The "patent bargain" is the foundation of patent law – in exchange for a 20 year monopoly, the inventor discloses their invention to the world. Inherent in this is a quid pro quo, that the disclosure must be commensurate with the reward. Under UK/EU patent law, this is tested under the related concepts of sufficiency, enablement and plausibility. But what does this mean in practice? The UK Court of Appeal recently opined on the topic in the case of Generics UK and others v AstraZeneca.
Farxiga® (dapagliflozin), which is indicated for the treatment of diabetes, has been a wildly successful drug for AstraZeneca, earning them $6 billion in sales in 2023. It is covered by EP1506211, which claimed the compound and its use as an anti-diabetic. Aside from disclosing its synthesis and an assay to determine activity, the patent did not provide any in vivo or pre-clinical data. A very similar compound was disclosed in an earlier ('prior art') patent document alongside millions of other compounds.
Plausibility
While sufficiency is a statutorily prescribed requirement, plausibility developed in case law. In the UK, it requires that the application (when read together with the common general knowledge, CGK) must positively make it plausible that the invention will achieve the claimed technical effect. This has since been called "ab initio plausibility". It can be distinguished from ab initio implausibility – which reverses the test, asking that the technical effect is not implausible in light of the application and CGK (aka, there is no legitimate reason to doubt it).
In this case, while the description drew a link between inhibition of a protein called SGLT2 and reduction in glucose levels, and that dapagliflozin was an SGLT2 inhibitor, it did not provide data that dapagliflozin had sufficient activity as an SGLT2 inhibitor. Accordingly, it was not plausible from the patent itself that the compound could be used to treat diabetes.
On appeal, Arnold LJ confirmed that the test in the UK is ab initio plausibility and not implausibility. The test could not be met here, as there was a lack of positive experimental data or theoretical reasoning to support the effect of dapagliflozin. (Although his Lordship found against AstraZeneca even if the alternative standard was adopted). Accordingly, the patent was found to be insufficient.
Both the Court of Appeal and first instance judge said it's tempting to get swayed by knowledge that dapagliflozin later turned out to be successful. But such post-priority date knowledge cannot be used to shape the interpretation of what is actually disclosed in the patent. To do otherwise would undermine the patent bargain.
In reaching this conclusion, Arnold LJ was not persuaded that the decision of the Enlarged Board of Appeal of the European Patent Office (in G 2/21) justified departure from previous case law.
Arbitrary selections
This case also considered arbitrary selections. Making a purely arbitrary selection from the prior art is not inventive, just as the mere identification of a new chemical compound is not an invention. The rationale is that such choices are not a real technical advance, unless it is demonstrated that such selections have some different or improved property compared to the earlier ones described (and this is made plausible in the document).
This test is higher than plausibility, because it requires the selection to be a technical advance, and so AstraZeneca could not succeed here given the court's findings above. The disclosure in the patent was the same as that in the prior art and so there could be no advance. Accordingly, the patent was also bad for obviousness.
Outcome
This case is a harsh reminder of what happens when a patent is filed without any data supporting the therapeutic use (the key is here is 'any', as the threshold for plausibility should not be an onerous one). Patents should always be filed with some kind of technical rationale, and this could possibly include in silico data if it is convincing enough (or eg reasoning by reference to the prior art, although care should be taken not to make it seem that the invention was obvious).
AstraZeneca's permission to appeal to the Supreme Court was refused as was their further application to the court.
It will be interesting to see how the Enlarged Board's decision in G 2/21 is applied by the EPO going forward, and in particular if and how it will be adopted by the Unified Patent Court.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
