House of Lords
|Session 2005 - 06|
Publications on the Internet
PDF Print Version
Synthon BV (Appellants) v. Smithkline Beecham plc (Respondents)
LORD BINGHAM OF CORNHILL
1. I have had the privilege of reading in draft the opinion of my noble and learned friend Lord Hoffmann. I am in full agreement with it and would, for the reasons he gives, allow the appeal and restore the decision of Jacob J.
2. Paroxetine is a compound used to treat depression and related disorders. It has for some time been marketed in the form of its hydrochloride hemihydrate salt under the name Paxil or Seroxat. These proceedings arise out of the more or less simultaneous discovery in about 1997 by the appellants Synthon BV, a Dutch pharmaceutical company, and the respondents, Smithkline Beecham plc ("SB"), a UK pharmaceutical company, that a different paroxetine salt, paroxetine methanesulfonate ("PMS"), has properties which make it more suitable for pharmaceutical use. It is more stable, less hygroscopic and much more soluble, so that it can be prepared in higher concentrations.The Synthon disclosure
3. On 10 June 1997 Synthon filed an international application under the Patent Cooperation Treaty for a patent which claimed a broad class of sulfonic acid salts including PMS. This was published on 17 December 1998. The specification said that a known useful salt of paroxetine was the hydrochloride in various forms but that each of them had, to a greater or lesser extent, disadvantages for safe handling and formulation. The object of the invention was to provide a compound with improved characteristics. It then set out, by reference to a formula with a number of variables, a class of compounds which were said to exhibit good stability and high solubility. These included PMS.
4. The specification, as is customary in patents for chemical compounds, then narrowed its focus to a preferred group within the class which was said to exhibit a very high degree of solubility and then, by way of illustration, to a particular compound in that group. That compound was PMS, which features in the first example of the preparation of a salt of paroxetine suitable for pharmaceutical use. Under the heading "Example 1", the specification describes how to make PMS in crystalline form.
5. The notion of crystalline form may require some explanation. The same substance may exist in different solid forms, depending upon the arrangement of its molecules. In crystalline form the molecules arrange themselves in an organised pattern called a lattice which gives the crystal a distinctive shape. On the other hand, in an amorphous form or an oil, the molecules are randomly distributed and the substance has no particular shape. Some substances have only one crystalline form. They are called monomorphic. But others have a variety of patterns into which the molecules may arrange themselves. They are polymorphic. Different crystalline forms can be distinguished by a number of conventional tests. Infra-red radiation (IR) will result in a spectrum of readings of absorbance which are characteristic of that particular crystal. X-ray diffraction (XRD) will likewise give a characteristic series of readings. The IR and XRD readings are the crystal's fingerprint. Some compounds form one or more types of crystals which include, as part of their crystalline structure, molecules of the solvent from which the crystal has been precipitated. They are called solvates or, if the solvent is water, hydrates.
6. Example 1 teaches how to make PMS from a solution of paroxetine, prepared in accordance with a procedure disclosed in a previous US patent, by adding methane sulphonic acid. This is the standard method of producing a salt by adding an acid to a base. In the example, crystallisation is obtained by the use of a seeding crystal which induces precipitation of crystals from the solution. This is said to produce a 99.5% yield of crystals having a 98% purity. The characteristics of the crystals are described in Table 1, giving their melting point (142o to 144oC), DSC (differential scanning calorimetry) curve, IR spectrum and NMR (nuclear magnetic resonance) readings which map the hydrogen and carbon atoms in the structure of the crystal. A note after Table 1 draws attention to the fact that "the compounds of the invention are crystalline, with defined melting points, DSC curves and IR spectra" but that they may be polymorphic and exist in other crystalline forms. The crystals of the acid addition salts with organic sulphonic acids, like PMS, are "substantially free of bound organic solvents", that is to say, they are not solvates, but some "may contain crystallisation water and also unbound water, that is to say water which is other than water of crystallisation." The particular example in Table1, however, is said to be a crystal of 98% purity and therefore not a hydrate.
7. Example 1, as I have said, requires a seeding crystal to start the process of crystallisation and is preceded by a description of how such a seeding crystal had been obtained. It involved dissolving paroxetine in hot ethanol, adding methanesulfonic acid, cooling and then freezing the mixture and evaporating it to reduce it to an oil. After being left for a month, a waxy solid was obtained, part of which was dissolved in EtOAc and the rest used to precipitate crystals from the solution in a freezer.
The SB patent
8. After Synthon had filed its application but before it was published, SB filed a document dated 6 October 1998 which gave it priority for a UK patent application filed on 23 April 1999. The patent was published on 10 May 2000 as UK Patent No 2 336 364. SB, like Synthon, appears at first to have thought that PMS was a novel compound. The SB specification began by saying that it had been surprisingly discovered. The title of the patent is "Paroxetine Salt". But during the course of prosecution, it appeared that there was prior art in which PMS had been identified as one of many paroxetine salts suitable for a method of treatment patented in the United States. SB therefore confined its claim to a particular form of crystalline PMS. It is described in claim 1, upon which other claims are dependent:
9. It will be seen that the claim identified a particular crystalline form by reference to its IR and XRD peaks. If, as the specification said was possible, PMS turned out to be polymorphic, no other forms of crystal were claimed. And it is of particular importance to notice that the IR peaks in claim 1 are different from those in Table 1 of the Synthon application. A person skilled in the art, reading both documents, would think that they identified different polymorphs.
10. Although the specification does not spell out the advantages of PMS or sulfonate salts in general in the way that the Synthon application does, the specification makes it clear that the inventive step was the discovery of the sulfonate salt of paroxetine as an alterative to the hydrochloride salt.
11. The specification suggests a variety of solvents which may be used for dissolving paroxetine before mixing it with commercially available methanesulfonic acid and goes on to say that "the salt may be isolated in solid form by conventional means from a solution thereof obtained as above." Examples of non-crystalline and crystalline solids are given. As for the crystalline form, the specification says:
A number of other alternative methods are then given over the following pages. But there is no suggestion that any of them involved an inventive step. They are all described as involving commonly used solvents and conventional methods (eg "vigorous stirring is particularly useful").
12. On 7 March 2001 Synthon commenced proceedings to have the SB patent revoked on the ground that the crystalline form of PMS described in claim 1 was not new. Section 1(1)(a) of the Patents Act 1977 provides that a patent may be granted only for an invention which is new. Section 2(1) provides that an invention shall be taken to be new if it does not form part of the state of the art. Section 2(2) and (3) define the state of the art:
13. Synthon do not rely on section 2(2). They accept that the crystalline form of PMS identified in claim 1 of the patent in suit had not been "made available to the public", whether by description or in any other way, before the priority date. They rely on section 2(3), claiming that the invention in claim 1 was disclosed by their own patent application.
14. In order to make good their case, Synthon had to satisfy the judge on two points. The first was that their application disclosed the invention which had been patented as claim 1. I shall call this requirement "disclosure". The second was that an ordinary skilled man would be able to perform the disclosed invention if he attempted to do so by using the disclosed matter and common general knowledge. I shall call this requirement "enablement". If both these requirements are satisfied, the invention is not new. I shall later have to discuss the law on disclosure and enablement and the relationship between them in some detail, but for the moment that is enough to explain the course which the proceedings took before the judge.
15. For the purposes of disclosure, Synthon relied upon the fact that their application disclosed the existence of PMS in crystalline form. An immediate difficulty, however, was that the differences in the IR spectra suggested that it was not the same crystalline form as was claimed in the patent. Synthon nevertheless pressed on with experiments designed to show that they also satisfied the requirement of enablement. That means, as I have said, that the ordinary skilled man would be able to perform the invention. Synthon, however, left nothing to chance and engaged Sir Jack Baldwin FRS, Waynflete Professor of Chemistry in the University of Oxford and one of the foremost organic chemists in the world, together with Dr Robert Adlington, whom Professor Baldwin described as the best practical organic chemist he had ever worked with, to conduct the experiments. They were given in a sealed bottle a sample of the mixture of paroxetine dissolved in ethanol mixed with methanesulfonic acid mixed with ethanol described in the Synthon application as having been used to make seeding crystals and asked to reproduce the experiment. The result was a complete failure. The method failed to produce any crystals at all. Eventually, after a good deal of skilled manipulation of a kind not described by Synthon's application, Professor Baldwin and Dr Adlington produced some crystals. These turned out not to have the IR spectrum predicted by Synthon in Table 1. Instead, they had the spectrum described in the patent in suit.
16. Faced with these results Synthon had to retreat and regroup. When it came to the trial, they advanced new arguments. First, they called evidence to show that the IR spectrum in Table 1 of their application was the result of a mistaken reading in their own laboratory. They submitted that the totality of the evidence, including the IR readings of the crystals obtained by Professor Baldwin, showed that PMS was monomorphic. Any PMS crystal would have the characteristics described in the patent in suit. The judge accepted this submission. He therefore found that a disclosure of a crystalline form of PMS necessarily meant that it would be the form described in the patent, even though a person who had read the patent and set out to make the form described in the application might have thought he was making something different.
17. Secondly, Synthon put forward a new argument on enablement. They said that although the method described in the application did not produce seeding crystals, that did not mean that the ordinary skilled man would not be able to crystallise PMS. The trouble with the described method turned out to be that it had used an unsuitable solvent. Ethanol produced a reaction which inhibited crystallisation. But the ordinary skilled man, if not confined to a particular solvent, would try another. Crystallisation was an art rather than an exact science and commonly involved some routine trial and error before results were achieved. For these propositions, Synthon relied mainly upon the evidence of SB's own expert, Dr Ward. They also relied, as evidence of what would have been thought within the abilities of the ordinary skilled man, upon the terms of SB's own specification, which described the process of crystallisation as conventional and capable of being effected with a variety of solvents and in a number of different ways. They abandoned reliance upon any particular method disclosed by their application and relied upon it only for the information it contained about the chemistry of PMS which the skilled man might find useful in choosing solvents and methods of crystallisation.
18. Jacob J accepted this argument as well. He found that the reader of the application, seeking to crystallise PMS, would be able to overcome any problems within a reasonable time. The crystals he made would then inevitably be the crystals described in the patent in suit. The judge therefore held that the matter contained in the application did disclose the existence of the very product which was the subject of the SB patent and that the making of that product was enabled. The patent was therefore invalid.
19. Before I discuss what the Court of Appeal made of these findings, I must say something about the law. I have said that there are two requirements for anticipation: prior disclosure and enablement.
20. The concept of what I have called disclosure has been explained in two judgments of unquestionable authority. The first is Lord Westbury LC in Hill v Evans (1862) 31 LJ(NS) 457, 463:
21. The second authoritative passage is in the judgment of the Court of Appeal (Sachs, Buckley and Orr LJJ) in General Tire and Rubber Co v Firestone Tyre and Rubber Co Ltd  RPC 457, 485-486: