On 10 October 2008 the Appellant (“GSK”) filed application SPC/GB08/046 for a supplementary protection certificate for “an oil in water emulsion comprising squalene, DL-α-tocopherol and polysorbate 80”, an adjuvant known as AS03 which is protected by European Patent (UK) No 0 868 918.

On 18 August 2011 GSK filed application SPC/GB11/043 for a supplementary protection certificate for “an adjuvanted influenza vaccine comprising an influenza virus component which is an influenza virus antigen from an influenza virus strain that is associated with a pandemic outbreak or has the potential to be associated with a pandemic outbreak, wherein the adjuvant is an oil in water emulsion comprising squalene, DL-α-tocopherol and polysorbate 80”, a vaccine comprising an antigen and AS03 which is protected by European Patent (UK) No 1 618 889.

In both applications GSK relied upon marketing authorisation EU/1/08/453/001 for a pre-pandemic influenza vaccine against the H5N1 subtype of influenza A virus marketed by GSK under the trade mark Prepandrix.

By a decision dated 19 December 2012 (BL O/506/12), Dr C.L. Davies, the Deputy Director of the UK Intellectual Property Office, acting on behalf of the Comptroller, decided that neither application was allowable as it stood since AS03 was not an “active ingredient” of Prepandrix, although she was prepared to give GSK an opportunity to amend the applications. GSK has appealed against this decision. It is common ground that the resolution of the appeal depends on an issue of interpretation of Article 1(b) of European Parliament and Council Regulation 469/2009/EC of 6 May 2009 concerning the supplementary protection certificate for medicinal products (codified version) (“the SPC Regulation”) upon which a preliminary ruling from the Court of Justice of the European Union is required.

The SPC Regulation includes the following recitals:

Articles 1, 2 and 3 of the SPC Regulation provide:

The following account is based upon evidence given by Professor Geert Leroux-Roels in an expert report submitted by GSK prior to the hearing before the Deputy Director and upon scientific literature which he exhibited to that report. Professor Leroux-Roels is an experienced and distinguished scientist in the field of immunology. Having held a number of positions at Ghent University between 1983 and 1995, he has been Director of CEVAC – The Centre for Vaccinology – Ghent University Hospital since 1995. He is an author of over 250 publications in the field. His evidence was not tested in cross-examination, but there is no reason to think that it is not technically accurate. Any inaccuracies in the following summary will be attributable to me.

Immunity is the general term used to describe the body’s ability to resist infection by a particular pathogen (a microorganism that typically causes a disease upon infection or in the context of failing immunity). There are, broadly speaking, two main types of immunity: innate immunity and adaptive immunity.

Innate immunity describes the various resistance mechanisms that are first encountered by a pathogen upon infection. These are general and non-specific responses which are present in all individuals at all times and are not affected (nor enhanced) by repeated exposure to a particular pathogen. Aspects of the innate immune response include, for example, anatomical barriers to infection, the “complement system” (the coating of a pathogen by proteins which facilitates its removal by phagocytes, which engulf the coated pathogen) and the responses that are triggered by recognition of a pathogen-associated molecular pattern (PAMP). PAMPs are molecules that are specifically associated with groups of pathogens. They are recognised by receptors, such as Toll-like receptors (TLRs), expressed extra- or intracellularly in antigen-presenting cells (APCs) of the innate immune system. APCs are cells (such as dendritic cells, monocytes and macrophages) that display foreign antigen complexes with major histocompatibility complex (MHC) on their surface. Recognition of PAMPs by TLRs triggers signals, including the secretion of cytokines, which help initiate immune responses.

Adaptive immunity occurs after the innate response and describes the response of antigen-specific lymphocytes to a particular antigen, including the development of immunological memory. Unlike innate immunity, adaptive immunity can be boosted by repeated exposure to a particular antigen. It involves both humoral responses and cellular responses.

The adaptive immune system comprises B and T cells. B and T cells are antigen-specific lymphocytes which are named according to their sites of development in the Bone marrow and Thymus respectively.

B cells recognise antigens via B cell receptors or BCRs, and respond by proliferating and producing and secreting antibodies. B cells are divided into two main classes: conventional B cells (which have highly diverse antigen receptors and are generated in bone-marrow) and B-1 cells (which have less diverse antigen receptors and are self-renewing in peritoneal and pleural cavities).

T cells have highly variable receptors (T cell receptors or TCRs), which recognise complexes consisting of antigens bound to MHC molecules, for example on the surface of APCs of the innate immune system. Facilitator (CD4⁺) or “helper” T cells (e.g. Th1, Th2 and Tfh cells) primarily act by secreting cytokines which stimulate other parts of the immune system, such as macrophages to clear the pathogen. Adaptive effector (CD8⁺) or “killer” T cells secrete cytokines and cytotoxic factors to kill cells infected with pathogens.

Both B and T cells have so-called “memory” cells, which mediate immunological memory and are more sensitive and therefore respond more rapidly upon re-exposure to a particular antigen.

Adaptive immunity is an important aspect of vaccination as it is responsible for long-term immunity to particular pathogens. Over the last two decades an increased understanding of both innate and adaptive immunity and how they interact has led to significant advances in both vaccinology and adjuvant technology.

Vaccination has been one of the most important contributions to public health in the past century. Its immediate aim is the protection of an individual by the generation of a long lasting (ideally permanent) and protective immunity. Its ultimate aim is the eradication of particular diseases. Eradication (or near eradication) has been achieved for several human diseases, most famously for smallpox. Examples of well known diseases for which there are vaccines include rabies, diphtheria, pertussis (whooping cough), tetanus, polio, measles, mumps and rubella.

Vaccines are designed to produce an immune response which, when established as part of an individual’s adaptive immunity, recognises a particular pathogen (or parts thereof) and protects that individual against future infection by that pathogen. Accordingly, vaccination involves striking a balance between inducing a protective immune response whilst at the same time ensuring that no disease or other adverse side effects occur as a result of vaccination.

Generally speaking, vaccines contain three main components:

An antigen, a term which is derived from “antibody-generator”.

An adjuvant, which is usually co-administered with the antigen to stimulate and/or modulate the immune response to the antigen.

An excipient, which primarily relates to the formulation of the antigen and adjuvant in a particular solvent, but which can also convey other properties such as stability and preservation.

In classical vaccination the antigen tends to be well established and characterised and remains unchanged, for example the diphtheria and tetanus toxoid vaccines, which have been in use for over 40 years. Such vaccinations take place at predictable time points, for example, during infant years, adolescence or prior to travel, and therefore supply of these vaccines can be managed to meet what is a predictable demand.

By contrast to the well-established and stable nature of the antigens mentioned above, certain viruses, for example influenza, are prone to variation. In the case of the influenza virus, mutations in the original virus strain can result in the appearance of a variant strain. Different influenza virus strains are named according to their type (e.g. influenza A or B), the species from which the virus was isolated (omitted if human), location of isolate, the number of the isolate, year of isolation, and, in the case of influenza A viruses, the haemagglutinin (H1 to H16)) and neuraminidase (N1 to N9) subtypes. For example the 220^th isolate of an H5N1 subtype influenza A virus in Hong Kong in 1997 is designated: influenza A/chicken/Hong Kong/220/1997 (H5N1). A “clade” (or subclade) describes a group (or subgroup) of related viruses descendent from a common ancestor.

In the case of diseases like influenza, vaccines can be developed in the pre-pandemic phase in order to “prime” the population for an eventual pandemic outbreak. Priming not only achieves initial immune protection, but can also result in a better response to subsequent booster vaccinations.

In the case of influenza, the virus in the eventual pandemic outbreak may be a variant of the virus originally identified in the pre-pandemic phase because of the highly variable nature of the influenza virus. Therefore, ideally, an influenza vaccine developed in the pre-pandemic phase should be able to induce the broadest protection possible. Cross-reactivity (an adaptive immune response that is able to recognise not only the vaccine antigen, but also related variants of the antigen) is therefore a valuable property of such a vaccine.

If a pandemic outbreak occurs, typically large amounts of vaccine product must be readily available within a short period of time to satisfy the global demand. This can be achieved by a vaccine that induces the necessary immune protection with the lowest amount of vaccine possible. So-called “antigen sparing” is therefore another important factor in the context of pandemic vaccines.

Vaccine antigens are selected so that when an immune response is produced against the antigen, it will protect the individual against future infection by particular pathogens.

Many vaccines contain whole pathogens (e.g. viruses or bacteria), which is feasible when suitable versions can be administered to induce the necessary immune response without leading to disease or side effects. The administration of the whole pathogen tends to be more immunogenic and stimulates both innate and adaptive immune responses. This has the advantages of eliciting stronger and potentially broader immune responses and can produce longer lasting protection. However, whole pathogen vaccines have the disadvantages that they can have poor safety profiles and produce adverse side effects.

Where whole pathogen vaccines are not feasible, other approaches are needed and vaccines may comprise inactive split parts or sub-units of pathogens or recombinantly derived antigens. In recent decades there have been significant advances in peptide (and carbohydrate) engineering and recombinant technology that have enabled the large-scale manufacture of highly specific, safe, tolerable and highly purified antigens. However, one of the main challenges to modern vaccines that comprise such antigens is that the antigens may not be sufficiently immunogenic or potent and may induce only a weak immune response.

The concept that the immune response to antigens can be improved by the addition of certain compounds into the vaccine formulation was demonstrated approximately 100 years ago when aluminium salts were first used in vaccine formulations. Such compounds are referred to as “adjuvants” (derived from the Latin word “adjuvare” which means “to help or aid”).

The last few decades have seen extensive research and development in the field of adjuvants. There has been a surge in adjuvant technology which has been assisted by a greater understanding of innate and adaptive immunity and their close interaction at the molecular level in the response to a pathogen.

It is important to be clear as to exactly what an adjuvant does and what it does not do. It is something which aids the process of generating an antigen-specific antibody response, but it does not generate antibodies itself. The main way in which an adjuvant works is to amplify the immune response.

What an adjuvant does is to provide a quantitative change in the immune response which in turn leads to a qualitative change in the immune response. The first thing the adjuvant can do is to stimulate cytokine production, a general immune response that is not specific to any antigen. The second thing the adjuvant can do is to stimulate the antigen presentation of APCs. This leads to stimulation of the adaptive immune response, which leads to a greater generation of antigen-specific antibodies.

These quantitative effects are of value in themselves since they enable antigen-sparing. In addition, however, they lead to a qualitative effect which is greater cross-reactivity. When the body is exposed to an antigen, it normally produces what is described as a polyclonal response. That is to say, it does not just generate one antibody to the antigen, rather it generates a group of different antibodies which will be specific to different parts (epitopes) of the antigen. Within that polyclonal group, a small proportion will bind not only to that antigen, but also to related antigens. That is referred to as cross-reactivity. Because adjuvants cause more antibodies to be produced, this can result in increased cross-reactivity.

Developing a new adjuvant involves considerable research and development. An adjuvant must interact with the immune system to elicit an improved response to the antigen, but equally it must be safe.

AS03 has the following main components:

polysorbate 80 (polyoxy-ethylene sorbitan-20 monooleate) - a surfactant;

DL-α-tocopherol – a vitamin E analogue; and

squalene – an essential intermediate in the biosynthesis of cholesterol, steroid hormones and vitamin D.

AS03 is an oil-in-water (o/w) emulsion adjuvant system, distinguished from other o/w emulsions by the presence of α-tocopherol. By comparing AS03 with an adjuvant formulation in which α-tocopherol was replaced by an equal volume of squalene, α-tocopherol has been shown to significantly and positively influence the kinetics and the magnitude of the cytokine responses in vitro and in vivo, the antigen loading in monocytes, the granulocyte recruitment to the draining lymph nodes, and the magnitude of the antibody response to the model antigen HBsAg. Thus AS03 is superior to previous o/w emulsion adjuvants.

There is good evidence that the presence of AS03 in a pre-pandemic influenza vaccine, namely Prepandrix, has three benefits:

increased immunogenicity, including a high antibody response;

antigen-sparing; and

cross-reactivity.

These benefits are not expected to be specific to the use of AS03 in a pre-pandemic influenza vaccine, but they are of particular advantage in that context.

This vaccine comprises an antigen known as A/Indonesia/05/2005 (H5N1) like strain used (PR8-IBCDC-RG2) together with AS03. Studies have demonstrated that the presence of the AS03 is an important factor in ensuring that the vaccine satisfies the criteria for licensing by the United States Food and Drug Administration and the European Medicines Authority.

The issue arising on this appeal involves consideration of a number of decisions of the CJEU, which I shall take in chronological order.

In Case C-31/03 Pharmacia Italia SpA [2004] ECR I-10001 an SPC application had been made in Germany for the active ingredient cabergoline, which was protected by a basic patent filed in 1981. The application was based on a marketing authorisation for cabergoline granted for the human medicinal product Dostinex in Germany in June 1994. By virtue of the transitional provision contained in Article 19(1) of Regulation 1768/92/EEC (the predecessor of the SPC Regulation), an SPC could only be granted for a product if, on the date the Regulation entered into force, it was protected by a basic patent and “the first authorization to place it on the market as a medicinal product in the Community was obtained after” 1 January 1988. The first authorisation for Dostinex in the Community had been granted in the Netherlands in October 1992, but there had been an earlier authorisation for cabergoline as the active ingredient of a veterinary medicinal product called Galastop granted in Italy in January 1987. In these circumstances the Bundesgerichtshof (Federal Court of Justice) referred the following question to the Court of Justice:

The applicant argued that it was the date of first authorisation to place the product on the market for human use which was relevant, whereas the Commission and the United Kingdom contended that the relevant date was that of the first authorisation to place the product on the market for either human or veterinary use. Advocate General Jacobs advised the Court of Justice to adopt the latter interpretation. In his Opinion he considered Article 3(d) as well as Article 19(1), saying (footnote omitted):

In its judgment the Court of Justice followed the Advocate General’s advice, holding:

In Case C-431/04 Massachusetts Institute of Technology [2006] ECR I-4089 the applicant had applied in Germany for an SPC for the product carmustine, either in combination with a polymeric biodegradable matrix called polifeprosan or alternatively on its own. The applicant relied on a marketing authorisation for the medicinal product Gliadel, which was used for the treatment of human brain tumours. Gliadel comprised carmustine as its active ingredient and polifeprosan as an excipient. Carmustine was a highly cytotoxic substance which was already covered by an earlier marketing authorisation for the treatment of brain tumours with inert excipients. Polifeprosan was a new excipient that enabled the slow release of carmustine from a disc implanted in the cranium after surgical removal of the tumor, thereby permitting the delivery of a higher but constant dose of carmustine. Accordingly to the applicant, the combined use of carmustine and polifeprosan extended the life expectancy of patients by several months. Polifeprosan was the subject of a patent which the applicant relied on as the basic patent for the application.

The application was refused by the Deutches Patent- under Markenamt (German Patent and Trade Mark Office) on the basis that (i) no SPC could be granted for the combination of carmustine and prolifeprosan since that was not a combination of active ingredients within the meaning of Article 1(b); and (ii) no SPC could be granted for carmustine on its own since the marketing authorization relied on was not the first authorization to market carmustine contrary to Article 3(d). As I understand it, the applicant appealed against holding (i), but not holding (ii). In these circumstances the Bundesgerichtshof referred the following questions to the Court of Justice:

Advocate General Léger advised the Court of Justice to rule that Article 1(b) should be interpreted as including a combination of an active ingredient (such as carmustine) with an excipient which is necessary for the therapeutic efficacy of the active ingredient (such as polifeprosan). He did so on the basis of the kind of teleological approach to interpretation contended for by GSK in the present case, saying that this was just the kind of costly innovation that the Regulation was designed to protect. Nevertheless the Court of Justice declined to follow that advice, and ruled that such a combination was not within the scope of Article 1(b) on its proper interpretation.

The following passages in the Court’s judgment are particularly relevant to the present issue:

The Court of Justice’s reasoning in these passages can be summarised as follows:

the expression “active ingredient” was generally accepted not to include substances forming part of a medicinal product which did not have an effect (which in context must mean a therapeutic effect) of their own on the human or animal body;

this understanding was consistent with paragraph 11 of the Explanatory Memorandum in the Commission’s original Proposal for the Regulation, indeed it was apparent from the Memorandum that “product” was understood to mean an “active ingredient” in the strict sense;

accordingly, a substance which did not have any therapeutic effect of its own and which was used to obtain a certain pharmaceutical form of the medicinal product was not an “active ingredient”;

therefore an alliance of such a substance with one that did have a therapeutic effect of its own was not a “combination of active ingredients”;

it was immaterial that the substance without any therapeutic effect of its own rendered possible a pharmaceutical form of the medicinal product necessary for the therapeutic efficacy of the substance which did have a therapeutic effect;

indeed, a test which involved considering whether a substance without any therapeutic effect of its own rendered possible a pharmaceutical form of the medicinal product necessary for the therapeutic efficacy of another substance which did have a therapeutic effect would create legal uncertainty and inhibit the attainment of a uniform solution at Community level.

Although the Court did not refer to its earlier judgment in Pharmacia, and Pharmacia did not dictate the decision in MIT, the Court’s decision in MIT was consistent with the earlier decision.

In Case C-202/05 Yissum Research and Development Company of the Hebrew University of Jerusalem v Comptroller-General of Patents [2007] ECR I-2839 the applicant applied for an SPC for the product calcitriol either alone or in combination with an ointment base. The applicant relied upon (i) a second medical use patent, the claims of which were directed to the use of calcitriol in topical treatment of skin disorders including psoriasis, and (ii) a marketing authorization for Silkis ointment, which comprised calcitriol as the active ingredient with various carriers and was authorised for the topical treatment of psoriasis. The application was refused by the Comptroller because there were two earlier marketing authorisations for medicinal products containing calcitriol as the active ingredient, namely Calcijex and Rocaltrol. Calcijex was authorised for the management of hypocalcaemia in patients undergoing dialysis for chronic renal failure. Rocaltrol was authorised for administration to patients with chronic renal failure or post-menopausal osteoporosis.

On the applicant’s appeal, I referred three questions to the Court of Justice ([2004] EWHC 2880 (Pat)]). Two of those were subsequently withdrawn in the light of the Court’s judgment in MIT. The remaining question was as follows:

The Court of Justice gave its answer to that question by reasoned order on the basis that the answer to it could be clearly deduced from the existing case law, and in particular Pharmacia and MIT. In its order the Court held:

It can be seen from this that the Court of Justice confirmed that the concept of “product” in Article 1(b) of the Regulation was to be interpreted strictly and could not include the therapeutic use of the active ingredient or even whether the medicinal product was for human use or for veterinary use.

In Case C-130/11 Neurim Pharmaceuticals (1991) Ltd v Comptroller-General of Patents [2012] ECR I-0000 melatonin, a naturally occurring hormone, had been marketed by Hoechst under the trade mark Regulin for regulating the seasonal breeding activity of sheep pursuant to a patent applied for in 1987 and a marketing authorisation granted in 2001 (“the Regulin MA”). Neurim marketed melatonin under the trade mark Circadin for the treatment of insomnia in humans pursuant to a patent applied for in 1992 and a marketing authorisation granted on 28 June 2007 (“the Circadin MA”). Neurim applied to the UK Intellectual Property Office for an SPC in respect of Circadin. The IPO refused the application on the ground that it did not comply with Article 3(d) of the SPC Regulation since the Circadin MA was not the first authorisation to place melatonin on the market, the Regulin MA was. I dismissed Neurim’s appeal ([2010] RPC 22, [2010] RPC 22), but the Court of Appeal referred five questions to the Court of Justice ([2011] EWCA Civ 228, [2011] RPC 19).

The Court interpreted the first and third questions as asking, in essence, whether Articles 3 and 4 of the SPC Regulation were to be interpreted as meaning that the existence of an earlier marketing authorisation for a veterinary medicinal product precluded the grant of an SPC for a different use of the same product. It answered those questions as follows:

As I observed in AstraZeneca AB v Comptroller-General of Patents, Trade Marks and Designs [2012] EWHC 2840 (Pat) at [52]-[53], the Court’s judgment in Neurim is problematic for two reasons.

First, it appears that the Court was intending to depart from its earlier judgments and order in Pharmacia, MIT and Yissum. This is not clear, however, since it did not refer to those decisions. Thus one does not know if those decisions are to be regarded as having been overruled, or as qualified in some unspecified manner.

Secondly, it does not appear that the Court was intending to depart from its earlier judgments in Case C-195/09 Synthon BV v Merz Pharma GmbH & Co KGaA [2011] ECR I-0000, [2012] RPC 3 and Case C-427/09 Generics (UK) Ltd v Synaptech Inc [2011] ECR I-0000, [2012] RPC4, since it cited Synthon at [20]. It is not clear to me, however, how Neurim is to be reconciled with those decisions. The reasoning which the Court relied on in Neurim, namely that the research required to obtain a patent and marketing authorisation for a second medical use of an active ingredient justifies the grant of an SPC for the second medical use despite the fact that the same active ingredient has already been lawfully marketed as a medicinal product, seems to me to be equally applicable to Generics and Synthon.

Case C-11/13 Bayer CropScience is a recent reference from the German Bundespatentgericht (Federal Patent Court) to the CJEU. In this case Bayer applied for an SPC for isoxadifen. Isoxadifen is a so-called “safener”, a substance which is added to herbicides and other plant protection products to eliminate or reduce their phytotoxic effects on certain plants. The safener works in combination with the herbicide to allow beneficial application of the herbicide. Without the safener, in many cases the herbicide would destroy the crop in addition to the weed or pest. The question referred is whether a safener is an “active substance” within the meaning of Article 1(3) and 1(8) of European Parliament and Council Regulation 1610/96/EC of 23 July 1996 concerning the creation of a supplementary protection certificate for plant protection products (“the Plant SPC Regulation”), and hence eligible for an SPC as a product under Article 3(1).

The Federal Patent Court has set out a number of competing considerations in its reference, while indicating its own provisional view in favour of refusing the application.

First, the language of the Plant SPC Regulation supports the view that a safener is an active substance. In particular, “active substances” are defined under Article 1(3) of the Plant SPC Regulation as “substances or microorganisms including viruses, having general or specific action”. Furthermore, “plant protection products” are broadly defined in Article 1(1) as products intended inter alia to protect plants against all harmful organisms or to influence the life processes of plants (other than as a nutrient).

Secondly, however, MIT decided that a substance that is used to obtain a pharmaceutical form of a medicinal product is not covered by the concept of an “active ingredient” under the SPC Regulation, even if it renders possible the therapeutic efficacy of the substance in the medicinal product which does have therapeutic effects. This supports a strict interpretation of “active substance” which is limited to substances which have a direct therapeutic effect or plant protection action of their own. It also indicates that merely because the safener is an integral part of the plant protection product, it does not thereby necessarily constitute an active substance.

The Bundespatentgericht recognises, however, that a factual distinction can be drawn between MIT and the facts before it, since the mechanism of action of a safener is different from that of a biodegradable matrix. In particular, a safener appears to have a more complex mechanism of action which is of more direct effect.

Furthermore, by contrast, in Case C-420/10 Söll [2012] ECR I-0000 the CJEU ruled that the concept of “biocidal products” under Article 2(1)(a) of European Parliament and Council Directive 98/8/EC of 16 February 1998 concerning the placing of biocidal products on the market (“the Biocidal Directive”) follows from the concept of “active substance” as defined in Article 2(1)(d) of that Directive and includes indirect action.

Thirdly, European Parliament and Council Regulation 1107/2009/EC of 21 October 2009 concerning the placing of plant protection products on the market, which replaces Directive 91/414/EEC, clearly distinguishes between the terms “active substance” and “safener”. It indicates that active substances have general action against harmful organisms or on plants, whereas safeners are added to the plant protection product to influence its effects.

It is reasonable to consider that “active substance” should have the same meaning under the Plant SPC Regulation. In particular:

Paragraph 66 of the Explanatory Memorandum to the Plant SPC Regulation states expressly that the term “active substances” is taken from Directive 91/414/EEC.

There is repeated reference to Directive 91/414/EEC (now to be read as Regulation 1107/2009/EC) in the Plant SPC Regulation.

There is a clear link or connection between the plant authorisation legislation and the Plant SPC Regulation.

When Directive 91/441/EEC was repealed and replaced with Regulation 1107/2009/EC, the Plant SPC Regulation was not amended and the definition of “active substance” under Article 1(3) remains unchanged.

Fourthly, the substantive requirements for an authorisation of a safener are largely the same as those for an active substance. It would be consistent with the objectives of the Plant SPC Regulation to grant a certificate to compensate for the delay (and resulting reduced period of patent protection) caused by such authorisation requirements.

Finally, the lack of searchability of safeners, which do not appear on any databases of active substances, suggests that they are not intended to be treated in the same way.

The Deputy Director decided that, applying the reasoning of the Court of Justice in MIT, AS03 was not an “active ingredient” within the meaning of Article 1(b) of the SPC Regulation whether in its own right or in combination with the antigen contained in Prepandrix because AS03 did not have a therapeutic effect on its own. AS03 did not itself confer any immunity, whether against influenza or any other condition. The fact that AS03 enhanced the therapeutic effect of the antigen, irrespective of the actual antigen involved and the immunological protection sought, was not sufficient.

GSK contends that the Deputy Director’s decision was wrong. Counsel for GSK argued in summary as follows.

First, he relied upon various paragraphs of the Explanatory Memorandum to the Commission’s Proposal for Regulation 1768/92/EEC, and in particular paragraph 11, as showing that it was intended to apply to all new products which were the subject of innovative research save only for minor variants such as a new dose, the use of a different salt or ester or a different pharmaceutical formulation.

Secondly, he relied upon the primary purpose of the SPC Regulation, which was described by the Court of Justice in Case C-482/07 AHP Manufacturing BV v Bureau voor de Industriële Eigendom [2009] ECR I-7295 at [30] as follows:

In this regard, counsel for GSK argued that giving effect to this purpose had been key to the Court’s reasoning in Neurim and that that decision showed that Article 1(b) should not be narrowly interpreted.

Thirdly, he argued that the present case was to be distinguished from MIT. In this regard, he argued that MIT was concerned with an excipient which had no physiological effect on the body, whereas the present case was concerned with an adjuvant which did have physiological effects on the body and which thereby enhanced the therapeutic effect of the antigen.

The Comptroller supports the Deputy Director’s decision. Counsel for the Comptroller argued in summary as follows.

First, she relied on the language of the SPC Regulation. In this regard she argued that the SPC Regulation makes a clear distinction between a “medicinal product” and a “product” (compare Article 1(a) with Article 1(b)). The former may have an effect on a physiological function, but it is only the latter which can be the subject of an SPC. As a consequence, although an adjuvant may fall within the definition of “medicinal product”, it does not mean that that the adjuvant is an “active ingredient” or “product” within the meaning of the SPC Regulation. If a substance with general and indirect physiological effects - which could be positive or negative or might or might not be related to the indication which was the subject of the marketing authorisation - also fell within the concept of “active ingredient”, then it would result in a definition which is too broad and uncertain, leading to divergent results in different Member States.

Secondly, she relied on the Court of Justice’s decision in MIT. In this regard, she disputed that the present case could be distinguished from MIT. She pointed out that the Court’s judgment made barely any reference to excipients, and that neither the ruling nor the reasoning was confined to excipients. She argued that the ruling and the reasoning were directly applicable to the present case. She also pointed out that in that case the Court had been considering a situation in which the presence of the polifeprosan was necessary to the therapeutic efficacy of the medicinal product. She argued that, if inventive merit was the only consideration, the Court would have reached the opposite conclusion.

Thirdly, she relied on European Parliament and Council Directive 2001/83/EEC of 6 November 2001 on the Community Code relating to medicinal products. In this regard, she pointed out Directive 2001/83/EEC draws a clear distinction between “active substances” on the one hand and “excipients” on the other. “Active substance” is defined in Article 1(3a) as:

Furthermore, the Directive expressly treats an adjuvant as a type of excipient. Thus Annex 1, Part I, Module 3 to the Directive provides at section 3.2.2.1 as follows (emphasis added):

Counsel for the Comptroller argued that it is reasonable to consider that “active ingredient” should have the same meaning in the SPC Regulation, for at least the following reasons:

There is repeated reference to Directive 2001/83/EC in the SPC Regulation.

The SPC Regulation sits at the interface between the regulatory and patent systems, and serves to compensate for the delay involved in obtaining marketing authorisations and the resulting effective loss in patent protection. As a result, the protection is strictly limited to the product which has obtained authorisation to be placed on the market as a medicinal product under the Directive (see Recital (10) and Article 3(b)).

It would be perverse if a substance was not considered an active substance for the purposes of regulatory approval, but was nevertheless considered an active ingredient for the purposes of SPC protection.

Although the Directive has been amended on several occasions and the SPC Regulation has been codified, no amendments have been made to the definition in Article 1(b) or otherwise to indicate that the distinction that exists between active ingredients and adjuvants within the regulatory framework would not also apply to the SPC Regulation.

In addition, counsel for the Comptroller pointed out that the then European Agency for the Evaluation of Medicinal Products’ Note for Guidance on Requirements for the Evaluation of New Adjuvants in Vaccines dated 25 April 2002 stated that “because the adjuvant is not the active ingredient, it is an individual vaccine/adjuvant combination which will be licensed”.

Fourthly, she relied upon the other purposes of the SPC Regulation, reflected in recitals (7)-(10), which the Court of Justice described in AHP as follows:

In this regard counsel for the Comptroller argued that in order to give effect to these purposes it was necessary to interpret Article 1(b) strictly, as the Court of Justice had done in MIT.

As noted above, the parties are agreed that this issue should be referred to the Court of Justice for a preliminary ruling. I also agree that this is necessary, for three reasons. First, in the light of the arguments summarised above, I do not consider that the answer is acte clair. Secondly, it is clear that there is a divergence of interpretation between the different national authorities on this point. Thus the patent offices in Austria, Italy, Luxembourg and Spain have granted SPCs in respect of both AS03 and the combination of the antigen and AS03 based on the 918 patent, while the Portuguese patent office has refused an SPC in respect of AS03 alone but granted one in respect of the combination and the Swedish patent office, like the Deputy Director, has refused both applications. In addition, the patent offices in Cyprus and Italy have granted SPCs for the combination based on the 889 patent. Thirdly, the present case raises a similar issue to the issue in Bayer CropScience.

Accordingly, I shall refer the following questions to the Court of Justice:

In case it assists, I shall state my own view on these questions. While I acknowledge the force of counsel for GSK’s argument that the primary purpose of the SPC Regulation is to reward innovative research of the kind that led to the development of AS03 and Prepandrix, I find the arguments presented by counsel for the Comptroller more convincing. In particular, I consider that the SPC Regulation was intended to provide a simple and predictable system that could be operated by the competent authorities of the Member States, and in particular the national patent offices, in a uniform manner. To achieve those objectives, it is necessary to have bright-line rules. Article 1(b) is such a rule. In my view the Court of Justice was correct to hold in Pharmacia, MIT and Yissum that it should be strictly interpreted. The result of a strict interpretation is to deny extended protection for what may well be meritorious inventions, but the price of not adopting a strict interpretation is a level of uncertainty and inconsistency which in my opinion is unacceptable. Accordingly, I would answer both questions no.

It is worth adding that, as counsel for the Comptroller pointed out, GSK may be able to obtain an SPC for the antigen contained in Prepandrix based on the 889 patent and the marketing authorisation for Prepandrix. Thus refusal of the appeal would not necessarily leave GSK bereft of protection.

Finally, I would observe that this is the third time in six months that I have had to refer questions of interpretation of the SPC Regulation to the CJEU. I do so with considerable regret. That this should be necessary demonstrates the dysfunctional state of the SPC system at present. This is primarily due to the poor drafting of the SPC Regulation and to the failure of the European Commission, Council and Parliament to revise it to address the problems which have emerged. Matters have not been assisted, however, by the fact that the Court of Justice’s recent case law interpreting the SPC Regulation has not provided the level of clarity and consistency that is required.