Important developments for the medical devices and medicinal products industry: clarification of EU law about the borderline between these groups of products is currently in the making at the European Court of Justice. The German Oberlandesgericht Frankfurt am Main referred a reference for preliminary ruling to the European Court of Justice, which was lodged there on 20 June 2011 under number C-308/11. The case concerns the borderline between medicinal products in the meaning of applicable EU law and medical devices, and it turns on the ever problematic demarcation question: “What does pharmacological action mean?”. Although the particulars of the dispute are as yet difficult to ascertain and I was not able to find any in public sources, we do know the three questions that were referred to the European Court in order to obtain a highest instance interpretation of concepts of EU law:
- For the purpose of defining the term ‘pharmacological action’ in Article 1(2)(b) of Directive 2001/83/EC, as amended by Directive 2004/27/EC, can recourse be had to the document compiled under the auspices of the European Commission to provide guidance in distinguishing between medicinal products and medical devices (the ‘Medical Devices: Guidance document’), which states that there must be an interaction between the molecules of the substance in question and a cellular constituent, usually referred to as a receptor, which either results in a direct response or blocks the response of another agent?
- If the first question is answered in the affirmative: does the term ‘pharmacological action’ require that there should be an interaction between the molecules of the substance in question and cellular constituents of the user, or is it sufficient if there is an interaction between the substance in question and a cellular constituent which does not form part of the human body?
- In the event that the first question is answered in the negative or that neither of the two definitions proposed in the second question is appropriate: which alternative definition should be used instead?
It will be a while before we will first see the opinion of the Advocate General in this matter and subsequently the Court’s judgement. I expect about nine months for the opinion of the AG and then another three to six months for the Court’s judgment.
Borderlines and pharmacological action
Let’s take a step back in the mean time and look at borderlines again. As I have argued on this year’s DIA Euromeeting, the use of the term borderLINE is not helpful from a scientific perspective, because borderline questions are in the end scientific questions, and no so much legal questions and for that reason it would be more appropriate to speak about a border area, in which the edges of two or more regulatory regimes (like in this case medicinal products and medical devices) overlap. Judges, as I have shown in that presentation, are not at all well-equipped to decide scientific questions.
What do we know already about how the European Court interprets ‘pharmacological action’ and what factual support it requires for its interpretation? We know that the legal criterion is that a product, for it to constitute a medicinal product by function, must “significantly affect the metabolism and thus modify the way in which it functions” (see BIOS Naturprodukte, point 21). The question then is: “Does ‘significantly affect the metabolism and thus modify the way in which it functions’ require that there should be an interaction between the molecules of the substance in question and cellular constituents of the user, or is it sufficient if there is an interaction between the substance in question and a cellular constituent which does not form part of the human body?”
To ask this question is to answer it: that depends on whether the interaction between the substance and the cellular constituent that does not form part of the body is still capable of restoring, correcting or modifying physiological functions by exerting a pharmacological, immunological or metabolic action. It is difficult to answer this question without any knowledge of the product concerned, but excluding interactions between substance and cellular constituent removed from the body could have consequences for the Advanced Therapy Medicinal Products Regulation. This regulation relies on Directive 2001/83/EC for its core definition of medicinal product (see article 2 (a)) and the core criterion of ‘substantial manipulation’ (article 2 (c)) presumes that often procedures will be applied to cellular constituents that do not form part of the body.
My guess therefore is that the European Court will bounce this question right back in the way I just did: it will repeat its case law and then leave it to the national court to decide on the facts if the interaction between the substance and the cellular constituent that does not form part of the body is still capable of restoring, correcting or modifying physiological functions by exerting a pharmacological, immunological or metabolic action.
Furthermore, the first question of the preliminary reference misrepresents the MEDDEV in that it suggests that the MEDDEV requires “an interaction between the molecules of the substance in question and a cellular constituent usually referred to as a receptor, which either results in a direct response, or which blocks the response to another agent.”, because the question is phrased that the MEDDEV “states that there must be an interaction” (bold and underlining added). In the text of the MEDDEV (see p. 6) the element describing ‘interaction’ is however immediately followed by this statement:
“Although not a completely reliable criterion, the presence of a dose-response correlation is indicative of a pharmacological effect.”
Wait a second, all my lawyer alarms are going off now. Not a completely reliable criterion? Correlation? Indicative? There is the wiggle room in science that the European Court will use as described above. The fact that the German judge posed the question as a positive “must” while science in these cases only speaks about “not completely reliable criterion”, “correlation” and “indicative” underlines my thesis that judges are really not that well equipped to answer scientific questions, especially when there is correlation involved, which they all too often mistake for causality, as a Dutch judge recently did in a case involving a dispute about the outcome of a statistical analysis of clinical trial data. So, again, knowing the European Court a bit, it will rephrase the first question to be able to give a useful answer, state that the causal link implied does not exist, and then repeat the criteria from its case law set out above, to finally hand it down again to the Frankfurt am main Oberlandesgericht to apply the criterion that did not change at all to the actual facts of the case, which it just should have done in the first place like the Hecht Pharma case prescribes (points 38-45).
This case does show however that there is demand for a supra-national EU capability to make borderline product calls between the different regulated products regimes, which sources close the Recast drafting process tell me is likely to make it politically. If that is instated cases like this could actually be decided on an EU level.
Status of MEDDEVS
Some further remarks about this case: the status of MEDDEVs – can one have recourse to these documents in legal proceedings? The first preliminary question obviously refers to MEDDEV 2.1/3 on Borderline products, drug-delivery products and medical devices incorporating,as integral part, an ancillary medicinal substance or an ancillary human blood derivative, last updated in December 2009. In addition there is a Manual on borderline and classification in the Community Regulatory framework for medical devices containing guidance on a number of specific borderline cases, last updated in August this year. In response to a Parliamentary Question about MEDDEV 2.1/3 the Commission answered in 2008:
“Moreover, in order to ensure a uniform approach within the Community, the Commission services have drafted a Manual that represents the agreed position of the Medical Device Expert Group ‘Borderline and Classification’ on specific borderline products or categories of products. This Manual will be regularly updated in the light of the discussions of this group. Once endorsed by the Medical Device Expert Group ‘Borderline and Classification’ the Manual will be published on the Commission’s website.
In order to update the MEDDEV guidance 2.1/3, the Commission services have undertaken a broad written consultation of stakeholders through the involvement of the abovementioned expert group, which includes experts from Member States’ competent authorities, the relevant Commission services, the European Medicines Agency and industry trade associations. These stakeholders were asked to send their written contributions on the draft revised guidance by the end of January 2008.
Due to the participation of the aforementioned interested parties and of experts from competent authorities, it is anticipated that this guidance will be followed within the Member States and, therefore, will ensure uniform application of the directives.”
So, it is clear that the MEDDEV and the Manual were both drafted with participation of the authorities of the Member States and that they are supposed to adhere to them because they are stakeholders in it. Does that impose an obligation for member states to act according to the MEDDEV? That will really first of all depend on national law governing policy documents because as we have seen: the EU has no competence under the current rules to decide in this matters. That is done by national authorities and judges.
The status of the MEDDEV is clearer to me than the status of the Manual although they address the exact same thing. The MEDDEV is a document drafted and released by the Commission itself after consultation with interested parties, as the Commission page on MEDDEVs explains. Such documents are not binding on the Commission but may serve to help the Commission meet its obligation to give reasoned decisions, as has been established in competition law cases (see for example Corsica Ferries, point 66). The documents can however never overrule or pre-empt the European Court’s or national courts’ interpretation of the EU statutes concerned, as these documents make quite clear themselves. MEDDEVs therefore only bind the Commission and then only insofar as the Commission uses them to give reasons for its decisions. One earlier case that was filed as a direct action claiming that the Court of First Instance should order the Commission to make a notified body act as set out in a MEDDEV failed quite spectacularly (sorry, the actual judgment in this case is only available in French).
The Manual is much more interesting, because the group behind that (the Medical Device Expert Group ‘Borderline and Classification’) consists of industry, trade associations and notified bodies. The Commission merely publishes the manual on its website if al of these parties agree. Even if this case clarifies the status of MEDDEV, it will not clarify the status of the Manual, which deals with the exact same questions and is also a reflection of positions of representatives of Competent Authorities, Notified Bodies, industry and other interested parties in the medical devices sector.
However, my guess here is that the European Court will say that it is for the German court to decide under German law to what extent it and German authorities are bound by MEDDEVs and Manuals they helped write, provided of course that they can never act contrary to EU law in the application of that law because Germany has the duty to ensure effective application of EU law on its territory.
An interesting case, although there is not much known about it yet until the AG opinion comes out. That should contain a good description of the facts underlying this case. In the mean time I could not help myself and had to write this speculative post. If any reader of this blog has more information about this case, I would be much obliged! Discussion about the outcome of the case is more than welcome too.