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

A broad prior disclosure of a chemical class does not anticipate a later claim to one member unless it gives an individualised description of that member. Performing the general disclosure must necessarily infringe the later claim.

A selected compound is not obvious merely because it lies within a previously disclosed class. The decisive question is whether the selection is arbitrary or instead makes a credible technical contribution. Under the Patents Act 1977, the former domestic rules for selection patents are not a separate test. The court should adopt the established European Patent Office approach to novelty and inventive step.

Dr Reddy's Laboratories (UK) Ltd appealed against Floyd J’s dismissal of its application to revoke Eli Lilly’s European patent (UK) No 0 454 436 for olanzapine: [2008] EWHC 2345.

The appellant alleged that a 1978 Lilly specification covering an immense class of thienobenzodiazepines anticipated or rendered obvious the claimed compound. It also alleged obviousness over a 1980 article reporting related compounds and sought, contingently, to rely on insufficiency. The central issues were the correct approach to a claimed compound selected from an earlier broad chemical class, and whether the judge had erred in rejecting the asserted routes to olanzapine.

1. Appeal dismissed. Floyd J was entitled to hold the patent valid. Jacob LJ and the Master of the Rolls gave substantially concurring reasons, with Richards LJ agreeing with both judgments.

2. The 1978 specification did not anticipate olanzapine. Although it contained a formula broad enough to include the compound, it neither named nor individually described it. A general disclosure can be performed in numerous ways that do not produce the claimed compound. It therefore does not necessarily result in infringement of the later claim.

3. The Court rejected the contention that the pre-1977 domestic rules for “selection patents” supplied a distinct and continuing test. The governing questions under the Patents Act 1977 are novelty and inventive step, construed consistently with the European Patent Convention. The established European Patent Office approach was to be followed: a selection from a wider class is obvious if it is arbitrary, but may be patentable where it discloses a real and credible technical advance.

4. Olanzapine was not an arbitrary selection from the 1978 disclosure. The patent identified its beneficial therapeutic and side-effect profile by experimental and early clinical material. The earlier specification made only unsupported, highly general assertions about a vast and scientifically unpredictable class. The patent’s comparison with closely related compounds also supported the conclusion that it made a technical contribution.

5. The challenge based on the 1980 article also failed. The first-instance judge was entitled to prefer evidence that the skilled team would pursue promising compounds already identified, rather than undertake the proposed systematic structure-activity exercise. The appellant had not shown any misunderstanding of evidence, failure to consider relevant material, or conclusion outside the range open to the judge.

6. The contingent insufficiency argument and Lilly’s respondent’s-notice point did not require determination.

• Court of Appeal (Civil Division): dismissed Dr Reddy’s appeal and upheld the validity of Eli Lilly’s patent: [2009] EWCA Civ 1362.
• High Court, Chancery Division, Patents Court: Floyd J dismissed the application for revocation: [2008] EWHC 2345.