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 – Harmonised classification and labelling of the substance silanamine, 1,1,1-trimethyl-N-(trimethylsilyl)-, hydrolysis products with silica; pyrogenic, synthetic amorphous, nano, surface treated silicon dioxide – Criteria for classification of a substance in the hazard class ‘Specific target organ toxicity-repeated exposure’ – Appropriateness of the classification – Absence of public consultation on the opinion of ECHA’s Committee for Risk Assessment – Interinstitutional Agreement on Better Law-Making – Absence of impact assessment )

In Case T‑449/22,

Evonik Operations GmbH, established in Essen (Germany), represented by T. Delille and N. Kuśnierkiewicz, lawyers,

applicant,

v

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

defendant,

supported by

European Chemicals Agency (ECHA), represented by W. Broere and J.‑P. Trnka, acting as Agents,

intervener,

THE GENERAL COURT (Sixth Chamber),

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

Registrar: S. Spyropoulos, Administrator,

having regard to the written part of the procedure,

having regard to the measure of organisation of procedure of 15 December 2023,

further to the hearing on 8 February 2024,

gives the following

Judgment

1 By its action under Article 263 TFEU, the applicant, Evonik Operations GmbH, seeks 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 it concerns the harmonised classification and labelling of the substance silanamine, 1,1,1-trimethyl-N-(trimethylsilyl)-, hydrolysis products with silica; pyrogenic, synthetic amorphous, nano, surface treated silicon dioxide (‘silanamine’).

I. Background to the dispute

2 The applicant is a company incorporated under German law which manufactures silanamine.

3 Silanamine is a type of synthetic amorphous silica (‘SAS’), with a molecular formula of ‘[SiO2]n-[OSi(CH3)3]m’, which has been processed to become hydrophobic. SAS, including surface treated forms, are used in various products, such as medicinal and pharmaceutical products, food and cosmetics, and also in a variety of industrial applications, as reinforcement and thickening agents in various systems such as elastomers, resins and inks.

4 Silanamine is also an active substance within the meaning of Regulation (EU) No 528/2012 of the European Parliament and of the Council of 22 May 2012 concerning the making available on the market and use of biocidal products (OJ 2012 L 167, p. 1).

5 On 17 December 2018, the Agence nationale de sécurité sanitaire de l’alimentation, de l’environnement et du travail (National Agency for Food, Environmental and Occupational Health and Safety, ANSES, France; ‘the French competent authority’) submitted to the European Chemicals Agency (ECHA), in accordance with Article 37(1) 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), a proposal for the harmonised classification and labelling of silanamine (‘the classification proposal’) in the hazard class ‘specific target organ toxicity – repeated exposure’ (‘the hazard class STOT RE’) of Category 2. The classification proposal was marked ‘version 2’.

6 Between 4 March and 3 May 2019, several of the parties concerned submitted their comments on the classification proposal, in accordance with Article 37(4) of Regulation No 1272/2008.

7 On 5 December 2019, on the basis of Article 37(4) of Regulation No 1272/2008, ECHA’s Committee for Risk Assessment (‘RAC’) adopted an opinion (‘the RAC Opinion’). In that opinion, RAC proposed the classification of silanamine in the hazard class STOT RE Category 2, with hazard statement code ‘H373’ (lungs, inhalation), as well as in the hazard class acute toxicity of Category 2 via the inhalation route (H330) with an acute toxicity estimate of 0.45 milligram per litre (mg/L).

8 Between 3 and 17 February 2020, a targeted public consultation was organised on the classification of silanamine in the hazard class acute toxicity.

9 On 16 February 2022, on the basis of the RAC Opinion, the European Commission adopted the contested regulation. By that regulation, silanamine was added to Table 3 of Part 3 of Annex VI to Regulation No 1272/2008, with harmonised classification and labelling in the hazard class STOT RE Category 2, hazard statement code ‘H373’ (lungs, inhalation) and pictogram hazard code ‘GHS 08Wng’ (‘the contested classification’).

10 On the other hand, the contested regulation stated that the classification of silanamine in the hazard class acute toxicity Category 2, recommended in the RAC Opinion, should not be included in Table 3 of Part 3 of Annex VI to Regulation No 1272/2008, as the new scientific information received had been assessed by the Commission, which had found that it required further assessment by the RAC.

II. Forms of order sought

11 The applicant claims that the Court should:

– annul the contested regulation in so far as it concerns the contested classification;

– order the Commission to pay the costs.

12 The Commission, supported by ECHA, contends that the Court should:

– dismiss the action;

– order the applicant to pay the costs.

III. Law

13 In support of its action, the applicant puts forward four pleas in law:

– the first plea alleges manifest errors of assessment and breach of the criteria for classification of a substance in the hazard class STOT RE Category 2;

– the second plea alleges breach of the harmonised classification and labelling procedure;

– the third plea alleges breach of the Commission’s obligation to examine the appropriateness of the contested classification;

– the fourth plea alleges absence of any impact assessment.

A. Preliminary considerations on harmonised classification and labelling of substances in the hazard class STOT RE

14 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 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 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’.

15 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).

16 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.

17 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.

18 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, the 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.

19 Furthermore, it should be recalled 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 (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 [ECHA], 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). 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).

20 The hazard class STOT RE is provided for in Section 3.9 of Part 3 of Annex I to Regulation No 1272/2008.

21 In particular, Section 3.9.1.1 of Annex I to Regulation No 1272/2008 provides as follows:

‘[Hazard class STOT RE] means specific toxic effects on target organs occurring after repeated exposure to a [substance] or mixture. All significant health effects that can impair function, reversible and irreversible, immediate and/or delayed are included. However, other specific toxic effects that are specifically addressed in sections 3.1 to 3.8 and 3.10 are not included here.’

22 With regard to the hazard categories, it follows from Section 3.9.2.1 and Table 3.9.1 of Annex I to Regulation No 1272/2008 that the classification for the hazard class STOT RE is divided into two categories, depending upon the nature and severity of the effect(s) observed, namely Category 1 and Category 2.

23 More specifically, Category 2 covers substances that on the basis of evidence from studies in experimental animals can be presumed to have the potential to be harmful to human health following repeated exposure. Substances are classified in that category on the basis of observations from appropriate studies in experimental animals in which significant toxic effects, of relevance to human health, were produced at generally moderate exposure concentrations.

B. Preliminary observations on the intensity of the Court’s review

24 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).

25 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).

26 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).

27 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).

28 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).

29 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).

30 It is in the light of those considerations that the applicant’s pleas in law must be examined.

C. The first plea in law, alleging manifest errors of assessment and breach of the criteria for classification of a substance in the hazard class STOT RE Category 2

31 In the context of the first plea, the applicant submits, in essence, that the contested classification is vitiated by manifest errors of assessment and does not meet the criteria for classification in the hazard class STOT RE Category 2.

32 That first plea is divided into three parts. The first alleges that silanamine produces no specific toxic effects on the lungs, the second alleges that the observed effects are based on a read-across that does not meet the criteria for classification in the hazard class STOT RE Category 2 and the third alleges that silanamine produces no adverse health effects.

33 In the light of the circumstances of the present case, the Court considers it appropriate to examine, first of all, the third part, next, the first part and, lastly, the second part of the first plea.

1. The third part, alleging that silanamine has no adverse health effects

34 In the context of the third part of the first plea, the applicant submits, in essence, that the contested classification is not based on adverse health effects within the meaning of Section 3.9 of Annex I to Regulation No 1272/2008. In particular, it argues that the RAC Opinion does not identify any effect justifying such a classification. It adds that none of the conclusions in the RAC Opinion meet the criteria laid down in Section 3.9 of Annex I to Regulation No 1272/2008.

35 Having regard to the applicant’s arguments, it is appropriate, first, to examine the effects considered to support classification of a substance in the hazard class STOT RE Category 2, secondly, to verify whether the RAC Opinion identified the effects which were considered to support the contested classification and, thirdly, to analyse whether the effects identified could justify that classification.

(a) Preliminary considerations concerning the effects considered to support classification of a substance in the hazard class STOT RE Category 2

36 As is apparent from paragraph 21 above, under Section 3.9.1.1 of Annex I to Regulation No 1272/2008, the hazard class STOT RE means specific toxic effects on target organs occurring after repeated exposure to a substance or mixture; in addition, all significant health effects that can impair function, reversible and irreversible, immediate and/or delayed are included.

37 In accordance with Section 3.9.1.2 of Annex I to Regulation No 1272/2008, hazard class STOT RE identifies substances which are a specific target organ toxicant and, as such, may present a potential for adverse health effects in people who are exposed to them.

38 Pursuant to Section 3.9.1.3 of Annex I to Regulation No 1272/2008, those adverse health effects include consistent and identifiable toxic effects in humans, or, in experimental animals, toxicologically significant changes which have affected the function or morphology of a tissue/organ, or have produced serious changes to the biochemistry or haematology of the organism and those changes are relevant for human health.

39 Section 3.9.2.7 of Annex I to Regulation No 1272/2008 specifies the rules relating to the effects considered to support classification for specific target organ toxicity following repeated exposure.

40 In particular, Section 3.9.2.7.1 of Annex I to Regulation No 1272/2008 states that reliable evidence associating repeated exposure to the substance with a consistent and identifiable toxic effect demonstrates support for the classification.

41 Section 3.9.2.7.3 of Annex I to Regulation No 1272/2008 provides an indicative list of effects considered to support classification of a substance in the hazard class STOT RE.

42 Moreover, in accordance with Section 3.9.2.9.1 of Annex I to Regulation No 1272/2008, in studies conducted in experimental animals, reliance on observation of effects alone, without reference to the duration of experimental exposure and dose/concentration, omits the fundamental concept of toxicology, namely that all substances are potentially toxic, and what determines the toxicity is a function of the dose/concentration and the duration of exposure. In most studies conducted in experimental animals the test guidelines use an upper limit dose value.

43 In that context, Section 3.9.2.9.2 of Annex I to Regulation No 1272/2008 states that, in order to help reach a decision about whether a substance is to be classified or not, and to what degree it is to be classified (Category 1 or Category 2), dose/concentration guidance values are provided for consideration of the dose/concentration which has been shown to produce significant health effects. The principal argument for proposing such guidance values is that all substances are potentially toxic and there has to be a reasonable dose/concentration above which a degree of toxic effect is acknowledged. Also, repeated-dose studies conducted in experimental animals are designed to produce toxicity at the highest dose used in order to optimise the test objective and so most studies reveal some toxic effect at least at this highest dose. What is therefore to be decided is not only what effects have been produced, but also at what dose/concentration they were produced and how relevant is that for humans.

44 Thus, in accordance with Section 3.9.2.9.3 of Annex I to Regulation No 1272/2008, in animal studies, when significant toxic effects are observed that indicate classification, consideration of the duration of experimental exposure and the dose/concentration at which those effects were seen, in relation to the suggested guidance values, can provide useful information to help assess the need to classify (since the toxic effects are a consequence of the hazardous property(ies) and also the duration of exposure and the dose/concentration).

45 Moreover, in accordance with Section 3.9.2.9.4 of Annex I to Regulation No 1272/2008, the decision to classify at all can be influenced by reference to the dose/concentration guidance values at or below which a significant toxic effect has been observed.

46 Table 3.9.3 of Annex I to Regulation No 1272/2008 thus gives guidance values to assist in Category 2 classification.

47 For inhalation tests, the guidance values depend on the form of the substance. As regards, in particular, dust, they are laid down in the last line of Table 3.9.3 of Annex I to Regulation No 1272/2008.

48 It is in the light of those considerations that it is necessary to examine whether the RAC has identified the effects considered to support the contested classification.

(b) The effects identified in the RAC Opinion which are considered to support the contested classification

49 The applicant maintains that, although the hazard class STOT RE is not based on an exhaustive list of effects considered to support classification of a substance, the RAC must specify the effects on which it bases its opinion. In the present case, however, the RAC Opinion identifies no effect which meets the criteria in Section 3.9.2.7 of Annex I to Regulation No 1272/2008.

50 The Commission, supported by ECHA, disputes those arguments.

51 In the present case, it should be noted that, in the part relating to the hazard class STOT RE, the RAC Opinion contains a section entitled ‘Assessment and comparison with the classification criteria’, which contains, on the one hand, a table summarising the relevant studies and, on the other hand, an analysis of those studies.

52 With regard to the table summarising the relevant studies, the RAC Opinion sets out the results of the repeated-dose inhalation toxicity studies with the three forms of hydrophobic SAS referred to in the classification proposal and in the open literature focusing specifically on the lungs.

53 The table summarising the relevant studies is divided into three headings containing information relating, under the first heading, to the studies and animal species used; under the second heading, to the methods and test substances used (the dose and concentration of the substance used and the test period are indicated under that heading); and, under the third heading, to the results recorded.

54 With regard to the analysis of the relevant studies, the RAC Opinion first sets out the results recorded in the studies, referring, in particular, to respiratory distress that escalated to moderate dyspnoea at the mid dose; irregular breathing; transient inflammation, especially in the alveolar region, and local injury of the lungs and, in some cases, of the mediastinal lymph nodes and, more rarely, the nose; interstitial fibrosis; histiocytosis in lung-draining mediastinal lymph nodes; fibrogenesis and structural remodelling of the lung tissue that could progress to fibrosis.

55 Secondly, the RAC Opinion concludes that ‘regarding the effects observed, some alterations in pulmonary function (breathing) are consistent among the majority of the repeated dose inhalation studies with hydrophobic SAS’. It adds ‘hydrophobic SAS induced treatment-related effects reflecting inflammation of lung tissue (main mechanism of toxicity identified), associated with a morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling)’. It found that, ‘given further remaining uncertainties on whether or not there was fibrosis in key study #2’, the proposed classification was warranted.

56 It is clear from those assessments in the RAC Opinion that the effects which were considered to support the contested classification are alterations in pulmonary function (breathing), hydrophobic SAS induced treatment-related effects reflecting inflammation of lung tissue (main mechanism of toxicity identified), associated with a morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling), and fibrosis.

57 It follows that, contrary to what the applicant maintains, the RAC identified in its opinion the effects considered to support the contested classification.

58 Although the applicant maintains that the RAC took into account, for the purposes of identifying the effects, studies which were not part of the classification proposal, it should be recalled that, in accordance with Section 3.9.2.1 of Annex I to Regulation No 1272/2008, substances are classified as hazard class STOT RE by the use of expert judgment, on the basis of the weight of ‘all evidence available’. The reference to ‘all evidence available’, without any limitation, is a sufficient basis for considering that the RAC could, without making any error, take into account all available studies, including those which were not part of the classification proposal.

59 It is therefore necessary to analyse whether the effects identified in the RAC Opinion could justify the contested classification.

(c) Justification for the contested classification on the basis of the effects identified in the RAC Opinion

60 The applicant states that the RAC Opinion is based on conclusions which do not meet the criteria laid down in Section 3.9 of Annex I to Regulation No 1272/2008. According to the applicant, the assessments concerning fibrosis are not supported by any studies. Moreover, the other effects observed in the studies do not alter lung function, in that they are an adaptive reaction to the deposition of large numbers of inhaled particles, and are not linked to any guidance value. Furthermore, their relevance to human health is contradicted by epidemiological studies.

61 The Commission, supported by ECHA, disputes those arguments.

62 In the present case, it should be noted that, as is apparent from paragraph 56 above, fibrosis is one of the effects identified in the RAC Opinion which was considered to support the contested classification.

63 Fibrosis is referred to in Section 3.9.2.7.3(e) of Annex I to Regulation No 1272/2008, in the indicative list of effects considered to support classification of a substance in the hazard class STOT RE.

64 However, it should be noted that, as the applicant maintains, fibrosis could not, as such, justify the contested classification. It follows from the RAC Opinion that the existing data were not sufficient to conclude that there was, in accordance with Section 3.9.2.7.1 of Annex I to Regulation No 1272/2008, a link between repeated exposure to silanamine and fibrosis. In that regard, it is sufficient to note that the RAC Opinion states that ‘the reported interstitial fibrosis and other serious adverse histopathological findings reported in the A6.4.3_01 study, became questionable in the light of the Weber et al. (2018) re-evaluation of the findings of the study’ and that, ‘following re-evaluation it was concluded that there was no fibrosis detected’. The RAC Opinion acknowledges in that regard that fibrogenesis is ‘the main finding in the Weber et al. (2018) … study’ and that ‘exposure-related fibrogenesis and structural remodelling of the lung tissue, which are reversible[,] … cannot be excluded as an adverse effect that could progress to fibrosis, if exposure persists and in the presence of another detrimental pathology, such as infection’. It adds that ‘interstitial fibrosis is also reported in [a one]-yr study with monkeys by Dow Corning (1972) (study #4, reviewed in Becker et al. (2013) and ECETOC, 2006) …, but [that] very few study details are available’ and that ‘the original results … are not available’. In addition, it notes that, ‘in the 13-week rat study by Wacker (1998) (study #6), reviewed in ECETOC (2006), no indication of increased birefringence (typical for interstitial fibrosis) was reported’ and that, ‘unfortunately, the original results of the Wacker (1998) study (rated as reliable guideline study by ECETOC) are not available’.

65 However, it is also clear from paragraph 56 above that the RAC Opinion is based on three other effects, namely alterations in pulmonary function (breathing), hydrophobic SAS induced treatment-related effects reflecting inflammation of lung tissue (main mechanism of toxicity identified), associated with a morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling), concerning which it states that ‘the vast majority of the effects disappeared during recovery, showing clear signs of reversibility’, that ‘only the inflammation effects could be regarded as adaptive (compensatory) changes’, but that ‘the adversity of the consequences and the clinical toxicity (i.e. impaired breathing) upon cessation of exposure is still present’.

66 The applicant does not dispute the existence of those effects.

67 However, in the first place, the applicant submits, in essence, that those effects do not justify the contested classification.

68 On the one hand, the applicant states that, in the comments addressed to the ‘Competent Authorities for REACH and CLP’ expert group (‘the CARACAL’), the industry provided evidence that the ‘effects described in the lung after inhalation of AEROSIL® R 974 were the typical adaptive reaction of the lung to the deposition of large numbers of inhaled particles’ and that the ‘deposition of large numbers of foreign material in the lung … triggers an inflammatory response intended to remove the deposited material’, that ‘this inflammatory response [could] persist for a longer time after the end of an inhalation exposure to particulate material’, ‘but [that] the inflammatory response [was] considered normal if not progressing and [was] correlated to an effective particle clearance’. In the applicant’s view, an inflammatory response will typically be observed after repeated exposure to the high dose/concentration used in repeated-exposure inhalation studies but this is not a harmful health effect and there is no observed impairment of lung function.

69 In that regard, it is sufficient to point out that, as stated in paragraph 36 above, under Section 3.9.1.1 of Annex I to Regulation No 1272/2008, all significant health effects that can impair function, reversible and irreversible, immediate and/or delayed are included in the hazard class STOT RE.

70 Accordingly, even assuming that the effects identified are an adaptive reaction of the lungs to the deposition of large numbers of inhaled particles and are therefore reversible, the RAC could, under Section 3.9.1.1 of Annex I to Regulation No 1272/2008, base the contested classification on such effects.

71 In any event, it should be noted that the RAC Opinion does acknowledge that there are hydrophobic SAS induced treatment-related effects reflecting inflammation of lung tissue. However, that opinion states that such effects are associated with others, namely morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling) and that ‘only the inflammatory effects could be regarded as adaptive (compensatory) changes’.

72 Contrary to the approach which it has adopted in relation to fibrosis, the applicant has not put forward any argument that calls into question the effects entailing a morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling).

73 It follows that, even assuming that the hydrophobic SAS induced treatment-related effects reflecting inflammation of lung tissue cannot support the contested classification, the applicant does not establish that this is also the case so far as concerns the morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling).

74 In that context, in the absence of any evidence adduced by the applicant to make the assessments used in the RAC Opinion implausible in accordance with the case-law cited in paragraph 29 above, the RAC cannot be criticised for considering that the contested classification was justified at the very least on the basis of the morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling).

75 On the other hand, the applicant criticises the RAC for failing to link the effects identified to the guidance values.

76 In that regard, it should be noted that, as is apparent from paragraph 43 above, Section 3.9.2.9.2 of Annex I to Regulation No 1272/2008 provides that the argument for proposing guidance values is that all substances are potentially toxic and there has to be a reasonable dose/concentration above which a degree of toxic effect is acknowledged. As stated in paragraphs 46 and 47 above, the guidance values for Category 2 for inhalation tests with dusts are provided for in the last line of Table 3.9.3 of that annex.

77 In the present case, it should be noted that the RAC Opinion states that ‘in the case of silanamine (SAS-HMDS), the effective dose in the various studies presented in the Table above mostly point to classification in Category 2, although in two studies (#1 and #6) Category 1 could also be supported and in study #2 the effective dose is close to the cut-off for Category 1’.

78 It should also be noted that the table summarising the relevant studies, first, indicates, as is apparent from paragraph 53 above, the dose and concentration of the substance used and the test period and, secondly, sets out the results and how they meet the criteria for classification as hazard class STOT RE Categories 1 and 2.

79 It follows from those assessments that the RAC compared the doses/concentrations used in the various studies with the guidance values referred to in Table 3.9.2, relating to Category 1, and in Table 3.9.3, relating to Category 2, of Annex I to Regulation No 1272/2008, in order to conclude, in essence, that the doses/concentrations used in most of the studies corresponded to the guidance values for classification in Category 2.

80 In addition, it should be noted that, where the criteria for classification as hazard class STOT RE Categories 1 and 2 which are set out in the table summarising the relevant studies are met, an accompanying footnote states that Haber’s rule for inhalation applies. In accordance with Section 3.9.2.9.5 of Annex I to Regulation No 1272/2008, that rule requires that, in toxicity studies of a duration other than the standard 90 days, the guidance values are to be extrapolated, taking into account the fact that the effective dose is, essentially, directly proportional to the exposure concentration and the duration of exposure.

81 It follows that, contrary to what the applicant claims, the RAC, for the purposes of the contested classification, took into account the dose/concentration and duration of exposure in relation to which the effects were observed.

82 As regards the applicant’s argument that the RAC erred in consistently referring in its opinion to guidance values as being ‘classification criteria’, it is sufficient to state in that regard that, in addition to the fact that such an argument confirms that guidance values were taken into account, those values may influence the decision whether or not to classify a substance, as is apparent from paragraphs 44 and 45 above. Accordingly, the RAC did not err in referring in its opinion to guidance values for the purposes of the contested classification.

83 In the second place, the applicant states that the relevance of the effects identified on human health is contradicted by certain epidemiological studies, namely the Taeger et al. (2002), Morfeld et al. (2014), Morfeld et al. (2016) and Mei Young et al. (2022) studies, which the RAC failed to take into consideration.

84 In that regard, it should be noted that, in accordance with Section 3.9.2.4 of Annex I to Regulation No 1272/2008, the weight of evidence of all data (Section 1.1.1 of Annex I to Regulation No 1272/2008), including human incidents, epidemiology, and studies conducted in experimental animals, is used to substantiate specific target organ toxic effects that merit classification.

85 Moreover, under Section 1.1.1.4 of Annex I to Regulation No 1272/2008, relating to the weight of evidence determination, where evidence is available from both humans and animals and there is a conflict between the findings, the quality and reliability of the evidence from both sources are to be evaluated in order to resolve the question of classification. Generally, adequate, reliable and representative data on humans (including epidemiological studies, scientifically valid case studies as specified in the abovementioned annex or statistically backed experience) are to have precedence over other data. However, even well-designed and conducted epidemiological studies may lack a sufficient number of subjects to detect relatively rare but still significant effects, to assess potentially confounding factors. Therefore, positive results from well-conducted animal studies are not necessarily negated by the lack of positive human experience but require an assessment of the robustness, quality and statistical power of both the human and animal data.

86 In the present case, as noted by the Commission, and the applicant does not dispute this, the reference document relating to the RAC Opinion states, inter alia, that ‘even if there is no long-term respiratory health effect in the available epidemiological study in workers (section 10.9_Table 50), uncertainties are present in this publication (nature of the silica, duration and level of the exposure) leading to inadequate evidence’ and that ‘in this context, the epidemiological study cannot be used as a proof of no effect and cannot rule out the pulmonary effect reported in rats’.

87 In accordance with the case-law cited in paragraph 28 above, the Commission must be allowed a broad discretion with regard to the evaluation of the scientific studies available and as regards the choice of which studies must take precedence over others, irrespective of their chronology.

88 Against that background, it should be noted that, in the context of the weight of evidence evaluation, the RAC took into account, in accordance with Section 1.1.1.4 of Annex I to Regulation No 1272/2008, the data on humans which it regarded as adequate but considered that they did not have sufficient weight of evidence to call into question the results of the animal studies.

89 It follows from the foregoing considerations that none of the applicant’s arguments is such as to establish that the RAC Opinion is based on conclusions which do not meet the criteria set out in Section 3.9 of Annex I to Regulation No 1272/2008.

90 As to the remainder, while the applicant maintains that the RAC has failed to demonstrate to what extent the effects observed constituted adverse health effects within the meaning of Section 3.9 of Annex I to Regulation No 1272/2008, it should be noted that a reading of the RAC Opinion – in particular under the heading relating to the results recorded in the various studies in the table summarising those studies – and of the analysis of those studies as a whole makes it possible to understand the reasons why the RAC considered that those effects constituted effects capable of justifying the contested classification, in accordance with the criteria laid down in Section 3.9 of Annex I to Regulation No 1272/2008.

91 It is apparent from paragraph 78 above that the heading relating to the results recorded in the various studies in the table summarising those studies sets out the results and how they meet the criteria for classification as hazard class STOT RE Categories 1 and 2. Moreover, as is apparent from paragraph 77 above, the RAC Opinion states, in the context of the analysis of the relevant studies, that ‘in the case of silanamine (SAS-HMDS), the effective dose in the various studies presented in the Table above mostly point[s] to classification in Category 2, although in two studies (#1 and #6) Category 1 could also be supported and in study #2 the effective dose is close to the cut-off for Category 1’. It is clear from a combined reading of those paragraphs that the RAC, having identified the effects in the relevant studies, checked whether they had been observed at guidance values corresponding to Tables 3.9.2 and 3.9.3 of Annex I to Regulation No 1272/2008. As is apparent from paragraph 45 above, in accordance with Section 3.9.2.9.4 of that annex, the decision to classify at all can be influenced by reference to the dose/concentration guidance values at or below which a significant toxic effect has been observed.

92 It follows from all of the foregoing that the applicant is wrong to claim that the RAC Opinion is based on conclusions which do not meet the criteria set out in Section 3.9 of Annex I to Regulation No 1272/2008.

93 The third part of the first plea must therefore be rejected as unfounded.

2. The first part, alleging that silanamine produces no specific toxic effects on the lungs

94 In the context of the first part of the first plea, the applicant submits that, for the purposes of classifying a substance in the hazard class STOT RE, the substance must produce specific toxic effects, namely effects which result from its intrinsic properties. In the applicant’s view, that interpretation is confirmed by the 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). In the present case, however, silanamine does not produce specific toxic effects, in that the effects observed result not from its intrinsic properties but from the deposition of large numbers of inhaled particles in the context of the relevant studies and from the high dose/concentration administered in those studies.

95 The Commission, supported by ECHA, disputes those arguments.

96 As a preliminary point, it should be recalled that, as is apparent from paragraph 36 above, in accordance with Section 3.9.1.1 of Annex I to Regulation No 1272/2008, the hazard class STOT RE means specific toxic effects on target organs occurring after repeated exposure to a substance or mixture.

97 With regard to the concept of ‘specific toxic effects’, it should be noted that, although Regulation No 1272/2008 does not define that concept, it must be interpreted as referring to toxic effects specifically affecting a target organ as a result of repeated exposure to the substance in question.

98 On the one hand, it is clear from the last sentence of Section 3.9.1.1 of Annex I to Regulation No 1272/2008 that the specific toxic effects that are addressed in Sections 3.1 to 3.8 and 3.10 of that annex are not included in Section 3.9.

99 In addition, under Section 3.9.1.6 of Annex I to Regulation No 1272/2008, non-lethal toxic effects observed after a single-event exposure are classified as described in Section 3.8, entitled ‘Specific target organ toxicity – single exposure’, and are therefore excluded from Section 3.9 of that annex.

100 It follows that each section of Annex I to Regulation No 1272/2008 addresses the specific toxic effects associated with the type of hazard.

101 On the other hand, as is apparent from paragraph 37 above, in accordance with Section 3.9.1.2 of Annex I to Regulation No 1272/2008, the hazard class STOT RE identifies substances or mixtures which are a specific target organ toxicant and, as such, may present a potential for adverse health effects in people who are exposed to them.

102 Moreover, according to the note under Table 3.9.1 of Annex I to Regulation No 1272/2008, attempts are to be made to determine the primary target organ of toxicity and classify for that purpose, as hepatotoxicants, neurotoxicants, etc. It is necessary to carefully evaluate the data and, where possible, not include secondary effects (a hepatotoxicant can produce secondary effects in the nervous or gastro-intestinal systems).

103 It follows from the foregoing that the purpose of the hazard class STOT RE is to classify substances where, following repeated exposure, a target organ is affected by their toxicity.

104 In that context, it should be noted that, as the Commission argues, the term ‘specific’ in the concept of ‘specific toxic effects’ provided for in Section 3.9.1.1 of Annex I to Regulation No 1272/2008 refers to the target organ.

105 On the other hand, contrary to what is argued by the applicant, there is nothing in Regulation No 1272/2008 to suggest that the toxic effects produced by a substance must be specific to that substance.

106 As stated in paragraph 103 above, a substance must be classified as hazard class STOT RE if, following repeated exposure, a target organ is affected by its toxicity, irrespective of whether or not the toxic effects which that substance produces are specific to it.

107 Moreover, the analogy drawn by the applicant with the hazard classes for reproductive toxicity and carcinogenicity is not such as to call that assessment into question.

108 With regard to reproductive toxicity, the various provisions of Section 3.7 of Annex I to Regulation No 1272/2008 support the interpretation of the concept of ‘specific toxic effects’ given in paragraph 97 above.

109 In that regard, it is sufficient to note that Section 3.7.2.2.1 of Annex I to Regulation No 1272/2008 states that 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 are not to be so classified if such an effect is produced solely as a non-specific secondary consequence of other toxic effects.

110 The same applies to Section 3.7.2.3.4 of Annex I to Regulation No 1272/2008, which states that data from animal studies ideally must provide clear evidence of specific reproductive toxicity in the absence of other systemic toxic effects.

111 With regard to carcinogenicity, it should be noted that, while it is clear from Section 3.6.2.2.1 of Annex I to Regulation No 1272/2008 that classification as a carcinogen is intended to be used for substances which have an intrinsic property to cause cancer, it does not provide any information on the concept of ‘specific toxic effects’.

112 Nor is such an interpretation apparent from the 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).

113 In paragraphs 138 to 141 of the 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), the Court interpreted the concept of ‘intrinsic properties’, which relates to the cause of the hazard, but did not rule on the concept of ‘specific toxic effects’, which refers to the result of the hazard.

114 In the present case, it should be noted that, as is apparent from paragraph 9 above, the contested regulation included silanamine in Part 3 of Annex VI to Regulation No 1272/2008 in the hazard class STOT RE Category 2 with hazard statement code ‘H373’ (lungs, inhalation).

115 That listing was recommended by the RAC on the basis of several studies which identified silanamine as a specific toxicant affecting the lungs. In particular, the RAC noted, in summarising the classification proposal, that, in the preliminary 14-day study (A6.3.3) with the substance SAS-DDS (Aerosil R974), the target organ was clearly the lung, since at all doses respiratory distress, dyspnoea and histological changes to the lung related to alveolar inflammation were observed. It added that in the subchronic 90-day study (A6.4.3) with the substance SAS-DDS (Aerosil R974), the target organ was also the lung. It noted that the French competent authority had declared that the lung was the primary target organ after exposure to the substance SAS-DDS (Aerosil R974). Against that background, the RAC analysed the relevant repeated dose toxicity studies of hydrophobic SAS, from the classification proposal and the open literature, focusing mainly on the effects on lungs, and concluded that, on the basis of the weight of the evidence of all the available data, it was appropriate to endorse the proposal for classification of silanamine as hazard class STOT RE Category 2 with hazard statement code ‘H373’ (lungs, inhalation).

116 The RAC was therefore right to identify, in accordance with Section 3.9.1.1 of Annex I to Regulation No 1272/2008, toxic effects caused by silanamine which specifically affect the lungs because of repeated exposure.

117 However, it should be borne in mind that harmonised classification and labelling under Regulation No 1272/2008 concern the transmission of information on the hazards linked to the substances’ intrinsic properties (see 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 136 and the case-law cited).

118 The concept of ‘intrinsic properties’, while it does not appear in Regulation No 1272/2008, must be interpreted in its literal sense as referring to the properties which a substance has in and of itself (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).

119 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 15 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).

120 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 cited in paragraph 19 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 (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).

121 It is in the light of those considerations that it is appropriate to examine the applicant’s arguments that, in essence, the effects observed result not from the intrinsic qualities of silanamine but from the deposition of large numbers of inhaled particles in the context of the relevant studies and from the high dose/concentration administered in those studies.

(a) The applicant’s arguments alleging that the effects observed result from the deposition of large numbers of inhaled particles in the context of the relevant studies

122 The applicant argues that the effects observed are an adaptive reaction of the lungs to the deposition of large numbers of inhaled particles resulting from the dusty form of silanamine administered in the relevant studies and are common to all particulate substances. It maintains that, in the context of the public consultation, the industry argued that ‘the effects observed with AEROSIL® R 974 represent markers of typical inflammatory responses of the rat lung after continued high exposures to particles’. It adds that, in the comments submitted to the CARACAL, the industry stated, in a discussion paper entitled ‘Thought Starter. New mechanistic study to provide more comprehensive data and a more accurate assessment for particulate substances for regulatory purposes’, that ‘all respirable particles without exception will show unspecific inflammatory effects in the lung at the end of exposure at concentrations up to 200 mg/m 3 ’. It notes that, in that context, the industry also explained that ‘there [was] an unspecific particle-related inflammation which [was] not associated with irreversible clinical effects from inhalation exposure to surface-treated SAS’, that ‘all effects described in the lung after inhalation of AEROSIL® R 974 were the typical adaptive reaction of the lung to the deposition of large numbers of inhaled particles and thus [were] not a reaction to specific chemical properties of the material’, that ‘the deposition of large numbers of foreign material in the lung, in general, trigger[ed] an inflammatory response intended to remove the deposited material’, that ‘this inflammatory response [could] persist for a longer time after the end of an inhalation exposure to particulate material’, ‘but [that] the inflammatory response [was] considered normal if not progressing and [was] correlated to an effective particle clearance’. However, neither the RAC nor the Commission took that evidence into account for the purposes of the contested classification.

123 The applicant notes that, in the light of those considerations, in the 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), the Court held, in relation to the classification of titanium dioxide, that an accumulation of particles in the lung in sufficient quantities to bring about a significant impairment of particle clearance mechanisms, which could be verified only when certain quantities of particles were inhaled, could not be regarded as forming part of the intrinsic properties of the particles at issue, that the quantity of inhaled particles was one of the key elements of the toxicity observed, even if it is accepted that the properties of particles, such as their size, form and poor solubility, may have played a role in their accumulation in the lung, and that the toxicity observed was not exclusive to titanium dioxide particles but was common to other poorly soluble low-toxicity particles. The applicant submits that that judgment should be applied by analogy to the present case.

124 In the present case, it should first be noted that the RAC classified silanamine as hazard class STOT RE Category 2, in particular on the basis of the following inhalation studies: A6.3.3, Degussa (1968), A6.4.3_01; Reuzel et al. (1991), Degussa (1987), Weber et al. (2018), ECETOC (2006), Becker et al. (2013), Dow Corning (1972), Wacker (1998) and Degussa (1962).

125 It should next be noted that, as is apparent from paragraphs 54 and 55 above, after setting out the results recorded in the various studies, the RAC Opinion concludes, in essence, that the effects which were considered to support the contested classification are alterations in pulmonary function (breathing), hydrophobic SAS induced treatment-related effects reflecting inflammation of lung tissue (main mechanism of toxicity identified), associated with a morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling), and fibrosis.

126 Lastly, it should be recalled that, as stated in paragraph 71 above, the RAC Opinion acknowledges that there are hydrophobic SAS induced treatment-related effects reflecting inflammation of lung tissue, but explains that such effects are linked to others, namely the morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling) and that ‘only the inflammatory effects could be regarded as adaptive (compensatory) changes’.

127 It must be stated that the applicant fails to establish that the RAC and, consequently, the Commission committed a manifest error of assessment in taking into account, on the basis of the studies referred to in paragraph 124 above, the morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling).

128 In its argument, the applicant merely claims that, in the course of the procedure, the industry argued that the effects observed in the studies resulted from the deposition of particles inhaled into the lungs which caused an inflammatory response, but it fails to show that the assessments contained in the RAC Opinion and accepted by the Commission in the contested regulation concerning the morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling) cannot support the contested classification in that they are also inflammatory effects resulting from the deposition of large numbers of inhaled particles in the context of the studies taken into consideration.

129 In the absence of evidence to render implausible the factual assessments used in the RAC Opinion and accepted by the Commission in the contested regulation in that regard, in accordance with the case-law cited in paragraph 29 above it is not the Court’s role to substitute its assessment of complex facts for that made by the RAC and accepted by the Commission.

130 Nor can the applicant validly maintain that neither the RAC nor the Commission took account of the industry’s arguments relied on in that regard in the course of the procedure.

131 In that respect, it should first of all be noted that, during the public consultation, the industry argued that the French competent authority had justified its classification proposal based on ‘effects reported for a 90-day rat inhalation toxicity study with AEROSIL® R 974, which was carried out by the contract laboratory TNO (TNO, 1987)’, and that ‘the effects observed with AEROSIL® R 974 represent[ed] markers of typical inflammatory responses of the rat lung after continued high exposures to particles’.

132 It follows that, in the context of the public consultation, the industry argued, in essence, that the effects observed in the study taken into consideration by the French competent authority represented markers of typical inflammatory responses of the rat lung after continued high exposures.

133 It should next be noted that it is apparent from Annex 2 to the RAC Opinion that, in response to comment No 1 made during the public consultation, the RAC found as follows:

‘[The] RAC reviewed various inhalation studies with silanamine, both acute and chronic. In all studies a prominent and consistent clinical symptom observed in animals was respiratory distress. Histopathological findings supported local inflammation, as well as congestion and oedema. Indications of tissue remodelling (increase in collagen content) and tissue injury (increased LDH, NAG activity) were also available, despite the fact that the main histopathological finding of fibrosis was reviewed (Weber et al., 2018) and downgraded to fibrogenesis. These adverse findings, although reversible and adaptive (for inflammation), explain and account for the breathing difficulties, which may persist after the end of exposure. The response to silanamine repeated exposure via inhalation appears to be quite substance specific. All in all, the respiratory distress varying from difficulties in breathing to slight dyspnoea and shortness of breath is attributed to silanamine exposure via inhalation. Clearly symptoms are observed both after acute and repeated exposure, with considerably lower doses after repeated exposure. … In addition, SAS is indeed cleared from the lung through the lymph nodes, although not that rapidly and when cleared, inflammation progresses from the lung to the mediastinal lymph nodes.’

134 In response to comment No 2 made in the context of the same public consultation, the RAC also stated as follows:

‘Despite the fact that the histopathological effects induced in the lung by hydrophobic SAS-inhalation are reversible, and could be regarded as adaptive, at least for the part that included inflammation, they cannot be considered not adverse, as a persistent and consistent clinical symptom of respiratory distress is observed both in acute and repeated dose inhalation studies at various doses. Breathing disorders are not expected to disappear immediately after cessation of exposure and in the context of another pathological condition, such as an infection, could prove to be detrimental to health. Therefore, they cannot be ignored.’

135 It is clear from those assessments that the RAC examined the industry’s comments that the effects observed in the study taken into consideration by the French competent authority represented markers of typical inflammatory responses of the rat lung after continued high exposures, but considered, in essence, on the basis of the various inhalation studies with silanamine, both acute and chronic, that, while there might be adaptive effects related to inflammation, a persistent and consistent clinical symptom of respiratory distress was observed in both acute and repeated inhalation studies at various doses, which was not adaptive.

136 This is also clear from the RAC Opinion itself, which indicates, as stated in paragraph 62 above, that there are hydrophobic SAS induced treatment-related effects reflecting inflammation of lung tissue (main mechanism of toxicity identified), associated with a morphological tissue reaction (hypertrophy, lung injury, partial hyperplasia of the bronchiolar epithelium, collagen remodelling), concerning which it is explained that ‘the vast majority of the effects disappeared during recovery, showing clear signs of reversibility’, that ‘only the inflammation effects could be regarded as adaptive (compensatory) changes’, but that ‘the adversity of the consequences and the clinical toxicity (i.e. impaired breathing) upon cessation of exposure is still present’.

137 Lastly, it should be noted that it is apparent from recital 5 of the contested regulation that the Commission considered that the classification of silanamine as hazard class STOT RE, recommended in the RAC Opinion, should be included in Annex VI to Regulation No 1272/2008, since no new information that would have required further assessment for that classification had been received.

138 It follows that the Commission also took into account the industry’s arguments, but considered, in essence, that they were not supported by any recent scientific development which would have called into question the soundness of the conclusions set out in the RAC Opinion.

139 In that regard, it should be noted that the document ‘Thought Starter. New mechanistic study to provide more comprehensive data and a more accurate assessment for particulate substances for regulatory purposes’ provided by the industry in the comments submitted to the CARACAL is, as its name indicates, only an initial reflection on the ongoing mechanistic study.

140 In any event, according to the document ‘Thought Starter. New mechanistic study to provide more comprehensive data and a more accurate assessment for particulate substances for regulatory purposes’, the mechanistic study aims ‘to clarify acute inhalation toxicity for all particulate substances’. In addition, the Organisation for Economic Co-operation and Development (OECD) Guidelines Nos 403 and 436 referred to in that document also relate to acute toxicity studies. While that document states that a comparable situation arises for classification in the hazard class STOT RE, the only argument it contains in that regard merely notes that ‘all respirable particles without exception will show unspecific inflammatory effects in the lung at the end of exposure at concentrations up to 200 mg/m 3 ’. The results of the study provided by the industry in October 2021 confirm, moreover, that ‘the research study was based on OECD 436 test design with additional endpoints’.

141 According to Section 3.1.1.1 of Annex I to Regulation No 1272/2008, acute toxicity means serious adverse health effects (that is to say, lethality) occurring, inter alia, after a single or short-term inhalation exposure to a substance or mixture.

142 Thus, in contrast to the hazard class STOT RE, the classification of a substance in the hazard class acute toxicity is based on a single or short-term exposure and relates to lethality.

143 In addition, unlike with the hazard class STOT RE, Section 3.1.2.3 of Annex I to Regulation No 1272/2008 sets out specific considerations for classification of substances as acutely toxic by the inhalation route.

144 In particular, Section 3.1.2.3.2 of Annex I to Regulation No 1272/2008 states that of particular importance in classifying for inhalation toxicity is the use of well-articulated values in the highest hazard categories for dusts and mists. Inhaled particles between 1 and 4 microns mean mass aerodynamic diameter will deposit in all regions of the rat respiratory tract. This particle size range corresponds to a maximum dose of about 2 mg/l. In order to achieve applicability of animal experiments to human exposure, dusts and mists would ideally be tested in this range in rats.

145 The analogy drawn by the applicant with the case which gave rise to the 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), is unfounded.

146 First of all, the hazard at issue in the case giving rise to the 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), was carcinogenicity Category 2, by the inhalation route, whereas, in the present case, it is the hazard class STOT RE which is at issue. Consequently, the classification criteria laid down in Section 3.6 of Annex I to Regulation No 1272/2008 and in Section 3.9 of that annex are not the same. Moreover, the substance, the reasoning set out in the RAC Opinion and the scientific studies taken into consideration are not the same. In particular, in the context of the hazard class STOT RE, the intrinsic properties of the substance must be taken into account in the light of the guidance values concerned. As is apparent from paragraph 42 above, Section 3.9.2.9.1 of Annex I to Regulation No 1272/2008 provides that, in studies conducted in experimental animals, reliance on observation of effects alone, without reference to the duration of experimental exposure and dose/concentration, omits the fundamental concept of toxicology, that is to say, all substances are potentially toxic, and what determines the toxicity is a function of the dose/concentration and the duration of exposure. It is also clear, in essence, from paragraphs 43 to 45 above that, in the framework relating to the hazard class STOT RE, the toxic effects are a consequence of the hazardous property(ies) and also the duration of exposure and the dose/concentration and that guidance values are proposed in that regard which may influence the decision whether or not to classify. Furthermore, as stated in paragraphs 46 and 47 above, within that framework, guidance values are expressly provided for inhalation tests with dusts, which are referred to, as regards Category 2, in the last line of Table 3.9.3 of Annex I to Regulation No 1272/2008.

147 Next, it should be noted that, in the case which gave rise to the 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), it followed from the actual wording of the RAC Opinion and the regulation which was contested that the carcinogenic hazard at issue was ‘non-intrinsic in a classical sense’ or ‘particular’ in nature. The RAC Opinion expressly described the carcinogenic hazard in question as ‘non-intrinsic in a classical sense’ and the contested regulation contained a ‘Note W’ from which it followed that the Commission had found it necessary to accompany the contested classification and labelling with a description of the ‘particular toxicity of the substance’ (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 144).

148 Lastly, the ‘non-intrinsic in a classical sense’ or ‘particular’ nature of the carcinogenic hazard resulted from several elements referred to in the RAC Opinion and in the regulation which was contested. In particular, the carcinogenicity hazard referred to in the contested classification and labelling was linked solely to certain respirable titanium dioxide particles, when they were present in a certain form, physical state, size and quantity. Moreover, it occurred only in lung overload conditions, that is to say, when large quantities of particles were inhaled, giving rise to a significant impairment of particle clearance mechanisms in the lung. The contested regulation thus expressly provided that carcinogenicity was associated with the inhalation of respirable titanium dioxide particles and with the retention and poor solubility of those particles in the lungs and contained a ‘Note W’ stating that carcinogenicity ‘ar[ose] when respirable dust [was] inhaled in quantities leading to significant impairment of particle clearance mechanisms in the lung’. Furthermore, the contested regulation stated that the carcinogenicity hazard covered by the contested classification and labelling corresponded, according to the actual wording of the RAC Opinion, to ‘particle toxicity’, the reason for which was ‘the deposited particles, but not solutes of [titanium dioxide] molecules’ (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 145 to 153).

149 It was in that specific context that the Court held that an accumulation of particles in the lung in sufficient quantities to bring about a significant impairment of particle clearance mechanisms, which could be verified only when certain quantities of particles were inhaled, could not be regarded as forming part of the intrinsic properties of the particles at issue (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 158), with the result that that assessment must be read in the light of the specific circumstances of the case which gave rise to the 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).

150 However, there are no such specific circumstances in the present case. In particular, the classification of silanamine is not linked to a specific particle size as was the case with titanium dioxide, for which the size was specified in the contested classification and labelling, in the expression ‘in powder form containing 1% or more of particles with aerodynamic diameter ≤ 10 μm’ (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 9). Moreover, there is no assessment or quantified data in the file to suggest that the effects identified were observed in lung overload conditions. Consequently, the findings of the Court in paragraph 158 of the 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), cannot be applied in the present case.

151 It follows from the foregoing considerations that all of the applicant’s arguments alleging that the effects observed result from the deposition of large numbers of inhaled particles in the context of the relevant studies must be rejected.

(b) The applicant’s arguments alleging that the effects observed result from the high dose/concentration administered in the relevant studies

152 The applicant maintains that the effects observed are the result of the ‘extremely high’ doses administered in the studies used to support the contested classification. It states that, in the course of the procedure, the industry identified issues related to the use of high doses/concentrations as well as the influence of particle sizes in existing toxicity testing guidelines for particulate substances. It claims that, in the comments submitted to the CARACAL, the industry identified, in the discussion paper entitled ‘Thought Starter. New mechanistic study to provide more comprehensive data and a more accurate assessment for particulate substances for regulatory purposes’, issues related to the use of high doses/concentrations as well as to the influence of particle sizes in existing guidelines, noting that ‘many confounding factors influence[d] the particle size distribution of particles reaching the test organism under the defined conditions of inhalation toxicity testing required by OECD guidelines’. It adds that, in that context, the industry highlighted the fact that a ‘situation ar[ose] considering the guidance values for [the hazard class] STOT classification (repeated dose inhalation toxicity)’, where ‘all respirable particles without exception … show unspecific inflammatory effects in the lung at the end of exposure at concentrations up to 200 mg/m 3 ’. It states that those findings are ‘commonly shared in the scientific community’, referring to a comparison of available toxicological reports of repeated exposure inhalation studies on various powders which is ‘currently [ongoing] and [the] results [of which were to] be available after the date of submission of the present application’. However, neither the RAC nor the Commission took that evidence into account for the purposes of the contested classification.

153 Moreover, the applicant notes that those effects do not result from the substance in the form or physical state in which it is placed on the market or in which it is reasonably possible to expect that it will be used.

154 As a preliminary point, as recalled in paragraph 43 above, under Section 3.9.2.9.2 of Annex I to Regulation No 1272/2008, for the purposes of the classification of a substance as hazard class STOT RE, the argument for proposing such guidance values is that all substances are potentially toxic and there has to be a reasonable dose/concentration above which a degree of toxic effect is acknowledged.

155 It is also apparent from Section 3.9.2.9.2 of Annex I to Regulation No 1272/2008 that what is to be decided is not only what effects have been produced, but also at what dose/concentration they were produced and how relevant that is for humans, since repeated-dose studies conducted in experimental animals are designed to produce toxicity at the highest dose used in order to optimise the test objective and so most studies will reveal some toxic effect at least at this highest dose.

156 It follows from those considerations that, in the framework relating to the hazard class STOT RE, guidance values are provided because repeated-dose studies conducted in experimental animals are designed to produce toxicity at the highest dose used in order to optimise the test objective and there has to be a reasonable dose/concentration above which a degree of toxic effect is acknowledged.

157 In the present case, on the one hand, it should be noted that, as is apparent from paragraph 79 above, the RAC compared the doses/concentrations used in the various studies with the guidance values referred to in Tables 3.9.2 and 3.9.3 of Annex I to Regulation No 1272/2008 in order to conclude, in essence, that the doses/concentrations used in most of the studies corresponded to the guidance values for classification in Category 2.

158 Contrary to what the applicant maintains, it is not apparent from the file that the doses administered in the studies used to support the contested classification are ‘extremely high’. In fact, it is clear from the table summarising the relevant studies, which sets out, under the third heading, the results of those studies and how they meet the classification criteria, that the doses used in the various studies corresponded to the guidance values referred to in Tables 3.9.2 and 3.9.3 of Annex I to Regulation No 1272/2008.

159 On the other hand, as recalled in paragraphs 80 and 81 above, the RAC also applied Haber’s rule for inhalation, which demonstrates that it also considered the duration of exposure at which the effects were observed in the studies taken into consideration.

160 It follows from those considerations that the RAC took into account the guidance values provided for by Regulation No 1272/2008, when making the contested classification.

161 It must be stated that the applicant does not establish that the RAC and, consequently, the Commission committed a manifest error of assessment in that regard.

162 In its argument, the applicant merely claims that, in the course of the procedure, the industry identified issues related to the use of high doses/concentrations as well as the influence of particle sizes in existing toxicity testing guidelines for particulate substances, but does not demonstrate that the assessments set out in the RAC Opinion and reproduced by the Commission in the contested regulation cannot support the contested classification.

163 In any event, on the one hand, it should be noted that it is clear from paragraphs 154 to 156 above that, in the framework relating to the hazard class STOT RE, Regulation No 1272/2008 takes account of the fact that repeated-dose studies conducted in experimental animals are designed to produce toxicity at the highest dose used in order to optimise the test objective and that guidance values are specifically provided to establish a reasonable dose/concentration above which a degree of toxic effect is acknowledged.

164 On the other hand, the evidence put forward by the industry in that regard in the course of the procedure cannot call that finding into question.

165 With regard to the document ‘Thought Starter. New mechanistic study to provide more comprehensive data and a more accurate assessment for particulate substances for regulatory purposes’, in accordance with paragraphs 139 to 144 above, it is not relevant to the hazard class STOT RE.

166 As regards the comparison of available toxicological reports of repeated exposure inhalation studies on various powders, it should be noted that, as stated by the applicant itself, that comparison was not available on the date on which the contested regulation was adopted.

167 According to case-law, in an action for annulment brought under Article 263 TFEU, the lawfulness of EU measures must be assessed on the basis of the facts and the law as they stood at the time when those measures were adopted (see judgment of 21 May 2015, Rubinum v Commission , T‑201/13, not published, EU:T:2015:311, paragraph 84 and the case-law cited).

168 In any event, the applicant’s argument based on the comparison referred to in paragraph 166 above in no way detracts from the RAC’s assessment, which relates solely to the effects of silanamine and not to the effects of other substances.

169 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.

170 In that context, the RAC cannot be criticised for taking into account effects resulting from the dose/concentration administered in the relevant studies.

171 As regards the applicant’s arguments that the effects do not result from the substance in the form or physical state in which it is placed on the market or in which it is possible reasonably to expect it to be used, 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 19 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.

172 It follows from the foregoing considerations that none of the applicant’s arguments alleging that the effects observed result from the high dose/concentration administered in the relevant studies can succeed.

173 The first part of the first plea must therefore be rejected as unfounded.

3. The second part, alleging that the effects observed are based on a read-across that does not meet the criteria for classification in the hazard class STOT RE

174 In the context of the second part of the first plea, the applicant submits that, in relying on studies relating to the substance SAS-DDS, the RAC Opinion did not comply with the criteria laid down in Section 3.9.2.10.3 of Annex I to Regulation No 1272/2008, relating to read-across. On the one hand, that substance has not been subject to harmonised classification and, on the other hand, the RAC failed to demonstrate that a validated structure activity relationship between that substance and silanamine and an expert judgment-based extrapolation existed.

175 The Commission, supported by ECHA, disputes those arguments.

176 In that regard, it should be recalled that, in accordance with Section 3.9.2.10.3 of Annex I to Regulation No 1272/2008, a substance that has not been tested for specific target organ toxicity may, where appropriate, be classified on the basis of data from a validated structure activity relationship and expert judgment-based extrapolation from a structural analogue that has previously been classified together with substantial support from consideration of other important factors such as formation of common significant metabolites.

177 It is apparent from Section 3.9.2.10.3 of Annex I to Regulation No 1272/2008 that the use of read-across in the context of the STOT RE hazard assessment is subject to several conditions: first, that a substance has not been tested for specific target organ toxicity; secondly, that there is a structural analogue that has previously been classified and data from a validated structure activity relationship, as well as an expert judgment-based extrapolation, and, thirdly, that the extrapolation of the data has substantial support from consideration of other important factors.

178 In the present case, it should be noted that it is clear from the table summarising the relevant studies that the RAC took into consideration studies using the substance SAS-DDS, namely A6.3.3, Degussa (1968), A6.4.3_01, Reuzel et al. (1991), Degussa (1987), Weber et al. (2018), Becker et al. (2013), Degussa (1962), A6.3.1, Degussa (1964), A6.4.1, A6.5 and Degussa (1969), and studies using the substance SAS-HMDS, such as ECETOC (2006), Becker et al. (2013), Dow Corning (1972) and Wacker (1998).

179 It should next be observed that, in the section entitled ‘Read-across between the different types of amorphous silica’, the RAC stated that silanamine was the result of the reaction of synthetic amorphous silica treated with hexamethylsilazane (HMDS) and that the surface modification of the hydrophilic silica with dichlorodimethylsilane resulted in a dimethylsilyl-surface modified silica, abbreviated SAS-DDS.

180 Moreover, the RAC noted that, since SAS-DDS was structurally similar to silanamine and had the same physical, chemical and toxicological properties as silanamine, it was used as a source substance in the classification proposal and in its opinion. It added that several characteristics, such as chemical composition, particle size and shape, surface chemistry, surface area, solubility and rate of dissolution, hydrophobicity, zeta potential, dispersibility and dustiness, support the use of SAS-DDS as a source substance. It noted that a similar grouping approach had been widely accepted and used in the open literature for both human health and environmental hazards. It concluded that SAS-HMDS and SAS-DDS were sufficiently similar surface-modified SAS for a read-across.

181 Finally, it should be pointed out that, in its conclusion on the hazard class STOT RE, the RAC stated that, on the basis of the weight of the evidence of all of the available data, it supported the proposal to classify silanamine in that hazard class.

182 In the first place, it should be noted that the RAC Opinion took into account both the results of studies using SAS-DDS and the results of studies using SAS-HDMS and that the contested classification is based on the weight of the evidence of all of the available data.

183 In accordance with Section 1.1.1.3 of Annex I to Regulation No 1272/2008, a weight of evidence determination means that all available information bearing on the determination of hazard is considered together, in particular information from application of the category approach (grouping, read-across).

184 Accordingly, the results of the studies using SAS-DDS were taken into account in the RAC Opinion, as part of the determination of the weight of the evidence of all of the available data.

185 In the second place, it should be noted that the RAC took into account studies using SAS-DDS, in finding that that substance was sufficiently similar to silanamine.

186 Although the applicant does not dispute that fact, it maintains, first, that SAS-DDS has not been subject to harmonised classification, with the result that it could not be considered to be ‘a structural analogue that has previously been classified’ within the meaning of Section 3.9.2.10.3 of Annex I to Regulation No 1272/2008.

187 In that regard, it should be noted that it is clear from recital 16 of Regulation No 1272/2008 that responsibility for the identification of hazards of substances and mixtures and for deciding on their classification should mainly lie with manufacturers, importers and downstream users of those substances or mixtures.

188 It should also be borne in mind that the hazard class STOT RE is not provided for in Article 36(1) of Regulation No 1272/2008, which contains the list of hazards in respect of which substances are normally subject to classification and labelling.

189 It should also be noted that, pursuant to Article 9(1) of Regulation No 1272/2008, manufacturers, importers and downstream users are to evaluate the information available on substances by applying to it the criteria for classification for each hazard class in Parts 2 to 5 of Annex I to that regulation, so as to ascertain the hazards associated with the substance.

190 It follows that, with regard to the hazard class STOT RE, substances are normally subject to self-classification, which is deemed to meet the criteria for classification in that hazard class.

191 In so far as, for the purposes of the weight of evidence determination, all available information bearing on the determination of hazard must be considered and substances meeting the criteria for the hazard class STOT RE are not normally subject to harmonised classification, the expression ‘structural analogue that has previously been classified’, within the meaning of Section 3.9.2.10.3 of Annex I to Regulation No 1272/2008, must be interpreted as covering both source substances which have been subject to harmonised classification and those which have been subject to self-classification.

192 In the present case, it should be noted that, as the Commission points out, SAS-DDS has been subject to self-classification on several occasions. The ECHA classification and labelling inventory database shows that that substance was notified by eight notifiers, three of which explicitly referred to the lungs as the target organ.

193 It follows that, contrary to what the applicant maintains, SAS-DDS could be considered to be ‘a structural analogue that has previously been classified’, within the meaning of Section 3.9.2.10.3 of Annex I to Regulation No 1272/2008, for the purposes of the classification of silanamine.

194 Although the applicant maintains that self-classification cannot benefit from a presumption of validity, it should be noted, as the Commission observed, that, as is apparent from paragraphs 179 and 180 above, the RAC assessed, in the section entitled ‘Read-across between the different types of amorphous silica’, the relevance of the data relating to SAS-DDS and concluded that it was sufficiently similar in order to be used as a source substance.

195 The applicant submits, secondly, that the RAC failed to demonstrate that the data related to a validated structure activity relationship between SAS-DDS and silanamine and that an expert judgment-based extrapolation existed.

196 However, it must be stated, as the Commission did at the hearing, that such an argument was raised for the first time at the reply stage.

197 In that respect, it should be pointed out that, under Article 84(1) of the Rules of Procedure of the General Court, no new plea in law may be introduced in the course of proceedings unless it is based on matters of law or of fact which come to light in the course of the procedure.

198 According to the case-law, Article 84(1) of the Rules of Procedure is also applicable to complaints or arguments (see judgment of 14 July 2021, AQ v eu-LISA , T‑164/19, not published, EU:T:2021:456, paragraph 59 and the case-law cited).

199 However, a plea, or a complaint, which amounts to amplifying a plea or complaint put forward previously, whether directly or by implication, in the originating application and which is closely connected therewith must be declared admissible (see judgment of 24 September 2019, Yanukovych v Council , T‑301/18, not published, EU:T:2019:676, paragraph 74 and the case-law cited). In order to be regarded as the amplification of a plea or a complaint previously advanced, a new argument must display a sufficiently close connection with the pleas or complaints initially set out in the application in order to be regarded as forming part of the normal evolution of debate in proceedings before the Court (see judgment of 20 November 2017, Petrov and Others v Parliament , T‑452/15, EU:T:2017:822, paragraph 46 and the case-law cited).

200 In the present case, it should be noted that it is not at all apparent from the application that the argument relating to data from a validated structure activity relationship and to the existence of an expert judgment-based extrapolation was referred to therein. Indeed, the second part of the plea put forward in the application related, in essence, to the requirement of a structural analogue that has previously been classified.

201 Consequently, the applicant’s argument relating to data from a validated structure activity relationship and to the existence of an expert judgment-based extrapolation is new and must therefore be declared inadmissible.

202 In any event, it should be noted, on the one hand, that it is clear from the RAC Opinion and from the accompanying reference document, which sets out the detailed scientific grounds for that opinion, that the RAC examined the read-across in detail and set out the justifications permitting extrapolation and, on the other hand, that, in accordance with the case-law cited in paragraph 28 above, the Commission enjoys a broad discretion with regard to the evaluation of scientific studies.

203 In the light of the foregoing considerations, the second part and, consequently, the first plea in law in its entirety must be rejected as unfounded.

D. The second plea in law, alleging breach of the harmonised classification and labelling procedure

204 In the context of the second plea, which is divided into two parts, the applicant claims that the contested regulation was adopted in breach of the procedure laid down in Article 37(4) of Regulation No 1272/2008. According to the first part, the RAC did not adopt its opinion within the 18-month deadline laid down in that article. According to the second part, the RAC did not give the parties concerned the opportunity to comment on its opinion, in accordance with the same article.

1. The first part, alleging failure to meet the deadline laid down for the adoption of the RAC Opinion

205 In the context of the first part of the second plea, the applicant submits that Article 37(4) of Regulation No 1272/2008 lays down a legally binding 18-month deadline for the adoption of the RAC Opinion. It states that this is an essential procedural requirement, the infringement of which must lead to the annulment of the contested act. In the present case, however, the date of adoption of the RAC Opinion is uncertain. Accordingly, the RAC failed to meet the deadline laid down in Article 37(4) of Regulation No 1272/2008 when it adopted its opinion.

206 The Commission, supported by ECHA, disputes those arguments.

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

208 It should also be noted that, under Article 37(1) of Regulation No 1272/2008, a competent authority may submit to ECHA a classification proposal, which must follow the format set out in Part 2 of Annex VI to that regulation and contain the relevant information provided for in Part 1 of Annex VI to that regulation.

209 Part 1 of Annex VI to Regulation No 1272/2008, entitled ‘Introduction to the list of harmonised classifications and labelling’, sets out, inter alia, the information relating to each entry. Part 2 of that annex, entitled ‘Dossiers for harmonised classification and labelling’, lays down general principles for preparing dossiers to propose and justify harmonised classification and labelling. In accordance with that part, a dossier for harmonised classification and labelling contains the proposal, which includes the identity of the substance concerned, the justification, which contains a comparison of the available information with the criteria contained in Parts 2 to 5, taking into account the general principles in Part 1 of Annex I to that regulation, and the justification for other effects at EU level.

210 It follows from those considerations that the RAC must adopt its opinion on any proposal submitted, in particular, in accordance with Article 37(1) of Regulation No 1272/2008, that is to say, any proposal which complies with the format set out in Part 2 of Annex VI to that regulation and contains the relevant information provided for in Part 1 of Annex VI to that regulation.

211 In that context, it must be considered that the 18-month period for the adoption of the opinion provided for in Article 37(4) of Regulation No 1272/2008 starts to run from the date of receipt of a classification proposal which complies with the requirements laid down in Parts 1 and 2 of Annex VI to that regulation.

212 In the present case, it should be noted that it is apparent from the file that the classification proposal submitted by the French competent authority on which the RAC adopted its opinion is dated 17 December 2018 and contains the words ‘version number: v2’. As can be seen from the Registry of CLH intentions until outcome, the French competent authority submitted a classification proposal for ‘accordance check’ on 6 December 2017 and the ‘final submission date’ was 17 December 2018. It follows that the latter date is the date on which the version of the classification proposal was deemed to comply with the necessary requirements.

213 It should also be noted that the RAC Opinion taken into account in the contested regulation is dated 5 December 2019. Since the version of the classification proposal complying with the requirements laid down was submitted on 17 December 2018, the date of 5 December 2019 should be regarded as falling within the 18-month period provided for by Article 37(4) of Regulation No 1272/2008.

214 It is nevertheless clear that, as the applicant submits and as is apparent from paragraph 8 above, the RAC organised a targeted public consultation which took place between 3 and 17 February 2020 and which related to the hazard class acute toxicity Category 2. According to the minutes of the 52nd meeting of the RAC of 4 May 2020, the RAC ‘took note of the outcome of the ad hoc consultation’, ‘reviewed the data on the studies and addressed industries[’] comments in detail but did not change its earlier classification conclusion as a result’. The RAC also asked the ‘Rapporteurs to finalise the opinion with the outcome of the ad hoc consultation and provide it to [the ECHA Secretariat]’, so that the latter could ‘make an editorial check of the opinion documents in consultation with the Rapporteurs’, ‘forward the adopted opinion and its annexes to [the Commission] and publish it on the ECHA website’. Comments from parties concerned during the targeted public consultation were thus compiled in Annex 3 to the RAC Opinion.

215 Although Annex 3 to the RAC Opinion contains the comments of the parties concerned dating from February 2020, the published version of the opinion, as is apparent from paragraph 213 above, is dated 5 December 2019.

216 In that context, it should be noted that, as the applicant submits, the answer to the question of what is the actual date of adoption of the RAC Opinion is not self-evident.

217 Nevertheless, it must be held that, although the RAC Opinion dates from 4 May 2020, the date of the minutes of the 52nd meeting of the RAC, it is still within the 18-month period provided for in Article 37(4) of Regulation No 1272/2008.

218 In any event, even assuming that the RAC Opinion was not adopted within the 18-month deadline provided for in Article 37(4) of Regulation No 1272/2008, failure to meet that deadline could be regarded as a procedural irregularity.

219 In that regard, it should be borne in mind that, according to the case-law, a procedural irregularity could entail annulment of the act ultimately adopted only if, were it not for that irregularity, the procedure could have led to a different result (see judgment of 25 October 2005, Germany and Denmark v Commission , C‑465/02 and C‑466/02, EU:C:2005:636, paragraph 37 and the case-law cited).

220 In the present case, it should be noted that compliance with the 18-month deadline for the adoption of the RAC Opinion could not have led to a different result. As can be seen from paragraphs 9 and 10 above, the contested regulation included silanamine in Table 3 of Part 3 of Annex VI to Regulation No 1272/2008, in the hazard class STOT RE, but specified that that substance should not be included in the hazard class acute toxicity, recommended in the RAC Opinion, as the new scientific information received had been assessed by the Commission, which had found that it required further assessment by the RAC.

221 The RAC Opinion proposed the classification of silanamine in the hazard class STOT RE in the version of the opinion of 5 December 2019 and that proposal was not the subject of the targeted public consultation which took place in February 2020.

222 Moreover, as is apparent from the RAC Opinion, that proposal was based on the information contained in the classification proposal submitted by the French competent authority and on studies from the open literature.

223 As to the remainder, the applicant does not establish that the targeted public consultation could also have a bearing on the hazard class STOT RE.

224 Accordingly, contrary to what the applicant maintains, the classification of silanamine in the hazard class STOT RE would not have led to a different result in the absence of the targeted public consultation.

225 It follows that, even assuming that the RAC failed to meet the 18-month deadline laid down in Article 37(4) of Regulation No 1272/2008, this could not have had an effect on the inclusion of silanamine in Table 3 of Part 3 of Annex VI to Regulation No 1272/2008, in the hazard class STOT RE, by the contested regulation, so that such an irregularity cannot lead to the annulment of that regulation.

226 The first part of the second plea must therefore be rejected as unfounded.

2. The second part, alleging the absence of any public consultation on the RAC Opinion

227 In the context of the second part of the second plea, the applicant submits 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 applicant’s right to comment on the RAC Opinion, or even on the draft of that opinion, derives from the wording of Article 37(4) of Regulation No 1272/2008, read in conjunction with recital 52 of that regulation. That right is all the more justified where, as in the present case, the classification proposed by the RAC is different from that referred to in the proposal for the harmonised classification and labelling of silanamine. The applicant considers that, had the parties concerned been consulted on the RAC Opinion, the result of the classification of silanamine would have been different. It argues that such an irregularity constitutes an infringement of essential procedural requirements which must lead to the annulment of the contested regulation.

228 The Commission, supported by ECHA, disputes those arguments.

229 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).

230 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).

231 In the present case, the contested regulation lays down measures of general application, including the contested classification. As is apparent from paragraph 9 above, Article 1 of that regulation lays down a measure of general application concerning the inclusion of silanamine 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.

232 Against that background, the procedural rights which the applicant enjoys 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).

233 In that regard, it should be recalled that, as stated in paragraph 207 above, Article 37(4) of that regulation 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’.

234 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 18 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.

235 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).

236 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 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).

237 Accordingly, in the present case, the applicant had the right to comment on the proposal for harmonised classification and labelling and to be heard in that regard before the RAC, and availed itself of that right. As is apparent from paragraph 6 above, between 4 March and 3 May 2019, a public consultation on the proposal for harmonised classification and labelling of silanamine was organised. Moreover, as is apparent from the content of the RAC Opinion itself, the RAC took into consideration the comments submitted by the parties concerned, which were compiled in Annex 2 to that opinion.

238 In those circumstances, and in accordance with the case-law cited in paragraphs 230 and 232 above, the applicant did not have a right to be heard or consulted on the RAC Opinion.

239 The applicant’s other arguments cannot call that conclusion into question.

240 In the first place, with regard to the applicant’s argument that the right to be heard on the RAC Opinion is more justified where the classification proposed by the RAC is different from that referred to in the classification proposal, it must be considered that, quite apart from the fact that the RAC is not bound by the classification proposal, it follows from paragraph 236 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.

241 In the second place, in so far as the applicant argues that consultation of the parties concerned on the draft opinion is particularly necessary where the opinion takes into account studies which had not been taken into account in the classification proposal, it is sufficient to point out that, as is clear from paragraph 58 above, the reference to ‘all evidence available’ in Section 3.9.2.1 of Annex I to Regulation No 1272/2008, without any limitation, is a sufficient basis for considering that the RAC may take into account studies which do not form part of the classification proposal.

242 It follows from the foregoing that an infringement of Article 37(4) of Regulation No 1272/2008 has not been demonstrated in the present case.

243 In the light of the foregoing considerations, the second part and, consequently, the second plea in its entirety must be rejected as unfounded.

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

244 The applicant submits 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 principle of sound administration, in that it failed to verify the quality of the classification proposal dossier and failed to take into account the breach of procedure committed by the RAC when it adopted its opinion.

245 The Commission, supported by ECHA, disputes those arguments.

246 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 by inclusion of that substance together with the relevant classification and labelling elements in Table 3 of Part 3 of Annex VI and, where appropriate, the specific concentration limits or M-factors.’ To that end, it must take into account, first of all, the proposal submitted pursuant to Article 37(1) to (3) of that regulation, next, the RAC Opinion and, lastly, the observations made during the public consultations, in accordance with Article 37(2) and (4) of that regulation, although those elements, in particular the RAC 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).

247 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 28 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 applicant does not dispute.

248 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).

249 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.

250 In the present case, as is apparent from the first indent of recital 2 and from recital 5 of the contested regulation, the Commission ensured that the reasoning in the RAC Opinion was full, consistent and relevant. It is apparent from recital 5 of the contested regulation that the Commission considered that additional information pertaining to the acute inhalation toxicity of silanamine had been received after the RAC Opinion had been forwarded to it. The Commission considered that the classification of that substance as hazard class acute toxicity by inhalation Category 2, recommended in the RAC Opinion, should not be included in Annex VI to Regulation No 1272/2008, as, following assessment of the new scientific information, it required further assessment by the RAC. However, the Commission stated that the classification of that substance as hazard class STOT RE Category 2, recommended in the RAC Opinion, should be included in Annex VI to that regulation, since no new information that would require further assessment for that classification had been received.

251 As to the remainder, 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 silanamine was appropriate.

252 In so far as the applicant argues that the Commission wrongly concluded that the contested classification was appropriate, although the classification proposal dossier was of insufficient quality, it should be noted that, in accordance with Section 6.1 of ECHA’s ‘Guidance on the preparation of dossiers for harmonised classification and labelling’, the ECHA Secretariat performs an accordance check of the classification proposal dossier in order to ensure that it complies with the legal requirements, namely that it includes all the information needed for the RAC to consider the classification proposal and to deliver an opinion.

253 In the present case, it should be noted that, as is apparent from paragraph 212 above, on 6 December 2017 the French competent authority submitted a classification proposal for accordance check, before submitting the final version of that classification proposal on 17 December 2018.

254 It follows that ECHA performed an accordance check on the classification proposal to ensure that it complied with the requirements laid down in Article 37(1) of Regulation No 1272/2008.

255 Moreover, it should be noted that it follows from the RAC Opinion that the RAC was able to examine the classification proposal and that, as is clear from paragraph 8 above, between 4 March and 3 May 2019, several parties concerned submitted, in accordance with Article 37(4) of Regulation No 1272/2008, comments on the classification proposal which were compiled in Annex 2 to the RAC Opinion.

256 It follows that both the RAC and the parties concerned have been able to take a position on the classification proposal, the compliance of which with Article 37(1) of Regulation No 1272/2008 has been verified by ECHA.

257 Against that background, the Commission cannot be criticised for concluding that the harmonisation of the classification and labelling of silanamine was appropriate within the meaning of Article 37(5) of Regulation No 1272/2008, although the classification proposal was of insufficient quality.

258 Nor can the applicant’s claims that the Commission wrongly concluded that the harmonisation of the classification and labelling of silanamine was appropriate, although the RAC Opinion was adopted in breach of procedure, succeed.

259 For the same reasons as those set out in paragraphs 205 to 243 above, it is necessary to reject the applicant’s arguments alleging failure to meet the 18-month deadline and the absence of public consultation on the RAC Opinion.

260 Moreover, with regard to the applicant’s arguments concerning the targeted public consultation, it is sufficient to note that, irrespective of whether the RAC could organise such a public consultation, as is apparent from paragraphs 8 and 10 above, that public consultation concerned acute toxicity and the Commission excluded the classification of silanamine in that hazard class. As to the remainder, the applicant does not establish that the targeted public consultation was also related to the hazard class STOT RE.

261 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.

262 For the same reasons, it must be concluded that the Commission did not breach the principle of sound administration, and the applicant does not moreover put forward any specific argument in that regard.

263 In the light of the foregoing considerations, the third plea must be rejected as unfounded.

F. The fourth plea in law, alleging the absence of an impact assessment

264 The applicant submits that the Commission infringed 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 silanamine 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 silanamine.

265 The Commission, supported by ECHA, disputes those arguments.

266 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.’

267 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’.

268 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).

269 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).

270 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).

271 In the present case, it should be noted that, in accordance with the case-law cited in paragraph 268 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.

272 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 18 and 234 above).

273 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.

274 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 14 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).

275 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).

276 Furthermore, the applicant’s 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.

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

Costs

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

279 Since the applicant has been unsuccessful, it must be ordered to bear its own costs and to pay those incurred by the Commission, in accordance with the form of order sought by the latter.

280 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 ECHA must bear its own costs.

On those grounds,

THE GENERAL COURT (Sixth Chamber)

hereby:

1. Dismisses the action;

2. Orders Evonik Operations GmbH to bear its own costs and to pay those incurred by the European Commission;

3. Orders the European Chemicals Agency (ECHA) to bear its own costs.

Costeira

Kancheva

Zilgalvis

Delivered in open court in Luxembourg on 27 November 2024.

V. Di Bucci

S. Papasavvas

Registrar

President

Table of contents

I. Background to the dispute

II. Forms of order sought

III. Law

A. Preliminary considerations on harmonised classification and labelling of substances in the hazard class STOT RE

B. Preliminary observations on the intensity of the Court’s review

C. The first plea in law, alleging manifest errors of assessment and breach of the criteria for classification of a substance in the hazard class STOT RE Category 2

1. The third part, alleging that silanamine has no adverse health effects

(a) Preliminary considerations concerning the effects considered to support classification of a substance in the hazard class STOT RE Category 2

(b) The effects identified in the RAC Opinion which are considered to support the contested classification

(c) Justification for the contested classification on the basis of the effects identified in the RAC Opinion

2. The first part, alleging that silanamine produces no specific toxic effects on the lungs

(a) The applicant’s arguments alleging that the effects observed result from the deposition of large numbers of inhaled particles in the context of the relevant studies

(b) The applicant’s arguments alleging that the effects observed result from the high dose/concentration administered in the relevant studies

3. The second part, alleging that the effects observed are based on a read-across that does not meet the criteria for classification in the hazard class STOT RE

D. The second plea in law, alleging breach of the harmonised classification and labelling procedure

1. The first part, alleging failure to meet the deadline laid down for the adoption of the RAC Opinion

2. The second part, alleging the absence of any public consultation on the RAC Opinion

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

F. The fourth plea in law, alleging the absence of an impact assessment

Costs

* Language of the case: English.