AIPPI Rapid Response Report: Debating Lyrica's recurring pain on plausibility, abuse and infringement

Just over a year ago, AIPPI hosted a Rapid Response seminar following Mr Justice Arnold finding that the material claims of Warner-Lambert’s patent for the use of pregabalin in the treatment of pain were invalid and in any event not infringed by Actavis’ skinny label pregabalin product Lecaent (Warner-Lambert v Actavis [2015] EWHC 2548 (Pat)).  At that stage, the panel were also in possession of six other judgments from Arnold J concerning pregabalin and one from the Court of Appeal.   Steven Willis (Bristows) reports:  

"Since that first rapid response seminar, Arnold J has issued a further judgment concerning Warner-Lambert’s right to amend the Patent after the trial and whether that amounted to an abuse of process (Generics (UK) t/a Mylan v Warner-Lambert [2015] EWHC 3370 (Pat)).  In addition, last month, the Court of Appeal upheld Arnold J’s findings on the validity of the Patent and the amendment/abuse of process issues but disagreed with him, obiter, on issues concerning the construction and infringement of Swiss Type Claims (Warner-Lambert Company LLC v Generics (UK) Ltd (t/a Mylan) & Ors [2016] EWCA Civ 1006).  

The same panel of experts  - Claire Baldock (Boult Wade Tenant), Brian Cordery (Bristows) and Stuart Baran (3 New Square) - that presented at the first seminar reconvened last week to revisit the issues and grapple with new ones raised by the two most recent judgments. Claire covered the plausibility/insufficiency aspects of the Court of Appeal judgment, Brian addressed the amendment and related abuse of process issues and Stuart tackled the thorny area of the construction and infringement of Swiss type claims.  

Plausibility/Insufficiency

Claire reminded the audience of the relevant background to the insufficiency issues, some of which are set out below:

• The Patent in suit contains Swiss form claims to pregabalin in the treatment of pain (Claim 1) and in the treatment of neuropathic pain (Claim 3).
• Pain is difficult to define both clinically and in patent terms.  Paragraph [0003] of the Patent states that the invention is directed to chronic pain disorders.  It then provides a list of conditions which the invention is said to include but not be limited to.  Paragraph [0006] of the Patent then refers to the invention as being the use of pregabalin in "the treatment of pain as listed above".  
• The Patent refers to neuropathic pain as being that "which is caused by injury or infection of peripheral sensory nerves”.  The patent does not draw a distinction between peripheral neuropathic pain and central neuropathic pain.  
• The Patent contained data relating to animal models of inflammatory pain.  None of these models were established models of neuropathic pain.  
• Inflammatory pain contains a central sensitisation component which is present in neuropathic pain but not central neuropathic pain.   

Claire went on to summarise some of the parties’ key submissions on plausibility/insufficiency:

• The Defendants argued that Warner-Lambert must rely on the data in the animal models for the purposes of plausibility.
• Warner-Lambert responded that central sensitisation component provided the requisite link between the animal models and the claimed forms of pain.  
• Mylan contended that not all forms of pain which fell within the claims contained a central sensitisation component.  Furthermore, it was CGK that neuropathic pain included both peripheral neuropathic pain and that central neuropathic pain did not have a central sensitisation component.  
• Warner-Lambert argued that the skilled person would in any event have understood claim 3 (i.e. to neuropathic pain) to have been limited to peripheral neuropathic pain.  

Claire considered the Court of Appeal’s construction of the claim and noted the finding that not every use of a term will be understood to be a definition:

although it is often said that a specification can act as its own dictionary, not every use in the specification of a term found in the claim will be understood by the reader to be a definition. On no basis can paragraph [0006] or [0003] or the two paragraphs read together be taken to be a definition of what the patentee means by the term "pain". They are statements exemplifying the broadest statement of invention, namely that pregabalin is suitable for the treatment of pain.

This contributed to the Court's finding that the skilled person would have viewed the claims as being broad claims to efficacy concerning the treatment of all kinds of pain.  Although the skilled person would adopt a narrow meaning of a particular claim if there were CGK reasons as to why the wider meaning would cover implausible embodiments, this was not such a case.  The Patent taught that pregabalin is likely to treat inflammatory pain.  Inflammatory pain has a central sensitisation component.  Therefore, the skilled person would know that pregabalin might work for peripheral neuropathic pain but not for central neuropathic pain because of the lack of a central sensitisation component.  The claim was therefore not plausible across its scope.  Having failed to overcome the plausibility hurdle, the patentee was not entitled to rely on post-published evidence to show sufficiency.

Claire raised some interesting points to consider at the end of her talk:

1. In light of Kitchen LJ’s treatment of paragraphs [0003] and [0006], what is/is not to be recognised as a definition in a Patent?  Claire considered that in light of this decision, Patent Attorneys must be particularly precise with definitions and the definitions must be reasonable based on the state of the art at the time.
2. At what level of generality should the plausibility test be applied?  In the event that certain embodiments of a claim are plausible at the priority date, should it be possible to “fill in the gaps” with later filed evidence (bearing in mind that pregabalin works in central neuropathic pain).
3. Should evidence of technical effect only be rejected where nothing plausible is covered by the claim?
4. Is it sometimes the case that having no data in your Patent is better than having the “wrong” data?  Claire referred to the HGS v Lilly [2011] UKSC 51 and Actavis v Lilly [2015] EWHC 3294 (Pat) cases as examples of patents which did not have data in them and yet were found to be plausible.      

Amendment / Abuse of Process

In light of Arnold J’s first instance finding on plausibility/sufficiency, Warner-Lambert sought to amend its claim to neuropathic pain to that which is "caused by injury or infection of peripheral sensory nerves”. Arnold J refused the application to amend and considered it an abuse of process.  This was upheld by the Court of Appeal.

Brian Cordery reminded the audience of the general principles relating to amendments set out in Henderson v Henderson (1843) 3 Hare 100 and Johnson v Gore Wood [2000] UKHL 65 which prevent re-litigation in circumstances not amounting to strict res judicata.   Brian considered that the same principles apply to applications to adduce new evidence on appeal to that which was considered at first instance (see e.g. Ladd v Marshall [1954 EWCA Civ 1; MMI Research v Cellxion [2012] EWCA Civ 7).

Brian also reminded the audience of the three categories of amendments recognised by Jacob LJ in Nikken v Pioneer [2005] EWCA Civ 906:

(a) before a trial; (b) after trial, at which certain claims have been held valid but other claims held invalid, the patentee simply wishing to delete the invalid claims (I would include here also the case where the patentee wishes to re-write the claims so as to exclude various dependencies as in Hallen v Brantia [1990] FSR 134. There the patentee is in effect continuing to claim which he had claimed before but in a much smaller way); and (c) after a trial in which all claims have been held invalid but the patentee wishes to insert what he hopes are validating amendments.  

Brian pointed out that categories (a) and (b) are generally considered acceptable whereas category (c) will generally amount to an abuse of process.  Brian considered whether this was altered by the Court of Appeal decision in Samsung v Apple [2014] EWCA Civ 250 where the Court of Appeal held that an application to centrally limit a Patent following a first instance finding of invalidity did not amount to an abuse of process per se, although as Brian noted, the Court was careful to point out that asserting such a Patent in the UK could still amount to an abuse:

"we believe that whether or not a patentee's attempted reliance before this court upon a patent which has been amended pursuant to a central amendment application made after trial constitutes an abuse of process must depend upon all the circumstances, including whether it would be necessary to remit the case for retrial and, if so, what the consequences of that would be.”

Having considered the law, Brian revisited the procedural history of the pregabalin litigation noting that the Defendants had raised the distinction between neuropathic and central neuropathic pain in their reply expert report.  They had also referred to it in their skeleton argument.  Warner-Lambert did not object to the argument being raised at this stage.  Warner-Lambert instead focussed its case on arguing that the skilled person would construe neuropathic pain as being limited to peripheral neuropathic pain, which was ultimately rejected by Arnold J at first instance, a  rejection which was upheld on appeal.   

Brian set out some of the arguments relied upon by Warner-Lambert as to why it should have been allowed to amend claim 3 post-trial, which include: 

• that Warner-Lambert had been the victim of procedural unfairness by the Defendants raising the peripheral/central distinction so late in proceedings; 
• that the proposed amendment should not be considered to fall within Jacob LJ’s category (c) but instead was more akin to a category (b) amendment as it simply sought to excise invalid matter from an otherwise valid claim; and
• the amendments would not in any event require a retrial.

Brian highlighted the Court of Appeal’s reluctance to interfere with a first instance judgment on this issue.  This case ultimately depended on two issues:  (i) had the subject matter of the claim as proposed to be amended been the subject of a finding (such that a retrial would not be require)?; and (ii) in light of the procedural history, had Warner-Lambert been hampered from proposing an amendment to Claim 3 at an earlier stage in proceedings?  The Court held that notwithstanding the procedural issues, Warner-Lambert could and should have proposed its amendment at the start of trial.   Had it done so, Mylan could have run a different attack on the validity of the claim.

Brian’s take home points included:

• Propose amendments at an early stage in proceedings, preferably before trial;
• Keep possible amendments in mind as the trial progresses; and
• carry out a mock trial with a view to unearthing any unexpected issues.

Brian concluded by reminding the audience that Warner-Lambert has petitioned the Supreme Court for permission to appeal and that the Supreme Court will be hearing issues relating to the construction of Swiss-type claims next year in the Actavis v Lilly (pemetrexed) case.

Construction/Infringement

IPKat readers will recall that construction and infringement issues first arose in these proceedings within the context of Arnold J’s refusal of Warner-Lambert’s application for an interim injunction and the Defendants’ application to strike out Warner-Lambert’s case under s60(2).  Arnold J held that the claim required subjective intention, on the part of the manufacturer, that pregabalin would be used in the treatment of pain at the point of manufacture and as a result there was no arguable case on infringement.  Similarly, the absence of a downstream act of manufacture led Arnold J to strike out the s60(2) infringement claim.  

On appeal, Floyd LJ had disagreed with Arnold J’s “subjective intention” approach and set out, obiter, his test of "reasonable foreseeability”.  The refusal of the interim injunction was upheld (as the Court was reluctant to interfere with Arnold J’s assessment of the balance of convenience but the s60(2) strike out was overturned.  Arnold J attempted to grapple with Floyd LJ’s test in the first instance decision on the merits of the case, but in doing so expressed serious doubts as to its correctness.  Due to the uncertainty and notwithstanding the Court of Appeal’s finding that the patent was invalid, the Court of Appeal sought to clarify the position, albeit that the Court’s comments are once again obiter.  

Stuart Baran reminded the audience that Swiss-Type claims were a fudge designed by the EPO to get around the methods of treatment exclusion from patentability that was present in the EPC at that time.  As such, Stuart suggested that the focus should be on providing a logical and fair outcome rather than water-tight reasoning.  The latter may not be feasible in light of the “legal fiction” of Swiss-type claims.  Bearing in mind that the novelty in the claim derives from the new use of the known drug and has nothing to do with “manufacture” or “preparation”, Stuart emphasised that a fair and logical outcome would result in the patentee being rewarded for its investment in bringing a new treatment to patients and generic pharmaceutical companies being allowed to market the drug in question for the non-patented indications.  

Stuart provided an overview of the approach in some other European jurisdictions to what constitutes an infringement of Swiss-type claims.  The German Courts appear to have adopted the “only packaging will do” approach i.e. there must be some direction to the new use on the SmPC or PiL in order for there to be a finding of infringement.  Other forms of encouragement will not suffice.  In Spain, the Courts have required some positive encouragement on the part of the generic.  Finally, in France the Courts have arguably adopted a more patentee friendly approach and require the Gx to take positive steps to prevent infringement.  In light of the obiter comments from the Court of Appeal, the UK appears to be most closely aligned with the French approach.

Stuart then turned his attention to Floyd LJ’s test for infringement of Swiss-Type claims, that of whether the manufacturer knew or could reasonably foresee that the drug would intentionally be used for treating the new indication.  Stuart pointed out that there had been some disagreement following Floyd LJ’s first articulation of this test, as to whether the manufacturer needed to foresee that a given box of tablets would be intentionally administered for the patented indication.  This interpretation of the test had led Arnold J to a finding of non-infringement.  However, Floyd LJ has now clarified that it is sufficient that the manufacturer foresees that some of the drug that it puts on the market will be so administered.  The second mental element, i.e. that the drug will be “intentionally administered” is only intended to exclude “accidental treatment” i.e. where a patient is administered drug X for disease Y1 and unbeknown to the patient/physician, the patient is suffering from, and is treated for, disease Y2.  

Stuart pointed out that the presence of these two mental elements is not the end of the inquiry.  The Court appears to have recognised that in circumstances where a generic pharmaceutical company has taken all reasonable steps to ensure that its drug is used only for the non-patented indication(s), it does not seem fair that it should be found to infringe by virtue of it being reasonably foreseeable that some of its drug will be used in the patented indication due to flaws in the regulatory system.  As such, a generic can avoid a finding of infringement if it has taken “all reasonable steps”.  Although it is unclear exactly what will constitute “all reasonable steps” and it is likely to depend in part on the circumstances of each case, Stuart suggested that the following steps should be considered:

• Alerting doctors/HAs/pharmacists in advance
• Avoiding supply through certain channels
• Avoiding oversupply vs. old indication market 
• Steps to update clinical software; and 
• Co-operating with NHS to issue guidance.

Stuart then turned his attention to the question of indirect infringement pursuant to s60(2).  As s60(2) requires the supply of “means relating to an essential element of the invention,” it is necessary to consider what the invention in a Swiss form claim is.  Stuart highlighted that s125 of the Patents Act states that an invention “shall, unless the context otherwise requires, be taken to be that specified in the claim.”  As such, notwithstanding the wording of the claim, it is clear that the invention in a Swiss form claim is ascribing the novel therapeutic purpose to the medicament.  The invention has nothing to do with “manufacture” or “preparation”.  This is a case where the context does otherwise require. By adopting this approach, the Court of Appeal were able to reach a finding that there was no requirement for a downstream act of manufacture and as such, the ascribing of purpose by, e.g. the pharmacist in applying a label to the product, was relevant to the question of infringement.  

Stuart concluded by suggesting that relief provides the best opportunity for the Court to take the necessary steps to reach a fair outcome.  For example, Stuart suggested that it was perfectly possible to analyse the market share data / % of use for a particular indication and award financial relief accordingly.  However, there was some discussion about whether this was possible in light of the different financial models relating to generic and originator pharmaceutical products.  

On the question of interim injunctions based on second medical use patents, Stuart suggested that it was difficult to envisage a scenario in which the grant of an interim injunction precluding the generic pharmaceutical company from marketing its product would be appropriate.  However, in the event that an interim injunction is awarded, Stuart considered that it would be important for the English Courts to be more prescriptive as to the terms of the injunction than they had arguably been in the past.  Stuart noted that other European Courts have adopted this approach to injunctions.  

In questions which followed the seminar, the panel considered the difference in the infringement analysis of Swiss type claims and post-EPC 2000 claims in light of Floyd LJ’s approach to the former.  Notwithstanding that it is settled law that the former should be considered purpose-limited process claims and the latter purpose-limited product claims and as a result different sections of the Patents Act will be engaged, it was agreed that the analysis is likely to differ very little in practical terms, as the “ascription of purpose” will still amount to an infringing act, “divided” acts of infringement (i.e. where different entities carry out different parts of the claim) will still suffice and the reasonable foreseeability test imparted by the inclusion of the word “for” in the claim will still apply.  It is plain that further case law will be required to crystallise Floyd LJ’s obiter comments in Warner-Lambert.  

The panel also agreed that an ideal solution would include regulatory law reform.  However, despite the apparent agreement of the judiciary (see Arnold J at paras 722-726 of Warner-Lambert v Actavis [2015] EWHC 2548 (Pat)), it does not appear that such reform is imminently on the horizon."