Eli Lilly & Co. v Human Genome Sciences Inc., 31 July 2008 [2008] EWHC 1903 (Pat)

Eli Lilly sought to revoke a European patent owned by Human Genome Sciences on various grounds, including failure to disclose an invention capable of industrial application. The patent disclosed the DNA and protein sequence of ‘Neutrokine-a’, which is a member of the tumour necrosis factor (TNF) cytokine family. The sequence was discovered using bioinformatics and a patent was filed before any practical use had been identified. The patent speculated that it might be used for a very large number of possible uses, based on the known uses of other TNF cytokines.  

There was little UK authority in this area of patent law (just Chiron v Murex) and the Judge, Mr Justice Kitchin, reviewed relevant EPO case law and US law (which does not have an identical requirement of industrial application but has a similar requirement of utility). The Judge set out a series of principles for finding an invention to be capable of industrial application. These are set out below.  

On the basis of these principles, the Judge found that the skilled person would have considered that Human Genome Sciences’ patent did not identify any industrial application other than by way of speculation. The Judge considered that to have included such an astonishing range of applications was no better than to have included none at all. The common general knowledge of the skilled person also did not disclose a practical way of exploiting Neutrokine-a and therefore the claimed inventions were not capable of industrial application at the date of grant of the patent.  

Industrial application has rarely been in issue in UK patent disputes in the past and is an objection that has been given a new lease of life due to certain speculative biotechnology patents. Similar issues are less likely to arise for more recent patent applications as scientific techniques have become more sophisticated and researchers follow up bioinformatics data with functional studies. However, on a more general note, it is clear that the specification must disclose how the invention can be exploited in practice for it to be patentable.  

It is also interesting to consider the approach taken by the Judge in relation to the obviousness of the patent. An obviousness attack of the classic type, based on prior art, failed. The Judge, however, went on to apply the “plausibility threshold” test referred to in Conor v Angiotech and set out in the EPO Technical Board of Appeal case, John Hopkins University School of Medicine. He held that as the invention was entirely speculative, its teaching was implausible and so there was no inventive step. This introduces the concept of implausibility into the law of obviousness and it is not clear that this is correct.