T 1210/05 (2-(2-pyridylmethylsulfinyl)benzimidazole crystals/TAKEDA) 24-01-2008

1. The appeal is admissible.

2. State of the art - Prior disclosure of a poster

2.1 Document (1) is a true copy of a poster that was presented and inspected at the oral proceedings before the first instance (not contested by the appellant). It refers to the preparation and characterization of two polymorphs and two pseudopolymorphs of lansoprazole (see introductory part).

The preparation of the polymorphic A-form may be performed by three methods, one of them consisting in stirring solvated crystals of lansoprazole (ethanolate/hydrate) in water at room temperature for two hours, the crystals then being removed by filtration and dried in vacuum (see document (1) under "Methods: Preparation of polymorphic forms", point 2).

The various forms of lansoprazole, namely polymorphic A-form and B-form, and pseudopolymorphic forms are characterized by thermal analysis, FT-IR spectroscopy, X-ray diffraction and scanning electron microscopy (see document (1), under "Results and Discussion").

2.2 It is not contested that the method of preparation mentioned above falls within the process defined in claim 1 of the patent in suit and the board concurs with the relevant finding in the decision under appeal.

Hence, the question of whether or not that method was made available to the public before the priority date of the patent in suit by display of a poster identical to document (1) at the Edinburgh Congress, is crucial for deciding on the present appeal as the underlying decision of the opposition division is primarily based on a finding that a poster identical to document (1) was so displayed.

2.3 The first point is to decide on whom lies the burden of proof in these appeal proceedings regarding the public availability of a poster identical to document (1).

2.3.1 The respondent contended that since the opposition division on the basis of a correct evaluation of the evidence had decided to revoke the patent due to prior disclosure, the burden of proof lies with the proprietor of the patent to demonstrate that the reasons given by the opposition division for revoking the patent were not correct.

2.3.2 The respondent thus appears to suggest that as the opposition division had decided that a poster identical to document (1) was displayed at the Edinburgh Congress, and that hence the patent lacked novelty, by challenging the finding of lack of novelty in appeal proceedings, the appellant has the burden of proving that a poster identical to document (1) was not displayed at the Edinburgh Congress. In other words the appellant has to prove a negative.

2.3.3 Before the opposition division it is the opponent who bears the burden of proof as regards demonstrating that the patent does not fulfil the requirements of the EPC. Appeal proceedings do not result in a shift in the burden of proof in the way suggested by the respondent - that is towards a requirement that the appellant prove a negative. It is clear that in the case where the opposition division has revoked a patent, the appellant must argue before the board why this decision was wrong - (see Case Law, 5th ed. 2006, page 440, section 5.2, para.2). In the present case the appellant has so argued. Thus, the board is in a position to examine the issues of standard of proof required as regards prior public disclosure and then to evaluate the evidence on file.

2.3.4 Since according to the jurisprudence of the boards of appeal each party carries the separate burden of proof of any fact they allege (see T 270/90, OJ EPO 1993, 725), the burden of proof concerning the precise content of the poster displayed at the Edinburgh Congress lay and still lies with the respondent as the party which alleges the prior disclosure of a poster identical to document (1).

2.4 The second point is to decide the standard of proof to be applied in the present case.

2.4.1 According to the jurisprudence of the boards of appeal, in prior public use cases, where practically all the evidence in support of an alleged prior public use lies within the power and knowledge of the Opponent, the latter has to prove his case "up to the hilt" (see T 472/92, OJ EPO 1998, 161, point 3.1).

2.4.2 In this case the board holds that the standard of proof to be applied for ascertaining whether a disclosure was made available to the public through displaying a poster during a congress is the same as for public prior use. Therefore, the opponent has the burden to prove beyond reasonable doubt that a poster identical to document (1) was displayed at the Edinburgh Congress (see T 97/94, OJ EPO 1998, 467, point 5.1, and T 472/92; T 574/00 dated 9 June 2004, point 2.2).

2.5 Turning to the question of whether the evaluation of evidence by the opposition division was actually in conformity with the relevant case law, the following has to be considered:

2.5.1 Each of the three elements of a public prior disclosure, namely its date, the circumstances in which it took place and its content, has to be proven. In the present case the key issue is whether on the poster displayed at the Edinburgh Congress, the following text appeared in relation to the preparation of lansoprazole:

"solvated crystals of lansoprazole (ethanolate/hydrate) were stirred in water at room temperature two hours, then crystals removed by filtration and dried in vacuum."

This text appears on document (1).

2.5.2 As to the reasoning for the finding of the opposition division that a poster identical to document (1) was displayed at the Edinburgh Congress, the board observes the following:

The relevant content of document (1), the affidavits of the witnesses Ms Kotar-Jordan and Ms Kramar, documents (2) and (3), as well as their oral testimony (documents (8) and (9)), and the abstract, document (4), are not in contradiction which each other. However, that is not sufficient for meeting the required standard of proof.

2.5.3 With the exception of Ms Kotar-Jordan's testimony, document (8), none of this evidence refers to the critical passage of document (1). Ms Kramar, who was not involved in the design and the preparation of a poster for the Edinburgh Congress (second page of document (9)), could not remember the detailed wording of it (last page of document (9)). Document (4) is a supplement to the European Journal of Pharmaceutical Sciences which was distributed at the Edinburgh Congress (see document (6)). Although document (4) has the same title and lists the same authors as document (1), it does not refer to any poster and is silent on the method by which two of polymorphs and more pseudopolymorphs of 2-[[[3-methyl-4-(2,2,2-trifluoroethoxy)-2-pyridyl]methyl] sulfinyl]-1H-benzimidazole were prepared (see document (4), two first paragraphs). Furthermore, document (4) makes no mention of FT-IR spectra, the subject-matter of Figures 3, 4 and 8 of document (1).

Documents (2) and (3) do not enable the board to determine any features of the poster displayed at the Edinburgh Congress. The respective assertions of Ms Kotar-Jordan and Kramar set out therein, namely:

"I displayed during the afore-mentioned Congress the attached poster of KRKA, entitled "Study of Polymorphism of a Novel Antiulcer Drug" in a poster session for participants" and

"During the afore-mentioned Congress the attached poster of KRKA, entitled "Study of Polymorphism of a Novel Antiulcer Drug" was displayed in a poster session for participants of the Congress"

are unsubstantiated allegations.

2.5.4 Under these circumstances, documents (2) to (4) and the testimonies of Ms Kotar-Jordan and Kramar, documents (8) and (9), cannot, contrary to the statements in the paragraph bridging pages 6 and 7 of the Reasons of the decision under appeal, be considered as corroborative evidence supporting a finding that a poster identical to document (1) was displayed at the Edinburgh Congress. Equally, documents (4) and (2) and the oral testimony of Ms Kotar-Jordan, document (8) do not constitute an unbroken chain of evidence in respect of the fact in question (see page 11 of the opposition division's decision).

2.5.5 Hence, the finding of the opposition division rests exclusively on the testimony of Ms Kotar-Jordan, document (8). This remains so even when taking into account that she could produce, from her possession, the original of a poster of which document (1) is a copy, on which the critical method for preparing lansoprazole was expressly described in the relevant context set out. No independent evidence (in writing or by other persons) is available to support Ms Kotar-Jordan's testimony.

2.5.6 This does not mean that her written and oral testimony was per se insufficient. However, there must be good reasons for treating this evidence alone as having established the facts beyond any reasonable doubt.

2.5.7 As to these reasons the following is found in the decision under appeal:

"Mrs Kotar-Jordan has clearly stated that it was the original poster, which was presented at the Congress. She was also the appropriate person to testify on this issue since she drafted the poster and was in the possession of the poster all the time. The Opposition Division does not have any indications which would put doubts on the credibility of Mrs Kotar's testimony." (page 8)

and

"..both witnesses provided testimony on the basis of

their personal knowledge. They credibly answered the questions as to the date, the subject and the circumstances for the presentation to the public of the poster D1. Under these circumstances, the Opposition Division is of the opinion that Mrs. Kotar's testimony could only then be refused, if it thought, that Mrs. Kotar was lying. However, the Opposition Division does not have any indications for this" (page 10).

2.5.8 The Board finds this reasoning to be faulty. It starts by accepting as true the matter that has to be proven by the testimony whose credibility is to be evaluated. Furthermore, it is certainly not the case that a person only does not tell the truth if he is being dishonest. A person can be honestly mistaken in his recollection of an event, particularly if the event took place some time ago. It is worth noting that the Edinburgh Congress took place more than seven years before Ms Kotar-Jordan's affidavit, document (2), and more than eight years before she was heard as a witness. Hence, the board does not agree with the opposition division's conclusions that Ms Kotar-Jordan's testimony must be true as there was no evidence that she was lying.

2.6 It follows from these considerations that it has not been sufficiently proven, i.e. beyond all reasonable doubt, that a poster with the content of document (1) (in particular a specific method for preparing the form A of lansoprazole - see point 2.1, above) was displayed to the public at the Edinburgh Congress (or at any other place or date before the priority date of the patent in suit). Thus, the finding in the decision under appeal, that a poster identical to document (1), belongs to the state of the art within the meaning of Article 54(2) EPC, cannot stand. The conclusion of the board is based upon an objective assessment of the facts, evidence and arguments and did not necessitate a hearing of the witnesses by the board itself.

2.7 As the board held it appropriate and no party objected, to decide itself on this issue (Article 114(1) EPC), the further evidence adduced by the parties during the appeal proceedings has to be considered.

2.7.1 The invoice, document (13), is for the "Design for a poster on STUDY OF POLYMORPHISM ... 1 80 m in length" (see document (14)). Document (13) is dated 18 September 1996. This may be an indication that a poster with that title and dimensions was prepared for display at the Edinburgh Congress.

It should be noted that document (13) alone does not show that only one single poster was made by the Biro Stavek Company for the Edinburgh Congress. No conclusions can be drawn from document (13) as to the content of the poster, in particular whether it described the relevant method for preparing the A-form of lansoprazole. Therefore, the invoice is not, either on its own or in combination with other evidence adduced by the opponent, sufficient to establish the identity of document (1) with a poster displayed at the Edinburgh Congress (cf. point 2.5.3, above).

2.7.2 The same is true for Ms Kotar-Jordan's second affidavit, document (18). Here, she states that between September 1996 and August 1997 seven more posters, of which she was a co-author, were shown at international symposia, the majority of them relating to lansoprazole. This document cannot, therefore, serve as corroborating evidence in respect of the exact content of a poster displayed at the Edinburgh Congress.

2.7.3 Therefore, the further evidence submitted by the opponent does not persuade the board that a poster identical to document (1) was disclosed at the Edinburgh Congress.

2.7.4 That being so, the documents and affidavits submitted by the proprietor in support of his position on this issue need not be considered.

2.8 In conclusion, it cannot be established beyond reasonable doubt that the poster displayed at the Edinburgh Congress had the content of document (1); hence the content of document (1) is not considered as prior art under Article 54(2) EPC.

3. Novelty - document (5)

3.1 Document (5) discloses a method for the production of 2-(2-pyridylmethylsulfinyl)benzimidazole compounds of formula (II) by oxidation of the corresponding 2-(2-pyridylmethylthio)benzimidazole of formula (i) with hydrogen peroxide in the presence of vanadium compounds (see page 2, lines 27 to 62).

The desired compound (II) produced by the reaction described above is usually separated out as crystals from the reaction mixture, so that the crystals can be collected by filtration after decomposition of the excess of hydrogen peroxide remaining after the reaction by addition of an aqueous solution of sodium thiosulfate, but the crystals may also be collected by extraction with a solvent such as chloroform if necessary, followed by concentration.

The crystals thus collected can be purified if necessary by a routine method such as recrystallization and chromatography (see page 4, lines 24 to 30). According to Example 4, purification is achieved by multi-crystallisation steps using a mixture of ethanol-water.

3.2 The claimed subject-matter as defined in claim 1, differs from the disclosure of document (5) in that the solvate with water and C1-C6 alcohol is suspended, left standing or stirred in water. The respondent's contention that the wording of claim 1 did not exclude the use of ethanol in addition to water is at variance with a proper understanding of claim 1 and is unfounded.

3.3 In view of the above, the claimed subject-matter is novel over document (5) within the meaning of Article 54 EPC.

4. Request for remittal

4.1 At the oral proceedings before the board, the respondent requested the case to be remitted to the first instance in order to have his case examined by two instances in respect of inventive step, since this had not been discussed before the opposition division.

4.2 In exercising its discretionary power under Article 111(1) EPC, the board takes into account the circumstances of the case and the issue of procedural economy. Article 111(1) EPC does not give the parties the right to have each request examined by two instances (see T 133/87 dated 23 June 1988, not published, point 2 and T 5/89, OJ EPO 1992, 348, point 5.5).

4.3 The board observes, first, that the present situation has not changed with respect to that prevailing before the first instance, in that the claims according to the appellant's main request are those granted and document (5) was cited against inventive step in the statement of the grounds of opposition. The respondent was not, therefore, confronted with a fresh case. Furthermore, the respondent had to expect that inventive step could be discussed on the basis of document (5), since the main line of the appellant's attack was against the acceptance of the content of document (1) as forming part of the state of the art.

4.4 In view of the above and for reason of procedural economy, the board finds it appropriate to exercise its discretion not to allow the request of remittal submitted by the respondent so that the case can be expected to be ready for a final decision at the conclusion of the oral proceedings (see Article 15(6) of the Rules of Procedure of the Boards of Appeal, OJ EPO 2007, 536).

5. Inventive step

5.1 The board concurs with the parties that document (5) represents the closest prior art, since it relates to a method for producing 2-(2-pyridylmethylsulfinyl)benzimidazole compounds (see page 2, line 3 of document (5)).

5.2 Thus, starting from document (5), the technical results or effects successfully achieved by the claimed subject-matter are to be determined for defining the objective technical problem to be solved.

5.2.1 According to the patent in suit, when the process of preparation of 2-(2-pyridylmethylsulfinyl)benzimidazole compounds as described in document (5) is followed, water and ethanol can hardly be eliminated from the end product and the resulting crystals inevitably contain a fair amount of water and ethanol. These solvents turned out to be difficult to remove by vacuum drying without detracting from the stability of the compound (see page 2, paragraphs [0004]-[0006] of the patent in suit). Thus, although this is not stated in the text of document (5), the benzimidazole compound provided by the process of document (5) is a solvate containing one molecule of water and one of ethanol per molecule of benzimidazole (see page 2, lines 30-31 of the patent in suit).

5.2.2 Example 1 of the patent in suit relates to the production of 2-[[3-methyl-4-(2,2,2-trifluoroethoxy)pyridine-2-yl] methylsulfinyl]benzimidazole. In order to render clearer the distinguishing step between this example and the comparative example disclosed in the patent in suit, i.e. Comparative Example 1, the board has artificially divided the process of example 1 into three steps:

- 13.0 g of 2-[[3-methyl-4-(2,2,2-trifluoroethoxy)pyridine-2-yl] methylsulfinyl]benzimidazole monohydrate, monoethanolate obtained according to Reference Example 6 (see page 7 of the patent in suit), was dissolved in a solution heated at about 60ºC made of 75 ml of ethanol-water mixture (9:1) and 70 myl of 25% aqueous ammonia solution. Insolubles were removed by filtration and the filtrate was ice-cooled to give wet crystals of 2-[[3-methyl-4-(2,2,2-trifluoroethoxy)pyridine-2-yl] methylsulfinyl]benzimidazole monohydrate, monoethanolate solvate.

- The obtained wet crystals were suspended in 53 ml of water and the suspension was stirred. The emerged crystals were recovered by filtration, washed with water,

- and then were dried in vacuum to give 2-[[3-methyl-4-(2,2,2-trifluoroethoxy)pyridine-2-yl] methylsulfinyl]benzimidazole as white needles, having a water content of 0.01% and an ethanol content: 63 ppm.

5.2.3 The patent in suit also contains a Comparative Example 1 which differs from its Example 1 in that the second step of suspension in water is not carried out. Following this process 2-[[3-methyl-4-(2,2,2-trifluoroethoxy)pyridine-2-yl] methylsulfinyl]benzimidazole is obtained as white needles. Water content: 0.12%. Ethanol content: 360 ppm.

5.2.4 Comparative Example 1 of the patent in suit is not within the teaching of document (5) and, therefore, appears "at first glance" not to be a basis for determining the technical problem to be solved.

However, in the present case, Comparative Example 1 of the patent in suit derives from the product obtained according to Reference Example 6 of the patent in suit (see point 5.2.2), which is, in fact, Example 4 of document (5). Due to the drying treatment disclosed in Comparative Example 1 of the patent in suit (40ºC for 20 hours under vacuum), it is clear that the resulting product of Comparative Example 1 cannot have more water and ethanol than that of Reference Example 6, (i.e. Example 4 of the document (5)). Thus the product obtained according to Example 4 of document (5) contains at least as much water and ethanol as the 2-[[3-methyl-4-(2,2,2-trifluoroethoxy)pyridine-2-yl] methylsulfinyl]benzimidazole obtained according to Comparative Example 1 of the patent in suit.

It is also pointed out, that Comparative Example 1 and Example 1 of the patent in suit differ only in that in Example 1 the crystals were suspended in water. The noticeable effect obtained (see respective contents of water and ethanol in the solvate) can thus be attributed exclusively to this difference which distinguishes the closest prior art from the currently claimed subject-matter. Therefore, Comparative Example 1 of the patent in suit provides a basis for assessing if there is an improvement provided by the claimed process (see T 496/02, not published in the OJ EPO, dated 11 January 2005, point 4.5.2 and T 480/02, dated 28 March 2007, not published in the OJ EPO, point 3.6). Example 1 in that respect shows that a dramatic decrease of the water and ethanol contents in the 2-[[3-methyl-4-(2,2,2-trifluoroethoxy)pyridine-2-yl] methyl sulfinyl]benzimidazole is obtained by suspending the monoethanolate, monohydrate solvate in water as compared to Comparative Example 1 of the patent in suit.

5.2.5 The respondent contended that a single example in the patent in suit cannot be used to show that the technical problem underlying the present patent has been solved, since a salt is formed by addition of aqueous ammonia for the making of the solvate. The wording of the claims relates only to the free base and not the salts thereof. He also disputed the fact, that the claimed process solved the technical problem without using an aqueous solution of ammonia (see point 5.2.2, above).

5.2.6 However, the presence of a very minor amount of ammonia, i.e. 70 microliters of a 25% aqueous solution of ammonia (see point 5.2.2) when crystallizing the solvate does not mean, in the absence of evidence, that a salt of the compound is formed and if it is so, this salt would not be formed in a considerable amount.

Hence the respondent's contention that the technical problem alleged in the patent in suit was not solved in the absence of the aqueous ammonia solution is not substantiated.

5.2.7 Finally, the respondent submitted that the water content and the alcohol content in the final product are dependent of the drying conditions as mentioned on page 13, lines 22-29 of the originally filed description. Since the wording of the claims is silent on these conditions the respondent concluded, that the technical problem alleged by the appellant could not be solved over the whole scope of the claims.

5.2.8 Even if the board accepted that the content of water and ethanol varies depending on the conditions used in the suspending and drying procedures, this is a matter of the difference in the degree of desolvation. In view of the significant decrease of the amounts of water and ethanol (see Example 1 of the patent in suit and point 5.2.2) and in view of the upper limits mentioned for these amounts in claim 1, the board is satisfied that there exists a broad range of conditions for which the degree of desolvation is improved, so that the respondent's contention that, in the absence of the specification of the process conditions, the problem is not solved over the whole claimed area, is not convincing.

5.2.9 In view of the above, the technical problem addressed and successfully solved by the claimed subject-matter in view of document (5) as the closest prior art is to be seen in the provision of a process for the transformation of solvate crystals of compounds of formula (i) (see point II above) containing water and C1-6 alcohol, into crystals having a water content of not higher than 500 ppm and a C1-6 alcohol content of not higher than 200 ppm.

5.3 It remains to be decided whether or not the claimed solution is obvious in view of the prior art cited.

5.3.1 The prior art does not give any guidance to suspend a solvate hydrate and ethanolate of compound of formula (i) in water. Still less does it give any hint that such a process step would have solved the technical problem defined above.

5.3.2 Moreover, starting only from document (5), that is to say without knowing the teaching of the patent in suit, the person skilled in the art would not find therein any hint that the compounds obtained in this document are solvates and not pure crystalline forms of the 2-[[3-methyl-4-(2,2,2-trifluoroethoxy)pyridine-2-yl] methylsulfinyl]benzimidazole obtained after drying, since document (5) nowhere mentions the formation of solvates (see point 5.2.1 above). He would thus not even envisage to suspend the said crystals in water to solve the problem underlying the patent in suit.

5.3.3 The respondent argued that in Example 4 of document (5) the term "washed" embraced the features of claim 1, "stirred, suspended and left-standing". Since there was a general trend to remove any trace of solvent from the crystalline compound, it would have been obvious for the person skilled in the art to suspend crystals in water, all the more given the fact that the suspension of the crystals in water was a technique commonly used in the art. Hence, the obtained result is to be considered as a bonus effect.

5.3.4 The board observes that the term "washed" used in Example 4 of document (5) relates to the washing of the crystals with ice-cooled ethanol-water mixture (8:2). Therefore, already for this reason the argument of the respondent is at variance with the facts. Furthermore, the respondent's allegation that it was common general knowledge to suspend crystals in water is unsubstantiated. To the contrary, it is surprising that the suspension of a solvate hydrate in water causes the removal of the water contained in said solvate.

5.3.5 For these reasons, document (5) does not render claim 1 of the patent as granted obvious. It follows that the subject-matter of claim 1 involves an inventive step within the meaning of Article 56 EPC. The same applies to dependent claims 2 to 6.