T 0429/96 (Serine protease inhibitors/AMGEN) 31-05-2001

1. The appeal is admissible.

Admissibility of the grounds of opposition under Articles 83 and 56 EPC

2. In the notice of opposition (cf points 5 and 6), in addition to objection to novelty, objections were raised and substantiated by the respondent under Articles 56 and 83 EPC in respect of some particular aspects of the invention. However, the patent as granted was attacked as a whole and its revocation was requested on grounds of Article 100(a)(b) EPC. This has led to the revocation of the patent by the opposition division and to the subsequent filing upon appeal of an amended claim request aimed at restoring patentability. The board finds that the said grounds of opposition were sufficently substantiated in the notice of opposition. They are thus valid throughout the opposition-appeal proceedings and have to be examined in respect of the amended claims now on file within the legal and factual framework of the present appeal.

Articles 123(2),(3) and 84 EPC

3. Claims 1 to 14 for the non-AT Contracting States are identical with the corresponding granted claims 19, 9 to 18 and 24 to 26 in that order, while claims 1 to 17 for the Contracting State AT are identical with the corresponding granted claims 19, 9 to 18, 24 to 28 and 31. in that order. The respondent does not raise any formal objections to the claims at issue and the board also sees none.

Sufficiency of disclosure (Article 83 EPC)

4. The respondent argues that the production of variants of the inhibitor of claim 1 is not sufficiently disclosed because in the description (page 14, lines 14 and 17 of the published application as filed) reference is made to two U.S. patent applications which were not available to the public at the filing date of the patent in suit (also because one application number was missing). In the respondent's opinion, these cross-referenced applications contained the necessary information for the skilled person to substitute by recombinant DNA methods one or more amino acid residues in the inhibitor to create analogs thereof.

In the board's judgement, however, a document incorporated by reference into the text of a European patent application has rather to become available to the public at the latest on the publication date of this European patent application, in order to be taken into account for the purpose of Article 83 EPC (see decision T 737/90 of 9 September 1993, points 3 and 5 of the "Reasons"). This requirement is fulfilled by cross-referenced "U.S. Patent Application Serial No. 678 822 filed December 6, 1984" which can be easily established to be the priority document of the International patent application published as W0 86/03519 on 19 June 1986 that corresponds to the European patent No. 0 205 475 (application No. 8 5905 953.7, date of filing 4 December 1985) published on 30 December 1986, ie on the same day as the application underlying the patent in suit.

As for the other U.S. patent application, to which incomplete reference is made, the additional question is whether it was retrievable without undue effort on the basis of the information provided in the patent in suit (cf the rationale of decision T 737/90, supra; see point 5). It can easily be established on the basis of the whole information given, including the correlation with the other reference, that this is the second priority document of the quoted International patent application, namely the U.S. Patent Application Serial 803 471 filed on 2 December 1985, of Pradip K. Bandyopadhyay et al. with title "Recombinant methods for isolating serine proteases inhibitors and DNA sequences useful for same".

Therefore, it must be concluded that the skilled person had access to the technical contents of both documents at the publication date of the application underlying the patent in suit via the easily retrievable WO 86/03519 (EP 0 205 475). This provides the necessary information enabling the substitution of one or more amino acid in the inhibitor to create analogs thereof. In view of these findings, the board is satisfied that the patent in suit enables the skilled person to arrive at analogs of the claimed serine protease inhibitor.

5. In the respondent's view, undue burden would be required for the skilled person to select, among the great many variants covered by claim 2, those exhibiting the required property of inhibiting at least one serine protease.

In spite of the considerable amount of theoretically possible variations of the amino acid sequence, in the board's opinion, there is still likely to be a strong structural similarity between all the variants covered by the present claim 2. This view is confirmed when considering the passage on page 8, lines 5 to 19 of the patent in suit, which gives examples of variants exhibiting specific combinations of amino acid alterations. Therefore, it can be seen that all the variants share a substantial number of amino acid residues. The situation here, where the claimed products are limited to those having a certain structural relationship to one another, and a testable narrowly defined activity, must be distinguished from a situation where either the structure or the activity is not defined in a disputed claim, so that it can be said that some substances which it would be desirable to make fall within the claim, but the description gives no guidance as to how they can be made (cf decision T 301/87 (OJ EPO 1990, 335).

6. In conclusion, the requirements of Article 83 EPC are fulfilled.

Novelty Document (1)

7. The respondent argues that the serine protease inhibitor of claim 1 lacks novelty over document (1), disclosing a preparation called HUSI-I made from human seminal plasma and having a strong affinity for granulocytic elastase and cathepsin G.

In the board's view, however, the HUSI-I preparation disclosed by document (1) is a heterogeneous (page 361, line 1: "several multiple forms") and degraded (page 361, line 3: "due partly to proteolytic degradation") preparation. This finding is confirmed by post-published document (8) taken as an expert opinion, wherein it is stated on page 80, right-hand column that "Proteolytic modification of MPI [another name for HUSI-I] occurred extensively by seminal plasma proteases". In fact, document (1) gives the amino acid sequence of two HUSI-I degradation products predominating in the mixture. Therefore, it must be concluded that the source material and the preparation method disclosed by document (1) do not enable the skilled person to arrive at an isolated undegraded inhibitor according to claim 1 at issue.

Document (2)

8. In the respondent's view, document (2) is also novelty-destroying for claim 1 at issue because it discloses an antileukoproteinase from saliva or parotid secretion which is identical to HUSI-I.

However, the board observes that the preparation of document (2) is highly impure since it comprises "several distinct protein bands" once subjected to agarose gel electrophoresis (see page 1326, bottom of left-hand column). The concentration of the inhibitors in this impure preparation is also too low to appear as distinct protein bands (see passage bridging left-hand and right-hand column on page 1326 and Figure 3b). There is therefore neither a teaching in this document as to how to arrive at the purified and undegraded protein of claim 1 at issue, nor any unambiguous evidence that antileukoproteinase is identical with this undegraded serine protease inhibitor. In view of this, the subject-matter of claim 1 is novel over document (2). As a further consequence, the board has to dismiss the respondent's contention that claims 15 and 16 for the Contracting State AT lack novelty over document (2), disclosing all the steps (a) to (d) stated in these claims. As seen above, there is no evidence before the board that the result of applying steps (a) to (d) to the starting material of document (2) is the purified, undegraded serine protease inhibitor of claim 1 at issue.

Document (3)

9. As for the bronchial mucus inhibitor (BMI) described in document (3), which the respondent also views as anticipating the claimed inhibitor, the board notes that it exhibits a N-terminal Tyr (page 586, bottom of right-hand column), a 99 amino acid chain length (see Table on page 587), no Trp residue (ibidem) and 12 to 14. Cys residues (page 588, left-hand column), while the claimed inhibitor has a N-terminal Ser, a 107 amino acid chain length, one Trp moiety and 16 Cys residues. These discrepancies, especially the N-terminal Tyr moiety as opposed to the Ser residue, suggest that the protein of document (3) is degraded or is a different one. It is true that the polyacrylamide gel electrophoresis (see page 585, right-hand column, penultimate paragraph) shows only one band, however, the respondent does not dispute that electrophoresis takes place under reducing conditions, where the protein fragments are cross-linked through -S-S- bridges.

Document (4)

10. The inhibitor of trypsin and chymotrypsin disclosed in this document has a molecular weight between 3,000 and 6,500 (see the Summary). It is smaller than the claimed inhibitor of 12 kD (see patent in suit, page 4, line 23). Therefore, it cannot affect the novelty of the claimed inhibitor.

Documents (5) and (6)

11. These documents have as co-author the author of document (1) (Prof. H. Fritz) and pre-date this document by six and four years, respectively. They relate to early characterization attempts of HUSI-I and cervical mucus derived inhibitor (CUSI). They do not provide any more technical information than document (1), according to which these inhibitors are mixtures of fragments.

12. In conclusion, owing to the techniques and/or source material of the prior art, any attempt to isolate and characterize the protease inhibitors yields either a different and shorter inhibitor (document (4)) or merely mixtures of degraded protein fragments (documents (1) to (3), (5) and (6)), rather than an isolated polypeptide consisting of a single unfragmented polypeptide chain as required by claim 1 in suit. The provision of this unfragmented serine protease inhibitor is a true technical achievement conferring novelty on present claim 1. Therefore, the respondent's contention that claim 1 merely relates to the provision of a correct amino acid sequence for an otherwise known protein, does not convince the board. Since claims 2 to 14 for the non-AT Contracting States and claims 1 to 17 for the Contracting State AT all rely on the novel protein of claim 1, there is no need to consider their novelty separately from that of claim 1.

Inventive step

13. The parties consider document (1) as representing the closest prior art and the board agrees as well. Document (1) relates to the best attempt before the priority date of the patent in suit to isolate and characterize the inhibitor HUSI-I, known to be present in human seminal plasma and having a strong affinity for granulocytic elastase and cathepsin G. The disclosure of document (1), however, does not lead to an isolated serine protease inhibitor consisting of a single unfragmented polypeptide chain as required by claim 1 in suit because the "HUSI-I" disclosed therein is a heterogeneous and degraded preparation (see point 7 supra). The board is satisfied that the patent in suit solves the problem of providing such an isolated serine protease inhibitor consisting of a single unfragmented polypeptide chain. It has thus to be established whether or not the claimed protein follows in an obvious way from the prior art. In the board's view, document (1) does not suggest that it is possible to identify and isolate the claimed "native" inhibitor, much less teaches a purification process that would yield that protein. Consequently, the subject-matter of claim 1 fulfils the requirements of Article 56 EPC. Since claims 2 to 14 for the non-AT Contracting States and claims 1 to 17 for the Contracting State AT all rely on the inventive inhibitor of claim 1, there is no need to consider their inventive step separately from that of claim 1.

14. The respondent argues that the skilled person could have easily further purified HUSI-I of document (1) and sequenced other fragments to arrive at the whole sequence stated in claim 1 at issue, as done in Example 3 of the patent in suit.

In the board's view, though, it has first to be noted that the problem to be solved by the subject-matter of claim 1 is not the provision of the amino acid sequence but of a native, undegraded serine protease inhibitor. No prior art discloses the measures to be taken (eg starting material, succession of process steps) by the skilled person for arriving at this molecule.

Even if it were assumed, for the sake of reasoning, that finding the complete amino acid sequence of HUSI-I was the final task aimed at, the board observes that the skilled person was faced with serious difficulties in reconstructing the actual protein because, as seen above (point 7), the preparation of document (1) was a mixture of degraded multiple inhibitory active forms. Therefore, the skilled person had not only to sequence the fragments but also to find out to which HUSI-I active form each of the fragments belonged, an arduous, if not impossible, task.

Adaptation of description

15. No objections are raised by the respondent to the amendments to the description effected to bring it into line with the claims, exception made for the wording "purified forms of protease inhibitors" on page 4, lines 39 to 41 of the description, which, in the respondent's view, implies a false distinction vis-à-vis the prior art also disclosing "purified forms of protease inhibitors". As emphasized under point 12 above, however, this is not the case. Therefore, the board sees no objections in this respect.