Saturday, 23 February 2013

T 713/08 – Too Select

The present examination appeal decision is not surprising at all, but it nicely demonstrates the – rather strict – way in which A 123(2) is applied by the EPO nowadays.

Claims 1, 12, 13, 26, and 27 as filed read (my emphasis):
1. A method for modulating the activity of at least one neurotrophin and/or a pro-neurotrophin in a cell or an organism, such as an animal, comprising administering to said animal a sufficient amount of an agent capable of
(i) binding to a receptor of the Vpsl0p-domain receptor family and/or
(ii) interfering with binding between a receptor of the Vpsl0p-domain receptor family and a neurotrophin and/or proneurotrophin and/or
(iii) modulating the expression of a receptor of the Vpsl0p-domain receptor family.

12. The method according to any of the preceding claims, wherein the agent is selected from proteins, peptides, polypeptides, antibodies, antisense RNA, antisense-DNA or organic molecules, SiRNA.

13. The method according to any of the preceding claims, wherein the agent is capable of inhibiting binding of said neurotrophin or said pro-neurotrophin to the receptor.

26. A method for treating a disease or disorder in an individual, comprising administering to said individual a sufficient amount of an agent as defined in any of the claims 1-25.

27. The method according to claim 26, wherein the disease or disorder is selected from one or more of the following diseases or disorders: inflammatory pain, diseases or disorders of pancreas, kidney disorders, lung disorders, cardiovascular disorders, various types of tumours, psychiatric disorders or neuronal disorders.
Claim 1 of the main request before the Board read:
Use of an agent capable of inhibiting the binding of a proneurotrophin to a Vpsl0p-domain receptor by binding to an extracellular part of the receptor, an intracellular part of the receptor, or a segment of the transmembrane part of the receptor in the manufacture of a medicament for treating and/or preventing neurological diseases and disorders in an animal.
The Board found this claim not to comply with the requirements of A 123(2):

[7] In accordance with established case law of the Boards of Appeal, the relevant question to be decided in assessing whether an amendment adds subject-matter extending beyond the content of the application as filed is whether the proposed amendment is directly and unambiguously derivable from the patent application as filed.

[8] In a primary line of argumentation the appellant has referred to a combination of various claims as originally filed which when taken in combination supported claim 1. However, the board notes that when combining the various embodiments of the claims […] it transpires that the specific subject-matter of claim 1 can only be derived therefrom when making a variety of specific selections from the lists of embodiments which are the subject of these claims. By way of example the board refers to claim 12 as filed in which antibodies are listed among other candidate inhibitory binding compounds; to claim 13 as filed, in which a choice has to be made between pro-neurotrophin and neurotrophin and to claim 27 as filed where neuronal diseases are merely one of the numerous diseases listed to be treated. 

[9] Accordingly, the claims as originally filed cannot provide a basis for claim 1 to comply with the requirements of A 123(2).

As the Board found that the description did not provide such a basis either, the request was dismissed.

Should you wish to download the whole decision, just click here.

The file wrapper can be found here.


Anonymous said...

Interestingly, the Board could not have used the same logic had the claims been drafted as, e.g.:

12. The method according to any of the preceding claims, wherein the agent is a protein.
12a. The method according to any of the preceding claims wherein the agent is a peptide.
12b. The method according to any of the preceding claims, wherein the agent is a polypeptide.

such that all possible embodiments were claimed as a unique combination of dependent claims, without any "selections".

However, Applicants are encouraged by the EPO claims fee structure to combine alternatives within a single claim, rather than have distinct claims to each variant.

Should we now start including explicit "embodiments" text in the Description of new specification, reciting explicitly each combination selected from alternatives in the claims? It would make specifications longer and less readable, but would avoid brutal application of the two-lists principle.

MaxDrei said...

I'm not a chemist, so I look in on the field of "selection inventions" as a quizzical outsider.

But it seems to me quite simple. If you write in the app as filed that you can choose from A to Z and from 1 to 99 to solve the problem, and any combination from your two lists is as good as any other selection, that's one thing. Quite different is to say to the reader later on, during prosecution and after the patent issues, that you should select a particular member of the A to Z list (say "X") because X delivers an effect that A-W, Y and Z do not.

That difference in teaching the skilled reader is where the offence against Art 123(2) lies, does it not? To allow such amendments is to destroy the notion that the First to File prevails over another fellow who was in reality second to file on X. By selecting X in a prosecution amendment you can steal from the other fellow the legal status of who was "First to File" on Selection X.

So is the EPO on Art 123(2) "brutal"? No way. Comments please.

Myshkin said...

I agree with the anonymous commenter that decisions like this invite applicants to include ridiculously long automatically generated lists of different combinations of features into the description. For example 26 x 99 = 2574 automatically generated embodiments added to the hypothetical application discussed by MaxDrei, allowing the applicant to amend to any combination of A-Z and 1-99.

In my view the solution is to be relatively relaxed on Art. 123(2), i.e. allow the applicant to make multiple selections, but strict on Art. 56 when examining the resulting combinations. If the description fails to mention any effect arising out of the particular combination claimed, the two chosen features should be examined separately.

So using MaxDrei's example, if none of A-Z and none of 1-99 is inventive on its own, then a combination P-65 can be claimed by making two selections, but can only be inventive if the description specifically mentions that this combination achieves a particular technical effect.

(A problem of this approach might be that some prior art documents in chemistry or so might now be too novelty destroying.)

Btw, why do Boards never object to claims 1-10, each claim depending on any of the previous claims? The applicant is now seeking protection inter alia for all combinations of claims 1, p and q with 2 <= p < q <= 10. Any choice of p and q is a selection from two lists 2-10, so undisclosed. Seeking protection for combinations not even disclosed should at least be a clarity problem.

MaxDrei said...

Very good observations from Myshkin. Re-reading my earlier comment, i see that it confuses the distinction between Art 54, Art 56 and Art 123(2). We need to be scrupulous about the distinction between what one can derive (directly and unambiguously) from the disclosure, what is obvious from the disclosure, and whether the disclosure that the inventor relies upon for satisfying Art 56 EPC is indeed present in the application as filed.

I really dislike the idea of having to recite all 2574 permutations explicitly. Any law that encourages such drafting should be got rid of toot sweet. Myshkin gives us a better way.