Tuesday 9 October 2012

T 990/09 – In The Archives


This is an appeal againt the decision of the OD to revoke the opposed patent.

In what follows, the Board deals with the novelty of the main request on file.

The passage is insteresting because of the way in which the Board deals with Internet archive documents.

[6] Claim 1 is directed to a protein-free and serum-free medium for the cultivation of mammalian cells. The medium comprises an ultrafiltered soy hydrolysate, at least 40% of which has a molecular weight less than or equal to 500 daltons […]. As no endotoxin content is specified regarding the soy hydrolysate in claim 1, said technical feature is to be ignored for the novelty assessment.

[7] The [opponents] have argued that such a medium was described in document E4 and that, consequently, claim 1 was not novel. Document E4 which benefits from a priority date (25 September 2000) that is prior to the filing date claimed for the patent at issue (27 September 2000) is cited under the provisions of A 54(3) EPC.

[8] The priority document of the patent at issue fails to describe a medium comprising an ultrafiltered soy hydrolysate, at least 40% of which has a molecular weight less then or equal to 500 daltons. Indeed, only hydrolysates, at least 40% of which have a molecular weight from 200 to 500 daltons are described […]. Moreover, according to the priority document it is an essential feature of the medium that it has an endotoxin content of less than 500 endotoxin-unities per gram. Therefore, claim 1 of the main request is not entitled to its claimed priority date and the relevant date to be considered for the novelty assessment is the international filing date, i.e. 27 September 2000, which is later than the priority date of the post-published document E4. Consequently, document E4 belongs to the state of the art pursuant to A 54(3).

[9] The relevant passage of document E4, relied on by the [opponents], is Example 1, which describes a medium referred to as the ‘SF-medium’, used for the cultivation of Vero cells […]. This medium consisted of a mixture of DMEM, Ham’s F12 and L-Gln which was supplemented with a solution of one of two protein hydrolysates referred to respectively as the ‘hy-soy/UF, Quest 5X59100’ and the ‘hy-pep 1510, Quest’ […]. The content of a deep frozen ampoule of Vero cells – which are mammalian cells – was thawed and added to 9 ml of SF-medium. After centrifugation for 10 min at 1000 rpm, the pellet was resuspended in SF-medium, transferred to a Roux bottle and incubated at 37 ºC and 7% CO2 for at least 15 minutes. Therefore, the [opponents] have argued, that document E4 clearly and unambiguously describes a protein-free and serum-free medium for the cultivation of mammalian cells comprising a protein hydrolysate.

For the Board it remains to be assessed whether a skilled person at the priority date of document E4 was in a position to establish that one of the two protein hydrolysates was an ultrafiltered soy hydrolysate, at least 40% of which had a molecular weight less than or equal to 500 daltons.

[10] The question to be answered is whether there is sufficient evidence on file to convincingly establish, by applying the balance of probabilities as the standard of proof, that an ultrafiltered soy hydrolysate, at least 40% of which having a molecular weight of less than or equal to 500 daltons, was available to the public at the priority date of document E4 under one of the two designations ‘hy-pep 1510’ and ‘hy-soy/UF’.

[11] An answer to this question is provided by the two product information web pages from ‘Quest international’ of document E68, bearing the address ‘http://web.archive.org/web/20000308080227/www.sheffield-products.com/products/5...’ (emphasis added by the Board). As derivable from the address, and as argued by in particular [opponent 1] at the oral proceedings, said web pages were archived on 8 March 2000. This has not been contested by the appellant which at the same oral proceedings has similarly indicated that the another web page of document E68 bearing the address ‘http://web.archive.org/web/20030518162635/http://www.sheffield-products.com/phar...’ (emphasis added by the Board) was archived on 18 May 2003.

[12] As the two product information web pages mentioned in point 11 above were archived on 8 March 2000 they were made publicly available on the web at a date prior to the priority date of document E4. They describe ‘HyPep 1510’ as an ultrafiltered soy hydrolysate, 25,4% of which having a molecular weight less than 200 daltons and 57,5% of which having a molecular weight comprised in the closed interval of 200 to 500 daltons.

[13] Therefore, it can be concluded that an ultrafiltered soy hydrolysate, at least 40% of which having a molecular weight less than or equal to 500 daltons, has been commercialised by ‘Quest international’ before the priority date of document E4.

[14] The [patent proprietor] has argued that it was possible that the trademark ‘HyPep 1510’ had been used by ‘Quest International’ to designate a protein hydrolysate having different technical features. It has also contended that ‘Quest’ could have prepared a special protein hydrolysate to serve the only purpose of providing the SF-supplement of document E4. In the absence of any supporting evidence, theses arguments are not tenable.

[15] Additionally also document E79 was considered to be highly relevant for the novelty assessment. This document is a declaration, dated 27 November 2008, from Anthonie Kunst, the then R&D Director Proteins of Kerry Bio-Sciences, a company which was the successor of ‘Quest International’ (Quest).

According to this declaration, ‘HyPep 1510’, which is an ultrafiltered soy hydrolysate (see point 8 of E79), has been commercially available from ‘Quest’ and its successor since 1997 (see point 5 of document E79). Document E79 also states that the ‘HyPep 1510’ was available in 1999 and earlier in a form wherein 82,9% of it had a molecular weight of less than or equal to 500 daltons (see point 7 of document E79). Whereas document E79 is silent as to the molecular weight distribution of the ‘HyPep 1501’ available in 2000 - because the declaration was made in response to questions restricted to the ‘HyPep 1510’ sold prior to September 1999 - (see point 4 of E79), there is no evidence on file that the composition of ‘HyPep 1510’ produced between September 1999 and the priority date of document E4 (25 September 2000) had changed. While it is true that the catalogue number used along with the commercial designation ‘HyPep 1510’ has varied over the time (‘IPL:5X59053’ on the web page archived on 18 May 2003 of document E68 and ‘IPL:5Z10493’ on the web pages of the same document archived on 8 March 2003), this does not necessarily mean that the product has changed. It could also represent a change in the presentation of the product only. In view of the evidence on file, the Board concludes that document E79 rather confirms that ‘HyPep 1510’ described in the two product information web pages of document E68 archived on 8 March 2000 (see point 11 above) is the same as the one referred to in document E4.

[16] The appellant has argued that in the present case the ‘up to the hilt’ standard of proof should be applied rather than the standard of the balance of probabilities. The Board disagrees and asserts that the ‘up to the hilt’ standard, which was first developed in decision T 472/92, is to be applied for cases of public prior use where practically all the evidence in support of an alleged public prior use lay within the power and knowledge of the opponent(s) (see the Case Law, 6th Edition, 2010, Chapter VI.H, Section 4.3.1, page 558).

[17] In view of the above remarks, the Board decides that document E4 describes a medium according to claim 1, which consequently lacks novelty. Therefore, the main request does not comply with the requirements of A 54(3).

Should you wish to download the whole decision, just click here.

The file wrapper can be found here.

NB: This decision has already been presented on Le blog du droit européen des brevets (here).

3 comments:

DrZ said...

Not a massive surprise as the Guidelines since April 2010 have stated internet disclosure are measured on the balance of probability (Part C Chapter IV 6.2 ff.)

DrZ said...

The URL using a RESTful type format would (in my mind at least) put the date as determined by the board on the balance of probabilities.

I'm of course assuming that it can be easily verified that the web archive in question used the same dating structure consitiently

or said...

Yes, DrZ, I think there is some consensus that T 1134/06 had gone too far when it adopted the “beyond all reasonable doubt” standard.