Human Genome Sciences Inc v Eli Lilly and Company, United Kingdom Supreme Court, Judgment, Law, casemine.com

Citation Codes

Equivalent Citations

citation codes

Advocates

Appellant Simon Thorley QC Michael Tappin QC (Instructed by Powell Gilbert LLP)
Respondent Andrew Waugh QC Thomas Mitcheson (Instructed by Field Fisher Waterhouse LLP)

See more information ...

Click here to read the full judgment

Human Genome Sciences Inc v. Eli Lilly and Company

Smart Summary

Factual and Procedural Background

The dispute concerns European Patent (UK) No. 0,939,804, which claims the nucleotide sequence encoding a novel human protein referred to as “Neutrokine-α,” together with related amino-acid sequences and antibodies. The application was filed on 25 October 1996 by Company A and granted by The Office on 17 August 2005. Company B opposed the grant before The Office’s Opposition Division, which revoked the patent. Company A appealed to The Office’s Technical Board of Appeal (“the Board”), which reinstated the patent in decision T 0018/09.

Parallel revocation proceedings were started by Company B in the High Court of England and Wales. Judge [Kitchin] revoked the patent for lack of industrial applicability (Article 57 EPC) and for insufficiency. The Court of Appeal (led by Judge [Jacob] and joined by Judge [Hallett] and Judge [Lewison]) affirmed that decision. Company A appealed to the United Kingdom Supreme Court (“the Court”). Company B filed a cross-appeal on insufficiency.

Legal Issues Presented

Whether the disclosure in the patent application met the requirement of industrial applicability under Articles 52 and 57 of the European Patent Convention when assessed as at October 1996.
Whether Claim 1 was insufficient because the patent did not enable a skilled person to put the invention into practice without undue burden.

Arguments of the Parties

Appellant's Arguments (Company A)

The Board’s decision in T 0018/09 correctly applied consistent jurisprudence: identifying a new member of a protein superfamily with known shared functions plausibly establishes industrial applicability.
The High Court and Court of Appeal adopted an unduly stringent standard by requiring identified therapeutic uses or experimental data at the filing date.
Claim 1 is directed only to isolated nucleic-acid sequences; no functional limitation (“Neutrokine-α activity”) is imported, so the patent is sufficient.

Respondent's Arguments (Company B)

The courts below properly held that the patent supplied nothing more than speculative predictions; without concrete data it merely reserved a research field.
Pleiotropic variability inside the TNF ligand superfamily prevents any reliable inference of function from mere sequence homology.
If industrial applicability is nevertheless recognised, Claim 1 is insufficient because it implicitly requires demonstration of “Neutrokine-α activity,” which cannot be achieved without an extensive research programme.

Table of Precedents Cited

Precedent	Rule or Principle Cited For	Application by the Court
T 0870/04 BDP1 Phosphatase/Max-Planck	Industrial applicability demands a “practical application” and not a speculative objective.	Supreme Court accepted the principle but distinguished the facts: unlike BDP1, Neutrokine-α belonged to a family with consistently shared functions.
T 1329/04 Factor-9/John Hopkins	Plausibility may suffice; post-file evidence cannot be the sole basis for industrial applicability.	Relied on to show that Article 57 can be met if the application makes a plausible attribution based on family membership.
T 0604/04 PF4A Receptors/Genentech	Structural assignment to a known family and pharmaceutical interest can satisfy Article 57.	Supported the proposition that sequence and family data can demonstrate a “real possibility” of exploitation.
T 0898/05 Hematopoietic Receptor/ZymoGenetics	“Immediate concrete benefit” is fulfilled if a product is plausibly shown to be usable as a research tool.	Formed the backbone of the Supreme Court’s rationale that assays and research usage constituted industrial applicability.
T 1452/06 Serine Protease/Bayer	Lack of credible functional assignment can defeat Article 57.	Distinguished on grounds that Neutrokine-α had credible functional assignment via the superfamily.
T 1165/06 IL-17 Related Polypeptide/Schering	Membership of a cytokine family with characteristic motifs makes similar activity plausible.	Used to reinforce that common cysteine spacing and other motifs justified plausible function here.
Chiron Corp v Murex Diagnostics Ltd [1996] RPC 535	General observations on industrial applicability in UK law.	Cited but regarded as providing no decisive guidance on the specific biotechnology context.
Brenner v Manson 383 US 519 (1966)	Utility must be specific and substantial.	Referred to as persuasive but not controlling; Court noted differences between US and EPC frameworks.
Fisher v Lalgudi 421 F 3d 1365 (2005)	Requirement that invention deliver a “well-defined and particular benefit.”	Cited similarly to Brenner; acknowledged but not followed.
Generics (UK) Ltd v H Lundbeck A/S [2009] UKHL 12	Need for UK doctrine to align with EPO jurisprudence.	Major support for following the Board’s line of authority.
Conor Medsystems Inc v Angiotech Pharmaceuticals Inc [2008] UKHL 49	National courts may evaluate evidence differently yet should apply common EPC principles.	Used to justify departure from lower courts’ factual evaluations where legal principle was mis-applied.

Court's Reasoning and Analysis

The Court emphasised the imperative of harmonising UK law with well-established Board jurisprudence. Synthesising multiple Board decisions, it distilled a series of guiding points:

The patent must disclose a plausible, specific possibility of exploitation that is derivable from the description and common knowledge.
When a novel protein is credibly assigned to a family whose members share a recognised biological function, that shared function can found industrial applicability without in-vitro or in-vivo data.
Plausibility, not experimental proof, is the threshold; subsequent confirmation may reinforce but is not essential.
Use as a research tool within the pharmaceutical industry constitutes a valid industrial application.

Applying those principles, the Court noted unchallenged findings that: (i) all known TNF-ligand superfamily members are expressed on activated T-cells and co-stimulate T-cell proliferation; (ii) the patent disclosed Neutrokine-α’s sequence, tissue distribution—including expression in B- and T-cell lymphomas—and its superfamily membership. Accordingly, a skilled person in 1996 would plausibly expect Neutrokine-α to share the family’s immune-cell regulatory activities, enabling immediate use in established assays and in diagnostic/therapeutic research relating to lymphoid disorders. The lower courts erred by demanding identified therapeutic end-points rather than recognising the sufficiency of a credible research application.

On insufficiency, the Court agreed with Judge [Kitchin]’s construction that Claim 1 covered isolated nucleic-acid sequences per se without any functional limitation. Because the sequences were fully disclosed and reproducible with routine techniques, no undue burden existed, and the claim was sufficient.

Holding and Implications

The appeal was allowed; the finding of lack of industrial applicability was set aside. The cross-appeal on insufficiency was dismissed. The matter was remitted to the Court of Appeal for resolution of outstanding issues.

Implications: the decision aligns UK law with Board jurisprudence, confirming that plausible assignment of function through family membership and recognised research utility can satisfy Article 57 for biological inventions. It lowers the evidential threshold required at the filing date, thereby offering earlier certainty for innovators while maintaining safeguards against purely speculative claims.