The challenge of patenting combination claims at the EPO

A patent relating to combination claims is the subject of the latest referral to the CJEU on the interpretation of the SPC regulation (IPKat). Even before you get to the question of whether SPCs are available for combination therapies, demonstrating the patentably per se of combination therapies is a challenge in and of itself. As recent Boards of Appeal case law confirms, a claimed therapeutic combination must not have been the subject of accidental anticipation (T 2056/17), must show more than a mere additive technical effect, and must not have been "obvious-to-try" out of a number of possible alternatives (T 2168/17). 

Legal Background - Combination claims  

The ongoing questions of whether combination SPCs are precluded by one or more sub-provisions of the SPC Regulation reflects the high development cost and value of combination therapies. Combination claims also face a high bar patentability bar in Europe. In particular, in order to embody inventive character, a combination invention must not merely result in an aggregation of the features of the underlying components (Case Law of the Boards of Appeal, I-D-9.2.1). In practice, this requires the patent applicant to demonstrate that the combined features produce a synergistic as opposed to mere additive effect. Whether or not the applicant can rely on post-published evidence to support a synergistic effect is currently the subject of a pending referral to the Enlarged Board of Appeal (EBA) (G2/12, IPKat). In the absence of an unexpected synergistic effect, combinations of known drugs for use in therapy are often considered obvious-to-try by the EP patent Examiners.  

Combined therapeutic-effect destroys the novelty of a combination claim, regardless of regulatory status (T 2056/17)

In the recent appeal case T 2056/17 the patent (EP2695618) in question related to pharmaceutical composition for use in the treatment of epilepsy, comprising lacosamide (LCM) and levetiracetam (LVT). The granted claim specified that the LCM and LVT are provided in separate dosage forms, and administered simultaneously and/or subsequently.  The prior art cited against the patent (D2, Jatuzis et al.) was a report of clinical trials evaluating the effect of LCM add-on therapy to baseline treatment with antiepileptic drugs (AEDs) including LVT.  

The Board of Appeal found that, given that the AEDs were provided as a baseline treatment, the dose of the AEDs (including LVT) had to have been sufficient to provide a treatment effect. Furthermore, as LCM and the AEDs were administered together during part of the trial, the drugs could be assumed to have been present together at therapeutic levels in the patients. However, the patentee argued that D2 only disclosed the use of LCM as an adjunctive therapy, i.e. a therapy used to assist the primary therapy of AEDs. Particularly, the patentee pointed out that "adjunctive and combination therapy have different meanings in the context of marketing authorisations: they implied different clinical set-ups and required different experimental evidence. Therefore, D2 did not allow the reader to draw any conclusion on a combination therapy".

Cat-Owl combination

The Board of Appeal did not agree with the patentee's argument. Particularly, the Board could find no basis for interpreting the feature "pharmaceutical combination" narrowly or restricted to the meaning used in the specific context of marketing authorisations. The Board concluded that the design of the clinical trials in D2 implied the combined effect of LCM and LVT. The combination claim was therefore found to lack novelty in view of the report of the previous clinical trial. 

The inventiveness of combination claims (T 2168/17)

In the appeal case T 2168/17, the patent related to a combination of an anti-BLyS antibody and an anti-CD20 antibody (e.g. rituximab) for use in the treatment of lupus (e.g. SLE). The closest prior art was identified as a document disclosing the treatment of SLE with rituximab. Particularly, the document provided data showing that rituximab reduced peripheral B cells in SLE patients, where it is known that SLE pathology is partly associated with peripheral B cell levels. The prior art did not disclose the use of rituximab in combination with an anti-BLyS antibody (which also targets B cells). The data provided in the application as filed demonstrated that the use of either an anti-BlyS antibody or rituximab alone depleted B cells, but concluded that rituximab monotherapy might not be adequate by itself to treat SLE. The prior art therefore suggested combining rituximab with, among other things, an agent that blocks the BLyS system. 

The patentee argued that in view of the number of different options that the skilled person could try in a potential rituximab combination therapy, it would not have been obvious to try an anti-BlyS antibody in particular. The Board of Appeal did not agree, and found that there was no reason in the prior art for the skilled person to have rejected the use of an anti-Blys antibody as one of a number of the obvious alternatives (compare the recent UK High Court decision in Bayer v Teva on whether the selection of a formulation salt was obvious to try, IPKat). The Board of Appeal also considered whether it would have been obvious to a skilled person that the claimed combination would have a particularly beneficial effect. However, inventive step based on an unexpected technical effect was rejected in view of the known mechanism of action of anti-Blys antibodies. The combination claim was therefore found to lack inventive step in view of the prior art. 

Final thoughts

Neither of the decisions in T 2056/17 or T 2168/17 are particularly surprising given the facts of the cases, and the case law on combination claims. These cases are a reminder that, given that there will usually be mechanistic justification for combining two drugs based on the underlying science, demonstrating the patentability of a combination therapy can be an uphill battle.