Warner-Lambert Appeal: Swiss form claims & skinny labels
The long awaited Supreme Court decision in Warner-Lambert v Actavis was handed down on 14 November 2018.
Full decisionLink to full decision
A reminder of the facts
Pregabalin is sold by Pfizer (ex Warner-Lambert) as Lyrica® for three labelled indications: epilepsy, generalised anxiety disorder and neuropathic pain. The basic patent, which disclosed the epilepsy and anxiety indications, expired in 2013.
The patent in suit, EP 0 934 061 (as centrally limited at the European Patent Office - EPO), was directed to pregabalin (marketed as Lyrica®) for the treatment of pain. Two claims were alleged to be infringed, claims 1 and 3. Claim 1 was a Swiss form claim to the "use of [pregabalin] for the preparation of a pharmaceutical composition for treating pain". Dependent claim 3 was limited to treating neuropathic pain.
Following expiry of the basic patent, generic manufacturers prepared to launch generic pregabalin with a “skinny label” omitting the neuropathic pain indication. Such an approach is specifically permitted under EU pharmaceutical regulators. Some also took further steps to inform pharmacists and health professionals that the drug was not to be prescribed for the treatment of pain.
The lower courts' decisions
Following interim proceedings, the full trial on the merits was heard before the Patents Court in 2015. The generic manufacturers sought revocation of the patent on the grounds of lack of inventive step and insufficient disclosure. Warner-Lambert brought a counter-claim against Actavis for infringement.
At first instance, the judge (Mr Justice Arnold) rejected the arguments on inventive step, but ruled claims 1 and 3 of the patent invalid for insufficiency. Specifically, the judge found that the claims were sufficiently disclosed in respect of inflammatory pain and peripheral neuropathic pain, but not central neuropathic pain. He also ruled that, even if the claims had been found valid, they would not have been infringed: either directly (as there was no act of manufacture in the UK), or indirectly (as there was no “supply of means relating to the invention” by an upstream manufacturer). He also rejected a request by Warner-Lambert to amend the claim post-trial to try and address the validity issue, considering it an abuse of process.
The Court of Appeal upheld the Patents Court’s findings both on validity and abuse of process. The court did not decide on the infringement issue, but indicated in obiter comments that it differed from the Patent’s Court’s test on infringement. As regards direct infringement, the court considered that if the manufacturer knows or can reasonably foresee that the pharmaceutical will be used for the patented use, there is prima facie infringement. The court considered this could be negated where “the manufacturer takes all reasonable steps within his power” to prevent that use – while not stating explicitly what those steps should be, the message was that reliance on a “skinny label” may not be enough. As regards indirect infringement, in the court’s view, “preparation” for the purposes of a Swiss-form claim could include a packaging step and/or a labelling step, the latter potentially being carried out by a downstream pharmacist.
The issues before the Supreme Court
In summary, the Supreme Court dismissed Warner-Lambert's appeal that the patent was sufficiently disclosed, and upheld the generics’ appeal that none of the disputed claims were sufficient.
The Supreme Court also unanimously held that, if the claims had been found valid, they would not have been infringed.
The Supreme Court was also unanimous on construction and agreed with the lower courts that the post-trial amendment sought by Warner-Lambert was an abuse of process.
The case details
Construction of the claims, in particular claim 3 (neuropathic pain)
Warner-Lambert had attempted to argue before the Court of Appeal that the term "pain" should be construed by a skilled person as limited to specific types of pain listed in the description. They also argued, based on expert testimony, that the term "neuropathic pain" in claim 3 would be understood as limited to peripheral neuropathic pain. However, the Court of Appeal dismissed this argument and ruled that claim 1 covers all pain and claim 3 covers all neuropathic pain, whether peripheral or central.
Sufficiency of disclosure
The sufficiency requirement in patent law is designed to ensure that the patentee, in exchange for obtaining a monopoly, must provide a full disclosure of their invention so that third parties following the teachings of the patent can carry out the invention without an undue burden – this is sometimes called the “patent bargain”.
Lack of sufficient disclosure is a ground for revocation of the patent both in national revocation proceedings and in opposition proceedings before the EPO. However, the UK courts have generally adopted a higher threshold for sufficiency than the EPO in recent years, and continued this approach in this case.
UK patent case law (and that of the EPO) recognises two lines of reasoning why a patent lacks sufficient disclosure. The first – “classical insufficiency” – is that there is not sufficient information in the patent to enable the skilled person to practice the invention.
The second line of reasoning for insufficiency is where an invention is found not to be enabled across the entire scope of the claim. This is based on the principle that the claim must not exceed the technical contribution to the art. This is often known as “Biogen insufficiency” as it follows the House of Lords’ 1997 decision in Biogen v Medeva.
Although it is possible to provide additional evidence to support sufficiency of disclosure after filing of the patent application, the application as filed must itself include sufficient information to make the claimed therapeutic use plausible across the scope of the claim. The requirement that the invention be “plausible” from the application as filed originated in the case law of the EPO Boards of Appeal in relation to inventive step, but the UK Courts have adopted it as part of the test for sufficiency. However, the intention is the same in both cases: to prevent applicants filing applications having over-broad, speculative claims without scientific support across their scope, and attempting to justify them solely by the use of post-filing data.
In relation to medical use claims, as the invention is the new use of the product, the Supreme Court considered it essential that the application as filed contained some disclosure how or way the known product can be expected to work in the new application – otherwise, it would be possible to patent purely speculative claims to any conceivable use of a drug without having invented anything at all.
A majority of the Supreme Court opined that plausibility would be established by the application as filed containing reasonable scientific grounds that the product would have the claimed therapeutic effect: experimental data are not essential. However, a bare assertion that a product is efficacious for the claimed medical indication, without any credible scientific reasoning why, would not meet the plausibility threshold.
Dissenting, Lord Mance considered such a test imposed too high a standard and considered it necessary for the patentee to disclose reasons for regarding the claimed therapeutic effect as plausible only when the skilled person reading the patent would be sceptical about it in the absence of such disclosure.
The Supreme Court thus dismissed Warner-Lambert's appeal that the claim to the use of pregabalin for the treatment of central neuropathic pain was sufficiently disclosed. The Supreme Court went even further, and upheld Actavis’ and Mylan's appeal that the treatment of any kind of neuropathic pain was also insufficiently disclosed.
The Supreme Court may have established a higher bar to sufficiency of disclosure for pharmaceutical inventions and firmly placed plausibility as a key aspect of sufficiency.
Infringement of Swiss-style second medical use claims
Even though claims 1 and 3 had been found invalid for insufficiency, the Supreme Court nevertheless considered the issue of infringement in obiter remarks. In particular, they considered the proper interpretation of Swiss-form claims.
The Supreme Court recognised the considerable difficulty of applying to Swiss-form patents the statutory provisions on infringement of a patent in the UK, which they were not designed to accommodate. This misgiving has been expressed by UK patents courts in the past, most notably by the House of Lords in Merrell Dow v Norton. In particular, the judges considered the issue of the “mental element” of second medical use patents. The court noted that while section 60(1)(c) of the Patents Act, which relates to direct infringement (by acts relating to a direct product of a process claim) is a “strict liability” provision which has no mental element, section 60(2), which relates to indirect infringement, requires proof of knowledge that the act would infringe.
The Supreme Court considered the infringement questions raised a number of competing objectives:
- To provide reasonable protection to the second medical use patentee, by preventing competitors from marketing the drug for the patented use.
- To allow generic manufacturers to lawfully market the drug for the non-patented uses (thereby avoiding the second medical use patent effectively becoming a second patent for the drug itself), and the public the benefit of lower prices for these uses.
- To provide reasonable legal certainty for all parties involved in the pharmaceutical supply chain (manufacturers, wholesalers, pharmacists and prescribers).
The differing opinions of the Supreme Court judges take these competing objectives into account, but they reached different conclusions as to how.
Lord Sumption and Lord Reed agreed that the intention of the alleged infringer is irrelevant and that the sole criterion of infringement is whether the product as it emerges from the manufacturing process, including its packaging and any labelling or accompanying leaflet, is presented as suitable for the uses which enjoy patent protection.
Lord Mance agreed that the test depends on the objective appearance and characteristics of the product as it is prepared, presented and put on the market, but leaves open the possibility (i) that in rare cases the context may make it obvious that these are not to be taken at face value, and (ii) that there may be circumstances in which the generic manufacturer should positively exclude use for the patent-protected purpose.
Lord Hodge and Lord Briggs preferred the view of Mr Justice Arnold at first instance that the test is whether the alleged infringer subjectively intended to target the patent-protected market.
As regards the case on indirect infringement, the Supreme Court disagreed with the Court of Appeal’s view that the invention in a Swiss form claim, which is a purpose-limited claim to a process of manufacture, could somehow be put into effect by a pharmacist dispensing a product upon prescription. In this regard, they preferred the view of the first instance Patents Court.
Clearly there were split views by Supreme Court judges on this matter and an apparent reluctance to commit themselves in obiter remarks. Most acknowledged this as the difficulty of applying the existing tests for infringement in UK patent law to the use element in Swiss-type claims. Therefore further litigation where the infringement issue is ultimately pursued as far as the Supreme Court may be necessary to resolve the split of views exposed in the Supreme Court by this decision