Developments in English Patent Law
01 October 2008
Life is rarely dull in the English courts that get to hear patent litigation, but the first half of 2008 has already proved eventful for them.
Trevor Cook, Bird & Bird
The House of Lords has told the lower courts that they got the law of inventive step wrong in Conor v Angiotech. The Court of Appeal in Lundbeck v Generics has reined in some of the excessive interpretations placed on the Biogen approach to insufficiency in a judgment which, ironically, was authored by the very member of the House of Lords who had written the main judgment in Biogen, but who was on this occasion sitting in the Court of Appeal. The Court of Appeal has also aligned English law with that of the EPO as to second medical use claims in Actavis v Merck. And at first instance Lilly v Human Genome Sciences has provided the first judgment in inter partes litigation in Europe on a so-called "gene patent", which has been held invalid for, among other things, a lack of specific, substantial and credible utility.
Of these the cases most likely to have the widest consequences are Conor and Lundbeck. In its judgment of 9 July 2008 in Conor Medsystems v Angiotech Pharmaceuticals the House of Lords completed the long march by the English courts back from the early years of this decade, when practitioners might have been forgiven for having the impression that to the English courts everything was obvious. Here, the House reversed judgments of the Patents Court in 2006 and the Court of Appeal early in 2007 that had both found the UK designation of Angiotech's patent for a stent coated with the drug taxol to be obvious and thus invalid for lacking inventive step. In so doing it agreed with, and indeed quoted various passages from, a decision of the District Court of the Hague in 2007 that had rejected such an attack in respect of the Dutch designation of the patent. The judgment is as significant for what it does not say as for what it does. It eschews comment on existing tests for, or approaches to, inventive step, or formulating any of its own, no doubt recognising the dangers of these given the irreducible subjectivity of obviousness determinations. Instead it focuses on the specific issue in this case with which the House of Lords disagreed with the lower courts. But in so doing it also puts its seal of approval on a trend in the English courts to find fewer patents obvious, and it will also make it even more difficult than it had already become to found an attack on inventive step based on an "obvious to try" argument.
Lord Hoffmann, giving the lead judgment of the House of Lords, identified the principle in this case as being how one identified the concept embodied in the invention for the purposes of determining inventive step. In essence, the lower English courts had erred by formulating the inventive concept in such a way as to incorporate insufficiency type concerns in their analysis of inventive step. The Angiotech patent was directed to providing a solution to the problem of restenosis, a condition encountered with ordinary stents (tubular metal scaffolds inserted into an artery to keep it open) where the injury caused to the inner layer of an artery by their insertion could produce an exaggerated healing response that restored the original constriction in the artery that the stent was meant to treat. The patent gave only limited experimental data in support of its assertion that the invention would prevent restenosis. But Lord Hoffmann held that obviousness should be determined by reference to the patent claim "and not to some vague paraphrase based upon the extent of his disclosure in the description". There was no requirement that the specification must demonstrate by experiment that the invention will work or explain why it will work.
This was not one of the extreme cases where a patent lacked inventive step because it disclosed nothing more than speculation that did not go beyond what was obvious. Here there was teaching in the specification, based on a particular assay, indicating a likely benefit of using taxol to prevent or treat restenosis, and so it passed the threshold test of making the invention plausible. Having done so there was no reason why the question of obviousness should be subject to a different test according to the amount of evidence which the patentee presents to justify a conclusion that his patent will work. Here the claim in issue was to a stent coated with taxol, the novelty of which was unchallenged. The inventiveness lay not in discovering how to make it but in the claim that such a product would have a particular property, namely, the prevention or treatment of restenosis. Thus the only relevant question was whether it was obvious to use a taxol-coated stent for this purpose. The test that the lower courts should have applied, but did not, was whether it could be shown on the basis of the prior art "that the skilled person would have an expectation of success sufficient to induce him to incorporate taxol in a drug eluting stent?" From what the trial judge at first instance had said in his judgment it was apparent that he would have answered this particular question in the negative, an answer with which the House of Lords also agreed on the basis of the evidence before the trial judge. Thus they reinstated the patent.
Lord Hoffmann observed that the notion of "obvious to try" was useful only in cases in which there was fair expectation of success, and that how much of an expectation was needed depended upon the particular facts of the case. Another member of the House of Lords, Lord Walker, although agreeing, as with other members of the House, with Lord Hoffmann, pointed out that the origins of this particular expression lay in a 40-year-old case that was concerned with "a fairly low tech process", and he observed that with the increase since then in the volume of high-technology research, especially in the fields of pharmaceuticals and biotechnology, the expression had taken on a life of its own, the implication being that it was unlikely to be a useful approach in the future in such fields. Here he echoed and quoted from a view expressed by the retired Patents Court Judge, Sir Hugh Laddie (himself not thought to be a great friend of patents while a Judge), who had observed that it produced an irrational test in that if the reward for finding a solution to a problem and securing a monopoly for that solution was very high, then it might well be worthwhile for large players to examine all the potential avenues that were "obvious to try" to see if one gave the right result, even though the prospects of any one avenue succeeding were remote.
In its judgment of 10 April 2008 in Lundbeck v Generics the English Court of Appeal explained the limits of an insufficiency attack keyed to the issue of the technical contribution to the art, as referred to in T-0409/91 EXXON/Fuel Oils in the EPO and Biogen v Medeva in the UK House of Lords. The claim here was to a single chemical that was later commercialised as the SSRI escitalopram, and that although not explicitly disclosed in the prior art would have been known from it to exist. The prior art compound, of known therapeutic utility, was a racemate (subsequently commercialised as citalopram), namely a 50:50 mixture of two optical isomers. It was known that either one or the other optical isomers or possibly both were responsible for the therapeutic effect of the racemate, but the likelihood, by analogy with other SSRIs, was that one particular optical isomer was. However, this optical isomer could not readily be made at the priority date and the patent taught people for the first time how to do so.
The Patents Court, having held that the disclosure of the racemate did not deprive a claim to the optical isomer of novelty, had then held that inventing one way of synthesising a known desideratum only justified a claim to that synthetic route, and not a product claim to the optical isomer that it enabled. That judgment would have had the effect of rendering the patent protection nugatory, as other methods of making the optical isomer had been developed later. On appeal Lord Hoffmann explained how his judgment in Biogen v Medeva, where the claim was to a specific product made by a certain process, had been misapplied in coming to this conclusion. He held that the Patents Court approach was wrong and so upheld the validity of the product claim, because here the product was itself the invention, even though the process for making it was all that was inventive. For the purpose of enabling it the disclosure of only one such process would suffice.
Another member of the Court of Appeal, Lord Justice Jacob, while agreeing with Lord Hoffman, also explained how product claims for known desiderata were still likely to be insufficient. He gave the example of what is sometimes called a "free beer" claim. As the EPO Guidelines pointed out, a patentee could not normally frame his claim simply by reference to known desirable properties of a product, in particular if they only amounted to claiming the underlying technical problem. He gave a nice example:
61. So, for example, if a man finds a particular way of making a new substance which is 10 times harder than diamond, he cannot just claim "a substance which is 10 times harder than diamond." He can claim his particular method and he can claim the actual new substance produced by his method, either by specifying its composition and structure or, if that cannot be done, by reference to the method ... but no more. The reason he cannot claim more is that he has not enabled more - he has claimed the entire class of products which have the known desirable properties, yet he has only enabled one member of that class. Such a case is to be contrasted with the present where the desirable end is indeed fully enabled - that which makes it desirable forms no part of the claim limitation.
It is interesting to compare the outcome to date in this case in the UK with that of a parallel dispute in Germany, where, following its traditionally expansive approach to the prior art when assessing novelty, rather than the narrow, "photographic" approach of the EPO (and the English Court), the German Patent Court held that disclosure of the racemate deprives the optical isomer of novelty. That decision is itself under appeal to the Federal German Supreme Court. However, things are not yet over in England as the House of Lords has agreed to hear n appeal from the English Court of Appeal decision.
The autumn of 2008 has also started off well for patents in the English Courts. On 8 October 2008 Lord Neuberger, another member of the House of Lords sitting for the occasion in the Court of Appeal, gave the judgment of the Court in Symbian v Comptroller General. This took the Court of Appeal back to the vexed issue of computer implemented inventions, after its decision, less than two years ago, in the joined cases Aerotel and Macrossan. This decision disappointed many commentators at the time, and the UK Intellectual Property Office (UKIPO) relied on the stepwise approach that it set out as providing a basis for refusing most applications for patents for computer implemented inventions that came its way, including the one in issue in Symbian. However, whereas the subject matter in Macrossan had been a business method and that in Aerotel could be presented as a new combination of hardware elements, the subject matter in Symbian involved a computer implemented invention that made a real technical contribution, in terms of both faster and more reliable computer operation. It thus presented the Court with a much better fact pattern than Aerotel and Macrossan for analysing the issues associated with computer implemented inventions and for determining to what extent the English courts, and thus the UKIPO, could align themselves with the accepted EPO practice in this area when considering how to deal with these in the context of the statutorily excluded subject matter of "programmes for computers (...) as such."
The Court of Appeal has risen to the challenge, deftly tracing a path through the EPO case law and that of the English courts to show how most of it is in fact capable of reconciliation and that much of the apparent difference between the two lies in the differing staged approaches that have been developed in an effort to try to deal with the same question in a step-wise way. It instead considered that the case law established that the only real issue in common to both jurisdictions was whether the claimed invention provided a technical contribution to the art. In the present case the claimed invention, a programme that employed an improved method for accessing data from dynamic link libraries (DLLs) resulted in a computer that ran the claimed programme operating better than a similar prior art computer that did not. Accordingly, the Court of Appeal held that this did indeed provide the requisite technical contribution and so did not fail as statutorily excluded subject matter. The result should be to make it easier in future to secure claims for computer-implemented inventions in the UK, and also to preserve such claims when granted by the EPO against challenge in the English courts on grounds other than the traditional ones encountered in most other areas of technology, such as obviousness. We must now hope that computer-implemented inventions can at last enter the mainstream of UK patent litigation with less of a cloud over them than before.