Actavis UK Ltd v Novartis AG

An invention is obvious where the difference from the prior art is a technically motivated step which the skilled person would expect to work without real or apparent difficulty. Commercial uncertainty about implementing the idea does not confer inventiveness.

The Pozzoli/Windsurfing structure and the EPO problem-and-solution approach are aids to the statutory inquiry, not substitutes for it. The latter is especially useful where the patent's stated problem requires no reformulation. The “could-would” inquiry asks whether the skilled person would arrive at the idea, not whether commercial considerations would lead to its physical implementation.

Novartis appealed against Warren J's decision, [2009] EWHC 41 (Ch), that its patent for a sustained-release formulation of fluvastatin was invalid for obviousness. The patent presented fluvastatin's high solubility as an obstacle to making such a formulation by conventional techniques.

The judge found that the skilled team would consider sustained release, expect to produce a clinically effective formulation using known techniques, and regard improved patient compliance as a benefit. The supposed technical obstacle was therefore illusory. The central issue was whether the invention nevertheless involved an inventive step because the skilled person lacked sufficient reason or expectation to implement the idea.

1. Appeal dismissed. Jacob LJ, with whom Stanley Burnton and Lloyd LJJ agreed, held that claim 1 was obvious and upheld the order invalidating the patent.

2. The ultimate statutory question was whether the invention was obvious to the person skilled in the art. The Pozzoli/Windsurfing structure merely oriented the tribunal. Its fourth stage, asking whether the differences from the prior art would have been obvious without knowledge of the invention, was the decisive statutory stage. Obviousness meant technical, rather than commercial, obviousness.

3. The EPO problem-and-solution approach was a useful but non-exclusive aid. Its weakest feature was the retrospective reformulation of an objective technical problem, which could introduce hindsight or distort inventions involving recognition of a problem. Reformulation was inappropriate where, as here, no unknown prior art had changed the patent's perspective and the patentee's stated problem remained suitable.

4. The “could-would” approach did not require proof that the skilled person would physically develop and commercialise the idea. It required proof that the skilled person would arrive at the invention because the prior art prompted that course. Whether the idea would then be implemented depended on commercial considerations irrelevant to technical obviousness.

5. The patent's asserted obstacle was illusory. The skilled person would not regard fluvastatin's solubility as preventing or materially impeding a sustained-release formulation. The person would consider such a formulation, expect conventional techniques to produce one with some clinical efficacy, and have a technical motivation in improved patient compliance.

6. Under the Pozzoli/Windsurfing approach, the only material difference from the prior art was the idea of making a sustained-release formulation, for which there was motivation and no real or apparent difficulty. The problem-and-solution approach produced the same conclusion: the stated problem was to produce such a formulation, and standard methods would clearly work. The problem could not properly be recast as obtaining improved medical effects which the patent neither identified as its problem nor promised as its solution.

1. Court of Appeal (Civil Division): By [2010] EWCA Civ 82, unanimously dismissed Novartis's appeal and upheld the finding that the patent was invalid for obviousness.
2. High Court, Chancery Division, Patents Court: Warren J held in [2009] EWHC 41 (Ch) that the patent was invalid for obviousness.