Court of Justice of the European Communities (including Court of First Instance Decisions)

JUDGMENT OF THE GENERAL COURT (Sixth Chamber)

27 November 2024 (*)

( Environment and protection of human health - Regulation (EC) No 1272/2008 - Classification, labelling and packaging of certain substances and mixtures - Delegated Regulation (EU) 2022/692 - Classification and labelling of the substance N-carboxymethyliminobis (ethylenenitrilo)tetra(acetic acid) and its pentasodium and pentapotassium salts (DTPA) - Criteria for classification of a substance as a reproductive toxicant Category 1B - Intrinsic property to specifically harm reproduction - Non-specific secondary consequence of other toxic effects - Possible influence of maternal toxicity - Appropriateness of the classification - Absence of public consultation on the opinions of ECHA’s Committee for Risk Assessment - Interinstitutional Agreement on Better Law-Making - Absence of impact assessment )

In Case T‑453/22,

BASF SE, established in Ludwigshafen am Rhein (Germany),

Dow Europe GmbH, established in Horgen (Switzerland),

Nouryon Functional Chemicals BV, established in Amsterdam (Netherlands),

represented by J.-P. Montfort and P. Chopova-Leprêtre, lawyers,

applicants,

v

European Commission, represented by M. Farley and R. Lindenthal, acting as Agents,

defendant,

supported by

French Republic, represented by B. Fodda, B. Travard and M. de Lisi, acting as Agents,

and by

European Chemicals Agency (ECHA), represented by A. Deloff-Bialek, A. Hautamäki and M. Heikkilä, acting as Agents,

interveners,

THE GENERAL COURT (Sixth Chamber),

composed of M.J. Costeira (Rapporteur), President, M. Kancheva and P. Zilgalvis, Judges,

Registrar: P. Cullen, Administrator,

having regard to the written part of the procedure,

having regard to the measures of organisation of procedure of 23 November 2023 and the Commission’s reply lodged at the Registry of the General Court on 8 December 2023,

further to the hearing on 2 February 2024,

gives the following

Judgment

1        By their action under Article 263 TFEU, the applicants, BASF SE, Dow Europe GmbH and Nouryon Functional Chemicals BV, seek the annulment of Commission Delegated Regulation (EU) 2022/692 of 16 February 2022 amending, for the purposes of its adaptation to technical and scientific progress, Regulation (EC) No 1272/2008 of the European Parliament and of the Council on classification, labelling and packaging of substances and mixtures (OJ 2022 L 129, p. 1; ‘the contested regulation’), in so far as concerns the harmonised classification and labelling of the substance N-carboxymethyliminobis (ethylenenitrilo)tetra(acetic acid) (‘DTPA-H5’), and its pentasodium (‘DTPA-Na5’) and pentapotassium salts (‘DTPA-K5’) in the hazard class for reproductive toxicity, Category 1B.

 Background to the dispute

2        The applicants are manufacturers, importers or suppliers in the EU market of the substances DTPA-H5, DTPA-Na5 and DTPA-K5 (together, ‘DTPA’).

3        The substances constituting DTPA are related and correspond to N-carboxymethyliminobis (ethylenenitrilo)tetra(acetic acid), that is to say, DTPA-H5, to its pentasodium, that is to say, DTPA-Na5, and to its pentapotassium salts, that is to say, DTPA-K5.

4        DTPA is used as a chelating agent, in particular, in the pulp and paper industries and in the laundry detergents, cleaners, soaps and textiles industries. The function of a chelating agent is to bind metal ions and, by doing so, to form stable, water-soluble metal-chelate complexes. Amongst other things, chelating agents enable water hardness to be controlled, by limiting the concentrations of calcium and magnesium ions, and enable of barium sulphate scale in oil processing plants to be removed.

5        On 21 November 2015, Dow Chemical Company Ldt submitted a proposal to the European Chemicals Agency (ECHA) for harmonised classification and labelling of DTPA-Na5, in particular, for the hazard class for reproductive toxicity, Category 2, with the hazard statement code ‘H361d (Oral)’ (suspected of damaging the unborn child), on the basis, essentially, of the study entitled ‘Study of the Prenatal Toxicity of Trilon C Liquid in Wistar Rats after Oral Administration (Gavage)’ (BASF, 1994) (‘the BASF study’).

6        On 1 April 2016, Akzo Nobel Functional Chemicals BV, now Nouryon Functional Chemicals, submitted two proposals to ECHA for harmonised classification and labelling, concerning DTPA-H5 and DTPA-K5 respectively, proposing the same hazard class, category and statement code as those referred to in paragraph 5 above, on the basis of the BASF study.

7        The three classification proposals referred to in paragraphs 5 and 6 above (‘the proposals for harmonised classification and labelling of DTPA’) were submitted on the basis of Article 37(2) of 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), in the version in force on the date of adoption of the contested regulation, which is to say the version amended by Commission Delegated Regulation (EU) 2021/1962 of 12 August 2021 correcting Annex VI to Regulation No 1272/2008 (OJ 2021 L 400, p. 16).

8        Between February and July 2016, public consultations concerning the proposals for harmonised classification and labelling of DTPA were organised. Several Member States and interested parties submitted observations.

9        On 9 June 2017, on the basis of Article 37(4) of Regulation No 1272/2008, ECHA’s Committee for Risk Assessment (‘the RAC’) adopted, by consensus, three opinions relating to, first, DTPA-H5, second, DTPA-Na5 and, third, DTPA-K5 (together, ‘the first RAC opinions’). In those first opinions, the RAC proposed that each of the three substances be classified as, inter alia, toxic for reproduction, Category 1B, with the hazard statement code ‘H360D’ (may damage the unborn child).

10      In the course of 2018 and 2019, the authors of the proposals for harmonised classification and labelling of DTPA sent, through their lawyers, letters to the European Commission expressing their disagreement with the classification proposal referred to in the previous paragraph and providing, amongst other things, additional information about alleged differences in zinc kinetics and physiology between rats and humans.

11      On 10 January 2020, further to a request from the Commission to that effect, ECHA sent a note to the RAC asking for its first opinions to be reassessed in the light, in particular, of the additional information referred to in the previous paragraph, in so far as the classification of DTPA as toxic for reproduction and the setting of specific concentration limits were concerned.

12      In the course of 2020, a targeted public consultation regarding the ‘developmental toxicity’ of DTPA, in the light, in particular, of the information referred to in paragraph 10 above, was organised. One Member State submitted observations.

13      On 11 June 2020, the RAC adopted, by consensus, an additional opinion on the reproductive toxicity of the substances constituting DTPA, that is to say, DTPA-H5, DTPA-Na5 and DTPA-K5 (‘the additional RAC opinion’). In that opinion, the RAC maintained its proposed classification of each of the three substances as, in particular, toxic for reproduction, Category 1B, with the hazard statement code ‘H360D’ (may damage the unborn child), but accepted the proposal of the authors of the proposals for harmonised classification and labelling of DTPA to introduce a specific concentration limit of 3%.

14      The Commission received, from the applicants amongst others, additional information in response to the additional RAC opinion and, as is apparent from recitals 3 and 4 of the contested regulation, did not find it sufficient to cast doubt on the scientific analysis contained in that opinion.

15      On 16 February 2022, on the basis of the additional RAC opinion, the Commission adopted the contested regulation. By that regulation, the substances constituting DTPA, that is to say, DTPA-K5, DTPA-H5 and DTPA-Na5, were inserted into Annex VI, Part 3, Table 3, to Regulation No 1272/2008, with harmonised classification and labelling in the hazard class for reproductive toxicity, Category 1B (Presumed human reproductive toxicant), with the hazard statement code ‘H360D’ (may damage the unborn child) and a specific concentration limit of 3% (‘the contested classification’). The contested regulation also introduced harmonised classification and labelling for those substances in the hazard classes for acute toxicity, Category 4, and STOT RE, Category 2, and also, in the case of DTPA-K5 and DTPA-H5, in the hazard class for eye irritant, Category 2.

16      Pursuant to the second paragraph of Article 2 of the contested regulation, the amendments to Annex VI to Regulation No 1272/2008, in so far as the contested classification is concerned, are to apply from 23 November 2023 onwards.

17      In addition, the applicants or the undertakings affiliated to them are ‘registrants’ of DTPA, within the meaning of Article 3(7) of 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 Article 2(13) of Regulation No 1272/2008.

18      Moreover, following recommendations from the competent German authority, the applicants, as registrants of DTPA, adopted decisions to self-classify DTPA as a reproductive toxicant, Category 2.

 Forms of order sought

19      The applicants claim that the General Court should:

–        annul the contested regulation in so far as the contested classification is concerned;

–        order the Commission to pay the costs.

20      The Commission, supported by the French Republic and ECHA, contends that the Court should:

–        dismiss the action;

–        order the applicants to pay the costs.

 Law

21      In support of their action, the applicants put forward six pleas in law:

–        the first plea in law alleges breach of the classification criteria according to which the substance must have the intrinsic, specific property to produce an adverse effect on reproduction and that effect must not be regarded as a non-specific secondary consequence of other toxic effects;

–        the second plea alleges a failure to take account of the possible influence of maternal toxicity;

–        the third plea, which is subsidiary to the first two pleas, alleges that the available data did not justify the classification of DTPA in Category 1B;

–        the fourth plea alleges a breach of the Commission’s duty to examine the appropriateness of the contested classification;

–        the fifth plea alleges an absence of public consultations regarding the RAC opinions;

–        the sixth plea concerns the absence of an impact assessment.

 Preliminary considerations on harmonised classification and labelling of substances in the hazard class for reproductive toxicity

22      As a preliminary point, it should be noted that, in accordance with recital 1 and Article 1(1) of Regulation No 1272/2008, the purpose of that regulation is to ensure a high level of protection of human health and the environment as well as the free movement of chemical substances, mixtures and certain specific articles on the EU market. As is apparent from, in particular, recitals 5 to 8, 10 and 27 of that regulation, the objective of that regulation is to determine the intrinsic properties of the substances which must lead to their classification as hazardous products, so that the hazards posed by those substances (and by mixtures containing such substances) can be correctly identified and notified. To that end, in accordance with Article 1(1)(a) thereof, the purpose of that regulation is, inter alia, to ‘[harmonise] the criteria for classification of substances and mixtures, and the rules on labelling and packaging for hazardous substances and mixtures’.

23      In addition, it is apparent from recitals 4 to 8 of Regulation No 1272/2008 that the EU legislature intended to contribute to the global harmonisation of criteria for classification and labelling, not only at the level of the United Nations, but also through the incorporation of the internationally agreed Globally Harmonised System of Classification and Labelling of Chemicals (‘GHS’) criteria into EU law. To that effect, Annex I to that regulation reproduces verbatim almost all of the GHS provisions (judgment of 22 November 2017, Commission v Bilbaína de Alquitranes and Others, C‑691/15 P, EU:C:2017:882, paragraph 42).

24      As regards the classification of hazardous substances and mixtures, it should be recalled that, according to Article 3 of Regulation No 1272/2008, a substance or a mixture fulfilling the criteria relating to physical hazards, health hazards or environmental hazards, as laid down in Annex I thereto, is hazardous and is to be classified in relation to the respective hazard classes provided for in that annex.

25      In that regard, Regulation No 1272/2008 provides, in Title V, for a harmonisation procedure throughout the European Union of the classification and labelling of substances, which concerns substances meeting the criteria set out in Annex I for the hazards listed in Article 36(1) of that regulation, including for the hazard of reproductive toxicity. That regulation also lays down, in particular in Articles 5, 9 and 13, a self-classification obligation imposed on manufacturers, importers and downstream users, which relates to substances and mixtures.

26      The procedure for harmonisation of classification and labelling of substances is triggered, by the competent authority of a Member State or by manufacturers, importers or downstream users of a substance, by the submission of a proposal for harmonised classification and labelling of that substance before ECHA, in accordance with Article 37(1) and (2) of Regulation No 1272/2008. Next, RAC is to ‘adopt an opinion on any proposal submitted … giving the parties concerned the opportunity to comment’, and ECHA is to ‘forward this opinion and any comments to the Commission’, in accordance with Article 37(4). Lastly, where the Commission finds that the harmonisation of the classification and labelling of the substance concerned is appropriate, it adopts a delegated act, in accordance with Article 37(5) and Article 53a of that regulation, in order to amend Annex VI thereto by including in Table 3 of Part 3 of that annex the substance in question together with the relevant classification and labelling elements.

27      Furthermore, it should be recalled that Regulation No 1272/2008 concerns the assessment of hazards of substances and that that assessment must be distinguished from the risk assessment provided for in Regulation No 1907/2006. Hazard assessment constitutes the first stage of the process of risk assessment, which is a more specific concept. Thus, an assessment of the hazards linked to the intrinsic properties of a substance must not be limited in the light of specific circumstances of use, as in the case of a risk assessment, and may be properly carried out regardless of the place where the substance is used (laboratory or elsewhere) or the possible levels of exposure to the substance (see, to that effect, judgment of 21 July 2011, Nickel Institute, C‑14/10, EU:C:2011:503, paragraphs 81 and 82).

28      As regards the hazard of reproductive toxicity, Article 36(1)(d) of Regulation No 1272/2008 provides that, if a substance meets the criteria set out in Annex I to that regulation for the hazard of reproductive toxicity, it will normally be subject to harmonised classification and labelling. Those criteria are defined in Section 3.7 of Part 3 of Annex I to Regulation No 1272/2008.

29      In particular, in the first paragraph of Section 3.7.1.1 and points (a) and (b) of the second paragraph of that section of Annex I to Regulation No 1272/2008, it is provided that ‘reproductive toxicity’ is subdivided into two ‘headings’ of adverse effects, the first relating to adverse effects on sexual function and fertility in adult males and females and the second to adverse effects on development of their offspring.

30      As regards hazard categories, it follows from Section 3.7.2.1.1 and Table 3.7.1(a) of Annex I to Regulation No 1272/2008 that classification for reproductive toxicity is divided into two categories, namely Category 1, which is subdivided into Categories 1A and 1B, and Category 2. In particular, Category 1B corresponds to presumed human reproductive toxicants and Category 2 corresponds to suspected human reproductive toxicants.

 Preliminary observations on the intensity of the Court’s review

31      As regards the intensity of the Court’s review, it should be recalled that, in accordance with settled case-law, if the Commission is to be able to classify a substance pursuant to Regulation No 1272/2008, account being taken of the complex scientific and technical assessments which it must undertake, it must be recognised as enjoying a broad discretion (see judgment of 22 November 2017, Commission v Bilbaína de Alquitranes and Others, C‑691/15 P, EU:C:2017:882, paragraph 34 and the case-law cited).

32      However, the exercise of that discretion is not excluded from review by the Court. It has consistently been held that, in the context of such a review the Courts of the European Union must verify whether the relevant procedural rules have been complied with, whether the facts accepted by the Commission have been accurately stated and whether there has been a manifest error in the appraisal of those facts or a misuse of powers (see judgment of 18 July 2007, Industrias Químicas del Vallés v Commission, C 326/05 P, EU:C:2007:443, paragraph 76 and the case-law cited).

33      In particular, where a party claims that the institution competent in the matter has committed a manifest error of assessment, the Courts of the European Union must verify whether that institution has examined, carefully and impartially, all the relevant facts of the individual case on which that assessment was based. That duty to act diligently is inherent in the principle of sound administration and applies generally to the actions of the EU administration (see judgment of 22 November 2017, Commission v Bilbaína de Alquitranes and Others, C‑691/15 P, EU:C:2017:882, paragraph 35 and the case-law cited).

34      In addition, the limits to review by the Courts of the European Union do not affect their duty to establish whether the evidence relied on is factually accurate, reliable and consistent and also whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it (see, to that effect, judgment of 6 November 2008, Netherlands v Commission, C‑405/07 P, EU:C:2008:613, paragraph 55 and the case-law cited).

35      Furthermore, as regards the evaluation of scientific studies, the Court has already held that the Commission must be allowed a broad discretion with regard to that assessment, as well as the choice of studies which must take precedence over others, irrespective of their chronology. Thus, it is not sufficient for the applicant to rely on the age of a scientific study to call into question its reliability, but it is also necessary for the applicant to provide sufficiently precise and objective evidence to argue that any recent scientific developments would call into question the soundness of the conclusions of such a study (see, to that effect, judgment of 24 October 2018, Deza v Commission, T‑400/17, not published, EU:T:2018:712, paragraph 95).

36      It must be added that, in order to establish that the administration committed a manifest error in assessing complex facts such as to justify the annulment of the contested act, the evidence adduced by the applicant must be sufficient to make the factual assessments used in the act implausible. Subject to that review of plausibility, it is not the Court’s role to substitute its assessment of complex facts for that made by the institution which adopted the act (see judgment of 5 July 2023, TIB Chemicals v Commission, T‑639/20, not published, EU:T:2023:374, paragraph 35 and the case-law cited).

37      It is in the light of those considerations that the applicants’ pleas in law must be examined.

 The first plea in law, alleging breach of the classification criteria according to which the substance must have the intrinsic, specific property to produce an adverse effect on reproduction and that effect must not be regarded as a non-specific secondary consequence of other toxic effects

38      In their first plea in law, the applicants maintain, essentially, that the contested classification is inconsistent with the criteria for classification as a reproductive toxicant, established in Article 36(1) of Regulation No 1272/2008 and in Section 3.7.2.2.1 of Annex I to that regulation, inasmuch as those criteria require that (i) the substance have the ‘intrinsic, specific property to produce an adverse effect on reproduction’ and (ii) that effect should not be regarded as a ‘non-specific secondary consequence of other toxic effects’.

39      The Commission, supported by the French Republic and ECHA, disputes those arguments and contends, moreover, that the applicants’ first plea is not ‘credible’.

 The ‘credibility’ of the first plea

40      The Commission raises the lack of ‘credibility’ of the applicants’ first plea on the ground that they dispute the classification of DTPA in the hazard class for reproductive toxicity, even though they had themselves submitted a classification proposal to that effect.

41      It is apparent from the documents before the Court that Dow Chemical Company was indeed the author of the proposal for harmonised classification and labelling of DTPA-Na5 for, inter alia, the hazard class for reproductive toxicity, Category 2 (see paragraph 5 above). It is also apparent from the file that Akzo Nobel Functional Chemicals, now Nouryon Functional Chemicals, was the author of two proposals for harmonised classification and labelling concerning, respectively, DTPA-H5 and DTPA-K5 for the same hazard classes, categories and statement codes (see paragraph 6 above). In addition, as noted in paragraph 18 above, the applicants, following recommendations from the competent German authority, adopted decisions to self-classify DTPA as a reproductive toxicant, Category 2.

42      However, those proposals for classification and those decisions to self-classify DTPA cannot have an impact on the applicants’ right to bring an action before the Court seeking annulment of the harmonised classification and labelling of the substances which are the subject of those proposals and of those self-classifications. Such circumstances are not capable of altering the applicants’ standing and interest in bringing proceedings for annulment of the contested classification which underlie the present action.

43      Accordingly, there is nothing to prevent the examination of the applicants’ first plea.

44      The Court considers it appropriate to examine, first of all, the applicants’ arguments relating to the interpretation of the criteria for classification as a reproductive toxicant, next, their arguments relating to an incorrect application of the criteria for classification as a reproductive toxicant and, lastly, their arguments relating to the ‘inconsistent’ and ‘confusing’ positions taken by the RAC, ECHA and the Commission on the interpretation of the classification criteria.

 The applicants’ arguments relating to the interpretation of the criteria for classification as a reproductive toxicant

45      The applicants submit that the criteria set out in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 must be interpreted as requiring a direct relation between an ‘intrinsic, specific property’ of a substance and primary, principal or direct adverse effects on reproduction. Indeed, in their submission, the word ‘intrinsic’ must be interpreted in its literal sense, as referring to the ‘properties which a substance has in and of itself’, as the Court recently held in paragraphs 138 to 142 of its judgment of 23 November 2022, CWS Powder Coatings and Others v Commission (T‑279/20, T‑283/20 and T‑288/20, under appeal, EU:T:2022:725). Moreover, the word ‘specific’ means ‘connected with one particular thing only’, which is to say that it designates a distinctive, inherent characteristic of a given substance.

46      As a preliminary point, it is appropriate to limit the examination of the criteria for classification as a reproductive toxicant solely to the heading at issue in the present case, namely that of adverse effects on development of the offspring, which is referred to in the first paragraph of Section 3.7.1.1 and point (b) of the second paragraph of that section of Annex I to Regulation No 1272/2008 (see paragraph 29 above).

47      The criteria for the harmonised classification and labelling of a substance for the hazard of reproductive toxicity, including under the heading of adverse effects on development of the offspring, are provided for in Section 3.7.2.2 of Annex I to Regulation No 1272/2008, under the title ‘Basis of classification’, read in conjunction with Article 36 of that regulation. It follows, in particular, from Section 3.7.2.2.1 of Annex I to that regulation that that classification requires two cumulative classification criteria to be satisfied.

48      As regards the first classification criterion, the wording of Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 is not exactly the same in the various language versions of that regulation. The French-language version of Section 3.7.2.2.1 of Annex I to that regulation provides that classification as a reproductive toxicant applies to substances which ‘possèdent la propriété intrinsèque de nuire spécifiquement à la reproduction’ (possess the intrinsic property of specifically harming reproduction) and does not apply to substances which ‘ne produisent cet effet que comme conséquence secondaire et non spécifique d’autres effets toxiques’ (produce this effect only as a secondary and non-specific consequence of other toxic effects). On the other hand, the English-language version of that section is worded as follows: ‘classification as a reproductive toxicant is intended to be used for substances which have an intrinsic, specific property to produce an adverse effect on reproduction and substances shall not be so classified if such an effect is produced solely as a non-specific secondary consequence of other toxic effects.’

49      Thus, as regards the first classification criterion, the French-language version of Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 appears to require that the substance possess an ‘intrinsic’ property and that that property have the effect of ‘specifically’ harming reproduction. The English-language version appears to require that the substance have an ‘intrinsic, specific’ property which has the effect of harming reproduction.

50      In a context in which various language versions of Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 may give rise to varying interpretations, it should be recalled that, according to settled case-law, the need for a uniform interpretation of EU law makes it impossible for the text of a provision to be considered, in case of doubt, in isolation; on the contrary, it requires that it be interpreted also in the light of the versions existing in the other official languages and by reference to the purpose and general scheme of the rules of which that provision forms part (see judgments of 17 September 2009, Vorarlberger Gebietskrankenkasse, C‑347/08, EU:C:2009:561, paragraph 26 and the case-law cited, and of 15 March 2018, Deichmann, C‑256/16, EU:C:2018:187, paragraph 49 and the case-law cited).

51      The applicants submit that the first classification criterion must be interpreted as requiring the substance to have an ‘intrinsic, specific property’, that is to say, a property which ‘the substance has in and of itself’ and which corresponds to a ‘distinctive, inherent characteristic’ of that substance.

52      As regards the concept of ‘intrinsic … property’ referred to in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008, it should be noted that the Court has already examined a similar concept in relation to the criterion for classification of a substance as carcinogenic, laid down in Section 3.6.2.2.1 of that annex, which requires a substance to have an intrinsic property to cause cancer. In that context, the Court has held that, although the concept of ‘intrinsic property’ did not appear in Regulation No 1272/2008, it had to be interpreted in its literal sense as referring to the properties which a substance has in and of itself (see, to that effect, judgment of 23 November 2022, CWS Powder Coatings and Others v Commission, T‑279/20, T‑283/20 and T‑288/20, under appeal, EU:T:2022:725, paragraph 138).

53      That interpretation of the concept of ‘intrinsic property’ is consistent with the objectives and purpose of harmonised classification and labelling under Regulation No 1272/2008, from which it follows that only the properties specific to a substance must lead to its classification as a hazardous product, so that the hazard associated with such properties can be correctly identified and notified. That interpretation is also consistent with the GHS criteria, which are incorporated into EU law (see paragraph 23 above), of which Section 1.1.1.6(b) and footnote 1 and Section 1.1.3.1.1 make, inter alia, a distinction between the intrinsic properties of a substance, to which the hazard classification process relates, and other properties which are not specific to the substance (see, to that effect, judgment of 23 November 2022, CWS Powder Coatings and Others v Commission, T‑279/20, T‑283/20 and T‑288/20, under appeal, EU:T:2022:725, paragraphs 139 and 140).

54      Moreover, that interpretation is consistent with the fact that the harmonised classification and labelling under Regulation No 1272/2008 relate to hazard assessment, and not risk assessment, which is provided for by Regulation No 1907/2006. As is apparent from the case-law referred to in paragraph 27 above, an assessment of the hazards linked to the intrinsic properties of a substance must not be limited in the light of specific circumstances of use, as in the case of a risk assessment, and may be properly carried out regardless of the place where the substance is used or the possible levels of exposure to the substance (see, to that effect, judgment of 23 November 2022, CWS Powder Coatings and Others v Commission, T‑279/20, T‑283/20 and T‑288/20, under appeal, EU:T:2022:725, paragraph 141).

55      The case-law cited in paragraph 27 above can be applied in the present case, as regards the concept of ‘intrinsic … property’ referred to in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008, for the purposes of the classification for the hazard of reproductive toxicity.

56      However, the applicants are wrong to claim that Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 requires the substance to have an ‘intrinsic, specific property’. Even if the wording of the English-language version of Section 3.7.2.2.1 may be interpreted as requiring the substance to have an ‘intrinsic [and] specific property’, such an interpretation is contradicted by the French-language version of the same section, from which it is clearly apparent that the adverb ‘spécifiquement’ (specifically) is not linked to the ‘substance’, but to ‘reproduction’, that is to say, that the term ‘specific’ is linked to the type of health hazard referred to therein and, therefore, to the effects caused by the intrinsic property of the substance.

57      Furthermore, the interpretation which results from the wording of the French-language version of Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 is consistent with the general scheme and purpose of that regulation and, in particular, with the harmonised classification and labelling in the hazard class for reproductive toxicity. In that regard, it should be recalled that Regulation No 1272/2008 concerns the assessment of the hazards linked not to the ‘intrinsic [and] specific properties’, but to the ‘intrinsic properties’ of a substance, in accordance with the case-law cited in paragraph 27 above.

58      In addition, as is apparent, in particular, from Sections 3.7.1.1 and 3.7.2.3.4 of Annex I to Regulation No 1272/2008, the harmonised classification and labelling in the hazard class for reproductive toxicity is intended to identify toxic effects ‘touchant spécifiquement la reproduction’ (specifically affecting reproduction) (‘specific reproductive toxicity’, in the English-language version of Section 3.7.2.3.4 of that annex). Moreover, it should be noted that the second classification criterion laid down in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 requires that the adverse effect on reproduction should not be considered to be a ‘non-specific’ secondary consequence of other toxic effects (see paragraph 49 above). It follows that the term ‘specific’ is always linked to toxic effects.

59      Therefore, in accordance with the case-law cited in paragraph 50 above, the first classification criterion referred to in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 must be interpreted as requiring that the substance have an intrinsic property and that that property have the effect of specifically harming reproduction and, so far as is relevant in the present case, specifically harming development of the offspring.

60      Thus, contrary to what the applicants claim, the first classification criterion referred to in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 cannot be interpreted as requiring the substance to have a ‘specific [and] intrinsic property’.

61      Furthermore, the applicants claim that Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 contrasts the ‘intrinsic [and] specific property’ with the other properties of the substance that produce effects as ‘a non-specific secondary consequence’. That second criterion requires the ‘intrinsic, specific property’ of the substance to have ‘primary effects on reproduction’, as opposed to secondary or indirect consequences resulting from other mechanisms, such as excessive exposure or maternal toxicity.

62      In that regard, it should be noted that it follows from Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 that the second classification criterion requires that the effect of specifically harming reproduction, or even of specifically harming development of the offspring, should be considered not to be a ‘non-specific secondary consequence of other toxic effects’.

63      Moreover, it follows from the criteria relating to the hazard categories for reproductive toxicants set out in Table 3.7.1(a) of Annex I to Regulation No 1272/2008 that, both for classification in Category 1B and for classification in Category 2, the data are to provide clear evidence either that there is an adverse effect on development in the absence of other toxic effects or, if occurring together with other toxic effects, that that adverse effect on development is not considered to be a secondary non-specific consequence of those other toxic effects.

64      Moreover, as regards the concept of adverse effects on development of the offspring which are not considered to be a secondary non-specific consequence of other toxic effects, it follows from Section 3.7.1.4 of Annex I to Regulation No 1272/2008 that adverse effects on development correspond to ‘any effect which interferes with normal development of the conceptus, either before or after birth, and resulting from exposure of either parent prior to conception, or exposure of the developing offspring during prenatal development, or postnatally, to the time of sexual maturation’ and that ‘developmental toxicity essentially means adverse effects induced during pregnancy, or as a result of parental exposure’.

65      Furthermore, as regards maternal toxicity, Section 3.7.2.3.5 of Annex I to Regulation No 1272/2008 provides that, ‘generally, the presence of maternal toxicity shall not be used to negate findings of embryo/foetal effects, unless it can be clearly demonstrated that the effects are secondary non-specific effects.’ In addition, Section 3.7.2.4.3 of that annex provides that ‘classification shall not automatically be discounted for substances that produce developmental toxicity only in association with maternal toxicity, even if a specific [maternally mediated] mechanism has been demonstrated’.

66      It follows from the abovementioned provisions that the hazard of reproductive toxicity, under the heading of adverse effects on development of the offspring, covers all adverse effects on development of the offspring which occur after exposure to a substance of either parent prior to conception or of the developing offspring during prenatal development, or postnatally (see paragraph 64 above). In addition, the presence of maternal toxicity cannot be used to negate findings of embryo/foetal effects, unless it can be clearly demonstrated that the effects are secondary and non-specific (see the previous paragraph).

67      In that context, the second criterion referred to in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 must be interpreted as not precluding adverse effects on development of the offspring, inter alia, when they occur after exposure to a substance of either parent prior to conception or when they are observed in parallel with maternal toxicity, unless it can be clearly demonstrated that those adverse effects are a non-specific secondary consequence of other toxic effects.

68      Thus, contrary to what the applicants claim, the second criterion referred to in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 cannot be interpreted as limiting the effects relevant for toxicity on development of the offspring solely to ‘primary effects on reproduction’.

69      In the light of the foregoing, the applicants’ arguments relating to the interpretation of the criteria for classification as a reproductive toxicant under the heading of adverse effects on development of the offspring must be rejected.

 The applicants’ arguments relating to an incorrect application of the criteria for classification as a reproductive toxicant

70      The applicants argue that DTPA does not have, ‘in and of itself’, the capacity to cause reproductive toxicity hazards ‘on its own’. Indeed, in their submission, there is no direct relation between any ‘intrinsic, specific property’ of DTPA and the adverse effects on the basis of which the contested classification was justified. Admittedly, it is true that DTPA has the intrinsic, specific property to bind itself to metal ions, and zinc in particular, such that it could, under extreme exposure conditions, induce zinc deficiency. Nevertheless, DTPA is incapable of producing primary effects on reproduction, given that it is not readily bioavailable and given that it cannot pass through the placenta and reach the foetus.

71      Further, the applicants maintain that the effects on the development of the foetus, which were observed in the BASF study and on which the contested classification was based, are secondary, non-specific effects not directly related to DTPA. DTPA is merely a trigger for zinc deficiency and it is that deficiency, and not DTPA, that could lead to subsequent impacts on the foetus. Similar non-specific secondary effects could be produced by any other chelating agent or by any external source of zinc deficiency, such as an unbalanced diet. Correction of zinc deficiency with a zinc supplement would be sufficient to prevent any developmental toxicity in the offspring. Therefore, the observed effects depend exclusively on the amount of zinc in the body or expelled from it, and not on the amount of DTPA to which the pregnant mother has been exposed. In other words, other circumstances, and not DTPA, led to the effects in question, which are all related to zinc deficiency.

72      In the present case, as is apparent from the first RAC opinions, the RAC proposed the harmonised classification and labelling of DTPA as toxic for reproduction Category 1B, on the basis, in essence, of the BASF study, which it considered to be the key developmental toxicity study in rats. On the basis of that study and other support studies identified in the opinions, the RAC concluded that there were sufficient data to support the hypothesis that the teratogenicity (that is to say, the production of anomalies or foetal malformations) resulting from the dosing of rats with DTPA was the result of an induced deficiency of zinc which subsequently had an impact on the foetus.

73      As regards the mode of action by which DTPA produces toxicity in the body, the first RAC opinions, first of all, state that that mode of action, common to other chelates, consisted of removing or adding ions, thereby altering the homeostasis of metal ions in the body, that is to say, the ionic balance of the body. Next, the first RAC opinions state that DTPA was a chelate binding to bivalent metals and, principally, to zinc and that that binding effect limited the absorption of zinc into the body of rats, thus causing zinc deficiency. That zinc deficiency triggered mechanisms in the maternal body which sought to reserve all the available zinc for that body and to restrict the supply of zinc to the foetus. Finally, zinc deficiency in the foetus caused a number of adverse effects, in particular malformations and skeletal variations in the foetus.

74      In addition, in its first opinions, the RAC considered that the mode of action described in paragraph 73 above was plausible and potentially relevant for humans, in that it was based on the chelation of zinc by DTPA, which notably reduced the concentrations of zinc available to the foetus.

75      Moreover, in its first opinions, the RAC considered that there was clear evidence of an adverse effect on development that was not a non-specific secondary consequence of other toxic effects.

76      Furthermore, it is apparent from its additional opinion that the RAC maintained the proposal for harmonised classification and labelling of DTPA as toxic for reproduction Category 1B, but that it accepted the proposal of the authors of the proposals for harmonised classification and labelling of DTPA to introduce a specific concentration limit of 3% (see paragraph 13 above).

77      On the one hand, the applicants submit that the classification criteria are not satisfied, given that DTPA does not ‘in and of itself’ have the capacity to produce ‘primary’ adverse effects on offspring, since it is unable to reach the foetus. Those arguments cannot succeed.

78      It should be noted that, for the reasons mentioned in paragraphs 47 to 69 above, the applicants’ arguments are based on a misinterpretation of the criteria laid down in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008. It in no way follows from those criteria that classification as a reproductive toxicant, under the heading of adverse effects on development of the offspring, is limited solely to substances capable of reaching the foetus. By contrast, it follows that that classification requires that the substance have an intrinsic property that produces an effect specifically harming reproduction, such an effect being capable of occurring after exposure to the substance of either parent prior to conception or of the developing offspring during prenatal development, or postnatally (see paragraph 66 above).

79      In addition, the contested classification is based on the mode of action of the toxicity of DTPA found by the RAC in its first opinions, on the basis, inter alia, of the BASF study. According to that mode of action, first of all, DTPA binds itself to zinc, next, that binding effect of DTPA limits the absorption of zinc into the body, thereby causing zinc deficiency which triggers mechanisms in the maternal body which seek to reserve all the available zinc for that body and to restrict the supply of zinc to the foetus and, lastly, the insufficient zinc supply to the foetus causes a number of adverse effects on the foetus (see paragraphs 72 to 74 above).

80      Thus, it is apparent from the first RAC opinions that toxicity on development of the offspring covered by the contested classification is the result of zinc deficiency which, in turn, is a consequence of the binding property of DTPA. The mode of action of toxicity found in the RAC opinions therefore displays a causal relationship between that intrinsic property of DTPA and the adverse effects on the foetus, which are identified in the BASF study and upheld in the RAC opinions. Therefore, as the Commission submits, it is the binding property of DTPA which triggers a chain of events which has an adverse effect on development of the offspring.

81      In the light of the foregoing, it is necessary to reject the applicants’ arguments that the contested classification breaches the classification criteria because DTPA does not have, in and of itself, the capacity to cause reproductive toxicity hazards on its own.

82      On the other hand, the applicants submit that the adverse effects on the foetus found in the RAC opinions are caused not by DTPA, but by zinc deficiency, in respect of which DTPA is only the trigger. According to the applicants, those effects are not directly linked to DTPA, but are non-specific secondary effects, within the meaning of Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008, which could be produced by any other chelating agent or by any external source of zinc deficiency, such as an unbalanced diet.

83      In the present case, it is apparent from the first RAC opinions that the RAC considered that there was clear evidence of an adverse effect on development that was not a non-specific secondary consequence of other toxic effects (see paragraph 75 above).

84      The applicants’ arguments do not call that finding into question. That line of argument is based on the incorrect premiss that ‘non-specific secondary’ effects, within the meaning of Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008, correspond to all effects that are not direct effects, caused by substances capable of reaching the foetus. For the reasons set out in paragraphs 67 and 68 above, such a requirement does not arise from the criteria laid down in Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008. Moreover, as has been noted in paragraph 80 above, it is the binding property of DTPA that is the cause of the adverse effects on the foetus, which are established in the first RAC opinions.

85      Furthermore, in so far as the applicants’ arguments seek to call into question the RAC’s assessments on the possible influence of maternal toxicity, they must be examined below, in the context of the second plea.

86      In addition, the applicants’ argument that zinc deficiency may also be caused by other sources, such as other chelating agents or an unbalanced diet, cannot call into question the RAC’s assessment, which relates solely to the effects of DTPA, and not to the effects of other sources of zinc deficiency.

87      In that regard, it should be noted that the mere fact that a substance may have an intrinsic hazardous property which is common to other substances cannot call into question the validity of the harmonised classification and labelling of that substance. The harmonised classification and labelling procedure, pursuant to Title V of Regulation No 1272/2008, relates only to the substance which, first of all, forms the subject matter of the classification proposal submitted to the Commission, on the initiative of the persons identified in Article 37 of Regulation No 1272/2008; next, forms the subject matter of the RAC opinion on the proposal submitted; and, lastly, forms the subject matter of a delegated act adopted by the Commission, where the latter considers that harmonisation is appropriate (see paragraph 26 above). In the present case, the contested regulation covers DTPA, in accordance with the proposals for classification, the classification procedure and the RAC opinions which gave rise to it (see paragraphs 5 to 15 above), so that the legality of that regulation, so far as the contested classification is concerned, cannot be affected by the possible existence of other sources of zinc deficiency.

88      In the light of all of the foregoing, the applicants’ arguments relating to an incorrect application of the criteria for classification as a reproductive toxicant must be rejected. Consequently, it has not been demonstrated, in the present case, that the contested classification was adopted in breach of the classification criteria referred to in Article 36(1) of Regulation No 1272/2008 and Section 3.7.2.2.1 of Annex I to that regulation, in that they require that the substance have the intrinsic property of specifically harming reproduction and that that effect not be considered to be a non-specific secondary consequence of other toxic effects.

 The applicants’ arguments relating to the ‘inconsistent’ and ‘confusing’ positions taken by the RAC, ECHA and the Commission on the interpretation on the classification criteria

89      The applicants submit that the contested classification is the result of the ‘inconsistent’ and ‘confusing’ positions taken by the RAC, ECHA and the Commission on the interpretation of the criteria for classification as a reproductive toxicant. In particular, ECHA’s interpretation of the concept of ‘secondary, non-specific effects’ conflicts with the wording of Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 and breaches the principle of legal certainty.

90      In so far as that line of argument of the applicants seeks to demonstrate that DTPA does not meet the criteria for classification as a reproductive toxicant, it is merely a repetition of the arguments already examined above and must therefore be rejected, as is apparent from paragraph 88 above.

91      Moreover, in so far as the applicants’ line of argument seeks to establish the existence of ‘inconsistent’ and ‘confusing’ interpretations by the RAC, ECHA and the Commission, it is ineffective. In particular, the applicants submit that ECHA adopted positions which contradict its own guidance concerning the concept of an ‘intrinsic, specific property’ and the concept of ‘non-specific secondary consequences’. However, the applicants do not explain to what extent the contested classification is based on such allegedly contradictory positions of ECHA. By contrast, it is apparent from the contested regulation that the contested classification is based on the RAC opinions. Similarly, the contradictions between the RAC opinions and the positions taken by ECHA relied on by the applicants, even if they were established, cannot in themselves affect the legality of that classification. The applicants do not put forward any evidence to support the conclusion that the alleged contradictory or erroneous interpretations by ECHA, expressed during the procedure which led to the adoption of the contested regulation, formed the basis of the contested classification and could therefore have had an impact on the legality of that classification. Those arguments are therefore ineffective.

92      Similarly, it is necessary to reject the applicants’ argument that ECHA relied on a concept of ‘non-specific secondary effects’ which is contrary to the wording of Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008. The applicants do not put forward any evidence to support the conclusion that a possible erroneous position taken by ECHA might have formed the basis for the contested classification and might therefore have an impact on the legality of that classification.

93      Consequently, and in any event, no breach of the principle of legal certainty has been demonstrated in the present case, and the applicants have not put forward any specific argument in that regard.

94      In the light of all of the foregoing, the applicants’ arguments relating to the ‘inconsistent’ and ‘confusing’ positions taken by the RAC, ECHA and the Commission must be rejected and, consequently, the first plea must also be rejected as in part unfounded and in part ineffective.

 The second plea in law, alleging a failure to take account of the possible influence of maternal toxicity

95      The applicants argue that the contested classification is based on an incorrect application and interpretation of Article 36(1) of Regulation No 1272/2008 and of Section 3.7.2.2.2 of Annex I to that regulation in that the RAC opinions wrongly limited the assessment of the possible influence of maternal toxicity in the BASF study to the mere quantification of the severe and visible effects in the mother. Regulation No 1272/2008 does not establish any causal link between the severity of the maternal toxicity and the severity of the visible clinical symptoms in the mother herself. In so doing, the RAC failed to take account of the fact that the zinc deficiency induced by DTPA caused a phenomenon of disruption of the maternal homeostasis, which should have been identified as maternal toxicity triggering secondary effects within the meaning of Sections 3.7.2.3.5 and 3.7.2.4.1 of Annex I to the regulation, even if the consequent symptoms were observed solely or principally in the foetus. According to the applicants, given that zinc depletion during pregnancy may not result in any visible clinical effects in the mother, it was necessary to analyse the levels of zinc available to the mother rat and her foetus and then compare the zinc needs during pregnancy at each specific moment in the foetus’ development. No such evaluation took place in the present case. If the influence of the disruption of maternal homeostasis had been properly taken into account, the necessary conclusion would have been that DTPA had induced zinc deficiency that, during pregnancy, could have triggered non-specific developmental effects on foetuses. That could not have supported the classification of DTPA as toxic for reproduction.

96      The Commission, supported by the French Republic and ECHA, disputes those arguments.

97      As a preliminary point, it should be noted that Section 3.7.2.2.2 of Annex I to Regulation No 1272/2008 provides that, in the evaluation of toxic effects on the developing offspring, it is important to consider the possible influence of maternal toxicity.

98      As regards the presence of maternal toxicity, Section 3.7.2.3.5 of Annex I to Regulation No 1272/2008 provides, as part of the weight of evidence evaluation, as follows:

‘It is important to try to determine whether developmental toxicity is due to a specific maternally mediated mechanism or to a non-specific secondary mechanism, like maternal stress and the disruption of homeostasis. Generally, the presence of maternal toxicity shall not be used to negate findings of embryo/foetal effects, unless it can be clearly demonstrated that the effects are secondary non-specific effects. This is especially the case when the effects in the offspring are significant, e.g. irreversible effects such as structural malformations. In some situations it can be assumed that reproductive toxicity is due to a secondary consequence of maternal toxicity and discount the effects, if the substance is so toxic that dams fail to thrive and there is severe inanition, they are incapable of nursing pups; or they are prostrate or dying.’

99      As regards the influence of maternal toxicity on the development of the offspring, Section 3.7.2.4.2 of Annex I to Regulation No 1272/2008 provides as follows:

‘Based on pragmatic observation, maternal toxicity may, depending on severity, influence development via non-specific secondary mechanisms, producing effects such as depressed foetal weight, retarded ossification, and possibly resorptions and certain malformations in some strains of certain species. However, the limited number of studies which have investigated the relationship between developmental effects and general maternal toxicity have failed to demonstrate a consistent, reproducible relationship across species. Developmental effects which occur even in the presence of maternal toxicity are considered to be evidence of developmental toxicity, unless it can be unequivocally demonstrated on a case-by-case basis that the developmental effects are secondary to maternal toxicity. Moreover, classification shall be considered where there is a significant toxic effect in the offspring, e.g. irreversible effects such as structural malformations, embryo/foetal lethality, significant post-natal functional deficiencies.’

100    As regards an association between developmental toxicity in the offspring and maternal toxicity, Section 3.7.2.4.3 of Annex I to Regulation No 1272/2008 provides as follows:

‘Classification shall not automatically be discounted for substances that produce developmental toxicity only in association with maternal toxicity, even if a specific [maternally mediated] mechanism has been demonstrated. In such a case, classification in Category 2 may be considered more appropriate than Category 1. However, when a substance is so toxic that maternal death or severe inanition results, or the dams are prostrate and incapable of nursing the pups, it is reasonable to assume that developmental toxicity is produced solely as a secondary consequence of maternal toxicity and discount the developmental effects. Classification is not necessarily the outcome in the case of minor developmental changes, when there is only a small reduction in foetal/pup body weight or retardation of ossification when seen in association with maternal toxicity.’

101    In the present case, it is apparent from the file and from the Commission’s response to the measures of organisation of procedure of 23 November 2023 that, in the first RAC opinions, the RAC described the adverse effects observed in the BASF study following the administration of two different doses of DTPA, namely 1 000 milligrams per kilogram (mg/kg) and 400 mg/kg respectively. First, it found that, as regards the highest dose, although maternal toxicity had been observed in pregnant females, the effects on mothers were not sufficiently severe to explain the severe adverse effects observed in the foetus. Second, as regards the lower dose, no maternal toxicity was observed and the adverse effects in the foetus were less severe. It was on the basis of that combination of effects, namely (i) severe effects in the foetus, observed in the presence of low maternal toxicity following the administration of the higher dose and (ii) less severe effects in the foetus, observed in the absence of maternal toxicity following the administration of the lower dose, that the RAC considered that there was clear evidence of an adverse developmental effect that was not a non-specific secondary consequence of other toxic effects (see paragraph 75 above).

102    The position adopted by the RAC in its first opinions was summarised in the minutes of the 41st meeting of the RAC, which took place between 29 May and 9 June 2017. It follows from those minutes that the RAC agreed to classify DTPA in Category 1B on the basis of serious malformations and other developmental effects (including a delay) also observed at lower doses, higher than historical controls and without severe maternal toxicity in the BASF study, supported by two additional studies in rats and mice showing constant developmental toxicity. According to the RAC, the proposed mode of action (namely developmental toxicity, induced by a maternal zinc deficiency) was deemed plausible and relevant for humans. Also according to the RAC, the understanding of the mode of action supports the fact that developmental toxicity is not a non-specific consequence of a maternal effect.

103    The applicants’ arguments must be examined in the light of those considerations.

104    In the first place, it should be noted that it is apparent from the first RAC opinions that the RAC took into account the possible influence of maternal toxicity, but concluded that (i) the effects of DTPA on the foetus observed during the BASF study were severe and that (ii) maternal toxicity was either not severe or was non-existent (see paragraphs 101 and 102 above).

105    In that regard, it should be noted that it follows from Section 3.7.2.3.5 of Annex I to Regulation No 1272/2008 that, especially when the effects in the offspring are significant, the presence of maternal toxicity is not to be used to negate findings of embryo/foetal effects, unless it can be clearly demonstrated that the effects are secondary and non-specific (see paragraph 98 above). Similarly, it follows from Section 3.7.2.4.3 of that annex that classification is not to be automatically discounted for substances that produce developmental toxicity in the offspring only in association with maternal toxicity, even if a specific maternally mediated mechanism has been demonstrated (see paragraph 100 above).

106    It follows that, in the present case, the mere fact that there are maternal toxicity effects at the same time as developmental effects does not establish that those latter effects are a secondary non-specific consequence of maternal toxicity (see, to that effect and by analogy, judgment of 23 February 2022, Chemours Netherlands v ECHA, T‑636/19, not published, EU:T:2022:86, paragraph 50).

107    In the second place, the applicants’ argument that the RAC wrongly limited the assessment of the possible influence of maternal toxicity to the severity of the visible clinical symptoms in the mother cannot succeed, given that it follows, in particular, from Sections 3.7.2.3.5, 3.7.2.4.2 and 3.7.2.4.3 of Annex I to Regulation No 1272/2008 that the severity of maternal toxicity is relevant for the purposes of determining the possible influence of that toxicity on development of the offspring. It follows from Section 3.7.2.4.2 of that annex that the assessment of the influence of maternal toxicity on the development of the offspring via non-specific secondary mechanisms must take into account the severity of that toxicity (see paragraph 99 above). Moreover, it follows from Sections 3.7.2.3.5 and 3.7.2.4.3 of that annex that reproductive toxicity may be considered a secondary consequence of maternal toxicity in cases where the substance is so toxic that maternal death results or that dams fail to thrive and there is severe inanition (see paragraphs 98 and 100 above).

108    In the light of the foregoing, it cannot be held that the RAC committed a manifest error of assessment in so far as it took into account the severity of maternal toxicity when assessing the possible influence of that toxicity on development of the offspring and as it concluded that the maternal toxicity observed during the BASF study was either not severe or was non-existent (see paragraphs 101, 102 and 104 above).

109    In the third place, while it does indeed follow from Section 3.7.2.3.5 of Annex I to Regulation No 1272/2008 that disruption of homeostasis is regarded as a non-specific secondary mechanism which may cause developmental toxicity in the offspring (see paragraph 98 above), the fact remains that the disruption of homeostasis alleged by the applicants has not been demonstrated in the present case.

110    In that regard, it is apparent from the first RAC opinions that the mode of action by which the DTPA produces toxicity in the body consists of removing or adding ions, thereby altering the homeostasis of metal ions in the body, that is to say, the ionic balance of the body (see paragraph 73 above). Moreover, it is apparent from those opinions that the zinc deficiency induced by DTPA triggers mechanisms in the maternal body which seek to reserve all of the available zinc for that body and to restrict the supply of zinc to the foetus (see paragraph 73 above). In particular, the RAC noted that it appeared that, during periods of zinc deficiency, the maternal liver sequestered zinc via the induction of proteins binding zinc, which restricted the supply of zinc to the foetus. As the Commission points out, zinc deficiency induced by DTPA and the subsequent mechanism of reserving available zinc for the mother’s body to the detriment of the foetus cannot correspond to a situation in which the mother’s homeostasis is disrupted, that is to say, to a disruption of the regulatory mechanism that enables the ionic balance of the mother’s body to be maintained.

111    The existence of a disruption of homeostasis is also not apparent from the file, the applicants’ arguments being, moreover, based on the unsubstantiated hypothesis of a disruption or severe maternal toxicity without any visible clinical effects in the mother, but with an impact on the bone formation of the foetus.

112    In the fourth place and consequently, it is necessary to reject the applicants’ argument that it is necessary to analyse the levels of zinc available for female rats and their foetuses at each moment of the foetus’ development, given that that argument is based on the unsubstantiated hypothesis of severe maternal toxicity without any visible clinical effects in the mother.

113    In the light of all of the foregoing, the second plea must be rejected as unfounded.

 The third plea in law, alleging that the available data did not justify the classification of DTPA in Category 1B

114    By the third plea in law, which is subsidiary to the first and second pleas, the applicants argue that the contested classification was adopted in breach of Table 3.7.1(a) of Annex I to Regulation No 1272/2008, in that the available scientific data did not justify the classification of DTPA in Category 1B of reproductive toxicants. The Commission made a manifest error of assessment and exercised its discretion inappropriately, given that, if it had taken all the relevant information into account, it would have had to classify DTPA as a reproductive toxicant of Category 2, at most.

115    In the first place, the applicants submit that the contested classification is not based on scientific data that establish a ‘strong presumption’ that the substance has the capacity to interfere with reproduction in humans, as is required by Table 3.7.1(a) of Annex I to Regulation No 1272/2008. Even if it had been confirmed that DTPA’s mode of action could in theory be relevant to humans, that would not have been sufficient to satisfy that criterion, since the Commission did not carry out an analysis of the ‘interspecies’ differences between humans and rats in terms of the need for zinc or of the amount of zinc available during the development of the foetus. Those differences should have led the RAC to conclude that there were serious doubts about the relevance to humans of the reproductive toxicity effects observed in the BASF study. The additional RAC opinion offered no answer to that and, moreover, noted some uncertainties as to the extent of the interspecies differences. In addition, the level of exposure to DTPA that would have been necessary to reach a comparable level of zinc depletion in humans to that observed in rats was unrealistic, and indeed impossible, given the very high doses administered in the studies on rats and given that DTPA was not even used in products that were intended to be ingested.

116    In the second place, the applicants submit that the available scientific data cannot be regarded as ‘clear evidence’ that DTPA could produce adverse effects on the development of the foetus ‘in the absence of other toxic effects’, as is required by Section 3.7.2.1.1 and Table 3.7.1(a) of Annex I to Regulation No 1272/2008. The BASF study merely showed some adverse effects resulting from zinc depletion but did not address the zinc status of pregnant rats. Moreover, the other ‘supporting studies’, mentioned in the RAC opinions, were unable to provide clear evidence that classification in Category 1B had been justified, since they involved intravenous administration at extremely high dosage levels. Similarly, the Kalkwarf study, mentioned in the additional RAC opinion, was irrelevant because it was a pharmacological study carried out in exceptional circumstances, in the context of a treatment given following an acute radioactive metal poisoning incident.

117    The Commission, supported by the French Republic and ECHA, disputes those arguments.

118    As a preliminary point, it should be noted that, in the context of the classification for reproductive toxicity, Category 1B corresponds to presumed human reproductive toxicants and Category 2 corresponds to suspected human reproductive toxicants, in accordance with Section 3.7.2.1.1 and Table 3.7.1(a) of Annex I to Regulation No 1272/2008 (see paragraph 30 above).

119    Table 3.7.1(a) of Annex I to Regulation No 1272/2008 provides, inter alia, that substances are classified in Category 1B when ‘there is evidence from animal studies, possibly supplemented with other information, to provide a strong presumption that the substance has the capacity to interfere with reproduction in humans’. Such data from animal studies are to ‘provide clear evidence’ of the existence of ‘an adverse effect on … development in the absence of other toxic effects, or if occurring together with other toxic effects the adverse effect on reproduction is considered not to be a secondary non-specific consequence of other toxic effects’. It is also stated in that table that, ‘however, when there is mechanistic information that raises doubt about the relevance of the effect for humans, classification in Category 2 may be more appropriate.’

120    In addition, Section 3.7.2.3.2 of Annex I to Regulation No 1272/2008 provides that ‘if it is conclusively demonstrated that the clearly identified mechanism or mode of action has no relevance for humans or when the toxicokinetic differences are so marked that it is certain that the hazardous property will not be expressed in humans then a substance which produces an adverse effect on reproduction in experimental animals should not be classified.’

121    In the light of the provisions cited above, the case-law cited in paragraph 31 above and the fact that the choice of the hazard category appropriate for the classification of a substance considered toxic for reproduction involves complex scientific and technical assessments, the Commission must be allowed a broad discretion in that regard.

122    In the present case, the contested regulation established the harmonised classification and labelling of DTPA in the hazard category 1B of reproductive toxicants, in accordance with the RAC opinions (see paragraphs 13 and 15 above).

123    The contested regulation therefore classified DTPA in a hazard category higher than Category 2, which was that proposed by the applicants, as authors of the classification proposals submitted to ECHA and registrants of DTPA (see paragraphs 5 to 7 and 17 above).

124    Furthermore, it should be recalled that it follows from the examination of the first plea in law that it has not been demonstrated in the present case that the contested classification was adopted in breach of the classification criteria referred to in Article 36(1) of Regulation No 1272/2008 and Section 3.7.2.2.1 of Annex I to that regulation, according to which the substance must have the intrinsic property of specifically harming reproduction and that effect should not be considered to be a non-specific secondary consequence of other toxic effects (see paragraphs 81, 84 and 88 above).

125    In addition, it is apparent from the examination of the second plea that, contrary to what the applicants claim, the RAC opinions on which the contested classification was based took into account the possible influence of maternal toxicity, in accordance with Section 3.7.2.2.2 of Annex I to Regulation No 1272/2008 (see paragraph 113 above).

126    It is in the light of those considerations that the applicants’ arguments in the context of the third plea, raised in the alternative to the first and second pleas, must be examined.

127    In the first place, the applicants submit that the contested classification is not based on data from animal studies that establish a ‘strong presumption’ that the substance has the capacity to interfere with reproduction in humans. The Commission failed to take into account either the significant differences between DTPA’s mode of action in rats and that mode of action in humans or the fact that the adverse effects were observed at a level of exposure to DTPA which is unrealistic or impossible to reach in humans.

128    First of all, it follows from paragraph 74 above that, in its first opinions, the RAC considered that the mode of action described in paragraph 73 above, consisting of developmental toxicity in the foetus induced by a maternal zinc deficiency, was plausible and potentially relevant to humans.

129    Next, it is apparent from its additional opinion that the RAC concluded that the relevance to humans of the toxicity found in the BASF study could not be ruled out. In that additional opinion, in the chapter entitled ‘Susceptibility of pregnant women to DTPA-induced zinc deficiency’, the RAC, inter alia, analysed the additional information provided by the authors of the classification proposals, referred to in paragraph 10 above, and considered, on the basis of a set of data and scientific studies detailed in that opinion, that the additional information received did not rule out the relevance to humans of the hazard to development of the offspring identified in the BASF study.

130    Lastly, it is apparent from recitals 3 and 4 of the contested regulation that the Commission considered that other additional information, received following the additional RAC opinion, was not sufficient to cast doubt on the scientific analysis set out in that opinion.

131    It should be noted that the RAC’s assessments on the relevance to humans of toxicity for development of the offspring identified in the BASF study appear plausible in the light of the provisions for classification as a reproductive toxicant in Category 1B, which are recalled in paragraph 119 above. Those provisions require (i) the existence of data from animal studies which are to provide clear evidence of an adverse effect on development that is considered not to be a secondary non-specific consequence of other toxic effects and (ii) that there be no mechanistic information capable of raising doubt about the relevance of the effect for humans.

132    The RAC considered (i) that the adverse effects of DTPA on development of the offspring, observed in animal studies, were not a secondary non-specific consequence of other toxic effects (see paragraphs 123 and 125 above) and (ii) that there was insufficient evidence to raise doubt about the relevance of DTPA’s mode of action for humans (see paragraph 129 above).

133    Moreover, as the Commission submits, it follows from Section 3.7.2.3.2 of Annex I to Regulation No 1272/2008 that the level of confidence needed to dismiss the relevance to humans of reproductive toxicity observed in experimental animals must be particularly high. Thus, the substance should not be classified only in cases where it ‘is conclusively demonstrated that the clearly identified mechanism or mode of action has no relevance for humans’ or where ‘the toxicokinetic differences are so marked that it is certain that the hazardous property will not be expressed in humans’ (see paragraph 120 above). Neither of those two hypotheses has been demonstrated in the present case.

134    In any event, the Court cannot substitute itself for the Commission in those assessments as to the validity of the RAC’s conclusions in that regard, in view of the Commission’s discretion referred to in paragraph 121 above and the absence of a manifest error of assessment capable of affecting the plausibility of the RAC’s scientific conclusions; such an error has not been demonstrated, or even alleged, in the present case.

135    As regards the applicants’ argument that the studies on which the RAC opinions are based were conducted at levels of exposure to DTPA which are unrealistic or impossible to reach in humans, it is sufficient to recall that Regulation No 1272/2008 concerns the assessment of hazards of substances and that that assessment must be differentiated from the risk assessment provided for in Regulation No 1907/2006. As is apparent from the case-law cited in paragraph 27 above, an assessment of the hazards linked to the intrinsic properties of a substance must not be limited in the light of specific circumstances of use, as in the case of a risk assessment, and may be properly carried out regardless of the possible levels of exposure to the substance.

136    The applicants are therefore wrong to claim that the contested classification is not based on evidence from animal studies which provided a strong presumption that the substance had the capacity to interfere with reproduction in humans.

137    In the second place, the applicants submit that the available scientific data cannot be regarded as ‘clear evidence’ that DTPA could produce adverse effects on the development of the foetus ‘in the absence of other toxic effects’, as is required by Section 3.7.2.1.1 and Table 3.7.1(a) of Annex I to Regulation No 1272/2008.

138    First of all, it is necessary to reject the applicants’ argument that such proof is not apparent from the BASF study, in so far as that study showed certain harmful effects resulting from zinc depletion, but did not examine the ‘zinc status’ of pregnant rats.

139    On the one hand, as the Commission contends, the reliability of the BASF study is undisputed in the present case. Both the RAC and the authors of the classification proposals, as well as the applicants, considered that the BASF study was reliable with a score of 1, corresponding to ‘reliable without restriction’, in the Klimisch scoring system (as described in the article of Klimisch, H.J., Andreae, M. and Tillmann, U., ‘A Systematic Approach for Evaluating the Quality of Experimental Toxicological and Ecotoxicological Data’, Regulatory Toxicology and Pharmacology, Elsevier, 1997, Vol. 25, pp. 1 to 5). Similarly, both the RAC and the authors of the classification proposals, as well as the applicants, relied, in order to propose a classification of DTPA as a reproductive toxicant, on the mode of action of the toxicity observed during the BASF study (see paragraphs 5, 6 and 72 above).

140    On the other hand, even if it were established that the BASF study did not contain an examination of the ‘zinc status’ of pregnant rats, that is to say, that the study had not measured the quantities of zinc in pregnant rats and therefore did not allow conclusions to be drawn on the level of zinc depletion, that is not sufficient to undermine the credibility of that study for the purposes of the contested classification, in view of the circumstances referred to, in particular, in paragraphs 131 to 135 above.

141    In addition, the applicants’ argument concerning ‘zinc status’ overlaps with other arguments which have already been rejected, namely the argument concerning the existence of other external sources of zinc deficiency, the argument concerning the relevance of DTPA’s mode of action for humans and the argument relating to the level of exposure (see paragraphs 86, 127 and 135 above).

142    Next, it should be noted that the applicants’ argument based, in essence, on the inadequacy of the other ‘supporting studies’ on which the RAC relied is ineffective. The key study on which the RAC relied, and indeed the authors of the classification proposals, is the BASF study (see paragraph 72 above). Thus, without prejudice to the ancillary role of the other studies mentioned in the RAC opinions, those studies did not form the basis for the contested classification, contrary to what appears to emerge from the applicants’ line of argument.

143    For the same reasons, it must be held that the applicants’ argument concerning the Kalkwarf study is also ineffective, since that study is mentioned in the additional RAC opinion only in relation to an estimate of the amount of zinc lost by the human body.

144    Lastly, it should be recalled that it follows from Section 3.7.2.1.1 and Table 3.7.1(a) of Annex I to Regulation No 1272/2008 that classification of a substance in Category 1B of reproductive toxicity may be based on data from animal studies, provided that those data provide clear evidence of an adverse effect, inter alia, on development, either in the absence of other toxic effects or if occurring together with other toxic effects, but that the adverse effect on reproduction is not considered to be a secondary non-specific consequence of other toxic effects.

145    In that regard, it follows from the examination of the first plea in law that the RAC concluded, without the applicants calling into question the plausibility of that conclusion, that the adverse effects of DTPA were not a non-specific secondary consequence of other toxic effects (see paragraph 88 above). In addition, it follows from the examination of the second plea in law that, contrary to what the applicants claim, the RAC took into account the possible influence of maternal toxicity, but concluded that that toxicity was either not severe or was non-existent (see paragraph 113 above).

146    The applicants are therefore wrong to claim that the available scientific data cannot be regarded as ‘clear evidence’ that DTPA could produce adverse effects on the development of the foetus in the absence of any other toxic effects.

147    Accordingly, the third plea in law must be rejected as in part unfounded and in part ineffective.

 The fourth plea in law, alleging infringement of the Commission’s obligation to examine the appropriateness of the contested classification

148    The applicants submit that the Commission failed to comply with its obligation to examine the appropriateness of the contested classification, in accordance with Article 37(5) of Regulation No 1272/2008. The Commission also breached the principles of proportionality and sound administration. First, the Commission failed to confirm that the RAC opinions were accurate, reliable and consistent. Secondly, the Commission failed to assess the appropriateness of the classification proposed by the RAC in the light of all the relevant information, including in particular that provided by the authors of the proposals for harmonised classification and labelling of DTPA. Thirdly, the Commission failed to consider that the contested classification did not meet the fundamental objective referred to in Section 3.7.1.4 of Annex I to Regulation No 1272/2008, which is to provide a hazard warning for pregnant women, and for men and women of reproductive capacity, about possible effects of the substance. The intake of DTPA that would be needed for a comparable level of zinc depletion to be reached in humans as that observed in rats is unrealistic.

149    The Commission, supported by the French Republic and ECHA, disputes those arguments.

150    Under Article 37(5) of Regulation No 1272/2008, ‘the Commission shall without undue delay adopt delegated acts in accordance with Article 53a [of that regulation], where it finds that the harmonisation of the classification and labelling of the substance concerned is appropriate, to amend Annex VI [thereto] by inclusion of that substance together with the relevant classification and labelling elements in [Table 3 of Part 3 of that annex] and, where appropriate, the specific concentration limits or M-factors.’ To that end, the Commission must take into account, first of all, the proposal submitted pursuant to Article 37(1) to (3) of that regulation, next, the RAC’s opinion and, lastly, the observations made during the public consultations, in accordance with Article 37(2) and (4) thereof, although those elements, in particular the RAC’s opinion, are not binding on the Commission (judgment of 5 July 2023, TIB Chemicals v Commission, T‑639/20, not published, EU:T:2023:374, paragraph 204).

151    As is clear from the wording of Article 37(5) of Regulation No 1272/2008, the Commission is to adopt delegated acts for the purposes of the inclusion of a substance in Annex VI to Regulation No 1272/2008 if ‘it finds that the harmonisation of the classification and labelling of the substance concerned is appropriate’. However, neither that article nor any other provision of Regulation No 1272/2008 sets out the criteria which must be taken into account by the Commission in order to find that the harmonisation of the classification and labelling of a substance is ‘appropriate’. It follows from the case-law cited in paragraph 31 above that the Commission must be recognised as enjoying a broad discretion, in the light of the complex scientific and technical assessments which it must undertake if it is to be able to classify a substance. It follows that the Commission has a broad discretion in determining the appropriateness of a proposal for harmonised classification and labelling pursuant to Article 37(5) of that regulation, which, moreover, the applicants do not dispute.

152    It should also be noted that the Court has already held, in a judgment concerning a decision granting authorisations under Regulation No 1907/2006, that, where the Commission endorsed the opinion of one of the ECHA committees in order to justify an authorisation decision, it had to ensure that the reasoning in the opinion was full, consistent and relevant (see judgment of 7 March 2019, Sweden v Commission, T‑837/16, EU:T:2019:144, paragraph 68 and the case-law cited). That case-law can be applied in the context of the procedure for harmonisation of classification and labelling of substances, governed by Regulation No 1272/2008.

153    In the present case, in the first place, it should be noted that it follows from the 16th indent of recital 2 and from recital 3 of the contested regulation that the Commission endorsed the RAC opinions on DTPA. In addition, as is apparent from recitals 3 and 4 of the contested regulation, the Commission considered that additional information, received in response to the additional RAC opinion, was not sufficient to cast doubt on the scientific analysis set out in that opinion.

154    Accordingly, it must be concluded that the Commission exercised its broad discretion, pursuant to Article 37(5) of Regulation No 1272/2008, in considering that that harmonisation of the classification and labelling of DTPA was appropriate.

155    In the second place, it should be noted that there is nothing in the file to indicate that the Commission failed to verify that the RAC opinions were factually accurate, reliable and consistent. The applicants’ arguments alleging ambiguities or contradictions in the RAC opinions or divergent positions between the RAC and ECHA cannot succeed.

156    Thus, first of all, the applicants merely assert that the RAC’s reasoning in its first opinions does not support its conclusion that ‘DTPA salts are able to induce developmental toxicity’, without, however, indicating, with a minimum degree of precision, how that reasoning is lacking. Next, the applicants reiterate the argument that the Kalkwarf study cannot support the relevance of the effects in humans, an argument which has already been rejected as ineffective for the reasons referred to in paragraph 143 above. Lastly, the applicants merely claim, in very general terms, that the Commission should have concluded that the RAC opinions were ambiguous and contradictory. As to the remainder, the applicants merely reiterate some of the arguments already raised in the context of the first plea, which have to be rejected, as is apparent from paragraph 94 above.

157    In the third place, it has not been demonstrated in the present case that the Commission did not examine the classification proposed by the RAC in the light of all of the relevant factors, and the applicants do not, moreover, indicate any specific factor which should have been taken into account by the Commission. In addition, contrary to what the applicants claim, the Commission took into consideration both the information provided by the authors of the proposals for harmonised classification and labelling of DTPA following the first RAC opinions (see paragraphs 10 to 12 above) and the additional information provided by those authors following the additional RAC opinion (recitals 3 and 4 of the contested regulation).

158    In the fourth place, as regards the applicants’ argument that the contested classification is not appropriate for attaining the objective of health protection, it should be recalled that the purpose of Regulation No 1272/2008 is to ensure a high level of protection of human health and the environment as well as the free movement of chemical substances, mixtures and certain specific articles on the EU market. The objective of the regulation is to determine the intrinsic properties of the substances which must lead to their classification as hazardous products, so that the hazards posed by those substances (and mixtures containing such substances) can be correctly identified and notified (see paragraph 22 above). It follows that the contested classification, in so far as it seeks to identify and notify a reproductive toxicity hazard under Regulation No 1272/2008, must be regarded as an instrument for improving the protection of health, in accordance with the objectives and purpose of that regulation.

159    In the fifth place, the applicants’ argument that zinc depletion in humans comparable to that observed in rats is unrealistic must be rejected for the reasons set out in paragraph 135 above.

160    In view of the foregoing, it must be concluded that an infringement of Article 37(5) of Regulation No 1272/2008 has not been established in the present case.

161    For the same reasons, it must be concluded that the Commission did not breach the principles of proportionality and sound administration, and the applicants do not moreover put forward any specific argument in that regard.

162    The fourth plea must therefore be rejected as in part unfounded and in part ineffective.

 The fifth plea in law, alleging the absence of public consultations regarding the RAC opinions

163    The applicants submit that the Commission infringed Article 37(4) of Regulation No 1272/2008, in so far as that provision lays down an obligation for the RAC to organise public consultations on its opinions, and not solely on classification proposal dossiers. The applicants’ right to comment on RAC opinions, and even on draft RAC opinions, flows from their right to be heard and from the wording of Article 37(4) of that regulation, read in conjunction with recital 52 of that regulation. That right is all the more justified in cases such as the present one, in which the classification proposed by the RAC diverges from that referred to in the proposals for harmonised classification and labelling of DTPA.

164    The Commission, supported by the French Republic and ECHA, disputes those arguments.

165    It should be recalled that, according to Article 41(2)(a) of the Charter of Fundamental Rights of the European Union, the right to good administration includes the right of every person to be heard before any individual measure which would affect him or her adversely is taken. Respect for the right to be heard is, in all proceedings initiated against a person which are liable to culminate in a measure adversely affecting that person, a fundamental principle of EU law which must be guaranteed even in the absence of rules governing the proceedings in question. That principle requires that the addressees of decisions which significantly affect their interests should be placed in a position in which they can effectively make known their views on the accusation made against them forming the basis of the contested measure (see, to that effect, judgment of 19 December 2019, Probelte v Commission, T‑67/18, EU:T:2019:873, paragraph 86 and the case-law cited).

166    By contrast, in the case of acts of general application, neither the process of drafting them nor those acts themselves require, in accordance with the general principles of EU law, such as the right to be heard, consulted or informed, the participation of the persons affected. That is not the case if an express provision of the legal context governing the adoption of that act confers such a procedural right on a person affected (see, to that effect, judgment of 19 December 2019, Probelte v Commission, T‑67/18, EU:T:2019:873, paragraph 87 and the case-law cited).

167    In the present case, the contested regulation lays down measures of general application, including the contested classification. As is apparent from paragraph 15 above, Article 1 of that regulation lays down a measure of general application concerning the inclusion of the substances constituting DTPA, that is to say, DTPA-K5, DTPA-H5 and DTPA-Na5, in the ‘list of harmonised classification and labelling for hazardous substances’, which is set out in Table 3 of Part 3 of Annex VI to Regulation No 1272/2008.

168    Against that background, the procedural rights which the applicants enjoy in the procedure for harmonised classification and labelling are those expressly provided for in Regulation No 1272/2008 (see, to that effect, judgment of 5 July 2023, TIB Chemicals v Commission, T‑639/20, not published, EU:T:2023:374, paragraph 180 and the case-law cited).

169    In that regard, Article 37(4) of Regulation No 1272/2008 provides that ‘the [RAC] shall adopt an opinion on any proposal submitted pursuant to paragraphs 1 or 2 [of that article] within 18 months of receipt of the proposal, giving the parties concerned the opportunity to comment’ and that ECHA ‘shall forward this opinion and any comments to the Commission’.

170    Article 37(4) of Regulation No 1272/2008 must be interpreted in the light of the procedure for harmonisation of classification and labelling of substances, referred to in Article 37 of that regulation. As stated in paragraph 26 above, that procedure takes place in several stages, namely, first of all, the submission of a classification proposal; next, the adoption of an opinion by the RAC ‘giving the parties concerned the opportunity to comment’; subsequently, the forwarding by ECHA of that opinion and all comments to the Commission; and, lastly, the adoption by the Commission of a delegated act, where it considers that harmonisation of the classification and labelling of the substance concerned is appropriate.

171    It follows that the public consultation provided for in Article 37(4) of Regulation No 1272/2008 is intended to allow interested parties to comment on the classification proposal and thus possibly to contribute elements not mentioned in that proposal, so as to allow the RAC to take account, in its opinion, of the comments and elements presented by the interested parties during that phase (see, to that effect and by analogy, judgment of 9 June 2021, Exxonmobil Petroleum & Chemical v ECHA, T‑177/19, not published, EU:T:2021:336, paragraph 236 and the case-law cited).

172    Therefore, it should be noted that, although Article 37(4) of Regulation No 1272/2008 provides for the possibility of submitting comments on the proposal for harmonised classification and labelling, that regulation does not, however, provide for the possibility for the parties concerned to submit observations on the RAC’s opinion (see, to that effect, judgment of 5 July 2023, TIB Chemicals v Commission, T‑639/20, not published, EU:T:2023:374, paragraph 184).

173    Accordingly, in the present case, the applicants had the right to comment on the proposals for harmonised classification and labelling and to be heard in that regard before the RAC, and availed themselves of that right. Between February and July 2016, public consultations concerning the proposals for harmonised classification and labelling of DTPA were organised (see paragraph 8 above). Moreover, as is apparent from the content of the first RAC opinions themselves, the RAC took into consideration the comments submitted by several Member States and parties concerned, which were compiled as an annex to those opinions. In addition, in the course of 2020, a targeted public consultation regarding the ‘developmental toxicity’ of DTPA-H5, DTPA-Na5 and DTPA-K5 was organised. Furthermore, as is apparent from the content of the additional RAC opinion, the RAC took into consideration the comments submitted by a Member State (see paragraph 12 above).

174    In those circumstances, and in accordance with the case-law cited in paragraphs 166 and 168 above, the applicants did not have a right to be heard or consulted on the RAC opinions.

175    That conclusion is not called into question by the applicants’ argument that the right to be heard on the RAC opinions is more justified where the classification proposed by the RAC is different from that referred to in the classification proposals. Quite apart from the fact that the RAC is not bound by the classification proposal, it follows from paragraph 171 above that the public consultation provided for in Article 37(4) of Regulation No 1272/2008 is not concerned with the RAC opinion, but, rather, is intended to enable the RAC to take into account in its opinion the comments submitted by the parties concerned.

176    It follows from the foregoing that an infringement of Article 37(4) of Regulation No 1272/2008 and of the right to be heard has not been demonstrated in the present case.

177    The fifth plea in law must therefore be rejected as being unfounded.

 The sixth plea in law, alleging the absence of an impact assessment

178    The applicants submit that the Commission infringed point 13 of the Interinstitutional Agreement of 13 April 2016 between the European Parliament, the Council of the European Union and the Commission on Better Law-Making (OJ 2016 L 123, p. 1; ‘the Interinstitutional Agreement on Better Law-Making’), as well as breached the principle of sound administration, on the ground that it did not carry out an impact assessment of the proposal for harmonised classification and labelling of DTPA before the adoption of the contested regulation. The Commission’s commitment to carry out an impact assessment is also apparent from its ‘Better Regulation Guidelines’ of 3 November 2021. Furthermore, it follows from the judgment of 3 December 2019, Czech Republic v Parliament and Council (C‑482/17, EU:C:2019:1035), applicable in the present case by analogy, that the Commission undertook to carry out impact assessments where its legislative initiatives were expected to have significant economic, environmental or social impacts, as is the case with the harmonised classification and labelling of DTPA.

179    The Commission, supported by the French Republic and ECHA, disputes those arguments.

180    It should be noted that the Interinstitutional Agreement on Better Law-Making, which establishes a series of initiatives and procedures with a view to improving the way in which the European Union legislates, makes provision for impact assessments, in points 12 to 18 thereof, as one of the ‘tools for better law-making’. In particular, point 13 of that interinstitutional agreement provides as follows:

‘The Commission will carry out impact assessments of its legislative and non-legislative initiatives, delegated acts and implementing measures which are expected to have significant economic, environmental or social impacts. The initiatives included in the Commission Work Programme or in the joint declaration will, as a general rule, be accompanied by an impact assessment.

In its own impact assessment process, the Commission will consult as widely as possible. The Commission’s Regulatory Scrutiny Board will carry out an objective quality check of its impact assessments. The final results of the impact assessments will be made available to the European Parliament, the Council and national Parliaments, and will be made public along with the opinion(s) of the Regulatory Scrutiny Board at the time of adoption of the Commission initiative.’

181    In addition, Chapter 4 of the ‘Better Regulation Guidelines’ of 3 November 2021 sets out the objectives of an impact assessment and the elements which it must contain. Those guidelines have been supplemented by Tool No 7 in the Commission’s ‘Better Regulation Toolbox’.

182    As regards the legislative process, the Court of Justice has already held that an obligation to carry out an impact assessment in every circumstance does not follow from the wording of points 12 to 15 of the Interinstitutional Agreement on Better Law-Making (see, to that effect, judgment of 3 December 2019, Czech Republic v Parliament and Council, C‑482/17, EU:C:2019:1035, paragraph 82).

183    First, points 12 to 15 of the Interinstitutional Agreement on Better Law-Making show that the European Parliament, the Council of the European Union and the Commission recognise the contribution of impact assessments in improving the quality of EU legislation and that those assessments are a tool to help the three institutions concerned reach well-informed decisions. Secondly, those points stipulate that impact assessments must not lead to undue delays in the law-making process or prejudice the co-legislators’ capacity to propose amendments, for which it is moreover provided that additional impact assessments may be carried out when the Parliament and the Council consider it to be appropriate and necessary. Thirdly, those same points note that the Commission will carry out impact assessments of its legislative initiatives which are expected to have significant economic, environmental or social implications. Fourthly, it is stated that the Parliament and the Council, when examining the Commission’s legislative proposals, are to take full account of the Commission’s impact assessments (see, to that effect, judgment of 3 December 2019, Czech Republic v Parliament and Council, C‑482/17, EU:C:2019:1035, paragraph 83).

184    It follows that the preparation of impact assessments is a step in the legislative process that, as a rule, must take place if a legislative initiative is liable to have such implication (see, to that effect, judgment of 3 December 2019, Czech Republic v Parliament and Council, C‑482/17, EU:C:2019:1035, paragraph 84).

185    In the present case, it should be noted that, in accordance with the case-law cited in paragraph 182 above, it does not follow from point 13 of the Interinstitutional Agreement on Better Law-Making that the Commission is required, in all circumstances, to carry out an impact assessment of its delegated acts.

186    Moreover, such an obligation does not follow from Article 37 of Regulation No 1272/2008, governing the procedure for harmonised classification and labelling, which does not provide for such an assessment at any of the stages of that procedure (see paragraphs 26 and 172 above).

187    By contrast, it follows from the provisions of Article 76(1)(c) and (d) of Regulation No 1907/2006 that, in the context of the procedure for the harmonisation of classification and labelling of substances governed by Regulation No 1272/2008, it is the RAC which is to be responsible for preparing the opinion of ECHA and that the Committee for Socio-economic Analysis is not involved. In the absence of any express provision for the involvement of the latter, it can be deduced that the legislature did not wish to include the socioeconomic impact in that procedure.

188    Furthermore, it should be noted that the objective pursued by Regulation No 1272/2008, namely to ensure a high level of protection of human health and the environment (see paragraph 22 above), may justify adverse economic consequences, and even substantial adverse consequences, for certain traders. In that context, the protection of public health, which the contested regulation is intended to guarantee, must take precedence over economic considerations (see, to that effect and by analogy, judgment of 11 September 2002, Pfizer Animal Health v Council, T‑13/99, EU:T:2002:209, paragraph 456 and the case-law cited).

189    It follows from the foregoing that, in the context of the procedure for harmonisation of classification and labelling which led to the adoption of the contested regulation, the Commission was under no obligation to carry out an impact assessment of that regulation under point 13 of the Interinstitutional Agreement on Better Law-Making (see, to that effect and by analogy, judgment of 5 July 2023, TIB Chemicals v Commission, T‑639/20, not published, EU:T:2023:374, paragraph 206).

190    Furthermore, the applicants’ argument alleging breach of the principle of sound administration, in that the Commission infringed the Interinstitutional Agreement on Better Law-Making, must be rejected. It follows from the previous paragraph that the Commission was under no obligation to carry out an impact assessment before adopting the contested regulation. Consequently, the absence of such an impact assessment cannot constitute a breach of the principle of sound administration.

191    In the light of all of the foregoing, the sixth plea must be rejected as unfounded and, consequently, the action must be dismissed in its entirety.

 Costs

192    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.

193    Since the applicants have been unsuccessful, they must be ordered to bear their own costs and to pay those incurred by the Commission, in accordance with the form of order sought by the latter.

194    Under Article 138(1) of the Rules of Procedure, the Member States and institutions which have intervened in the proceedings are to bear their own costs. Under Article 1(2)(f) of the Rules of Procedure, the term ‘institutions’ means the institutions of the European Union referred to in Article 13(1) TEU and the bodies, offices or agencies established by the Treaties, or by an act adopted in implementation thereof, which may be parties before the General Court. Under Article 100 of Regulation No 1907/2006, ECHA is a body of the European Union. It follows that the French Republic and ECHA must each bear their own costs.

On those grounds,

THE GENERAL COURT (Sixth Chamber)

hereby:

1.      Dismisses the action;

2.      Orders BASF SE, Dow Europe GmbH and Nouryon Functional Chemicals BV to bear their own costs and to pay those incurred by the European Commission;

3.      Orders the French Republic and the European Chemicals Agency (ECHA) to each bear their own costs.

Delivered in open court in Luxembourg on 27 November 2024.

*      Language of the case: English.

© European Union
The source of this judgment is the Europa web site. The information on this site is subject to a information found here: Important legal notice. This electronic version is not authentic and is subject to amendment.