Dr Reddy's Laboratories (UK) Ltd v Eli Lilly & Company Ltd

A generic chemical formula covering many compounds will not normally anticipate a later claim to an individual compound where arriving at that compound requires choices from lists of some length. The compound must be disclosed in an individualised form or as the inevitable result of a prescribed method.

Novelty and inventive step remain distinct. A selection need not satisfy the traditional selection-patent requirements to be inventive, although any advantage may be relevant to obviousness. Obviousness is assessed by the structured inquiry into the skilled person, common general knowledge, differences from the prior art and whether those differences required invention.

The claimant sought revocation of the defendant’s patent for olanzapine, an antipsychotic compound selected from a broad class disclosed in earlier patent documents and related scientific papers. The principal attacks alleged lack of novelty, obviousness and insufficiency.

The court considered the disclosure of a broad Markush formula, the significance of individual compounds disclosed in Chakrabarti 1980 and Chakrabarti 1989, and an allegedly inconsistent disclosure in Schauzu. It also considered whether olanzapine was obvious as a result of structure-activity research and whether its commercial success assisted the patentee.

1. Novelty. Under Patents Act 1977, section 2, anticipation requires both disclosure and enablement. A prior disclosure must contain subject matter which, if performed, would necessarily infringe the later patent. A generic disclosure does not normally disclose every individual member of the class.
2. A general chemical formula with multiple substituents selected from lists of some length does not normally provide an individualised disclosure of a particular compound. The skilled person’s ability to enumerate the theoretical compounds is artificial and does not amount to a clear and unambiguous technical teaching. The 235 Provisional therefore did not anticipate olanzapine. Chakrabarti 1980 disclosed ethyl olanzapine and a general observation about alkyl substitution, but did not disclose olanzapine itself. Schauzu was neither clear nor unambiguous because of inconsistencies between its title, formula and related material.
3. Selection patents. The traditional principles in I.G. Farbenindustrie’s Patent and E.I. Du Pont de Nemours & Co.’s (Witsiepe’s) Application did not govern the novelty analysis. Provided the compound was novel on conventional grounds, obviousness was assessed according to ordinary principles. An advantage could be relevant to obviousness but was not an essential prerequisite for inventive step.
4. Inventive step. Applying the approach restated in Pozzoli v BDMO SA and the fact-sensitive guidance in H. Lundbeck A/S v Generics (UK) Limited, olanzapine was not obvious over the 235 Provisional, Chakrabarti 1980, Chakrabarti 1989 or Schauzu. The skilled team would not have been led without hindsight to the necessary combination of substitutions. In particular, the evidence and prior art favoured compounds with a halogen substituent and gave no reliable reason to pursue the proposed research programmes or to change ethyl olanzapine to olanzapine.
5. Commercial success did not assist the patentee because the broad earlier patent would have prevented third-party manufacture and sale. The insufficiency attacks did not arise because olanzapine was held novel and non-obvious without reliance on selection-patent principles.
6. The attacks on validity failed. The court was to hear counsel on the form of order.

First-instance decision. The judgment records no prior appellate decision in the same proceedings.