Novartis v Generics: "Obvious to try" not the true test

The Court of Appeal has today handed down a judgment on a recurring question in the chemical field -- the validity of patents for one stereoisomeric form of a racemic compound (non-chemists may prefer the cats-paw analogy below). 

The decision, Novartis AG v. Generics (UK) Limited (t/a Mylan) [2012] EWCA Civ 1623, is an appeal from the High Court judgment of Floyd J where he held that a patent (and SPC) for the drug rivastigmine was invalid (reported last year by the IPKat's friends at the SPC Blog here).

Enantiomers of a simple molecu

As the IPKat understands them, enantiomers are like a cat's left and right paws: molecules which are identical other than being mirror images of one another. And just like one's paws, left and right versions usually appear in equal quantities (in the laboratory at least)  - a racemic mixture. To round off the analogy, nature sometimes endows one paw - or one enantiomer - with much more activity or effectiveness than the other (the IPKat can testify to this having received many a swat from Merpel's favoured left paw). For a less feline and more rigorous explanation one can refer to Mr Justice Floyd’s summary at paragraphs 20-27 of his judgment.

The patent in suit was for the (-) enantiomer of a compound referred to as RA7 in the prior art. The prior art disclosed RA7 and taught that it was effective against Alzheimer’s disease. It would have been immediately apparent from the structural formula that the compound was chiral, so the only question before Floyd J. was whether it was obvious to investigate the compound further and resolve the mixture into its left- and right-pawed versions. He held that it was obvious, and the patent and its SPC were thus invalid.

In the Court of Appeal, Lord Justice Kitchin upheld this judgment (with Munby and Lewison LLJ concurring). The trial judge had analysed the steps from the prior art to the invention as being:

Unlike the molecules of RA7, a lion's
mirror-image paws can prove difficult
to separate, even for the skilled team

"... putting the matter in the most generous way to Novartis: (a) the choice of RA₇ (b) its resolution into its enantiomers and (c) the preparation of a pharmaceutical containing the (-) enantiomer." [92]

In his decision the trial judge had found that each of these steps was obvious on the evidence before him. 

Novartis, in their appeal, argued at length on the issue of whether it was "obvious to try" to resolve the enantiomers. They relied on evidence that there was no clear indication that resolving the enantiomers would be fruitful and would lead to a better drug, and this issue takes up much of the Court of Appeal's judgment. 

In short, Kitchin LJ endorsed the line of case law most recently set out in MedImmune v Novartis [2012] EWCA Civ 1234, where both he and Lewison LJ (both sitting in this Court also) cautioned against substituing any other test, such as was it "obvious to try", for the fundamental question "was the invention obvious?". In doing so they harked back to the original "obvious to try" case in the UK, Johns-Manville Corporation's Patent [1967] RPC 479, where Diplock LJ said:

"I have endeavoured to refrain from coining a definition of "obviousness" which counsel may be tempted to cite in subsequent cases relating to different types of claims. ... The correctness of a decision upon an issue of obviousness does not depend upon whether or not the decider has paraphrased the words of the Act in some particular verbal formula."

So having disposed of the "obvious to try" argument, which when put at its strongest by Novartis would have only encouraged the skilled person to try where there was a confident expectation of success, Kitchin LJ held [at 58] that the trial judge had correctly applied the law, and that it required no insight or invention to investigate the compounds which the prior art had highlighted as having potential, and then as to the next step, 

"the skilled team would consider that resolution of the racemate might bring practical benefits and would see resolution as a routine step."

This decision can be compared with some of the earlier forays by Generics (UK) to the Court of Appeal when they failed to convince the Court to invalidate similar patents for enantiomers of other compounds.

In those cases, the resolution of the enantiomers was either difficult and uncertain (Generics v. Daiichi) or the skilled person simply didn’t know how to resolve the enantiomers at the priority date (Lundbeck v. Generics). In this case, in contrast, it was trivial for the skilled person to resolve the mixture into its two enantiomers using standard techniques and the evidence pointed strongly to this being a common-sense and routine step when investigating a compound for further development.

postscript - Thanks to Tony McStea for setting the IPKat straight on a subtlety of sterochemistry which had eluded him (the IPKat that is, not Tony) in the initial version of this post. The IPKat has now carefully covered his tracks, or pawprints.