JUDGMENT OF THE GENERAL COURT (First Chamber)

22 November 2023 (*)

(REACH – Substance homosalate – Exclusive use for the manufacture of cosmetic products – Checking of the conformity of registrations – Request for additional toxicity studies – Article 41 of Regulation (EC) No 1907/2006 – Prohibition of animal testing – Article 18(1)(d) of Regulation (EC) No 1223/2009 – Article 2(4)(b) and Article 14(5)(b) of Regulation No 1907/2006 and Section 3 of Annex XI thereto – Obligation to state reasons – Manifest error of assessment – Error of law)

In Case T‑656/20,

Symrise AG, established in Holzminden (Germany), represented by R. Cana and E. Mullier, lawyers,

applicant,

supported by

Cruelty Free Europe (CFE), established in Brussels (Belgium), represented by S. Vandamme, V. McClelland, lawyers, and P. Moser, Barrister,

by

European Federation for Cosmetic Ingredients (EFfCI), established in Brussels (Belgium), represented by R. Cana and E. Mullier, lawyers,

and by

PETA International Science Consortium Ltd, established in London (United Kingdom),

and

PETA Science Consortium International eV, established in Stuttgart (Germany),

represented by R. Dereškevičiūtė, lawyer, D. Scannell and S. Love, Barristers,

interveners,

v

European Chemicals Agency (ECHA), represented by W. Broere, L. Bolzonello and A. Deloff-Bialek, acting as Agents,

defendant,

THE GENERAL COURT (First Chamber),

composed, at the time of the deliberations, of H. Kanninen, President, M. Jaeger (Rapporteur) and N. Półtorak, Judges,

Registrar: I. Kurme, Administrator,

having regard to the order of 23 February 2021, Symrise v ECHA (T‑656/20 R, not published, EU:T:2021:99),

having regard to the written part of the procedure, in particular:

–        the orders of 11 August 2021 granting CFE, PETA International Science Consortium, PETA Science Consortium International and EFfCI leave to intervene in support of the form of order sought by the applicant;

–        the decision of EFfCI not to lodge a statement in intervention,

having regard to EFfCI’s waiver of its right to participate in the hearing,

further to the hearing on 22 November 2022,

gives the following

Judgment

1        By its action under Article 263 TFEU, the applicant, Symrise AG, seeks annulment of Decision A-009-2018 of the Board of Appeal of the ECHA of 18 August 2020 (‘the contested decision’), relating to the compliance check of its registration dossier for the substance homosalate (‘the substance at issue’), dismissing its action against the decision of 13 March 2018 by which the ECHA had requested it to provide, inter alia, studies on vertebrate animals (‘the initial decision’).

 Background to the dispute

2        The applicant imports the substance at issue for which it is the lead registrant under Regulation (EC) No 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/EEC, 93/67/EEC, 93/105/EC and 2000/21/EC (OJ 2006 L 396, p. 1 and corrigendum OJ 2007 L 136, p. 3) (‘the REACH Regulation’).

3        The substance at issue is an organic single-component chemical substance used exclusively as an ingredient in cosmetic and personal care products as an ultraviolet filter.

4        The substance at issue is listed in Annex VI to Regulation (EC) No 1223/2009 of the European Parliament and of the Council of 30 November 2009 on cosmetic products (OJ 2009 L 342, p. 59) (‘the Cosmetics Regulation’) as an ultraviolet filter permitted in cosmetic products at a maximum concentration of 10% in the ready-for-use preparation.

5        On 2 November 2016, the ECHA carried out a compliance check of the applicant’s registration dossier, on the basis of Articles 41 and 50 of the REACH Regulation.

6        On 17 November 2016, the ECHA submitted its proposal for an initial decision to the applicant, in which it was requested to conduct a subchronic toxicity study (‘the OECD 408 Study’), a standard prenatal development toxicity study (‘the OECD 414 Study’), an extended one-generation toxicity study (‘the OECD 443 Study’) and a study to identify degradation products. On 9 January 2017, the applicant submitted its observations on the proposed decision. As such, it submitted adaptations concerning the OECD 408, OECD 414 and OECD 443 studies, on the basis of Section 1.2 of Annex XI to the REACH Regulation. With regard to the study to identify degradation products, the applicant stated that it was prepared to carry it out under certain conditions.

7        On 7 September 2017, the ECHA notified the competent authorities of the Member States of its proposal for an initial decision, in accordance with Article 51(1) of the REACH Regulation. The competent authorities of two Member States in turn proposed amendments, in accordance with Article 51(2) of the REACH Regulation.

8        On 10 November 2017, the applicant submitted its comments on those proposed amendments, in accordance with Article 51(5) of the REACH Regulation.

9        On 13 March 2018, the ECHA adopted the initial decision, following the unanimous agreement of the ECHA Member State Committee and in accordance with Article 51(6) of the REACH Regulation. In the initial decision, the ECHA requested the applicant to carry out four studies, while rejecting the adaptations proposed by the latter. The first, the OECD 408 Study, was to be carried out via the oral route on rats or rabbits. The second, the OECD 414 Study, and the third, the OECD 443 Study, were to be performed via the oral route on rats. Finally, the fourth was a study to identify degradation products. Nevertheless, the ECHA gave the applicant the opportunity, in accordance with Annexes VI to XI to the REACH Regulation, to propose other adaptations to the requested studies, justifying them scientifically, in accordance with the general rules of Annexes XI and VI to X to the REACH Regulation. The initial decision set a time limit of 42 months and 7 days within which that information had to be provided, expiring on 20 September 2021.

10      On 12 June 2018, the applicant brought an action against the initial decision before the Board of Appeal of the ECHA and sought its annulment in so far as it required it to carry out the OECD 408, OECD 414 and OECD 443 studies and in so far as it made the production of those studies subject to a time limit of approximately 42 months.

11      On 18 August 2020, the Board of Appeal issued the contested decision, dismissing, in essence, the applicant’s action against the initial decision.

 Forms of order sought

12      The applicant, supported by PETA International Science Consortium Ltd and PETA Science Consortium International eV (together ‘PETA’), claims that the Court should:

–        annul the contested decision;

–        order the ECHA to pay the costs.

13      Cruelty Free Europe (CFE) claims that the Court should annul the contested decision.

14      The ECHA contends that the Court should:

–        dismiss the action;

–        order the applicant to pay the costs.

 Law

15      In support of its action, the applicant relies on three pleas in law. The first plea alleges errors of assessment and interpretation of the REACH Regulation committed by the ECHA in requiring tests on vertebrate animals to be carried out. The second plea alleges a number of manifest errors of assessment by the ECHA in requiring the OECD 443 Study to be conducted via the oral route and a breach of the duty to state reasons. The third plea alleges a number of errors and infringements committed by the ECHA in setting the time limit for submitting the required information.

 The first plea in law, alleging errors of assessment and interpretation committed by the ECHA in requiring tests on vertebrate animals to be carried out

16      In support of its first plea in law, the applicant puts forward three parts, the first alleging errors of assessment and interpretation of the REACH Regulation committed by the ECHA in requiring vertebrate animal testing, the second alleging an error of assessment committed by the ECHA in considering that the risk associated with worker exposure was not covered by the Cosmetics Regulation and the third alleging an error of assessment committed by the ECHA in failing to take account of the fact that the safety of the substance at issue had been assessed under the Cosmetics Regulation.

17      By the first part, the applicant claims that the ECHA infringed Article 2(4)(b) and recital 13 of the REACH Regulation by requiring tests to be carried out on vertebrate animals in respect of the substance at issue, which is used exclusively in cosmetic products.

18      First of all, the applicant claims that Article 2(4)(b) of the REACH Regulation does not exclude cosmetic ingredients from its scope, but that that scope is ‘without prejudice’ to vertebrate animal testing, which falls within the scope of the Cosmetics Regulation. Article 18(1)(d) of the Cosmetics Regulation lays down a ban on animal testing, with no exceptions for vertebrate animals, while Article 18(2) of that regulation lays down a complex procedure allowing the European Commission to authorise derogations from the bans in certain specific cases and under strict conditions.

19      Therefore, according to the applicant, the only logical interplay between the relevant rules of the REACH Regulation and those of the Cosmetics Regulation is that information with a view to establishing the safety of cosmetic ingredients is mandatory under the REACH Regulation only if that safety requirement does not fall within the scope of the Cosmetics Regulation. Thus, the information requirements of the REACH Regulation apply to a substance, used exclusively as a cosmetic ingredient, only when they do not concern tests to establish effects on human health.

20      In addition, the applicant claims that the safety of cosmetic ingredients from the point of view of environmental protection does not fall within the scope of the Cosmetics Regulation and, therefore, that vertebrate animal testing to establish the safety of a substance for the environment do not fall within its scope either, since the REACH Regulation applies not ‘without prejudice’ but directly and fully, even if the substance at issue is used exclusively as a cosmetic ingredient.

21      The applicant states that, on numerous occasions during the procedure leading to the adoption of the initial decision, it expressly set out its concerns regarding the studies on vertebrates, since the substance at issue is used exclusively in cosmetic products.

22      The applicant considers that the applicability of the Cosmetics Regulation is a question of law and that the REACH Regulation does not provide for any possibility or obligation to make an adaptation of the registration dossier in order to waive the animal testing requirements on the ground that they do not apply to a given substance. According to the applicant, in other words, the purpose of adaptations is to adjust the response to the standard information requirements applicable for a given effect whereas, in the present case, its position is that the standard information requirements were not applicable and that, consequently, there was no need to waive them by invoking an adaptation.

23      Next, the applicant points out that the REACH Regulation and the Cosmetics Regulation may apply to the same substance. In that regard, it reiterates that the REACH Regulation contains express provisions relating specifically to vertebrate animal testing falling within the scope of the Cosmetics Regulation, namely Article 2(4)(b) and recital 13 thereof. According to the applicant, if the REACH Regulation were to apply to substances used exclusively in cosmetic products, it should be ‘without prejudice’ to vertebrate animal testing, which falls within the scope of the Cosmetics Regulation.

24      In that context, the applicant claims that, according to the case-law of the Court of Justice, if the EU legislature has expressly provided that the provisions of an act of the Union are to apply ‘without prejudice’ to those of a directive, it has by that very fact meant that the obligations arising from that act of the Union ‘cannot directly affect those flowing’ from the directive. The applicant thus considers that, in the present case, as the REACH Regulation applies ‘without prejudice’ to the provisions relating to tests on vertebrate animals which fall within the scope of the Cosmetics Regulation, the provisions of the REACH Regulation, such as the testing requirements, cannot directly affect the obligations relating to tests on vertebrate animals flowing from the Cosmetics Regulation.

25      The applicant adds that the fact that the provisions on animal testing in the REACH Regulation cannot affect the prohibition on such testing laid down by the Cosmetics Regulation is, moreover, acknowledged by the ECHA in its factsheet on the interface between the REACH Regulation and the Cosmetics Regulation (‘the factsheet’). The factsheet expressly acknowledges that ‘registrants of substances that are exclusively used in cosmetics may not perform animal testing to meet the information requirements of the REACH human health endpoints’. According to the applicant, that means that the information requirements of the REACH Regulation only apply to a substance used exclusively as a cosmetic ingredient when they do not concern testing for human health endpoints.

26      Moreover, the applicant considers that the judgment of 21 September 2016, European Federation for Cosmetic Ingredients (C‑592/14, EU:C:2016:703) should not be applied to the prohibition on carrying out tests. It submits that that judgment relates to a factual and legal situation that is fundamentally different from that at issue in the present case.

27      Finally, the applicant claims that the ECHA’s argument relating to the consequences for substances which are used both in cosmetic products and in medicinal products, the latter being governed principally by Directive 2001/83/EC of the European Parliament and of the Council of 6 November 2001 on the Community code relating to medicinal products for human use (OJ 2001 L 311, p. 67), is irrelevant. First, the substance at issue is used exclusively as a cosmetic ingredient and does not give rise to dual use, so that its use in pharmaceutical products is purely hypothetical. Secondly, the present case only concerns the interaction between the REACH Regulation and the Cosmetics Regulation. Thirdly, the ECHA has neither explained nor demonstrated why and how the applicant’s observations ‘would have significant and … unacceptable consequences’ under the pharmaceutical legislation. In any event, the applicant points out that the REACH Regulation provides, inter alia, that registration obligations do not apply to substances used in medicinal products for human and veterinary use falling within the scope of Directive 2001/83.

28      The interveners agree with the applicant’s arguments.

29      In addition, CFE submits that all the data contained in the information file drawn up pursuant to Article 11 of the Cosmetics Regulation are mentioned in the product safety report drawn up pursuant to Article 10 of that regulation and may therefore lead to a prohibition. In support of that argument, CFE produced an expert statement in Annex 3 to its statement in intervention.

30      By the second part, the applicant claims that the tests required on vertebrate animals in order to establish the safety of the substance at issue for human health cannot be regarded as falling outside the scope of the Cosmetics Regulation on the ground that they are justified by the risks associated with worker exposure.

31      The applicant adds that the ECHA has drawn an artificial distinction between the different categories of human health endpoints. According to the applicant, the scope of the Cosmetics Regulation covers the safety of cosmetic products for human health without being limited to certain categories of persons, namely end users.

32      In support of that claim, the applicant refers, first, to Article 3 of the Cosmetics Regulation, secondly, to Annex I to that regulation and, thirdly, to Commission Implementing Decision 2013/674/EU of 25 November 2013 on Guidelines on Annex I to the [Cosmetics Regulation] (OJ 2013 L 315, p. 82).

33      By the third part, the applicant claims that the safety of the substance at issue was assessed under the Cosmetics Regulation.

34      In that regard, the applicant claims that the studies on which the Scientific Committee on Consumer Safety (‘the SCCS’) relies in order to assess the safety of a cosmetic product, which were completely omitted both in the initial decision and in the contested decision, cover developmental toxicity and reproductive toxicity, namely the two effects on human health for which the ECHA required the OECD 408, OECD 414 and OECD 443 studies. Thus, according to the applicant, the ECHA disregarded the case-law according to which it was required, when exercising its discretion, to examine carefully and impartially all the relevant elements of the case in question and to take into consideration all the relevant elements and circumstances of the situation which the contested measure was intended to govern.

35      The applicant adds that the assessment of the substance at issue in respect of its use as a cosmetic product constitutes the most relevant and appropriate assessment of safety for human health, including for workers. First, the applicant points out that consumer exposure to the substance at issue is intentional and that it is the most significant, the most repetitive and the most prolonged. Secondly, that exposure is not influenced by any type of personal protective equipment or physical barriers which may be required in order to comply with good laboratory practice and which constitute standard occupational hygiene conditions. Thirdly, consumer exposure would be the exposure of the most vulnerable groups. In that regard, the applicant observes that the ECHA introduces an exposure-related element which it applies only to cosmetic ingredients, which is not based on any provision of the REACH Regulation and which is not linked to any adaptation under Annex XI to that regulation. The ECHA was obliged to take into consideration the applicant’s explanations regarding worker exposure in comparison with consumer exposure.

36      Furthermore, the applicant points out that the manufacture of cosmetic products is subject to compliance with extremely strict good practice under the Cosmetics Regulation.

37      Thus, according to the applicant, the information cannot be described as ‘lacking’ and cannot be ‘adapted’ unless it is required from the outset. In that regard, it considers that the question of whether or not the REACH Regulation requires testing of cosmetic ingredients is a question of law and not a question of submitting an adaptation under that regulation.

38      The applicant adds that the substance at issue, which is used exclusively as an ingredient in cosmetic products, is exempt from the standard information requirements of the REACH Regulation not ipso facto, but as a matter of law.

39      The applicant claims that that is not a matter to be dealt with in the context of an adaptation, which is confirmed by the fact that any ‘Cosmetics adaptation’ is not one of the possibilities for adaptation recognised by Annex XI to the REACH Regulation. In its view, that is a matter that arises from the uniform and consistent application of the law. It therefore considers that its dossier is compliant, since the requirements for information on human health endpoints are covered by the Cosmetics Regulation.

40      The applicant also challenges the ECHA’s assertion that Implementing Decision 2013/674 is ‘irrelevant’.

41      Furthermore, according to the applicant, the Commission’s answer to a parliamentary question, relied on by the ECHA, is, at most, an opinion of that institution.

42      Finally, the applicant considers that, in any event, assuming that worker exposure is relevant, consumer exposure is, in the present case, the worst possible human health exposure. The applicant considers that all the potential risks to human health were assessed by the SCCS in accordance with the Cosmetics Regulation when examining consumer exposure.

43      Like the applicant, CFE submits that, in the disproved hypothesis that there is a risk linked to the substance at issue, consumers are in fact more exposed than workers. Products containing the substance at issue are designed to be applied repeatedly to the skin whereas, for workers, the objective is to avoid all contact, which is why extensive personal protective equipment would be used and other risk management measures would be in place. Therefore, according to CFE, if the substance at issue is safe for consumers, as the SCCS concluded, it must also be safe for workers.

44      PETA submits that, if the distinction made by the ECHA and the Board of Appeal were to be followed, that would allow the person responsible, within the meaning of Article 4 of the Cosmetics Regulation, to claim that no animal testing had been carried out on a cosmetic product even though a full gamut of animal tests had been carried out, pursuant to the REACH Regulation, on a substance contained in that product, which is contrary to Article 20(3) of the Cosmetics Regulation.

45      The ECHA disputes the arguments of the applicant and the interveners.

 The examination carried out by the Board of Appeal in the contested decision

46      As a preliminary point, in paragraphs 54 to 56 of the contested decision, the Board of Appeal held that, in the absence of provisions establishing the primacy of one regulation over the other, the REACH Regulation and the Cosmetics Regulation could apply to the same substance and that, therefore, those two regulations had to be interpreted and applied in a consistent and compatible manner.

47      Next, the Board of Appeal examined the REACH Regulation and the Cosmetics Regulation and their relationship.

48      As regards the relevant rules of the REACH Regulation, after examining, in particular, Article 2(4)(b) and Article 14(5)(b) of, and Section 3 of Annex XI to, that regulation, the Board of Appeal concluded, in paragraph 93 of the contested decision, that that regulation contained no provision exempting a registrant from carrying out studies on vertebrate animals solely because the substance for which registration was sought was used only as an ingredient in a cosmetic product, while specifying that, for the purposes of obtaining an exemption, the registrant had to demonstrate that the conditions for adaptation set out in Section 3 of Annex XI to the REACH Regulation, read in conjunction with Article 14(5)(b) of that regulation, were satisfied.

49      With regard to the relevant rules of the Cosmetics Regulation, the Board of Appeal examined, in particular, Article 18(1)(b) and (d) and (2) of the Cosmetics Regulation.

50      In the first place, in paragraphs 101 to 104 of the contested decision, the Board of Appeal held that Article 18(1)(d) and (2) of the Cosmetics Regulation provided for a prohibition on animal testing. The Board of Appeal stated, in particular, first, that that provision prohibited, as from certain dates, the performance of studies on vertebrate animals if they were carried out ‘in order to meet the requirements of the [Cosmetics Regulation]’ and, secondly, that that condition did not, in itself, prohibit the performance of that type of study. In that regard, the Board of Appeal observed that, in the absence of an explicit provision to that effect, Article 18(1)(d) and (2) of the Cosmetics Regulation could not be interpreted as prohibiting the performance of the studies requested on the basis of the REACH Regulation, adding that such an interpretation would not ensure a consistent and compatible application of those two regulations.

51      In the second place, the Board of Appeal considered, in paragraphs 105 to 107 of the contested decision, that Article 18(1)(b) and (2) of the Cosmetics Regulation provided for a prohibition on placing on the market of cosmetic products containing ingredients which, from certain dates, had been the subject of animal testing ‘in order to meet the requirements of [the Cosmetics Regulation]’. In that regard, the Board of Appeal noted that, in paragraph 39 of the judgment of 21 September 2016, European Federation for Cosmetic Ingredients (C‑592/14, EU:C:2016:703), the Court of Justice had held that a study on vertebrate animals was carried out ‘in order to meet the requirements of [the Cosmetics Regulation]’ only if it was included in the product safety report drawn up in accordance with Article 10 of that regulation to demonstrate the safety for end users of that product containing the substance tested as an ingredient.

52      In that context, the Board of Appeal considered, in paragraphs 108 and 109 of the contested decision, that the results of a study carried out in order to comply with the standard information required by the REACH Regulation could confirm the safety of the cosmetic product containing the substance which was the subject of the application for registration, as demonstrated in the context of the product safety report drawn up in accordance with Article 10 of the Cosmetics Regulation. In the present case, the Board of Appeal considered that those results should be included in the information file drawn up on the basis of Article 11 of the Cosmetics Regulation.

53      However, the Board of Appeal observed, in paragraphs 110 and 111 of the contested decision, that the results of studies carried out on vertebrate animals could call into question the safety of a cosmetic product containing a substance by contradicting the report on the safety of the product drawn up in accordance with Article 10 of the Cosmetics Regulation. In the present case, the Board of Appeal considered that, if the safety of the cosmetic product could no longer be guaranteed, it was possible that the product could no longer be placed on the market. The Board of Appeal noted that that was not the automatic consequence of carrying out a study on vertebrate animals in order to comply with the standard information requirements imposed by the REACH Regulation, but the consequence of the outcome of that study and the choice made by the legislature in Articles 3 and 18 of the Cosmetics Regulation, according to which a cosmetic product must be safe for end users, provided that vertebrate animals are not sacrificed in order for its safety to be established.

54      Thus, in paragraph 112 of the contested decision, the Board of Appeal held that it was not the conduct of the studies required under the REACH Regulation, but rather the use of the results of those studies which was liable to lead to the prohibition on placing on the market provided for in Article 18(1)(b) and (2) of the Cosmetics Regulation.

55      In the light of those factors, the Board of Appeal concluded, in paragraph 116 of the contested decision, that the Cosmetics Regulation did not prohibit the registrant of a substance used, exclusively or among other uses, as an ingredient of a cosmetic product from carrying out vertebrate animal testing for the purposes of complying with the information required for the registration of that substance under the REACH Regulation.

56      As regards the relationship between the REACH Regulation and the Cosmetics Regulation, first, the Board of Appeal concluded, in paragraph 117 of the contested decision, that the REACH Regulation did not provide for any exemption for registrants of a substance from carrying out tests on vertebrate animals on the ground that that substance was used solely in cosmetic products. According to the Board of Appeal, even in the present case, in the light of Section 3 of Annex XI to the REACH Regulation, and Article 14(5)(b) of that regulation, registrants are required to prove that the conditions for adaptation, within the meaning of those provisions, are met. Secondly, the Board of Appeal considered that that conclusion was not called into question by the Cosmetics Regulation on the basis of the examination carried out in paragraphs 94 to 116 of the contested decision.

57      Thus, in paragraphs 118 to 122 of the contested decision, the Board of Appeal held, in the first place, that, in the present case, the applicant had not submitted an adaptation in accordance with Section 3 of Annex XI to the REACH Regulation, read in conjunction with Article 14(5)(b) thereof, and, therefore, that the ECHA was not required to assess whether the conditions for such an adaptation were satisfied. In the second place, given that it was not disputed that the cosmetic products containing the substance at issue were formulated in the European Union, the Board of Appeal inferred that workers other than professionals within the meaning of Article 2(1)(f) of the Cosmetics Regulation could have been exposed to that substance, with the result that at least one of the conditions laid down in Section 3 of Annex XI to the REACH Regulation, namely no exposure or negligible exposure, was not satisfied. Consequently, the Board of Appeal concluded that, contrary to the applicant’s contention, the ECHA had not committed an ‘error of assessment’ by asking it to supplement its registration dossier with the OECD 408, OECD 414 and OECD 443 studies or by proposing valid adaptations.

58      As regards the fact that the ECHA did not take into account the circumstance that the safety of the substance at issue had been assessed under the Cosmetics Regulation, the Board of Appeal held, in paragraphs 127 to 130 of the contested decision, first, that, even assuming that the substance at issue did not constitute a risk to workers’ health, the applicant could not refrain from providing the OECD 408, OECD 414 and OECD 443 studies, unless it submitted a valid adaptation and, secondly, that Article 3 of the Cosmetics Regulation ensured the safety of end users, so that other risks due to exposure to the substance at issue, such as that of workers producing cosmetic products containing it, were not covered.

 The merits of the first plea in law

59      In the context of the first plea in law, it is necessary to examine the three parts together and to determine whether the Board of Appeal, in the contested decision, erred in law by misinterpreting the REACH Regulation and the Cosmetics Regulation.

60      First, the Board of Appeal relied, in essence, on the absence of provisions establishing the primacy of one regulation over the other, so that the REACH Regulation and the Cosmetics Regulation must be interpreted and applied in a consistent and compatible manner. Secondly, the Board of Appeal noted that the Cosmetics Regulation did not prohibit the registrant of a substance used, exclusively or among other uses, as an ingredient of a cosmetic product from carrying out tests on vertebrate animals for the purposes of satisfying the information required for the registration of that substance under the REACH Regulation. Thirdly, the Board of Appeal considered that Article 3 of the Cosmetics Regulation ensured the safety of end users, so that other risks due to exposure to the substance at issue, such as that of workers producing cosmetic products containing it, were not covered by the Cosmetics Regulation.

61      In that regard, it should be borne in mind that, according to settled case-law, for the purposes of interpreting a provision of EU law, account must be taken not only of the terms of that provision but also of its context and the objectives pursued by the legislation of which it forms part (see judgment of 27 April 2023, Castorama Polska and Knor, C‑628/21, EU:C:2023:342, paragraph 36 and the case-law cited).

62      In the present case, the substance at issue is a component used in the manufacture of cosmetic products and falls within the scope of both the REACH Regulation and the Cosmetics Regulation.

63      It is therefore necessary to determine, in the first place, the relationship between the REACH Regulation and the Cosmetics Regulation and, in the second place, which of those regulations guarantees the prevention of risks to workers’ health arising from exposure to the substance at issue.

–       The relationship between the REACH Regulation and the Cosmetics Regulation

64      In the first place, Article 2 of the REACH Regulation sets out the limits and conditions of application of that regulation. Article 2(1) of that regulation states that it does not apply to certain substances. Article 2(2) of that regulation states that certain wastes are not substances falling within its scope. Article 2(3) of that regulation states that Member States may provide for exemptions in specific cases for certain substances where this is necessary in the interests of defence. Article 2(4) of that regulation states that it is to apply without prejudice to the provisions of EU law relating to the workplace and the environment and to the Cosmetics Regulation as regards vertebrate animal testing falling within the scope of that regulation, which repealed Council Directive 76/768/EEC of 27 July 1976 on the approximation of the laws of the Member States relating to cosmetic products (OJ 1976 L 262, p. 169), Article 38 of which provides that references to the repealed directive are to be construed as references to that regulation. Article 2(5) of that regulation provides that the provisions of Titles II, V, VI and VII thereof are not to apply where a substance is used in medicinal products for human or veterinary use or in foodstuffs or feedingstuffs. Article 2(6) and (9) of that regulation specifies that certain titles are not applicable to certain substances and polymers, and Article 2(7) and (8) of that regulation exempts certain substances and intermediates from the application of certain titles.

65      Thus, it follows from Article 2 of the REACH Regulation that the legislature expressly provided, first, for the circumstances in which that regulation was not to be applied and, secondly, for the circumstances in which that regulation applied ‘without prejudice’ to other legislation.

66      Article 2(4)(b) of the REACH Regulation provides that it applies without prejudice to the Cosmetics Regulation as regards vertebrate animal testing falling within the scope of the latter.

67      That is also in line with recital 13 of the REACH Regulation, which states that that regulation should apply without prejudice to the prohibitions and restrictions laid down in the Cosmetics Regulation.

68      Accordingly, it must be concluded that the Board of Appeal held, in paragraphs 65 to 76 of the contested decision, without committing an error of law, that the expression ‘without prejudice’ in Article 2(4) of the REACH Regulation should not be interpreted as exempting registrants of substances used as ingredients in cosmetic products from all the information requirements arising from that regulation.

69      In the second place, it is appropriate to examine Article 18(1)(d) of the Cosmetics Regulation, which defines the system of prohibition of animal testing ‘in order to meet the requirements of [that regulation]’.

70      The applicant bases its argument on the premiss that vertebrate animal testing is prohibited by Article 18(1)(d) of the Cosmetics Regulation.

71      In that context, it should be noted that it follows from paragraphs 36 to 39 of the judgment of 21 September 2016, European Federation for Cosmetic Ingredients (C‑592/14, EU:C:2016: 703), to which the Board of Appeal refers, that the expression ‘in order to meet the requirements of [the Cosmetics Regulation]’ in Article 18(1)(b) of the Cosmetics Regulation must be interpreted as prohibiting reliance on the results of animal tests contained in the product safety report drawn up within the meaning of Article 10 of that regulation in order to place that product on the EU market.

72      By contrast, as the Court of Justice stated in paragraph 38 of the judgment of 21 September 2016, European Federation for Cosmetic Ingredients (C‑592/14, EU:C:2016:703), vertebrate animal testing, the results of which appear in the information file established within the meaning of Article 11 of the Cosmetics Regulation, is not sufficient to render applicable the prohibition on placing on the market laid down in Article 18(1)(b) of that regulation.

73      It should also be noted that the expression ‘in order to meet the requirements of [the Cosmetics Regulation]’, interpreted by the Court of Justice and appearing in Article 18(1)(b) of the Cosmetics Regulation, is reproduced verbatim in Article 18(1)(d) of that regulation.

74      It follows that Article 18(1)(d) of the Cosmetics Regulation must be interpreted as meaning that only the tests which appear in the product safety report drawn up in accordance with Article 10 of that regulation to demonstrate the safety of that product are prohibited.

75      Furthermore, Article 25(1)(g) of the Cosmetics Regulation provides that a cosmetic product may be withdrawn from the market or recalled in the event of non-compliance with the animal testing requirements referred to in Article 18 of that regulation. It therefore follows from the provisions of Article 25(1)(g) of the Cosmetics Regulation, read in conjunction with those of Article 18(1)(d) of that regulation, that the animal tests leading to the withdrawal of a cosmetic product from the market are those set out in the product safety report drawn up in accordance with Article 10 of that regulation.

76      Consequently, it was without committing an error of law that, in paragraph 102 of the contested decision, the Board of Appeal concluded that the expression ‘in order to meet the requirements of [the Cosmetics Regulation]’ in Article 18(1)(d) of that regulation did not, in itself, prohibit animal testing.

77      That conclusion is not called into question by the applicant’s argument that the factsheet expressly acknowledges that ‘registrants of substances that are exclusively used in cosmetics may not perform animal testing to meet the information requirements of the REACH human health endpoints’.

78      It suffices to note that, in the present case, the applicant confines itself to quoting a sentence from the factsheet without specifying that, first, the following sentence states that, ‘where there is a risk to a worker directly or indirectly involved in the manufacture of a substance, animal testing is permitted’ and, secondly, that sheet specifies that, on a case-by-case basis, animal testing need not be carried out if, in order to meet the requirements of the REACH Regulation, the registrant does not need to assess the risks arising from worker exposure and if the substance is only used in cosmetic products. Moreover, in response to a question put by the General Court at the hearing, the applicant acknowledged the existence of the following sentence in the factsheet.

79      Accordingly, the applicant’s argument is lacking in fact and must be rejected.

–       The regulation applicable to ensure the prevention of risks to workers’ health arising from exposure to the substance at issue

80      As regards the applicant’s argument that, first, the Cosmetics Regulation also covers risks to workers’ health arising from exposure to the substance at issue and, secondly, the Board of Appeal failed to take account of the fact that the safety of that substance had been assessed under that regulation, it should be noted, as a preliminary point, that the REACH Regulation and the Cosmetics Regulation overlap as regards one of the objectives which they pursue, namely that of protecting human health.

81      First, Article 3 of the Cosmetics Regulation provides that ‘a cosmetic product made available on the market shall be safe for human health when used under normal or reasonably foreseeable conditions of use’. Secondly, Article 1(1) and (3) of the REACH Regulation states that its purpose ‘is to ensure a high level of protection of human health and the environment’ and that it ‘is based on the principle that it is for manufacturers, importers and downstream users to ensure that they manufacture, place on the market or use such substances that do not adversely affect human health or the environment’.

82      Moreover, although the Cosmetics Regulation does not contain an express and literal definition of the term ‘human health’, Article 3 thereof nevertheless provides that ‘a cosmetic product made available on the market shall be safe for human health when used under normal or reasonably foreseeable conditions of use’. Thus, safety for human health, in that regulation, is defined with reference to the way in which a product is ‘used’ and, as such, to ‘normal or reasonably foreseeable conditions of use’. Given the meaning of those concepts in everyday language, the human health in question is that of end users and professionals who use cosmetic products under normal or reasonably foreseeable conditions of use. Workers involved in the formulation of an ingredient for a cosmetic product or in the formulation of the product itself, by contrast, are exposed to that ingredient throughout the manufacturing cycle and under conditions that differ from the use of the final cosmetic product.

83      In that context, it is also appropriate to examine the Board of Appeal’s interpretation of Article 14(5)(b) of the REACH Regulation in conjunction with that of Section 3 of Annex XI to that regulation in the contested decision.

84      In that regard, first of all, it should be recalled that Article 14(5)(b) of the REACH Regulation provides that ‘the chemical safety report need not include consideration of the risks to human health from  … end uses … in cosmetic products within the scope of [the Cosmetics Regulation]’.

85      Moreover, Section 3.1 of Annex XI to the REACH Regulation recognises the possibility for a registrant to waive the tests to be carried out in accordance with Sections 8.6 and 8.7 of Annex VIII and Annexes IX and X to that regulation and, therefore, the OECD 408, OECD 414 and OECD 443 studies requested in the present case, on the basis of the scenario(s) developed as part of the chemical safety report. However, on the basis of Section 3.2 of Annex XI to the REACH Regulation, in all cases the registrant must provide a sufficient description and sufficient justification, which is based on a thorough and rigorous exposure assessment established in accordance with Section 5 of Annex I to that regulation. In addition, a number of criteria must be met, including documented evidence from the manufacturer or importer that the results of the assessment covering all relevant exposures throughout the life cycle of the substance demonstrate no or negligible exposure in all manufacturing scenarios and for all identified uses referred to in Section 3.5 of Annex VI to the REACH Regulation.

86      Furthermore, the following should be noted.

87      In the first place, the chemical safety report drawn up in accordance with the REACH Regulation does not have to take into account the risks covered by the Cosmetics Regulation, in accordance with Article 14(5)(b) of the REACH Regulation.

88      In the second place, the registrant of a substance may, in principle, waive the studies requested under certain conditions, provided for in Section 3 of Annex XI to the REACH Regulation, namely that it is proven that the results of the assessment covering all the exposures concerned throughout the life cycle of the substance establish the absence of exposure or negligible exposure in all manufacturing scenarios and for all uses.

89      In the third place, the registrant of a substance used solely in cosmetic products is not exempt, under Section 3 of Annex XI to the REACH Regulation, from providing the standard information relating to the intrinsic properties of that substance. That is because the risks to human health covered by the Cosmetics Regulation and the REACH Regulation are not identical. The Cosmetics Regulation covers risks to human health resulting from the end uses of a cosmetic product containing a substance (see paragraph 82 above), whereas the REACH Regulation covers risks to human health relating to all exposures throughout the life cycle of the substance (see paragraph 88 above), including, in particular, risks to which workers are exposed.

90      Thus, it was without committing an error of law that the Board of Appeal held, in paragraphs 85 to 92 of the contested decision, that the exposure to which the end-users of a cosmetic product were subjected could not be considered to be an exposure falling within Section 3 of Annex XI to the REACH Regulation, so that the assessment of such exposure did not have to be contained in the chemical safety report drawn up within the meaning of Article 14(5)(d) of that regulation. However, Section 3 of Annex XI to the REACH Regulation establishes that, in order for a registrant to waive the tests to be carried out in accordance with Sections 8.6 and 8.7 of Annex VIII and Annexes IX and X to that regulation, the registrant must demonstrate that the results of the assessment covering all relevant exposures throughout the life cycle of the substance establish no or negligible exposure in all manufacturing scenarios and for all uses.

91      Consequently, in view of the finding made in paragraph 68 above, it follows from the joint interpretation of Article 14(5)(b) of the REACH Regulation and Section 3 of Annex XI to that regulation that registrants of a substance used solely in cosmetic products are not exempted from providing the standard information as regards the intrinsic properties of that substance.

92      It follows that it was without committing an error of law that the Board of Appeal concluded, first, in paragraphs 129 and 130 of the contested decision, that Article 3 of the Cosmetics Regulation ensured the safety of end users, so that other risks due to exposure to the substance at issue, such as that of workers producing cosmetic products containing it, were not covered and, secondly, in paragraphs 85 to 92 of the contested decision, that the registrant of a substance used solely in cosmetic products remained subject to the obligation to submit the standard information provided for in Annex IX to the REACH Regulation or any adaptations based on Annex XI to that regulation as regards the exposure of workers to that substance.

93      That conclusion is not called into question by the other arguments raised by the applicant.

94      As regards the arguments relating to Annex I to the Cosmetics Regulation and to Implementing Decision 2013/674, the following must be stated.

95      Section 8 of Part A of Annex I to the Cosmetics Regulation states that the chemical safety report shall include, ‘without prejudice to Article 18 [of that regulation], the toxicological profile of substance contained in the cosmetic product for all relevant toxicological endpoints’, so that ‘all significant toxicological routes of absorption shall be considered’.

96      However, it should be noted that, in the first place, Part A of Annex I to the Cosmetics Regulation is entitled ‘Cosmetic product safety information’ and that, in the second place, Sections 5 and 6 thereof, entitled, respectively, ‘Normal and reasonably foreseeable use’ and ‘Exposure to the cosmetic product’, refer to the use of the cosmetic product under normal and foreseeable conditions and to the exposure of end users.

97      Thus, as noted in paragraph 92 above, in that context Part A of Annex I to the Cosmetics Regulation must be interpreted as referring to the safety of a cosmetic product for its end users.

98      Consequently, Section 8 of Part A of Annex I to the Cosmetics Regulation cannot be interpreted as covering the risks to the safety of workers involved in the formulation of an ingredient for a cosmetic product or in the formulation of the product itself.

99      With regard to Implementing Decision 2013/674, it should be noted that it cannot concern workers involved in the formulation of an ingredient for a cosmetic product or in the formulation of the product itself, in so far as they are not in the situation of normal or reasonable use of the cosmetic product, referred to in its recital 1, according to which ‘it is essential that cosmetic products made available on the Union market be safe for human health when used under normal and reasonably foreseeable conditions of use’.

100    Furthermore, the danger to human health resulting, inter alia, from the manufacturing process and packaging, to which Part A of the annex to Implementing Decision 2013/674 refers, does not concern workers, but rather end users. In that regard, point 3.4 of that annex, entitled ‘Impurities, traces, information about the packaging material’, states that ‘a trace is a small quantity of an unintended substance in the finished product’, that ‘traces are to be evaluated with regard to safety of the finished product’ and that, ‘because substances may migrate from the packaging to the formulation, the relevant characteristics of the packaging material are to be considered’. Furthermore, point 3.4.3 of that annex states that ‘the combination of packaging material, formulation of the cosmetic product and contact with the external environment may have an impact on the safety of the finished product’. Consequently, the hazard concerned by Implementing Decision 2013/674 is not that to the health of the worker due to exposure to a substance contained in a cosmetic product, but rather that to the health of the end user or the professional who uses that product under normal conditions.

101    As regards the applicant’s argument relating to the SCCS, it is sufficient to note that the document entitled ‘Notes of Guidance for the Testing of Cosmetic Ingredients and Their Safety Evaluation’, drawn up by that committee, partially reproduces the factsheet and, in particular, the part stating that, even if a substance is registered solely as an ingredient of a cosmetic product, animal testing may be carried out in order to assess the risk to workers’ health. Thus, the applicant’s argument that the SCCS reviews all risks to human health, including that of workers, cannot be accepted, in so far as the partial reproduction of the factsheet demonstrates that the SCCS itself considers that certain risks to human health are beyond its control and, consequently, beyond the control of the Cosmetics Regulation.

102    As regards the applicant’s argument that exposure of the end users of a cosmetic product is the most appropriate, it starts from the premiss that the consumer of a cosmetic product comes into more contact with a substance contained in that product than a worker involved in its manufacture.

103    However, that assertion is not supported by any evidence and remains a mere allegation. Moreover, as the ECHA points out, it is possible that the occasional application via the dermal route of sunscreens containing the substance at issue in relatively high concentrations is safe for consumers, but that does not mean that it is safe for a worker’s health to ingest or inhale on a daily basis a quantity in lower concentrations of the substance at issue over a long period. Furthermore, it should be noted that, if the interpretation proposed by the applicant were to be accepted, the ECHA could not ask a registrant to carry out tests on animals in order to assess the risks of a substance to workers’ health on the basis that that substance, which is safe for consumers, would also be safe for workers, with the result that the ECHA would have to consider that workers’ health is protected without any certain and specific scientific benchmark in that regard.

104    As regards the applicant’s argument that the health of workers is ensured by compliance with extremely strict good practice under the Cosmetics Regulation, it should be recalled that, in accordance with the examination carried out above, the Cosmetics Regulation does not concern the safety of workers, but only that of end users and professionals. In addition, and in any case, it is also clear from the above review that the actual safety of workers’ health can only be deduced from valid tests or adaptations that confirm it.

105    Furthermore, the examination carried out by the Board of Appeal in paragraphs 127 to 130 of the contested decision cannot be called into question by PETA’s argument that the distinction made by the ECHA and the Board of Appeal runs counter to Article 20(3) of the Cosmetics Regulation.

106    In that regard, it is sufficient to note that, in response to a question put by the Court at the hearing, PETA confirmed that that argument had not been raised before the Board of Appeal in the proceedings which led to the contested decision. Accordingly, that argument must be rejected as inadmissible (see, to that effect, judgment of 20 September 2019, BASF Grenzach v ECHA, T‑125/17, EU:T:2019:638, paragraph 475).

107    Finally, as regards CFE’s argument that it is apparent from the statement of an expert, annexed to its statement in intervention, that all the data in the information file drawn up pursuant to Article 11 of the Cosmetics Regulation are referred to in the product safety report drawn up pursuant to Article 10 of that regulation, it must be held that, apart from the doubts expressed by the ECHA as to its admissibility, it cannot be accepted.

108    It should be noted that, in his statement reproduced in Annex 3 to CFE’s statement in intervention, the expert called upon by CFE admittedly observes that, in his view, in the first place, it would be bad practice for the producer of a substance not to take into consideration all the information concerning the safety of that substance and that, in the second place, Article 10(1) of the Cosmetics Regulation appears to require that all the relevant data available and any new data confirming or containing new information be taken into account. However, contrary to what CFE maintains, that expert does not assert that, in practice, any data resulting from a study on vertebrate animals would automatically be included in the product safety report drawn up in accordance with Article 10 of the Cosmetics Regulation.

109    Furthermore, and in any event, as the Board of Appeal observed in paragraphs 110 and 111 of the contested decision, without committing an error of law, if a study carried out on vertebrate animals were to call into question the safety of a cosmetic product by contradicting the report on the safety of the product drawn up in accordance with Article 10 of the Cosmetics Regulation, the prohibition on placing that product on the market would not be the consequence of the carrying out of that study, conducted to comply with the standard information requirements imposed by the REACH Regulation, but the consequence of the results of that study and the choice made by the legislature in Articles 3 and 18 of the Cosmetics Regulation, according to which a cosmetic product must be safe for end users, provided that vertebrate animals are not sacrificed in order for its safety to be established.

110    In the light of all those considerations, the first plea in law must be rejected.

 The second plea in law, alleging, in essence, several manifest errors of assessment committed by the ECHA in requiring the OECD 443 Study to be conducted via the oral route and a breach of the duty to state reasons

111    In support of its second plea in law, first of all, the applicant points out that the substance at issue is used as an ingredient in cosmetic products and, therefore, that the appropriate route of human exposure is the dermal route. It claims, first, that the REACH Regulation states that the OECD 443 Study must be conducted via the likely route of human exposure, secondly, that 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 2008 L 353, p. 1), indicates that substances should be labelled and classified according to the potential route of human exposure and, thirdly, that Commission Regulation (EC) No 440/2008 of 30 May 2008 laying down test methods pursuant to Regulation No 1907/2006 (OJ 2008 L 142, p. 1) also seems to indicate that the choice of route of administration takes into account, in particular, the most appropriate routes of human exposure. For example, the most appropriate route of exposure, including worker exposure, would be the dermal route, and that is the route via which the OECD 443 Study should have been conducted.

112    Next, in the first place, the applicant submits that the ECHA is wrong when it considers, without any basis, that the appropriate route must be chosen so as to ‘maximise the likelihood’ of obtaining results since ‘using the dermal route would therefore be unlikely to lead to sufficient foetal exposure to give meaningful results’. The applicant submits that that approach is without support by any reference to potentially relevant legal provisions or guidance documents. In that context, it adds that the fact that the dermal route, compared to the oral route, induces a potentially lower rate of absorption of the substance, between 2 and 10% of the substance being found in the body, does not make the oral route the ‘likely route of human exposure’.

113    In the second place, the applicant claims that the ECHA acted in breach of the duty to state reasons in that it requested that the OECD 443 Study via the oral route be carried out by making only vague observations, even though that was not the most likely route of human exposure. Furthermore, according to the applicant, since the ECHA had departed from the most appropriate route of exposure, it should have justified its choice in the light of the situation. In that regard, the applicant considers that the ECHA’s justifications concerning worker exposure are inoperative, in that workers are protected by measures and by the good laboratory practice in force.

114    In the third place, the applicant claims that, in the statement of defence, the ECHA misrepresented its arguments. In that regard, the applicant observes that the ECHA claims that it argues that ‘the [OECD 443 Study] should always be carried out by the most likely route of human exposure’, whereas it maintains that the likely route of human exposure is ‘the default in selecting the appropriate route’. According to the applicant, at no time did it claim that other routes could not be appropriate and that they could not be used to carry out an OECD 443 Study.

115    In the fourth place, the applicant contests the ECHA’s reference to other factors to justify its choice to impose the oral route, such as the design of the study and the known properties of a substance, without determining in what specific way the design of the study or the known properties of the substance at issue justify the use of the oral route in this particular case. The applicant does not dispute the fact that other factors may influence the choice of the most appropriate route for carrying out tests. In its view, however, it must be noted that those other factors can only be invoked to depart from the default route of administration, namely the likely route of human exposure, if they have actually given rise to an assessment in the specific case and if the reasons justifying their invocation have been provided. The applicant considers that the ECHA did neither in the present case and that, moreover, the arguments put forward by the ECHA in the defence confirm that it did not carry out an examination of the concrete manner in which the other elements were relevant in the present case.

116    In the fifth place, the applicant claims that the ECHA attempts to justify the choice of the oral route by stating that it cannot be ruled out that the substance ‘may be ingested or inhaled during the manufacture of cosmetic products’. According to the applicant, in so doing, the ECHA completely disregards the actual manufacturing process and the regulatory occupational hygiene conditions, in so far as, during the manufacturing process, no significant exposure of workers could occur, taking into account the protective measures and good laboratory practice in force. Therefore, any potential exposure of workers by inhalation or ingestion could only occur by accident or misuse. Furthermore, in an OECD 443 Study, it would not be short-term acute exposure to a substance, such as would occur in the event of accidental inhalation or ingestion, that would be analysed, but, on the contrary, long-term chronic exposure to a substance and its effects on offspring. The applicant therefore fails to see how that could be relevant to worker exposure.

117    Finally, the applicant points out that the ECHA claims that the use of the oral route is justified in so far as the study via the dermal route is ‘a complex study, which can cause much stress to vertebrate animals, in a way which is unlikely to produce useful results’. In that regard, in the first place, the applicant confirms that the study can be conducted via the dermal route. In the second place, it submits that the allegations relating to the stress allegedly caused to vertebrate animals and the usefulness of the results are unfounded.

118    The ECHA disputes the applicant’s arguments.

 The examination carried out by the Board of Appeal in the contested decision

119    First of all, the Board of Appeal noted that, in the initial decision, the ECHA had considered that the oral route was the most appropriate route for identifying the hazardous effects of substances on reproduction, as indicated in the ECHA’s ‘Guidance on information requirements and Chemical Safety Assessment’, and that, in so far as the substance at issue was a liquid, the test had to be carried out via the oral route.

120    Next, in response to the applicant’s arguments, the Board of Appeal indicated that column 1 of Section 8.7.3 of Annex IX to the REACH Regulation provided that an OECD 443 Study should, if necessary, be carried out using the most appropriate route of administration with regard to the most likely human exposure. In that regard, the Board of Appeal noted that column 1 of Section 8.7.3 of Annex IX to the REACH Regulation referred to test method B.56, as established in Commission Regulation (EU) No 900/2014 of 15 July 2014 amending, for the purposes of its adaptation to technical progress, Regulation No 440/2008 (OJ 2014 L 247, p. 1). The Board of Appeal added that point 18 of Test Method B.56 required that ‘selection of the route should take into consideration the route(s) most relevant for human exposure’ and stated that, ‘although the protocol is designed for administration of the test chemical through the diet, it can be modified for administration by other routes (drinking water, gavage, inhalation, dermal), depending on the characteristics of the chemical and the information required’.

121    Finally, taking into account test method B.56, the Board of Appeal considered that the most likely route of human exposure could not be the only element to be taken into consideration when deciding on the route of administration of the substance at issue for the OECD 443 Study and that other elements had to be taken into consideration, such as the design of the study and the known properties of a substance.

122    In the light of those factors, in paragraphs 172 to 174 of the contested decision, first, the Board of Appeal rejected the applicant’s claim, holding that, admittedly, in its current use in cosmetic products for application to the skin, the most likely route of exposure to the substance at issue for humans was the dermal route, but that, however, exposure to the substance at issue was possible by other routes, for example in the process of formulating cosmetic products containing it. Secondly, the Board of Appeal noted that an OECD 443 Study required exposure of the foetus to the substance tested and that absorption into the body by dermal exposure was normally low. In fact, the Board of Appeal pointed out that existing data showed that only 2 to 10% of the substance administered via the dermal route was systematically present in laboratory animals. Consequently, the Board of Appeal considered that it was unlikely that administration of the substance at issue by the dermal route in an OECD 443 Study would lead to sufficient exposure of the foetus to give significant results, whereas administration via the oral route would have maximised the probability of obtaining useful results enabling the potential toxicity of the substance at issue on reproduction to be assessed.

 The merits of the second plea in law

123    As a preliminary point, in the first place, as regards the scope of the review carried out by the Board of Appeal, it is clear from the case-law that the purpose of an action before the Board of Appeal against a decision of the ECHA in the context of the evaluation of a substance may be limited to examining whether the evidence submitted by the applicant is capable of demonstrating that that decision is vitiated by errors. Therefore, in the context of such an action, the applicant may not confine itself to claiming that the result of the assessment on which that decision is based should have been different, but it must put forward arguments to show the existence of errors vitiating the scientific assessment on which the decision in question is based (judgment of 20 September 2019, BASF Grenzach v ECHA, T‑125/17, EU:T:2019:638, paragraph 86).

124    In the second place, with regard to the review of the scientific assessments contained in an ECHA decision carried out by the Board of Appeal, it should be noted that it is not limited to checking whether there are manifest errors. On the contrary, in that regard, relying on the legal and scientific expertise of its members, the Board of Appeal must examine whether the arguments put forward by the applicant are capable of demonstrating that the considerations on which the ECHA’s decision is based are vitiated by errors (judgment of 20 September 2019, BASF Grenzach v ECHA, T‑125/17, EU:T:2019:638, paragraph 89).

125    In the third place, the EU Courts review the legality of a decision of the Board of Appeal concerning an action against a decision taken in respect of the assessment of a substance. According to the case-law, that review is limited when it comes to assessing highly complex scientific and technical factual elements. With regard to such assessments, the EU Courts are limited to checking whether they are vitiated by a manifest error, a misuse of powers or whether the decision-maker manifestly exceeded the limits of its power of review (see judgment of 20 September 2019, Germany v ECHA, T‑755/17, EU:T:2019:647, paragraph 120 and the case-law cited).

126    It should be noted that column 1 of Section 8.7.3 of Annex IX to the REACH Regulation states that the 443 OECD Study is intended to be carried out by the most appropriate route of administration with regard to the most likely human exposure. However, without being explicitly contradicted on this point, the ECHA claims that, as the Board of Appeal found in paragraphs 169 and 170 of the contested decision, point 18 of Test Method B.56 clearly indicates that the most likely route of human exposure is one of the factors, but not the only one, to be taken into consideration when deciding on the route of administration, depending on the characteristics of the substance concerned and the information sought. Thus, on the basis of that premiss, the Board of Appeal was entitled to consider that the most likely route of administration could not be the only factor to be taken into account in order to determine, in the present case, the method for carrying out the OECD 443 Study.

127    Furthermore, first, with regard to the applicant’s argument raised against the justification relating to other possible exposures to the substance at issue and, in particular, to worker exposure, it should be noted that, in paragraph 92 above, it was noted that, even in the case of a substance used solely in cosmetic products, if those products were formulated in the European Union, the registrant was required to comply with the obligations arising from the REACH Regulation, in so far as the safety of workers was not covered by the rules laid down in the Cosmetics Regulation. That factor is sufficient to conclude that the justification provided by the Board of Appeal, according to which there are other possible exposures to the substance at issue, is not vitiated by a manifest error of assessment.

128    Secondly, as regards the applicant’s argument raised against the justification relating to the fact that exposure of the foetus would be more likely if the substance at issue were administered via the oral route rather than via the dermal route, the applicant merely states that that justification does not make the oral route the ‘likely route of human exposure’, without providing any additional decisive evidence capable of calling into question the contested decision.

129    As regards the applicant’s argument that the ECHA’s approach lacks any basis in potentially relevant legal provisions or guidance documents, it should be noted that it is not capable of calling into question the data relating to the very low rate of absorption of the substance at issue by the body in the event of administration via the dermal route. Accordingly, it is without committing a manifest error of assessment that the Board of Appeal was able to conclude that the oral route was likely to produce more useful results.

130    Thirdly, in relation to the breach of the duty to state reasons alleged by the applicant, it is sufficient to note that, in the light of the assessment set out above, in particular in paragraphs 119 to 122 above, the contested decision is reasoned and that the Board of Appeal provided sufficient justification to make it possible to understand the reasons which led it to decide that the OECD 443 Study should be carried out via the oral route and not via the dermal route.

131    Fourthly, as regards the applicant’s argument, referred to in paragraph 114 above, that the ECHA misrepresented its arguments in the defence, it is sufficient to note that such a misrepresentation, even if proven, would have no bearing on the legality of the contested decision. Accordingly, that argument cannot succeed.

132    The second plea in law must therefore be rejected.

 The third plea in law, alleging, in essence, several manifest errors of assessment and infringements committed by the ECHA in determining the time limit for the submission of the required information

133    By its third plea in law, the applicant claims, in essence, that, in determining the 42-month deadline set out in the contested decision, the ECHA committed manifest errors of assessment and infringed Article 25 of the REACH Regulation.

134    In the first place, the applicant claims that a longer period, namely 54 months, is necessary in view of the time required to carry out the required studies. The applicant points out that that circumstance was brought to the ECHA’s attention in the context of the process leading to the adoption of the initial decision and the action leading to the adoption of the contested decision.

135    Furthermore, according to the applicant, the ECHA is required, in each compliance decision, to set ‘adequate time limits’ for each individual case. In that regard, the applicant submits that it put forward numerous reasons to explain why the time limit of 42 months laid down in the initial decision was not adequate. The ECHA committed a manifest error of assessment by ignoring its explanations.

136    The applicant adds that the Board of Appeal’s assertion that the problems of laboratory capacity are no longer relevant since the expiry of the 2018 time-limit for registration is unfounded.

137    In the second place, the applicant alleges infringement of Article 25 of the REACH Regulation in that the ECHA considered that the OECD 408, OECD 414 and OECD 443 studies should not be carried out sequentially, even though it had accepted that sequential procedure in the initial decision.

138    In that regard, the applicant claims that it is possible, given the results of one of the required studies, that the other studies are no longer necessary. Thus, carrying out, first of all, the OECD 408 Study, then the OECD 414 Study and, finally, the OECD 443 Study, which requires the use of more vertebrate animals than the first two, would be a usual way of proceeding in view of the principle of the absence of any other solution and in order to avoid carrying out tests on vertebrate animals which are unnecessary under Article 25 of the REACH Regulation.

139    As regards the ECHA’s argument that none of the OECD 408, OECD 414 or OECD 443 studies allows one of the other two to be omitted or adapted, the applicant claims that, if the OECD 408 Study were to reveal a large-scale effect on the parameters most relevant from the human point of view, that would be sufficient to require classification of the substance at issue, with the consequence that the OECD 443 Study would not be required in order to analyse that aspect in greater detail. Furthermore, in this particular case, the alleged indications of harmful effects on which the ECHA based its request for the OECD 443 Study could, in fact, be evaluated and could be the subject of a definitive assessment in the context of the OECD 408 Study and the OECD 414 Study. In those circumstances, the applicant submits that, since the question of whether the OECD 443 Study is, in the present case, actually required under column 1 of Section 8.7.3 of Annex IX to the REACH Regulation is controversial, it should be authorised – having regard to Article 25 of that regulation and the obligation to avoid unnecessary testing on vertebrate animals – to carry out the OECD 414 Study first, since the results of that study could demonstrate that it is unnecessary to carry out the OECD 443 Study in the present case.

140    In addition, the applicant claims that, if the results of the OECD 414 Study were to show effects leading to the classification of the substance at issue in category 1B of toxicity for reproduction, that would mean that the OECD 443 Study was unnecessary.

141    The applicant claims that, taking the view that the sequential nature of the studies is not scientifically justified, the contested decision requires it to start all the studies immediately so that the results are available before the expiry of the time limit set at 42 months. That could lead to the sacrifice, in addition to 100 vertebrate animals for the OECD 408 Study, of approximately 3 201 vertebrate animals for the OECD 414 and OECD 443 studies which, depending on the results of the OECD 408 Study, might not be necessary.

142    In the third place, the applicant alleges a manifest error of assessment on the part of the ECHA with regard to the time limit imposed, which does not take into account the possible need to conduct specific investigations or additional in-depth studies following the conduct of the OECD 414 Study.

143    The applicant adds that the ECHA deliberately misrepresented its arguments by stating that it did not understand their relevance to the 2018 time-limit for registration in so far as the OECD 414 and OECD 443 studies would not be required for registrations in the tonnage band of 1 to 100 tonnes per year subject to that time-limit. In that regard, the applicant submits that it did not claim that the limited availability of the contract research organisations (‘CROs’) was linked to the fact that they had to carry out precisely the OECD 414 and OECD 443 studies in the context of the proximity of the registration deadline for 2018. On the contrary, it argued that the CROs generally had limited capacity, whatever the studies considered, due to the large number of studies required for the 2018 registrations.

144    Next, the applicant contests the ECHA’s assertion that the time-limits set are based on ‘the ECHA experts’ experience and consultations with [CROs]’, since in the present case the ECHA has not produced the slightest piece of evidence that the time limit of 42 months is in fact based on that experience and those consultations.

145    Finally, the applicant points out that the fact that the ECHA does not in any way justify or substantiate its assertions relating to the time limit contrasts with its attitude to the explanations and arguments which it itself put forward as to why a longer time limit was necessary. In that regard, the applicant states that it submitted its own calculations of the time required and an estimate of the time required according to a CRO. In addition, it explained that the other laboratories refused to produce statements or to commit themselves in any way whatsoever as to their availability or to a specific time limit in the absence of a formal order to carry out the studies in question.

146    The ECHA disputes the applicant’s arguments.

 The examination carried out by the Board of Appeal in the contested decision

147    As regards the argument that a time limit of 42 months was too short to carry out the studies sequentially and to submit an adaptation on the basis of the results of the OECD 414 Study rather than an OECD 443 Study, the Board of Appeal considered, in paragraphs 181 and 182 of the contested decision, that it was not obliged, pursuant to Article 41 of the REACH Regulation, to wait for a registrant to draw up an adaptation for its registration dossier. In addition, the Board of Appeal considered that there was no scientific justification for the studies required by the initial decision to be carried out sequentially, one after the other.

148    As regards the argument that the laboratories did not have the capacity to carry out the studies requested in the initial decision because of the proximity of the time-limit for registration for 2018, the Board of Appeal held, in paragraph 185 of the contested decision, that the proceedings before it had suspensive effect and that that time-limit was no longer applicable. Accordingly, the Board of Appeal rejected the applicant’s request.

 The merits of the third plea in law

149    It is necessary to examine, first of all, the applicant’s argument that the Board of Appeal committed a manifest error of assessment in considering that the period of 42 months fixed by the initial decision was sufficient.

150    In that regard, it should be noted that the Board of Appeal concluded, in paragraphs 181 and 182 of the contested decision, that the time limit set by the initial decision was not too short on two grounds, namely that, according to its practice, the ECHA was not obliged, in accordance with Article 41 of the REACH Regulation, to wait for a registrant to draw up an adaptation for its registration dossier and that there was no scientific justification for the studies required by the initial decision to be carried out sequentially, one after the other.

151    In the first place, it should be noted that the applicant does not dispute the fact that the ECHA is not obliged, pursuant to Article 41 of the REACH Regulation, to wait for a registrant to draw up an adaptation for its registration dossier.

152    In the second place, with regard to the fact that there are no scientific reasons justifying the sequential conduct of the OECD 408, OECD 414 and OECD 443 studies, the following should be noted.

153    First, the ECHA argues that none of the three studies can allow one of the other two to be omitted or adapted.

154    Secondly, even if it accepts that the OECD 414 Study could, in theory, show such severe effects in terms of developmental toxicity that the substance at issue would then meet the criteria for classification as a category 1A or 1B reproductive toxicant, the ECHA nevertheless emphasises that the fact that that substance is classified as such would not preclude its effects on fertility from being studied by means of an OECD 443 Study.

155    The applicant contests the ECHA’s arguments, arguing, in essence, that the conduct of an OECD 443 study following an OECD 414 study is not mandatory in all cases. However, it acknowledges that the advisability of conducting an OECD 443 Study cannot be ruled out.

156    In that context, it should be borne in mind that, according to settled case-law, in the context of an appeal against a decision adopted by the Board of Appeal, where the assessment of highly complex scientific and technical factual elements is involved, the review carried out by the EU Courts is limited. With regard to such assessments, the EU Courts are limited to reviewing whether they are vitiated by a manifest error, a misuse of powers or whether the author of the decision clearly exceeded the limits of its power of review (see judgment of 20 September 2019, Germany v ECHA, T‑755/17, EU:T:2019:647, paragraph 120 and the case-law cited).

157    Thus, in the light of the explanations provided by the ECHA, the applicant’s arguments and the case-law cited in paragraph 156 above, it must be concluded that it was without committing a manifest error of assessment that the Board of Appeal was able to consider that there were no scientific reasons justifying the sequential conduct of the OECD 408, OECD 414 and OECD 443 studies.

158    With regard to the applicant’s argument that carrying out the OECD 408, OECD 414 and OECD 443 studies in a non-sequential manner would result in the sacrifice of more animals and would be contrary to Article 25 of the REACH Regulation, it should be noted that that article, entitled ‘Objectives and General Rules’, states in paragraph 1 that, ‘in order to avoid animal testing, testing on vertebrate animals for the purposes of this Regulation shall be undertaken only as a last resort’.

159    According to the case-law, a registrant has, in general and therefore in particular in the event that the ECHA issues it with a decision asking it to supplement its registration dossier with a study involving animal testing, not simply the possibility but the obligation to generate information obtained by means other than animal testing ‘whenever possible’ and to undertake such testing ‘only as a last resort’ (judgment of 21 January 2021, Germany v Esso Refining, C‑471/18 P, EU:C:2021:48, paragraph 132).

160    It suffices to note that, according to Article 25 of the REACH Regulation, in the event that the ECHA asks a registrant to supplement its registration dossier, the latter is obliged to produce ‘whenever possible’ information obtained by means other than tests on animals and to carry out such tests ‘only as a last resort’.

161    In addition, it is clear from the assessment carried out above that it was without committing a manifest error of assessment that the Board of Appeal concluded that the OECD 408, OECD 414 and OECD 443 studies did not have to be carried out sequentially. Consequently, the ECHA’s request to carry out those studies non-sequentially is not contrary to Article 25 of the REACH Regulation.

162    Finally, the applicant’s argument concerning problems of laboratory capacity must be rejected. The action before the Board of Appeal had suspensory effect on the initial decision, so that any capacity problems of the laboratories during 2018 are irrelevant, in so far as, in paragraph 192 of the contested decision, the Board of Appeal set a new deadline of 25 February 2024 for the applicant to carry out the OECD 408, OECD 414 and OECD 443 studies.

163    Accordingly, the third plea in law must be rejected, and the action in its entirety must be dismissed.

 Costs

164    Under Article 134(1) of the Rules of Procedure of the General Court, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the successful party’s pleadings.

165    Since the applicant has been unsuccessful, it must be ordered to pay the costs, in accordance with the forms of order sought by the ECHA, including those relating to the proceedings for interim relief.

166    In accordance with Article 138(3) of the Rules of Procedure, CFE, PETA and EFfCI are to bear their own costs.

On those grounds,

THE GENERAL COURT (First Chamber)

hereby:

1.      Dismisses the action;

2.      Orders Symrise AG to pay its own costs as well as those incurred by the European Chemicals Agency (ECHA), including those relating to the interim proceedings;

3.      Orders Cruelty Free Europe (CFE), PETA International Science Consortium Ltd, PETA Science Consortium International eV and European Federation for Cosmetics Ingredients (EFfCI) each to pay their own costs.

Kanninen Jaeger Półtorak

Delivered in open court in Luxembourg on 22 November 2023.

V. Di Bucci M. van der Woude

Registrar President

*      Language of the case: English.