Cipla Ltd. & Ors v Glaxo Group Ltd.

This judgment is given in four actions to revoke UK patent 2 235 627 which stands in the name of Glaxo Group Limited (“Glaxo”). The claimants in the four actions are Cipla Limited and Neolab Limited (HC03 C00429), Generics (UK) Limited (HC03 C02619), IVAX Pharmaceuticals (UK) Limited (HC03 C03055) and Arrow Generics Limited (HC03 C03946). Each of the claimants is engaged in the manufacture or sale of generic pharmaceuticals and was separately represented in the proceedings until a comparatively late stage, when all instructed a single leading counsel for the purpose of the conduct of the trial.

Although the grounds of invalidity relied on by each of the claimants did not traverse exactly the same ground in each case, on 17 December 2003 I ordered that all four claimants should serve a consolidated statement of the Grounds of Invalidity relied on, and this was done on 30 December 2003, a further ground of invalidity being added by amendment on 13 January 2004.

When the four actions came on for trial, Dr Costello and Professor Page gave evidence first. They had been instructed by Cipla Limited and Neolab Limited. Although the actions had not been consolidated, an order had been made that the evidence in one should stand as the evidence in the others, and Mr Watson QC indicated that Professor Kay, who had been instructed by Generics (UK) Limited and Ivax Pharmaceuticals (UK) Limited, would not be called. Dr Auty and Dr MitchellHeggs gave evidence on matters not covered by the other two experts. Expert evidence was given by Dr Crompton and Professor Johnson for Glaxo, who also called Dr Palmer (the named inventor), Professor Partridge, and Professor Ruffin (witnesses of fact).

Dr Rapeport, a senior manager with Glaxo, was also cross-examined on his witness statement which related not to the issues in the case but to a very unsatisfactory incident which occurred shortly before the trial. I shall deal with that matter at the conclusion of this judgment.

In general terms, the patent in suit relates to an asthma treatment consisting of an inhaler that administers simultaneous doses of two drugs, salmeterol and fluticasone propionate. Salmeterol is a β2-agonist and fluticasone propionate a steroid suitable for administration by inhalation. The only issue in this case was whether this combination of β2-agonist and steroid was obvious to include in an inhaler at the priority date. A number of citations were relied on, but the principal basis for the allegation of obviousness turned out to be a previous combined inhaler containing salbutamol (a β2-agonist) and beclomethasone dipropionate (a steroid) sold by Glaxo well before the priority date under the name ‘Ventide’ which was established to form part of the common general knowledge of the skilled person. Two publications, “Fixed dose combination therapy in asthma. The positive case for future therapy” by Ruffin in Mechanisms in Asthma: Pharmacology, Physiology and Management 1988, 427-435 (Ruffin (1988)) and “The Drug Therapy of Asthma: Directions for the 21st Century” by Barnes, published in “Directions for New anti-Asthma Drugs”, Agents and Actions supplements, Volume 23, 293-313 (Barnes (1988)) were also particularly relied on.

The principal argument relied on by Glaxo to repel what appeared on the surface to be a strong argument of obviousness was that there existed at the priority date such a prejudice in the relevant art against regular β2-agonist therapy and to combination therapy by inhalation for asthma, that the combination claimed could not at the date be said to be obvious to the skilled person, albeit that Ventide was still on sale. In support of the contention that there was invention here, Glaxo also pointed to the enormous success of inhalers according to the invention which are sold under the mark ‘Seretide’: it is common ground that they have effected a very important change in prescribing practice and that Seretide is by all accounts a very successful treatment, both medically and commercially. It is one of GlaxoSmithKline’s best-selling drugs.

The question of the extent of the prejudice against regular β2-agonists and against combination therapies occupied most of the hearing. A large number of articles in the scientific press and elsewhere were referred to, and in order to keep some order in my discussion of them I use the conventional method of reference, that is Author’s name (year) with a letter following the year if there is more than one publication by that author in that year.

I shall return to the history of asthma treatments in greater detail below, but the principal milestones in modern asthma therapy down to the priority date in 1989 are as follows. Treatment of asthma by inhaled β-adrenoreceptor agonists, or β-agonists, can be traced back to a report of the treatment of asthma by the use of inhaled adrenaline at Guy’s well before the second World War. Administration of a β-agonist relieves the symptoms of asthma by relaxing the bronchial smooth muscle and so opening the airways. β-agonists act on the so-called β receptors on the walls of smooth muscle cells to affect the biochemical processes within the cell. There are at least two types of β receptors in human tissue. Adrenaline acts on the β receptors in the bronchial smooth muscle to relax it, but acts on the cardiac receptors to increase heart rate. Notwithstanding this undesirable effect β-agonists entered clinical practice after the introduction in the 1950’s of the pressurised metered-dose inhaler, which permitted an accurately metered dose of the drug to be administered via the airways directly to the affected muscle. The first drug to be widely used in this manner was the non-selective β-agonist isoprenaline, and its undisciplined use in the 1960’s is suspected to have caused an increase in the number of asthma deaths during that decade. In 1967 the Committee on the Safety of Medicines issued a warning to all doctors that there might be a link between the observed increase in asthma deaths and the overuse of isoprenaline-containing inhalers. Sales of the inhalers fell and so did the asthma death rate.

Salbutamol was one of the first β-agonists to be selective in its activity. It did not target the β₁ receptors in cardiac tissue, but selected the β₂ receptors of the bronchial smooth muscle. It accordingly gave symptomatic relief against an asthmatic attack but had much less effect upon the heart. Salbutamol and terbutaline, another β₂-agonist, were very well established by 1989 in the symptomatic treatment of asthma. It should be noted that some physicians certainly thought that regular (that is, non-symptomatic) use of bronchodilators was undesirable for a number of reasons. I shall return to this topic below.

While this development in the symptomatic treatment of asthma was going on, it came to be appreciated that asthma is in fact an inflammatory disease. Glucocorticosteroids are of course important anti-inflammatory drugs and their use in the treatment of asthma had started in the 1950’s. The available drugs at that time, prednisolone and hydrocortisone, had problems for long term therapy unless used in very low doses because of their side-effects. In the early 1970’s beclomethasone diproprionate (‘BDP’) was introduced and it became apparent very rapidly that inhaled BDP was a valuable asthma therapy.

BDP was administered either four times or twice daily (‘b.i.d.’, the accepted abbreviation for bis in diem). Compliance with a four-times-daily schedule was difficult for some patients, the more so as treatment with a steroid did not relieve immediate symptoms. I shall deal with the problems of compliance in greater detail below. It is sufficient to observe that compliance with a twice-per-day regimen is obviously easier for the patient, if only because the inhaler did not have to be carried about but could be left next to the toothpaste tube.

At this stage also, it should be noted that salbutamol alone was a comparatively short-acting β2-agonist, and would not suffice for overnight symptomatic treatment. There is no doubt that by 1989 many patients were prescribed BDP twice daily and instructed to use their salbutamol inhaler only when necessary to alleviate symptoms of bronchoconstriction.

The patent is entitled ‘Inhalation medicaments for treating respiratory disorders’, and the invention is said to relate to the use of a bronchodilator drug in combination with a steroidal anti-inflammatory drug for the treatment of respiratory disorders such as asthma. The nature of this unpleasant condition is then briefly described in outline in terms that certainly represent the general understanding of the condition at the priority date:

At page 2 line 23 the patent starts its discussion of Ventide. It points out that salbutamol has ‘usually a four to six hour duration of action, which is too short either to control nocturnal asthma or for convenient maintenance of the disease in some patients. Having referred to BDP (page 1 line 31) the specification observes that the benefits of reduced inflammation may not be immediately apparent. There follows this important passage:

Having identified salmeterol and fluticasone propionate as the bronchodilator and steroid respectively, the specification observes that

No data are given in support of this assertion. It does not appear that any clinical trials had in fact been performed at this stage. There are no data offered in support of any other property of the combination, the specification concerning itself only with the manner in which the metered dose inhalers containing the combination are to be manufactured. There is, in my view, force in the claimants’ contention that the only teaching of the document is that the two drugs should be combined in an inhaler, for subsequent clinical trial.

There was a minor dispute about the precise import of the words ‘compatible and complementary’ in this passage. In context, it is plain that they refer to the effects just described: that is, treatment of the underlying inflammation by the steroid and of the constriction of the airways by the bronchodilator. There was a suggestion that they also suggested that there was some kind of ‘synergistic’ effect between the two active ingredients. I do not think that they do indicate that there is such an effect, but in any event I shall consider the question of synergy below.

Claim 1 of the patent in suit is to

Only this claim is said to possess independent validity. It will be seen at once that where simultaneous administration is concerned it covers an inhaler adapted to administer appropriate doses of the two active ingredients. It also appears to cover the provision of two inhalers packaged together to provide sequential administration. The words ‘for ... sequential administration’ will normally be taken to mean only ‘suitable for’ such administration, and since the claim is not limited to any particular dosage regime, it potentially covers two inhalers used as required. Since this was certainly obvious at the priority date, Mr Waugh QC for Glaxo was content, if necessary, for the words ‘or sequential’ to be removed by amendment. Without such an amendment, the claim is only partially valid. The question is whether what remains is obvious.

The patent claims priority from two applications and it was not disputed that it is entitled to a priority date of 8 September 1989, unless the words ‘compatible and complementary’ introduce a suggestion that there are ‘surprising synergistic or other advantages or interactions’. If they do, the claimants contend that the patent is insufficient and loses priority, since there is no suggestion of any such interaction in the priority documents. So far as loss of priority is concerned, the point is a bad one. The invention (whose scope is to be ascertained from the claim) is not in any way defined by reference to any synergistic effect. It covers combinations where there is no synergistic effect, if that is possible, and combinations where there is such an effect. The failure to refer to such an effect in the priority documents cannot mean that their disclosure does not support the invention claimed in the patent. In any event, only one further relevant publication would have been let in, and this was abandoned by Mr Watson QC for the claimants.

Apart from the foregoing, no substantial questions of construction arise. I can turn to the obviousness of the invention.

Ten matters forming part of the state of the art at the priority date are pleaded:

As I have indicated above, the specification discloses a proposed new drug formulation. The active ingredients are known, and it seems to me that the specification is plainly addressed to a pharmaceutical manufacturer with an interest in marketing an inhaler. In Catnic v Hill & Smith [1982] RPC 183 at 242, Lord Diplock described the addressee of the specification as the person ‘likely to have a practical interest in the invention’. It is of course well settled that this practical interest may involve a collection of skills and knowledge that would not normally be found in a single individual, and such cases are often described as ones in which the addressee of the specification is a team. In this case the addressee is making metered dose inhalers with a view to clinical trial and, if the clinical trial is successful, ultimate sale. It includes clinical pharmacologists and formulation specialists, and obviously it consults with physicians. I do not accept the contention that the patent is addressed to physicians alone. Physicians lack all the skills necessary to make a workable product: but they will advise the manufacturer as to the appropriateness of the proposed product in clinical practice.

In this case, the discussion has centred on the knowledge and prejudices to be imputed to the notional clinical member of the team. The skilled person is a legal construct, related no doubt to the reasonable man and other similar personifications of legal criteria. It is through his eyes that the specification must be read and he provides the touchstone for obviousness and insufficiency of description. There are technologies where no great knowledge is to be attributed to the skilled man, and others (such as genetic engineering) where to attribute an unrealistically low level of attainment to the skilled man would prejudice industrial development. There are two classic statements. The first is that of Fletcher Moulton LJ in British Ore Concentrate Syndicate Ltd v Minerals Separation Ltd (1909) 26 RPC 124, 138:

The second is in General Tire v Firestone [1972] RPC 457 at 482:

Laddie J has described the difference between the ‘plodding unerring perceptiveness of all things obvious [of] the notional skilled man and the personal characteristics of real workers in the field’ at length in Pfizer Ltd’s Patent [2001] FSR 201, paragraphs 62 ff. As he says, the skilled person has no idiosyncratic preferences or dislikes, but he does possess the preferences or dislikes that may be objectively identified as characteristic of the real workers in the field.

Invention may lie in overcoming the preconceptions of the skilled person in a particular field of technology. As Jacob J put it in Union Carbide v BP Chemicals [1998] RPC 1, 13:

A good example is to be found in the Dyson vacuum cleaner case, Dyson Appliances Ltd v Hoover Ltd [2002] RPC 22, [2001] EWCA Civ 1440. The judge had held that

In the Court of Appeal, Sedley LJ put it this way:

Such a prejudice may be a merely commercial one (‘this device won’t sell’) or it may be a technical one (‘this won’t work and it is not worth bothering with’). A twenty-year monopoly is conferred for overcoming a prejudice of the second kind, but not for overcoming a commercial prejudice (see Hallen v Brabantia [1989] RPC 307 (Aldous J)). A technical prejudice must be general: it is not enough that some persons actually engaged in the art at the material time labour under a particular prejudice if a substantial number of others do not. A prejudice which is insufficiently widespread for it properly to be regarded as commonly shared will not, in my view, be attributed to the notional skilled person. As Jacob J put it in Union Carbide v BP (above at page 16):

This is the obverse of the distinction between the common general knowledge and that which is known only to some who enjoy advantages that it would be unfair to attribute to every person working in the field (see, for example, Beloit v Valmet [1997] RPC 489 (CA)).

At the same time, the skilled man should not be taken to represent some sort of lowest common denominator of persons actually engaged in the field, possessed only of the knowledge and prejudices that all of them can be said to possess. The common knowledge of different groups employed on the same tasks in different organisations is likely to be different, for the reasons mentioned by Aldous J in Beloit. It is unlikely that the expert witnesses will be truly representative of the skilled person, as not only may they be too well qualified but they will come to the case with personal prejudices or preferences that must be discounted: see for example the discussion of the question whether there was a bias in favour of or against d.c. and a.c. coupled amplifiers in PCME Ltd v Goyen Controls Co UK Ltd [1999] FSR 801 at paragraph 61, and in the passage in Pfizer to which I have referred.

The statutory provision, following Article 56 EPC, is:

The state of the art is defined in section 2(2) as follows:

These provisions do not permit what is sometimes called the mosaicing of individual documents or prior uses said to form part of the state of the art, unless it can be shown that the skilled person, confronted with a particular citation, would turn to some other citation to supplement the information provided by the first. Such cases are not common, and the problem most frequently arises when the patent in suit itself makes assumptions about what its disclosure enables a skilled man to do. To take an example, one might consider a prior use consisting of the sale of a particular pharmaceutical preparation. The defendant says that the nature of the active ingredient can be ascertained by analysis of the preparation. Whether that active ingredient forms part of the state of the art depends upon identifying the experimental techniques available to the skilled person. The use of such techniques may involve reference to books and journals, and the explanation of the results may require reference to particular publications. If it can be shown that this analysis is the kind of analysis that would be normally performed, then it is part of the common general knowledge even if only a fraction of the material can be ascertained without reference to the books. Laddie J gives a further example in paragraph 66 of his judgment in Pfizer:

The analysis whether an invention is obvious in the light of the state of the art proceeds, therefore, on the basis of the individual citations, read alone in the light of the common general knowledge. An approach very frequently adopted is set out in the much-cited passage in the judgment of the Court of Appeal in Windsurfing International v Tabur Marine [1985] RPC 59:

The High Court of Australia has detected a shift in what it calls the Grundnorm in English law relating to obviousness in a case cited by Glaxo in the present case, AB Hässle v Alphapharm [2002] HCA 59, 194 ALR 485, a shift said to be exemplified by the observation of Laddie J that the skilled person ‘is assumed to have read and understood all the available prior art’ (see Cairnstores v AB Hässle [2002] EWHC 309 (Ch) at [94]). That statement of Laddie J is, with respect, only concerned with the point decided in Windsurfing International v Tabur Marine [1985] RPC 59 at 71, that the skilled man is to be assumed to take an interest in the prior art cited, and thus that obscurity of a citation is not to be counted against it. To allow documents to be discounted on the ground of obscurity will run against the policy of the law that any person must be free to enter a library and exploit what he finds there, provided that he exercises no invention in that exploitation.

On the basis of the decision of the High Court in the Alphapharm case, I was asked to approach the final Windsurfing question in this case on the basis of the reformulation of the so-called Cripps question by Graham J in Olin Mathieson Chemical Corp v Biorex Laboratories Ltd [1970] RPC 157, as follows (with the emphasis favoured by their Honours):

This formulation is derived from the ‘Cripps question’ (Sharp & Dohme v Boots Pure Drug Co Ltd 45 RPC 154) but it is not identical to it. Graham J’ s formulation contains an implicit definition of the word ‘obvious’. Graham J says that he used the words ‘led directly as a matter of course to try’ to incorporate the meaning of the word ‘obvious’. It omits any reference to the therapeutic quality of the compounds, because the claim with which Graham J was concerned was not limited by any therapeutic effect.

It is worth noting that Graham J in his judgment approached the words ‘in the expectation that it might well’ as follows:

It seems to me clear that Graham J’s question was formulated in the light of the particular case before him. Had the evidence been that the notional research team would, in following a strategy dictated by the common general knowledge, investigate either the direct - CF₃ substitution at the 2 position or (less likely) the reduction of Smith’s formula VIII the question formulated in the manner set out above might have been misleading, because the substitution might be considered to have been part of an obvious course of action, without any expectation of success at all. I say ‘might’ because there will certainly come a point where the result of an entirely rational strategy of investigation based exclusively on the common general knowledge cannot be considered obvious. It is a question of fact in every case. Both the Scylla of considering nothing obvious except that to which the skilled man is driven and the Charybdis of considering every invention obvious that can be decomposed into a sequence of obvious steps must be avoided. The former is unfair to industry because it stifles natural development. The latter is unfair to inventors and not countenanced by English patent law (see British Westinghouse v Braulik 27 RPC 209 at 230 (Fletcher Moulton LJ) Non-Drip Measure Co Ltd v Stranger’s Ltd (1943) 60 RPC 135 at 142 (Lord Russell) and Technograph Printed Circuits Ltd v Mills & Rockley (Electronics) Ltd [1972] RPC 346 at 362 (Lord Diplock). The problem with the Cripps question, and with its various reformulations, is that it tends to obscure the actual question, which is whether the invention was obvious or not. The form propounded by Graham J is, with respect, open to a number of further objections if it is separated from the facts of the particular case. The first is that it seems to assume that if there are a large number of obvious steps that might be taken in the light of a particular citation, some of them will not be obvious because they would not be suggested to be carried out directly: they may only be carried out if the ones tried earlier fail. That is unacceptable: the obviousness of an invention cannot depend upon the order in which the various courses of action obvious in the light of a particular citation are attempted - see Brugger v Medic-Aid Ltd [1996] RPC 635 at 660-1 (Laddie J). Second, the import of the final italicised phrase is unclear. Is something a priori obvious as a possible course of action necessarily to be rejected on the ground that the notional skilled man has only a modest hope that it will produce something useful?

In my judgment, evidence that the skilled person will be led to try something because there is a reasonable expectation that it will produce a useful result establishes one possible route to a possible finding of obviousness. But if it is obvious to try that thing for other reasons, there need be no superadded requirement that there should also be some expectation of success. This was made clear in the decision of the Court of Appeal in Norton Healthcare v Beecham Group plc CA 19 June 1997 (unrep) where Aldous LJ said:

Mr Waugh submitted that if English law did not in every case require consideration of the prospect of success of a particular step (he put it as high, I think, as requiring the expectation of a good result) it was out of step not only with the law of Australia exemplified in AR Hässle v Alphapharm which I have discussed above but also with the law as applied in the EPO. To support this submission, he referred to the long discussion of obviousness in the C.I.P.A. Guide to the Patents Acts, paragraphs 3.21, 3.33-4 and to the ‘Case Law of the EPO’ at I.D.6 paragraphs 6.1 and 6.2. The approach of the EPO to the question of obviousness is highly structured. It has three stages:

The whole methodology is encapsulated in the decision of Technical Board of Appeal 3.3.1 in T 31/84 MILES/Test Device. The Board remitted an opposition to the Opposition Division for re-examination on the basis of a new claim 1, with the following direction:

I am not persuaded that this is substantially different from the Windsurfing approach, subject to one qualification. The EPO will consider obviousness on the basis of the closest prior art only. Every pleaded starting point, however remote, needs to be considered in coming to a conclusion on obviousness in the domestic context, but I suspect that the need to concentrate on the ‘best’ citation has a result that is not much different. When one reaches the final step (the last Windsurfing step seems to be the same) the factors to be taken into account in assessing obviousness are not, so far as I can see, much different. The summaries of decisions contained in the ‘Case Law of the Boards of Appeal’, sections 6.1-2 clearly indicate the wide variety of factors that the EPO consider relevant to an assessment of obviousness. Interestingly, the question of the expectation of success seems to be considered particularly relevant where the course of action in question is long and consists of much labour, and understandably it is in the field of genetic engineering and biotechnology that the question becomes important (see section 6.2). Obviousness is a question of fact, a so-called ‘jury question’, and I see no basis for the suggestion that UK law is out of step with the principles applied in the EPO.

I propose to analyse the question of obviousness along Windsurfing lines. Before I do so I must consider the common general knowledge in the art, and the question of the prejudices from which it is contended that the skilled person suffered.

I have discussed the skilled team above. The claimants contend that the following matters formed part of the common general knowledge at the priority date.

Asthma has two aspects, bronchospasm and bronchial inflammation. Both require to be treated. This is obviously common general knowledge, and it forms the basis of the dispute as to the correct therapy to which I shall come.

Bronchospasm is treated by a bronchodilator. By 1989 the leading class of bronchodilators were the β₂-agonists of which the most popular was salbutamol. There is no doubt that during this period it was prescribed both for regular and for symptomatic (p.r.n.) use. It had the particular shortcoming to which I have referred, that it had a comparatively short duration of action, and thus could not protect against nocturnal asthma.

By 1989, it was generally known that Glaxo had developed a new β₂-agonist called salmeterol which was in clinical trials. Its main point of interest was that it had a considerably longer duration of action than salbutamol. It had been the subject of conferences and even a press article quoting Professor Page (who gave evidence for the claimants) in the Independent in April 1989:

I have quoted this article at length because it describes in a digestible form two of the matters causing concern to asthma specialists. The first is compliance with steroid therapy, and the second is the masking of a deteriorating condition of the airways by the use of bronchodilators. It is obvious that two kinds of bronchodilators will have this effect: long acting compounds, as the article describes, but also short acting compounds taken regularly rather than in response to the onset of an attack.

The new material had been discussed at a conference in November 1987 the proceedings of which were published in 1989 under the title ‘Allergy and asthma, new trends and approaches to therapy’ attended by representatives of Glaxo, one of whom, a Mr Coleman, is reported as having said this:

The last sentence is some coat-trailing for an anti-inflammatory effect that it was hoped would be exhibited by the compound. In fact, I believe that no clinically significant effect that cannot be ascribed to its longer-lasting bronchodilator effect has been convincingly demonstrated, and the suggestion that there was such an effect is disputed by Professor Page, who considered that the claim should not have been made. But the importance of salmeterol was immediately appreciated by the physicians who were interested in this field. I am also satisfied that while they were all aware of it, they were not confident of the place, if any, it would finally occupy in the armamentarium.

There are other pre-launch publications that need to be considered. Salmeterol was recognised as a very important drug for Glaxo and thus it was covered in the business press as well. The publications are as follows.

Ullman & Svedmyr (1988) ‘Salmeterol, a new long acting inhaled β₂ adrenoceptor agonist: comparison with salbutamol in adult asthmatic patients’ Thorax 43: 674-678. This document is pleaded. It describes the attractive feature of salmeterol, that is, its long acting nature as appearing to be ‘an important therapeutic advance’. It also contains a warning about the risk that regular therapy with the drug might produce a rebound increase in bronchial reactivity and notes also the risk of tachyphylaxis (i.e. a decrease in response to a given dose over time). The evidence goes a long way towards showing that this paper was common general knowledge in itself, as Dr Crompton accepted:

There was some suggestion that Ullman & Svedmyr (1988) was only suggesting the use of salmeterol once per day, in the evening, for the control of nocturnal symptoms. This is obviously wrong. The reference to ‘maintenance treatment’ is clearly a reference to more frequent use, and the whole is plainly concerned with the use of the drug at all times of the day and night. It is to be noted that those of Ullman & Svedmyr (1988)’s patients who were on steroids were kept on steroids for the duration of the trial.

Ullmann & Svedmyr (1988) refer to four further papers, of which the most important is Bradshaw & al (1987) below. I consider that the extent to which a skilled person can be expected to follow cross-references depends on a number of factors. On the face of it, he should be taken to follow immediate citations, but obviously he cannot be taken to undertake a citation hunt, following every lead in every paper, or there would be no end to it. This, however, is a direct citation on the central properties of salmeterol, and I think that he should be taken to follow it, although Dr Crompton did not, in fact, do so, being insufficiently interested.

Bradshaw & al (1987) ‘The Design of Salmeterol, a long-acting selective β₂ Adrenoceptor Agonist’ British Journal of Pharmacology 92: 590P. This paper is not pleaded. It describes the design of the salmeterol molecule in response to the need ‘for a longer acting inhaled bronchodilator for maintenance treatment of the disease and to control nocturnal asthma’. This is one of a number of early papers originating with Glaxo that appear to have been presented at a meeting of the British Pharmacological Society in September 1987.

SCRIP No 1177 (February 1987) observes that products highlighted by the analysts include ‘salmeterol, a long-acting beta₂-agonist for asthma; and the steroids, fluticasone and cloticasone, believed by Bell Lawrie to be in Phase III clinical trials for dermatological uses; Greenwell Montagu analysts speculate that Glaxo may also develop these steroids for use in asthma.’ By SCRIP no 1184 (4 March 1987) the news had hardened as a result of an extensive presentation by Glaxo:

SCRIP No 1243 contains a report of the meeting of the British Pharmacological Society at which Bradshaw & al (1987) was presented. By April 1988, SCRIP No 1301 is reporting that both salmeterol and fluticasone propionate are in Phase III clinical trials as aerosol formulations. Dr Costello said that there was a great deal of excitement about the drug in the field, and it was known to rival manufacturers.

The 1987 conference was two years before the priority date of the patent in suit (September 1989) and in the two years that followed it is clear that salmeterol became a molecule of great interest. I conclude that salmeterol was certainly part of the common general knowledge by the priority date. Although it had not emerged from clinical trials, it was clearly a molecule of great importance and accordingly of interest to any skilled person as a sound basis for further development.

The next matter is the treatment of the underlying inflammation. Again, there is no doubt that it was generally accepted that bronchial inflammation in asthma should be treated with an anti-inflammatory agent, and that the most popular class of anti-inflammatory agents for this purpose was the inhaled corticosteroids. Beclomethasone dipropionate (Becotide inhaler) was one of the most successful, at least in the United Kingdom.

In 1989, the correct treatment of asthma was a matter of considerable debate. There is no doubt that the claimed combination contemplates regular treatment by inhalation, at least once per day (I take this from the reference to nocturnal asthma) and most likely twice per day. Glaxo contend that regular use of a β₂-agonist was so strongly disapproved of by physicians in this field that it was a real barrier to the acceptance of any therapy involving regular administration of a β₂-agonist.

At the outset, it should be noted that salbutamol was at all material times sold as suitable for use in regular therapy. There was some explanation of this, centred on the regulatory difficulty in changing the ‘package insert’, but I agree with the observation that if regular therapy with β₂-agonists was perceived to be really harmful this would have been changed. This, however, is only a straw in the wind, and the substantial question is what the notional specialist physician, who forms part of the addressees of the specification for this purpose, thought.

First, it is necessary to consider therapy using β₂-agonists alone. I have no doubt that at the priority date specialist physicians considered that treatment with β₂-agonist alone was normally inappropriate except in the mildest cases. There is a fundamental problem in the treatment of asthma with β₂-agonists. This is that the underlying inflammation is not treated, and may consequently deteriorate to the point at which the inhaled β₂-agonist will not provide symptomatic relief. Certainly, I think that by the priority date it was thought by specialists in the field that the best mode of treatment was based upon the regular use of steroids together with symptomatic relief using β₂-agonists.

It is this fact that explains why there was a controversy as to the correctness of prescribing the drug for regular administration. I think that the overall position is well described by Professor Partridge, who was involved in the preparation of the Guidelines for the management of asthma in adults that were published in the BMJ in 1990. The Guidelines were presented under the auspices of the British Thoracic Society, the Research Unit of the Royal College of Physicians, the King’s Fund Centre and the National Asthma Campaign. Prof. Partridge put it this way:

Thus Professor Page and Dr Costello (both of whom gave evidence for the claimants) doubted the usefulness of regular β₂-agonist therapy in general and regular salmeterol treatment in particular in their paper ‘Why β-agonists should not be used regularly’ Respiratory Medicine (1992) 86, 477-479. This paper is concerned both with use of the drugs in monotherapy and with their use in combination with steroid therapy, although there is no doubt that some of their reasons for not using β-agonists regularly can relate only to monotherapy.

The ‘Guidelines for management of asthma in adults: I-chronic persistent asthma’ (1990) BMJ 301:651-653 represent the direction in which professional thought was moving at the priority date. The guidelines divide the treatment of asthma into a number of steps, with patients starting treatment at a step most appropriate to the severity of their condition. The steps are:

There was considerable discussion of the guidelines. I think that it is plain that they deprecate regular treatment with β₂-agonists at an early stage, although the description of stage 4 suggests a use of the β₂-agonist that has become in effect regular, the rate of symptomatic use making the distinction from regular use difficult to draw. The reference to night time symptoms certainly suggests the use of the β₂-agonist prophylactically at night time. The publication of the guidelines more or less coincided with the publication of Sears & al (1990) Lancet 326: 521, an important paper. The message of this paper is that regular use of β₂-agonists, whether or not in combination with inhaled corticosteroids, involves an increased risk over their symptomatic use:

Dr Palmer, the sole inventor of the patent in suit, put the point at its highest when he referred, somewhat hyperbolically, to a ‘gathering storm’ in the period 1988-9. He referred to three publications, ‘the Tattersfield paper, the Crane paper on Fenoterol and then an editorial in Thorax...’. Taking these in reverse order, the Thorax editorial (a paper by Mitchell published in February 1989) was concerned in particular with enhanced bronchial hyperresponsiveness and tachyphylaxis caused by regular, as opposed to symptomatic, use of β₂-agonists. It points out that that the population of asthma sufferers is large and that small effects can be of great significance and concludes as follows:

The Crane paper on fenoterol is Crane & al. (1989) ‘Prescribed Fenoterol and Death from Asthma in New Zealand 1981-83: Case-controlled Study’ Lancet 1:917. Fenoterol is a short-acting β₂-agonist less selective than salbutamol, and the hypothesis advanced in the paper is that ‘the use of fenoterol by MDI [metered dose inhaler] increases the risk of death in severe asthma’. The Tattersfield paper is (I think) Vathenen & al. (1988) ‘Rebound Increase in bronchial Responsiveness after Treatment with inhaled Terbutaline’ Lancet 1: 554. This was one of a trio of papers, with Kerrebijn & al. (1987) ‘Effect of long-term treatment with inhaled corticosteroids and beta-agonists on the bronchial responsiveness in children with asthma’ J Allergy Clin Immunol 79: 653 and Kraan & al (1985) ‘Changes in bronchial hyperreactivity induced by 4 weeks of treatment with antiasthmatic drugs in patients with allergic asthma: a comparison between budesonide and terbutaline’ J Allergy Clin Immunol 76: 628. The summary of this paper concludes that ‘budesonide [a corticosteroid] improves bronchial hyperreactiviy, possibly by a dampening of late allergic reactions, whereas treatment with terbutaline [a β₂-agonist] may lead to a temporary increase of bronchial hyperreactivity, possibly as a result of β-receptor desensitization.’

I formed the clear view that there was increasing concern about the use of β₂-agonists throughout the 1980’s. My impression is that Crane & al (1989) and rather more importantly Sears & al (1990) had a significant effect on the attitude of some consulting physicians to the use of β₂-agonists. The position adopted by the Guidelines represented the logical culmination of such developments down to their publication in 1990.

However, I am inclined to regard the Guidelines as representing the culmination of one line of thought only. That this is the case is demonstrated by a number of factors. First, none of the papers excited a worry about regular low dose use of β₂-agonists. There are two instances of ‘for and against’ pairs of papers in this case. The first is Lindsay (1988) ‘Fixed dose combination therapy in the treatment of asthma-the case against it’ and Ruffin (1988) ‘Fixed dose combination therapy in asthma. The positive case for future therapy’ in ‘Mechanisms in Asthma: Pharmacology, Physiology and Management’ 421 and 427. Ruffin (1988) is cited against the patent. These papers were prepared for a conference that took place in Australia. I shall return to Ruffin (1988) in its place, but for present purposes it is necessary only to note that Ruffin is supporting treatment using a combination inhaler containing a β₂-agonist and a steroid. If the steroid is to be administered regularly, necessarily so too will the β₂-agonist be. The proceedings of the conference do not record this objection being raised. Dr Crompton was cross-examined on this:

This is not, I think, evidence of a widespread unwillingness among the relevant class of physicians to contemplate regular therapy with β₂-agonists. In fact, the point goes further. There is no doubt that in 1987 regular therapy with β₂-agonists was being employed in the United Kingdom Thus, Cochrane (1987a) ‘Inhaled Steroids and their use in the treatment of asthma’ Bronchus 2: 3: 8 observes that ‘a recent survey of two general practices in London has suggested that in the UK, at least, there is now a move to use regular prophylactic therapy earlier in the treatment of asthma, and that in 75% of the mild to moderate asthma patients the second-line drugs after regular inhaled bronchodilators are inhaled corticosteroids (two-thirds) and disodium cromoglycate (one-third).’ In his ‘Colour Atlas of Asthma’ published in 1989, Dr Cochrane proposes a number of algorithms for assessment of the treatment of various classes of asthma, three of which involve regular inhaled β₂-agonist therapy, albeit in two of those classes the regular regimen is preceded by inhaled corticosteroids. Dr Costello, who gave evidence, had proposed something rather similar in his book.

In his paper Cochrane (1987b) ‘The difficult chronic asthmatic’ British Journal of Diseases of the Chest 81: 313 Dr Cochrane had made the following observation:

Professor Barnes wrote the paper that I allowed to be cited by a late amendment, Barnes (1988) ‘The drug therapy of asthma: directions for the 21st century’ in Directions for New Anti-Asthma Drugs, Agents and Actions supplements 23: 293. He remarks that inhaled corticosteroids ‘should probably be introduced much earlier in the management of asthma and β-adrenoceptor agonists, which have no effect on the inflammatory response, should be used for symptomatic control as required’.

None of these publications, nor Dr Crompton’s evidence that I have quoted above, suggests that there was a general refusal to use regular β₂-agonist therapy. Dr Mitchell-Heggs gave evidence, entirely credible, that he was prescribing regular β₂-agonists with inhaled corticosteroids in 1989. The facts are far more consistent with there being no uniformity of view in the profession, but that one view was in 1989 becoming more dominant.

Objectively, the later history of this matter does not support the view that there was any general prejudice against regular therapy with β₂-agonists. There is clear evidence of regular therapy with β₂-agonists well after the priority date. Even Sears & al (1990), described by Dr Crompton as having poured fuel on the embers of a debate that had been smouldering, should not be regarded as having completely turned the general view of the profession. Dr Sears’ views on regular β₂-agonist therapy and similar views expressed by Professor Page in the Lancet were criticised in the letters pages by (inter alios) representatives of the Glaxo group for generalising from the results for fenoterol to all β₂-agonists.

The existence of a spectrum of opinion, and of prescribing practice, is also shown by the succession of publications on which Professor Page was cross-examined. Page & Costello (1992) ‘(2) Why β-agonists should not be used regularly’ Respiratory Medicine 86: 477 is one member of the second pair of deliberately controversial papers. The other is Twentyman & Higenbottam (1992) ‘(1) β-agonists can be used safely and beneficially in asthma’ Respiratory Medicine 86: 471, the pair of papers being published under the heading ‘Controversies in respiratory medicine: regular inhaled β-agonists - clear clinical benefit or a hazard to health?’. These two papers come to conflicting conclusions even on the basic question whether the studies (including Kerrebijn & al (1987), Kraan & al (1985) and Vathenen & al (1988)) do in fact show that regular administration of β-agonists does increase bronchial responsiveness. Twentyman & Higenbottam (1992) say it does not: Page & Costello (1992) say that it does. I cannot resolve this dispute, but it was even in 1992 a dispute.

There is in Page & Costello (1992) a passage suggesting that it is necessary to take a more nuanced view of the opinions of the relevant physicians than merely to attribute an outright dismissal of regular β-agonist therapy:

“Airway hyperresponsiveness that characterizes bronchial asthma has two distinct components: an increased sensitivity of the airway as reflected by a leftward shift of the dose-response curve...as well as excessive airway narrowing as evidenced by an exaggerated or absent maximal response plateau. ...This latter component of the airway hyperresponsiveness has been considered by some investigators to be ‘the most hazardous feature of airway hyperresponsiveness’ as patients are still at risk of undergoing bronchoconstriction, even if the airway sensitivity is shifted. It is therefore of considerable importance that in contrast to glucocorticosteroids, neither regular use of short-acting bronchodilators...or the long acting bronchodilator salmeterol protect against excessive airway narrowing, leading some investigators to conclude that ‘the insufficient efficacy against this component of airway hyperresponsiveness of monotherapy with long-acting agonists is a drawback in the treatment of asthma.

Against this background, it is relevant to ask why should any clinician prescribe regular β-agonist therapy? The British Thoracic Society guidelines for the treatment of chronic asthma specifically recommends that β-agonist should be used as required rather than regularly, and suggests that patients who need to inhale a bronchodilator more than once daily require inhaled anti-inflammatory drugs. In this regime, therefore, there appears to be no place for regular inhaled β-agonists. The choice for prescribers has become more complicated since the arrival of the longer-acting β-agonists such as salmeterol and formoterol, with the implication that their prolonged bronchodilator effect might be sufficient in itself to control asthmatic symptoms; indeed, the ‘symptomatic control’ argument may have some validity in the short-term, but the evidence presented here suggests that their long-term regular use is undesirable. A case has been made for an anti-inflammatory effect of salmeterol in man...; when scrutinized, however, this appears to be a functional antagonist rather than a true anti-inflammatory effect...and subsequent detailed histological and lavage studies on the effects of regular salmeterol treatment do not support an anti-inflammatory action. The debate amongst specialists, and the publicity given to the postulated ‘anti-inflammatory’ effects of salmeterol must have confused general practitioners and others not directly involved, and indeed may have led to some believing that inhaled long-acting β-agonists might be a panacea for asthma, with consequential and regrettable under-use of inhaled corticosteroids or cromoglycate.

The Committee on Safety of Medicines concur that the case for the anti-inflammatory effects of β-agonist has not been made. We suggest, therefore, that the use of regular inhaled β-agonist has no place in treatment regimes for asthma, and is indeed potentially hazardous.”

This passage is primarily aimed at monotherapy with β-agonists. Basing itself on the Guidelines recommendation that patients who inhale a β-agonist more than once per day are in need of anti-inflammatory drugs, it points out that there should be no scope for regular therapy with β-agonists at all: they should only be used as required, and only in the short term. Long term regular use is undesirable, because it shows a failure to control the underlying inflammation with anti-inflammatories. This passage was employed to suggest that the article is not only concerned with monotherapy (which is true) but that it was evidence of a firm objection among physicians that regular therapy was unacceptable. I would draw precisely the opposite conclusion. The article has a polemic aspect, and I do not believe it would be necessary if regular therapy with β-agonists were not an established feature of the clinical scene, whether in combination with steroid therapy or as a monotherapy.

The last part of the passage I have quoted above represents a real fear that did arise after the introduction of salmeterol in 1990. It had been claimed that salmeterol had anti-inflammatory properties, and it was feared that GP’s would be discouraged from prescribing inhaled corticosteroids as well. Dr Costello gave unchallenged evidence that Glaxo changed their patient information to reflect this concern.

It is difficult to get over the conclusions of the article by Drazen & al. (1996) ‘Comparison of regularly scheduled with as-need use of albuterol [salbutamol] in mild asthma’ New England Journal of Medicine 335: 841:

There are a number of other factors that suggest that this was a continuing dispute, with strong positions being taken on each side, albeit that the majority may well have been opposed to regular β-agonist therapy. For example, the Committee of the Safety of Medicines did not set up their working party to investigate the literature on β-agonists until about 1991 (the only reference to this I can find is in the transcript of the ‘Public Eye’ television programme transmitted on 31 May 1991), and it seems that the FDA in the United States did not convene a meeting to consider the question until late 1991. I am satisfied that I should not ascribe any rooted objection to regular therapy to the skilled person when I work out his relevant characteristics. He or she might well view the possibility critically, but this would be a matter that could be resolved by clinical trial. Dr Crompton was too willing to attribute his own views, genuinely held and for good reasons, to all practitioners in the art, and he was, I think, too reluctant to accept that others might not hold those views, or hold them less dogmatically.

Mr Waugh QC was inclined to suggest that the only class of physicians whose views should be attributed to the skilled man were those at the forefront of asthma research and so those who entertained most doubt about regular β₂-agonist therapy and about combination therapy. This approach is not consistent with the evidence. The patentees cast their net much more widely (see paragraph 104 below) and I am satisfied that any drug company would want to have an idea of what the general approach to a particular therapy was, if it could.

I have referred to the fact that bronchial hyperreactivity on the rebound was mentioned as a possible problem in the published literature. Dr Crompton did not attach any clinical significance to it. Tachyphylaxis I have also referred to: there was little evidence about it and it cannot have been a genuine issue at the date. There was some attempt to turn various theories about the action of β-agonists into aspects of the common general knowledge, such as their effect on mast cells, or Professor Page’s theory relating to the ‘asthma paradox’ in Page (1991) ‘One explanation of the asthma paradox: inhibition of natural anti-inflammatory mechanism by β₂-agonists’ Lancet 337: 717. Professor Page was obviously very doubtful about the risks of β₂-agonists and this article emphasises in particular the risks of monotherapy, but this is unsurprising given that asthma morbidity and mortality rates were still increasing, although the disease was in principle preventable. It was never shown that this kind of scientific speculation could ever have formed part of the common general knowledge of the addressee.

Finally, I should turn to the problem of masking. There is no doubt that regular use of a β-agonist tends to mask the frequency of attacks. Thus, regular use makes it more difficult for the prescriber to satisfy him- or herself that the correct dosage of anti-inflammatories is being administered. This prevents the frequency of inhaler use from being used as a marker for disease progression. This is undoubtedly true, but (at least with short-acting β-agonists) nocturnal symptoms and breathlessness are also useful. I think that this was common general knowledge at the date. Nonetheless, the importance of this problem is most important when a β-agonist is administered as a monotherapy.

I have dealt at length with the common general knowledge at the date because it is crucial to the proper approach to the question of obviousness, to which I now turn. As I have indicated, each of the matters relied on as a starting point for the allegation of obviousness must be considered individually. The first matter is the prior use of Ventide.

Ventide is a combination of salbutamol and beclomethasone dipropionate in a metered dose inhaler. It was launched in 1983, and by 1989 it was selling 250,000 units a year. It was far from being a front-line treatment for asthma, and I do not think that any of the witnesses had any affection for it, although Dr Mitchell-Heggs saw a use for it and prescribed it. The reason that Ventide was prescribed was to overcome compliance problems, which I must now discuss. Although this is a topic which falls naturally into the common general knowledge, it is best considered in the context of this particular product.

The problem of compliance is a well recognised one. It is difficult for the prescriber to persuade certain patients to take their medicine, particularly when the medicine does not have an obvious effect. Thus, when Ventide was launched in 1983, GP magazine carried a piece of ‘advertorial’:

The claimants’ evidence in chief indicated that compliance was indeed a problem, and that a solution was a combination inhaler. Dr Auty, who was with Fisons at the relevant time, said that prescribing both corticosteroid and β-adrenoceptor stimulant was known to have benefits, as is common ground. He identifies the problem of compliance, and observes that some patients would think that only the β₂-agonist was working, and would not take the corticosteroid as prescribed. He says that the problem of achieving compliance with the corticosteroid regimen was of particular concern in the treatment of asthmatics. He says that Ventide was a very well known product, and that although well received, there were two criticisms of it. First, that the frequency of dosing (3-4 times per day) was dictated by the duration of action of the β₂-agonist component and the salbutamol element was insufficient to last through the night. Second, it was not possible to give the individual patient that patient’s ideal dose of both components. He expresses the view that it was the view amongst a large part of the field, especially practising clinicians, that better compliance with an imperfect dosage regimen was preferable than poor compliance withan individually tailored regimen. Dr Costello said that although he did not like Ventide, GP’s would use it ‘to get steroids into patients’.

This is confirmed by Glaxo’s own documents. In 1988, Dr Palmer, the inventor of the patent in suit, wrote a ‘Clinical Expert Report’ to the Medicines Control Agency in support of the extension of the Ventide product licence to disks and disks (paediatric) for use in so-called diskhalers. This is a regulatory document, and can safely be taken to express an honest view. Under the heading ‘Combined Therapy’ he says:

Dr Palmer stood by these statements, as was inevitable:

There was some consistency among the witnesses as to the circumstances in which Ventide might be used, although it would not be used as a first-line therapy. Dr Costello, who did not like it, said that it might be used by GP’s to get steroids into patients, and Dr Mitchell-Heggs explained in his witness statement that compliance was a real problem. In cross-examination he identified two groups of patients who drew advantage from the use of the combination therapy:

In his report, Dr Crompton had described Ventide as an irrational product. He accepted under cross-examination that its use in an attempt to improve compliance was not irrational. To use combination products for such a purpose had been suggested by others. For example, Ruffin (1988) is a report of a paper delivered at a conference. Included is a report of the ensuing discussion, during which Dr Barnes is reported as having said

Finally, Dr Crompton was himself a co-author of a paper, McDonald & al (1988) ‘Evaluation of the combination inhaler of salbutamol and beclomethasone dipropionate in the management of asthma’ Current Medical Research and Opinion 11;2:116. It is necessary only to consider the Introduction and a concluding paragraph.

The evidence from Dr Crompton as to this is, first, that it was done as a favour to Glaxo:

The evidence that follows demonstrates, I am afraid, what happens to the reliability of publications when a particular department is beholden to a sponsor of its research. After an intervention by me caused by my failure to understand Dr Crompton’s reply, the cross-examination resumed:

There is no reason for Dr Crompton to put his name to a paper supporting a treatment that he considers irrational unless either he was negligent in permitting his name to be added to the paper or he felt sufficient pressure, professional and financial, to override his scientific doubts. I do not think he was negligent, and against my better judgment I found myself in sympathy with him in the circumstances in which he appears to have found himself. I feel deep unease at the fact that funding considerations can have this effect on those who appear to be conducting disinterested research. What this evidence does convince me of is that there was no reason to suppose that there was any relevant prejudice against either combination products or regular β₂-agonist therapy.

Against this background, two answers are relied on by the claimants as encapsulating the case of obviousness over Ventide. The first is that of Professor Page. He is the Professor of Pharmacology at King’s, and he is not a clinician. He has, however, had a long experience of asthma therapy, and has published very extensively about it. He said this:

Professor Page was challenged on his evidence that the ‘updating’ of Ventide was obvious. He had not suggested it in his article. In fact, his article looks further into the future. As the extract above makes clear, he is talking about not the combination of two separate drugs having respectively prophylactic and symptomatic effects, but the design of a single molecule with both these qualities. Such a molecule would be open to the criticisms advanced against the combination.

Dr Crompton’s answer followed a challenge by Mr Waugh QC to Mr Watson’s use of the word ‘obvious’ in a question:

In the light of the whole of the evidence, I think this series of answers contains the right approach. The combination was an obvious combination, as a combination. It was not obvious that it would be safe, or that it would represent any advance on what had gone before. Whether this would be the case depended upon the outcome of clinical trials which was not foreseeable, although it was a reasonable prediction for the skilled man having regard to the known properties of salmeterol (which was part of the common general knowledge at the date) and any steroid previously used for prophylactic therapy that the combination would be both safe and efficacious. Any skilled person would find fluticasone propionate in looking for up-to-the-date members of the class of steroids as candidates for incorporation in his new Ventide. A. good reason for such a combination is the need to ensure compliance. The ‘prejudice’ discussed at such length above was, in fact, a doubt relating to questions that could only be resolved by clinical trial. This extends to such matters as the significance of the long-acting effects of salmeterol. The skilled man would understand this: and accordingly the combination was entirely obvious in the light of the prior manufacture, use and sale of Ventide and the publication of the various data associated with that preparation.

That salmeterol would be the obvious substitution for salbutamol follows from the most basic consideration, that a drug that only needs to be taken twice a day will encourage compliance. That this was the case appears to be the universal message of the evidence. It is also Glaxo’s internal view of the justification for the product:

The foregoing findings are sufficient to dispose of this case, but the claimants put their case of obviousness on a selection of the citations to which I have referred in paragraph 22 above.

The case on Ullman & Svedmyr (1988) adds nothing to the case on Ventide, once it is accepted that the paper is common general knowledge.

The case on Dr Barnes’s article (Barnes (1988)) is rather more straightforward. Dr Barnes is an acknowledged expert in this field. This article was introduced by late amendment, Mr Watson saying that he did not wish to advance any case on the basis of evidence in chief. Thus, all the evidence on this citation comes from Glaxo. The article concludes:

It may be observed that elsewhere in the article Dr Barnes advocates the reduction in the use of β-agonists to use for symptomatic control as required. It must be accepted that nonetheless this is a clear general instruction to use combination inhalers, which by the priority date included as one of a limited number of available combinations salmeterol and an inhaled corticosteroid. Dr Crompton accepted this. A flavour of the cross-examination is to be obtained from this exchange, when it was being put that a combination of salmeterol and beclomethasone dipropionate is obvious in the light of this disclosure:

The message from Dr Barnes in Ruffin (1988) is to the same effect and, it seems to me, equally clear. The only real objection that can be advanced to this argument of obviousness is the prejudice alleged to exist that I have considered. I conclude that on the basis of this publication also, the patent is invalid for obviousness.

The case on these publications is in substance the contention that the patent is invalid in the light of Ullman & Svedmyr (1988). They add little to the case overall. The evidence relied on has two aspects: the exchange that I have set out in paragraph 102 above and the fact that combinations were common general knowledge as a possible way of presenting drugs with complementary effect, even if Ventide is not part of the common general knowledge:

I confess that absent any strong prejudice against regular therapy with β-agonists I would consider that this was, in fact, sufficient to make the invention of the patent in suit obvious, but manifestly the case is stronger on the footing of the Ventide prior use.

In the result the invention of the patent in suit is obvious. The claimants’ case is, in my view, overwhelmingly strong. This is a decision arrived at on what I believe to be a conventional approach, and I have not felt it necessary to examine the effect of the interesting decision of the Privy Council in Ancare New Zealand Limited’s Patent [2003] RPC 8 (page 139). Had I considered that scientific opinion was wholly out of accord with what was being done in the market, it would have been necessary to do so. I do not read this case as throwing any doubt on the proposition that invention may lie in overcoming a prejudice in the art.

It is sometimes thought that a patent may be saved from a finding of obviousness if a combination otherwise obvious has some unexpected advantage, and, in particular, an advantage caused by an unpredictable cooperation between the elements of the combination. I do not consider that such an approach is in general justified. There is a limited class of cases in which the patentee has identified an advantageous feature possessed by some members only of a class otherwise old or obvious, has described the advantageous effect in his specification and has limited his claim to the members of the class possessing this advantageous feature. Such a claim may be justified on the basis of what is called selection. Unexpected bonus effects not described in the specification cannot form the basis for a valid claim of this kind. I think that the matter is described with complete correctness by Jacob J in Richardson-Vicks’ Patent [1995] RPC 568 at 581:

If a synergistic effect is to be relied on, it must be possessed by everything covered by the claim, and it must be described in the specification. No effect is described in the present specification that is not the natural prediction from the properties of the two components of the combination. Although there was some considerable evidence in the case intended to show synergy between the elements of this combination, Mr Waugh QC rightly abandoned any reliance on synergy per se as justifying a finding of non-obviousness (see paragraph 152 of his closing skeleton argument). Instead, he refers to the real, great, success of Seretide as demonstrating that a prejudice had indeed been overcome.

This argument is hopeless. At all material times down to the priority date, salmeterol was protected by patent. The question of any commercial use of the combination product was thus under the control of the patentees and subject to the satisfactory completion of the clinical trial process. But Mr Waugh QC puts the point:

It was on this issue that Dr Palmer, the inventor, gave evidence. Mr Watson QC cross-examined him on the basis that the invention of the patent in suit was what Glaxo call a ‘line extension’, that is, a further presentation of existing active ingredients which may provide the basis for further patent protection. No doubt there was something in this, but in this case I did not regard this evidence as helpful. It has to be remembered that a perfectly valid patent may be written by a person who does not stir from his armchair, thinks it is all obvious and does no experiments to confirm his hunch. I suspect he may even be permitted to invent the experiments, provided they work and give the promised result. Because the assessment of obviousness has to be objective, there is always a risk in paying too much attention to what the patentee does. I have myself found it of assistance when, for example, the experts suggest that a particular course of action is the obvious one to try first, and it appears that that is what the patentee indeed did. There was no evidence of that kind here. Of course, what Glaxo record internally may for example reflect on the alleged prejudice against β₂-agonist regular therapy, and I have taken that into account. It is also true that Mr Watson QC suggested to Dr Palmer the evidence that so far as he was concerned, the question was essentially a marketing one:

Even were I to accept Mr Watson’s suggestion that for Glaxo the idea of the combination (the line extension) is essentially a marketing concept in this case, it is of limited assistance in deciding the objective question. While therefore I consider that there is some force in it I prefer to leave it out of account in coming to my decision on obviousness.

Much of this judgment has concentrated on the evidence of Dr Crompton, and to a somewhat lesser extent that of Professor Page, Dr Auty and Dr Costello. I should make it clear that Dr Crompton and Professor Page were both on the dogmatic side, so much so in Dr Crompton’s case that I have felt it necessary to criticise his evidence in specific respects. Dr Costello and Dr Auty were, if I may put it this way, more objective. There is one final matter to which I should refer.

In his first report, Professor Page referred to the good relations that subsisted between him and Glaxo. In his reply report, he reveals that he was telephoned by Dr Garth Rapeport, who is rather grandiloquently described as the ‘Senior Vice President, Centre of Excellence for Drug Discovery, Respiratory and Inflammatory Diseases’ at GlaxoSmithKline. It is accepted that he telephoned Professor Page and informed him that he was disappointed with his involvement in the case. He said that Professor Page could no longer expect Glaxo to be friends with him and that Glaxo could no longer fund his research activities beyond the existing research grants. Professor Page told Dr Rapeport that he was expressing only his genuinely-held views.

Dr Rapeport was very upset that Professor Page had not told him of his involvement in the case. As he puts it:

No apology was proffered by Glaxo for this. Dr Rapeport attended at my insistence, and he offered an apology in the witness box. I have seriously considered whether to refer this matter to the Attorney-General with a view to proceedings for contempt of court. I have decided not to do so. My reasons, put shortly, are that Dr Rapeport acted on his own initiative, and that his reasons for his action were silly. There was no conflict of interest. Dr Rapeport said that Professor Page would be treated in future on the same basis as any other eminent academic with interests in respiratory medicine, and I am content to leave the matter there. I wish it to be clearly understood that expert witnesses are relied on by the court to express a disinterested view. Any pressure, and any act which may have the effect of placing pressure, on a witness may be a contempt of court and dealt with accordingly. If the witness’s actions have genuinely created a risk of the disclosure of confidential information, or if there is genuinely a conflict of interest, then the matter must be dealt with in open court on the application of the party which considers itself to be wronged.