Generics (UK) Limited v AstraZeneca AB

For a chemical compound to make a technical contribution, the patent must make it plausible, in light of the common general knowledge, that the compound has the asserted useful property. A bare assertion of activity is insufficient. An assay described in the patent does not establish plausibility unless the disclosure states, or otherwise makes clear, that the claimed compound was tested and achieved relevant results. In vitro activity must support a reasonable prospect of useful in vivo efficacy through a known or demonstrated relationship with the disease mechanism. A compound selected from a previously disclosed genus is an arbitrary selection where it has no different or improved property, and the patent does not make such a property plausible. Later success cannot cure an inadequate disclosure.

Three generic pharmaceutical companies sought declarations that two supplementary protection certificates relating to dapagliflozin were invalid, and that the underlying patent was invalid. The live grounds concerned inventive step and insufficiency.

The patent claimed dapagliflozin as a compound and for use in treating or delaying diabetes. The prior disclosure, WO 01/27128 A1, described a very large genus of C-aryl glucosides said to inhibit SGLT2 and to be useful in treating diabetes, but did not individually disclose dapagliflozin or report assay results. The central issues were whether the patent made the asserted activity and therapeutic utility plausible, and whether dapagliflozin made a technical contribution over the genus disclosed in WO 01/27128 A1.

1. Applicable approach. The court applied the majority approach in Warner-Lambert. Plausibility is not a free-standing validity requirement, but a standard requiring the specification to disclose reasonable scientific grounds for expecting the asserted utility to work. A bare assertion or mere possibility is insufficient. Experimental data are not essential, but the patent must disclose either relevant results or adequate a priori reasoning.
2. Assay disclosure. The patent described an SGLT2 assay but stated only that the activity of the compounds of the invention may be determined using it. It did not disclose that dapagliflozin had been tested, identify the inhibitor used, or state any result or potency. The court rejected the argument that the skilled person could assume that a pharmaceutical company would have obtained supporting data.
3. Therapeutic efficacy. Similarities between dapagliflozin and phlorizin did not permit a reasonable prediction of SGLT2 inhibitory activity, given the material structural differences and the known sensitivity of activity to small structural changes. The Tanabe Seiyaku material disclosed strict structural requirements for activity and gave the skilled team reason to doubt that dapagliflozin would be active. The patent therefore did not make plausible either a useful in vivo reduction in blood or plasma glucose or efficacy in treating diabetes.
4. Technical effect and selectivity. The relevant technical effect for the use claim was sufficient efficacy to treat or delay diabetes. For the compound claim, the court also considered the proposed utility of reducing blood or plasma glucose sufficiently for use as an experimental tool. Selectivity for SGLT2 over SGLT1 was not an independent requirement; the focus was efficacy.
5. Arbitrary selection. The absence of classical obviousness did not dispose of the arbitrary-selection case. The patent had to disclose a real technical advance over WO 01/27128 A1. AstraZeneca did not contend that dapagliflozin had any different or improved property from the compounds in that disclosure. It was therefore an arbitrary selection and made no technical contribution over the prior art.
6. Disposition. The patent was invalid. The supplementary protection certificates, which depended on it, were consequently invalid.