Scientific opinion on toxicity of pyrethroid common metabolites

Abstract

The EFSA Panel on Plant Protection Products and their Residues was requested by the European Commission, to conclude based upon available evidence if metabolites 3‐phenoxybenzoic acid (PBA) and 3‐(4′‐hydroxyphenoxy)benzoic acid (PBA(OH)), common to several pyrethroid compounds, have genotoxic properties, if they share the (neuro)toxicity profile of their parent compounds, and if evidence allows to conclude on their health‐based guidance values. Available body of evidence consisted of studies from regulatory dossiers submissions, as well as from public literature. In addition, the data gap for short‐term toxicity profile of PBA was addressed by read‐across. Assessment revealed that PBA and PBA(OH) do not raise a concern with respect to genotoxicity. As regards general toxicity, PBA and PBA(OH) have different qualitative (no neurotoxic mechanism) and quantitative (higher NOAELs) toxicity compared to the parent pyrethroid compounds. For both metabolites, acceptable daily intake (ADI) and acute reference dose (ARfD) values were derived at 0.1 mg/kg body weight (bw) per day and 1 mg/kg bw, respectively.

Summary

3-phenoxybenzoic acid (PBA) and 3-(4′-hydroxyphenoxy)benzoic acid (PBA(OH)) are common metabolites from several pyrethroid compounds. They are frequently identified in commodities of plant origin, but few toxicity data were available in the past to conclude whether they share the same (neuro)toxicity effect with their parent compounds.

Ester hydrolysis is an initial step in pyrethroid metabolism, which determines, within other factors, to a significant extent the toxicokinetic and toxicodynamic characteristics of the individual pyrethroids. Even if PBA and PBA(OH) are important distal metabolites for both type I and II pyrethroids, the metabolism from the parent pyrethroid to these metabolites contains many sequential steps and finally hydroxylated metabolites (e.g. PBA and PBA(OH)) may be variably conjugated with amino acids, sulfates, sugars or sugar acids. Accordingly, PBA/PBA(OH) as aglycons are frequently only minor metabolites of the pyrethroids.

As regards genotoxicity, data available for PBA and PBA(OH) do not raise a concern with respect to gene mutation. Data available for PBA(OH) do not raise a concern for chromosomal damage, but positive results for PBA were observed in vitro. However, results from the in vivo micronucleus assay were negative with sufficient evidence of bone marrow exposure. Therefore, no concern for chromosome damage is concluded also for PBA.

Only few data on toxicity of PBA and PBA(OH) were available in the past to reliably conclude if they are of similar, higher or lower toxicity than their parent compounds. For the purpose of this Scientific Opinion, data (provided by Member States and industry) from old toxicity studies (conducted in 1970s and 1980s) with PBA and PBA(OH) and newly conducted short-term toxicity studies (28-day rat studies) with PBA(OH) and 3-phenoxybenzaldehyde (PBAld) were evaluated. An NOAEL of 106.9 mg/kg bw per day was set for PBA(OH) and of 98.9 mg/kg bw per day for PBAld. In both studies, the effects observed were lower body weight gains and food consumption and increased liver weights associated with minimal-to-mild centrilobular hepatocellular hypertrophy. The observed findings were not related to a neurotoxic mechanism. The benchmark dose modelling (BMD) for the reference point in the 28-day study with PBAld confirmed the NOAEL of 98.9 mg/kg bw per day.

No short-term study was submitted for PBA, but a read-across was conducted from PBA(OH) and PBAld, with the conclusion, following a weight of evidence (WoE) approach, that PBA can be assumed to have similar toxicological properties.

Based upon available experimental data and considerations, the PPR Panel concluded that PBA and PBA(OH) do not share the same leading hazard properties (neurotoxicity) with their parent pyrethroid compounds and that they are less toxic (higher NOAELs in 28-day studies with PBA(OH) and PBAld). Health-based guidance values (HBGV) were derived for PBA and PBA(OH). The acceptable daily intake (ADI) has been derived at 0.1 mg/kg bw per day, based upon the overall NOAEL of 100 mg/kg bw per day from the 28-day studies with PBAld (NOAEL of 98.9 mg/kg bw per day) and PBA(OH) (NOAEL of 106.9 mg/kg bw per day) and the uncertainty factor (UF) of 1,000. The acute reference dose (ARfD) has been derived at 1 mg/kg bw based upon the same overall NOAEL and UF of 100.

Introduction

Pyrethroids are a class of synthetic insecticides with a principal mode of action via modulating voltage-gated sodium channels (VGSC or Na_V) of axons in nervous system of insects, causing neurotoxic effects (Soderlund et al., 2002; Soderlund, 2012). Synthetic pyrethroids are divided into two classes, with Type I compounds (e.g. permethrin) not containing a cyano moiety and Type II compounds (e.g. deltamethrin) containing a cyano moiety in alpha-carbon of the pyrethroid structure. Type II pyrethroids are more neurotoxic to insects than type I compounds, due to molecularly distinct interactions with the primary site of action, the Na_V. All pyrethroids can produce neurotoxic effects in mammalian cells, although there are differences between the two types regarding their potency, the primary mode of action and the neurotoxic syndromes in rodents. The characteristic clinical signs seen in mammals following exposure to type I pyrethroids include the occurrence of fine tremors, hyperexcitability and myoclonus (T-syndrome). For the type II pyrethroids, clinical signs seen in mammals following exposure include sinuous writhing (choreoathetosis), salivation, hyperactivity and clonic/tonic convulsions (CS-syndrome).

The basic pyrethroid structure is composed of two parts: an acid and an alcohol moiety (Figure 1). The ester link between the two moieties is of prime importance to understand the toxicokinetic properties of a pyrethroid, because it is a target for carboxylesterases to a variable extent (e.g. pending the molecular affinity) and the hydrolysis is a key step for the inactivation of a pyrethroid.

[IMAGE OMITTED. SEE PDF]

There are also other structural characteristics that may be of significance for toxicokinetic and toxicodynamic behaviour of pyrethroids. In addition to cis-trans isomers, all pyrethroids contain one to three chiral centres resulting in two to eight enantiomers (Figure 2). Each pyrethroid and each enantiomer may demonstrate (but not necessarily) some unique or quantitatively distinct features in toxicodynamic molecular interactions and toxicities as well as in toxicokinetic characteristics.

[IMAGE OMITTED. SEE PDF]

Ester hydrolysis is an initial step in pyrethroid metabolism, which determines, among other factors, to a significant extent the toxicokinetic and toxicodynamic characteristics of an individual pyrethroid. Even if PBA and PBA(OH) are important distal metabolites for both type I and II pyrethroids, the metabolism from the parent pyrethroid to these metabolites contains many sequential steps and finally hydroxylated metabolites (e.g. PBA and PBA(OH)) may be variably conjugated with amino acids, sulfates, sugars or sugar acids. Accordingly, PBA/PBA(OH) are frequently only minor aglycon metabolites of the pyrethroids.

In summary, principal metabolites and pathways for most pyrethroids are the following (not in the order of abundance, but of the intact parent as the first metabolism target):

• cleavage (hydrolysis) of the ester linkage of the parent;
• oxidation of the parent at the 2′- and 4′-phenoxy positions of the alcohol moiety;
• oxidation of the parent at suitable groups (e.g. methyl) of the acid moiety;
• conversion of the cyano group (for type II pyrethroids) to SCN ion and CO₂;
• conversion of 3-phenoxybenzyl alcohol to 3-phenoxybenzaldehyde and further to PBA;
• conjugation of the resulting carboxylic acids, phenols, etc. with glucuronic acid, sulfate and/or glycine, and
• (rarely reported) ether hydrolysis of the alcohol moiety.

It is clear that hydrolysis of the ester link is the primary metabolic step for many pyrethroids and is of particular importance, because the outcome is the inactivation of the parent, i.e. termination of the insecticidal effect. Moreover, final products are simple or substituted (e.g. fluoro-atom in cyfluthrin) 3-phenoxybenzoic acid (PBA) and its further metabolised derivatives. Chemical structures of PBA and PBA(OH), the metabolites which are subject to this Scientific Opinion, and in addition PBAld, used for read-across purpose, are provided under Annex A.

The principal enzymes to metabolise pyrethroids are carboxylesterases (CES) executing the hydrolytic cut of the ester link between two parts of the molecule (Figure 1), and NADPH-dependent oxidative enzymes, principally cytochrome P450 (CYP) enzymes (Scollon et al., 2009; Hedges et al., 2020). Alcohol and acid moieties produced by CES enzymes in microsomes, cytosol and plasma may be further metabolised by other enzymes, mainly P450 enzymes in in liver, intestine and other tissues. Oxidative enzymes may metabolise also the uncleaved parent pyrethroid, in addition to suitable targets in the cleaved alcohol and acid moieties. The result is usually a large number of metabolites, which is subject to a demanding analytical identification and quantification. Thus, it is not surprising that in many in vitro metabolism studies, the parent disappearance (clearance) has been the selected approach, and in this case, metabolites produced remain experimentally undefined, even if the pathway is well recognised and clearance predictions are rather reliable (see e.g. Song et al., 2019; Mallick et al., 2020a,b).

Despite considerable knowledge on the metabolism of pyrethroids as outlined above, only relatively scarce information is available on the toxicological profile(s) of pyrethroid metabolites. This holds true also for PBA and its related metabolites and intermediates, even if those metabolites are common to most pyrethroids and constitute an important part of residues of pyrethroids in the environment and are used for biomonitoring of exposure to pyrethroids. Consequently, there is a necessity to explore toxicological characteristics of PBA-related metabolites in relations to their parent pyrethroids and to decide whether additional studies or other actions would be needed regarding residue definition, biomonitoring or any other regulatory considerations.

Background and terms of reference as provided by the requestor

The metabolites 3-phenoxybenzoic acid [PBA or 3-PBA] and 3-(4′-hydroxyphenoxy)benzoic acid [PBA(OH) or 4-OH-PBA] can form during the metabolism and breakdown of pyrethroid active substances including the following substances (some of which are no longer approved in the EU): gamma-cyhalothrin, lambda-cyhalothrin, alpha-cypermethrin, cypermethrin, beta- and zeta-cypermethrin, deltamethrin, etofenprox, tau-fluvalinate, fenvalerate, esfenvalerate, permethrin, fenpropathrin, cyfluthrin, beta-cyfluthrin, acrinathrin.

In the EFSA Conclusions on the active substances gamma-cyhalothrin (2014)1 and lambda-cyhalothrin (2014),2 a data gap was concluded by EFSA for toxicological information to assess the toxicity profile of those metabolites. EFSA also concluded that depending on the toxicological profile of the metabolites, a cumulative assessment taking into account exposure from the use of multiple substances may need to be considered (similar to that carried out for the triazole derivative metabolites3 following a mandate sent by the European Commission).

Furthermore, in the EFSA Conclusions on the active substances alpha-cypermethrin (2018)4 and cypermethrin (2018),5 EFSA concluded that the metabolites are unlikely to be of higher toxicity than the parent, but that this conclusion might need to be revised once the confirmatory information submitted in the context of lambda-cyhalothrin are peer reviewed. Therefore, the consumer dietary risk assessment could not be finalised for those substances.

Confirmatory information on gamma-cyhalothrin and lambda-cyhalothrin as required in the respective approval/renewal of approval were submitted by the applicants within the deadlines prescribed in the relevant Implementing Regulations, and were evaluated by the respective Rapporteur Member States. The assessments, in compliance with Guidance Document SANCO 5634/20096, were circulated to the applicants, the other Member States and EFSA for comments, all of which were collated in Technical Reports.

In the Technical Report on gamma-cyhalothrin (2019)7 EFSA found that the genotoxic potential of the metabolites PBA and PBA(OH) either regarding mutagenicity or clastogenicity could not be concluded and proposed a peer review consultation, for which the use of a Threshold of Toxicological Concern (TTC) approach for these metabolites would also merit a peer review discussion. However, some Member States did not agree and considered that the existing data already confirm that there is no concern for genotoxicity for the metabolites PBA and PBA(OH).

For the Technical Report on lambda-cyhalothrin (2020)8 EFSA evaluated additional studies submitted as confirmatory information, however an overall conclusion on the toxicity profiles of the metabolites could not be determined. Again, EFSA indicated a need for further peer review.

In the light of some divergence of views between the Member States experts involved in the peer review process, the need to avoid unnecessary testing on vertebrate animals, and for ensuring harmonised assessments for the pyrethroid active substances, including for ongoing or future renewals, the Commission mandates EFSA to conduct the following tasks:

Step 1: preparatory work

The Commission asks EFSA, in cooperation with Member States, to collect all available evidence relevant for the completion of steps 2 and 3 considering information for all pyrethroid active substances for which these metabolites occur.

Step 2: toxicological characterisation

The Commission asks the Panel on Plant Protection Products and their Residues:

• to perform a literature review of published scientific literature to ensure the database is complete;
• to assess the available evidence in relation to the toxicity profile of the metabolites PBA and PBA(OH) and establish a toxicity profile for the metabolites PBA and PBA(OH), including the genotoxic potential, in particular indicating whether the metabolites are of lower or comparable toxicity to the parent substances. If possible, health-based reference values for the metabolites to be used in risk assessment should be derived.

Step 3: consideration of the impact on the consumer risk assessment

As soon as possible, in parallel to finalising the Opinion of the Panel, EFSA should indicate to the Commission whether or not a follow up review of existing MRLs would be likely to become necessary, taking into account all available elements that possibly have an impact on the consumer risk assessment. Based on that, the Commission can then consider to request a follow up opinion under Article 43 of Regulation (EC) No 396/2005 (including data and scope to be considered) i.e. to carry out an assessment of dietary exposure to them from the use of multiple active substances.

Once the toxicological assessment in Step 2 is completed by the Panel, the Commission asks EFSA to provide a conclusion to determine if the metabolites should be included in the residue definitions for risk assessment, and if so to conclude on the appropriate definitions (crops, livestock and processed commodities, where necessary).

All information collected in Step 1 and identified in Step 2 (literature survey) should be considered. The issues and divergent views identified in the peer review process of the confirmatory information recently assessed (e.g. EFSA Technical Reports) and/or during the renewal evaluations should be taken into account by the Panel and EFSA. The current state of the art on assessment of toxicity, read-across and other alternative in silico and non-testing methods and tools should be considered, where appropriate.

Interpretation of the terms of reference

The objective of this mandate is to assess the information on toxicity (including genotoxicity) of metabolites PBA and PBA(OH), in relation to their parent pyrethroid compounds. The relevant information has been provided by the Member States competent authorities and data owners from industry. Beginning the evaluation, the PPR Panel agreed to include in the assessment of short-term toxicity also PBAld, since bridging for general toxicity from PBAld to PBA was proposed in the dossier for renewal of approval of lambda cyhalothrin. This was based on the circumstance that no studies of longer duration are available for PBA. Additionally, a literature review was performed for PBA, PBA(OH) and PBAld to complete the data set to be assessed by the PPR Panel.

The PPR Panel interpreted that the terms of reference (ToR) require the Panel to provide advice on:

• The genotoxicity of PBA and PBA(OH), discussed controversially in the evaluation of the active substances lambda-cyhalothrin and gamma-cyhalothrin.
• Toxicological endpoints relevant for hazard characterisation of PBA and PBA(OH) according to Regulation (EU) 283/2013 for active substance and relevant for setting of health-based guidance values (HBGVs). These include in vivo and in vitro data. Beside the available data, read-across and other alternative in silico and non-testing methods and tools can be applied, if necessary.
• Comparison of relevant general toxicity endpoints (see point 2) of PBA and PBA(OH) to their parent pyrethroid compounds as listed in the mandate.
• Derivation of HBGV for PBA and PBA(OH), pending sufficient information on their (geno)toxicity.

The PPR Panel decided to consult the cross-cutting working group (ccWG) genotoxicity of the EFSA Scientific Committee for questions on gene mutation and chromosome damage, for which divergent views were identified in the peer review of the active substances lambda-cyhalothrin and gamma-cyhalothrin.

Problem formulation

In line with the ToRs and the draft framework for protocol development for EFSA's scientific assessments (EFSA, 2020a), the following questions will be addressed:

• Based on available evidence, can the conclusion on genotoxicity of common metabolites PBA and PBA (OH) be reached?

  • ○ Do PBA and PBA (OH) induce gene mutation in mammalian cells?
  • ○ Do PBA and PBA (OH) induce numerical and structural chromosome aberrations (in vitro/in vivo)?

• Based upon available evidence, can the general toxicity of PBA and PBA(OH) be assessed?

  • ○ Can hazard characterisation/intrinsic toxicological properties of PBA and PBA(OH), relevant for derivation of specific HBGVs, be addressed?

• Based upon available evidence, are PBA and PBA(OH) less, equal or more toxic, compared to general toxicity of their parent compounds?

• Based upon available evidence, can specific HBGVs for PBA and PBA (OH) be derived?

  • ○ If specific HBGVs for PBA and PBA (OH) can be derived, which data have been considered to support the decision?
  • ○ If specific HBGVs for PBA and PBA (OH) can be derived, how are uncertainties reflected, if captured?

Definition of the methods

The following methodology was identified by the PPR Panel WG to address the problem formulation questions:

• Tools for data collection:

  • – Member States and Industry task force9 consultation, conducted by EFSA Pesticides Peer Review (PREV) Unit in April–May 2021.
  • – Review of available literature on toxicity studies conducted with PBA, PBA(OH) and PBAld through an EFSA outsourced project.10
  • – Technical hearing with experts (Rapporteur Member State, Greece) for renewal of active substance lambda-cyhalothrin (December 3, 2021).11

• Tools for data validation and evidence appraisal:

  • – Comparison of regulatory studies protocols (genotoxicity, general toxicity) against OECD Testing Guidelines (TGs) criteria (Appendices A and B).
  • – Critical appraisal of public literature against defined criteria according to OHAT/NTP Risk of Bias tool.12

• Tools for data extraction:

  • – Collection of relevant toxicological endpoints in an Excel file (Annex E).

• Tools for data evaluation:

  • – Evaluation of genotoxicity studies by the EFSA ccWG on Genotoxicity, according to criteria from OECD TGs (Appendix A, Annex C).
  • – Evaluation of absorption, distribution, metabolism and excretion (ADME) and general toxicity studies by the experts of the PPR Panel WG on pyrethroids, according to criteria from OECD TGs (Appendix B, Annex D).
  • – Evaluation of the applicability of read-across and other alternative in silico and non-testing methods by the experts of the PPR Panel WG and according to the Guidance on the establishment of the residue definition for dietary risk assessment (EFSA PPR Panel, 2016).

• Tools for uncertainty analysis:

  • – Listed sources of uncertainties, following principles according to the EFSA Guidance on Uncertainty Analysis in Scientific Assessments (EFSA Scientific Committee, 2018).

Data and methodologies

Data collection

In its mandate, the European Commission asked EFSA, in cooperation with Member States, to collect all available evidence relevant for toxicological characterisation of PBA and PBA(OH).

For this purpose, EFSA approached rapporteur Member States (RMS) responsible for the evaluation of the relevant active substance (a.s.) forming the two respective metabolites to provide EFSA with the original studies/literature articles retrieved from dossiers of the relevant active substances. In parallel, an Industry task force (Pyrethroids Common Metabolites Group (PCMG)) was consulted to collect available data related to PBA and PBA(OH), including newly conducted studies, not yet available to the public and to provide relevant public literature submitted to the Member States in the dossiers for approval or renewal of respective active substances.

For the evaluation of the genotoxicity studies, the EFSA ccWG on genotoxicity was consulted. Mutagenicity in bacterial cells, for which no disagreement was noted in the past, was not considered in this Scientific Opinion. Studies on gene mutation in mammalian cells, as well as chromosome aberration, are evaluated in Appendix A and Annex C.

For general toxicity of PBA and PBA(OH), the PPR Panel WG conducted the assessment. The evaluation of the studies is included in Appendix B and Annex D. In order to compare the toxicological endpoints available for PBA and PBA(OH) with toxicological endpoints of their parent compounds, a comparison file has been created (Annex E).

Additionally to study reports, quantitative structure–activity relationship (QSAR) and read-across documents were submitted. Since experimental genotoxicity data were available for PBA and PBA(OH), no QSAR evaluations for genotoxicity were considered necessary and were not further evaluated. As regards QSAR and read-across for general toxicity, an independent read-across assessment from PBA(OH) and PBAld to PBA has been done by the PPR Panel, since no experimental toxicity studies of longer duration were submitted for this metabolite.

Apart from regulatory studies, data collection considered also public literature. Public literature included in the dossiers has been provided by Member States and PCMG. In addition, EFSA outsourced a review of available literature of toxicity studies conducted with PBA, PBA(OH) and PBAld.

Data validation and evidence appraisal

Relevant regulatory studies (genotoxicity, general toxicity) were validated against the criteria of respective OECD TGs.

The relevance criteria for public literature were restricted to information on intrinsic toxicological properties of metabolites under assessment. Epidemiological studies, biomonitoring investigations and information on toxicokinetic behaviour of parent compounds were excluded.

Reliability criteria for public literature were defined according to OHAT/NTP Risk of bias tool,13 as captured in the review protocol developed by the PPR Panel WG (Annex A to the External Scientific Report (Greene et al., 2021)14).

Data evaluation

Experimental studies for gene mutation and chromosome damage of PBA and PBA(OH) were evaluated by the ccWG on genotoxicity and results are summarised in Appendix A and Annex C.

Studies on general toxicity (regulatory studies and reliable public literature) of PBA, PBA(OH) and PBAld were evaluated by the PPR Panel WG (Appendix B and Annex D). Relevant endpoints (clinical signs, target organs, LD50s, NOAELs, etc.) were listed for the metabolites and respective potential parent molecules in Annex E, to allow direct comparison of toxicity. In order to investigate if PBA and PBA(OH) have also a neurotoxic MoA as their parent compounds, clinical signs were grouped in those addressing general toxicity and those being signs of neurotoxic mechanism. Data (metabolism, toxicological endpoints) for 16 potential parent pyrethroid compounds were collected from the respective regulatory documents (draft assessment reports (DARs)/renewal assessment reports (RARs)/EFSA conclusions/JMPR evaluations).

In order to explore qualitative and quantitative short-term toxicity for PBA, for which no fully acceptable studies of longer duration are available, read-across from PBA(OH) and PBAld (for which recent 28-day rat studies are available) was applied. This has been done by expert judgement, supported by considerations of structural similarity.

Assessment

Metabolism

In in vivo rat toxicokinetic (TK) studies (mostly conducted according to OECD TG 417), information on identities and quantities of metabolites analysed usually 48 h after the administration of a test pyrethroid with a radioactive label on alcohol and acid moieties are contained. Relevant TK and metabolite data have been extracted from the DARs/RARs of the concerned 16 active substances and salient pieces of information shown in Annex B (chemical structures of the 16 active substances and the common metabolites are provided in Annex A).

Evaluation of the pyrethroid metabolism following ester cleavage demonstrated that PBA-based metabolites and conjugates with glucuronic acid, glycine or sulfate are predominant metabolites in the pyrethroid metabolic pathway. There are some exceptions such as etofenprox and esfenvalerate, which produce less PBA metabolites than other pyrethroids. Most in vivo studies present the metabolic profile in variable ways and details, thus hamper the collection of precise information on relative abundances of metabolites or even about their exact identities. Admittedly, also natural biological variability provides an additional difficulty to the assessment.

The recent physiologically based pharmacokinetic (PBPK) simulation studies based on known physiological characteristics of rat and human and on physico-chemical properties of parent pyrethroids, as well as on in vitro metabolism and clearances have provided more reliable information – especially for clearance via different enzymes and pathways rather than formation and abundance of specific metabolites. An extensive series of in vitro and PBPK modelling studies on several common pyrethroids (Song et al., 2019; Hedges et al., 2019a,b, 2020; Mallick et al., 2020a,b) demonstrate that hydrolysis, ester cleavage, is of prime importance and even more so in humans. These data demonstrate that deltamethrin, cis- and trans-permethrin, esfenvalerate, cyhalothrin and cyfluthrin are metabolised principally by hydrolytic enzymes, i.e. microsomal and cytosolic isoforms of carboxylesterases (CES1 and 2), usually constituting >90% of clearance on the basis of simulations. Only esfenvalerate and cis-permethrin were metabolised by CYP enzymes to a significant extent.

On the basis of quantitative and semiquantitative information on pyrethroid metabolism in rats (Annex B), several conclusions on the presence and amounts of PBA and PBA(OH) produced by in vivo metabolism of pyrethroids should be considered:

• For most pyrethroids (acrinathrin, gamma-cyhalothrin, cypermethrin, deltamethrin, fenpropathrin, permethrin), PBA and/or PBA(OH) are not quantified as major metabolites, even if the ester hydrolysis pathway is major. The principal reason is that PBA and/or PBA(OH) are further metabolised into (mainly) sulfo- and glucuronide-conjugates, which constitute the major metabolite fractions excreted.
• For esfenvalerate, fenvalerate and etofenprox, ester hydrolysis as such may be less significant than the hydroxylation of the parent. Also in this case, PBA and PBA(OH) are not quantified as major metabolites.
• For cyfluthrin and beta-cyfluthrin, the parent pyrethroid molecule contains a fluorosubstitution at 4-position of the alcohol portion. Thus, 4-F-PBA and 4-F-PBA(OH) and their conjugates are major metabolites.

Genotoxicity

Studies investigating gene mutation in mammalian cells and in vivo /vitro chromosome damage with PBA and PBA(OH) were assessed by the ccWG on genotoxicity as presented in Appendix A and Annex C. All studies were considered reliable without restrictions.

The in vitro gene mutation assays (murine lymphoma assays (MLA) and Hypoxanthine-guanine-phosphoribosyl transferase (HPRT)) conducted with PBA and PBA(OH) were considered negative with high relevance of the test system and of the study results. Based on this, and also considering the negative results reported in bacterial tests, PBA and PBA(OH) do not raise a concern for gene mutation induction.

In relation to the assessment of chromosome damage:

• the first in vitro micronucleus test (Bohnenberger, 2014) (Chinese hamster V79 cells) conducted with PBA was considered positive, with high relevance of the test system and of the study results.
• the second in vitro micronucleus test (Donath, 2019a) (human peripheral blood lymphocytes) conducted with PBA was considered equivocal, with limited relevance of the test system and of the study results.
• the in vivo micronucleus test (Report 2015/1001101; Report 2017/1021042) (mouse bone marrow) with PBA was considered negative, with high relevance of the test system and of the study results. The additional information on PBA determination and quantification in (mouse) plasma samples (Report 2017/1005933; Report 2017/1005933; Report 2017/1194603; Report 2015/1003984) provided sufficient evidence for systemic exposure to PBA.
• the in vitro micronucleus test (human peripheral blood lymphocytes) conducted with PBA(OH) (Donath, 2019b) was considered negative, with high relevance of the test system and of the study results.
• the in vivo micronucleus test (Report BFI1048) (mouse bone marrow) with PBA(OH) was considered negative, with high relevance of the test system and of the study results. Clinical signs of toxicity and information on PBA quantification in plasma samples provided sufficient evidence for systemic exposure to PBA(OH).

The data available for PBA(OH) do not raise a concern with respect to chromosome damage. Limited evidence for chromosomal damage with PBA was observed in vitro; however, results from the in vivo micronucleus assay were negative with sufficient evidence of bone marrow exposure; therefore, no concern for chromosome damage is concluded for PBA.

Overall, PBA and PBA(OH) do not raise a concern with respect to genotoxicity.

General toxicity profile of PBA and PBA(OH)

Regulatory studies

In two acceptable acute oral toxicity studies in rats and mice (Report 13687; Report AL-460-099), PBA showed LD₅₀ > 2,000 mg/kg bw and no neurotoxicity effects were observed. Toxicity of PBA via intraperitoneal (i.p.) route was investigated in one study (Report AT-80-0217) resulting in LD₅₀ of 154 (m) and 169 (f) mg/kg bw. One acceptable study with PBA(OH) by gavage (Report 155209) and one supplementary study by i.p. (Report AT-80-0217) revealed LD₅₀ > 2,000 mg/kg bw and 745 mg/kg bw, respectively.

For completeness, also acute toxicity studies for PBAld have been evaluated, but not taken into account for the definition of general toxicity profile of PBA and PBA(OH), for which acute toxicity studies were available. Two studies considered as supplementary information (Report 1985/1002156 and Report 1985/1002153) on oral toxicity of PBAld reported the lowest LD₅₀ of 1,222 mg/kg bw; the LD₅₀ by i.p and by inhalation were 415 mg/kg bw (Report AT-80-0217, supplementary information) and 0.27 mg/mL (Report 1984/1002182 and Report 1985/1002160, supplementary information), respectively. Dermal LD₅₀ for PBAld was > 5 g/kg bw in two studies (Report 1985/1002155 and Report 1985/1002154, supplementary information). PBAld was not a skin irritant in rat (Report 1985/1002159 and Report 1985/1002158, supplementary information) or an eye irritant (one study in rabbits) (Report 1985/1002157, supplementary information). Discordant results were obtained in two skin sensitisation studies in Guinea pigs, that were considered not acceptable.

Concerning repeated-dose administration, only three studies, one for PBA (Report CY-470-007), acceptable with limitations as regards neurotoxicity but not acceptable for systemic toxicity, one for PBA(OH) (Report 511694) and one for PBAld (Report 510795), both acceptable, were available. In the 7-day neurotoxicity study via gavage with PBA (Report CY-470-007), the NOAEL for neurotoxicity was set at 375 mg/kg bw per day (highest dose tested). The NOAEL in the 28-day rat study with PBA(OH) (Report 511694) was set at 106.9 mg/kg bw per day. For PBAld (Report 510795), the NOAEL for the 28-day rat study was set at 98.9 mg/kg bw per day. In both studies, the effects considered for NOAEL setting were lower body weight gain and food consumption, as well as liver weight and histopathology findings. The decrease in body weight gain at least in one sex in the 28-day studies with PBAld (Report 510795) and PBA(OH) (Report 511694) was observed at the highest dose during the first days of treatment. The observed findings were not related to neurotoxic mechanism.

Studies with PBA, PBA(OH) and PBAld investigating general toxicity are presented in Appendix B (detailed assessment) and in Annexes D and F.

Published literature

Thirteen studies from published literature (all in vitro) were appraised and assigned to Tier 1 (low RoB). From these studies, four thematically covered interaction with endocrine system, one study was related to reproductive parameters, three studies were experiments on cells related to immune response and five studies were allocated to heterogeneous categories (genotoxicity, hepatotoxicity, cardiac toxicity).

The four studies thematically allocated to interaction with endocrine system dealt with oestrogenic activity. While oestrogenic activity was concluded for PBA in Wielogórska et al. (2015) and Jin et al. (2010), anti-oestrogenic activity for PBA was concluded in Sun et al. (2014). No (anti)oestrogenic activity of PBA was measured in Laffin et al. (2010). In all studies, different assays were explored.

In one study related to reproductive parameters (Yuan et al., 2010), no effects on sperm motility were identified.

Three in vitro studies with cell types related to immunological responses (Wang et al., 2017: murine macrophage cell line; Gabbianelli et al., 2009: polymorphonuclear neutrophils; He et al., 2018: human promyelocytic leukaemia cells) reported some immunotoxic effects for PBA (Gabbianelli et al., 2009; Wang et al., 2017; He et al., 2018) and PBAld (Gabbianelli et al., 2009). These effects were mostly related to cell cycle arrest, generation of reactive oxygen species (ROS), cell viability and apoptosis.

The effects indicated in the Tier 1 ROB studies were all observed in single in vitro systems. They cannot be directly translated in the in vivo situations. Although the design of the 28-day rat studies with PBA and PBAld (OECD TG 407) does not allow for identification of reproductive, endocrine or specific immunotoxic effects, there are no strong indications from the in vitro studies to look for correlates in vivo.

One in vivo study (Uterotrophic assay, Laffin et al., 2010) was assigned by Greene et al. (2021) to Tier 3 (high RoB), based upon one key criterion not fulfilled. This criterion was related to blinding of personnel, which has not been done. However, the PPR Panel considered that blinding of personnel is not critical for the endpoint like uterine weight, which is being assessed by objective methodology (using the balance to measure organ weight). Therefore, the in vivo part of Laffin et al. (2010), related to uterine weight, was finally assigned Tier 1 (low RoB). In this study, no statistically significant increase in uterine weight (which is the OECD TG 440 criteria to declare a positive response) was observed in treated animals at 1, 5 and 10 mg/kg bw per day, compared to control animals. Based on the in vivo study results, it is concluded that PBA was not oestrogenic in adult female Sprague Dawley rats following in vivo treatment. However, several requirements of the OECD TG 440 were not met in this study (see Section 5.5).

Complementary information from bioactivity models

In the ToxCast pathway models (CompTox Chemicals Dashboard, US EPA), PBA was inactive for both oestrogenic and androgenic agonism and antagonism.15 PBA(OH)16 was not screened sufficiently in ToxCast in order to derive oestrogen receptor (ER) or androgen receptor (AR) model scores.

Both PBA and PBA(OH) were inactive in the Collaborative Modelling Project for Androgen Receptor Activity (COMPARA).

The Collaborative Oestrogen Receptor Activity Prediction Project (CERAPP) model includes a combination of multiple models to predict ER activity of a common set of 32,464 chemical structures (i.e. QSAR models and docking approaches). While PBA was predicted inactive in the CERAPP for ER agonism, antagonism and binding, PBA(OH) was predicted to be very weak or weak oestrogen receptor agonists or antagonists and to have weak binding activity. There is no data in the literature; therefore, only the ‘consensus’ QSARs model was populated (due to lack of empirical data).

Read-across analysis

The following lines of evidence were taken into account when concluding on read-across for short-term toxicity of PBA, for which no 28-day rat study is available.

• According to OECD QSAR Toolbox 4.5, the structural similarity of the three compounds is high (79% between PBA and PBA(OH) and 77% between PBA and PBAld).
• Considering the functional groups (aldehyde, carboxylic acid and hydroxy acid), there are no indications for differences in short-term toxicity profile between PBAld, PBA and PBA(OH). Although potential differences in reactivity are recognised (different functional groups), it is not expected that the molecule containing carboxylic acid as functional group would be more toxic than the compound with the aldehyde moiety.
• In the metabolic pathway observed in rats, PBAld is transformed to PBA (acid) and this to PBA(OH) (hydroxy acid). PBA is an intermediate metabolite between PBAld and PBA(OH).
• None of the three metabolites show marked specific neurotoxic effects in available studies, typical for pyrethroid parent compounds.
• Results from 28-day rat studies show comparable reference points (around 100 mg/kg bw per day) and qualitative toxicity (effects on body weight (gain), liver weight and hypertrophy) for PBAld and PBA(OH).

Based upon these arguments, the PPR panel concludes that PBA can be considered to have comparable qualitative and quantitative short-term toxicity as PBAld and PBA(OH).

General toxicity comparison of PBA and PBA(OH) to parent compounds

Toxicity profiles of PBA and PBA(OH) were compared to the parent compounds (Annexes D and E).

Data on acute toxicity after oral and i.p. administration are available for PBA and PBA(OH).

The oral LD₅₀ values exceeded 2,000 mg/kg bw for the PBA and PBA(OH), towards a range from about 50 to > 2,000 mg/kg bw for the parent compounds. The clinical observations derived from the studies with metabolites did not indicate neurotoxicity.

There are not many data available to evaluate the systemic toxicity of metabolites towards parent compounds in the case of repeated dosing. The reports available and acceptable are limited to 28-day studies with PBA(OH) and PBAld. No fully acceptable data in this regard are available for PBA.

In the recently conducted 28-day rat studies with PBA(OH) and PBAld, the NOAELs were 106.9 and 98.9 mg/kg bw per day, respectively, mainly based on body weight decrease and liver toxicity. The observed findings were not related to a neurotoxic mechanism.

Table 1 reports the identified NOAELs and LOAELs for the repeated dose 28-day oral toxicity studies in rat with the concerned 16 parent compounds. Taken together, a range of 4–75 mg/kg bw per day is obtained for LOAEL and of 1–50 mg/kg bw per day for NOAEL.

Table 1 NOAELs and LOAELs in the 28-day oral toxicity rat studies with the concerned 16 parent compounds (sources: DARs/RARs)

Compound (reference)	Study/doses (mg/kg bw per day)	LOAEL (mg/kg bw per day)	NOAEL (mg/kg bw per day)
Acrinathrin (EFSA, 2013)	Rat – 28-day study (corn oil) 5, 25 and 100	5 Mortality and premature sacrifices with lymphocyte depletion in spleen, hypertrophy of zona glomerula in adrenal gland and cortical tubular dilatation and/or degeneration in kidneys. Prostration, decrease in activity, dyspnoea, piloerection, ptyalism and ptosis. Lower body weight gain/weight loss, with lower food intake. Lower absolute and relative spleen weights in M. Lower absolute kidney weight and relative liver weight in M. Higher absolute and relative adrenal gland weights in F. In kidneys of M: acute tubular necrosis, papillary necrosis, cortical tubular degeneration. Splenic lesions in both sexes: lymphocyte depletion in the marginal zone around periarteriolar lymphoid sheaths.	n.d. < 5
Beta-cyfluthrin (EFSA, 2020b)	Rat – 28-day study (cremophor/water) 0.25, 1, 4 and 16	4 Mortality, clinical signs, reduced body weight development, increased liver weight	1
Cyfluthrin (EFSA, 2020b)	No data
Lambda-cyhalothrin (EFSA, 2014c)	Rat – 28-day study (diet) 250 ppm (corresponding to 30 mg/kg bw)	30 ↓ bw gain ↑ hepatic aminopyrine-N-demethylase activity* ↑ SER proliferation* *considered to be an adaptive response to increased liver workload.	n.d
Gamma-cyhalothrin (EFSA, 2014b)	Rat – 28-day study (diet) 0.22/0.85,4.2/4.5, 8.8/8.2 (M/F)	8.8 Signs of toxicity included increased activity and reactivity, crouched posture, splayed hindlimbs, tip-toe gait, knuckling, noisy and laboured breathing, incoordination, salivation and lacrimation. Lower overall weight gain and consistently lower food consumption and clinical chemistry	4.2
Cypermethrin (EFSA, 2018a)	Rat – 28-day (in fact 35) 0, 1.25, 5, 12.5, 37.5, 75	75 Clinical signs of neurotoxicity characterised by piloerection, nervousness, uncoordinated movements, ataxia, splayed gait and hypersensitivity. Body weight was reduced and signs of anaemia were present.	37.5
Alpha-cypermethrin (EFSA, 2018b)	Rat – 6 weeks 0, 25, 100, 200, 400 and 800 ppm	32.2 (400 ppm) ↓ food intake ↓ body weight abnormal gait, ↑ liver weight, haematological findings	16.8 (200 ppm)
Rat – 28-day (open literature) study (soy bean oil) gavage: 0, 4, 8 or 12	12 Increased relative adrenal weight	8
Beta-cypermethrin (EFSA, 2014a)	Rat – 28-day range-finding study (diet): 0, 10, 25, 50, 100	25 reduced food ↓ bw and bwg, ↓ serum glucose, ↓ bwg	10
Rat – 28-day range-finding study (diet): 0, 10, 25, 50 and 100/150	100 Clinical signs, decreased body weight, body weight gain and food consumption and organ weight changes	50
Zeta-cypermethrin (EFSA, 2008a)	Rat – 28-day range-finding study (diet) 50, 100, 300, 600, 900, 1,500 ppm	48 (600 ppm) Transient decreased body weight, decreased body weight gain, decreased food consumption. Sign of neurotoxicity at 900 ppm (69 mg/kg bw per day); mortality at 1500 ppm (102 mg/kg bw per day)	26 (300 ppm)
Deltamethrin (Deltamethrin Rapporteur Assessment Report for public consultation (2018))	Rat – 28-day study (diet as a 1% premix in maize starch) 200 ppm (corresponding to approx. 13 mg/kg bw per day)	13 Lower body weight	n.d.
Esfenvalerate (EFSA, 2014)	No data
Etofenprox (EFSA, 2008b)	No data
Fenpropathrin (JMPR, 2012)	Rat – 28-day range-finding study (diet), 5, 50 or 500 ppm (corresponding to 0.08, 0.40, 4.04 and 38.4 and 0, 0.08, 0.41, 4.40 and 42.8 mg/kg bw per day for M and F, respectively)		38.4 (500 ppm)
4-week range-finding study 500, 700 (1,000 or 1,500) ppm (corresponding to 44 and 65 mg/kg bw per day for M; 51 and 85 mg/kg bw per day for F)	44 (500 ppm) Body tremors, at week 4 increases in red blood cell and neutrophil counts and increased haemoglobin concentration with reduced mean corpuscular volumes. High incidence of mortality or severe clinical signs at 1000 and 1,500 ppm led to termination of these groups on day 3–4	n.d.
Fenvalerate (JMPR, 2012)	No data
Permethrin (EFSA, 2016)	Rat – 28-day study (diet) 0, 200, 500, 1,000, 2,500, 5,000 or 10,000 ppm	2,500 ppm Clinical signs transient, increases in liver weight without histopathological or biochemical changes, reduced body weights in M	110 (1,000 ppm)
Tau-fluvalinate (EFSA, 2010)	No data

Results of short-term toxicity studies available for PBA(OH) and PBAld are reported in Table 2. The individual values of NOAEL were three- to fourfold higher than the mean of parent compounds and eight- to ninefold higher for LOAEL.

Table 2 NOAELs and LOAELs in the oral toxicity 28-day rat studies with the PBA(OH) and PBAld (sources: DARs/RARs)

Compound	Study/doses (mg/kg bw per day)	LOAEL (mg/kg bw per day)	NOAEL (mg/kg bw per day)
PBA(OH)	Rat – 28-day study 0, 28-day oral study in Rats 0, 27.4, 110,1, 474 for M; 25.8, 106.9499.6 for F (Report 511694)	474.0 Decreased body weight (gain), decreased food consumption, liver weight, as well as liver histopathological findings	106.9
PBAld	Rat – 28-day study, 0, 25.8, 98.9 and 474.3 for M and 25.7, 99.7 and 302.4 for F (Report 510795)	302.4 Decreased body weight (gain), decreased food consumption, liver weight as well as liver histopathology findings	98.9

Endocrine activity

There were no indications for interaction of metabolites with the androgenic system. Based upon different observations from in vitro, in vivo and in silico results for oestrogenic activity of metabolites, the following lines of evidence are provided:

• The ToxCast model for oestrogen activity is the regulatory accepted model for prediction of oestrogenic properties (ECHA and EFSA, 2018).

  In the ToxCast model, PBA was concluded inactive for both oestrogenic agonism and antagonism. No information on PBA(OH) are available in ToxCast model. While PBA was predicted inactive in the CERAPP model for oestrogen receptor agonism, antagonism and binding, PBA(OH) was predicted to be very weak or weak oestrogen receptor agonists or antagonists and to have weak-binding activity.

• E-screen assay for PBA revealed no effect on MCF-7 cells proliferation (Laffin et al., 2010). At the same concentrations tested, PBA and two parent compounds, permethrin and beta-cypermethrin, but not PBAld, induced significant MCF-7 cells proliferation (Jin, 2010); however, no cytotoxicity was assessed in this study. Lambda cyhalothrin caused increase in MCF-7 cells proliferation and no cytotoxicity was assessed (Zhao, 2008). Fenpropathrin and deltamethrin were also tested in MCF-7 cells, showing either no or enhanced proliferation, respectively; cytotoxicity was assessed in the study (Andersen, 2002).

  The E-screen assay, originally developed by Soto et al. (1995), was included in the first version of the OECD Conceptual Framework for Testing and Assessment of Endocrine Disrupters published in 2012. However, due to failed validation, it was not included in the updated version of the framework published in 2018 (OECD, 2018) as well as in the ECHA-EFSA guidance for identification of endocrine disruptors (ECHA and EFSA, 2018).

• In the uterotrophic assay with PBA (Laffin et al., 2010), a non-statistically significant increase in uterus weight was observed in animals at the high dose tested (10 mg/kg bw per day). In this study, a lower number of animals per group (4 instead of 6) was used and the maximum tolerated dose was not tested; therefore, the study did not fulfil the requirements from OECD TG 440. In the submitted regulatory 28-day rat studies with PBA(OH) and PBAld, no statistically significant increase in uterine weight was observed up to approximately 500 mg/kg bw per day.

HBGVs derivation

For parent pyrethroid compounds, the NOAELs utilised for derivation of ADI ranged from 0.25 to 7.5 mg/kg bw per day and from 0.8 to 100 mg/kg bw in the case of derivation of ARfD (Table 3).

Table 3 ADI/ARfD values for the concerned 16 parent pyrethroid compounds

	ADI (mg/kg bw per day) Critical study	UF	ARfD (mg/kg bw) Critical study	UF
Acrinathrin	0.01 (rat acute oral neurotoxicity study)	100	0.01 (rat acute oral neurotoxicity study)	100
Beta-cyfluthrin	0.01 (4 weeks rat study)	100	0.01 (4 weeks rat study)	100
Cyfluthrin	0.003 (mouse pharmacology study)	100	0.02 (rat acute oral neurotoxicity study)	100
Lambda-cyhalothrin	0.0025 (rat multigeneration study of cyhalothrin)	200	0.005 (1-year dog study)	100
Gamma-cyhalothrin	0.0012 (from rat multigeneration of cyhalothrin)	400	0.0025 (1-year dog study with lambda-cyhalothrin)	200
Cypermethrin	0.005 (2-year rat study, supported by the DNT study)	100	0.005 (DNT study)	3,000
Alpha-cypermethrin	0.00125 (DNT study)	200	0.00125 (DNT study)	200
Beta-cypermethrin	0.0016 (DNT study)	300	0.0016 (DNT study)	300
Zeta-cypermethrin	0.04 (2-year dog study and 1-year dog)	200	0.125 (rat developmental study and acute neurotoxicity study)	100
Deltamethrin	0.01 (90-day and 1-year dog study, supported by 2 years rat study)	100	0.025 (90-day dog study)	100
Fenpropathrin	0.03 (multigeneration reproduction study in rats, teratogenicity studies in rats and the 1-year feeding study in dogs)	100	0.03 (published study measuring motor activity at the time of peak effects following a single oral dose in rats, teratogenicity studies in rats)	100
Fenvalerate	0.02 (multigeneration reproduction study in rats (esfenvalerate) and long-term toxicity studies (fenvalerate))	100	0.2 (acute oral rat and mouse and acute neurotoxicity study)	100
Esfenvalerate	0.0175 (acute neurotoxicity study in rat)	100	0.0175 (acute neurotoxicity study in rat)	100
Tau-fluvalinate	0.005 (2-year rat study)	100	0.05 (28-day rat neurotoxicity study, rat developmental study)	100
Permethrin	0.05 (1-year dog)	100	1.5 (acute rat neurotoxicity study)	100
Etofenprox	0.03 (2-year mouse study supported by 2-year rat study)	100	1 (rabbit developmental study)	100

For most pyrethroid parent compounds, rat was the most sensitive species and neurotoxic effects the most relevant effects to set HBGVs. Only in few cases reference values were derived based upon dog studies (ARfD for gamma and lambda cyhalothrin and deltamethrin and ADI for zeta-cypermethrin, deltamethrin and permethrin). In the DARs/RARs, this was argued as a result of slightly higher susceptibility of dogs, the dose spacing applied or the use of rat studies as supportive to derive HBGVs. There is no firm evidence that dog is a more susceptible species than rat as regards the toxicity of pyrethroids.

As regards PBA-related metabolites, the overall NOAEL was 100 mg/kg bw per day (considering an NOAEL of 98.9 mg/kg bw per day in the 28-day rat study with PBAld and 106.9 mg/kg bw per day in the 28-day rat study with PBA(OH), based on effects on body weight (gain) and food consumption as well as liver weight and histopathology). This was supported by the BMD analysis (Annex F). According to the EFSA guidance document on residue definition (EFSA, 2016), an additional UF of 10 should be applied for chronic HBGVs if the metabolites are considered qualitatively different from the parent compounds and only short-term study is available for metabolites to compensate for lack of studies with other endpoints (long-term toxicity/carcinogenicity, reproductive toxicity). Applying the UF of 1,000, the ADI for PBA and PBA(OH) is derived at 0.1 mg/kg bw per day.

As regards the ARfD, a value of 1 mg/kg bw is established, derived from the overall NOAEL of 100 mg/kg bw per day (considering the NOAEL of 98.9 mg/kg bw per day in the 28-day rat study with PBAld and the NOAEL of 106.9 mg/kg bw per day in the 28-day rat study with PBA(OH), based on decrease in body weight gain during the first week of treatment) and applying the standard UF of 100.

Uncertainties

The following sources of uncertainty were identified:

• Primary goals of the OECD TG 417 are the description of the mass balance of active substance (urine, faeces, tissues, etc.) and the detection of major metabolites (> 5%), while other important toxicokinetic characteristics and details of metabolite profiles often remain rather undefined and uncertain. These uncertainties are compounded by uneven quality of studies, variable and sometimes obsolete analytics and missing data and inadequate reporting.
• The presence of metabolites in rat metabolism, expressed as percentages, is based on the total radioactivity recovered (the total dose). However, it must be emphasised that often the original sources are unequivocal, and for this reason, metabolite percentages are only rough approximation.
• Comparative in vitro metabolism (CIVM) studies have been employed only sporadically. There is uncertainty in the interspecies differences and especially extrapolation to humans.
• Modified PBA metabolites, to which consumer might be first exposed, may display different toxicological potencies as compared with PBA and PBA(OH).
• The quality and reporting of old studies (conducted 1970–80s), especially if these predate any OECD Guidelines, is a limiting factor in toxicological assessment.
• For most pyrethroid parent compounds, rat was considered the most sensitive species for derivation of HBGV. However, in few cases, HBGV were derived from dog studies. No dog studies are available for PBA and PBA(OH).
• As no 28-day study was available for PBA, read-across from PBA(OH) and PBAld has been conducted.
• There is only expert judgement that the respective functional groups (carboxylic acid, hydroxy group) have comparable toxicity, not substantiated by evidence in this case.
• In the Guidance document on residue definition (EFSA PPR, 2016), no specific consideration is provided on how common metabolites should be assessed.
• Toxicological studies are always subject to inter- and intra-studies variability. Comparing the NOAELs of different studies does not take into account different biological and environmental variables specific for the studies as conducted.
• Only 28-day studies are available with PBA(OH) and PBAld, and no studies of longer duration, assessing different endpoints (long-term toxicity/carcinogenicity, reproductive toxicity).
• The available evidence of oestrogenic properties of PBA and PBA(OH) contains several uncertainties (lack of testing for cytotoxicity in the E-screen assay, failed validation of the E-screen assay, not standard study design for the uterotrophic assay).

It is concluded that the listed standard uncertainties are sufficiently covered by the UFs as applied.

Conclusions

For most pyrethroids, information was made available to the PPR Panel to understand their metabolism in vivo and their transformation to PBA and PBA(OH). Even if the ester hydrolysis is a major rat metabolic pathway, PBA and/or PBA(OH) are rarely quantified as major metabolites.

Based upon available experimental data, PBA and PBA(OH) do not raise a concern with respect to genotoxicity.

PBA and PBA(OH) showed lower acute oral toxicity (based on LD₅₀) and different hazard (no neurotoxic mechanism) compared with their parent pyrethroid compounds.

There are limited data available to evaluate the toxicity of common metabolites towards parent pyrethroid compounds. In the case of repeated dosing, the fully acceptable studies are limited to 28-day rat study for PBA(OH). In addition, a 28-day study with PBAld was considered for read-across purpose.

In the repeated dose studies, the NOAEL was 98.8 mg/kg bw per day for PBAld and 106.9 mg/kg bw per day for PBA(OH), while the NOAELs obtained in the equivalent studies of parent compounds ranged from 1 to 100 mg/kg bw per day. Also in repeated-dose studies, no neurotoxic effects were observed with PBAld and PBA(OH). Based upon read-across considerations, it was concluded that PBA can be considered to have comparable qualitative and quantitative short-term toxicity as PBAld and PBA(OH).

It can therefore be concluded that on the basis of the available (limited) data, PBA and PBA(OH) have different qualitative (no neurotoxic mechanism) and quantitative (higher NOAELs) toxicity compared to the parent compounds. For both metabolites, the ADI has been derived at 0.1 mg/kg bw per day, based upon the overall NOAEL of 100 from the 28-day rat studies with PBAld and PBA(OH) and an UF of 1,000. The ARfD has been derived for PBA and PBA(OH) at 1 mg/kg bw based upon the overall NOAEL of 100 from 28-day rat studies with PBAld and PBA(OH) and UF of 100.

Documentation as provided to EFSA

• Bohnenberger, 2014. Report 2014/1168736. 3-Phenoxybenzoic acid (metabolite of BAS 310 I, Alpha-Cypermethrin): Micronucleus test in Chinese Hamster V79 cells in vitro, 2014.
• Donath, 2019. Report 146489 LLT-0258. In vitro mammalian micronucleus assay in human lymphocytes with analytical standard of PBA(OH), 2019.
• Donath, 2019. Report 146489 LLT-0261. In vitro mammalian micronucleus assay in human lymphocytes with 3-phenoxybenzoic acid, 2019.
• Report 2015/1001101. 3-phenoxybenzoic acid (metabolite of BAS 310 I, Alpha-cypermethrin) Micronucleus Assay in bone marrow Cells of the Mouse, 2014.
• Report 2015/1032402. ANALYTICAL REPORT 3-phenoxybenzoic acid (metabolite of BAS 310 I, Alpha-Cypermethrin): plasma analysis for external uses, 2015.
• Report 2017/1021042. REPORT AMENDMENT Number 1. 3-phenoxybenzoic acid (metabolite of BAS 310 I, Alpha-Cypermethrin): Micronucleus Assay in Bone Marrow Cells of the Mouse, 2017.
• Report BFI1048. 3-(4-Hydroxyphenoxy) Benzoic Acid - Oral (Gavage) Mouse Micronucleus Test, 2021.
• Report 2017/1005933. GLP-020/16 Validation of the Determination of 3-phenoxybenzoic acid (metabolite of BAS 310 I, Alpha-Cypermethrin) in Plasma (Mouse), 2017.
• Report 2017/1005933. GLP-021/16 Concentration Analysis of 3-phenoxybenzoic acid (metabolite of BAS 310 I, Alpha-Cypermethrin) in Plasma (Mouse), 2017.
• Report 2017/1194603. VALIDATION REPORT 3-phenoxybenzoic acid (metabolite of BAS 310 I, Alpha-Cypermethrin) in polyethylene glycol 400 using HPLC-UV, 2018.
• Report 2015/1003984. Analytical report - 3-Phenoxybenzoic acid (metabolite of BAS 310 I, AlphaCypermethrin) – Concentration control analysis in Polyethylene glycol 400, 2015.
• Report 13687. m-phenoxybenzoic acid: acute oral LD50 – Rat, 1995.
• Report CGA55186/0460 (563–001). R41207 (3-PHENOXYBENZOIC ACID): ACUTE ORAL TOXICITY TO RATS, 1981.
• Report AL-460-099. CL 206128: Acute oral toxicity study in mice, 2000.
• Report 155209. Acute oral toxicity (up and down procedure) in the rat with PBA(OH), 2015.
• Report 1985/1002156. m-PBAD-III Lot 1 - Acute oral toxicity study in rats with meta-phenoxybenzaldehyde, 1991.
• Report 1985/1002153. m-PBAD Lot 450 - Acute oral toxicity study in rats (14 day) with M-PBAD lot #450, 1991.
• Report FE-470-006. Toxicity of WL 43775 intermediates: Acute toxicity skin and eye irritancy and skin sensitization of 3- Phenoxybenzaldehyde, 1977.
• Report AT-80-0217. Acute intraperitoneal toxicity of fenvalerate metabolites in mice, 1979. Study assessed by JMPR (2012).17
• Report 1984/1002182. m-PBAD-III - Four hour acute aerosol inhalation toxicity study in rats of M-PBAD III, 1991.
• Report 1985/1002160. m-PBAD-III Lot 1 - Four hour acute aerosol inhalation toxicity study in rats of M-PBAD III, 1991.
• Report 1985/1002155. m-PBAD Lot 450 - Acute dermal toxicity test in rabbits with meta-phenoxybenzaldehyde. Public study, 1991.
• Report 1985/1002154. m-PBAD Lot 1 - Acute dermal toxicity test in rabbits with meta-phenoxybenzaldehyde, 1991.
• Report 1985/1002159. m-PBAD Lot 450: Primary dermal irritation study in rabbits: M-PBAD lot #450. Public study, 1991.
• Report 1985/1002158. m-PBAD-III Lot 1 - Primary dermal irritation study in rabbits: M-PBAD-III lot #1, 1991.
• Report 1985/1002157. m-PBAD Lot 450 - Acute eye irritation test with metaphenoxybenzaldehyde, 1991.
• Report 1979/1001663. m-PBAD - Guinea pig contact dermal irritation/sensitization, 1991.
• Report CY-470-007. A neurotoxicity study on the pyrethroid metabolite 3-Phenoxybenzoic acid (3-PBA), 1979.
• Report 511694. 3-(4-hydroxyphenoxy) benzoic acid - A 28 Day Oral (Diet)Study in Rats, 2021.
• Report 510795. 3-Phenoxybenzaldehyde - A 28 Day Oral (Dietary)Toxicity Study in the Rat, 2020.
• Report 1980/1001707. m-PBAD – Ninety-day subchronic oral toxicity study of meta-phenoxybenzaldehyde in albino rats, 1991.
• Report 1980/1001708. m-PBAD - Histopathologic evaluation of tissues and organs from albino rats in a 90-day subchronic oral dosing study with metaphenoxybenzaldehyde, 1991.
• Voges, 2019. Report 146483 LLT-0259. In vitro mammalian cell gene mutation assay (thymidine kinase locus/TK+/−) in mouse lymphoma L5178Y cells with analytical standard of PBA(OH), 2019.
• Voges, 2019. Report 146486 LLT-0260. In vitro mammalian cell gene mutation assay (thymidine kinase locus/TK+/−) in mouse lymphoma L5178Y cells with 3- phenoxybenzoic acid, 2019.
• Woods, 2021. Report 8459167 FMC-55651. 3-(4-Hydroxy-phenoxy)-benzoic acid [PBA(OH)]: In Vitro HPRT Mutation Test Using Chinese Hamster Ovary Cells, 2021.
• Woods, 2021. Report 8459166 FMC-55652. 3-Phenoxybenzoic acid (PBA): In Vitro Hprt Mutation Test Using Chinese Hamster Ovary Cells, 2021.

Abbreviations

• ADI
• Acceptable daily intake
• ADME
• Absorption, distribution, metabolism and excretion
• ARfD
• Acute reference dose
• BMD
• Benchmark dose modelling
• Bw
• Body weight
• CYP
• Cytochrome P450
• DAR
• Draft assessment report
• (D)RAR
• (Draft) Renewal assessment report
• F
• Females
• HBGVs
• Health-based guidance values
• HPRT
• Hypoxanthine-guanine-phosphoribosyl transferase
• LD50
• Lethal dose, 50%
• LOAEL
• Lowest observed adverse effect level
• M
• Males
• MLA
• Murine lymphoma assay
• MTD
• Maximum tolerated dose
• NOAEL
• No observed adverse effect level
• TG
• Testing guidelines
• TK
• Toxicokinetic
• UF
• Uncertainty factor
• WG
• Working group

Appendix A – Genotoxicity studies

A.1 Introduction

The EFSA cross-cutting working group (ccWG) on genotoxicity was asked to review the available evidence and provide advice on the genotoxic potential of the common pyrethroid metabolites 3-phenoxybenzoic acid (PBA) and 3-(4′-hydroxyphenoxy)benzoic acid (PBA(OH)), in the context of a mandate (M-2021-00031) received by the Plant Protection Products and their Residues (PPR) Panel from the European Commission for the assessment of the toxicity profile of PBA and PBA(OH).

Specifically, the ccWG on genotoxicity was asked the following questions:

• Based upon available evidence, can genotoxicity of common metabolites PBA and PBA(OH) be reached?

  • ○Do PBA and PBA(OH) induce gene mutation in mammalian cells?
  • ○Do PBA and PBA(OH) induce numerical and structural chromosome aberrations (in vitro/in vivo)?

Study reports on gene mutation and chromosome aberration, collected by the Pesticides Peer Review (PREV) Unit through Member States and industry18 consultation, were made available to the ccWG genotoxicity.

A.2 Data and methodologies

The assessment of the data has been conducted independently by the ccWG genotoxicity and submitted to the PPR Panel for its consideration and decision.

Evaluation of data quality for hazard/risk assessment includes evaluation of reliability and relevance (Klimisch et al., 1997; OECD, 2005; ECHA, 2011; EFSA Scientific Committee, 2011). Accordingly, the reliability was scored using numerical values, where 1 corresponded to ‘Reliable without Restriction’, and the relevance of the test system and study results was categorised into high, limited or low relevance. The assessment of the studies was performed in line with the EFSA Scientific Committee Guidance on genotoxicity testing strategies (EFSA Scientific Committee, 2011), and follow up publications (EFSA Scientific Committee, 2017, 2021). Furthermore, potential minor and/or major deviations from relevant OECD test guidelines were also considered. A tabular format was used to transparently structure the outcome of the evaluations and to provide a possibility to consider the relevance of study results in a weight-of-evidence approach. The studies were grouped based on genetic endpoints or test systems. After being evaluated by the ccWG genotoxicity, the results were presented as positive, negative, equivocal or inconclusive (Table 1 under Annex C).

A.3 Assessment

A.3.1 Endpoint: gene mutation in mammalian cells

Voges, 2019a

In vitro gene mutation assay, locus TK+/− (Mouse lymphoma L5178Y cells) with PBA

In Voges (2019a), the test substance PBA (CAS 3739-38-6; purity 99.6%) was tested for mutagenicity at the thymidine kinase locus in mouse lymphoma cells (L5178Y) following the OECD TG 490 (2016) under good laboratory practice (GLP) conditions.

The test item was tested without metabolic activation up to 2.5 mM (4-h exposure) and 0.7 mM (24-h exposure) and with metabolic activation up to 2.6 mM (4-h exposure).

Statistically significant increases in mutant frequency over the negative control and a positive concentration response trend were reported in both experiments with metabolic activation. Moreover, a positive concentration response trend was reported in one experiment without S9. However, the global evaluation factor (GEF) was not exceeded in any experimental condition, and therefore, in line with OECD TG 490, the overall outcome of the study was considered negative.

The study was considered reliable with high relevance of the test system and overall result.

Woods, 2021a

In vitro gene mutation assay, locus hprt (CHO-K1 cells) with PBA

In Woods (2021a), the test substance PBA (CAS 3739-38-6; purity 99.9%) was tested for mutagenicity at the hprt locus in CHO-K1 cells following the OECD TG 476 (2016) under GLP conditions.

The cells were exposed to the test item for 3 h at concentrations up to a maximum of 2,000 μg/mL (corresponding approximately to 9.37 mM) without metabolic activation and 1,000 μg/mL (corresponding approximately to 4.67 mM) with metabolic activation.

The test was considered negative as no statistically significant increase in mean mutant frequency was observed in any experimental condition.

The study was considered reliable with high relevance of the test system and overall result.

Voges, 2019b

In vitro gene mutation assay, locus TK+/− (Mouse lymphoma cells L5178Y) with PBA(OH)

In Voges (2019b), the test substance PBA(OH) (CAS 35065–12-4; purity 97.2%) was tested for mutagenicity at the thymidine kinase locus in Mouse lymphoma cells (L5178Y) following the OECD TG 490 (2016) under GLP conditions.

The test item was tested without metabolic activation at concentrations up to 2.6 mM (4-h exposure) and 1.0 mM (24-h exposure) and with metabolic activation up to 3.3 mM (4-h exposure).

Statistically significant increases in mutant frequency over the negative control and a positive concentration response trend were reported in both experiments, with and without metabolic activation. However, GEF was not exceeded in any experimental condition, and therefore, in line with OECD TG 490, the overall outcome of the study was considered negative.

In both experiments, with and without metabolic activation, at the highest concentrations, the authors reported an increase of more than 40% in the number of small colonies, that could indicate potential clastogenic effects. However, the GEF was not reached.

The study was considered reliable with high relevance of the test system and overall result.

Woods, 2021b

In vitro gene mutation assay, locus hprt (V79 cells) with PBA(OH)

In Woods (2021b), the test substance PBA (CAS 35065–12-4; purity 98%) was tested for mutagenicity at the hprt locus in V79 cells following the OECD TG 476 under GLP conditions.

The test item was tested up to a maximum concentration of 2,000 μg/mL (corresponding approximately to 8.69 mM) for 3 h, with and without metabolic activation.

In the first experiment without metabolic activation, a positive concentration-response trend was reported; however, this observation was not confirmed in the second experiment. The experiment with metabolic activation gave negative results, and so the test item was considered not mutagenic under the experimental conditions used.

The study was considered reliable with high relevance of the test system and overall result.

A.3.1.1 Conclusions on gene mutation studies

Overall, also considering the negative results reported in bacterial tests, PBA and PBA(OH) do not raise a concern for gene mutation induction.

A.3.2 Endpoint: chromosome damage

Bohnenberger, 2014

In vitro Micronucleus test (Chinese hamster V79 cells) with PBA

In Bohnenberger (2014), the test substance PBA (CAS 3739-38-6; purity 100%) was tested for the capacity to induce chromosome damage with an in vitro Micronucleus test in Chinese hamster V79 cells following the OECD TG 487 (2010) under GLP conditions.

A short treatment (4 h + 20 h of recovery) in the absence and presence of metabolic activation was applied at concentrations ranging from 535.5 to 2,142 μg/mL (corresponding approximately to 2.50–10 mM). A continuous treatment (24 h + 0 h of recovery) without metabolic activation was carried out using six concentrations ranging from 133.9 to 803.3 μg/mL (corresponding approximately to 0.62–3.75 mM).

Statistically significant increases in the frequencies of micronuclei in mononucleated cells were observed after short treatment with and without metabolic activation. The increases were not concentration-related and not clearly reproducible in separate experiments. Therefore, they were considered not biologically relevant.

A statistically significant increase in the frequency of micronuclei outside the historical control range and reproducible was observed following 24-h continuous treatment without recovery. At the highest concentrations, a strong reduction in cell proliferation was observed with concurrent decrease in the frequency of micronuclei. A clear concentration-response relationship was not observed, which was possibly due to the severe cytotoxicity at the higher concentrations. The test was considered positive after continuous treatment without metabolic activation.

The study was considered reliable with high relevance of the test system and overall result.

Donath, 2019a

In vitro micronucleus test (human peripheral blood lymphocytes) with PBA

In Donath (2019a), the test substance PBA (CAS 3739-38-6; purity 99.6%) was tested for the capacity to induce chromosome damage with an in vitro micronucleus test in human peripheral blood lymphocytes following the OECD TG 487 (2010) under GLP conditions.

PBA was tested up to 9 mM applying a short treatment (4 h + 40 h of recovery) in the absence and presence of metabolic activation and up to 1.75 mM with a continuous treatment without metabolic activation (44 h + 0 h of recovery). Cytochalasin B was used to focus the analysis of micronuclei in binucleated cells.

The test was considered equivocal after short treatment with and without metabolic activation because of a statistically significant concentration-related response (trend test). Severe cytotoxicity was observed following continuous treatment without metabolic activation; no increase in the frequency of micronuclei was reported in this experimental condition.

The study was considered reliable with limited relevance of the result.

Report 2015/1001101, Report 2017/1021042

In vivo micronucleus test (bone marrow), Mouse NMRI with PBA

In Report 2015/1001101 and Report 2017/1021042, the test substance PBA (CAS 3739-38-6; purity 100%) was tested for the capacity to induce chromosome damage with an in vivo micronucleus test in the bone marrow of Mouse NMRI (7 M/group, 5 M/controls), following the OECD TG 474 (2014) under GLP conditions.

Animals were treated by gavage with a single dose of PBA at 500–1000 and 2000 mg/kg bw and sacrificed 24 and 48 h after dosing. Few clinical signs of toxicity related to treatment were observed. No biologically relevant increase of micronuclei was observed. No evidence of toxicity to the target tissue was detected. The results reported in Report 2015/1032402 showed that the test material was identified in plasma, but not quantified. However, additional evidence was available on a quantitative analysis of PBA concentration in plasma from mice (Report 2017/1005933). The study was performed in compliance with GLP applying UHPLC–MS–MS, suitable for determination of PBA in plasma as demonstrated in a previous separate validation study (Report 2017/1005933). The calibration curve and the calibration control met the specifications of the acceptance criteria (details reported in Table 2 under Annex C). Following treatment of mice with 2000 mg/kg bw PBA, concentrations ranging from 89.7 to 168.5 mg/L were detected in six samples of plasma. These data provided sufficient experimental evidence for the presence of PBA in plasma.

In addition, an analytical report was available on the concentration control analysis of PBA in the vehicle polyethylene glycol 400 (PEG400) by HPLC-UV (Report 2015/1003984). In a separate validation study (Report 2017/1194603), it was demonstrated that HPLC-UV was an appropriate analytical method to determine PBA in PEG 400. The GLP study Report 2015/1003984 showed that in the negative vehicle samples, no test item was detected.

In conclusion, the micronucleus in vivo test was considered negative with high relevance of the test system and high relevance of the study result.

Donath, 2019b

In vitro micronucleus test (Human peripheral blood lymphocytes) with PBA(OH)

In Donath (2019b), the test substance PBA(OH) (CAS 35065–12-4; purity 97.2%) was tested for the capacity to induce chromosome damage with an in vitro micronucleus test in human peripheral blood lymphocytes following the OECD TG 487 (2010) under GLP conditions.

PBA(OH) was tested up to 10 mM applying a short treatment (4 h + 40 h of recovery) in the absence and presence of metabolic activation and up to 2.0 mM with a continuous treatment without metabolic activation (44 h + 0 recovery). Cytochalasin B was used to focus the analysis of micronuclei in binucleated cells.

A statistically significant increase in the frequency of micronuclei was observed after short treatment in the absence of metabolic activation. The increase was not concentration-related (trend test not statistically significant) and was within the range of negative historical controls. On this basis, the increase of micronuclei was considered not biologically relevant and the result negative.

The study was considered reliable with high relevance of the test system and overall result.

Report BFI1048

In vivo micronucleus test (peripheral blood reticulocytes) with PBA(OH)

In Report BFI1048, the test substance PBA(OH) (CAS 35065–12-4; purity 98%) was tested for the capacity to induce chromosome damage with an in vivo micronucleus test in the peripheral blood reticulocytes of Crl:CD-1 Mice (6 M/group) following the OECD TG 474 (2014) under GLP conditions.

Animals were treated by gavage with PBA(OH) at 500, 1000, 1,250 and 2000 mg/kg bw. Two administrations were given 24 h apart, and blood samples were taken 48 h after the second dose. The analysis of micronuclei was performed up to 1,250 mg/kg bw because mortality was observed at 2000 mg/kg bw. Clinical signs of toxicity were reported at 1250 mg/kg bw. No evidence of toxicity of the target tissue was reported since no changes in the percentage of reticulocytes were observed. No increase in the frequency of micronuclei was detected by flow cytometric analysis of peripheral blood reticulocytes. The presence of PBA(OH) in plasma was confirmed by LC–MS/MS analysis. Based on these results, the test was considered negative.

The study was considered reliable with high relevance of the test system and overall result.

A.4 Conclusions

In relation to the assessment of genotoxicity, PBA did not induce gene mutations in mammalian cells. Limited evidence of chromosomal damage was observed in vitro. The results from an in vivo micronucleus assay were negative with sufficient evidence of bone marrow exposure. Therefore, there is no concern for genotoxicity.

The data available for PBA(OH) do not raise a concern with respect to genotoxicity.

Appendix B – General toxicity studies

B.1 Introduction

The WG, in the context of a mandate (M-2021-00031) received by the Plant Protection Products and their Residues (PPR) Panel from the European Commission, was asked to the assess of the toxicity profile of PBA and PBA(OH).

B.2 Data and methodologies

The criteria for acceptance of the studies were in agreement with guidance/GLP in place at date of study conduction, availability of information such as purity of the chemical tested, characteristics of the animals and their randomisation, route of administration, statistical approach applied, if any, and type of study.

In principle, studies with no information about the purity of the chemical tested were considered as ‘supplementary’. Studies with GLP deviations were considered as ‘acceptable with limitations’. Studies with severe limitations (e.g. GLP/GL deviations and no information on purity, poor reporting, etc.) were considered as ‘not acceptable’.

Clinical signs were attributed either to general toxicity (A) or to neurotoxic effects (B). Clinical signs related to pyrethroid caused neurotoxicity and reported in acute toxicity studies of pyrethroids (obtained from DARs/RARs) were mainly tremors, convulsions, salivation and choreoathetotic movements (Annex E).

For the 28-day rat study with PBAld (Report 510795), in addition to an NOAEL setting, BMD analysis has been conducted for tentatively most sensitive endpoints (Annex F).

A tabular format (Annexes D and F) was used to transparently structure the outcome of the evaluations and to provide a possibility to consider the relevance of study results in a weight-of-evidence approach.

B.3 Assessment

B.3.1 Endpoint: Acute and local toxicity studies

Results of acute and local toxicity studies conducted with PBA, PBA(OH) and PBAld have been summarised in Annex D (Table 1).

B.3.2 Endpoint: Repeated administration

Information and observations from the repeated administration studies conducted with PBA, PBA(OH) and PBAld have been summarised in Annex D (Table 2). Study summaries are provided as follows.

Report CY-470-007

Neurotoxicity study in rat with PBA

PBA (25, 77, 375 mg/kg bw) was administered to Wistar rats by oral gavage for 7 consecutive days. Animal were sacrificed 3 weeks from the starting of treatment. The 77 mg/kg bw dose is equivalent of a neurotoxic oral dose of the parent compound, cypermethrin (WL 43467), tested in the same experiment (150 mg/kg bw per day).

No information on followed test guidelines is reported. No individual values for measured parameters are included in the study report.

In addition to clinical signs and body weight, beta-galactosidase and beta-glucuronidase were measured in the sciatic/posterior tibial nerves and trigeminal ganglia.

No signs of general toxicity were reported, although one female died in each dosed group and one male in the top dose group. Body weight was reported only in a figure, showing slight decrease at the highest dose in both sexes, and in females at 77 mg/kg.

Sparse axonal degeneration was observed only for the parent compound cypermethrin.

An NOAEL of 375 mg/kg bw (highest tested dose) was established for males and females.

The study was considered acceptable with limitations for neurotoxicity but not acceptable for systemic toxicity.

Report 511694

28-day oral toxicity study in rat with PBA(OH)

Wistar male and female rats were provided with diet ad libitum containing PBA(OH) (98%) at 300, 1200 or 6,000 ppm (corresponding to 27.4, 110.1 and 474.0 mg/kg bw per day for males and 25.8, 106.9 and 499.6 mg/kg bw per day for females) for 28 days. The control group received blank diet. The study was conducted under GLP conditions and according to OECD 407 (2018) and OPPTS 870.3050 (2000).

There were no deaths during the course of this study. There were no clinical observations, ophthalmology, coagulation, urinalysis or gross (macroscopic) pathology findings related to administration of PBA(OH).

At 6,000 ppm, a decrease in body weight (> 10%, not statistically significant) and body weight gain (> 10%, statistically significant in males only, starting from the first days of treatment) was observed in both sexes. Slightly lower food consumption was also noted in both sexes (≥ 10% at the end of the study).

Haematological findings at 6,000 ppm in males included statistically significant increase in lymphocytes (+29%) and large unstained cells (+9%). These findings were not observed in females.

Clinical chemistry findings for males at 6000 ppm included statistically significant increased albumin (+9%), statistically significantly decreased globulin levels (−18%) and consequently, statistically significantly increased albumin–globulin ratio (+34%). Females receiving 6,000 ppm showed statistically significant increased urea (+41%).

Males receiving 1,200 ppm or 6,000 ppm had statistically significant higher phosphate levels (56% and 62%, respectively) when compared with the controls, whereas females at 1,200 ppm had a statistically significant decrease in phosphate levels at 1,200 ppm but not at 6,000 ppm. In the absences of any other findings at 1,200 ppm and given the wide range of phosphate values across all study groups including controls, the increased mean phosphate level noted for both males and females at 1,200 ppm are considered to be unrelated to treatment.

Pathology findings at 6,000 ppm were limited to the liver and included mean body weight adjusted liver weight increases of approximately 12% and 14%, relative to control values, for males and females, respectively. Histologically, an increased incidence of diffuse hepatocellular hypertrophy (minimal) was observed at 6,000 ppm for males (2/10) and females (5/10).

Changes in functional observational battery (FOB) and motor activity at 6,000 ppm were not statistically significant, occurred only at the maximum tolerated dose (MTD) and were not considered as evidence of neurotoxicity.

Based on lower body weight (gain), lower food consumption values and liver weight and histopathology findings at 6,000 ppm, the NOAEL was established at 1,200 ppm (equivalent to 110.1 mg/kg bw per day in males and 106.9 mg/kg bw per day in females).

The study was considered acceptable and reliable.

Report 1980/1001707 (Histopathology: Report 1980/1001708)

90-day oral toxicity study in rat with PBAld

PBAld (97%) was administered by gavage to Sprague–Dawley rats at 50, 150 or 300 mg/kg bw per day over a period of 90 days.

Study was predated to the OECD guideline, but comparable to OECD No. 408, equivalent to EC test method B.26 Reg. EC n°440/2008. No GLP conditions.

Six animals (2M/4F) in control group, 4 (2M/2F) in 50 mg/kg bw per day group, 3 (1M/2F) at 150 mg/kg bw per day group, 6 (2M/4F) at 300 mg/kg bw per day succumbed before study termination due to mis-gavaging or aspiration of test substance.

It is not clear from the report if the number of examined animals reported in the gross lesion tables takes into account also these deaths.

The large intestine showed evidence of parasitic infection (nematodes), mainly in colon (incidence of 20% and 15% in males and females, respectively, at 300 mg/kg bw per day, and 10% and 20% in control males and females, respectively), the liver contained evidence of parasite migration (inflammation, focal granulotomas).

Based on the poor quality of reporting and unclear infectious status of animals, the study is considered not acceptable.

Report 510795

28-day oral toxicity study in rat with PBAld

Han Wistar male and female rats were provided with diet containing PBAld (95%) at 0, 300, 1,200, 4,000 (females only) or 6,000 (males only) ppm for at least 28 days, corresponding to dosages of 25.8, 98.9 and 474.3 mg/kg bw per day for males and 25.7, 99.7 and 302.4 mg/kg bw per day for females.

The study was performed in accordance with the OECD Guideline 407, EPA Health Effects Test Guideline OPPTS 870.3050 and GLP.

There were no deaths during the course of this study.

There were no clinical observations, ophthalmoscopy findings, changes in functional observations or urinalysis considered to be related PBAld.

Decrease in food consumption (> 10%) was observed at 6,000 and 4,000 ppm throughout the course of the study (statistical analysis was not performed).

Body weight (gain) decrease (> 10%) was observed at 6,000 and 4,000 ppm starting from the first days of treatment. Body weight decrease was statistically significant in females only. Body weight gain decrease was statistically significant in both sexes and throughout the entire study.

A statistically significantly increased relative liver weight was observed for the 6,000 ppm males (27%), accompanied by minimal to mild centrilobular hepatocellular hypertrophy in all animals. A similar trend in liver weight increase was also present in the high-dose females, but was of a lower magnitude (5%) and without statistical significance. Minimal severity grade centrilobular hepatocellular hypertrophy was observed in two females at 1,200 and all females at 4,000 ppm.

A lower absolute, body weight adjusted and body weight relative values for ovary was observed from 1,200 ppm females.

Slight changes in some haematology, coagulation, clinical chemistry parameters were noted at the top dose (males: −19% fibrinogen, +8% albumin, −6% total protein, −30% globulin, +56% albumin/globulin ratio, −40% cholesterol, −5% Ca; females: −24% cholesterol).

The study was considered acceptable and reliable. The NOAEL derived from this study is 1,200 ppm (98.9/99 mg/kg bw per day in males and females, respectively), due to the decreased body weight (gain) accompanied by decreased food consumption, liver weight and histopathology findings.

To confirm the reference point, a BMD modelling was performed for effects on body weight gain, ovary weight and liver findings observed in females at 1,200 ppm (see Annex F). The BMDs were: 104/131 mg/kg bw per day for body weight and 96.7/83.4 mg/kg bw per day for liver hypertrophy, in males and females, respectively. No BMDL could be identified for changes in ovary weights (absolute, relative and adjusted) because of lack of dose–response.

Based on the findings above, the NOAEL of 1,200 ppm in males and females, which corresponds to overall achieved dosages of 98.9 and 99.7 mg/kg bw per day, respectively, was confirmed.

B.4 Conclusions

PBA, PBA(OH) and PBAld showed lower acute oral toxicity (LD₅₀) than the parent pyrethroid compounds. Information on dermal LD₅₀ and LC₅₀ is available only for 3-Phenoxybenzaldehyde.

There is no information on local toxicity with the exception of PBAld, which is either not or moderate skin irritant and not an eye irritant.

There are not many data available to evaluate the toxicity of metabolites towards parent compounds and in the case of repeated dosing the acceptable studies are limited to 28 days, and only with PBA(OH) and PBAld.

The NOAELs for PBA(OH) and PBAld in 28-day rat studies were close to 100 mg/kg bw per day, while the NOAELs obtained in the equivalent studies of parent compounds ranged from 0 to 100 with a mean of 20–30 mg/kg bw per day.

It can therefore be concluded that on the basis of the available data, the metabolites considered have a lower toxicity compared to the parent compounds.

List of Annexes

Annexes can be found in the online version of this output (‘Supporting information’ section):

Annex A – Chemical names and structures

Annex B – Pyrethroids metabolism

Annex C – Genotoxicity studies with pyrethroid metabolites

Annex D – General toxicity studies with pyrethroid metabolites

Annex E – Pyrethroids and common metabolites endpoints

Annex F – Benchmark Dose Analysis of selected endpoints from the 28-day oral study on 3-phenoxybenzaldehyde (Report 510795)

References

Andersen HR, Vinggaard AM, Rasmussen TH, Gjermandsen IM and Bonefeld‐Jørgensen EC, 2002. Effects of Currently Used Pesticides in Assays for Estrogenicity, Androgenicity, and Aromatase Activity in Vitro. Toxicology and Applied Pharmacology, 179, 1–12. [DOI: https://dx.doi.org/10.1006/taap.2001.9347]