Background to the complaint

1. The complainant, a pharmaceutical company, submitted an application to the European Medicines Agency (EMA) for an EU marketing authorisation[1] of a medicinal product for treating age-related macular degeneration[2] (AMD), which is a cause of low vision for people over 50 years of age. (See the annex for a timeline).

2. The medicinal product in question contains the active substance bevacizumab, also used in the medicinal product Avastin, which is authorised for marketing in the EU. Avastin has been used as an ‘off-label’ treatment[3] for AMD. As such, Avastin is a less costly alternative to the authorised medicinal products for treating AMD, Lucentis[4] and Eylea[5].

3. The complainant submitted its application for marketing authorisation on the basis of a “well-established medicinal use”[6], which enables applicants to benefit from a derogation from the requirement to provide the results of toxicological and pharmacological tests or the results of clinical trials. It may be applied for where the efficacy and safety of a medicinal product have been well established. The application for marketing authorisation may then be based on data from the existing scientific literature. In this case, the complainant sought to rely primarily on the efficacy and safety data from the off-label use of the product Avastin.

4. EMA’s Committee for Medicinal Products for Human Use (CHMP), which carries out the scientific assessment of medicinal products for which marketing authorisation is sought, recommended, by a majority of votes, to refuse the complainant’s application. In particular, the CHMP took the view that, in order for an authorisation to be considered, the medicinal product should undergo a comparative clinical trial in which it is compared with Avastin. The trial design should be sensitive enough to show potential differences in efficacy and safety. The CHMP considered that the applicant failed to establish in a robust manner that its medicinal product and Avastin (on whose safety and efficacy data from off-label use the applicant relied) are similar.

5. The complainant requested the re-examination of the CHMP’s negative opinion.

6. As part of the re-examination procedure, an ad hoc scientific advisory group (‘advisory group’[7]) meeting was convened by the CHMP. The advisory group consists of experts and provides answers to questions from the CHMP, on a consultative basis.[8] The CHMP remains responsible for issuing a final opinion.

7. After the advisory group meeting, the complainant raised concerns with EMA that one of the experts appointed to that group had not declared an interest related to consultancy work for a pharmaceutical company. EMA contacted the expert in question to check and, based on the reply received, concluded that it was not necessary for him to declare the consultancy contract with the pharmaceutical company, as it concerned a medical device, and not a medicinal product. According to EMA, the consultancy contract could not be considered as a ‘competing interest’ in the context of the complainant’s application for authorisation of its medicinal product.

8. The CHMP subsequently adopted a negative opinion based on the re-examination procedure, and the European Commission, which is responsible for taking final decisions on applications for marketing authorisations, subsequently decided to reject the application.

9. The complainant then submitted a complaint to EMA about alleged conflicts of interest of the advisory group members, the nature of the questions posed by the CHMP to the advisory group, and how the results of the advisory group meeting were reported. EMA replied to the complainant, concluding that, following a careful investigation of the claims submitted, there was no basis to support a “new expert consultation” as requested by the complainant.

10. Dissatisfied with EMA’s reply, the complainant turned to the European Ombudsman.

The inquiry

11. The Ombudsman opened an inquiry into: (i) how EMA dealt with risks of conflicts of interest of the external experts consulted and (ii) whether the report on the advisory group meeting was complete, in particular as regards the recording of minority views.

12. In the course of the inquiry, the Ombudsman received the reply of EMA to the Ombudsman’s questions[9] and, subsequently, the comments of the complainant on EMA's reply.

13. The Ombudsman inquiry team also inspected documents on how EMA followed up on the concerns raised by the complainant, and aspects of the advisory group’s meeting report.[10] The complainant provided comments on the Ombudsman’s inspection meeting report.[11]

Alleged conflicts of interest of the external experts

Arguments presented to the Ombudsman

By the complainant

14. The complainant contended that its right to an unbiased re-examination of its application for market authorisation for the medicinal product in question had been breached in part through the selection of the members of the advisory group.

15. In the complainant’s view, only two out of eight advisory group members were free from conflicts of interest. The complainant argued that it is not desirable to have an advisory group to contain so few experts with no potential conflicts of interest. The objective should be to have no experts with potential conflicts of interest, and to consider experts with conflicting or competing interests only if they possess specific expertise that would otherwise not be available.

16. In addition, the complainant contended that one of the three experts that EMA considered not to have any conflict of interest had worked in a consultancy role for a pharmaceutical company and that this interest should have been declared. Had it been declared, this expert’s participation would have been restricted to that of an expert witness, according to the complainant. As the interest was not declared, the complainant argued that EMA should have considered this a breach of trust.[12]

17. The complainant described the pharmaceutical company that hired the expert as a consultant as one of the leading and most widely known pharmaceutical companies in the world, whose main business is to develop medicinal products, although it is also engaged in developing and marketing medical devices. The complainant also argued that medical devices can be developed to support medicinal treatment modalities of AMD and could, in that capacity, be considered to compete with its medicinal product. Therefore, involvement in the development of a medical device could give rise to a conflict of interest.

18. According to the complainant, a consultancy position with the pharmaceutical industry should be considered a conflict of interest for an expert, independent of the scope of the consultancy services provided by the expert to the industry. The complainant referred in particular to guidance[13] that EMA has issued for declaring interests. This guidance includes declaring interests related to consultancy services to pharmaceutical companies.

19. The complainant also criticised how EMA verified the claim by the expert in question that he provided consultancy only related to a medical device. According to the complainant, EMA should have inspected the contract between the company and the expert.

By EMA

20. EMA said it strives to have a sufficiently broad pool of potential experts available, while avoiding risks of conflicts of interest.

21. How EMA deals with conflicts of interest is set out in the EMA policy on the handling of competing interests of scientific committees’ members and experts[14]. According to this policy, experts are not excluded from a meeting only because they have a connection with a rival product to the medicinal product that is assessed by the experts. However, experts with such ‘competing interests’ participate as ‘expert witnesses’, meaning that their role is limited to testifying and giving advice in reply to questions only. Importantly, expert witnesses do not participate in the adoption of the conclusions of the advisory group.

22. EMA explained that they distinguish between experts who are employed by pharmaceutical companies and those who are or were engaged as consultants. EMA also distinguishes between work related to medical devices and to medicinal products.

23. EMA indicated that, according to its internal rules applicable at the time, experts were required to declare interests in the pharmaceutical sector only if they related to medicinal products and not if they related to medical devices.[15] Under the current policy (which took effect as of January 2023), experts are also required to declare interests related to medical devices. However, EMA said that it would not exclude from advisory groups experts that have had consultancy contracts related to medical devices in the past.[16]

24. EMA trusts that experts provide a full and sincere account of their interests when they complete and sign their declaration of interests. Experts are required to update their declaration of interests as needed and when any (declarable) interest arises. At the start of every scientific meeting, such as the advisory group meeting in question, the chair asks the experts to declare any interest that was not declared at the time when they submitted their declaration of interest and which could potentially create a conflict of interest with regard to items on the meeting agenda.

25. EMA also referred to the time constraints involved, as a re-examination of a marketing authorisation application must be carried out within 60 days of the CHMP’s original decision.^[17] In order to convene an advisory group within this timeframe, EMA is thus subject to strict time limits and it manages this by using internal milestones to ensure that the CHMP can timely adopt an opinion within the legal deadline. In this case, EMA said it contacted 40 potential experts for the purposes of convening an advisory group meeting but a number of them were unavailable. However, it was not possible to delay the advisory group meeting, because of those internal milestones.

26. EMA selected eight experts for the advisory group in question. Based on their declarations of interest, EMA determined that five had competing interests that required them to participate as expert witnesses and were therefore excluded from the adoption of the conclusions.

27. When EMA becomes aware that an expert’s declaration of interests may be incomplete or inaccurate, for example where it is alerted to this by an interested party, EMA said it verifies such concerns by consulting publicly available information, as needed. EMA may also contact the expert concerned to obtain information or clarifications, if it considers this necessary. In this case, EMA reviewed the concern submitted by the complainant, which was accompanied by a journal article. EMA also contacted the expert concerned to ascertain the nature of the consultancy work referred to by the complainant, and found that the interest related to a medical device for the indication of cataracts. EMA said that such a medical device cannot be considered as a rival product to the medicinal product in question, as the indication of cataracts is a condition that is manifestly different to the indication of AMD. In accordance with EMA’s policy, there was no requirement for this interest to be declared.

28. As regards the alleged breach of trust, EMA said that based on the outcome of its assessment, there was no requirement to launch a ‘breach of trust procedure’[18] as the interest was not declarable in the first instance.

The Ombudsman's assessment

How EMA assesses risks of conflicts of interest

29. EMA has to strike a balance between having the necessary expert input to advise its decision-making bodies within the time limits established, on the one hand, and ensuring the credibility of resulting decisions is not undermined by conflicts of interest, on the other.

30. Prior to any involvement in EMA’s activities, nominated experts need to be included in its experts database. In order to be included in the database, the proposed expert must submit the following documents: (a) a nomination form, (b) a duly completed electronic declaration of interest, (c) a confidentiality undertaking and (d) a Curriculum Vitae.

31. EMA has two documents that set out its internal rules related to conflicts of interest of experts. The first document, the Policy on the handling of competing interests of scientific committees[19] (‘Policy’), sets out how EMA deals with potential conflicts of interest of experts, amongst others[20]. The second document, the Procedural guidance on inclusion of declared interests in the European Medicines Agency’s electronic declaration of interests form (for scientific committees’ members and experts) (‘Guidance’), aims to help experts complete EMA’s electronic declaration of interest form by highlighting key aspects of the declarable interests and clarifying what information is required in which section of the form.

32. Where the onus is on experts to declare accurately their relevant interests, it is important that the rules on declarable interests and guidance on how to complete the declaration of interest are clear and unambiguous.

33. EMA’s determination that the expert in question had not erred - when the expert did not declare his consultancy to a pharmaceutical company related to a medical device - was based on its interpretation of the Policy and Guidance as it existed at the time. The complainant interpreted these documents differently. The Ombudsman inquiry team also noted that some other experts appointed to the advisory group had drawn different conclusions as regards what was necessary to declare. Some had declared consultancy work for pharmaceutical companies related to medical devices, whereas at least one other had not. At the time of the advisory group meeting, the Policy and the Guidance lacked clarity as regards whether experts needed to declare consultancy work on medical devices.

34. The Ombudsman is of the view that EMA should have all the relevant information regarding interests of experts in pharmaceutical companies, both regarding medicinal products and medical devices, from the onset. It is then up to EMA to assess these interests, in accordance with its Policy, and to mitigate any risks of conflicts of interest, including potential risks.

35. While the Ombudsman’s inquiry was ongoing, EMA introduced changes to the Policy, which clarify what EMA considers to be declarable interests. In particular, the Ombudsman notes that experts are now clearly required to declare consultancies and strategic advisory work in the medical device sector.  The Guidance already included the requirement for experts to declare interests related to consultancy services to pharmaceutical companies, without distinguishing between whether the consultancy services related to medicinal products or medical devices.[21] With the recent revision of the Policy, the Policy and the Guidance are now consistent as to what interests experts are meant to declare. The Ombudsman welcomes this.

How EMA dealt with the concerns about a supposed undisclosed conflict of interest of an expert and the complainant’s subsequent complaint

36. The complainant was concerned that one of the three experts on the advisory group had not disclosed an interest that gave rise to a conflict of interest, and alerted the EMA shortly after the advisory group meeting (prior to the CHMP vote on the re-examination of its opinion). Later, after the CHMP’s vote on the re-examination of its opinion, the complainant submitted a complaint to EMA as regards the manner in which it had organised the advisory group, including as regards how EMA had dealt with risks of conflicts of interest.

37. The Ombudsman’s inquiry showed that EMA had limited its assessment in relation to the alert to determining whether the expert in question was required to have declared the interest in the first place, and not whether the interest could give rise to a conflict of interest. The Ombudsman’s inquiry also demonstrated that EMA had not verified the concerns raised, notably by consulting other possible sources of publicly available information, nor did it ask the expert in question for the consultancy contract in question.

38. In reply to the complainant’s subsequent complaint, EMA eventually provided adequate reasoning as to why the interest of the expert in question was deemed to be compatible with his full involvement in the advisory group. Specifically, EMA found that the expert’s consultancy role for the pharmaceutical company on a medical device related to cataracts could not be considered as a rival product to the complainant’s medicinal product for treating AMD.

39. The Ombudsman finds that EMA should have provided this reasoning already in relation to the concerns relating to a supposed undeclared conflict of interest.

40. The inquiry showed that EMA receives ‘alerts’ about ’supposed undisclosed conflicts of interest’ of experts approximately once or twice per year. Given such concerns could have profound implications for EMA activities in which experts participate, the Ombudsman is of the view that EMA should verify the information obtained in such alerts by taking into account the information enclosed in the alert, as well as other relevant publicly available information. This should be adequately documented. The Ombudsman agrees with the complainant that EMA could ask the experts in question to submit documentary evidence related to undisclosed interests. To address this and other issues identified in the inquiry, EMA could consider adopting guidance for staff dealing with alerts related to supposed undisclosed conflicts of interest.

The expert group meeting report

Arguments presented to the Ombudsman

By the complainant

41. The complainant argued that the manner in which the advisory group formulated its meeting report was flawed, misleading and one-sided.

By EMA

42. EMA noted that, according to the rules of procedure for the organisation of an advisory group[22], where consensus cannot be reached, the conclusion reached by the majority of experts together with any divergent positions within the group must be recorded and explained in the minutes. There are no formal voting procedures for advisory groups, as they perform an advisory role only and do not adopt opinions. Experts’ names are not linked to positions expressed.

43. EMA said that the meeting report of the advisory group contained answers to the CHMP (subdivided in four sub questions) and a justification for each answer. The conclusion reached by the majority of members together with any divergent positions within the advisory group was recorded and explained. For one answer, there was no consensus, and the divergent position expressed by an expert was recorded in the meeting report.

44. The chair of the advisory group is responsible for drafting the meeting report with administrative support from the EMA secretariat. The meeting report was approved by means of a series of emails between the three experts who were entitled to participate in the adoption of the advisory group conclusions, taking into account the views of the expert witnesses who participated in the meeting.

45. The chair then presented the advisory group meeting report to the CHMP.[23] It was also shared with the complainant.

The Ombudsman's assessment

46. The Ombudsman’s inquiry confirmed that only the views of the non-conflicted experts were reflected in the conclusions of the advisory group provided to the CHMP.

47. The Ombudsman’s inquiry also confirmed that the divergent position of one expert, which could be interpreted as being favourable to the complainant’s position, was expressed by the expert who the complainant alleged to have an undeclared interest.

48. The record of the meeting implied that more than one expert had expressed the diverging opinion that was recorded in the advisory group’s meeting report. The report says: “There was however divergence on this point where other experts considered that no additional data of safety and efficacy of [the complainant’s medicinal product] were needed, and that there are very limited safety issues with Avastin IVT based on Italian experience as long as sterile ‘fractionation’ of multi-dose vials or preferably single dose intravitreal preparations can be ensured” (emphasis added). By referring to ‘experts’ in plural in the recording of the minority opinion, this could have led to the mistaken interpretation that expert witnesses (who should not be involved in deciding on the opinion) had been involved in the adoption of the conclusions of the advisory group meeting. However, during the inquiry, EMA clarified that the diverging opinion reflected the view of only one expert (with no potential competing interest) at the meeting, and that reference to ‘experts’ in the plural referred to the fact that his position was shared by other experts in the scientific community who were not at the meeting.

49. The Ombudsman finds this formulation unfortunate in the context of the organisation of the advisory group and trusts that, in future, EMA will assist the chairs of advisory groups  in formulating their conclusions in a clear and unambiguous manner.  

Conclusion

Based on the inquiry, the Ombudsman closes this case with the following conclusion

While the Ombudsman has identified some issues in the handling of the matter, overall there was no maladministration by the European Medicines Agency.

The complainant and the European Medicines Agency will be informed of this decision.

 

Emily O'Reilly
European Ombudsman

Strasbourg, 18/12/2023

 

 

Annex - timeline

The inquiry team drew up the following timeline of the procedure.

Date	Actions of complainant	Actions of EMA
10 March 2020	Application for marketing authorisation submitted to EMA under Article 10a of Directive 2001/83/EC, based on well-established medicinal use.
11 November 2021		The Committee for Medicinal Products for Human Use (CHMP) issued a negative opinion, recommending that the European Commission to reject the application.
26 November 2021	Requested a re-examination of the CHMP’s negative opinion.
14 January 2022	Supplemented request with the submission of detailed grounds for re‑examination. Asked the CHMP to consult with external experts.
17 January 2022		Re-examination procedure launched. EMA started organising the appointment of an ad hoc expert group (AHEG) for the purposes of seeking specialised input.
14 February 2022		EMA sent the Re-examination Assessment Report of the lead CHMP Rapporteur to the AHEG and to the complainant.
16 February 2022		AHEG meeting took place.
21 February 2022	The complainant sent an alert by email to EMA in relation to alleged undeclared interests of one of the three experts deemed not to have ‘competing interests’.
22 February 2022		EMA requested clarification from the expert in question. EMA received clarifications from the expert that the interest related to a medical device. EMA assessed this clarification and determined that this interest related to medical devices and not to medicinal products. In accordance with EMA’s policy, EMA found there was no interest that needed to be declared, and that this would not have excluded or restricted the expert from participating in the AHEG.
24 February 2022		CHMP re-examined the complainant’s application and confirmed its opinion not to recommend granting a marketing authorisation.
31 March 2022	The complainant submitted a complaint to EMA in relation to the organisation and conduct of the AHEG meeting, requesting a new expert consultation
20 April 2022		EMA conducted an assessment of the complaint, found there was no conflict of interest and provided a detailed reply to the complaint.
29 April 2022		The European Commission adopted a decision rejecting the application for a marketing authorisation.

 

[1] For detailed information on the process for dealing with applications for EU marketing authorisations for medicinal products, see: https://www.ema.europa.eu/en/human-regulatory/marketing-authorisation/obtaining-eu-marketing-authorisation-step-step.

[2] The macula is the central region of the retina, which is the light-sensing layer at the back of the eye. The macula is responsible for the vision needed to see detail for everyday tasks such as driving, reading and recognising faces.

[3] Off-label use refers to any intentional use of an authorised product not covered by the terms of its marketing authorisation. This may for example be the use for a different indication, use of a different dosage, dosing frequency or duration of use, use of a different method of administration, or use by a different patient group (e.g. children instead of adults).

[4] See: https://www.ema.europa.eu/en/medicines/human/EPAR/lucentis.

[5] See: https://www.ema.europa.eu/en/medicines/human/EPAR/eylea.

[6] Pursuant to Article 10(1)(a)(ii) 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 L 311, 28.11.2001, p. 67–128, see: https://eur-lex.europa.eu/legal-content/en/TXT/?uri=CELEX%3A32001L0083.

[7] EMA refers to such ad hoc expert groups using the acronym ‘AHEG’.

[8] The possibility to convene such a group is foreseen under Article 56(2) of Regulation (EC) No 726/2004. The organisation of this group is subject to rules of procedure that are adopted by the CHMP in accordance with Article 61(8) of the same Regulation. See: the Mandate, objectives and rules of procedure for the scientific advisory groups (SAGs)and ad-hoc expert groups (EMA/117014/2010) of 20 May 2010; available at:

https://www.ema.europa.eu/en/documents/other/mandate-objectives-rules-procedure-scientific-advisory-groups-sags-ad-hoc-experts-groups_en.pdf

[9] See: https://www.ombudsman.europa.eu/en/opening-summary/en/164029.

[10] https://www.ombudsman.europa.eu/doc/inspection-report/179321.

[11] https://www.ombudsman.europa.eu/doc/correspondence/179322.

[12] EMA’s breach-of-trust procedure sets out how it deals with incorrect or incomplete declarations of interest by scientific experts and committee members, see for more details: https://www.ema.europa.eu/en/documents/other/european-medicines-agency-breach-trust-procedure-competing-interests-disclosure-confidential_en.pdf.

[13] Namely the procedural guidance on inclusion of declared interests in EMA’s electronic declaration of interests form for scientific committees’ members and experts, see: https://www.ema.europa.eu/en/documents/regulatory-procedural-guideline/procedural-guidance-inclusion-declared-interests-european-medicines-agencys-electronic-declaration_en.pdf. This includes in point 4.2:
“In section 2.2 of the e-DoI, please declare all consultancy advice (including training on a one-to-one basis) provided to one or more pharmaceutical companies. Employment by a consultancy company should be indicated under section 2.1 of the e-DoI”. See also, in annex 2: “Consultancy to a pharmaceutical company shall mean: any activity where the concerned expert provides advice (including training on a one to one basis) to a pharmaceutical company regardless of contractual arrangements or any form of remuneration.”

[14] Based on a previous version of the EMA Policy from January 2021, with reference EMA/MB/89351/2020. The policy has since been updated: https://www.ema.europa.eu/en/about-us/how-we-work/handling-competing-interests.

[15] According to the EMA policy on the handling of competing interests of scientific committees’ members and experts (as applicable at the time of the evaluation of the medicinal product in question), only members and alternates of the EMA Committee for Advanced Therapies needed to declare interests in the medical device sector that could affect their impartiality. See Section 2 of the policy (titled “Scope”), that provides that “The policy focuses on competing interests in the pharmaceutical industry and, in particular on medicinal products. The policy also implements the specific requirement from Article 22 of the ATMP Regulation that members of the Committee for Advanced Therapies and alternates shall not have financial or other interests in the biotechnology sector and medical device sector that could affect their impartiality”.

[16] The same goes for members of Scientific Advisory Groups (SAGs). AHEGs are a subcategory of SAGs.

[17] As provided in Article 9(2), second subparagraph, of Regulation (EC) No 726/2004.

[18] EMA has a breach-of-trust procedure in place, which sets out how it deals with incorrect or incomplete declarations of interests by scientific experts and committee members: https://www.ema.europa.eu/en/about-us/how-we-work/handling-competing-interests.

[19] The version of 11 June 2020 that applied at the time of the AHEG

[20] As well as scientific members, although that is not relevant for this inquiry.

[21] Under point 4.2 of the Procedural guidance on inclusion of declared interests in the European Medicines Agency’s electronic declaration of interests form for scientific committees’ members and experts. See: https://www.ema.europa.eu/en/documents/regulatory-procedural-guideline/procedural-guidance-inclusion-declared-interests-european-medicines-agencys-electronic-declaration_en.pdf

[22] Part 3.7 of the rules of procedure, point 14. See footnote 10 for a hyperlink.

[23] The AHEG meeting report was eventually included in the European Public Assessment Report (EPAR) for the medicinal product in question, which is public.