Idenix Pharmaceuticals Inc v Gilead Sciences Inc & Ors

A patent claim must be justified across substantially its whole scope by a plausible technical contribution. Plausibility is a low threshold, but the specification and common general knowledge must provide a real reason for supposing that the promised technical effect will occur. Speculative assertions and invitations to screen compounds do not suffice.

A patent is insufficient if it neither makes the claimed effect plausible across the claim nor enables the skilled person to perform the invention without undue burden. A broad class may be claimed where a disclosed principle supports a reasonable prediction across that class. It may not be claimed where identifying effective members requires a substantial research project.

An amendment adds matter if it selects a narrower class which was neither disclosed nor clearly and unambiguously derivable from the application.

The proprietor of a European patent claimed that the defendants infringed claims covering a large family of nucleoside analogues intended to treat hepatitis C and other Flaviviridae infections. The defendants admitted no liability and counterclaimed for revocation on grounds including lack of novelty, obviousness, insufficiency and added matter.

Arnold J held in [2014] EWHC 3916 (Pat) that the defendants' sofosbuvir product fell within the claims, but that the patent was invalid for lack of novelty, inventive step and sufficiency. He also found added matter in one granted claim and in a proposed amendment. The proprietor appealed on almost every issue, while the defendants challenged the rejection of their added-matter attack on claim 1 as granted.

The principal questions were whether the claimed antiviral effect was plausible, whether the compounds could be made and the invention performed across the claims without undue burden, and whether the granted and proposed claims added matter.

1. The appeal was dismissed unanimously. The judge was entitled to find that it was not common general knowledge at the priority date that the relevant methyl-hydroxy nucleoside analogues had potential efficacy against hepatitis C. Common general knowledge comprises information generally regarded by the bulk of those engaged in the field as a good basis for further action. It can include reliable material which the skilled person knows exists and would consult as a matter of course.

2. The patent was invalid for obviousness. A technical effect relied upon as the inventive contribution must be shared by substantially everything within the claim and must be plausible from the specification read with the common general knowledge. Plausibility is a low threshold directed against speculative claiming, but there must be a real reason for supposing that the promised effect will occur. The patent contained no informative experimental data or rationale supporting antiviral activity across the immense claimed class. Its statement that the compounds might inhibit the virus and could be screened amounted to speculation rather than a technical contribution.

3. The patent was also insufficient. The disclosure did not make it plausible that the invention would work across the claims. Nor did it enable the skilled medicinal chemist to synthesise the compounds bearing the critical methyl-fluoro configuration without undue burden. The specification gave no meaningful guidance on the difficult stereospecific fluorination step. The skilled person faced numerous possible routes, uncertain prospects and a task dependent upon skill and luck. The judge properly treated the sustained unsuccessful work of a representative chemist, together with the expert and documentary evidence, as relevant to that objective assessment.

4. A patent may require routine trials, ordinary trial and error and, in appropriate circumstances, a literature search. It must not require prolonged research or a substantial research project. The court inclined to agree that selecting, synthesising and testing compounds from billions of possibilities would impose such a project, but did not finally decide that alternative breadth-based insufficiency issue.

5. Claim 1 as granted did not add matter because its compound class was clearly and unambiguously disclosed as a distinct preferred embodiment in the application. Claim 4 did add matter because its particular selection of eight bases was not disclosed. The proposed amendment also added matter by creating a narrower class which was neither disclosed nor clearly and unambiguously derivable from the application. The judge rightly refused to preserve the systemically deficient claim without amendment.

1. Court of Appeal (Civil Division): The proprietor's appeal was dismissed unanimously. The defendants' respondent's notice concerning added matter in claim 1 as granted was rejected. [2016] EWCA Civ 1089.
2. High Court, Patents Court: Arnold J held that the defendants' product fell within the patent claims, but revoked the patent for lack of novelty, lack of inventive step and insufficiency. He rejected most of the added-matter case, while finding added matter in claim 4 and in the proposed amendment. [2014] EWHC 3916 (Pat).