The time and financial expense of bringing a novel drug to market is enormous and the process is fraught with risk. Very often, a putative novel chemical entity showing promising efficacy in pre-clinical or animal models will fail to reach the clinic due to unforeseen toxicology or dosing issues. As such, the repertoire of usable drugs in the clinic is finite and comparatively limited.
On the other hand, if a drug which has successfully negotiated the approval process and has reached the clinic can be found to have other clinical uses, the risks, timescales and costs of obtaining regulatory approval for such uses is significantly less. Thus, there is strong impetus in the pharmaceutical industry to repurpose their existing medicinal products to the treatment of second or further medical conditions (or “indications”), thereby expanding their drug portfolio in the most efficient way. As a result, more therapies can be made available to the public in a shorter period of time.
Classic examples of repurposed drugs include aspirin (a non-steroidal anti-inflammatory repurposed as an anticoagulant useful in preventing heart attack, stroke and DVT), thalidomide (a treatment for leprosy repurposed as a treatment of multiple myeloma), viagra (a treatment for angina repurposed as a treatment for erectile disfunction) and, more recently, remdesivir (a hepatitis C medicament repurposed to treat COVID-19). The latter example really shows the potential power of the system – a life-saving drug for a novel respiratory disease was in use in the clinic within a matter of months rather than the decade or so it sometimes takes to develop a new treatment from scratch.
Society at large is keen to have as many pharmaceutical treatments available as possible. To further incentivise the pharmaceutical industry to investigate new indications for known medicaments and bring those discoveries to the clinic, governments worldwide have clauses in their patent laws which exempt second medical uses from the usual novelty criteria for patentable inventions, thereby providing a route to a 20 year monopoly protection for the further medical use of the drug. The conventional wisdom is that this monopoly gives the pharmaceutical company the chance to recover the costs of bringing a new therapy to the market and so encourages investment.
However, critics point out that this system is overly generous – giving the same compensation to the development of the original drug as the new use – and, if left unfettered, open to abusive practices by innovator companies who can game the system to “evergreen” patent protection for their original medicament and build patent thickets around their drug to keep generic competition off the market.
In Europe second or further medical uses of known compounds are claimed in patents either as a Swiss style claim – “use of compound X in the manufacture of a medicament for the treatment of disease Y” – or, more recently, in the form of “compound X for use in the treatment of disease Y” as set out in the EPC 2000 (Article 54(5)). Patentability is determined by the novelty and inventive step of the proposed treatment. It has been established that novel and advantageous dosage regimes within the same disease, and even the specific treatment of selected patient groups within the known target population may be claimed.
Available case law from the European patents courts has so far been based on Swiss style claims and it has become clear that some European states consider such claims to be purpose-limited process claims and others consider such claims to be purpose-limited product claims. It is expected that the modern medical use format will be interpreted universally as purpose-limited product claims. These differences in interpretation of Swiss style medical use claims have implications for the heading under which infringement may be alleged (direct and/or indirect) and as such may have particular implications for the outcome of a case depending on the underlying facts, for instance the identification of alleged infringing parties, the impact of their geographical location and the timepoint at which assessments of intention are made.
As will be explained, the jurisdictions which view such claims as purpose limited process claims seem to be more friendly to generic defendants than innovator claimants, whereas the reverse is true in those jurisdictions which view such claims as purpose limited product claims.
In an attempt to address the issue of evergreening and patent thickets keeping generic competition out of the market, the EU introduced provisions (primarily Article 11 of Directive 2001/83/EC) which permits a generic applicant to carve out from the summary of product characteristics (SmPC) and other product information for their generic medicinal product any medical indications, dosage regimes or patient groups that are the subject of patent protection. In other words, at least in principle, a generic company is able to obtain authorisation for a generic version of an innovator company’s branded product and be able market that generic product for the original indication without being at risk of allegations that the product is to be used to treat the new indication for which the original product has been repurposed, thereby infringing the repurposed drug patent.
Although Directive 2001/83/EC is a piece of EU legislation, similar provisions are in place in the non-EU states of UK, Switzerland and Norway.
In reality, although the concept of a skinny label is an elegant solution to the issue from a perspective of patent law, the effects of the label on the relevant players in the health services is marginal at best and so use of the generic in the patented indication is not reliably prevented.
In most European states healthcare professionals usually (sometimes encouraged by electronic prescribing systems) prescribe a drug to a patient by the international non-proprietary name (INN) of the drug and no mention is made on the prescription of the condition to be treated. Only in certain circumstances, usually verifiable clinical reasons, will the healthcare authorities permit a practitioner to prescribe the branded product. In the pharmacy, the pharmacist is usually obliged by law and/or by insurers/funding schemes to fill the prescription with the most cost-effective version of the drug as defined by INN. Without knowledge of the condition being treated, it is therefore common for a generic product to be taken by a patient to treat a patented medical indication – so called cross-label use.
Of course, any instance of cross-label use eats into the market share of innovator companies and strangles their revenue streams. The situation on the ground is clearly unsatisfactory for innovator companies seeking to recoup their investment in the novel second medical uses they have brought to market. As such, a number of such companies have launched litigation against generics at the patent courts in various European states in an attempt to push the boundaries of skinny labelling practice to include the requirement that generic companies take further action in the market to prevent cross-label use.
These courts have tried to find a suitable balance between the rights of the innovators and the need for generics to reach the market as soon as possible. The results have been mixed on account of the precise nature of the healthcare system in each country and the approaches to medical use claim interpretation adopted by the individual courts.
In Germany it is settled law that Swiss style second medical use claims are purpose limited product claims (Pemetrexed, BGH X ZR 29/15; Kollagenase I, FSC, GRUR 2014, 461), as are EPC 2000 type claims. Thus, infringement actions have tended to focus on direct infringement though the handling of products per se at all levels of the supply chain, including the end use and the courts have needed to look beyond the inherent suitability of the product in question for the claimed therapeutic use to the intent, knowledge and behaviour of the alleged infringer (generic manufacturer) and the actual outcomes of the alleged infringer’s activities to satisfy the “for use” criteria.
Previous case law based on a “manifest arrangement” test (much like the UK’s current “outward presentation” test, discussed below) heavily favoured generics, and so long as the generic product was marketed with a skinny label, the generic manufacturer was safe even in the face of evidence of significant cross-label use.
More recent case law has swung back towards the innovators. Following Rabattvertrag (GRUR-RR 2015, 330) Ostrogenblocker (OLG Dusseldorf, I-2 W 6/17) and Fulvestrant (OLG Dusseldorf, I-2 U 27/18) it has become clear that the behaviour and knowledge of the alleged infringer will be considered and a skinny label will not be adequate to avoid a finding of infringement if the alleged infringer acts to promote cross-label use (e.g. engages in a tender process which would inevitably result in cross-label use), or if sufficient evidence can be provided that there is or will be significant cross-label use, the alleged infringer is, or at least should be, aware of the extent of or risk of cross-label use and the alleged infringer does not take the steps which could be reasonably expected from him to minimize the risk of infringing use.
Interestingly, despite the German courts swinging behind the innovator companies in finding infringement they seem to be equally keen to ensure that the generic product remains available for the non-patented uses. Indeed, rather than handing down orders to remove the generic product from the market entirely, the German courts have instead ordered generic companies to actively educate the various players in the drug prescribing system about the patented use and the correct prescription of the branded product in that context, to encourage the appropriate updating of drug prescribing databases, and to include warning notices on the drug packaging and labels with respect to the prohibited use of the drug for the patented second medical use (within the bounds of regulatory law).
MSD v Teva (Ribavarin – ECLI:NL:GHDHA:2017:2807) and Novartis v Sun (Zolendronic acid – ECLI:NLGHDHA:2015:1769) are the most relevant case law in the Netherlands in connection with the infringement of repurposed drug patents and skinny labels. These cases place the current balance heavily in favour of the innovator company.
In MSD v Teva it was determined that are policy reasons for interpreting Swiss style claims as purpose limited product claim rather than in line with their literal wording, and as such capable of being indirectly infringed.
The Supreme Court further found that despite a skinny label a generic manufacturer/distributor will still directly and indirectly infringe a Swiss style medical use claim if he foresees or should have foreseen that the product will be used for the purpose that is the subject of the Swiss-type claim. For foreseeability, it is enough that it is apparent to the skilled person that product would be suitable for such use.
Thus, in the Netherlands, as confirmed in Novartis v Sun, simply carving out a patented indication from the product SmPC will almost certainly not be sufficient to avoid infringement. However, in MSD v Teva it was made clear that infringement may be avoided if the generic company takes all measures that can be reasonably required to prevent the product from being dispensed downstream for the claimed indication. Given that such dispensing practices are driven by insurance funds who usually tender without consideration of patent rights, it is difficult to see how much a generic company can do to improve its position in reality.
The relevant case law in France involves two preliminary injunction actions by Warner-Lambert and Pfizer against generic companies marketing pregabalin.
In the decision of December 2016 it was made clear that the French courts interpret Swiss style second medical use claims as purpose limited (manufacturing) process claims. As there is usually no downstream manufacture of the pharmaceutical in question, infringement actions tend to focus on the direct infringement of process claims by the handling the direct product thereof in the form in which it leaves the manufacturer and the representations the manufacturer makes as to its potential use. As such, less emphasis seems to be given to how the product is in fact eventually put to use.
In the decision of October 2015, the Paris District Court found in favour of the generic company (Sandoz) indicating that a proper skinny label accompanied by reasonably comprehensive steps to educate practitioners and pharmacists about patented indications was enough to avoid a finding of infringement even in light of evidence of cross label use or even significant likelihood thereof. However, it may be that evidence of inducement or encouragement to prescribe cross-label would be enough to tip the balance in the other direction.
The Italian courts interpret Swiss style claims as purpose limited product claims and as such capable of being indirectly infringed.
Very recent case law in Italy (Everolimus PI 2022) has moved the balance towards innovator companies from the position that was very pro-generic. Now it is no longer sufficient to rely solely on a skinny label and it is now the responsibility of generic companies to work with the Italian Agency for Medicinal Products (AIFA) to notify all relevant parties involved in the purchase and use of their products about any patented indications, to inform them that such products should not be used in the protected indication, and maintain e-prescribing software which prevents cross-label dispensing. It used to be innovator companies that shared the burden with the AIFA. Actions by the generic company that would be reasonably likely to result in cross-label use (e.g. participation in non-discriminatory tenders) will lead to probable findings of infringement.
Diverging case law from the Court of Appeal means the current situation is unclear. The previous position of the courts (Madrid Court of Appeal) favoured the generic companies quite heavily and with a suitable skinny label infringement was unlikely to be found unless the actions of the generic company clearly amounted to the promotion of cross-label prescribing.
More recent case law from the Court of Appeal of Barcelona has found that a skinny label is insufficient when it is probable that cross-label use will take place. Prescribing and dispensing practices in Spain make this a real risk and so it was found that generic companies must “contribute fairly to avoid the prohibited outcome” The defendant generic company was ordered to inform customers that their product must not be prescribed or dispensed for the patented indication, and not to supply their product if, despite such steps, they would have reasonable grounds to suspect that it would, nevertheless, be used for the patented indication.
Greece and Sweden
It these countries there is a strict drug substitution policy and so skinny labels are disregarded entirely and have no effect. Presently, we are not aware of any case law in these countries relating to infringement of second medical use patents within this system of mandatory cross-label prescribing, and the consequent liability of generic companies.
Unusually, prescribing in Denmark is by brand name rather than INN. Nevertheless, a substitution scheme is in place in pharmacies which encourages the dispensing of the cheapest available version of the drug unless the physician has indicated that substitution is not to occur. Each drug has a defined substitution group. This scheme is administered by the Danish Medicines Agency (DKMA) and following pregabalin litigation the DKMA has taken it upon itself to closely monitor these substitution groups and ensure drugs for patented indications are not placed in a substitution group and to ensure physicians prescribed the branded drug for patented indications. This minimises the extent of cross-label prescribing in Denmark.
The United Kingdom
Medicament prescribing in the UK is generally by INN but there is no obligation to do so. There is no formal system of substitution and so if the branded drug is prescribed, the pharmacist must dispense that product. There is however no mention of the medical indication on the prescription and if a physician prescribes by INN for a patent protected indication, cross-label use will occur. To try to minimise such outcome the NHS provides guidance to clinicians requiring them to prescribe by reference to brand name when prescribing for certain patent protected medical indications and e-prescribing software is maintained to reflect this. Such actions appear to be effective.
Case law in the UK has meandered recently between generic and innovator positions, but following the Actavis decision in the Supreme Court (Warner-Lambert v Generics  UKSC 56) it seems to be the case that a product with a properly written skinny label will be sufficient to avoid infringement.
Thus, a “subjective intention” test (i.e. where the manufacturer intended the product to be used for treating a patented indication) and a “reasonable foreseeability” test (i.e. where it is known or reasonably foreseeable that some of the product will be used for treating a patented indication) were both rejected for an “outward appearance test”.
That said, the point is not entirely settled because of minority dissent amongst the judges with two of the five preferring a “subjective intention” test. The “reasonable foreseeability” test was however soundly rejected for being so far in favour of the innovator position that it would offer a de facto extension of the expired patent for the original use until the expiry of the patent for the new one.
The “subjective intention” test was not favoured by the majority because liability would turn on an acting party’s state of mind, liability could found for one party on the basis of the actions of separate actors (e.g. the manufacturer, distributor or pharmacist), and it raised practical evidential problems in showing these states of mind. Instead, the more objective nature of the “outward appearance” test was favoured, in particular because it reflected the critical feature of Swiss-form claims in that they are directed to the process of manufacture and not at subsequent use of the product, and a test that could be applied, once and for all, at the factory gates is appropriate.
The fact that the comments on infringement were strictly obiter and the fact that they were not unanimous, with some of the judges favouring the “subjective intention” test, has led many commentators to speculate that there will be some opportunity for future court cases to consider the wider actions of the generic defendant when assessing infringement and for infringement to be found where compelling evidence can be provided of clear intent by the generic company for its skinny-labelled product to be used to treat the patented indication.
Recent pro-generic legislation in Switzerland has clarified the situation. Swiss style medical use claims are considered to be purpose limited process claims and infringement by a manufacturer or supplier will be found only if use of the product to treat the patented indicated is promoted explicitly. Knowledge of actual use, or foreseeable use, is not enough.
Although we are not aware of any particular case law on the interpretation of medical use claims and/or skinny labels in Norway, the matter is somewhat academic in view of the progressive drug prescribing practices there. Unusually, in Norway prescriptions include mention of the medical indication being treated in the form of a specific code and the Norwegian Medicines Authority maintains a comprehensive computer system showing pharmacists which indications require branded products to be dispensed. Thus, inadvertent cross-label use is ruled out.
The variability in the interpretation of Swiss type medical use claims and the effectiveness of skinny labels to control prescribing practices amongst European countries makes for significant uncertainty in the delicate balance between generic and innovator rights in Europe. This uncertainty is beneficial to neither side of the equation. It will be having a chilling effect on innovation towards new therapies and will be delaying the release of much needed low-cost generic treatments for established therapies.
The move to the EPC2000 style medical use format in 2010 should mean that the next generation of skinny labelling litigation will be brought across a more consistent legal framework and so hopefully lead to greater alignment in outcome.
For the EU countries, the dawn of the UPC represents a golden opportunity for the CJEU to harmonise the approach to skinny labelling and second medical use patents across much of the continent. That said, which side of the issue the UPC judges will come down on remains to be seen. Both Germany and the Netherlands are pro-innovator and Italy has recently moved in that direction, so perhaps this is the line they will take despite France’s current pro-generic position. Moreover, the case law in Germany, the Netherlands and Italy is based on a medical use claim interpretation that matches the EPC2000 format and so is perhaps more relevant than the French approach.
Of course, eventual harmonisation will require a test case to work its way through the Unified Patent Court and up to the CJEU, and so could be the best part of a decade away. Whether the UK patents courts would remain on its differing trajectory or would fall in behind the CJEU is also an open question.
More broadly, drug regulatory law could be changed to make the policing of patented repurposed drugs more effective. For instance, the situation might be improved by allowing drug labels to expressly exclude a patented use rather than simply not mentioning it. More fundamentally, and with the potential to be more effective, the law could be changed to require clinicians to prescribe by medical indication as well as by drug name. With effective IT support this would better allow pharmacists to dispense the correct product. Indeed, this is how it works in Norway.
At a bare minimum, generic companies considering entering the market with a skinny-labelled product will need to draft their carve outs diligently, making sure they consider both the originally defined indications submitted to the official medicines agencies by the innovator company, but also the wording in the patent claims in question.
Generic companies must also coordinate with their marketing and sales divisions to ensure the Company’s conduct in those areas does not contradict the messaging from the skinny label.
It would be prudent for generic companies to take reasonably comprehensive steps in advance of market entry, and on an ongoing basis, to engage with all players in the market to educate them of the patented indication and the fact that the generic product is not to be used therein. A still further measure to minimise risk is to engage with the IT systems underpinning the prescribing and dispensing of drugs in each country to ensure IT processes are in place to minimise cross label dispensing.
The present position of the European patent courts means innovator companies can be bold in protecting their European market positions. In general terms, there does seem to be growing appetite in the patent courts to entertain claims against generic companies where it can be shown that the carve out is not complete and/or where the conduct or actions of a generic company can be considered to undermine their position on the skinny label. This is not clear cut in all countries and so action is perhaps limited to situations where cross-label use is significantly eroding market share and revenue, or be limited to the most innovator friendly jurisdictions, e.g. the Netherlands or Germany.
The contributions of both generic and innovator companies to our healthcare systems are essential, and a balance must be struck between their competing interests to ensure the system works at its best. At the moment, that balance differs between countries in Europe and the resulting uncertainty serves only to undermine each party’s business and the public’s access to medicines. Hopefully, the expected alignment in medical use claim interpretation and the harmonising hand of the UPC/CJEU will eventually lead to greater consistency. More fundamental changes to the system through modifications to European pharmaceutical regulatory laws would be welcome also, but may be more difficult to achieve. In the meantime, the development of “smarter” prescribing practices could abrogate some of the difficulties caused by the present legal framework. Until such developments occur, generic companies have some guidance from the courts on how to stay the right side of the line in specific countries and the innovator companies can be bold in protecting their positions where legitimate infractions are occurring, perhaps optimising their position through forum selection.