A Swiss approach to the EPO’s gold standard

Reading time: 5 minutes

Case No. O2016_016 | Decision of 6 June 2019

The patent at stake is MundiPharma‘s EP 2 425 825 B9; see EPO Register and Swissreg. EP’825 pertains to a combination of oxycodone hydrochloride (in an amount of 10 to 150g) and naloxone hydrochloride (in an amount of 1 to 50g), in a weight ratio of 2:1.

This combination is an analgesic drug that is available as modified-release tablets under the trade name Targin® by MundiPharma.

Develco’s logo

Develco sought annulment of EP(CH)’825 for various reasons, i.e. undue extension of subject-matter, non-enablement and invalidity of the priority claim. Please see this Blog here for a brief summary of the main hearing in this matter, and some relevant prior art Develco has asserted against patentability of the claimed subject-matter.

MundiPharma’s logo

Defendant / patentee MundiPharma countersued for infringement. MundiPharma not only countersued the plaintiff, but rather also attempted to involve the German branch of Develco as well as Konapharma. This attempt failed.

It was evident from the pleadings at the main hearing that the expert opinion of the judge-rapporteur had held that the patent was invalid for undue extension of subject-matter. Thus, it does not come as a big surprise in the decision that EP(CH)’825 is held invalid for undue extension of subject-matter. Consequently, the decision does not address the other grounds of invalidity asserted by Develco; and the counterclaim for infringement is moot, too.

In my perception, the detailed assessment of undue extension of subject-matter is nothing out of the ordinary. It’s all about the EPO’s ‘gold standard’ (G 2/10 in ¶4.3, with reference to G 3/89 and G 11/91), i.e.

[…] any amendment […] can […] only be made within the limits of what a skilled person would derive directly and unambiguously, using common general knowledge, and seen objectively and relative to the date of filing, from the whole of [the application as filed].

What is interesting, though, is how the decision addresses the so-called essentiallity-test (‘Wesentlichkeits-Test’), in particular how this test fits with the ‘gold standard’. In brief, in T 331/87 the Board held that the replacement or removal of a feature from a claim may (referred to in the Case Law of the Boards of Appeal as a ‘might’) not be in breach of Art. 123(2) EPC if the skilled person would directly and unambiguously recognise that

  1. the feature was not explained as essential in the disclosure;
  2. it was not, as such, indispensable for the function of the invention in the light of the technical problem it served to solve; and
  3. the replacement or removal required no real modification of other features to compensate for the change.

It is important to note that even this Board apparently (in view of the conditional language used; see mark-up above) did not consider compliance with the above three criteria as a sufficient condition for compliance with Art.123(2) EPC in any given case. Accordingly, the present decision holds that the only thing that really matters is the ‘gold standard’ discussed above, which cannot be deviated from. With reference to the Swiss Supreme Court decision 4A_109/2011, 4A_111/2011 (¶4.3.1), however, the decision further holds that in a case where not even the criteria of the essentiallity test are fulfilled, it must be assumed that an undue extension of subject-matter occured.

What suprises me, though, is that the decision holds (¶31):

Otherwise this would again amount to a materiality test, which was expressly rejected in G 2/98.

Frankly, I don’t see that in G 2/98, at least not expressly (‘ausdrücklich’). One may well conclude from the gist of G 2/98, in particular ¶8.3, that it does not endorse a distinction of technical features which are related to the function and effect of the invention and technical features which are not. But there is no discussion of the essentiallity test at all in G 2/98. Else, if there was, the essentiallity test would surely not be referred to anymore in the most recent edition of the Case Law of the Boards of Appeal and the Guidelines.

But still, my personal take-away message from this decision is that the essentiallity test is of no avail in proceedings at the FPC (until someone comes along and proves me wrong).

The decision is not yet final / may still be appealed to the Supreme Court.

Reported by Martin WILMING


Case No. O2016_016 | Decision of 6 June 2019

Develco Pharma Schweiz AG
MundiPharma Medical Company

Panel of Judges:

  • Dr. Daniel M. ALDER
  • Dr. Tobias BREMI
  • Dr. Hannes SPILLMANN


  • Dr. Tobias BREMI

Court Clerk:


Representative(s) of Develco:

  • Andrea MONDINI (TIMES)
  • Dr. Cornelia HOFFMANN (SBMP), assisting in patent matters

Representative(s) of Mundipharma:

  • Dr. Simon HOLZER (MLL)
  • Dr. Dirk BÜHLER (Maiwald), assisting in patent matters
  • Dr. Andreas LEDL (Maiwald), assisting in patent matters


Don’t be misled by the patent no. on the bottom of the title page. EP 2 245 825 is a typo; this should read EP 2 425 825.
Fullscreen view (new tab)


Fullscreen view (new tab)


Enter your name and email address below to get notified of new posts by email.

Print Friendly, PDF & Email

One Reply to “A Swiss approach to the EPO’s gold standard”

  1. Although the wording “essentiality test” has been deleted from H-V, 3.1, it is still present in the Guidelines.

    The essentiality test has been heavily criticized in a series of decisions starting with T 2311/10. I can also quote T 2282/10 or T 1118/10.

    In T 2311/10, the “problem” to be taken into account in step 2) is the problem as defined originally in the application and not the “objective problem” when it comes to asses inventive step. This is correct since should the latter be taken into account it would made the assessment of Art 123(2) dependent from the state of the art, which is not acceptable. The Board also expressed considerable doubts if this criteria is compatible with the definition of the “same invention“ in G 2/98 and G 1/03 and G 2/03. The Board also considered that in the third criterion, the notion of “real modification” is anything but clear.

    In T 1852/13 the Board criticised again the “essentiality test” and reminded of decisions doing the same, like T 910/03 which linked his considerations with G 2/98 (the same invention when it comes to acknowledging priority), T 681/08 (point 2 of the reasons), T 1180/08 (point 2), T 1003/09 (point 4.1), T 2146/09 (point 11), T 2311/10 (point 2.9) and T 2561/11 (point 8.1)

    Independently of T 910/03, other Boards have also reached the conclusion drawn in T 910/03, for example T 1202/03 (point 2.2), T 765/07 (point 1.4), T 1643/10 (point 2.22) and T 755/12 (point 2.3) or T 46/15 (point 2.8.1).

    T 910/03 and the subsequent decisions questioning the essentiality test did not in fact lead to the test being abandoned as in T 1852/13 the Board of Appeal was aware of more than twenty decisions taken by the Boards of Appeal between 2006 and 2012, in which the test was used.

    But the “essentiality test” does not yet appear to be dead as in T 2100/14, the Board came to the same conclusion as the examining division which had applied the three point essentiality test in order to conclude that the deletion of two features contravened Art 123(2), see Point 1.2.

    As long as the three point test is in the Guidelines, examining and opposition divisions can and will use it. It is not known if and when the “three point test” will disappear from the Guidelines.

    I still consider it a good tool when deciding if one is in the presence of an intermediate generalisation. If for instance 2 features, DE, are inextricably linked, then the three point test is a useful tool. If D can be used independently of E than separating them is allowable.

    It is a tool, and as such it should not be overrated, and the gold standard has to be respected in the end. Not inextricably linked=gold standard.

    It might sound strange, but if I it is possible to decide that D and E have no features in common, i.e. would be not unitary, then they can be dealt with separately and there is no added matter.

    Just another tool!

Leave a Reply

Your email address will not be published.