T 0811/01 (IL-1 inhibitor/BIOGEN) 06-08-2004

Claims as granted

Article 83 EPC

1. For the purpose of considering whether a European patent does or does not disclose the invention, the subject-matter of a particular claim, in a manner sufficiently clear and complete to be carried out by a person skilled in the art (Article 100(b), Article 83 EPC), the Board has to be satisfied firstly that the patent specification certainly puts the skilled person in possession of at least one way of putting the claimed invention into practice, and secondly that the skilled person can put the invention into practice over the whole scope of the claim.

2. In the present case, the opposition division (page 3 of the decision) considered that the indication of a particular value for the specific activity only made sense in the context of a particular assay using a particular cell line, since different assays using different cell lines would lead to different results and alter the definition of the compound of claim 1. The non-availability to the public of the cell line EL-4.6.1c10 used in the assay for determining the specific activity mentioned in feature (f) of claim 1 as granted prevented the skilled person from reproducing the invention.

3. The EL-4/CTLL assay disclosed in Example 4 of the patent in suit (page 10, column 15, lines 18 to 40), which led to the specific activity mentioned in feature (f) of claim 1, determines the inhibition obtained with a given IL-1 inhibitor on the IL-1 mediated interleukin-2 (IL-2) production by the EL-4.6.1c10 cell line. The only significant feature of the EL- 4.6.1c10 cell line in the context of the EL-4/CTLL assay used in the patent in suit is its ability to produce IL-2 upon stimulation with IL-1. Thus, even if one concurred with the opposition division that the value of the specific activity given by various assays might be different (this is indeed shown in the paragraph bridging columns 10 and 11 of the patent in suit), in the context of the EL-4/CTLL assay, the replacement of a cell line having the feature mentioned above, such as the EL-4.6.1c10 cell line, by another one exhibiting the same feature should be without influence on the value obtained, as far as the production of IL-2 upon stimulation by IL-1 is not the limiting factor of the reaction.

4. The question to be answered for determining whether the requirements of Article 83 EPC are met by the subject-matter of claim 1 is, therefore, whether the EL-4.6.1c10 cell line or a cell line having the ability of producing IL-2 upon IL-1 stimulation was publicly available at the priority date of the patent in suit.

5. Whether the EL-4.6.1c10 cell line was, as argued by the appellant, publicly available from the group of scientists having prepared it as disclosed in document (17), is besides the point, because document (17) in fact, contrary to the allegation of the appellant, does not describe the preparation of this cell line. In this document, the preparation of the cell line EL-4.6.1 is described, by sub-cloning of the parental cell NIH EL-4 leading to EL-4-16, then submitting the obtained EL-4-16 cell line to mutagenesis with methane sulfonate and culture in presence of 5-bromo-2-deoxyuridine and ouabain (bridging paragraph between pages 3662 and 3663). The preparation of a further sub-clone, EL-4BUrOUr6.1b, is also described (page 3664, left column, lines 17 to 21 and right column, line 12), but nothing can be retrieved form document (17) concerning the cell line EL-4.6.1c10 used in the patent in suit.

6. In document (24), however, the preparation of NOB-1 cell line (abstract, page 9, left column, last paragraph), which is a sub-clone of the mouse EL-4.6.1 cell line of document (17), is described. This cell line constitutively produces very little IL-2, but in response to IL-1 produces high concentration of IL-2 (abstract) and it is said, on page 10 (heading "Discussion"), to meet all the requirements for the determination of IL-1 activity. It was deposited at and available from PHLS tissue culture collection (page 8, left column), the name of which has been changed since then, according to the appellant's indication in his statement of grounds of appeal, into "European Collection of Cell Cultures" (ECACC), an international depository authority under the Budapest Treaty. The respondent has submitted no evidence to the contrary or suggesting that the availability to the public of said NOB-1 cell line was in some way subjected to special restrictions.

7. Therefore, the skilled person, taught by the patent in suit that the only relevant feature of the cell line used in the EL-4/CTLL assay is its ability to produce IL-2 upon IL-1 stimulation, had at the priority date of the patent in suit at least one publicly available functional equivalent to the EL- 4.6.1c10 cell line used in the patent and was well able to determine feature (f) of claim 1.

8. The respondent had also objected that the product obtained after Step 6 of the purification process depicted in Figure 9 of the patent in suit and which is embraced by the subject- matter of claim 1, because of its specific activity of 1250 U/mg, does not exhibit the features of

(i) being substantially pure

(ii) migrating as a single band on SDS-PAGE

(iii) being substantially free of apolipoprotein A1

(iv) being substantially free of retinol binding protein, as requested by claim 1.

This is allegedly shown by the disclosure of document (12), since

(a)the product run on lane 6 of Figure 1 of document (12), which corresponds to the product obtained after Step 6 of the purification process depicted in Figure 9 of the patent in suit, contains, besides IL-1 INH, apolipoprotein A1 as a contaminant,

(b)a comparison of the values of the protein concentration and the specific activity given in Table 1 of document (12) which depicts, as Figure 9 of the patent in suit, the purification process of IL-1 INH, leads to the conclusion that the product of Step 6 of said process still contains 91.1% contaminants and is hence not substantially pure as required by claim 1.

9. Dr Mazzei, one of the inventors of the patent in suit and one of the authors of document (12), has indicated in his declaration (document (18), point 10) that, whereas in document (12) the specific activity reported for the protein preparation in Step 6 of Table 1 has been obtained, as in Figure 9 of the patent in suit, using two negative immunosorption columns (one containing anti-retinol binding protein antibodies and the other anti-apolipoprotein A1 antibodies), the preparation run on SDS-PAGE in lane 6 of Figure 1 of document (12) has been treated by a single immunosorption using only antibodies directed against retinol binding protein. This assertion is corroborated by the legend of Figure 1 of document (12) on page 685. There is thus no difference between Table 1 of document (12) and Figure 9 of the patent in suit, which both depict the various steps of the purification process of IL-1 INH. In contrast to this, the results depicted in Figure 1 of document (12) from lane 6 onwards do not relate to a product similar to that obtained from Step 6 onwards of the process disclosed in Figure 9 of the patent in suit or in Table 1 of document (12). Furthermore, the product of Step 6 of Figure 9 of the patent in suit or of Table 1 of document (12), which has been treated with anti-lipoprotein A1 antibodies, displays a specific activity of only 1250 U/mg, ie a specific activity very close to the lowest limit mentioned in feature (f) of claim 1 (1.2x10/3 U/mg). On the other hand, apolipoprotein A1 is said on page 9 (column 13, lines 1 to 10) of the patent in suit to be one of the two major contaminants representing, before Step 6, 90% of the protein content and in point 13 of document (18) one pass on anti-apolipoprotein A1 antibodies is said the remove 50 to 60% of said apolipoprotein A1. The specific activity of the product run on lane 6 of Figure 1 of document (12), which has not been submitted to a pass through a column containing anti-apolipoprotein antibodies, can hence be expected to be lower than 1.2x10/3 U/mg. Therefore, the product run on lane 6 of Figure 1 of document (12) is not encompassed by the subject-matter of claim 1 and the argumentation of the respondent is basically based on an incorrect interpretation of the teaching of document (12).

10. Furthermore, this line of argumentation is in contradiction with Figure 1 of document (12), in which the product run on lane 6, although less pure than the product obtained after Step 6 of the process of the patent in suit or of Table 1 of document (12), migrates as a single band on SDS- PAGE. Indeed, the presence of a single band on SDS-PAGE does not necessarily imply that only a single molecular species is present in the band, as the respondent assumes on page 17 of his statement of facts and arguments supporting the opposition filed on 8 October 1997.

11. The argument mentioned above (point 8) under point (b) is based on a calculation made in the statement of facts and arguments supporting the opposition (pages 10 to 11, point 4.1.1.2. and page 16, point 5.1.1.1) using the values given in Table 1 of document (12) and Figure 9 of the patent in suit for the protein concentration and the total IL-1 INH activity. This calculation leads to the conclusion that the IL-1 INH of Step 6 (ie after the immunosorption on anti-apolipoprotein A1- antibodies and anti-retinol binding protein antibodies) still contains more than 91.1% contaminants and cannot be considered as "substantially pure", this term being characterized in the patent in suit (page 4, column 4, lines 46 to 51) by two attributes:

(a)substantially free of retinol binding protein and apolipoprotein A1

(b) migration as a single band on SDS-PAGE.

12. It has already been shown (cf supra point 10) that the attribute (b) is fulfilled by the product run on lane 6 of Figure 1 of document (12) and that the product of Step 6 of the purification process described in the patent in suit and in Table 1 of document (12), being even more pure, should also migrate as a single band.

13. As far as the attribute (a) is concerned, the meaning of the expression "substantially free of retinol binding protein and apolipoprotein A1" has to be determined. The patent in suit does not give any numerical value of this expression. However, Example 1(e) shows what is meant by this expression. The product obtained at the end of Step 6 has a specific activity of 1250 U/mg and migrates as a single band (page 9, column 13, lines 29 to 36) and is hence encompassed by the subject-matter of claim 1. It has been subjected to a single pass on immunosorption using anti-retinol binding protein and anti-apolipoprotein antibodies, a treatment which is said in point 13 of document (18) to result in an almost complete absence of retinol binding protein and in the removal of 50 to 60% of the apolipoprotein A1. This is an illustration of what the expression "substantially free of retinol binding protein and apolipoprotein A1" as used in the patent in suit means. Since the product of Step 6 serves to exemplify the meaning of the feature "substantially free of retinol binding protein and apolipoprotein A1", the question as to whether this product, which also migrates as a single band in SDS-PAGE (Figure 1 of document (12), lane 6), is "substantially pure" in the meaning of the patent in suit is superfluous.

14. Furthermore, this line of argumentation is in contradiction with the teaching of the patent in suit (page 9, column 13, lines 1 to 36) which mentions that the two major contaminants representing at least 90% of the protein content are present in the preparation obtained after the gel filtration on Ultrogel AcA54 (Step 5 of the purification process) in the form of retinol biding protein and apolipoprotein A1, ie before the immunosorption on antibodies raised against apolipoprotein A1 and retinol binding protein. Since in document (18), Dr Mazzei indicates (points 12 and 13 of the declaration) that a single pass on anti-retinol binding protein and anti-apolipoprotein A1 antibodies removes substantially all the retinol binding protein and 50 to 60% of the apolipoprotein A1, the protein preparation of Step 6 can no longer contain 91.1% contaminants, as had been argued by the respondent.

15. The discrepancy between the results obtained from the calculation of the respondent and the logic behind the steps of the purification process (Figure 9 of the patent in suit or Table 1 of document (12)) suggests, as possible explanations, an interference of the remaining contaminants or of the buffers used in Steps 6 and 7 with the EL-4/CTLL assay. Indeed in Step 6, the column is developed with a phosphate buffer, whereas in Step 7 the elution from the Phenyl-Sepharose column is made using a gradient of NaCl in Tris buffer. Phosphate and Tris ions may have a different influence on the determination of the specific activity. On the other hand, IL-1 INH migrates in SDS-PAGE more slowly on lane 6 than on lane 7 of Figure 1 of document (12). The product run on lane 6 is, however, the same as on lane 7, except for the presence of apolipoprotein A1 as contaminant (because it has not been passed through a column of anti-apolipoprotein A1 antibodies). This suggests some kind of tight interaction between IL-1 INH and apolipoprotein A1 which modifies the migration behaviour of the former. Such a tight interaction modifying the migration behaviour of IL-1 INH can reasonably be expected to also have an impact on the specific activity of IL-1 INH, so that the basis for the calculation made by the respondent is rather hypothetical.

16. The respondent had also objected that even with the step of hydrophobic chromatography on Phenyl-Sepharose (Step 7 of the purification process) no preparation substantially free of retinol binding protein and apolipoprotein A1 can be obtained, since, according to lane 7 of Figure 1 of document (12), a product containing three bands is obtained, the two contaminant bands being assumed to be retinol-binding protein and apolipoprotein A1. The Board does not share the respondent's opinion and considers that the very faint contaminating bands seen on lane 7 of Figure 1 of document (12) are encompassed by the definition of the expression "substantially free of retinol binding protein and apolipoprotein A1" as used in the patent in suit (cf supra point 14) which does not imply the total absence of retinol binding protein and apolipoprotein A 1.

17. The respondent had further argued during the opposition procedure that claims 8 to 12 directed to the production of IL- 1. INH by recombinant DNA technology are the paraphrase of a mere desire and provide no technical teaching. Documents (13) and (14) were cited to show that the sequencing of the polypeptides and the establishment of a cDNA library caused difficulties. The Board cannot concur with this view. In document (13), three IL-1 receptor antagonists (x, Alpha and Beta) are disclosed which have the same amino acid sequence (x-form), from which the Alpha- and the Beta-forms are glycosylation variants (page 340, left column, heading "Discussion"). These IL-1 receptor antagonists are said on page 340 (right column, last paragraph) to "quite likely be the same protein as the IL-1 INH isolated from urine of febrile patients", ie the IL-1 INH of the patent in suit. On page 338 (left column, first full paragraph), the three forms of the inhibitor are said to be "directly sequenceable". This teaching is not in contradiction with the information mentioned on page 337 (right column, last paragraph), according to which of ten samples only two yielded significant sequence information and three preparations of the Beta-protein yielded no sequence information, because the failure can be explained by the poor yield of the fractions from the reverse-phased HPLC column (the last step of the purification procedure (page 337, left column, last paragraph)) used for the sequence determination. It can be assumed that the failure was due to an insufficient concentration of the inhibitor in these fractions. Document (14) discloses the expression of a cDNA of a human IL-1 receptor antagonist isolated from monocytes (page 342, right column first paragraph) which is the IL-1 INH described in document (13). A cDNA library has been constructed and screened with mixed oligonucleotide probes, this process leading finally to the isolation of a cDNA coding for the IL-1 INH (page 343, left column, second paragraph and paragraph bridging the left and right columns). The nucleotide sequence of the cDNA and of the corresponding polypeptide are given in Figure 2. There is no indication in document (14) that unexpected difficulties were encountered during the completion of this work. Therefore, the Board considers that the sequencing of the polypeptide in order to prepare hybridisation probes, the preparation of a cDNA library and its screening with the probes were feasible for the skilled person at the priority date of the patent in suit without undue burden of experimentation or the use of inventive skill.

18. In view of the foregoing, the Board is of the opinion that the subject-matter of the claims as granted meets the requirements of Article 83 EPC.

Article 111 EPC

19. The opposition division, having considered that the patent in suit did not fulfil the requirements of Article 83 EPC, has not dealt in its decision with the objections raised by the opponent under Article 56 (cf supra section II). Decision G 9/91 (OJ EPO 1993, 408) in point 18 after indicating that "the purpose of the appeal procedure inter partes is to give the losing party the possibility of challenging the decision of the opposition division on its merits", further states that "it is not in conformity with this purpose to consider grounds for opposition on which the decision of the opposition division has not been based". Accordingly, the Board remits the case for further prosecution by the opposition division, including exercising its discretion as to whether any other issues raised in the original opposition should be taken up ex officio pursuant to Rule 60 EPC.