Fordham 2016: Will purposive construction of Swiss-type claims save second-medical use patents?

Brian Cordery sets the stage
for second medical use

After hearing the views of the IP judiciary, the Fordham audience was treated to a session devoted to second medical use claims, with Brian Cordery (Bristows LLP) moderating the panel.  Brian set the scene explaining that there are two types of second medical use inventions in Europe - Swiss-type claims which are purpose-limited process claims (according to the English) and EPC 2000 claims (purpose-limited product claims).  Swiss-type claims take the form:

"use of a drug X in the manufacture/preparation of a medicament for the treatment of disease Y"

EPC 2000 claims take the form of:

"Drug X for the treatment of disease Y"

The European Patent Office's Technical Board of Appeal (TBA) have told us that these claims are not synonymous in scope (see T 1780/12 - University of Texas) and you cannot amend from one to another. Swiss-type claims have been outlawed since 2010, but there will still be Swiss-type claims kicking around until about 2035 (right around when Brian hopes to retire).

Last year was the most important year, so far, for second-medical use claims with Warner-Lambert's Lyrica (pregabalin) litigation undoubtedly the biggest case (see numerous Kat reports here).  With the UK Court of Appeal hearing argument starting on 24 May 2016, it is hoped that there will be a decision before the summer break.  Brian considers the Alimta (pemetrexed) case as potentially more interesting (see summary of decisions here). It concerns Eli Lilly's patent with Swiss-type claims to the use of pemetrexed disodium and Vitamin B12 for the treatment of cancer.  Actavis sought to market a pemetrexed product but in a different salt form.  However, when Actavis's product is reconstituted with saline pemetrexed and the alternative salt would be present in solution as separate ions and thus fall within the claim.  The  Court of Appeal considered that the reconstitution would be an act of infringement, as would be the administration to the patient.  Brian noted that, contrary to the suggestions made in the earlier judiciary session, the English and the Germans are not always in harmony.  In pemetrexed, the English court (Arnold J) granted a declaration of non-infringement (DNI), whereas the Germans did not.  Both courts were reversed on appeal with the German court granting the DNI and the English court refusing it with respect to the indirect infringement claim.  Both cases will be going to the Supreme Court.  A third case to watch out for is Novartis's Aclasta (zoledronic acid) patent with Swiss-type claims for the use of zoledronic acid in a once-yearly i/v administration for the treatment of osteoporosis.  The preliminary injunction was refused at first instance, granted by the Court of Appeal in January 2015 with a decision on the issue of direct infringement expected late Spring 2016.

Brian commented that perhaps Mr Justice Arnold (who has heard the Warner-Lambert and Lilly cases), is not looking at the claims from the right perspective.  He suggested that you have to be purposive with these cases and identify the invention (see Eisai G5/83).  Getting too bogged down on the "manufacture" element (literal) as opposed to the invention (purposive) is not the right way to go.  Lord Justice Floyd in Lilly has adopted a purposive approach in relation to indirect infringement.  In Warner-Lambert, Lord Justice Floyd suggested, also in relation to indirect infringement, that you could construe the claim as a multi-step process claim.  Mr Justice Arnold has not.  Is there a distinction to be drawn between the fact that Lilly concerns a reconstituted product and Warner-Lambert a white pill (so the focus is on the active manufacture)?  Surely not - as such distinction is arbitrary.  This is why the construction of a claim as a multi-step process claim could be preferred in that it avoids this arbitrary distinction, it is more in line with the purpose of the law and more closely aligns with the two forms of claim, while not abusing the language of the claim (see section 125 Patents Act 1977).  But if the manufacture element is all important, then one has to look at what the manufacturer knows or could reasonably foresee about the intended use.  This then potentially causes a myriad of enforcement difficulties.

While the courts are wrestling with the correct approach to these types of cases - both on construction and infringement - we are starting to see cause for optimism via non-legal routes.  For example, tendering authorities splitting tenders between patented and non-patented indications.

Sir Robin wants to make the payers pay

Sir Robin Jacob (UCL) considers second medical use a very serious subject and, indeed, the biggest topic on this year's Fordham agenda.  Sir Robin's view is that patent law is currently inadequate to deal with this issue.  The judgments that are coming out on second medical use are almost branches of metaphysics, given the complexity of the analysis - what does "manufacture" mean and so forth.  But underlying this is a much more serious problem - that of policy.  It is never self-evident that the first use of a medicine will be its only use.  Also, it is not the case that every side-effect is a bad side-effect.  The problem is that you often cannot obtain any sort of patent on a second medical use.  Why?  Because, for example, a doctor may be treating cancer in a patient who also has Parkinsons.  Then the doctor notes that the patient's Parkinsons is also starting to improve.  The doctor then publishes the results.  The publication is then prior art.  Who will invest $500 million to follow-up on this research without patent protection?  A further problem is that, even if you have patent protection, who do you sue?  Big pharma will never want to sue a doctor for prescribing the product to a patient.  That is not a good PR story.  Is there a solution?  The ideal would be to institute rules whereby the prescription says what the medicine is being used to treat (i.e., the indication), with everything else flowing from there.  Some think that this would impact on patient confidentiality in that the pharmacist would learn of the patient's condition.  However, Sir Robin commented, they often already do. Sir Robin stated that he does not consider the generic companies a suitable target either, since injunctions will almost always be impossible (although it may be legitimate to get them to include information by way of labelling or advertising alerting pharmacists as to what indication their product is to be dispensed for).  So who and what does that leave?  The payers.  It could be solved politically, Sir Robin noted, but emotions and irrationality get in the way.  Sir Robin warned that if you do not make someone pay, then the value  of these second medical uses will be largely unexplored to the detriment of patients.

Ever the optimist, Mark Ridgway sees
solutions on the horizon by way of
purposive construction of
second medical use claims

Mark Ridgway  (Allen & Overy LLP) who co-led the Lyrica litigation in the UK on behalf of Warner-Lambert reminded the audience that whether a medical use patent can be effectively enforced has wider repercussions than just second medical uses, since first medical uses can also be affected.  For example, if a first use is protected by a Swiss-type claim and a second use emerges that is unpatented, the patent on the first use is undermined if it cannot be enforced effectively and cross-label dispensing occurs.  The incentives for drug discovery are thereby even more greatly reduced than in Sir Robin's example (there is a positive disincentive for a patentee to seek further authorized indications, as they could undermine the original).  On a positive note, however, Mark observed that, in Sir Robin's example of where a doctor's publication is prior art, it will not always be the case that a medical report, which might be anecdotal, will render a patent obvious.  In response to Brian's question as to whether Mark was optimistic that we are nearing a solution on second medical use, Mark stated that he was.  After a very busy year in Europe's courts, Mark said, we have a clearer idea as to what can or cannot be done.  Although there is not yet a consistent view as to how these claims will be construed, the courts across Europe will hopefully arrive at a sensible construction soon, since there needs to be a way of making these patents enforceable.  However, Mark concluded, we will need to await the Court of Appeal to see how things develop.

Jurgen Dressel's focus is on
the patient - they must be
at the heart of the solution 

Jürgen Dressel (Head of Global Patent Litigation Strategy, Novartis), emphasized that in this debate we should not lose sight of the patient.  We have many stakeholders in the game, he continued, but the patient is paramount.  Although stakeholders adopt short term budgeting, we need to think about the long term implications of not incentivizing these types of innovations.  We must remember, Dressel continued, the patient who will not obtain a treatment because a company will not spend the money to develop a medicine if there is insufficient exclusivity.  2015 was a really crucial year for second medical use cases, with the German, Dutch and UK courts at the frontline grappling with these issues.  These cases represent a significant opportunity to explore this area and test the boundaries.  Dressel urged all stakeholders - innovators, generics, doctors and insurers - to collaborate to find a way forward in order to create a segment of the market where generics can come to the market for the non-patented indication (skinny label) without impacting the innovator's second-medical use market.  Dressel noted that there is starting to be a drive for collaboration and a feeling that each side will play its part.  The Lyrica case has "really woken everyone up" to these issues.  Dressel recalled that he once spoke to his general practitioner about these issues and asked him how he prescribes Lyrica.  His doctor explained that he does not prescribe by brand (in Germany a prescribing medical practitioner can tick a box to prescribe by brand).  The reason he does not is because it is too much bureaucracy for him and he operates under a tight budget.  Dressel stated that Warner-Lambert tried to address these practical bureaucratic issues by improving the prescribing software to make it easier for the doctor to prescribe by indication.  However, when you read Arnold J's decision (in the UK) he explains that GP software systems have three levels of warning message on screen when a doctor comes to prescribe a medicine.  Are doctors actually going to pay attention to these different levels?  They are likely only going to pay attention to the top level which is reserved for serious health issues - not patent issues.  The solution needs to be reframed by focusing on the doctor's perspective and by making it easier for doctors to prescribe by indication.  The discussion also needs to be reframed by emphasizing the value behind clinical inventions and not just seeing them as a method for innovators to fill product.  At the end, Dressel concluded, it needs to be about the patient.

Judge Kalden points to Article 69
as a potential solution to address
enforcement problems with 
Swiss-type claims

Judge Rian Kalden (Court of Appeal of the Hague), speaking in her personal capacity, focused on the construction of Swiss claims.  She agrees that patent law falls short in currently providing solutions in this area.  However, if we do not have any other solutions on how to deal with second medical use, including from the regulatory arena, then we need to focus on the claims.  Rian examined the way in which the Swiss-type claims could be interpreted to provide greater protection.  Rian reminded the audience that the reason we are in this situation is that in Europe there is a protection problem when the compound is not new, given there is an exclusion for methods for medical treatment.  The solution was the creation of the Swiss-type claim with the manufacture element rendering the claim a purpose-limited process claim.  However, it is a legal fiction.  In reality, it concerns a new form of medical treatment  - the manufacture is neither new or inventive.  The invention really lies in the new use.  The EPC 2000 claim form has not really changed anything - it is a purpose-limited product claim (Article 54(5) EPC).  There are certain problems with Swiss-type claims.  Who is the manufacturer?  It is unlikely to be one and the same person.  For example, the white pills containing the active ingredient could be produced in India, packaged in Bulgaria and distributed elsewhere in Europe.  What is manufacturing?  Is it just making tablets or does it go as far as advertising (as some have suggested)?  If the manufacturer intends to sell for the patented use, but the pharmacy is not dispensing for the patented use - is this still infringement (Article 64(3))?  Whose actual or constructive knowledge counts in this chain?  If the manufacturer only intends non-patented use and the pharmacy knowingly dispenses for the patented use, is it not an infringement as he is not manufacturing?  Indirect infringement is problematic as there is no manufacturing downstream - this makes it almost impossible to find infringement.  It is the manufacturing element of the Swiss-type claim that causes this problem.

Rian stated that T-1780/12 is referred to a lot by the Swiss and Dutch courts as a process claim.  The case was actually a case about double patenting.  A Swiss-type claim was granted and then a divisional was granted with EP 2000 claims for exactly the same thing.  The Examining Division refused to grant the claim because it related to the same subject matter.  The Technical Board of Appeal said that the two are different claim formats:

"It is generally accepted as a principle underlying the EPC that a claim to a particular physical activity (e.g. method, process, use) confers less protection than a claim to the physical entity per se, see decision G 2/88 (supra, reasons, point 5). It follows that a purpose-limited process claim also confers less protection than a purpose-limited product claim. The scope of protection sought by the invention claimed pursuant the present main request is thus noticeably different from the scope of protection conferred by claims 1, 24 and 25 of EP1..."

However, the Enlarged Board in G2/08 said that it seemed clear that no such difference was intended.  The Enlarged Board referred to the preparatory document, which explained that the provisions filled a loop hole in the legislation.   In particular, point 4 to the explanatory notes to MR/18/00 confirm the clear intention that:

"The new Article 54(5) EPC eliminates any legal uncertainty on the patentability of further medical uses. It unambiguously permits purpose—related product protection for each further new medical use of a substance or composition already known as a medicine. This protection is equivalent, as far as the further uses are concerned, to that offered by the 'Swiss type claim'. In contrast to previous Article 54(5), now Article 54(4) EPC, providing broad (generic) protection for use in a medical method for the inventor of such use for the first time, new Article 54(5) is expressly limited to a specific use. This limitation is intended to match as closely as possible the scope of protection to the scope provided by a 'Swiss type claim'."

There is therefore a strong argument to construe Swiss-type claims as if they were purpose-limited product claims and therefore ignore the manufacturing element.  In doing so, Rian explained, the right balance is struck between the legitimate interests of the patentee and legal certainty of third parties.  Further it is consistent with Article 69 EPC and the Protocol.  In Kirin Amgen, the House of Lords stated that the skilled person should be asked what the invention is that is being protected.  You do this by construing the claim purposively.  Since everyone knows that Swiss-type claims are a legal construct and understands what they actually doing - protecting medical use - Rian concluded that this is a potential solution.

Marleen outlines a menu of remedies,
but are any good enough?

Marleen van den Horst (BarentsKrans) outlined the wide ranging scope of potential claims being brought and remedies being sought to try to forge a path through these issues.  Do we base claims on direct or indirect infringement, Marleen asked.  It depends on intent, knowledge and forseeability.  What kind of precautionary measures are patentees, generic suppliers or others required to take to prevent infringement?  How far do you have to go?  Do you just need to send letters, take legal action or stop delivery altogether?  It is unclear, Marleen continued, what is required to satisfy the requirement that either party has done everything to inform the market.  However, in Marleen's view, the patentee should have the burden to protect their market.  Different approaches have been adopted in different jurisdictions.  Marleen explained that the Dutch Medicines Evaluation Board (MEB) only carved out the patented indication in the the printed version of medical information, but not online.  After being sued by Warner-Lambert they took the action to show the carve out ounline as well.  In Denmark pharmacies have been sued.  In Germany, health insurers were sued on the basis that they do not designate between patented and non-patented indications.  Is this effective?  Even if there is a distinction in the tender, this does not mean that the markets or pharmacies are aware of what indication the medicine has been prescribed for.  There were actions against the ministries of health in some countries which resulted in prescription rules being amended to get doctors to prescribe by indication to differentiate between the patented and non-patented indications.  Remedies have been granted against generic suppliers based on injunctions claimed on direct or indirect infringement.  Other remedies include orders to restrict generic supplies to certain amounts of products and to open the books of the generic companies.  The question is whether these remedies are effective, reasonable and proportionate and permissible in that they do not violate European competition law or the Charter of Fundamental Rights.

Christine finds the German remedies
for second medical use limited, but
remains positive

Dr Christine Kanz (Hoyng Rokh Monegier) stated that in Germany we had originally seen a tendency to look at marketing materials to assess whether a claim was being infringed.  The situation in German had been very stable for a decade, Christine explained.  The first Ribavin case in 2004 held that the only thing you can consider is the packaging material.  If the materials do not mention the patented indication, then the defendant is not infringing.  This would hold true even if patients that fell within the non-patented subgroup also fell within the patented subgroup.  In the second Ribavin case in 2013, the Dusseldorf Regional Court held there was no infringement as there must be "a direct and specific link between the act of evident preparation" and the manufacture and distribution of the product.  The link must "clearly urge the user to use the item in a patent protected way."  However, the court held that you cannot consider marketing material to establish this link - promotional flyers and activities are irrelevant, as are actual prescriptions of the patented indication. Then the German Warner-Lambert case came along.  The Warner-Lambert case showed us that maybe the whole issue cannot be solved by patent law alone, but also regulatory, social and medical law.  In this case, generics participated in tenders for rebate agreements relating to the active ingredient, but the tenders were not restricted to the non-patented indications.  The Regional Court of Hamburg held that the defendant was prohibited from entering rebate agreements without declarations that their pregabalin products could not be used for the treatment of the patented indication (neuropathic pain).  The appeal is pending.  Christine stated that she found the remedy the court granted quite limited in effect.  The generic only has to tell the health insurers that they can only offer the non-patented indication.  Then what?  It does not go far enough.  The tenders are still unlimited, although the generic may not win the tender if the health insurers are looking for someone who can supply throughout all indications.  Warner-Lambert also found that unsatisfactory.  They therefore started an action in the German tender courts arguing that tenders should not be unlimited, but should distinguish between patented and non-patented indications.  The Court essentially agreed, saying that  patent law should not stand behind social law - they have different objectives.  This decision was confirmed by the Higher Regional Court of Dusseldorf in 2015.  The lessons coming out of the German experience is that second medial use claims are not a matter that can be solved by patent law only, commented Christine.  Do we really have to be careful with doctors on the subject?  This matter does not impact the doctor's freedom of therapy.  Its only about whose drug they prescribe.  Why can't pharmacists see what the prescription is for, Christine asked.  Christine concluded by stating that she is very optimistic about the future.  2015 saw good movement on a number of complex issues and 2016 will provide further clarification.

Discussion

Steve Tepp from the audience then asked the question as to how product liability can be addressed if there is off label use and no way to ascertain what indication the product is being prescribed for.  Dressel responded that to associate product liability with a specific drug with a specific manufacture would be a difficult issue and it is an issue.  As a matter of principle, this would be a troublesome exercise as it depends on the country and, of course, has pharmacovigilance implications in terms of collecting and reporting patient data associated with the active ingredient.  Cordery also pointed out that the issue of "cross-label" dispensing (which arises in relation to usage patents) is not the same as "off-label" use (which is where the product is not licensed for an indication at all).  The two should not be confused.

The second medical use panel looking forward to more clarity coming
from the courts in 2016.  But should we only rely on the courts for a solution?

The AmeriKat wonders whether by prescribing and dispensing guidelines/procedures not being amended to ensure that generic products are only prescribed/dispensed for non-patented indications, it does not cause problems under pharmacovigilance legislation (namely Regulation No 726/2004 and Directive 2001/83)?  Certainly things are made more difficult if no-one knows or is difficult to ascertain what the various products are being used to treat, but that is not unique to cases where there are medical use patents in play.  On this point, AmeriKat also recalls the submissions from AIPPI's Spanish Group to the 2014 question on second medical use. In it they recommended a list of measures to be considered to ensure the effective protection of investments in developing new uses for known compounds which included establishing a public system for identifying the final use of a product, reimbursement being conditional on proving final use of the medicament, the specification of the indication for which the product is prescribed and updating software systems on both the patient and pharmacy end. The Group concluded that such measures would constitute "a positive step forward from the perspective of pharmacovigilance and for monitoring possible malpractices in off-label medication."

The Amerikat was also struck by the purposive construction approach and its impact on enforcement.  The argument for the approach keeps in mind the specific nature of the origins of Swiss-type claims - a legal fiction created to provide protection for new medical uses of known compounds.   The "specific nature" language is also present in the Enforcement Directive.  Recital 17 of the Enforcement Directive provides that:

"The measures, procedures and remedies provided for in this Directive should be determined in each case in such a manner as to take due account of the specific characteristics of that case, including the specific features of each intellectual property right and, where appropriate, the intentional or unintentional character of the infringement." 

By ignoring the specific features of the Swiss-type claim on construction, are the courts also ignoring the specific features of the right when it comes to providing effective, proportionate and dissuasive remedies under Article 3?  If they are, then the remedies so far ordered may be failing to protect the medical use, resulting in a chilling effect to innovation contrary to the objectives of the Enforcement Directive as set out in Recitals 1 and 3 (e.g., "without effective means of enforcing intellectual property rights, innovation and creativity are discouraged and investment diminished").  With the current consultation on the Enforcement Directive ongoing (submissions due on 15 April 2016), the AmeriKat wonders if more will be said be said about the role of the Directive as a means for ensuring more effective and proportionate enforcement of second medical use claims.