In the first patent case to be considered by the new UK Supreme Court (formerly the House of Lords), the court considered the validity of a patent of Human Genome Sciences for the nucleotide sequence of the gene which encodes for Neutrokine-α, a newly discovered member of the tumor necrosis factor (TNF) family of cytokines. The question at issue was whether the newly described sequence had any industrial application (under Article 57 EPC) or, as was held at first instance and on appeal, the patent described no more than the starting point for a research project because its disclosure as to the function of the sequence was too vague and speculative.
It was accepted that HGS had found the protein and shown that it was a member of the TNF superfamily, known to be mediators in immune regulation and inflammatory response. The problem was that the application described a wide range of purported uses - so wide as to be "puzzling" and indeed "fanciful" - leaving the reader, perhaps, with the conclusion that the inventors had "no idea" as to the activity of Neutrokine-α .
Not so. The Supreme Court found the patent to be valid (Human Genome Sciences v Eli Lilly,  UKSC 51). While the drafting of the patent specification was open to criticism, the reader would not actually have been diverted from its central teaching in the light of what was known of the common properties of the TNF ligand superfamily.
This was almost the end of a long saga of litigation through the EPO Board of Appeal and then three levels of the UK courts. Another chapter of the saga is set out above under "Court of Appeal allows reach-through claims". The important question before the Supreme Court was that of timing of filing of a patent application in this field, and how much research is enough to support filing of a patent application. Is it necessary to conduct tests to provide experimental data to prove that a protein (or its antagonists) has therapeutic use? Such tests take time, risk disclosure and may be hard to fund if there is no patent protection.
At the end of the day, the Supreme Court followed the jurisprudence of the EPO Boards of Appeal, carefully analyzing a number of cases of new neucleotide sequences, some of which satisfied Article 57 (and Article 56) and some not. The dividing line is very fact-specific. Here, we set out four examples upon which the Court relied, two on each side of the line.
The Supreme Court agreed with the EPO Board of Appeal in T0898/05 that the function of a protein and the nucleic acid encoding it can be seen at different levels. These include: (i) the biochemical activity of the protein, i.e. its molecular function; (ii) the function of the protein in cellular process, i.e. its cellular function; and (iii) the influence of those cellular processes within a multicellular organism, this being its biological function in a broad sense. For the purpose of industrial application under Article 57, none of these levels is more fundamental than the others.
Note that these cases are not about sufficiency of disclosure. They are not about whether the patent specification provides adequate experimental instructions for the skilled person to be able to reproduce the claimed polypeptides. They stand or fall on whether there is an adequate and plausible description that the novel sequence belongs to a family or class for which there is some anticipated use. If they fall, it can be through lack of industrial application (Article 57) or lack of inventive step for failure to solve any technical problem (Article 56).
The drafting of the patent was the subject of significant criticism, which is easy with hindsight. The applicants over-egged the pudding by listing such a range of putative therapeutic uses as to undermine credibility. Doubtless the authors of the patent application wished to convey that the newly identified member of the TNF family offered the prospect of being an improvement over known members. This was not necessary. It would have been enough to attribute to the new member the known functions of the family.
Similarly, in Conor v Angiotech para 52, the "tribulations of litigation" were said to have been brought on the patentees by themselves because the very lengthy patent sought to cover as much ground as possible but in doing so risked making the patent "so unfocused as to end up with nothing capable of resisting a challenge to its validity".
There is danger, particularly before the EPO, in drafting a patent application with lists of functions, features or parameters without giving at least "educated guesswork" as to hierarchies or sub-groups among these, e.g. pointing to how they might be grouped or combined to solve different problems. Long lists of features or ranges of parameters, for example, do little to serve as support for claims.
At the risk of mixing metaphors, if you use too much egg, you cannot pick out the cherries
Not Patentable - Mere speculative indication of possible objective that might or might not be achievable
T 0870/04 BDP1Phosphatase/ Max- Planck (11 May 2005) BDP-1, a new polypeptide, said to be a member of the so-called PTP-PEST family. The application suggested that PTPPESTs played an important role in certain specified cellular functions, and were possible “candidate anti-cancer proteins”. It also disclosed that BDP-1 was expressed in most tissues and cell lines, particularly in epithelium origin cell lines and in cancer cell lines. The application did not explicitly disclose the specific nature and the possible significance of the suggested roles for BDP1. It stopped short of suggesting, let alone identifying, an anti-cancer activity for BDP1 or a therapeutic use of BDP1 as a tumour-suppressor agent. There was no evidence as to whether BDP1 played a passive role or an active role in cancer. Nor could the identification of BDP1 as a PTP-PEST be taken as any clear indication of its function or use, as the prior art did not attribute clear functions to PTPPESTS as a class.
Patentable - Reasonably credible indication of function
T 0604/04 PF4A receptors/Genentech (16 March 2006) Certain polypeptide members of the PF4AR family of chemokine receptors. It was not certain that the claimed polypeptides were receptors for members of the PF4A family, but certain structural features make it plausible that this was indeed the case. The patent provided a structural characterisation which enabled their assignment to the category of receptors which bind members of the PF4A family and, insofar, indicated what their function might be. The function was at best incompletely understood and the data did not show to which ligands the polypeptides bind, but the skilled person would have understood that any role of a given chemokine was reflected in its receptor and it was clear that chemokines as a family were not only interesting in research but also important for the pharmaceutical industry irrespective of whether or not their role had been clearly defined.
T 1329/04 Factor9/John Hopkins (28 June 2005) The application sought to identify a new member of the TGF-β superfamily, the known family having been described as having influence on a wide variety of differentiation processes such as adipogenesis and myogenesis. The application described TGF-9 and alleged it was a member of the superfamily. But there was a significant structural feature that was not identical between TGF-9 and the members of the TGF-β superfamily. No functional characterisation of TGF-9 was forthcoming in the application. There was not enough evidence in the application to make at least plausible that a new member of the superfamily had been found. Postpublished evidence may not serve as the sole basis to establish that the application solves the problem it purports to solve.
T 0898/05 Hematopoietic receptor/ZymoGenetics (7 July 2006) This patent application disclosed and claimed the nucleotide sequence and the encoded amino acid sequence of the cytokine Zcytor1. It disclosed the tissue distribution and claimed that Zcytor1 had various therapeutic roles of which a fair number of possibilities were given. No experimental evidence was provided to support these claimed roles or uses. Based in part on computer-assisted sequence homology studies, the Zcytor1 receptor was identified as a putative member of the hematopoietin receptor family and it was assigned a role in proliferation, differentiation and/or activation of immune cells, from which was derived the possible therapeutic role for its ligands. The application gave no experimental evidence for the suggested role of the receptor or its ligands. Later evidence, however, confirmed this sort of ‘educated guess’, which was ‘reasonably credible’.