The Supreme Court today handed down its judgment in Eli Lilly v HGS overturning the lower courts' finding that HGS's patent lacked industrial applicability. The case is now likely to return to the Court of Appeal to determine the issues of obviousness and insufficiency.


The patent claims the nucleotide sequence of the gene which encodes for a novel protein - "Neutrokine-a". The patent was revoked by the Court of First Instance, and the revocation upheld by the Court of Appeal, on the ground that the invention claimed in the patent is not capable of industrial application. This ground has not previously been used successfully to revoke a patent but came under consideration because of the unusual means of making the discovery disclosed by the patent.

When an inventor makes a machine, he/she makes it for a purpose and there is no doubt what the industrial application is. Similarly, in the days before high-speed DNA sequencing machines, only the sequences of those proteins already known to be useful were identified; the industrial application was known before the invention was made.

However, the advent of high-speed sequencing changed everything. Researchers could generate sequence data, compare it to prior known sequence data and have a guess at what the encoded protein might be useful for. The patent which is the subject of this litigation was one such patent. It contained no relevant data as to function, no definite indication of any particular function for the protein and no definite indication of any particular industrial application. What it did have was correct sequence data, and good enough homology to a prior known family of proteins called the 'Tumour Necrosis Factor (TNF) family' for it to be identified as a new member of that family.

The TNF family was known to be a diverse one, with each member of the family known up until the date of the patent having both overlapping and differing functions; the common theme was that they were all involved in the immune system. This pleiotropy made it impossible to predict with any certainty what the function of a new member of the family might be, but it did give some confidence that a new member would also be involved in the immune system. That degree of confidence has been judged to be sufficient to render the discovery capable of industrial application. This is despite many years of research that were required after the filing of the patent before a function could be ascribed to the encoded protein, and the many years of research since then required to create a useful product derived from the gene sequence.

Today's judgment

The Supreme Court acknowledged the experience and skill of the courts below and acknowledged that the evidence before the European Patent Office (EPO) and Kitchin J was different. It also acknowledged that Lords Walker and Hoffmann have previously (in Generics v Lundbeck and Conor v Angiotech, respectively) confirmed that EPO decisions need not be followed in all cases. The Court then confirmed that it be "inappropriate to interfere with [Kitchin J's] decision" unless it was one he could not reasonably have reached.

With that background, Lord Neuberger then turned to the size and value of the UK's biosciences sector and commented that it is important for bioscience companies to be able to decide at what stage to file for patent protection, given the dependence of their funding on patent portfolios.

The BioIndustry Association (BIA) had intervened in the Supreme Court appeal and had said that if the Supreme Court were to "agree with the reasoning Court of Appeal there is at least a risk that it will "make it appreciably harder for patentees to satisfy the requirement of industrial applicability in future cases." If that were so, it is suggested that this "would cause UK bioscience companies great difficulty in attracting investment at an early stage in the research and development process"."

Lord Neuberger then analysed the jurisprudence of the EPO and the findings of Kitchin J and came to the conclusion that Kitchin J had not properly applied the principles of the EPO. In particular, Kitchin J had failed to give proper weight to the principles that:

  1. "If all known members [of the TNF family] have a 'role in the proliferation, differentiation and/or activation of immune cells' or have a 'function in controlling physiology, development and differentiation of mammalian cells', assigning a similar role to the protein may suffice."
  2. A "plausible" or "reasonably credible" claimed use, or an "educated guess", can suffice.
  3. Such plausibility can be assisted by being confirmed by "later evidence", although later evidence on its own will not do.
  4. The requirements of a plausible and specific possibility of exploitation can be at the biochemical, the cellular or the biological level.

Had Kitchin J, then the Court of Appeal, applied these principles correctly then they would have found that the patent is capable of industrial application.

Further thoughts on future biotech patents

The following passage from Lord Walker's judgment is telling, and should be borne in mind by future prospective patentees in the Biotech sector.

"I have to say that all my instincts, as an appellate judge, are for dismissing this appeal. The issue is one of multi-factorial evaluation of evidence, a task which has already been carried out twice, with the same result, by a very experienced patent judge, and a division of the Court of Appeal presided over by a Lord Justice with even more experience in the field of patents. Their task was to evaluate the evidence against a statutory test expressed in simple terms, whose meaning is not necessarily made much clearer by elaborate judicial exposition.... Nevertheless the powerful and sustained analysis and reasoning in the judgments of Lord Hope and Lord Neuberger has persuaded me, against my inclination, that this appeal must be allowed. There is nothing that I can usefully add to their reasoning, except to repeat that there are two strong policy arguments for allowing the appeal. The first is to reduce the risk of a chilling effect on investment in bioscience (though here the arguments are certainly not all one way). The other is to align this country's interpretation of the European Patent Convention more closely with that of other contracting states. To my mind these considerations justify this Court in taking what would otherwise be a questionable course."

Clearly HGS's patent falls just on the right side of the "Industrial applicability" line. Future prospective biotech patentees should take heed of Lord Walker's comments and must be careful to ensure they too stay on the right side of that line if they want to avoid such an attack. The EPO principles cited by Lord Neuberger (and listed above) need to be kept in mind when deciding when to file patents on biotech products and thought must be given to the "plausibility" of any suggested role for the claimed product.

Such prospective patentees should remember that 'lack of industrial applicability' remains a real weapon in any patent defendant's armoury.