INNsight articles by Duncan Curley

The unsuccessful attack mounted by Dr Reddy’s, Teva and others in 2005 on Eli Lilly’s US patent to the blockbuster anti-psychotic drug product Zyprexa® has been widely reported.  The validity of US Patent No. 5,229,382 claiming the API in Zyprexa® - olanzapine  - was upheld at both District Court level and subsequently (in December 2006) by the Federal Circuit.  The fight then moved to Europe, with the German Federal Patent Court rendering a nullity decision in 2007.  This decision is presently on appeal, but meanwhile the European battleground over olanzapine has moved to the UK.   

European Patent (UK) No. 0,454,436 expires in 2010, but if the patent is valid, supplementary protection certificate No. SPC/GB96/058 will extend legal protection for olanzapine until 26 September 2011.  Last month, the UK Patents Court heard the evidence and arguments in the case brought by Dr Reddy’s for the revocation of European Patent (UK) No. 0,454,436 B1 and the SPC.   Actavis participated passively in the UK case in support of Dr Reddy’s, but Teva were perhaps notable by their absence.  

A number of the arguments that were traversed previously in the US case were rehearsed again before Mr Justice Floyd at the hearing in July.  Judgment is not expected for a number of weeks, but because olanzapine is a blockbuster, the trial has already attracted a good deal of interest from the generic sector, to the extent that the court was forced to open up an additional public gallery during the trial.  This article is intended to provide an indication of some of the likely points of difference between the US case and the UK case, in anticipation of the judgment in a few weeks time. 

A brief history of the development of the tricyclic anti-psychotics goes some way to explaining why a number of the prior art citations in the olanzapine case were quite old, relative to the priority date of the patent (25 April 1990).  Following the discovery in the 1950s of the first drugs that were effective in treating schizophrenia, it became apparent that a number of these compounds gave rise to adverse side effects, such as tardive dyskinesias (involuntary, repetitive movements) in patients.  Throughout the 1960s and 1970s, many three ring (tricyclic) compounds were made and tested for anti-psychotic activity in the hope of obtaining a molecule that would not have these extrapyramidal side effects.  

In the US case (and initially in the UK case), Dr Reddy’s relied upon the disclosure of US patent no. 4,115,574, that issued in 1978.  The ‘574 patent described a (then) new family of tricyclic compounds, the novel element being a thiophene, 5-membered sulphur-containing ring instead of a 6-membered benzene ring.  These compounds were said to be useful in the treatment of psychotic conditions such as schizophrenia.  Judge Richard L. Young recorded in his decision (at US District Court level) that although olanzapine was “one of the millions of compounds within the scope of claim 1 of the ‘574 patent”, the ‘574 patent expressed a preference for halogen-containing compounds and olanzapine does not have a halogen substituent.  

In the UK case, Dr Reddy’s placed more reliance on the disclosure in the provisional GB patent specification for the same class of compounds (published on 22 November 1978).   Olanzapine was said to be within a preferred group of compounds defined in this document.  In the European Patent Office, it is possible in certain circumstances to obtain a granted patent to a chemical compound that is selected from a general class that is disclosed in the prior art.  However, Dr Reddy’s argued that under UK law, Lilly had not made a valid selection of olanzapine from the general class of thienobenzodiazepines described in the provisional GB patent specification, in order to obtain their later patent to olanzapine.   Dr Reddy’s argued that the later patent is therefore invalid for lack of novelty and/or obviousness.  Lilly on the other hand argued that olanzapine has surprising advantages that were not disclosed in the GB patent specification and that were not predictable from the prior art.  Accordingly, the later patent discloses an inventive selection of olanzapine from the broader class of compounds. 

It is perhaps on the “selection patent” issue that Dr Reddy’s UK case differs most from the previous US case.  Dr Reddy’s argued that the criteria for a valid selection patent (as set out in an old case from the 1930s) were not met in relation to European Patent (UK) No. 0,454,436 B1.  There have only been a few cases on selection patents in the UK, but it seems that this line of attack on the validity of the olanzapine patent may be the most promising for Dr Reddy’s, who argued that the Judge is bound by precedent to follow the existing UK jurisprudence - however old it may be - and despite the fact that the approach adopted in the UK caselaw appears to be somewhat at variance with the practice of the EPO.    

In both the US and the UK case, Dr Reddy’s relied upon two articles published in J. Med. Chem. in 1980 and 1989 that were authored by a number of chemists who were then based at Lilly’s research centre located at Windlesham in Surrey.  These articles were referred to as “Chakrabarti 1980” and “Chakrabarti 1989”.  They provided synthesis details and structure-activity relationships for series of tricyclic compounds as potential anti-psychotics.  Dr Reddy’s primary case on obviousness seems to have been based on Chakrabarti 1980.  They relied on a number of pointers within this article which (they argued) would lead the skilled person to synthesise olanzapine.  This attack was rejected in the US case, although in part because Lilly successfully relied upon a number of secondary indicia of non-obviousness, such as the commercial success of olanzapine.   In the UK, the courts afford significantly less weight to secondary indications of non-obviousness such as commercial success; the UK courts focus much more upon the primary evidence of the expert witnesses.   The experts retained by both Dr Reddy’s and Lilly in the UK case were subjected to extensive cross-examination by the other party.  Points appear to have been scored by both sides and it remains to be seen how the Judge will assess this evidence in his judgment. 

Finally, Dr Reddy’s prior art “joker in the pack” in the UK revocation case is a short article published in Pharmazie in 1983, known in both the US and the UK litigation by the name of one of its authors, “Schauzu”.   A person familiar with the chemical structure of olanzapine might be inclined to do a “double take” when first reading this article.  It seems to disclose olanzapine, with all of the corresponding substituents in the correct position on the tricyclic chemical structure.  However, on closer examination of the structural formula, a carbon atom has been substituted for a nitrogen.   The title of this article nevertheless suggests that this may have been a mistake.  The interpretation of the disclosure of this prior art document was a matter for the respective experts to address in their evidence, although it is worth noting that the legal treatment of Schauzu as an anticipatory disclosure of olanzapine gives rise to the possibility of a different outcome in the UK case compared to the earlier US case, where this line of attack appears to have been put differently to the Judge (and was rejected). 

As is usual in patent cases concerned with blockbuster pharmaceuticals, many other issues were debated by the parties in court.  A full report on the UK judgment will follow on GenericsWeb in due course.