Lipitor and Paroxetine patent decisions handed down by the UK High Court and House of Lords - by Ralph Cox of Stringer Saul LLP
In the Lipitor case, Ranbaxy v Warner-Lambert ([2005] EWHC 2142 (Pat)) the High Court held that a claim to a racemic mixture is infringed by use of the pharmaceutically active enantiomer alone. The Court also commented on the drawbacks to the European Patent Office’s problem/solution approach to obviousness.
Warner-Lambert had 2 European patents that covered atorvastin, its cholesterol synthesis inhibitor sold under the brand “Lipitor”. Ranbaxy applied for a declaration of non-infringement of the earlier “633” patent and for revocation of the later “281” patent. The non-infringement argument was that the 633 patent described the synthesis of and claimed the racemate for atorvastin only and so should not cover the optically pure enantiomer that it wanted to sell. Mr Justice Pumfrey disagreed. Though the 633 patent referred throughout to the racemate, the question on construction was why the patentee would wish to exclude the effective enantiomer. There was every difference between a surprising omission for which there might be a reason and one that deprived the 633 patent of any commercial effect. Accordingly its claim covered the enantiomer as well.
The construction aspect of the decision was based on the findings of fact that it was common general knowledge in the field that pharmaceutical activity generally lay in one enantiomer only, that there was a tendency to resolve racemic mixtures to obtain the single enantiomers where practicable and that the techniques involved were routine. While offering some further comfort to the Paroxetine case discussed below on the degree of precision needed in drafting, it raises a question mark over the inventiveness of patents for isomers. It has been accepted practice at the UK Patent Office that such patents require strong evidence of surprising superiority of the claimed isomer over the other isomers, one form of such evidence being that all activity lies in that isomer. Unless it can be argued that the decision is limited to the particular facts of this case, isomer patents may become more difficult to obtain and/or uphold in the future.
The second interesting aspect to the decision is its rejection of the European Patent Office’s approach to obviousness. This so-called problem/solution approach involves identifying the closest prior art, establishing the technical problem to be solved and considering whether, from the starting point of the closest prior art and the problem, the claimed invention is obvious. Mr Justice Pumfrey considered that this approach had 2 difficulties. First, it assumes that it cannot be obvious to start from something that is not the closest prior art, which offends the principle that the skilled person can start with any prior available material. Secondly, as here, the reformulation of the problem can obscure that which is objectively obvious. Here the cited prior art (the application for the 633 patent) specifically listed 7 metal salts that could be used and there was nothing inventive in selecting the calcium salt, the one claimed in the 281 patent.
A third aspect of the case is the salutary reminder of the care that needs to be taken by patentees in building up a portfolio of patents covering a particular product. Here, as well as Warner-Lambert’s 281 patent being obvious in light of its 633 application, the 281 patent was anticipated by a second application (WO 89/07598). Although the 598 application was not published until after the priority date of the 281 patent, it was filed by Warner-Lambert
beforehand and so was available as an anticipatory reference under section 2(3) of the Patents Act 1977.
Since the House of Lords decision in Asahi’s Application ([1991] RPC 485), an anticipating disclosure has had to be enabling (i.e. allow the invention to be performed). However, there was some uncertainty as to what was necessary over and above the fact of disclosure to meet this requirement. Guidance has now been given by Lord Hoffmann in his decision in the Paroxetine case, Synthon v Smithkline Beecham ([2005] UKHL 59).
SKB’s patent claimed crystalline paroxetine methanesulfonate (PMS) identifiable by characteristic peaks in an infra red (IR) spectrum. Synthon had filed an earlier application for a broad class of sulfonic acid salts including PMS. The question was whether this earlier application was an “enabling disclosure” capable of invalidating SKB’s patent, especially given that the example for making crystalline PMS in Synthon’s application specified use of a solvent that would prevent the procedure from working and gave a table of IR peaks which, due to a laboratory error, were incorrect and different to those given in SKB’s claim.
The “disclosure” requirement for anticipation
Lord Hoffmann held that, to be an anticipating disclosure, “the prior art must disclose subject-matter which, if performed, would necessarily result in an infringement of the patent” (anticipation and infringement being opposite sides of the same coin). Further, as patent infringement does not require awareness that one is infringing, nor does anticipation. Hence, while someone following Synthon’s example might well be puzzled to find their IR spectrum was different to that expected, they would have made the crystals and they would necessarily infringe SKB’s patent because PMS in fact had only one crystalline form with the IR spectrum claimed by SKB. SKB’s PMS was therefore prior disclosed.
Lord Hoffmann then went on to assess whether the disclosure was also enabling, a separate and distinct requirement involving a consideration of whether the skilled person would in fact have been able to perform or work the prior disclosure. This contrasts with the role of the skilled person in relation to disclosure, which is to construe what is meant by the prior art. Lord Hoffmann also clarified that it is the invention disclosed in the prior art not the invention in the patent under attack that has to be enabled.
Also on enablement, the skilled person is assumed to be willing to make trial and error experiments to get the prior disclosure to work but the degree of knowledge, skill and perseverance expected of the skilled person would vary with the field. In this case, the first instance judge (Mr Justice Jacob as he then was) had found that the skilled man would have tried an alternative solvent from his common general knowledge when the specified one failed and so would have made PMS crystals within a reasonable time. Lord Hoffmann was reluctant to interfere with a finding of fact by an experienced IP judge and accordingly the disclosure in Synthon’s application was enabling and SKB’s patent anticipated.
From the perspective of IP lawyers and those seeking to invalidate patents, this decision brings clarity to the law of anticipation and what is required in the prior art relied on. From the point of view of patentees it is something of a mixed bag. On the positive side, as the test for
enablement on sufficiency is the same as that for anticipation, there is some comfort to patentees that errors in their specifications will not necessarily lead to the patent being invalid for insufficiency. Not so good is that “incorrect” prior art may still anticipate even if the person following that art has no awareness of what he is doing. Anticipation by “inevitable result” has long been a part of the case law on lack of novelty but this decision arguably broadens the scope of that doctrine to the detriment of patentees.
This article was first published in Pharma Patent Bulletin, December 2005
1.6 Riskfaktorer Venös tromboembolism (VTE) är en multifaktoriell sjukdom medinteraktion mellan ärftliga och förvärvade riskfaktorer [64,203,205,211,252]. Ett bekymmer i riskfaktorstudier är att närvaron av ett statistisktsamband inte nödvändigtvis innebär ett kausalt förhållande. Det tycksockså vara så att det ofta krävs att flera riskfaktorer samverkar för att entrombos ska bi