Language of document : ECLI:EU:C:2019:710

OPINION OF ADVOCATE GENERAL

HOGAN

delivered on 11 September 2019(1)

Case C178/18 P

MSD Animal Health Innovation GmbH,

Intervet international BV

v

European Medicines Agency (EMA)

(Appeal — Access to documents — Regulation (EC) No 1049/2001 — Documents held by the European Medicines Agency (EMA) and submitted in the context of an application for marketing authorisation for a veterinary medicinal product for veterinary use — Decision to give a third party access to the documents — General presumption of confidentiality — Article 4(2) — Exception relating to the protection of commercial interests — Article 4(3) — Protection of decision-making process)






I.      Introduction

1.        The present appeal concerns one of two appeals in which a party attempts to overturn a decision as a result of which access to documents has been granted by a European institution or agency. By their appeal, MSD Animal Health Innovation GmbH (‘MSD’) and Intervet international BV (‘Intervet’) (‘the appellants’) ask the Court to set aside the judgment of the General Court of the European Union of 5 February 2018, MSD Animal Health Innovation and Intervet international v EMA (T‑729/15, EU:T:2018:67, ‘the judgment under appeal’). In that judgment the General Court dismissed the appellants’ application for annulment of the European Medicines Agency (‘EMA’) decision EMA/785809/2015 of 25 November 2015 (‘the contested decision’). That decision granted access to toxicity study reports containing information submitted in the context of an application for marketing authorisation (‘MA’) for the veterinary product Bravecto on the basis of Regulation (EC) No 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents. (2) The requester of access to these documents is another pharmaceutical company which is or might be a competitor of the appellants.

2.        The central question presented by this appeal is whether toxicity study reports prepared as part of an application for an MA in respect of new veterinary products to the EMA constitute commercially confidential information which is protected by the first indent of Article 4(2) of Regulation No 1049/2001. This appeal was heard together with the appeal in Case C‑175/18 P, PTC Therapeutics International v European Medicines Agency.

3.        Both appeals are the first appeals where the question of access to documents submitted in the course of MA applications has come for consideration before this Court. The importance of these appeals in respect of the European Union’s regime on access to documents regime and the pharmaceutical and veterinary products industries cannot, I think, be overstated. While the present appeal concerns veterinary products — as distinct from the pharmaceutical products for human use at issue in Case C‑175/18 P, PTC Therapeutics International v EMA — the legal issues raised in both appeals are, to all intents and purposes, so similar as to be all but identical. As these issues relating to the operation of the freedom of information regime provided for in Regulation No 1049/2001 are, however, dealt with at greater length in my Opinion already delivered this morning in the companion case C‑175/18 P, PTC Therapeutics International v EMA, the present Opinion should really be read in conjunction with that earlier Opinion.

4.        Before considering any of these detailed legal issues, it is first necessary to set out the relevant legal provisions.

II.    Legal context

 International law

Agreement on Trade-Related Aspects of Intellectual Property Rights (‘TRIPs Agreement’)

5.        Article 39(3) of the TRIPs Agreement which is part of the Agreement establishing the World Trade Organisation (WTO), signed by the European Community and subsequently approved by Council Decision No 94/800/EC of 22 December 1993 (3) provides as follows:

‘3. Members, when requiring, as a condition of approving the marketing of pharmaceutical or of agricultural chemical products which utilise new chemical entities, the submission of undisclosed test or other data, the origination of which involves a considerable effort, shall protect such data against unfair commercial use. In addition, Members shall protect such data against disclosure, except where necessary to protect the public, or unless steps are taken to ensure that the data are protected against unfair commercial use.’

 EU law

A — Regulation No 1049/2001

6.        Article 4 of Regulation No 1049/2001 provides for exceptions to the right of access to documents. Paragraphs 2, 3 and 6 are relevant for the present case. They provide as follows:

‘2. The institutions shall refuse access to a document where disclosure would undermine the protection of:

–        commercial interests of a natural or legal person, including intellectual property,

–        court proceedings and legal advice,

–        the purpose of inspections, investigations and audits,

–        unless there is an overriding public interest in disclosure.

3. Access to a document, drawn up by an institution for internal use or received by an institution, which relates to a matter where the decision has not been taken by the institution, shall be refused if disclosure of the document would seriously undermine the institution's decision-making process, unless there is an overriding public interest in disclosure.

Access to a document containing opinions for internal use as part of deliberations and preliminary consultations within the institution concerned shall be refused even after the decision has been taken if disclosure of the document would seriously undermine the institution’s decision-making process, unless there is an overriding public interest in disclosure.

6. If only parts of the requested document are covered by any of the exceptions, the remaining parts of the document shall be released.’

B — Regulation (EC) No 726/2004 (4)

7.        Article 39(10) refers to Article 13 and Article 13a of Directive 2001/82/EC of the European Parliament and of the Council of 6 November 2001 on the Community code relating to veterinary medicinal products.

8.        Article 73 provides:

‘Regulation [ No 1049/2001] shall apply to documents held by the Agency.

The Agency shall set up a register pursuant to Article 2(4) of Regulation [ No 1049/2001] to make available all documents that are publicly accessible pursuant to this Regulation.

The Management Board shall adopt the arrangements for implementing Regulation [No 1049/2001] within 6 months of entry into force of this Regulation.

…’

C — Directive 2001/82 (5)

9.        The first subparagraph of Article 12(3), in particular its point (j), has the following wording/is worded as follows:

‘The application for marketing authorisation shall include all the administrative information and scientific documentation necessary for demonstrating the quality, safety and efficacy of the veterinary medicinal product in question. The file shall be submitted in accordance with Annex I and shall contain, in particular, the following information:

(j) results of:

–        pharmaceutical (physico-chemical, biological or microbiological) tests,

–        safety tests and residue tests,

–        pre-clinical and clinical trials;

–        tests assessing the potential risks posed by the medicinal product for the environment. This impact shall be studied and consideration shall be given on a case-by-case basis to specific provisions seeking to limit it.

…’

10.      Article 13(1) (6) provides as follows:

‘By way of derogation from point (j) of the first subparagraph of Article 12(3), and without prejudice to the law relating to the protection of industrial and commercial property, the applicant shall not be required to provide the results of the safety and residue tests or of the pre-clinical and clinical trials if he can demonstrate that the medicinal product is a generic of a reference medicinal product which is or has been authorised under Article 5 for not less than eight years in a Member State or the Community.

A generic veterinary medicinal product authorised pursuant to this provision shall not be placed on the market until ten years have elapsed from the initial authorisation of the reference product.

The first subparagraph shall also apply when the reference medicinal product was not authorised in the Member State in which the application for the generic medicinal product is submitted. In this case, the applicant shall indicate in the application the Member State in which the reference medicinal product is or has been authorised. At the request of the competent authority of the Member State in which the application is submitted, the competent authority of the other Member State shall transmit, within a period of 1 month, confirmation that the reference medicinal product is or has been authorised together with the full composition of the reference product and if necessary other relevant documentation.

…’

A.      Facts

11.      The appellants both belong to the global animal health business unit of Merck & Co., Inc. The first appellant, MSD, was the sponsor of the toxicology studies underpinning the study reports which were submitted in the form of detailed non-clinical study reports to the EMA when Intervet applied for an MA for the veterinary medicinal product ‘Bravecto’. Bravecto consists of chewable tablets used to treat tick and flea infestations in dogs.

12.      An application for an MA for Bravecto was submitted to the EMA on 20 November 2012. That application was granted by the European Commission on 11 February 2014 on the basis of the opinion of the EMA drawn up by its Committee for Medicinal Products for Veterinary Use.

13.      On 7 August 2015, the EMA received a request for access to documents in accordance with Regulation No 1049/2001. The request referred to five toxicity study reports contained in the MA application file. The EMA decided to give access to three of those reports, namely Study C45151 (28-day dermal toxicity study in rats), Study C88913 (28-day dermal toxicity study in rats), and Study C45162 (28-day oral toxicity study in rats). I propose for convenience to describe these as either ‘the batch 1 study reports’ or ‘the reports at issue’. As I have already stated, the request for access came from another pharmaceutical company.

14.      By email dated 24 August 2015, the EMA informed the appellants of its decision to allow access and sent the batch 1 study reports as redacted by the EMA to the appellants for their comments, including justifications for any redactions they might propose. The appellants proposed further redactions on the basis of the first indent of Article 4(2) and/or Article 4(3) of Regulation No 1049/2001.

15.      The EMA only accepted redactions regarding details on the concentration range of the active substance and on the internal reference standard used for the analytical tests, as well as references to future development plans. By its decision EMA/671379/2015 of 9 October 2015, the EMA informed the appellants that it had decided to grant access to the requester and enclosed the documents in the form in which it intended to grant access.

16.      By email of 19 October 2015, the appellants objected to the fact that the ‘vast majority’ of their redactions had been rejected by the EMA and argued for the first time that each of the reports at issue fell under a general presumption of confidentiality.

17.      As further correspondence and a conference call could not resolve the parties’ differences, the EMA took the contested decision. It decided to uphold its decision of 9 October 2015 to grant access to the reports at issue as redacted. Attached to that decision were justification tables that had been prepared by the EMA taking into account the updated comments of the appellants.

18.      The EMA maintained that the appellants’ comments were neither specific to the proposed redactions nor of a scientific nature. It argued that the presumption of confidentiality claimed by the appellants for the reports at issue was contrary to the transparency provisions of the TFEU and the provisions of Regulation No 1049/2001. It pointed out that any potential unfair commercial use of the information would be prevented due to data and market exclusivity provisions contained in Article 13 and Article 13a of Directive 2001/82. The EMA further concluded — for essentially the same reasons — that the protection afforded by Article 39 of the TRIPS Agreement was not undermined by the release of the information. In so far as the appellants had argued that there might be a circumvention of the existing data and market exclusivity, the EMA found that this was always a risk and referred to other EU and national legislation providing for remedies in that event. It considered that to refuse access on these grounds would result in practice in an almost full paralysis of the access to documents as provided for in Regulation No 1049/2001. The EMA stated further that it only considered information as commercially confidential if such information is not publicly available or otherwise accessible from publicly available sources. At the same time, it noted that it must be demonstrated that access to such information would undermine the economic interest or competitive position of the appellants.

19.      The appellants challenged the contested decision in an action for annulment before the General Court. By a separate document, they brought an application for interim relief. This was granted by order of the President of the General Court. (7) The EMA’s appeal against that order was dismissed by order of the Vice-President of the Court of 1 March 2017. (8)

III. The judgment under appeal

20.      In support of their action before the General Court, the appellants raised five pleas, alleging that (1) the reports at issue are protected by the first indent of Article 4(2) or Article 4(3) of Regulation No 1049/2001 pursuant to a general presumption of confidentiality, (2) the reports at issue constitute commercially confidential information protected by the first indent of Article 4(2) of Regulation No 1049/2001, (3) the reports at issue are protected by Article 4(3) of Regulation No 1049/2001 on the ground that their disclosure would undermine the EMA’s decision-making process, (4) no balancing exercise has been carried out in respect of the relevant interests and, (5) no proper balancing exercise has been carried out in respect of the competing interests. The General Court dismissed the action in its entirety on the following grounds.

 General presumption of confidentiality

21.      The General Court held that no general presumption of confidentiality existed in respect of the batch 1 study reports. It reached this view for the following reasons.

22.      Applying criteria identified by this Court for recognising a general presumption, (9) it found that none of those criteria were fulfilled in the present case. First, the batch 1 study reports did not relate to an ongoing administrative or judicial procedure so that the objectives that a general presumption might serve, namely, the overriding need to ensure that the procedures at issue operate correctly and to guarantee that their objectives are not jeopardised, were not at issue here. (10) The fact that data could be reused in connection with new MA applications did not change this assessment. (11) Second, the General Court found that Regulation No 726/2004, which contains the rules applicable to an MA procedure, did not contain specific rules restricting access to documents. 

23.      The General Court also noted that the EMA had adopted document EMA/110196/2006 entitled ‘European Medicines Agency policy on access to documents (related to medicinal products for human and veterinary use)’ as well as the policy document EMA/127362/2006. That policy document contains a table, which is not legally binding, which gives guidance for the application of Regulation No 1049/2001 to categories of documents. This table is updated on a continuous basis taking into account the interpretation given by this Court on Regulation No 1049/2001 as well as further experiences gained by the EMA. (12)

24.      The General Court further held that the fact that the batch 1 study reports were drawn up by the appellants rather than being derived from the assessment report that the EMA had drawn up does not justify greater protection. (13)

25.      It also expressed doubts as to whether non-clinical and clinical studies were in themselves confidential as they may be limited to satisfying a regulatory scheme prescribed by the EMA and contain no new material. (14) The General Court held that the provisions of Article 39(2) and (3) of the TRIPS Agreement could not be relied on in favour of a general presumption of confidentiality as they do not give absolute precedence to the protection of intellectual property rights over the principle of disclosure. (15) It observed that the protection of data according to Article 39(10) of Regulation No 726/2004 (16) and the exception contained in the first indent of Article 4(2) of Regulation No 1049/2001 which provides for the protection of commercially confidential information contained in an MA application fulfilled the requirements under Article 39(3) of the TRIPS Agreement. (17) For the very same reason it held that the disclosed data was sufficiently protected against misuse, the possibility of third parties using the documents to circumvent data exclusivity in breach of those provisions notwithstanding. (18)

26.      The General Court also confirmed the EMA’s position that the fact that data might be reused in connection with new MA applications does not in itself constitute a ground to consider that that information is confidential or that its disclosure might undermine the decision-making process within the meaning of Article 4(3) of Regulation No 1049/2001. (19)

 The specific application of the first indent of Article 4(2) of Regulation No 1049/2001 in the present case

27.      In addition to the arguments that were relevant and assessed with regard to the general presumption and that are also relevant for the specific assessment, the General Court found, on the basis of a strict interpretation of the exception contained in the first indent of Article 4(2) of Regulation No 1049/2001, that it must be shown that the reports at issue contain elements which may, if access was granted to them, ‘seriously’ undermine the commercial interests of the appellants.(20) It further held that an assessment on whether the EMA has infringed the first indent of Article 4(2) of Regulation No1049/2001 must be made, amongst others, in the light of this. (21) It further held that for the entire batch 1 study reports to be covered by the exception in the first indent of Article 4(2) of Regulation No 1049/2001, all the information contained therein would have to be commercially confidential information. The Court nevertheless considered the arguments adduced by the appellants in this regard, namely that those reports disclose regulatory know-how, clinical assessment abilities and MSD’s innovative strategic approach to running its safety studies, to be vague and unsubstantiated. (22) It further held that ‘commercial interests’ as referred to in the first indent of Article 4(2) of Regulation No 1049/2001 are not necessarily determined by reference to the financial value of the information to which access is to be granted. (23)

28.      The General Court also found that the appellants could not rely on the exception provided for in the first subparagraph of Article 4(3) of Regulation No 1049/2001 because the procedure for granting an MA for Bravecto was closed at the time that the request for access to the reports at issue was made. (24) Moreover, the appellants had not submitted any evidence showing in what way the EMA’s decision-making process would be seriously undermined if the reports were relied on for future applications, particularly as the EMA ensured, that the data not relating to the already authorised indication and the data relating to future development projects would remain confidential. (25)

29.      With regard to the appellants’ pleas that no proper balancing of the diverging interests had taken place, the General Court addressed two scenarios. If the appellants’ argument was aimed at the balancing of its interests in keeping the reports at issue confidential with the ‘overriding public interest in disclosure’ mentioned in Article 4(2) and (3) of Regulation No 1049/2001 which becomes relevant only after the existence of commercially confidential information had been ascertained, this exception was not applicable here as, indeed, the batch 1 study reports had not been recognised as commercially confidential information. (26) If, on the other hand, the appellants’ argument required the EMA to balance the appellants’ interest in confidentiality against the public interest in disclosure already at the time of establishing whether data constitutes commercially confidential information, the General Court considered that this would be contrary to Article 73 of Regulation No 726/2004. That provision leads to the application of Regulation No 1049/2001 which only allows for exceptions to the obligation of transparency in the cases set out in Article 4(2) and (3) of Regulation No 1049/2001. (27)

IV.    The appeal

30.      The appellants put forward five grounds in support of their appeal. By their first ground, they claim that the General Court erred in law by not recognising the existence of a general presumption of confidentiality with respect to the batch 1 study reports. The second ground alleges an infringement of the first indent of Article 4(2) of Regulation No 1049/2001. In their third ground of appeal, the appellants claim that the batch 1 study reports should be protected under Article 4(3) of Regulation No 1049/2001. By their fourth and fifth grounds, the appellants allege that the General Court has erred in law when, given that the first, second and third pleas in law had demonstrated that Article 4(2) and (3) of Regulation No 1049/2001 was engaged, the EMA should have balanced these private interests with overriding public interests in disclosure, if any. The appellants claim that, had the EMA done so, it would have found that there are no overriding public interests in favour of disclosure.

31.      In line with the request by the Court, I propose to confine my Opinion to the first, second and third grounds of appeal.

V.      Assessment

A.      First ground of appeal: infringement of the first indent of Article 4(2) of Regulation No 1049/2001 due to a failure to recognise a general presumption of confidentiality for toxicity study reports such as the batch 1 study reports

32.      In their first ground of appeal, the appellants argue that the General Court erred in law in so far as it rejected the submission that the batch 1 study reports were protected by a general presumption of confidentiality.

1.      Arguments of the parties

33.      In the appellants’ opinion, the General Court misapplied the criteria for recognising the existence of a general presumption of confidentiality.

34.      The appellants agree with the General Court that one of the criteria for recognising a general presumption in respect of the exceptions contained in Article 4(2) and (3) of Regulation No 1049/2001 is that there are specific rules relating to access to documents in the procedure to which the documents pertain. In their opinion, however, the General Court misconstrued Article 73 of Regulation No 726/2004 as being such a provision. Rather, they consider that Article 13 of Directive 2001/82 should be taken into account. Article 13(1) of Directive 2001/82 (28) describes the generic approval route for medicinal products. According to the appellants, that article is based on the assumption that competitors never get access to the reports, as they are not needed for an MA of a generic medicinal product. They further infer that it is an inherent and essential characteristic of the MA regime that the documents submitted as part of an MA dossier to which the toxicity study reports belong must be kept confidential. (29)

35.      The appellants point out that the public interest is satisfied by the disclosure requirements contained in Regulation No 726/2004 and in particular by the publication of the European Public Assessment Report (‘EPAR’) pursuant to Article 38(3) of Regulation No 726/2004. (30) That publication requirement and the rules described in point 34 of this Opinion strike a balance between the interests of stimulating research by protecting first movers, the need to stimulate competition in research and marketing of products, the provision of an MA system in which the applicants are encouraged to disclose information, and the desire to limit testing on animals. In their opinion, the effet utile of Regulation No 726/2004 thus requires the recognition of a general presumption of confidentiality in order to comply with that balance. (31)

36.      The appellants further argue that the General Court erred in law when it made the following suggestions with regard to cases in which a general presumption of confidentiality had been recognised by the Court: (i) access to documents was typically limited to ‘parties concerned’ or ‘complainants’ and (ii) all those cases involved ‘ongoing administrative or judicial proceedings’ in so far as this implied that they are necessary criteria. On the other hand they claim that the General Court failed to take into account the fact that a general presumption of confidentiality may be applied where the documents concerned contain commercially sensitive information by their nature. (32) In this respect they claim that data which may be used unfairly must be considered confidential. (33)

37.      The appellants further reproach the General Court for relying on the EMA’s policy documents as sources of law. (34)

38.      The appellants argue that the General Court erred in holding that any general presumption of confidentiality ends once the MA decision has been adopted. Rather, the possibility of the information being reused should have been taken into consideration.

39.      Furthermore, the appellants state that an interpretation of Regulation No 1049/2001 consistent with the TRIPs Agreement would have led to the recognition of a general presumption of confidentiality. (35)

40.      The EMA contests all of these arguments. In particular, it claims that none of the criteria relevant in the past for the recognition of a general presumption of confidentiality are fulfilled and that the way in which the guarantee against ‘unfair commercial use’ within the meaning of Article 39(3) of the TRIPs Agreement is envisaged is the granting of a certain period of data exclusivity. It thus maintains that the provisions of Article 39(10) of Regulation No 726/2004 fulfil the requirements of the TRIPs Agreement. Moreover, the first indent of Article 4(2) of Regulation No 1049/2001 effectively protects commercially confidential information against unfair use. (36) However, in the EMA’s opinion, not the entirety of the content of the batch 1 study reports or indeed the entirety of toxicity study reports in general constitute commercially confidential information.

2.      Assessment of arguments in relation to a general presumption of confidentiality

(a)    Purported reliance on the EMA’s internal policy documents

41.      The appellants’ claim that the General Court erred in law by relying on the EMA’s own policy document as a source of law is incorrect. I agree that if the General Court had treated this document as a source of law and had based its decision accordingly, this would have been an error of law. But is this really what happened before the General Court?

42.      According to settled case-law, a complaint directed against a ground included in a decision of the General Court purely for the sake of completeness cannot in itself lead to the decision being set aside and is therefore nugatory. (37) While the General Court mentioned that the EMA had adopted arrangements pursuant to Article 73 of Regulation No 726/2004 for implementing Regulation No 1049/2001 and that it had adopted the document ‘European Medicines Agency policy on access to documents (related to medicinal products for human and veterinary use)’, (38) as well as a document entitled ‘Output of the European Medicines Agency policy on access to documents related to medicinal products for human and veterinary use’, (39) it is apparent that the General Court did not rely on these documents in its reasoning.

43.      The General Court held in paragraph 38 of the judgment under appeal that ‘in view of all the foregoing, it must be held that there is no general presumption of confidentiality …’. The use of the word ‘moreover’ at the beginning of paragraph 39 of the judgment makes it clear that the General Court did not base its reasoning on the EMA’s policy rules described thereafter.

44.      In summary, therefore, it is clear that the documents referred to in paragraphs 39 and 40 of the judgment under appeal are only mentioned for the sake of completeness. They are not central to the rationale for the General Court’s ruling.

(b)    General presumption of confidentiality

45.      In points 36 to 40 of my Opinion on the appeal in Case T‑718/15, PTC Therapeutics v EMA I have set out some general principles regarding the freedom of information regime which are also relevant here. In that Opinion I expressed the view that an institution to which a request for access to documents is directed and which wishes to refuse such access must explain how disclosure of the document in question could specifically and actually compromise the interest protected by the exception. (40) I have also set out in that Opinion that, according to the case-law, it is open to the EU institution concerned to base its decision in that regard on general presumptions which apply to certain categories of documents, as considerations of a generally similar kind are likely to apply to requests for access relating to documents of the same nature. (41)

46.      The General Court found that a general presumption of confidentiality could not be recognised because none of the criteria that were present in other cases in which a general presumption of confidentiality had been recognised were present in the case at issue. (42) Those criteria were (i) ‘that the documents requested belong to the same kind of documents or be documents of the same nature’ (43) (ii) that a general presumption is required to ensure that the procedures at issue operate correctly and to guarantee that their objectives are not jeopardised by the intervention of third parties, (44) (iii) that the set of documents concerned were clearly defined by the fact that they all belonged to a file relating to ongoing administrative or judicial proceedings, (45) and (iv) that there are specific rules provided for by a legal measure relating to a procedure before an EU institution that limit access to certain documents. (46)

47.      Although the General Court stated itself that those criteria depended on the type of case — thus implying that they might or might not be relevant, depending on the case at hand — it applied all of them in the present case and found that, because they were not fulfilled, a general presumption of confidentiality could not be recognised. (47)

48.      In my view, however, according to the case-law of this Court, none of the above criteria are necessary criteria for the recognition of a general presumption of confidentiality. (48) The second and third criteria relate to an ongoing administrative or judicial procedure. However, both in Case C‑477/10, Commission v Agrofert Holding (49) and in Case C‑404/10, Commission v Éditions Odile Jacob (50)general presumptions of confidentiality were recognised although the proceedings were no longer ongoing and in Case C‑562/14, Sweden v Commission, this Court recognised a general presumption of confidentiality although there were no specific rules dealing with access to documents in the procedure at issue. (51)

49.      For my part I cannot help thinking that the General Court applied the wrong test and thus came to the wrong conclusion on the question whether a general presumption of confidentiality should be recognised with respect to toxicity study reports.

The ClientEarth test

50.      The governing considerations regarding whether a general presumption should be recognised for a new category of documents, (52) were articulated by the Court in paragraph 80 of the judgment of 4 September 2018, ClientEarth v Commission (C‑57/16 P, EU:C:2018:660), a case which, as it happens, was decided after the judgment of the General Court in this case, but prior to the oral hearing in this appeal: (53)

‘… recognition of a general presumption in respect of a new category of documents presupposes that it has first been shown that it is reasonably foreseeable that disclosure of the type of document falling within that category would be liable actually to undermine the interest protected by the exception in question. Furthermore, as general presumptions constitute an exception to the rule that the EU institution concerned is obliged to carry out a specific and individual examination of every document which is the subject of a request for access and, more generally, to the principle that the public should have the widest possible access to the documents held by the institutions of the European Union, they must be interpreted and applied strictly.’ (54)

Application of the ClientEarth test

51.      In the context of the present appeal, the interest to be protected according to the first indent of Article 4(2) is the commercial interest of the applicant for an MA. The category of documents that this case concerns are study reports of the kind contained in batch 1, namely, toxicity study reports. For all the reasons I have set out in my Opinion in Case C‑175/18 P, PTC Therapeutics International v EMA, the question is whether it is reasonably foreseeable that the granting of access to toxicity study reports would be liable to undermine commercial interests (within the meaning of the first indent of Article 4(2) of Regulation No 1049/2001) in respect of applicants for an MA who have submitted such reports.

52.      The General Court only deals with this question in passing when it states in paragraph 44 of the judgment under appeal that ‘it is not clear that non-clinical and clinical studies are in themselves confidential. Those non-clinical and clinical studies may be limited to satisfying a regulatory scheme prescribed by the EMA and contain no new material’. (55) Yet, the criterion distilled by this Court from its previous case-law and prescribed by its judgment in Case C‑57/16 P, ClientEarth v Commission — which, admittedly was not to hand when the judgment under appeal was delivered, is not whether information is new, but rather whether it is reasonably foreseeable that access to such information might undermine the commercial interests of the parties that compiled this information. I consider that, when that criterion is correctly applied, the answer, with respect to toxicity study reports, can only be in the affirmative. It follows that I consider the interests in question would thereby be undermined if access to such reports were permitted.

53.      I take this view for the following reasons. It is in the nature of toxicity tests to be expensive and time consuming. It is not really disputed that Bravecto was tested on more than 2 000 dogs (56) with the longest toxicity test lasting 90 days and involving daily administration. The preparation of this study was undoubtedly costly and time consuming. In any event, in the nature of things, however, toxicity studies of this kind are likely to be expensive to prepare and involve a considerable amount of time, money, effort and expertise.

54.      As was pointed out by the appellants in their appeal, the EPAR only contains a high-level summary of the length of the studies, on what doses were administered and a summary of the effects observed as well as the conclusions drawn by the regulator. (57) By contrast, the toxicity studies themselves contain detailed information that might allow a competitor not only to draw on the results but also on the method on how a study ought to be conducted. This, in my opinion, is not only the case with regard to the reports at hand, but also with regard to toxicity studies in general.

55.      All of this means that toxicity studies require know-how and a considerable financial input. Because there is a real possibility — perhaps even a likelihood — that competitors can learn from them or even use them, I consider that toxicity studies constitute a category of document, access to which undermines, in principle, the interest protected by the exception for commercially confidential information contained in the first indent of Article 4(2) of Regulation No 1049/2001.

56.      It follows, therefore, that just as in the case of the clinical study reports at issue in Case C‑175/18 P, PTC Therapeutics International v EMA, toxicity study reports thus fulfil the criteria for the recognition of a general presumption of confidentiality.

Data exclusivity and protection under Article 4(2) of Regulation No 1049/2001

57.      It may be convenient at this juncture to address the arguments raised by the EMA to the effect that there is already sufficient protection in place which will prevent access to the documents in question, namely toxicity study reports, from undermining the commercial interests of the applicant for an MA that submitted those reports. That protection includes on the one hand ‘data exclusivity’ and on the other hand the protection provided by the exceptions to access contained in Article 4(2) of Regulation No 1049/2001 applied to the specific case, rather than by applying a general presumption of confidentiality on the basis of the same provision.

58.      The concept of data exclusivity is relevant in the case of the so-called ‘generic approval route’. It allows a pharmaceutical company to apply for an MA with a more limited set of documents. Thus, under Article 39(10) of Regulation No 726/2004 in conjunction with the first subparagraph of Article 13(1) of Directive 2001/82, the applicant for an MA shall not be required to provide the results of the safety and residue tests or of the pre-clinical and clinical trials if the MA relates to a generic medicinal product and the reference medicinal product has been on the market for no less than 8 years in a Member State of the European Union. This covers toxicity study reports which form part of the pre-clinical trials. The rationale for this is to prevent the unnecessary duplication of testing as well as access to veterinary medicinal products at lower prices.

59.      The second means of protection of information which was raised by the EMA and which was subsequently relied on in its reasoning by the General Court is the protection under the first indent of Article 4(2) on the basis of an examination of each individual document and the information contained therein, rather than the application of a general presumption of confidentiality on the basis of the same provision. The exception in the first indent of Article 4(2) is then only applied to that part of the information that can be specifically shown to be confidential rather than relying on a general presumption where all the information is protected unless it can be shown that the reasons for which a general presumption was accepted do not apply to the specific information or that the public interest counter-exception applies.

60.      The EMA relies on Article 4(6) of Regulation No 1049/2001 in support of this assertion. That provision states that if only parts of the requested document are covered by any of the exceptions, the remaining parts of the document shall be released.

61.      The EMA claims that a toxicity study cannot be considered confidential in its entirety and that, if, in accordance with Article 4(6) of Regulation No 1049/2001 only parts of that document are covered by the exception in the first indent of Article 4(2) of that regulation, the remaining parts of the document must be released.

62.      While that argument of the EMA may be correct in the abstract, it cannot be accepted in the context of the application of a general presumption. If it were to be accepted, it would mean that no such presumption could ever exist in the light of Article 4(6) of Regulation No 1049/2001. The very nature of a general presumption of confidentiality is that an institution is not required to examine each and every document in a file or each and every part of a document in order to assess whether it is confidential or not. It is only where it is demonstrated that the general presumption does not apply to a specific document or a part thereof that an institution or agency such as the EMA is required to assess individually whether access should in fact be granted in respect thereof.

63.      As far as data exclusivity is concerned, it must be observed that Article 39(10) of Regulation No 726/2004 when read in conjunction with the first subparagraph of Article 13(1) of Directive 2001/82 has the effect of only preventing the use of the respective data for an application using the generic route within the European Union. (58) There is, however, no provision to prevent such a use in third countries.

64.      I therefore consider that the requirements for the recognition of a general presumption of confidentiality are fulfilled with respect to toxicity study reports.

65.      In the event that the Court does not agree with this assessment, I propose now to look further at one of the criteria scrutinised by the General Court, namely, whether there are specific rules with regard to the grant (or refusal) of access provided for by a legal measure relating to a procedure conducted before an EU institution or agency which have to be balanced with the transparency rules contained in Regulation No 1049/2001. (59) I propose to take this step despite the fact that I consider it sufficient for the recognition of a general presumption that toxicity study reports fulfil the test prescribed in Case C‑57/16 P, ClientEarth v Commission. (60)

Balancing exercise between Regulation No 1049/2001 and specific legislation governing access to documents, including Article 39(3) of the TRIPs Agreement

66.      The Court has recognised a general presumption of confidentiality in a number of cases in which the principles set out in Regulation No 1049/2001 and differing rules specific to the procedure at issue had to be reconciled and thus be interpreted in a consistent and uniform manner. The Court has accordingly recognised a general presumption of confidentiality on the basis of an interpretation of Regulation No 1049/2001 in the light of Regulations (EC) No 1/2003 (61) and (EC) No 773/2004, (62) in cases of state aid (63) and for information gathered during merger proceedings. (64)

67.      It should be pointed out, however, that, although the parties to the dispute agree that the existence of specific legislative rules regarding access to documents is a relevant criterion so far as the recognition of a general presumption is concerned, it is by no means an essential prerequisite to such recognition.

68.      As already mentioned in point 48 of this Opinion, there have also been cases in which the Court recognised a general presumption of confidentiality in instances where there were no special rules in place governing access to documents. (65)

69.      I agree with the General Court’s findings in paragraphs 33 to 36 of the judgment under appeal that Regulation No 726/2004 does not contain any provision restricting the use of documents obtained by the requester following a freedom of information request.

70.      A more detailed assessment of that issue may be found in points 85 to 86 of my Opinion which was delivered today in Case C‑175/18 P, PTC Therapeutics International and it is probably unnecessary for present purposes to repeat that exercise here.

71.      I agree with the General Court that Regulation No 726/2004 does not contain any specific provision restricting access to documents. The first paragraph of Article 73 of Regulation No 726/2004, on the other hand, specifically provides that Regulation No 1049/2001 applies to documents held by the EMA.

72.      The wording of the first paragraph of Article 73 of Regulation 726/2004 is broad and unqualified. The fact that certain information has to be published proactively (66) does not mean that access to further information should be in any way restricted, particularly in the light of Article 73 of that regulation.

73.      I do not, however, agree with the General Court’s findings, with regard to Article 39(3) of the TRIPs Agreement. (67) In my view, that provision calls for an interpretation of the first indent of Article 4(2) of Regulation No 1049/2001 which ensures that undisclosed data that was submitted as a condition of approving the marketing of a pharmaceutical product is protected against disclosure. In my opinion, this is only possible if a general presumption of confidentiality is recognised for toxicity study reports that were submitted in an application procedure for an MA.

74.       As it happens, a more detailed discussion of this issue is contained in points 87 to 97 of my Opinion in Case C‑175/18 P, PTC Therapeutics International which was delivered today and which is equally applicable here, as both the clinical study reports that were the subject of that Opinion and the toxicity study reports that are at issue here were submitted as a requirement in order to obtain an MA. (68)

75.      With regard to the TRIPs Agreement, the Court has ruled, that, while it is not directly applicable, in the areas governed by the TRIPs Agreement, the Union rules must nonetheless be interpreted in a manner consistent with the TRIPs Agreement in so far as it is possible to do so. (69) Taking into account the fact that Article 39(3) of the TRIPs Agreement requires the protection of data against disclosure, then, if no effective steps are otherwise taken to ensure the protection of the data against unfair commercial use, the question arises whether the data protection conferred by Article 39(10) of Regulation No 726/2004 in conjunction with Article 13 of Directive 2001/82 confers such effective protection.

76.      Article 39(3) of the TRIPs Agreement deals with the treatment of undisclosed test or other data, the submission of which involves a considerable effort in approval proceedings, for example MA proceedings for a pharmaceutical product.

77.      Toxicity study reports fall squarely within the ambit of Article 39(3) of the TRIPs Agreement. First, applicants for MAs must submit their toxicity study reports to a regulatory body, namely the EMA. Second, the approval process concerns pharmaceutical products. Third, the pharmaceutical product by definition contains a new chemical entity, because if it were otherwise, the generic approval route would be possible, assuming the relevant time limits have lapsed. Fourth, the carrying out of toxicity studies involves considerable effort. Fifth, save for the (relatively limited) amount disclosed by EPARs, (70) the data is hitherto undisclosed to the general public.

78.      An argument that the disclosure is necessary to protect the public by reason of overriding public interests to that effect (71) has never been considered by the EMA because it had decided that toxicity study reports do not constitute confidential information. It follows, therefore, that under Article 39(3) of the TRIPs Agreement the issue is whether sufficient steps have been taken to protect such data against disclosure (except where necessary to protect the public) and to ensure that the data at issue are protected against unfair commercial use. 

79.      For my part, I find myself obliged to conclude that the first subparagraph of Article 13(1) of Directive 2001/82 does not provide effective protection as it applies only within the territory of the European Union/European Economic Area (EEA). It must be noted that other members of the TRIPs Agreement have the same obligation to protect such data. In order to make that system of protection effective worldwide, those rules would not only have to protect data that is submitted within the framework of their own authorisation procedure, they would also have to apply to data that has been submitted for that purpose in a third country or to another authority in a third country. Looking, as an example, at the EU rules, Article 13(1) of Directive 2001/82 does not provide such protection. (72)

80.      All of this means that there is a real risk that the non-redacted part of a toxicity study report which was released to a requester under the provisions of Regulation No 1049/2001 might subsequently be used outside of the European Union in an application for an MA for a generic product even prior to the end of the data exclusivity period.

81.      I therefore consider that the General Court erred in law in failing to give an interpretation of Article 4(2) of Regulation No 1049/2001 which was consistent with the requirements of Article 39(3) of the TRIPs Agreement in circumstances where it was possible to do so and where such an interpretation would not have been contra legem.

Article 4(3) as the basis of a general presumption

82.      The appellants also rely on the first subparagraph of Article 4(3) of Regulation No 1049/2001 in favour of a general presumption of confidentiality. Their argument is that a general presumption is necessary to guarantee that the objectives of the MA procedure are not jeopardised and to ensure the integrity of the conduct of the bilateral procedure. However, as the General Court rightly points out in paragraph 45 of the judgment under appeal, the procedure at issue is one for granting an MA to Bravecto and this was closed at the time the request for access to documents was made. Further procedures in which the appellants may want to rely on toxicity study reports again, are clearly outside the ambit of the first subparagraph of Article 4(3) of Regulation No 1049/2001.

83.      The appellants also rely on the second subparagraph of Article 4(3) of Regulation No 1049/2001. However, applying the criteria prescribed by this Court in Case C‑57/16 P, ClientEarth v Commission, (73) I do not consider that it is reasonably foreseeable that the disclosure of toxicity study reports — in general — would be liable seriously to undermine the EMA’s decision-making process, which is the interest that is protected by the exception in question.

84.      I therefore agree with the reasoning of the General Court that Article 4(3) of Regulation No 1049/2001 cannot serve as the basis for a general presumption of confidentiality of toxicity study reports.

Documents drawn up by the appellants

85.      The General Court in paragraph 43 of the judgment under appeal deals with the appellants’ claim that the EMA should have considered that the batch 1 study reports are data generated and assembled by them. However, the General Court’s finding that the fact that toxicity study reports like the batch 1 study reports are not drawn up by the EMA but rather by the appellants makes no difference to the application of Regulation No 1049/2001 is correct. As the Court held in Commission v Breyer, (74) the wording of Article 2(3) in conjunction with Article 1(a) of Regulation No 1049/2001 is clear. That regulation applies to all documents, irrespective of whether they are drawn up or received by the institution. (75) Thus, the statement of the General Court that, for the application of the first indent of Article 4(2) of Regulation No 1049/2001 the decisive factor is whether the information concerned is confidential, irrespective of the origin of the document, is correct.

3.      Conclusions on the first ground of appeal

86.      I therefore consider, on the grounds set out above, that the General Court erred in law in concluding that there was no general presumption in favour of the non-disclosure of toxicity study reports. This does not, however, mean that the decision of the EMA should necessarily be annulled. Although, in contrast to its judgment in the companion case, Case T‑718/15, PTC Therapeutics International v EMA, (76) the General Court did not rely on this in the present case, the institution concerned is not required to base its decision on a general presumption, even if such a presumption exists. (77) It can, as it quite clearly did in the present case, carry out a specific examination of the requested documents and reach a conclusion based upon that specific examination.

87.      It is therefore necessary in any event to consider the arguments raised by the appellants in respect of their second and third grounds of appeal. These arguments overlap to some extent with the arguments which have already been dealt with in the first ground of appeal, because they are directed at the specific prejudice and damage to their commercial interests which the appellants claim they will suffer if disclosure of the reports at issue is permitted. I will now consider these issues.

B.      Second ground of appeal: infringement of the first indent of Article 4(2) of Regulation No 1049/2001

88.      The principal question, which must be answered under this ground of appeal, is whether the disclosure of the batch 1 study reports would undermine the protection of the appellants’ commercial interests within the meaning of the first indent of Article 4(2) of Regulation No 1049/2001.

1.      Arguments of the parties

89.      The appellants argue that the General Court failed to draw attention to countervailing principles of law, amongst others, the TRIPs Agreement. They also argue that the General Court applied an erroneous interpretation of this Court’s judgment in LPN and Finland v Commission. (78)The General Court thus erred in law in weighing the commercial interests of the appellants against the (general) public interest in disclosure. Rather, they consider that the question whether their commercial interests were undermined should have been assessed separately before assessing whether an overriding public interest might prevent disclosure. (79)

90.      The appellants argue that the General Court applied an erroneous standard when it held that the first indent of Article 4(2) of Regulation No 1049/2001 required that the interests of the appellants must be ‘seriously undermined’. (80). They also claim that the General Court committed further errors in law in using an unrealistic test of commercial confidentiality and in failing to find that there was a reasonably foreseeable prospect that the batch 1 study reports may be misused by a competitor. (81) The appellants argue that, due to these misconceptions, the General Court did not properly engage with the witness statements submitted by them.

91.      The EMA contests this and claims that, in order to justify application of the exception provided in the first indent of Article 4(2) of Regulation No 1049/2001, it is necessary to prove that disclosure of the batch 1 study reports is actually likely to undermine the commercial interests of the appellants, that risk of a protected interest being adversely affected must be of a specific nature that is directly linked to the disclosure and not be merely hypothetical. (82) The EMA insists that access to the documents held by it can be refused only if one or more of the exceptions of Article 4 of Regulation No 1049/2001 apply to the whole of the content of the document concerned. (83) In its opinion, the redaction of (i) details on the concentration range of the active substance, (ii) details on the internal reference standard used for analytical tests and (iii) references to future development plans ensure that the batch 1 study reports as redacted do not contain commercially confidential information as protected by the first indent of Article 4(2) of Regulation No 1049/2001 and that some of the information is in any case available because it is published in the EPAR for Bravecto. (84) The EMA further argues that the appellants’ investments are adequately protected by means of the data exclusivity applied according to Article 39(10) of Regulation No 726/2004. (85)

2.      Assessment of the arguments as to whether access to the batch 1 study reports infringes the first indent of Article 4(2) of Regulation No 1049/2001

(a)    Must the particular interest protected by the Article 4(2) exception be weighed against the general public interest in disclosure of documents?

92.      The General Court states in paragraph 65 of the judgment under appeal:

‘It must also be noted that the system of exceptions laid down in Article 4 of Regulation No 1049/2001, particularly in Article 4(2), is based on a weighing of the opposing interests in a given situation, that is to say, on the one hand, the interests which would be favoured by the disclosure of the documents in question and, on the other, those which would be jeopardised by such disclosure. The decision taken on a request for access to documents depends on which interest must prevail in the particular case (judgments of 14 November 2013, LPN and Finland v Commission, C‑514/11 P and C‑605/11 P, EU:C:2013:738, paragraph 42, and of 23 September 2015, ClientEarth and International Chemical Secretariat v ECHA, T‑245/11, EU:T:2015:675, paragraph 168).’

93.      This question is addressed in some detail in points 111 to 128 of my Opinion in Case C‑175/18 P, PTC Therapeutics International v EMA that has already been delivered today. In that Opinion, I examine how the General Court in the judgment under appeal in that case erred in law by finding that an agency can weigh the general public interest in transparency against the private interest of the party relying on one of the exceptions in Article 4(2) of Regulation No 1049/2001. I would note that the authorities relied on are, however, different in the present judgment under appeal. The General Court relied for this purpose on the judgment of this Court in LPN and Finland v Commission. (86)

94.      While the passage cited above in point 92 is a correct rendition of the content of paragraph 42 of the judgment in LPN and Finland v Commission, it is, I think, with respect, somewhat taken out of context. In paragraph 41 of that judgment, this Court held: ‘Under the exception relied upon by the Commission, namely the third indent of Article 4(2) of Regulation No 1049/2001, the institutions must refuse access to a document where disclosure would undermine the protection of the purpose of inspections, investigations and audits, unless there is an overriding public interest in disclosure.’

95.      That passage confirms my reasoning in my Opinion in Case C‑175/18 P, PTC Therapeutics International v EMA (87) to the effect that the question of whether one of the exceptions contained in Article 4(2) of Regulation No 1049/2001 is applicable must be addressed first and in a manner which is independent of public interest concerns regarding disclosure. It is only where such an exception is applicable and the case comes within Article 4(2) of that regulation, that the second question comes into play, namely, whether an overriding public interest exists. This second question constitutes, in effect, a counter-exception to the exceptions laid down in Article 4(2) of Regulation No 1049/2001.

96.      Paragraphs 69 and the last sentence of paragraph 87 of the judgment under appeal — although in the latter case, this is only a subsidiary argument to the General Court’s finding that the reports at issue do not contain confidential information — all suggest that the General Court wrongly applied Article 4(2) of Regulation No 1049/2001 in the manner I have just described.

97.      For these reasons, I find that the General Court committed an error of law when it held that the public interest must be weighed against the interests of the party relying on the protection of its confidential commercial information against disclosure. This error in law has, I fear, tainted the General Court’s reasoning in this case.

(b)    Must the disclosure ‘seriously’ undermine the protection of the commercial interests of the appellants for the purposes of bringing the Article 4(2) exceptions into play?

98.      In paragraph 68 of the judgment under appeal, the General Court states

‘Consequently, in order to apply the exception provided for by the first indent of Article 4(2) of Regulation No 1049/2001, it must be shown that the documents at issue contain elements which may, if disclosed, seriously undermine the commercial interests of a legal person.’ (88)

99.      As I have set out in points 104 to 109 of my Opinion delivered today in Case C‑175/18 P, PTC Therapeutics International v EMA, Article 4(2) of Regulation No 1049/2001 does not contain the word ‘seriously’. In so far as the General Court used this elevated standard to assess whether the disclosure of the batch 1 study reports undermines the protection of the commercial interests of the appellants — as indeed paragraph 69 of the judgment under appeal suggests — it thereby fell into error.

100. I now propose to show how these errors of law might have influenced the General Court’s assessment of the appellants’ arguments, raised in support of their claim that the batch 1 study reports in their entirety constitute commercially confidential information.

(c)    Criteria for information to be considered as commercially confidential

101. One of the General Court’s arguments in rejecting the appellants’ claim that the batch 1 study reports constitute commercially confidential information is, that the information they contain is not ‘novel’ or ‘innovative’. (89) It further holds that ‘the infringement of the protection of the commercial interests of a person as referred to in Article 4(2) of Regulation No 1049/2001 is not necessarily determined by reference to the financial value of the information subject to disclosure’. (90)

102. The EMA and the General Court criticise the lack of novelty in the batch 1 study reports. I cannot, however, agree that ‘novelty’ is in itself an indispensable prerequisite before a document can be considered to be commercially sensitive for the purposes of the first indent of Article 4(2) of Regulation No 1049/2001. (91) If a potential competitor can usefully draw on the batch 1 study reports — even after they have been redacted by the EMA — and thus save money that the appellants had to spend in order to conduct their toxicity studies and draw up those reports, those studies, in my opinion, constitute commercially confidential information belonging to the appellants. This is true irrespective of whether the information in the reports is novel. The arguments raised by the General Court against this assessment are, to my mind, not convincing. (92) While it is true that any competitors would have to conduct their own studies, the point here is that their reliance on the disclosed part of the appellants’ studies might make it easier and less costly for them to carry out those studies successfully by relying closely on the approach taken by the appellants. (93) This is also the appellants’ argument when they claim that the reports at issue might be used as a ‘road map’. Moreover, the General Court’s reliance on the protection provided by means of data exclusivity is, I think, ultimately, unconvincing. As I have already set out above, data exclusivity does not prevent use of the data in third countries.

103. Thus, even though there might be cases in which a document that is financially valuable might not be considered as commercially confidential information, the fact that this information is not readily available together with the fact that its use gives an advantage to its user — which in turn places at a disadvantage the private or legal person that had to invest much time, effort and money to draw it up in the first place — means that the batch 1 study reports must be considered as commercially confidential information for the purposes of the exception contained in the first indent of Article 4(2) of Regulation No 1049/2001.

(d)    Level of foreseeability with regard tothe undermining of commercial interests

104. The General Court takes issue with the appellants’ arguments concerning the question of whether the undermining of commercial interests of the appellants is foreseeable in several instances. It considers them to be ‘too vague and unsubstantiated’ with regard to the ‘road map’ argument, (94) ‘vague and imprecise’ with regard to the influence the disclosure would have on a competing company obtaining an MA, (95) and it considers the appellants’ explanations as to why the entire batch 1 study reports should be considered as confidential as ‘vague and generic’. (96)

105. In paragraph 81 of the judgment under appeal the General Court points out that ‘it was all the more necessary to adduce precise and proper explanations since, as has been pointed out in paragraph 64 above, the exceptions provided for in Article 4 of Regulation No 1049/2001 derogate from the principle that the public should have the widest possible access to the documents and must therefore be interpreted and applied strictly’.

106. While this is, of course, correct, what is essential in this context, and as the General Court accepted itself, (97) is that what has to be shown according to the case-law is how disclosure of that document could specifically and actually undermine the interest protected by Article 4(2) of Regulation No 1049/2001. In addition, the risk of the interest being undermined must be a reasonably foreseeable one. It must also be one which is not purely hypothetical, (98) although, of course, it is not necessary to demonstrate that the risks must be certain. The General Court’s argument that the risks described by the appellants are not specific are based on the General Court’s preconception that the information contained in the batch 1 study reports is not confidential because competitors have to carry out their own toxicity study reports. This, however, in my judgment, is too narrow a view for all the reasons set out elsewhere in this Opinion and in the Opinion already delivered today in Case C‑175/18 P, PTC Therapeutics International v EMA. The ability to follow an approach and to benchmark one’s own work against the successful work of another is clearly a real advantage to potential competitors. The appellants have pointed to the reality — indeed, the obviousness — of this risk. The very fact that the requester in the present case is another pharmaceutical company in its own way speaks volumes as to the nature of the risk.

107. The General Court’s understanding that the appellants’ commercial interests must be ‘seriously’ undermined as well as its assumption that it should weigh those interests against the public interest considerations underlying Regulation No 1049/2001 might well have influenced the assessment of the General Court in its judgment. Paragraph 69 of the judgment under appeal supports this finding. It is clear from paragraph 87 of the judgment under appeal, while presented only as a subsidiary argument, that the General Court carries out such a weighing exercise when it holds that ‘the non-disclosure of all studies in order to prevent the authorities of a third country granting access to its market to a manufacturer without that manufacturer being required to submit its own studies would amount to depriving the public of the right — granted by EU law — to have access to documents containing information relating to authorised medicinal products’. (99)

108. Summing up therefore on this point, I consider that the General Court’s assessment was premised on certain assumptions which are not legally correct. It is not necessary that the appellants’ commercial interests are ‘seriously’ undermined, nor does the information confidential treatment of which is sought, have to be ‘novel’ and the commercial interests of the appellants, once asserted, may only be disregarded if an overriding public interest in disclosure is found, rather than being weighed against the public interests in disclosure underlying Regulation No 1049/2001. I therefore consider that the second ground of appeal is well founded.

C.      Third ground of appeal: infringement of Article 4(3) of Regulation No 1049/2001

1.      Arguments of the parties

109. Despite the fact that the application procedure for an MA for Bravecto was concluded, the exception contained in the first subparagraph of Article 4(3) of Regulation No 1049/2001 relating to matters where a decision has not been taken was raised by the appellants. In addition the appellants rely in that regard on the second subparagraph of Article 4(3) of Regulation No 1049/2001.

110. At the oral hearing the appellants underlined that they believe that the first subparagraph is applicable because, at the time of the contested decision of the EMA to grant access to the batch 1 study reports, an application for a further fluralaner-based medicinal product, a spot-on solution for cats and dogs, was pending and the EMA had been informed by the appellants that an application for a further fluralaner-based medicinal product for the treatment of poultry would be submitted. The reports at issue were reused in both of these applications.

111. As regards the second subparagraph of Article 4(3) of Regulation No 1049/2001, the appellants’ claim that opinions on results of the studies are contained in the batch 1 study reports and that they were prepared for the internal deliberations of the EMA. If such documents were disclosed, applicants for MAs would ‘take steps’ to protect their valuable data to the greatest degree possible, to the point of avoiding the centralised application procedure for an MA.

112. The EMA contests these arguments.

2.      Assessment of the arguments in relation to Article 4(3) of Regulation No 1049/2001

113. In so far as the appellants rely on the first subparagraph of Article 4(3) of Regulation No 1049/2001, I refer to point 82 of this Opinion. As the application for an MA was clearly closed at the time the request for access was made, the requirements of that provision are not fulfilled. There is no possibility of a more extensive interpretation of the first subparagraph of Article 4(3) of Regulation No 1049/2001 because the second subparagraph of Article 4(3) clearly deals with that set of facts and it does so in a manner which is different to that provided by the provision of the first subparagraph of Article 4(3) of Regulation No 1049/2001.

114. With respect to the appellants’ arguments under the second subparagraph of Article 4(3) of Regulation No 1049/2001, the General Court in paragraphs 108 to 110 of the judgment under appeal would appear to be dealing with questions of confidentiality. It found that, as information not relating to the already authorised indication (100) as well as such data relating to future development projects is redacted, the EMA’s decision-making process with regard to such future indications is not affected. While this reasoning is unclear, I agree with the General Court’s finding in paragraph 111 of the judgment under appeal that the appellants have not submitted any evidence from which it might be concluded that the alleged undermining of the decision-making process would be ‘serious’. One might add that it is not even clear in which way that decision-making process could be affected at all.

115. As regards the argument that applicants for an MA might be unwilling to entrust information to the EMA for fear of its being disclosed under Regulation No 1049/2001, this is really an argument which goes to the question of the confidentiality of the information which is protected by the first indent of Article 4(2) of Regulation No 1049/2001, rather than to the question of the decision-making process protected by Article 4(3) of Regulation No 1049/2001. Consequently, I consider that the General Court did not err in law when it found that the first subparagraph of Article 4(3) of Regulation No 1049/2001 is inapplicable because the MA application process was terminated and thus the decision in the matter had already been taken. I further consider that the General Court did not err in law when it found that it would not seriously undermine the EMA’s decision-making process to give access to the reports at issue.

116. It follows that the third plea must be rejected.

VI.    Overall conclusions

117. For all of the reasons set out above, I believe, that, with respect, the General Court erred in law in so far as it concluded that there was no general presumption under the first indent of Article 4(2) of Regulation No 1049/2001 that access to the reports at issue should be refused. In any event, I also consider that the General Court erred in law in so far as it concluded that access to the reports at issue would not compromise the appellants’ commercial interests for the purposes of the first indent of Article 4(2) of Regulation No 1049/2001.

118. Under Article 61 of the Statute of the Court of Justice of the European Union, if the appeal is well founded, the Court of Justice shall quash the decision of the General Court and may itself give final judgment in the matter, where the state of the proceedings so permits, or refer the case back to the General Court for judgment.

119. In my opinion, this is not a case where the Court of Justice may give final judgment in the matter as it requires the legal assessment of complex issues of fact. I propose therefore to set aside the judgment under appeal and refer the case back to the General Court for judgment following a re-examination of the reports at issue in the light of the above reasoning.

VII. Conclusion

120. For those reasons and without prejudging the Court’s assessment of the other pleas put forward in this appeal, I propose that the Court should:

(1)      Set aside the judgment of the General Court of the European Union of 5 February 2018, MSD Animal Health Innovation and Intervet international v EMA (T‑729/15, EU:T:2018:67);

(2)      Refer the case back to the General Court;

(3)      Reserve the decision as to costs.


1      Original language: English.


2      OJ 2001 L 145, p. 43.


3      Council Decision concerning the conclusion on behalf of the European Community, as regards matters within its competence, of the agreements reached in the Uruguay Round multilateral negotiations (1986-1994) (OJ 1994 L 336, p. 1).


4      Regulation of the European Parliament and of the Council of 31 March 2004 laying down Community procedure for the authorisation and supervision of medicinal products for human and veterinary use and establishing a European Medicines Agency (OJ 2004 L 136, p. 1).


5      OJ 2001 L 311, p. 1.


6      As amended by Directive 2004/28/EC of the European Parliament and of the Council of 31 March 2004 (OJ 2004 L 136, p. 58).


7      Order of 20 July 2016, MSD Animal Health Innovation and Intervet international v EMAMSD Animal Health Innovation and Intervet international v EMAMSD Animal Health Innovation and Intervet international v EMAMSD Animal Health Innovation and Intervet international v EMAMSD Animal Health Innovation and Intervet international v EMA (T‑729/15 R, not published, EU:T:2016:435).


8      EMA v MSD Animal Health Innovation and Intervet internationalEMA v MSD Animal Health Innovation and Intervet internationalEMA v MSD Animal Health Innovation and Intervet internationalEMA v MSD Animal Health Innovation and Intervet internationalEMA v MSD Animal Health Innovation and Intervet international (C‑512/16 P(R), not published, EU:C:2017:149).


9      For the criterion that the documents requested belong to the same category of documents or be documents of the same nature, it relies on the judgments of 1 July 2008, Sweden and Turco v Council (C‑39/05 P and C‑52/05 P, EU:C:2008:374, paragraph 50), and of 17 October 2013, Council v Access Info Europe (C‑280/11 P, EU:C:2013:671, paragraph 72). For the criterion that a general presumption may be recognised to ensure that the integrity of the conduct of the procedure can be preserved by limiting intervention by third parties, it relies on the Opinion of Advocate General Wathelet in LPN and Finland v Commission (C‑514/11 P and C‑605/11 P, EU:C:2013:528, points 66, 68, 74 and 76). For the criterion that the documents must belong to a set of documents which were clearly defined by the fact that they all belonged to a file relating to ongoing administrative or judicial proceedings, it relies on the judgments of 29 June 2010, Commission v Technische Glaswerke Ilmenau (C‑139/07 P, EU:C:2010:376, paragraphs 12 to 22); of 21 September 2010, Sweden and Others v API and Commission (C‑514/07 P, C‑528/07 P and C‑532/07 P, EU:C:2010:541, paragraph 75), and of 27 February 2014, Commission v EnBW (C‑365/12 P, EU:C:2014:112, paragraphs 69 and 70). For the criterion that there are special rules regarding access to documents, it relies on the judgment of 11 June 2015, McCullough v Cedefop (T‑496/13, not published, EU:T:2015:374, paragraph 91), and the Opinion of Advocate General Cruz Villalón in Council v Access Info Europe (C‑280/11 P, EU:C:2013:325, point 75).


10      The judgment under appeal, paragraphs 32 and 45.


11      Ibid., in paragraph 46.


12      Ibid., paragraphs 39 and 40.


13      Ibid., paragraphs 34 to 38.


14      Ibid., paragraph 44.


15      Ibid., paragraph 50.


16      In conjunction with Article 13(1) of Directive 2001/82.


17      Judgment of 5 February 2018, MSD Animal Health Innovation and Intervet international v EMAMSD Animal Health Innovation and Intervet international v EMAMSD Animal Health Innovation and Intervet international v EMAMSD Animal Health Innovation and Intervet international v EMAMSD Animal Health Innovation and Intervet international v EMA (T‑729/15, EU:T:2018:67, paragraphs 50 and 51).


18      Ibid., paragraph 52.


19      Ibid., paragraph 46.


20      I am aware that the word ‘seriously’ contained in paragraph 68 of the English version of the judgment under appeal is not contained in all the linguistic versions (this is namely the case for the French and German versions). However, the language of the proceedings is English and thus, the English version is the only authentic version, as well as the version on which the appellants relied in their pleadings. I will therefore rely on this language version of the judgment under appeal in my reasoning.


21      Ibid., paragraph 68 and 69.


22      Ibid., paragraphs 70 to 77,


23      Ibid., paragraphs 82 and 89.


24      Ibid., paragraph 102.


25      Ibid., paragraph 110.


26      Ibid., paragraph 122.


27      Ibid., paragraph 127.


28      Which is applicable in accordance with Article 39(10) of Regulation No 726/2004.


29      Paragraphs 41 to 49 of the appeal.


30      Paragraph 50 of the appeal.


31      Paragraphs 46 and 51 to 53 of the appeal.


32      Paragraph 55 of the appeal.


33      Paragraph 68 of the appeal.


34      Paragraph 55 of the appeal.


35      Paragraphs 58 to 62 of the appeal.


36      Paragraphs 95 to 98 of the EMA’s response.


37      Judgments of 9 June 2011, Comitato ‘Venezia vuole vivere’ and Others v Commission (C‑71/09 P, C‑73/09 P and C‑76/09 P, EU:C:2011:368, paragraph 34 and the case-law cited), and of 21 December 2011, A2A v Commission (C‑318/09 P, not published, EU:C:2011:856, paragraph 109).


38      EMA/110196/2006.


39      EMA/127362/2006.


40      See point 39 of my Opinion in PTC Therapeutics International v EMA (C‑175/18 P and the case-law cited).


41      Ibid., in point 59 and the case-law cited.


42      The rejection of the General Court’s criteria — criteria which were also used in the General Court’s judgment of the same day —is set out in points 63 to 67 of my Opinion on the appeal in that case (namely PTC Therapeutics v EMA, T718/15) which is also being delivered today. Although the arguments in PTC Therapeutics v EMA (T‑718/15) are set out in relation to clinical study reports, they apply in the same way to toxicity study reports like the ones at issue in the present appeal. The arguments of the respective parties in this case (MSD Animal Health Innovation and Intervet international v EMA, T‑729/15) and the PTC Therapeutics case are very similar and so are the judgments under appeal and the General Court’s judgment in PTC Therapeutics v EMA (T‑718/15) with respect to the general presumption of confidentiality.


43      Paragraph 25 of the judgment under appeal.


44      Ibid., paragraph 26.


45      Ibid, paragraph 28.


46      Ibid., paragraph 29.


47      Ibid.,paragraphs 30 to 37.


48      Except possibly for the first criterion, which the General Court does not, however, discuss in any detail and, which only means that a category must be identified to which a general presumption applies.


49      Judgment of 28 June 2012 (EU:C:2012:394).


50      Judgment of 28 June 2012 (EU:C:2012:393).


51      Judgment of 11 May 2017 (EU:C:2017:356).


52      Five categories are set out in the judgment of 4 September 2018, ClientEarth v Commission (C‑57/16 P, EU:C:2018:660, paragraph 81). They are: (i) documents on the Commission’s administrative file with regard to State aid (see judgment of 29 June 2010, Commission v Technische Glaswerke Ilmenau (C‑139/07 P, EU:C:2010:376), (ii) documents lodged in proceedings before the Courts of the European Union while they are pending, (judgment of 18 July 2017, Commission v Breyer (C‑213/15 P, EU:C:2017:563, as well as case-law cited in paragraph 41 of that judgment), (iii) documents exchanged between the Commission and notifying parties or third parties in the course of merger control proceedings (judgment of 28 June 2012, Commission v Éditions Odile Jacob (C‑404/10 P, EU:C:2012:393), (iv) documents relating to an infringement procedure during its pre-litigation stage (judgment of 14 November 2013, LPN and Finland v Commission (C‑514/11 P and C‑605/11 P, EU:C:2013:738), and (v) documents relating to proceedings under Article 101 TFEU (judgment of 27 February 2014, Commission v EnBW (C‑365/12 P, EU:C:2014:112)). It must be stressed that those five categories do not constitute a closed group. I would also note that this has not been argued before the Court.


53      Indeed, the case was referred to and the parties had the opportunity to comment on it in the oral hearing.


54      Emphasis added.


55      Emphasis added.


56      Although this is not necessarily the number relating to the toxicity studies.


57      Paragraph 29 of the appeal.


58      This effect applies for 8 years after the reference medicinal product has been authorised in a Member State or the European Union.


59      Paragraph 29 of the judgment under appeal.


60      Judgment of 4 September 2018 (EU:C:2018:660, paragraph 80).


61      See Article 27(2) and Article 28 of Council Regulation of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty [now Articles 101 and 102 TFEU] (OJ 2003 L 1, p. 1).


62      See Articles 6, 8, 15 and 16 of Commission Regulation of 7 April 2004 relating to the conduct of proceedings by the Commission pursuant to Articles 81 and 82 of the EC Treaty (OJ 2004 L 123, page 18). The latter regulations contain restrictive rules for the use of documents relating to competition proceedings under Article 81 EC (now Article 101 TFEU). They confer access to the file in those proceedings to the ‘parties concerned’ and the ‘complainants’ whose complaints the Commission intends to reject subject to specific further provisos. Judgment of 27 February 2014, Commission v EnBW (C‑365/12 P, EU:C:2014:112, paragraphs 86 to 92).


63      Judgment of 29 June 2010, Commission v Technische Glaswerke Ilmenau (C‑139/07 P, EU:C:2010:376, paragraph 61) on the basis of Article 6(2) and Article 20 of Council Regulation (EC) No 659/1999 of 22 March 1999 laying down detailed rules for the application of Article 93 of the EC Treaty (OJ 1999 L 83, p. 1), meanwhile replaced by Council Regulation (EU) 2015/1589 of 13 July 2015 laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union (codification) (OJ 2015 L 248, page 9). These provisions envisage that certain information in the review procedure is to be submitted to the Member States, while such a provision does not apply in respect of interested parties.


64      Judgment of 28 June 2012, Commission v Agrofert Holding (C‑477/10 P, EU:C:2012:394, paragraph 64) on the basis of Articles 17 and 18(3) of Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings (OJ 2004 L 24, p. 1) and Article 17 of Commission Regulation (EC) No 802/2004 of 7 April 2004 implementing Council Regulation No 139/2004 on the control of concentrations between undertakings (OJ 2004 L 133, p. 1), the latter two provisions concern safeguarding the rights of the defence.


65      See judgment of 11 May 2017, Sweden v Commission (C‑562/14 P, EU:C:2017:356). I would note, however, that this case concerned the exception contained in the third indent of Article 4(2) of Regulation No 1049/2001.


66      See Article 35(6), Article 36, Article 37(3) and Article 38(2) and (3) of Regulation No 726/2004.


67      Paragraphs 47 to 50 of the judgment under appeal.


68      The reasoning of the General Court was equally similar in paragraphs 47 to 51 of the judgment under appeal and paragraphs 61 to 65 of the judgment of 5 February 2018, PTC Therapeutics International v EMA (T‑718/15, EU:T:2018:66).


69      See point 87 of my Opinion of today’s date in PTC Therapeutics International (C‑175/18 P and the case-law cited).


70      There might be specific cases in which this is different.


71      According to Article 39(3) of the TRIPs Agreement and the counter-exception under Article 4(2) of Regulation No 1049/2001.


72      The first subparagraph of Article 13(1) of Directive 2001/82 relates to a reference medicinal product which has been authorised in a Member State or the European Union.


73      See above in point 51 of this Opinion and judgment of 4 September 2018 (EU:C:2018:660, paragraph 80).


74      Judgment of 18 July 2017 (C‑213/15 P, EU:C:2017:563).


75      Ibid., in paragraph 35 and 36.


76      Judgment of 5 February 2018, PTC Therapeutics International v EMA (T‑718/15, EU:T:2018:66, paragraph 70), and point 98 of my Opinion in PTC Therapeutics International v EMA (C‑175/18 P).


77      Judgment of 14 November 2013, LPN and Finland v Commission (C‑514/11 P and C‑605/11 P, EU:C:2013:738, paragraphs 66 and 67).


78      Judgment of 14 November 2013 (C‑514/11 P and C‑605/11 P, EU:C:2013:738).


79      Paragraph 71 of the appeal.


80      Ibid., in paragraph 71.


81      Ibid, in paragraphs 71 and 74


82      In paragraphs 31 and 42 of the EMA’s response.


83      Ibid., in paragraph 105.


84      Ibid., in paragraphs 18 to 19.


85      Ibid., in paragraphs 35 to 37.


86      Judgment of 14 November 2013 (C‑514/11 P and C‑605/11 P, EU:C:2013:738, paragraph 42).


87      In point 128


88      Emphasis added. As explained in footnote 20, the adverb ‘seriously’ does not appear in all language versions. However, it appears in the only language version that is authentic, namely the language of procedure (which is English in this case).


89      See, for example, paragraphs 74 to 76 of the judgment under appeal.


90      In paragraphs 82 and 89 of the judgment under appeal.


91      See also point 147 of my Opinion in PTC Therapeutics International v EMA (C‑175/18 P).


92      In paragraph 84 of the judgment under appeal.


93      See also point 150 of my Opinion in PTC Therapeutics International v EMA (C175/18 P).


94      In paragraph 80 of the judgment under appeal.


95      In paragraph 87 of the judgment under appeal.


96      In paragraph 81 of the judgment under appeal.


97      In paragraph 93 of the judgment under appeal.


98      Judgment of 3 July 2014, Council v In’t Veld (C‑350/12 P, EU:C:2014:2039, paragraph 52 and the case-law cited).


99      In spite of this, in paragraph 128 of the judgment under appeal the General Court states the opposite with regard to its assessment under the fourth plea.


100      This being the indication related to the chewable tablets for dogs, as opposed to other indications like the abovementioned spot-on solution for dogs and cats.