T 1505/23 (SIRPalpha signalling blocking antibodies/OSE) 03-07-2025

Main request (patent as granted)

Remittal - Article 111(1) EPC

1. The issue of remittal arose in these appeal proceedings in relation to one of the two alternative embodiments of the subject-matter of claim 1 as granted, more specifically in relation to the treatment of hepatocellular carcinoma (HCC) which was present in claim 1 as granted but had been deleted in the set of claims held allowable by the opposition division, since the opposition division had decided in relation to the HCC alternative solely on the issue of added subject-matter but not on other issues, including inventive step.

The appellant-proprietor requested remittal of the case to the opposition division for consideration inter alia of the inventive step of the claimed HCC alternative.

2. The board, however, decided not to remit the case to the opposition division for the following reasons.

Novelty and inventive step of the subject-matter of the granted claims, i.e. the claims of the present main request, including the treatment of HCC and melanoma, had already been discussed by the parties in the opposition proceedings (see e.g. notice of opposition points 5.2 and 5.4; points 4.1. to 4.3 of the appellant-opponent's letter dated 20 March 2020).

In accordance with Article 12(3) RPBA, both the appellant-proprietor and the appellant-opponent presented their complete appeal case. While novelty objections were only raised in relation to the treatment of HCC, a full discussion of inventive step was provided for both (alternative) therapeutic indications, namely the treatment of HCC and the treatment of melanoma. Document D12 was identified as the closest prior art for the treatment of HCC (see points 3.1.1 and 3.2.2 of the appellant-proprietor's statement of grounds of appeal and point E.8 of the appellant-opponent's statement of grounds of appeal), and document D10 was identified as the closest prior art for the treatment of melanoma (see point 3.2.1 of the appellant-proprietor's statement of grounds of appeal and point E.2 of the appellant-opponent's statement of grounds of appeal).

Moreover, in their respective replies to the statements of grounds of appeal, both parties addressed inventive step concerning the treatment of HCC (see appellant-proprietor's letter dated 1 March 2024, point 3.3.5; appellant-opponent's letter dated 1 March 2024, point 2.1.2) and the treatment of melanoma (see appellant-proprietor's letter dated 1 March 2024, point 3.3.1; appellant-opponent's letter dated 1 March 2024, point 2.1.1).

In view of these submissions by the parties, the board was in a position to decide the issue of inventive step also in relation to the HCC alternative. Accordingly, no special reasons within the meaning of Article 11 RPBA presented themselves that would have justified remittal to the opposition division.

Claim construction - claim 1

3. As set out by the Enlarged Board of Appeal in decision G 1/24 (see Order), the claims are the starting point and the basis for assessing the patentability of an invention under Articles 52 to 57 EPC. The description and drawings shall always be consulted to interpret the claims when assessing the patentability of an invention under Articles 52 to 57 EPC, and not only if the person skilled in the art finds a claim to be unclear or ambiguous when read in isolation.

The board understands the Order of decision G 1/24 as confirming that the claims form the decisive basis for assessing patentability. The second sentence of the Order is not understood by the board as implying that a claim should be interpreted in a manner that imports limitations from the description which are not supported by the wording of a claim. In particular, features disclosed only in embodiments in the description cannot be used to narrow the scope of a claim that is otherwise clear and broader in its wording.

In this context, the board notes that decision G 1/24 expressly refers to the principles developed in the case law of the boards of appeal, which remain applicable. One such principle is that the description cannot be relied on to read into the claim an implicit restrictive feature not suggested by the explicit wording of the claim (see Case Law of the Boards of Appeal, 10th ed., 2022, II.A.6.3.4 and the decisions cited therein).

This approach is also consistent with Reasons point 20 of decision G 1/24, in which the Enlarged Board of Appeal held that the correct response to any unclarity in a claim is amendment. It is thus not the task of a board to reconcile discrepancies between the claims and the description by way of interpretation.

4. Claim 1 is drafted as a purpose-limited product claim pursuant to Article 54(5) EPC. It defines a compound - either an antibody or an antigen-binding fragment thereof - through two functional features: specific binding to signal-regulatory protein alpha (SIRPalpha) and the ability to block the interaction between SIRPalpha and CD47 without any limitation on the precise molecular mechanism of interaction. The patent discloses in the description that blocking SIRPalpha signalling induces myeloid-derived suppressor cell (MDSC) differentiation into non-suppressive, natural killer-like (NK-like) cells; however, this technical effect is not reflected in the wording of claim 1 and therefore cannot be considered a limiting technical feature of the subject-matter of claim 1.

The treatment of HCC or melanoma in a patient constitutes a functional technical feature of the claim.

Particularities of the implementation of the treatment are not specifically defined in claim 1. However, it is apparent from the dependent claims that the treatment encompasses administration of the active ingredient - i.e. the antibody or antigen-binding fragment thereof - either alone or in combination with other therapeutic agents. Thus, the subject-matter of claim 1 encompasses both monotherapy and combination therapy (see dependent claim 4). The conditions to be treated encompass all forms of melanoma and HCC, including both primary and metastatic disease (see dependent claim 2).

5. According to established case law (e.g. T 609/02, Reasons point 9; G 2/21, OJ EPO 2023, A85, Reasons point 76), in second medical use claims the therapeutic effect is a functional technical feature that must be achieved by the claimed compound. Thus, the compound must be causally involved - either directly or indirectly - in bringing about the claimed therapeutic effect.

In the case of combination therapies, the compound's causal involvement in achieving a therapeutic effect in this method for treatment by therapy does not necessarily require the compound to be effective when administered alone. A therapeutic effect resulting from, e.g., a synergistic interaction with another agent or treatment may also satisfy this requirement of causal involvement, provided that the claimed compound makes a verifiable and meaningful contribution to the therapeutic outcome. In the context of a combination therapy, as encompassed claim 1 (see point 4. above), the anti-SIRPalpha antibody (or its fragment) may therefore be regarded as causative for the therapeutic effect, insofar as it contributes to the observed improvement in treatment efficacy.

6. The mechanism of action proposed in the patent - the differentiation of MDSCs into non-suppressive cells to reduce MDSC-induced immunosuppression - is not a technical feature of the claim. As explained in point 3. above, a principle in the established case law is that the description cannot be relied on to read into the claim an implicit restrictive feature not suggested by the explicit wording of the claim.

Inventive step - Article 100(a) and Article 56 EPC

Treatment of melanoma

Closest prior art document D10

7. It was undisputed that document D10 can be taken to be the closest prior art for assessing the inventive step of the subject-matter relating to the treatment of melanoma.

8. Document D10 reports that it was well-documented that SIRPalpha acts to inhibit the phagocytosis and in vivo clearance of CD47-expressing cells. This supported the idea that SIRPalpha-CD47 interactions create a barrier for antibody-mediated tumour cell elimination and provided a rational basis for targeting SIRPalpha-CD47 interactions to potentiate the clinical effects of cancer therapeutic antibodies (page 18342, right-hand column).

The authors postulate that interactions between CD47, expressed broadly on normal and tumour cells, and the myeloid inhibitory immunoreceptor SIRPalpha would negatively regulate phagocyte-mediated antibody-dependent cell-mediated cytotoxicity (ADCC) induced by cancer therapeutic antibodies, and that targeting of SIRPalpha-CD47 interactions would comprise a generic strategy to improve antibody therapy against cancer (page 18343, left-hand column, paragraph 1). Because of its much more limited tissue distribution compared with CD47, the authors also consider SIRPalpha to be the preferred target for potential future therapeutic interventions (page 18345, left-hand column, last full sentence).

Page 18343, left-hand column, paragraph 1 further provides a study in which the well-established B16F10 mouse metastatic melanoma model was used to investigate the role of SIRPalpha signalling in antibody-mediated tumour elimination. Melanoma cells expressing CD47 were intravenously injected into both wild-type and SIRPalpha-mutant mice, the latter lacking the cytoplasmic tail of SIRPalpha and thus being incapable of transducing inhibitory signals. Figure 1 shows that, in the absence of therapeutic anti-melanoma gp75 tumour antigen antibody TA99, both groups developed comparable tumour burdens in the lungs, indicating that SIRPalpha signalling does not influence tumour growth or metastasis per se. However, when treated with suboptimal doses of antibody TA99, wild-type mice exhibited only minimal tumour reduction, whereas SIRPalpha-mutant mice showed near-complete abrogation of tumour formation.

Page 18344, right-hand column, first full paragraph discloses that anti-SIRPalpha antibodies that block interaction with CD47 significantly enhance trastuzumab (anti-HER2) ADCC resulting in the killing of SKBR-3 breast cancer cells. This enhanced cancer cell killing is linked to the presence of the anti-SIRPalpha antibody which is thus causally involved in the treatment (see point 5. above).

The authors of document D10 conclude that SIRPalpha-derived inhibitory signals form a barrier to effective antibody-dependent tumour cell elimination in vivo. The findings provided a strong rational basis for combining therapeutic antibodies against cancer cells with antagonists of the SIRPalpha-CD47 interaction, such as the monoclonal antibodies against SIRPalpha described in the scientific paper. This is anticipated to enhance the clinical efficacy of cancer-targeting therapeutic antibodies and/or reduce the need for chemotherapy of other non-specific treatment regimens (page 18346, left-hand column, last paragraph).

Difference, its technical effect, and objective technical problem

9. The board considers the mouse model used in document D10 to be a valid model for metastatic melanoma. As stated in D10, this model is a well-established mouse metastatic B16 melanoma model. The appellant-proprietor's argument that the model did not reflect the tumour microenvironment of melanoma overlooks the fact that document D10 does not aim to model primary cutaneous melanoma, but rather focuses on the immune-mediated clearance of metastatic melanoma cells - a therapeutic target also encompassed by the scope of claim 1 (see point 4. above). The experimental design and conclusions are consistent with that objective.

10. The subject-matter of claim 1 differs from the metastatic melanoma model treatment disclosed on page 18343, left-hand column, paragraph 1 of document D10 in that an antibody or an antigen-binding fragment thereof specifically binding to SIRPalpha and blocking the interaction between SIRPalpha and CD47 is used to inhibit SIRPalpha signalling.

11. The technical effect of using a SIRPalpha-CD47 interaction-blocking anti-SIRPalpha antibody or antigen-binding fragment thereof, as claimed, instead of a SIRPalpha gene-silencing mutation as used in the metastatic melanoma experiment in document D10, is that it provides a therapeutically practical approach to inhibiting SIRPalpha-CD47 signalling in patients.

12. The objective technical problem is the provision of a therapeutic reagent for silencing the SIRPalpha signal for use in melanoma therapy.

The claimed solution is the use of an antibody or an antigen-binding fragment thereof specifically binding to SIRPalpha and blocking the interaction between SIRPalpha and CD47.

Obviousness

13. The board considers that, starting from the metastatic melanoma experiment in document D10, the skilled person would have had an incentive to use anti-SIRPalpha antibodies capable of blocking the interaction between SIRPalpha and CD47 as used in the subsequent SKBR-3 breast cancer cell experiment also for treating metastatic melanoma, at least for a combination therapy.

Due to the theory provided in document D10, i.e. that SIRPalpha-CD47 interactions create a barrier for antibody- mediated tumour cell elimination and that targeting the receptor-ligand interaction potentiates the clinical effects of cancer therapeutic antibodies, as well as the experimental evidence in document D10, there was a reasonable expectation of success that an antibody or an antigen-binding fragment thereof specifically binding to SIRPalpha and blocking the interaction between SIRPalpha and CD47 would be useful in the treatment of CD47+ cancers such as metastatic melanoma.

14. The appellant-proprietor argued that document D10 focused on enhancing ADCC by combining the SIRPalpha-CD47 blockade with tumour-targeting antibodies (e.g. trastuzumab for HER-2+ breast cancer or TA99 for gp75). D10 provided no data or suggestion that anti-SIRPalpha antibodies could be effective as monotherapy.

In contrast, the patent disclosed a novel and unexpected mechanism: blocking SIRPalpha signalling induced MDSC differentiation into non-suppressive, NK-like cells. Reducing MDSC-mediated immunosuppression and enabling effective immune responses using anti-SIRPalpha antibodies alone opened a new therapeutic avenue. The therapeutic effect in document D10, by contrast, relied entirely on ADCC enhancement via antibody combinations.

15. However, the fact that document D10 only discloses treatment in combination with other therapeutic antibodies is irrelevant since claim 1 is not limited to monotherapy (as is also evident from the dependent claims). As shown by the reduced tumour load in mutant animals lacking the SIRPalpha cytoplasmic tail (see Figure 1 of document D10), the absence of SIRPalpha signalling in SIRPalpha mutant mice had a significant causative effect on the inhibition of metastatic melanoma formation, compared to animals treated with phosphate-buffered saline (PBS) or anti-gp75 antibody TA99 alone.

Treatment with anti-SIRPalpha antibody also enhanced the ADCC toward trastuzumab-opsonised SKBR-3 breast cancer cells (Figure 5 of document D10) compared to controls.

16. The board also notes that the MDSC differentiation-based mechanism of action described in the patent is not reflected in the wording of claim 1 and therefore cannot be considered a limiting technical feature of the claimed subject-matter, nor relied upon to establish a new clinical situation for the purposes of assessing inventive step, even if it is disclosed in the description and supported by experimental data.

17. Furthermore, claim 1 encompasses the treatment of melanoma, irrespective of its clinical presentation, including both solid primary tumours and metastatic manifestations.

18. Consequently, the subject-matter of claim 1 as far as it relates to the treatment of melanoma lacks an inventive step in view of the teaching in document D10.

Document D49

19. With its statement of grounds of appeal, the appellant-opponent filed document D49 in the context of inventive step for the first time.

However, since the board did not rely on the disclosure of document D49 for the assessment of inventive step, there was no need to decide on the admission of document D49.

Treatment of HCC

Closest prior art - document D12

20. It was undisputed that document D12 could be taken to represent the closest prior art for assessing inventive step of the subject-matter relating to the treatment of HCC.

21. Document D12 describes liver transplantation as the best treatment for unresectable, early stage HCC; however, HCC recurrence was common - due to either occult metastasis development or circulating tumour cell engraftment. In vitro and in vivo experiments were conducted to assess whether blocking SIRPalpha-CD47 interaction enhances peritoneal cavity macrophage phagocytosis of cancer cells. Mouse macrophages were incubated with labelled Hepa1-6 (HCC cell line) and CMT93 (adenocarcinoma cell line) tumour cells and treated with SIRPalpha-CD47 interaction-blocking anti-SIRPalpha antibodies or control antibodies, while in vivo assays involved peritoneal injection of tumour cells with anti-SIRPalpha or anti-CD47 antibodies. Blocking SIRPalpha-CD47 interaction with anti-SIRPalpha antibodies significantly increased macrophage phagocytosis. The authors conclude that these findings suggest that targeting SIRPalpha-CD47 interaction could be a promising strategy for preventing HCC recurrence in liver transplant recipients.

While CD47 knockdown in tumour cells also enhanced phagocytosis, treatment with anti-CD47 antibodies failed to do so. The study explains this unexpected result by showing that anti-CD47 antibody treatment led to a marked decrease in the expression of CD80, CD86, and major histocompatibility complex (MHC) class II on macrophages, indicating functional impairment. It is likely that this downregulation compromised the macrophages' ability to effectively engage and destroy tumour cells.

Difference, its technical effect, and objective technical problem

22. Since document D12 merely shows enhanced phagocytosis of Hep1-6 cells, the subject-matter of claim 1 is taken to differ from the disclosure in document D12 by the effective treatment of HCC.

23. The technical effect is the effective treatment of HCC.

24. The objective technical problem is how to effectively treat HCC.

Obviousness

25. The authors of document D12 concluded from their experiments that SIRPalpha-CD47 interaction could be a therapeutic target for preventing HCC recurrence in liver transplantation recipients.

Due to this suggestion and the observation of increased phagocytosis of HCC cells as shown in document D12, the skilled person would have been motivated to use anti-SIRPalpha antibodies blocking the interaction between SIRPalpha and CD47 to treat HCC and, by doing so, would have arrived at the subject-matter of claim 1.

26. The appellant-proprietor argued that document D12 lacked methodological detail, mechanistic explanation, or reproducibility.

If deemed reproducible at all, the teaching of document D12 would be limited to the prevention of recurrence in liver transplant recipients, but not treating an existing HCC.

27. The board is not persuaded by the appellant-proprietor's arguments. The key findings in document D12 - namely, that inhibition of CD47-SIRPalpha signalling with anti-SIRPalpha antibodies enhances macrophage phagocytic activity against Hepa1-6 hepatoma cells in the peritoneal cavity - are presented with sufficient methodological detail and precision to be regarded as reproducible by the person skilled in the art. Therefore, the teaching of document D12 is considered a credible basis for assessing inventive step in the context of therapeutic strategies targeting HCC, especially in the context of metastatic forms.

The distinction drawn by the appellant-proprietor between "prevention" and "treatment" is not convincing in view of the disclosure in document D12. While the authors of document D12 frame their findings in terms of recurrence prevention, the experimental evidence clearly demonstrates a therapeutic intervention. Mice were actively inoculated with Hepa1-6 hepatoma cells, and administration of anti-SIRPalpha antibodies led to a significant enhancement of macrophage-mediated phagocytosis of these tumour cells. This immune-mediated clearance of viable HCC cells constitutes a direct anti-tumour effect.

The model used in document D12 simulates a clinically relevant scenario involving residual or circulating tumour cells, which are a known source of recurrence and are particularly relevant in the context of metastatic disease. Accordingly, the skilled person would have understood document D12 as relating to a treatment strategy for HCC, including especially metastatic forms, rather than a purely prophylactic approach.

28. The appellant-proprietor also argued that the invention targeted modulation of the tumour microenvironment, particularly reducing MDSCs and increasing non-suppressive NK-cells, a mechanism not addressed in document D12.

29. While it is evident that the tumour microenvironment in the peritoneal cavity may differ from that of the liver, the model investigated in document D12 remains biologically relevant for studying immune-mediated clearance of metastatic HCC cells. The observed enhancement of macrophage phagocytosis upon anti-SIRPalpha antibody treatment in this model strongly indicates a therapeutic effect against metastatic HCC cells.

Only those effects that arise from the distinguishing technical features reflected in the wording of a claim can support an inventive step.

30. Therefore, the board concluded that the subject-matter of claim 1, insofar as it relates to the treatment of HCC, lacks an inventive step in view of the disclosure in document D12.

Conclusion

31. Both claimed alternatives of claim 1 lack an inventive step. Therefore, the ground for opposition of Article 100(a) EPC in combination with Article 56 EPC prejudices maintenance of the patent as granted.

Auxiliary requests 1 to 21

Inventive step - Article 56 EPC

32. Claim 1 of auxiliary requests 1 to 6 is identical to claim 1 of the main request. Claim 1 of auxiliary requests 7 to 12 has been restricted to the treatment of melanoma, while claim 1 of auxiliary requests 13 to 21 has been restricted to the treatment of HCC.

Thus, the considerations presented with regard to inventive step in relation to the treatment of melanoma (see points 7. to 17. above) and/or HCC (see points 20. to 30. above) also apply to claim 1 of auxiliary requests 1 to 21.

33. Consequently, the subject-matter of claim 1 of auxiliary requests 1 to 21 lacks an inventive step under Article 56 EPC.

Admission of documents D37 to D48, D50 and D51

34. The appellant-proprietor requested the admission of (i) documents D37 and D38, which were first filed during the opposition proceedings but not admitted by the opposition division; (ii) documents D39 to D48, submitted for the first time with the statement of grounds of appeal; and (iii) documents D50 and D51, first filed with the reply to the appellant-opponent's statement of grounds of appeal.

Documents D37 to D48 were submitted in support of the argument that the anti-SIRPalpha antibody P84, as used in certain examples of the patent, possesses blocking activity.

Document D50 was filed in response to an objection under Article 123(2) EPC.

Document D51, a dictionary definition of the term "antineoplastic" was submitted to support the interpretation of terminology used in the priority application, i.e. to address an objection by the appellant-opponent regarding the priority date of claims 1 to 4 of the main request.

35. The present decision is based solely on the ground of opposition under Article 100(a) EPC in combination with Article 56 EPC relating to inventive step.

The board's findings concerning inventive step are based exclusively on the assessment of documents D10 (for melanoma) and D12 (for HCC).

The question of whether antibody P84, as used in some examples of the patent, is capable of inhibiting SIRPalpha-CD47 interaction was not relevant to the assessment of inventive step. Since claim 1 is drafted as a purpose-limited product claim pursuant to Article 54(5) EPC, its subject-matter is restricted to antibodies or antigen-binding fragments that specifically bind to SIRPalpha, block its interaction with CD47, and achieve the claimed therapeutic effect (see points 3. to 6. above). Thus, there was no need to decide on the admission of documents D37 to D48.

As the ground under Article 100(c) EPC is not part of the present decision, there was no need to decide on the admission of document D50.

Similarly, the question of priority was not relevant to the outcome of the appeal, and therefore there was no need to decide on the admission of document D51.