Up in the air again with SPCs

26 July 2017

The past nine months have seen a new bout of referrals to the Court of Justice of the EU (CJEU) and other important national judgments concerning Supplementary Protection Certificates (SPCs), the European system for providing additional patent term to reflect delays caused by having to seek regulatory authorisation for medicinal and plant protection products.  The referrals reflect the continuing difficulty in being able to pin down exactly the circumstances in which a patent holder should be able to obtain an SPC, how long the additional protection should last, and aspects relating to enforcement.  In addition to these referrals, there has been a judgment in Norway that illustrates the hurdles in applying SPC law to biologics.

In one referral, Teva UK Ltd & Ors v Gilead Sciences Inc [2017] EWHC 13 (Pat), the UK court has asked the CJEU yet again to answer the question of what “protected by the basic patent” means in practice.  The CJEU has provided multiple answers previously but still there are areas of uncertainty.  This is at least in part because too much may be being asked of this requirement in determining when and for how long there should be the reward of additional term.

Thus, if a party has an SPC for a product with active ingredient "A", under what circumstances should they be able to get an SPC for combination product "A+B", and how is that implemented?  Is an SPC to "A+B" justified based on a patent claim of the form "compound A and a further pharmaceutically active compound" (as is the situation in the present case); or "compound A and a compound having activity Y"; or "compound A and compound B"; or some other permutation?  And with what rationale? 

In referring the question, Arnold J suggests that not only must the active ingredient(s) fall within the scope of at least one claim, but they should also constitute the core inventive advance of the patent.  Such would seem to make the assessment of what is "protected by the basic patent" a matter of substantive examination, essentially involving an inventive step test.

Arnold J used this approach in a further recent case, Sandoz Ltd & Anor v G.D. Searle LLC & Anor [2017] EWHC 987 (Pat), which centred on the question of whether or not a 'Markush' chemical formula (a skeleton formula using letters to represent lists of possible substituents at various positions) is sufficient for an active ingredient to be "protected by the basic patent".  The active ingredient of the SPC, darunavir, was not explicitly disclosed in the patent but, nevertheless, Arnold J ruled that darunavir was identified by the Markush structural formula, regardless of its breadth, and the ability of darunavir to function as an HIV protease inhibitor embodied the inventive advance of the patent.

The approach was also applied to the question of whether "A+B" can be the subject of an SPC if there is already an SPC to "A" in Teva UK Ltd & Ors v Merck Sharp & Dohme Corporation [2017] EWHC 539 (Pat).  Although the SPC was declared invalid for other reasons, Arnold J concluded that an SPC should be available for combination products if the combination ("A+B") is inventive over the single product ("A").
One implication of this approach is that, the more closely the system is required to reflect "merit", the more complex and unpredictable it is likely to become in practice.  Is this acceptable, or should a simpler system be adopted, even if it might close the door to a further SPC in some circumstances?

In a further referral in Abraxis Bioscience LLC v The Comptroller-General of Patents [2017] EWHC 14 (Pat), Arnold J asks whether a further SPC may be available for a new formulation of an active ingredient that already has an earlier SPC based on an earlier marketing authorisation (MA).  In the Neurim case (C-130/11, ECLI:EU:C:2012:489), the CJEU ruled that it is permissible for a further SPC to be granted to cover a 'different application' of a previously authorised product (at least when the earlier MA is for a veterinary product and the later MA is for a human medicinal product, with different indications).  However, it is unclear just how far Neurim can be extrapolated to other scenarios, and how broadly 'different application' can be interpreted beyond therapeutic use.
Arnold J suggested that an SPC in this particular case for a new formulation should not be granted.  However, whether the answer from the CJEU is yes or no, the reach of Neurim could still be unclear depending on how far the CJEU chooses to opine.  For example, if a new formulation were refused on the basis that an earlier MA exists, would this preclude an SPC for a patented invention that is neither a new therapeutic use nor a new formulation, such as a new dosage regime? Thus, what would constitute a new 'formulation'?  Is it simply the presence of specified components, or can it comprise new ratios of components or a different dosage regime?  Whatever the nature of the CJEU's judgment, it is possible that the answer will bring additional such questions.
In Merck Sharp & Dohme Corporation v The Comptroller-General of Patents, Designs and Trade Marks [2016] EWHC 1896 (Pat), Arnold J referred a question about whether an End of Procedure (EoP) Notice – generated as part of the 'decentralised procedure' for obtaining an MA – can be used instead of an MA when applying for an SPC, and if not, whether the MA can be validly submitted later to remedy the application.  These questions have been answered differently by national patent offices: as at the date of the referring decision, three had refused an SPC, four had granted an SPC and a further state had refused an SPC on other grounds.  The importance of this decision for the Applicant lies in the fact that the EoP Notice was issued only three days before the expiration of the patent, and waiting for the MAs to issue would have made it too late to apply for an SPC.

Arnold J considered that an EoP Notice is not equivalent to an MA, for several reasons, including that: the SPC Regulation never refers to EoP Notices, only to MAs; an EoP Notice has no legal effect; an MA is required before a product can be placed on the market; and each member state still must decide whether to grant an MA after issuance of an EoP Notice and may have good reason for not doing so.  Arnold J also considered that it is not possible to correct an application by submitting the later-issued MA when it didn't exist "at the date of that application" (SPC Regulation, Art. 3).  Arnold J seemed clear in his decision, so his choice to refer these questions to the CJEU may reflect a desire for equitability and consistency between territories, seeking an authoritative and harmonising ruling that will have effect on all corresponding SPC applications.

Transitional provisions and the 'specific mechanism' are the subject of a further referral (Pfizer Ireland Pharmaceuticals, Operations Support Group v Orifarm GmbH, Case C-681/16).  The ‘specific mechanism’, incorporated into the Acts of Accession of new EU member states (‘Accession states’) since 2003, provides a derogation from the EU’s principle of free movement of goods, allowing holders of patents or SPCs to stop parallel imports of patented products from Accession states, providing that the patent or SPC was filed at a time when protection could not be obtained in the Accession state for that product.  Such provisions provide a taste of what might be to come in attempting the opposite task of unravelling UK SPCs from the EU context.
In the referral, there are a series of questions (which do not appear to be as clear as they could be), regarding relative timings of patent filing and publication, SPC filing and paediatric extension availability.  The first question relates to the previous inability to obtain an SPC in an Accession territory by virtue of a lack of a basic patent in that territory, but does not make it clear whether this is a matter of non-availability of the appropriate patent protection, or a failure of the party to seek protection in the Accession territory, which seem to be very different scenarios demanding different answers.  The second question appears to imply that the scenario relates to the non-availability of a patent, and essentially asks whether the party should have filed a relevant patent application (once possible) at a time when the corresponding German (for example) patent application had not been published.  Such a scenario appears to be contrary to the express wording of the 'specific mechanism' provisions (which refers to the filing date), and also appears illogical and inequitable, as the prior art landscape may have changed considerably in the intervening months, even if the corresponding patent application had not yet been published.  It seems unlikely that the CJEU will absolve patentees of the need for forward thinking, preferably with a crystal ball, in terms of pursuing patent protection when it is available at the filing date of the patent/SPC in question, but it would seem incompatible with the purpose of the 'specific mechanism' to expect them to devise a patenting strategy based on a different filing date and hence different prior art landscape.
How long?

There is another referral (Incyte Corporation v Szellemi Tulajdon Nemzeti Hivatala, Case C-492/16) about patent offices refusing to implement a change in SPC term arising from the Seattle Genetics (Case C-471/14) decision that concluded that the relevant date for calculating SPC term is the MA notification date, not the decision date.  Some patent offices are being slow to implement this.  It appears possible that there will be a decision deeming the date to have been corrected automatically, or subject to application from the SPC holder (so that the notification date can be captured).
Enforceable scope
Finally, the Norwegian courts have put into effect the decision of the Court of Justice of the European Free Trade Association States (EFTA Court) concerning the breadth of SPC scope relative to the product approved in the relevant MA.  One of the Norwegian Courts of Appeal concluded that an apparent difference in efficacy meant that a vaccine could not be considered to be the same product as that authorised, with fatal consequences for the SPC.  This decision was later upheld by the Norwegian Supreme Court.  Such an approach appears to lead to very considerable uncertainty and lack of clarity, as SPC validity could be subject to future unpredictable developments.  But on a more positive note, amino acid sequence identity is not required for the authorised entity and a further entity to potentially be considered the same product.  It appears that a more equitable approach may be for the SPC scope not to extend to a product that has significantly different properties, for example different efficacy; rather than for the SPC to be incurably invalid if the wording of the SPC product encompasses a product that has significantly different properties.  No doubt this is an issue that will be considered further. 

As ever, SPCs continue to pose difficult questions, with few concrete answers.  For current thinking and advice in this ever-changing field please contact our SPC team.

For more information, please contact your normal Potter Clarkson representative.

Back to all updates