Celltrion Inc. v Genentech, Inc & Anor

Patent claims must be construed according to their ordinary meaning, informed by the skilled person and the specification. Numerical terms such as about allow fact-sensitive flexibility, but a specified ingredient or concentration remains a limitation where the skilled person would understand that it was deliberately included.

For novelty, a prior disclosure must provide an individualised description of the claimed subject matter. Selection from ranges or lists is assessed in context, including the number and interdependence of alternatives, overlap and any pointers. Equivalents of the claimed invention are irrelevant to novelty. Obviousness is a multifactorial assessment; no fixed percentage threshold governs reasonable expectation of success.

Celltrion sought revocation of a patent owned by Genentech and Novartis concerning a liquid formulation of omalizumab. The patent claimed approximately 150 g/L antibody in histidine, arginine-HCl and 0.04% polysorbate 20, at pH 6, together with dependent claims.

The attacks alleged lack of novelty, lack of inventive step, lack of technical contribution, insufficiency and added matter. The principal prior art included the Liu formulation disclosure, the Xolair formulation and Shiraki. The central issues were the construction of the claims, whether Liu individually disclosed the claimed combination, whether equivalents were relevant to novelty, and whether the claimed formulation was obvious or insufficiently supported.

1. Construction. Claim 1 was limited to a liquid formulation manufactured, stored, transported and administered as a liquid. It therefore required sufficient stability between manufacture and administration. “About 150 g/L” was construed as approximately 145–155 g/L. The specified 0.04% polysorbate 20 was a deliberate limitation and was not construed to include 0.02%. “Histidine” meant histidine-HCl because the formulation was defined by the contents of the solution, including 0.2M arginine-HCl.
2. Novelty. Liu did not anticipate claim 1. Its disclosure required a selection from a concentration range and a list of polysorbates. The skilled team would not derive an individualised description of the claimed combination. The assessment was not merely a numbers exercise, but the effective number of alternatives, their independence, the absence of preferences and the nature of the disclosure all pointed away from anticipation.
3. Equivalents. Equivalents of the claimed invention were irrelevant to novelty. The scope of protection may be wider for infringement, but the Formstein defence prevents a patentee enforcing an equivalent which would itself be anticipated by the prior art. The alternative novelty argument therefore had no basis in English law.
4. Inventive step. Shiraki would have encouraged the skilled team to add arginine to Liquid Xolair and to try formulations broadly resembling the Dalby Screen, with a reasonable expectation of obtaining a stable, administrable liquid formulation. However, there was no reason to select approximately 150 g/L omalizumab or 0.04% polysorbate 20. No fixed 50% threshold applied to reasonable expectation of success; obviousness remained a multifactorial assessment, including motivation and the factors identified in Actavis Group PTC EHF v ICOS Group [2019] UKSC 15.
5. Technical contribution and other grounds. The technical contribution was the complete formulation, not the isolated antibody concentration. The patent plausibly disclosed a liquid formulation suitable for administration, made a technical advance over the lyophilised Xolair product and had narrowly supported claims. The insufficiency attack failed because claim 1 required histidine-HCl. The added-matter attack failed because the application disclosed the formulation as such, not merely products made by the disclosed processes.
6. Disposition. The patent was valid and infringed.