DISPUTES over patent validity rarely turn on the requirement that an invention must be capable of industrial application. In most cases industrial applicability is uncontroversial. However, for some bioscience inventions, particularly gene patents with uncertain applications, it presents a significant hurdle. This was the focus of the Supreme Court's first decision regarding patents, Human Genome Sciences, Inc. v Eli Lilly and Co. [2011] UKSC 51. The case concerned a European patent for the polypeptide Neutrokine-α, its encoding gene, and antibodies to it. In 1996, Human Genome Sciences Inc (HGS) used emerging bioinformatics techniques to identify the gene and determine that Neutrokine-α was similar to proteins in a group called the tumour necrosis factor (TNF) ligand superfamily. Members of this family are involved in regulating inflammation and immune responses by mediating cell activation, proliferation, differentiation and death. Based on this association, but without any supporting experimental evidence, HGS filed a patent predicting that the invention could be used for diagnosis and treatment of a broad range of immune system-related disorders. The patent was granted in 2005.

Eli Lilly, a competitor working on antibodies to Neutrokine-α, sought revocation of the patent in the European Patent Office (EPO) and the UK High Court. In June 2008, the Opposition Division of the EPO revoked the patent for lack of inventive step and added matter. The main argument in all subsequent litigation was that the invention was not capable of industrial application because its function and uses were speculative. This led to revocation by the High Court (Kitchin J.) in July 2008, with additional negative findings on inventive step and insufficiency. In October 2009, an EPO Board of Appeal found that the invention was capable of industrial application, but in February 2010 the Court of Appeal endorsed the High Court's decision. In November 2011, the Supreme Court was therefore faced with deciding between the EPO and national courts. Somewhat controversially, it followed the EPO.

Lords Neuberger and Hope delivered the main judgments in the Supreme Court. Their complementary reasons appear to be motivated by three considerations. First, the desirability of aligning UK and EPO jurisprudence on interpretation of the European Patent Convention 1973. Second, a marked reluctance to interfere with concurrent lower court findings, except where they are legally inconsistent with EPO jurisprudence. Third, policy concerns specific to the bioscience industry that require a clear and certain standard for industrial applicability (implicitly, a low-level standard, as discussed at the end of this note). In support, Lord Walker echoed all three considerations and Lord Clarke noted the second. Lord Collins agreed without comment.

It was accepted in all of the decisions that the patent specified that Neutrokine-α was a member of the TNF ligand superfamily. A skilled addressee would have known that members of this family were expressed on activated T-cells and played a role in co-stimulating T-cell proliferation, and that some members had an effect on B-cell proliferation. Beyond these shared features there was great complexity.

The essential point of distinction between the courts arose from how they treated this information about Neutrokine-α's membership of a particular protein group with several known features. For the High Court and the Court of Appeal, it was insufficient to satisfy industrial applicability. The information did not reveal how the invention could be used to solve any particular problem and merely inspired a research project into its possible applications. For the Supreme Court (and the EPO Board), this information satisfied the requirement of industrial applicability set out in the EPO's jurisprudence, which does not require further evidence as to any therapeutic role of the invention.

Each decision gave detailed consideration to a series of EPO Board cases from 2005–07 concerning gene/protein patents. Kitchin J. extracted nine points of guidance (endorsed by the Court of Appeal) and Lord Neuberger identified 15. The areas where these lists differ indicate distinct perspectives. Kitchin J. emphasised that industrial applicability is not satisfied by speculative indications requiring a research programme or further investigation of the invention itself, noting that the purpose of a patent grant is not to reserve an unexplored field of research, nor to give the patentee unjustified control. Lord Neuberger, by contrast, highlighted that a “plausible use” at either the biochemical, cellular or biological level is sufficient, and may be established by membership of a family with some known functions. A disclosure that is important to the pharmaceutical industry may also be sufficient in itself. These remarks, based primarily on ZymoGenetics (T 0898/05), were reiterated by Lord Hope and led the Supreme Court to conclude that the lower court decisions were inconsistent with EPO authority. The EPO Board's decision in the instant case was used as further support. It is unsurprising that this was consistent with the authorities, given that the same Board (Galligani, Julià and Rennie-Smith) also decided ZymoGenetics and at least one member was involved in each of the other cases.

The Supreme Court's decision is overall a pragmatic one, in the sense that it puts the UK courts on the same course as the EPO and does not jeopardise early generation gene/protein patents. On the other hand, it indicates a divergence from the more stringent US standard of “specific, substantial and credible use”.

An intervention by the UK BioIndustry Association (BIA) was influential in the decision. Public interest interventions are permitted by Rules 26(1)(a) Supreme Court Rules 2009 and 52.12A Civil Procedure Rules 1998, but are rare in intellectual property litigation. The Supreme Court implicitly supported BIA's submission that patent portfolios are a necessary precursor to securing investment for research and development in the bioscience industry. To avoid chilling such investment, a low threshold for industrial applicability (and other validity requirements) inevitably follows. The requirement of “plausible use”, satisfied by even the most general prediction of biological function, is such a threshold. It is unfortunate that equal consideration was not given to the downstream effects of allowing broad patents at an early stage. Even if the patent's legitimacy hinges on common features of a protein family, the protection extends to all claims. This allows the patentee to claim beyond their actual technical contribution and has the consequence of threatening productive independent research by competitors.