Making a case for protein claims

Eli Lilly & Co v Human Genome Sciences Ltd was a case before the UK's Court of Appeal.  It concerned a patent which covered Neutrokine-alpha protein, antibodies to it and the polynucleotide encoding it.  The description of the patent set out a large number of potential applications for the claimed entities.

The judge in the High Court (the UK's first instance court) had revoked the patent for lack of industrial applicability, insufficiency and lack of inventive step.  The Court of Appeal have now upheld his decision.

Under UK patent law, an invention must be "susceptible of industrial application" if it is to be patentable.  This means that the invention must be capable of being made or used in some kind of industry: for example it can be usable to treat or diagnose a particular disease.

In the area of gene and protein sequences, it is established law that an isolated gene sequence is patentable if the patent application discloses uses and/or effects for the protein which it encodes.  One way of satisfying this requirement is to show that the protein has an amino acid sequence homologous to that of a well-characterised family of proteins which have a known function and/or effect, so that the uses of the new protein can be readily predicted.

The Court of Appeal believed that a patent application, as filed, must establish that it is at least "plausible" that the claimed invention solves a technical problem.  Merely proposing a problem to be studied further is not enough.  Later filed experimental evidence cannot be used as the sole basis for establishing that the problem has been solved.  Whether these criteria are met depends of course on the facts.

In this case, the judges found that the application as filed had not shown that it was plausible that Neutrokine-alpha had any practical application.  They said that "there must be some real reason for supposing" that statements about potential uses are true.  Here, the application suggested that Neutrokine-alpha was probably a member of the TNF superfamily and went on to speculatively list a large number of possible downstream effects and utilities (some of which were contradictory).  However, it was shown by earlier relevant documents and expert testimony that existing members of the TNF superfamily did not always act in the same way.  In addition, publications from after the date of the patent application indicated that there had been difficulties with establishing the activities and effects of Neutrokine-alpha.

For these reasons, the Court of Appeal agreed with the first instance judge that the claims lacked industrial applicability.

This case underlines the importance of including as much experimental data as possible when filing a patent application on a new protein.  If you do not have experimental data, and the protein cannot be shown to be part of a well-characterised family with well-understood functions, then you should consider carrying out further work before filing a patent application.

A UK or European patent application must contain, at its filing date, enough information to make it plausible that the invention is industrially applicable and solves a technical problem.  It must provide reasonable certainty, and not just a possibility, of an alleged use.

If you have any queries concerning the issues raised in this case, please contact Rhiannon Turner .