News
T 0295/22: Apremilast decision highlights hopes and challenges for broad second medical use claims
January 2025
This case provides insight to the EPO’s interpretation of administration features in second medical use claims. Whilst the door appears to be open for broader second medical use claims, they can present inventive step challenges. Further, a particularly strict approach was adopted regarding added matter.
The Board in T 0295/22 revoked Amgen’s patent EP 2962690 directed to apremilast (Otezla), a single enantiomer inhibitor of phosphodiesterase 4 (PDE4) which is indicated for the treatment of psoriatic arthritis, psoriasis and Behcet’s disease when taken orally. The claims pursued by the patent proprietor during the appeal proceedings were found to lack inventive step over US 6020358 (D1).
D1 is a genus patent exemplifying the racemate of apremilast among the twenty compounds investigated for PDE4 inhibitory activity. D1 discusses that both the racemates of the compounds and the individual isomers – optical purity of > 95% is mentioned – are part of the invention; and that oral administration is envisaged, among other routes. None of the exemplary pharmaceutical compositions included racemic apremilast.
The patent in suit contains data showing that apremilast in 98% e.e. has superior PDE4 selectivity and aqueous solubility compared to the racemate, together with an outstanding therapeutic index when orally administered. During opposition proceedings, the patent had been maintained in amended form with claim 1 directed to stereomerically pure apremilast with greater than 94% e.e. for use as an orally administered medicament. Amgen and numerous opponents appealed and during the appeal proceedings Amgen attempted to pursue analogous claims which were broadened to greater than 60% e.e. apremilast.
Administration in second medical use claims
The Board considered whether the administration route of a drug can constitute a characterizing feature of the claimed subject-matter. If a known substance or composition is found to have a therapeutic use, it may be possible to claim the product limited by its use as a medicament as a first medical use claim. If it is then discovered that the same product may be used in a different therapeutic use, it may be possible to protect the product limited by this further “specific use” as a second medical use claim. G 2/08 confirms that the “specific use” of Article 54(5) EPC is not limited to a different disease, but may include a new dosage regimen, or patient sub-population (e.g. distinguished by pathological status). The Guidelines for Examination suggest however that a different administration route of a drug would not be patentable unless the route were related to a specific therapeutic indication. Here the Board found that, based on the conclusions in G 2/08 and contrary to the Guidelines, the feature of ‘oral administration’ does equate to a specific use within the meaning of Article 54(5) EPC without being confined to a particular medical indication, and so characterized the claims.
Accordingly, the Board decided that the claims differed from D1 by oral administration of the compound. The patentee had demonstrated that oral administration was associated with an unexpectedly high selectivity, reduced propensity for emesis and improved solubility. However, unfortunately for the patentee, the Board noted that the prior art showed that positive results had been obtained following oral administration of the PDE4 inhibitors, cilomilast and roflumilast. Furthermore, another an analogue of the claimed compound had been described as having a favourable therapeutic index versus emesis. Accordingly, while acknowledging the Board decided that it would already have been obvious for a skilled person to arrive at the claimed subject matter because advantageous effects could be expected to result from the prior art. Accordingly, the unexpected advantages were seen as “bonus effects” that could not support inventive step.
D1 did not contain any human or animal data concerning therapeutic effect, so the patentee argued that it failed to provide an enabling disclosure of the use of the compounds as a medicament. While the Board concluded (citing T 230/01) that it had not been unequivocally proven by the patentee that D1 lacks enablement for the medical use of apremilast, in the context of inventive step the Board did appear to take into account the lack of experimental data in D1, although in this case it was insufficient to tip the balance in Amgen’s favour.
Strict approach to selections in added matter
The EPO commonly adopts the two-lists principle for added matter, whereby subject-matter that derives from choosing one option from a list (one selection) is treated as being disclosed, but subject-matter that comes from choosing options from two or more independent lists (multiple selections) is considered to not be disclosed in individualised form. What constitutes a selection is of primary importance, e.g. narrowing a long list to the preferred options is generally considered not to involve a selection.
Here the Board took a relatively harsh approach in assessing what constituted a selection. Different definitions of “stereomerically pure” were provided in the description, with the only definition directly corresponding to e.e.s presented as a convergent list from the broadest (60% e.e.) to more preferred (higher) e.e.s. Commonly the EPO considers introduction of the broadest option of a convergent list from the description not to involve a selection, as its scope is the same as the whole list being used with the more preferred options present as optional features. However, in light of the plurality of definitions and the lack of a stated preference, the Board considered 60% e.e. to constitute a selection.
In taking this approach the Board sidestepped a tricky clarity issue from first instance: The OD had considered there to be a contradiction in a claim between the term “stereomerically pure” and relatively low enantiomeric purities such as 60% e.e., whilst being content that higher e.e.s were consistent with it. Where this line should be drawn is likely to be challenging to decide.
Conclusions and practice points
Although this decision provides useful precedent for pursuing second medical use claims directed to new administration routes of a known drug, it also suggests that such claims may be difficult to get granted unless there existed a known prejudice/significant obstacles in the prior art to delivering the drug via the claimed route. The decision highlights that when drafting European applications to dosage regimens it may be prudent to define the subject-matter both with and without reference to the disease, to give greatest flexibility and potentially the broadest scope of protection.
It will also be interesting to see if there will be an increase in the application of bonus effect case law following this decision. In December last year, we speculated that the “bonus effect” case law may be increasingly relied upon in the interest of aligning conclusions on inventive step more closely with those reached by the UPC[1]. From what we have seen so far, the UPC’s inventive step framework appears to be more holistic and focuses on whether a skilled person would have been motivated to consider the claimed solution and implement it as a next step in developing the prior art[2]. Although the Board in T 0295/22 did not cite the UPC as a reason for their decision, this case adds to an increasing body of recent case law in which a technical advantage was deemed insufficient to support an inventive step. That said, the chances of an opponent successfully relying on a “bonus effect” is likely to be fact dependent, as we have also seen the same Board in a different composition consider a “swallowable” tablet of deferasirox as being inventive over a dispersible form of the drug[3] (T 0526/21). In the latter case, despite acknowledging that providing a swallowable tablet to improve the ease of administration was obvious per se, the Board considered that a skilled person would not have had a reasonable expectation of success of achieving good bioavailability when formulating deferasirox as a swallowable film coated table. In T 0526/21, therefore, the bioavailability of the swallowable tablet was considered to be unexpected and indicative of an inventive step. Accordingly, despite the opponents’ successful reliance on the bonus effect in T 0295/22, we should not downplay the importance of data in demonstrating an unexpected technical effect for supporting an inventive step.
[1] https://www.hgf.com/news/with-the-epo-and-upc-adopting-different-frameworks-for-inventive-step-is-there-potential-for-conflicting-decisions/
[3] Reasons, 3.5.2 and 3.5.3
For any questions relating to the above, please contact the authors, Nick Howe, Robert Scanes, or Hsu Min Chung.