T 2006/08 (Conjugated factor IX/BIOVITRUM) 18-10-2011

Main and sole request - Claims as granted

Article 100(b) EPC

1. In the decision under appeal, the opposition division came to the conclusion that the claims as granted did not fulfil the requirements of Article 83 EPC, because all examples of the patent concerned factor VIII and not factor IX, being that factor IX was only briefly mentioned in the general part of the description. In view of the large structural and functional differences between these two factors, it could not be expected that the conclusions reached for factor VIII could be prima facie applied to factor IX without undue experimentation.

2. Claim 1 as granted is directed to "a process for improving the in vivo function of a factor IX by shielding exposed targets of factor IX", the process then being defined by the method steps (a) to (d) (cf. Section I supra). According to the established case law, an objection for lack of sufficiency of disclosure should be based on serious doubts - substantiated by verifiable facts - that the process itself, as defined by the method steps, could not be successfully applied to factor IX without undue burden. Moreover, the claimed process implies that an improved in vivo function of factor IX is achieved, since this is the purpose of the claimed process and thus, a limiting feature of the claimed subject-matter. From the argumentation put forward by the opposition division as well as by the respondent, it is however not apparent to the board to what extent the undisputedly existing structural and functional differences between factors VIII and IX may have an impact in the extrapolation of the method disclosed in the patent-in-suit for factor VIII to factor IX.

2.1 As for the method, the board considers that, although no experimental details are provided for factor IX in the patent-in-suit, no undue experimentation would be required to carry out the method steps (a) to (d) with factor IX. The patent-in-suit provides sufficient information (cf. paragraph [0020] et seq., Examples 3 to 5 of the patent-in-suit) and there is also prior art on file which demonstrates that no special contribution from a skilled person would be required for immobilizing factor IX on (group-specific) adsorbent anionic-exchange materials. The PEGylation of factor IX was also well-known in the art (cf. document D6 and paragraphs [0004] and [0008] of the patent-in-suit).

2.2 As for the effect or the presence of an improved in vivo function of factor IX, it was known from the prior art that, while the specific activity of a PEGylated protein usually decreases, a PEGylated protein also has a longer half-life and a decreased immunogenicity compared to the non-PEGylated protein. This is specifically disclosed in document D6 in relation to factor IX (cf. page 4 lines 10 to 16 of document D6). These properties provide an overall improvement in the in vivo function of the protein. The degree of modification allows a trade off between advantages (longer half-life and lower immunogenicity) and disadvantages (decreased activity) (cf. inter alia paragraphs [0002] to [0004] of the patent-in-suit). Accordingly, it is plausible that the claimed process does achieve an improvement of the in vivo function of factor IX. It is worth noting here that the improvement defined in claim 1 has to be achieved over the in vivo function of a native, non-conjugated factor IX.

Respondent's experimental evidence (Annex II)

3. In reply to the appellant's grounds of appeal, the respondent submitted an experimental report (Annex II) to support its argumentation (cf. Section III supra). According to the respondent this report shows that the resulting specific activities for factor IX are significantly lower than those obtained for factor VIII when both are PEGylated using NHS-chemistry (sic) and when compared to the respective untreated proteins. As regards this experimental evidence, the board has the following observations:

3.1 Indeed, it is shown in Annex II that, after PEGylation of recombinant factor VIII and factor IX, the specific activities are 49.9% and 55% for factor VIII and 13.5% and 21.8% for factor IX - all in comparison to the corresponding un-conjugated starting material. However, the experiments performed in Annex II have not been performed in accordance with the process of claim 1. In particular, the critical first step of this process, namely the immobilization of the protein by interaction with a group-specific adsorbent carrying anionic-exchange ligands, i.e. step (a) of claim 1 (cf. paragraphs [0011] and [0018] of the patent-in-suit), has not taken place.

3.2 While in Annex II it is shown that the decrease of specific activity upon PEGylation is larger for factor IX than for factor VIII, this result cannot serve as evidence that the performance of step (a) of the process of claim 1, previous to carrying out the PEGylation step, does not have a positive influence on the specific activity of the conjugated or PEGylated factor IX (cf. point 2.2 supra). Accordingly, the board considers that the results presented in Annex II do not allow to conclude that an improved in vivo function of factor IX (decreased immunogenicity and longer half-life compared to non-PEGylated factor IX; higher specific activity compared to PEGylated factor IX with no previous immobilization) cannot be achieved in a straightforward way by applying the process of claim 1.

3.3 It is also common general knowledge that the degree of purity of a protein might well influence its specific activity resulting from a PEGylation method (cf. inter alia paragraph [0010] of the patent-in-suit). Although both factor VIII and factor IX used in Annex II are recombinant products, different techniques have been used for their production (Advate manufacturing process for factor VIII and CHO cell-lines for factor IX) and there is no information in Annex II on their respective degree of purity.

4. It follows from these considerations that the respondent's argumentation as regards the objection of lack of sufficiency of disclosure is not supported by verifiable facts and thus, the claimed subject-matter is considered to fulfil the requirements of Article 83 EPC.

Entitlement to the claimed priority

5. The respondent argues that the claimed subject-matter is not entitled to the claimed priority (cf. Section X supra). No comments have been made by the opposition division in the decision under appeal on the merits of this objection.

6. Claims 1 to 16 of the priority document are directed to "a process for chemical modification of a polypeptide by covalent binding a biocompatible polymer to the said polypeptide", wherein said process is characterized by method steps identical to those of granted claim 1. Claim 3 of the priority document further specifies the polypeptide as being factor IX. However, claims 1 and 3 are not restricted to processes resulting in an improvement of the in vivo function of factor IX.

7. Nevertheless, it is clearly derivable from the section "Summary of the invention" present in the priority document that the purpose of immobilizing a polypeptide on a group-specific adsorbent prior to the conjugating reaction is to retain the specific activity of the chemically modified polypeptide. This is also the purpose mentioned in the patent-in-suit. Moreover, by reference to the known prior art, the priority document also acknowledges the advantageous effects of PEGylation, namely an increased in vitro stability, an improved in vivo half-life and a reduction of the immunogenicity of the modified or PEGylated polypeptide (cf. page 1, last two paragraphs under the section "Background of the invention" of the priority document). In fact, this is an implicit disclosure that the process disclosed in the priority document actually intends to improve the in vivo function of a polypeptide such as factor IX. Therefore, the board concludes that the subject-matter of the claims as granted is at least implicitly, but directly and unambiguously, disclosed in the priority document.

8. As for the absence in the priority document of any example of the disclosed process using factor IX, the same reasons as those given above regarding Article 83 EPC apply also to the disclosure of the priority document.

9. Therefore, the claimed subject-matter is considered to be entitled to the claimed priority date.

Article 100(a) EPC in connection with Article 54(2) EPC

10. The respondent holds that the subject-matter of the granted claims is anticipated by the disclosure of document D5 (cf. Section X supra). The board, however, agrees with the conclusions of the opposition division in the decision under appeal and considers that the disclosure of document D5, which does not mention factor IX at all, cannot anticipate the subject-matter of the granted claims which is directed to a process for improving the in vivo function of factor IX (cf. Section I supra). Decision T 26/85 (supra), referred to by the respondent and concerning the novelty of overlapping ranges, does not apply to the present case in which the relevant issue concerns the novelty of a specific disclosure (factor IX) over a generic disclosure (polypeptides) and wherein the specific disclosure has not been disclosed at all.

11. Moreover, document D5 refers only to a "specific binder substance" or to a second substance that "specifically binds" to the domain of a first bioactive substance which is responsible for the activity of this first substance and wherein this first substance - which is defined as any molecule having a desired activity, typically a protein or a glycoprotein - is to be conjugated with a polymer. However, there is no reference to "a group-specific adsorbent carrying anionic-exchange ligands" and, even though the definition of "a group-specific adsorbent carrying anionic-exchange ligands" may be broadly interpreted (infra), it does certainly not embrace any of the (second) substances mentioned in document D5 (cf. inter alia pages 7 and 8 of document D5).

12. Thus, the claimed subject-matter is considered to fulfil the requirements of Article 54 EPC.

Article 100(a) EPC in connection with Article 56 EPC

13. Document D5 or, alternatively, document D8 have been cited by the respondent as closest prior art (cf. Section X supra). Document D8 was published on May/June 1996 - after the priority date of the patent (29 September 1995) and, since the claimed priority is considered to be valid (cf. points 5 to 9 supra), this document is not prior art citable under Article 54(2) EPC. The board, however, agrees with the appellant that document D6 is the most suitable starting point for the discussion of inventive step (cf. Section IX supra).

14. Both documents D5 and D6 disclose methods for improving the in vivo function of proteins by means of PEG conjugation. However, only in document D6 are such methods disclosed specifically in relation to factor IX, while factor IX is not even mentioned in document D5. Accordingly, document D6, disclosing methods for improvement of the in vivo function of factor IX, is considered by the board to represent the closest prior art.

15. The underlying technical problem may be seen in the provision of an alternative method for improving the in vivo function of factor IX. The process of claim 1 provides a solution to this problem and differs from the method disclosed in document D6 in that, previous to the PEGylation, factor IX is immobilized in a column comprising a group-specific adsorbent carrying anion-exchange ligands. By immobilizing factor IX prior to its conjugation, the negative impact of PEGylation on the biological function of factor IX can be advantageously avoided or reduced. Although in the patent the effect is not demonstrated for factor IX, it is clearly shown for factor VIII. Example 3 shows that, adsorbing recombinant Factor VIII to Q Sepharose**(TM) FF (a strong anion-exchanger on a Sepharose/Agarose matrix) before conjugation with PEG, results in a higher retention of its specific activity in comparison to the PEGylation of factor VIII carried out in Example 2 without prior immobilization of factor VIII. Other advantages of both economical and clinical nature are also referred to in the patent-in-suit (cf. paragraph [0013] of the patent-in-suit).

16. Although the deficiencies and problems associated with protein PEGylation were well-known to a skilled person, in particular the impairment of the in vivo function of the PEGylated protein, the board is convinced that, in view of the prior art on file, the solution disclosed in the patent was not obvious to a skilled person. Indeed, a skilled person faced with the problem underlying the patent-in-suit and looking for appropriate solutions would most likely have turned its attention to the teachings of documents D4 and D5, both disclosing protein PEGylation methods with protection of the protein active site by shielding it with a specific ligand.

17. Document D4 discloses the PEGylation of serine proteases, wherein the active site of these proteases is protected by shielding it with benzamidine, a specific inhibitor of the serine proteases of the trypsin family, which is immobilized on a highly hydrated insoluble polysaccharide (Sepharose) (cf. page 3, lines 5 to 29 of document D4).

17.1 It could be argued that, since benzamidine binds to several proteases belonging to the group of serine proteases, benzamidine is a "group-specific adsorbent". Moreover, since benzamidine has an amidino group in its structure, it could also theoretically function as an "anionic-exchange ligand". Thus, benzamidine could be seen as being encompassed within the general definition of "a group-specific adsorbent carrying anionic-exchange ligands" used in claim 1(a) as granted and broadly characterized in paragraphs [0011] and [0028] of the patent-in-suit. If this was the case, the solution as claimed would be obvious in the light of document D4, since factor IX, which is not mentioned at all in document D4, was however known to be a serine protease of the trypsin family.

17.2 However, the appellant convincingly argued that benzamidine has a very basic pK around 11 and thus, would not be suitable for use as an anionic-exchange ligand in a process in which the function of a protein is to be preserved (cf. Section IX supra). There is also no evidence on file to show or support that benzamidine has ever been used as an anionic-exchange ligand. Therefore, benzamidine is considered not to fall within the general definition used in the method step (a) of claim 1.

17.3 Moreover, the method of document D4 is, conceptually, completely different from the claimed process. Whereas the former relies on a direct and specific binding of a substance (benzamidine) to the active site of a protein (serine proteases), the latter does not specifically target the active site of the (serine protease) factor IX but relies on the presence of accessible negative charged amino acids on its surface to interact - in a more general manner - with the "group-specific adsorbent carrying anionic-exchange ligands". As stated in paragraph [0028] of the patent, group-specific adsorbents "often bind less strongly and elution can be performed under milder conditions than with mono-specific adsorbents, the latter binding to a single or a very small number of polypeptides" - as is the case for benzamidine.

18. Document D5 discloses a process for the preparation of a conjugate between a polymer and a first substance having a biological activity, the process comprising binding the first substance with a second substance that specifically binds the domain mediating the biological activity of the first substance, conjugating a polymer to the first substance having the second substance bound thereto, and freeing the second substance from the first substance having the polymer conjugated thereto (cf. page 5, line 20 to page 6 line 3 and claim 1 of document D5).

18.1 Document D5 explicitly states that "[t]he invention relies upon the use of a second substance that specifically recognizes a domain that mediates the desired biological activity of a first substance which is to be derivatized. The second substance can be viewed as a specific binder substance" (cf. page 6, lines 14 to 18) and further exemplifies the second substance as being "an antigen or antidiotypic antibody, receptor or anticytokine antibody, antibody, enzyme substrate, receptor or ligand. In each case the second substance specifically binds to the first substance to shield the domain of the first substance which is responsible for the activity of the first substance" (cf. page 7, lines 8 to 14). Accordingly, the method of document D5, as that of document D4, involves shielding the active site of the (first) biologically active substance to be conjugated, and requires, as an essential feature, that said shielding is made by binding a specific (second) binder substance to the active site of the first substance.

18.2 Factor IX is not mentioned at all in document D5 and, as for document D4, the board is convinced that there is no hint or indication in document D5 that could lead a skilled person to the claimed subject -matter in an obvious manner. The method of document D5 is conceptually and technically more closely related to that described in document D4 than to that of claim 1 (cf. point 17.3 supra).

19. Thus, the board considers the claimed subject-matter to fulfil the requirements of Article 56 EPC. Neither the disclosure in document D6 alone or in combination with any of documents D4 or D5 render the solution proposed by the claimed subject-matter obvious.

Reimbursement of the appeal fee (Rule 103(1)(a) EPC)

20. The appellant alleges that the opposition division committed a substantive procedural violation (Article 113(1) EPC) because it reached a decision which was based on a wrong application of established legal principles. Moreover, the opposition division, having issued twice a favourable opinion on sufficiency of disclosure, came to a negative conclusion only at the oral proceedings based on the same facts and evidence that were on file before. Finally, the patentee was deprived of the possibility of providing experimental data to overcome the alleged deficiency of the patent and it was left with the only possible remedy of appeal (cf. Section IX supra).

21. Although the opposition division expressed an opinion on the issue of sufficiency of disclosure in favour of the patentee in two communications (cf. point 5 of the communication dated 3 November 2006 and point 8 of the communication dated 26 October 2007 annexed to the summons to the oral proceedings), this opinion was clearly labelled as being preliminary and non-binding. It is also worth noting that, following the aforementioned communications of the opposition division, the opponent filed submissions on 28 February 2007 and on 8 February 2008, respectively, in which it clearly stated that its objections raised under Article 83 EPC were maintained and some further developments on that issue were added. To this extent, the appellant could have legitimately expected that the opponent would try to reverse the preliminary and non-binding opinion of the opposition division during the oral proceedings.

22. A preliminary, provisional (positive) opinion does not prevent a party to make its complete case. It is the responsibility of a party to ensure that the facts and evidence filed are not only unequivocally clear but also as complete as possible. If a party decides to retain or not to file further evidence to support its case, it runs the risk that an adverse decision may be issued based only on the available (incomplete) evidence on file.

23. The alleged wrong indications of the opposition division which the appellant considered to be contrary to the Guidelines and the established case law may well amount to an error of judgment by the opposition division but they do not constitute a procedural non-compliance or violation, let alone a substantial one.

24. Thus, the request for the reimbursement of the appeal fee is rejected.