As second medical use patents become ever more popular, Nicola Dagg and Eibhlin Vardy of Allen & Overy take a look at the problems innovators face enforcing them.
The inventive use of existing medicines to treat additional indications (such as other diseases or patient groups) is of tremendous benefit to patients and health systems alike. It is an efficient use of existing resources and knowledge, because the active ingredient has often already been tested in clinical trials and used by patients. As a result, much is known about the drug’s profile. The importance of second medical use patents, which protect such new uses and provide an incentive for the considerable research and development costs involved, is widely recognised. However, innovators may face serious difficulties in enforcing them. The basic problem is that the product itself – the drug – is not new, and where a generic is on the market for other uses cost-saving mechanisms within health systems nudge medical practitioners and pharmacists towards the generic, potentially cutting the patentee out of the loop.
In Europe, when patent protection for a branded medicine expires generic companies may obtain marketing authorisation (MA) for the generic under the abridged procedure by demonstrating that their products are bioequivalent (Directive 2001/83/EC). They can exclude patented subject matter relating to a second medical use from the authorisation and their labels – a so-called “skinny label”. Similarly, the patented use will not be mentioned in the patient information leaflets; the generic’s promotional material is unlikely to mention the patented use at all. However, this does not prevent the generic being dispensed for the patented indication as both medical practitioners and pharmacists know the products are equivalent.
Medical practitioners are often encouraged to prescribe by international non-proprietary name (INN). In England, for example, the effect of guidance from various health-care bodies is that drugs should normally be prescribed by INN where possible. This is reinforced by prescribing software systems, which may be configured to warn the clinician if he or she is prescribing a more expensive drug when a cheaper alternative is available. Clinicians are audited annually by clinical commissioning groups and are likely to be asked for an explanation if they prescribe a more expensive, branded drug. Their performance targets also include the cost of prescriptions issued. An estimated 83 per cent of prescriptions are written generically, and neither the brand name nor the clinical indication is likely to be on the prescription. The process by which the pharmacist is reimbursed by the NHS also provides a substantial financial incentive to dispense the generic, because where the prescription is written by INN, the same amount is paid for a generic product as for a branded one, so the profit margin is higher where the wholesale price of the generic is less.
As a result, profits from the second medical use, which in some cases represents by the far the largest volume of the drugs dispensed, may be diverted to the generic company almost without it lifting a finger. The question arises whether in such circumstances the skinny label is enough to protect the generic from a claim of patent infringement or whether it must do more to prevent the generic being dispensed for the patented second medical use. The potential problem for the generic is, of course, that it does not control the prescribing, dispensing, or reimbursement system; so to what extent should it be held responsible for it?
The public policy benefits of second medical use patents are now clearly recognised in European patent law through the European Patent Convention (EPC) 2000. Article 54(5) EPC 2000 provides for the patentability of any substance or composition for any specific use or treatment, provided that such use is novel and inventive. This is consistent with the position in many other jurisdictions worldwide (notable exceptions include parts of Latin America and the Caribbean, along with India).
However, there is currently little European case law on how a second medical use patent is infringed. Two recent decisions have supported the idea that it is not enough for the generic to rely on its skinny label. In January 2015 the Hague Court of Appeal granted an interim injunction against SUN in relation to infringement of Novartis’s second medical use patent for zoledronic acid for osteoporosis. SUN had obtained a “skinny” MA for zoledronic acid for Paget’s disease, a disease that accounted for only about 5 per cent of the market for zoledronic acid, whereas about 95 per cent of that market was accounted for by osteoporosis, Novartis’s patented indication. SUN obtained a contract to supply a health insurer, VGZ. The Court said it was “inescapable” that zoledronic acid would be prescribed for osteoporosis; the fact that this was the consequence of VGZ’s preference policy did “not exonerate SUN” – prima facie, it was up to SUN to “ascertain that effective measures would be taken” to prevent it. Similarly, at the beginning of April 2015 the Hamburg Regional Court granted Warner-Lambert interim injunctions against five generics prohibiting them from entering into discount agreements with statutory health insurers for generic pregabalin unless Warner-Lambert’s patented indication (for neuropathic pain) was excluded, again recognising that the skinny label alone does not avoid infringement.
As with all patents the test for infringement revolves around the wording of the patent claim. Claims to second medical use under the EPC 2000 are in the format “product X for treating indication Y”, Y being the second medical use. Earlier European patents use the more esoteric “Swiss form” developed by the Enlarged Board of Appeal of the EPO (G5/83 EISAI). Many such patents still exist. It claims “use of substance X for the preparation of a medicament (or pharmaceutical composition) for treating indication Y”. Hence it can be regarded (at least for the purposes of validity) as a purpose-limited claim to a process whereas the EPC 2000 claim is a purpose-limited claim to a product (Technical Boards of Appeal of the EPO 1780/12). In each case it is the second medical purpose rather than the process or product which is novel and inventive.
Both the decisions referred to above involved Swiss form claims and were based on “contributory” infringement – the principle that supply of the generic medicine amounted to the supply of an essential element of the invention. In the circumstances the generic companies’ knowledge that the generic would be dispensed for the patented use was, rightly in our view, a sufficient basis for infringement in injunction proceedings. The generic could not hide behind the “skinny” label in circumstances where everybody knows that the generic will be prescribed.
“Direct” infringement is also relevant. In the case of a Swiss form (process) claim direct infringement arises where a person disposes of any “product obtained directly by means of the [claimed] process” (UK Patents Act 1977 section 60(1)(c)), whereas for EPC 2000 (product) claims infringement arises under section 60(1)(a), more simply, where a person “disposes of the product” (ie, the product “for” the second medical use). Whether a second medical use claim is directly infringed hangs on the interpretation of the wording of the patent claim, particularly the meaning of “for” in “for treating indication Y”, which appears in both types of claim. So the basic question is in what circumstances the medicine will be regarded as being “for” treating the patented indication. The generic may argue that, based on its “skinny” label, the product is only “for” the unpatented use. But is this really tenable in circumstances where everyone knows that it will in fact be dispensed “for” the patented use? This issue was considered in an action by Warner-Lambert in the English High Court for a preliminary injunction against Actavis based on its second medical use patent for pregabalin for neuropathic pain (Lyrica) in January 2015 in circumstances where the major commercial market for pregabalin was for pain. Judge Arnold J’s approach was that “subjective intent” by the manufacturer was required and could not be inferred from the generic’s knowledge that in practice the drug would be dispensed for the patented use. Some sort of explicit targeting of the patented indication by the generic seemed to be necessary. In our view such an approach is not required by the wording of the claims and would make it practically impossible to enforce second medical use patents, enabling the generic to hide behind the “skinny” label while reaping the profits from the patented use, flying in the face of the public interest in the development of second medical uses. At the time of writing this interim decision is subject to appeal.
So what “more” should the generic company be required to do to avoid infringement? In some cases it may be able to clear a path for itself before the courts by means of a timely revocation action. If not, as Arnold J recognised in the Warner-Lambert case referred to above, it should contact the innovator at an early stage asking it to acknowledge non-infringement and, if this is refused, seek a declaration of non-infringement in order to determine the infringement issue promptly. Infringement may be avoided if the generic company takes sufficient steps to prevent the generic being prescribed for the second medical use. This may include: obtaining changes to the relevant clinical commissioning software systems and prescribing software so that they only permit prescribing by generic/INN name for the specific active ingredient for non-patented indications; a modest “design around” with a removable label/sticker specifying that the drug is not to be dispensed for the patented indication; putting an obligation on the pharmacist to make a reasonable check on the purpose for which it was prescribed; and imposing contractual restrictions on pharmacist customers, for example limiting the volumes of the generic that may be dispensed in a particular market to volumes reflecting realistic non-patented use. The generic should also approach the relevant health authorities in good time to discuss how the reimbursement categories under the NHS Drugs Tariff, which determine how pharmacists will be reimbursed when dispensing the medicines, can be dealt with.
The underlying problem is, of course, that neither party has much direct influence over the prescribers. A creative solution was adopted by the English High Court in the Warner-Lambert case referred to above. On application by Warner-Lambert it made an interim order against NHS England requiring it to provide guidance to NHS service providers. The guidance basically provides that medical practitioners should prescribe only by reference to the brand name for the patented indication and by reference to the generic name for non-patented ones, and where they have been told it is for the patented indication, pharmacists should only dispense the branded product. NHS health providers were also required to consider amending the operation of electronic prescription systems to warn prescribers only to prescribe the branded medicine for the patented indication. This was agreed to be the “best solution to the problem”, though it has yet to be seen how effective it will actually be in practice.
The order was granted pursuant to the Court’s jurisdiction under S.37(1) Senior Courts Act 1981 as developed by Arnold J in Cartier v British Sky Broadcasting (2014, English High Court), which related to blocking orders against internet service providers. The Court has broad jurisdiction to grant an injunction against a third party which is not itself a wrongdoer but is “mixed up in wrongdoing”. Here NHS England was in a unique position to assist based on its central role in giving guidance within the NHS generally, and it was also a very cost-effective solution. At the time of writing it is an interim order granted broadly on the basis that infringement of the patent is arguable at this stage. A permanent order of this sort will require a finding of infringement.
Mr Justice Arnold commented that such cases involve “collision between the policy of incentivising important medical research by granting second medical use patents… and other policies and practices which form part of the United Kingdom’s healthcare systems”. The question arises as to whether patent litigation is the right framework in which to address such issues of policy. On the face of it there is much to be said for a regulatory solution which could, for example, dispense with the need to argue about the state of mind of the generic, which ought to be irrelevant in a highly regulated health environment. Any Europe-wide regulatory solution is, however, unrealistic in the short to medium term, and is likely to have its own complexities, involving changes to the prescription, dispensing and reimbursement processes in health-care systems, which vary considerably between countries. It remains to be seen whether there is a sufficient will among governments to achieve this.