INNsight articles by Duncan Curley

In the latest of a series of English court decisions that seem to indicate a trend in favour of patentees, the Japanese company Daiichi Sankyo has claimed a comprehensive victory over Generics [UK] in the battle over the remaining UK legal protection to the API levofloxacin, a member of the quinolone class of anti-microbial agents.   Levofloxacin was originally launched as Cravit® in Japan in 1993 and is marketed in Europe as Tavanic®.  European Patent (UK) No. 0,206,283 B1 (“the Patent”) expired on 20 June 2006, but supplementary protection certificate SPC/GB97/085 (“the SPC”) continues in force and should protect levofloxacin from generic competition until 19 June 2011.   The attack by Generics [UK] so late in the life of the levofloxacin patent estate may be due to the increasing market success of Tavanic® as an anti-infective in Western markets.   

Generics [UK] challenged the SPC, firstly on the basis that the relevant claims in the basic patent (the Patent) designated in the application for the SPC were invalid.  An SPC depends for its life on valid patent claims that protect the active ingredient.   This attack therefore involved consideration of some of the standard attacks on the validity of a patent, including obviousness, lack of novelty, insufficiency and added subject-matter.   The second attack was based on a stand-alone technical point, essentially to the effect that the SPC should not have been granted, because the UK marketing authorisation relied upon by Daiichi in its application for the SPC was not the first authorisation to place the product on the market as a medicinal product, as it was required to be, pursuant to Council Regulation (EEC) No. 1768/92.  Generics [UK] contended that the first authorisation was in fact an authorisation for ofloxacin dated 16 March 2000, since the API ofloxacin in fact consisted of a 50:50 mixture of two enantiomers, one of which was levofloxacin.  This second attack is noteworthy, since it highlights another emerging trend in the patent caselaw, which is an increasing readiness by generic pharmaceutical manufacturers to target SPCs as part of an offensive legal strategy.

Turning first to the attacks on the validity of the Patent, one could be forgiven for thinking that Generics [UK] should have been pushing at an open door in attacking yet another enantiomer patent.  However, this case emphasises the danger of trying to generalise, by way of a broad statement of principle about the likely validity (or not) of patents claiming particular stereochemical forms of a previously-known drug, merely on the basis that many in the pharmaceutical field were examining enantiomers by the relevant priority date.  In particular, in Generics [UK] v Daiichi Sankyo, the Judge found that the quinolone field was unusual, in that the medicinal chemists researching this field in 1985 were more intent on developing new chemical entities, rather than examining the relative activities of enantiomers of existing molecules.   In doing so, the Judge relied on evidence brought by Daiichi from an expert who was working in the quinolone field at the relevant time.  This evidence of the skilled person’s “mindset” or motive to resolve proved to be critical in the Judge’s assessment of obviousness.  

Generics [UK]’s argument on obviousness proceeded on the assumption that a chemical intermediate to ofloxacin disclosed in a prior art document was an obvious target for resolution in 1985.  Once this intermediate had been obtained in enantiomerically pure form, it was then a routine matter to obtain the enantiomers of ofloxacin (including levofloxacin) by means of well known chemical reactions.  

The first difficulty with this obviousness case as presented by Generics [UK] was that all of the experts agreed that the starting point for the skilled person would have been ofloxacin itself, and not the intermediate.   Assuming that the skilled person was unsuccessful in resolving ofloxacin, what would he have done next?  The Judge was not persuaded that the skilled person would have attempted to resolve one of the intermediates and proceed that way.  In light of the preference of the skilled person working in the quinolone field for new chemical entities, the Judge thought that the skilled person might well have given up on the resolution project at this point and re-commenced his search for new molecules instead.  

Nevertheless, on the assumption that the skilled person would choose the racemic intermediate as his starting point, Generics [UK] then sought to prove by means of experiments that there were obvious ways in which the skilled medicinal chemist would have sought to resolve it.  Resolution was then (in 1985) what the Judge described as an empirical art (Daiichi’s expert witness said that resolution of racemates was a “black art”).   As such, it was not possible to predict with any degree of certainty how the enantiomers of a known racemate could be prepared.   The methods available for resolution in 1985 included the use of an enantiomerically pure resolving agent, that would react with each of the target enantiomers in the racemic mixture so as to produce physically separable diastereoisomers.  Generics [UK] relied on a textbook which disclosed menthyl chloroformate as a suitable resolving reagent that was worth trying.  They alleged that this book was part of the skilled person’s common general knowledge, but the Judge rejected this, concluding that the book in question was one of a number of specialised source or reference books available in some (but not all) pharmaceutical companies.  A second method was separation by chiral HPLC, but the Judge did not accept that preparative HPLC was used routinely by medicinal chemists in 1985.  

There were three other prior art citations but in relation to each of them, the Judge concluded that there was no additional information that would have led the skilled person to resolve ofloxacin.   In conclusion, the Judge found that it was not obvious to resolve ofloxacin and to produce levofloxacin.   All the other attacks on the validity of the Patent (lack of novelty, insufficiency and added subject matter) also failed. 

That left the stand-alone attack on the levofloxacin SPC.   The key to this aspect of the case was whether the earlier marketing authorisation for ofloxacin could also be considered an authorisation to place levofloxacin on the market as the active ingredient in a medicinal product.   The Judge held no – ofloxacin, levofloxacin and the R(+) enantiomer of ofloxacin all had different therapeutic effects.    As such, ofloxacin was itself a distinct active ingredient, albeit made up of two components, levofloxacin and the (less active) R(+) enantiomer of ofloxacin.   The attack on the SPC therefore failed. 

In conclusion, Generics [UK]’s main case can be categorised as an obvious to try case, involving several “what if” steps towards a so-called obvious desideratum (in this case, the active enantiomer of ofloxacin).   Such cases are inherently difficult, because they require the Judge to embark on a series of findings as to the plausibility of various steps, with the attendant danger of a charge of a hindsight – fatal to a conclusion of obviousness.  We have seen similar such obvious to try cases founder recently (escitalopram, olanzapine) and it seems that if the generic firms are going to turn the tide of recent defeats, a return to more conventional obviousness attacks should be the way forward.