Takeda UK Ltd v F Hoffmann-La Roche AG (Rev 1)

For novelty, enablement concerns what the prior art discloses, not every characteristic of the historical product. A claim lacks novelty where the prior art enables the skilled person, without undue burden, to make a product falling within the claim. The reproduced product need not be identical in every irrelevant respect.

Patent claims must also provide reasonable certainty. A claim is insufficient where reliable measurement techniques within its wording can produce materially different results, placing the same product inside or outside the claim without the skilled person knowing which result is correct. A claimed technical contribution must justify the breadth of the monopoly.

Takeda sought revocation of Roche’s European patent concerning highly fucosylated glycosylated antibodies. Roche counterclaimed that vedolizumab, the active antibody in Takeda’s Entyvio product, infringed.

The issues included claim construction, infringement, novelty over Bihoreau, Shinkawa, Ferrara and the prior use of Simulect, obviousness including lack of technical contribution, and insufficiency caused by uncertainty in the claimed LCMS measurement method.

1. Disposition. Vedolizumab fell within the relevant claims. The claims were nevertheless invalid and the patent ought to be revoked. They lacked novelty, lacked a technical contribution and were insufficient.
2. Construction. The fucose calculation required all fucosylated glycans found in the sample to be included in the numerator. “Man 4” and “Man 5” were paradigm examples of higher mannoses, so other higher mannose species had to be included in the denominator. The claim’s reference to a functionally active FcR-binding Fc part excluded antibodies in which mutations such as LALA or LAGA made that part cease to be functionally active.
3. Infringement. Vedolizumab satisfied the fucose requirement. The evidence showed some Fc receptor binding and it was more likely than not that the LAGA mutation alone did not account for the absence of ADCC. Vedolizumab therefore fell within the relevant claims.
4. Novelty. Applying Synthon BV v SmithKline Beecham plc [2006] RPC 10, enablement required the prior disclosure to enable a practical result. It was enough that the skilled person could make a version falling within the claimed features without undue burden. Bihoreau, Shinkawa and Ferrara anticipated claims 1 to 3. Simulect anticipated claims 1, 3, 4, 6 and 7, but not claim 2. Bihoreau did not anticipate claims 6 or 7.
5. Obviousness and technical contribution. The patent did not contribute the idea that CHO cells could produce highly fucosylated antibodies, or that increasing fucose could reduce ADCC. Those matters formed part of the common general knowledge or prior art. The wider assertion that 99% fucosylation reduced ADCC to background was not adequately disclosed, plausible or established as true.
6. Insufficiency. A skilled person could use either micro-bore or nano-bore LCMS with QTOF and obtain precise results, yet the techniques could yield different fucose values which placed the same product on opposite sides of the claim boundary. That uncertainty invalidated the claims.