 119 Confidential2024Expert Statement; Subject: Sodium Trifluoroacetate; Finding: Retroesophageal right subclavian arteryTFA Task ForcePosition paper or review/third‐party assessment 120 L. Graff and S. Drechsler2024Trifluoroacetic Acid (TFA): public environmental monitoring data (groundwater, surface water, drinking water, soil, air and sediment)TFA Task ForceEnvironmental fate and behaviour assessment considered to fall outside the scope of the mandate 121 Confidential2012Combined Repeated Dose Toxicity Study with the Reproduction/Developmental Toxicity Screening Test by Oral Route (Gavage) in RatsTFA Task ForceTest item: no TFA/TFA Na 127 Confidential2020Repeated Dose 90‐Day Oral Toxicity Study of TFSK TFAK REACTION MASS in Rats (Dose Formulation Analysis Report)TFA Task ForceTest item: no TFA/TFA Na 128 Confidential2020Repeated Dose 90‐Day Oral Toxicity Study of TFSK TFAK REACTION MASS in Rats (Pathology Report)TFA Task ForceTest item: no TFA/TFA Na 129 Confidential2020Repeated Dose 90‐Day Oral Toxicity Study of TFSK TFAK REACTION MASS in RatsTFA Task ForceTest item: no TFA/TFA Na 135 K. Arai et al.1989Variation of TFAA in Liver Cells After an Inhalation of HalothaneDETest item: no TFA/TFA Na 136 Holaday and Cunnah1976Distribution and Elimination of trifluoroacetic acid in man and the rat.DEIn vivo study conducted through a route of administration other than oral. 137 A. Mäkeläinen and P. Rosenberg1974The Effects of Halothane and Methoxyflurane Metabolites on Lipolysis In VitroDELipolysis study considered to fall outside the scope of the mandate 138 P. H. Rosenberg1971Decrease in Reduced Glutathione and NADPH and Inhibition of Glucose‐6‐phosphate Dehydrogenase Activity Caused by Metabolites of Fluroxene and HalothaneDEIn vivo study conducted through a route of administration other than oral 142 Confidential2012[Thiadiazole‐5‐14C] Flufenacet: Supportive Experiment for the Identification of Metabolites in the Urine of the RatDETest item: no TFA/TFA Na 143 Confidential2016Mutagenicity Study of Flufenacet Metabolite M2 in the Salmonella Typhimurium Reverse Mutation Assay (In Vitro)DETest item: no TFA/TFA Na 144 Confidential2016In Vitro Assessment of the Clastogenic Activity of Flufenacet Metabolite M2 in Cultures Human Peripheral LymphocytesDETest item: no TFA/TFA Na 145 Confidential2007The Non Relevance of the Environmental Metabolites of Flurtamone; TFMBA and TFADEPosition paper or review//third‐party assessment 146 Confidential2008Toxicological data on Trifluoroacetic acid (TFAA)DE, TFA Task ForcePosition paper or review//third‐party assessment 153 Confidential2022Development of a bioanalytical method to determine Vitamin A concentrations in Rabbit Serum using LC–MS/MSTFA Task ForceTest item: no TFA/TFA Na 154 Confidential2022Scientific Validation of a Bioanalytical Method for the Determination of Vitamin A concentrations in Rabbit Serum using LC–MS/MSTFA Task ForceTest item: no TFA/TFA Na 157 Confidential2023Development of a Bioanalytical Method to Determine Lactate concentrations in Rabbit aqueous/vitreous humour using LC–MS/MSTFA Task ForceTest item: no TFA/TFA Na 158 Confidential2024Additional Method Development for a Bioanalytical Method for the Determination of Lactate concentrations in Rabbit aqueous/vitreous humourTFA Task ForceTest item: no TFA/TFA Na 159 Confidential2024Scientific Validation of a Bioanalytical Method for the Determination of Lactate concentrations in Rabbit Aqueous/Vitreous humour using LC–MS/MSTFA Task ForceTest item: no TFA/TFA Na 165 Confidential2011P2255: Test strategy to determine a long‐term DNEL for irritation effects by inhalation for workers exposed to TFATFA Task ForceConference proceedings 166 Confidential2025Position paper_TFA – Review of TRVs EFSA PREV 3rd WG Meeting on TFA ToxicityTFA Task ForcePosition paper or review//third‐party assessment 167 Confidential1986Test d'AmesTFA Task ForceStudy not in English ‐ French 168 Confidential2007Evaluation of Skin Sensitization potential in mice using the local lymph node assay (LLNA)TFA Task ForceTest item: no TFA/TFA Na 170 Confidential1997A Summary of the Potential Toxicity of Trifluoroacetic AcidEFSA, DEPosition paper or review//third‐party assessment Studies excluded from the body of evidence of TFA according to the eligibility criteria. Joint Assessment of Commodity Chemicals No.31; 1,1,1,2‐Tetrafluoroethane (HFC‐134a) Cas No. 811‐97‐2 Immunotoxicity in Retrospective Evaluations of Immunological Endpoints Across the major retrospective evaluations by the US EPA (2013), Gehen et al. (2014) and CRD (2015), immunotoxicity endpoints, including TDAR responses and immune‐related biomarkers from repeated‐dose studies, were consistently not identified as determinants in regulatory decision‐making. Most pesticides showed no immunotoxic effects even at the highest doses tested, and when effects did occur, they generally appeared only at doses that also produced systemic toxicity. TDAR NOAELs reported in the US EPA and Gehen et al. data sets were several to 100‐fold higher than the reference points used in risk assessment, confirming that TDAR was not the more sensitive endpoint. In the EU review of 198 active substances (CRD, 2015), which had limited overlap with the US EPA data set, reproducible immunotoxic effects were identified for only 21 substances, and just one case involved an immune biomarker in the derivation of a health‐based guidance value. For the substances further examined for potential immunotoxicity, the margins between immunotoxicity NOAELs and ADI‐setting NOAELs were generally wide, with the few exceptions based on isolated biomarker changes rather than consistent evidence. Taken together, these findings indicate that immunotoxicity has historically not been a limiting factor in regulatory toxicology and hardly ever determines the reference point for risk assessment. It needs to be noted that investigated TDAR NOAELs relate to assessment of immunotoxicity in adult animals not been exposed in utero previously. Advice from the EFSA Working group on Genotoxicity 10 April 2025 The potential of TFA and TFA Na to induce chromosomal damage was evaluated with the chromosome aberration test performed in human peripheral blood lymphocytes and CHL cell line in the presence and absence of metabolic activation. Negative results were obtained in three studies of limited relevance carried out using TFA Na, while a highly relevant study conducted with TFA was positive. The potential of TFA to induce chromosomal damage was also investigated in two in vitro micronucleus tests, conducted in the presence and absence of metabolic activation using the human lymphoblastoid TK6 and CHO‐K1 cell lines. Negative results were obtained in both studies, considered of high relevance. The WG noted that the former data gap on aneugenicity (EFSA, 2023a) is now fulfilled by applying the in vitro micronucleus test. Overall, the studies carried out with TFA Na report no induction of chromosomal damage, while inconsistencies were observed among the studies performed with TFA showing a positive response in one chromosome aberration test and negative results in two in vitro micronucleus tests. In addition, the WG noted that studies with TFA showed considerable variability in cytotoxicity levels, both within individual studies (e.g. between range‐finding and main experiments) and across different studies. The reasons for these inconsistencies remain unclear but could be related to the physico‐chemical characteristics of TFA (a strong acid) (Morita et al., 1989) and technical factors modulating the effects of TFA. In contrast, no such variability was observed with TFA Na. Considering (i) the substantial uncertainty surrounding the mechanisms underlying these findings and (ii) that exposure to TFA in its acid form is unlikely under physiological conditions, the WG assigns greater weight to the consistently negative results of chromosomal aberration studies conducted with the TFA salt than to those observed with the acid. Overall, the weight of evidence does not raise a concern for chromosomal damage of TFA and its Na salt. Annex 1: Protocol of the toxicological assessment. Annex 2: List of studies collected in the framework of the targeted call for data. Annex 3: Outcome of the Risk of Bias for non‐guideline in vitro and ex vivo studies in the body of evidence. Annex 4: Outcome of the Risk of Bias for non‐guideline in vivo studies in the body of evidence. Annex 5: Outcome of the assessment for human studies. Annex 6: Data extraction of non‐guideline in vivo animal oral toxicity studies. Annex 7: Data extraction of non‐guideline in vitro toxicity studies. Annex 8: Outcome of the public consultation. Annex 9: Consolidated list of studies outside the window of call for data. Annex 10: Assessment of immunotoxicity endpoints in available toxicity studies. Annex 11: Benchmark Dose (BMD) Modelling. Annexes 1 to 11 are available under the Supporting Information section on the online version of the scientific output. EFSA (European Food Safety Authority) , Cavelier, A. , Coja, T. , Gall, A. , Hernández‐Jerez, A. , Vleminckx, C. , Bergsma, E. , Binaglia, M. , Crivellente, F. , Colas, M. , Lanzoni, A. , Louisse, J. , De Jesus, R. R. , Grillo, A. , Mangas, I. , & Parra Morte, J. M. (2026). Scientific Report on consumer health‐based guidance values for trifluoroacetic acid. EFSA Journal, 24(7), e10227. 10.2903/j.efsa.2026.10227 Approved: 29 June 2026 Correspondence: Ask a Question The declarations of interest of all scientific experts active in EFSA's work are available at https://open.efsa.europa.eu/experts. Note: A plain language summary of this scientific opinion is available at https://efsa.onlinelibrary.wiley.com/doi/10.2903/j.efsa.2026.p240701 Notes According to OECD guideline (OECD, 2021) and ECHA's definition from restriction intention (ECHA, 2023), PFAS are identified based on their perfluorinated chemical structure. Regulation (EC) No 1107/2009 of 21 October 2009 of the European Parliament and of the Council concerning the placing of plant protection products on the market and repealing Council Directives 79/117/EEC and 91/414/EEC. OJ L 309, 24.11.2009, p. 1–50. REACH registration dossier on TFA including the prenatal developmental toxicity study in rabbits (OECD TG 414). Regulation (EC) No 1272/2008 of the European Parliament and of the Council of 16 December 2008 on classification, labelling and packaging of substances and mixtures, amending and repealing Directives 67/548/EEC and 1999/45/EC, and amending Regulation (EC) No 1907/2006. OJ L 353, 31.12.2008, pp. 1–1355. ECHA Registry of CLH intentions until outcome – Trifluoroacetic acid: https://echa.europa.eu/it/registry‐of‐clh‐intentions‐until‐outcome/‐/dislist/details/0b0236e188e6e587. May damage fertility. Suspected of damaging the unborn child. ECHA Harmonised classification and labelling consultations: https://echa.europa.eu/fr/harmonised‐classification‐and‐labelling‐consultation/‐/substance‐rev/80001/term. Harmful if swallowed. Toxic if inhaled. May damage fertility. Suspected of damaging the unborn child. https://echa.europa.eu/documents/d/guest/rac77_final_minutes_en. Regulation (EC) No 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety. OJ L 31, 1.2.2002, p. 1–24. Refer to the EC Mandate and follow‐up communications at: https://open.efsa.europa.eu/questions/EFSA‐Q‐2024‐00502. On 7 January 2021, a notification according to Article 56 of Regulation 1107/2009 was submitted from the TFA task force to EFSA, the European Commission and Member States concerning potentially harmful or unacceptable effects of TFA. The CLH dossier was drafted by the German federal agency, BAuA – Federal Institute for Occupational Safety and Health. According to their chemical structures (https://echa.europa.eu/hot‐topics/perfluoroalkyl‐chemicals‐pfas). Commission Regulation (EU) No 283/2013 of 1 March 2013 setting out the data requirements for active substances, in accordance with Regulation (EC) No 1107/2009 of the European Parliament and of the Council concerning the placing of plant protection products on the market. OJ L 93, 3.4.2013, pp. 1–84. Directive (EU) 2020/2184 of the European Parliament and of the Council of 16 December 2020 on the quality of water intended for human consumption. http://data.europa.eu/eli/dir/2020/2184/oj. https://environment.ec.europa.eu/news/new‐eu‐rules‐limit‐pfas‐drinking‐water‐2026‐01‐12_en. Please refer to CLH report, Section 6. Data Sources (electronic page 9) https://echa.europa.eu/documents/10162/0dae2d2a‐2b4c‐b63e‐c669‐bd3a76197aa1. https://open.efsa.europa.eu/questions/EFSA‐Q‐2024‐00502. Methods of analysis reports supporting other toxicological studies were also submitted (#149, #152, #155, #156, #160, #161); only methods of analysis supporting key studies were assessed in the current output. https://echa.europa.eu/documents/d/guest/rac77_final_minutes_en. Study submitted under Comment 47 of the ECHA public consultation: https://echa.europa.eu/documents/10162/e7cc66e0‐aa3f‐fd05‐b35d‐fe2647db3fbc: Anonymous, 2025. SODIUM TRIFLUOROACETATE: A PRENATAL DEVELOPMENTAL TOXICITY STUDY IN RABBITS WITH 14‐DAY RECOVERY. A conversion factor of 0.84 was applied to adjust for the difference in relative molecular weight between TFA Na and TFA, i.e., calculated from the ratio of TFA [114 g/mol] to TFA Na [136 g/mol]. Self‐task Mandate of the Plant Protection Products and their Residues (PPR) panel for a Scientific Opinion on waiving of the dog studies in the regulatory process of agrochemicals approval (EFSA‐Q‐2024‐00199). https://open.efsa.europa.eu/question/EFSA‐Q‐2024‐00199. The WG TFA considered that HCD on negative control should not be used to show proficiency of the lab to measure thyroid hormones since the proficiency should come from HCD on positive control. The WG TFA would consider more appropriate to state the following: “The values of T4 in the concurrent controls are within the min–max range of the HCD provided. No positive controls were submitted to confirm the proficiency of the lab for hormonal measurements”. US Environmental Protection Agency (EPA). (2011). Integrated approaches to testing and assessment strategy: use of new computational and molecular tools. FIFRA Scientific Advisory Panel Consultation, May 24—26, 2011. Arlington (VA): FIFRA Scientific Advisory Panel Meeting. EFSA Scientific Committee. (2025). Risks to human and animal health from the presence of bromide in food and feed. EFSA Journal, 23(1), e9121. https://doi.org/10.2903/j.efsa.2025.9121. US Environmental Protection Agency (EPA). (2011). Integrated approaches to testing and assessment strategy: use of new computational and molecular tools. FIFRA Scientific Advisory Panel Consultation, May 24—26, 2011. Arlington (VA): FIFRA Scientific Advisory Panel Meeting. EFSA Scientific Committee. (2025). Risks to human and animal health from the presence of bromide in food and feed. EFSA Journal, 23(1), e9121. https://doi.org/10.2903/j.efsa.2025.9121. US Environmental Protection Agency (EPA). (2011). Integrated approaches to testing and assessment strategy: use of new computational and molecular tools. FIFRA Scientific Advisory Panel Consultation, May 24—26, 2011. Arlington (VA): FIFRA Scientific Advisory Panel Meeting. AOP 402: Thyroid peroxidase (TPO) inhibition leads to periventricular heterotopia formation in the developing rat brain. Available at: https://aopwiki.org/aops/402. REFERENCES Associated Data Supplementary Materials Annex 1: Protocol of the toxicological assessment. Annex 2: List of studies collected in the framework of the targeted call for data. Annex 3: Outcome of the Risk of Bias for non‐guideline in vitro and ex vivo studies in the body of evidence. Annex 4: Outcome of the Risk of Bias for non‐guideline in vivo studies in the body of evidence. Annex 5: Outcome of the assessment for human studies. Annex 6: Data extraction of non‐guideline in vivo animal oral toxicity studies. Annex 7: Data extraction of non‐guideline in vitro toxicity studies. Annex 8: Outcome of the public consultation. Annex 9: Consolidated list of studies outside the window of call for data. Annex 10: Assessment of immunotoxicity endpoints in available toxicity studies. Annex 11: Benchmark Dose (BMD) Modelling.