Provisional text
OPINION OF ADVOCATE GENERAL
EMILIOU
delivered on 6 June 2024 (1)
Joined Cases C‑119/22 and C‑149/22
Teva BV,
Teva Finland Oy
v
Merck Sharp & Dohme Corp.
(Request for a preliminary ruling from the markkinaoikeus (Market Court, Finland))
and
Merck Sharp & Dohme Corp.
v
Clonmel Healthcare Limited
(Request for a preliminary ruling from the Supreme Court (Ireland))
(Reference for a preliminary ruling – Medicinal products for human use – Supplementary protection certificate (SPC) – Regulation (EC) No 469/2009 – Products consisting in a combination of active ingredients – Conditions for grant – Article 3 – Point (a) – The product is ‘protected’ by a basic patent – Point (c) – The product ‘has not already been the subject of a certificate’ – Proper tests for the assessment of those conditions)
I. Introduction
1. The present requests for a preliminary ruling, from the markkinaoikeus (Market Court, Finland) and the Supreme Court (Ireland) respectively, concern the conditions for the grant of a supplementary protection certificate (‘SPC’) for medicinal products, set out in Article 3 of Regulation (EC) No 469/2009 (2) (‘the SPC Regulation’). In essence, those courts wonder whether, and to what extent, an SPC may be granted for a combination of active ingredients used in such a product, where a previous SPC has already been granted for one of those ingredients. In that context, they seek guidance on the interpretation of two of those conditions, namely that such a combination be ‘protected by a basic patent in force’ (Article 3(a)) and that it ‘has not already been the subject of [an SPC]’ (Article 3(c)).
2. As will be explained in this Opinion, these issues can hardly be considered as new. In fact, they have already been the subject of several decisions of the Court, including the judgments in Actavis I, (3)Actavis II (4) and Teva I. (5) Despite (or, as some uncharitable commentators might say, because of) those decisions, the national authorities tasked with granting SPCs and the courts called upon to review their validity still struggle to make out, with certainty, the tests governing the conditions at issue. That uncertainty leads to difficulties and divergences of assessment with respect to the SPC eligibility of certain subject matters (combinations of active ingredients being one of them).
3. In that context, the referring courts seek to know, with respect to the assessment of each condition, what the proper test should be. They enquire, in that regard, about certain ambiguous parts of the judgments in Actavis I, Actavis II and Teva I and how the first two judgments interact with the third. Their requests provide the Court with a new opportunity to set the matter straight, hopefully once and for all.
II. Legal framework
4. Article 3 of the SPC Regulation, entitled ‘Conditions for obtaining [an SPC]’, provides as follows:
‘[An SPC] shall be granted if, in the Member State in which the application referred to in Article 7 is submitted and at the date of that application:
(a) the product is protected by a basic patent in force;
(b) a valid authorisation to place the product on the market as a medicinal product has been granted …;
(c) the product has not already been the subject of [an SPC];
(d) the authorisation referred to in point (b) is the first authorisation to place the product on the market as a medicinal product.’
III. Facts, national proceedings and the questions referred
A. Case C‑119/22
5. Merck Sharp & Dohme Corp. (‘Merck’) is a pharmaceutical company. It is the holder of European patent EP 1 412 357, granted by the European Patent Office (EPO) on 22 March 2006 for, inter alia, Finland, with a priority date of 5 July 2002 (‘the basic patent in Case C‑119/22’). That patent was valid until 5 July 2022.
6. The title of the patent in question is ‘Beta-amino tetrahydroimidazo(1,2-a)pyrazines and tetrahydrotriazolo(4,3-a)pyrazines as dipeptidyl peptidase inhibitors for the treatment or prevention of diabetes’. The ‘Summary of the invention’ section of the description of the patent in question states that the invention according to the basic patent is directed at substances which are inhibitors of the dipeptidyl peptidase-IV enzyme (‘DP-IV inhibitors’) and which, as such, are useful for the treatment or prevention of diseases in which the dipeptidyl peptidase-IV enzyme is involved, such as diabetes and particularly type 2 diabetes. Furthermore, that section states that the invention is also directed at pharmaceutical compositions comprising those substances and the use of those substances and compositions in the treatment or prevention of such diseases in which the dipeptidyl peptidase-IV enzyme is involved.
7. That patent contains a total of 30 patent claims. Claim 1 is a product claim relating to a substance, drafted in the form of a Markush formula; Claims 15, 26 and 28 refer more specifically to certain substances falling under that formula, represented in the form of chemical structural formulas, including a substance that later became known as ‘sitagliptin’; furthermore, Claims 20, 25 and 30 relate to (a) combinations consisting in one of the claimed substances with one or more other substances, selected from a group listed in those claims, which are also used for the treatment of diabetes, as well as to (b) pharmaceutical compositions containing such a combination. In particular, Claim 30 relates to a pharmaceutical composition that is comprised of a combination of one of the substances claimed in the patent and a substance known as ‘metformin’ (which is a drug in the public domain, also used for the treatment of diabetes). (6)
8. On 21 March 2007, Merck obtained a marketing authorisation with respect to the medicinal product ‘Januvia’, a drug that is used to treat type 2 diabetes and contains sitagliptin as its sole active ingredient.
9. On 31 August 2008, Merck obtained another marketing authorisation with respect to the medicinal product ‘Janumet’, another drug that is used to treat type 2 diabetes but contains sitagliptin and metformin hydrochloride (which is a pharmaceutically acceptable metformin salt) as a combination of active ingredients.
10. On 13 March 2012, Merck obtained an SPC in Finland (namely SPC No 343) for sitagliptin, on the basis of (i) the basic patent at issue in Case C‑119/22 and (ii) the marketing authorisation for ‘Januvia’. That SPC expired on 23 September 2022.
11. On 20 March 2012, Merck was granted another SPC in Finland (namely SPC No 342) for the combination of sitagliptin and metformin, on the basis of (i) the same patent and (ii) the marketing authorisation for ‘Janumet’. That SPC was valid until 8 April 2023.
12. Subsequently, Teva BV and Teva Finland Oy (collectively, ‘Teva’), a pharmaceutical firm which produces generic medicines, brought an action before the markkinaoikeus (Market Court) against Merck for a declaration of invalidity of the second SPC (SPC No 342). Teva submits that that certificate was issued in breach of the conditions laid down in Article 3 of the SPC Regulation.
13. Specifically, Teva submits, inter alia, that the contested SPC was issued in breach of Article 3(a) of the SPC Regulation on the ground that the ‘product’ for which it was granted, namely the combination of sitagliptin and metformin, was not ‘protected’ (within the meaning of that provision) by the basic patent in Case C‑119/22.
14. Teva also submits that SPC No 342 was granted in breach of Article 3(c) of the SPC Regulation. Since a first SPC (in this case SPC No 343) had already been granted in Finland for sitagliptin, that provision precluded the grant of another SPC for that active ingredient combined with metformin.
15. Merck opposed the form of order sought by Teva and requested that the action be dismissed. With respect to Article 3(a) of the SPC Regulation, Merck submitted that Teva’s argument is founded on an incorrect test for the assessment of that condition. Under the proper test, the combination of sitagliptin and metformin was indeed ‘protected’ (within the meaning of that provision) by the basic patent in Case C‑119/22. With respect to Article 3(c) of that regulation, Merck submitted that the prior grant of an SPC for sitagliptin (SPC No 343) did not preclude the grant of another SPC for a combination of sitagliptin with metformin (SPC No 342), as that combination is, for the purposes of that provision, a ‘product’ different and distinct from sitagliptin alone.
16. It is in those circumstances that the markkinaoikeus (Market Court) decided to stay the proceedings and to refer the following questions to the Court of Justice for a preliminary ruling:
‘(1) What criteria must be applied to determine when a product has not already been granted [an SPC] within the meaning of Article 3(c) of [the SPC Regulation]?
(2) Must the assessment of the condition set out in Article 3(c) of the SPC Regulation be regarded as being different from the assessment of the condition set out in Article 3(a) of that regulation, and if so, in what way?
(3) Must the statements on the interpretation of Article 3(a) of the SPC Regulation in the judgments of the Court in [Teva I ] and [Royalty Pharma (7)] be regarded as relevant to the assessment of the condition in Article 3(c) of the SPC Regulation and, if so, in what way? In that connection, particular attention should be paid to the statements made in those judgments regarding Article 3(a) of the SPC Regulation, specifically:
– the essential meaning of patent claims; and
– the assessment of the case from the point of view of a person skilled in the art and in the light of the prior art at the filing date or priority date of the basic patent.
(4) Are the concepts “core inventive advance”, “central inventive step” and/or “subject matter of the invention” of the basic patent relevant to the interpretation of Article 3(c) of the SPC Regulation and, if any or all of those concepts are relevant, how are they to be understood for purposes of interpreting Article 3(c) of the SPC Regulation? For the purposes of applying those concepts, does it make any difference whether the product in question consists of a single active ingredient (“mono-product”) or a combination of active ingredients (“combination product”) and, if so, in what way? How is the latter question to be assessed in a case in which the basic patent contains, on the one hand, a patent claim for a mono-product and, on the other hand, a patent claim for a combination product, the latter patent claim relating to a combination of active ingredients consisting of the active ingredient of the mono-product plus one or more active ingredients from the known prior art?’
B. Case C‑149/22
17. Merck is also the holder of European patent EP 0 720 599, granted by the EPO on 19 May 1999 for, inter alia, Ireland, with a priority date of 21 September 1993 (‘the basic patent in Case C‑149/22’). That patent expired in September 2014.
18. The title of the patent in question is ‘Hydroxy-substituted azetidinone compounds useful as hypocholesterolemic agents’. The description of the patent indicates that certain substances known as ‘azetidinones’ have the effect of inhibiting the absorption of cholesterol into the bloodstream at the borders of the intestinal villus in the small intestine. As such, those substances are useful for the treatment and prevention of atherosclerosis. (8)
19. Claims 1 to 8 of that patent relate to single molecules, including a substance known as ‘ezetimibe’. By contrast, Claims 9, 12, 15 and 16 address uses of ezetimibe in combination with other molecules, including statins (which are substances also used to treat high cholesterol). (9) Furthermore, Claim 17 refers to a combination of ezetimibe and one of the statins listed in that claim, including ‘simvastatin’ (which is a substance in the public domain).
20. In 2003, Merck obtained a marketing authorisation with respect to a medicinal product named ‘Ezetrol’, a cholesterol-lowering medicine which contains ezetimibe as its sole active ingredient.
21. That same year, Merck was granted an SPC in Ireland (namely SPC No 2003/014) for ezetimibe, on the basis of (i) the basic patent at issue in Case C‑149/22 and (ii) the marketing authorisation for ‘Ezetrol’. That SPC expired in April 2018.
22. In 2004, Merck obtained a marketing authorisation with respect to another medicinal product named ‘Inegy’, a drug that is also a cholesterol-lowering medicine, but contains ezetimibe and simvastatin as a combination of active ingredients.
23. In 2005, Merck was granted another SPC in Ireland (namely SPC No 2005/01) for the combination of ezetimibe and simvastatin, on the basis of (i) the basic patent at issue in Case C‑149/22 and (ii) the marketing authorisation for ‘Inegy’. That SPC expired in April 2019.
24. After the expiry of the SPC for ezetimibe, but while the SPC for ezetimibe and simvastatin was still valid, Clonmel Healthcare Limited (‘Clonmel’), a pharmaceutical company which produces generic medicines, launched a generic version of ‘Inegy’.
25. Taking the view that the production and marketing of that generic drug infringed the second SPC, Merck brought infringement proceedings against Clonmel before the High Court (Ireland), seeking a prohibitive injunction and damages.
26. As part of its defence in those proceedings, Clonmel claimed that the SPC for ezetimibe and simvastatin was invalid on the ground that it was granted in breach of the conditions laid down under Article 3 of the SPC Regulation. Essentially, Clonmel argued, first, that the combination of active ingredients in question was not ‘protected’ (within the meaning of Article 3(a) of that regulation) by the basic patent in Case C‑149/22, and, second, that the grant of the first SPC for ezetimibe precluded, under Article 3(c) of that regulation, the grant of a second SPC for that active ingredient in combination with simvastatin.
27. On 29 November 2019, the High Court concluded that the SPC for ezetimibe and simvastatin was invalid under Article 3(a) and Article 3(c) of the SPC Regulation and thus made an order revoking that SPC. On 24 February 2021, the Court of Appeal (Ireland) upheld that decision.
28. On 24 May 2021, Merck sought leave to appeal the judgment of the Court of Appeal, which was granted by the Supreme Court on 4 August 2021.
29. Observing that the core issue in the main proceedings concerns the validity of the SPC granted for the combination of ezetimibe and simvastatin which, in turn, depends on the proper interpretation of Article 3(a) and (c) of the SPC Regulation, and that that matter is (still) not clear in the Court’s case-law, the Supreme Court decided to stay the proceedings and to refer the following questions to the Court of Justice for a preliminary ruling:
‘(1) (a) For the purpose of the grant of [an SPC], and for the validity of that SPC in law, under Article 3(a) of [the SPC Regulation], does it suffice that the product for which the SPC is granted is expressly identified in the patent claims, and covered by it; or is it necessary for the grant of an SPC that the patent holder, who has been granted a marketing authorisation, also demonstrate novelty or inventiveness or that the product falls within a narrower concept described as the invention covered by the patent?
(b) If the latter, the invention covered by the patent, what must be established by the patent holder and marketing authorisation holder to obtain a valid SPC?
(2) Where, as in this case, the patent is for a particular drug, ezetimibe, and the claims in the patent teach that the application in human medicine may be for the use of that drug alone or in combination with another drug, here, simvastatin, a drug in the public domain, can an SPC be granted under Article 3(a) of [the SPC Regulation] only for a product comprising ezetimibe, a monotherapy, or can an SPC also be granted for any or all of the combination products identified in the claims in the patent?
(3) Where a monotherapy, drug A, in this case ezetimibe, is granted an SPC, or any combination therapy is first granted an SPC for drugs A and B as a combination therapy, which are part of the claims in the patent, though only drug A is itself novel and thus patented, with other drugs being already known or in the public domain[,] is the grant of an SPC limited to the first marketing of either that monotherapy of drug A or that first combination therapy granted an SPC, A+B, so that, following that first grant, there cannot be a second or third grant of an SPC for the monotherapy or any combination therapy apart from that first combination granted an SPC?
(4) If the claims of a patent cover both a single novel molecule and a combination of that molecule with an existing and known drug, perhaps in the public domain, or several such claims for a combination, does Article 3(c) of [the SPC Regulation] limit the grant of an SPC;
(a) only to the single molecule if marketed as a product;
(b) the first marketing of a product covered by the patent whether this is the monotherapy of the drug covered by the basic patent in force or the first combination therapy[;] or
(c) either (a) or (b) at the election of the patentee irrespective of the date of market authorisation?
And if any of the above, why?’
IV. Procedure before the Court
30. Written observations in Case C‑119/22 have been submitted by Teva, Merck, Ireland, the French, Latvian, Hungarian and Netherlands Governments and the European Commission. Written observations in Case C‑149/22 have been submitted by Merck, Clonmel, Ireland, the French, Hungarian, Netherlands and Polish Governments and the Commission.
31. By decision of the President of the Court dated 17 January 2023, Cases C‑119/22 and C‑149/22 were joined for the purposes of the hearing and the judgment.
32. Teva, Clonmel, Merck, Ireland, the French, Latvian, Hungarian and Netherlands Governments and the Commission were represented at the hearing that took place on 8 March 2023.
V. Analysis
33. As indicated in the introduction, the present cases concern the conditions under which SPCs may be granted in the European Union for combinations of active substances used in medicinal products. Before dissecting the questions referred to the Court, I find it appropriate to provide the reader, who may not be familiar with the intricacies of this complex area of law, with a basic overview of the background and relevant rules.
34. When one (typically, a pharmaceutical company) discovers, through research, that a given substance (or family of substances, or combination of substances, and so on) has a certain effect on the human body, which makes it useful in the treatment, prevention or management of a certain disease or condition (or several diseases or conditions, and so on), one may, in certain circumstances, protect the fruits of that research against competition through the patent system. In Europe, one may, in particular, (10) apply for a ‘European patent’ before the EPO in Munich (Germany), pursuant to the centralised procedure established by the EPC. (11) For such a patent to be granted, certain conditions must be met (‘patentability requirements’): the idea of using the substance in question as a medicine must constitute an ‘invention’ that is, inter alia, ‘new’ and involves an ‘inventive step’. (12) For present purposes, it suffices to say that, assuming those requirements are met, such a patent is granted (generally for a period of 20 years) (13) and provides its holder with certain exclusive rights (essentially, a trade monopoly) over the patented ‘invention’ in several or all of the European countries which are party to the EPC. (14) As such, the patent holder may prevent third parties from making and offering for sale a drug using the ‘invention’ in question within the territory of those States. In exchange for that 20-year protection against competition, the patentee must ‘disclose’ its invention in the patent, (15) that is to say, divulge it to the public so that everyone is free to use it (including by preparing a copy or generic version (16) of such a drug) after the expiry of the patent. (17)
35. Nevertheless, before the patent holder can put such a patented, pharmaceutical invention on the market as a medicinal product in the European Union, it must obtain a marketing authorisation for that product from the competent authorities. (18) In that context, extensive pre-clinical tests and clinical trials need to be carried out to demonstrate the safety and efficiency of the product. Hence, the procedure for granting such an authorisation usually takes years. Effectively, the period during which the patent holder is able to market its invention under the protection of the patent and reap the benefits of its monopoly is reduced accordingly.
36. In that context, the EU legislature saw fit to compensate pharmaceutical companies for those regulatory delays by granting them, in certain circumstances, an additional period of marketing exclusivity. To that effect, it created the SPC system.
37. That system is not a straightforward patent extension mechanism such as those that exist in other jurisdictions, whereby the validity of a patent is directly extended for a given number of years. Indeed, while the SPC was designed to serve as an extension to patent rights (for, as its very name indicates, it provides a protection that is ‘supplementary’ to the latter), the legal reality is unfortunately more complex than that. In fact, an SPC is a sui generis intellectual property right, with a somewhat sophisticated subject matter and mode of operation.
38. Essentially, an SPC may be granted for a ‘product’ which is ‘protected’ by a given patent (referred to, in that context, as the ‘basic patent’) (19) and in respect of which a marketing authorisation has been issued for the first time. (20) Pursuant to the definition laid down in Article 1(b) of the SPC Regulation, such a ‘product’ can be either the ‘active ingredient’ (21) or the ‘combination of active ingredients’ as such of a medicinal product. The holder of a European patent which ‘protects’ such a ‘product’ may file, with respect to the latter, an application for an SPC within six months of the date on which the marketing authorisation was granted. (22) Since the SPC is a national title, the patent holder must do so individually before the NPOs of all the Member States for which that European patent was issued and that authorisation obtained. Once granted, an SPC takes effect at the end of the 20-year term of the ‘basic patent’ for a period which is commensurate to the time it took to obtain that marketing authorisation and, in any event, cannot exceed 5 years.(23) During its lifespan, the SPC confers the same exclusive rights as the basic patent (as defined by the law applicable to that patent), but solely with respect to the specific ‘product’ for which that certificate was issued. (24)Effectively, the SPC ensures the patentee up to five extra years of trade monopoly over the ‘product’ in question.
39. The SPC system, and the (sort of) patent extension it allows, while prima facie limited in its effect, is, in fact, of singular importance in practice. Indeed, that system has to be seen in the light of the tremendous (and diverging) economic interests of the stakeholders of the pharmaceutical sector. On the one hand, the business model of pharmaceutical companies that develop novel medicines (‘the originator industry’) is heavily dependent on the patent-conferred monopolies over their medicines (and the high revenues they derive from them). Logically, those companies seek to extend those monopolies as much as possible. On the other hand, the business model of manufacturers of generic medicines consists in putting on the market the generic equivalents of successful originator drugs, something from which they may also derive considerable revenue. However, those companies cannot lawfully do so while such a monopoly exists. Those diverging economic interests explain why the period surrounding the expiration of patents for originator medicines is often conflictual and why, in particular, the grant of SPCs frequently leads to litigation.
40. The cases in the main proceedings are illustrative of the foregoing considerations. For the purposes of my analysis, the relevant features of those cases are essentially the same, and may be summarised as follows.
41. In each case and at some point in time, Merck discovered that a certain family of substances (pyrazines in Case C‑119/22 and azetidinones in Case C‑149/22) had a certain effect on the human body (pyrazines inhibit the dipeptidyl peptidase-IV enzyme and azetidinones inhibit the absorption of cholesterol into the bloodstream). Subsequently, in the 1990s, Merck was granted a European patent (covering various Member States) for an invention consisting in the innovative idea of using substances belonging to that family, in the light of their effect on the human body, as a medicine to treat certain diseases or health conditions (pyrazines in relation to, inter alia, diabetes, and azetidinones in relation to high cholesterol and atherosclerosis). The patent in question also referred to the idea of using those substances in combination with other substances which were already used to treat those diseases or health conditions at the time the application for that patent was filed (metformin in relation to diabetes and statins in relation to high cholesterol and atherosclerosis).
42. Merck developed and obtained a marketing authorisation for a first drug to treat the diseases or health conditions concerned, containing a substance belonging to the patented family in question as its sole active ingredient (sitagliptin in Case C‑119/22 and ezetimibe in Case C‑149/22). For the sake of simplicity, in the rest of this Opinion I will usually refer to that substance as ‘A’. Subsequently, to compensate for the time it took to obtain that authorisation, Merck was granted a first SPC for A (in various Member States).
43. Later, Merck developed and obtained a marketing authorisation for another medicinal product for treatment of the same diseases or health conditions, which had, as active ingredients, A in combination with one of the other substances that was already used to that effect, as envisioned in Merck’s patent (metformin in Case C‑119/22 and statins in Case C‑149/22). For the sake of simplicity, in the rest of this Opinion I will usually refer to that other known substance as ‘B’, and, thus, to that combination of active ingredients as ‘A+B’. Afterwards, Merck applied for, and was granted (again, in various Member States), a second SPC for A+B.
44. That second SPC (in its national version issued, respectively, in Finland and Ireland) is at the heart of the cases in the main proceedings. Essentially, manufacturers of generic medicinal products (Teva in Case C‑119/22 and Clonmel in Case C‑149/22), which were prevented from producing and marketing generic versions of Merck’s medicines by that further extension of its monopoly, are challenging the validity of that SPC before the referring courts.
45. Those manufacturers submit, in each case, that the second SPC was invalid on the ground that it was issued contrary to the cumulative (25) conditions laid down in Article 3 of the SPC Regulation. (26) Their challenge to the validity centre on the conditions set out in Article 3(a) and Article 3(c) of that regulation. I recall that the first provision requires that ‘the product is protected by a basic patent in force’, and the second that ‘the product has not already been the subject of [an SPC]’. In their view, the contested SPC did not meet those requirements, which Merck contests.
46. The disagreement of the litigants rests on a diverging understanding of those conditions. In that context, on the one hand, the first and second questions in Case C‑149/22, (27) which it is appropriate to examine together, concern Article 3(a) of the SPC Regulation. On the other hand, all the questions in Case C‑119/22, and the third and fourth questions in Case C‑149/22, which it is also appropriate to examine together, concern Article 3(c) of that regulation. As indicated in the introduction, those numerous questions are, at their core, about the proper tests to apply in order to assess each condition and, ultimately, whether (and if so, to what degree) one condition or the other (or both of them) precludes the grant of an SPC for a combination of active ingredients (A+B), particularly in the scenario, at issue in the main proceedings, where the patent holder previously obtained an SPC for one of those ingredients (A).
47. Although that may seem counterintuitive, I will start by addressing the questions related to Article 3(c) of the SPC Regulation. Indeed, several interveners submitted, before the Court, that the key to solving the present cases lies therein. I disagree and wish to provide, in that respect, some (arguably well-needed) clarifications and a (relatively) swift answer from the outset (Section A). In my view, the key to solving those cases lies, in reality, in the proper construction of Article 3(a) of that regulation, which calls for a more substantial discussion (Section B).
A. Article 3(c) of the SPC Regulation (first to fourth questions in Case C‑119/22 and third and fourth questions in Case C‑149/22)
48. I recall that, as part of the cumulative conditions for grant of an SPC, Article 3(c) of the SPC Regulation requires that ‘the product has not already been the subject of [an SPC]’.
49. In that regard, the referring courts wish to know (i), generally, the proper test to determine whether that condition is met in any given case and (ii), specifically, whether, under the proper test, that condition precludes the grant of an SPC for a combination of active ingredients (A+B) where an SPC has already been granted for one of those ingredients (A).
50. The test to be followed to assess the condition laid down in Article 3(c) of the SPC Regulation is (supposed to be) straightforward. As Merck submits, and as it stems from the clear wording of that provision, it consists in (i) defining the ‘product’ for which the SPC application under review is made or for which the contested SPC has been granted and (ii) verifying whether the patent holder had already obtained an SPC, at a prior date, for the same ‘product’.
51. For the purposes of, inter alia, that provision, I recall that Article 1(b) of the SPC Regulation provides a definition of the concept of ‘product’, which refers to ‘the active ingredient or combination of active ingredients of a medicinal product’ (emphasis added).
52. In each of the main proceedings, the contested SPC was granted for A+B. Under the definition restated in the previous point, such a combination of active ingredients constitutes, in itself, a ‘product’. Thus, as Merck submits, for the purposes of the assessment of the condition laid down in Article 3(c) of the SPC Regulation, following the test explained in point 50 above, (i) the relevant ‘product’ is the A+B combination and (ii) the examiner is to verify whether the patent holder has already obtained an SPC for that combination. Here, it is undisputed that no prior SPC had been granted for A+B. Therefore, that condition is fulfilled.
53. That conclusion is not (or at least should not be) called into question by the fact that, in each case, Merck had obtained a prior SPC for A. Indeed, pursuant to the definition laid down in Article 1(b) of the SPC Regulation, A, as a single active ingredient, is a ‘product’ different from A+B, as a combination of active ingredients. Thus, under Article 3(c) of that regulation, the grant of an SPC for A should not preclude the grant of another SPC for A+B. (28)
54. That being said, the doubts of the referring courts with respect to the interpretation of Article 3(c) of the SPC Regulation stem from two decisions of the Court, namely the judgments in Actavis I and Actavis II, mentioned in the introduction, where the Court departed from the straightforward logic described above.
55. The cases that led to those judgments had a certain factual proximity with the present ones. In each of those previous cases, (i) a first SPC had been granted to a patent holder for an active ingredient (A) on the basis of a patent and a marketing authorisation for a medicinal product containing that ingredient and (ii) a second SPC had then been granted for the combination of that active ingredient with another active ingredient in the public domain (A+B). In each case, a UK court had to determine the validity of the second SPC and referred to the Court, for that purpose, several questions pertaining to the proper interpretation of, inter alia, Article 3(a) and Article 3(c) of the SPC Regulation.
56. The Court followed essentially the same reasoning in both judgments. In essence, it took the view that Article 3(c) of the SPC Regulation precluded the grant of the SPC for A+B. In that regard, it pointed out that the basic patent at issue ‘protected’ (within the meaning of Article 3(a) of that regulation) only A, since that active ingredient was the ‘core inventive advance’ (to use the terms of the Court in the judgment in Actavis I) (29) or the ‘sole subject matter of the invention’ (to use the terms of the Court in the judgment in Actavis II) (30) under that patent. By contrast, B was a known active ingredient in the public domain. In such circumstances, the grant of an SPC for the combination of A+B would be tantamount to the grant of a second SPC for A, contrary to Article 3(c) of that regulation. (31)
57. The Court added that such a construction of Article 3(c) was consistent with the objectives pursued under the SPC Regulation. Indeed, the SPC system was designed to compensate patent holders for the regulatory delays they have to face (as explained in points 35 and 36 above) before they can first market their pharmaceutical inventions. In the cases in question, in the Court’s view, only A constituted such an invention; thus, the first SPC for A had already fulfilled that function. By contrast, if the patent holder could obtain a new SPC each time it placed on the market its invention (A) ‘in all possible forms’, including in the form of combination drugs including other known ingredients (B), it would unduly favour the interests of the pharmaceutical industry to the detriment of the generic manufacturers and, ultimately, of public health (whereas the legislature sought, in the SPC Regulation, to take into account, and balance, all those interests). (32) It could also facilitate ‘evergreening’ strategies, (33) by which pharmaceutical companies could prolong, in an excessive manner, their monopoly by marketing a first medicine containing A, then a medicine containing A+B, then another one composed of A+C, and so on. (34)
58. Understandably, Teva, Clonmel, the intervening governments (save for the Hungarian) and the Commission submit that the Court should follow the same construction of Article 3(c) of the SPC Regulation in the present cases. For my part, I do not think so.
59. To be clear, I have a great deal of sympathy for the pragmatic and teleological reasoning of the judgments in Actavis I and Actavis II. In fact, I concur with the policy considerations put forward by the Court. While, as stated in point 38 above, the SPC Regulation expressly provides for the possibility to obtain SPCs for combinations of active ingredients, it would have been contrary to the spirit of that regulation to admit it there. In certain circumstances, which will be further discussed in the next section, it would be excessive to allow patent holders to obtain an SPC for such combinations. Nevertheless, in my view, the construction of Article 3(c) of the SPC Regulation endorsed in those judgments is not the appropriate course of action in this respect.
60. On the one hand, it seems to me that Article 3(c) of the SPC Regulation is not open to such a purposive interpretation. Indeed, that provision is neither ambiguous nor vague as to the nature of the condition therein. For the purposes of that provision, the definition of ‘product’ in Article 1(b) of that regulation is also limpid in the fact that an ‘active ingredient’ and a ‘combination of active ingredients’ are two different things. Furthermore, in its judgment in Santen, (35) the Court took the view that that definition is ‘strict’. In my view, by ignoring that definition in the judgments in Actavis I and Actavis II, the Court, however laudable its intention, overrode the clear wording of that regulation.
61. Moreover, in those judgments, the Court also distorted the system laid down in Article 3 of the SPC Regulation. Indeed, while that article sets out four cumulative conditions, each with its own logic and purpose and which should accordingly be assessed independently of one another, the Court ended up conflating two of them. Indeed, it essentially required the examiner to verify what a patent ‘protects’ (A, B and/or A+B?) in order to decide whether two SPCs granted on that basis concern the same ‘product’. By doing so, the Court imported into Article 3(c) of that regulation an analysis which pertains, by its nature, to Article 3(a) thereof. (36) That state of affairs creates a regrettable confusion, with national authorities wondering whether they are expected to carry out the same analysis under the two conditions, or a different one. (37)
62. On the other hand, as the Hungarian Government submits, Article 3(c) of the SPC Regulation, even interpreted as in the judgments in Actavis I and Actavis II, is in fact insufficient to prevent SPCs being granted for combinations of active ingredients contrary to the spirit of that regulation. In the first place, the application of that provision would depend on the prior grant of an SPC for one of the ingredients included in that combination and, in the second place, the prohibition laid down in that provision, following the judgment of the Court in Biogen, (38) can easily be circumvented by pharmaceutical groups. (39)
63. Therefore, in the present case, the Court should, in my view, endorse the straightforward and literal construction of Article 3(c) of the SPC Regulation stated in point 52 above. It should restore the integrity of the system laid down therein by reserving any discussion as to what a patent ‘protects’ for the assessment of the condition laid down in Article 3(a) of that regulation. The considerations of the Court in the judgments in Actavis I and Actavis II which pertained to that issue (particularly the concepts of ‘core inventive advance’ and ‘subject matter of the patent’) should be examined in that context.
64. Contrary to what Teva, Clonmel and the Lithuanian Government submit, such a literal application of Article 3(c) of the SPC Regulation would not open a freeway for abuse of the SPC system. In fact, like the Hungarian Government, I take the view that the legitimate policy concerns touched upon by the Court in the judgments in Actavis I and Actavis IIwith respect to the grant of SPCs for combinations of active ingredients can be addressed in a way which is both more efficient and more respectful of the wording and system of the SPC Regulation, through proper construction, and strict application, of the latter condition, as will be explained in the next section.
B. Article 3(a) of the SPC Regulation (first and second questions in Case C‑149/22)
65. Article 3(a) of the SPC Regulation states, I recall, that, for a given ‘product’ to be eligible for an SPC, it must be ‘protected by a basic patent in force’.
66. In each of the cases in the main proceedings, there is no dispute that the relevant ‘basic patent’ (that is to say, the European patent designated by Merck for the purposes of the procedures for the grant of the second, contested SPC) was ‘in force’ at the time at which that SPC was applied for.
67. By contrast, the litigants are very much in dispute over whether the ‘product’ for which that SPC was granted (which, I recall, is A+B, as a combination of active ingredients) (40) was ‘protected’ by that patent. The answer to that issue depends on what ‘protected’ means.
68. Since that term is not defined in the SPC Regulation, national courts have on several occasions sought guidance from the Court as to its construction, often in relation to ‘products’ consisting in combinations of active ingredients, as in the present cases. The resulting case-law evokes The Long and Winding Road described by The Beatles. In a first series of decisions, the Court provided somewhat diverging explanations on the matter (Section 1). As that meandering line of decisions became a source of legal uncertainty, a few years ago the Court, sitting in Grand Chamber, attempted to clarify its case-law and provide a definitive test in its judgment in Teva I (Section 2). Nevertheless, uncertainty remains. Indeed, that judgment (and the test laid down therein) require important clarifications (Section 3). Should the Court provides those clarifications in the future judgment in the present cases, patent practitioners and national authorities involved in SPC matters will, hopefully and eventually, reach the ‘door’ to which that ‘road’ leads.
1. The case-law of the Court before the judgment in Teva I
69. The very first case in which the Court was asked to clarify the meaning of the term ‘protected’ used in Article 3(a) of the SPC Regulation was Farmitalia. (41) It is not necessary to recall the facts of that case; it suffices to say that the Court provided a straightforward answer, pursuant to which, ‘in the absence of Community harmonisation of patent law, the extent of patent protection can be determined only in the light of the non-Community rules which govern patents’. Accordingly, whether a given ‘product’ is ‘protected’ by a patent within the meaning of that provision was to be determined not by EU law, but solely by (national or EPC) patent law. (42)
70. Thus, in the judgment in Farmitalia, the Court seemingly deferred the matter entirely to the (national or international) rules of patent law that concern the scope (or ‘extent’) of the protection conferred by a patent (a crucial issue, particularly in infringement proceedings), such as Article 69 of the EPC. Under that provision and its explanatory protocol, the ‘extent of the protection’ conferred by a European patent depends on the patent claims, which must be construed in the light of the description (and potential drawings) contained in the patent. Applying such an ‘extent of protection’ test in the context of Article 3(a) of the SPC Regulation would mean that a ‘product’, including a combination of active ingredients (A+B), would be regarded as ‘protected’ by a patent if it is covered by the claims, so construed.
71. However, despite that first answer, in subsequent cases, the Court laid down two (seemingly autonomous, EU-law-based) tests to determine whether a given ‘product’ is ‘protected’ by a patent within the meaning of Article 3(a) of the SPC Regulation.
72. On the one hand, the Court laid down, for that purpose, what I will call an ‘identification’ test in its judgments in Medeva (43)and Eli Lilly. (44) In those judgments, after initially indicating that the matter was left to national law (and expressly referring to the ‘extent of protection’ rules and, in particular, Article 69 of the EPC in the latter judgment), the Court ruled that a ‘product’ may be regarded as ‘protected’ by a patent within the meaning of Article 3(a) of the SPC Regulation only if it is ‘specified in the wording of the claims’ (Medeva) (45) or, at least, ‘identified’ in the claims, either expressly or ‘implicitly but necessarily and specifically’ (Eli Lilly). (46) Just as the ‘extent of protection’ test discussed in point 70 above, the ‘identification’ test depends on the wording of the claims of the basic patent, while being more demanding in that regard. (47) Under the latter test, in particular, a combination of active ingredients (A+B) would be eligible for an SPC only if it were expressly referred to (by chemical name or structural formula), or at least identifiable, with a sufficient degree of specificity, in those claims.
73. On the other hand, in parallel, the Court laid down what I will call an ‘invention’ test in its judgments in Actavis I and Actavis II. As discussed in the previous section, in each of those cases, on the basis of the same patent, (i) a first SPC had been granted to a patent holder for an active ingredient (A) and (ii) a second SPC had then been granted for the combination of that active ingredient with another active ingredient in the public domain (A+B). The validity of the combination SPC was contested. In the relevant parts of each judgment, the Court indicated that, in its view, the basic patent ‘protected’ only A (and, thus, not A+B). The Court did not refer to the claims of that patent, but rather based its reasoning on the fact that A was the ‘core inventive advance’ (Actavis I) (48) or the ‘sole subject matter of the invention’ (Actavis II) (49) under that patent. Furthermore, the Court hinted at the fact that the combination of A+B could only have been regarded as ‘protected’ if it had been a ‘separate innovation’ (presumably, from A). (50) In sum, in those judgments, the Court seemed to indicate that, irrespective of whether the claims of the basic patent meet the ‘identification’ test (51) discussed in the previous points (which it did not even mention), a given ‘product’, especially a combination of active ingredients, is eligible for an SPC only if it corresponds to the ‘invention’ for which that patent was issued.
2. The ‘definitive test’ set out in the judgment in Teva I
74. In the light of the uncertainties that arose from the diverging indications given in the judgments discussed in the previous section, the Court, sitting in Grand Chamber, seized the opportunity afforded to it by another preliminary reference concerning Article 3(a) of the SPC Regulation to clarify its case-law; it did so in the judgment in Teva I.
75. Once more, the case that led to that decision concerned the SPC eligibility of a combination of active ingredients. In essence, Gilead Sciences Inc. was granted a patent for an invention consisting in the use of a family of substances for the treatment of HIV. The claims of that patent referred, inter alia, to (i) one of those substances (A) and (ii) a pharmaceutical composition containing such a substance ‘and optionally other therapeutic ingredients’. Gilead Sciences Inc. developed and obtained a marketing authorisation for a medicinal product containing A+B as a combination of active ingredients (B being a substance in the public domain, also useful for the treatment of HIV). It subsequently obtained an SPC for that combination, the validity of which was challenged by Teva UK Ltd before the UK courts. In that context, the national judge wondered whether the ‘identification’ test laid down in the judgments in Medevaand Eli Lilly was satisfied in the circumstances at issue and whether, in addition to (or maybe instead of) that test, the ‘invention’ test stemming from the judgments in Actavis I and Actavis II had to be satisfied for the A+B combination to be regarded as ‘protected’ by the basic patent under Article 3(a) of the SPC Regulation.
76. Again, the Court began, in paragraphs 31 to 33 of the judgment in Teva I, by indicating that the matter was left to national patent law and, more specifically, to the ‘extent of protection’ rules, such as Article 69 of the EPC. Nevertheless, after a detailed outline of its reasoning, which will be examined in the next section, the Court presented its definitive test in paragraph 57 and the operative part of that judgment, as follows:
‘… a product composed of several active ingredients [(52)] with a combined effect is “protected by a basic patent in force” within the meaning of [Article 3(a) of the SPC Regulation] where, even if the combination of active ingredients of which that product is composed is not expressly mentioned in the claims of the basic patent, those claims relate necessarily and specifically to that combination. For that purpose, from the point of view of a person skilled in the art and on the basis of the prior art at the filing date or priority date of the basic patent:
– the combination of those active ingredients must necessarily, in the light of the description and drawings of that patent, fall under the invention covered by that patent, and
– each of those active ingredients must be specifically identifiable, in the light of all the information disclosed by that patent.’
3. The uncertainties that arise from the judgment in Teva I and the clarifications needed
77. Despite the judgment in Teva I, there is still controversy around what needs to be satisfied for a ‘product’ to be regarded as ‘protected’ by a patent within the meaning of Article 3(a) of the SPC Regulation. The present cases are illustrative in that regard.
78. In all fairness, Teva, Clonmel and Merck (as well as the other interveners) do agree on one thing. It seems clear that, contrary to what is indicated in the judgment in Farmitalia, the term ‘protect’ used in Article 3(a) of the SPC Regulation is not defined by reference to the (national or international) rules governing patents, such as Article 69 of the EPC. While the Court, unfortunately, again hinted at that at the beginning of its judgment in Teva I, it simply cannot be the case; otherwise, it would have stopped its reasoning after paragraph 33 of that judgment.
79. The fact that it did not do so, but rather continued and added a series of considerations and requirements with respect to the construction of Article 3(a) of the SPC Regulation, demonstrates that in reality the Court has given an autonomous, EU-law meaning to the term ‘protected’ used in that provision (which only partially coincides with the understanding of the ‘extent of protection’ of a patent under Article 69 of the EPC).
80. That approach is, in my view, sound. After all, it is settled case-law that the terms of a provision of EU law which, like Article 3(a) of the SPC Regulation, makes no express reference to national (or international) law for the purposes of determining its meaning and scope must, as a rule, be given an independent and uniform interpretation throughout the European Union, taking into consideration the context in which they are used and the objectives pursued by the instrument of which they are part. (53)
81. That is all the more necessary here given the fact that Article 3(a) of the SPC Regulation sets out a requirement that is crucial for the grant of SPCs in the Member States. In that regard, what is at stake is not so much the imperative that SPCs be granted under the same conditions across the European Union; (54) indeed, the scope of a European patent is determined, under Article 69 of the EPC, the same way in all the Member States. (55) Rather, an autonomous construction of the term ‘protected’ used in Article 3(a) of the SPC Regulation, in the light of the specific context and objective pursued by that instrument, is necessary (56) to ensure that SPCs are issued only where that is in line with the spirit of that instrument. Simply applying, for the purposes of that provision, the rules on the ‘extent of protection’ of patents, such as Article 69 of the EPC, would not always guarantee such an outcome, as I shall explain below.
82. Thus, to determine whether a ‘product’ is ‘protected’ by a patent, for the purposes of Article 3(a) of the SPC Regulation, an autonomous test must be carried out which (partially) differs from the ‘extent of protection’ test applied as a matter of patent law. (57) It is also clear that the former is more stringent than the latter. Effectively, certain ‘products’ may be regarded as protected by a patent from the point of view of patent law, while not being regarded as ‘protected’ by that patent for the purposes of Article 3(a) of that regulation. A clear acknowledgment by the Court of that (for now implicit) state of affairs in the future judgment in the present cases would already constitute a welcome clarification of the judgment in Teva I.
83. The controversy begins there. Indeed, as the referring court in Case C‑149/22 underlines, there are two ways to understand the test laid down in paragraph 57 (and the operative part) of the judgment in Teva I and the interaction of the latter with the Court’s previous case-law.
84. Pursuant to a first reading, defended by Merck, in the judgment in Teva I, the Court endorsed and refined the ‘identification’ test set out in the judgments in Medevaand Eli Lilly. As such, a ‘product’, including a combination of active ingredients, is to be regarded as ‘protected’ by a basic patent within the meaning of Article 3(a) of the SPC Regulation where it is ‘expressly mentioned in the claims of that patent’ (first scenario) or where ‘those claims relate to that product necessarily and specifically’ (second scenario). In the first scenario (which was not at issue in the case that led to the judgment in Teva I), the condition laid down in that provision is automatically satisfied. Indeed, the Court did not require any further assessment of whether a ‘product’ so expressly mentioned constitutes the ‘core inventive advance’ or the ‘subject matter of the invention’ under the patent. Thus, the Court implicitly dismissed the ‘invention’ test set out in Actavis I and Actavis II and overruled those last judgments. By contrast, in the second scenario (which was at issue in the case that led to the judgment in Teva I), the two indents that appear at the end of paragraph 57 of that judgment gain significance; they constitute a sort of ‘subtest’ designed ‘for [the] purpose’ (58) of determining whether the claims of the patent may be regarded as relating ‘necessarily and specifically’ to a ‘product’ which they do not mention expressly. (59)
85. Pursuant to a second reading of the judgment in Teva I, defended by Teva, Clonmel, the intervening governments and the Commission, the Court endorsed both the ‘identification’ test and the ‘invention’ test. It refined and turned them into the new, two-prong test set out in paragraph 57 of that judgment which must be followed in every case to determine whether a ‘product’ is ‘protected’ by a patent for the purposes of Article 3(a) of the SPC Regulation. It means that, to be regarded as such, the ‘product’ must not only be expressly mentioned in the claims or, at least, ‘specifically identifiable’ by a person skilled in the art (second prong), but also ‘fall under the invention’, that is to say correspond to the innovation for which the patent was granted (first prong).
86. The referring court in Case C‑149/22 invites the Court to clarify which is the correct reading of the judgment in Teva I. Besides the fact that the equivocal nature of that judgment apparently led to diverging decisions at national level, it is crucial for the outcome of the main proceedings. Indeed, that national court has established that the combination of ezetimibe and simvastatin, which is the subject of the second, contested SPC, was ‘expressly mentioned’ in a claim of the relevant basic patent. (60)
87. In my view, the second reading of the judgment in Teva I is the correct one. Indeed, while I do agree that, taken at face value, paragraph 57 and the operative part of that judgment could be read in the manner suggested by Merck (which, admittedly, is unfortunate), the following reasons have convinced me otherwise.
88. At the beginning of the judgment in Teva I (specifically, in paragraph 30), the Court announced what it intended to do. At the invitation of the referring court, it was to clarify whether, for a ‘product’ to be regarded as ‘protected’ by a patent, within the meaning of Article 3(a) of the SPC Regulation, it suffices that it pass the ‘identification’ test set out in the judgments in Medevaand Eli Lilly, or whether, on top of that, ‘an additional criterion’ must be satisfied. That was, obviously, a reference to the ‘core inventive advance’ and ‘subject matter of the invention’ mentioned in the judgment in Actavis I and Actavis II, which the national judge expressly mentioned. (61)
89. The answer given by the Court in the follow-up reasoning is, admittedly, not entirely straightforward; rather, it goes slightly back and forth. However, when one puts the pieces of the puzzle together, one can see the full picture, and it is, in my view, a clear one.
90. On the one hand, the Court spent a significant part of its reasoning restating that for a ‘product’ to be regarded as ‘protected’ by a basic patent within the meaning of Article 3(a) of the SPC Regulation, it must pass the ‘identification’ test as initially set out in the judgments in Medevaand Eli Lilly. The Court repeated the statement found in the latter judgment whereby such a ‘product’ may only be regarded as so ‘protected’ if it is ‘expressly mentioned in the claims of that patent’ or, at least, ‘those claims relate to that product necessarily and specifically’. (62)
91. Furthermore, the Court clarified the circumstances under which the claims of a patent can be regarded as relating ‘necessarily and specifically’ to a ‘product’ (something that it had very much left open to interpretation in the judgment in Eli Lilly). That is the case where that ‘product’ is ‘specifically identifiable’ by a person skilled in the art in the light of all the information disclosed by the basic patent (especially by the description) (63) and of the prior art at the date of filing of the patent application or priority date of that patent. That is, in fact, the second prong of the test laid down at the end of paragraph 57 and the operative part of the judgment in Teva I. (64)
92. The Court also explained the rationale behind that requirement. In the light of the purpose of the additional period of exclusivity granted by an SPC, which is to ‘ensure that the investments put [by the basic patent holder] into … research are covered’, that prong of the test is designed to ensure that SPCs are granted only for ‘products’ which have been developed, through such research, when the application for the basic patent was filed. Indeed, it would run counter to the purpose of SPCs if a patent holder could, on the basis of broadly formulated claims (including generic functional definitions covering a vast family of substances), obtain such a certificate for a substance which was not known at the time of filing the application, but was developed later, as the result of further research, potentially carried out by a third party. (65)
93. On the other hand, in my view, several statements in the judgment in Teva I clearly indicate that for a ‘product’ to be regarded as ‘protected’ by a patent, within the meaning of Article 3(a) of the SPC Regulation, it does not suffice that it is ‘expressly mentioned in the claims of that patent’ or that ‘those claims relate to that product necessarily and specifically’, and that the ‘additional criterion’ foreshadowed in paragraph 30 of that judgment must, indeed, be satisfied.
94. Specifically, in paragraph 43 of the judgment in Teva I, the Court indicated that the claims of the basic patent must also ‘be construed in the light of the limits of [the invention covered by the patent], as it appears from the description and the drawings of that patent’.
95. In that regard, the Court added, in paragraph 46 of that judgment, that ‘the subject matter of the protection conferred by an SPC must be restricted to the technical specifications of the invention covered by the basic patent, such as claimed in that patent’. Unfortunately, the meaning of that statement is lost in the English language version of the judgment due to a mistake in translation. For patent practitioners, the terms ‘technical specifications of the invention’ are at best meaningless and at worst confusing. (66) In fact, in the French language version of that judgment (which is the language in which it was drafted), that paragraph refers to the ‘caractéristiques techniques’ of the invention, that is to say, its ‘technical features’. (67)
96. Accordingly, as Teva, Clonmel, the intervening governments and the Commission submit, it stems from those paragraphs that, to determine whether a ‘product’ is ‘protected’ by a basic patent, one must also (i) identify in the claims, read in the light of the description and drawings of that patent, the subject matter of that patent (the ‘invention’ and its technical features, which is a narrower concept than the ‘extent of the protection’ granted by the patent around that subject matter) (68) and (ii) determine whether that ‘product’ corresponds to that ‘invention’. (69)
97. Subsequently, the Court concluded that a ‘product’ may only be regarded as ‘protected’ by a patent where it ‘necessarily falls under the invention’ for which that patent was granted. That is, effectively, what the first prong of the test laid down at the end of paragraph 57 and the operative part of the judgment in Teva I is about.
98. As those interveners observe, that test requires more than a mere reference to the ‘product’ in the claims of the basic patent. In particular, where the ‘product’ for which an SPC is sought is a combination of active ingredients, the claim(s) relating to that combination must be read in the light of the description and drawings of that patent to determine whether that combination corresponds to the ‘invention’ for which that patent was granted. Thus, that test is a confirmation and a refinement of the ‘invention’ test set out in the judgments in Actavis I and Actavis II.
99. Contrary to what Merck submits, that finding is not, in my view, called into question by the fact that, in paragraph 31 of its judgment in Royalty Pharma the Court indicated, at the request of the referring court, that the concept of ‘core inventive advance’ (used in the judgment in Actavis I) is not (or is no longer) relevant in the context of Article 3(a) of the SPC Regulation.
100. Indeed, while the Court dismissed those terms, it restated, in the same paragraph, that ‘the subject matter of the protection conferred by an SPC must be restricted to the technical [features] of the invention covered by the basic patent’, confirming that the nature and scope of the ‘invention’ are decisive in deciding whether a ‘product’ is ‘protected’ by a patent within the meaning of that provision. Therefore, in my view, the Court wanted, in the paragraph at issue, to validate a change of terminology rather than of substance. It substituted the concept of ‘invention’ for that of ‘core inventive advance’ (something it already did, in fact, in the judgment in Actavis II), probably because the former is known to patent practitioners and thus carries the message it wants to send better than the latter. (70) Nevertheless, the idea behind all those terms is, in essence, the same: whether or not the ‘product’ corresponds to the ‘invention’ for which the patent was granted.
101. The reasons given by the Court as to the purpose of the first prong of the test set out in the judgment in Teva I confirm, in my view, that interpretation. In paragraph 40 of that judgment, the Court indicated that it would be contrary to the objective of the SPC Regulation, which is to support innovation in the pharmaceutical sector, and the balance of interests that it seeks to achieve in that regard to grant an SPC for a ‘product’ which does not fall under the ‘invention’ covered by the basic patent, ‘inasmuch as such an SPC would not relate to the results of the research claimed under that patent’. Some explanations are required in that regard.
102. Essentially, that first prong was designed with combinations of active ingredients in mind. In that regard, it is clear from the definition of the concept of ‘product’ set out in Article 1(b) of the SPC regulation that, as Merck emphasises, that regulation envisions the grant of SPCs for such combinations. Indeed, it stems from the Explanatory Memorandum that the drafters of the SPC system intended to reward the development of ‘new combination[s] of substances containing a new or known product’. (71) Nevertheless, I would like to distinguish, with respect to combinations of active ingredients, two situations.
103. On the one hand, the idea of using certain active ingredients (new or known) in a particular combination may constitute an ‘invention’ which is ‘new’ and involves an ‘inventive step’ and, as such, is eligible for a patent. That is the case where those active ingredients, when combined, display an innovative ‘synergistic effect’, going beyond their mere additive effect, (72) useful for the treatment of certain diseases or health conditions. Usually, where that is the case, the combination is the subject matter of a dedicated patent, disclosing the innovative effect of that combination. As the intervening governments submit, those are the types of ‘new combination[s]’ that the EU legislature had in mind and intended to encourage under the SPC system. (73) As such, they should be rewarded with an SPC when developed.
104. On the other hand, as helpfully described by the Hungarian Government in its submissions, it appears to be a well-established practice that, where pharmaceutical companies file applications for patents relating to the development of new, single active ingredients, those companies include in those applications, in addition to the main claims dedicated to the ingredient(s) in question (A), one or several (dependant) claims (74) for the use of that (or those) ingredient(s) in combination with other known substances (A+B, A+C, and so on) as particular ‘embodiments’ of the invention. Usually, the description does not disclose any innovative ‘synergistic effect’ inherent to those combinations. Often, that description does not even substantiate their suitability (ability to function well together, absence of dangerous side effects, and so on). In fact, such combination claims can be entirely speculative and added for the sole purpose of broadening the extent of the protection conferred by the patent under Article 69 of the EPC.
105. Although, as Merck submits, that practice seems accepted by the EPO (and, indeed, it appears that a large number of European patents granted by that office contain such combination claims), (75) it is clear that, as the other interveners reply, it would go against the objectives of the SPC Regulation and the balance of interests it seeks to achieve to grant SPCs for combinations of active ingredients in the second situation.
106. Indeed, in such a situation, the A+B combination is not, in itself, the new and inventive idea that is the result of the research disclosed in the patent; A is. Thus, the development of A should be rewarded by an SPC. By contrast, the combination of A+B should not be so rewarded simply because a speculative claim for that combination was added in the patent for A.
107. Merck’s argument that the grant of an SPC for such a combination of active ingredients presupposes the development and marketing of a medicinal product including that combination, which requires research and testing (to obtain the marketing authorisation), does not call that interpretation into question. As explained, in the scenario described in point 104 above, the ‘invention’ disclosed in the basic patent is A. The fact that, after the filling of the patent application, further research proved the safety and usefulness of the A+B combination should not be taken into account. In addition, as the Hungarian Government recalls, the SPC system was designed to reward not all pharmaceutical research giving rise to the marketing of a new medicinal product, but the kind of research leading to the discovery of, inter alia, new combinations of active ingredients, as understood in point 103 above, namely those displaying a ‘synergistic effect’. (76) By contrast, the (sometimes very relative) innovation consisting in putting in a single pill A and another known drug (B), each having an independent effect on the human body, to facilitate the administration of a combination therapy against a given disease does not deserve that reward. As the Court stated in the judgments in Actavis I and Actavis II, the objective pursued by the SPC Regulation is not to compensate a patent holder for the delay in the marketing of its invention in all its possible commercial forms, including in the form of such single-pill combinations. (77)
108. Thus, it is necessary to separate the first type of combination of active ingredients from the second one. It seems all the more necessary that, as the Court pointed out in paragraph 42 of the judgment in Teva I (with a reference to the judgment in Actavis II), if the patent holder could obtain multiple SPCs on the basis of the same invention (A) in the form of single-pill combinations, it would facilitate ‘evergreening’ strategies, by which pharmaceutical companies could prolong, in an excessive manner, their monopoly by marketing a first medicine containing ‘A’, then a medicine containing ‘A+B’, then another one composed of ‘A+C’, and so on. (78)
109. The ‘invention’ test set out in the judgments in Actavis I and Actavis II and refined in the judgment in Teva I is suitable and proportionate in that regard. Indeed, it ensures the proper balance of allowing the grant of SPCs for combinations that deserve supplementary protection (and thus encourages innovation in that regard), while avoiding multiple SPCs being granted for single active ingredients in a slightly modified, new ‘combination’ package.
110. Conversely, and despite what Merck submits, the ‘identification’ test is ill-suited for such a purpose. Admittedly, as it limits the grant of SPCs to combinations of active ingredients which are expressly mentioned in the claims of the basic patent or are at least ‘specifically identifiable’ therein, that test does restrict the possibility for the patent holder to obtain multiple SPCs for A in single-pill combinations. However, it would only (very) partially contribute to that objective. Indeed, as Teva submits, it would simply create incentives for pharmaceutical companies, when drafting their patent applications, to include therein a standard list of active substances (diuretics, antibiotics, etc.) that may be combined with the substances that are the subject matter of those applications. (79)
111. I am not convinced either by Merck’s argument that such ‘evergreening’ attempts are more theoretical than real because, under Article 13(2) of the SPC Regulation, the protection conferred by any SPC based on the same basic patent is in any event limited to five years after the expiration of that patent.
112. Indeed, on the one hand, as Clonmel submits, given the (economic) interests involved, the possibility for a patent holder to obtain even just a few months of extra protection by filing multiple applications for SPCs for A in single-pill combinations can hardly be regarded as a trivial matter. (80) On the other hand, it is fairly easy for pharmaceutical companies to circumvent the five-year limit set out in Article 13(2) of the SPC Regulation. Indeed, since that limit applies only to SPCs granted on the basis of the same patent, such a company simply has to obtain, at intervals in time, various patents concerning the same active ingredient (for the family of substances including it, for specific substances in the family, for a specific use thereof, and so on), some of which include speculative claims for the use of that active ingredient in combination with other substances, and to file applications for SPCs on the basis of those various patents. (81)
113. My conviction is not shaken either by Merck’s contentions that the ‘invention’ test is complex, tantamount to an ‘inventive-step’ analysis (whereas the SPC system was designed to be ‘simple’),(82) and, as such, generates an uncertainty that could lead to diverging decisions being reached by NPOs on essentially the same facts, contrary to the objective of uniformity pursued under that regulation.
114. Undoubtedly, merely following the ‘identification’ test to determine whether a ‘product’ is ‘protected’ by a patent within the meaning of Article 3(a) of the SPC Regulation would (sometimes) be simpler than applying, in addition, that ‘invention’ test. However, it would lead to outcomes conflicting with the spirit of that instrument, as explained in this Opinion. Furthermore, the legal certainty argument must be relativised. Single active ingredients will easily pass the latter test; it is only with respect to ‘products’ consisting in combinations of active ingredients that further analysis is required and, even in that respect, it seems to me that, save some borderline cases, patent practitioners will be able to predict when those combinations are eligible for an SPC and when they are not.
115. Indeed, the first prong of the test set out in the judgment in Teva I is not as complex as Merck presents it. With respect to the implementation of that test, I recall that the Court specified, in paragraph 48 of that judgment, that a ‘product’ ‘necessarily falls under the invention which is the subject matter of the patent’ where ‘a person skilled in the art can understand without any doubt, on the basis of their general knowledge and in the light of the description and drawings of the invention in the basic patent, that the product to which the claims of the basic patent relate is a [feature] (83) required for the solution of the technical problem disclosed by that patent’.
116. While I admit that the explanation given in the previous point echoes the ‘problem-and-solution’ approach used to assess the existence of an ‘inventive step’ under Article 56 of the EPC, it is, in reality, not a matter of determining whether the combination of A+B meets the patentability requirements. In fact, it is an ex post assessment of what the patent (especially the description) discloses. Does the patent describe, as invention, the use of A+B given their combined, synergistic effect on the human body, to solve a certain technical (medical) problem, so that (following the terms of the Court in the judgment in Teva I) the combination of A and B would be ‘a feature required’ for the solution of that technical problem? Or does the patent rather describe, as the invention, the idea to use certain single substances (including A), given their (individual) effect on the human body, to treat certain diseases or health conditions, while adding that those substances could also be used in combination with other substances (B, C and so on), without disclosing any ‘synergistic effect’ inherent to that combination? In that case, the combination of A+B (or C, and so on) is not a ‘required feature’ of the invention. Although it would be for the referring courts to verify, it seems to me that the main proceedings correspond to the second scenario. (84)
VI. Conclusion
117. In the light of all the foregoing considerations, I propose that the Court answer the question referred by the markkinaoikeus (Market Court, Finland) and the Supreme Court (Ireland) as follows:
(1) Article 3(a) of Regulation (EC) No 469/2009 of the European Parliament and of the Council of 6 May 2009 concerning the supplementary protection certificate for medicinal products
must be interpreted as meaning that to be regarded as ‘protected by a basic patent’ within the meaning of that provision, a ‘product’ must not only (i) be expressly mentioned or at least ‘specifically identifiable’ in the claims but also (ii) fall under the invention which is the subject matter of that patent.
(2) Article 3(c) of Regulation No 469/2009
must be interpreted as meaning that it does not preclude the grant of a supplementary protection certificate (SPC) for a combination of active ingredients where a previous SPC had been granted for one of those ingredients. The concepts of ‘core inventive advance’ and ‘subject matter of the invention’ are irrelevant for the purposes of the assessment of the condition laid down in that provision.