Generics (UK) Ltd v Daiichi Pharmaceutical Co Ltd & Anor

For novelty, a prior disclosure of a racemate does not disclose its individual enantiomers where carrying out the disclosure produces only the racemate and does not necessarily infringe the later patent. Obviousness is assessed as a single question in the light of the skilled person, common general knowledge, the available research routes, the effort involved and the expectation of success. An investigation may be obvious to try only where there is a fair expectation of success. Priority is assessed by asking whether the claimed invention is enabled and directly and unambiguously disclosed in the earlier application as a whole. Later-discovered properties cannot be used to bolster inventive step where they were neither disclosed nor foreshadowed.

GUK sought declarations that an SPC for levofloxacin was invalid and that specified claims of Daiichi’s patent were invalid. The challenges concerned priority, novelty, obviousness, added matter, insufficiency and whether earlier authorisations for racemic ofloxacin were the first authorisations for levofloxacin.

The patent claimed levofloxacin and a process for making it. The principal prior art disclosed racemic ofloxacin and its synthesis. Other publications concerned the enantiomers of structurally different quinolones. The central issues were whether the patent was entitled to its claimed priority dates, whether the prior art disclosed or made obvious levofloxacin, whether claim 5 added matter, and whether the 1997 authorisations were the first authorisations for the product.

1. Priority. The claims were directly and unambiguously disclosed, and enabled, by the first and second priority documents. The absence of information about reduced toxicity and increased solubility was immaterial because those matters were not features of the claimed inventions. However, those later-discovered properties could not be relied on to bolster inventive step where they were not disclosed, foreshadowed or predictable at the priority date.
2. Novelty. The 005 Patent and Drugs of the Future disclosed racemic ofloxacin and its manufacture, including the use of N-methylpiperazine. They did not teach or suggest resolution into the enantiomers. Performing those disclosures produced racemic ofloxacin, not levofloxacin, and therefore did not necessarily infringe the claims. The novelty attack based on the Bayer application also failed because the patent retained its first priority date.
3. Obviousness. Applying the structured approach in Pozzoli v BDMO, the skilled person would have considered it worthwhile investigating whether ofloxacin could be resolved relatively easily. But resolution was unpredictable, the skilled person would first have considered resolving ofloxacin itself, and there was no sufficient basis for assuming success with an intermediate or with the particular experimental route proposed. The skilled person would have redirected efforts to new molecules if routine resolution failed. The experiments therefore represented a research programme with a highly uncertain outcome, not an obvious route.
4. The Riker publications did not make the claims obvious. The structural differences between flumequine and ofloxacin, and the absence of a reliable understanding of quinolone binding, prevented a prediction about the behaviour of ofloxacin’s enantiomers. Gerster IP was available to the public and remained part of the state of the art; information did not cease to be prior art merely because it was not indexed or later retrieved. Even so, its unusual resolving agent, different intermediate and uncertain applicability did not make production of levofloxacin obvious.
5. Added matter and insufficiency. Claim 5 was not an impermissible intermediate generalisation. The claimed final reaction formed part of each disclosed process and was immediately recognisable as applicable to intermediate (V) however that intermediate had been made. The insufficiency attack failed in this court in light of the Court of Appeal’s decision in Lundbeck; the court nevertheless found that neither resolving ofloxacin nor resolving intermediate (V) by the proposed route was obvious.
6. SPC. The 1985 and 1990 authorisations concerned ofloxacin, an active ingredient or combination of active ingredients with properties different from those of levofloxacin and the R(+) enantiomer. They were not authorisations to place levofloxacin on the market. The 1997 authorisations were therefore the first authorisations for levofloxacin, consistently with the scheme and purpose of the SPC Regulation.
7. The claim for declarations of invalidity of the SPC and claims 1–2 and 5–7 of the patent failed. The court would hear argument on the form of order if not agreed.