T 0975/16 (LAG-3 on regulatory T cells/JOHNS HOPKINS) 28-10-2021

Reasons for the Decision

Admittance of documents D24 to D36 and the appellants' lines of arguments provided in their letter of 27 August 2021

1. Documents D24 to D29 were filed with the appellants' statement of grounds of appeal; documents D30 to D33 were filed with respondent's reply. Their admittance is thus governed by Article 12(4) RPBA 2007. No objections to admittance were raised by the parties, and the board has no reason to hold these documents inadmissible ex officio.

2. Since the appellants' lines of argument submitted by letter of 27 August 2021 as well as documents D34 to D36 are considered to be a legitimate reaction to issues newly raised in the board's communication (see section VIII.), the board decided to admit them into the appeal proceedings (Article 13(2) RPBA 2020).

Main request

Amendments - (Article 123(2) and 76(1) EPC)

3. This claim request was auxiliary request 1 underlying the decision under appeal. It was considered to meet the requirements of Article 123(2) EPC by the opposition division. On appeal, the respondent contested that claims 1, 4, 7, 8 and 10 met the requirements of Article 123(2) EPC and 76(1) EPC.

4. The board considers that the decision under appeal was correct on this point. However, in view of the board's conclusion on novelty (see point 20 below), there is no need for the board to give reasons for holding so, except for the subject-matter of claim 8, which is identical to the subject-matter of claim 1 of auxiliary request 2.

5. Regarding the features of claim 8, the opposition division considered that the feature "isolated polypeptides" found basis in paragraph [56] of the application as filed. The combination of CD223 inhibitory antibodies and an anti-tumour vaccine found basis in paragraphs [45], [46], [48] and [56] of the application as filed. Finally, the feature "in a patient with regulatory T cells which suppress an immune response to an anti-cancer vaccine" found basis in paragraph [17] of the application as filed. The respective passages (same numbering) were also present in the parent application.

6. The respondent argued that claim 8 referred to an undisclosed combination of features.

7. Under the "gold standard" (see G 2/10, OJ 2012, 376), an amendment is allowed under Article 123(2) EPC only if it remains 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 (see G 3/89, OJ 1993, 117; G 11/91, OJ 1993, 125). After the amendment, the skilled person may not be presented with new technical information (see Case Law of the Boards of Appeal, 9th edn., 2019, II.E.1.3.).

8. Paragraph [17] of the application as filed and the parent application discloses that the invention provides a method for overcoming suppression of an immune response to an anti-cancer vaccine by administering an antibody which binds to CD223 to a cancer patient with Treg cells which suppress an immune response to an anti-cancer vaccine. An anti-cancer vaccine is also administered to the patient. The antibody increases the response of the cancer patient to the anti-cancer vaccine. That such an antibody has to be an inhibitory antibody is directly and unambiguously derivable from paragraph [45] of the application as filed (as well as the parent application).

Similar disclosure to paragraph [17] is provided in paragraph [56] (see also the identical paragraphs in the parent application), which in addition specifies that the anti-cancer vaccines can be isolated tumour antigens or polypeptides comprising one or more epitopes of tumour antigens. Finally, paragraph [46] of the application and the parent application refers to the use of an inhibitory agent that binds to CD223 protein or mRNA and also discloses co-administration of an anti-tumour vaccine.

9. Therefore, the board agrees with the opposition division's finding that the subject-matter of claim 8 does not provide the skilled person with technical information not contained in the original application documents. Consequently, the subject-matter of claim 8 of the main request does not extend beyond the content of the application as filed (Article 123(2) EPC) or the parent application as filed (Article 76(1) EPC).

Claim construction

10. In accordance with established case law, if a term used in a claim has (per se) a clear (ordinary) technical meaning, the description cannot be used to interpret such a term differently. In case of a discrepancy between the claim and the description, the unambiguous claim wording prevails. Hence, it must be interpreted as it would be understood by the person skilled in the art without the help of the description (see Case Law of the Boards of Appeal, 9th edn., 2019, II.A.6.3.1, especially decisions T 2221/10 and T 197/10). The appellants' argument that the claims had to be read in the context of the description thus fails.

11. The board considers that the expression "anti-cancer vaccine comprising isolated tumour antigen or polypeptides comprising one or more epitopes of tumour antigens" is clear. A person skilled in the art of vaccines would have understood this terminology to include vaccine compositions comprising tumour antigens isolated from their natural environment. This is not limited to a composition comprising only purified proteins, etc. but also includes, for instance, compositions comprising (small amounts of) tumour tissue isolated from a patient's body and comprising the tumour antigen.

Novelty (Article 54 EPC) - claims 1, 4, 7, 8 and 10

12. The opposition division held that the subject-matter of claims 1, 4, 7, 8 and 10 was novel over the disclosure in document D3 and that the subject-matter of claims 1 to 16 was novel over the disclosure in document D7.

13. The respondent contested the opposition division's decision on novelty over document D3. It also maintained that document D7 anticipates the subject-matter of claims 1, 4, 7 and 10.

14. Document D3 discloses the treatment of small amounts of tumour tissue before reimplantation with anti-LAG-3 (CD223) antibodies (Ab) to prevent T cell inhibition induced by LAG-3 and allow induction of cellular and humoral immunity against it (see page 9, lines 19

to 29).

15. The board considers that utilisation of the term "anti-cancer vaccine" in compound claims requires only that the disclosed isolated tumour antigens or polypeptide (fragments) be in a form "suitable" for use as an anti-cancer vaccine (see also Case Law of the Boards of Appeal, 9th edn., 2019, I.C.8.1.5).

16. Accordingly, document D3 anticipates the subject-matter of claims 1, 4, 7 and 10 of the main request.

17. The appellants' counter-argument, which was solely based on claim construction, cannot succeed in view of the claim construction adopted by the board (see points 10 and 11 above).

18. The respondent argued that page 9, lines 22 to 29 of document D3 disclosed the combination of an antibody to LAG-3 and an anti-cancer vaccine comprising isolated tumour antigens for administration to a patient to induce immunity against the cancer.

19. However, document D3 fails to disclose the treatment of the patient subgroup characterised by having Treg cells which suppress an immune response in response to an anti-cancer vaccine (see page 8, line 6; page 9, lines 19 to 29 and page 15, lines 15 to 17), and the respondent's argument does not address this difference. Consequently, the board does not find the respondent's line of argument persuasive.

20. The board concludes from the above that the subject-matter of claims 1, 4, 7 and 10 of the main request lacks novelty in view of the disclosure in document D3, while the subject-matter of claim 8 is novel over the disclosure in document D3.

21. In view of the board's conclusion on lack of novelty of the subject-matter of claims 1, 4, 7 and 10 over the disclosure in document D3 (see above), there is no need to consider the respondent's objection to these claims based on document D7.

Auxiliary request 1

Novelty (Article 54 EPC) - claim 1

22. The appellants did not contest that the subject-matter of the claim is anticipated by the disclosure in document D3 for the same reasons as set out above for claim 7 of the main request (see points 14 to 21 above).

Auxiliary request 2

Amendments (Article 123(2) and 76(1) EPC) - claim 1

23. Claims 1 and 2 of auxiliary request 2 are identical to independent claim 8 and its dependent claim 9 of the main request. The subject-matter of claim 1 does not extend beyond the content of the application as filed for the same reasons as set out above for claim 8 of the main request (see points 5 to 9 above).

Claim construction

24. Claim 1, drafted as a second medical use claim pursuant to Article 54(5) EPC, is directed to the treatment of a purposely selected subgroup of cancer patients characterised by the presence of Treg cells which suppress an immune response in response to an anti-cancer vaccine. The claimed treatment is restricted to patients refractory to anti-cancer vaccines due to Treg cells. The respondent did not contest that the patients represent a subgroup of cancer patients.

As set out by the appellants, the patent discloses how these patients with suppressive or "anergic" Treg cells can be identified, e.g. using the differentially expressed markers FoxP3, LAG3, CD25 and/or IL-10 (see Examples 2, 4 and 5 as well as Figures 2, 3 and 6 of the patent).

Inventive step (Article 56 EPC)

Closest prior art

25. Regardless of which document is ultimately considered to be the closest prior art, pursuant to Article 56 EPC, the claimed invention must not be obvious to the person skilled in the art having regard to any prior art, subject to Article 56, second sentence, EPC.

26. The purpose or objective of a purpose-limited product claim under Article 54(5) EPC is, generally, the therapeutic indication recited in the claim. In the case at hand, this is the treatment of cancer. The board therefore concurs with the opposition division and the respondent that document D6 represents a suitable springboard for assessing inventive step (see Case Law of the Boards of Appeal, 9th edn., 2019, I.D.3.1).

Indeed, this document discloses the same purpose and has the most relevant technical features in common with the current claim. Document D23, on the other hand, investigates the consequences of the absence of LAG-3 (= CD223) in knock-out mice, describing that these animals exhibit a defect in the natural killer cell, rather than the T cell, compartment (see abstract).

27. Review article D6 summarises the knowledge on LAG-3 (= CD223) as an important negative regulator of T cell homeostasis (see abstract; page 619, left-hand column). Soluble anti-LAG-3 antibodies which block LAG-3-MHC-II interactions are described to upregulate CD4 T cell clone activation. LAG-3 is expressed as a co-receptor on T cells and modulates effector T cell activity as well as Treg cell suppressor activity (see page 619, right-hand column; page 620, left-hand column). Document D6 concludes that "This suggests that LAG-3 blockade through systemic administration of mAb or sLAG-3 could be used for breaking tolerance to human cancer antigens for cancer immunotherapy, as shown recently for CTLA-4 [25]" (see page 621, right-hand column, last sentence).

28. This suggestion is based on a successful approach disclosed in document D13 (reference [25] in

document D6) which blocks CTLA-4, another T cell inhibitory receptor, with an anti-CTLA-4 antibody (see page 621, left-hand column of document D6).

Difference, its technical effect and problem to be solved

29. The subject-matter of claim 1 differs from the teaching in document D6 in that a subgroup of cancer patients, i.e. patients having Treg cells which suppress an immune response in response to an anti-cancer vaccine, is treated.

30. The technical effect arising from this difference is that Treg cell-mediated tolerance to the anti-cancer vaccine is broken in these anti-cancer vaccine tolerant patients such that an immune response is generated.

31. The board does not concur with the technical problem as formulated by the opposition division and the respondent for the following reasons.

32. "[P]roviding further means for cancer immunotherapy" or "providing an alternative means for cancer immunotherapy" as defined by the opposition division and the respondent is too generic and fails to take into consideration the difference with the teaching in document D6 and the technical effect provided.

33. The technical problem as defined by the appellants, i.e. generation of a targeted immune response to treat cancer in patients with Treg cells that suppress the immune response to an anti-cancer vaccine, on the other hand, comprises a pointer to the solution. This is contrary to established case law, under which the technical problem addressed by an invention has to be formulated in such a way that does not contain pointers to the solution or partially anticipate the solution (see Case Law of the Boards of Appeal, 9th edn., 2019, I.D.4.3.1.).

34. The board considers that the objective technical problem can be defined as the provision of a further cancer therapy in anti-cancer vaccine tolerant patients.

35. The claimed solution is the use of the combination of an antibody which binds and blocks CD223 and an anti-cancer vaccine comprising an isolated tumour antigen in those anti-cancer vaccine tolerant patients that have Treg cells which suppress an immune response in response to an anti-cancer vaccine.

Obviousness

36. The question to be answered is whether starting from the teaching in document D6 it was obvious to use the combination of an antibody which binds and blocks CD223 and an anti-cancer vaccine comprising an isolated tumour antigen to treat cancer in the subgroup of patients characterised by having Treg cells which suppress an immune response in response to an anti-cancer vaccine and if so, whether the skilled person had a reasonable expectation of success.

37. At the filing date of the application, it was common general knowledge that immune tolerance to cancer may be based on different mechanisms.

38. For instance, document D28 discloses that "In cancer patients, the absence of efficient tumor-specific immunity can be related to inadequate APC (antigen presenting cell) function or to T cell tolerance/ignorance towards tumor antigens" (see Introduction, second sentence).

39. Document D31 reviews a number of successful therapies that combine an immunostimulatory antibody targeting one of the different inhibitory pathways with a cancer vaccine.

40. Finally, document D13 reveals CTLA-4 to be a critical immunoregulator on activate T cells and shows that blocking CTLA-4 with specific antibodies results in autoimmune manifestations and cancer regression (see abstract). Blocking CTLA-4 with antibodies is reported to break tolerance against self antigens and to result in enhanced anti-tumour immunity in response to vaccination with tumour-antigens (see page 8372, paragraph bridging left- and right-hand columns). However, the patients treated in document D13 had never been immunised against the gp100 antigen used for vaccination (see page 8372, right-hand column, last paragraph), nor are they described as having failed to respond to an anti-cancer vaccine before.

41. Considering the skilled person's common general knowledge about the complex mechanisms involved in the induction of immunological tolerance (see e.g. document D31 or D35), and considering that various mechanisms of tolerance are responsible for immunosuppression in cancers (see point 37 above), the respondent's argument that the combination of the anti-cancer vaccine with an anti-LAG-3 antibody would have been expected to augment the immune response regardless of the subgroup of cancer patients is not persuasive for the reasons set out below.

42. The application is the first document to show that LAG-3 is expressed at higher levels on Treg cells and that its expression alone is sufficient to convert cells from activated effector T cells into immunosuppressive Treg cells (see Examples 2 to 9). The application also shows that LAG-3-deficient mice have increased numbers of CD4+ as well as CD8+ T cells compared to wild-type mice, while the CD4:CD8 T-cell ratio was unchanged. Suppressive Treg cells can be identified by the differential expression of LAG-3, IL-10, CD25 and/or FoxP3 (see Examples 2, 4 and 5 as well as Figures 2, 3 and 6). The suppressive activity of induced Treg cells is shown to be directly proportional to LAG-3 expression (see Example 5), and LAG-3 expression alone is shown to be sufficient to induce the suppressive phenotype of Treg cells (see Example 8).

Finally, inhibitory anti-CD223 antibodies are shown to eliminate the in vitro (see example 10) and in vivo (see example 12) suppressive activity of the Treg cells, and ectopic LAG-3 expression on auto-antigen-specific T cells is shown to be sufficient to prevent the onset of autoimmune diabetes in a transgenic mouse model in vivo (see Examples 10 to 12).

43. Without this knowledge, the skilled person had no reason to use the combination of an antibody which binds and blocks CD223 and an anti-cancer vaccine comprising an isolated tumour antigen in the claimed subgroup of patients, let alone a reasonable expectation of success that treating them with a combination of CD223 blocking antibodies and isolated tumour antigens would overcome immunosuppression.

44. The board therefore considers that the teaching of document D6 alone, or even in combination with the teaching of document D13, would not have led the skilled person to treat a subgroup of cancer patients characterised by the presence of immunosuppressive Treg cells with a combination of an antibody which binds and blocks CD223 and an anti-cancer vaccine comprising an isolated tumour antigen.

45. Consequently, the subject-matter of claim 1 and, a fortiori, dependent claim 2 of auxiliary request 2 involves an inventive step (Article 56 EPC).

Disclosure of the invention (Article 83 EPC)

46. Under established case law (see Case Law of the Boards of Appeal, 9th edn., 2019, II.C.7.2.), the requirement of sufficiency of disclosure is considered fulfilled for a claim to a second medical use if the application discloses the suitability of the compound/composition for the claimed therapeutic application.

47. The disclosure of the application showing that LAG-3 is necessary and sufficient for the suppressive function of Treg cells has been summarised above (see point 42). The application also shows an about 50% increase in the number of dividing cells in CD223-/- mice in vivo and an increase in the number of CD4+ and CD8+ cells but also of B cells and macrophages (see Example 1). The person skilled in tumour immunology moreover knew that Treg cells are also located in tumours and that they influence CD8 T cell responses (see e.g. review articles D35 and D36).

48. The application furthermore provides a mechanism on how anti-LAG3 antibodies can break tolerance making it plausible that an antagonistic anti-LAG-3 antibody is able to block Treg cell-mediated tolerance to cancer-antigens in vivo and is thus suitable for treating cancer in a patient with Treg cells which suppress an immune response to an anti-cancer vaccine. The board therefore sees no need for any additional in vitro or in vivo data, such as an effect of an anti-CD223 antibody on CD8+ cells or the expansion of antigen-specific T cells in an immune tolerant, in vivo tumour model.

49. Moreover, as evidenced e.g. by document D6 (see "Abstract" and "Concluding remarks"), the skilled person would have extrapolated LAG-3-related mouse in vivo evidence to humans. No evidence to the contrary has been provided by the respondent.

50. The board furthermore considers that based on the evidence on file, the skilled person was in a position to provide appropriate inhibitory anti-CD223 antibodies without undue burden. Again, no evidence to the contrary has been presented by the respondent.

51. Finally, inventive step is acknowledged on the basis of the subgroup of patients claimed and not because some key information on CD8+ cells, antigen-specific T cells or intercellular mechanisms is missing in the prior art (see points 27 to 45 above). The board therefore does not agree with the respondent's argument that such information, if missing from the prior art, is also missing from the application.

52. The board concludes that the requirements of Article 83 EPC are met.