England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM CHANCERY DIVISION
MR JUSTICE PUMFREY

		Royal Courts of Justice Strand, London, WC2A 2LL Friday 14th December 2001

LORD JUSTICE ALDOUS:

1. The appellants G.D. Searle & Co, the Pharmacia Corporation and Pfizer Inc. are the proprietors of European Patent (UK) 0 679 157. I will refer to them as the patentees. They alleged that that patent had been infringed by Merck & Co Inc. and Merck Sharp & Dohme Limited by amongst other acts the manufacture and sale of a non-steriodal anti-inflammatory (NSAID) sold under the trademark Vioxx, but referred to by Merck as MK-966.
2. Merck accepted that they had carried out the allegedly infringing acts, but denied infringement and submitted that the patent was invalid upon a number of grounds. Pumfrey J upheld those submissions. His decision is challenged by the patentees in this appeal.
3. The appeal involves issues which require knowledge of a difficult area of chemistry. I am indebted to the judge for his full and clear exposition of the chemistry, the issues involved and his reasons for coming to the conclusions that he did. It has lightened the considerable burden involved in this appeal.

The Patent

4. The background - The patent was published on 2nd November 1995 and claims priority from a U.S application filed on 15th January 1993. It is entitled "Novel 3,4-Diaryl Thiophenes and Analogs Thereof Having Use as Antiinflammatory Agents". Such compounds would be recognised by the addressee of the specification as NSAIDs. They are well known drugs which include aspirin, paracetamol and ibuprofen. As is commonly known they can cause gastro-intestinal ulceration. Research in 1971 suggested these drugs had the ability to inhibit synthesis of a prostaglandin produced by the body during inflammation. This is because known NSAIDs directly inhibited what is called COX. The addressee of the specification would have been appraised of the common general knowledge of those working in that field. The judge summarised that knowledge in these passages of his judgment.

"5. I take the following account of the development of the understanding of the action of NSAID's from Professor Flower's evidence, which was unchallenged on this issue. There are two major classes of anti-inflammatory compounds, the steroids and the NSAID's. In 1971, Sir John Vane and his colleagues discovered that aspirin, indomethacin and sodium salicylate had the ability to inhibit the synthesis of prostaglandins, and suggested that this was the major mode of operation of these drugs. Prostaglandins are chemicals which play a part in inflammation. There are a number of prostaglandins, and they are responsible for many of the body's regulatory functions. The ones in which Vane was interested were those which were produced by the body during inflammation. They are synthesised in the body from arachidonic acid and other fatty acids via an enzyme then called prostaglandin H synthase, then fatty acid cyclooxygenase, and finally COX. Vane showed that NSAID's directly inhibited COX in cell-free tissue extracts in vitro. This discovery explained the tendency of the NSAID's, to cause stomach irritation and ulceration. The drugs cause the irritation because they inhibit the synthesis of the prostaglandin which is responsible for regulating the secretion of acid in the stomach and the secretion of the mucus which protects the lining of the stomach against being digested.

6. This elucidation of the action of NSAID's led immediately to a method of screening potential drugs for aspirin-like activity without having to perform in vivo experiments. Provided that a source of COX was available, the enzyme could be exposed to the drug, and its ability to synthesise prostaglandin from arachidonic acid could be measured. If the drug successfully inhibited COX, little arachidonic acid would be used to synthesise prostaglandin and the amount produced could be measured. At this time, all the enzyme available had to be derived from animal tissue; so there was dog spleen COX, rabbit brain COX, and so on.

…

10. Once the discovery of COX had been made, those responsible for the development of NSAID's, having available an assay for the activity of a compound, were in a position easily to investigate both new NSAID's and existing drugs.

…

15. By the mid-1980's a number of companies (including the defendants) had run their investigation of novel NSAID's down. During the later 1970's and the 1980's it appears that at least two groups were working on the activity of NSAID's in the inhibition of the COX enzyme. At DuPont, a long investigation into the properties of existing compounds had started in 1979/80, as Dr Galbraith described. This work consisted at that time of screening the compounds in vivo and the principal area of concentration was diaryl heterocycles. Such compounds are the subject of the two Haber patents relied on by the defendants in these proceedings as rendering the patent invalid. The diaryl heterocycles were (as far as the evidence went) 2,3 diaryl materials (Figure A)

Figure A: 2,3 diaryl compounds

16. Among the results of this work was a compound called DuP 697, a description of which was published in 1989 in a patent applied for in 1984 relied on by the defendants ("Haber 2"). This compound is also the subject of a paper also relied on by the defendants to support their allegation of invalidity by Gans and others. DuP 697 was reported as having low gastric irritancy but good anti-inflammatory characteristics. The authors of this Gans paper hypothesised that its good gastric characteristics were due to tissue selectivity and its lack of an acid functional group in the molecule, a group possessed by the largest class of NSAID's.

17. During the 1980's, other work was being carried out in relation to the prostaglandins and inflammation, particularly by Dr Needleman's group both before and after he joined the claimants in 1989. By 1989, it had been demonstrated that some COX enzyme was manufactured by the body in response to inflammatory stimuli. This means that the COX responsible for the inflammatory response was not a purely constitutive enzyme (i.e. present all the time in the body). In 1988 Dr Needleman had hypothesised that there might be two different COX enzymes. Then in 1991 it was reported by a number of groups that there was in fact not one but two COX enzymes. The paper of one of these groups (Xie) is relied on by the defendants. In 1992 Dr Needleman's group confirmed that there was a second COX enzyme which came to be called COX-2. The gene for this enzyme was cloned by late 1992, so making available supplies of recombinant COX-2 to all those who had access to the cloned gene."

5. There was a dispute before the judge as to whether the skilled addressee's common general knowledge would include knowledge of the existence of the isozymes COX 1 and COX 2 and that one of the isozymes was inducible and was a target for NSAID research. His conclusion was not challenged during the appeal. He said:

"24. I think that it is clear that those working at the forefront of NSAID research were in all probability aware of the existence of Cox II and of its importance. Indeed, it appeared from Dr Galbraith's evidence under cross-examination that all the information with which I am concerned was discussed at that conference by himself and the other speakers. The list of those attending survives, and although it represents a good cross-section of well-known chemical companies I cannot say that the attendees are representative of every company who might be interested in the subject matter of the patent. But a year elapsed from the Winter Prostaglandin conference to the priority date of the patent in suit, and another year until the application resulting in the patent in suit. By the application date for the patent in suit, I have no doubt that the Cox I/Cox II relationship, its importance and the fact that Cox II was a target for NSAID development was common general knowledge. On the whole of the evidence, and paying attention to that of Professor Flower, I think that it was also part of the knowledge of the skilled man, as a good basis for further action, at the claimed priority date."

6. The Specification – The specification starts with a general description of the invention. It states that the invention is in the field of anti-inflammatory pharmaceutical agents and relates to compounds and methods for treating inflammation and inflammation-related disorders. It says: "More specifically this invention relates to selected effective and safe compounds having anti-inflammatory and/or analgesic activity without erosion of the stomach." It continues with a short explanation of the background:

"Prostaglandins play a major role in the inflammation process, and the inhibition of prostaglandin production, especially production of PGG₂, PGH₂ and PGE₂, has been a common target of anti-inflammatory drug discoveries. However, common non-steriodal anti-inflammatory drugs (NSAIDs) that are active in reducing prostaglandin-induced pain and swelling associated with the inflammation process, are also active in affecting other prostaglandin-regulated processes not associated with the inflammation process. Thus, use of high doses of most common NSAIDs can produce severe side-effects, including life-threatening ulcers, that limit their therapeutic potential. An alternative to NSAIDs is the use of corticosteriods which have even more drastic side-effects, especially when long-term therapy is involved.

Previous NSAID's have been found to prevent the production of prostaglandins by inhibiting enzymes in the human arachidonic acid/prostaglandin pathway, including the enzyme cyclooxygenase (COX). Recently, the sequence of another heretofore unknown enzyme in the human arachidonic acid/prostaglandin pathway has been reported by T. Hia and K. Nielson, Proc. Natl. Acad, Sci, USA, 89, 7384 (1992) and named "cyclooxygenase II (COX II)" or "prostaglandin G/H synthase II". The discovery of an inducible enzyme associated with inflammation provides a viable target of inhibition which more effectively reduces inflammation and produces fewer and less drastic side effects. Cyclooxygenase II is inducible by cytokines or endotoxins and such induction is inhibited by glucocortoids (J. Masferrer et al, Proc. Natl. Acad. Sci. USA, 89, 3917 (1992)). The 6-methoxy-2-naphthylacetic acid metabolite of nabumetone has been found by E. Meade et al to selectively inhibit the COX II enzyme (J. Biol. Chem. 268, 6610 (1993)). In addition, Futaki et al (Gen. Pharmac., 24, 105 (1993)) has reported that N-(2-cyclohexyloxy-4-nitrophenyl) methanesulphonamide is anti-inflammatory and lacks gastric side effects.

The substituted thiophene compounds disclosed herein selectively inhibit cyclooxygenase II over cyclooxygenase I and relieve the effects of inflammation. These compounds, in addition, do not display substantial inhibition of cyclooxygenase I and produce a reduced amount of side effects."

7. Next the specification refers to a number of prior art patents describing anti-inflammatory activity which are said to "show continuing efforts to find a safe and effective anti-inflammatory agent."
8. There follows a heading "Description Of The Invention". That description starts on page 4 with a recitation of the formula of claim 1 which is said to depict "a class of compounds useful in treating inflammation-related disorders". It then states that compounds of the formula would be useful for the treatment of inflammation in a subject, and for the treatment of other inflammation-associated disorders. A number of disorders such as arthritis and headaches are given. The passage concludes in this way:

"The compounds are useful as anti-inflammatory agents, such as for the treatment of arthritis, with the additional benefit of having significantly less harmful side effects.

The present invention also includes compounds which selectively inhibit cyclooxygenase II over cyclooxygenase I and do not significantly inhibit one or more other arachidonic pathway steps, such as thromboxane B₂ (TXB₂) production. Importantly, thromboxanes cause blood platelet aggregation and have vasoconstriction properties. Thus a lack of effect in the regulation of non-inflammation related thromboxane production is further evidence of the beneficial selectivity of the present compounds.

Preferably the compounds of the present invention have a thromboxane B₂ inhibition lC₅₀ of greater than about 1.5 µM, as determined by a whole cell assay and preferably over 10 µM. The inhibition of the production of TXB2 by a whole cell assay is a better indicator of potential in vivo behavior as the assay also incorporates such factors as cell transport.

More preferably the compounds also have a selectivity ratio of cyclooxygenase II inhibition over cyclooxygenase I inhibition of at least 50 and preferably of at least 100. Such preferred selectivity may indicate an ability to reduce the incidence of common NSAID-induced side effects, such as ulcers."

9. As the judge pointed out the specification then lists preferred and more preferred compounds which appear to correspond with subsidiary claims. That is followed by a number of definitions.
10. Pages 11-31 set out methods of synthesizing compounds of the invention followed by a description of two in vivo and two in vitro tests. These were described by the judge in this way:

"32. The text may be picked up at page 31, where the biological evaluation of the compounds are described. Two in vivo and two in vitro tests are described. The in vivo tests are well established tests carried out on rats to evaluate NSAID's, and form part of the common general knowledge. It should be noted that such evaluations are performed only in respect of those compounds which have already demonstrated promise in vitro. In summary, the carrageenan foot pad oedema test is to determine whether the compound reduces the inflammation-induced swelling induced in a rat's paw caused by injection of a compound (carrageenan). The carrageenan-induced analgesia test consists of inflaming the paw, again by injection of carrageenan, and seeing whether the compound increases the time which it takes for the rat to remove its paw from a source of pain. A number of these evaluations were carried out for the purpose of this action. The results reported in the specification were in respect of 5 compounds only, and as will be apparent from Table I on page 32 of the specification not all 5 compounds were assayed on both tests. Example 1 is shown as analgesic but not as anti-inflammatory. Examples 2, 13 and 14, which are within the claim, are shown as having an anti-inflammatory effect, while the comparison example 4 shows comparable anti-inflammatory characteristics. It should be noted that Example 4 is said to be a comparison example. The compound is 3,4-bis (4-methoxyphenyl) thiophene. This compound does not fall within claim 1 because neither of the substituent aryl groups is itself substituted with methylsulphonyl or sulphamyl.

33. A very large amount of evidence was provoked by the in vitro tests which the specification now describes. The tests as described depend upon having available the gene for either human or mouse Cox I and Cox II.

A 2.0 kb fragment containing the coding region of either human or murine COX-I or human or murine COX-II was cloned into a BamHI site of the baculovirus transfer vector pVL1393 to generate the baculovirus transfer vector. Recombinant baculoviruses were isolated by transfecting 4µg of baculovirus transfer vector DNA into SF9 cells (2 x 10e8) along with 200 ng of linearized baculovirus plasmid DNA by the calcium phosphate method. Recombinant viruses were purified by three rounds of plaque purification and high titer (10E7 - 10E8 pfu/mI) stocks of virus were prepared. For large scale production SF9 insect cells were infected in 10 liter fermentors (Bioprocess group) (0.5 x 106/mI) with the recombinant baculovirus stock such that the multiplicity of infection was 0.1. After 72 hours the cells were centrifuged and the cell pellet homogenized in Tris/Sucrose (50 mM: 25%, pH 8.0) containing 1% CHAPS. The homogenate was centrifuged at 10,000xG for 30 minutes, and the resultant supernatant was stored at -80ºC before being assayed for COX activity.

34. The identification of the gene coding for human or murine COX had been identified by the claimed priority date. Professor Flower's evidence was that Needleman's hypothesis of the existence of the two isozymes was accepted by 1992 and the cloning of the human gene had been achieved in 1992. This passage of the specification reinforces my view that Professor Flower's evidence as to what was generally known is correct, and that the genes were available.

35. The assay for Cox I and Cox II activity which is described in the specification is in fact that which was used by the claimants at the relevant time in assessing Cox I/Cox II activity. The basic idea of the assay is that the enzyme is pre-incubated with the compound under test. During the period of incubation, the enzyme will be inhibited if the compound is a COX inhibitor. The enzyme is then exposed to its substrate, arachidonic acid, and the amount of prostaglandin synthesised by the enzyme in a fixed period is measured. A good inhibitor will mean less prostaglandin. The compound is tested at a number of concentrations and the final figure which is used for comparison purposed between different compounds is the concentration of compound which is needed to produce 50% inhibition, the so-called ID₅₀ or IC₅₀. Table II of the patent on page 33 shows the results of such tests. Fourteen compounds of the patent are shown as having been tested, mainly by reference to mouse Cox I and Cox II, although one of the results (Example 1) relates to human COX. The specification does not provide any interpretation of these results. The only guide to their interpretation is the passages to which I have already referred. There is no comparison with, for example, an established NSAID such as indomethacin which is used, incidentally, to stop the synthesis of prostaglandin during the assay. It seems to follow that this test is essentially a comparative test which is intended to reveal differences in Cox I and Cox II activity.

36. The final test to which the patent relates is the whole blood assay for thromboxane B₂ activity. This is an assay for Cox I inhibition. Professor Flower describes it as follows. This experiment tests for eicosanoid (i.e. prostaglandin derivative) synthesis as a consequence of blood platelet (red blood cell) aggregation. Briefly, blood is taken from a healthy volunteer. A sample is measured out together with the test compound and the blood is allowed to clot naturally in a 37°C water bath. During the clotting process, platelets are activated and arachidonic acid is released. Large amounts of prostanoids (that is, prostaglandin derivatives) will be produced by the platelets, as a result of Cox I acting on the arachidonic acid substrate. These prostanoids can be recovered (mainly in the form of thromboxane B₂ or TXB₂) in the serum. The presence in the blood of Cox inhibitors will reduce the TXB₂ formation. This is a very convenient way of assessing the activity of potential inhibitors in man and precise IC₅₀ values can be obtained.

37. Thus, the idea is that the thromboxane B₂ assay is that it provides a way of detecting Cox I inhibition alone. Professor Flower emphasises, and all the witnesses agreed, that it is not possible directly to compare the results of the thromboxane B₂ assay with that of the Cox I/Cox II comparative assay. The specification says nothing about the interpretation of the thromboxane B₂ assays that are set out. In general terms, however, it may be seen that the higher the IC₅₀ , the more of the compound is needed to achieve a 50% reduction in prostanoid synthesis, and thus the less effective is the compound as a Cox I inhibitor. Seen in the context of the passage in the specification to which I have referred in paragraph 26 above, this test enables the skilled man to identify those compounds included within the claim which do not significantly inhibit one or more "other arachidonic pathway steps, such as thromboxane B₂ production." Two preferable ranges for IC₅₀ in the thromboxane B₂ assay are indicated at page 5 line 5: greater than 1.5µM and greater than 10µM. Table III thus shows three groups of compounds: 1, 13 and 14, which show IC₅₀'s comfortably in excess of 10µM; compounds 10 and 11, which show IC₅₀'s in excess of 1.5µM, and compounds 2, 4 and 7 which are potent inhibitors of Cox I with very low IC₅₀'s.

38. This table may be compared with Table II, which shows relative Cox I/Cox II activity in that assay. Example 1 is highly Cox II selective over Cox I. Example 13 is also highly Cox II selective over Cox I. Example 14 is Cox II selective, as are compounds 10, 11 and 13. Example 4, the comparison example, shows little or no Cox II selectivity. If there is Cox II selectivity in example 7, it is obscured because in this assay example 7 is a potent inhibitor both of Cox II and of Cox I. The specification draws no conclusions from these results, which are merely presented to the reader. They may be related to the preferred ranges of selectivity set out on page 4 at line 9: Examples 1, 10 and 11 are certainly over 100, and example 14 may be: it is certainly over 50. Examples 3 and 13 show a selectivity of 50 or above. The remainder are Cox II selective, with the exception of example 4 (where they are comparable) and example 7, where no conclusions can be drawn."

11. Claim 1 was conveniently numbered and underlined by the judge to highlight the parts relevant to the issues of infringement. It is in this form:

"Claim 1 of the unamended patent is as follows. I have added labels to the features for ease of reference. This subdivision of the features was given to me by the claimants without dissent from the defendants. I have also underlined those features of the claim which are relied on for the allegation of infringement.

I

A compound of Formula I

wherein Y is selected from S, O and NR¹;

wherein R¹ is selected from hydrido and C₁-C₆ alkyl;

wherein X is one or more substituents selected from

(a) (i) hydrido

(ii) halo

(iii) cyano

(iv) nitro

(v) hydroxy

(vi) acyl

(vii) lower alkyl substituted at a substitutable position with a substituent selected from halo, hydroxyl, amino, acylamino, lower alkylamino, lower alkyl(acyl)amino, acyl, aryl optionally substituted with hydroxyl, a heterocyclic group, hydroxyimine and lower alkoxyimine,

(viii) lower alkonyl optionally substituted at a substitutable position with cyano,

(ix) amino optionally substituted at a substitutable position with a radical selected from acyl and lower alkylsulfonyl,

(x) sulfo,

(xi) sulfamoyl optionally substituted with a substituent selected from the group consisting of lower alkyl, halo(lower)alkyl, aryl, hydroxyl, lower alkylamino(lower)alkyl, a heterocyclic group and (esterified carboxy)lower alkyl,

(xii) N-containing heterocyclicsulfonyl,

(xiii) a heterocyclic group optionally substituted at a substitutable position with a substituent selected from the group consisting of hydroxyl, oxo, amino and lower alkylamino,

provided that when Y is O or NR¹ then X cannot be hydroxyalkyl,

(b) S(O)_nR⁵, wherein R⁵ is C₁-C₆ alkyl optionally substituted at a substitutable position with fluoro, and n is 0, 1 or 2,

(c) C(R⁶)(OR⁸)(R⁷) wherein R⁶ and R⁷ independently are selected from CF₃, CF₂H, CFCl₂, CF₂Cl, CClFH, CCl₂F, CF₃CF₂ and C₁-C₂ alkyl, and wherein R⁸ is selected from hydrido, C₁ -C₄ alkyl, (C₁-C₃ alkyl)C(O) and CO₂R⁹ wherein R⁹ is C₁-C₄ alkyl,

(d) C(O)ZR⁴, wherein Z is O, N, or S, and R⁴ is selected from hydrido, C¹-C⁵ alkyl and aryl, and when Z is N then R⁴ is independently taken twice, and

(e) C(R⁹) (NHR¹¹) (R¹⁰), wherein R⁹ and R¹⁰ are independently selected from CF₃, CF₂H, CFCl₂, CF₂Cl, CClFH, CCl₂F, and R¹¹ is selected from hydrido and C₁-C₃ alkyl, and

wherein R² and R³ are independently selected from

aryl or heteroaryl, wherein the aryl or heteroaryl radical is optionally substituted at a substitutable position with a radical selected from halo, lower alkyl lower alkoxy, lower alkylthio, lower alkylsulfinyl, lower alkylsulfonyl, nitro, amide, amino, lower akylamino, sulfamyl and lower alkylsulfonylamino,

provided that at least one of R² and R³ is substituted with methylsulfonyl or sulfamyl;

or a pharmaceutically acceptable salt thereof."

12. The amendment sought fell into two categories. First, the deletion of the words "aryl or heteroaryl, wherein the aryl or heteroaryl radical is" and the substitution of "phenyl" in the second part of integer (e). Second, limitation to R² and R³ being "4- methysulphonyl phenyl".
13. Claims 13-19 are limited to pharmaceutical compositions comprising a therapeutically-effective amount of anti-inflammatory compounds referred to in claims 1-11.
14. Claims 20-28 are to use of a compound as claimed in other claims for treating inflammation or an inflammation-associated disorder. Claims 29, 30 and 31 specify a particular disorder.
15. The patentees allege infringement of claims 1, 5, 6, 7, 13, 20 and 27-31. They do not make a positive case that the remaining claims have independent validity.
16. In that part of the judgment which dealt with the specification, the judge resolved what he called a fundamental dispute between the parties as to what the invention actually provided. He recorded that the patentees' case as opened was that "the classes of compounds claimed were classes of compounds all of which could reasonably be expected to have anti-inflammatory properties, which they possessed by reason of being Cox inhibitors. The claimants contended, however, that the specification did not suggest that the claimed class of compounds had gastric sparing qualities; and, regardless of whether selectivity of Cox II over Cox I provided gastric sparing qualities or not, did not suggest that the claimed compounds were Cox II selective." That is not the sole case that was advanced in this Court.
17. Mr Kitchin QC, who appeared for the patentees, drew attention to section 125 of the 1977 Act which provides that an invention "shall, unless the context otherwise requires be that specified in a claim." He then drew to our attention claims 1-12 which claim chemical compounds. He submitted that they could be used for any purpose because they were claimed without limitation as to use. That was emphasised by the terms of claims 13 to 26 which claimed compounds which were therapeutically-effective as an anti-inflammatory and claims 21-30 which had to be effective for the particular complaints set out. It followed that the invention of claim 1 was the compounds themselves.
18. Mr Kitchin is correct that claims 1 to 12 do not include any limitation as to use. Thus when construed without recourse to the rest of the specification, the invention claimed is to the chemical compounds set out. But that construction makes the invention inconsistent with, amongst other passages, the description of the invention in the specification. It states that "A class of compounds useful in treating inflammation-related disorders is defined by Formula 1". I will deal with this submission and the other submissions on construction later in this judgment in the context in which they arise. But I will first decide whether the judge was right to accept Merck's submission which is set out in paragraph 42 of his judgment as to what was the "invention" or "technical contribution" in the specification.

"42. On the other hand the defendants contended that the invention of the specification was a class of compounds substantially all of which were both anti-inflammatory and had significantly less harmful side effects than the existing NSAID's. They further said that the specification taught only two mechanisms for reducing harmful side effects: to provide Cox II selectivity, and further to provide Cox I inactivity. They submitted that while gastric-sparing qualities might arise from other causes, so far as the specification was concerned the teaching was such that the addressee would understand that the inventor's contribution lay in a class of compounds which possessed Cox II/Cox I selectivity at least when assayed in the manner described in the specification. However, apart from a specific teaching in respect of the thiophenes at page 3 line 29 (see paragraph 27 above) the patent neither identifies the members of the claimed class which possess Cox II selectivity nor those which do not inhibit Cox I. The defendants contend that accordingly the teaching of the patent is that all the claimed classes, or at least all the thiophenes, possess Cox II selectivity, and moreover produce a reduced amount of side effects. This question as to the teaching of the specification is fundamental to the dispute between the parties. …

19. The judge went on to set out the evidence of the witnesses. He concluded:

46. I have come to the conclusion that the defendants' contention is correct. The class is presented as a class of compounds which have anti-inflammatory and/or analgesic activity with fewer and less drastic side effects, the reduction in side effects being due to Cox II selectivity. Of the particular passages in the specification on which the claimants relied, I consider that the passage bridging pages 4 and 5 of the patent supports the construction which I think is the right one. As a matter of language, it says that all the compounds of the present invention selectively inhibit Cox II over Cox I, and some of them also do not inhibit one or more other arachidonic pathway steps (i.e. Cox I). The experimental results and the manner in which they are presented are also consistent with this view. All the compounds tested in vivo show either anti-inflammatory or analgesic qualities. The comparison compound 4 does not exhibit Cox II selectivity, but those within the claim do, with the exception of example 7 where the result is not determinate. Those compounds tested with a high degree of Cox II selectivity also have poor Cox I activity in the thromboxane B₂ assay. So apart from example 7, which is equivocal, the experimental results demonstrate that the compounds tested within the claim are Cox II selective. The preferred ratios for Cox II over Cox I activity do not throw any light on the problem, since they leave open the question whether the class includes compounds which exhibit no Cox II selectivity at all. The diffident statement that "Such preferred selectivity may indicate an ability to reduce the incidence of common NSAID-induced side effects, such as ulcers" again does not indicate that the class includes compounds which are not Cox II selective, and must be read in the context of the clear statement in respect of the thiophenes to which I have already referred and the general statement that the compounds have fewer and less drastic side effects. The whole thrust of the specification is towards Cox II selectivity."

20. I agree with the judge. Nobody reading the specification could believe that the "invention" was the compounds claimed in claim 1. The specification makes clear that the patentees had found a class of compounds that could be made which at least had anti-inflammatory action. It was that contribution that merited a 20 year monopoly. In my view the only question capable of argument is whether the compounds in the class were chosen merely for their anti-inflammatory action or because in addition they had reduced side-effects due to them being Cox II selective.
21. The Court has to read the specification with the common general knowledge of the skilled person. As found by the judge, such a person would have been aware of the work done, shortly before the application in the US from which patent priority is claimed, to discover a second Cox enzyme. He would know of the importance of the Cox I/Cox II relationship and that the inhibition of Cox II had since 1992 became a target of NSAID development. He would therefore not be surprised that the patent was directed to hitting that target.
22. The specification starts with the statement that the invention is in the field of anti-inflammatory pharmaceutical agents and particularly to selective effective and safe compounds having anti-inflammatory and/or analgesic activity without erosion of the stomach. That would bring Cox II selectivity to the forefront of the reader's mind.
23. The specification then explains that the common NSAIDs can produce severe side-effects including life threatening ulcers. They had been found to inhibit Cox. There follows a short explanation of discovery of Cox II. That "discovery of an inducible enzyme associated with inflammation provides a viable target of inhibition which more effectively reduces inflammation and produces fewer and less side effects." One such compound is mentioned which is said selectively to inhibit the Cox II enzyme.
24. There follows this paragraph:

"The substituted thiophene compounds disclosed herein selectively inhibit cyclooxygenase II over cyclooxygenase I and relieve the effects of inflammation. These compounds, in addition, do not display substantial inhibition of cyclooxygenase I and produce a reduced amount of side effects."

25. Mr Kitchin submitted that that passage was only referring to the substituted thiophenes not to any furan or pyrrole derivatives. He is wrong. The invention is entitled "Novel 3,4 Diaryl Thiophenes and Analogs Thereof …". That passage is referring generally to compounds of claim 1.
26. The reader would understand that the target was anti-inflammatory compounds with reduced side-effects, such as stomach ulcers, resulting from Cox II selection. That target had been achieved by the substituted thiophene compounds disclosed in the specification: in particular those in formula in claim 1. That would be confirmed in the passage on page 4 lines 45 to 57 which informs the reader that the compounds of claim 1 are useful in treating inflammation and a range of inflammation-associated disorders, some of which are set out. The compounds are said to be useful anti-inflammatory agents and have the additional benefit of having less harmful side-effects. The reader would believe that the specification was teaching that they were less harmful than the commonly made NSAIDs because they were Cox II selective. That is confirmed by the tests for activity which are designed to determine whether the particular examples were anti-inflammatory and Cox II selective, and the passages at the top of page 3 of the specification which give preferred levels of Cox II selectivity. The technical contribution disclosed in the specification is a new class of anti-inflammatory compounds which have reduced side-effects as they are Cox II selective.

Infringement

27. As the judge said:

"174. The compound sold by Merck as the active ingredient of Vioxx is 3-phenyl-4-(4-methylsulphonylphenyl)-2-(5H) furanone. It is referred to as MK-966 (Figure B).

Figure B: MK-966

It is common ground that this compound is not among those explicitly listed in claim 1. In aqueous conditions, however, MK-966 undergoes a chemical process called tautomerisation, and it is from this fact that the allegation of infringement really springs. Before turning to the details of the allegation of infringement, I should first describe tautomerism."

28. After explaining the basics of tautomerisation the judge said:

"The same is true of tautomerism in aromatic molecules. Returning to the defendants' compound MK-966, this undergoes tautomerisation in aqueous solution. Its tautomeric companion is 3-phenyl-4-(4-methylsulphonylphenyl)-2-hydroxyfuran (Figure C).



Figure C: tautomerism of MK-966"

29. He continued in paragraph 183 as follows:

"The hydroxyfuran is an acid, and will dissociate to some extent into an anion (a negatively charged ion) and a hydrogen ion (or proton).


Figure D: formation of enolate

The charges are shown by the plus and minus signs. The negatively charged ion is the enolate ion."

30. Infringement is governed by section 60 of the Patents Act 1977. Thus a person infringes if, but only if, he does an infringing act in the United Kingdom without the consent of the patentee. The relevant infringing acts are:

"60(1) (a) where the invention is a product, he makes, disposes of, offers to dispose of, uses or imports the product or keeps it whether for disposal or otherwise;

…

(2) Subject to the following provisions of this section, a person (other than the proprietor of the patent) also infringes a patent for an invention if, while the patent is in force and without the consent of the proprietor, he supplies or offers to supply in the United Kingdom a person other than a licensee or other person entitled to work the invention with any of the means, relating to an essential element of the invention, for putting the invention into effect when he knows, or it is obvious to a reasonable person in the circumstances, that those means are suitable for putting, and are intended to put, the invention into effect in the United Kingdom."

31. The patentees submitted that Merck infringed because claim 1 and certain other claims when properly construed, covered MK-966 as sold, namely the furanone or keto. That submission depends upon the construction of those claims and I will refer to it as the issue of primary infringement. Their alternative submission alleges secondary infringement. They submitted that Merck had supplied MK-966 which was a means, relating to an essential element of the invention, for putting the invention into effect when Merck knew, or it was obvious to a reasonable person in the circumstances, that those means were suitable for putting and were intended to put the invention into effect in the United Kingdom.
32. Primary Infringement - This issue on infringement starts from acceptance by both parties (a) that MK-966 is a solid, in which the active ingredient is the keto form: (b) that the words of claim 1 do not explicitly mention the keto form, and (c) the enol form is covered by claim 1 as X can be hydroxy.
33. Mr Kitchin QC made two submissions. First that the word "hydroxy" was a "term of art" which included the keto form. Thus the keto form (MK-966) was within the ambit of claim 1 when construed literally. Second the claim should be construed according to the principles laid down in the Protocol on Interpretation. If so, the word "hydroxy" should be interpreted to include the keto tautomer as that would give fair protection to the patentees and also reasonable certainty to third parties. I will deal with those submissions in turn.
34. I reject the first submission. The word "hydroxy" has a precise chemical meaning. The keto form by definition does not exist where X is hydroxy. There is nothing in the specification to suggest that this word should be given anything other than its normal chemical meaning. Further the evidence to which we were referred did not suggest to the contrary.
35. There was evidence which suggested that the skilled reader of the specification would assume that when the hydroxy (enol) form was referred to, the keto was embraced as well. That Mr Kitchin submitted reflected the common general knowledge that the keto and enol forms existed in solution in equilibrium with the keto form predominating. They are indivisible in solution. Thus a description of one tautometric compound was a complete description of all forms. In support he referred us to the evidence of Professor Baldwin. He said that a chemist would understand, given the enol form, that there existed a corresponding keto form and that their properties reflected the co-existence of all tautomers. It was implicit from the depiction of the enol form that the keto form existed as well. It was for those reasons that he read claim 1 as including MK-966.
36. That evidence does not alter the meaning of the word "hydroxy" in the context of claim 1. At best it is evidence which suggests that the Court, when seeking to find the ambit of the claim as required by the Protocol, should construe the claim as covering the keto form. In my view the judge was right when he said – "The meaning of 'hydroxy' is clear and unambiguous. It means hydroxyl substitution and the skilled man knows that the compound with a hydroxyl substitution is a different chemical species from its keto tautomer."
37. The judge went on to consider whether the conclusion he reached was consistent with the Protocol on Interpretation. He said:

"197. I must be satisfied that the construction which I place on the claim satisfies the requirements of the Protocol by providing fair protection for the patentee and a reasonable degree of certainty for third parties. Were the claim to be construed so as not to cover a compound which is necessarily present with a compound expressly claimed, and which would be prepared with that compound, I think that there is no offence against this principle. I believe that it is helpful to consider the phrase used by Fox LJ which is quoted by Hoffmann J in Improver, "within the ambit of the language". My conclusion is that the words used in the claim have a clear and unambiguous meaning which does not admit of variants, and that the words are not used figuratively so that the correct conclusion is that the claim does not cover the furanone tautomer of the 2-hydroxyfuran expressly claimed.

198. It follows from my conclusions that claim 1 is not infringed by the importation and sale of MK-966. Claim 13 is not infringed by the importation and sale of pharmaceutical preparations containing it."

38. Both parties referred this Court to the Protocol questions considered in Wheatley v Drillsafe Ltd [2001] RPC 133 and sought to persuade us that they should be answered in a way that suited their submissions. For example Mr Kitchin submitted that a negative answer should be given to the first question. The keto form was the smallest possible variant from the claimed enol (hydroxy). When a person put into practice the invention he would obtain the keto in addition to the enol form. The two forms were tautomers of each other and were intertransmissible.
39. Mr Young QC, who appeared for Merck, submitted that the first Protocol question should be answered in the affirmative. He drew attention to the difference in properties between the enol and keto. The enol had no or no substantial anti-inflammatory activity whereas the keto form had. To that Mr Kitchin countered – once administered all the tautomers would be present and the effect would be the same.
40. The Protocol questions are normally a useful tool to arrive at the middle ground required by the Protocol. But this is a case where the difficulties in application outweigh the advantages. As the judge said:

"196. … It is not altogether easy to apply the principles of construction articulated in these cases [Improver and STEP v Emson] to a claim in which every term is unambiguous and devoid of any question of degree (except as to the number of carbons embraced by the phrase "lower alkyl" and similar matters) but which would nonetheless be read as necessarily involving the presence of other species which are not mentioned but which would play a role in the chemical properties of the compound claimed. The third Improver question also presents a difficulty. It is in the nature of class claims that not every member of every class will be exemplified and described. If there is no description of a hydroxy-substituted compound, there will be nothing in the specification likely to help to decide whether the patentee intended as a matter of words to exclude the tautomer, but, since the presence of the tautomer is a matter of common general knowledge the fact that there is no reference to suggests a deliberate exclusion."

41. In my view this is a case where the Protocol questions cannot be used without modification. For that reason, I conclude that the Court has no alternative but to seek the middle way required by the Protocol by considering the effect upon the patentees and the public of reading claim 1 as covering the keto form. In particular, ascertaining what would be fair for the patentee and whether that would unfairly impinge upon the required certainty for the public.
42. Merck supported the reasoning of the judge. The claim was clear: the keto form was not the same as the hydroxy: they were structurally different compounds. The patentees had chosen the words used to define their monopoly and to construe claim 1 to include the keto would amount to using the claim as a mere guideline. To support that submission they drew attention to the fact that the hydroxy form was not directly detectable and was unstable in the sense that it converted predominantly to the keto or enolate tautomers. It was therefore the minor tautomer. That being so, the recognised convention was that reference to the minor tautomer meant that tautomer to the exclusion of all others. Further there was every reason to believe that that convention was adopted, as there was no indication in the specification that the patentee intended to cover other tautomers. The absence of such words as "or a tautomer thereof" was a clear indication that the patentee did not intend to include tautomers.
43. Merck also drew attention to page 14 of the specification which showed part of a synthesis. There step G was said to involve "reduction of a furanone to give the anti-inflammatory furans of the present invention." Thus the furanone was described by the patentees as an intermediate not a compound within claim 1.
44. The patentees submitted that Merck's approach did not take full account of the fact of tautomerisation. Everybody accepted that the enol was a compound within claim 1 and that those compounds had anti-inflammatory activity in the body. The skilled person would have known that the enol had tautomers which would exist in equilibrium in the body and that the tautomers were in practice inseparable with the result that the properties of the tautomers reflected all of those present. It followed that the enol would exist in solution together with the keto and that it was the composite which provided the effect.
45. Mr Kitchin accepted that it was not possible to separate the enol, but he relied on passages in Professor Baldwin's evidence which suggested that the tautomeric solution could be made through the enol. Mr Kitchin did not dispute the general convention of specifying the minor tautomer, but submitted that that was not definitive in the context of construction of claim 1 according to the Protocol. He also did not dispute that one of the methods of producing a compound within claim 1 was to use furanone as an intermediate. However, that did not mean that when claim 1 included the enol, it was not appropriate to include within the ambit of the claim the keto tautomer as well.
46. I, like the judge, have not found the issue of infringement easy. However I prefer the approach of the patentees to the more literalist approach of Merck. The compounds of claim 1 are a class which are said to have activity in the body. Thus to give fair protection to the patentee, it is reasonable to take into account the existence of the compound in the body. If so, it is accepted that the enol will exist in an inseparable equilibrium with the keto. It is the composite that will have the effect. To restrict the claim to the minor tautomer would not appear to achieve any useful purpose so far as the patentees are concerned.
47. To construe claim 1 as covering the keto tautomer would not, in my view, prevent there being a reasonable degree of certainty for third parties. Of course third parties would appreciate that the keto form was not specifically mentioned. But they would know that that did not necessarily mean that the claim would be construed as limited to the mentioned compounds. If the contrary were the position, then claims to chemical compounds would have to be construed using literal interpretation outlawed by the Protocol. They would also know that the inevitable result of using the enol was the formation of the other tautomers. They would therefore not be surprised if manufacture and sale of the other tautomers, which would form the enol in solution, would infringe. I therefore conclude that claim 1 includes the keto form and MK-966 infringes and the tautomers would together provide the activity.
48. Secondary Infringement – In view of the conclusion on primary infringement this issues does not arise and it would not be appropriate to indicate a view on the interesting submissions of counsel based upon an assumption that the issue of primary infringement had been decided in Merck's favour.

Insufficiency

49. Section 72(1)(c) provides for revocation of a patent on the ground that:

"(c) the specification of the patent does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art."

50. From the words of that subsection it is clear that if the invention is a novel compound, the specification must contain sufficient information to enable the skilled person to make that compound. What amounts to sufficient information will depend upon the circumstances. If the invention is a compound having a specific quality, such as the ability to stick two materials together or alleviate pain, the specification must contain sufficient information to enable manufacture of that compound having that attribute.
51. The judge accurately set out the principles behind section 72(1)(c) in paragraphs 60 and 61 of his judgment:

"60. It is a generally recognised principle of patent law that the patentee is not entitled to protection wider than the contribution which he has made to the art. It goes without saying that the patentee's claim should not cover that which is old or obvious. If it were otherwise, the patentee would be able to prevent others from doing what was old or obvious at the priority date, and this is intuitively wrong. Of course, this principle must be qualified. Merrell Dow v Norton [1996] RPC 76 establishes that it applies only to matters rendered old or obvious by information disclosed before the priority date. Other forms of protection (rather than invalidity) are provided by the statute for those who have used the invention before the priority date but in a manner which conveys nothing.

61. When one turns to the extent of the patentee's contribution to the art, the problem becomes more difficult. The essential underlying concept is that of the enabling disclosure. The patentee may not obtain a monopoly for matter which he has not told the public about and enabled them to do for themselves on the basis of what he has disclosed in the specification. This concept is manifested in two substantive requirements for the specification and claims. Subsection 14(3) of the Act requires that the specification of the application shall disclose the invention in a manner clear enough and complete enough for it to be performed by a person skilled in the art. Subsection 14(5) of the Act, mirroring Article 84 EPC, requires that the claim be supported by the description. It has never been the law that the claim must be co-extensive with the embodiments specifically disclosed by the patentee in his specification. Protection limited in this way would in all probability be illusory. When the patentee makes an invention, he may not appreciate its ramifications, and, when he files an application, he has in effect a year in which to develop and refine his invention before describing it in a further application which will result in the patent. The process of development and refinement may result in further applications along the way, but the application which will result in the patent must be made within a year. If, and only if, the claim is supported by the matter disclosed in the earliest application will it be entitled to priority from that application; and the intervening applications may form the bases of other claims to priority. The claims of the patent itself must be supported by the description given in the application for the patent, and the patentee may not add further to the disclosure of the application during the process of prosecution. As Lord Hoffmann explained in Biogen v Medeva [1997] RPC 1 at page 46, when the Act provides in section 72 for revocation of the patent on the ground that the specification of the patent does not disclose the invention clearly enough and completely enough for it to be performed by a man skilled in the art, it gives effect not merely to the substantive requirement of subsection 14(3) but to that of subsection 14(5)(c) as well."

52. The judge then cited passages from the speech of Lord Hoffman in Biogen v Medeva Plc [1997] RPC 1 including this passage which appears on page 54:

"Section 72(1)(c) of the 1977 is not only intended to ensure that the public can work the invention after expiration of the monopoly. It is also intended to give the court in revocation proceedings a jurisdiction which mirrors that of the Patent Office under section 14(3) or the E.P.O. under article 83 of the EPC, namely, to hold a patent invalid on the substantive ground that, as the E.P.O. said in Exxon/Fuel Oils (T 409/91) [1994] O.J. E.P.O. 653, paragraph 3.3, the extent of the monopoly claimed exceeds the technical contribution to the art made by the invention as described in the specification. In the 1949 Act, this function was performed by another ground for revocation, namely that the claim was not "fairly based on the matter disclosed in the specification" (section 32(1)(i)). The requirement of sufficiency was therefore regarded as serving a narrower purpose. But the disappearance of "lack of fair basis" as an express ground for revocation does not in my view mean that general principle which it expressed has been abandoned. The jurisprudence of the E.P.O. shows that it is still in full vigour and embodied in articles 83 and 84 of the EPC, of which the equivalents in the 1977 Act are section 14(3) and (5) and section 72(1)(c)."

53. The judge concluded:

"67. This seems to me to be a clear statement that the applicable principle is that the claim must equiparate (to use Graham J's phrase in Olin Mathieson v Biorex [1970] RPC 157) with the invention, and, if it does not do so, the claim will not be enabled across its full width, because it will cover matter which owes nothing to the invention disclosed. It follows that if the claim covers compounds which do not satisfy the representations made about them in the specification, it is likely to be invalid, because, again, the claim will be covering compounds which owe nothing to the teaching of the specification. This is not to say that when the claim reflects a new feature uniting the class as a class, and relating the desirable quality of the compounds to this feature, it is bad if a few members of the class do not possess the desirable quality or possess it in an attenuated form. But the qualities of the class must be related to the features of the claim. If compounds having the features of the claim may or may not possess the qualities which the patent says unify the class, it cannot be said that the claim reflects a true class at all. It is just a generalised description of a large number of chemical compounds. Such a claim is not analogous to a claim to a new principle, since the patentee has given no information, such as a structure/activity relationship, which enables the reader of the specification to draw any conclusions as to the properties of any particular compound without further experiment. All he has done is to describe the scope of the claim with spurious precision."

54. I do not believe it is helpful to assume that the test of sufficiency under section 72(1)(c) of the Act is that applied in Olin Mathieson which was decided under the 1949 Act. The issue in that case was whether the claim was fairly based (see section 32(1)(i) of the Patents Act 1949). There was no dispute that the claim was enabled by the specification in that the allegation of inactivity was not pressed. The true test is that laid down in the section 72(1)(c) of the 1977 Act. Further I cannot endorse, as a general proposition of law, the judge's conclusion that if a claim covers compounds which do not satisfy the representations made about them in the specification it is likely to be invalid. That was the law under the 1949 Act in that a patent could be revoked if it was obtained upon a false representation. The test of sufficiency under the 1977 Act is concerned with enablement. The specification must contain an enabling disclosure of the invention. If the invention requires satisfaction of representations, then the specification will not be sufficient unless those representations are satisfied.
55. With that introduction to the law, I return to Mr Kitchin's submission mentioned in paragraph 17 above. He is correct that claim 1 does not contain any explicit restriction as to use or the attributes of the compounds claimed. When read literally, it covers the claimed compounds which may or may not be suitable for pharmaceutical use. But that does not mean that the "invention" is of that width. As section 125 makes clear, the word "invention" is not restricted to meaning that specified in the claim if the context otherwise requires. An example where the context requires another meaning is section 14(3) and section 72(1)(c) which require the specification to disclose the invention in a manner which is clear enough and complete enough for the invention to be performed. 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.
56. Where the claimed invention is to a class of compounds, the same principle applies and, as was made clear by the House of Lords in Biogen, is that the disclosure in the specification must enable the invention to be performed to the full extent of the monopoly claimed. Thus if the invention is a selection of certain compounds, in order to secure an advantage or avoid some disadvantage, not only must the specification contain sufficient information on how to make the compounds, it must also describe the advantage or how to avoid the disadvantage. Further the compounds monopolised by the claim must all have that advantage or avoid the disadvantage. The same principle applies where the claim is to a class of compounds. To be sufficient, the specification must identify the characteristics of the class and a method of manufacture. Further all the claimed compounds must in substance have the characteristics of the class.
57. That view of the law is consistent with that expressed by the EPO as Lord Hoffmann pointed out in Biogen. He said at page 49 line 10:

"Since Genentech I/Polypeptide expression the E.P.O. has several times reasserted the well established principles for what amounts to sufficiency of disclosure. In particular, in Exxon/Fuel Oils (T 409/91) [1994] O.J. E.P.O. 653, paragraph 3.3, the Technical Board of Appeal said of the provision in the European Patent Convention equivalent to section 14(5)(c) of the Act:

"Furthermore, Article 84 EPC also requires that the claims must be supported by the description, in other words, it is the definition of the invention in the claims that needs support. In the Board's judgment, this requirement reflects the general legal principle that the extent of the patent monopoly, as defined by the claims, should correspond to the technical contribution to the art in order for it to be supported, or justified.""

58. That was reiterated in the decision of the Technical Board of Appeal in T939/92 (AgreEvo) [1996] OJ EPO 309.

"2.4.2 … it has for long been a generally accepted legal principle that the extent of the patent monopoly should correspond to and be justified by the technical contribution to the art (see T 409/91, OJ EPO 1994, 653, reasons Nos. 3.3. and 3.4, and T 435/91, OJ EPO 1995, 188, reasons Nos. 2.2.1 and 2.2.2). Now, whereas in both the above decisions this general legal principle was applied in relation to the extent of the patent protection that was justified by reference to the requirements of Articles 83 and 84 EPC, the same legal principle also governs the decision that is required to be made under Article 56 EPC, for everything falling within a valid claim has to be inventive. If this is not the case, the claim must be amended so as to exclude obvious subject-matter in order to justify the monopoly.

…

2.5.4 It follows directly from these considerations that a technical effect which justifies the selection of the claimed compounds must be one which can be fairly assumed to be produced by substantially all the selected compounds (see also e.g. T 131/87 of 7 September 1989, No.8 of the reasons T 742/89 of 2 November 1992, No.7.4 of the reasons, T 626/90 of 2 December 1993, No.4.3.2 of the reasons, and T 741/91 of 22 September 1992, No 4.2 and 4.3 of the reasons)."

59. It is also consistent and only consistent with the conclusions reached by the House of Lords in Biogen. They held that section 72(1)(c) required an enabling disclosure across the width of the claim. If so the word "invention" in section 72(1)(c) must include the technical contribution as that is the teaching of the specification which enables. As Lord Hoffmann said at page 47:

"'The concept of an enabling disclosure is central to the law of patents. For present purposes, it touches the matters in issue at three different points. First, as we have seen, it forms part of the requirement of "support" in section 5(2)(a). Secondly, it is one of the requirements of a valid application in section 14. And thirdly, it is essential to one of the grounds for the revocation of a patent in section 72. I shall start with section 14. Subsection (3) says:

"The specification of an application shall 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."

This is plainly a requirement of an "enabling disclosure". In addition, subsection (5)(c) says that the claim or claims shall be "supported by the description". It was by reference to subsection (3) that Lord Oliver of Aylmerton, who gave the leading speech in Asahi, reasoned at page 536 that a description would not "support" the claims for the purpose of subsection (5)(c) unless it contained sufficient material to enable the specification to constitute the enabling disclosure which subsection (3) required: "the Act can hardly have contemplated a complete application for a patent lacking some of the material necessary to sustain the claims made". By parity of reasoning, he said that "support" must have the same meaning in section 5(2)(a).

The absence of an enabling disclosure is likewise one of the grounds for the revocation of a patent specified in section 72(1). Paragraph (c) says that one such ground is that -

"the specification of the patent does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art."

This is entirely in accordance with what one would expect. The requirement of an enabling disclosure in a patent application is a matter of substance and not form. Its absence should therefore be a ground not only for refusal of the application but also for revocation of the patent after grant. Similarly, the same concept is involved in the question of whether the patent is entitled to priority from an earlier application. This is not to say that the question in each case is the same. The purposes for which the question is being asked are different. But the underlying concept is the same.'

60. He continued at page 54 line 6:

"Section 72(1)(c) of the 1977 is not only intended to ensure that the public can work the invention after expiration of the monopoly. It is also intended to give the court in revocation proceedings a jurisdiction which mirrors that of the Patent Office under section 14(3) or the E.P.O. under article 83 of the EPC, namely, to hold a patent invalid on the substantive ground that, as the E.P.O. said in Exxon/Fuel Oils (T 409/91) [1994] O.J. E.P.O. 653, paragraph 3.3, the extent of the monopoly claimed exceeds the technical contribution to the art made by the invention as described in the specification. In the 1949 Act, this function was performed by another ground for revocation, namely that the claim was not "fairly based on the matter disclosed in the specification" (section 32(1)(i)). The requirement of sufficiency was therefore regarded as serving a narrower purpose. But the disappearance of "lack of fair basis" as an express ground for revocation does not in my view mean that general principle which it expressed has been abandoned. The jurisprudence of the E.P.O. shows that it is still in full vigour and embodied in articles 83 and 84 of the EPC, of which the equivalents in the 1977 Act are section 14(3) and (5) and section 72(1)(c).'"

61. The patent in this case claims a class of compounds. There is no technical contribution in a list of compounds which a skilled person would know how to make at the priority date. The 20 year monopoly was granted because of the disclosure in the specification that the class of compounds claimed had the quality disclosed in the specification. The invention or technical contribution justifying the monopoly claimed can only be that quality. I have already decided that the judge was right when he held that the specification would be read by the skilled person as disclosing that the claimed class of compounds had anti-inflammatory and/or analgesic effect with fewer and less drastic side-effects, the reduction in side-effects being due to Cox II selectivity. It is that disclosure which is the technical contribution and invention.
62. Merck carried out experiments which they alleged showed that a substantial number of compounds falling within claim 1 were inactive as anti-inflammatory compounds and were not Cox II selective. Thus it was alleged claim 1, even as amended, was insufficient. That allegation was upheld by the judge. The patentees submitted in this Court that the judge's conclusion should be reversed because he had gone wrong in principle and in fact. They also took a pleading point and for that reason I must first review the state of the pleadings.
63. The relevant part of the pleaded case of insufficiency was in these terms:

"5. The specification of the Patent does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art. In particular

(a) the specification of the patent fails to identify which (if any) compounds among those classes of 3,4 diaryl thiophenes, furans and pyrroles claimed:

(i) are Cox II selective or enable the skilled addressee of the patent to predict which compounds are Cox II selective;

(ii) exhibit anti-inflammatory activity or enabled the skilled addressee of the patent to predict which compounds exhibit anti-inflammatory activity;

(iii) exhibit reduced gastric side effects when administered to patients or enable the skilled addressee of the patent to predict which compounds will exhibit reduced gastric side effects.

(b) the specification of the patent fails to teach the skilled addressee how to synthesise the substituted pyrrole and furan derivatives of the invention and in particular the hydroxyfurans of the invention.

In the premises the claims of the Patent extend beyond the technical contribution to the art (if any) of the specification of the Patent."

64. The allegation in paragraph (b) was only pressed before the judge in respect of one class of compounds and for the purpose of this appeal can be disregarded.
65. Further and better particulars were sought of "all compounds which the defendants intend at trial to allege are … claimed by the patent in suit but which (a) are not Cox II selective and/or (b) do not exhibit anti-inflammatory activity and/or (c) do not exhibit reduced gastric side-effects when administered to patients."
66. The response was contained in a document called Supplemental Further and Better Particulars. The relevant part was as follows:

"2. Without prejudice to the generality of the plea, hereunder the Defendants repeat the groups of compounds listed in paragraph 1(a), 1(b), 1(d)-(f) above as examples of those compounds which are claimed in the patent but which are not COX II selective and do not exhibit reduced gastric side effects.

Further compounds which R³ is a phenyl group substituted in the 2 position by an alkyl group and substituted in the 6 position by any group occupying a space larger than flourine; or compounds in which R³ is a non phenylaryl group substituted in the 2 position by any group occupying a space larger than flourine, are claimed in the Patent but are not COX II selective and do not exhibit reduced gastric side effects."

67. I do not need to set out the groups, as there followed on 29th January 1999 the Notice of Experiments which was followed on 26th March by a Statement of Case of the use to be made of the experiments. The relevant part of that document is as follows.

"1. The Defendants refer hereunder to the Supplemental Further and Better Particulars of the Particulars of Objections of 21 September 1998.

2. In their Notice of Experiments the Defendants have alleged that they will provide the following facts in relation to the in vitro experiments the results of which are recorded in Annexes 1(a) and 1(b) herein:

2. That compounds of the alleged invention lack h-COX II selectivity unless one of R2 or R3:

(a) is substituted with methylsulfonyl or sulfamyl; and in particular unless,

(b) one of R2 or R3 is 4-methylsulfonylphenyl or 4-sulfamylphenyl; and,

(c) there are no additional substituents on R2 (other than the 4-methylsulfonylphenyl or 4-sulfamylphenyl groups).

3. That prior art compounds 72 – 75 and 78 have lower IC50 against h-COX II than against h-COX I selective and that prior art compounds 74, 75 and 78 have a h-COX II selectivity ratio greater than 50.

4. That of the compounds exemplified in the patent in suit, examples 3, 5, 8 and 9 (that is to say compounds 3, 4, 7 and 8) are inactive as measured by either h-COX I or h-COX II inhibition in vitro.

5. That of the compounds listed in Annex I those in which the aryl or heteroaryl group R3 is substituted with a bulky group or are distributed, are inactive as measured by either h-COX I or h-COX II inhibition in vitro (see for example compounds 27 – 34 and 37).

6. That of the compounds listed in Annex I those in which the thiophene or furan ring is substituted at the carbon atom opposite R2 by a bulky group are inactive as measured by either h-COX I or h-COX II inhibition in vitro (see for example compounds 4 and 54 – 60).

7. That of the 3,4 diarylthiophenes and furans listed in Annex I those in which the thiophene or furan ring is substituted at the carbon atom adjacent R2 by a bromine group or any group that occupies a space as large or larger than a bromine group are inactive as measured by either h-COX I or h-COX II inhibition in vitro (see for example compounds 38 – 40a and 43 – 47).

3. The Defendants have disclosed the results of all COXII/COXI in vitro testing using the techniques advanced in the patent in suit) of the compounds falling within claim 1 of the patent in suit. At the trial of this action they intend to rely on those results which are the subject of the Notice of Experiments."

68. Exhibit RJF 25 to the witness statement of Professor Flower contains a table which sets out every compound either the subject of an experiment performed for these proceedings or an experiment performed by one or other of the parties in the normal course of their research. It is a useful table, but must be approached with care as the comments made do not always represent all the relevant information.
69. Exhibit RJF 25 splits the compounds into classes A to F. Class A contains 22 compounds which are within claim 1. They are numbered B1 to B22. All were COX II selective and were designated by Merck as active. They included examples 1, 2, 6, 7, and 10 to 15. Class B consisted of 15 compounds numbered B23 – B37. They were said to be compounds where one of R² and R³ was not substituted with methylsulphonyl or sulfamyl and to lack activity. Thus class B contains compounds falling within one of the classes set out in Merck's statement of case.
70. The patentees submitted that to rebut the attack of insufficiency based upon classes B to F of exhibit RJF 25 it was sufficient to show that one compound falling within each class had anti-inflammatory activity. That submission was based upon the reading of the specification which I have rejected. They realised that the Court might read the specification differently and therefore submitted in the alternative that the attack of insufficiency could be rebutted by showing that only one compound within each class did have anti-inflammatory action and was COX II selective.
71. The task of this Court is to ascertain whether the technical contribution in the specification applies to the class of compounds claimed. It was Merck's case that certain subclasses of the claimed class did not have the attributes asserted for the claimed class. That being so, the extent of the monopoly claimed exceeded the technical contribution. Thus the Court had to consider the subclasses relied on by Merck and to decide whether, on the balance of probabilities, they met the technical contribution. I do not accept that it necessarily follows from the evidence that if one compound in a subclass showed some activity, the conclusion sought to be drawn by Merck should be rejected. The evidence must be considered as a whole. Any evidence of activity must be weighed against the rest of the evidence.
72. The in vitro experiments were succinctly described by the judge in this way:

"71. I can turn to the details of the experiments. The ELISA plates are made of translucent plastics material and contain 96 "wells" in which reagents are successively added to the compound under test. This is a standard technique which utilises an antibody which can bind specifically to PGE2. It results in a colour change which can be measured using a spectrophotometer and is inversely proportional to the amount of PGE2 present. The ELISA used is based on a proprietary kit produced by Cayman Chemicals. There are a number of prescribed concentrations of compound, expressed as micro moles or µM, which enables the results of the assay on different compounds to be compared. If a compound is insufficiently soluble to pass into solution, the results are said to be valueless because not all the compound is in solution and the nominal concentration is not achieved.

72. The concentrations selected by the defendants were 0.0064µM, 0.032µM, 0.16µM, 0.8µM, 4µM, 20µM, 100µM and 1000µM. Each successive concentration is five times the preceding one (except, obviously, between 100 and 1000µM). There was precipitation in all of the defendants' experiments at 1000µM. There is some dispute as to the extent to which there was precipitation in the others."

73. The judge pointed out that in the patent, activity had been assessed either at 10µM or 100µM. He concluded that the criterion to be adopted, when deciding whether the extent of the monopoly exceeded the technical contribution, was that provided by the patent itself. That conclusion was criticised by the patentees. They submitted that the tests set out in the patent were carried out to show activity, not to provide a definition. In any case the skilled person would not follow the tests blindly. He would know that a negative result would not mean that the compound was inactive. He would be able to sensitise his assays and would do so, if he obtained a negative result. He would also realise that the tests in the specification had an arbitrary cut-off and were designed to find commercially attractive compounds, not to differentiate between those that were active, and those that were not. The judge should have held that a compound was "active" if it demonstrated appreciable "activity" using any acceptable test.
74. I reject those submissions of the patentee. The task of the court is to ascertain from the specification what is said by the patentee to be the technical contribution. I have already decided it was a class of compounds which were active in the sense that they were anti-inflammatory and COX II selective. That activity was established in the specification by the tests carried out. To ascertain whether any compound had that activity the skilled person must be taken to be seeking to put the teaching of the specification into effect. He cannot be expected to adopt a different criterion from that selected by the patentee. Thus Merck cannot be criticised for carrying out the tests used in the specification. The judge was right to test whether the extent of the monopoly claimed exceeded the technical contribution by adopting the criteria which were used in the specification to explain the technical contribution. I accept the evidence of Professor Flower that a negative result in one type of assay does not establish that the compound tested is inactive as it might be found to be active in another assay. So what? Insufficiency is concerned with whether the monopoly claimed exceeds the technical contribution. That must be judged by the same criteria as those which set the standard of the contribution.
75. I conclude that the judge was right to reject the patentee's criticisms that Merck had failed in their experiments to adjust the conditions to avoid precipitation at 1000µM, and that they had wrongly adopted a standard of IC₅₀ less than 100µM. I also reject the suggestion that the compounds tested were not representative of the subclasses because they could have been selected with a view to making the technical contribution look bad. The patentees had ample opportunity to do experiments in reply to demonstrate that the compounds were not representative. They had the equipment and the knowledge to do the experiments, but failed to do them.
76. I come next to the results of the experiments conducted for these proceedings and those done during research by the parties as listed in exhibit RJF 25.
77. As to class B, it was the patentees' case that two of the 15 compounds in the class did not support the proposition for which Merck contended. The judge considered the evidence relating to those two compounds and concluded "… that substitution at the 4 positions of R² and R³ by a methylsulphonyl or sulphamyl group is essential to COX II selectivity". The patentees' submissions in this Court were similar to those made to the judge. Having considered them, I have no doubt that the judge was entitled to come to the conclusion he did.
78. Class C contains 5 compounds. The judge held that 4 were inactive, but that one, compound C1, was accepted to show COX II selectivity. C1 was a compound tested by the patentees during their research. Thus the experiment was not repeated in the presence of Merck. I suspect that the judge's conclusion on C1 was based upon an answer (Ev. 11, 1491) by Professor Baker in cross-examination to the effect that C1 had activity. However, Professor Baker went on, having looked at the graph that showed the results of the assay, to point out that the dose—responsive curve was not absolutely normal. That being so, the judge's conclusion – "I think the only conclusion which one can draw from these five examples is that it is beginning to look as though activity will be rather more likely if R₂ is a phenyl ring" – may be understated in favour of the patentees.
79. The judge's conclusion on class D compounds was:

"96. Class D is intended to show that substitution at the 4 position in R₂ is essential, a move to either the 2 or 3 positions resulting in an inactive molecule. This the experiments did show: there were no counterexamples. The claimants say that the result is to be explained by insolubility, and those otherwise analogous compounds in class A which did work are to be explained by the fact that sufficient got into solution to work. In fact, a check of the results show that these compounds showed no activity at any concentration whether there was precipitation or not. This experiment establishes that a compound in which the methylsulphonyl or sulphamyl group is not at the 4 position will be inactive."

80. The only criticism advanced concerned the criteria used by Merck, and I have already rejected that as a basis for discounting the results of the experiments. The judge was entitled to come to his conclusion on this class.
81. The judge's conclusion on class F was:

"98. Class F was a somewhat miscellaneous class, intended to show that various bulky groups on the aryl ring not substituted at the 4 position with methylsulphonyl or sulphamyl would prevent inhibitory activity. The claimants complain, with some justification, that Class F represented a moving target. There were three classes of compounds. The Notice of Experiments states that the fact to be proved is

…compounds in which the aryl or heteroaryl group R₃ is substituted with a bulky group, or are disubstituted, are inactive as measured by either h-Cox I or h-Cox II inhibition in vitro (see for example compounds 27 - 34 and 37).

In his report, Professor Baker relied not only upon this experiment but upon three other compounds. Thus his class F in its entirety consists of compounds B50 to B63 inclusive. In support of the fact, however, Professor Baker also relied on a number of compounds in Class A (which, it will be remembered, the defendants considered to be active and Cox II selective). His evidence was this:

Section F of the Table contains a list of compounds that differ from selective COX-2 molecules only in that they have a large group on the other aryl group (R³). Comparison of Sections A and F indicates that, whilst there is some space around the second aryl group, such that fluoro (B1), methyl (B6), methoxy (B3), thiomethyl (B13) and amino (B14) can be tolerated, anything significantly larger cannot be accommodated in the enzyme.

The examples which he gives of the effect of such substitutions is complicated, and he summarises them in his report thus:

(a) A second 4-methylsulphonylphenyl or sulphamylphenyl group is not tolerated by either COX-1 or COX-2 (see compounds B50 & B51). Further, a 4-methylsulphinylphenyl group is not tolerated (B58);

(b) Even relatively small groups are not tolerated in the 2-position (see B52, B55 and B56). This effect may be aggravated if there is a substituent in the X position opposite the 4-methylsulphonylphenyl because this may cause the twisting of the aromatic ring (see B62).

(c) Whilst there appears to be more space around the 4-position, moderately large alkyl or alkylamine groups in these positions result in inactive compounds against either form of the enzyme (see B53 and B54).

(d) The 2, 6 or 3,5 di-substituted compounds for example in which R3 is 2, 6-dichlorophenyl and 3,5-difluorophenyl are inactive (see B60 and B61).

Professor Flower introduced a number of additional compounds into the discussion, with the result that RJF 25 contains 30 compounds in class F. It should be recalled that the compounds found by Professor Flower in the defendants' documents have a number beginning C and it is not suggested that Professor Baker was aware of these compounds when he wrote his report. With his C compounds, Professor Flower produces counter-examples to each of these propositions of Professor Baker's. In fact, there is a clear counter-example to (a) in compound C13 as Professor Baker accepted in cross-examination. Professor Baker himself qualified paragraph (b) in the light of compound C9. His evidence was that the results collected in RJF 25 represented an attempt at an SAR for these molecules, which would necessarily require refinement, but this is not one of the facts that the Notice of Experiments was put forward to show. There is no doubt that many of the compounds in question were inactive; others less so. The claimants pointed to C3, C5, C6, C10, C11 and C12 as compounds where there was activity below but close to 50% at 100µM or 1000µM. On the other hand, some compounds were plainly inactive, among which B55 is to be included. The same goes for paragraph (d). The evidence was thin, but the compounds disubstituted at the 2,6 and 3,5 positions actually tested were inactive."

82. Merck accepted that C13 showed activity, but submitted that it was wrongly inserted into class F by Professor Flower as the compound did not have the bulky group of the class. Thus the conclusion reached by the judge that - "many of the compounds in question were inactive; others less so" – could be generous to the patentees. Merck may be right, but I do not believe it right or necessary to differ from the conclusion reached by the judge.
83. The judge went on to consider classes G and H and then to arrive at these conclusions:

"100. It is possible to draw a number of conclusions from the very extensive evidence which was available in respect of classes F, G and H. The first is that it is quite clear that it is not possible to predict the degree of activity of the compounds in general. The second is that it is likely that the defendants have identified classes of compounds within the claims in which activity is very unlikely. In particular, it seems clear that large substituents at X will be inactive, and this is more likely if there are two X substituents, as the claim permits. Since X may be acyl, a substantially unbounded class of substituents, and aryl, a very large class also, and since both on Professor Baker's unchallenged evidence include substituents equal in size to or bigger than the whole of the rest of the molecule, I am satisfied that what can loosely be called the bulky group point is established. From classes B, C and D I am satisfied that unless there is a 4-methylsulphonylphenyl or 4-sulphamylphenyl substitution at either the 3 or 4 position of the heterocycle, it is very much more likely than not that the compound will be on the patent's own terms inactive. Finally, Class E does establish that disubstitution of the 4-substituted phenyl ring will substantially reduce the prospects of finding an active compound."

84. Merck also did in vivo experiments on rats using the compounds of Examples 1, 2, 5, 7, 11 and 15. The patentees also did tests on the compound of Example 1.
85. Before the judge and before us the patentees submitted that, at best, no positive conclusion could be drawn from the results as to whether the compounds lacked anti-inflammatory action or were gastric sparing. They submitted that the information provided by the tests was insufficient to enable any conclusion to be drawn and they criticised the failure to use DMSO as the solvent.
86. I have already held that Merck cannot be criticised for carrying out tests using the method disclosed in the specification. That being so, the patentees' experiments which used DMSO as the solvent should not be taken as establishing that the compounds of Example 1 had the activity asserted in the specification. DMSO has extraordinary properties as it is a carrier which penetrates membranes. Even so, its use only produced marginally better results than those produced by Merck. Certainly it would not be appropriate to conclude, from the patentees' tests on the compound of Example 1, that Merck's results on the other Examples should be discounted.
87. We were taken with the assistance of counsel through the results of the in vivo experiments. I reject the detailed criticisms advanced by the patentees. The judge was entitled to come to the conclusion he did. This Court should not come to a different conclusion unless it is shown that he went clearly wrong. He did not. His conclusion was:

"112. The in vivo tests did demonstrate that there were three examples of the patent in suit which failed to demonstrate any anti-inflammatory effect, and a preferred compound (compound 68) which was a gastric irritant. They did not demonstrate that example 1 of the patent in suit was inactive as an anti-inflammatory, but did show that it was certainly not attractive."

88. The judge summarised his conclusion on insufficiency in this way:

"115. In coming to a conclusion on the question of the right criterion to apply in deciding whether or not a compound exhibits activity in the in vitro tests, I have gone over a great amount of the ground covered by the objection of insufficiency. I am satisfied that the experimental evidence shows that the specification provides the skilled man with a class comprising many compounds which may or may not be Cox II selective with diminished side effects. The effect of the experiments taken as a whole is to demonstrate that the class in question is unpredictable in its qualities and contains a substantial number of compounds which do not have one or other or both of these qualities. It is not possible to point to any characteristic which underpins success, and avoids failure. As I have indicated, the feature which immediately distinguishes the compounds of claim 1 from those constituting the state of the art is (subject to the disclosure of Brown) the 3,4 disubstitution of the heterocycle as opposed to the 2,3 disubstitution displayed in, for example, DuP 697. But it is clear that this does not give success on its own; and I am satisfied on the evidence that it has not proved possible to provide a general description of a class having the desired characteristics. As will be apparent when I consider the history of the claimants' development, I do not consider that the work they had done by the priority date nor the later work which they did justify any suggestion that it was a reasonable prediction that the class of claim 1 either as granted or as amended would possess anti-inflammatory properties.

116. The invention of claim 13 must be considered separately. Claim 13 covers "a pharmaceutical composition comprising a therapeutically-effective amount of an anti-inflammatory compound, said compound selected from a compound of Claim 1, or a pharmaceutically-acceptable salt thereof". This is a claim, in effect, to any compound of claim 1 which works. Necessarily, the specification is open to the same objections in respect of the invention specified by this claim as it is in respect of that of claim 1. Claim 20 stands or falls with claim 1 so far as this objection is concerned. I conclude that the unamended specification is insufficient in respect of the invention of claims 1, 13 and 20.

117. So far as the amended specification is concerned, the position is to some extent different. The amendment to claim 1 restricts that claim to compounds in which one of R₂ or R₃ is a 4-methylsulphonylphenyl group and the other a substituted phenyl group, and in which there is only one substituent X. The experimentally demonstrated fact that it is likely that substitution by a methylsulphonyl or sulphamyl group at the 4 position of a phenyl ring is essential to activity is accordingly accommodated by the amendment. The excessive range of possible X substituents, which includes substituents larger than the rest of the molecule and is unbounded in number is not catered for. As I have found that on the balance of probabilities a large number of these compounds will also be inactive, the specification is still insufficient in respect of the invention specified by the amended claim 1, and so also in relation to that of claims 13 and 20."

89. I agree with the judge. He could have gone further and pointed out that the claims covered an enormous number of compounds and that upon his findings many hundreds if not thousands did not have the quality of the class. For example, the list of possible substitutions for X is extremely large and a very large number would be inactive because they are too large. As pointed out by Professor Baker, many of that group can be bigger than the rest of the molecule. The term "aryl" is unbounded and the term "aryl" introduces groups which are at least as large as the rest of the molecule itself. Further the conclusions of the judge in respect of classes F, G and H bear upon the amended claim as well.
90. I conclude that claims 1, 13 and 20 of the patent as granted and as sought to be amended are invalid as being insufficient. As the patentees do not assert independent validity of the other claims, the patent as a whole is invalid.

Priority Date

91. The patent claims priority from a US application filed on 15th January 1993. Merck alleged that that claim was not soundly based having regard to the changes appearing in the application which was filed. If they are right, the earliest priority date is the 14th January 1994 and application No. 94918259 (Merck Frosst) is of earlier priority. It follows from section 2(3) of the 1977 Act that the Merck Frosst application formed part of the state of the art for novelty. As the Merck Frosst application disclosed MK-966, the patent must be invalid having regard to the conclusion reached that MK-966 infringes. In fact the patentees accept that the disclosure in Merck Frosst invalidates claims 1, 5-8, 13, 20, 27-31, if the patent is not entitled to the priority date of the US application.
92. Priority is governed by section 5 of the Act. The relevant parts of that section are:

"5. – (1) For the purposes of this Act the priority date of an invention to which an application for a patent relates and also of any matter (whether or not the same as the invention) contained in any such application is, except as provided by the following provisions of this Act, the date of filing the application.

(2) If in or in connection with an application for a patent (the application in suit) a declaration is made, whether by the applicant or any predecessor in title of his, complying with the relevant requirements of rules and specifying one or more earlier relevant applications for the purposes of this section made by the applicant or a predecessor in title of his and each having a date of filing during the period of twelve months immediately preceding the date of filing the application in suit, then –

(a) if an invention to which the application in suit relates is supported by matter disclosed in the earlier relevant application or applications, the priority date of that invention shall instead of being the date of filing the application in suit be the date of filing the relevant application in which that matter was disclosed, or, if it was disclosed in more than one relevant application, the earliest of them;

(b) the priority date of any matter contained in the application in suit which was also disclosed in the earlier relevant application or applications shall be the date of filing the relevant application in which that matter was disclosed or, if it was disclosed in more than one relevant application, the earliest of them.

…

(4) The foregoing provisions of this section shall apply for determining the priority date of an invention for which a patent has been granted as they apply for determining the priority date of an invention to which an application for that patent relates."

93. Section 5 stems from Articles 87 and 88 of the EPC. They are relevant as section 130(7) provides that section 5 is to have, as near as practicable, the same effect in the UK as the corresponding provisions of the EPC.
94. The leading authority on the construction of section 5 is the speech of Lord Hoffmann in Biogen. Having set out section 5 of the Act he said at page 46:

"In Asahi Kasei Kogyo KK's Application [1991] R.P.C. 485 this House decided that for matter to be capable of supporting an invention within the meaning of section 5(2)(a) it must contain an "enabling disclosure", that is to say, it must disclose the invention in a way which will enable it to be performed by a person skilled in the art. This construction has not been challenged by the appellants before your Lordships' House. It is however important to notice the relationship between the requirement of "support" in section 5(2)(a) and certain other provisions of the Act which share the concept of an enabling disclosure.

The concept of an enabling disclosure is central to the law of patents. For present purposes, it touches the matters in issue at three different points. First, as we have seen, it forms part of the requirement of "support" in section 5(2)(a). Secondly, it is one of the requirements of a valid application in section 14. And thirdly, it is essential to one of the grounds for the revocation of a patent in section 72. I shall start with section 14. Subsection (3) says:

"The specification of an application shall 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."

This is plainly a requirement of an "enabling disclosure". In addition, subsection (5)(c) says that the claim or claims shall be "supported by the description". It was by reference to subsection (3) that Lord Oliver of Aylmerton, who gave the leading speech in Asahi, reasoned at page 536 that a description would not "support" the claims for the purpose of subsection (5)(c) unless it contained sufficient material to enable the specification to constitute the enabling disclosure which subsection (3) required: "the Act can hardly have contemplated a complete application for a patent lacking some of the material necessary to sustain the claims made". By parity of reasoning, he said that "support" must have the same meaning in section 5(2)(a)."

95. He went on at page 47 line 50:

"The need for an enabling disclosure to satisfy the requirements of support under section 5(2)(a), valid application under section 14 and sufficiency under section 72(1)(c) has, I think, been plain and undisputed since the decision in Asahi. What has been less clear is what the concept of an enabling disclosure means. Part of the difficulty has been caused by a misinterpretation of what the Technical Board of Appeal of the E.P.O. said in Genentech I/Polypeptide expression (T 292/85) [1989] O.J. E.P.O. 275. This was a patent for a plasmid suitable for transforming a bacterial host which included an expression control sequence or "regulon" which could enable the expression of foreign DNA as a recoverable polypeptide. The Examining Division was willing to grant a patent only in respect of the plasmids, bacteria and polypeptides known at the date of application. The Technical Board of Appeal allowed the appeal, saying that the Examining Division had taken too narrow a view of the requirement of enabling disclosure:

"What is also important in the present case is the irrelevancy of the particular choice of a variant within the functional terms `bacteria', `regulon' or `plasmid'. It is not just that some result within the range of polypeptides is obtained in each case but it is the same polypeptide which is expressed, independent of the choice of these means. ... Unless variants of components are also embraced in the claims, which are, now or later on, equally suitable to achieve the same effect in a manner which could not have been envisaged without the invention, the protection provided by the patent would be ineffectual ... The character of the invention this time is one of general methodology which is fully applicable with any starting material, and is, as it was already stated, also independent from the known, trivial, or inventive character of the end-products."

In other words, the applicants had invented a general principle for enabling plasmids to control the expression of polypeptides in bacteria and there was no reason to believe that it would not work equally well with any plasmid, bacterium or polypeptide. The patent was therefore granted in general terms.

In Mölnlycke AB v. Procter & Gamble Ltd. [1992] F.S.R. 549, however, Morritt J. interpreted this decision to mean that it was a general rule of European patent law that an invention was sufficiently disclosed if the skilled man could make a single embodiment. This interpretation was followed by Aldous J. in Chiron Corporation v. Organon Teknika Ltd. [1994] F.S.R. 202, although I think I detect in his judgment some surprise that the E.P.O. should have adopted such a mechanistic and impoverished approach to the concept of enabling disclosure. As we shall see, he applied the same rule in the present case.

In fact the Board in Genentech I/Polypeptide expression was doing no more than apply a principle of patent law which has long been established in the United Kingdom, namely, that the specification must enable the invention to be performed to the full extent of the monopoly claimed. If the invention discloses a principle capable of general application, the claims may be in correspondingly general terms. The patentee need not show that he has proved its application in every individual instance. On the other hand, if the claims include a number of discrete methods or products, the patentee must enable the invention to be performed in respect of each of them.

Thus if the patentee has hit upon a new product which has a beneficial effect but cannot demonstrate that there is a common principle by which that effect will be shared by other products of the same class, he will be entitled to a patent for that product but not for the class, even though some may subsequently turn out to have the same beneficial effect: see May & Baker Ltd. v. Boots Pure Drug Co. Ltd. (1950) 67 R.P.C. 23, 50. On the other hand, if he has disclosed a beneficial property which is common to the class, he will be entitled to a patent for all products of that class (assuming them to be new) even though he has not himself made more than one or two of them."

96. Lord Hoffmann went on to consider the technical contribution to the art which had been disclosed in Biogen I, the document from which priority was claimed. He concluded that that contribution did not justify a claim to a monopoly of the width of claim 1 of the patent and that therefore the claim was not entitled to the priority date of Biogen I.
97. In the past it has been possible to find decisions of the EPO which differed from the principles explained by Lord Hoffmann, but they have been recently reviewed by the Enlarged Board of Appeals in G 002/98 [2001] EPO OJ 11. That decision is consistent with the explanation given by Lord Hoffmann.
98. The judge held:

"132. So far as the unamended patent in suit is concerned, on the other hand, at least one methylsulphonyl or sulphamyl substitution, not necessarily on a phenyl ring but on an aryl ring (which includes naphthyl, phenyl substituted with a lower alkyl, and also other two-membered rings, three-membered rings and heteroaryl rings) is a defining feature of the claim. In my judgment, claim 1 of the patent in suit is not entitled to priority for three reasons. First, because there is no disclosure of a sulphamyl substitution in R² or R³ in the priority document other than in the list of Formula I', and certainly no disclosure of it as a defining feature of the invention. Second, there is no disclosure of a methylsulphonyl substitution in R² or R³ as a defining feature of the invention, but only as defining a preferred subclass in combination with a 4-fluorophenyl substitution, a subclass which does not include a hydroxy substitution at X. Third, a monopoly defined by one or other of these substitutions as an invariant feature across the whole width of the claim cannot be fairly based on a disclosure which does not hint at their materiality either textually or as a matter of necessary implication but only, at best, as a result of well-informed assessment of probabilities.

133. The amended claim is no longer open to objection on the ground either that it covers a sulphamyl substituent at R² or R³, or that it covers R² or R³ substituted with methylsulphonyl other than as 4-methylsulphonylphenyl. The objection that there is no disclosure in the priority document of such a substitution otherwise than in the combination I have already described, and the consequence is the same."

99. Mr Kitchin took us to the priority document which discloses the classes of compounds; the broader being designated Formula 1¹ and the narrower being Formula 1. He then explained the features of the compounds of claim 1 could be found in the priority document. The range of substitutions for X in claim 1 is the same as in the priority document. R³ of the priority document may be substituted phenyl as can be R² and that either R² or R³ can be substituted with methysulphonyl. That being so, claim 1, the patentees submitted, was a permissible narrowing based upon the priority document.
100. I accept that once the patentees come to believe that the class of compounds claimed in claim 1 had particular activity, it was possible to trace in the priority document where compounds within claim 1 were disclosed. That reflects the extreme breadth of the disclosure in the priority document but it does not follow that it provides support for claim 1. That can be demonstrated by the fact that formula 1¹ does not require a 4-methysulphonyl or a 4-sulfamyl to be essential. It is that feature which is sought to be introduced by amendment as an essential characteristic of claim 1. Formula 1 of the priority document does require a 4-methysulphonylphenyl group at the 3 or 4 position in certain compounds, but such compounds also have the other aryl as a 4-fluorophenyl group with the X substitution not including the hydroxy.
101. In my view the judge was right that the invention (technical contribution) of the priority document is not the same as that disclosed in the application. The invention of the priority document, namely the broad class of compounds in formula 1¹ and class in formula 1 is not the same as in the application. The disclosure of the broad class and the subclass without more cannot support the selection of another subclass as having anti-inflammatory activity and COX II selectivity which is the invention of the application. Of course it is possible to claim in a patent a narrower range of compounds than that disclosed in the priority document. But the fact that it is a narrower range does not mean that there is support for that range. What is required, as Lord Hoffman pointed out in Biogen, is that the priority document must contain sufficient material for the priority document to constitute the enabling disclosure of claim 1.
102. I have already held that the specification of the patent which essentially is in the same form as the application document does not sufficiently disclose the invention (section 72(1)(c)). The application is even more deficient. It does not disclose the class of compounds claimed, nor does it disclose that the invention is that class which has anti-inflammatory activity and is COX II selective. The support is deficient.
103. The judge was right to conclude that the application is not entitled to its priority date. That being so, it is accepted that claims 1, 5-8, 13, 20 and 27-31 lack novelty. Independent validity is not asserted for the other claims.

Novelty

104. Section 72(1)(a) provides for revocation if the invention is not a patentable invention. Section 1 sets out the conditions that need to be satisfied. The material one, contained in subsection (1)(a), is that the invention must be new. Section 2 states that the invention shall be taken to be new if it does not form part of the state of the art, meaning that it has not "been made available to the public".
105. Merck alleged that all the claims, as granted, which are asserted to have independent validity lacked novelty having regard to the disclosure in US Patent 3743656 (Brown) which was published in 1973. Their submission, advanced before the judge and in this Court, involved a two-pronged attack. First that the anti-inflammatory compounds disclosed in Brown anticipated the claims as granted, and second that the compounds disclosed in Brown as intermediates also anticipated the claims.
106. If Brown is to anticipate a claim, there must be clear and unmistakable directions in Brown to make a compound within the relevant claim (see General Tire and Rubber Co. v Firestone Tyre and Rubber Co. Ltd [1972] RPC 457). With that in mind I turn to the disclosure of Brown.
107. Brown discloses an invention for novel heterocyclic aromatic compounds. They are said to exhibit "pharmacological activity for example anti-inflammatory activity as shown by tests on warm-blooded animals and/or are intermediates in the preparation of other substituted furans, thiophenes and pyrroles."
108. As the judge pointed out the class of compounds disclosed in Brown is huge. What is alleged by Merck is that there is a substantial overlap between that class and the class claimed by the patentees.
109. In claim 1 of Brown the invention is said to provide compounds of what is referred to as Formula 1:



110. Ar and Ar¹ are aryl radicals. They may (see column 1 line 52) be in the 2 and 4, the 2 and 3, 3 and 4, 3 and 5, or 2 and 5 positions. Further each of the aryl groups can be a phenyl and phenyl substituted by halogen, lower alkyl, lower alkoxy, nitro, amino, tryfluromethyl, mercapto and methylsulfonyl (see column 1 line 63-70). R is an aliphatic acid containing 2 to 6 carbon atoms or a derivative thereof and Z can be hydrogen or alkyl.
111. The judge held that claim 7 of the patent had been anticipated by Brown in that claim 7 covered a compound disclosed in Brown were Z was hydrogen; Ar and Ar¹ were phenyl radicals substituted as suggested and R¹ was an aliphatic group.
112. Mr Young submitted that the overlap between the disclosure of Brown and claim 7 was substantial. Claim 7 covered all compounds of Brown where Ar is phenyl substituted with methysufonyl; Ar¹ is any of the options suggested save for phenyl substituted with triluoromethyl or mercapto; Z is either hydrogen or alkyl and R¹ is an aliphatic group.
113. The patentees submitted that Merck's submissions were based on a wrong construction of claims 1 and 7. Claim 1 did not include any compound with an aliphatic acid group. That the judge accepted. He said:

"144. The first question is accordingly one of construction of the patent in suit. Does the claim allow X to be an aliphatic acid? The argument proceeds in stages. First, the claim expressly contemplates that one of the substituents (or the substituent, so far as the amended claim is concerned) may be an acyl group (feature a(vi) of the claim as I have labelled them) or may be a "lower alkyl group substituted … with a substituent selected from … acylamino, … lower alkyl(acyl)amino, acyl …" (feature a(vii) of the claim). The terms "acyl" and "acylamino" are normally well defined. The primer gives the necessary structural diagrams.

The acyl group is shown thus (R is an alkyl group, that is, what is left after removing a single hydrogen atom from a hydrocarbon).



The acylamino group is shown thus (R1 is an alkyl group, R2 is an alkyl group or hydrogen):



145. At page 9 line 56 of the specification, the term "acyl" is given an extended meaning:

The term "acyl", whether used alone, or within a term such as "acylamino", denotes a radical provided by the residue after removal of a hydroxyl ion from an organic acid. Suitable "acyl" and acyl moiety in the terms "acylamino" and "lower alkyl(acyl)amino" may be carboxy,…"

At page 10 line 6, the specification continues

The terms "carboxy" or "carboxyl", whether used alone or with other terms, such as "carboxyalkyl", denotes –CO₂H.

Taken together, these two passages may be summarised as saying that acyl equals an acid without its -OH; but in the term "lower alkyl(acyl)amino" it may mean carboxy, that is, an acid which has not had its -OH removed. It therefore includes acid substituents of this general type:



146. Such acids are called carbamic acids, as Professor Baldwin pointed out. Are such acids aliphatic? The answer is that they are not, because of the nitrogen atom in the chain. But Mr Young points out that the "more preferred" class of compounds described in the patent in suit at page 5 line 28 is stated to fall within formula I and undoubtedly expressly contemplates substitution of a lower alkyl itself substituted with carboxy (lines 29-31). This passage corresponds to claim 7 of the patent in suit, which is to a class of compounds "of claim 1" and which therefore on the face of it fall within claim 1. He submits that one way or another X=(lower alkyl substituted with carboxy) and hence X=aliphatic acid is contemplated by claim 1. Mr Kitchin acknowledged the difficulty, saying that there is a clear inconsistency between claim 1 and claim 7 in this respect, and that claim 7 covers at least one class of compounds which do not fall within claim 1."

114. The judge went on to hold that claim 7 was anticipated because claim 7 expressly included carboxy. That the patentees disputed. They accepted that claim 7 mentioned carboxy as a possibility for X, but drew our attention to the opening words of claim 7, which claimed a "compound of claim 6" and also that claim 6 was appendent to claim 1. It followed, they submitted, that claim 7 should be construed as claiming a compound within claim 1. If so, it could not include an aliphatic acid.
115. Merck submitted that as the carboxy was specifically mentioned in claim 7, it must be included within claim 1 and that was borne out by construing claim 1 in the context of the teaching in the specification. If so, claim 1 as granted was anticipated. There was an alternative, namely that claim 7 should be construed as a claim to compounds of the type claimed in claim 1, but differing in at least one respect namely that it included the carboxy.
116. As claim 7 specifically states that X is selected from a lower alkyl substituted at a suitable position with a substitution selected from - … carboxy, it is difficult to construe it as excluding the carboxy. However to construe claim 1 as excluding the carboxy appears to set up tension between claims 1 and 7. That the judge appreciated. He held that there was a clear inconsistency between claim 1 and claim 7 and concluded that claim 7 covered at least one class of compounds which does not fall within claim 1. I agree, if he is right as to the construction of claim 1. That turns upon whether claim 1 allows X to be an aliphatic acid.
117. Claim 1 states that X can be "aryl or may be a lower alkyl substituted at a suitable position with … acylamino, … lower alkyl(acyl)amino_, acyl ….". As recorded by the judge:

"145. At page 9 line 56 of the specification, the term "acyl" is given an extended meaning:

The term "acyl", whether used alone, or within a term such as "acylamino", denotes a radical provided by the residue after removal of a hydroxyl ion from an organic acid. Suitable "acyl" and acyl moiety in the terms "acylamino" and "lower alkyl(acyl)amino" may be carboxy,…"

At page 10 line 6, the specification continues

The terms "carboxy" or "carboxyl", whether used alone or with other terms, such as "carboxyalkyl", denotes –CO₂H."

118. Mr Young submitted that the last sentence of the definition of "acyl" should be read as – "suitable 'acyl' (and acyl moiety in the terms 'acylamino' and 'lower alkyl(acyl)amino') may be carboxy."
119. If he is correct, then the judge's conclusion in paragraph 145 of his judgment set out above was wrong.
120. The judge was right to construe claim 7 as including a compound where X included the carboxy. That being so, there would, at first sight, appear to be a conflict between claims 1 and 7 unless the definition of acyl includes carboxy. Mr Young's suggested construction of the definition at page 9 line 56 has the attraction that claims 1 and 7 can be read as being consistent. If so Brown anticipates both claims 1 and 7. However the wording is not clear. As claim 1 is invalid on other grounds, there is no need to resolve this issue. I therefore endorse the conclusion of the judge that claims 7 and 20 lack novelty.
121. Having regard to that conclusion, there is no need to consider whether the judge was correct when he held claims 1 and 20 anticipated by the disclosure of intermediaries in Brown. I decline to add to this long judgment by doing so.

Obviousness

122. The judge held the patent both as amended and as unamended invalid as being obvious. He rightly set out the law in this way:

"139. … The objection of obviousness assumes that there is no anticipation of the claim, but that nonetheless there has been no inventive activity on the part of the patentee which justifies the grant of a patent. Obviousness is a question of fact, and the factual analysis is habitually approached in the manner described by Oliver LJ in Windsurfing International v Tabur Marine [1985] RPC 59 at 73.

There are, we think, four steps which require to be taken in answering the jury question. The first is to identify the inventive concept embodied in the patent in suit. Thereafter, the court has to assume the mantle of the normally skilled but unimaginative addressee in the art at the priority date and to impute to him what was, at that date, common general knowledge in the art in question. The third step is to identify what, if any, differences exist between the matter cited as [forming part of the state of the art] and the alleged invention. Finally, the court has to ask itself whether, viewed without any knowledge of the alleged invention, those differences constitute steps which would have been obvious to the skilled man or whether they require any degree of invention.

There are a number of arguments which are frequently advanced to fend off an attack of obviousness which need to be considered with care. The first (and perhaps the commonest) is that where there is a number of possible courses of action for the skilled man to pursue in the light of a particular disclosure, no particular course is obvious. This is wrong: all of the courses of action which present themselves without the exercise of invention are obvious (see Brugger v Medicaid [1996] RPC 635). The second is that the test in such a case is whether the skilled man could, rather than would, adopt the course of action which would bring him within the claim. A test formulated in this way almost invites the tribunal to consider irrelevant considerations, such as commercial attractiveness (Hallen v Brabantia [1991] RPC 195). In each case, the only question is whether something within the claim was technically obvious to the skilled man in the light of the disclosure relied on. Evidence of what was actually done in the art at the relevant time can be of assistance, but what people actually did may be explicable on many grounds having nothing to do with technical obviousness. For this reason it has been said that such evidence has to be kept in its place and it is necessarily valueless unless the persons whose activities are under examination can be shown both to be aware of the prior art and possessed of the common general knowledge (see Molnlycke v Procter & Gamble [1994] RPC 49 and Hoechst Celanese v BP [1997] FSR 547)."

123. Mr Kitchin realised that this Court was unlikely to reverse the decision of the judge, on what has been called a jury question, unless he could show that the judge had gone wrong in principle. He submitted that he had done just that. First he submitted that the judge had failed to realise that to find the invention obvious, it was necessary to conclude that there was some obvious reason or purpose to make the compounds of claim 1. In support he referred us to this passage in the judgement of Laddie J in Hoechst Celanese Corp. v BP Chemicals Ltd [1997] RPC .. at page 573:

"Before a step from the prior art can be held to be obvious there must be some reason why the man skilled in the art would wish to take it. If he has a problem and the step would occur to him as a solution to it, then he has a reason. But there is no requirement that it be demonstrated that the step would have been expected to produce significant commercial advantages. The problem might be very small. The courts will assume that he may just want an alternative way of achieving essentially the same result as in the prior art. Thus were workshop modifications, none of which would be expected to produce significant technical or commercial benefits are still obvious. To adopt an example sometimes given by Jacob J., if it is known to make a 5-inch plate, it is obvious to make a 5¼ -inch plate. Technicians and businessmen frequently want to make trivial variations in established or known products. Similarly if the prior art discloses two wooden parts held together by screws it would be obvious to glue them, even if so doing would not be expected to advance the industry. The notional addressee is likely to want to use materials readily at hand to make essentially the same thing as is disclosed in the prior art. That is sufficient motivation and the use of those materials is, accordingly, obvious. When the defendants argue that Hingorani or any of his readers is entitled to use any "natural extension" or "obvious variant" of his concept, they are correct if by that they mean the type of workshop modification or alternative discussed above. But it was not and could not be suggested by any witness that changing the medium from aqueous to organic and changing the resin was a mere workshop variant of what is set out in Hingorani." (checked)

124. That statement of the law was, I expect, apt on the facts of that case, but should not be followed generally. A step from the prior art, albeit made without reason, can still be obvious. The judge categorises such as step as workshop modifications and, in so doing, introduces a test not in the statute, namely whether the step from the prior art was a workshop modification. The statutory test is obviousness and any modification which is obvious will not be patentable, whereas one which is not obvious will be. The true test, as made clear in Windsurfing, is to ask whether the invention was obvious. Whether or not there is a reason for taking the step from the prior art may well be an important consideration, but that does not mean that it is an essential requirement of a conclusion of obviousness. In any case the judge in these proceedings did consider whether there was a reason for taking the step from the prior art and concluded that there was, namely a natural desire to investigate the analogs and the structural activity relationship of such compounds. I will return to this later.
125. Second, Mr Kitchin submitted that the judge had failed to realise that before a claim, such as claim 1, could be held obvious, it was necessary for the court to hold that the step from the prior art to the claim would serve a useful purpose. That he said was the conclusion reached by Graham J in Olin Mathison Chemical Corporation v Biorex Laboratories Limited [1970] RPC 157. That case is an example of the way patent actions were fought prior to Oliver LJ's judgment in Windsurfing. Many hours were spent arguing about the question to be considered. That is no longer desirable. Use of the structured approach advocated by Oliver LJ in Windsurfing avoids such considerations. Whether or not a useful purpose would be served may be relevant, but that cannot in all cases be a requirement before a finding of obviousness results.
126. Mr Kitchin also submitted that the judge failed to take proper account of the expert evidence and instead substituted his own ideas for theirs. I reject that submission. The judge correctly summarised the evidence of the patentees' experts and was entitled to conclude as he did. I will come back to this submission as I believe that the evidence of Professor Baker was more persuasive on the essential issue than that of Professor Baldwin.
127. It was also submitted that the judge had taken account of commercial factors, whereas he should have decided whether the invention was technically obvious. Further he had been guilty of hindsight reasoning. I can find no basis in the judgment to support those criticisms.
128. I have not been able to discern any error of principle in the judge's conclusions on obviousness and therefore would not reverse his finding. In fact I have concluded that he was right and will give my reasons for that conclusion.
129. The primary attack on obviousness arose from the Gans paper which was published in 1990. Its disclosure was correctly summarised, as the parties accepted, in paragraph 159 of the judgment.

"159. Gans. Gans is a 1990 paper which describes the anti-inflammatory activity of DuP 697 in rats. It points out that there is a need for NSAID's which are safer than existing drugs, and states that DuP 697 does not produce either intestinal or gastric ulcers in rats at single doses of up to 400 mg/kg. This apparent gastric safety is attributed to its distinct chemical structure as a non-acidic thiophene, and also to the possibility that it may be tissue selective, that is, it may inhibit prostaglandin synthesis only in certain tissues and not in gastrointestinal tissue. This paper shows, therefore, that at 1990 DuP 697 was an interesting compound, but that the possibility that there might be two isoforms of COX had not been appreciated, and there is nothing, therefore, about Cox II selectivity. The claimants contend that the evidence is that the lack of irritancy of DuP 697 is attributed in the paper to its lack of a carboxylic acid moiety: this is certainly one reason advanced, but tissue selectivity is advanced as an alternative, or additional, reason."

130. The second task, advocated in Windsurfing, is to assume the mantle of the unimaginative skilled person. I have already set out what was the common general knowledge of such a person. It is sufficient to recall that such knowledge included the importance of COX II selectivity and that COX II selectivity was a target for NSAID development.
131. The third task set by Windsurfing is to identify what differences exist between the disclosure in Gans and the alleged invention. In this case, there is no dispute that DuP 697 had a 2,3 substitution pattern around the heterocyclic ring, whereas the compounds of claim 1 have a 3,4 substitution pattern. DuP 697 is the 2,3 isomer of Example 2. As the judge said:

"157. The substantial allegation of obviousness is based on disclosures of DuP 697 and its properties. A comparison with Example 2 of the patent in suit shows why:



DuP 697 is the 2,3 isomer of Example 2 of the patent in suit. The Defendant's case is simple. They say that if one wishes to produce an SAR of DuP 697, then it is obvious to make its 3,4 isomer, which falls within the claim. Routine investigation would also traverse a great number of compounds within the claim."

132. The final task is to decide whether, viewed without knowledge of the alleged invention, the change from 2,3 substituted to the 3,4 substituted pattern was obvious. The patentees' case that the change was not obvious was supported by the evidence of Professor Baldwin. The judge accurately summarised that evidence in this way:

"164. Professor Baldwin's view of medicinal chemists working in industry was that they basically try to copy. When they get a lead, he suggested that they mounted "some sort of chemical war" on the molecule to make something better. Generally, I think his view was that they take the line of least resistance: that is, they go first for the things that are easy to make. The evidence was that the 2,3 compounds were easier to synthesise than the 3,4 compounds. I do not think that I misrepresent his evidence if I say that his view was that invention might lie in departing from the line of least resistance. I have no doubt that he may be correct. His view was that the change from 2,3 to 3,4 substitution was sufficiently radical that it would not immediately commend itself to a skilled team seeking to make more anti-inflammatory compounds having reduced side effects."

133. These passages from his cross-examination in Evidence 5 illustrate Professor Baldwin's view:

"A. Yes. I think my characterization from my contact with these people, as consultant for the pharmaceutical industry, is that they basically try to copy. They find something that someone else has discovered and that gives them what they call a lead. Then they mount some sort of chemical war on that molecule to change it into something that is better. The theme that they largely follow, as I have observed, is the route of least work, of least effort, which is actually understandable because many of the substances that they may wish to make are actually not very easy to make. Their job is to produce compounds that can be tested in a screen that some other people are running to see if they can improve on the so-called lead structure. I am not criticizing them, but it is inevitable that they take the path of least resistance and they follow the small changes. If they do not work out, they are forced to do more and more difficult things until eventually they do something that actually leads to an invention of a new structure. They do not do that initially. They initially try to, if you like, massage what is there with modest changes and test them to see if they can see a route leading away from the first activity to improved activity.

…

I would have thought a medicinal chemist, when confronted with that data would say, "Let us really try some interesting variations on this basic structure and see if we can get away from the problems they are having with it, and we can do that because we have chemistry that is already laid out by Du Pont to construct, in a very simple fashion, 2,3 diaryl thiophenes with many variations permissible by that chemistry." Alternatively, they would say, "We are going to try something totally different and change the positions of the aryl group." First of all, they would have to create new chemistry to assemble these types of molecules before they can actually do anything. I think, in practical terms, there is a big barrier to actually making substantive changes.

Q. But the route to prepare 3,4s, that was known from … Well, we have Brown referring to a route, or two routes, as to how to make the 3, 4s.

A. It was possible to make them, yes, but they are more complex than the route to the Du Pont compound. The Du Pont synthetic scheme is extremely simple.

Q. Monsanto and Merck, as you probably know, both came up with 3, 4s in the course of their, we say, routine work leading up to what would be the next phase of a proprietary compound. That is part of our case.

A. You say "routine work". I do not believe it was routine."

134. It should be noted that Professor Baldwin's view was that the idea of moving away from 2,3 compounds was not routine, by which I understand him to mean that it was not obvious. I do not read his evidence as suggesting that if the idea of trying other substitutions was obvious, then the 3,4 substitution with methylsulphonyl was not an obvious one to try.
135. Professor Baldwin gave his opinion of what a medicinal chemist would think and do. Professor Baker was such a person, although more highly skilled and imaginative than the notional unimaginative skilled person. He set out in his witness statement the role of such a chemist in a passage that was not challenged in cross-examination. It was reproduced in full by the judge.

"13. It is clear that the skills of the medicinal chemist are central to the task of developing a new drug. Professor Baker described the role of the medical chemist in the development of a modern drug as follows, in a part of his evidence which was not challenged.

The starting point for a drug discovery programme is to identify a pharmacological effect or an important feature of pharmacology which allows the establishment of an in vitro or in vivo assay. The assay will be dependent on the identification of an active site on an enzyme or target receptor which will be responsible for controlling a biological response. The basis of drug therapy, in its broadest terms, is the use of discrete chemical compounds to bind to the target thereby blocking the normal action of the enzyme or receptor. Once a target has been identified and, notwithstanding that the target would be associated with many different aspects of pharmacology, the fundamental aspects of medicinal chemistry programmes would follow similar lines. It is particularly important that a suitable in vitro assay is developed so that it can be used for assessing the biological activity of candidate compounds.

Once one or more lead compounds have been identified, it is normal procedure for a patent and literature search to be undertaken to find out what was known about such compounds including how they can be synthesised and what other similar compounds might have been investigated. It would be then necessary to identify the key features of the lead compounds which determine their ability to bind to the enzyme or receptor target.

To identify the key features of a lead compound it is common practice to synthesise and screen a number of structurally closely analogous compounds, which could be isomers or compounds differing by say one functional group at a time. In such a screening programme one would expect such closely analogous compounds to possess similar biological properties to the lead compound albeit that there will be differences in potency and/or specificity. The medicinal chemist would be particularly seeking molecules which would have good bioavailability and would also have in mind the question of toxicity. Some types of molecules are known to be associated with toxicity and the programme would be directed in such a way to avoid potential problems. The overall aim is, as far as possible, to develop a perfect compound with good activity, selective action, good in vivo properties, and a lack of any adverse effects in patients.

The process described by Professor Baker as the identification of the key features of the lead compounds which determine their ability to bind to the enzyme or receptor target is called producing an SAR or structure activity relationship. Elsewhere in his report Professor Baker says that it is standard practice for the medicinal chemist to generate an SAR."

136. There can be no doubt that at the priority date of the patent, the target of new NSAIDs which were COX II selective was known. Further DuP 697 was a lead product as it showed reduced side effects. That being so, the medicinal chemist would, according to Professor Baker, identify the key features and "To identify the key features of a lead compound it is common practice to synthesise and screen a number of structurally close analogous compounds, which could be isomers or compounds differing by say one functional group at a time."
137. In his witness statement Professor Baker stated that the ordinary medicinal chemist would have approached the task of finding an effective alternative to DuP 697 by investigating its close analogs. He went on to give reasons why by 1992 the 3,4 analog would have been tested as a matter of course.
138. I must come to the cross-examination of Professor Baker, but before doing so, draw attention to the crucial difference between his evidence and that of Professor Baldwin. Professor Baldwin was a distinguished organic chemist who had experience in industry. He did not believe it was obvious to investigate the isomers of DuP 697, but did not dissent from the view that the 3,4 could have been an obvious one to try if the isomers were to be investigated. Professor Baker was of the view that it was routine to investigate a lead compound, such as DuP 697, and that the 3,4 substituted ones were compounds which were analogous and obvious ones to look at. The difference between them is whether the step of investigating analogous compounds was obvious.
139. In his cross-examination Professor Baker maintained his view that it was obvious to make the 3,4 substituted form of DuP 697. However, Mr Kitchin submitted that his evidence should be disregarded because his opinion was tainted by a mosaic of documents not permissible in law. He relied on this piece of the cross-examination (Ev. 11, 1412 and 1413).

"Q. I am obliged, professor. Is it not right that a reader of this document, looking at Haber 2, will understand that the inventor perceived that sulphur in the heterocycle is very important because it is common to every molecule?

A. No. I think that is not the correct conclusion you draw. If we could make five-membered ring compounds, which are very useful templates, without a sulphur or other heteroatom, we would do so. Therefore, the preparation of thiophenes is because they are convenient, unable to make them, and they have certain chemistry which allows you to get the compound itself; but it does not mean you make a thiophene, you necessarily believe the sulphur has an importance. In fact, as we would know, the sulphur does not have a great importance in the binding of these molecules to the receptor.

Q. We know that now.

A. We would guess it then and it is confirmed now. We were talking about intuition at the time and the conclusions drawn, and I am giving what you I believe are those conclusions.

Q. Looking at Haber 1 and comparing it to Haber 2, the sulphur in the heterocycle is consistent throughout, is it not?

A. It is the same man and it is the same group making a series of compounds. That is the way that chemists often work, yes.

"Q. In each case he has assembled his molecule. He has synthesised his molecule with the aryl groups at the 2 and 3 positions.

A. Yes, because he would have to devise a new synthesis of those if he moved it around. In the terms of his own work and importance of that work, he may not wish to have the time to do such development. Someone reading this would also have in mind the question of Brown, and other publications, which does mention 3,4 substitution."

140. I do not accept that that passage affects the clear evidence given by Professor Baker, that the medical chemist would investigate the isomers. If so, the 3,4 substituent was an obvious one to try: but the important point is that even if Brown at all times was in his mind, that could not have any bearing upon whether it was routine work to investigate analogous compounds such as the 3,4 substituted pattern. Thus assuming Mr Kitchin's criticism is valid, it does not impinge upon the issue which split the experts. In any case, I do not accept the criticism. The witness was taken to the other documents and asked to comment. It was therefore not surprising that he referred to Brown.
141. For my part I would have preferred the evidence of the medicinal chemist, Professor Baker, as to whether it was routine to investigate close compounds, to that of Professor Baldwin. Certainly the judge was entitled to do so and to come to the conclusion he did in paragraph 171 of his judgment:

"171. Structurally, there is a clear similarity between DuP 697 and its close 3,4 disubstituted analogues. It is a matter of reasonable prediction that they will have similar activity. I think that a medical chemist wishing to investigate the structure/activity relationship of DuP 697 would think of making its 3,4-diaryl analogues, with a view to seeing whether they are active. I also think that confronted with DuP 697 and required to develop a novel compound of similar activity, the 3,4-diaryl substitution is one of the first things which would occur to the medical chemist. The claimants point to DuPont. Why, they say, if it was so obvious to use the 3,4-diaryl substitution, did not DuPont do so. The answer to this is, I think, threefold. First of all, DuP 697 was a good compound: it was not commercialised not because it had poor bioavailability but because it has too long a lifetime in the body. Second, it was comparatively straightforward to synthesise. Third, Dr Galbraith's evidence suggests that there was a degree of inertia in this field, which was only dissipated with the discovery of the inducible Cox II enzyme. All the evidence gave me the clear picture that the 3,4 diaryl compounds were obvious to try for any skilled person knowing of DuP 697, and the ones which were most obvious to try (Examples 1 and 2 of the specification) are both Cox II selective, anti-inflammatory, and gastric sparing. It follows that in my judgment claim 1 both as granted and as sought to be amended is obvious in the light of Dr Galbraith's disclosure, which provides the impulse to test for Cox II selectivity. Given also that the existence of the inducible isoform of the COX enzyme was in my view common general knowledge at the priority date, it was obvious to investigate the compound of Gans, which was said to be gastric sparing, to see why."

142. Before the judge Merck relied on what happened at the patentees and at Merck to support their submissions on obviousness. The patentees relied on published documents of Du Pont to show that it had not been routine or obvious to Du Pont to investigate the 3,4 substituted isomer of DuP 697.
143. The judge correctly set out the facts of what happened, but does not appear to have relied upon them as important when deciding whether the step from DuP 697 to the claim was obvious. That appears to reflect the need to be cautious when being asked to draw inferences from secondary evidence of that type.
144. Merck also contended that the claims of the patent as granted and amended were obvious having regard to the disclosure in Brown. The judge held claims 1 and 20 lacked novelty over Brown. He went on to hold that claims 1, 7 and 20 of the amended patent were obvious having regard to the disclosure in Brown.
145. I have already concluded that claims 1, 7 and 20 of the unamended patent were anticipated. I also believe that the limitation sought to be introduced by amendment does not introduce invention. The judge was right when he said:

"154. Against this, the claimants submit that the disclosure is a disclosure of a less-preferred method. This is, of course, irrelevant. It cannot matter if the disclosure is less preferred, since we are here concerned with a description of the compounds themselves, not with determining whether there are clear and unambiguous directions to do or make something within the claim. Second, they submit that there is no disclosure of the 4– substitution of the methylsulphonyl group. This is correct. I do not think that the 4– substitution is expressly disclosed, but it is plainly obvious, since any person wishing to substitute one of the Ar rings of Brown with a methylsulphonyl group in accordance with Brown's specific teaching will, on Professor Baldwin's evidence, go for the easiest substitution, which is at the 4 position. It is not suggested that Brown does not give sufficient directions for the intermediates to be prepared."

Added Matter

146. In view of the conclusion I have reached, there is no need to deal with the issues raised. The allegations that the proposed amendments add matter and are therefore not allowable are irrelevant as the Court will not allow an amendment which will leave a patent invalid. As to the allegation that the limitation in claim 1 to R² and R³ being substituted with methylsulfonyl or sulfamyl added matter, I believe that the judge was right for the reasons given in paragraphs 137 and 138 of his judgment.

Industrial Application

147. Mr Young submitted that the patent was invalid as it was not capable of industrial application (see section 1(1)(c) of the 1977 Act). He submitted that claim 1 covered compounds that were not useful and for that reason they were not capable of industrial application. To support his submission, he referred to the judgment of Morritt LJ in Chiron Corp. v Murex Diagnostics Ltd [1996] RPC 535 at 607. Mr Kitchin relied on the EPO Guidelines. He submitted that the judgement in Chiron should be read as being consistent with those guidelines. He may be right, but that issue is best left to a case where it is an essential issue.

Conclusion

148. The patent is invalid. That being so, it cannot be infringed. The appeal must be dismissed. I see no point in making an order on the respondent's Notice, but believe it right to hear counsel on this.

LORD JUSTICE SEDLEY:

149. I agree with both judgments.

LADY JUSTICE ARDEN:

150. There are a number of factors which make this case remarkable even by the standards of patent law. First, as is apparent from the description of the patent in suit in the judgment of Aldous LJ, claim 1 sets out a formula with numerous permutations. Indeed we were told that there were literally trillions of formulae comprised within claim 1. Second, at the trial each side called several experts in support of its contentions and among them on each side were some of the leading experts in the field. The matters on which they gave evidence and indeed disagreed were not only complex but related to matters at the very edge of knowledge in biochemistry. Third, one of the issues in this case is whether the patent in suit on its true interpretation includes a tautomer which is not specified in the patent in suit but into which one of the specified substituents is automatically converted when in aqueous solution and with which such substituent exists in constant equilibrium. Pumfrey J describes tautomerism from the technical point of view in paragraphs 175 and 185 of his judgment. The tautomer is a variant of an invention which (according to counsel's researches) has not previously considered by the courts.
151. I gratefully adopt the statement of the facts in this case as set out in the judgment of Aldous LJ. I concur with the observations in paragraph 3 of his judgment.

Issues

152. I deal with the issues on this appeal under the following headings:-

1) Interpretation.

2) Direct infringement.

3) Indirect infringement.

4) Insufficiency .

5) Priority.

6) Added matter.

7) Novelty.

8) Obviousness.

9) Industrial application.

10) Amendment.

The appellate function

153. The question of the interpretation of the patent in suit is in the final analysis a question of law for the court. All the other questions in this case (apart from amendment) are mixed questions of law and fact. In the case of questions of fact, the Court of Appeal must, unless it is satisfied that the judge has not directed himself in law correctly, only interfere if it is satisfied that the judge was not entitled to come to the conclusion on the facts that he did. If the conclusion to which he came was open to him on the evidence, the Court of Appeal should not interfere. The reasons for this have been explained in several cases, including Biogen Inc v Medeva plc [1997] RPC 1 per Lord Hoffmann, with whom the remainder of the House agreed. Lord Hoffmann explained that:

"It is because specific findings of fact, even by the most meticulous judge, are inherently an incomplete statement of the impression which was made upon him by the primary evidence. His expressed findings are always surrounded by a penumbra of imprecision as to emphasis, relative weight, minor qualification and nuance (as Renan said, la vérité est dans une nuance), of which time and language do not permit exact expression, but which may play an important part in the judge's overall evaluation. It would in my view be wrong to treat Benmax as authorising or requiring an appellate court to undertake a de novo evaluation of the facts in all cases in which no question of the credibility of witnesses is involved. Where the application of a legal standard such as negligence or obviousness involves no question of principle but is simply a matter of degree, an appellate court should be very cautious in differing from the judge's evaluation." (page 45)

154. So far as the question of amendment is concerned this Court is only concerned with the question whether the judge was correct that the requirements for a valid amendment were satisfied since the question of whether as a matter of discretion the amendment should be allowed has yet to be determined by the judge.

1. Interpretation of the patent in suit

155. The judge held that the invention in the patent in suit lay in the provision of a class of chemicals which was not merely anti-inflammatory but also had fewer and less drastic side effects due to Cox II selectivity (judgment, paragraph 46). The judge's findings on sufficiency, priority and added matter all turned on his interpretation of the patent in this respect.
156. A second issue on interpretation arises, namely whether the patent covers the tautomer of the furan OH in formula l. This is relevant to the issue of infringement.
157. In Dyson v Hoover the court applied the statement of the approach to construction set out in the judgment of Aldous LJ in Wheatley v Drillsafe Ltd [2001] RPC 133:

"18. ….. Section 125 of that Act provides that an invention shall be that specified in the claim "as interpreted by the description and any drawings … shall be determined accordingly." The extent of protection is not only important when considering whether an alleged infringement falls within the claim, but also when considering validity. A patent will be invalid if the extent of protection includes within it the prior art or something which was obvious having regard to the prior art.

19. Section 125(3) requires the Protocol on Interpretation of Article 69 of the EPC to be applied. It states:

"Article 69 should not be interpreted in the sense that the extent of the protection conferred by a European Patent is to be understood as that defined by the strict, literal meaning of the wording used in the claims, the description and drawings being employed only for the purpose of resolving an ambiguity found in the claims. Neither should it be interpreted in the sense that the claims serve only as a guideline and that the actual protection conferred may extend to what, from a consideration of the description and drawings by a person skilled in the art, the patentee has contemplated. On the contrary, it is to be interpreted as defining a position between these extremes which combines a fair protection for the patentee with a reasonable degree of certainty for third parties."

The Protocol outlaws what can be termed strict literal and also liberal interpretation using the claims as a guideline. The correct approach is to achieve a position between those extremes "which combines a fair protection for the patentee with a reasonable degree of certainty for third parties."

158. It is important to analyse the process of interpretation required for Protocol questions from the position of the court. There are similarities and differences between the court's approach to Protocol questions and its approach to the interpretation of contracts and statutes. In construing contracts and statutes, the court does not take a purely literal meaning of words, if indeed such meanings exist. On the other hand when construing a contract the court aims to find the intention of the parties as expressed in the document in the light of the factual matrix of the transaction and, in the case of a statute, the intention of the legislator as expressed in the enactment. In the case of a contract, the factual matrix to which the court can have regard includes all relevant information that would have been available to the parties. In the case of a statute the court can in limited circumstances also look at material outside the statute such as a Law Commission report setting out the mischief to which a reforming measure was directed or indeed in some cases to parliamentary material itself. Under the Protocol, intention is relevant in only a limited sense. It is necessary to ascertain what the patentee contemplated (and this process seems to bear some similarity with the approach to the interpretation of a will where the court puts itself "in the testator's armchair"). However, this is only one step in the interpretation process. The final step in the process requires the court to find the interpretation which balances the competing consideration of fair protection for the patentee with a reasonable degree of certainty for third parties.
159. There are of course other similarities and differences between on the one hand the court's approach to the interpretation of contracts and statutes and on the other hand the court's approach to the interpretation of patents. In particular, the purposive interpretation of the patent is from a scientific point of view since the court approaches the question of construction in the light of the common general knowledge of persons skilled in the relevant art at the date of the filing of the patent. In addition, the approach to interpretation provided for by Article 69 is no doubt one which is to serve the general purposes of the European Patent Convention, in particular strengthening co-operation between the signatories to the Convention in respect of patent inventions.
160. It is occasionally possible for the court to fill gaps in contracts by the process of interpretation, for instance where a term is implied to give a contract business efficacy. It is perhaps less common to fill a gap in a statute by interpretation. Under the Protocol it may be necessary to fill a gap or more generally to depart from the precise terms of the patent in suit but only where the court is satisfied that this can be done as a matter of interpretation.
161. The Protocol introduces concepts of reasonableness in relation to certainty and fairness in relation to the degree of protection to be given to the patentee. Reasonableness and fairness require the exercise of judgment but argument was not addressed to the relevant criteria to be used when forming this judgment or the weight to be given to them and they have yet to be articulated by the courts. The weight which the law attaches to (say) reasonable certainty as opposed to fairness to the patentee may well be guided by policy considerations such as that identified by Lord Hoffmann in Biogen Inc v Medeva plc:

"…. care is needed not to stifle further research and healthy competition by allowing the first person who has found a way of achieving an obviously desirable goal to monopolise every other way of doing so (see Merges and Nelson On the Complex Economics of Patent Scope [1990] 90 Columbia Law Review 839)."

See further for example the observations of Lord Wilberforce on selection patents in Du Pont (Witsiepe's Application) [1982] FSR 303 at 309. On the other hand it would be unfair to the patentee to construe a patent beyond its intended ambit (see per Aldous LJ in Wheatley v Drillsafe Ltd, above, at pages 141 to 143). To do so may make his patent invalid.

162. An analogous problem to that of filling gaps in statutes arises in patents in relation to what are known as "variants". In this field the court traditionally derives assistance from the questions identified by Hoffmann J in Improver Corporation v Remington Consumer Products Ltd [1990] FSR 181:-

"(1) Does the variant have a material effect upon the way the invention works? If yes, the variant is outside the claim . If no-

(2) Would this (i.e. that the variant had no material effect) have been obvious at the date of the publication of the patent to a reader skilled in the art? If no, the variant is outside the claim. If yes-

(3) Would the reader skilled in the art nevertheless have understood from the language of the claim that the patentee intended that strict compliance with the primary meaning was an essential requirement of the invention? If yes, the variant is outside the claim.

On the other hand a negative answer to the last question would lead to the conclusion that the patentee was intending the work or phrase to have not a liberal but a figurative meaning (the figure being a form of synecdoche or metonymy) denoting a class of things which included the variant and the literal meaning, the latter being perhaps the most perfect, best-known or striking example of the class."

Hoffmann J went on to explain that the first two questions involved

considerations of fact, but the last was a question of interpretation.

163. The courts use the "Improver" questions as a tool of analysis and the disciplined approach it offers promotes consistency and transparency in this important field. The habits of those who draft patents are no doubt conditioned by the law's requirements, and that is another reason for following a structured approach. The terms of the Protocol have the beneficial effect of impelling the drafter of a patent towards using the maximum precision in his patent. On the other hand, if he is too precise, he may find himself impaled upon the third Improver question. The emphasis the Protocol places on the language of the patent is to be noted: the court has held that the court cannot ignore a step inserted into a patent (Société Technique de Pulverisation [STEP] v Emson Europe Ltd. [1993] RPC 513). However, it bears repetition that the Improver questions are no more than guidelines: the final conclusion on interpretation must be found by asking the Protocol questions. But they make it clear that the touchstone for a variant to be within a patent is not that it is necessary so that the monopoly granted by the patent is effective. On the contrary the variant must be immaterial, obvious and consistent with the language of the patent.
164. The identification of a previously unmade chemical compound would not itself be patentable. It is possible to create different compounds without any inventive step. Patentability requires there to be a technical contribution, that is the compounds must solve a technical problem. As the Technical Board of Appeal of the European Patent Office said in AgrEvo UK Limited (T939/92 [1996] OJ EPO 309), the selection of chemical compounds "…. in order to be patentable, must not be arbitrary but must be justified by a hitherto unknown technical effect which is caused by those structural features which distinguish the claimed compounds from numerous other compounds", and in addition "a technical effect which justifies the selection of the claimed compounds must be one which can be fairly assumed to be produced by all the selected compounds."
165. I thus reject the initial starting point of the appellant, that the mere identification of a hitherto unknown class of chemicals can be patentable even if the class does not have novel unifying properties.
166. The appellant's main submission, however, is that the compounds described in claim 1 have hitherto unknown anti-inflammatory and analgesic effects. The patent made no promise that the compounds would be gastric sparing or Cox II selective though by applying the information provided by the assays in the specification the skilled reader would be able to find compounds which were gastric sparing or Cox II selective. The appellant submits that this is the true interpretation of the invention. Thus the title page of the patent refers to the invention as "novel 3, 4, - diaryl thiophenes and analogs thereof having use as anti-inflammatory agents". The passage in the specification headed "Background of the Invention" explains that prostaglandins play a major role in the inflammation process and that the inhibition of prostaglandin production has been a common target for anti-inflammatory drug discovery. There is then a reference to the fact that the drugs usually given (called NSAIDs) can produce severe side effects, including life threatening ulcers, that limit their therapeutic potential. The specification then explains that recently a form of inducible Cox II enzyme has been discovered which produces fewer and less drastic side effects. The thiophene compounds disclosed in the patent selectively inhibit Cox II over Cox 1 and relieves the effects of inflammation while producing reduced side effects. Under the heading "Description" the specification states that a class of compounds "useful in treating inflammation related disorders" is defined in formula l.
167. According to the appellant, its interpretation of the patent is supported by the language of the patent. The specification states that the invention "also includes compounds which selectively inhibit [Cox II] over [Cox 1] and do not significantly inhibit one or more other arachidonic pathway steps …." (This is referred to below as "the bridging passage"). Moreover, there are statements in the specification to the effect that capacity to inhibit TXB 2 and to inhibit Cox II over Cox 1 are preferable and more preferable respectively. Moreover the specification states that "such preferred activity may indicate [not "indicates"] an ability to reduce the incidents of common NSAID - induced side effects such as ulcers".
168. The respondent submits that the claimed technical contribution is that the compound should be gastric sparing or Cox II selective. According to the respondent, the patent promises reduced side effects, and those side effects referred to in the specification concern stomach erosion. The respondent relies on the common general knowledge as found by the judge (and as to which there is no real dispute), and in particular that a skilled person would know of the discovery of Cox II and that the inhibition of Cox I was a viable target for developing NSAIDs. The respondent points in particular to a sentence under "Background of the Invention" which states that "The discovery of an inducible enzyme associated with inflammation provides a viable target of inhibition which more effectively reduces inflammation and produces fewer and less drastic side effects." The concern of the patent is with continuing efforts to find a safe and effective anti-inflammatory agent, "safe" meaning without harmful stomach erosion. Indeed at the time of the grant of the patent it was thought that the solution to the problem of gastric erosion lay solely in Cox II selectivity and the inhibition of Cox 1.
169. The respondent also contends that the examples given by the patent are of Cox II selectivity. The in vitro tests demonstrate Cox 1 and Cox II activity and all but two of the in vivo tests showed Cox II selectivity and one of those which did not show Cox 1 selectivity was the comparator. The judge found that the results in the other such test were equivocal.
170. The judge held that the patent related to a class of compounds having anti-inflammatory and/or analgesic activity with fewer and less drastic side effects, the reduction in side effects being due to Cox II selectivity. He considered that the bridging passage supported the respondent's case, on the grounds that the lack of inhibition of one or more of the arachidonic pathway steps was an additional feature. He considered that the examples in the patent supported his interpretation. He held that the specification promised reduced side effects. He considered that the whole thrust of the specification was towards Cox II selectivity.
171. In my judgment, the judge's interpretation is correct. There are three candidates for the technical contribution of this patent: anti-inflammatory/analgesic quality, gastric sparing quality and Cox II selectivity. A literal construction of the patent supports the first of these possibilities. There is no actual promise of reduced side effects by Cox II selectivity. Indeed the specification is careful to say that Cox II selectivity "may" indicate an ability to reduce the incidence of common NSAID reduced side effects such as ulcers. On the other hand, as explained above, the court cannot stop at the literal construction. Account has to be taken of the fact that the problem to which the patent directs itself is that of the severe side effects of NSAIDs: see the passage headed "Description" in the specification, for example. There is no other side effect identified apart from gastric erosion. Side effects and gastric erosion are a recurring theme of the specification. This would support interpreting the technical contribution as a reduction in gastric irritancy. However the common general knowledge as found by the judge was that Cox II was a viable target for reduction in gastric irritancy. He found that a skilled person would know that this was the most likely route for finding improved drugs for anti-inflammatory purposes. There are references in the specification to Cox II selectivity and moreover the examples support the theory that the technical contribution was to be obtained by Cox II selectivity. The specification recites the very recent developments in the field. The bridging passage and the passages starting "preferably" and "more preferably" are identifying subsets, for example in the case of the bridging passage compounds which are both Cox II selective and Cox I inactive. As I read the patent, a reduction in gastric irritancy by Cox II selectivity was also the technical contribution which the patentee aimed to make. To restrict the patentee to this technical contribution reduces the scope of the patent and this safeguards the position of third parties. Accordingly, applying the Protocol, I consider that the judge's interpretation is the proper interpretation of the patent. As I have said, this conclusion has implications for other issues.
172. The next question is whether the patent includes the keto tautomer of the explicitly claimed enol form of hydroxy (OH), an expressly permitted substituent for X in formula 1. I am, of course, examining the patent at this stage on the basis that it is valid. The judge concluded that the patent did not include the keto tautomer (judgment paragraph 195). The appellant contends that the respondent's molecule, MK966, fell within claim 1 as it was a tautomer of two compounds called enol and enolate which fell within claim 1. The appellant goes on to argue that the patent in suit is infringed in one of two ways: either through the manufacture abroad of MK966 and its importation into this country, or when MK966 is administered to the patient. MK966 is produced in tablet form. It is manufactured abroad and the appellant contends that the presence of enol and enolate forms of the compound during production infringes claim 20. The first way of putting the case would involve a direct infringement within section 60 (1) (a) of the Patents Act 1977 ("the 1977 Act"). The other way of putting the case involves indirect infringement contrary to section 60 (2) of the 1977 Act.
173. The judge was impressed by the fact that the specification set out the compounds to which it applied in considerable detail but made no mention of any tautomer. In this case the specified tautomer is in fact the minor tautomer, and the keto form is the major tautomer. The position might have been different if the formula had been more general in its terms. The judge held that the skilled man would know of tautomers as part of his common general knowledge but "the natural conclusion to draw when one species is mentioned to the exclusion of the other is that the species not mentioned is not intended to be covered. It is not for the skilled person to speculate why this should be so, as STEP v Emson points out" (judgment paragraph 195).
174. The judge did not consider that the Protocol questions led to any different conclusion. If the wording was precise and unambiguous, it accorded with the patentee's intention and third parties were safeguarded (see paragraph 197).
175. The judge expressed difficulty with the third Improver question though his reasoning suggests that he would have answered it "yes". It is not clear to me that he asked the first and second Improver questions, but if he did it would appear that he answered them "no" and "yes" respectively. The respondent contends that the answer to the first question is "yes, given that the specified tautomer is at any point in time only one-sixty-three billionth part of the MK966 molecule, is undetectable (its presence has to be inferred) and has no active pharmaceutical effect, where the major tautomer was the active ingredient of MK966. (The enol is in fact continuously tautomerising so up to 50% of the relevant ingredient in MK966 could be affected over the period that MK966 might be expected to be in the body). I do not consider that the respondent makes the right comparison: the first Improver question involved deciding whether the molecule into which enol tautomerises in aqueous form is material to claim 1 in the patent in suit. Even though the formula in claim 1 is highly specific, the answer to that question must be "no" because it is an integral and indivisible part of a substituent in claim l in aqueous form. On Improver question (2), the respondent argues that the keto form was an entirely different type of molecule and would behave differently from the enol form in its ability to bind to the substance. Here the second Improver question taken literally does not really aid interpretation because the skilled reader of the patent who used the enol would have no choice about having the keto form as well if the enol was placed in aqueous solution. The real issue is whether the skilled person would have known how the tautomer would work and the answer to that is "yes". Accordingly, I turn to Improver question (3).
176. On this question, on the facts of this case, I do not agree with the judge's holding that "the natural conclusion" to draw is that constituents not mentioned were expressly excluded. That cannot be a general rule. It would mean that the terms of a patent could never include an unspecified variant. The position here is that the tautomer was absolutely inseparable from the specified substituent, in aqueous form (which was within the patent) and while the claims are indeed very extensive and specific they do not (save for one reference under compound 16) use terms which would mean that tautomers were excluded. The tautomer could not affect the claims in the patent (if valid). The patent must be given a purposive interpretation and to that end the literal meaning must give way to one which in the context more accurately conveys the patentee's meaning without removing reasonable certainty from third parties. Since the skilled person on the judge's finding knows that the keto form exists in equilibrium with the specified enol in aqueous form and accordingly that it cannot be divided up from the enol in that form, third parties are not in my judgment unreasonably prejudiced by the interpretation that enol includes its tautomer. In my judgment, such an interpretation can be reached as a matter of interpretation of the wording of the patent. In my judgment the word "hydoxy" is descriptive and covers both the enol form of that name and the keto form into which the enol form is constantly and ineluctably interconverting when in solution.

2. Direct Infringement

177. The judge held that the claim for infringement failed.
178. As regards direct infringement, it follows from my conclusion that, as the patent in suit includes the enol and the enolate, the respondent infringed claims 1 and 13 by the importation and sale within the United Kingdom of MK966.

3. Indirect Infringement

179. In the circumstances it is not necessary for me to deal with the arguments of indirect infringement. In summary, the appellant puts its case on indirect infringement in two ways. First it is said that section 60 (2) is infringed when MK966 is administered to the patient and is dissolved as an enol in the body. Alternatively it is said that it is converted into the enol form through the metabolic process.
180. Leaving aside the question of whether any relevant finding of fact by the judge is reviewable on appeal, this part of the appellant's case raises difficult questions as to the construction of section 60 (2) of the 1977 Act. This provides:

"Subject to the following provisions of this section, a person (other than the proprietor of the patent) also infringes a patent for an invention if, while the patent is in force and without the consent of the proprietor, he supplies or offers to supply in the United Kingdom a person other than a licensee or other person entitled to work the invention with any of the means, relating to an essential element of the invention, for putting the invention into effect when he knows, or it is obvious to a reasonable person in the circumstances, that those means are suitable for putting, and are intended to put, the invention into effect in the United Kingdom."

181. This provision, we are told, provides a statutory basis for secondary infringement and it was designed to incorporate into United Kingdom law article 26 of the Community Patent Convention, even though that convention has not yet come into force. It applies to all sorts of patents. It is clear that the provision is intended to have a wide application given the open-textured nature of the words used. This can also be seen from the fact that a person may be liable for indirect infringement even though the person supplied uses the product for a private purpose and is thus not himself responsible for an infringement (see section 60 (5), (6)). However, if the appellant is right, section 60 (2) can be applicable where what happens is a purely natural result of the administration of non-infringing matter which is the only active ingredient. The proposition that section 60 (2) has such sweeping effect is on the face of it a surprising conclusion. As the question does not arise in the light of my conclusion on direct infringement, I express no view on it.

4. Insufficiency

182. The judge held that the patent as granted and as proposed to be amended was insufficient because it covered compounds which were not selective Cox II inhibitors or were not gastric sparing or did not have anti-inflammatory activity.
183. The applicable principles of law were common ground. To be valid a patent must be supported by enabling disclosure. This means that the specification must disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art (see section 5 (2) (a) as construed in conjunction with section 14 (3) and section 72 (1) (c) of the 1977 Act in Biogen Inc v Medeva plc (above)). In addition the specification must enable the invention to be performed to the full extent of the monopoly claimed. Accordingly

"if the invention discloses a principle capable of general application the claims may be in correspondingly general terms. The patentee need not show that he has proved his application in every individual instance. On the other hand if the claims include a number of discrete methods or products the patentee must enable the invention to be performed in respect of each of them. Thus if the patentee has hit upon a new product which has a beneficial effect but cannot demonstrate there is a common principle by which the effect will be shared by other products of the same class, he will be entitled to a patent for that product but not for the class, even though some may subsequently turn out to have the beneficial effect: see May and Baker Ltd v Boots Pure Drug Co Ltd (1950) 67 RPC 23, 50. On the other hand if he had disclosed a beneficial property which is common to the class he will be entitled to a patent for all the products of that class (assuming them to be new) even though he has not himself made more than one or two of them" (per Lord Hoffmann in Biogen Inc v Medeva plc at pages 48 – 49).

184. The court approaches the specification through the eyes of a person who is skilled in the field, not through the eyes of a leading expert in the field. The reader of a patent may be expected to carry out normal routine trials (Mentor Corporation v Hollister Inc. [1993] RPC 7).
185. In relation to insufficiency, the onus is on the defendant. In the light of my conclusions on the interpretation of the patent, the respondent had at trial to establish that a substantial number of compounds within claim 1 did not possess the unifying characteristic of Cox II selectivity. For this purpose the respondent carried out a large number of tests. (The tests were also relevant to their arguments on priority and added matter.) The judge was satisfied that the respondent discharged the onus of proof on it. For instance in relation to class F of these compounds, 14 were tested of which 10 were inactive and 4 had marginal activity. The appellant had carried out his own tests which on the respondent's criteria showed that out of 23 compounds, 12 were active and nearly all of the remainder were inactive. Mr Kitchin attacks these results in two ways. First he says that the case which the appellant had to meet at trial was proved simply that all of the class failed to meet their criteria and that they had not achieved this. Second, various aspects of the tests were criticised. There was a failure to sensitise or optimise the assays. It was said that no skilled person would follow the instructions and specification blindly. The critical criteria chosen by the respondent (IC 50 and 100 micromolar units) were arbitrary. The respondents were testing the compounds for commercial products. The defendants had given no explanation as to why they had chosen the particular compounds which they had chosen. Moreover, on some of the results, there was precipitation showing that not all of the compounds had dissolved and so the results were not reliable. However, this occurs only at 1000 micromolar units. The appellant challenges the validity of the respondent's experiments even if it fails on the question of interpretation of the patent.
186. I have considered the appellant's submissions and agree with Aldous LJ for the reasons that he gives that the appellant have not discharged the onus on it of showing that the judge's findings of fact and conclusions with regard to insufficiency were not findings and conclusions which he was entitled to make and reach respectively.

5. Priority

187. On 15 January 1993 the appellant filed a priority document for the patent in suit. This is headed "novel 3, 4 – diaryl thiophenes and analogs thereof having use as anti-inflammatory agents". The priority document makes it clear that the compounds are to treat inflammation and pain in mammals and in particular the invention related to compounds "having anti-inflammatory and/or analgesic activity without erosion of the stomach and therefore more effective and safe." The judge held that the patent in suit lacked priority because the invention was not disclosed in the priority document and because certain compounds in the priority document were not Cox II selective and/or gastric sparing. This raises the issues as to the judge's findings on the experiments with which I have dealt above.
188. The appellant's primary submission on this point is that the judge went wrong in principle in identifying the invention. This was the only basis on which it could be said that the priority document failed. As I have found against the appellant on the interpretation of the patent, it follows that the challenge to the judge's finding on priority must fail.
189. The resolution of this issue is important because it is common ground that the respondent's compound MK966 was published on 10 January 1994, just four days before the date of the application for the patent in suit and has priority over claims 1, 5 to 8, 13, 20 and 27 to 31 in the patent in suit if the latter loses its priority.

6. Added Matter

190. Section 72 (1) of the 1977 Act provides:

"72. (1) Subject to the following provisions of this Act, the court or the comptroller may on the application of any person by order revoke a patent for an invention on (but only on) any of the following grounds, that is to say-

…….

(c) the specification of the patent does not disclose the invention clearly enough and completely enough for it to be performed by a person skilled in the art;

(d) the matter disclosed in the specification of the patent extends beyond that disclosed in the application for the patent, as filed, ….. "

191. Accordingly, matter is improperly added to the description of an invention in the application for a patent so as to give rise to an objection to validity under section 72 (1)(d) if it is matter relevant to the invention (see per Aldous J in Bonzel v Intervention Ltd (No.3) [1991] RPC 553, 574).
192. The judge held that the patent in suit (as granted and as proposed to be amended) contained added matter. I can deal with this matter shortly. The amendments to the patent makes it necessary to have as a substituent a 4-methylsulfonyl phenyl (or 4-sulfamyl phenyl) group. This was not required in the priority documentation or the patent application. The patent claims relates to a much smaller class of compounds. What the patentee had done was in effect to make a new selection of compounds. This could be done only by fresh application. In the light of my interpretation of the patent in suit, the challenge to the judge's conclusion on the point must fail.

7. Novelty

193. In order to be patentable, an invention must be "new" (1977 Act, s.1 (1), section 2 (1) and (2) of the 1977 Act provide:

"2. (1) An invention shall be taken to be new if it does not form part of the state of the art.

(2) The state of the art in the case of an invention shall be taken to comprise all matter (whether a product, a process, information about either, or anything else) which has at any time before the priority date of that invention been made available to the public (whether in the United Kingdom of elsewhere) by written or oral description, by use or in any other way."

194. Information is available for this purpose if it is available to the public anywhere. A patent may be said to be anticipated by a prior invention which forms part of the prior art. For a claim to be anticipated by prior disclosure, the prior disclosure must contain a clear description of , or clear instructions to make, something which would infringe the patentee's claim if carried out after the grant of the patentee's patent. In the time-hallowed words of Sachs LJ in General Tire and Rubber Co v Firestone Tyre and Rubber Co Ltd [1972] RPC 457:

"A signpost, however clear, upon the road to the patentee's invention will not suffice. The prior inventor must be clearly shown to have planted his flag at the precise destination before the patentee." (at page 486)

195. The judge held that the patent in suit (as granted and as proposed to be amended) was anticipated by US Patent 3,743,656, which I shall call Brown.
196. Brown was patented on 3 July 1973. Brown concerns diaryl furan and thiophene lower alykanic acids and derivatives which are pharmacologically efficacious as anti-inflammatory agents. The invention is said to consist of "furans, thiophenes .... and an aliphatic acid group."
197. The judge held that Brown anticipated claim 7 of the patent in suit. The judge held that claim 7 covered the aliphatic acid substituent but that claim 1 (somewhat inconsistently) did not do so. The appellant contends that the judge was wrong on this point. The Brown disclosure was of a less preferred compound. Moreover Brown did not disclose the 4 – methylsulphonyl phenyl substitution. However, the judge rejected the first ground on the basis that whether or not the disclosure was less preferred was irrelevant. The Brown disclosure was sufficiently clear. In addition the judge rejected the second ground on the basis that the 4- substitution was obvious. So far as this is concerned, since obviousness was a question of fact and as will be seen from the discussion under the next issue the judge was entitled to reach the conclusion that he did, I also reject this ground.
198. The respondent challenges the judge's conclusion that claim 1 was not also anticipated by Brown. This issue involves a question of the interpretation of the term "acyl" as stated in the specification. This specification states that "Suitable acyl and acyl moiety in the terms acylamino and lower alkyl (acyl) amino may be carboxy …..". The argument of Mr David Young QC, for the respondent, depends on the words "in the terms acyl amino and lower alkyl (acyl) amino" qualifying only the term "acyl moiety" and not the term "acyl". The judge did not proceed on this construction. Moreover it can be said that as a matter of literal construction there is no reason why the inclusion of carboxy should be so limited. Mr Young submits that his construction is to be preferred because it is only acyl moieties that are present in acylamino and lower alkyl (acyl) amino. On this basis, claim 1 is also anticipated by Brown. The immediately preceding sentence however states that "The term 'acyl', whether used alone, or within a term such as "acylamino", denotes a radical provided by the residue after the removal of hydoxyl form organic acid." This may suggest that the term "acylamino" can also apply to "acyl". In the light of my earlier conclusions it is not necessary to resolve this point.
199. Mr Young's alternative argument was that claim 1 must necessarily be anticipated by Brown if claim 7 is so anticipated (which is not now seriously disputed). I do not accept this argument. The fact that Brown anticipates claim 7 does not necessarily mean that it anticipates claim 1 by reason only that claim 7 is a compound of claim l with variations. Claim 7, on the judge's finding, is structurally different since it includes aliphatic acids. In those circumstances it seems to me artificial to say that Brown anticipates claim 1 merely because the starting point for making a compound in claim 7 is the compound in claim 1. Accordingly, I reject this argument.
200. The judge also held that Brown rendered claims 1 and 20 within the amended patent obvious because of the point as to 4 methylsuphonyl phenyl substitution mentioned above. For the reasons given below, in my judgment the judge's conclusions on obviousness cannot be challenged. That conclusion precludes any argument that although Brown anticipates the patent in suit, the latter is none the less valid as a selection patent.

8. Obviousness

201. The judge held that the patent in suit (as granted and as proposed to be amended) was rendered obvious by Brown and Du Pont 697.
202. Section 3 of the 1977 Act provides:

"3. An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the state of the art by virtue only of section 2(2) above (and disregarding section 2(3) (above)."

203. The basic test of obviousness is that outlined by Oliver J giving the judgment of the Court of Appeal in Windsurfing International Inc v Tabur Marine (Great Britain) Ltd [1985] RPC 59:

"There are, we think, four steps which require to be taken in answering the jury question. The first is to identify the inventive concept embodies in the patent in suit. Thereafter, the court has to assume the mantle of the normally skilled but unimaginative addressee in the art at the priority date and to impute to him what was, at that date, common general knowledge in the art in question. The third step is to identify what, if any, differences exist between the matter cited as being "known or used" and the alleged invention. Finally, the court has to ask itself whether, viewed without any knowledge of the alleged invention, those differences constitute steps which would have been obvious to the skilled man or whether they require any degree of invention." (pages 73 – 74).

204. Earlier in his judgment Oliver J warned against adding human qualities to the hypothetical skilled man:

"The hypothetical Skilled Man is, no doubt, (together with his cousins the Reasonable Man and the Officious Bystander) a useful concept as setting a standard and, in the instant case, as providing the touchstone by which the question of obviousness may be judged by the equally hypothetical Juror; but he must not be allowed to obscure the nature of the inquiry which the words of the statute require, and one cannot help feeling that his image may lead to confusion if one seeks to attribute to him human qualities either of constitutional idleness or of perception beyond the knowledge and skill in the field in which he is hypothetically supposed to operate." (page 71).

205. This approach was confirmed in Mölnycke A B v Proctor & Gamble Ltd (No.5) [1994] RPC 49, at 115 per Sir Donald Nicholls VC and by Lord Hoffmann in Biogen Inc v Medeva plc.
206. The combination of two previously known technologies may be obvious. The new application must involve an inventive step (Du Pont v ICI (Witsiepe's application), above).
207. As Windsurfing makes clear, obviousness is a question of fact. Accordingly, the Court of Appeal will rarely interfere. As Aldous LJ said in "Wheatley v Drillsafe:

"Obviousness is a jury-type question and therefore this Court should be slow to differ from the decision of the judge unless he erred in principle. In this case I believe he did, in that he failed to differentiate between what was known and what was common general knowledge." (paragraph 52)

(Mance and Sedley LJJ agreed. (paragraphs 67 and 70)).

208. The principal case of obviousness is that Du Pont 697 was part of the prior art and was known to be an anti-inflammatory compound with gastric sparing qualities. As it turns out the only difference between Du Pont 697 and formula 1 in the claims of the patent in suit is the position of the aryl groups on the hetero cycle. In Du Pont 697 they are at the 2, 3 position whereas in claim 1 they are at the 3, 4 position. The appellant contends that the two molecules are structurally completely different. However, the essential issue on obviousness is whether it would have been obvious for a skilled person with the common general knowledge to have tried a 3, 4 substitution on Du Pont 697. The judge held that in January 1992 Dr Galbraith gave a paper at the Winter Prosterglantin Conference in Keystone which indicated that the reason why Du Pont 697 was an anti-inflammatory with gastric sparing qualities was Cox II selectivity. The judge held that this presentation was a part of the common general knowledge a year later and would therefore have been part of the common general knowledge at the date of the filing of the priority document of the patent in suit.
209. The appellant criticises the judge's conclusion on obviousness in relation to Du Pont 697. The judge's process of reasoning is criticised saying that he substituted his own views for those of the experts. It is said that he treated steps as obvious where there was no reason to try them and that he took account of commercial factors and motivation, which are not relevant in questions of obviousness.
210. However, the crucial evidence on which the judge relied was that of Professor Baker. Professor Baker's view was that it would have been obvious for a skilled person to have moved to the 3, 4 substitution pattern. The appellant contends that his view improperly "mosaiced" the Brown patent with the common general knowledge and that this is not permissable for obviousness purposes. We have been shown Professor Baker's expert reports and also the transcript of his cross-examination. It is correct that he based his view on the fact that Brown used the 3, 4 substitution but he only did so in part. This did not affect the view originally expressed in his expert's report on this point. The judge clearly preferred Professor Baker's expert evidence on this point to that of Professor Baldwin who disagreed with Professor Baker on the obviousness of moving to a 3, 4 substitution pattern.
211. In my judgment, the judge was entitled to accept Professor Baker's evidence over that of Professor Baldwin on this point and that his decision to do so is not reviewable on appeal.
212. A number of other objections are made to the judge's findings on obviousness. It is said that the judge's findings that Dr Galbraith had attributed the gastric sparing qualities of Du Pont 697 to Cox II selectivity was against the weight of the evidence. However, this was a matter for the judge. Moreover, contrary to the appellant's submission the judge was entitled to reject Professor Baldwin's view that chemists were waging some kind of chemical war on the molecule in order to discover a new NSAID. Likewise, the judge was entitled to take the view that there was a sufficient prospect of success in the 3, 4 substitution to justify investigating it. Lastly, the appellant criticises the judge for having taken into account the research done by the parties and by Du Pont. In fact Du Pont's research was focussed on the 2, 3 substitution and did not proceed to the 3, 4 pattern. A research chemist at the appellant thought of pursuing research on the 3, 4 pattern in relation to the Du Pont 697 molecule but this research was not approved by the appellant. Merck's research did in fact move the 3, 4 substitution but this was the product of detailed research and to some extent to intuition. In my judgment, the judge was entitled to consider whether his findings on obviousness were consistent with what actually happened and he did no more than this. In other words, in my judgment, he correctly applied the principles as regards obviousness and accordingly his decision on this point is not capable of being reviewed on appeal.

9. Industrial Application

213. In the circumstances, the question whether the enolate or enol forms of claim 1 have any industrial application does not arise.

10. Amendment

214. The appellant applied to amend the patent in the course of the proceedings to remove certain possible substituents of the patent. The judge held that the amendments were not objectionable in principle as they did not add matter.
215. In my judgment the judge was correct. The effect of the proposed amendment is to make methylsuphonyl phenyl mandatory in the 4 position, and as this is one of the options in US patent 432974 mentioned as an optional substituent for the 4 position in the priority document the proposed amendment would not add matter but be a disclaimer which the patentee is entitled to make.

Disposition of the appeal and cross-appeal

216. In my judgment the appeal must be dismissed. I agree with Aldous LJ that it is not in the event necessary to make any order on the respondent's notice.
Order: Appeal dismissed; permission to appeal to the House of Lords refused; appellants to pay 90 per cent of the respondents' costs in this court; the order of the judge on costs reduced 10 per cent; interim order for costs of £100,000. (Order does not form part of the approved judgment)