Judgments - Generics (UK) Limited and others (Appellants) v H Lundbeck A/S (Respondents)

(back to preceding text)

49.  Mr Simon Thorley QC in his conspicuously clear submissions for Generics (UK) Ltd. and the other appellants in this appeal submitted that the passages from Lord Hoffmann’s speech to which I have referred in paragraph 47 above represent the ratio of Biogen Inc. v. Medeva plc. He reinforces this with a further submission that that case was, like the present, concerned with a simple claim to a novel product. In his submissions for Medeva plc in Biogen Inc. v. Medeva plc, Mr Thorley did indeed refer to claim 1 as a product claim, but both the opposing submissions for Biogen Inc. and the speeches in this House appear to me to have preferred to describe the position in more complex terms. At p.40 in particular, Lord Hoffmann said that “The claim is to a product, a molecule identified partly by the way in which it has been made (“recombinant DNA”) and partly by what it does (the words following “characterised by”). It generalises what Professor Murray had done in two ways. First, as to the results he had achieved. ….. Secondly, there was generalisation of the method which he had used. ……. The claim was for any method of making a DNA molecule which would achieve the necessary expression".

50.  As to Mr Thorley’s submission that the claim in Biogen Inc. v. Medeva plc was as a matter of factual analysis related to a novel product, the recombinant DNA molecule, which could be patented as such, Mr Waugh QC for Lundbeck challenged this, and submitted with reference to an article by Summers summarised by Aldous J at first instance - [1995] RPC 25, 57 lines 39-47 - that Professor Murray had invented nothing new when he made such a molecule, and that his only invention consisted in the development of the notion that such a molecule could be used to create the HBV antigens, which would in turn cause a patient’s immune system to produce the desired antibodies. I see considerable force in Mr Waugh’s submission on this point, but perhaps even more relevant in my view is the fact that nowhere in Biogen Inc. v. Medeva plc do the speeches treat or discuss the claim as a simple claim in respect of a novel product or address the issue that would on that basis arise, as to whether such a claim can or should be restricted in scope by reference to the inventive step involved in its creation.

51.  This is most apparent when one looks at the passages in Lord Hoffmann’s speech at pp.48-49, where he corrects an error which had crept into the first instance decisions of Mölnlycke AB v. Procter & Gamble Ltd. [1992] FSR 549, Chiron Corp. v. Organon Teknika Ltd. [1994] FSR 202 and Biogen Inc. v. Medeva plc itself. The error was to treat an invention consisting of a product as sufficiently disclosed for the purposes of s.14(3) if the description and specification enabled a skilled person to make a single embodiment, rather than to perform it across its full width or to its full extent as was, Lord Hoffmann held, the correct approach. It is particularly noticeable that, while in the Mölnlycke case issue was joined as to whether the disclosure must enable “all possible embodiments” or whether (as Morritt J there held) only one would suffice (pp.595 and 600), it was “not in dispute that under the Patents Act 1977 it is not a ground of revocation if the specification fails to disclose the best method of performing the invention” (p.600). Admittedly, Morritt J had (at pp.594-5) held non-compliance with s.14(5)(c) to be no basis for revocation under s.72 (a conclusion not accepted in this House in Biogen Inc. v. Medeva plc), but that is a separate point. What the description discloses must under s.14(5)(c), read with s.14(3), enable a skilled person to make the patented product across its full width or to its full extent. This does not mean that it must also enable the skilled person to make it by all possible methods.

52.  It seems to me therefore that the Court of Appeal was not in the present case bound by the reasoning or result in Biogen Inc. v. Medeva plc to arrive at a conclusion that the present claims 1 and 3 were invalid in so far as they extended in scope to any method of making escitalopram other than that devised by Lundbeck. The question whether this was the correct approach in principle was open to the Court to consider and determine for itself. Lord Hoffmann in paras. 26 and 42 of his judgment in the present case expressed understanding for and sympathy with the judge’s instinctive reaction to the inherent breadth of a product claim. Kitchin J’s reaction was, as he himself made very clear, based on the fact that there was nothing inventive about the idea that escitalopram might have valuable properties by itself or therefore about the aim of separating, preparing or testing it. This is not a case where someone found or made a substance which no-one had previously thought of or thought would have any value. The only inventive step was the means by which Lundbeck managed to separate or prepare escitalopram; and this involved no general or common principle of which the appellants made any use when they claimed to produce and market the (+) enantioner in competition with escitalopram. One reaction to these circumstances might be that it is surprising that the product escitalopram should be regarded as in any relevant sense novel (or in the language of s.1(1)(a) new) at all. But that reaction is precluded by the definition of novelty accepted in Synthon. Hence, the stark issue identified at the outset of this speech. Is patent protection in respect of a new product qualified where the only inventive step involved in making the product available consists in the method by which it is made available, and its description and specification disclose only that inventive method and superior methods are found by others which owe nothing to that method?

53.  As Lord Hoffmann and Jacob LJ observed in their judgments, both the Convention and the Patents Act 1977 distinguish between an invention consisting in a product and an invention consisting in a process (see e.g. articles 52-57, 167 and s.60). As I have said, the significance of the distinction does not appear to me to be spelled out in a manner which resolves the present issue unequivocally. But it can at least be said to be surprising if so significant a qualification exists with regard to the protection available to a product as opposed to a process that there is no positive trace of it in the Convention or statutory language. S.60 of the Act, providing that there is infringement if, where the invention is a product, a person “makes, disposes of, offers to dispose of, uses or imports the product or keeps it whether for disposal or otherwise” would also fit uneasily with an approach according to which the patent would be invalid if the product was made by a method owing nothing to the inventive step. The appellants rely on article 83 and s.72(1)(c) as involving a requirement that not merely the invention in all its embodiments, but also all the methods of making it sought to be protected, should be sufficiently disclosed by its description and specification. Under s.125 the words “patented invention” mean an invention for which a patent is granted, unless the context otherwise requires, and in Pharmacia Corp. v. Merck & Co. Inc. [2001] EWCA Civ 1610; [2002] RPC 41, para. 55, Aldous LJ held that the context otherwise required in respect of s.72(1)(c) and that “There the use of the word “invention” must include the technical contribution which supports the monopoly claimed, with the result that those sections require an enabling disclosure of that technical contribution". However, in illustrating this, with reference to the House’s decision in Biogen, he took only as examples a claimed invention consisting of “a class of compounds", or “a selection of certain compounds", making it clear that he was speaking of claims covering different products or different embodiments of a single invention, rather than expressing any different view, about the sufficiency of a single method of achieving a patented product, to that which had been common ground before Morritt J in Mölnlycke (para. 51 above).

54.  Apart from the submission (which I have rejected) that the appeal must succeed on the basis of Biogen, the appellants’ approach does not therefore in my opinion find direct support in either the statutory language or any United Kingdom authority, whatever may be the principled arguments that can be advanced in its favour. I would find this alone surprising, if the appellants’ approach is a good one. Further, I can foresee that the appellants’ approach, however principled, could well add in practice to the issues which may arise as to the validity or proper scope of patent claims to what under Synthon are novel products prepared by inventive methods. Finally, and in my view conclusively, this is, as Lord Hoffmann underlined in the Court of Appeal, an area where there is clear jurisprudence of the European Patent Office and of its Boards of Appeal: Kawasaki Steel Corp. [1994] OJEPO 695; T 0233/93 E I Du Pont (28 October 1996); T 1195/00 Alcan International Ltd. (24 May 2004); and T 0803/01 Novartis AG (9 September 2003). It is true that in each of these cases the issue was one of obviousness, and in none was an objection of insufficiency raised. But that, as Lord Hoffmann said, is itself very significant.

55.  The Board of Appeal in the Kawasaki Steel case concluded that “a product which can be envisaged as such with all characteristics determining its identity together with its properties in use, i.e. an otherwise obvious entity, may become nevertheless non-obvious and claimable as such if there is no known way or applicable (analogy) method in the art to make it and the claimed methods for its preparation are therefore the first to achieve this in an inventive manner". It could not sensibly have given such unequivocal endorsement to the patentability of a product in such circumstances, had it envisaged that the patent would be liable to revocation in so far as it purported to cover other methods owing nothing to the inventive method(s) described in the claim. The passage quoted by Lord Hoffmann at p.49 in Biogen from the Board of Appeal’s decision in Exxon/Fuel Oils has never been applied to a simple product claim such as the present, and a reading of the full text from which it is taken shows that it too was dealing with a situation where the description did not support all the inventions or all the embodiments of the invention in respect of which the patent claim was made.

56.  I would add that Mr Waugh sought to gain further support from the fact that in American and Australian litigation about escitalopram no suggestion of insufficiency has been raised. I do not myself find it profitable to try to assess why parties have not raised arguments of law in jurisdictions which have evidently different patent legislation and case-law to our own. The Australian High Court does not even accept the correctness of the conclusion in Biogen that the description and specification must amount to an enabling disclosure across the full width (and not merely in relation to one among other embodiments) of the invention: the amusing comments on Biogen made in para. 67 of the judgment of Gleeson CJ, McHugh, Gummow, Hayne and Heydon JJ in Lockwood Security Products Pty. Ltd. v. Doric Products Pty. Ltd. [2004] HCA 58 stress the independence of Australian from United Kingdom patent law and show that there is very little scope to argue any point at all on insufficiency in Australia.

57.  For the reasons I have given in paragraphs 41 to 55, I would however dismiss this appeal.


My Lords,

58.  This appeal raises a point of principle relating to product claims in patents, and it also requires consideration of the ambit of the reasoning of this House in Biogen Inc v Medeva plc [1997] RPC 1.

The factual and technical background

59.  1-(3-dimethylaminopropyl)-1-(4'-fluorophenyl)-1,3-dihydroisobenzofuran-5-carbonitrile is an organic compound which was first synthesised by the respondent, H. Lundbeck A/S, in 1972. The compound, whose generic name is citalopram, was patented by the respondent, but that patent expired many years ago. Citalopram was found to have a significant antidepressant effect, by virtue of its action as a selective serotonin reuptake inhibitor (“SSRI”) in the brain. It was launched on the market in 1989, under various brand names, and has proved very successful. At least in terms of volume, it is currently the world’s highest selling branded antidepressant.

60.  As manufactured in accordance with the teaching of this patent, citalopram was known to be a racemate. In other words, it existed as a mixture of two types of molecule, known as enantiomers, in equal proportions. Enantiomers have precisely the same chemical formula, and precisely the same three-dimensional, stereochemical, structure, save that one is the mirror-image of the other. Many organic compounds which have therapeutic or other effects have enantiomers and exist as a racemate, either because it is how they are formed in nature, or (as in the case of citalopram) because it is the form in which they are manufactured. Each of the two enantiomers can be conventionally distinguished from the other in one of three different ways, namely (i) by a prefix of (+) and (- ), which is based on the direction in which it rotates polarised light, (ii) by a prefix of D or L which is based on the chemical glyceraldehyde, which itself exists in two enantiomeric forms, and (iii) by a prefix of R or S, which depends on priority rules, the detail of which is not necessary to explain.

61.  It has long been known that two enantiomers can have different properties from each other. Thus, where a particular racemate has a therapeutic effect, it may transpire that the effect is attributable more to one enantiomer than to the other, or that one of the enantiomers has a toxic, or other, side-effect which is not shared by the other. The only way in which it is possible to tell whether the effects of the two enantiomers of a particular racemate differ, and if so how, is by obtaining relatively pure forms of each enantiomer and comparing them experimentally. At any rate as yet, it is impossible to predict such differences in advance.

62.  Accordingly, the notion of obtaining pure forms of each enantiomer of a racemate, which has a therapeutic, or other beneficial, effect, is obvious. However, the ease with which one can obtain relatively pure forms of each (or either) enantiomer varies from one racemate to another. In the case of citalopram, it proved particularly difficult. The respondent appears to have taken seven years of hard work, between 1980 and 1987, to manufacture a relatively pure form of each of the two enantiomers of citalopram. This was achieved by finding a way of separating, or resolving, the two enantiomers of a diol (which was one of the intermediate substances in the manufacture of citalopram) and then proceeding separately with the manufacture of each of the enantiomers of citalopram.

63.  Having obtained separate samples of the two enantiomers, the respondent then carried out experiments to compare them. These experiments established that virtually the whole of the therapeutic effect of citalopram as an SSRI lay in the (+)-enantiomer, which has the generic name escitalopram. Subsequent research has now established that the (-)-enantiomer actually inhibits the therapeutic effect of the (+)-enantiomer.

64.  Having isolated the (+)-enantiomer and established that it was a substantially more effective SSRI than the racemate, the respondent applied for, and obtained, the patent the subject of the present appeal, EP (UK) 0,347,066. This patent (which I shall call “the Patent”) has a priority date of 14 June 1988, and primarily claims escitalopram, i.e. (+)-1-(3-dimethylaminopropyl)-1-(4'-fluorophenyl)-1,3-dihydro- isobenzofuran-5-carbonitrile. As my noble and learned friend Lord Walker of Gestingthorpe explains, the Patent also makes other claims, including the process by which escitalopram had been manufactured. Esciptalopram has been successfully marketed under the brand name Cipralex since 2002.

These proceedings

65.  The three appellants, Generics (UK) Ltd, Arrow Generics Ltd, and Teva UK Ltd and Teva Pharmaceutical Industries Ltd, all wish to market generic citalopram (i.e. the racemate), and they started these proceedings in 2005, claiming revocation of the Patent. The claims came before Kitchin J, who, in a clear and full judgment, decided that, although the attacks based on lack of novelty and on obviousness failed, the attack on the Patent’s claim to the (+)-enantiomer succeeded on the ground of insufficiency: see [2007] EWHC 1040 (Pat), [2007] RPC 32. The Court of Appeal upheld the Judge’s conclusions on lack of novelty and obviousness, but reversed his decision on insufficiency: see [2008] EWCA Civ 311, [2008] RPC 19. Accordingly, they concluded that the main claim of the Patent, namely the claim to the (+)-enantiomer, was valid.

66.  The only point in issue on this appeal is that on which the Court of Appeal disagreed with Kitchin J, namely whether the claim to the (+)-enantiomer was insufficient. Accordingly, we are proceeding on the basis that the enantiomer is a new product, in particular as against the racemate, and on the basis that the obtaining of the enantiomer was not obvious. The question is whether the claim fails on the ground of insufficiency.

The statutory framework

67.  As Lord Walker observed in Synthon BV v SmithKline Beecham plc [2005] UKHL 59, [2006] RPC 10, para 57, “[t]he law of patents is wholly statutory", although “the courts have shown an inclination to enrich the bare simplicity of the stautory text with their own explanatory commentary", which “has over the years done much to clarify the abstract generalities of the statutes and to secure uniformity in their application". Nonetheless, as he went on to say in the following paragraph, “it is salutary to be reminded, from time to time, that the general concepts which are the common currency of patent lawyers are founded on a statutory text, and cannot have any other firm foundation.”

68.  The current statute governing the validity of patents is, of course, the Patents Act 1977, which must be read together with the European Patent Convention (“the EPC”). Indeed, all the provisions of the 1977 Act of central relevance for present purposes have been specifically framed “as nearly as practicable” to have “the same effects in the United Kingdom as the corresponding provisions of the [EPC] have in the territories to which [it applies]": see section 130(7) of the 1977 Act.

69.  The distinction between product claims and process claims, which is, as Lord Walker says, at the heart of this appeal, is effectively taken for granted in the 1977 Act, but it is implicit in section 60 which is concerned with infringement. It specifically refers to cases “where the invention is a product” and to cases “where the invention is a process". As one would expect the same concepts are referred to in the EPC - see arts 52 to 57 and 167.

70.  Section 1(1) of the 1977 Act (reflecting art 52 of the EPC) provides that a “patent may be granted only” if the invention it claims satisfies four requirements. Those requirements are that it “(a) … is new", (b)… involves an inventive step", “(c) … is capable of industrial application", and (d) … is not excluded by subsections (2) and (3)…". There has never been any suggestion by the appellants that paras (c) or (d) apply in this case; and they no longer seek to rely on paras (a) and (b), now that Kitchin J has concluded that, as at June 1988, escitalopram was both new and inventive (the antithesis of obvious), and the Court of Appeal has upheld those conclusions.

71.  Furthermore, it is not, and could not be, suggested by the appellants that a patent cannot be granted for a specified molecule or specified molecules, or a substance comprising specified molecules. Such a suggestion would be completely inconsistent with what has been universally assumed by patent law experts, the U.K. and European Patent Offices and the courts; it would also undermine an enormous number of patents (many of which have been unsuccessfully challenged) granted under the 1977 Act. Indeed, such a suggestion would be inconsistent with the reasoning in the very case on which the appellant primarily relies, namely Biogen [1997] RPC 1. Any such suggestion would also be inconsistent with the statutory history set out at [2008] RPC 19, paras 43 to 46 by Lord Hoffmann in the Court of Appeal in this case.

72.  That is not, of course, the end of the issue in this case, as section 1(1) is effectively negative in nature, and the fact that product claims can extend to specific molecules does not mean that the product claim to escitalopram in this case is valid. However, it demonstrates that the appellants have to look elsewhere to find grounds for establishing the invalidity of the Patent’s principal claim. Now that their arguments on lack of novelty and obviousness have been disposed of, the appellants’ case is, as mentioned, based on an allegation of insufficiency.

73.  However, at any rate on the face of it, that allegation is not assisted by the statutory provision concerned with that topic, namely section 14 of the 1977 Act. Section 14(3) (reflecting art 83 of the EPC) requires the specification of a patent to “disclose the invention in a manner which is clear enough and complete enough for the invention to be performed by a person skilled in the art". And section 14(5)(c) (reflecting art 84 of the EPC) requires any claim to “be supported by the description". There is no suggestion that the description in the Patent does not enable a person equipped with the appropriate degree of skill and knowledge to manufacture escitalopram. Nor is there any suggestion that the description does not make good the contention that escitalopram has substantially enhanced therapeutic effectiveness as an SSRI over the prior art, namely citalopram.

74.  Of course, sections 1 and 14 are concerned with the grant of patents, whereas it is section 72 (reflecting art 100 of the EPC) which is concerned with the revocation of patents, and which is therefore the section directly in point on this appeal. Section 72(1) provides that a patent can “only” be revoked on certain specified grounds. These grounds include “(a) the invention is not a patentable invention", and “(c) the specification … does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art". Section 72(1)(a) reflects section 1(1) though it may also go further. Section 72(1)(c) appears only to reflect section 14(3), but, as explained by Lord Hoffmann in Biogen [1997] RPC 1, 47, it also extends to what is covered by section 14(5)(c).

The reasoning of the courts below

75.  In a sense, it was at this point that the reasoning of the Court of Appeal in this case ended. At [2008] RPC 19, para 36, Lord Hoffmann said that “[w]hen a product claim satisfies the requirements of section 1 of the 1977 Act, the technical contribution to the art is the product and not the process by which it was made, even if that process was the only inventive step". Accordingly, as sections 1 and 14 appeared to be satisfied by the patent, he concluded that the claim to escitalopram was valid.

76.  To the same effect, Jacob LJ said at [2008] RPC 19, para 52, that, as at June 1988, the pure (+)-enantiomer, as a product, was “novel and non-obvious", and if “one asks the straightforward question ‘Does the patent enable the skilled man to make it?’ the answer is an equally straightforward ‘Yes’. So, in the language of art 83, the patent ‘discloses the invention in a manner sufficiently clear and complete for it to be carried out.’”

77.  The different view formed by Kitchin J was not based on any disagreement with this approach as far as it goes, but on reasoning which is helpfully summarised in his judgment at [2007] RPC 32, paras 264 and 265. He described the obtaining of the purified enantiomers as “an obviously desirable goal", and said that, accordingly, the “inventive step” was “not deciding to separate the enantiomers …, but finding a way it could be done". He went on to say that the technical contribution made by the Patent was not to find a new product, but to find a way of making a product, namely a single enantiomer of citalopram, through the medium of isolating the diol intermediate.

78.  Accordingly, the Judge concluded that, as the specification disclosed that the respondent had found only one way to make the (+)-enantiomer, it would be a monopoly disproportionate to the technical contribution if the Patent effectively covered all ways of making the enantiomer, which would be the effect of the product claim. The principle he relied on was succinctly encapsulated in a short sentence virtually at the end of his judgment, namely “The first person to find a way of achieving an obviously desirable goal is not permitted to monopolise every other way of doing so".

79.  The sole authority upon which Kitchin J relied in support of this analysis was the speech of Lord Hoffmann in Biogen [1997] RPC 1. I propose first to consider whether his conclusion is justified on the basis of any principle or authority other than what was said in this House in Biogen [1997] RPC 1, and then to address the reasoning in Biogen [1997] RPC 1.

The insufficiency argument apart from Biogen [1997] RPC 1

Continue  Previous