Finding new uses for existing medicines is an important area of pharmaceutical research which boasts many success stories. The Supreme Court’s recent ruling in the Lyrica case relating to the use of pregabalin for pain is, however, disappointing news for patent owners in this area. The ruling potentially makes it both more difficult to obtain patents on new therapeutic uses and also more difficult to enforce them.
Warner Lambert (now a subsidiary of Pfizer) was the proprietor of a Swiss-form, second medical use patent for pregabalin (brand name: Lyrica). The drug was originally used as an anti-convulsant, but it was later found to be effective to treat pain. Claim 1 claimed pregabalin for the treatment of pain and claim 3 claimed the treatment of neuropathic pain specifically. Generics UK (trading as Mylan) and Actavis manufactured a generic form of pregabalin. In 2014, they each applied for Warner Lambert’s patent to be revoked in part, arguing that its claims were implausible with the result that the patent failed for insufficiency. Warner Lambert countered by bringing infringement proceedings against Actavis, and the dispute eventually made its way to the Supreme Court. On 14 November 2018, the Supreme Court confirmed the lower courts’ decision that the patent was invalid (though the analysis was slightly different) and also held that, even if the relevant claims had been valid they would not have been infringed.
The case raises three key practical issues:
Plausibility – how early can I file?
Plausibility (an aspect of insufficiency) is the idea that it must be plausible or credible that the patented invention will work across the full breadth of the patent claim. The plausibility rule is there to prevent speculative patenting and has particular relevance to second medical use patents where the invention consists simply of the idea that a particular drug will be effective to treat a new indication. At the time of application for the patent this efficacy will not yet have been proved through clinical trials: the plausibility rule tries to ensure that such patents are not granted merely on the basis of a guess, which could enable the applicant to reserve a field of research without having made a genuine technical contribution. In practice it may amount to the question of how much evidence needs to be included in the patent specification – or at how early a stage in the research process is it safe to file for a patent.
The Supreme Court’s ruling introduces a higher threshold for plausibility, which is likely to increase the difficulty of patenting second medical uses in the future and may also result in challenges to existing patents. Giving the lead judgment, Lord Sumption emphasises that, although there does not need to be definite proof of efficacy for the new indication, there should be disclosure of reasonable scientific grounds for expecting that the drug might well work for the new indication. He comments that these grounds are what mark the difference between speculation and a contribution to the art. However, they need not necessarily consist in experimental data; it may be theoretical analysis. Lord Sumption also indicates that these grounds should emerge from the disclosure in the patent specification itself rather than placing too much reliance on the common general knowledge. We have yet to see how the ruling will be interpreted in practice and how much difference it will really make, but the message to applicants is clear – try to ensure that a clear explanation of the grounds are included on the face of the patent.
Does a skinny label avoid infringement?
The generic producers must be free to sell their product for the non-patented uses – in the case of Lyrica this was for general anxiety disorder and epilepsy. They will do so under a "skinny label", whereby the patient information leaflet included with the medicine will state that the product is indicated for the non-patented uses. The question in the patentee’s mind is whether this will be enough in practice to prevent the generic being sold for the patented use, especially given health service policies encouraging the prescription of generic products generally. The test for infringement adopted by the Court of Appeal had involved an element of foreseeability so that (broadly) the generic manufacturer might be liable for infringement if it was obvious that the product was being sold for the patented use but did nothing to stop this. This approach was, however, roundly rejected by all the Lords as being impractical and moving too far in favour of the patentee. The Lords were not, however, able to agree on a test for infringement. Two Lords thought it should be a question of whether the generic manufacturer subjectively intended to target the patent-protected market; two thought that the sole criterion for infringement should be whether the product as it emerges from the manufacturing process, including any labelling or accompanying leaflet, is presented as suitable for the patent-protected uses – they called this the “outward presentation” test. Finally one of the Lords adopted this outward presentation test with the proviso that in rare cases the context may make it obvious that this is not to be taken at face value. The Lords accepted that none of these tests really gave watertight protection to the patentee. However, there were other issues at stake including the public interest in the free exploitation of the non-patented product and legal certainty for third parties.
As a result it is not clear how courts will approach infringement in the future. Although a foreseeability test seems effectively to have been ruled out, it is unclear whether a skinny label alone will necessarily avoid infringement. In addition to the skinny label, generic manufacturers will wish to avoid any suggestion in their marketing or otherwise that they are targeting the patented market.
I want to narrow my patent to avoid infringement – how late can I leave it?
The High Court (Arnold J) had held that the patent was plausible in relation to peripheral neuropathic pain but not in relation to central neuropathic pain. Warner Lambert wished to amend the claim to peripheral pain only in an attempt to save the patent. However, it waited until the end of the trial before applying to limit the claim at which point an application to amend would have required a further trial. The High Court ruled that the application was too late and constituted an abuse of process because it could and should have been raised either before or even during the trial. This issue ultimately became irrelevant, because the Supreme Court, applying the higher plausibility threshold, held that the patent was not plausible for any neuropathic pain. However, as they had heard argument on the point the Supreme Court considered it and upheld Arnold J’s decision. The message to patentees is, of course, to play it safe where possible and apply at an early stage; the application can be made conditional on the claim otherwise being held invalid.
A clear win for the generics and the NHS?
Clearly, the end result of the four years of proceedings in this dispute has brought a win for the generics: the judgment will be regarded by some as a welcome step in guarding against speculative patenting; in addition, a foreseeability test for infringement, which might have involved a burdensome obligation on the generics to take steps to prevent use of their products for the patented use, has almost certainly been avoided. The NHS also gains from cheaper generic prices and may now have a claim against the patentee for inflated prices paid for the drug during the period of the litigation. Only time will tell whether there might be a “chilling” effect on the willingness of pharmaceutical companies to invest in research on second medical uses or whether the Court has got the balance right.
Case: Warner Lambert Company LLC v Generics (UK) and Actavis Group PTC EHF, Supreme Court 14 November 2018.