T 2094/13 (Treatment of amyloidogenic diseases/JANSSEN) 12-05-2016

Admission of documents D88, D95, D96, D101 and D110 into the proceedings

1. Documents D95 and D96 were submitted by the appellant with its letter dated 10 May 2013 in reply to the respondents' submissions and thus shortly before the oral proceedings before the opposition division. Document D101 was filed with the appellant's statement of grounds of appeal. Documents D88 and D110 were both submitted by respondent II, the former during the opposition proceedings with its letter dated 10 April 2013 and the latter in reply to the appellant's statement of grounds of appeal.

2. None of the respondents objected to the admission of documents D95 and D96 and the board decided to take them into account (Article 114(2) EPC).

3. It is within the discretionary power of the board to hold inadmissible facts, evidence or requests which could have been presented in the first-instance proceedings or were late-filed in those proceedings (Article 114(2) EPC and Article 12(4) RPBA). The appellant requested that documents D88 and D110 not be admitted into the proceedings. The board notes that respondent II wanted to rely on them to address an issue raised in the preliminary view expressed by the board in point 28 of its communication pursuant to Article 15(1) RPBA. Under these circumstances, respondent II's arguments based on evidence disclosed in documents D88 and D110 could not have been presented earlier and the board consequently admitted the former and kept the latter in the appeal proceedings.

4. Respondent I requested the exclusion from the appeal proceedings of document D101 relating to a decision from the England and Wales High Court (Patents Court) concerning the UK equivalent of the patent under consideration. The board notes, however, that it is established practice that the boards of appeal take into consideration relevant decisions of courts of the contracting states (see also Case Law of the Boards of Appeal of the EPO, 7th edition 2013 (hereinafter "CLBA"), III.H.3.1). Accordingly, the board confirmed that document D101 remained in the proceedings (Articles 12(2),(4) RPBA).

Main request

Introduction in the invention

5. Although today there are still a number of different hypotheses about the exact etiology of how Alzheimer's disease (AD) develops in humans, one of the established pathogenic hallmarks at the priority date of the patent in suit was - and still is today - the presence of amyloid deposits, also known as senile plaques, in the brain of AD patients (see e.g. document D88, page 631, column 1, first paragraph).

5.1 Abeta is the principal constituent of amyloid deposits and a natural proteolytic peptide fragment of 39 to 43 amino acids in length of an amyloid precursor protein (APP). Several mutations within APP, e.g. at position 717 of the protein sequence, have been correlated with the pathogenesis of AD, and are all commonly thought to increase the amount of pathogenic Abeta in the brain, in particular its long forms, e.g. Abeta1-42 or Abeta1-43 (see paragraphs [0003] and [0031] of the patent in suit; the board notes that the numbers 1-42 or 1-43 in relation to Abeta indicate that the peptide has a length of either 42 or 43 amino acids respectively (see paragraph [0033] of the patent in suit)).

5.2 The patent in suit relates to the use of compositions comprising antibodies against the amyloid beta peptide (Abeta) for the treatment of diseases characterised by amyloid deposits, e.g. AD.

Sufficiency of disclosure

6. It is established case law of the boards of appeal that when assessing medical use claims attaining the claimed therapeutic effect is a functional technical feature of the claims. Accordingly under Article 100(b) EPC, unless this is already known to the skilled person at the priority date, the patent in suit must disclose the suitability of the product to be manufactured for the claimed therapeutic application. Clinical trials are not required to establish suitability. It may suffice that in vitro or in vivo data directly and unambiguously reflect the therapeutic effect on which the claimed therapeutic application relies or, alternatively, an established relationship between the physiologic activities of the compound under consideration and the claimed disease (see e.g. CLBA, II.C.6.2 and decision T 609/02 of 27 October 2004 cited therein).

7. The subject-matter of claim 1 is pharmaceutical compositions comprising antibodies of a human IgG1 isotype binding to Abeta and a carrier or diluent for use in preventing or treating diseases characterised by amyloid deposits. The compositions for the claimed therapeutic application therefore comprise as embodiments such antibodies of a human IgG1 isotype which are (i) monoclonal, i.e. all bind to one single epitope on Abeta, or (ii) polyclonal, i.e. bind to an undefined number of different epitopes of undefined location within Abeta. Moreover, the epitope(s) recognised by the antibodies referred to in the claim can be situated within (iii) soluble or (iv) aggregated forms of Abeta.

8. Thus, the question to be assessed in the context of Article 100(b) EPC in the present case is whether or not the patent in suit or the prior art provides information disclosing the suitability of the compositions comprising antibodies as referred to in claim 1 for the claimed therapeutic application.

9. It is uncontested that the patent in suit does not disclose explicit experimental evidence that administered compositions comprising anti-Abeta antibodies attain a therapeutic effect in the treatment of diseases characterised by amyloid deposits, such as AD. It was also common ground between the parties that such evidence was not derivable from any of the cited prior art documents. Under these circumstances, the patent in suit must disclose evidence to the skilled person, having due regard of common general knowledge, that compositions comprising anti-Abeta antibodies reduce or prevent amyloid deposits. Failure to demonstrate such an effect necessarily results in the lack of the suitability of the anti-Abeta antibodies in the claimed therapeutic application.

10. The patent in suit discloses in several examples the immunisation of so-called PDAPP mice with either full-length aggregated Abeta or conjugated and non-conjugated fragments thereof (see e.g. paragraph [0112]). PDAPP mice are transgenic for the human APP gene, having a point mutation at position 717 of the protein sequence (see point 5.1 above). Expression of this gene inevitably causes the formation of Abeta amyloid deposits or plaques in mouse brains at an age of six months (see paragraph [0061]). Since these plaques were one of the established pathogenic hallmarks of AD at the relevant date of the patent in suit (see point 5 above), in the board's opinion - contrary to the view of the respondents - the skilled person would consider the PDAPP mice as a suitable animal model for evaluating the effectiveness of therapeutic agents in the treatment of human AD.

11. The following experimental data of the patent in suit are of particular interest:

11.1 Example I discloses that the immunisation of PDAPP mice with human aggregated Abeta1-42 prevents Abeta plaque formation, while non-immunised control mice develop plaques (see paragraphs [0061], [0069] to [0071] and [0076]).

11.2 Example II reports a dose-dependent formation of anti-Abeta antibodies upon immunisation of PDAPP mice and control mice with human aggregated Abeta1-42 (see paragraphs [0077] and [0081]).

11.3 Example III shows that the immunisation of PDAPP mice with human aggregated Abeta1-42 also reduces amyloid plaque deposits already established in the brain (see figure 7), including a reduction in the total amount of detectable Abeta and Abeta1-42 in certain regions of the brain (see tables 2 and 3 on pages 14 and 15). An immunohistochemical analysis of the brains after immunisation further indicates that activated phagocytic microglia and monocytes, i.e. cell-mediated processes, are involved in plaque removal (see paragraphs [0091] and [0095]).

11.4 Example IV reports that a reduction of established amyloid plaques, including a reduction in total Abeta in the brain of PDAPP mice, is also achieved by immunisation with either aggregated rodent Abeta1-42 or a conjugated Abeta1-5 fragment derived from the N-terminus of human Abeta (see paragraphs [0112], [0116], [0117], [0120] and [0121]). Example IV also reports, however, that the immunisation of PDAPP mice with other conjugated human Abeta fragments, i.e. Abeta1-12, Abeta13-28, Abeta33-42 and an aggregated Abeta25-35 peptide (see paragraphs [0107] and [0112]), does not significantly reduce either the amount of established amyloid plaques or the total amount of Abeta in the brain of PDAPP mice (see paragraphs [0120] and [0121]). In this context, the board notes that the lack of a therapeutic effect after immunisation with the Abeta fragments cited appears not to be due to the absence of anti-Abeta antibodies, since polyclonal antibodies binding to Abeta are detectable in the serum and on cerebral Abeta plaques (see paragraphs [0123], [0124]).

12. Accordingly, it can be derived from the above that the patent in suit reports that upon immunisation, certain Abeta antigens induce an anti-Abeta antibody response combined with a cellular immune response which in turn prevents or reduces cerebral amyloid plaque formation, whereas other Abeta antigens induce solely an anti-Abeta antibody response in PDAPP mice without affecting plaque formation.

13. On the one hand, the board considers that the skilled person after analysing the data disclosed in the patent in suit (see points 11.1 to 11.4 above) could thus have arrived at the conclusion that the immunisation of PDAPP mice with certain Abeta immunogens, i.e. aggregated human and rodent Abeta1-42 and a conjugated human N-terminal derived Abeta1-5 fragment, elicits polyclonal anti-Abeta antibodies which are suitable for the claimed therapeutic application. Moreover, on the basis of these data the skilled person could have also considered that monoclonal anti-Abeta1-5 antibodies were suitable for this application, since the Abeta1-5 fragment, due to its small size, encompasses only a very limited number of potential epitopes to which the antibodies can bind.

14. On the other hand however, the board notes that the experimental data also demonstrate that certain other Abeta immunogens, in particular fragments derived from the central or C-terminal region of Abeta, fail to induce therapeutically effective polyclonal antibodies, although the antibodies bind to cerebral amyloid plaques (see point 11.4 above). Since, due to their small size, the fragments Abeta1-12, Abeta13-28, Abeta25-35 and Abeta33-42 encompass only a very limited number of potential epitopes, the board considers that the failure of these polyclonal antibodies also extends to monoclonal antibodies binding to the same fragments.

15. The appellant has submitted that the skilled person would have considered anti-Abeta antibodies binding to soluble Abeta in general to be therapeutically suitable, since it was commonly known that antibodies binding to soluble proteins caused their removal from the body and thereby reduced the available amount of Abeta as the principal constituent of cerebral amyloid plaques.

16. The board does not agree. As reported in example IV of the patent in suit, all of the fragments Abeta1-12, Abeta13-28, Abeta25-35 and Abeta33-42 elicit anti-Abeta antibodies which bind to full-length Abeta1-42 in an ELISA assay, i.e. to soluble Abeta. However, none of these antibodies significantly reduces the total amount of Abeta in the brain or affects the amount of Abeta cerebral deposits (see paragraph [0124]). In the board's opinion, the skilled person would therefore derive from these data that antibodies, although binding to soluble Abeta and therefore able to remove Abeta from the blood, have no significant therapeutic effect in the treatment of AD.

17. The appellant further argued that the skilled person would conclude from the disclosure of example VI, and in particular table 6 on page 20 of the patent in suit, that monoclonal antibodies directed against Abeta1-12, Abeta13-28 and Abeta33-42, i.e. the N-terminal, central and C-terminal regions of Abeta are therapeutically suitable.

18. The board notes that example VI in paragraph [0130] and table 6 of the patent in suit only suggest the use of a variety of monoclonal antibodies binding inter alia to Abeta1-12, Abeta13-28 and Abeta33-42 for the treatment of amyloid deposits in PDAPP mice. However, no experimental data demonstrating a reduction of Abeta cerebral plaques by passively administered monoclonal antibodies, let alone by the ones mentioned in table 6, are disclosed in the patent in suit. As set out above (see points 11.4 and 16), example IV demonstrates that polyclonal antibodies binding to Abeta1-12, Abeta13-28 and Abeta33-42 fail to reduce amyloid plaques and total Abeta in the brain of PDAPP mice. The actual experimental data of example IV are therefore at odds with the suggestions reported in example VI and the arguments of the appellant must therefore fail.

19. The appellant also submitted that according to e.g. decision T 63/06 of 24 June 2008 (see point 3.3 of the Reasons) a presumption existed that a patent, once granted sufficiently disclosed the claimed invention and that the opponents bore the burden of proof for establishing insufficiency of disclosure.

20. While in general the board agrees with the appellant regarding the principles established in this decision, the situation in the present case is different since, as outlined in point 11.4 above, the patent in suit itself already discloses experimental evidence that antibodies binding to the central or C-terminal region of Abeta, which are encompassed as embodiments in claim 1, are not suitable for the claimed therapeutic application. Under these circumstances, no additional experimental evidence from the respondents is required, as they can rely on the evidence provided by the patent in suit itself. This shifts the burden of proof back to the appellant, whose argument must therefore fail.

21. Lastly, the appellant argued that the skilled person, in view of the disclosure of the patent in suit that polyclonal anti-Abeta antibodies were effective in the claimed therapeutic application, would have considered this as sufficient evidence for a generalised antibody-mediated concept in the treatment of AD, without the need for major conceptual leaps.

22. While it is established case law that claims can validly cover broad subject-matter, the question of the allowability of a broad claim versus the requirement of sufficiency of disclosure is one which is strictly assessed on a case-by-case basis, influenced by the extent to which the information in the patent in suit could be used to develop further embodiments without a major conceptual leap (cf. e.g. decision T 636/97 of 26 March 1998, point 4.5 of the Reasons).

23. A generalised anti-Abeta antibody-mediated concept for the claimed therapeutic application is not derivable from the patent in suit, since the patent in suit discloses evidence that antibodies binding to the central and the C-terminal region of Abeta are not suitable (see points 11.4 and 14). Also, the patent in suit provides neither the rationale for the disclosed failure of these antibodies nor information enabling the skilled person to nevertheless obtain therapeutically effective antibodies of a human IgG1 isotype directed against the central and the C-terminal region of Abeta. Accordingly, also this argument of the appellant must fail.

24. The board therefore concludes, in view of the patent in suit disclosing experimental evidence that anti-Abeta antibodies encompassed by the subject-matter of claim 1 are not suitable for the claimed therapeutic application, that the patent in suit fails to sufficiently disclose the claimed medical use as required by Article 100(b) EPC.

Auxiliary requests 1 to 3

25. The subject-matter of claim 1 of auxiliary request 1 differs from that of the main request in that the feature "antibody to Abeta" has been replaced by the feature "antibody to Abeta13-28".

26. The board notes that the patent in suit discloses in example IV, that anti-Abeta antibodies binding to Abeta13-28 fail to demonstrate suitability for the claimed therapeutic application (see point 11.4 above).

27. The subject-matter of claim 1 of auxiliary request 2 differs from that of the main request in that the feature "and wherein the antibody binds specifically to the dissociated form of Abeta peptide without binding to the aggregated form" has been added at the end of the claim.

28. The board notes firstly that the patent in suit suggests only in general terms antibodies binding specifically to soluble Abeta without binding to aggregated Abeta, and provides no experimental data demonstrating the suitability of specifically these antibodies for the claimed therapeutic application.

29. Secondly, as pointed out in point 16 above, the patent in suit discloses in example IV that polyclonal antibodies binding at the same time to soluble Abeta and aggregated Abeta in amyloid deposits are not suitable for the claimed therapeutic application. Although the binding properties of the antibodies of example IV therefore differ from the compositions comprising the antibodies according to claim 1, neither the patent in suit nor the cited prior art documents disclose evidence that antibodies binding solely to soluble Abeta are suitable for the claimed therapeutic application. Nor has that not been argued by the appellant.

30. The subject-matter of claim 1 of auxiliary request 3 differs from that of the main request in that the feature "wherein the patient is asymptomatic optionally wherein the patient is under 50, and/or has inherited risk factors indicating susceptibility to Alzheimer's disease" has been added at the end of the claim.

31. The board notes that neither the patent in suit nor any of the cited prior art documents reports evidence that restricting the patient group as claimed has an effect on the suitability of the compositions comprising anti-Abeta antibodies for the therapeutic application. This too, has not been argued by the appellant.

Therefore, the reasoning set out in points 7 to 23 and the conclusion in point 24 apply mutatis mutandis to the patent in suit in the context of claims 1 of auxiliary requests 1 to 3. Accordingly, the patent in suit does not sufficiently disclose the subject-matter of claims 1 of these requests, contrary to the requirements of Article 83 EPC.