T 2160/12 (Peptide binding to MHC class-I/IMMATICS BIOTECHNOLOGIES) 18-05-2017

Main request

Amendments (Article 123(2) EPC)

1. In the following, the references are to passages and claims of the application as filed.

2. The subject-matter of claim 1 is derivable from claims 1 and 2, in combination with page 30, second and third paragraphs of the application.

3. The subject-matter of claims 2 to 4 and 6 to 12 is disclosed in claims 4, 6 to 8, 11, 14 and 19 to 22, respectively.

4. The subject-matter of claim 5 is derivable from claim 9 in combination with page 34, second and third paragraphs of the application.

5. The subject-matter of claim 7 is derivable from claim 7 in combination with page 16, last paragraph.

6. Therefore the board is satisfied that the subject-matter of claims 1 to 12 of the main request meets the requirements of Article 123(2) EPC.

Clarity and support (Article 84 EPC)

7. Present claims 10 to 12 are purpose-limited product claims, like claim 9 on which these claims depend. In the board's view, since all the claims now belong to the same claim category, the clarity objection raised by the examining division in the impugned decision has become moot. The examining division did not raise any further objections under lack of clarity and support against claims 1 to 12 and the board too has none.

8. Accordingly, the claims of the main request meet the requirements of Article 84 EPC.

Introduction to the invention

9. The invention concerns the peptide "C20-001" which is characterised by the amino acid sequence "ALSNLEVTL" (SEQ ID NO. 1) and peptides having a length of 9 to 16 amino acids comprising this sequence. This peptide is derived from the tumour-associated antigen "chromosome 20 open reading frame 42" (C20orf42), which is a focal adhesion protein and involved in the attachment of the actin cytoskeleton to the plasma membrane or in integrin-mediated cellular processes. Moreover, C20orf42 is over-expressed in the majority of colon and lung carcinomas (see Table 1, line 1 on page 9 and page 10, second paragraph, page 11, lines 1 to 3 of the application).

10. The peptide C20-001 binds to major histocompatibility complex (MHC) class-I molecules, which are exposed on the surface of most cells having a nucleus (see page 1, last paragraph, Figure 4 and legend thereto on page 7, third paragraph of the application). The complex of MHC class-I and bound peptide is recognised by cytotoxic T cells expressing on their surface CD8 co-receptors and T-cell receptors (TCRs); the former bind to MHC class-I while only the latter bind to MHC class-I-peptide-complexes. The binding of both receptors to the same MHC class-I-peptide complex activates the cytotoxic T cells, which then lyse the cells presenting the complex, for example, tumour cells (see page 2, first and last paragraphs to page 3, first paragraph of the application).

Sufficiency of disclosure (Article 83 EPC)

11. Claim 1 is directed to peptides comprising a sequence of SEQ ID NO. 1 or retro-inverso peptidomimetics thereof having an overall length of between 9 and 16 amino acids, which are further characterised by the functional feature "which induces T cells cross-reacting with said peptide".

12. Accordingly, peptides comprising a sequence of SEQ ID NO. 1 having a length of 11 to 16 amino acids and retro-inverso peptidomimetics of SEQ ID NO. 1 having a length of 9 to 16 amino acids which all induce cross-reactive T cells are embodiments of claim 1 and will be considered in the following.

13. It is established case law of the boards of appeal that a claimed invention must be disclosed in the application as a whole in a manner sufficiently clear and complete for it to be carried out by the skilled person, taking also into account his common general knowledge. Furthermore, when a technical effect is a feature of a claim, whether or not this effect is achieved by substantially all embodiments of the claim is a question of sufficiency of disclosure (see Case Law of the Boards of Appeal of the EPO, 8th edition 2016 (hereinafter "CLBA"), II.C, II.C.4.4).

14. In the impugned decision the examining division essentially pointed out two deficiencies with regard to insufficiency of disclosure of the invention defined in claim 1, namely (i) the inability of peptides or retro-inverso peptidomimetics thereof which are longer than 10 amino acids to bind to MHC class-I molecules and (ii) the inability of retro-inverso peptidomimetics in general, i.e. independent of their length and amino acid sequence, to bind to MHC class-I molecules. It was common ground between the examining division and the appellant that those peptides claimed which had a length of either 9 or 10 amino acids did bind to MHC class-I molecules and that this was a prerequisite for the induction of cross-reactive T cells.

15. It has therefore to be assessed whether the application discloses evidence that the peptides or retro-inverso peptidomimetics under consideration (see point 12 above) bind to MHC class-I molecules or whether the skilled person reading the application in the light of his common general knowledge would consider that these peptides or retro-inverso peptidomimetics were suitable for binding to MHC class-I molecules.

16. The application does not explicitly disclose that the peptides or retro-inverso peptidomimetics under consideration bind to MHC class-I molecules. It reports on page 3, third paragraph that "For a peptide to trigger (elicit) a cellular immune response, it must bind to an MHC-molecule. This process is dependent on the allele of the MHC-molecule and specific polymorphisms of the amino acid sequence of the peptide. MHC-class-I-binding peptides are usually 8-10 amino acid residues in length and usually contain two conserved residues ("anchors") in their sequence that interact with the corresponding binding groove of the MHC-molecule. In this way each MHC allele has a "binding motif" determining which peptides can bind specifically to the binding groove" (emphasis added).

17. The application further discloses that the peptide "C20-001" characterised by the amino acid sequence "ALSNLEVTL" (SEQ ID NO. 1) consisting of 9 amino acids binds to MHC class-I molecules (see Table 1, Figure 4 and legend thereto on page 7, third paragraph).

18. Document D19 is a standard textbook in the field of immunology and therefore represents the common general knowledge of the skilled person at the filing date of the application, as was not contested by the examining division. It discloses on page 129, second paragraph that "Peptides that bind to MHC class-I molecules are usually 8-10 amino acids long. Longer peptides are thought to be able to bind, particularly if they can bind at their carboxy terminus, but are subsequently cleaved by exopeptidases present in the endoplasmic reticulum, which is where MHC class I molecules bind peptides. The peptide lies in an elongated conformation along the cleft; variations in peptide length seem to be accommodated, in most cases, by a kinking in the peptide backbone. However, two examples of MHC class I molecules in which the peptide is able to extend out of the cleft at the carboxy terminus suggest that some length variation can also be accommodated in this way" (emphasis added).

19. In the board's view, the skilled person, reading that "MHC-class-I-binding peptides are usually 8-10 amino acid residues in length" in the paragraph of the application cited in point 16 above in the light of his common general knowledge as represented by document D19 (see point 18 above) would derive from the application that also peptides with more than 10 amino acids bind to MHC class-I molecules. This is so because document D19 explicitly discloses that these peptides may bind to MHC class-I molecule, in particular at their C-terminal end, variations in length being accommodated either by "kinking", i.e. forming a curve or sharp twist, or by extending out of the binding cleft. Furthermore, the board is of the opinion that the wordings "in most cases" or "suggest that some length variation can also be accommodated in this way" in document D19 would indicate to the skilled person that more than the two explicitly mentioned examples of longer class I-binding peptides are known in the prior art and that therefore the binding of peptides longer than 10 amino acids is not an exception. It follows from this that the examining division's argument that the binding of peptides longer than 10 amino acids to MHC class-I molecules is "impossible" does not convince the board.

20. The reasons set out in point 19 above apply equally to the retro-inverso peptidomimetics according to claim 1 exceeding a length of 10 amino acids and to the invention defined in claim 8.

21. The examining division further took the view that there was no guarantee that the exopeptidases disclosed in document D19 which cleaved longer peptides into shorter ones would cleave in such a manner that the peptides according to claim 1 were generated. Rather, their cleaving action could result in peptide fragments having little or nothing in common with that encoded by SEQ ID NO. 1.

22. The board is not convinced by this argument. Document D19 discloses that the exopeptidases cleave peptides only "subsequently" to their binding to MHC class-I proteins (see point 18 above). Further, as mentioned in point 17 above, the application discloses that a peptide consisting of the sequence of SEQ ID NO. 1 binds to MHC class-I molecules. It follows from this that peptides once bound to MHC class-I are protected from the proteolytic activity of the exopeptidase, since otherwise MHC-bound peptides would not exist. Therefore, MHC-bound fragments of peptides having little or nothing in common with the sequence of SEQ ID NO. 1 would not be generated by the exopeptidase.

23. In a second line of argument with regard to insufficiency of disclosure of the retro-inverso peptidomimetics according to claim 1, the examining division held that the application disclosed that they bound to MHC class-II molecules, but not that they bound to MHC class-I molecules. A similar objection was raised against the peptide nucleic acids (PNAs) referred to in claim 4, i.e. that the application did not disclose an example of these molecules encoding the peptides according to claim 1.

24. Article 83 EPC stipulates that an invention must be disclosed "in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art". Thus, it does not require a claimed invention to have actually been carried out by the applicant. Moreover, according to Rule 42(1)(e) EPC, the description must describe in detail at least one way of carrying out the invention claimed. Consequently, the presence of an example is not mandatory in the application. Therefore, just because an application mentions an effect - in this case the binding of retro-inverso peptidomimetics to MHC class-I molecules - but does not experimentally disclose that it actually occurs is not a sufficient reason for the board to doubt that the effect exists. Thus, for this reason alone the examining division's objections with regard to the retro-inverso peptidomimetics according to claim 1 or to the PNAs according to claim 4 do not convince the board.

25. Furthermore, the board notes that document D20 discloses the binding of retro-inverso peptidomimetics to MHC class-I molecules (see abstract). Moreover, document D21 reports that PNAs are standard molecules in the field of molecular biology which are structurally closely related to DNA or RNA molecules because they differ only in their backbone structure (see abstract, Figure 6). In view of this close structural relationship to DNA and RNA the board has no doubts that PNAs are able to encode the claimed peptides. Therefore, for these reasons too the objections of the examining division fail.

26. Accordingly, the board concludes that the main request meets the requirements of Article 83 EPC.

Novelty (Article 54 EPC)

27. The board agrees with the examining division's view that none of prior art documents cited are detrimental to the novelty of the subject-matter of any of claims 1 to 12 and thus the main request meets the requirements of Article 54 EPC.

Remittal (Article 111(1) EPC)

28. Since the examining division has not addressed the issue of inventive step, the board, exercising its discretion under Article 111(1), second sentence, EPC, and in accordance with the appellant's request, remits the case to the examining division for further prosecution.

Violation of the right to be heard and reimbursement of the appeal fee (Article 113(1) and Rule 103(1)(a) EPC)

29. The appellant requested reimbursement of the appeal fee under Rule 103(1)(a) EPC, arguing that points 3 to 5 and 7 of the decision under appeal contained two new objections under Article 83 EPC against the invention defined in claim 1 which had not been communicated to it before, and that this had deprived it of an opportunity to comment on them before the decision was taken; that was in breach of Article 113(1) EPC.

30. According to Rule 103(1)(a) EPC, the appeal fee shall be reimbursed where the board deems an appeal to be allowable, if such reimbursement is equitable by reason of a substantial procedural violation.

31. The present appeal is being allowed because the board has overturned the examining division's finding concerning sufficiency of disclosure on which the impugned decision was essentially based (see point 26 above). The fact that the case is being remitted to the examining division for further prosecution is no impediment to a possible refund.

32. A violation of Article 113(1) EPC may constitute a substantial procedural violation justifying reimbursement of the appeal fee. Article 113(1) EPC stipulates that decisions of the European Patent Office may only be based on grounds or evidence on which the parties concerned have had an opportunity to present their comments. The right to be heard is an important procedural right intended to ensure that no party is caught unawares by grounds and evidence, in a decision turning down its request, on which that party has not had the opportunity to comment (see e.g. R 2/14, point 6 of the Reasons). "Grounds or evidence" under Article 113(1) EPC are to be understood as meaning the essential legal and factual reasoning on which the decision is based (see e.g. T 951/92, OJ EPO 1996, 53, point 3(v) of the Reasons).

33. With regard to the first objection allegedly made for the first time in the impugned decision, the examining division's arguments in point 3.1 of its communication dated 28 March 2012 (hereinafter the "communication"), immediately preceding the impugned decision read as follows "The peptides of claims 1 are peptides having between 9 to 16 residues or its retro-inverso peptidomimetic and must bind to T cell corss-reacting the said peptide. Sufficiency of disclosure is not met, because it is impossible that peptides longer than 10 residues bind to the MHC complex and therefore being capable of activating T cells".

34. The corresponding objection in the impugned decision (see point 3 of the reasons) reads "The examining division considers that the requirements of sufficiency of disclosure are not met, because it is impossible that all the peptides between 9 to 16 amino acids bind to the MHC molecules, in particular those peptides longer than 10 amino acids and therefore they cannot induce T cells" (emphasis added).

35. In the board's view, both statements raise the same issue, i.e. that peptides having a length of between 11 and 16 amino acids cannot bind to MHC class-I molecules, a property which however is needed to induce cross-reactive T cells as required by the claim. In other words, the examining division held that the invention defined in claim 1 could not be carried out over the whole ambit of the claim. Indications for this are the cited ranges of "9 to 16 residues" and "peptides longer than 10 residues" in the communication (see point 33 above) and the ranges "all the peptides between 9 to 16 amino acids" and "in particular those peptides longer than 10 amino acids" cited in the impugned decision (see point 34 above).

36. The appellant argued that the objection in the impugned decision was new, because the statement "all the peptides between 9 to 16" (see point 34 above) implied that the examining division's view was that peptides of 9 and 10 amino acids also did not bind to MHC class-I molecules, a position which differed from that taken in the communication (see point 33 above). A further indication that the objection had substantially changed was derivable from point 5, second paragraph of the impugned decision, which stated that "As regards binding to MHC class I, it is noted that the two peptides extended out of the cleft mentioned in the textbook seem to be an exception of what the textbook discloses, namely that MHC class I molecules only bind those peptides having between 8 to 10 amino acids. The application lacks sufficiency of disclosure as to any peptide comprising SEQ. ID. No.: 1 and having more than 9 (length of the peptide of SEQ. ID. No.: 1) amino acids and binding to MHC class I molecules" (emphasis added). In particular, the indication of "more than 9" implied that the examining division held that peptides of 10 amino acids in length also did not bind.

37. The board does not agree. In the communication, the examining division indicated that it was "impossible that peptides longer than 10 residues bind to the MHC complex". Against this background, the reference to "all peptides" in conjunction with "in particular those peptides longer than 10 amino acids" in point 3 of the reasons of the decision (see point 34 above) can, in the circumstances of the present case, be interpreted only in the sense that peptides having 11 to 16 amino acids in length do not bind to MHC class-I molecules and that this was the reason why it was held "impossible that all the peptides between 9 to 16 amino acids bind to the MHC molecules". In other words, the board understands the term "in particular" in this context as the announcement of the reason given by the examining division as to why not all peptides between 9 and 16 amino acids bind to the MHC molecules, and not in the sense that the examining division was expressing its view that peptides with a length of 9 and 10 amino acids would not bind.

38. This interpretation is not changed by the statement "more than 9 [...] amino acids" in point 5 of the reasons (see point 36 above), because the passage as a whole further indicates "that MHC class I molecules only bind those peptides having between 8 to 10 amino acids". Moreover, the expression "peptides longer than 10 amino acids" in point 3 of the reasons (see point 34 above) addresses the same issue, albeit in other words. Both phrases cited imply that the examining division took the view that peptides with 10 amino acids bind to MHC class-I molecules, which contradicts the single statement "more than 9 [...] and binding to MHC class I molecules" (see point 36), since the above-cited wording of point 3 excludes 10. Therefore, in the board's view, the number "9" in point 5 of the reasons appears rather to be a clerical error and should have read "10".

39. In a further line of argument with regard to insufficiency of disclosure of the retro-inverso peptidomimetics according to claim 1, the examining division held in point 7 of the impugned decision that "The application also lacks disclosure as to any retro-inverso peptidomimetic binding to MHC class I molecules. Only the description indicates that some retro-inverso peptidomimetics have been shown to bind to MHC molecules (reference to Meziere et al on p. 30 of the description and available under http:// www.jimmunol.org/content/159/7/3230.full.pdf+html). However, this document relates to retro-inverso peptidomimetics binding to MHC class II molecules, whereas the present peptide binds to class I molecules". The examining division had not raised this objection in any earlier communication.

40. Therefore the board notes that the appellant was not given an opportunity to comment on this objection before the examining division took its decision. Accordingly, the appellant's right to be heard under Article 113(1) EPC has been violated in this respect.

41. However, in the board's view, this does not warrant reimbursement of the appeal fee. For such reimbursement to be equitable, there must be a causal link between a substantial procedural violation and the filing of the appeal (see CLBA, IV.E.8.6.1).

42. In the present case, the board cannot see any such causal link between the violation of the appellant's right to be heard and the need to file an appeal, because it would have had to file the appeal anyway, in view not only of the examining division's first objection under Article 83 EPC (in respect of which the board found no violation of the right to be heard see points 33 to 38 above) but also of the objection under Article 84 EPC in relation to claims 10 to 12. The second objection indicated in point 39 above was raised only additionally against the retro-inverso peptidomimetics referred to in claim 1, since the first objection - that peptides longer than 10 amino acids did not bind to MHC class-I molecules - was also directed against them (see point 7, first paragraph of the impugned decision). It was therefore incidental in the sense that - leaving aside the particular formulation - the impugned decision is based on objections on which the appellant had the opportunity to comment, and to reverse that decision it would have had to file an appeal, and also pay the appeal fee.

43. Lastly, the fact that the examining division's reasoning on sufficiency of disclosure has not convinced the board (see point 26 above) amounts to an error of judgement which, according to established jurisprudence, does not constitute a procedural violation (see CLBA, IV.E.8.4.5).