JUDGMENT OF THE GENERAL COURT (Second Chamber)

5 July 2023 (*)

(Environment and protection of human health – Regulation (EC) No 1272/2008 – Classification, labelling and packaging of substances and mixtures – Delegated Regulation (EU) 2020/1182 – Classification of dioctyltin dilaurate [1]; stannane, dioctyl-, bis (coco acyloxy) derivs. [2] – Read-across – Burden of proof – Manifest error of assessment – Impact assessment)

In Case T‑639/20,

TIB Chemicals AG, established in Mannheim (Germany), represented by K. Fischer, lawyer,

applicant,

v

European Commission, represented by R. Lindenthal, S. Delaude and A. Dawes, acting as Agents,

defendant,

supported by

Republic of Austria, represented by J. Schmoll, acting as Agent,

by

Kingdom of Sweden, represented by O. Simonsson, C. Meyer-Seitz, A. Runeskjöld, H. Shev, M. Salborn Hodgson, H. Eklinder and R. Shahsavan Eriksson, acting as Agents,

and by

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

interveners,

THE GENERAL COURT (Second Chamber),

composed of A. Marcoulli, President, S. Frimodt Nielsen and J. Schwarcz (Rapporteur), Judges,

Registrar: V. Di Bucci,

having regard to the written part of the procedure,

having regard to the fact that no request for a hearing was submitted by the parties within three weeks after service of notification of the close of the written part of the procedure, and having decided to rule on the action without an oral part of the procedure, pursuant to Article 106(3) of the Rules of Procedure of the General Court,

gives the following

Judgment

1        By its action under Article 263 TFEU, the applicant, TIB Chemicals AG, seeks the partial annulment of Commission Delegated Regulation (EU) 2020/1182 of 19 May 2020 amending, for the purposes of its adaptation to technical and scientific progress, Part 3 of Annex VI to Regulation (EC) No 1272/2008 of the European Parliament and of the Council on classification, labelling and packaging of substances and mixtures (OJ 2020 L 261, p. 2; ‘the contested regulation’), in so far as it concerns the susbstance dioctyltin dilaurate; [1] stannane, dioctyl-, bis(coco acyloxy) derivs. [2] (EC No 222-883-3 [1] 293-901-5 [2] and CAS No 3648-18-8 [1] 91648-39-4 [2]) (‘DOTL’).

 Legal background

2        In accordance with Article 1(1) thereof, the purpose of Regulation 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), 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 within the EU market. To that end, in accordance with Article 1(1)(a) thereof, that regulation seeks, inter alia, to harmonise the criteria for classification of substances and mixtures, and the rules on labelling and packaging for hazardous substances and mixtures.

3        As regards the classification of hazardous substances and mixtures, Article 3 of Regulation No 1272/2008 provides that a substance or a mixture fulfilling the criteria relating to physical hazards, health hazards or environmental hazards, laid down in Parts 2 to 5 of Annex I, is hazardous and is to be classified in relation to the respective hazard classes provided for in that annex.

4        In that regard, first, Regulation No 1272/2008 provides, in Title V thereof, for a procedure for the harmonisation, throughout the European Union, of the classification and labelling of substances, which concerns substances which fulfil the criteria set out in Annex I for the hazards referred to in Article 36(1) of that regulation. In Article 36(3) of that regulation, other situations are provided for, in particular ‘where a substance fulfils the criteria for other hazard classes or differentiations than those referred to in paragraph 1’. Second, Regulation No 1272/2008, in particular Articles 9 and 13 thereof, requires manufacturers, importers and downstream users to classify substances and mixtures.

5        First of all, the procedure for harmonising the classification and labelling of substances laid down in Title V of Regulation No 1272/2008, as relevant in the present case regarding the substance DOTL (see paragraph 14 below), is initiated by manufacturers, importers and downstream users of a substance or by the competent authority of a Member State by the submission of a proposal to the European Chemicals Agency (ECHA), in accordance with Article 37(1) and (2) of Regulation No 1272/2008. A ‘competent authority’ is defined in Article 2(24) of that regulation as ‘the authority or authorities or bodies established by the Member States to carry out the obligations arising from this Regulation’.

6        Next, the ECHA Committee for Risk Assessment (‘the RAC’) is to adopt an opinion on the proposal submitted, giving the parties concerned the opportunity to comment, and the ECHA is to forward that opinion and any comments to the European Commission, in accordance with Article 37(4) of Regulation No 1272/2008.

7        Lastly, where the Commission finds that the harmonisation of the classification and labelling of the substance concerned is appropriate, it is to adopt a delegated act, in accordance with Article 37(5) and Article 53a of Regulation No 1272/2008, to include the name of the substance in question and the relevant classification and labelling elements in Table 3.1 of Part 3 of Annex VI to that regulation.

8        That harmonised classification and labelling of substances, pursuant to Title V of Regulation No 1272/2008, seeks to determine the intrinsic properties of the substances (and mixtures containing them) which must lead to their classification as hazardous products, so that their hazards can be correctly identified and notified.

9        The hazard class ‘reproductive toxicity’ covers adverse effects on sexual function and fertility in adult males and females as well as developmental toxicity in their offspring. Article 36(1)(d) of Regulation No 1272/2008 provides that, if a substance meets the criteria set out in Annex I to that regulation for the hazard of reproductive toxicity, it will normally be subject to harmonisation of classification and labelling. Those criteria are set out in Section 3.7, specifically in Section 3.7.1.1, of Part 3 of Annex I to that regulation.

10      The hazard class ‘specific target organ toxicity (repeated exposure)’ covers specific toxic effects on target organs arising from repeated exposure to a substance or mixture. The criteria which must be fulfilled for the purposes of Article 36(3) of Regulation No 1272/2008 in order for a substance to be subject to harmonisation of classification and labelling are set out in Section 3.9, specifically in Section 3.9.1.1, of Part 3 of Annex I to Regulation No 1272/2008.

 Background to the dispute

11      The applicant manufactures and sells DOTL, which is used, inter alia, as a food contact and in drinking water pipes, in the European Union.

12      In 2013, the applicant, acting as the lead registrant, submitted to the ECHA an application for registration of DOTL (‘the REACH registration dossier on DOTL’), in the context of a joint submission pursuant to Article 11 of Regulation (EC) No 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/EEC, 93/67/EEC, 93/105/EC and 2000/21/EC (OJ 2006 L 396, p. 1) (‘the REACH Regulation’).

13      The REACH registration dossier on DOTL does not contain any data on tests performed as regards the endpoints ‘reproductive toxicity’ and ‘repeated dose toxicity’. Those endpoints were evaluated via a read-across from dioctyltin oxide (EC No 212-791-1 and CAS No 870-08-6; ‘DOTO’) and lauric acid.

14      On 18 September 2015, the Kemikalieinspektionen (Chemicals Agency, Sweden) (‘KEMI’) notified the ECHA of its intention to submit a proposal for the harmonised classification and labelling of DOTL as toxic for reproduction. That initial intention was first withdrawn by KEMI and then reactivated on 14 August 2017. KEMI’s dossier on the harmonised classification and labelling of substances in accordance with Articles 36 and 37 of Regulation No 1272/2008 (‘the CLH dossier’) contained a proposal for classification as ‘Repr. 1B’ (presumed human reproductive toxicant category 1B) and ‘STOT RE 1’ (specific target organ toxicity – repeated exposure, category 1), with the respective hazard statement codes ‘H360D’ and ‘H372’ (immune system), as well as with the respective labelling requirements and a justification and background information in the report drawn up with a view to the constitution of that dossier (‘the CLH report’).

15      On 17 October 2017, KEMI’s CLH report was published on the ECHA’s website and interested parties and the competent authorities of the Member States were given the opportunity to comment until 1 December 2017. The applicant and a number of other private entities as well as public authorities submitted comments.

16      After learning, in September 2018, that the RAC had issued its opinion on a harmonised classification and labelling of DOTL on the basis of read-across from dioctyltin dichloride (EC No 222-583-2 and CAS No 3542-36-7; ‘DOTC’), the applicant submitted a testing proposal to the ECHA in order to provide further evidence to show that the RAC’s assessment was flawed. The subject of that testing proposal was the performance of a prenatal developmental toxicity study (OECD TG 414) in rats, in order to demonstrate that there was no effect or indication of reproductive toxicity with respect to DOTL. As a result, the REACH registration dossier on DOTL was updated with this testing proposal on 11 September 2018.

17      On 14 September 2018, the RAC adopted its opinion on the proposed harmonised classification and labelling of DOTL. The ECHA forwarded the RAC’s opinion to the Commission, together with the unchanged classification proposal.

18      The applicant explained in two letters – the first sent to the European Ombudsman on 6 March 2019 and the second sent to the Commission’s Directorate-General (DG) for Internal Market, Industry, Entrepreneurship and SMEs on 14 June 2019 – that the RAC’s opinion that DOTL should be classified as toxic for reproduction, carcinogenic or mutagenic was flawed, because it contradicts scientific data provided by the applicant and scientific literature to which the applicant referred. The Ombudsman rejected the applicant’s arguments in its letter of 4 April 2019 and DG Internal Market, Industry, Entrepreneurship and SMEs rejected them in its letter of 24 July 2019. In addition, the applicant contacted the competent German authority, the Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (Federal Institute for Occupational Health and Safety, Germany).

19      On 17 October 2019, the ECHA sent the applicant a communication concerning the identity of DOTL and questioning whether that substance should have a new name.

20      On 19 May 2020, the Commission adopted the contested regulation, recital 1 of which states as follows:

‘Table 3 of Part 3 of Annex VI to Regulation … No 1272/2008 contains the list of harmonised classification and labelling of hazardous substances based on the criteria set out in Parts 2 to 5 of Annex I to that Regulation.’

21      Recital 2 of the contested regulation is worded as follows:

‘Proposals to introduce harmonised classification and labelling of certain substances and to update or delete the harmonised classification and labelling of certain other substances have been submitted to the [ECHA] pursuant to Article 37 of Regulation … No 1272/2008. Based on the opinions … on those proposals issued by the [RAC], as well as on the comments received from the parties concerned, it is appropriate to introduce, update or delete the harmonised classification and labelling of certain substances. Those RAC opinions are:

…

–        Opinion of 14 September 2018 concerning [DOTL];

…’

22      By the contested regulation, DOTL was inserted into Part 3 of Annex VI to Regulation No 1272/2008 with the following harmonised classification: hazard class ‘Repr. 1B’ (presumed human reproductive toxicant category 1B) and ‘STOT RE 1’ (specific target organ toxicity – repeated exposure, category 1) with the respective hazard statement codes ‘H360D’ and ‘H372’ (immune system) and the corresponding labelling.

 Forms of order sought

23      The applicant claims that the Court should:

–        partially annul the contested regulation, in so far as it concerns DOTL;

–        order the Commission to pay the costs.

24      The Commission contends that the Court should:

–        dismiss the action;

–        order the applicant to pay the costs.

25      The Republic of Austria, the Kingdom of Sweden and the ECHA support the Commission’s position and contend that the action should be dismissed. The ECHA also requests the Court to order the applicant to pay the costs.

 Law

 Admissibility of the action inasmuch as the applicant disputes the lawfulness of the classification of DOTL as a specific target organ toxicant, repeated exposure category 1 (STOT RE 1)

26      The Commission submits that, even though the applicant seeks the partial annulment of the contested act in so far as it concerns DOTL and the relevant classification and labelling elements, it does not substantiate its challenge to the lawfulness of the classification of DOTL as a specific target organ toxicant, repeated exposure category 1 (STOT RE 1).

27      In response to the questions posed by the Court by way of measures of organisation of procedure, the applicant submits that it also justified its challenge to the lawfulness of the classification of DOTL as a substance designated as ‘STOT RE 1’. It reaches that conclusion on the basis that it criticised the Commission, the RAC and KEMI for relying on a read-across from DOTC, which is classified as ‘STOT RE 1’, when classifying DOTL as a substance in the same class. It is apparent from its own read-across from DOTO, which is classified as ‘STOT SE 2’ and ‘H371’, corresponding to specific target organ toxicity, single exposure, that it maintains, rather, that the latter classification, relating to acute but reversible effects, should apply to DOTL.

28      Under the first paragraph of Article 21 of the Statute of the Court of Justice of the European Union, which is applicable to proceedings before the General Court by virtue of the first paragraph of Article 53 of that statute, and under Article 76(d) of the Rules of Procedure of the General Court, an application must contain the subject matter of the proceedings, the pleas in law and arguments relied on and a summary of those pleas in law. Those particulars must be sufficiently clear and precise to enable the defendant to prepare its defence and the Court to rule on the action, if necessary without any other supporting information (see order of 13 May 2020, Lucaccioni v Commission, T‑308/19, not published, EU:T:2020:207, paragraph 34 and the case-law cited).

29      Contrary to the requirements of the case-law referred to in paragraph 28 above, the particulars as are directly apparent from the application are not sufficiently clear in so far as they take issue with the lawfulness of the classification of DOTL as a specific target organ toxicant, repeated exposure, category 1 (STOT RE 1). The fact that the applicant referred to the proposed classification of DOTL as a ‘STOT SE 2’ and ‘H371’ substance, due to the risks of harmful effects on the immune system, is not sufficient to alter that conclusion. Nor can that conclusion be called into question by the explanatory information subsequently provided by the applicant in response to the Court’s questions. Accordingly, the applicant’s head of claim relating to the lawfulness of the classification of DOTL as a specific target organ toxicant, repeated exposure, category 1 (STOT RE 1), must be rejected as inadmissible.

 The scope of judicial review

30      With regard to the intensity of the Court’s review, it should be borne in mind that, according to 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).

31      However, the exercise of that discretion is not excluded from judicial review. It follows from settled case-law that in the context of such a review the EU judicature must verify whether the relevant procedural rules have been complied with, whether the facts admitted by the Commission have been accurately stated and whether there has been a manifest error of appraisal 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).

32      In particular, where a party claims that the institution competent in the matter has committed a manifest error of assessment, the EU judicature 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).

33      Furthermore, the limits to the review by the EU judicature do not affect its duty to establish whether the evidence relied on is factually accurate, reliable and consistent, 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).

34      Furthermore, as regards the assessment of scientific studies, the Court has already held that the Commission must be allowed a broad discretion as regards that assessment, as well as the choice of studies that must prevail over others, irrespective of their chronology (see judgment of 24 October 2018, Deza v Commission, T‑400/17, not published, EU:T:2018:712, paragraph 95).

35      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 17 May 2018, BASF Agro and Others v Commission, T‑584/13, EU:T:2018:279, paragraph 94 and the case-law cited).

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

 The substance

37      The applicant claims that the contested regulation is unlawful in so far as DOTL is classified as ‘Repr. 1B’ and ‘STOT RE 1’, with the respective hazard statement codes ‘H360D’ and ‘H372’ (immune system) as well as with the respective labelling requirements.

38      The applicant raises, in essence, seven pleas in law in support of its claims, alleging, respectively:

–        infringement of Article 37(1) and (5) of, and Part 2 of Annex VI to, Regulation No 1272/2008, in that KEMI, the RAC and the Commission committed a manifest error of assessment in rejecting the proposal to perform a read-across from DOTO to DOTL;

–        infringement of Articles 5, 9 and 36 of, and Parts 1 and 3 of Annex I to, Regulation No 1272/2008, in that the Commission and the other competent authorities did not take certain information into account in a ‘weight-of-evidence’ approach and thus made a manifest error of assessment;

–        infringement of Regulation No 1272/2008, inasmuch as the read-across from DOTC to DOTL made by KEMI and the RAC is not based on scientific data;

–        failure to comply with the substantive classification requirements of Regulation No 1272/2008, in that the Commission and the other competent authorities erred in classifying DOTL as toxic for reproduction, ‘category 1B’ (presumed human reproductive toxicant);

–        failure to observe the principle of proportionality inasmuch as the classification of DOTL as ‘Repr. 1B’ is neither necessary nor appropriate;

–        infringement of Article 37(4) of Regulation No 1272/2008, breach of the right to be heard and failure to observe the principle of good administration;

–        failure, on the part of the Commission, to fulfil its obligation to carry out an impact assessment before adopting the contested act.

 The admissibility of the applicant’s first four pleas

39      The Commission contends that the first four pleas are inadmissible, since the application does not set out the essential points of law and of fact on which those pleas are based.

40      The applicant maintains that its first four pleas are admissible. While the pleas were set out in the body of the application, Annexes A.20 and A.21 to that application supplemented those pleas with scientific evidence.

41      It must be held, contrary to the Commission’s assertions, that the application complies with the requirements of the case-law referred to in paragraph 28 above, with the exception of the applicant’s claims concerning the lawfulness of the classification of DOTL as a substance coming within a ‘STOT RE 1’ classification (see paragraphs 26 to 29 above) and subject to the references to Annexes A.20 and A.21 (see paragraphs 42 to 45 below). As a result, and subject to those reservations, the first four pleas are sufficiently clear to be regarded as admissible.

 Admissibility of the applicant’s reference to the various annexes

42      So far as concerns the reference to Annexes A.20 and A.21 to the application (see paragraph 40 above), which is criticised by the Commission, it should be recalled that, whilst the body of the application may be supported and supplemented on specific points by references to extracts from documents annexed thereto, a general reference to other documents, even those annexed to the application, cannot make up for the absence of the essential arguments in law which, in accordance with Article 21 of the Statute of the Court of Justice of the European Union and Article 76 of the Rules of Procedure, must appear in the application (judgment of 17 September 2007, Microsoft v Commission, T‑201/04, EU:T:2007:289, paragraph 94).

43      Furthermore, it is not for the Court to seek and identify in the annexes the pleas and arguments on which it may consider the action to be based, since the annexes have a purely evidential and instrumental function (judgment of 17 September 2007, Microsoft v Commission, T‑201/04, EU:T:2007:289, paragraph 94).

44      Thus, an annex to the application may be taken into consideration only in so far as it supports or supplements arguments expressly set out by the applicant in the body of the application and in so far as it is possible for the Court to determine precisely what are the matters contained in the annex that support or supplement those arguments (see, to that effect, judgment of 17 September 2007, Microsoft v Commission, T‑201/04, EU:T:2007:289, paragraph 99).

45      In the light of that case-law, it is only in so far as Annexes A.20 and A.21 support or supplement arguments expressly set out in the body of the application, and in so far as it is possible for the Court to determine precisely what are the matters they contain that support or supplement those arguments, that they will be taken into account.

 The first plea in law, alleging infringement of Article 37(1) and (5) of, and Part 2 of Annex VI to, Regulation No 1272/2008, in that KEMI, the RAC and the Commission committed a manifest error of assessment in rejecting the proposal to perform a read-across from DOTO to DOTL

–       Arguments of the parties

46      The applicant submits that, under Part 2 of Annex VI to Regulation No 1272/2008, to which the second subparagraph of Article 37(1) of that regulation refers, one of the general principles for preparing dossiers to propose and justify the harmonised classification and labelling of a substance is that ‘any relevant information from registration dossiers shall be considered and other available information may be used’. Since the mandatory registration under the REACH Regulation results in an extensive collection of substance-specific data, information from the registration dossier is explicitly named as a primary tool for information gathering with respect to harmonised classification and labelling.

47      The applicant claims that the CLH report mentions the REACH registration dossier on DOTL as a data source. The fact that it, as the lead registrant for DOTL, had proposed that classification be based on a read-across from DOTO as the source substance was, however, neither mentioned nor considered to be an option. Nor did the RAC’s opinion consider such a possibility. The applicant submits that it adduced evidence to justify the use of a read-across from DOTO to DOTL, including a reference to an OECD TG 422 study.

48      According to the applicant, which is also the lead registrant for DOTO under the REACH Regulation, that substance is not classified for the endpoints ‘reproductive toxicity’ and ‘repeated dose toxicity’, as had already been demonstrated by the OECD TG 422 study on combined oral repeated dose toxicity with the reproduction/developmental toxicity screening test in rats, undertaken in 2004 and also included in the REACH registration dossier on DOTL (‘the 2004 Waalkens-Berendsen study’). Although that study did not show reproductive toxicity as regards DOTO, the ECHA demanded within a compliance check additional testing (ECHA decision of 18 August 2016). The tests produced following the ECHA’s request, namely the prenatal developmental toxicity study (OECD TG 414) (‘the PDTS study’) and the extended one-generation reproductive toxicity study (OECD TG 443) (‘the EOGRTS study’) confirmed that DOTO did not have any effects of being toxic for reproduction. The REACH registration dossier on DOTL refers to those additional studies, which were finalised in the meantime.

49      In the applicant’s view, the Commission and the other competent authorities relied on a manifestly erroneous read-across from DOTC to DOTL, based on in vitro studies, such as the 2015 and 2016 Nasshan studies, concerning the in vitro assessment of the metabolism of those two substances (respectively ‘the 2015 Nasshan study’ and ‘the 2016 Nasshan study’, and, together, ‘the Nasshan studies’), which are simple experiments on solutions accompanied by a subsequent chemical analysis, which were not carried out under conditions consistent with good laboratory practice (‘GLP’) and did not meet the requirements laid down by the REACH Regulation and Regulation No 1272/2008. The applicant listed what it considers to be serious errors in the RAC’s opinion. In its view, by relying solely on the RAC’s opinion, without an appropriate assessment and by omitting the relevant information, namely the available information contained in the REACH registration dossier on DOTL, the Commission and the other competent authorities infringed the provisions referred to in paragraph 46 above and committed a manifest error of assessment. The Commission also failed to comply with its duty of care in that regard.

50      In the reply, first, the applicant submits that the Commission acknowledges that it submitted, in the context of the REACH registration dossier on DOTL, an OECD TG 422 study on DOTO for the endpoints ‘reproductive toxicity’ and ‘repeated dose toxicity’. Thus, the ECHA and its expert bodies, such as the RAC, were able to see that data for DOTO were cited, which is only possible and permissible in the case of a read-across. Therefore, it was not decisive whether a formal justification of the read-across was presented, but it was necessary that such a read-across was done in a recognisable manner. According to the applicant, when KEMI submitted its proposal for harmonised classification and labelling in August 2017, the REACH registration dossier on DOTL already contained, in Section 7.8.1 (toxicity to reproduction), explicit information on a read-across from DOTO in the mandatory computer format for the purpose of REACH registration dossiers (IUCLID). It was also clearly stated that the data source of the 2004 Waalkens-Berendsen study had been used in a read-across from DOTO, whereas the CLH report cited only an OECD TG 408 study by Appel, M.J., and Waalkens-Berendsen, D.H, relating to the analysis of subchronic (13-week) oral toxicity study in rats, including a reproduction/developmental toxicity screening study, from 2004 concerning DOTC. Since it was ‘obvious’ that the registrant considered that a read-across from DOTO, and not from DOTC, was the appropriate approach, from a scientific point of view, it was therefore not feasible for the RAC and the Commission to neglect that information.

51      Second, the delay in obtaining the two reproductive toxicity studies (EOGRTS and PDTS), requested from the applicant by the ECHA in the context of the registration procedure for DOTL under the REACH Regulation, indicated, in its view, that important findings on reproductive toxicity could be expected. However, although the studies had not yet been completed in 2017, the RAC did not wait for those results, which proved to be negative and constituted scientific evidence in support of the non-classification of DOTO.

52      In response to the statements in intervention, the applicant reacted to the arguments of the Kingdom of Sweden by pointing out that the latter had recognised that the REACH registration dossier on DOTL referred to studies concerning DOTO. Furthermore, the Kingdom of Sweden acknowledges that DOTO was the main hydrolysation product of DOTL, although it considered that a read-across from DOTO at low pH values was not justified. The applicant disputes that there is a legitimate reason for denying a read-across from DOTO at a low pH value. At all pH levels, including in gastric conditions, DOTO is the main degradation product of DOTL (according to the 2015 Nasshan study). A correct application of the read-across from DOTO to DOTL would have led to a finding of only acute and reversible effects (STOT SE 2).

53      As regards the Republic of Austria’s contentions, the applicant submits that, in applying the read-across from DOTO to DOTL, it followed the ECHA’s recommendations. It is not decisive whether the applicant’s justification of the proposed read-across in the registration dossier was sufficiently detailed or not.

54      The Commission, supported by the Republic of Austria, the Kingdom of Sweden and the ECHA, disputes the applicant’s arguments.

–       Findings of the Court

55      As regards the references proposed by the applicant to data in the REACH registration dossier on DOTL (see paragraphs 46 and 47 above), first of all, it should be noted that the applicant acknowledges that that dossier ‘does not contain any data on tests performed on the endpoints “reproductive toxicity” and “repeated dose toxicity”‘ for that substance. Those endpoints were addressed by means of a read-across from DOTO. Thus, it cannot be considered that that dossier contained, as the applicant appears to claim, ‘an extensive collection of substance-specific data’ as regards DOTL.

56      Next, although the applicant had submitted, in the registration procedure for DOTL under the REACH Regulation, some evidence in support of a read-across from DOTO to DOTL, the Commission, the Republic of Austria, the Kingdom of Sweden and the ECHA claim that that evidence was insufficient and did not constitute a detailed justification for the proposed approach. In particular, in their view, the applicant’s claim that such a read-across was done ‘in a recognisable manner’ is insufficient to satisfy the requirements laid down in Section 1.5 of Annex XI to the REACH Regulation.

57      In that regard, first, given that DOTL was registered as a substance manufactured or imported in quantities of 100 tonnes or more per year per manufacturer or importer, as is apparent from Annex A.5 to the application, the information requirements for a registration dossier were set out, inter alia, in Article 12(1)(d) of the REACH Regulation, which provided, inter alia, that the registration dossier had to contain testing proposals for the provision of the information referred to in Annex IX to the REACH Regulation. However, since the REACH registration dossier on DOTL did not contain any study on the reproductive toxicity endpoint of that substance or any testing proposal for such a study (see paragraph 55 above), the requirements laid down in particular in Sections 8.6.2 (‘Sub-chronic toxicity study’) and 8.7.2 (‘Pre-natal developmental toxicity study’) of Annex IX to the REACH Regulation were not met.

58      Second, it is apparent from the third paragraph of the introductory part of Annex IX to the REACH Regulation that the registrant may propose to adapt the required standard information set out in column 1 of that annex, according to the general rules contained in Annex XI. In that event, ‘he shall clearly state the reasons for any decision to propose adaptations to the standard information under the appropriate headings in the registration dossier referring to the appropriate specific rule(s) in column 2 or in Annex XI’.

59      In that regard, the first paragraph of Section 1.5 of Annex XI to the REACH Regulation, as applicable on 14 August 2017 (see paragraph 14 above), provided as follows:

‘Substances whose physicochemical, toxicological and ecotoxicological properties are likely to be similar or follow a regular pattern as a result of structural similarity may be considered as a group, or “category” of substances. Application of the group concept requires that physicochemical properties, human health effects and environmental effects or environmental fate may be predicted from data for reference substance(s) within the group by interpolation to other substances in the group (read-across approach). This avoids the need to test every substance for every endpoint. …’

60      Next, the fourth paragraph of Section 1.5 of Annex XI to the REACH Regulation, as applicable on 14 August 2017, provided as follows:

‘…

In all cases results should:

–        be adequate for the purpose of classification and labelling and/or risk assessment,

–        have adequate and reliable coverage of the key parameters addressed in the corresponding test method referred to in Article 13(3),

–        cover an exposure duration comparable to or longer than the corresponding test method referred to in Article 13(3) if exposure duration is a relevant parameter, and

–        adequate and reliable documentation of the applied method shall be provided.’

61      In 2008, the ECHA also published ‘Guidance on information requirements and chemical safety assessments’, Chapter R.6 of which, entitled ‘QSARs and grouping of chemicals’, states (in point R.6.2.6.1) that it is necessary to provide justification for a read-across, including in a hypothesis based on the recognition of structural similarities. An explanation of the rationale for the prediction of properties and robust study summaries of the source studies is also necessary. Those conditions are further developed in the April 2013 Guidance on the conditions set out in Section 1.5 of Annex XI to the REACH Regulation, concerning, inter alia, the grouping of substances and the read-across approach, in particular in point 3.1(d).

62      Lastly, as regards the classification proposal included in the CLH dossier, the relevant date was 14 August 2017 (see paragraph 14 above). On that date, as the applicant itself acknowledges, the ECHA guidance on performing a read-across (Read-Across Assessment Framework, ‘the RAAF’), dated March 2017, which covered even more advanced requirements, was also in force.

63      The RAAF provides that ‘under REACH, registrants are required to submit a complete dossier with specific information complying with REACH information requirements’, that, ‘for each information requirement, registrants must indicate explicitly whether they are making an adaptation using read-across, and they must provide a comprehensive justification for the use of a read-across approach’. The RAAF also provides that the justification for the use of the read-across provided in the dossier will be assessed. Furthermore, it is provided that the ‘ECHA evaluates the documentation provided in the dossier, and does not undertake extra analysis or research to further develop the scientific justification that would be insufficient or the supporting documentation that would be incomplete’ (RAAF, page 8).

64      In the present case, the question is whether the material concerning the read-across from DOTO to DOTL, submitted by the applicant in the registration procedure for DOTL under the REACH Regulation, satisfies the relevant conditions.

65      In that regard, as the Commission, the Republic of Austria, the Kingdom of Sweden and the ECHA rightly submit, and as is also apparent from the applicant’s replies to the questions posed by the Court by way of measures of organisation of procedure, the section in question in the REACH registration dossier on DOTL referred only to general considerations on the oral absorption of DOTL and to the fact that that substance was considered to be immediately hydrolysed to DOTO and lauric acid in the gut, widely distributed in the body, hardly metabolised, and excreted mainly as unchanged DOTO. Although information relating to the hydrolysis of DOTL and DOTO in gastric conditions was submitted by the applicant, it in no way follows from this that the REACH registration dossier on DOTL contained explanations on the similarities of the toxicological properties between the two substances in question. As stated in paragraph 59 above, it was necessary to demonstrate the existence of certain structural similarities that made it possible to determine why the properties of DOTL could be predicted from those of DOTO, in the context of the proposed read-across. Similarly, the material to which the applicant referred did not contain adequate and reliable documentation of the read-across method used, as required by the fourth paragraph of Section 1.5 of Annex XI to the REACH Regulation (see paragraph 60 above).

66      It is true that the applicant also noted, in its written pleadings before the Court and in its reply to the questions posed by the Court by way of measures of organisation of procedure, additional points relating to the read-across from DOTO to DOTL. However, it must be stated that those points constitute neither a detailed justification for the proposed approach nor adequate and reliable documentation of the method used that would satisfy the requirements laid down in Section 1.5 of Annex XI to the REACH Regulation. The fact that, as the applicant submits, those points show that that read-across was proposed ‘in a recognisable manner’ or, in other words, in such a way that the fact that it was proposed is obvious to any scientific expert, is not decisive in the light of the requirements set out in paragraphs 57 to 63 above.

67      As regards the recognisable nature of the read-across made from DOTO, first, the applicant submits that the REACH registration dossier on DOTL was submitted in 2013, that it had been updated in July 2014 and that it fulfilled the required conditions at the time, in particular during use of the ‘IUCLID 5’ tool, which did not provide for a specific explanation of the read-across. In that regard, it should be noted that adequate and reliable documentation of the DOTO to DOTL read-across method applied had, in any event, to be provided, in accordance with the requirements laid down in the fourth paragraph of Section 1.5 of Annex XI to the REACH Regulation, so as to make it possible to envisage, as a result of any structural similarity between those substances, the relevance of that method for the assessment of DOTL. In addition to the fact that the REACH Regulation must be interpreted as requiring a read-across to be justified, inasmuch as it precisely provides the information to be indicated in that regard (see paragraphs 59 and 60 above), and given its objective, which is to ensure a high level of protection of human health and the environment, in accordance with the first recital thereof, it is apparent, moreover, from the ECHA’s guidance documents, as was noted in paragraphs 61 to 63 above, that it was necessary to provide justification for a read-across, including in a hypothesis based on the recognition of structural similarities.

68      Second, the applicant refers to the fact that, on 17 January 2017, the 2004 Waalkens-Berendsen study was added to the REACH registration dossier on DOTL, clearly showing the read-across from DOTO (see paragraphs 48 and 50 above). The applicant also refers to other studies, which clearly mention the read-across from DOTO to DOTL. While the applicant states, repeatedly, that that read-across was ‘clearly recognisable’, it is not apparent from the references made that it provided information in accordance with what was required by the REACH Regulation, and in particular Section 1.5 of Annex XI thereto (see paragraphs 57 to 63 above).

69      It is clear that the 2004 Waalkens-Berendsen study is relied on in so far as it relates to DOTO and its characteristics, since the applicant maintains that it was apparent from that study that DOTO was not toxic for reproduction. As the Commission submits, and in the absence of specific references having been made in that regard by the applicant, it was not possible to draw any conclusions from that study on the ‘likelihood of similar toxicology’ between DOTO and DOTL or to predict the properties of DOTL, without a study regarding that substance having been submitted.

70      Furthermore, as regards the same issue, so far as concerns the update of the REACH registration dossier on DOTL in September 2017, which contained certain evidence based on the hydrolysis process intended to justify the proposed read-across from DOTO to DOTL, that justification was rebutted in the CLH dossier on the basis of available studies which showed degradation products other than DOTO in simulated gastric conditions, namely at low pH.

71      As the Kingdom of Sweden points out, both KEMI and the RAC replied to the information contained in the REACH registration dossier on DOTL according to which DOTL was immediately hydrolysed to DOTO and lauric acid in water, stating that it was the degradation product formed in acidic conditions which was relevant for the purposes of assessing DOTL’s toxic effects after oral exposure and that the read-across from DOTO at low pH was not justified (KEMI stressed the differences as compared with neutral pH, citing the Nasshan studies).

72      By contrast, the applicant stated, in its observations on the Kingdom of Sweden’s statement in intervention, that the latter had admitted, first, that the approach of considering DOTO to be a degradation product of DOTL was correct at a neutral pH value and, second, as regards gastric conditions at low pH values, that there was no indication of a toxicological problem in that regard (according to the 2015 Nasshan study, see paragraph 52 above). There is thus a fundamental disagreement as to whether DOTO is a principal hydrolysis product of DOTL equally under gastric conditions. Although these arguments alone are not sufficient to establish the relevance of a read-across from DOTO to DOTL, in the absence of other required information (see paragraphs 59 and 60 above) and taking into account the presence of other products of the hydrolysis of DOTL in gastric conditions (see paragraphs 75 to 78 below), it must be held, in any event, that it was for the applicant to show that, even at a low pH value, DOTL would have been hydrolysed to DOTO in relevant volumes, contrary to KEMI’s estimations.

73      In that regard, first of all, the applicant presents before the Court an ambiguous position as regards whether the Nasshan studies referred to by KEMI satisfied the GLP conditions and met the requirements laid down by the REACH Regulation and Regulation No 1272/2008 (see paragraph 49 above). Indeed, departing from its initial criticisms, the applicant stated, in its reply to the questions posed by the Court by way of measures of organisation of procedure, in essence, that those studies could be regarded as comparable to studies carried out in accordance with GLP, on account of the various precautions taken, in so far as they concerned the physicochemical properties of DOTL and identified the hydrolysis products of that substance, in particular DOTO. However, it claims that the Nasshan studies are not suitable for classifying the human health and environmental risks of a substance. Given that the applicant’s position, presented in response to the questions posed by the Court by way of measures of organisation of procedure as regards the relevance of the Nasshan studies in terms of the evaluation of the hydrolysis of DOTL, is consistent with the position presented by the Commission and the ECHA, since the applicant is no longer reproducing its criticisms put forward in that regard in the application, those studies will subsequently be taken into account in so far as they identify hydrolysis products under simulated gastric conditions.

74      Next, in the context of the present plea, the references to the Nasshan studies are relevant in so far as, according to the applicant, the read-across from DOTO to DOTL was linked to considerations on the oral absorption of DOTL and in particular to the fact that that substance was considered to be immediately hydrolysed into DOTO and lauric acid in the gut (see paragraph 65 above and paragraph 75 below).

75      In addition to their relevance to the applicant’s argument referred to in paragraph 74 above, the Nasshan studies made it possible to assess the identity of the hydrolysis degradation products of DOTC and DOTL, respectively, at a low pH (pH 1.2, simulated gastric conditions, acidic), in the context of the read-across made by KEMI between those two substances. The identified breakdown products of DOTC and DOTL depend on the conditions of the experiment. According to the CLH dossier, at neutral and slightly acidic pH, both DOTL and DOTC hydrolyse to insoluble DOTO and to their respective ligand (lauric acid/chloride). By contrast, under simulated gastric conditions, the CLH dossier describes the breakdown as follows:

–        for DOTC: distannoxane ClOct2SnOSnOct2Cl (> 90% after 4 hours), unhydrolysed DOTC (< 10%);

–        for DOTL: distannoxane ClOct2SnOSnOct2Cl (14-16%), dioctyltin mono-chloro laurate (‘DOTLC’) (43-47%) and a non-assigned tin species (38-43%).

76      None of the evidence adduced by the applicant invalidates the abovementioned assessments, from which it is apparent that the hydrolysis product of DOTL is composed of three substances different from DOTO. In particular, as indicated in the summary and page 12 of the 2015 Nasshan study, as regards DOTL hydrolysis, the product consists, in addition to DOTLC and distannoxane, of ‘DOTO complex’, and not simply DOTO. In that regard, it is apparent from page 13 of the 2015 Nasshan study that the expression ‘DOTO complex’ describes ‘a hexane-extractable solution of DOTO in DOTLC or DOTL’ (unlike DOTO, which is not soluble in hexane). The CLH dossier further indicates that the non-assigned tin species (DOTO complex) may be associated with polymeric structures different from DOTO. That same consideration is apparent from the background document to the RAC opinion.

77      It is apparent from that study that, even though equal quantities of DOTLC and DOTO complex are present after 30 minutes (the quantities for each coinciding at 43%), even lower quantities are indicated for DOTO complex following a long incubation, after 4 hours (38% for DOTO complex compared to 47% for DOTLC).

78      Thus, the Nasshan studies do not support the approach of a read-across from DOTO to DOTL. In particular, the applicant has failed to demonstrate that, although several degradation products of a substance are formed, it is possible to conclude that no classification is warranted on the basis of a read-across from just one of those degradation products. Moreover, in its replies to the questions posed by the Court by way of measures of organisation of procedure, the applicant relies on certain ‘probabilities’ in that regard.

79      It should also be noted that the Commission correctly maintained that the information contained in the applicant’s contribution to the public consultation on the CLH dossier contradicted the proposed read-across from DOTO to DOTL, or, at least, weakened its relevance. Although the applicant claims that it had not withdrawn its proposed read-across, it is clear that it stated, in that public consultation, that ‘generating new data on the substance itself seems to be the only way to accurately characterise the reprotoxic potential and specific organ toxicity of DOTL and adequately address uncertainty of the substance classification’.

80      As regards the delay in obtaining the EOGRTS and PDTS studies (see paragraph 48 above) and the applicant’s argument that, despite the fact that those studies had not yet been completed in 2017, the RAC did not wait for their results (see paragraph 51 above), first, it must be stated that there was no reason to interpret the delay in carrying out certain studies as indicating that it was possible to ‘expect important findings with regard to reproductive toxicity’. Second, inasmuch as those studies concerned DOTO’s lack of toxic effects on reproduction, they are not directly relevant to the assessment of DOTL or the possibilities of demonstrating that a read-across between those substances was justified.

81      It should also be noted, as regards the general references made to Annexes A.20 and A.21 in order to demonstrate that the ECHA neglected relevant and important information, that, in the light of the case-law cited in paragraphs 42 to 44 above, it is not for the Court to seek and identify, in the annexes, the pleas and arguments on which it may consider the action to be based. In particular, the mere fact of setting out, without further clarification, a claim in the application, referring to the annexes for possible explanations of the arguments specifically supporting that claim, is not an approach consistent with the abovementioned case-law.

82      In those circumstances, and in the light of the case-law cited in paragraphs 30 to 35 above, the Commission did not make a manifest error of assessment in considering that the evidence submitted by the applicant regarding the read-across from DOTO to DOTL, in particular in the REACH registration dossier on DOTL, was not sufficiently relevant to the assessment in the context of the CLH dossier.

83      Lastly, and in any event, irrespective of the alleged relevance of the read-across from DOTO to DOTL, the fact remains that it is only if the Commission’s analysis of the read-across from DOTC to DOTL was also incorrect that it would be possible to conclude that there had been a manifest error of assessment as regards the harmonised labelling relating to DOTL. That matter is assessed in the context of the following pleas in law.

84      Accordingly, the first plea in law must be rejected.

 The second plea in law, alleging infringement of Articles 5, 9 and 36 of, and Parts 1 and 3 of Annex I to, Regulation No 1272/2008, in that the Commission and the other competent authorities did not take certain information into account in a ‘weight-of-evidence’ approach and thus made a manifest error of assessment

–       Arguments of the parties

85      The applicant submits that it explained, in Annexes A.20 and A.21 to the application, that the Commission and the other competent authorities had made serious mistakes regarding the scientific assessment. First, the Commission neglected relevant information from scientific literature, from basic scientific findings or from comments given by the applicant. Second, the Commission applied scientific literature or basic scientific findings incorrectly, or in a contradictory or illogical manner.

86      According to the applicant, the Commission therefore infringed Section 1.1.1.3 of Part 1 of Annex I to Regulation No 1272/2008, which states that all available information bearing on the determination of the hazard of a substance is to be considered together. It also infringed Sections 3.7.2.2 and 3.7.2.3 of Part 3 of Annex I to that regulation, in particular Section 3.7.2.3.1, which requires classification to be made ‘on the basis of an assessment of the total weight of evidence’ and which states that the weight attributed to the available data will be influenced by factors such as the quality of the studies, consistency of results and freedom from bias, and that the weight-of-evidence determination is based on both positive and negative results.

87      In the applicant’s view, the Commission also infringed Article 5(2) of Regulation No 1272/2008, which sets out the basic principles of a sound scientific assessment, namely the adequacy, reliability and scientific validity of the information. These principles are particularly important if a classification is not proposed on the basis of results of animal studies, but rather on the basis of analogy and read-across.

88      The applicant thus claims that, since the Commission and the other competent authorities in question did not correctly investigate, assess, consider and weigh the relevant available information, they infringed Article 5 (entitled ‘Identification and examination of available information on substances’), Article 9 (entitled ‘Evaluation of hazard information for substances and mixtures’) as well as Article 36 and Article 37(5) (those articles being entitled, respectively, ‘Harmonisation of classification and labelling of substances’ and ‘Procedure for harmonisation of classification and labelling of substances’) of Regulation No 1272/2008, read in conjunction with Parts 1 and 3 of Annex I to that regulation (entitled, respectively, ‘General principles for classification and labelling’ and ‘Health hazards’). In addition, by reason of the fact that the Commission merely relied on the RAC’s opinion without carrying out a sufficient assessment and without taking into account all the relevant information, it exercised its margin of discretion incorrectly and breached its duty of care.

89      The applicant considers that DOTL does not have properties that are harmful for reproduction and that only an acute adverse effect on the immune system must be taken into account. The registrant thus chose the following classification for DOTL: ‘STOT SE 2 H371’. According to the applicant, as part of the usual evaluation of the registration by the ECHA, the latter would have contacted the applicant, as the lead registrant of DOTL, if something appeared to be inconclusive, or it could have requested additional data. By the date of the present action, the DOTL dossier had not been the subject of such contact by the ECHA regarding the classification of DOTL. The applicant also claims that the ECHA did not respond to some of its testing proposals and entered into discussions on the substance identity of DOTL (also see paragraph 19 above).

90      Meanwhile, a subsidiary of the applicant, TIB Chemicals Corp. (established in Houston, Texas, United States), had commissioned, according to the applicant, a prenatal developmental toxicity study on DOTL (OECD TG 414) in order to meet the US and South Korean regulatory requirements. According to the applicant, the draft report of October 2020 relating to study No 8416027 by the independent institution Covance and the assessments of two independent institutions, namely, first, Covance’s assessment of 15 October 2020 and, second, Exponent’s assessment of 15 October 2020, prove that the DOTL should not be classified as ‘Repr. 1B’, but at most ‘Repr. 2’.

91      In the reply, the applicant maintains that, in the CLH dossier, KEMI classified DOTL as a sufficiently similar substance to DOTC in terms of toxicological effects, by basing that assessment essentially on partially similar degradation products and by taking into account their effects on reproductive toxicity. The applicant submits that it had studies carried out in order to analyse distannoxanes formed from DOTL, in which it is found that they occur solely in dimeric form, in a molecule of a mass of 1 400 daltons (Da), which cannot penetrate the cell membrane and, therefore, cannot have any toxic effects for reproduction.

92      Even though it set out that finding in its observations on the proposal for the harmonised classification and labelling of DOTL, the Commission and the RAC did not adequately acknowledge that finding. In its observations on the statement in intervention of the Republic of Austria, it adds that it was incorrect to compare dibutyltin dichloride (‘DBTC’) with dibutyltin dilaurate (‘DBTL’), since they have different modes of action.

93      The Commission, supported by the Republic of Austria, the Kingdom of Sweden and the ECHA, disputes the applicant’s arguments.

–       Findings of the Court

94      The applicant claims that the Commission and the other competent authorities did not take into account all the available information affecting the determination of the hazard of DOTL or that they applied that information incorrectly.

95      In that regard, first, the list of references in the CLH dossier demonstrates that KEMI took into consideration an extensive body of scientific evidence, including in vitro and in vivo analyses of elements such as the toxicity of the substance at issue, the hydrolysis products of certain substances in the context of the read-across approach and even the behaviour of distannoxanes (DOTL and DOTC transformation products). In addition, both the CLH dossier and the RAC’s opinion provided detailed descriptions of the available evidence, as well as extensive considerations of the relevant aspects for the classification of DOTL as toxic for reproduction category 1B. As regards the fact that the competent authorities did not consider the proposed read-across from DOTO to DOTL to be relevant information, it has already been found, in the context of the first plea, that that approach did not constitute a manifest error of assessment.

96      Second, as regards the applicant’s criticisms concerning the comparison of in vitro and in vivo studies, it should be noted that it is apparent from Annex A.13, which contains the RAC’s proposal, that classification in category 1B was not linked solely to in vitro studies, but also took into account certain in vivo analyses, as the Commission, referring to three animal studies, namely a combined reproduction screening study (OECD TG 421), an EOGRTS study (OECD TG 443) from 2011, and a PDTS study (OECD TG 414) from 2014, rightly submits. Two of these studies clearly demonstrated an adverse effect on development, which was considered not to be a secondary non-specific consequence of other toxic effects.

97      Third, as regards the applicant’s claim that the Commission applied scientific literature or basic scientific conclusions incorrectly or in a contradictory or illogical manner, it should be noted that the RAC set out the reasons why it took into consideration a read-across from DOTC to DOTL. In addition to pointing out that there were no specific data on DOTL as regards reproductive toxicity, the RAC stated that the read-across from DOTC took into account the structural similarities between the source substance and the target substance. The RAC particularly highlighted the fact that the two substances had common intermediates following hydrolysis under similar conditions to those present in the stomach. On the basis of the Nasshan studies, carried out in vitro, the RAC specified what those common intermediates of hydrolysis were, specifically referring to distannoxane (see paragraph 75 above), which enabled it to base the examination of the toxicity of DOTL on a read-across from DOTC. The RAC pointed out that the composition indicated reflected the situation after 4 hours, but that there were only minor differences compared to the situation after just 30 minutes. Furthermore, it found that although hydrolysis of both DOTC and DOTL could lead to DOTO in neutral pH conditions, the formation of DOTO was not favoured under low pH conditions.

98      Contrary to what the applicant claims, it cannot be considered that it was necessary to make references only to in vivo studies, or that preference should always be given to such studies over those carried out in vitro. Furthermore, as regards the applicant’s arguments relating to the joint reading of the Nasshan studies and of the study by Penninks, A.H. and Others, entitled ‘The absorption, tissue distribution and excretion of di-n-octyltin dichloride in rats’ (Toxicology, No 44, 1987, pp. 107-120, ed. Elsevier Scientific Publishers Ireland Ltd; ‘the Penninks study’), as well as its claims relating to the study by Meyer, A. and Others, entitled ‘Species-specific differences in the inhibition of human and zebrafish 11 ß-hydroxysteroid dehydrogenase 2 by thiram and organotins’ (Toxicology, No 301, 2012, pp. 72-78, ed. Elsevier Ireland Ltd; ‘the Meyer study’), seeking to criticise the conclusions drawn from those studies by the RAC and the Commission, they will be assessed in the context of the third plea.

99      As regards the applicant’s argument that it had studies carried out to analyse the distannoxanes (common DOTL and DOTC transformation products), in which it is found that they occurred solely in dimeric form, were unable to penetrate the cell membrane and, therefore, could not have any toxic effects for reproduction (since it was not, accordingly, bioavailable), it must be held that that matter has not been demonstrated, in the light of the contrary studies provided by the Commission, the Republic of Austria, the Kingdom of Sweden and the ECHA. More specifically, the conclusion of the document sent by the applicant to the ECHA in response to KEMI’s proposal, set out in Annex A.12, refers only to the molecular mass (1 554 Da) of the dimeric distannoxanes in question and states that that mass greatly exceeds that which still allows passage through membranes, with the result that such distannoxanes are not bioavailable. However, no source has been provided to support that information. Although a list of seven articles is given at the end of the document, none of them deals specifically with that matter. The same is true of the reference to the Harris, R.K., Sebalt, A. study (Journal of Organometallic Chemistry, 331, 1987, C9-C12), referred to on page 136 of Annex A.12, which is preceded by the statement that the dimeric structure is described therein. In any event, the graphics used and the conclusions drawn from them concern the size and structure of the substances in question, but do not relate to other elements which may possibly be relevant for assessing the bioavailability of distannoxanes (DOTL and DOTC transformation products) (see paragraph 103 below).

100    The Commission submits, in essence, that the distannoxanes present following the hydrolysis of DOTL and DOTC could be bioavailable, notwithstanding their size. In addition, it claims that, while the dimeric form of the distannoxanes proved to be predominant under the conditions used, the Davies, A.G. study, entitled ‘Difunctional distannoxanes, XR2SnOSnR2X’ (Journal of Chemical Reasearch, 2004, pp. 309-314) (‘the Davies study’) indicated that monomeric distannoxane, also referred to by KEMI as ‘the dimer with half the molecular weight’, could be present in significant concentrations at equilibrium in solution.

101    In that regard, the applicant is wrong to claim that it cannot be considered a sound scientific approach within a read-across to classify distannoxanes as bioavailable since it follows from ‘common scientific knowledge’ that there is a correlation between the size of a substance and its ability to cross a biological membrane, and that a substance does not generally have that capacity when its mass exceeds 1 000 Da.

102    The Commission is right to state that it is apparent from the ECHA Guidance on the Biocidal Products Regulation, in version 4.0 from 2017, that ‘although particles and large molecules (with molecular weights in the 1 000’s) would normally be considered too large to cross biological membranes, small amounts of such substances may be transported into epithelial cells by pinocytosis or persorption (passage through gaps in membranes left when the tips of villi are sloughed off) (Aungst and Shen, 1986)’.

103    Similarly, as is apparent from the evidence put forward by the Commission in the rejoinder and in its replies to the questions posed by the Court by way of measures of organisation of procedure, size is only one of the factors that influence the bioavailability of molecules in the intestine, the process being complex and bioavailability being difficult to predict or exclude, since account must also be taken of chemical changes resulting from metabolism. The examples given by the Commission in this regard relate to large mass molecules which, nevertheless, have significant bioavailability (cyclosporin: 1 200 Da, bioavailability of 29%; somatostatin: 1 600 Da, bioavailability of 10%). Confirmatory evidence on this issue was also submitted by the Republic of Austria in its statement in intervention. Among the factors cited, the Republic of Austria rightly refers to the fact that even the applicant’s toxicology expert, Ms Claudia Fruijtier-Pölloth, admits that dimeric distannoxanes may have some bioavailability.

104    As regards the more recent study reports from the Umweltbundesamt (Federal Environment Agency, Austria) to which the applicant referred, emphasising that they stated that ‘dimeric distannoxanes’ did not cause adverse effects, the fact remains that those reports were not submitted before the Court and cannot, therefore, be taken into consideration.

105    Accordingly, it cannot be ruled out that dimeric distannoxanes may have some bioavailability. It must be borne in mind that the RAC did indeed note, in its opinion of 14 September 2018, that the ‘dimer [was] at least in part bioavailable’. Moreover, as is apparent from the RAC’s opinion, the RAC makes reference, in the present case, to monomeric distannoxane, on the basis of its size. Concrete evidence in that regard was also provided in the RAC’s responses to the applicant’s observations of 18 September 2018. It was pointed out, with reference to the Davies study, that there was evidence of dissociation in solution, illustrating that monomeric distannoxane could be present in significant concentrations at equilibrium in solution.

106    Fourth, as regards the applicant’s criticism of the fact that the RAC took into account comparative evidence concerning an assessment of DBTC and DBTL, or of the erroneous nature of the RAC’s assessments, first of all, it should be noted that it is not apparent from the RAC’s opinion that the RAC’s comparison was intended to establish a read-across between those substances. The RAC merely took into consideration, in response to the comments of Member State competent authorities, certain analogies between those substances in the assessment of the read-across from DOTC to DOTL.

107    Next, KEMI acknowledged that, according to the Arcadis report, entitled ‘Assessment of the evaluative approach for organotin compounds under REACH’ (report prepared for the Ministerie van Infrastructuur en Milieu (Ministry of Infrastructure and the Environment, Netherlands), 2016), grouping of organotins needed to consider in greater detail the nature of the labile ligands and the chemistry associated with the relevant organotin substances. Consequently, instead of adopting a broader approach taking into account toxicological data for all dioctyltin compounds with different ligands, KEMI chose a specific read-across approach from DOTC to DOTL based on the similarity of the hydrolytic behaviour.

108    Furthermore, in so far as the applicant’s claims should be understood as criticising the scientific analysis relating to the analogy with DBTL and DBTC, based on the Milton et al. study  (Front. Pharmacol., No 8, p. 507, 2017), they must be rejected, since the applicant does not identify any manifest error of assessment on the part of the RAC. In particular, it is not apparent from its arguments that the RAC misread the abovementioned study, resulting in an incorrect assessment in so far as it relied on the fact that the structural relationship between DBTC and DBTL was similar to that between DOTC and DOTL, also taking into account the fact that those substances belong to two families of organotins of similar structure. In that regard, in order to explain the similar toxicological effects between, on the one hand, the elements containing the chlorine moiety (namely DOTC and DBTC) and, on the other hand, those including the laurate moiety (namely DOTL and DBTL), the RAC pointed out that the substitution of the first group by the second, in tin organic compounds, was generally not relevant for determining their toxicological properties.

109    Moreover, the RAC indicated that it was only for the purpose of ‘strengthening’ the read-across from DOTC to DOTL that additional information on DBTL and DBTC was taken into consideration, due to similar toxicological properties. Thus, the applicant misunderstands the RAC’s reasoning, in so far as it did not advocate a read-across from the butyl compound to the octyl compound.

110    Lastly, there is no need to analyse the studies that were finalised and submitted by the applicant in 2020 in order to meet the US and South Korean regulatory requirements, since they were not submitted in due time in the administrative procedure at issue, or the testing proposals which were allegedly not taken into consideration by the ECHA because of the exchanges that took place concerning the name of the substance concerned (see paragraphs 89 and 90 above). In that latter regard, it is apparent from the evidence only that the ECHA rejected as incomplete the applicant’s update of 25 July 2017 and that its objective was to ensure that the substance was correctly named (see paragraph 19 above). Similarly, as has been found in the context of the first plea in law (see paragraph 81 above), it should be noted, as regards the general references to Annexes A.20 and A.21 to the application, that, in the light of the case-law cited in paragraphs 42 to 44 above, it is not for the Court to seek and identify in those annexes the additional pleas and arguments on which it may consider the action to be based.

111    It must also be stated that the applicant has not substantiated its claim relating to an infringement of Article 9 of Regulation No 1272/2008. In the absence of the essential arguments in law which, pursuant to Article 21 of the Statute of the Court of Justice of the European Union, applicable to the procedure before the General Court in accordance with the first paragraph of Article 53 of that statute, and Article 76(d) of the Rules of Procedure, must appear in the application, those claims are inadmissible (see order of 13 May 2020, Lucaccioni v Commission, T‑308/19, not published, EU:T:2020:207, paragraph 34 and the case-law cited).

112    It should be noted that the applicant has not succeeded in showing that the Commission, in approving the analysis of KEMI and of the RAC, committed a manifest error of assessment and infringed Articles 5, 9, 36 and Article 37(5) of Regulation No 1272/2008, read in conjunction with Parts 1 and 3 of Annex I to that regulation, in the context of a weight-of-evidence approach, or that it failed to take account of all the relevant and available information concerning the determination of the hazard of DOTL.

113    Accordingly, the second plea in law must be rejected as in part inadmissible and in part unfounded.

 The third plea in law, alleging infringement of Regulation No 1272/2008 by applying an incorrect read-across

–       Arguments of the parties

114    According to the applicant, regardless of whether a read-across is applied under the REACH regime or under Regulation No 1272/2008, it always requires a sound scientific basis, which was not the case when the authorities applied the read-across using DOTC as a source substance. The authorities ignored the following basic scientific standards:

–        it is undisputed from a scientific standpoint, first, that only the toxicity of breakdown products is relevant, second, that the relevant breakdown product of the hydrolysis is dimeric distannoxane and, third, that, due to its size, dimeric distannoxane is not toxicologically active because it is not bioavailable;

–        the question that remains is therefore whether monomeric distannoxane is created during hydrolysis and whether monomeric distannoxane is bioavailable up to 20%, and not whether DOTC could be the bioavailable substance to that percentage; however, contrary to the suspicions of the authorities, according to the Davies study, dimeric distannoxane is stable in solution, whereas a monomeric distannoxane is not; DOTC is bioavailable and still exists in excess of 20% after one hour, which was not taken into account by the RAC or by the Commission;

–        additional evidence, provided by the expert opinion of Ms Fruijtier-Pölloth, shows that the read-across from DOTC is flawed and that the scientific quality of the respective RAC opinion is therefore insufficient.

115    The applicant submits that a read-across was only the second best method of determining toxicological properties, after testing the substance in the context of an in vivo or in vitro study. A read-across is considered to be equal to an in vivo study only for the purpose of avoiding animal studies and if it takes general scientific standards and relevant scientific literature into account. If scientific doubts occur, the read-across must be rejected or animal studies must be performed.

116    Since the read-across performed by KEMI and the RAC was not applied on a sound scientific basis and misjudged the weight of the available evidence, the authorities involved and the Commission infringed Articles 36 and 37 of Regulation No 1272/2008, read in conjunction with Section 1.1.1.3 of Part 1 and Sections 3.7.2.2 and 3.7.2.3 of Part 3 of Annex I to that regulation as well as Article 5(2) of that regulation, which contains basic principles of a sound scientific assessment.

117    In the reply, first, the applicant sets out the reasons why the read-across from DOTC to DOTL was not justified, relying in particular on the hydrolysis products of those substances and their degradation times.

118    Second, it notes the consequences of the Nasshan study as regards the hydrolysis of DOTL and DOTC, referring to the speed of the complete hydrolysis of DOTL, which takes less than 30 minutes and does not produce any DOTC.

119    Third, the applicant criticises what, in its view, constitutes incorrect citations of the Nasshan study, on the part of the Commission, in paragraph 34 of the defence, as well as in the CLH dossier and in the RAC’s opinion, in particular as regards the reference to a ‘non-assigned tin-species (38-43%)’, whereas the study referred to a ‘DOTO complex’.

120    Fourth, it invokes what it considers to be misconclusions from the Davies study regarding the assumption of the existence and relevance of monomeric distannoxanes, also referred to as ‘dimers with half the molecular weight’, as compared to ‘the dimer of dimers’. Distannoxanes are usually present in dimeric form and a monomeric form only occurs under special conditions, which does not support the assertion, in the present case, that they were decisive for reproductive toxicity.

121    Fifth, the applicant highlights alleged misconclusions drawn from the Penninks study, which was carried out in vivo on rats. The RAC used that study, which showed that 20% of DOTC administered to the vertebrate was absorbed by it, to support its hypothesis. In the Nasshan hydrolysis experiment, 10% of DOTC is degraded after 4 hours. The RAC concluded from this that tin compounds other than DOTC are also metabolised. However, it must be noted that the substance is not entirely broken down under those conditions and that bioavailable DOTC remains, and not, as the RAC noted, that ‘its metabolic products must also be bioavailable’. This assumption would only be correct in case the vertebrate does not absorb the test substance in the time between administration of DOTC and the lapse of 4 hours, and that those tin compounds formed as intermediates in both hydrolysis experiments (with DOTC as with DOTL) would be identical. However, that is not the case, according to the applicant, which criticises the assumption that the metabolic products formed from DOTL could have bioavailability and toxicity comparable to those of DOTC. It submits that the study contains only a statement on the absorbed DOTC, and nothing about the period of time during which the absorption took place or the absorbed substance (DOTC or degradation product distannoxane).

122    Sixth, it submits that the comparison between dibutyltin compounds and dioctyltin compounds is flawed. In particular, it was clearly shown during the public consultation on the harmonised classification of DBTL that dibutyltin and dioctyltin have completely different modes of action. It was also shown in the REACH registration dossier on DOTL, as well as in the Meyer study, that dibutyltin and dioctyltin compounds have different modes of action and give rise to different genetic modes of expression.

123    Seventh, it submits that the read-across from DOTO to DOTL is not invalidated by new studies.

124    Eighth, the applicant asserts that the Kingdom of Sweden is right to consider that a read-across always involves a certain degree of flexibility. However, the read-across from DOTC and DOTL is based on too many assumptions and uncertainties to be justified.

125    Lastly, ninth, in its observations on the Republic of Austria’s statement in intervention, the applicant states that only a few monomeric structures are known and that those structures are stabilised by very bulky ligands. Consequently, the degradation products are, typically, dimeric distannoxanes and not monomeric structures. This is confirmed by the findings of the Nasshan studies. Furthermore, it cannot be considered a sound scientific approach to classify distannoxanes as bioavailable, since there is a correlation between the size of a substance and its ability to cross a biological membrane.

126    The Commission, supported by the Republic of Austria, the Kingdom of Sweden and the ECHA, disputes the applicant’s arguments.

–       Findings of the Court

127    As a preliminary point, it should be noted that the applicant’s arguments put forward in the context of the third plea overlap in part with those relating to the second plea in law. Accordingly, the findings made in the context of the second plea will be taken into account in the examination of the present plea. It should also be noted that it is common ground between the parties that there is always a certain degree of flexibility in the application of the read-across method. However, the legislature has accepted that flexibility on account of the advantages, for society as a whole and for registrants, of not having to carry out costly animal studies for all substances.

128    As regards the read-across from DOTC to DOTL, Section 3.7.2.3.1 of Annex I to Regulation No 1272/2008 refers to information following from a read-across from a structurally similar substance. The abovementioned section also refers to Section 1.1.1 of Annex I to that regulation, which expressly mentions the ‘category approach (grouping, read-across)’ as one of the elements to be taken into consideration in a weight-of-evidence approach.

129    In the first place, as regards the applicant’s claim that an incorrect read-across was applied, Section 9.2 of the CLH report contains a detailed justification for the read-across from DOTC to DOTL, in accordance with the requirements of Section 1.5 of Annex XI to the REACH Regulation (see paragraphs 59 to 63 above). The RAC provided several reasons justifying the read-across beyond the structural similarity between DOTC and DOTL, including:

–        a comparison of hydrolysis products from DOTL and DOTC under different conditions (neutral/acidic);

–        toxicological properties of different dialkyltine compounds, which are closely related to DOTL and DOTC.

130    In the second place, it is necessary to reject as unfounded the applicant’s line of argument concerning the read-across from DOTC to DOTL inasmuch as that line of argument is based on the claim that the unhydrolysed DOTC (a breakdown product of DOTC, but not of DOTL), rather than distannoxane (the common breakdown product of DOTC and of DOTL), was responsible for DOTC’s toxicity for reproduction and that that was shown in the Nasshan studies (see paragraphs 118 to 121 above). Similarly, it was found that monomeric distannoxanes could be present in significant concentrations at equilibrium in solution (see paragraph 105 above).

131    First, as has been stated in paragraphs 101 to 105 above, it cannot be ruled out that there may be some dimeric distannoxanes produced in the hydrolysis of DOTC or DOTL which, despite their mass, cross cellular membranes, such that they are bioavailable and therefore have toxic effects on reproduction.

132    Furthermore, the applicant does not rule out the possibility that, at least exceptionally, distannoxanes in monomeric form, which are bioavailable, may appear. It must be held that the evidence adduced by the applicant does not render implausible the position of the RAC and of the Commission that monomeric forms occur in sufficiently large and stable quantities in the gastric environment.

133    Second, the Commission was right to maintain that the position of KEMI and of the RAC did not reflect the applicant’s assumption that the unhydrolysed DOTC (10% of DOTC was still present after 4 hours of its hydrolysis, according to the Nasshan studies) was the responsible toxic component for the effects observed with DOTC. Rather, they considered, on the basis of the Nasshan and Penninks studies, that the toxicological effects were attributable to the dimeric distannoxane, which could also dissociate and the characteristics of which were analysed in the response entitled ‘Response to comments document (RCOM) to the Opinion proposing harmonized classification and labelling at EU level of [DOTL]’. The RAC and KEMI diligently examined the question whether the toxicological effects observed in respect of DOTC in vivo (see paragraph 121 above) could be attributed to dimeric distannoxane or to unhydrolysed DOTC. KEMI did not agree that there was evidence for stating that the toxicological effects of DOTC in vivo were not attributed to the dimeric distannoxane, but to the 10% unhydrolysed DOTC.

134    On that point, the applicant’s criticisms regarding the Nasshan and Penninks studies are unconvincing. They do not rule out the possibility that, following hydrolysis, the toxic effects are at least partially linked to the fact that dimeric distannoxane is at least partially bioavailable.

135    In that regard, the Nasshan study on DOTC indicated that 90% of DOTC hydrolysed to dimeric distannoxane within 4 hours (see paragraph 75 above). It is also apparent from that study that, after 4 hours, 10% of the DOTC was not hydrolysed. The Nasshan study indicated, with regard to the absorption rate in relation to time, that the concentration of DOTC was already only 23% after one hour.

136    In the Penninks Study, which concerned the absorption of DOTC by rats, it was indicated, in the summary, that, after DOTC was administered orally to rats, 20% was absorbed by the organism and was systemically bioavailable. It is true that the Penninks study did not expressly identify the substance absorbed by the organism (unhydrolysed DOTC or degradation product distannoxane), because the measurement of absorption relied on radioactively labelled tin, which did not allow any conclusions on the nature of the absorbed molecules to be drawn. It is also common ground that that study did not refer to the precise period of time during which the absorption took place, since it concerned only daily periods.

137    In those circumstances, it cannot be ruled out, on the basis of the Penninks study, that, after the oral administration of DOTC to rats, 20% was absorbed and is bioavailable, which is higher than the 10% of unhydrolysed (and therefore bioavailable) DOTC after 4 hours, according to the Nasshan study, to which the applicant attributes the toxic effect. Therefore, it cannot be ruled out that the 20% of the degradation product absorbed, identified in the Penninks study, does not correspond solely to the unhydrolysed DOTC, but may include other degradation products, such as distannoxanes. The RAC’s assumption that a metabolite other than the unhydrolysed DOTC, namely distannoxane, is behind the toxic effect remains plausible (see paragraph 136 above), with the result that the applicant has not demonstrated a manifest error of assessment.

138    It should also be noted that the applicant acknowledges that it cannot be completely ruled out that, ‘other than DOTC’, other toxic substances may be formed during the hydrolysis in question. Contrary to what the applicant submits, it must be held that it has not been demonstrated that it is possible to disregard the potential effects of those other substances and the resulting risks, solely on the basis of the fact that, allegedly, DOTC is formed in significant quantities or is not completely degraded after hydrolysis. In addition, the applicant’s arguments are supported by assumptions relating to the lack of bioavailability of large mass molecules, which has already been rejected as not having been demonstrated.

139    Contrary to what the applicant maintains, neither KEMI nor the RAC assumed that DOTC was the probable toxic compound, but relied on the analysis that DOTL and DOTC shared a common hydrolysis product, namely distannoxane. In that regard, KEMI provided the details on which the read-across was based, in the CLH dossier, including assessments of the structural similarities between DOTC and DOTL and of their hydrolytic behaviour in neutral conditions (neutral pH) and in acidic conditions (low pH). Supporting evidence was drawn from information on comparisons of the behaviour of those substances, on the one hand, and the behaviour of DBTC and DBTL, on the other.

140    KEMI also responded to the applicant’s comments made in the context of the public consultation that the toxic effects were due to the unhydrolysed DOTC. It did not agree with the applicant that there was evidence for stating that, in vivo, the toxicological effects of DOTC were not attributed to the dimeric distannoxane, but to the 10% unhydrolysed DOTC. The RAC made the same assessment as KEMI and noted that the available information showed that dimeric distannoxane was in any event in part systematically available and therefore produced the toxicological effects observed after oral administration of DOTC. Yet, although the applicant has stated that it has a different opinion as to whether or not, in vivo, the toxicological effects observed in respect of DOTC could be attributed to dimeric distannoxane, it has not established that the authorities involved made a manifest error in the assessment of the read-across.

141    In those circumstances, its argument relating to the fact that DOTL does not produce DOTC by hydrolysis (see paragraph 118 above) must also be rejected as ineffective.

142    As regards the alleged incorrect citation (see paragraph 119 above), it has not been shown that this would have any effect on the conclusions reached by the Commission. As regards the applicant’s references to Ms Fruijtier-Pölloth’s scientific study, her analyses do not rule out the possibility of dimers being bioavailable either, since she states that ‘the dimers may form micellar structures with a polar core surrounded by lipophilic alkyl groups and may account for some bioavailability’.

143    Third, as already stated in paragraph 109 above, it was only for the purpose of ‘strengthening’ the read-across from DOTC to DOTL that additional information on DBTL and DBTC was taken into consideration, due to similar toxicological properties.

144    In any event, it does not appear plausible from the evidence submitted by the applicant that the similarities between the toxicological behaviour of dibutyltin and dioctyltin compounds ‘[had] long been considered obsolete’, such as to invalidate the RAC’s opinion. First, the evidence submitted for the first time at the stage of the replies to the questions posed by the Court by way of measures of organisation of procedure, namely the Summer, K.H., Klein, D., and Greim, H. report, entitled ‘Ecological and Toxicological Aspects of Mono- and Disubstituted Methyl-, Butyl-, Octyl-, and Dodecyltin- Compounds’ (2003) and the position of the Bundesinstitut für Risikobevertung (Federal Institute for Risk Assessment, Germany), without an express statement of reasons for the delay in submitting that evidence, as required by settled case-law (see, to that effect, judgment of 11 July 2019, BP v FRA, T‑838/16, not published, EU:T:2019:494, paragraph 132 and the case-law cited), must be regarded as being out of time and, therefore, inadmissible within the meaning of Article 85 of the Rules of Procedure. Such proposals of evidence cannot therefore compensate for the lack of evidence in the application on that point.

145    Second, as regards the Meyer study, which examined the inhibitory effect of different organotin compounds on the activity of one particular enzyme (11 β-hydroxysteroid dehydrogenase type 2), it should be noted that the applicant’s argument that the toxic effects of dibutyltin and dioctyltin compounds on reproduction in intact organisms are not similar (see also paragraph 122 above) does not demonstrate a manifest error of assessment on the part of the Commission.

146    In that regard, the RAC’s position that the read-across from DOTC to DOTL was strengthened by the analogies of toxicological behaviour between DBTC and DBTL cannot be regarded as entirely implausible in the light of the Meyer study. In particular, the role of the enzyme in question (11 β-hydroxysteroid dehydrogenase type 2) in mediating the effects of those dibutyltin and dioctyltin compounds is not apparent from the evidence adduced by the applicant or from the study in question.

147    The Meyer study provides no evidence that the inhibition of this enzyme is the only or the main mechanism for the developmental toxicity observed with different organotin compounds in animal studies. Nor is the relevance of those findings for the reproductive toxicity of different organotin compounds or the possibility of extrapolating them to DOTL apparent from that study.

148    Lastly, it should be noted, as regards the general references made by the applicant to Annexes A.20 and A.21, that, in the light of the case-law cited in paragraphs 42 to 44 above, it is not for the Court to seek and identify, in those annexes, the additional pleas and arguments on which it may consider the action to be based.

149    Accordingly, the third plea in law must be rejected.

 The fourth plea in law, alleging failure to comply with the substantive requirements of Regulation No 1272/2008 on classification, in that the Commission erred in classifying DOTL as toxic for reproduction, ‘category 1B’

–       Arguments of the parties

150    According to the applicant, the classification of a substance as ‘Repr. 1B’ requires it to fulfil the criteria set out in Section 3.7 of Annex I to Regulation No 1272/2008 for reproductive toxicity, category 1B, in accordance with Article 36(1)(d) of that regulation. Reproductive toxicity category 1B stands for ‘presumed human reproductive toxicant’ which, according to Table 3.7.1(a) of Part 3 of Annex I to Regulation No 1272/2008 requires that such classification is to be ‘largely based on data from animal studies’, which ‘shall provide clear evidence of an adverse effect’. However, when ‘there is mechanistic information that raises doubt about the relevance of the effect for humans, classification in Category 2 may be more appropriate’. Further, Sections 3.7.2.2 and 3.7.2.3 of Part 3 of Annex I to Regulation No 1272/2008 require classification to be made on the basis of the total weight of evidence which is to be influenced by different factors such as the quality of the studies or the consistency of results, or freedom from bias.

151    In its view, the burden of proving that DOTL fulfilled the abovementioned criteria lies with the Commission. Yet, neither the Commission nor the other authorities involved could provide scientific evidence establishing that DOTL was a ‘presumed human reproductive toxicant’. The other authorities in question relied on mere presumptions. Furthermore, the classification decision is not based on data from animal studies, even though such data were available by means of the read-across from DOTO. The Commission thus misjudged the weight of the available evidence.

152    Additional proof that the Commission’s classification decision was incorrect is demonstrated by the PDTS study commissioned by TIB Chemicals Corp., which shows that DOTL has no toxic effect on reproduction.

153    According to the applicant, since the Commission could not provide clear scientific evidence that DOTL fulfilled the classification for reproductive toxicity (category 1B) and STOT RE 1, it infringed Article 36(1)(d) and Article 37(5) of Regulation No 1272/2008, read in conjunction with Section 3.7 of Part 3 of Annex I to that regulation.

154    In the reply, the applicant claims that KEMI’s and the RAC’s evidence and assessments are based on assumptions and incorrect approaches, contain omissions and are not free from bias. It also maintains that, even though the Kingdom of Sweden seeks to raise doubts with respect to the new PDTS study, which was performed after the classification decision, two independent experts reached the conclusion that the classification of DOTL as ‘Repr. 1B’ was scientifically unconvincing. The results of the ‘draft study’ are confirmed by the ‘final study report’.

155    The Commission, supported by the Republic of Austria, the Kingdom of Sweden and the ECHA, disputes the applicant’s arguments.

–       Findings of the Court

156    As a preliminary point, it should be noted that the applicant’s arguments, in the context of its fourth plea, overlap significantly with those already analysed in the context of the first three pleas. Therefore, for the same reasons as those set out in the analysis of the third plea, they must be rejected.

157    Thus, it has already been held, both as regards the read-across from DOTC to DOTL and the failure to take into consideration the read-across from DOTO to DOTL, that the applicant has not demonstrated manifest errors of assessment in the scientific approaches taken by KEMI and the RAC. In addition, animal studies were available through a read-across from DOTC, and they were duly taken into account, in accordance with Annex XI to the REACH Regulation. In the light of that fact, and of the read-across made, the Commission is right to maintain before the Court that an obligation to carry out additional animal studies would not have been consistent with the principles governing the procedure for harmonised classification and labelling. In particular, according to Article 37(5) of Regulation No 1272/2008, the Commission is to adopt delegated acts, without undue delay, where it finds that the harmonisation of the classification and labelling of the substance concerned is ‘appropriate’. In the present case, the information already in its possession, based on the approach of a read-across from DOTC to DOTL, made adoption of the contested regulation possible.

158    Moreover, the applicant cannot criticise the RAC and the Commission for failing to take account of studies and expert statements which were not available at the time of the procedure for the harmonised classification and labelling of DOTL. There cannot be an obligation on the relevant authorities to await the completion of all studies which are ongoing at a given time as regards substances which are the subject of a harmonised classification and labelling procedure under Regulation No 1272/2008, since that would make it impossible to classify a substance and would thus be contrary to that regulation’s objective to ensure a high level of protection of human health and the environment.

159    Accordingly, the fourth plea in law must be rejected as unfounded.

 The fifth plea in law, alleging failure to observe the principle of proportionality inasmuch as the classification of DOTL as ‘Repr. 1B’ is neither necessary nor appropriate

–       Arguments of the parties

160    The applicant claims that the contested regulation infringes the principle of proportionality, enshrined in Article 5(4) TEU, since the classification of DOTL and the laying down of labelling obligations are not appropriate for the purpose of attaining the intended objective and there are less onerous means available.

161    In the present case, the classification of DOTL as ‘Repr. 1B’ was, according to the applicant, neither necessary nor appropriate, because the reliance on the read-across from DOTC gives no sound scientific justification for such classification, which is based only on assumptions and relies on considerations of precaution. However, in a situation when scientific uncertainties within a read-across exist which can be clarified and eliminated with the performance of an in vivo study, then the person affected must be given the opportunity to take such a measure.

162    The applicant submits that, since it is affected by the classification decision, which is binding, it must be given the opportunity of exoneration, because that constitutes a less onerous measure. Therefore, in order to be given such an opportunity, the applicant must be offered the possibility of performing an in vivo study in the proceedings under REACH. The legitimate objective of avoiding animal studies cannot, in its view, prevail in the present case. A simple reliance on the results of the read-across, which is inferior with regard to scientific robustness, and a denial of animal testing would not be appropriate and would infringe the principle of proportionality.

163    In the reply, the applicant submits, in essence, that, although there is a legitimate interest in avoiding animal testing, the principle of proportionality also demands that the applicant itself is given ‘the opportunity of exoneration, also if this can be proven under the regime of the REACH Regulation’. The generation of information on chemical substances and their properties is primarily carried out within the regime of the REACH Regulation, not within the framework of Regulation No 1272/2008. It is not proportionate to ignore that fact. Since the applicant can carry out studies and provide scientific evidence only under the REACH Regulation (not having such rights under Regulation No 1272/2008), its actions should also be taken into account in the context of the harmonised classification and labelling procedure. It refers, in that context, to the fact that the implementation of those two regulations is entrusted to the same agency (the ECHA). The more uncertain the data, the more necessary it is for the ECHA to clarify the situation by requesting additional data, even if it involves animal testing (in the light of Article 7(1) of Regulation No 1272/2008).

164    In its observations on the statement in intervention of the Republic of Austria, the applicant submits that it supports the objective of the REACH Regulation, which is to ensure a high level of protection of human health and the environment, but also emphasises the need to evaluate substances and their properties on the basis of sound scientific data. A classification decision infringes the principle of proportionality where, as in the case of DOTL, it is based on assumptions and relies on the precautionary principle, while the applicant is not given the opportunity of exoneration by performing an in vivo study in accordance with the REACH Regulation.

165    The Commission, supported by the Republic of Austria, the Kingdom of Sweden and the ECHA, disputes the applicant’s arguments.

–       Findings of the Court

166    According to settled case-law, the principle of proportionality is one of the general principles of EU law and requires that measures implemented through EU law provisions be appropriate for attaining the legitimate objectives pursued by the legislation at issue and must not go beyond what is necessary to achieve them (see judgment of 8 June 2010, Vodafone and Others, C‑58/08, EU:C:2010:321, paragraph 51 and the case-law cited).

167    In accordance with Article 37(5) of Regulation No 1272/2008, the Commission must act without undue delay where it considers that the harmonisation of the classification and labelling of the substance concerned is appropriate, in order to amend Annex VI to that regulation by the inclusion of that substance together with the relevant classification and labelling elements (Table 3.1 of Part 3 of Annex VI to Regulation No 1272/2008).

168    In the circumstances of the present case, contrary to the applicant’s submissions (see the end of paragraph 163), Article 7(1) of Regulation No 1272/2008 cannot be interpreted as requiring the ECHA to clarify the situation by requesting additional data from the applicant, by carrying out additional tests on animals. Indeed, it has already been found that such tests were neither necessary nor appropriate in situations where the assessment could be carried out in particular on the basis of a read-across (see paragraph 157 above). An obligation on the competent authorities to await the completion of all ongoing studies concerning substances subject to a harmonised classification and labelling procedure under Regulation No 1272/2008 would make it impossible to classify a substance, which is contrary to the objective of protecting human health and the environment (see paragraph 158 above).

169    Contrary to what the applicant maintains, the classification decision in the present case was not based solely on the precautionary principle, in the sense that it was based only on assumptions. As regards the specific studies carried out by the competent authorities, the applicant has not plausibly demonstrated that the measure adopted, namely the classification of DOTL as ‘Repr. 1B’, was not necessary and appropriate. The applicant is therefore not justified in claiming that the principle of proportionality would result in it being given an opportunity to be released from its obligations by conducting an in vivo study in accordance with the REACH Regulation. On the contrary, the basic principles of the REACH Regulation and Regulation No 1272/2008 seek, in principle, to avoid animal studies.

170    Accordingly, the fifth plea in law must be rejected as unfounded.

 The sixth plea in law, alleging infringement of Article 37(4) of Regulation No 1272/2008, breach of the right to be heard and failure to observe the principle of good administration

–       Arguments of the parties

171    According to the applicant, the Commission infringed Article 37(4) of Regulation No 1272/2008 and the applicant’s right to be heard, under Article 41 of the Charter of Fundamental Rights of the European Union, as well as its right to good administration. It was denied an adequate opportunity to comment meaningfully on the RAC’s opinion itself.

172    It submits that, since it was directly and individually affected by the harmonised classification and labelling of DOTL, it attempted to present its position. It was formally given the opportunity to submit observations on KEMI’s proposal, but its arguments were not taken into account in the RAC’s opinion. The applicant therefore turned to the Commission, the Ombudsman, and the Federal Institute for Occupational Health and Safety, without success.

173    The ECHA did not act in good faith and infringed its right to good administration by denying and delaying the cooperation provided for in the context of the procedure under the REACH Regulation and, consequently, undermined the applicant’s legitimate interest in proving, within the proceedings under Regulation No 1272/2008, that DOTL was of no concern.

174    According to the applicant, the ECHA rejected the implementation of the testing proposal, which it had submitted in the context of the registration procedure for DOTL under the REACH Regulation, in an inadmissible manner. The ECHA could easily have stayed the proceedings under Regulation No 1272/2008 and included the information gathered in the context of the registration procedure under the REACH Regulation.

175    In the reply, the applicant submits that its formal rights of participation under Regulation No 1272/2008 were limited to those provided for in Article 37(4) of that regulation. With its letters to the Ombudsman and to the Commission, the applicant tried every option it was aware of to be heard. It highlights alleged infringements of the right to good administration linked, first, to the fact that it had only limited rights under Regulation No 1272/2008 and, second, to the fact that the ECHA did not wait for and incorporate the results of procedures under the REACH Regulation and, moreover, refused to cooperate in that regard or delayed its cooperation.

176    The Commission, supported by the Republic of Austria, the Kingdom of Sweden and the ECHA, disputes the applicant’s arguments.

–       Findings of the Court

177    In the first place, it should be noted that, according to Article 41(2)(a) of the Charter of Fundamental Rights, 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 judgment of 19 December 2019, Probelte v Commission, T‑67/18, EU:T:2019:873, paragraph 86 and the case-law cited).

178    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 a procedural right on a person affected (see judgment of 19 December 2019, Probelte v Commission, T‑67/18, EU:T:2019:873, paragraph 87 and the case-law cited).

179    In the present case, the contested regulation lays down measures of general application, including the contested classification.

180    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 19 December 2019, Probelte v Commission, T‑67/18, EU:T:2019:873, paragraph 89).

181    In that regard, Article 37(4) of Regulation No 1272/2008 provides that the RAC is to 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 that the ECHA ‘shall forward [that] opinion and any comments to the Commission’. That article therefore provides for the right for the parties concerned to comment on the proposal for harmonised classification and labelling and to be heard in that regard before the RAC, of which the parties availed in the present case.

182    Indeed, the applicant itself states that it was formally given the opportunity to submit observations on KEMI’s proposal.

183    The applicant’s contribution to the public consultation on the CLH dossier was evaluated by KEMI and the RAC’s rapporteur, who responded to it.

184    In so far as the applicant’s criticisms should be understood as also referring to the fact that it was not able to submit observations on the RAC’s opinion, it should be noted that, although Article 37(4) of Regulation No 1272/2008 provides for the possibility of submitting comments on the proposal for harmonised classification and labelling, that regulation does not, however, provide for the possibility for the parties concerned to submit observations on the RAC’s opinion.

185    Lastly, it is apparent from the file and from the foregoing that the Commission properly assessed the RAC’s opinion.

186    Accordingly, it follows that the infringement of Article 37(4) of Regulation No 1272/2008 and breach of the right to be heard has not been demonstrated in the present case.

187    In the second place, contrary to what the applicant claims (see paragraph 173 above), there is no real basis in the file submitted to the Court for asserting that the ECHA did not act in good faith, or that it infringed the applicant’s right to good administration ‘by denying and delaying the cooperation in the proceedings under the REACH Regulation’ or by deliberately entering into exchanges on the identity of DOTL.

188    In that regard, under Article 41(1) of the Charter of Fundamental Rights, relating to the right to good administration, every person has the right to have his or her affairs handled impartially, fairly and within a reasonable time by the institutions, bodies, offices and agencies of the Union. According to the case-law, it is for the administration, in accordance with that principle, to examine carefully and impartially all the relevant aspects of a case and to gather all the factual and legal information necessary to exercise its discretion and to ensure the proper conduct and effectiveness of the procedures which it sets in motion (see judgment of 28 June 2016, AF Steelcase v EUIPO, T‑652/14, not published, EU:T:2016:370, paragraph 57 and the case-law cited).

189    First, it has already been found that the ECHA did take into consideration, in accordance with its obligations under Regulation No 1272/2008, all the relevant scientific studies, and it cannot be criticised for not having waited until certain other studies, that were still ongoing in the context of the registration procedure for DOTL under the REACH Regulation, had been finalised. Second, the facts relating to an evaluation of the name of the substance at issue did not constitute evidence of bad faith, as the applicant appears to claim, but established the existence of a detailed analysis carried out on the part of the ECHA.

190    Accordingly, the sixth plea in law must be rejected as unfounded.

 The seventh plea in law, alleging failure, on the part of the Commission, to fulfil its obligation to carry out an impact assessment before adopting the contested act

–       Arguments of the parties

191    The applicant claims, in essence, that, in adopting the contested regulation without having first carried out and documented an impact assessment, the Commission breached its commitments under the Interinstitutional Agreement between the European Parliament, the Council of the European Union and the European Commission on Better Law-Making of 13 April 2016 (OJ 2016 L 123, p. 1; ‘the Interinstitutional Agreement’), specifically point 13 thereof.

192    It should be noted that, according to the applicant, a classification of DOTL as ‘Repr. 1B’ would have a considerable impact on the use of that substance in so-called XPE drinking water pipes, that is to say, cross-linked polyethylene drinking water pipes, in which DOTL has been used for decades. Such a classification would result in a phase-out of that use, while substitutes are probably not currently available. Those consequences would have been taken into account if an impact assessment had been carried out.

193    The applicant submits that the Commission was required to carry out such an impact assessment under the second sentence of Article 295 TFEU, which states that interinstitutional agreements may have binding effect. Point 13 of the Interinstitutional Agreement indicates its intention to create legal effects for the future when it provides that the Commission ‘will carry out’ impact assessments. Thus, and in contrast to point 16 of the agreement which states that the Commission ‘may’ supplement the impact assessment, it has no discretion in that regard. The applicant can invoke that binding effect of the Interinstitutional Agreement, the objective of which, according to recital 2 thereof, is to strengthen transparency and safeguard the rights of EU citizens and the competitiveness of businesses. Furthermore, the obligation to carry out impact assessments is a general principle of sound legislation, emphasised by Article 191(3) TFEU, which concerns environmental law regulations.

194    The applicant maintains in the reply, first, that the Commission’s argument that there is a conflict between the Interinstitutional Agreement, as presented by the applicant, and Regulation No 1272/2008 is not convincing. Point 13 of the agreement explicitly refers to environmental impacts, not only to social and economic impacts. Moreover, the reduction of environmental effects is also one of the main objectives of Regulation No 1272/2008. Second, the reference to an ‘undue delay’ in Article 37(5) of that regulation, as regards the adoption of delegated acts, does not overrule the procedural requirement laid down in point 13 of the Interinstitutional Agreement. Third, the use, in the Interinstitutional Agreement, of the verbs in the future (in particular ‘will proceed’) does not mean that the Commission was not required to carry out an impact assessment, but rather that that was an unconditional obligation. Fourth, the Court of Justice merely stated, in paragraphs 82 and 85 of the judgment of 3 December 2019, Czech Republic v Parliament and Council (C‑482/17, EU:C:2019:1035), that there was no obligation to carry out an impact assessment in every circumstance, and it put an emphasis on the question whether the EU legislature was in a particular situation and whether it was able to exercise its discretion properly. Fifth, the three European institutions that are parties to the Interinstitutional Agreement are not the ‘beneficiaries’ of the requirement to perform an impact assessment, but are bound by that requirement and the absence of an impact assessment could render an EU act invalid.

195    The Commission contends that that plea is unfounded.

–       Findings of the Court

196    In that regard, it should be noted that point 13 of the 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.’

197    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 (judgment of 3 December 2019, Czech Republic v Parliament and Council, C‑482/17, EU:C:2019:1035, paragraph 82).

198    Those points show, first, 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. Second, 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. Third, 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. Fourth, 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 (judgment of 3 December 2019, Czech Republic v Parliament and Council, C‑482/17, EU:C:2019:1035, paragraph 83).

199    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 (judgment of 3 December 2019, Czech Republic v Parliament and Council, C‑482/17, EU:C:2019:1035, paragraph 84).

200    In addition, the Court of Justice has held that not carrying out an impact assessment cannot be regarded as a breach of the principle of proportionality where the EU legislature is in a particular situation requiring it to be dispensed with and has sufficient information enabling it to assess the proportionality of an adopted measure (judgment of 3 December 2019, Czech Republic v Parliament and Council, C‑482/17, EU:C:2019:1035, paragraph 85).

201    In the present case, it should be noted that, as is apparent from the case-law cited in paragraph 197 above, it does not follow from point 13 of the Interinstitutional Agreement that the Commission is required, in all circumstances, to carry out an impact assessment of its delegated acts.

202    The applicant merely claims, in paragraph 90 of the application, that classification of DOTL as ‘Repr. 1B’ would have a considerable impact on the use of that substance in so-called XPE drinking water pipes (that is to say cross-linked polyethylene drinking water pipes), in which DOTL has been used for decades. Such a classification would, according to the applicant, result in the phase-out of that use, while currently substitutes are probably not available.

203    It must be stated that the applicant relies on matters that have not been proven and which refer, moreover, to a specific issue, relating to a certain type of drinking water pipes, which do not make it possible to establish, without further explanation, significant economic, social or environmental consequences such as to require the Commission to carry out an impact assessment (see paragraph 198 above).

204    It should be noted, moreover, that such an obligation does not follow from the provisions of Article 37 of Regulation No 1272/2008, governing the procedure for harmonised classification and labelling, either, and that, on the contrary, such an analysis is not provided for at any of the stages of that procedure. According to Article 37(5) of that regulation, the Commission is to adopt delegated acts without undue delay where it finds that the harmonisation of the classification and labelling of the substance concerned is ‘appropriate’. 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’s opinion and, lastly, the observations made during the public consultations, in accordance with Article 37(2) and (4), although those elements, in particular the RAC’s opinion, are not binding on the Commission.

205    Therefore, in the context of the procedure at issue and subject to the findings made in paragraphs 196 to 203 above, the exercise of the Commission’s delegated power is based, in essence, on scientific data justifying a harmonised classification and labelling decision, both where it follows the RAC’s opinion and in the event that the Commission adopts a decision different from that proposed in that opinion.

206    It follows from the foregoing that the Commission was under no obligation to carry out an impact assessment under point 13 of the Interinstitutional Agreement in the context of the procedure for harmonised classification and labelling which led to the adoption of the contested regulation.

207    It follows that the applicant’s seventh plea in law must be rejected and the action must be dismissed in its entirety.

 Costs

208    Under Article 134(1) of the Rules of Procedure of the General Court, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the successful party’s pleadings. Since the applicant has been unsuccessful, it must be ordered to bear its own costs and to pay those incurred by the Commission.

209    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. According to Article 100 of the REACH Regulation, the ECHA is a body of the European Union. It follows that the Republic of Austria, the Kingdom of Sweden and the ECHA must bear their own costs.

On those grounds,

THE GENERAL COURT (Second Chamber)

hereby:

1.      Dismisses the action;

2.      Orders TIB Chemicals AG to bear its own costs and to pay those incurred by the European Commission;

3.      Orders the Republic of Austria, the Kingdom of Sweden and the European Chemicals Agency (ECHA) to bear their own costs.

Marcoulli Frimodt Nielsen Schwarcz

Delivered in open court in Luxembourg on 5 July 2023.

V. Di Bucci S. Papasavvas

Registrar President

Table of Contents

Legal background

Background to the dispute

Forms of order sought

Law

Admissibility of the action inasmuch as the applicant disputes the lawfulness of the classification of DOTL as a specific target organ toxicant, repeated exposure category 1 (STOT RE 1)

The scope of judicial review

The substance

The admissibility of the applicant’s first four pleas

Admissibility of the applicant’s reference to the various annexes

The first plea in law, alleging infringement of Article 37(1) and (5) of, and Part 2 of Annex VI to, Regulation No 1272/2008, in that KEMI, the RAC and the Commission committed a manifest error of assessment in rejecting the proposal to perform a read-across from DOTO to DOTL

– Arguments of the parties

– Findings of the Court

The second plea in law, alleging infringement of Articles 5, 9 and 36 of, and Parts 1 and 3 of Annex I to, Regulation No 1272/2008, in that the Commission and the other competent authorities did not take certain information into account in a ‘weight-of-evidence’ approach and thus made a manifest error of assessment

– Arguments of the parties

– Findings of the Court

The third plea in law, alleging infringement of Regulation No 1272/2008 by applying an incorrect read-across

– Arguments of the parties

– Findings of the Court

The fourth plea in law, alleging failure to comply with the substantive requirements of Regulation No 1272/2008 on classification, in that the Commission erred in classifying DOTL as toxic for reproduction, ‘category 1B’

– Arguments of the parties

– Findings of the Court

The fifth plea in law, alleging failure to observe the principle of proportionality inasmuch as the classification of DOTL as ‘Repr. 1B’ is neither necessary nor appropriate

– Arguments of the parties

– Findings of the Court

The sixth plea in law, alleging infringement of Article 37(4) of Regulation No 1272/2008, breach of the right to be heard and failure to observe the principle of good administration

– Arguments of the parties

– Findings of the Court

The seventh plea in law, alleging failure, on the part of the Commission, to fulfil its obligation to carry out an impact assessment before adopting the contested act

– Arguments of the parties

– Findings of the Court

Costs

*      Language of the case: English.