JUDGMENT OF THE GENERAL COURT (First Chamber)

22 November 2023 (*)

(REACH – Substance 2-ethylhexyl salicylate – Exclusive use for the manufacture of cosmetic products – Checking 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), Article 14(5)(b) of Regulation No 1907/2006 and Section 3 of Annex – Obligation to state reasons – Manifest error of assessment – Error of law)

In Case T‑655/20,

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,

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,

and by

Unilever U.K. Central Resources Ltd, established in London (United Kingdom),

and

Unilever ASCC AG, established in Schaffhouse (Switzerland),

represented by J.-P. Montfort, T. Delille and M. Pallu, lawyers,

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‑655/20 R, not published, EU:T:2021:98),

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, EFfCI, Unilever U.K. Central Resources and Unilever ASCC leave to intervene in support of the form of order sought by the applicant;

–        the decision by 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-010-2018 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 2-ethylhexyl salicylate (‘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 manufactures 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 which is used exclusively 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 5% in ready-for-use preparations.

5        On 8 December 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 19 December 2016, the ECHA submitted its proposal for an initial decision to the applicant, in which it was requested to conduct a standard toxicity study on pre-natal development (‘the OECD 414 Study’), an extended one-generation toxicity study (‘the OECD 443 Study’), a long-term study on aquatic invertebrate animals and a study involving aquatic testing on fish (‘the OECD 234 Study’). On 9 February 2017, the applicant submitted its observations on the proposed decision. In that regard, it submitted adaptations concerning both the OECD 414 Study, on the basis of Annex XI, Section 1.2, to the REACH Regulation, the OECD 443 Study, on the basis of Annex XI, Section 1.5, to that regulation, and the OECD 234 Study, on the basis of Annex XI, Section 1.1.2, to that regulation.

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 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 conduct four animal studies, while rejecting the adaptations proposed by the latter. The first, the OECD 414 Study, was to be conducted via the oral route on rats or rabbits. The second, the OECD 443 Study, was to be conducted via the oral route on rats. The third was a long-term study on invertebrate aquatic animals, to be conducted using the Daphnia magna method. Finally, the fourth, the OECD 234 Study, was to be conducted on the sexual development of fish. 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 414, OECD 443 and OECD 234 studies and in so far as it made the production of those studies conditional on a time limit of approximately 42 months.

11      On 18 August 2020, the Board of Appeal issued the contested decision dismissing 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), Unilever U.K. Central Resources Ltd and Unilever ASCC AG (together, ‘Unilever’) claim 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 five pleas in law. The first plea alleges errors of assessment and of interpretation of the REACH Regulation committed by the ECHA in requesting that tests on vertebrate animals be carried out. The second plea alleges a manifest error of assessment by the ECHA in relation to the OECD 443 Study and a breach of the duty to state reasons. The third plea alleges several manifest errors of assessment by the ECHA in requiring the OECD 443 Study to be carried out via the oral route and a breach of the duty to state reasons. The fourth plea alleges that the ECHA committed a manifest error of assessment and misinterpretation of Section 9.1, column 2, of Annex IX to the REACH Regulation by requiring the OECD 234 Study to be carried out, an infringement of the right to be heard and an infringement of Article 25 of the REACH Regulation. The fifth 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 raises 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 concerning 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 only apply to a substance, used exclusively as a cosmetic ingredient, when they do not concern testing to establish effects on human health.

20      In addition, the applicant submits 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 does 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 vertebrate animal testing, which falls 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 vertebrate animal testing 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 takes the view 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      Furthermore, first, CFE submits that all the data contained in the information file drawn up under Article 11 of the Cosmetics Regulation are mentioned in the product safety report compiled in accordance with 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      Secondly, Unilever claims that the ECHA committed an error of assessment and an error of interpretation of the REACH Regulation, in particular Article 2(4)(b) thereof, by requesting that studies on vertebrate animals be carried out in relation to a substance which was used exclusively in cosmetic products. Such a request would make it impossible for economic operators such as the members of its group to comply with the commitments they have made to put an end to those tests, which demonstrates that the animal welfare objectives of the REACH Regulation and the Cosmetics Regulation would not be met.

31      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.

32      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.

33      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).

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

35      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 414 Study and the OECD 443 Study. 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 and to take into consideration all the relevant elements and circumstances of the situation which the contested measure was intended to govern.

36      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.

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

38      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.

39      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.

40      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 a ‘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.

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

42      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.

43      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 considering consumer exposure.

44      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. In fact, products containing the substance at issue would be designed to be applied repeatedly to the skin whereas, for workers, the objective would be 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.

45      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 package of tests on animals 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.

46      Unilever adds that, in order to demonstrate that the exposure of workers to the substance at issue is negligible or harmless, it asked an independent toxicology consultancy to carry out an assessment of the exposure of workers at its factory in Buxtehude (Germany). Unilever claims that the use of the substance at issue in the factory is typical of the use made of it by users within the meaning of the REACH Regulation. It points out that the results of the expert report, attached to its statement in intervention, show that, if it is accepted, as is reasonable to do, that the Buxtehude plant complies with the EU health and safety regulations in force, the exposure of workers to the substance at issue is probably about eight times lower than that of consumers in the context of normal use of the product. Thus, according to Unilever, since the ECHA recognises that it is clear from the opinion of the SCCS that the risk to consumers resulting from exposure to the substance at issue is sufficiently low to be acceptable, there does not appear to be any valid reason relating to worker exposure to justify carrying out new tests on vertebrate animals, as that exposure is considerably lower than that considered safe for consumers.

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

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

48      As a preliminary point, in paragraphs 55 to 57 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.

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

50      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 94 of the contested decision, that that regulation did not contain any provision exempting a registrant from carrying out studies on vertebrate animals by reason solely of the fact that the substance for which registration was sought was used solely 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.

51      As regards 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.

52      In the first place, in paragraphs 102 to 105 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 the performance of 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.

53      In the second place, the Board of Appeal considered, in paragraphs 106 to 108 of the contested decision, that Article 18(1)(b) and (2) of the Cosmetics Regulation provided for a prohibition on the 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.

54      In that context, the Board of Appeal considered, in paragraphs 109 and 110 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 that 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.

55      However, the Board of Appeal observed, in paragraphs 111 and 112 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 that 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.

56      Thus, in paragraph 113 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 of the placing on the market provided for in Article 18(1)(b) and (2) of the Cosmetics Regulation.

57      In the light of those factors, the Board of Appeal concluded, in paragraph 117 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.

58      As regards the relationship between the REACH Regulation and the Cosmetics Regulation, in paragraph 118 of the contested decision, first, the Board of Appeal concluded 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 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 95 to 117 of the contested decision.

59      Thus, in paragraphs 119 to 123 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) of that regulation, 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 414 and OECD 443 studies or by proposing valid adaptations.

60      As regards the fact that the ECHA did not take into account that the safety of the substance at issue had been assessed under the Cosmetics Regulation, the Board of Appeal held, in paragraphs 128 to 131 of the contested decision, first, that, even assuming that the substance at issue did not constitute a risk to the health of workers, the applicant could not refrain from providing the OECD 414 and OECD 443 studies unless it had 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

61      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.

62      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.

63      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).

64      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.

65      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

66      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 the 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 applicable without prejudice to the provisions of EU law relating to the workplace and the environment and to the Cosmetics Regulation as regards tests on vertebrate animals 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 the REACH Regulation states that the provisions of Titles II, V, VI and VII thereof shall not 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.

67      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.

68      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.

69      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.

70      Accordingly, it must be concluded that the Board of Appeal did not err in law when it held, in essence, in paragraphs 65 to 77 of the contested decision, 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 obligations arising from that regulation.

71      In the second place, it is necessary 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]’.

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

73      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.

74      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), the 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.

75      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.

76      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.

77      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, 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.

78      Consequently, it was without committing an error of law that, in paragraph 103 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.

79      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’.

80      It is sufficient 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 concerning 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.

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

82      In addition, that conclusion is not called into question by Unilever’s argument that the contested decision makes it impossible for economic operators such as the members of its group to comply with the commitments they have entered into to put an end to animal testing. In that regard, first, it is sufficient to note that that argument is irrelevant for the purposes of interpreting the relationship between the Cosmetics Regulation and the REACH Regulation. Secondly, although the contested decision does require the applicant to carry out certain studies on vertebrate animals, the applicant is still in a position to avoid them by producing valid adaptations, so that the contested decision cannot be regarded as the direct source of an obligation to carry out studies contrary to a policy chosen by economic operators with regard to animal testing.

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

83      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, at the outset, that the REACH Regulation and the Cosmetics Regulation overlap as regards one of the objectives which they pursue, namely that of protecting human health.

84      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’.

85      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, as in the present case, in the formulation of an ingredient for a cosmetic product or in the formulation of the product itself are, by contrast, exposed to that ingredient throughout the manufacturing cycle and under conditions that differ from the use of the final cosmetic product.

86      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.

87      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]’.

88      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 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 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.

89      Furthermore, the following should be noted.

90      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.

91      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.

92      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 in relation 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 85 above), whereas the REACH Regulation covers risks to human health relating to all exposures throughout the life cycle of the substance (see paragraph 91 above), including, in particular, risks to which workers are exposed.

93      Thus, it was without committing an error of law that the Board of Appeal held, in paragraphs 86 to 93 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 which fell within the scope of 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.

94      Consequently, in view of the finding made in paragraph 70 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.

95      It follows that it was without committing an error of law that the Board of Appeal concluded, first, in paragraphs 130 and 131 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 86 to 93 of the contested decision, that the registrant of a substance contained 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.

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

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

98      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’.

99      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.

100    Thus, as noted in paragraph 95 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.

101    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.

102    As regards 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’.

103    Furthermore, the danger to human health resulting, in particular, 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.

104    With regard to the applicant’s argument concerning 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 controls 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.

105    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.

106    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.

107    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 noted 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.

108    Furthermore, the examination carried out by the Board of Appeal in paragraphs 128 to 131 of the contested decision cannot be called into question either 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, or by Unilever’s argument that tests carried out in one of its factories where the substance at issue is produced have produced results which demonstrate that there is no valid reason relating to the risks to the health of workers to justify carrying out new tests on vertebrate animals, their exposure being considerably lower than that of consumers, which is considered to be safe.

109    As regards PETA’s argument, it is sufficient to note that, in response to a question put by the Court at the hearing, the latter confirmed that it 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).

110    The same applies to Unilever’s argument, which concerns tests carried out after the date on which the contested decision was adopted. Moreover, the ECHA, in paragraph 72 of its observations on the statements in intervention, does not rule out the possibility that that type of information is capable of supporting an adaptation under Section 3 of Annex XI to the REACH Regulation, read in conjunction with Article 14(5)(b) of that regulation.

111    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 contained 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.

112    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.

113    Furthermore, and in any event, as the Board of Appeal observed in paragraphs 111 and 112 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 the placing of that product on the market would not be the consequence of the carrying out of that study, carried out in order 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 on condition that vertebrate animals are not sacrificed in order for its safety to be established.

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

 The second plea in law, alleging a manifest error of assessment by the ECHA with regard to the OECD 443 Study and breach of the duty to state reasons

115    In support of its second plea in law, the applicant raises two parts, the first alleging a manifest error of assessment by the ECHA in failing to take account of all the relevant information and the second alleging a breach of the duty to state reasons.

116    By the first part, first of all, the applicant claims that, pursuant to column 1 of Section 8.7.3 of Annex IX to the REACH Regulation, registrants are required to submit information concerning the OECD 443 Study only ‘if the available repeated dose toxicity studies (e.g. 28-day or 90-day studies, or [OECD] 421 or 422 screening studies) indicate adverse effects on reproductive organs or tissues or reveal other concerns in relation with reproductive toxicity’. The applicant maintains that, in view of the results of the OECD 421 Study attached to its registration dossier, the requirement to conduct the OECD 443 Study was unnecessary.

117    Moreover, in the first place, the applicant submits that the ECHA wrongly relied on irrelevant statistical effects to justify its request to conduct the OECD 443 Study.

118    In that context, the applicant points out that conclusions which are not statistically significant may be used, under the terms of the Organisation for Economic Co-operation and Development (OECD) document entitled ‘Guidance Notes for Analysis and Evaluation of Repeat-Dose Toxicity Studies’ (‘the OECD guidance document’), only ‘when considered in the light of the likely toxicological or pharmacological action of the compound, or when combined with results from other studies’. The applicant observes that the OECD guidance document emphasises that ‘this information may be useful when comparing results across studies and in the consideration of the overall significance or relevance of an observed effect i.e. in one study an effect may be only a trend whilst in another study it may be very clearly treatment-related’.

119    The applicant concludes that, in the contested decision, the ECHA ignored the limited value which the OECD guidance document attributed to statistically insignificant results and did not compare the results of the various studies in order to assess their overall relevance.

120    The applicant adds that the Board of Appeal considered that ‘an effect [could] be biologically relevant even if it [were] not statistically significant’ without having carried out a concrete assessment of the facts of the case, whereas the excessive value attributed by the ECHA to statistically irrelevant effects does not appear to comply with normal standards and guidelines. Furthermore, the ECHA did not put forward any reasons why the extended gestation period could in fact have been considered biologically relevant in this particular case.

121    In the second place, the applicant criticises the ECHA for having relied on the reduction in the average number of live births, the reduction in the birth index and the reduction in body weight, that is to say the ‘other effects’, and for having considered that those were ‘in themselves sufficient to justify the requirement to provide information on an [OECD 443 Study] under Column 1 of Section 8.7.3 of Annex IX [to the REACH Regulation]’. In that regard, the applicant submits that those ‘other effects’ occurred at high dose levels, which is obvious for pronounced systemic toxicity but less so for the reproductive system, and are therefore irrelevant.

122    The applicant adds that those ‘other effects’ relating to developmental toxicity should be addressed by means of the existing OECD 408 Study, combined with the required OECD 414 Study, which examines effects on offspring, in other words effects on development. However, the ECHA reportedly examined the two studies separately and formalistically considered that the required OECD 414 Study could ‘not be relied on to dispel all the existing concerns’ relating to ‘other effects’.

123    By its second part, the applicant submits that the ECHA failed to state the factual and legal grounds on which the contested decision is based.

124    The ECHA disputes the applicant’s arguments.

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

125    In its action against the initial decision, the applicant argued that the OECD 421 Study showed harmful effects. However, according to the latter, those effects could not lead to the conclusion that the information relating to the OECD 443 Study was standard information. According to the applicant, in the first place, the harmful effects identified in the OECD 421 Study concerned development and not reproduction and were not statistically significant. In the second place, the OECD 414 Study required in the initial decision addressed those effects and therefore rendered the OECD 443 Study unnecessary. In the third place, the OECD 408 Study demonstrated that the substance at issue had no harmful effects on reproductive organs.

126    In paragraphs 149 and 150 of the contested decision, the Board of Appeal rejected the first argument put forward by the applicant, holding that the OECD 421 Study showed effects, such as prolongation of gestation at 80 and 250 milligrams per kilogram (mg/kg) of body weight per day, which could be associated with reproductive toxicity. In addition, the Board of Appeal considered that the fact that those effects were not statistically significant was not sufficient to conclude that the OECD 421 Study did not demonstrate the existence of harmful effects, in so far as an effect may be biologically significant even if it is not statistically significant. Finally, the Board of Appeal considered that the OECD 421 Study revealed the existence of other effects that could be associated with reproductive toxicity, which were also statistically significant, consisting of a reduction in the average number of live births to 250 mg/kg of body weight per day, a reduction in the birth index to 80 and 250 mg/kg of body weight per day and a reduction in the weight of puppies to 250 mg/kg of body weight per day.

127    In paragraphs 153 to 155 of the contested decision, the Board of Appeal rejected the second argument put forward by the applicant, holding that, first, no OECD 414 Study existed for the substance at issue, secondly, some of the effects identified in the OECD 421 Study, such as the prolongation of gestation, were not investigated in an OECD 414 Study and, thirdly, the initial decision did not prevent the applicant from carrying out the OECD 414 Study and submitting an adaptation rather than carrying out the OECD 443 Study.

128    In paragraphs 158 and 159 of the contested decision, the Board of Appeal rejected the third argument put forward by the applicant, holding that the OECD 408 Study had not been submitted during the proceedings and that, therefore, that argument was not sufficiently supported by evidence. In addition, the Board of Appeal noted, first, that the OECD 421 Study was a study for the purpose of examining the effects of that substance following repeated short-term exposure and its effects on the reproductive capacity of males and females and, secondly, that the purpose of the OECD 408 Study was to assess toxicity in adult animals, including non-pregnant females, so that reproductive and developmental toxicity parameters were not examined. The Board of Appeal concluded that the absence of effects in the OECD 408 Study was not likely to address the concerns identified in the OECD 421 Study with regard to pregnant females and their puppies.

 The merits of the second plea in law

129    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. Consequently, in the context of such an action, the applicant may not limit 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).

130    In the second place, as regards the Board of Appeal’s review of the scientific assessments contained in an ECHA decision, it should be borne in mind 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).

131    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).

132    Moreover, it should be noted that, in essence, the applicant does not contest the fact that the OECD 443 Study may be requested by the ECHA, but submits that the OECD 421 Study sufficiently demonstrated that the OECD 443 Study was neither justified nor necessary in the circumstances of the case.

133    However, it follows from column 1 of Section 8.7.3 of Annex IX to the REACH Regulation that registrants are only required to submit information on the OECD 443 Study if the available repeated dose toxicity studies indicate harmful effects on reproductive organs or tissues or raise other concerns regarding reproductive toxicity. In that regard, the Board of Appeal, in confirming the ECHA’s assessment, considered that, for an OECD 443 Study to be carried out, it was sufficient for there to be concerns about reproductive toxicity.

134    In that context, it was without committing a manifest error of assessment that the Board of Appeal was able to consider that such concerns could arise even in the presence of effects which are not significant from a statistical point of view, but which are significant from a biological point of view, which, as the ECHA maintains, is justified by the OECD guidance document, cited by the applicant incompletely. In fact, that guidance document states that a finding that is not statistically significant may have biological significance when considered in the light of the likelytoxicological or pharmacological action of the compound or when combined with results from other studies. That confirms that the applicant’s argument that the Board of Appeal should have compared different studies cannot succeed. The guidance document states that ‘biological significance’ can be taken into account in the light of the likely toxicological or pharmacological action of the compound or when combined with results from other studies, which implies that it is sufficient for one of those two circumstances to exist for ‘biological significance’ to be taken into account.

135    Finally, even if the applicant contests that as a formality, the Board of Appeal justified the need to carry out the OECD 443 Study by, inter alia, the absence of an OECD 414 Study carried out on the substance at issue. Thus, contrary to what the applicant maintains, the contested decision is reasoned in that regard.

136    Accordingly, the second plea in law must be rejected in its entirety.

 The third 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

137    In support of its third 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 for 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 for 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 for human exposure and, thirdly, that Commission Regulation (EC) No 440/2008 of 30 May 2008 laying down test methods pursuant to Regulation (EC) 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 for human exposure. For example, the most appropriate route for exposure, including worker exposure, would be the dermal route, and that is the route via which the OECD 443 Study should have been conducted.

138    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, arguing that ‘using the dermal route would therefore be unlikely to lead to sufficient foetal exposure to give meaningful results’. The applicant submits that that approach lacks any basis in 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 1 and 3% of the substance being found in the body, does not make the oral route the ‘likely route of human exposure’.

139    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 oral study be carried out by making only vague observations, even though that was not the most likely route for human exposure. Furthermore, according to the applicant, since the ECHA had departed from the most appropriate route for 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.

140    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 the applicant submits 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.

141    In the fourth place, the applicant contests the reference made by the ECHA 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 for 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.

142    In the fifth place, the applicant submits 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, an OECD 443 Study does not investigate the short-term, acute exposure to a substance – such as that which would arise from accidental inhalation or ingestion – but rather the 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.

143    Finally, the applicant points out that the ECHA maintains 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.

144    The ECHA disputes the applicant’s arguments.

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

145    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.

146    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’.

147    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.

148    In the light of those factors, in paragraphs 171 to 173 of the contested decision, first, the Board of Appeal rejected the applicant’s claim, holding that, although the most likely route of exposure to the substance at issue for humans in its current use in cosmetic products for application to the skin was the dermal route, exposure to the substance at issue was nevertheless 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 1 to 3% 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 via the dermal route in an OECD 443 Study would lead to sufficient exposure of the foetus to give significant results, whereas administration by the oral route would have maximised the probability of obtaining useful results for assessing the potential toxicity of the substance at issue on reproduction.

 The merits of the third plea in law

149    It should be noted that the EU Courts are to carry out a review of 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 facts. With regard to such an assessment, 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 has 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).

150    It should be noted that column 1 of Section 8.7.3 of Annex IX to the REACH Regulation states that the OECD 443 Study is intended to be carried out by the most appropriate route of administration relating to the most likely human exposure. However, without being explicitly contradicted on that point, the ECHA submits that, as the Board of Appeal found in paragraphs 168 and 169 of the contested decision, point 18 of Test Method B.56 clearly indicates that the most likely route for human exposure is one of the elements, but not the only one, to be taken into consideration when deciding on the route of administration, depending on the characteristics of the substance at issue 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.

151    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 recalled that, in paragraph 95 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.

152    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.

153    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.

154    Thirdly, as regards 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 145 to 148 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.

155    Fourthly, as regards the applicant’s argument, referred to in paragraph 140 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.

156    The third plea in law must therefore be rejected.

 The fourth plea in law, alleging, in essence, a manifest error of assessment and interpretation of column 2 of Section 9.1 of Annex IX to the REACH Regulation by the ECHA in requiring the OECD 234 Study to be carried out, an infringement of the right to be heard and a breach of Article 25 of the REACH Regulation

157    In support of its fourth plea in law, the applicant raises three parts, the first alleging a manifest error of assessment and interpretation of column 2 of Section 9.1 of Annex IX to the REACH Regulation committed by the ECHA in requiring the OECD 234 Study to be carried out, the second alleging an infringement of the right to be heard and the third alleging a breach of Article 25 of the REACH Regulation.

 The first part of the fourth plea in law, alleging that the ECHA committed a manifest error of assessment and interpretation of column 2 of Section 9.1 of Annex IX to the REACH Regulation by requiring the OECD 234 Study to be carried out

158    In the first place, the applicant interprets Section 9.1 of Annex IX to the REACH Regulation on the basis of the texts in columns 1 and 2 thereof. Thus, according to the applicant, it is clear from a combined reading of those columns that column 1 lists the standard information to be provided and that column 2 allows adaptations to or derogations from the standard requirements if the chemical safety assessment, carried out in accordance with Annex I to the REACH Regulation, indicates that it is necessary to study the effects on aquatic organisms in greater depth, those adaptations or derogations being dependent on the chemical safety assessment of the substance.

159    The applicant states that column 1 of Annex IX to the REACH Regulation is entitled, like Annexes VII, VIII and X to that regulation, ‘Standard information required’ and that column 2 of Annex IX to the REACH Regulation is entitled ‘Specific rules for adaptation from column 1’. Furthermore, Annex XI to the REACH Regulation states that, ‘in addition to the specific rules set out in column 2 of Annexes VII to X, a registrant may adapt the standard testing regime in accordance with the general rules set out in Section 1 of this Annex’.

160    The applicant adds that the specific provisions of column 2 of Annex IX to the REACH regulation set out the rules for adapting the ‘standard information required’ by column 1 of that annex, as is made clear in the introductory paragraph of that annex, which states that column 2 lists the specific rules under which the registrant may propose to omit the standard information, replace it with other information, provide it at a later stage or adapt it in some other way, it being understood that, if the conditions to which that column makes adaptation proposals subject are met, the registrant is to clearly state that and the reasons for each adaptation proposal under the appropriate headings of the registration dossier.

161    Furthermore, the applicant claims that the provisions of column 2 of Section 9.1 of Annex IX to the REACH Regulation refer to the tests listed in column 1 of Section 9.1.6 of Annex IX to that regulation and not to ‘further studies’. In that regard, the applicant points out that, where the provisions in column 2 of the annexes to the REACH Regulation refer to studies other than those listed in column 1, that is clearly indicated. For example, according to the applicant, the last paragraph of column 2 of Section 8.6.2 of Annex IX to the REACH Regulation and the first paragraph of column 2 of Section 8.6.4 of Annex X to that regulation state that ‘further studies shall be proposed by the registrant or may be required by the Agency in accordance with Article 40 or 41’. Thus, the term ‘further’ indicates that at issue are other tests and further tests in relation to those indicated in column 1 of the sections in question.

162    The applicant states that, where the annexes to the REACH Regulation provide that the ECHA may require ‘further studies’ without defining their purpose, the choice of study falls within the ECHA’s discretion, as was confirmed by the Board of Appeal in another of its decisions. According to the applicant, it follows, a contrario, that where a column defines the purpose of the studies, as in the case of column 2 of Section 9.1 of Annex IX to the REACH Regulation, which cites ‘long-term toxicity testing’, the ECHA’s discretion in the context of a compliance check is limited to the studies cited.

163    In the second place, the applicant submits that the OECD 234 Study prejudges the assessment of the substance at issue. According to it, in the context of that assessment, it could very well be that no other information or information different from the OECD 234 Study would be requested.

164    In the third place, the applicant claims that, in the light of the chemical safety assessment, the request to carry out the OECD 234 Study was unnecessary. Furthermore, the applicant adds that the requirement for additional studies in the present case depended on the results of the OECD 443 Study requested in the initial decision, which was not the subject of an action before the Board of Appeal.

165    In the fourth place, the applicant contests the argument put forward by the ECHA that its proposed interpretation of column 2 of Section 9.1 of Annex IX to the REACH Regulation runs counter to the relationship between that regulation and Section 9.1.3 of Annex VIII to that regulation, relating to short-term toxicity tests on fish, since that interpretation would allow registrants to dispense with carrying out studies. In fact, in its defence, the ECHA argues that, if the applicant’s interpretation of column 2 of Section 9.1 of Annex IX to the REACH Regulation were to be accepted, that would render the requirements for information on aquatic toxicity in Annexes VIII and IX to that regulation meaningless in the case of poorly soluble substances.

166    The applicant considers that that statement is inaccurate having regard, first, to the operation of Annex VIII to the REACH Regulation, in so far as, in the case of substances of low water solubility, testing is not dispensed with, but the registrant must consider whether it is appropriate to replace short-term testing with long-term testing and, secondly, to the circumstances of the case in point.

167    Furthermore, in the light of the circumstances of the case and in particular the fact that the substance at issue is not very soluble, the applicant states that it considered whether it would be appropriate to carry out a long-term study under Section 9.1.6 of Annex IX to the REACH Regulation. However, having taken into consideration the low solubility of the substance at issue and the results of the short-term testing, as well as the fact that the substance at issue is readily biodegradable and has a high ‘log Kow’, the applicant concluded that, in that particular case, the chemical safety assessment did not ‘indicate the need for further analysis of the effects on aquatic organisms’ and relied on column 2 of Section 9.1 of Annex IX to the REACH Regulation not to carry out the long-term toxicity tests on fish.

168    Consequently, having concluded that there was no need to replace the short-term toxicity tests with long-term toxicity tests, the applicant carried out all the short-term aquatic toxicity studies required under Annexes VII and VIII to the REACH Regulation, in the present case on algae, invertebrates and fish.

169    The ECHA disputes the applicant’s arguments.

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

170    Before the Board of Appeal, the applicant claimed that the ECHA had committed several errors in the initial decision. In the first place, according to the applicant, the ECHA exceeded its powers by requiring it to carry out, in the context of a compliance checking procedure based on Article 41 of the REACH Regulation, the OECD 234 Study. In that regard, the applicant claimed that column 1 of Section 9.6.1 of Annex IX to the REACH Regulation concerning the standard information to be submitted referred only to the OECD 210 Study and not to the OECD 234 Study. In the second place, the applicant claimed that the ECHA could not request an OECD 234 Study in the context of a compliance checking procedure on the basis of Article 41 of the REACH Regulation, but only in the context of a substance evaluation procedure on the basis of Article 46 of that regulation. In the third place, the applicant argued that the initial decision infringed Article 25 of the REACH Regulation, in so far as the OECD 234 Study involved the sacrifice of a greater number of fish than the OECD 210 Study.

171    First of all, the Board of Appeal noted, in paragraphs 180 to 183 of the contested decision, that, in the initial decision, the ECHA had, in essence, established that the OECD 234 Study was to be regarded as standard information for the registration of the substance at issue on the basis of column 1 of Section 9.1.6.1 of Annex XI to the REACH Regulation, in so far as that study was an improved version of the OECD 210 Study, which made it possible to clarify whether that substance was an endocrine disruptor. The Board of Appeal observed that column 1 of Section 9.1.6.1 of Annex IX to the REACH Regulation contained an exhaustive list of studies that constituted standard information, which did not include the OECD 234 Study.

172    The Board of Appeal concluded, in paragraph 184 of the contested decision, that the ECHA had erred in considering the OECD 234 Study to be standard information within the meaning of column 1 of Section 9.1.6 of Annex IX to the REACH Regulation.

173    However, the Board of Appeal noted, in paragraphs 185 and 186 of the contested decision, that column 2 of Section 9.1 of Annex IX to the REACH Regulation allowed specific adaptations to be made to the standard information included in column 1, in so far as column 2 established that a registrant was required to submit a study different from the three studies referred to in column 1 in the event that the chemical safety report indicated that it was necessary to examine the effects of a substance on aquatic organisms in a manner other than that which those studies were able to guarantee.

174    Next, the Board of Appeal held, in paragraphs 187 and 188 of the contested decision, that, in the context of a compliance procedure provided for in Article 41 of the REACH Regulation, the ECHA could require a registrant to carry out a study other than the three studies referred to in column 1 of Section 9.1 of Annex IX to the REACH Regulation if the information available indicated that it was necessary to investigate the effects of a substance on aquatic organisms in a manner other than that which those three studies were able to guarantee. The Board of Appeal added that column 2 of Section 9.1 of Annex IX to the REACH Regulation referred, first, to other ‘long-term toxicity testing’ and, secondly, to the fact that it required further investigation of the effects of the substance at issue.

175    Thus, in the light of a combined interpretation of Article 41 of the REACH Regulation, relating to the conformity procedure for a dossier, and column 2 of Section 9.1 of Annex IX to that regulation, the Board of Appeal held, in paragraphs 189 and 190 of the contested decision, that the ECHA enjoyed a margin of discretion not only to determine whether a study different from those referred to in column 1 was necessary, but also to indicate which study should be carried out. In addition, the Board of Appeal noted that that study had to be necessary, proportional and in compliance with Article 25 of the REACH Regulation.

176    In addition, the Board of Appeal considered, in paragraph 191 of the contested decision, that the initial decision explained, first, that the information contained in the applicant’s file demonstrated that the substance at issue could be an endocrine disruptor, secondly, that the OECD 234 Study could clarify whether that was the case, unlike the three studies referred to in column 1 of Section 9.1.6 of Annex IX to the REACH Regulation and, thirdly, that, even though the OECD 234 Study involved using a higher number of vertebrate animals than the OECD 210 Study, that avoided a scenario in which the OECD 234 Study would have to be carried out after the OECD 210 Study had been carried out in order to clarify the question of endocrine disrupting properties.

177    Finally, the Board of Appeal noted, in paragraphs 192 to 194 of the contested decision, first, that the applicant had not contested before it the grounds for the initial decision and that it had therefore not examined them and, secondly, that the applicant had not proposed alternative solutions which would have made it possible to verify that the substance at issue did not produce endocrine disrupting properties by means of a less expensive study or with the use of fewer vertebrate animals. Accordingly, the Board of Appeal rejected the applicant’s arguments.

–       The merits of the first part of the fourth plea in law

178    In support of the first part of the fourth plea in law, the applicant raises, in essence, four complaints, the first alleging an error of law in the interpretation of column 2 of Section 9.1 of Annex IX to the REACH Regulation, the second alleging that the request to carry out the OECD 234 Study prejudges the outcome of the procedure, the third alleging that that request was unjustified and the fourth alleging an error of law in the interpretation of the relationship between Annexes VIII and IX to that regulation.

179    As a preliminary point, first, it should be noted that columns 1 and 2 of Section 9.1.3 of Annex VIII to the REACH Regulation are worded as follows:

Column 1 Standard information required

Column 2 Specific rules for adaptation from column 1

9.1.3. Short-term toxicity testing on fish: the registrant may consider long-term toxicity testing instead of short-term.

9.1.3. The study does not need to be conducted if: – there are mitigating factors indicating that aquatic toxicity is unlikely to occur, for instance if the substance is highly insoluble in water or the substance is unlikely to cross biological membranes, or – a long-term aquatic toxicity study on fish is available. Long-term aquatic toxicity testing as described in Annex IX shall be considered if the chemical safety assessment according to Annex I indicates the need to investigate further effects on aquatic organisms. The choice of the appropriate test(s) will depend on the results of the chemical safety assessment. The long-term aquatic toxicity on fish (Annex IX, Section 9.1.6) shall be considered if the substance is poorly water soluble.

180    Secondly, columns 1 and 2 of Section 9.1 of Annex IX to the REACH Regulation are worded as follows:

Column 1 Standard information required	Column 2 Specific rules for adaptation from column 1
9.1. Aquatic toxicity …	9.1. Long-term toxicity testing shall be proposed by the registrant if the chemical safety assessment according to Annex I indicates the need to investigate further the effects on aquatic organisms. The choice of the appropriate test(s) depends on the results of the chemical safety assessment.
9.1.6. Long-term toxicity testing on fish, (unless already provided as part of Annex VIII requirements). The information shall be provided for one of the Sections 9.1.6.1, 9.1.6.2 or 9.1.6.3.
9.1.6.1. Fish early life stage (FELS) toxicity test
9.1.6.2. Fish short-term toxicity test on embryo and sac-fry stages
9.1.6.3. Fish, juvenile growth test

181    In addition, it should be noted that the first, fourth and fifth sentences of the second paragraph of Annex IX to the REACH Regulation provide that ‘column 1 of this Annex establishes the standard information required for all substances manufactured or imported in quantities of 100 tonnes or more’ and that ‘column 2 of this Annex lists specific rules according to which the registrant may propose to omit the required standard information, replace it by other information, provide it at a later stage or adapt it in another way’, while specifying that, ‘if the conditions are met under which column 2 of this Annex allows an adaptation to be proposed, the registrant shall clearly state this fact and the reasons for proposing each adaptation under the appropriate headings in the registration dossier’.

182    Moreover, as regards the first complaint in the first part of the fourth plea in law, alleging an error of law in the interpretation of column 2 of Section 9.1 of Annex IX to the REACH Regulation, the relationship between column 2 and the corresponding column 1 must be clarified.

183    In that regard, the following should be noted.

184    The fourth sentence of the second paragraph of Annex IX to the REACH Regulation states that each column 2 lists the specific rules under which the registrant may propose to omit the required standard information, replace it with other information, provide it at a different stage or adapt it in some other way.

185    Nevertheless, with regard to column 2 of Section 9.1 of Annex IX to the REACH Regulation, it should be noted that it contains a specific rule according to which, if the chemical safety assessment of a substance justifies it, a registrant cannot simply refrain from providing the information due on the basis of the corresponding column 1, but is required to replace that information with other studies or to adapt that information.

186    Consequently, the applicant’s argument that a registrant should submit one of the studies referred to in column 1 cannot succeed. In the presence of elements which justify it, namely, in particular, the assessment of the chemical safety of a substance, the registrant is required to carry out the study which best meets the objective of that assessment, in accordance with column 2, even if that study is not among those referred to in column 1.

187    Such an obligation on the registrant is confirmed by the wording of column 2, which refers generically to ‘long-term toxicity testing’, necessarily implying that it is not only the studies expressly referred to in the corresponding column 1. Thus, the argument of a textual nature raised by the applicant, according to which column 2 of Section 9.1 of Annex IX to the REACH Regulation refers to the tests included in column 1 of Section 9.1.6 of that annex and not to ‘further studies’, whereas other provisions in column 2 of the various annexes refer to studies other than those listed in the corresponding column 1, cannot succeed. While it is true that that expression appears in column 2 of certain annexes to the REACH Regulation, the fact remains that, for the reasons set out above, the studies referred to in column 2 of Annex IX to that regulation cannot be regarded as being those referred to in the corresponding column 1.

188    Furthermore, it should be noted that if, as in the present case, in the context of a compliance checking procedure initiated on the basis of Article 41 of the REACH Regulation, a registrant who is required to provide a testing proposal for ‘long-term toxicity’ within the meaning of column 2 of Section 9.1 of Annex IX to that regulation fails to do so, the ECHA may indicate the test that it considers necessary in relation to the registration dossier.

189    Thus, it must be concluded that the Board of Appeal did not err in law in interpreting column 2 of Section 9.1 of Annex IX to the REACH Regulation as allowing a request to be made for the OECD 234 Study to be carried out. Consequently, the applicant’s first complaint cannot succeed.

190    As regards the second complaint in the first part of the fourth plea in law, alleging that the OECD 234 Study prejudged the assessment of the substance at issue, it should be noted that Article 47 of the REACH Regulation provides that ‘an evaluation of a substance shall be based on all relevant information submitted on that particular substance and on any previous evaluation under this Title’. Thus, the evaluation of a substance is based on all available information, including information resulting from the compliance check of a registration application carried out under Article 41 of the REACH Regulation, which appears in the same title as Article 47 of that regulation.

191    Therefore, contrary to what the applicant claims, the interpretation made by the ECHA and by the Board of Appeal is based on the premiss that, if the chemical safety assessment reveals the need to study the effects of a substance on vertebrate aquatic organisms beyond the information which could be offered by the studies provided for in column 1 of Annex IX to the REACH Regulation, that implies that other studies must be carried out on the basis of the rule established in the corresponding column 2, without that being considered as a circumstance prejudging the registration of the substance at issue, subject to the circumstance that those other studies are necessary, proportional and in compliance with Article 25 of the REACH Regulation.

192    It must therefore be concluded that the Board of Appeal did not commit a manifest error of assessment in considering that the OECD 234 Study could be requested in the context of a compliance check procedure under Article 41 of the REACH Regulation. Accordingly, the applicant’s second complaint cannot succeed and must be rejected.

193    As regards the third complaint in the first part of the fourth plea in law, alleging that the request to carry out the OECD 234 Study was unjustified, the applicant has not adduced any evidence to suggest that the contested decision is vitiated by a manifest error of assessment in that regard.

194    The applicant confines itself to asserting that the OECD 234 Study does not constitute a standard information requirement, so that the ECHA could not require it, even if the concern relating to possible endocrine disrupting properties was relevant according to the latter, by challenging the premiss of the ECHA’s approach, namely that the substance at issue was a potential endocrine disruptor. However, for the reasons set out in paragraphs 187 to 189, 191 and 192 above, in the context of the first and second parts of the present plea, that premiss is not vitiated by a manifest error of assessment, since the ECHA was able to consider, as the Board of Appeal concluded in paragraph 191 of the contested decision, that the information contained in the applicant’s file demonstrated that the substance at issue could be an endocrine disruptor and that the OECD 234 Study could clarify whether that was the case, unlike the three studies referred to in column 1 of Section 9.1.6 of Annex IX to the REACH regulation. That complaint should therefore be rejected.

195    As regards the fourth complaint in the first part of the fourth plea in law, alleging an error of law in the interpretation of the relationship between Annexes VIII and IX to the REACH regulation, it should be noted that it is raised by the applicant in response to an argument put forward by the ECHA to show that its interpretation of Annex IX to that regulation is incorrect, in the light of the context of that annex and, in particular, in relation to Annex VIII to that regulation.

196    In essence, the applicant considers that the wording of column 2 of Section 9.1.3 of Annex VIII to the REACH Regulation, which refers to the long-term aquatic toxicity study on fish provided for in Section 9.1.6 of Annex IX to that regulation, confirms that column 2 of Section 9.1 of Annex IX to that regulation does not concern ‘further studies’, but rather the long-term toxicity studies described in the corresponding column 1.

197    In that regard, it is sufficient to note that the fact that column 2 of Section 9.1.3 of Annex VIII to the REACH Regulation refers to column 1 of Section 9.1.6 of Annex IX to that regulation cannot have the consequence that column 2 of Section 9.1 of Annex IX to that regulation automatically refers to column 1 of Section 9.1 of Annex IX to that regulation.

198    Column 2 of Section 9.1.3 of Annex VIII to the REACH Regulation states that ‘the long-term aquatic toxicity study on fish (Annex IX, Section 9.1.6) shall be considered if the substance is poorly water soluble’, whereas column 2 of Section 9.1 of Annex IX of the same regulation states that ‘long-term toxicity testing shall be proposed by the registrant if the chemical safety assessment according to Annex I indicates the need to investigate further the effects on aquatic organisms’. In that regard, in so far as the wording of those two columns is different, the fact that column 2 of Section 9.1.3 of Annex VIII to the REACH Regulation refers to column 1 of Section 9.1.6 of Annex IX to that regulation cannot automatically entail the fact that column 2 of Section 9.1 of Annex IX to that regulation may be interpreted as the applicant maintains. Furthermore, the fact that column 2 of Section 9.1 of Annex IX to the REACH Regulation does not refer explicitly to the studies in the corresponding column 1 confirms that the studies envisaged in column 2 are not those in column 1 of Annex IX to that regulation.

199    In addition, it should be noted that the last sentence of the first paragraph of Annex VI to the REACH Regulation provides that ‘the Annexes shall thus be considered as a whole, and in conjunction with the overall requirements of registration, evaluation and the duty of care’. Taking into account that element of context, the ECHA’s argument that a combined reading of Section 9.1.3 of Annex VIII and Section 9.1 of Annex IX to that regulation gives rise, as regards standard information relating to aquatic toxicity tests on fish, to a set of progressive requirements depending on the quantity and properties of the substance registered is not vitiated by an error of law.

200    Accordingly, the fourth complaint cannot succeed and the first part of the fourth plea in law must be rejected in its entirety.

 The second and third parts of the fourth plea in law, alleging infringement of the right to be heard and breach of Article 25 of the REACH Regulation

201    By the second part of its fourth plea in law, the applicant alleges infringement by the ECHA of its right to be heard, guaranteed by Article 41 of the Charter of Fundamental Rights of the European Union. In that regard, the applicant submits that, if the argument based on the possibility of requiring the OECD 234 Study pursuant to column 2 of Section 9.1 of Annex IX to the REACH Regulation or, for that matter, a study additional to those referred to in column 1 of Section 9.1.6 of Annex IX to that regulation had been part of the decision-making process leading to the initial decision or of the action giving rise to the contested decision, it would have been able to exercise its right to be heard, namely its right to express itself and to make known its point of view.

202    By the third part of its fourth plea in law, the applicant claims that the contested decision contains an error of fact, in so far as it is foundthat it does not dispute the correctness of the conclusion of the initial decision that, even if the OECD 234 Study involved using a higher number of vertebrate animals than the OECD 210 Study, that avoided a scenario in which the OECD 234 Study would have to be conducted after the OECD 210 Study had been carried out in order to clarify the question of endocrine disrupting properties.

203    The applicant states that it contested that conclusion and submitted observations in the course of the procedure giving rise to the contested decision. Thus, by ignoring its observations and asking it to conduct the OECD 234 Study, which requires at least 720 vertebrate animals, in particular fish, as part of the compliance check procedure, the ECHA infringed Article 25 of the REACH Regulation.

204    The ECHA disputes the applicant’s argument.

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

205    As indicated in paragraph 176 above, the Board of Appeal considered, in paragraph 191 of the contested decision, that the initial decision explained, first, that the information contained in the applicant’s file demonstrated that the substance at issue could be an endocrine disruptor, secondly, that the OECD 234 Study could clarify whether that was the case, unlike the three studies referred to in column 1 of Section 9.1.6 of Annex IX to the REACH Regulation and, thirdly, that, even though the OECD 234 Study involved a higher number of vertebrate animals than the OECD 210 Study, the use of those animals would have avoided a scenario in which the OECD 234 Study would have had to be carried out to clarify the question of endocrine disrupting properties after the OECD 210 Study had been carried out.

206    The Board of Appeal then noted, in paragraphs 192 to 194 of the contested decision, first, that the applicant had not contested the grounds of the initial decision before it and that it had therefore not examined them and, secondly, that the applicant had not proposed alternative solutions which would have made it possible to verify that the substance at issue did not produce endocrine disrupting properties by means of a less expensive study or one requiring fewer vertebrate animals. Accordingly, the Board of Appeal rejected the applicant’s arguments.

–       The merits of the second and third parts of the fourth plea in law

207    As regards the second part of the fourth plea in law, the transcript of the hearing before the Board of Appeal, reproduced in Annex B.10 to the defence, states that, contrary to the applicant’s assertion, it had the opportunity to put forward its point of view on the question whether it was possible to require the OECD 234 Study to be carried out pursuant to column 2 of Section 9.1 of Annex IX to the REACH Regulation. Accordingly, that plea, alleging a breach of the right to be heard, cannot succeed.

208    As regards the third part of the fourth plea in law, it should be noted that, in the initial decision, the ECHA stated that an OECD 234 Study reduced the number of animals sacrificed and that, although it had acknowledged that that study was more expensive and required a greater number of animals than an OECD 210 Study, it considered that carrying out an OECD 234 Study at that stage of the procedure instead of an OECD 210 Study could avoid the need to carry out additional studies on animals at a later stage.

209    In its action before the Board of Appeal, the applicant submitted that the ECHA was attempting to justify departing from the rules on standard information by referring to animal welfare. In fact, according to the applicant, the ECHA argued that, by requesting that the 234 OECD Study be carried out rather than the OECD 210 Study, animal testing would be reduced.

210    The applicant also claimed that the ECHA’s argument that carrying out the OECD 234 Study ‘would have reduced animal testing’ should be rejected, since, in reality, even if the OECD 234 Study had been carried out, it would also have been necessary to carry out at least one sample of the OECD 210 Study.

211    However, the Board of Appeal cannot be criticised for concluding, in paragraph 192 of the contested decision, that the applicant had not contested the grounds of the initial decision.

212    It should be noted that the reasons given in the initial decision and reproduced in paragraph 191 of the contested decision refer to the desirability of reversing the order in which the OECD 210 Study and the OECD 234 Study were carried out in order to reduce the number of animals used. Thus, the Board of Appeal stated that, if the OECD 234 Study were carried out first, fewer animals would be sacrificed than if the OECD 210 Study were carried out first.

213    In its action before the Board of Appeal, the applicant argued that carrying out the OECD 234 Study would not reduce the number of animals sacrificed, stating that, if that study were carried out, at least a sample of the OECD 210 Study should be carried out. Even if, from a formal point of view, the applicant maintained that carrying out the OECD 234 Study would not be likely to reduce the number of animals sacrificed, in essence, the argument which it put forward in support of that assertion, namely that a sample of the OECD 210 Study should at least be carried out, demonstrates that carrying out the OECD 224 Study would make it possible not to carry out the OECD 210 Study in its entirety. Thus, the applicant’s argument does not call into question the finding, contained in the initial decision, that carrying out the OECD 234 Study would reduce the number of animals sacrificed. Accordingly, it is without committing a manifest error of assessment that the Board of Appeal was able to consider that the applicant had not contested that finding.

214    Accordingly, since the second and third parts of the fourth plea in law cannot succeed either, the fourth plea in law must be rejected in its entirety.

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

215    By its fifth 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.

216    In the first place, the applicant claims that a longer period, namely 46 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 as part of the process leading to the adoption of the initial decision and the action leading to the adoption of the contested decision.

217    Furthermore, according to the applicant, the ECHA is required, in each compliance decision, to specify ‘adequate time limits’ for each individual case. In that regard, the applicant claims that it has 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.

218    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.

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

220    In that regard, the applicant claims that carrying out the OECD 414 Study and then the OECD 443 Study sequentially is a customary way of proceeding in view of the principle of the absence of any other solution and in order to avoid carrying out vertebrate animal testing, which is unnecessary under Article 25 of the REACH Regulation.

221    With regard to the ECHA’s argument that neither the OECD 414 Study nor the OECD 443 Study can allow it to dispense with the other, the applicant claims that, in its particular case, the harmful effects revealed by the screening study on which the ECHA based its request for the OECD 443 Study could, in fact, have been assessed in the context of the OECD 414 Study, as regards both the reduction in the average number of live births and the prolonged gestation period. In those circumstances, the applicant claims 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 the uselessness of carrying out the OECD 443 Study in the present case.

222    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.

223    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 42-month time limit. That could lead to the sacrifice, in addition to the 1 301 vertebrate animals required for the OECD 414 Study, of approximately 1 900 vertebrate animals required for the OECD 443 Study, which, depending on the results of the OECD 414 Study, might not be necessary.

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

225    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 are not 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 contract research organisations (‘CROs’) was linked to the fact that those organisations were precisely to carry out OECD 414 and 443 studies close to 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.

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

227    Finally, the applicant points out that the fact that the ECHA does not in any way justify or substantiate its assertions concerning 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.

228    The ECHA disputes the applicant’s arguments.

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

229    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 210 and 211 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.

230    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 considered, in paragraph 214 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 fifth plea in law

231    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.

232    In that regard, it should be noted that the Board of Appeal concluded, in paragraphs 210 and 211 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.

233    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.

234    In the second place, as regards the fact that there are no scientific reasons justifying the sequential conduct of the OECD 414 and OECD 443 studies, the following should be noted.

235    First, the ECHA argues that, even if the OECD 414 Study and the OECD 443 Study were carried out sequentially, the result of neither of them would make it possible to dispense with the other. The OECD 414 Study aims to provide general information on the effects of prenatal exposure on the pregnant laboratory animal and the developing organism, but would not address the parturition, lactation and postnatal development of the fully exposed F1 generation into adulthood. Consequently, the ECHA maintains that the OECD 414 Study can neither invalidate the results of the OECD 421 Study, which triggered the obligation to conduct an OECD 443 Study under column 1 of Section 8.7.3 of Annex IX to the REACH Regulation, nor serve as a basis for adapting that information requirement.

236    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.

237    The applicant contests the ECHA’s arguments, arguing, in essence, that the performance 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.

238    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, when it is a question of assessing highly complex factual elements of a scientific and technical nature, the review carried out by the EU Courts is limited. As regards 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).

239    Thus, in the light of the explanations provided by the ECHA, the applicant’s arguments and the case-law cited in paragraph 238 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 carrying out the OECD 414 and OECD 443 studies sequentially.

240    With regard to the applicant’s argument that carrying out the 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’.

241    According to the case-law, a registrant has, in general and therefore in particular in the event that the ECHA sends it 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).

242    It is sufficient to note that, according to Article 25 of the REACH Regulation, in the event that the ECHA asks a registrant to complete its registration dossier, the latter is obliged to generate information obtained by means other than animal testing ‘whenever possible’ and to undertake such testing ‘only as a last resort’.

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

244    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 221 of the contested decision, the Board of Appeal set a new time limit of 25 February 2024 for the applicant to carry out the OECD 414 and OECD 443 studies.

245    The fifth plea in law must therefore be rejected, and the action in its entirety must be dismissed.

 Costs

246    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.

247    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 interim proceedings.

248    In accordance with Article 138(3) of the Rules of Procedure, CFE, PETA, EFfCI and Unilever 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, European Federation for Cosmetics Ingredients (EFfCI), Unilever U.K. Central Resources Ltd and Unilever ASCC AG 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

Table of contents

Background to the dispute

Forms of order sought

Law

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

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

The merits of the first plea in law

– The relationship between the REACH Regulation and the Cosmetics Regulation

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

The second plea in law, alleging a manifest error of assessment by the ECHA with regard to the OECD 443 Study and breach of the duty to state reasons

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

The merits of the second plea in law

The third 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

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

The merits of the third plea in law

The fourth plea in law, alleging, in essence, a manifest error of assessment and interpretation of column 2 of section 9.1 of Annex IX to the REACH Regulation by the ECHA in requiring the OECD 234 Study to be carried out, an infringement of the right to be heard and a breach of Article 25 of the REACH Regulation

The first part of the fourth plea in law, alleging that the ECHA committed a manifest error of assessment and interpretation of column 2 of section 9.1 of Annex IX to the REACH Regulation by requiring the OECD 234 Study to be carried out

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

– The merits of the first part of the fourth plea in law

The second and third parts of the fourth plea in law, alleging infringement of the right to be heard and breach of Article 25 of the REACH Regulation

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

– The merits of the second and third parts of the fourth plea in law

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

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

The merits of the fifth plea in law

Costs

*      Language of the case: English.