Pages 173-182

NATIONAL JUDGES' PRESENTATIONS

GB United Kingdom

GB United Kingdom - David KITCHIN - Senior judge of the Patents Court of the High Court - Some Recent Patent Cases in England and Wales

A review for the European Patent Judges' Symposium, September 2010

Such has been the weight of patent litigation in the UK that I must confine myself to a review of some of the recent decisions of the Court of Appeal. There are seven I would like to comment upon. They all raise interesting points, some procedural and some substantive.

Let me begin with Virgin Atlantic v Premium Aircraft.1 What order should a national court make when it has finally determined that the national designation of an EP patent is valid and infringed but there is an outstanding appeal to the Technical Board of Appeal? This was the question facing the Court of Appeal after allowing an appeal from a decision of the High Court and finding that Premium had infringed Virgin's EP UK patent for an aircraft seating system. Once again, it highlights the potential for conflict between the national courts and the EPO. Faced with the inevitable uncertainty created by the outstanding appeal, the court had to consider two particular issues: first, what approach should be adopted in relation to the claim for an injunction? Second, what should be done in relation to the claim for damages?

As for the injunction, the court distinguished the position from that of an application by a successful claimant for an injunction after trial but pending appeal to a superior court - in that case the balance of justice test is applied but a patentee will generally get his injunction if he gives a cross undertaking in damages. But where one is dealing with a final injunction with no possibility of appeal to a superior court, the position is very different. Now the claimant has established his right and the Court of Appeal considered it must be a very strong case for an injunction to be withheld. Indeed, in light of Article 3 of Enforcement Directive 2004/48/EC, it would have to be grossly disproportionate. And that remains the position despite a pending appeal to the EPO.

As for damages, the Court of Appeal considered UK law to be clear. A subsequent revocation of a patent by the Technical Board of Appeal does not have the effect of completely annulling a final order to the contrary of our national court2. Where issues of validity and infringement have been finally determined in the national court the matter is res judicata between the parties. If the patent is subsequently revoked in the EPO, that affects the future, in that any injunction will lapse. But the entitlement to damages remains. Though it may seem somewhat harsh, the alternative would introduce uncertainty and the prospect of long delays, neither of which would be good for business.

My next pair of cases reveals the flexibility of the English courts to the issue of bifurcation, at least in the context of general case management. Where a decision on infringement or validity is likely to be dispositive of the dispute it may be just and proportionate to hear one issue before other. The cases, Cook Biotech v Edwards Lifesciences3 and Medtronic Core Valve v Edwards Lifesciences4, both concerned heart valves. In Cook Biotech, the trial judge held that Cook's patent was invalid for obviousness in light of a piece of art called Anderson and was not infringed. In the course of his judgment, the trial judge also had to resolve a question about priority which had a bearing on what Edwards contended was other relevant art. Cook appealed against the findings of obviousness, non-infringement and on the priority question. The appeal was listed for three days. But at the outset of the hearing, it was agreed by the parties that the most efficient way forward was for the Court of Appeal to hear first the appeal against the judge's conclusion that the patent was obvious over Anderson, since if that part of the appeal failed, the rest would fall away. It duly did fail and so it proved unnecessary to hear submissions on all the other matters raised on the appeal, with the result that the appeal was disposed of in less than one day – with a substantial saving of court time and cost.

In the other case, Medtronic, the trial judge held Medtronic's heart valve device did not infringe Edwards' patent but found the patent valid. Edwards appealed against the finding of non-infringement and Medtronic cross appealed, but only contingently; if the Court of Appeal decided that Medtronic's device did not infringe, then it was not concerned with validity. This gave the Court of Appeal the opportunity to bifurcate the appeal by hearing all the argument on infringement first. Only if they concluded that the product infringed would it be necessary to hear the appeal about validity. The appeal on infringement duly failed, the parties were so informed and they agreed a consequential order. Once again, the appeal was disposed of in less than one day. In my view both cases clearly illustrate the benefit of a flexible system which allows the court to determine whether to try the issues of infringement and validity separately or together.

I now come to the substantive law and begin with a discussion of the "person skilled in the art". He appears three times in the EPC, in the Protocol to Art 69 directed to the scope of the claims; in Art 56 directed to inventive step - whether, "having regard to the state of the art, it is not obvious to a person skilled in the art"; and in Art 83 directed to sufficiency – whether the patent discloses the invention "in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art". The question which has not hitherto been answered in the UK is whether the person skilled in the art is necessarily the same for all three purposes. But it arose directly in a recent case concerning an invention for identifying hydrocarbon-bearing layers within sedimentary rock: Schlumberger v Electromatic Geoservices 5.

Prior to Geoservices' invention, oil exploration was carried out by seismic techniques. These can reveal a thin layer within sedimentary strata which might contain hydrocarbon, but they cannot tell the explorer that there is hydrocarbon present because the seismic properties of hydrocarbons cannot be distinguished from water or brine. The solution of the patent was to use a technique called CSEM (Controlled Source Electromagnetic) surveying on a layer previously identified using seismic methods. CSEM has the ability to distinguish hydrocarbons from water and brine. So a combination of the two techniques together can identify hydrocarbon in a way that either technique used on its own cannot. It was apparent from the specification that its implementation required seismics and CSEM expertise. It was also common ground that both were required to determine the scope of the claims. But the crucial question for the court was: what about the position before the invention was made? Must the prior art also be viewed through the eyes of a geophysicist with both seismics and CSEM skills? Despite finding that, prior to the invention, an exploration geophysicist would not have had anything but a vague knowledge of CSEM, the trial judge held that it did. Having so found, he revoked the patent for obviousness. The Court of Appeal took a different view and allowed the appeal. Following Luminescent Security Fibres/Jalon T 422/93, it held that the person skilled in the art for obviousness is not necessarily the same person skilled in the art for performing the invention once it has been made. Here part of the invention – the technical contribution - was bringing the two parts of the team together. As Jacob LJ put it:

"…some inventions are themselves art changing.  If a patentee says "marry the skills of two different arts to solve a problem," marrying may be obvious or it may not.  If it is not, and doing so results in a real technical advance then the patentee deserves and ought to have a patent. His vision is out of the ordinary." 

This brings me to two recent decisions on inventive step. They are further illustrations of what I believe to be a welcome trend in the UK - one of simplification and an attempt to identify whether the patentee has made a technical advance.

The first is Actavis v Novartis6. The alleged invention was a very simple one: a sustained release formulation of fluvastatin. At the priority date in 1996, fluvastatin was a well known statin available in immediate release form. Sustained release formulations (as a concept) were by that time well known and obviously desirable. Their particular advantage is that patients can use a much simplified dosage regime for their medicine and so are less likely to fail to take it. In other words, sustained release is good for patient compliance. Now the patent claimed that the problem with fluvastatin was that it was so highly soluble that it was perceived that the conventional ways for producing a sustained release formulation would not work. Against this background, the inventors claimed to have discovered that, contrary to what the skilled person would have thought, it is in fact possible to make a sustained release formulation of fluvastatin by entirely conventional techniques. The trial judge gave a rather lengthy judgment. He found the patent was wrong about the solubility of fluvastatin; the skilled man would not have thought its solubility was so high that a sustained release formulation could not be made; and the team would have been confident that it would have some efficacy. But he also held (i) it was not obvious - in the sense of a project with a fair expectation of success - that a formulation with improved therapeutic effects or fewer side effects could be made; and (ii) there was no motivation to produce a sustained release form. Nevertheless he held the invention was technically obvious. Not surprisingly, this rather complex analysis encouraged Novartis to appeal.

The Court of Appeal adopted a robust and straightforward approach: while the skilled person may indeed have thought it wasn't worth trying to make a formulation with improved therapeutic effects and fewer side effects, that was not the case in relation to sustained release. There was a technical motivation for that – patient compliance – and it was irrelevant whether or not it would be thought commercially worthwhile. The patent's claim about the solubility of fluvastatin being too high was illusory. Accordingly the patentee had done no more than to say: "you can make a sustained release formulation of fluvastatin" and that was obvious. It had made no technical advance.

The second, Dr Reddy v Eli Lilly7, saw the demise of an archaic and arcane element of UK patent law – selection inventions. The patent concerned an antipsychotic drug called olanzipine used for the treatment of schizophrenia. Lilly discovered it had the beneficial activity of an earlier antipsychotic called clozipine but significantly reduced nasty side effects. Dr Reddy attacked the patent, primarily on the ground of obviousness over an earlier patent which disclosed a Markush formula specifying a molecular structure with various possibilities for the different substituents and said that all the compounds were effective at treating schizophrenia. The formula included thousands of individual molecules amongst which was olanzipine, but it was not spelt out specifically. Dr Reddy said that the patent was bad because it was in truth a selection patent for a particular compound which had previously been disclosed as part of a class which were all described as being useful. In this regard it relied on the rules for selection inventions laid down as long ago as 1930 in the IG Farben case8. Here Dr Reddy said the selection was arbitrary and so failed the rules.

In rejecting this submission, the Court of Appeal declared that the IG Farben rules were no longer part of the living law, despite having been referred to by the House of Lords in 1982 in Du Pont9 and more recently by the Court of Appeal in Hallen v Brabantia10. They were not used by the EPO or the other members of the European Patent Oranisation. Instead, what one should be asking is whether the patentee has made a novel and non-obvious technical advance and provided sufficient justification for it to be credible. If no such advance can be found then, just as in AgrEvo, the patent will be obvious because it will be a mere arbitrary selection from a previously disclosed class. In this case, however, there was a real technical advance because the prior art did not suggest that all the compounds of the class would have the beneficial quality of causing fewer side effects. The selection of olanzipine represented a significant contribution to the art.

My final case is one of considerable importance, and the first in the UK to address in any substantive way the requirements of Arts 52(1) and 57. In Eli Lilly v HGS11 the Court of Appeal upheld the decision of the trial judge that HGS's patent in respect of a novel member of the TNF superfamily called TNFα was invalid on the basis that the claims were not susceptible of industrial application. The case is also notable for a further reason: the Technical Board of Appeal of the EPO had earlier come to the opposite conclusion and upheld the patent on appeal.

At the outset, the Court of Appeal emphasised the importance of co-operation between national courts and the EPO to ensure that commercially important cases are put on a fast track. But it also made clear that the nature of proceedings in England and the EPO is very different, with the trial process in England involving a detailed and intensive investigation and testing of the evidence. By contrast there is much less room for the testing of evidence in the EPO with the result that it is inevitably something of a "coarse filter". Whilst decisions of the TBA are therefore of immense importance on questions of law, it is not appropriate for an English court to give deference to their findings or evaluations of fact.

The Court of Appeal therefore gave detailed consideration to the legal principles articulated in the many that TBA decisions on Art. 57. It concluded the patentee has to disclose a use for his invention which is plausible in the sense of being more than merely "not incredible". There has to be some reason for supposing the statement made in the disclosure is true. Moreover the statement itself must be sufficiently precise. It is not good enough to say this protein or any antibody to it probably has a pharmaceutical use. Such a statement may be plausible, but is of no practical use. You are left to find out what that use is. Applying that test, the finding of the judge could not be challenged.

It has been suggested that this is an instance of the UK "going its own way". I strongly disagree. The Court of Appeal analysed the case law of the TBA with the greatest of care and sought to identify the principles which the TBA has developed. It arrived at a different conclusion not because of a different approach but because the trial judge found the patent disclosure to be speculative and contradictory – findings to which, it may be noted, no reference is made in the decision of the TBA.