By Sabrina Röttger-Wirtz
The independence and impartiality of administrative procedures is a core component of the right to good administration and within the EU is emphasized in Article 41 of the Charter of Fundamental Rights as well as Article 298 TFEU, as well as many pieces of secondary legislation. Conflict of interest policies and rules concerning the independence and impartiality of administrative procedures within the agencies are reoccurring topics in the literature and also on this blog. In the past, the European Medicines Agency (EMA) has been no stranger to conflicts of interest criticism, especially from the European Parliament. This led to an in-depth overhaul and reform of the EMA policy on conflicts of interest. However, apparently not all kinks are ironed out yet, as recent judgments of the Court and also a strategic inquiry of the Ombudsman have shown in the course of this year.
Impartiality and national interests
In March, the Court of Justice ruled case C-680/16P, where the Dr. August Wolff GmbH was challenging a Commission decision that limited the marketing authorization for its cream Linoladiol N. The origin of the dispute lies in Germany, where the German Federal Institute for Drugs and Medical Devices (BfArM) decided in September 2005 not to renew the marketing authorization due to concerns regarding the safety and effectiveness. The company appealed the decision in German administrative courts. While the German case was ongoing, the BfArM initiated a referral procedure under Article 31(1) of Directive 2001/83/EC, which brings the matter to the EMA’s main scientific committee, the CHMP. In this referral procedure, the CHMP appointed as rapporteur – a key figure in the scientific assessment tasked with drafting the report that forms the basis of the Committee opinion – an employee of BfArM, thus, the German authority that had triggered the Article 31 referral in the first place and was still involved in German court proceedings with the applicant. The Commission subsequently had followed the CHMP opinion and limited the marketing authorization of the product.
The Wolff GmbH claimed that the rapporteur had been biased, especially given that the German court procedures between the company and the BfArM were still ongoing, and that the referral procedure was, therefore, not conducted in an impartial manner. The Court, in contrast to the pervious General Court judgement, found in favor of the pharmaceutical company. It declared in paragraph 27 of the judgement that objective impartiality requires that: ‘there must be sufficient guarantees to exclude any legitimate doubt as to possible bias on the part of the institution concerned’. It compared the administrative procedure at BfArM and the referral procedure at the EMA and found that the procedures are the same concerning their purpose and the nature of the decision taken. Furthermore, it stressed that a rapporteur has an important role in the referral procedure and has responsibilities of his/her own. The Court therefore found an infringement of the objective impartiality requirement, concluding that the applicant could have legitimately considered that the rapporteur was pursuing national level interest. The Court found no procedural safeguard – such the co-rapporteur from another state or the fact that the opinion is formed by the whole committee – could disperse these doubts. Thus, on the basis of an infringement of the right to good administration enshrined in Article 41 of the Charter, the Court annulled the Commission decision in question.
The case is to a certain degree uncommon, as with regard to the EMA usually it is the relationship with the industry, rather than the relationship with the Member States that is cause for concern. However, the particular circumstances of the case make clear why it is an important issue to consider impartiality with regard to the Member States interest, especially in the EU’s composite administration, where administrative procedures are carried out with the formal or informal input of various actors on national and EU level. This is especially so as judicial accountability in such composite procedures is often flawed.
In this regard it is interesting to note that the recently adopted Regulation 2019/1381 introduced Member State representation in the Management Board of the European Food Safety Authority (EFSA) and strengthens the role of Member States in the appointment of the pool of expert on which the EFSA relies, which in itself is remarkable as the EFSA was created in the aftermath of the BSE crisis exactly to separate scientific advice form political influence. But then again, if the EFSA will carefully considers and prevents potential conflict of interest of key actors in its procedures, also with regard to Member State interest, it will not necessarily affect the impartiality of the scientific procedures.
Impartiality and conflicted experts
The Wolff case can be contrasted with the recently decided case T‑783/17 GE Healthcare A/S, which also arose in the context of a referral procedure. However, in GE Healthcare the alleged conflict of interest concerned an expert who had participated in an expert group, that had been consulted by an EMA scientific committee. The expert had previously been a consultant in a national class action against the applicant and others, and was at the time sued for libel by the applicant. The main question for the Court to determine the impartiality of the procedure was whether the expert had a decisive impact on either the conduct or the outcome of that procedure. The General Court did not see such an impact, as he was not a member of the involved EMA committees and a second expert group, not including the expert in question, was consulted as well.
When compared with the Wolff case one can conclude that a conflict of interest of a key actor in the procedure, like the rapporteur, cannot be counterbalanced through the fact that the opinion is formed by the scientific committee as a whole and thus reviewed by other committee members. However, this is different from the involvement of experts with conflicting interests, which have been consulted in the process, but have no formal role in the scientific opinion. Overall, these testing criteria and their application by the General Court and the Court of Justice seem to strike a fair balance between the need for very specific expertise which might entail pervious involvement with the product to some degree and the impartiality of the administrative procedure.
Conflicts of interest and scientific advice
This summer, also the European Ombudsman examined a specific aspect of conflicts of interests in EMA procedures. The Ombudsman issued a decision in a strategy inquiry regarding the pre-submission activities the EMA carries in the form of scientific advice and protocol assistance for future applicants. Scientific advice means that companies can consult with EMA experts which provide non-binding advice on how to best design studies which are required for a later marketing authorization. Also here, independence of the procedure was a concern as the CHMP adopts the pre-submission scientific advice (via the so-called Scientific Advice Working Party), but is also later responsible for the assessment of the marketing authorization application. The coordinators of the scientific advice provided to companies, can later become rapporteurs for the actual marketing authorization procedure, then being tasked with assessing the outcome of the studies. While this overlap of functions might be well-justified due to the specific expertise required from the coordinator and also the rapporteur, it could influence the impartiality of the rapporteur or at least the public perception thereof. Therefore, the Ombudsman recommended that, there should be a separation between the persons responsible for the scientific advice and the authorization procedure. At least one of the rapporteurs should not have prior contact with the applicant concerning the product in question.
Again Regulation 2019/1381 is interesting as comparator in this regard, as it also introduces a pre-submission advice mechanism for the European Food Safety Authority. The key difference is, however, that EFSA will be prevented from advising on study design, while advising only on rules applicable in the procedure and the content required from the applicants. This would prevent impartiality concerns as the Ombudsman raised them in case of the EMA.
Fine-tuning the EMA conflicts of interest policy in the absence of a coherent framework
The judgments and the Ombudsman decision discussed above will certainly lead to an even closer examination of the experts involved in the EMA administrative procedures and potential conflicts of interest. This will contribute to a further refinement of the existing EMA conflict of interest policy as well as the procedures with regard to rapporteur appointment and pre-submission scientific advice. However, it seems regrettable that such a refinement of the impartiality rules in administrative procedures are defined in sector specific contexts and shaped in a rather ad hoc manner. In the absence of an EU administrative procedure law European agencies are left to their own devices in determining the specificities of their procedures, falling back only on very general principles such as good administration. In the context of growing public mistrust, as exemplified in the glyphosate approval discussion, diverging approaches in the agencies regarding impartiality and independence policies, will not contribute to dispersing doubts concerning the objectivity of their assessments.
Sabrina Röttger-Wirtz is Assistant Professor of EU Law at Maastricht University. This blogpost is based on a case note written for issue 4/2019 of the European Pharmaceutical Law Review (forthcoming).