Current Intelligence - Lyrica decision upheld by Court of Appeal

Lyrica decision upheld by Court of Appeal

Christina Thompson and Dr Janet Strath 

Journal of Intellectual Property Law & Practice (2017) 12 (3): 158-160

The Court of Appeal dismissed Warner-Lambert’s appeal in the Lyrica/pregabalin matter, upholding Mr Justice Arnold’s finding that claims 1, 3, 4, 6, 13 and 14 of the patent at issue were insufficient and therefore invalid. The court also confirmed that Warner-Lambert’s amendment application after judgment was an abuse of process. The decision (i) provides further guidance on the infringement of Swiss-form claims, (ii) cements the role of ‘plausibility’ insufficiency in English patent revocation actions, highlighting that post-publication evidence cannot be relied upon unless the plausibility threshold has been passed and (iii) shows the importance of bringing a timely conditional application to amend patent claims. 

Legal context and facts

Warner-Lambert’s parent company, Pfizer Ltd, markets the drug pregabalin (trade name ‘Lyrica’) as a prescription-only treatment for neuropathic pain, epilepsy and generalized anxiety disorder (‘GAD’). Patent protection for pregabalin expired in 2013; however, Warner-Lambert still had a second medical use patent, with claims 1 and 3 in the Swiss form relating to the use for treatment of pain:
  1. Use of [pregabalin] or a pharmaceutically acceptable salt thereof for the preparation of a pharmaceutical composition for treating pain. 
  2. Use according to Claim 1 wherein the pain is neuropathic pain.
Mylan and Actavis commenced separate proceedings seeking revocation of Warner-Lambert’s patent, challenging its validity on the grounds of lack of inventive step and insufficiency. Actavis wished to enter the market for pregabalin and had obtained a marketing authorization for a generic version of the drug, limited to epilepsy and GAD, ie the non-patented indications (so-called ‘skinny label’). Warner-Lambert commenced patent infringement proceedings against Actavis arguing that Actavis’ cheaper generic drug (‘Lecaent’) was still likely to be dispensed for the treatment of pain, and that Actavis was liable for such sales.

On 28 May 2015, Floyd LJ upheld Arnold J’s refusal to grant interim relief to Warner-Lambert, but allowed Warner-Lambert’s appeal against the striking out of the indirect infringement claim under s 60(2) of the Patents Act 1977 (Warner-Lambert Company LLC v Actavis Group Ptc EHF [2015] EWCA Civ 556 – ‘Warner-Lambert CoA’). The judgment was particularly concerned with the correct interpretation of the word ‘for’ in Swiss-type claims, ie claims directed towards the use of a known therapeutic substance or chemical compound in the production of a medicine ‘for’ the treatment of a specific disease.

At first instance (Warner-Lambert Company LLC v Actavis Group Ptc EHF (Rev 1) [2015] EWHC 72 (Pat)), Arnold J had decided that ‘for’ meant ‘suitable and intended for’, so that the manufacturer infringes a second medical use patent only if the patentee can prove that it had been the manufacturer’s ‘subjective intent’ (in other words, his ‘wish or desire’) that the medicine would be used for the patented indication. Since Warner-Lambert had not alleged subjective intent on the part of Actavis, Arnold J had held that there was no serious question of direct infringement to be tried under s 60(1)(c). He had also dismissed the indirect infringement allegation under s 60(2), on the basis that infringement was only possible if there could be a person further down the supply chain who could do an infringing act (pharmacists do not use Lecaent to manufacture a pharmaceutical composition, as it has already been manufactured before they receive it). However, on appeal, Floyd LJ proposed a wider test, namely that it would be sufficient to prove that the manufacturer knew (including constructive knowledge) or could have reasonably foreseen the ultimate intentional use of the product for the patented indication. Adopting this interpretation, Floyd LJ found it ‘plain that Warner-Lambert has an arguable case of infringement’ under s 60(1)(c).

The actions came to trial in June and July 2015, again before Arnold J, who ruled that claim 3 encompassed both central and peripheral types of neuropathic pain, rejecting Warner-Lambert’s narrow construction of ‘neuropathic pain’ as limited to peripheral neuropathic pain (Generics (UK) Ltd (t/a Mylan) v Warner-Lambert Company LLC [2015] EWHC 2548 (Pat) – ‘the main judgment’). Accordingly, Arnold J concluded that a number of Warner-Lambert’s claims were invalid for insufficiency, including claims 1 and 3.

Mylan and Actavis, as well as Warner-Lambert, were given permission to appeal. Warner-Lambert subsequently made a conditional application to amend claim 3 to limit its scope to the treatment of ‘peripheral’ neuropathic pain ‘caused by injury or infection of peripheral sensory nerves’, which Arnold J had found plausible. In Generics (UK) Ltd (t/a Mylan) v Warner-Lambert Company LLC [2015] EWHC 3370 (Pat) (‘the abuse judgment’), Arnold J struck out Warner-Lambert’s post-trial attempt to amend claim 3—inspired by the judge’s favourable view of the validity of that invention—because it amounted to an abuse of process. The judge held that ‘if the amendment application was allowed to proceed, it could not be determined fairly without a second trial on validity’; thus, bringing the application after the patent had been found invalid amounted to unjust harassment of Mylan and Actavis.


On appeal, Warner-Lambert argued that the claims rejected as insufficient were sustainable across their breadth, even on the construction arrived at by Arnold J. Mylan and Actavis submitted that Arnold J had been correct in his construction and rejection of claims 1, 3, 4, 6, 13 and 14 as implausible, but had erred in finding that the claims 10, 11 and 12—which were limited to specific types of peripheral neuropathic pain—were plausible. Warner-Lambert also contended that the judge had wrongly deemed its amendment application to be an abuse of process, as the patent’s validity had already been established with respect to treating peripheral neuropathic pain and a second trial on that issue would not be required. The court also had to define the proper test of infringement to be applied to Swiss-form, second medical use claims.


Floyd LJ took the opportunity to provide further obiter guidance on the approach to Swiss-form, second medical use claims in England, in circumstances where a ‘skinny label’ is used. In doing so, he referred back to his comments in Warner-Lambert CoA and to Arnold J’s ‘forceful’ and ‘profound’ reservations. The comments were obiter as the court held that the allegedly infringing claims were invalid.

Acknowledging that there was no consistent approach to the treatment of Swiss-form claims in Europe, Floyd LJ recited his previous criticism of the position adopted in Germany (ie where the packaging of the product must indicate that the product is to be used for a patented indication and the advertising claims are ignored; see Warner-Lambert CoA at [81]) for being ‘too narrow’. He favourably referred to decisions in Spain and France, in which the courts looked at a plurality of factors, including the indications authorized and the actions of the generic manufacturer to prevent prescription for the patented indications (see Wyeth v Arafarma and Qualtec, Case 539/07, Madrid Court of Appeal; and Warner-Lambert v Sandoz, Case 15/58725, Tribunal de grande instance de Paris).

Floyd LJ then set out the test to apply when considering infringement by a manufacturer of a Swiss-form claim. The test is twofold, since the court should determine: 
  1. whether the generic manufacturer knew or could foresee that at least some of the prescriptions written generically for the drug to treat the patented indication would in fact be filled by the pharmacist with the generic product, 
  2. if so, whether the generic manufacturer has taken all reasonable steps in its power to prevent its generic product being used to treat the patented indication (see [217]). 
In relation to indirect infringement, Floyd J stated that there was ‘a danger in translating section 60(2) into a requirement for a “downstream act of manufacture”’, adding that the process of preparing the composition could continue through any packaging step performed by the manufacturer, including the labelling step performed by the pharmacist. By conceiving the act of manufacture as an ongoing process, terminating when the pharmacist dispenses the product to the patient, the generic manufacturer would be liable for infringement because it supplied the means of putting the invention into effect. However, adopting this view would also mean that the pharmacist would directly infringe the patent when applying the final label and dispensing the product, as Floyd LJ acknowledged when he agreed with the outcome in the recent Danish case of Warner-Lambert Company LLC and another v Krka d.d. and another, a decision of the Maritime and Commercial High Court dated 25 June 2015, in which pharmacists were found to have directly infringed a patent when they applied the label to the product and dispensed it, this being the final manufacturing step. This suggests that all parties in the supply chain of a generic product may be liable for both direct and indirect infringement.


Mylan and Actavis contended that the animal model results reported in the patent only provided support for the treatment of those types of pain for which the animal models were a plausible model.

In relation to claim 3, it was held that ‘neuropathic pain’ included peripheral neuropathic pain as well as central neuropathic pain, but that the animal tests only supported the former. There was no other data in the patent (or the common general knowledge) from which to predict use to treat central neuropathic pain. Claim 3 was thus insufficient and invalid, notwithstanding the fact that pregabalin had subsequently been authorized for central neuropathic pain. The court confirmed that later filed evidence could not render the disclosure in the patent plausible, rescuing implausible, speculative claims.

On appeal, the generic manufacturers had sought to reverse Arnold J’s finding that it was plausible that pregabalin was effective to treat peripheral neuropathic pain. They had submitted that the patent’s disclosure of the treatment of neuropathic pain was not plausible. This challenge was rejected by the court, with Floyd LJ reiterating that the test of plausibility has a very low threshold. In this case, whilst the test results in the patent did not enable ‘a firm prediction of efficacy to be made’, they were enough to suggest to the skilled person that pregabalin might be effective for peripheral neuropathic pain. Therefore, the disclosed technical effect was plausible.

Floyd LJ also found support for plausibility in the evidence that the skilled team would be encouraged by the disclosures in the specification to test pregabalin for the treatment of peripheral neuropathic pain. However, he rejected the suggestion that the test for plausibility should be aligned to that often used for obviousness (ie if the skilled team would be encouraged to try it with an expectation of success). That would be setting the bar too high.

Abuse of process

Arnold J had refused Warner-Lambert’s conditional application to amend claim 3 to narrow its scope and overcome the insufficiency finding, on the grounds that this constituted an abuse of process. The Court of Appeal agreed. The time for Warner-Lambert to validly make such conditional application was prior to the trial of the invalidity issues.

Practical significance

Floyd LJ upheld Arnold J’s conclusions on the insufficiency, and therefore invalidity, of all the patent claims at issue, also confirming that Warner-Lambert’s late application to amend the claims amounted to an abuse of process. Accordingly, the appeals of both parties were dismissed.

The decision clarifies that if a patentee wishes to preserve a fall-back position, particularly when faced with insufficiency arguments alleging excessive claim breadth, it must make a conditional application to amend prior to the trial, or at the very latest on day one of the hearing. Making the application after the parties have argued the invalidity issues and judgment has been handed down is too late. Patentees who are involved in revocation actions should consider throughout the life of the matter how a conditional application to amend the patent fits with the overall litigation strategy, carefully deciding when the application should be made in light of Floyd LJ’s comments in this decision.

© The Author(s) 2017. Published by Oxford University Press. All rights reserved.

No comments:

Post a Comment