Language of document : ECLI:EU:T:2019:892

JUDGMENT OF THE GENERAL COURT (Sixth Chamber)

19 December 2019 (*)

(Medicinal products for human use — Application for marketing authorisation for the medicinal product Fanaptum — iloperidone — Commission refusal decision — Regulation (EC) No 726/2004 — Scientific evaluation of the risks and benefits of a medicinal product — Obligation to state reasons — Manifest error of assessment — Proportionality — Equal treatment)

In Case T‑211/18,

Vanda Pharmaceuticals Ltd, established in London (United Kingdom), represented by M. Meulenbelt, B. Natens, A.‑S. Melin and C. Muttin, lawyers,

applicant,

v

European Commission, represented by L. Haasbeek and A. Sipos, acting as Agents,

defendant,

APPLICATION pursuant to Article 263 TFEU seeking the annulment of (i) the Commission Implementing Decision C(2018) 252 final of 15 January 2018 refusing marketing authorisation under Regulation (EC) No 726/2004 of the European Parliament and of the Council of 31 March 2004 laying down Community procedures 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) for the medicinal product for human use Fanaptum — iloperidone, and (ii) the opinion and the assessment report of the Committee for Human Medicinal Products of the European Medicines Agency (EMA) of 9 November 2017.

THE GENERAL COURT (Sixth Chamber),

composed of D. Spielmann (Rapporteur), acting as President, Z. Csehi and O. Spineanu‑Matei, Judges,

Registrar: P. Cullen, Administrator,

having regard to the written part of the procedure and further to the hearing on 8 July 2019,

gives the following

Judgment

 Background to the dispute

1        The applicant, Vanda Pharmaceuticals Ltd, is the holder in the United States of a marketing authorisation (‘MA’) for the medicinal product Fanaptum, which contains the active substance iloperidone (‘iloperidone’).

2        The medicinal product Fanaptum, indicated for the treatment of the symptoms of schizophrenia in adults, has a profile consistent with what is known as a ‘second generation’ anti-psychotic medicine. It has been marketed in the United States since 2010, and in Israel and Mexico since 2012.

3        On 4 December 2015 the applicant, pursuant to Article 4 of Regulation (EC) No 726/2004 of the European Parliament and of the Council of 31 March 2004 laying down Community procedures 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), as amended, submitted to the European Medicines Agency (EMA), an application for an MA for the medicinal product Fanaptum.

4        That application, submitted under Article 3(2)(a) of Regulation No 726/2004, was based on a dossier composed of administrative information, complete quality data, clinical and non-clinical data based on the applicant’s own tests and studies and bibliographic literature substituting and supporting certain tests or studies.

5        On 5 May 2017 the EMA convened a scientific advisory group to address questions raised by the Committee for Human Medicinal Products (‘the CHMP’). In that context, the applicant circulated a briefing note and delivered a presentation.

6        On 17 May 2017 the applicant gave an oral explanation before the CHMP. It addressed the outstanding issues by means of a presentation.

7        On 20 July 2017 the CHMP delivered a negative opinion and adopted an assessment report, recommending that an MA be refused for the medicinal product Fanaptum.

8        On 27 July 2017 the applicant requested the EMA to re-examine the CHMP’s negative opinion of 20 July 2017. On 26 September 2017 the applicant submitted detailed grounds in support of that request for re-examination.

9        On 30 October 2017 another scientific advisory group was convened to address questions raised by the CHMP during the re-examination procedure for iloperidone. In that context, the applicant delivered a presentation.

10      On 5 November 2017 the final joint assessment report was circulated to all CHMP members. On 6 November 2017 the applicant received the ‘Updated (co)-Rapporteurs Joint Assessment Report of the Grounds for the Re-examination Procedure’, which was dated 11 October 2017.

11      On 6 November 2017 the applicant received the minutes of the October 2017 meeting of the scientific advisory group. The applicant submitted its comments on those minutes on the same day.

12      On 7 November 2017 the applicant gave an oral explanation before the CHMP.

13      During that hearing, the CHMP and the applicant examined the detailed grounds for re-examination. The applicant explained its re-examination request by means of a presentation.

14      On 9 November 2017 a document entitled ‘Scientific conclusions and grounds for refusal’ containing the EMA opinion drawn up by the CHMP (‘the CHMP opinion’) and the CHMP scientific assessment report (‘the CHMP assessment report’) were sent to the applicant.

15      On 15 January 2018 the European Commission adopted Implementing Decision C(2018) 252 final refusing to grant an MA under Regulation No 726/2004 for ‘Fanaptum — iloperidone’, a medicinal product for human use (‘the Implementing Decision’), which was notified to the applicant on 16 January 2018.

16      Annex I to the Implementing Decision, entitled ‘Scientific conclusions and grounds for refusal presented by the European Medicines Agency’, which corresponds to the CHMP opinion, states, in particular, the following:

‘Considering all available non-clinical and clinical data (including the thorough QTc study, the overall clinical program[me] and the cases of cardiac-related/sudden unexplained death in clinical trials and post-marketing), [i]loperidone has a substantial and exposure-dependent arrhythmogenic potential. It is not considered that the risk minimisation measures proposed would appropriately address the risk identified in this specific case. Hence, the safety of [i]loperidone has not been sufficiently demonstrated.

Furthermore, [i]loperidone has a modest efficacy. In addition, it has shown a delayed onset of action, which is a significant concern in the treatment of acute exacerbation of schizophrenia. Therefore, and taking into account the overall safety and efficacy profile of [i]loperidone a patient population cannot be identified where the benefit of treatment is considered to outweigh the major safety concerns.

Based on the above, the risk-benefit balance of [i]loperidone is considered negative.’

 Procedure and forms of order sought

17      By application lodged at the Court Registry on 26 March 2018, the applicant brought the present action.

18      By document lodged at the Court Registry on 4 May 2018, the applicant submitted a reasoned request, in accordance with Article 66 of the Rules of Procedure of the General Court, requesting that certain data contained in an annex to the application be omitted from the documents relating to the present case to which the public might have access. On the basis of the information provided by the applicant, it was decided to grant that request.

19      By document lodged at the Court Registry on 13 July 2018, the Commission lodged a defence.

20      The reply was lodged at the Court Registry on 18 September 2018.

21      On 30 October 2018 the Commission lodged a rejoinder at the Court Registry, following which the written procedure was closed.

22      By document lodged at the Court Registry on 27 November 2018, the applicant requested a hearing, on the basis of Article 106(2) of the Rules of Procedure.

23      On a proposal from the Judge-Rapporteur, the General Court (Sixth Chamber) decided to open the oral procedure and put a question to the parties with a request for written answer as part of the measures of organisation of procedure provided for in Article 89 of the Rules of Procedure. The parties complied with those measures within the prescribed period.

24      The parties presented oral argument and answered the questions put to them by the Court at the hearing on 8 July 2019.

25      The applicant claims that the Court should:

–        declare the action to be admissible;

–        annul both the implementing decision and also the CHMP opinion and assessment report of 9 November 2017;

–        in the alternative, annul only the implementing decision;

–        order the Commission to pay the costs.

26      The Commission contends that the Court should:

–        dismiss the action as being inadmissible in part and, in any event, as being unfounded;

–        order the applicant to pay the costs.

 Law

 Admissibility

27      The Commission claims that the present action for annulment is inadmissible in so far as it is directed against the CHMP opinion and assessment report. It submits that, although, in accordance with established case-law, those two acts form an integral part of the Implementing Decision, the fact remains that they are preparatory measures for that decision. In the rejoinder, the Commission states that, although the applicant seems to assert in the reply that it never intended separately to request the annulment of the CHMP opinion and assessment report, such an intention is not clear from the form of order set out in the application.

28      The applicant observes that the Commission does not dispute the admissibility of the present action in so far as it is directed against the Implementing Decision. It maintains moreover that, contrary to what is suggested by the Commission, it does not maintain that the CHMP opinion and assessment report should ‘be subject to a separate and self-standing action’. Since, in accordance with the case-law, the CHMP opinion and assessment report form an integral part of the Implementing Decision, the contested act consists of those three acts.

29      According to settled case-law, only measures which definitively lay down an institution’s position upon the conclusion of that procedure constitute acts open to challenge under Article 263 TFEU, with the exception of ‘provisional measure[s] intended to pave the way for the final decision’ (judgment of 11 November 1981, IBM v Commission, 60/81, EU:C:1981:264, paragraph 10; see also judgments of 26 January 2010, Internationaler Hilfsfonds v Commission, C‑362/08 P, EU:C:2010:40, paragraph 52 and the case-law cited, and of 15 March 2017, Stichting Woonpunt and Others v Commission, C‑415/15 P, EU:C:2017:216, paragraph 44 and the case-law cited).

30      It is also apparent from the case-law that, in the context of the procedure for examining an MA application, the final scientific opinion — in the present case the CHMP opinion — constitutes provisional measure intended to prepare for the decision on that application. It is a preparatory measure which does not definitively lay down the Commission’s position and is therefore not a challengeable act within the meaning of the case-law cited in paragraph 29 above (see, to that effect, judgment of 18 December 2003, Olivieri v Commission and EMEA, T‑326/99, EU:T:2003:351, paragraph 53).

31      These considerations apply mutatis mutandis to the CHMP assessment report, which itself forms part of the CHMP opinion. Neither an opinion nor an assessment report of the CHMP can be regarded as a definitive measure, since their only purpose is to prepare the draft Commission decision in accordance with Article 10(1) of Regulation No 726/2004 and the final Commission decision, which must be adopted on the basis of Article 10(2) of Regulation No 726/2004.

32      In addition, it follows from the case-law that, in so far as a decision purely and simply confirms the EMA’s opinion, the content of that opinion, and also that of the assessment report on which it is based, are an integral part of the statement of reasons for that decision, with regard, in particular, to the scientific assessment of the medicinal product in question (judgment of 11 June 2015, Laboratoires CTRS v Commission, T‑452/14, not published, EU:T:2015:373, paragraph 60; see also, to that effect, judgment of 18 December 2003, Olivieri v Commission and EMEA, T‑326/99, EU:T:2003:351, paragraph 55).

33      Therefore, neither the CHMP opinion nor its assessment report, which are incorporated into the Implementing Decision, can be subject to a separate and self-standing action for annulment. Indeed, the forms of order seeking annulment of the CHMP opinion and report have no independent content and are in fact conflated with those directed against the Implementing Decision.

34      Consequently, and although the applicant stated at the hearing that it was not seeking, by the present action, a ‘separate’ annulment of the CHMP opinion and assessment report, the action for annulment must be declared to be inadmissible in so far as it is directed against the CHMP opinion and assessment report. However, the present action is admissible in so far as it is directed against the Implementing Decision (‘the contested decision’).

35      That being said, in assessing the statement of reasons for and merits of the contested decision, the CHMP opinion and assessment report must be taken into account. In so far as that decision purely and simply confirms the EMA’s opinion, the content of that opinion, and also that of the assessment report on which it is based, are an integral part of the statement of reasons for that decision, with regard in particular to the scientific evaluation of the medicinal product in question.

36      As regards the correct version of the assessment report that must be taken into account, which is the subject of debate between the parties, the report at issue is the CHMP report of 9 November 2017.

37      Since, as stated in recital 23 of Regulation No 726/2004, ‘exclusive responsibility for preparing the [EMA]’s opinions on all questions concerning medicinal products for human use should be vested in a Committee for Medicinal Products for Human Use’ and not in the individual members of that committee, in general, no account may be taken of the ‘Updated (co)-Rapporteurs Joint Assessment Report of the Grounds for the Re-examination Procedure’ of 11 October 2017, which, as the applicant claims, was circulated to all CHMP members on 5 November 2017 as the ‘final assessment report’.

38      As the Commission has correctly observed, that report is a document which sets out the views of the co-Rapporteurs and which has been prepared in order to facilitate scientific debate with the applicant and within the collegiate body of the CHMP. That document in no way reflects the final opinions of the CHMP, in so far as that body expresses itself through scientific opinions or recommendations adopted by consensus or by an absolute majority of its members.

39      However, while the provisional assessment reports drawn up under the evaluation procedure, including the documents drawn up by the co-Rapporteurs, such as, in the present case, the ‘Updated (co)-Rapporteurs Joint Assessment Report’ of 11 October 2017 drawn up under the re-examination procedure, must be distinguished from the CHMP’s final assessment report, it cannot at the outset be excluded that those provisional documents may provide the Court with guidance in certain respects. In particular, these documents may be taken into account for the purposes of determining whether the scientific findings finally adopted by the CHMP in the context of the examination procedure for the MA application at issue in the present case are vitiated by a failure to state reasons or manifest errors of assessment.

 Substance

40      The applicant puts forward five pleas in law in support of its action. The first plea alleges that the assessment of the risks of arrhythmia that might be caused by iloperidone is vitiated by a failure to state reasons, a manifest error of assessment and a breach of the principle of equal treatment. The second plea alleges that the assessment of the risk minimisation measures (‘RMMs’) proposed for iloperidone is vitiated by a failure to state reasons, a manifest error of assessment and a breach of the principle of proportionality, set out in Article 5(1) and (4) TEU, and of the principle of equal treatment. The third plea alleges that the assessment of the consequences of the delayed onset of iloperidone is vitiated by a failure to state reasons and a breach of the principle of proportionality. The fourth plea alleges that the obligation to identify a population in which iloperidone would outperform other products constitutes a breach of the principles of conferral and of proportionality (set out in Article 5(1) to (3) TEU), Article 12 and Article 81(2) of Regulation No 726/2004 as well as the principle of equal treatment. The fifth and final plea alleges that the overall risk-benefit assessment of iloperidone is vitiated by an inadequate statement of reasons and, in any event, is manifestly erroneous.

 Preliminary considerations on the nature and extent of judicial review

41      In the present case, it seems appropriate to set out, first of all, a number of general considerations concerning, first, the centralised procedure for MAs for medicinal products for human use (as governed in particular by Regulation No 726/2004) and, second, the nature and extent of the review that the Court is invited to carry out when an applicant for an MA challenges the scientific findings which led the competent authorities to propose that an MA be refused.

–       Review of the main features of the centralised procedure for MAs for medicinal products as governed by Regulation No 726/2004

42      It is apparent from the preamble to Regulation No 726/2004 (see in particular recital 19) that the centralised procedure for the authorisation of medicinal products provided for in that regulation is based on a scientific evaluation procedure being conducted by the EMA, which is to apply the highest possible standards, to address the quality, safety and efficacy of medicinal products. One of the main objectives of the authorisation regime provided for in that legislation is to ensure that patients are not administered medicinal products with a negative risk-benefit balance. In that respect, in accordance with Article 6(1) of Directive 2001/83/EC of the European Parliament and of the Council of 6 November 2001 on the Community code relating to medicinal products for human use (OJ 2001 L 311, p. 67), as amended by Directive 2012/26/EU of the European Parliament and of the Council of 25 October 2012 amending Directive 2001/83/EC as regards pharmacovigilance (OJ 2012 L 299, p. 1) ‘no medicinal product may be placed on the market of a Member State unless a marketing authorisation has been issued by the competent authorities of that Member State in accordance with [that] Directive or an authorisation has been granted in accordance with Regulation … No 726/2004, read in conjunction with Regulation (EC) No 1901/2006 of the European Parliament and of the Council of 12 December 2006 on medicinal products for paediatric use [(OJ 2006 L 378, p. 1)]’.

43      It is apparent from a combined reading of Article 1(28a) and Article 26 of Directive 2001/83, as well as Article 12 of Regulation No 726/2004, read in the light of recital 14 of that regulation, that an MA is to be refused where, after verification of the relevant particulars and documents, it proves that the risk-benefit balance of that medicinal product in question is not held to be favourable. In that respect, the positive therapeutic effects of the medicinal product concerned must be assessed in relation to the risks related to its use, that is to say, all risks with regard to the quality, safety or efficacy of the medicinal product as regards patients’ health or public health (see Article 1(28) of Directive 2001/83).

44      In that context, the applicant bears the burden of proving that the conditions for the authorisation of a medicinal product are met and must, in particular, provide scientific data in order to prove that the medicinal product is safe and effective (see, to that effect, judgment of 26 November 2002, Artegodan v Commission, T‑74/00, T‑76/00, T‑83/00 to T‑85/00, T‑132/00, T‑137/00 and T‑141/00, EU:T:2002:283, paragraphs 187 and 188). In that regard, Article 12(1) of Regulation No 726/2004 expressly provides that an MA is to be refused if, after verification of the particulars and documents submitted in accordance with Article 6 of that regulation, it appears that the applicant has not properly or sufficiently demonstrated the quality, safety or efficacy of the medicinal product. In other words, it is not for the authority responsible for the examination of an MA to prove that a product is not safe, but rather for the MA applicant to establish that the medicinal product in question has a favourable risk-benefit balance.

45      In addition, the decision whether or not to grant an MA, which must be founded on a high standard of public health protection, must be taken solely on the basis of the criteria of safety and efficacy arising from the relevant provisions of EU law. While it cannot be excluded that an MA applicant may rely on pre- and post-marketing data concerning third countries, no argument can be inferred, in absolute terms, from the fact that an MA has been granted in those countries (see, by analogy, as regards the grant of the status of orphan medicinal product, judgment of 9 September 2010, Now Pharm v Commission, T‑74/08, EU:T:2010:376, paragraph 57).

46      In that respect, it must be noted that, in the same way as the substantive conditions for the suspension or withdrawal of an MA, the conditions for granting an MA must be interpreted in accordance with the general principle that can be identified from case-law that the protection of public health must unquestionably take precedence over economic considerations (see judgment of 19 April 2012, Artegodan v Commission, C‑221/10 P, EU:C:2012:216, paragraph 99 and the case-law cited). Furthermore, the precautionary principle, which is a general principle of EU law, empowers, inter alia, the Commission to do no more than establish whether there is solid and persuasive evidence which may raise reasonable doubts regarding the safety of the medicinal product concerned or of the existence of a favourable risk-benefit balance (see, by analogy, regarding decisions to suspend, withdraw or vary an MA, judgment of 3 December 2015, PP Nature-Balance Lizenz v Commission, C‑82/15 P, not published, EU:C:2015:796, paragraph 23 and the case-law cited).

47      In practice, and as the parties agree, the process of evaluating an MA application is based on several stages of scientific assessment and discussions with the MA applicant. Initially, the data submitted in the application are assessed in parallel by two independent teams (the co-Rapporteurs’ teams), which report their initial conclusions and recommendations. Those are the subject of a first scientific evaluation by the CHMP. Following that first evaluation, the CHMP initiates a discussion with the applicant concerning the general evaluation, pointing out any gaps in the data and analysis presented and, where applicable, making one or more requests for information before setting out a final conclusion. The final conclusion on an MA application is therefore only adopted after several rounds of discussions in the course of which several provisional documents are drawn up, which may include reports from certain Rapporteurs. These provisional documents only reflect the progress of the evaluation at a given stage. Once the CHMP is satisfied with the evaluation, these reports are subject to critical peer review at the committee level and amended based on the results of the committee’s discussions.

48      In the evaluation process, ‘exclusive responsibility’ for preparing EMA opinions on all matters relating to medicinal products for human use is vested in the CHMP (see recital 23 and Article 5(2) of Regulation No 726/2004). Consequently, a distinction must be drawn between any documents drawn up by the co‑Rapporteurs — in the present case particularly the joint assessment report of 11 October 2017 (circulated to all CHMP members on 5 November 2017) to which the applicant intends, in part, to refer — and the final assessment report of the CHMP on which the contested decision is based.

–       The extent and scope of the judicial review

49      Clarity is required as to the scope and extent of judicial review, which in the present case is the subject of dispute between the parties. In its reply, the applicant submits that the Commission is seeking, by putting forward an excessively restrictive interpretation of the extent of judicial review, to adopt a ‘strategy of obfuscation’ intended to convince the Court not to examine the merits of the pleas and to prevent it from addressing several fundamental questions concerning the compatibility of the contested decision with EU law. The applicant submits that it is apparent from the case-law that the EU judicature is in a position to assess the legality of the scientific evaluation made by the CHMP (see, to that effect, judgment of 7 March 2013 Acino v Commission, T‑539/10, not published, EU:T:2013:110, paragraph 92 and the case-law cited) and, where appropriate, to determine whether the unlawfulness of that assessment constitutes a breach of an essential procedural requirement rendering the Commission’s decision unlawful (see, to that effect, judgment of 26 November 2002, Artegodan v Commission, T‑74/00, T‑76/00, T‑83/00 to T‑85/00, T‑132/00, T‑137/00 and T‑141/00, EU:T:2002:283, paragraph 197).

50      In that respect, according to the case-law, a distinction must, as a general rule, be drawn between the review that the EU judicature may be called upon to carry out, on the one hand, of the formal legality of the CHMP’s scientific opinion and, on the other hand, the Commission’s exercise of its discretion (see, to that effect, judgments of 7 March 2013, Acino v Commission, T‑539/10, not published, EU:T:2013:110, paragraph 92, and of 11 December 2014, PP Nature-Balance Lizenz v Commission, T‑189/13, not published, EU:T:2014:1056, paragraph 33; see also, to that effect and by analogy, judgment of 26 November 2002, Artegodan v Commission, T‑74/00, T‑76/00, T‑83/00 to T‑85/00, T‑132/00, T‑137/00 and T‑141/00, EU:T:2002:283, paragraph 199).

51      As regards the Commission’s exercise of its discretion, it must be recalled that it is apparent from settled case-law that, where the EU authorities have a broad discretion, in particular as to the assessment of highly complex scientific and technical facts in order to determine the nature and scope of the measures which they adopt, review by the EU judicature is limited to verifying whether there has been a manifest error of assessment or a misuse of powers, or whether those authorities have manifestly exceeded the limits of their discretion. In such a context, the EU judicature cannot substitute its assessment of scientific and technical facts for that of the EU authorities on which alone the FEU Treaty has placed that task (see, to that effect, judgments of 21 July 2011, Etimine, C‑15/10, EU:C:2011:504, paragraph 60; of 30 April 2015, Polynt and Sitre v ECHA, T‑134/13, not published, EU:T:2015:254, paragraph 52; and of 11 May 2017, Deza v ECHA, T‑115/15, EU:T:2017:329, paragraph 163).

52      It must be stated that the discretion of the EU authorities, which implies limited judicial review of its exercise, does not concern only the nature and scope of the measures to be taken but also, to some extent, the finding of the basic facts. However, even though such judicial review is of limited scope, it requires that the EU authorities which have adopted the act in question must be able to show before the EU judicature that in adopting the act they actually exercised their discretion, which presupposes the taking into consideration of all the relevant factors and circumstances of the situation the act was intended to regulate (see judgments of 8 July 2010, Afton Chemical, C‑343/09, EU:C:2010:419, paragraphs 33 and 34 and the case-law cited; of 30 April 2015, Polynt and Sitre v ECHA, T‑134/13, not published, EU:T:2015:254, paragraph 53; and of 11 May 2017, Deza v ECHA, T‑115/15, EU:T:2017:329, paragraph 164).

53      As regards judicial review of the CHMP opinion — and by extension of its assessment report — the Court cannot substitute its own assessment for that of the CHMP. Only the proper functioning of the CHMP, the internal consistency of the opinion and the statement of reasons contained in it can be subject to judicial review. As regards the last aspect, the review consists of examining whether these documents contain a statement of reasons from which it is possible to ascertain the considerations on which they are based and whether they establish a comprehensible link between the medical or scientific findings and the conclusions (see, to that effect, judgments of 7 March 2013, Acino v Commission, T‑539/10, not published, EU:T:2013:110, paragraph 93, and of 5 December 2018, Bristol-Myers Squibb Pharma v Commission and EMA, T‑329/16, not published, EU:T:2018:878, paragraph 99; see also, to that effect and by analogy, judgment of 26 November 2002, Artegodan v Commission, T‑74/00, T‑76/00, T‑83/00 to T‑85/00, T‑132/00, T‑137/00 and T‑141/00, EU:T:2002:283, paragraph 200).

54      However, it must be noted that, in the present case, the Commission has not departed from the CHMP opinion, since the content of that opinion, as well as that of the assessment report on which it is based, is an integral part of the reasoning behind that decision, in particular as regards the scientific evaluation of the medicinal product at issue (see paragraphs 16 and 35 above). The Commission has therefore endorsed the findings in that opinion. Consequently, it must be held that the judicial review to be undertaken by the Court, in particular its review of whether there has been a manifest error of assessment, must be carried out on the basis of all the considerations set out in that opinion and the assessment report mentioned above (see, to that effect, judgment of 5 December 2018, Bristol-Myers Squibb Pharma v Commission and EMA, T‑329/16, not published, EU:T:2018:878, paragraph 98).

 The first plea in law: the risk assessment of the arrhythmic potential of iloperidone is vitiated by a failure to state reasons, a manifest error of assessment and a breach of the principle of equal treatment

55      By its first plea, the applicant submits, in essence, that the contested decision is vitiated by a failure to state reasons, manifest errors of assessment and a breach of the principle of equal treatment with regard to the assessment of risks linked to the arrhythmogenic potential of iloperidone arising from the prolongation of the QT interval, a change in the electrical activity of the heart (‘QT interval’), which can result in an abnormal heart rate which can, in certain circumstances, be life-threatening.

56      In the first place, the applicant submits that the evaluation of the risks of arrhythmia carried out in the present case departed from the applicable guidelines, in particular the Note for Guidance on the Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-Antiarrhythmic drugs (CHMP/ICH/2/04; ‘the QT Guidance’), which defines the relevant parameters when evaluating the risks associated with medicinal products which may prolong the QT interval. In particular, the applicant asserts that, in diverging from the findings of the joint assessment report of 11 October 2017, the CHMP failed, on the one hand, to explain how the data and calculations provided by the applicant had been found not to be relevant and, on the other hand, to take account of the post-marketing experience in relation to iloperidone.

57      In the second place, the applicant considers that that risk assessment is not consistent with the current EMA practice at the material time and, consequently, infringes the principle of equal treatment. The applicant submits, in particular, that contrary to the approach which it took when evaluating other products designed for the treatment of schizophrenia, namely lurasidone and cisapride, the CHMP refused to consider the extensive post-marketing data which the applicant had provided and in particular the positive experience acquired for iloperidone on other markets, even taking into consideration the significant rate of ‘under-reporting’.

58      In the third place, the applicant asserts that, since the CHMP failed to state the reasons why it had diverged from the data submitted by the applicant on the two aspects mentioned in paragraph 56 above, the CHMP’s conclusions must be regarded as inadequately reasoned and, in any event, as being manifestly erroneous. In that context, the applicant states that, in the absence of a convincing statement of reasons, it can only assume that the refusal decision at issue in the present case was informed, first, by the unsubstantiated allegation of the Scientific Advisory Panel of October 2017, based on the personal opinion of one of its members, that iloperidone has caused a ‘very high’ number of sudden and unexpected deaths and, second, by the participation in that group of an expert who served as a consultant for a product competing with iloperidone.

59      The Commission submits that the first plea should be rejected since, in its submission, that plea tends more to show disagreement with the scientific findings forming the basis of the contested decision than to question a failure to state reasons.

–       Whether the obligation to state reasons was complied with

60      According to settled case-law, the statement of reasons required by the second paragraph of Article 296 TFEU must be appropriate to the act at issue and must disclose in a clear and unequivocal fashion the reasoning followed by the institution which adopted the measure in question, in such a way as to enable the persons concerned to ascertain the reasons for the measure and to enable the competent court to exercise its power of review. The requirements to be satisfied by the statement of reasons depend on the circumstances of each case, in particular the content of the measure in question, the nature of the reasons given and the interest which the addressees of the measure, or other parties to whom it is of direct and individual concern, may have in obtaining explanations. It is not necessary for the reasoning to go into all the relevant facts and points of law, since the question whether the statement of reasons for a measure meets the requirements of Article 296 TFEU must be assessed with regard not only to its wording but also to its context and to all the legal rules governing the matter in question (see judgments of 2 April 1998, Commission v Sytraval and Brink’s France, C‑367/95 P, EU:C:1998:154, paragraph 63 and the case-law cited, and of 28 March 2017, Rosneft, C‑72/15, EU:C:2017:236, paragraph 122 and the case-law cited).

61      As regards in particular decisions on an MA, Article 81(1) of Regulation No 726/2004, under which all decisions to grant, refuse, vary, withdraw or revoke an MA are to state in detail the reasons on which they are based, simply states expressly the general obligation to state reasons laid down in the second paragraph of Article 296 TFEU (see, to that effect, judgment of 10 April 2014, Acino v Commission, C‑269/13 P, EU:C:2014:255, paragraphs 121 and 122).

62      In the present case, the refusal to grant the MA at issue was decided on the basis of the CHMP assessment report and the CHMP opinion (the latter being included in Annex I to the contested decision under the heading ‘Scientific conclusions and grounds for refusal presented by the European Medicines Agency’), which, as recalled in the above preliminary observations (see in particular paragraph 32 above), form an integral part of that decision.

63      However, it is apparent from these documents that the scientific reasons, as examined over several dozen pages, why iloperidone was considered to present risks to patient safety, particularly in view of its arrhythmogenic potential, were clearly stated. The CHMP concluded, in particular, in its assessment report that the data submitted to it indicated that, despite the RMMs proposed, that medicinal product resulted in a prolongation of the QT interval, which could, in certain situations, be fatal for the patient.

64      In the CHMP opinion the reasons for the MA refusal include the following:

‘Considering all available non-clinical and clinical data (including the thorough QTc study, the overall clinical program[me] and the cases of cardiac-related/sudden unexplained death in clinical trials and post-marketing), [i]loperidone has a substantial and exposure-dependent arrhythmogenic potential. It is not considered that the risk minimisation measures proposed would appropriately address the risk identified in this specific case. Hence, the safety of [i]loperidone has not been sufficiently demonstrated.’

65      More generally, the CHMP opinion and assessment report clearly state the reasons on which the contested decision is based. They, inter alia, disclose in a clear and unequivocal fashion the reasoning followed by the institution which adopted the measure in question, in such a way as to enable the applicant to ascertain the reasons for the measure and to enable the Court to exercise its power of review. It must also be noted that the question of the arrhythmogenic potential of iloperidone has, in the present case, been at the heart of the concerns expressed by the CHMP regarding the safety of that substance, both in its first opinion dated 20 July 2017 and in its final assessment report of 9 November 2017 (drawn up as part of the re-examination procedure following the applicant’s request). The CHMP took, in particular, the view that, despite the RMMs proposed, including those submitted at the stage of the re-examination of the MA application, the risk of QT interval prolongation remained substantial. The CHMP had, in particular, expressed concern that the substance was broken down in the body by liver enzymes whose activity could be reduced in certain patients or by certain other medicines. As the magnitude of that risk was considered to be greater than the expected benefit, the CHMP confirmed its negative opinion regarding the MA for iloperidone.

66      It must consequently be held that the CHMP opinion and assessment report on which the contested decision is based are not vitiated by a failure to state reasons, in that they enabled the persons concerned to ascertain the reasons for the measure and enabled the Court to exercise its power of review.

67      It therefore appears that, by its arguments, the applicant in reality seeks to criticise the scientific conclusions reached by the CHMP as being manifestly erroneous and infringing the principle of equal treatment.

68      The applicant’s various complaints should be examined in the light of the foregoing preliminary observations.

–       The allegation that the iloperidone risk assessment is not consistent with the guidelines for QT/QTc prolongation

69      The applicant submits that the CHMP failed to comply with the applicable guidelines, and more specifically the QT Guidance, when evaluating the safety of iloperidone.

70      As a preliminary point, it should be noted that the authorities responsible for processing MA applications, which include the EMA, may be called upon to draw up guidance intended not only to direct the processing of those applications, but also to inform applicants, in order to provide transparency and foreseeability, of the parameters to be taken into account in the evaluation of the scientific and technical data they are to provide in support of their application.

71      Although such ‘guidance’ or ‘guidelines’ are not legally binding, they may be taken into account to a certain extent in the assessment of the risk-benefit balance of a medicinal product by way of additional evidence (see, by analogy, judgment of 16 October 2003, AstraZeneca, C‑223/01, EU:C:2003:546, paragraph 28). That is particularly so where, as in the present case, technically or scientifically complex issues are involved.

72      The present case involves the QT Guidance, the purpose of which is found in the statement, reiterated in the introduction to that guidance, that certain medicinal products have the adverse effect of delaying cardiac repolarisation — an effect that can be measured as prolongation of what is commonly referred to as the QT interval on the surface electrocardiogram (‘ECG’) — and, ultimately, of bringing about conditions which can lead to sudden death.

73      It is not disputed in the present case that the arrhythmogenic potential of a substance, which can be detected because of a significant prolongation of the QT/QTc interval, clearly constitutes a serious risk, to which particular importance should be attached when evaluating medicinal products before they are placed on the market. Paragraph 5.1 of the QT Guidance explains, to that effect, that ‘substantial prolongation of the QT/QTc interval, with or without documented arrhythmias, could be the basis for non-approval of a drug or discontinuation of its clinical development, particularly when the drug has no clear advantage over available therapy and available therapy appears to meet the needs of most patients’.

74      In the present case, by its arguments, the applicant seeks, essentially, to criticise the failure to take into account the QT Guidance with regard to three aspects, namely the definition of the worst case scenario, the taking into account of relevant data and the thresholds which must be applied when evaluating the safety of medicinal products.

75      In the first place, as regards the account taken of the ‘worst case scenario’ in the present case, the QT Guidance specifies in particular that ‘it is important to identify the “worst case scenario” for drugs that have demonstrated effects on the QT/QTc interval as part of a risk assessment (i.e., the QT/QTc interval measured in the target patient population at the time of the peak effect and under the highest blood levels that can be attained during therapy)’.

76      However, the CHMP did not make a manifest error of assessment when it concluded that, in the context of iloperidone’s risk assessment, the worst case scenario concerns patients who have also been prescribed medicinal products that mildly inhibit the main metabolic pathways for iloperidone. As explained by the Commission in its submissions, it is accepted that iloperidone’s arrhythmogenic potential increases with the blood concentration level of that substance. In other words, the slower the rate at which iloperidone is metabolised, the higher the patient’s exposure to that substance and, therefore, the greater the risk involved.

77      Contrary to the applicant’s assertion, in reaching that conclusion, the CHMP did not confine itself to theoretical reasoning, but took into account all the data that had been brought to its attention, and in particular the clinical and non-clinical scientific data that had been submitted to it (see paragraphs 81 to 88 below).

78      Furthermore, and in any event, although the definition of the worst case scenario is, as set out in the QT Guidance, important in assessing the risks presented by medicinal products resulting in an increase in the QT interval, it appears, in the present case, that the CHMP concluded that the risk-benefit balance of iloperidone was negative regardless of the population considered, since it was scientifically difficult to identify a priori a population who are poor metabolisers of iloperidone.

79      When invited at the hearing to provide explanations on the importance and scope of the definition of the ‘worst case scenario’, the Commission explained, without being contradicted on that point, that it was particularly difficult — if not impossible — to define a priori the factors inhibiting the metabolism of iloperidone as they were so numerous and unpredictable. In particular, it stated that the concomitant use of common or widespread consumer products (such as camomile, liquorice or even vitamin D) could slow down the metabolism of iloperidone — and therefore expose the persons concerned to the cardiac risks presented by that substance — even for patients who were in principle, in the light of their genetic profile, capable of rapidly metabolising that substance.

80      It follows that not only is there no manifest error of assessment in the definition of the sub-population for which the risk presented by iloperidone would be increased, but that, in addition, and in any event, the CHMP concluded that that risk exists regardless of the population or scenario envisaged.

81      In the second place, as regards the factors taken into account by the CHMP in reaching its conclusion on the arrhythmogenic potential of iloperidone, these comprise, as mentioned in the CHMP opinion, four data sources. The CHMP in fact referred to ‘all available non-clinical and clinical data (including the thorough QTc study, the overall clinical program[me] and also the cases of cardiac-related/sudden unexplained death in clinical trials and post-marketing)’.

82      First, as regards the pre-clinical data, the CHMP noted the following:

‘Iloperidone and the P88 metabolite revealed higher affinity for the hERG channel than other antipsychotic agents and concentration-dependently prolonged action potential durations in dog Purkinje fibers. Although no effects on ECG were seen in dog studies, [i]loperidone is considered to have a high torsadogenic potential, which constitutes a major safety risk for patients.’

83      Subsequently, as regards the thorough QTc interval study, the CHMP assessment report states as follows:

‘The thorough 2328 study is considered to show substantial QT prolongation. Patients were randomly assigned to receive iloperidone (ILO) 8 mg [twice daily], ILO 12 mg [twice daily] (the maximum recommended therapeutic dose), ILO 24 mg [once daily], ziprasidone 80 mg [twice daily] (positive control) or quetiapine 375 mg [twice daily] (negative control) in the absence (period 1) and presence of single (period 2) and dual (2D6 & 3A4 — period 3) metabolic inhibition.

The fact that no subjects in this study experienced a QT or QTc value of greater than 500 msec is not necessarily reassuring as this is a group of subjects with no risk factors and a normal QT interval at baseline, and the number of subjects in the trial was not large (about 30 per group). The observation that seven subjects in the iloperidone treatment [groups] experienced a change in QTc value of [more than] 60 msec at Tmax is evidence of a potential major safety issue.

It is also importantly noted, regarding the same study, that out of 94 patients exposed to [i]loperidone at different doses without metabolic inhibition (i.e. in Treatment Period 1) in the Secondary QTc population 43 and 2 patients respectively developed a prolongation of the QTcF of more than 30 and 60 msec.’

84      In addition, as regards the overall clinical programme, the CHMP assessment report makes the following statements:

‘Looking at the clinical safety data, Safety group 1, 4.5% of the patients treated with [i]loperidone, regardless of dose (4-24 mg/day), had an increase o[f] more than 60 msec at some time point during the clinical studies. In the ziprasidone group (160 mg/day) it was 1.6%.

3 patients had at some time-point a QTcF interval of more than 500 msec ([i]loperidone 10-16 mg/day group). This was not seen in the ziprasidone group although fewer patients were exposed to ziprasidone.

More deaths occurred in the [i]loperidone group than in any of the other groups and 6 of the deaths might be linked to QT prolongation (arrhythmia, sudden cardiac arrest and sudden death). Considering 4 423 patients have in the clinical trial program[me] been exposed to [i]loperidone, 0.14% of all treated patients died due to sudden death or death due to cardiac AE. This equates to a number needed to harm of 714. Or to put it differently for every 714 patients treated with [i]loperidone, one will die of a sudden death or a cardiac death.’

85      It follows from all that data that, contrary to the applicant’s assertions, it had been established that iloperidone could result in a risk of QT interval prolongation of more than 30 msec — and not an average level of between 5 and 30 msec — which was liable to expose patients to risks of ‘torsades de pointes’ and sudden death.

86      Finally, as regards the post-marketing data and in particular the cases of cardiac or sudden unexplained death which had been recorded in the United States, it should be recalled that the CHMP assessment report states as follows:

‘… from the Vanda global US post-marketing surveillance database through 24 August 2016 there were 33 deaths in total. 3 patients died during sleep, 6 were sudden deaths, 6 cases were of cardiac origin. The rest were suicides (6), unknown (7), 2 … died of other reasons and 3 died due to pulmonary embolism.’

87      Even if it were the case (which it is not) as the applicant claims in its reply that the number of cardiac deaths recorded in the United States cannot be regarded as ‘alarming’, the CHMP could, without exceeding the limits of its discretion when evaluating scientific data submitted to it, take the view that that number constituted evidence of iloperidone’s pro-arrhythmic potential and thus of the safety risk presented by that substance.

88      The fact that it is a ‘potential’ risk which is identified explains why the CHMP found it necessary to come to a negative opinion. Contrary to the applicant’s assertion, the CHMP cannot be required to establish the existence of a ‘significant actual risk’ such as a significant increase in cardiac mortality.

89      Having regard to all that data and those considerations, taken as a whole, it is plain that the CHMP committed no manifest error of assessment and that the conclusion it reached, that there was a population of patients who would be exposed to real and unacceptable safety risks through iloperidone treatment, was consistent with its medical and scientific findings.

90      In the third place, as regards the merits of the applicant’s claim that the risks of QT interval prolongation associated with iloperidone did not reach the ‘threshold levels of concern’ defined in the QT Guidance, it should be noted that that guidance provides that, ‘while increases in QT/QTc to [more than] 500 ms or of [more than] 60 ms over baseline are commonly used as thresholds for potential discontinuation, the exact criteria chosen for a given trial will depend on the risk-tolerance level considered appropriate for the indication and patient group in question’.

91      It follows that the QT Guidance does not define a threshold which would, in absolute terms, represent a risk enabling a conclusion that a medicinal product is unsafe, and even less does it define a threshold which, once exceeded, would justify refusing an MA. Therefore, while, as the CHMP moreover noted, ‘an increase of more than 60 msec from baseline when treating with a medicine is a concern and would usually lead to discontinuation’, that does not preclude the possibility that a medicinal product causing a prolongation of the QT interval below that figure may, in certain circumstances, present a safety risk.

92      In addition, and in any event, the CHMP opinion referred, inter alia, to the following conclusion:

‘In reviewing the results of the thorough QTc study, out of 94 patients exposed to [i]loperidone at different doses without metabolic inhibition in the Secondary QTc population 43 and 2 patients respectively developed a prolongation of the QTcF of more than 30 and 60 msec.’

93      That conclusion is based on various clinical data, mentioned in the CHMP assessment report, as follows:

‘Looking at the clinical safety data, Safety group 1, 4.5% of the patients treated with [i]loperidone regardless of dose (4-24 mg/day) has an increase o[f] more than 60 msec at some time point during the clinical studies. In the ziprasidone group (160 mg/day), it was 1.6%.

3 patients had at some time-point a QTcF interval of more than 500 msec ([i]loperidone 10-16 mg/day group). This was not seen in the ziprasidone group although fewer patients were exposed to ziprasidone.’

94      It is apparent from these considerations that both the CHMP opinion and the CHMP report clearly state, and in conformity with the QT Guidance, why the results of the clinical studies presented, which included different calculation and presentation methods, indicated that iloperidone use remained associated with a real risk of cardiac arrhythmia (torsades de pointes) which can cause sudden death.

–       The claim that the CHMP assessment report is not consistent with current EMA practice at the material time as no account has been taken of the positive post-marketing experience of iloperidone

95      The applicant submits that the CHMP refused to take into account post-marketing data relating to iloperidone and in particular the data collected in the United States following that substance being placed on the market. Such a refusal is allegedly not only contrary to the current EMA practice at the material time, but also infringes the principle of equal treatment, since the CHMP has in the past taken such data into account for the purposes of authorising other medicinal products.

96      That argument cannot succeed.

97      In the first place, the applicant’s claim that the EMA failed in the present case to take the post-marketing data into account cannot be accepted. It is apparent from the file that the CHMP did in fact take into account the experience acquired after iloperidone was placed on the market, in particular on the American market, but that it found that experience to be inconclusive.

98      Accordingly, paragraph 2.6 of the CHMP assessment report states, under the sub-heading ‘Post-marketing experience’:

‘From the Vanda global US post-marketing surveillance database through 24 August 2016 there were 33 deaths in total. 3 patients died during sleep, 6 were sudden deaths, 6 cases were of cardiac origin. The rest were suicides (6), unknown (7), 2 … died of other reasons and 3 died due to pulmonary embolism.’

99      However, that report states that the post-marketing data provided by the applicant is not considered to be reliable:

‘Concerning the post-marketing experience, it is difficult to conclude on the calculation made by the [a]pplicant on the excess mortality. This is due to difficulties in evaluating the matching and the assumed under-reporting rate. From the qualitative point of view, in the opinion of the assessor, and considering the age of the patient, time since initiation of treatment, and circumstances of the deaths, 15 cases can possibly be considered very likely associated with [i]loperidone. At least one fatal case could possibly have been preceded by ventricular arrhythmia and torsade[s] de pointes.’

100    As the Commission explained in its written pleadings and made clear at the oral hearing, the CHMP, prior to the qualitative examination of the data concerning the period post-marketing of iloperidone, evaluated the quantitative analysis of that same data which had been submitted by the applicant.

101    However, it was decided that that quantitative analysis was unreliable, due to two methodological flaws.

102    The first flaw identified related to the fact that the comparison put forward by the applicant to demonstrate the absence of excess mortality was based on non-comparable populations. The CHMP assessment report states in support of that as follows:

‘The available data and the lack of interchangeability (and to some extent comparability) between the populations compared does not allow to exclude, confirm or quantify an excess cardiac mortality in the clinical development program[me].’

103    The second flaw vitiating the quantitative analysis submitted by the applicant, was, in the view of the CHMP, due to the fact that the applicant’s choice of under-reporting level was arbitrary and could not be taken into consideration. The CHMP assessment report states as follows in that regard:

‘Similarly, it seems impossible to estimate the extent of under-reporting of fatal outcomes in association with iloperidone. Post-marketing spontaneous report data are not considered to provide substantial reassurance regarding cardiac safety. There are various reasons why very substantial underreporting of deaths in relation to iloperidone might be expected. Sudden cardiac death in general cannot definitively be shown to be as a result of iatrogenic QT prolongation and ventricular arrhythmia as there is no post mortem marker. Even where there is suspicion that this is a likely cause of death, a possible causal association with treatment might not be reported as iloperidone is well known to prolong the QT interval.’

104    It should be noted that, given the discretion enjoyed by the bodies responsible for examining the scientific data provided in support of an MA application, the CHMP’s findings and assessments can only be criticised if it is established that, having regard to the file, they bear no relation to the medical and scientific findings. The applicant has failed to demonstrate why in the present case it should be concluded such was the case. The applicant has therefore been unable to explain why the quantitative data it had provided in support of its MA application was capable of calling into question the assessment that iloperidone had arrhythmogenic potential and, therefore, the risk identified by the CHMP.

105    In that context, it should be recalled that the overriding need for safety in the field of public health (see, to that effect, judgment of 19 April 2012, Artegodan v Commission, C‑221/10 P, EU:C:2012:216, paragraph 99 and the case-law cited) logically presupposes that authorities can, when there are doubts concerning the reliability of information provided in support of an MA application, incline towards refusing an MA.

106    In the present case, the applicant has failed to establish that the CHMP’s conclusions on the post-marketing data recorded in third countries, in particular in the United States, were inconsistent or vitiated by an error of analysis. On that point, the CHMP assessment report contains a statement of reasons allowing an assessment of the considerations on which the CHMP opinion is based and establishes a comprehensible link between the relevant scientific findings and the conclusions reached.

107    In the second place, the applicant has failed to establish that the Commission had departed from the previous practice on the taking into account of post-marketing data provided in support of MA applications for certain medicinal products, in particular lurasidone and cisapride. In finding the post-marketing data provided by the applicant to be unconvincing, the CHMP did not refuse as a matter of principle to take into such data into consideration when evaluating iloperidone’s safety, but rather was concerned with determining scientific reliability.

–       The claim that the CHMP was unduly influenced by the comments of the ad hoc experts group of 30 October 2017

108    The applicant submits, in essence, that since the CHMP did not give reasons for its decision to refuse the MA, it can only ‘presume’ that the contested decision is explained by two factors, being, first, the unsupported allegation of the October 2017 scientific advisory group, which was itself based on the personal opinion of one of the group’s members that iloperidone has caused a ‘very high’ number of sudden and unexpected deaths and, second, the participation in that group of an expert who was a consultant for a company producing a medicinal product competing with the product containing iloperidone.

109    That argument cannot succeed.

110    In the first place, the claim that the CHMP was unduly influenced by the October 2017 scientific advisory group is not substantiated.

111    In that respect, it should be recalled that that group was convened following the applicant’s request to that effect under Article 62(1) of Regulation No 726/2004 in the context of the re-examination procedure.

112    As the Commission noted, it had not been established that that group of scientific experts, which had a purely advisory role and whose reports therefore did not bind the CHMP, had an undue influence in the drawing up of the CHMP assessment report.

113    In the second place, it is necessary to rule on the claim that one of the members of the October 2017 scientific advisory group had acted as a consultant for a competing product, even while iloperidone was being evaluated. As the Commission has pointed out, according to the EMA’s policy on the management of competing interests, experts who declare that they currently provide advisory services for a specific product are permitted to participate in meetings of a scientific advisory group or an ad hoc expert group, with the restriction that they cannot participate in the group when it is consulted on the specific declared product. In the present case, the two ad hoc expert groups were concerned with iloperidone and not the product covered by the expert’s declaration of interest (namely cariprazine). Therefore, after assessing the documents concerning declaration of interests in accordance with established procedures, it was concluded that there was no conflict of interest on the part of the expert in question and he was allowed to participate fully in the two ad hoc expert groups.

114    However, the applicant stated in its reply that it did not intend to positively allege a ‘conflict of interest’, but only intended to seek an explanation to understand the reasoning followed by the CHMP in the absence, in its view, of a convincing statement of reasons in the contested decision.

115    When invited at the oral hearing to clarify the exact scope of its allegations, the applicant confirmed that it was not seeking to claim an infringement of the principle of impartiality or any conflict of interest, which was noted in the minutes of the hearing, but that its arguments were to be understood from the viewpoint of a failure to state reasons.

116    Therefore, both the claim of undue influence of the October 2017 scientific advisory group on the CHMP and the claim of possible undue influence of an expert participating in that panel must be rejected as being unfounded.

117    Having regard to all those considerations, the first plea in law must be rejected as being unfounded.

 The second plea in law: the evaluation of the RMMs proposed for iloperidone is vitiated by a failure to state reasons, a manifest error of assessment and a breach of the principle of proportionality, as set out in Article 5(1) and (4) TEU, and of the principle of equal treatment

118    By its second plea, the applicant criticises, in essence, the statement that ‘it is not considered that the [RMMs] proposed would appropriately address the risk identified’ and which ultimately led the EMA to conclude that in the present case there was a negative risk-benefit balance. The applicant therefore disputes the evaluation of the RMMs proposed for iloperidone, which included the possibility of limiting the MA for that medicinal product to second line treatment. It submits that that evaluation is not only vitiated by a failure to state reasons and manifest errors of assessment (first complaint), but that it also infringes the principles of proportionality and of equal treatment (second complaint).

119    The Commission contests the applicant’s arguments and contends that the present plea should be rejected.

120    Before considering the various complaints put forward by the applicant in relation to the present plea, it should be explained that RMMs are generally intended to prevent or reduce the occurrence of adverse reactions, which are unavoidable, and are associated with exposure to a medicinal product, or to reduce their severity or impact on the patient should adverse reaction occur. These RMMs seek to optimise the safe and effective use of a pharmaceutical product throughout its life cycle. It is widely agreed, by those active in the pharmacovigilance field that not only the planning and implementation of RMMs but also the evaluation of their efficiency are key elements of risk management. Whether or not the RMMs proposed are sufficient can therefore prove to be crucial to any decision concerning an MA for a medicinal product.

121    In the present case, the RMMs proposed consisted not only of ‘routine’ measures, being classic information and warnings in the patient information leaflet and the summary of product characteristics, but also of more involved accompanying medical measures such as the use of genotyping and ECG monitoring. In addition to all the measures proposed, the applicant also asserted that consideration should be given to the use of iloperidone as a second line treatment.

–       The complaint that the evaluation of the RMMs is vitiated by a breach of the obligation to state reasons and manifest errors of assessment

122    The applicant submits that the evaluation of the RMMs it submitted in support of its MA application is vitiated by a failure to state reasons and is, in any event, manifestly incorrect. According to the applicant, the CHMP did not provide, in its assessment report, plausible reasons why the RMMs proposed, which included the possibility of limiting the MA to second line treatment (that is to say for cases where other products are not effective or are not tolerated by the patient), were not considered to be sufficient to manage the safety risks presented by iloperidone.

123    The applicant points out that, in order to manage the risks of QT interval prolongation, it submitted four types of RMM frequently used on the EU market, in particular for products intended for the treatment of schizophrenia. First, it claims that it has, as is common practice, included in the summary of product characteristics and in the patient information leaflet strict warnings and specific precautions for use stating that the product could prolong the QT interval and have adverse effects, also mentioning that sudden deaths have been reported. Second, it states that, in accordance with current practice and in line with the CHMP recommendations on that point, it has included precautions, or even contra-indications, to prevent the administration of iloperidone to higher risk patients, that is to say, patients inherently at risk of experiencing the adverse effects theoretically associated with QT interval prolongation. Third, it suggested, also in line with the CHMP recommendations, that the summary of product characteristics include a statement that genotyping should be performed for all patients prior to the initiation of treatment in order to identify patients with a certain genotype for whom iloperidone should be contra-indicated. Fourth, it agreed and proposed to limit initiation of iloperidone treatment to settings in which a cardiologist is available and to require that ECG monitoring be performed before and during iloperidone treatment.

124    The applicant maintains that the combination of those four RMMs and the proposal to limit the use of iloperidone to ‘second line’ treatment, which, as suggested by the scientific advisory group convened in May 2017, it submitted in the context of its re-examination request by means of an elaborate treatment algorithm, ought to have led the CHMP to conclude that the risks presented were managed adequately, which it failed to do. The CHMP’s conclusions allegedly establish no comprehensible link between the scientific findings, and are therefore vitiated by a failure to state reasons and manifest errors of assessment, in particular as regards the examination of genotyping measures and the use of ECG, and also as regards the proposal for second line use.

125    In the present case, it should be recalled that the CHMP took the view that the RMMs proposed were not sufficient to manage adequately the risk detected, namely the substantial arrhythmogenic potential of iloperidone.

126    The CHMP opinion, set out in Annex I to the contested decision, provides the following summary:

‘In consideration of the complex causal chain linking the exposure to [i]loperidone to events such as [t]orsades de [p]ointes, including unknown and stochastic elements and elements subject to an unpredictable variability, it is considered that the [RMMs] proposed would not adequately address in clinical practice the risk identified. For example, the proposal to perform ECGs at the estimated Tmax could miss the actual Tmax due to intrinsic or extrinsic factors, leading to an underestimation of QTcF prolongation.

In addition, whether the correct implementation of the complete set of measures would be feasible in all clinical settings is questionable due to practical reasons (for example, the availability of suitably trained cardiologists), as also mentioned by the experts in the ad-hoc meeting.’

127    The CHMP assessment report explains the reasons why the RMMs proposed were deemed to be insufficient.

128    It is apparent, in particular, that, at the stage of the first examination of the MA application at issue in the present case, the CHMP stated as follows:

‘Considering all available non-clinical and clinical data (including the thorough QTc study, the overall clinical program[me] and the cases of cardiac-related/sudden unexplained death in clinical trials and post-marketing), iloperidone has a substantial and exposure-dependent arrhythmogenic potential. The fact that the metabolism of [i]loperidone relies heavily on CYP3A4 and CYP2D6 increases the risks of drug-drug interactions and makes the medicine metabolism highly sensitive to genetic polymorphisms. [RMMs] such as CYP2D6 genotyping or extensive ECG monitoring are not considered sufficient to minimise this risk.’

129    It is also apparent from the CHMP assessment report that the CHMP, at the stage of re-examination of the MA application, maintained its conclusion that the RMMs proposed were not sufficient for the following reasons:

‘It is acknowledged that for a niche product intended for use in a small number of patients the proposed [RMMs] would seem to be feasible in some clinical settings in the EU but probably not all. However, the ability of the proposed measures to appropriately address the risks is questioned in presence of known and unknown sources of variability. Some examples of this can be provided, but a comprehensive list cannot by definition be formulated:

–        While the recommendation would be to perform the ECG at Tmax, this could be missed due to intrinsic or extrinsic factors.

–        The increase in exposure with non-contradicted inhibitors of [i]loperidone’s metabolism might be subject to a significant variability in [the] presence of a small safety margin.

Similarly, the proposal to lower the threshold for contraindicating [iloperidone] based on QT at baseline cannot be accepted due to the within-subject variability of this measurement in the population concerned.’

130    In the section entitled ‘Conclusion and Updated Risk Benefit Assessment’, in paragraph 5 of the CHMP assessment report, the CHMP concluded that:

‘Considering all available non-clinical and clinical data (including the thorough QTc study, the overall clinical program[me] and the cases of cardiac-related/sudden unexplained death in clinical trials and post-marketing), [i]loperidone has a substantial and exposure-dependent arrhythmogenic potential. It is not considered that the [RMMs] proposed would appropriately address the risk identified in this specific case. Hence, the safety of [i]loperidone has not been sufficiently demonstrated.’

131    It must be noted that, first, the CHMP opinion and assessment report on which the contested decision is based are not vitiated by an inadequate statement of reasons regarding the RMMs and that, second, the CHMP, in its assessment report, expressed, as required by the case-law, a set of plausible reasons in support of its conclusion that the RMMs proposed were not capable of countering the potentially adverse effects of iloperidone.

132    In the first place, as regards the first two categories of RMMs proposed, so called ‘routine’ measures, being the information and warnings contained in the summary of product characteristics and patient information leaflet, as the Commission stated in its defence, it is apparent from that information taken as a whole that, in view of the severity of the risk identified, the appropriateness of ‘simple’ RMMs, such as warnings in the summary of product characteristics and the patient information leaflet was excluded.

133    In the second place, with regard to the RMM consisting of the use of genotyping, it appears that, in essence, the CHMP considered it to be insufficient for two main reasons.

134    First, the CHMP concluded that genotyping could control only partially the risks created by exposure to iloperidone. As the Commission explained in its written pleadings, since it is accepted that the arrhythmogenic potential of that substance increases with exposure, it has become clear that, the higher the level of that substance in the blood, the greater the patient’s exposure to a risk of a potentially fatal cardiac arrhythmia. Conversely, the faster iloperidone is metabolised, the lower the risk. It is apparent from the data made available to the CHMP that iloperidone is mainly metabolised by two enzymes, namely cytochrome P450 3A 4 (CYP3A4) and cytochrome P450 2D 6 (CYP2D6), which can be inhibited by the consumption of medicinal products. In such a case, genotyping does not allow a sufficiently reliable identification to be made of patients for whom iloperidone consumption generates significant risks.

135    That analysis is clearly set out by the CHMP in its assessment report in the following terms:

‘The increases of Cmax up to 2.3 fold observed with metabolic inhibition are not insignificant and importantly this is a mean value; it does not describe the extent to which some individuals might show a much greater increase in iloperidone Cmax in the presence of strong CYP3A4 and CYP2D6 inhibitors. This will depend on the activity of the minor metabolic pathways for iloperidone, which could very well be highly variable. In the context of the risk assessment for drug induced [torsades de pointes] the population mean effect is less relevant than the worst case scenario. No data are available to establish the likely effect on drug levels in patients with low activity of the minor metabolic pathways for [i]loperidone exposed to mild CYP3A4 and CYP2D6 inhibitors.’

136    Second, even if exposure to iloperidone could be satisfactorily controlled by the use of genotyping, it is apparent from the CHMP assessment report that the view was taken that, although there was at first sight a clear link between QT interval prolongation (and therefore arrhythmogenic potential) and that exposure, such exposure was not the only factor linked to the use of iloperidone capable of resulting in effects such as torsades de pointes which can be fatal for the patient. In other words, the view was taken that controlling exposure to iloperidone, in particular by means of genotyping, would not mitigate adequately the risks associated with consumption of that medicinal product.

137    The CHMP therefore explained in the assessment report that:

‘The experts reflected that the causal chain linking the exposure to [i]loperidone to events such as [t]orsade[s] de [p]ointes is complex and includes some unknown or stochastic elements which by definition would be of very difficult control in any risk minimisation plan to be implemented in the clinical setting.’

138    In the third place, as regards the evaluation of the RMM consisting of the use of ECG, the conclusions reached with regard to the use of genotyping apply mutatis mutandis.

139    Indeed, the use of that RMM at ‘Tmax’, which consists of an estimate of the time elapsed until the maximum plasma concentration is reached, was regarded as not sufficiently effective given a number of intrinsic (related to the individual patient) and extrinsic (related to the availability in the clinical settings concerned of a cardiologist to monitor patients with schizophrenia) factors.

140    As regards the use of ECG at baseline, that is to say the stage when prescribing iloperidone is envisaged, the CHMP noted that it does not necessarily provide reliable information on the patient’s QT interval. The CHMP assessment report therefore states that ‘the proposal to lower the threshold for contraindicating [iloperidone] based on QT at baseline cannot be accepted due to the within-subject variability of this measurement in the population concerned’.

141    In other words, according to the CHMP, since the values obtained by means of ECG vary widely from one individual to another and are therefore unpredictable, the use of ECG monitoring does not appear to be a fully satisfactory measure to control the effects of consuming iloperidone. It therefore does not appear possible to define a value that could serve as a warning to the professionals concerned. Ultimately, according to the CHMP, there was therefore reason to doubt the reliability of the use of that instrument as a means of controlling the risks presented by iloperidone.

142    In the fourth place, regarding the proposal that iloperidone be used as a second line treatment, namely in situations where the patient’s treatment with another medicinal product has not been satisfactory, it is necessary to examine, further to the above considerations, whether there is a comprehensible link between the scientific findings brought to the EMA’s attention and the negative recommendation at issue in the present case.

143    In that regard, it is apparent from the file submitted to the Court that it was following the CHMP’s negative opinion delivered on 20 July 2017 and in support of its request for re-examination that the applicant proposed that iloperidone should be ‘indicated for the treatment of schizophrenia in adults as a second line option’.

144    With that in mind, the applicant had distinguished two sub-categories of the overall population for which it intended to demonstrate that iloperidone’s risk-benefit balance was positive, namely, first, schizophrenia patients who are stabilised and, second, patients who are in an acute schizophrenia episode and are unable to be fully stabilised by another medicinal product on another medicinal product because of tolerability issues.

145    However, the risk-benefit balance was held to be negative in view of the major safety issues for these two sub-populations (in the same way as for the overall population). In that respect, the CHMP assessment report establishes a comprehensible link between the medical or scientific findings and its conclusions regarding the use of iloperidone as a second line treatment.

146    If, as the parties both acknowledge, patients with schizophrenia, particularly those with akathisia, continue, to date, to have unmet medical needs, that concerns, according to the data provided during the re-examination procedure that is the subject of the present case, patients who would need access to a medicinal product with no or almost no propensity to cause such a disorder. However, that is not the case for iloperidone, which, as indicated by the CHMP in its assessment report, has ‘a low but not uniquely low propensity to cause [extrapyramidal symptoms] in general’.

147    In the fifth and final place, as regards the claim that the CHMP did not provide reasons why the RMMs proposed, taken as a whole and not separately, would not be sufficient to manage the safety risks presented by iloperidone, that plea is unfounded.

148    In that regard, it should be recalled that the CHMP opinion states that ‘in consideration of the complex causal chain linking the exposure to [i]loperidone to events such as [t]orsade[s] de [p]ointes, including unknown and stochastic elements and elements subject to an unpredictable variability, it is considered that the [RMMs] proposed would not adequately address in clinical practice the risk identified’.

149    As is also apparent from the position of the ad hoc group of 30 October 2017, it was concluded that, regardless of how the RMMs were combined, a number of patients would be exposed to a substantial safety risk associated in particular with the development of torsades de pointes. Among other considerations, that group noted the following:

‘The experts reflected that the causal chain linking the exposure to [i]loperidone to events such as [t]orsade[s] de [p]ointes is complex and includes some unknown or stochastic elements which by definition would be of very difficult control in any risk minimisation plan to be implemented in the clinical setting. In consideration of this and of the available data, the experts concluded by majority that it is not possible to design a set of [RMMs] that would appropriately address the risks identified and that the measures proposed would end up providing a false reassurance.’

150    In the light of all those considerations, it must be held that the conclusion that the RMMs proposed by the applicant with a view to the grant of an MA for iloperidone are insufficient is not vitiated by a failure to state reasons or manifest errors of assessment.

–       The complaint alleging a breach of the principle of proportionality

151    The applicant submits that the evaluation of the proposed RMMs is contrary to the principle of proportionality. It maintains that the summary rejection of those RMMs — and accordingly the refusal to grant an MA — goes beyond what is necessary to achieve the objective of balancing risks and benefits. That refusal is not the measure to minimise sufficiently the risks associated with iloperidone that is the least onerous. In the applicant’s submission, the CHMP could have imposed feasible RMMs to ensure that iloperidone is prescribed and administered in such a way as to manage risk and ensure an acceptable level of safety.

152    According to settled case-law, the principle of proportionality, now set out in Article 5 TEU, requires that acts of the EU institutions be appropriate for attaining the legitimate objectives pursued by the legislation at issue and do not exceed the limits of what is necessary in order to achieve those objectives; when there is a choice between several appropriate measures, recourse must be had to the least onerous, and the disadvantages caused must not be disproportionate to the aims pursued (see judgment of 4 May 2016, Pillbox 38, C‑477/14, EU:C:2016:324, paragraph 48 and the case-law cited).

153    In an area such as that at issue in the present case, in which the public authority concerned is called upon to undertake complex assessments, the legality of a measure adopted in that area can be affected only if the measure is manifestly inappropriate having regard to the objective which the competent institutions are seeking to pursue (see, to that effect, judgments of 4 May 2016, Pillbox 38, C‑477/14, EU:C:2016:324, paragraph 49 and the case-law cited, and of 16 March 2016, Dextro Energy v Commission, T‑100/15, EU:T:2016:150, paragraph 80 and the case-law cited).

154    In order to assess whether the principle of proportionality has been observed in the field of public health, account must be taken of the fact that human health and human life rank foremost among the assets and interests protected by the FEU Treaty (see, to that effect, judgment of 19 April 2012, Artegodan v Commission, C‑221/10 P, EU:C:2012:216, paragraph 99 and the case-law cited; see also, by analogy, as regards Member States’ observance of that principle in the field of public health, judgment of 8 June 2017, Medisanus, C‑296/15, EU:C:2017:431, paragraph 82 and the case-law cited).

155    The Court has held, having specific regard to the exclusive nature of the criteria of safety, efficacy and quality laid down in the context of the EU system for the harmonisation of the grant and management of MAs for medicinal products, that the assessment of the proportionality of a measure suspending or withdrawing an MA was to be guided by those criteria alone. It follows that the relevant interests in the context of the review of proportionality are the same as those related to the protection of public health, taken into consideration in the application of the relevant legislation (see, to that effect, judgment of 3 March 2010, Artegodan v Commission, T‑429/05, EU:T:2010:60, paragraph 128).

156    In the present case, it appears that the applicant’s arguments are essentially the same as those considered in paragraphs 125 to 150 above in the context of the considerations relating to the examination of the first complaint within the second plea, according to which the CHMP committed manifest errors of assessment in examining the RMMs proposed to address the safety problems associated with iloperidone.

157    Since the view was taken that those RMMs, either individually or in combination, were not sufficient ground for a conclusion that the risk-benefit balance was positive, a CHMP opinion recommending that the MA be refused for that medicinal product was the inevitable consequence. In other words, since, for a medicinal product with a negative risk-benefit balance, there is no less onerous alternative than providing a negative response to an MA application for that product, it cannot be claimed that the refusal of an MA, contained in the contested decision, is manifestly disproportionate.

–       The complaint alleging a breach of the principle of equal treatment

158    The applicant takes the view that the CHMP infringed the principle of equal treatment to the extent that it treated the RMMs proposed for iloperidone differently from the RMMs proposed for other medicinal products for the treatment of schizophrenia. As regards, in particular, sertindole, ECG monitoring and second line treatment were, according to the applicant, accepted as RMMs. Similarly, rather than refusing to grant an MA for cariprazine, the CHMP accepted that the safety concerns identified for that medicinal product should be reflected in the product information and in its specifications.

159    In that regard, it is well established that the principle of equal treatment requires that comparable situations not be treated differently and different situations not be treated in the same way unless such treatment is objectively justified (see judgments of 29 April 2004, Novartis Pharmaceuticals, C‑106/01, EU:C:2004:245, paragraph 69 and the case-law cited; of 4 May 2016, Pillbox 38, C‑477/14, EU:C:2016:324, paragraph 35 and the case-law cited; and of 9 September 2010, CSL Behring v Commission and EMA, T‑264/07, EU:T:2010:371, paragraph 113 and the case-law cited).

160    In the present case, since the scientific data relating to the safety of medicinal products is different, there can be no question of comparable situations.

161    As regards, in particular, the data relating to QT interval prolongation and, accordingly, to the arrhythmogenic potential of iloperidone, which were central to the evaluation of the risk-benefit balance in the present case, it is apparent from the data provided by the Commission, data which has not been seriously disputed by the applicant, that the scientific conclusions differ from those adopted for cariprazine and sertindole. With regard, in particular, to sertindole, the Commission explained that the data which had been presented had made it possible to exclude the existence of torsades de pointes indicators, which was not the case when evaluating iloperidone. As regards the comparability of the scientific evaluations for cariprazine and iloperidone, the Commission stated why there were objective differences in the results of the analyses regarding QT interval prolongation having regard to the safety thresholds defined in the QT Guidance.

162    Indeed, as is apparent from the CHMP assessment report, the experts consulted seem to have agreed that iloperidone has a significant arrhythmogenic potential due to QT interval prolongation, whereas that was not the case for cariprazine and sertindole.

163    Furthermore, in response to the written question put by the Court by way of measures of organisation of procedure, the Commission stated that for sertindole, the risk-benefit balance has not, following the suspension of its MA, been reassessed since 2002, and sertindole has not been found in any medicinal products authorised under the centralised procedure since that date. In response to a question raised at the hearing, it was confirmed that the parameters previously taken into consideration in the evaluation of the arrhythmogenic potential of certain substances were less stringent than those currently applicable and that the regulatory context applicable in that area had moved on since 1997.

164    It follows from all those considerations that the applicant’s claim regarding the alleged breach of the principle of equal treatment is unfounded and must be rejected.

165    The second plea should therefore also be rejected.

 The third plea in law: the evaluation of the consequences of the delayed onset of iloperidone is vitiated by a failure to state reasons and a breach of the principle of proportionality set out in Article 5(1) and (4) TEU

166    The applicant submits that the evaluation of the consequences of the delayed onset of iloperidone, which it claims to be internally inconsistent, is vitiated by a failure to state reasons and a breach of the principle of proportionality. It submits that, in considering that delayed onset of action was a ‘significant concern’ for the treatment of acute exacerbation of schizophrenia and, therefore, an additional ground for refusing an MA for that medicinal product, the CHMP failed to take into account three essential circumstances. First, it is apparent from the guidelines on clinical investigation of medicinal products, including depot preparations in the treatment of schizophrenia, published by the EMA on 20 September 2012 (EMA/CHMP/40072/2010 Rev. 1), which state that the short-term efficacy of a medicinal product may be demonstrated through a 6-week clinical trial, that immediate efficacy need not necessarily be demonstrated in order to attest to the efficacy of products intended for the treatment of that disease. Second, as regards the treatment of a chronic disease, the onset time of a medicinal product intended to treat that disease should not be decisive, especially when it is prescribed in second line, that is to say when the use of other products has been shown to be ineffective or when it has not been tolerated. Third, the CHMP itself acknowledged in its assessment report that the onset of efficacy ‘would not be considered in itself to be an impediment’ to the approval of iloperidone. The CHMP thus considered that the delayed onset would limit only the clinical situations in which the administration of that medicinal product might be considered.

167    The Court points out that, as is apparent from the CHMP opinion (see paragraph 16 above), the CHMP concluded that the delayed onset of iloperidone was ‘a significant concern in the treatment of acute exacerbation of schizophrenia’.

168    In the first place, as regards the claim that these conclusions are contradictory and therefore vitiated by a failure to state reasons, that is based on a flawed reading of the CHMP opinion and assessment report.

169    It is true that the CHMP stated in its assessment report that ‘the acknowledged delayed onset of efficacy would not be considered in itself to be an impediment for approval of iloperidone’.

170    However, that does not mean that the time period for onset of action cannot have an impact on the risk-benefit balance of a given medicinal product.

171    In the present case, the CHMP took the view in its assessment report that the delayed onset of the action of iloperidone could be clinically important in the treatment of acute exacerbations of schizophrenia, that is to say, in the treatment of patients in the second sub-population identified. That finding, in the circumstances of the present case, confirmed that the risk-benefit balance was negative.

172    The CHMP, in its assessment report, noted the following:

‘… [i]loperidone has a modest efficacy. In addition, it has shown a delayed onset of action, which is a significant concern in the treatment of acute exacerbation of schizophrenia. Therefore, … a patient population cannot be identified where the benefit of treatment is considered to outweigh the major safety concerns.’

173    Second, the claim that the CHMP infringed the principle of proportionality, in so far as it did not authorise — or even consider — a limited indication for iloperidone, is no more convincing.

174    In that regard, it must be recalled that the principle of proportionality requires that acts of the EU institutions do not exceed the limits of what is appropriate and necessary in order to achieve the legitimate objectives pursued by the legislation at issue; when there is a choice between several appropriate measures, recourse must be had to the least onerous, and the disadvantages caused must not be disproportionate to the aims pursued (see the case-law referred to in paragraph 152 above).

175    As regards, more specifically, the question of whether or not to approve an MA application for a medicinal product, the risk-benefit balance is at the core of the assessment by the authorities responsible for examining that request. If the conclusion is that the risk-benefit balance is negative, the MA application must be rejected. As the Commission has stated, the prescription of such a medicinal product must be prevented and cannot therefore be left to the discretion of health professionals.

176    Consequently, to the extent that the CHMP opinion on the risk-benefit balance of iloperidone was negative, it cannot validly be accused of having infringed the principle of proportionality by refusing to grant a ‘limited’ MA for that medicinal product, since such ‘limited’ MA does not constitute a ‘less onerous’ and ‘appropriate’ measure that could be contemplated in order to achieve the objectives pursued.

177    It follows from all the foregoing considerations that the conclusions reached regarding the consequences of the delayed onset of action of iloperidone are neither vitiated by a failure to state reasons nor are they contrary to the principle of proportionality.

178    Consequently, the third plea cannot succeed and must be rejected as being unfounded.

 The fourth plea in law: the obligation to identify a population in which iloperidone would outperform other products constitutes a breach of the principles of conferral and proportionality (set out in Article 5(1) to (3) TEU), Article 12 and Article 81(2) of Regulation No 726/2004 and the principle of equal treatment

179    The applicant submits, in essence, that, in requiring evidence of the ‘superiority’ of iloperidone over other second line medicinal products for schizophrenia, the contested decision infringes the principle of conferral, the principle of subsidiarity, the principle of equal treatment and Articles 12 and 81(2) of Regulation No 726/2004, those latter provisions establishing the precise grounds on which an MA application may be refused.

180    First, the applicant maintains that, in requiring such evidence — in particular by asking the applicant to ‘identify a population’ for which iloperidone has unique advantages by comparison with other products designed to treat schizophrenia, and even by comparison with all products available at the time of the relevant facts, taken together — the CHMP formulated and applied an additional condition for approval. The applicant submits that, although it indicated during the re-examination procedure that that condition was contrary to EU law, the CHMP maintained and applied it in its assessment report. Consequently, the CHMP failed to observe the principles of conferral and subsidiarity and the provisions of Articles 12(1) and 81(2) of Regulation No 726/2004, according to which an MA may be refused only if the applicant has not properly or sufficiently demonstrated the quality, safety or efficacy of the medicinal product.

181    Second, the applicant maintains that, by imposing, in an unprecedented fashion, in the field of second-generation medicinal products for the treatment of schizophrenia, additional requirements for the granting of an MA for iloperidone, the contested decision infringes the principle of equal treatment. In the applicant’s submission, the CHMP has thus far never required, in the assessment of such products, that they produce better results, in particular in terms of tolerance and therapeutic efficacy.

182    The Commission disputes the applicant’s arguments and contends that this plea should be rejected.

183    In the present case, it is clear that the arguments put forward by the applicant in support of the fourth plea are based on the false premiss that the CHMP imposed on it a condition not provided for by the law applicable to the granting of an MA, namely the demonstration of iloperidone’s superiority over other medicinal products designed to treat the symptoms of schizophrenia.

184    A careful reading of the reasons underlying the contested decision, in particular the CHMP assessment report, clearly shows that it is in response to the applicant’s argument that iloperidone met an ‘unmet medical need’ — in that it is intended for persons with a condition for which there is no fully satisfactory method of treatment — that the CHMP was led to carry out a comparative review of the therapeutic benefits of that medicinal product in relation to other second generation medicinal products designed to treat the symptoms of schizophrenia currently on the market. In particular, the CHMP was asked to clarify whether and to what extent iloperidone was less likely than other medicinal products currently available to cause extrapyramidal effects, including akathisia.

185    Consequently, both the complaints alleging a breach of the principles of conferral and of subsidiarity and those alleging a breach of the provisions of Regulation No 726/2004 are unfounded. As the Commission has correctly pointed out, the reason why the MA for iloperidone was not recommended by the CHMP was to be found in its negative risk-benefit balance and not, as the applicant implies, in the result of a comparison between that substance and authorised medicinal products.

186    As regards the claim that the CHMP’s evaluation of iloperidone was (i) in breach of the principle of equal treatment and (ii) much more demanding than those relating to other medicinal products for the treatment of schizophrenia, that is no more convincing.

187    In that respect, it should be pointed out, further to what was stated in the context of the examination of the third plea (see paragraph 160 above), that since the scientific data regarding the safety of medicinal products is different, there can be no question of comparable situations.

188    That is all the more so when the medicinal products concerned belong to different therapeutic classes, given that the relevant scientific considerations for assessing their safety and efficacy vary by definition from one group of diseases to another. In particular, as the Commission has stated, without being contradicted by the applicant, iloperidone and cisapride belong to different therapeutic classes. Iloperidone is an anti-psychotic medicinal product which falls within the therapeutic area of nervous system disorders. By contrast, cisapride is a medicinal product designed for digestive tract and metabolic disorders.

189    In any event, and even if the authorities responsible for evaluating the risk-benefit balance of a medicinal product in the context of an MA application submitted under the centralised procedure were to have been less strict in identifying the risks presented by other medicinal products, that cannot be used as an argument to justify granting an MA. Indeed, the evaluation of the efficacy, and safety of a medicinal product must be based on an objective examination of the scientific analyses submitted in support of an MA application and not on a comparative review of the evaluations carried out by those bodies. Furthermore, it is important that it be established that the medicinal product in question has efficacy and safety advantages which justify a positive risk-benefit balance.

190    In the light of the foregoing, the fourth plea must be rejected as being unfounded.

 The fifth plea in law: the overall risk-benefit assessment of iloperidone lacks an adequate statement of reasons and, in any event, is manifestly erroneous

191    The applicant maintains, in essence, that the overall risk-benefit evaluation of iloperidone carried out by the CHMP lacks an adequate statement of reasons and is, in any event, manifestly erroneous. Even if the errors which it has identified in the context of the first four pleas in the present action could not be considered manifest on their own, when taken together they resulted in an overestimation of the risks and an underestimation of the undeniable benefits of iloperidone (namely low akathisia and a reduction in the rate of relapse) and, accordingly, in the conclusion that there was a negative risk-benefit balance. The applicant claims that, apart from those benefits, the CHMP appears to have lost sight of the fact that schizophrenia is a serious disease and of the extent of the unmet needs of the patients concerned.

192    The Commission contends that the plea must be rejected.

193    It should be noted that, by the present plea, the applicant does not add, in essence, anything new compared to what has been put forward in the first four pleas. In fact the applicant merely argues that the CHMP appears to have lost sight of the seriousness of schizophrenia, the extent of unmet needs and the benefits of iloperidone in treating the symptoms (namely low akathisia and a significant reduction in the rate of relapse). It concludes that these elements should reasonably have led the CHMP to make a positive recommendation regarding the MA for that medicinal product.

194    In that respect, it is important to recall that the decision whether or not to authorise the placing on the market of a medicinal product must be based on a thorough examination of the quality, safety and efficacy of the product, and that such an examination must generally be based on an objective evaluation of its likely effects in the light of the scientific data submitted by the MA applicants. It is established that public health requirements must be recognised as fundamentally important (see in particular paragraphs 45 and 46 above).

195    In the present case, while it is admittedly common ground that a significant medical need continues to exist in pharmacological treatments currently available to treat the symptoms of schizophrenia, that cannot cause the authorities responsible for examining MA applications submitted to them to be, in their examination of the risk-benefit balance of the relevant medicinal product, less demanding as regards the parameters that must be taken into account when evaluating the safety of that medicinal product.

196    It follows from all the foregoing that the fifth plea must be rejected and, consequently, that the action must be dismissed in its entirety.

 Costs

197    Under Article 134(1) of the Rules of Procedure, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the successful party’s pleadings. Since the applicant has been unsuccessful, it must be ordered to pay the costs, in accordance with the form of order sought by the Commission.

On those grounds,

THE GENERAL COURT (Sixth Chamber)

hereby:

1.      Dismisses the action;

2.      Orders Vanda Pharmaceuticals Ltd to pay the costs.


Spielmann

Csehi

Spineanu-Matei

Delivered in open court in Luxembourg on 19 December 2019.


E. Coulon

 

      R. da Silva Passos

Registrar

 

      President

Table of contents


Background to the dispute

Procedure and forms of order sought

Law

Admissibility

Substance

Preliminary considerations on the nature and extent of judicial review

– Review of the main features of the centralised procedure for MAs for medicinal products as governed by Regulation No 726/2004

– The extent and scope of the judicial review

The first plea in law: the risk assessment of the arrhythmic potential of iloperidone is vitiated by a failure to state reasons, a manifest error of assessment and a breach of the principle of equal treatment

– Whether the obligation to state reasons was complied with

– The allegation that the iloperidone risk assessment is not consistent with the guidelines for QT/QTc prolongation

– The claim that the CHMP assessment report is not consistent with current EMA practice at the material time as no account has been taken of the positive post-marketing experience of iloperidone

– The claim that the CHMP was unduly influenced by the comments of the ad hoc experts group of 30 October 2017

The second plea in law: the evaluation of the RMMs proposed for iloperidone is vitiated by a failure to state reasons, a manifest error of assessment and a breach of the principle of proportionality, as set out in Article 5(1) and (4) TEU, and of the principle of equal treatment

– The complaint that the evaluation of the RMMs is vitiated by a breach of the obligation to state reasons and manifest errors of assessment

– The complaint alleging a breach of the principle of proportionality

– The complaint alleging a breach of the principle of equal treatment

The third plea in law: the evaluation of the consequences of the delayed onset of iloperidone is vitiated by a failure to state reasons and a breach of the principle of proportionality set out in Article 5(1) and (4) TEU

The fourth plea in law: the obligation to identify a population in which iloperidone would outperform other products constitutes a breach of the principles of conferral and proportionality (set out in Article 5(1) to (3) TEU), Article 12 and Article 81(2) of Regulation No 726/2004 and the principle of equal treatment

The fifth plea in law: the overall risk-benefit assessment of iloperidone lacks an adequate statement of reasons and, in any event, is manifestly erroneous

Costs


*      Language of the case: English.