IVAX Pharmaceuticals (UK) Ltd v Chugai Seiyaku Kabushiki Kaisha

The Honourable Mr Justice Kitchin:

Introduction

In this action the claimant (“Ivax”) seeks revocation of European Patent EP (UK) 0 230 932 (“the Patent”). It has an earliest priority date of 17 January 1986. The defendant (“Chugai”) is the proprietor of the Patent and counterclaims for infringement.

The validity of the patent was originally challenged on the grounds of lack of novelty and obviousness. Only the attack of obviousness is pursued. It is based upon two citations:

U.S. Patent 4,200,640 (“640”);

Japanese Patent Application 57 145 659 (“659”)

The amendment is opposed on the grounds that it adds matter over the application as filed. The statement of opposition also raised a lack of clarity objection but this has been overcome by a further application to amend.

There is no issue on infringement. The process for the production of generic nicorandil which Ivax proposes to use falls within claims 1, 3 and 6 of the Patent as granted and claim 1 of the Patent as proposed to be amended.

Independent validity is asserted for claims 1, 3 and 6 of the Patent as granted and claim 1 of the Patent as proposed to be amended.

The witnesses

I heard evidence from two experts on formulation. Professor Collett was called on behalf of Ivax and Professor Aulton on behalf of Chugai. Both are highly distinguished.

Professor Collett is Emeritus Professor at the School of Pharmacy and Pharmaceutical Sciences, University of Manchester, a post he has occupied since 2005. He is also Professor of Pharmaceutical Technology at the University of Sao Paulo and part-time Chief Scientific Officer for Quay Pharma UK. He was a development pharmacist at Boots in 1963 and Assistant Professor of Pharmacy at Temple University, Philadelphia between 1967 and 1969. Over the years he has acted as a consultant to a number of large pharmaceutical companies. His expertise is in the area of pharmaceutics – the science of turning a drug into a medicine. This includes formulation science and dosage form design.

Professor Aulton has been Professor of Pharmaceutical Technology at the Leicester School of Pharmacy at De Montfort University for the last 13 years. He had responsibility for teaching and research in the subject area of pharmaceutics within the university until moving to part-time employment at the end of September 2003. Since October 2003, he has had responsibility for pharmacy curriculum development. He originally joined the faculty in 1971. His research has concentrated mainly on pharmaceutical preparations in solid dosage forms and has involved extensive contact with other universities worldwide and the pharmaceutical industry.

Professor Aulton is the editor and part author of one of the leading undergraduate texts in the field of pharmaceutics, “Pharmaceutics: The Science of Dosage Form Design”, the first edition of which was published in 1988.

In addition I heard evidence from two distinguished medicinal chemists. Dr Wakefield was called on behalf of Ivax and Dr Newton on behalf of Chugai.

Dr Wakefield is a Chartered Chemist and Fellow of the Royal Society of Chemistry. Prior to his retirement from full-time employment in 1993, he spent most of his career at the University of Salford, with the title of Reader in Organic Chemistry from 1983. In 1984 he was a co-founder of Ultrafine, a company providing chemical services to the pharmaceutical industry and was an employee of that company until 2004. He has acted as consultant to many well known pharmaceutical and chemical companies, particularly in connection with the synthesis of pharmaceutically active substances.

Dr Newton obtained a BSc in Chemistry and Biology in 1964 from the University of London as an external student studying at Portsmouth Polytechnic. He remained at Portsmouth Polytechnic to carry out research into the stereochemistry of compounds with bridgehead nitrogen atoms and was awarded a PhD by the University of London in 1968. From 1971 to 1996, he worked for Glaxo, originally as a senior research chemist working for Allen and Hanburys (a Glaxo group company) in the chemical research department. In 1983 he was made director of chemistry at Allen and Hanburys with responsibility for the chemical research, analytical research and chemical development departments. When Glaxo combined Allen and Hanburys Research and Glaxo Laboratories to form Glaxo Group Research (GGR) he became director of the new chemistry division that included all of Glaxo’s UK chemical research, analytical chemistry and process research chemistry. In 1991, GGR reorganised chemistry into separate research and development groups and he became director of the new chemical research division. He has held a series of directorships and appointments at leading universities as Visiting Professor.

The only expert the subject of any criticism was Professor Collett. No attack was made on his honesty or integrity but it was said that he spent little time checking the accuracy of his reports, that he was forced to concede that much of what he said was unsupportable and that he was prone to speculation, even on matters outside his expertise. I think these criticisms are unduly severe. Professor Collett gave evidence in his reports on a number of matters which are of direct relevance to the issues which I have to decide. As I elaborate below, there are instances where he retreated from that evidence or qualified it under cross examination. However, in all cases when the basis of the challenge was put to him he readily acknowledged the extent to which he could not maintain the evidence he had originally given. In my judgment Professor Collett was, on occasion, a little casual in his approach to his written reports. But his evidence given in the course of cross examination was entirely candid. I believe he was doing his best to assist the court and I found his evidence helpful.

Technical background

The purpose of formulation is to convert an active pharmaceutical ingredient into an effective, safe, easy to administer and stable medicine. The broad process of formulation as of 1986 was not in dispute between the parties and can be summarised as follows.

During the course of formulation the dosage form is considered. A range of possible dosage forms are available. Orally administered formulations are, and were in 1986, the most preferred in terms of patient acceptance and most widely used. In terms of orally administered formulations, the main choice was between tablets and capsules. Tablets were the preferred route as they were generally cheaper and easier to manufacture than capsules. In the mid-1980s, the most popular choice of dosage form would have been an orally administered swallowable tablet. An alternative to a swallowable tablet was a sublingual tablet. This involves the drug being administered to the capillaries under the tongue. Some drugs are so difficult to formulate into tablets that the only choice is to opt for an alternative dosage form. In this context, gelatin capsules were known to avoid problems associated with poor powder flow, poor compaction and poor stability sometimes encountered in tablet formulations.

There were essentially three main processes by which pharmaceutical tablets were manufactured in 1986: wet granulation, dry granulation and direct compaction. The first two involve a granulation step that converts the original primary powder particles into intermediate granules that are compacted into tablets. The advantages of granulation are that it provides improved powder flow during processing and improved compaction properties during tabletting. Of the granulation methods available, wet granulation involves creating a granule via the addition of a liquid granulating fluid and subsequent drying of the formed granules. Dry granulation involves the creation of a granule in the dry state by a pre-compaction stage. Direct compaction involves compressing and consolidating the dry, powdered ingredients into a tablet without an intermediate granulation step.

The actual process by which these powders or granules are converted into solid tablets in a tabletting machine is known as “compaction”. During compaction, the powders or granules are fed from a hopper into dies on a tabletting machine where they experience a high applied stress that, if formulation and process conditions are correct, results in a solid tablet. Compaction actually consists of two stages: (i) “compression”, which is a volume reduction step in which particle rearrangement and removal of air takes place; and (ii) “consolidation”, which occurs as the applied pressure is increased and the solid entity of the tablet is produced as a result of the particles bonding together. The word “compression” is commonly used for the overall process of compaction.

Another aspect of formulation is the choice of excipients. These are used to convert the drug into a form which is suitable for administration to patients. The main excipients in a tablet formulation are (and were in 1986) classified as follows:

Fillers or diluents are used to add particular properties to a formulation such as bulk, improved compression, reduced segregation and to prevent over-granulation. Some commonly used diluents are lactose, microcrystalline cellulose and starch.

Disintegrants are included to enhance or promote the break up of the tablet once taken by the patient. Some commonly used disintegrants are microcrystalline cellulose and starch.

Binders are the active agents which provide the cohesive binding and deformation properties required to compress the formulation into tablet form. Binders are frequently hydrophilic polymers such as polyvinylpyrrolidone, methylcellulose and starch.

Lubricants and glidants are used to ease the ejection of the tablet from the die, to prevent sticking of the tablets to the punches, and to prevent excess wear on the dies and punches. Glidants also improve the flow properties of the granules or powders during the manufacturing process.

The whole development process is often broken down into two stages, namely pre-formulation and formulation although there is no sharp boundary between them and aspects of development described by one formulator as pre-formulation might be described by another as formulation.

The pre-formulation process begins when a drug shows sufficient pharmacological promise in animal models to warrant evaluation in man. The pre-formulation studies identify the physicochemical properties of the drug that could influence drug performance and development of an effective dosage form. The parameters studied include dissolution rate, chemical stability, aqueous solubility and pKa of the drug. The effect of various standard excipients, pH, oxygen, light and temperature on the chemical stability of the drug are also investigated and the results are used by the formulator to begin to design an appropriate dosage form.

Excipient compatibility studies are carried out using a form of accelerated stability testing. This is an important part of the process. At the start of the programme limited amounts of the active drug may be available and it is necessary to know which of the common excipients are compatible with it. Consequently an initial screen is likely to be conducted. In such a screen a small amount of drug is placed with a series of standard excipients in vials which are sealed and stored at a range of different temperatures and humidities. After storage the samples are assessed visually for signs of deterioration, by thermal analysis for evidence of interactions between the drug and excipients and by chemical analysis for evidence of degradation. Any excipient that shows incompatibility is discarded.

Before commencing pre-formulation studies, the formulator will liaise with the medicinal chemists involved in the drug’s development so as to obtain information on the known properties of the drug and the proposed development schedule. For example, medicinal chemists may have knowledge of the drug’s weaknesses (e.g. the fact that it is susceptible to degradation in particular conditions). A literature search on the active drug molecule is also likely to be performed.

At the formulation stage tablets containing various excipients are produced and their stability is once again assessed, usually at elevated temperatures and humidities. In addition, tests are conducted to assess the performance of the tablet itself. These include mechanical crushing tests to evaluate the physical strength of the tablet, disintegration tests to evaluate the ability of the tablet to break down when in contact with water, dissolution tests to measure the rate and extent of the dissolution of the drug and accelerated and long term shelf life and stability tests. As a result of all these studies an optimum formulation is produced.

The Patent

The Patent relates to a method for producing a stable preparation of nicorandil. More particularly, it relates to a method for producing a stable preparation containing nicorandil, a saturated higher aliphatic acid or a saturated higher alcohol (such as stearic acid or stearyl alcohol) both of which are solid at ordinary temperatures, and optionally fumaric acid, oxalic acid, salicylic acid, tartaric acid and/or glutaric acid.

The specification explains that nicorandil has various actions which make it useful as a curative for various types of angina pectoris whilst causing minimum side effects.

The object of the invention is then identified. It is to develop a method for providing a nicorandil preparation which is stable not only in humid conditions but also under the compressive pressure exerted by punching operations in tablet making.

The specification continues by describing something about the efforts which the inventors made to solve the problem. It says that the stability of nicorandil preparations was found to decrease as more pressure was exerted in compressing them into tablets. The inventors apparently noted the need to avoid the deformation of nicorandil crystals and distortion of the crystal lattices that take place under compressive force. The specification states that the lubricants magnesium stearate and calcium stearate are conventionally used to reduce the friction that occurs between powder particles when they are compressed. The inventors therefore tried mixing nicorandil with progressively increasing amounts (several to ten odd times the normally used amount) of magnesium stearate and calcium stearate. However, the nicorandil still proved unstable (page 2, lines 25-32).

The inventors found that the object of the invention was, however, achieved when nicorandil was mixed with at least 0.5% (on the basis of the weight of the preparation) of a saturated higher aliphatic acid or a saturated higher alcohol both of which are solid at ordinary temperatures, and optionally with at least 0.1% (on the basis of the weight of the preparation) of fumaric acid, oxalic acid, salicylic acid, tartaric acid and/or glutaric acid. This is said to result in a nicorandil preparation having remarkably improved stability (page 2, lines 33-43).

The specification contains eight examples and comparative examples. Example 1 compares tablets of a nicorandil preparation including 8% by weight of stearic acid with comparative tablets made under the same conditions except that stearic acid was replaced by the same amount of mannitol. The tablets also included mannitol, cornstarch, methyl cellulose and magnesium stearate. The tablets of the invention and those of the comparative sample were divided into two subgroups. The tablets in the first subgroup were vacuum dried to make them substantially water-free, whereas the tablets in the other subgroup remained undehydrated. All of the tablets were then put into glass bottles, capped and stored at 40°C for three months. The stability of the tablets was then evaluated in terms of the residual amount of nicorandil as a percentage of the initial weight before accelerated aging. The results are set out in Table 1. They show that the tablets of the invention were markedly more stable than those of the comparative example, whether dried or undried.

Similar results can be seen from the other examples. In examples 1 to 4, varying amounts of at least one of stearic acid, palmitic acid or stearyl alcohol were mixed with nicorandil. The later examples (5, 6, 7 and 8) additionally involved one of the organic acids, fumaric acid or salicylic acid. Example 8 is a slow-release formulation involving a two-layer formulation, in which the “lower layer” contained stearic acid (and additionally fumaric acid).

The relevant claims read as follows:

No issues of construction arise. Both parties agreed that, in the context of the Patent, “stable” means sufficiently stable to be awarded marketing authorisation from a regulatory authority.

The skilled addressee

There was no dispute as to the nature of the skilled addressee of the Patent. It is addressed to persons interested in formulating nicorandil into a pharmaceutically acceptable form. The work would be performed by a formulator with several years experience and he would be assisted by graduates and less experienced technicians and he would have access to analytical laboratories. He would liaise with a medicinal chemist at an early stage in the process to obtain information about the known properties of the drug and the proposed development schedule.

Common general knowledge

It was accepted that the technical background which I have set out was common general knowledge. Ivax placed particular emphasis upon the following. First, an active ingredient should be formulated as a tablet, if possible. Secondly, in any tablet formulation there is likely to be a lubricant. Thirdly, when used as a lubricant, a material would be used in the minimum quantity necessary. Fourthly, given the task of formulating a new active ingredient the formulator would want to find out what he could about the molecule from the medicinal chemist and would then carry out routine pre-formulation exercises.

Other aspects were, however, more contentious. The first concerned the candidates for use as a lubricant. There is no doubt that magnesium stearate was the lubricant of choice. Professor Collett accepted that around 90% of tablets would have contained magnesium stearate as the lubricant and the remainder one of a variety of alternatives such as stearic acid, hydrogenated castor oil, calcium stearate and PEG 400.

The approach to pre-formulation would, as I have mentioned, vary from formulator to formulator. Some formulators might choose to carry out pre-formulation with one excipient from each class. But others might just as easily choose two excipients from each class. This would be a routine course to take.

In 1986 the standard amount of a lubricant in a tablet formulation was frequently 1% or less, but larger amounts (up to 3 or 4%) were known to have been used, as I elaborate later in this judgment.

Before leaving the issue of lubricants it is convenient to mention one important matter which was not common general knowledge. There was no understanding that materials used as lubricants could be used to increase the stability of formulation against degradation caused by compaction or humidity.

Instead the formulator would have known he had a number of options for addressing instability caused by humidity. These included the following:

Changing to a different form of the active ingredient; so, for example, if the active ingredient was a base, then a salt form of the drug could be used;

Changing the manufacturing process and instead of using wet granulation using a non aqueous granulating fluid, such as ethanol or using direct compaction;

Drying all the excipients before use;

Ensuring that all the equipment in the manufacturing process was dry before use;

Coating the individual drug crystals with a polymer such as ethyl cellulose;

Improving the packaging of the final formulation by using foil wrappers, blister packs or adding a dessicant.

Degradation resulting from the application of pressure was an unusual phenomenon. Professor Aulton could only recall it occurring with indomethacin. This undergoes a polymorphic transition under the pressures experienced in tablet compaction. Faced with such a problem the formulator would have known he had the options of reducing the compaction pressure or using an alternative dosage form such as a soft or a hard gelatin capsule.

Obviousness - the unamended claims

Both parties agreed it is appropriate to address the question using the structured approach explained by the Court of Appeal in Windsurfing International Inc. v Tabur Marine (Great Britain) Ltd [1985] RPC 59. This may be summarised as follows:

The inventive concept

The inventive concept of claim 1 is producing a stable nicorandil preparation by mixing nicorandil with a saturated higher aliphatic acid or a saturated higher alcohol, both of which are solid at room temperatures, to produce a pharmaceutical preparation. No particular concentration of acid or alcohol is required, provided the formulation is stable. Claim 3 limits the inventive concept to the use of palmitic acid or stearic acid. Claim 6 requires the presence of at least 0.5% by weight of the acid or alcohol of claim 1 or claim 3.

Obviousness over 640

The 640 patent was published in 1980. It discloses nicorandil, a method of making it and its use for treating circulatory disease.

Example 7(a) is a sublingual tablet containing nicorandil prepared according to Example 2. It contains 0.6% (by weight) of magnesium stearate as the lubricant. Example 7(b) is a tablet prepared for internal use containing nicorandil prepared according to Example 1. It contains 0.5% (by weight) of magnesium stearate.

There are no other examples that contain any alternative lubricant. Further, there is no suggestion in 640 that the Example 7 formulations are unstable, either as a result of humidity or mechanical pressure during compaction. To the contrary, Experiment 2(II) suggests that when the compound prepared according to Example 2 was sublingually administered to dogs as a tablet containing 1-10mg of nicarnadil, coronary flow was increased.

Both Chugai and Ivax accepted that the Patent teaches that mixing nicorandil with 0.5% (by weight) stearic acid may result in a stable preparation and that I must assume for the purposes of this judgment that this teaching is correct.

Accordingly, the difference between the inventive concept of all the claims and the 640 patent disclosure is the use of 0.5% stearic acid in addition to or in place of the excipients mentioned in the 640 patent. On the teaching of the Patent this will produce a stable formulation, as both sides agreed.

In assessing the issue of whether or not this difference was obvious neither side suggested and there was no evidence that the skilled addressee would, as a matter of general knowledge or following a literature search, have known of what the Patent describes as being conventional practice on page 2, lines 15-22; that is to say, dealing with the instability of nicorandil by coating the nicorandil crystals with one or more fatty or waxy substances (such as stearic acid). Indeed, there was no evidence that, at the outset of the project, he would have known of the instability of nicorandil. I have to say I find this somewhat surprising but I must proceed on that basis.

In these circumstances the case of obviousness advanced on behalf of Ivax was very straightforward. It ran as follows:

The skilled addressee would have known that stearic acid was not just one of a number of lubricants, it was second only to magnesium stearate as the lubricant to use in excipient compatibility tests. Without invention he could (and probably would) select it as one of the two lubricants to use in pre-formulation. The other would be magnesium stearate.

If magnesium stearate failed materially so as to render it unsuitable to go forward to formulation then plainly stearic acid would be obvious.

But, even if it did not fail, stearic acid would (on the premises of the Patent) not have failed. There was no technical reason why it could not have been used as a lubricant.

It was therefore a matter of choice rather than invention whether the skilled man chose to go forward with stearic acid as opposed to magnesium stearate.

In substance, Ivax contended that it was technically obvious to formulate a tablet of nicorandil using stearic acid as a lubricant and, if used for that purpose, and in an appropriate amount to fulfil that purpose (say up to 1%), then it would in fact produce a stable formulation falling within the scope of the claims.

Alternatively, and as a secondary submission, if magnesium stearate proved unsuitable then stearic acid would have been an obvious alternative.

Chugai’s answer to this case was, in summary, as follows. As to the primary submission, the 640 patent examples describe the use of magnesium stearate as a lubricant. Moreover, the skilled person would use magnesium stearate unless there was a positive reason not to do so. The 640 patent does not describe any reason to change and there was no evidence that pre-formulation or formulation studies would lead the skilled person to conclude that magnesium stearate was unsatisfactory as a lubricant. Accordingly, it was simply not obvious to change to stearic acid. As to the secondary submission, if, during the course of pre-formulation or formulation, the problem of stability came to light then the skilled person would have no reason to attribute that problem to the magnesium stearate and would instead attempt to solve the problem by the one of the conventional means described in paragraph [48] above. It was not obvious that a stability problem could be solved by a change in lubricant.

In evaluating these rival contentions I must first set out some relevant principles:

All courses of action which present themselves without the exercise of invention are obvious: Brugger v Medicaid [1996] RPC 635 at 661.

Evidence as to what the skilled person could or would have done can be relevant, but, at the end of the day, the key question is whether something in the claim was technically obvious in the light of the disclosure relied upon: Hallen v Brabantia [1991] RPC 195 at pp.211-212 (CA); Pharmacia v Merck [2001] EWCA Civ 1610; [2001] RPC 41 at [122]; Asahi Medical v Macopharma [2002] EWCA Civ 466 at [26] - [27].

Whether or not there is a reason for taking the step from the prior art may be an important consideration, but it is not an essential requirement of a conclusion of obviousness: Pharmacia v Merck at [124].

Care must be taken in suggesting that a particular route was “obvious to try”. All will depend upon the facts of the case. So, for example, mere possible inclusion of something in a research programme on the basis you will find out more and something might turn up is not enough: St Gobain v Fusion Provida [2005] EWCA Civ 177 at [35]. But that does not mean that in every case the decision whether a claimed invention was obvious can be determined by deciding whether there was a reasonable expectation that a person might get a good result from trying a particular avenue of research: Pfizer’s Patent [2002] EWCA Civ 1 at [57];

If the claimed invention is obvious for one purpose then an added and unexpected benefit, however great, will not found a valid patent: Hallen v Brabantia at p.216.

I must now consider the application of these principles to the circumstances of this case. The starting point is the 640 patent. Ivax contended that the skilled person would carry out conventional pre-formulation studies and could, and probably would, select magnesium stearate and stearic acid as lubricants to use in excipient compatibility tests. Chugai contended that this was not an obvious step to take.

In my view, the evidence on this point was all one way. Professor Aulton accepted that the skilled person, reading the 640 patent, and without access to the work which lay behind it, would carry out the pre-formulation exercise. Professor Collett was of the same opinion. He said that when a new compound comes in, no matter what it is, the first thing to do is pre-formulation. The 640 patent told him nothing about the quality of the tablets made in Example 7 and he would go through the standard procedure. On the assumption that it was obvious to do pre-formulation studies then it was standard practice to test two excipients from each class and one of the obvious, if not the most obvious, lubricant to test in addition to magnesium stearate was stearic acid.

In these circumstances, are the claims of the Patent obvious? I have to say that I have found this a very difficult question to answer. In making the attack Ivax was content to assume that the skilled person would in fact proceed with magnesium stearate. Ivax was also content to assume that the Patent has made a technical contribution – namely the discovery that the lubricant stearic acid has stabilising properties when mixed with nicorandil crystals and so permits the production of stable tablets. These points may fairly be said to be powerful arguments in favour of the proposition that Chugai has made an invention which is not obvious.

Nevertheless, I feel compelled to reach the conclusion that the attack succeeds. In my judgment the insuperable difficulty which Chugai faces is that the claims are not limited to the technical advance which it has made. They cover the use of stearic acid in conventional amounts for the conventional purpose of lubrication. Put another way, I believe it was plainly technically obvious to use 1% of stearic acid as a lubricant in a nicorandil formulation. Following the routine pre-formulation work the skilled person would have known it would work for that purpose. Yet that is exactly what the claims cover. They are therefore invalid for obviousness.

The second limb of the obviousness case is based upon the contention that if, during the course of formulation, magnesium stearate failed materially then stearic acid would be an obvious alternative. In my judgment this adds nothing to the first limb. The evidence does not support the proposition that instability of the formulation would be attributed to the magnesium stearate, nor that stearic acid would be an obvious solution to that problem. In my judgment it was not obvious to reject magnesium stearate in favour of stearic acid. I can state my reasons quite shortly.

First, there was no evidence that magnesium stearate would show any incompatibility with nicorandil or any other excipients in excipient compatibility tests.

Thirdly, there was no common general knowledge that different lubricants might assist in solving problems of instability. On the contrary, the conventional ways to deal with instability are set out in paragraph [48] above.

Fourthly, at paragraphs [72] and [73] of his first report as served, Professor Collett put forward as a reason for obviousness that the skilled formulator would have avoided using magnesium stearate because nicorandil is a nitrate ester and such esters are prone to hydrolysis and that such hydrolysis can be catalysed by either acidic or alkaline conditions, such as might be created by magnesium stearate. Prior to giving his evidence Professor Collett amended the report to make clear that the skilled person would avoid using magnesium stearate assuming the results from stability studies revealed nicorandil to be susceptible to alkaline hydrolysis. Under cross examination he accepted that he really had no knowledge of these matters and was essentially relying on information provided to him by Dr Wakefield. In the event, Dr Wakefield accepted that there was no reason to suppose that magnesium stearate would cause stability problems, and this was a view shared by Dr Newton.

Obviousness over 659

The 659 application was published in 1982. It is directed to a method for the production of stable tablets of nicorandil.

The body of the application explains that although crystals of nicorandil are comparatively stable, the inventors (at Chugai) found that they became unstable when moulded into the form of small tablets suitable for administration and that consequently the nicorandil content of the tablets was liable to fall.

The inventors investigated the problem and found that light and oxygen had no effect on degradation. They also found that the problem could not be prevented by carrying out the formulation process under essentially water free conditions.

The examples show the benefit of pre-coating. In Example 1, stearyl alcohol was used as the wax like material and chloroform as the solvent. The resulting tablets (dried and undried) were stored at 40 °C for three months. Stability was evaluated in terms of residual nicorandil as a percentage of the initial weight as compared with tablets in which the stearyl alcohol was replaced with mannitol. The retention of nicorandil was markedly better in the case of the coated crystals.

In Example 2, the wax-like material was stearic acid, present at 0.56% by weight of the formulation. The tablets were made using the same process as in Example 1. The stability tests showed that the addition of stearic acid resulted in a significant increase in nicorandil retention. With dried tablets, 99.5% of the nicorandil remained after three months at 40 °C (compared with 73.1% for the control formulation). With non-dried tablets the equivalent figures were 96.4% and 39.6%.

It would be apparent to any skilled reader of the 659 application that the coating of nicorandil with the lubricant stearic acid does lead to a significant improvement in stability. Accordingly, the difference between the 659 disclosure and the inventive concept is that the 659 disclosure teaches a method in which the nicorandil crystals are coated with the stearic acid using a solvent such as chloroform whereas the invention of the Patent is that stability can be achieved simply by mixing the stearic acid with the nicorandil crystals.

I turn then to consider whether this difference requires any degree of invention. Ivax submitted the answer is “no” for the following reasons. Mixing nicorandil with stearic acid in solid form would have been obvious in the light of the teaching of the 659 application. The skilled man would want to find out if he could do away with chloroform. Mixing the nicorandil with stearic acid (possibly an excess of stearic acid) to enhance coating would be the work of an inquisitive but not an inventive addressee.

In his first report Professor Collett gave evidence which supported the submission advanced by Ivax. He said that the skilled formulator would note the addition of stearic acid produced a significant improvement in stability of the nicorandil formulation but would also be of the view that the process of dissolving it in chloroform was one to be avoided it all possible and an obvious alternative would be to see if the beneficial effects could be produced by direct mixing.

In cross examination, however, Professor Collett took a rather different line. He explained that the reader of the 659 application would understand that the pre-coating was necessary to solve the stability problem. He understood this to reflect the difference in nature between magnesium stearate and stearic acid. Magnesium stearate has a specific surface area five to ten times greater than stearic acid with the result that, in his words, “you cannot stop it mixing” and it coats extremely well. Stearic acid, on the other hand, is not such a good mixer. Hence the need for the solvent. The solvent was used to ensure that the stearic acid was brought into contact with the nicorandil crystals.

Professor Aulton maintained this position under cross examination and elaborated his reasoning. He understood the 659 application to be teaching that the inventors tried increasing the mount of magnesium stearate in the formulation with a view to reducing the friction between the particles on compression. When this did not work they moved to something totally different, which was coating. Mixing with a powder will result in powder particles spread over the surface of the crystals but, inevitably, not providing complete coverage. A solvent, on the other hand, will produce a complete coherent adhesive coating. If the skilled person was concerned about chloroform then he would consider alternative solvents such as ethanol or methylene chloride.

In the light of all this evidence I have reached the conclusion it was not obvious in the light of the 659 application to produce a stable formulation by mixing stearic acid with nicorandil. The application teaches that pre-coating is necessary to achieve a stable formulation. The reader would understand that mixing a lubricant such as magnesium stearate with the nicorandil crystals did not produce a satisfactory result. He would have no reason to suppose that mixing stearic acid would be any better. On the contrary, he would think it was likely to be worse because stearic acid is not such a good mixer. It would be less likely to produce a coat as a result of mixing than magnesium stearate. In summary, the 659 application teaches away from the invention claimed in the Patent.

The proposed amendment

The second is the inclusion of the limiting words: “in an amount of at least 3% of the total weight of the nicorandil-containing preparation”. This, Ivax contends, plainly results in the disclosure of additional subject matter, contrary to s.76(3)(a) of the Patents Act 1977.

The added matter objection has two limbs. First, it is said that in the unamended specification (and the application as filed), there is a teaching that stability can be achieved by using a product falling within either of two generic classes (saturated higher aliphatic acids and saturated higher aliphatic alcohols) on its own at low concentrations of about 0.5% based upon the weight of the formulation.

In the proposed amended specification, so it said, the teaching is materially different. Now there is a narrow class of aliphatic acids (palmitic acid and stearic acid) of which at least 3% must be used, whereas there is still a broad generic class of alcohols which can be used at the lower concentration of 0.5%.

Secondly, prior to the amendment there was no specific disclosure of a stable nicorandil preparation of 3% stearic acid. There was only the general disclosure that the object of the invention could be achieved by the use of any saturated higher aliphatic acid or saturated higher aliphatic alcohol with a lower limit of 0.5% set by claim 6. The only specific disclosures in relation to stearic acid and palmitic acid were in the examples and none of these disclosed the use of less than 4% stearic acid, and that was in conjunction with an organic acid.

Chugai responded that there was no added matter because the minimum amount of the identified sub-class of stearic acid or palmitic acid used by the inventors in the examples, where there is no additional organic acid present, is 3%. This is the minimum quantity disclosed as achieving the enhancement of activity without the addition of the optional organic acid. Moreover, the skilled person would appreciate there was no difference between stearic acid and palmitic acid when used as stability enhancing materials.

In considering the application of these principles and the rival contentions of the parties it is necessary to consider first the disclosure of the specification in its unamended form which corresponds, in all material respects, to the application as filed.

The general teaching of the unamended specification, as both parties agreed, was that 0.5% and above of the two generic classes of saturated higher aliphatic acids and saturated higher aliphatic alcohols can be used to achieve stability. Professor Aulton confirmed this in the course of his cross examination.

It is apparent from this table that the lowest concentration of stearic acid when used on its own and without the presence of any additional organic acid was 8%, and when used with an organic acid was 4%.

In my judgment this is more than a quantitative change. It is a qualitative change. The reader is being taught by the proposed amended specification something about the invention which is not disclosed in the unamended specification, namely there is a sub-class of two acids which behave in a different way to the broad generic classes of aliphatic acids and higher alcohols originally described.

Moreover, the reader is taught that palmitic acid and stearic acid will, for practical purposes, have the same stability enhancing characteristics. This is not expressly disclosed in the unamended specification. As I have indicated, there is no example showing the effect of stearic acid at 3%. Nevertheless, Chugai submitted that the skilled person would, on reading the unamended specification, consider that there that there was no difference between palmitic acid and stearic acid when used as stability enhancing materials. In support of this submission Chugai turned to the common general knowledge.

The principal matters upon which Chugai relied are as follows. First, the molecular structures of palmitic acid and stearic acid are very similar and, on the basis of this, the skilled person would expect their chemical and physical properties to be very similar. Second, stearic acid, as a product available to a formulator, is actually a mixture of stearic acid and palmitic acid, and may even contain as much as 60% palmitic acid. Third, the Handbook of Pharmaceutical Excipients of 1986 does not contain a reference to palmitic acid. Instead references to it are to be found under stearic acid. This reveals that the melting points of the commercial grade materials are, for practical purposes, indistinguishable.

The difficulty with all of the above is that whilst they could possibly give the skilled person grounds for believing that palmitic acid and stearic acid might behave in the same way as stabilising agents, he simply would not know without carrying out further research.

Professor Aulton suggested in his report that it is apparent from the experimental data in the Patent that the enhancement of stability is achieved when at least 3% of palmitic acid or 3% of stearic acid is mixed with nicorandil without the need for an additional organic acid. However, he accepted in cross examination that the skilled person would have known very little about palmitic acid in 1986, there was no evidence anywhere that palmitic acid and stearic acid had stabilising characteristics, the Patent gives no teaching that the two are interchangeable in terms of their efficacy as stabilisers and it is difficult to extrapolate between the examples. Based upon the Patent and the Handbook of Pharmaceutical Excipients, the highest he put it was that it “could” have been his understanding that they would work similarly. He would regard resolving the difference between them as a “challenge”.

Professor Collett explained that palmitic acid was not the sort of thing a formulator would know of. He looked for it in the Handbook of Pharmaceutical Excipients and could not find it. He looked for it in a variety of other sources too. Eventually he found it in the Merck Index. In his second report he explained that the examples in the Patent suggest that palmitic acid is more effective than stearic acid in its ability to stabilise nicorandil. He maintained under cross examination that it was his belief that it might be possible to mill palmitic acid to a smaller particle size than stearic acid with the result that it would give a better coating of the nicorandil cystals and so a greater stabilising effect.

In my judgment this evidence, considered as a whole, does not come near to establishing that there was a clear and unambiguous disclosure that palmitic acid and stearic acid were, for practical purposes, interchangeable or of the efficacy of a 3% concentration of stearic acid as a stabiliser.

For all these reasons I have reached the conclusion that the proposed amendments would offend against s.76(3)(a) of the Act.

Obviousness – proposed amended claim 1

In the light of my conclusion as to the allowability of the amendments this question does not arise. Nevertheless, it is right that I should state my conclusions as to the obviousness of proposed amended claim 1, albeit briefly.

The inventive concept is now limited to a process for the production of a stable nicorandil preparation which comprises mixing the nicorandil with palmitic acid or stearic acid in an amount of at least 3% by weight.

Obviousness over 640

In my judgment the amendment does make a material difference to the case of obviousness over the 640 patent.

I have found that it was obvious to carry out a process within the unamended claims because it was technically obvious to use stearic acid as a lubricant. However, it was, as I have found, common general knowledge that when used as a lubricant, a material should be used in the minimum quantity necessary. It was known, for example, that the addition of too much lubricant could adversely affect the dissolution rate of a tablet in the body and reduce tablet strength.

As to the amount of lubricant that would be used Professor Collett stated, at paragraph [42] of his first report, that the formulation textbooks and literature advise a range of concentrations of stearic acid. In particular he referred to Matsuda, Minamida, Hayashi as advocating a concentration range for stearic acid of 0.1 % - 5% of the overall formulation. He also referred to Modern Pharmaceutics as advising a concentration range for stearic acid of 1% - 4%. That is not an accurate characterisation of either of those publications and the Professor so accepted in cross examination. Indeed he said “this is not the best it of writing I have ever done” and “I would not be pleased if one of the students brought that in to me”. Instead he suggested that that the papers were advocating testing them over that range.

A number of standard textbooks were explored in evidence with both experts. In the light of that evidence I have reached the conclusion that it was known that some formulations contained up to 3 to 4% of magnesium stearate as a lubricant (this was, for example, described in Banker and Rhodes and the Handbook of Pharmaceutical Excipients) but that it was much more likely that the concentration of a lubricant in a new formulation would be no higher than 1 to 2%, and probably less than 1% (as described, for example, in Pharmaceutics: The Science of Dosage Form Design; York; Sprowles and Remington’s Pharamaceutical Science). Those formulations with the higher concentrations were likely to be older and, by 1986, there was an increasing perception that the concentration should be kept as low as possible.

In the light of this evidence I am unable to accept that it was obvious to formulate nicorandil using stearic acid as a lubricant at a concentration of 3%. I have no evidence that 3% would be necessary to achieve a satisfactory lubricating effect and there were good technical reasons for using no more than necessary for that purpose. It was submitted on behalf of Ivax that, on balance, as at 1986, a proposal to use 3% stearic acid would have been regarded as conventional. In my judgment, and for the reasons I have given, the evidence does not support that submission.

Obviousness over 659

I have rejected the allegation of obviousness against the unamended claims. The case against the proposed amended claim 1 fails for the same reasons.

Conclusion

My conclusions are as follows:

Claims 1, 3 and 6 of the Patent as granted are invalid for obviousness, but would have been infringed.

The amendments are not allowable.

Claim 1 as proposed to be amended would not have been invalid for obviousness and would have been infringed.