Bresagen Ltd and St Vincent's Hospital v the Austin Research Institute - [2003] APO 48

OFFICIAL NOTICE
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Application : No. 711144 in the name of Bresagen Ltd and St Vincent's Hospital

Title: Materials and Methods for the Management of Hyperacute Rejection in Human Transplantation

Action: Opposition under section 59 by The Austin Research Institute

Decision: Issued 06 November 2003.
Abstract

All claims found to be novel and inventive. While there were citations which explicitly taught the skilled worker to produce a knockout construct and animal for the gal T gene, they did not contain sufficient direction for the skilled worker to be able to put this teaching into effect. It was found that while the goal of producing a knockout construct and animal for the gal T gene was "obvious to try", the achievement of the goal was non-obvious.
However, the disclosure in the specification did not provide a "real and reasonably clear disclosure" of knockout constructs of the gal T gene other than in particular exons which the applicant had identified as suitable targets for homologous recombination. Similarly, claims involving methods of making knockout animals other than mice were not supported. The applicant had achieved the first step in making such animals but there was insufficient disclosure to enable subsequent steps to be performed which were necessary to produce such animals. The applicant was therefore not entitled to claim the entire field based on them reaching the first step.
PATENTS ACT 1990
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Re:Patent Application No. 711144 by Bresagen Ltd and St Vincent's Hospital and opposition thereto by The Austin Research Institute

BACKGROUND

Australian patent application 711144 was filed on 21 July 1998 as a divisional application of 15445/95 (696373) by Bresatec Ltd (who subsequently changed their name to Bresagen Ltd - hereafter referred to as Bresagen) and St Vincent's Hospital Melbourne Ltd (hereafter referred to as St Vincent's). Both the parent and current application claim priority from two US basic applications 188607 and 378615 filed on 27 January 1994 and 26 January 1995 respectively. The parent application was sealed on 26 November 1998 without a notice of opposition being filed.
The current application was advertised accepted on 7 October 1999, and on 7 January 2000 the Austin Research Institute (hereafter referred to as ARI) filed a notice of opposition. The evidentiary stages were completed on 4 February 2003 and the matter was set for hearing in Canberra on 25 and 26 June 2003. The applicant was represented by Ms Julia Baird of counsel assisted by Ms Robynne Sanders, solicitor and technical assistant from Peter Maxwell and Associates, Sydney. The opponent was represented by Mr Ben Fitzpatrick of counsel, assisted by Mr Mark O'Donnell, patent attorney of Blake Dawson Waldron Patent Services. Professor Ian McKenzie also attended most of the hearing on behalf of the opponent company.
SPECIFICATION
The specification concerns a method for the management of hyperacute rejection in human xenotransplantation. According to the specification, there is an acute shortage of human organs for transplantation. The use of xenografts (transplants between species) is one option for overcoming this short supply and the pig is the widely acknowledged beast of choice because its capillary size is similar to humans facilitating connection of xenografts to the human circulatory system. However, despite the occasional use in the past, immunological barriers have prevented the common use of viable xenografts. In particular, humans have an immediate, severe ("hyperacute") rejection (HAR) to pig organs that develops minutes to hours after transplantation.
The key epitope which causes this reaction appears to be galactose-a-1,3-galactose (the GAL epitope). This epitope is generated by an enzyme known as a-1,3 galactosyltransferase (the gal T enzyme). The specification suggested that if the gene encoding this enzyme (the gal T gene) could be inactivated or "knocked out" in a donor animal, that animal would not express the GAL epitope and the organs of that animal could be used more successfully in xenotransplantation.
The specification outlined general methods to generate such animals. From full length gal T genomic sequences, the applicants speculated that exons 4, 7, 8 and 9 would be suitable target regions for disruption of the gal T gene. The applicants chose the exon 9 site because it had a Sal I restriction enzyme site which enabled convenient manipulation of the gal T gene. The applicants inserted the genes coding for neomycin resistance (neo^R) or hygromycin resistance (hyg^R) into the Sal I site of gal T and inserted the disrupted gal T gene into the genome of mouse ES cells using the process of homologous recombination. From these cells, the applicants were then able to generate a transgenic chimeric mouse which was used to produce mice homozygous for the knocked out gal T gene using standard breeding techniques. The specification also suggested that a similar knockout strategy could be used for pigs.
The specification ends with 22 claims, 4 of which are independent. These are as follows:
1.A DNA construct when used for inactivating an a-1,3 galactosyltransferase gene, comprising a DNA sequence for porcine a-1,3 galactosyltransferase, wherein said DNA sequence carries a disruption of said DNA sequence such that functional a-1,3 galactosyltransferase is not encoded by said DNA sequence.
5.A DNA construct when used for inactivating an a-1,3 galactosyltransferase gene, comprising a DNA sequence for murine a-1,3 galactosyltransferase, wherein said DNA sequence carries a disruption of said DNA sequence such that functional a-1,3 galactosyltransferase is not encoded by said DNA sequence.
9.A method for generating a mammalian totipotent cell having at least one inactivated a-1,3 galactosyltransferase allele, said totipotent cell derived from a mammalian species having a functional a-1,3 galactosyltransferase gene comprising:
(a)providing a plurality of cells characterised as being totipotent cells of said mammalian species;
(b)introducing into said totipotent cells a DNA construct effective for inactivating said a-1,3 galactosyltransferase gene, said DNA construct comprising a DNA sequence for a-1,3 galactosyltransferase, wherein said DNA sequence carries a disruption of said DNA sequence such that functional a-1,3 galactosyltransferase is not encoded by said DNA sequence.
(c)identifying a totipotent cell or mitotic descendants of a totipotent cell having at least one inactivated a-1,3 galactosyltransferase allele.
15.A method for generating a non-human mammal lacking a functional a-1,3 galactosyltransferase gene, said mammal belonging to a species having a functional a-1,3 galactosyltransferase gene, comprising:
(a)providing a mammalian totipotent cell having at least one inactivated a-1,3 galactosyltransferase allele, said totipotent cell derived from a mammalian species having a functional a-1,3 galactosyltransferase gene;
(b)manipulating said totipotent cells such that mitotic descendants of said cell constitute all or part of a developing embryo;
(c)raising and breeding said neonate to obtain a non-human mammal homozygous for said inactivated a-1,3 galactosyltransferase and lacking the GAL epitope as determined by failure of somatic cells of said mammal to bind anti-GAL antibodies and IB4 lectin and resistance of said cells to lysis by human serum.
EVIDENCE
The evidence filed in support of the opposition comprises the statutory declarations of:
Julie Margaret Wilkie dated 5 July 2000 (with exhibits JMW-1-JMW-38);
Julie Margaret Wilkie dated 6 July 2000 (with exhibits JMW-39-JMW-40);
Julie Margaret Wilkie dated 9 August 2000 (with exhibits JMW-41-JMW-43);
Susan Sau Heng Wong dated 3 November 2000 (with exhibits SSHW-1-SSHW-2);
Dr Bruce Loveland dated 6 July 2000;
Professor Ian Farquar Campbell McKenzie dated 27 July 2000;
Dr Mauro Sergio Sandrin dated 6 July 2000 (with exhibits MSS-1-MSS-6);
Dr Joseph Albert Trapani dated 20 November 2000 (with exhibit JAT-1);
Dr Nicholas John Deacon dated 30 November 2000 (with exhibit NJD-1);
Dr Mauro Sergio Sandrin dated 3 November 2000 (with exhibit MSS-7);
Together with statutory declarations variously made by Carol Ellis Scott, Beverley Hore (2 declarations), Gretchen Sleeman, Kirsten Neilson (2 declarations), Margaret Pratt (2 declarations), establishing publication dates of certain documents.
The evidence in answer filed in the opposition comprises statutory declarations from:
Dr Allan Robins dated 31 August 2001 (with exhibits AR-1-AR-4);
Dr Richard Gordon Tearle dated 4 September 2001 (with exhibit RGT-1-RGT-2);
Dr Anthony John Frederick d'Apice dated 2 November 2001 (with exhibits AJFD-1-AJFD-7);
John Cooper Cox dated 8 January 2002
Professor Robert Charles Atkins dated 11 January 2002 (with exhibit RCA-1);
Associate Professor Emma Whitelaw dated 4 April 2002 (with exhibit EW-1);
Robynne Lyndsay Sanders dated 4 April 2002 (exhibiting redated copies of confidential exhibits not filed and RLS4).
The evidence in reply filed in the opposition comprises statutory declarations from:
Chris Cobbett dated 24 January 2003 (with exhibits CC-1-CC-2)
Alan Coleman dated 17 January 2003 (with exhibits AC-1-AC-4)
Professor Joseph F Sambrook dated 3 February 2003 (with exhibits JFS-1-JFS-3)
Dr Joseph Albert Trapani dated 3 February 2003 (with exhibits JAT-2-JAT-18).
GROUNDS OF OPPOSITION
The Statement of Grounds and Particulars listed 3 grounds of opposition:
1. Sub-section 59(a). The nominated person is not entitled to a grant of a patent for the invention because the alleged invention was communicated to the alleged inventors before the priority date;
2. Sub-section 59(b). The invention as claimed in claims 1-22 is not novel or inventive when compared with the prior art base.
3. Sub-section 59(c). The complete specification does not comply with the requirements of sub-section 40(2) or (3) of the Patents Act 1990
Some days prior to the hearing, the opponent advised the applicant that they would only be relying on the grounds of novelty and inventive step [section 59(c) - non-compliance with section 18(1)(b)(i) and section 18(1)(b)(ii)]. However, prior to the hearing, I raised a concern that many of the applicant's arguments about inadequate disclosure in certain citations might also apply to the scope of their own claims and that there may be a serious fair basis issue for them to address. In particular, whether the exemplified "knockout" mouse with a disruption in exon 9 of the gal T gene would support claims:
·to all knockout constructs in the porcine and murine gal T gene;
·to methods of generating knockout totipotent cells (or non-human mammals) for the gal T gene in species other than the exemplified mouse.
Neither the applicant nor the opponent was prepared to address this issue at the hearing. Nevetheless I considered the issue to be of importance and I advised the parties of my intention [pursuant to section 60(3)] to consider the issue in my decision. The applicant was allowed 1 month after the hearing to provide submissions and the opponent provided 1 month to respond. Both parties provided submissions as directed. The applicant then provided further submissions in response to the opponent's submissions. All these submissions are referred to in the decision where relevant.
DECISION
Novelty
The closest related art cited by the opponent was:
P.L. Dubowski, H.A. Vaughan, I.F.C. McKenzie and M.S. Sandrin "Characterisation of a cDNA clone encoding the pig a-1,3 galactosyltransferase: Implications for Xenotransplantation" Transplantation Proceedings, vol 25, no 5 (October), 1993 page 2921 (JMW-3)
P.L. Dubowski, H.A. Vaughan, I.F.C. McKenzie and M.S. Sandrin "Isolation of a cDNA clone encoding the pig a-1,3 galactosyltransferase" Transplantation Proceedings, vol 26, no 3 (June 1994), page 1335 (JMW-5) - note that although the abstract is published after the priority date of the current application, the information contained in the Abstract was orally disclosed at the Second International Congress of Xenotransplantation in Cambridge, UK in September 1993 (JMW-5)
D.K.C. Cooper, E. Koren, and R.Oriel "Genetically Engineered Pigs" The Lancet vol 342 (September 11, 1993) (JMW-6)
"Anti-pig IgM antibodies in human serum react predominantly with Gal(a-1,3)Gal epitopes" Proceedings of the National Academy of Science USA, (PNAS) 90: 11391-11395 (December 1993) (JMW-7)
Australian Patent Application AU-A-62792/94 by Austin Research Institute filed 15 March 1994, earliest priority date 16 March 1993 published 29 September 1994 (JMW-9)
M.S. Sandrin, H.A. Vaughan, P.L. Dubowski, and I.F.C. McKenzie "Studies on Human Naturally Occurring Antibodies to Pig Xenografts" Transplantation Proceedings, vol 25, no 5 (October), 1993 page 2917-2918 (JMW-27)
H.A. Vaughan, P.L. Dubowski, I.F.C. McKenzie and M.S. Sandrin "Biochemical Analysis if Pig Xenoantigens detected by Human Antibodies" Transplantation Proceedings, vol 25, no 5 (October), 1993 page 2919-2920 (JMW-28)
Each of these citations was published before the earliest priority date of the current application except for JMW-9. This document is an Australian patent application which, while published after the earliest priority date of the opposed application, has an earlier priority date than the opposed application and is therefore part of the prior art base of section 7(1)(c) for novelty ("whole of contents" novelty).
Five of the six disclosures relied on by the opponent are by the opponent themselves. Further, most of the citations were published within 5 months of each other. It is not surprising therefore that all of the disclosures are very similar. The opponent concentrated on the most detailed disclosures in JMW-3, JMW-7 and JMW-9 to present their case but their arguments could equally apply to any of the above documents. For convenience, I will refer to the documents as a whole except where there are specific issues concerning a particular document or a particular sub-set of documents.
All the documents relied on by the opponent disclose the identification of the GAL epitope as the major cause for HAR in xenotransplantation. They also disclose that the enzyme, gal T is responsible for the generation of this epitope. While none of the documents have generated knockout constructs of the gal T gene, some of the documents (JMW-3, 5, 6, 7, 9) explicitly suggested this as a potential approach to solve the problem of HAR. According to the opponent, this suggestion was "clear and unmistakable directions" (as per Flour Oxidising Co Ltd v Carr & Co Ltd (1908) 25 RPC 428) to produce a knockout donor animal. They argued that given this suggestion, a skilled worker armed with their common general knowledge could produce a knockout construct, cell or animal without the need for any further invention and hence there was an "enabling disclosure" for the purposes of novelty (Smith Kline and French Laboratories Application [1968] RPC 415 and Nicaro Holdings v Martin Engineering Co, 91 ALR 513 (at 531).
However, none of the citations actually produce a knockout construct or animal nor do they provide specific details about how either one could be achieved. The most detailed of the citations (JMW-9) discusses general methods for generating knockout animals (see page 8, lines 8-19) but it does not suggest which sites should be targeted within the gal T gene nor does it contain any genomic DNA sequence information for the gal T gene which would be needed for generating DNA constructs for homologous recombination. Such specific detail was also missing from the other citations. It would have required considerable work to obtain this detail, as the opponent acknowledged, and this detail was important to work the invention.
Given this, it is difficult to accept that the citations are the same for the "purposes of practical utility" (as per Hill v Evans (1862) 6 LT 90). While Nicaro Holdings v Martin Engineering Co, [supra] suggests that Hill v Evans [supra] does not require a literal disclosure and something less may suffice, it is well-established law that:
"a sign post, however clear, upon the road to the patentee's invention will not suffice. The prior inventor must be clearly shown to have planted his flag at the precise invention before the patentee"[General Tire and Rubber Company v Firestone Tyre and Rubber Co [1972] RPC 457 (at 485, 486)].
Having limited specific details of how to generate a knockout animal for the gal T gene, the skilled worker would have understood the "suggestion" to produce a knockout animal contained in the citations as an indication of the direction that they should proceed for future research (ie a signpost) rather than as a "clear and unmistakable direction" to produce the knockout animal.
I therefore conclude that none of the citations provided by the opponent deprive any of the claims of the opposed specification of their novelty.
Inventive Step
Statutory framework for inventive step
Section 18(1) of the 1990 Patents Act requires that a patentable invention (so far as claimed in any claim) has to contain an inventive step when compared with the prior art base as it existed before the priority date of the claim. Further definition of inventive step is provided by sub-sections 7(2) and (3):
7(2) For the purposes of this Act, an invention is to be taken to involve an inventive step when compared with the prior art base unless the invention would have been obvious to a person skilled in the relevant art in the light of the common general knowledge as it existed in the patent area before the priority date of the relevant claim, whether that knowledge is considered separately or together with the information mentioned in subsection (3).
(3) The information for the purposes of subsection (2) is:
(a) any single piece of prior art information; or
(b) a combination of any 2 or more pieces of prior art information;
being information that the skilled person mentioned in subsection (2) could, before the priority date of the relevant claim, be reasonably expected to have ascertained, understood, regarded as relevant and, in the case of information mentioned in paragraph (b), combined as mentioned in that paragraph.
Who was the primary notional skilled person?
At its broadest level, the problem addressed by the specification was how to overcome the problems of HAR in xenotransplantation. The skilled worker primarily faced with this problem was a medical researcher interested in immunology, and in particular xenotransplantation.
Documents relied on by the opponent for the purposes of inventive step
The opponent relied on JMW-3, JMW-5, JMW-6 and JMW-7 for the purposes of inventive step. These documents were all published (or orally disclosed in the case of JMW-5) before the priority date of the opposed specification and all explicitly taught that generating a donor animal whose gal T gene was knocked out was one approach to eliminate the expression of the GAL epitope on the donor organs making them more suitable for xenotransplantation.

Were these documents part of the prior art base for inventive step?

The applicant argued that JMW-3, JMW-5, JMW-6 and JMW-7 were not part of the prior art base for inventive step because the opponent had not established that these articles were such that the notional skilled person in Australia before the priority date could reasonably be expected to have ascertained them as required by section 7(3) of the Patents Act [1990] outlined above (also see Dyno Nobel Asia Pacific Ltd v Orica Australia Pty Ltd [1999] FCA 1369).
I note that JMW-5 is an article in the journal "Transplantation Proceedings" and was not published until after the priority date of the opposed specification. Although the information in the article was also apparently orally disclosed at a meeting in Cambridge U.K. in September 1993 (ie before the priority date), I have no evidence that a skilled worker would have routinely attended international conferences outside of Australia and thereby would have ascertained the disclosure in JMW-5 before the priority date. As a result, I agree with the applicant that the opponent has not established that JMW-5 could reasonably have been ascertained by the skilled worker at the priority date and I note that the opponent never pressed this issue in their evidence or submissions. As a consequence, this publication does not comply with the requirements of section 7(3) and is not part of the prior art base for inventive step.

With regard to JMW-3, JMW-6, and JMW-7, the applicant provided detailed reasons from their expert witnesses John Cox and Associate Professor Whitelaw explaining why they believed the skilled worker would not have ascertained such documents at the priority date (their declarations dated 8 January 2002 and 4 April 2002, respectively). I note that neither Cox nor Whitelaw had any particular interest at the relevant time in the field of xenotransplantation (Cox's primary interest appeared to be in vaccine development and Whitelaw was concerned with eukaryotic gene expression in transgenic animals). I therefore do not believe that either of them have the expertise to assert what documents the notional skilled worker would have ascertained at the priority date.

Despite this, even if I accept the applicant's picture of the typical research worker in the correct field (and I note this was disputed by the opponent), I cannot agree with the applicant's conclusion that documents JMW-3, JMW-6 and JMW-7 would not have been ascertained at the priority date. In the current case, the problem being faced by the notional skilled worker was the avoidance of the HAR reaction in xenotransplantation. If the focus of research was as narrow as the applicant suggested, then the notional skilled worker would need be a person who was interested in that particular problem, in other words they would at least need to be a medical researcher with a particular interest in immunological responses in xenotransplantation.
Cox and Whitelaw's general picture of the type of approach taken at the time suggests that the skilled worker would have been scanning for articles in:

·prestigious wide interest journals such as Nature;
·well-regarded general journals in their broad area of interest; and
·highly specialised journals in their specific area of interest.

The skilled worker would have also at least been attending key conferences in Australia in their specific field of interest. Applying this general picture to the specific medical researcher with a particular interest in immunological responses in transplantation, my view is that the skilled worker would have at least scanned for articles of interest in prestigious journals and in well-regarded general journals in the field of medicine or immunology. They would have also been interested in highly specialised journals in their specific field of interest (transplantation) and would have attended key conferences in Australia in transplantation (and possibly immunology).

In this type of search, the skilled worker would clearly have ascertained JMW-3, JMW-6 and JMW-7. JMW-3 is an abstract from an oral disclosure given at the conference proceedings of the Transplantation Society of Australia and New Zealand held in Canberra on 5-7 May 1993. This appears to be the key Australian conference in field of transplantation and it seems highly likely that the skilled worker would have attended such a key conference. In fact, the applicant's own witness and one of the inventors of the opposed specification, Dr d'Apice, admits to attending this conference (paragraph 29 in his declaration of 2 November 2001). Even if the skilled worker was unable to attend the conference, they still would have been interested in reading the proceedings. In doing this, the title of the paper "Characterisation of a cDNA clone encoding the pig a-1,3 galactosyltransferase: Implications for Xenotransplantation" would have immediately alerted their interest and they would have ascertained the document.

JMW-6 is from the general and highly regarded British medical research publication "The Lancet". A researcher with a strong interest in an applied medical research area such as transplantation would have been scanning this particular journal for articles of general interest. The title "Genetically engineered pigs" would have been of immediately interest to a transplantation expert because pigs were widely acknowledged as the donor of choice in xenotransplantation and genetic engineering of pigs had so far not been reported in the literature. The skilled worker would therefore have ascertained the article.
JMW-7 is from the highly respected scientific journal Proceedings of the National Academy of Science USA (PNAS). Sambrook, in Evidence in Reply [paragraph 20(iii)] notes that PNAS is rated in the top 20 of all scientific journals in all fields and has a very high impact factor based on the number of times that articles in the journal are cited by other workers in the field in a specific time interval. This suggests to me that it is well-regarded general journal and would be scanned by the skilled worker if it covered that worker's broad area of interest. Whitelaw, for example, admitted that she scanned PNAS for articles in her own field of interest, genetics. It is very large (Cox notes that it contains 15,000 pages per annum) but it is also published reasonably often in small volumes and the articles inside are classified into general areas of interest for easy reference including, I note, Immunology. This would be an area of general interest to the notional skilled worker interested in the problem of immunological responses in xenotransplantation and my view is that the current skilled worker would have routinely scanned that section of PNAS at the relevant date for articles of interest. In doing this, the title of JMW-7 would have immediately alerted the skilled worker's attention and they would have ascertained the article.

Having ascertained the documents JMW-3, JMW-6 and JMW-7, the notional skilled worker would have had no difficulties understanding these documents and regarding them as relevant to their problem of avoiding HAR in xenotransplantation (a subsidiary point that was not disputed by the applicant). As a consequence, I find that the documents JMW-3, JMW-6 and JMW-7 satisfy the requirements of section 7(3) and are all part of the prior art base for inventive step.

Did the citations teach a skilled worker to produce a knockout animal?

I note that the citations JMW-3, JMW-6 and JMW-7 each explicitly suggested that HAR could be overcome with the generation of a knockout animal donor. Despite this teaching, the applicant argued that the skilled workers would have been encouraged to pursue other immunological based approaches to overcome the problems of HAR in preference to the knockout solution encompassed by the claims of the opposed specification. JMW-7 had referred to some of these techniques (as did JMW-6) and the applicant noted that much of the research taking place immediately before the priority date was directed to these alternative techniques.
I do not agree that the skilled worker would have chosen the immunologically based approaches of treating HAR to the exclusion of generating a knockout animal. As the applicant conceded (paragraph 22 of Cox), the immunological solutions all seek to treat the effects of HAR rather than its cause. This required on-going treatment for the recipient of the xenotransplant, a problem which could be avoided if the donor animal did not express the GAL epitope in the first place. Because of this, the skilled worker would have a strong motivation to generate a knockout donor animal. In that context, I agree with the opponent's view that the immunological approaches were not considered to be "alternative" solutions but rather short term solutions until the longer term goal of knockout animals could be achieved.
The applicant also suggested that most scientists would have taught the skilled worker away from that particular solution because they believed at the priority date that producing a knockout animal for the gal T gene was an "unobtainable and inherently flawed approach" (see Cox, paragraph 23) given that:
1.it was generally believed that the gal T enzyme was necessary (or vital) for the functionality of that species; and
2.it was unknown at the priority date whether there was only one gene involved in the production of the GAL epitope and whether targeting a particular gene was likely to be effective.
There was also a suggestion that the skilled worker would not necessarily follow the teaching of the citations because they did not know whether the HAR was exclusively caused by the GAL epitope and therefore did not know whether a knockout animal could overcome the problems of HAR.
I am also not convinced by any of these arguments. There was an explicit teaching to make a knockout mouse in JMW-3, JMW-6 and JMW-7, at least two of which were in highly regarded and peer reviewed journals. JMW-7 even describes steps the opponent had already taken to make a knockout mouse for the gal T gene (last paragraph column 1, page 11395). It seems to me that there wouldn't have been such explicit teaching and these steps would never have been undertaken if the skilled worker had been taught away from the knockout approach as the applicant alleged.

It is also difficult to accept the applicant's argument that the gal T gene was considered vital because the prior art citations failed to mention the potential problem and the opponent did not provide any contemporaneous documentary evidence to support their position. Cox alleged that the skilled worker thought that the gal T enzyme may have played a key role in fertilisation. However, even if this were true, while it might affect the reproductive capacity of a homozygous knockout animal, it does not seem to impact on the viability of an individual homozygous knockout animal and such animals could always be generated by breeding from heterozygote parents if they themselves were infertile. In my view, while the skilled worker did not know the exact role of the gal T enzyme, there was no reason why they would have thought it was vital especially given that it was absent in higher primates. Given this, my view is that while success may not have been guaranteed in the production of knockout constructs of gal T, there was still a reasonable expectation of success in this endeavour.

The applicant noted that around one third of the papers referred to by Sambrook reported lethality in knockout mice. They argued that this demonstrated uncertainty in the field and a low rate of expected success which would have taught the skilled worker away from the solution. However, many of the reported experiments were directed to making a gene knockout to determine the gene's functionality. In such circumstances, the high rate of mortality may be more indicative of the gene's importance rather than the technique's lack of success. This would not have taught the skilled worker away from the invention.

Further, with respect to point 2 above, while there was a theoretical possibility that there was more than one gene encoding the gal T enzyme in mammals, there was no evidence that a skilled worker believed there to be more than one gene nor that, in any case, this was an insurmountable hurdle to overcome in producing a knockout animal.

In my view, the citations JMW-3, JMW-6 and JMW-7 all had an explicit teaching to generate a knockout animal as a way of overcoming the problems of HAR. There was nothing in any of the citations or the common general knowledge which would have taught away from this explicit teaching. This is especially true for JMW-7 which disclosed that specific steps had already been taken towards implementing the solution. I find that all the citations contain a clear direction for the skilled worker to produce a knockout animal.

Was the skilled worker a team?

Because each of the citations JMW-3, JMW-6 and JMW-7 contained an explicit teaching to generate a knockout animal for the gal T gene, they have provided an answer to the broad problem faced by the specification of how to overcome the problem of HAR in xenotransplantation. The only part of the problem remaining for the specification to solve was how a knockout animal could be generated for the gal T gene. In order to solve this problem, the skilled worker would need to have both skills in xenotransplantation and the production of transgenic animals. The applicant contended that there was no one scientist in Australia at the priority date with both sets of skills.
I accept the applicant's argument that a transplantation expert was unlikely to have had skills in knockout technology (and vice versa) at the relevant date as the two fields were quite distinct. However, it is well-established law that notional skilled worker can include a team of workers and that one person does not need to provide all the requisite skills in such a team (see, for example Minnesota Mining & Manufacturing Company v Tyco Electronics Pty Ltd (2002) AIPC ¶ 91-823 and ICI v Lubrizol 45 IPR 577). The applicant argued that there was no team in Australia at the priority date with the combined requisite skills. However, this belies the existence of their own team which was a collaboration between experts in immunology/transplantation (from St Vincent's) and experts in the knockout technology (from Bresagen).

In any case, there does not need to be an actual team with the requisite skills because the notional skilled worker is a hypothetical being. In Minnesota Mining & Manufacturing Company v Tyco Electronics Pty Ltd [supra], the court suggested that the nature of the problem determined whether a team would be involved. As noted above, the only part of the problem remaining for the specification to solve is how to generate a knockout animal for the gal T gene. The skilled worker in xenotransplantation would have recognised that they themselves had no skills in knockout technology and would need to form a team with an expert in the knockout technology field to solve the problem.

In ICI v Lubrizol [supra] decision, Emmett J noted that the team would have to be "normally" employed in that art.[1] From the evidence in this case, it is clear that Bresagen was often looking for collaborative research opportunities to capitalise on their expertise in producing knockout animals. Both the opponent and St Vincent's Hospital had independently tried to contact Bresagen to attempt to establish a collaborative project. As a consequence, my view is that a team of experts from the xenotransplantation and knockout technology fields was "normal" at the priority date. The notional skilled worker in this case is therefore a team which includes experts from both fields and which is consequently also primed with all the common general knowledge of both fields (see Minnesota Mining & Manufacturing Company v Tyco Electronics Pty Ltd [supra]).

Were the general methods for generating knockout animals common general knowledge at the priority date?
[1] While this decision was appealed (ICI v Lubrizol 49 IPR 513), the Full Bench did not overturn the single judge's conclusions in regard to the team.

The opponent argued that the general methods of generating knockout constructs and animals were part of the common general knowledge by 1993 and provided substantial evidence to that effect. Sambrook in evidence in reply (paragraph 13), in particular, cited a large number of publications (130 papers) which demonstrated the successful use of the technology, around 85% of which had been published before the priority date.
Sambrook's papers suggest that the technology was first reported in the literature in 1987 (ie: around 7 years before the priority date of the current application). However the importance of the technology must have been immediately recognised because by 1989, there were general articles and reviews in very prestigious general journals and in a broad interest genetics journal:
Mansour, SL et al (1988) "Disruption of the proto-oncogene int-2 in mouse embryo-derived stem cells: a general strategy for targeting mutations to non-selectable genes" Nature 336: 348-552
Capecchi, MR (1989) "Altering the genome by homologous recombination" Science 244: 1288-1292.
Capecchi, MR (1989) "The new mouse genetics: altering the genome by gene targeting" Trends Genet 5(3):10-6
Sambrook's evidence also shows a fast acceptance of the technology because the number of papers applying the technology appeared to grow exponentially in the years between 1988 and 1993:

year Number of papers

1988 1

1989 6

1990 6

1991 10

1992 26

1993 62

In fact, according to Sambrook (and this was not controverted by the applicant) the methodology had been developed to such an extent that it had been published in 1993 as a standard protocol in laboratory manuals such as "Guide to Techniques in Mouse Development", Methods in Enzymology, volume 225, Academic Press; and "Gene targeting: A practical approach", Joyner Oxford University Press.
In my view, the number of publications of the method by 1993, and the fact that the method has been published in key review articles in widely read journals such as "Nature" and "Science" and in standard text books such as Methods in Enzymology, suggests to me that the general methods were well-known by 1993.

The applicant claimed that Sambrook overstated the common general knowledge because of his eminence in the field of genetic engineering and also because his declaration was made in 2003, which is well after the priority date of the opposed specification. However, I do not believe the criticism is well-founded. My view of the common general knowledge is based the number and type of publications available at the priority date which were statements of fact provided by Professor Sambrook and not controverted by the applicant.

As an aside, I also observe that the applicant's argument may be inconsistent with their own evidence. Cox, for example, expressed his opinion in detail on the difficulties and technical barriers involved in generating knockout animals at the priority date. However, if Mr Cox has properly based his opinion on his own knowledge of the field at the priority date, then he has demonstrated that he clearly understood at the priority date the methodology involved in knockout technology, despite having no practical experience in the field nor even any apparent research interest in the transgenic animal field. The proper skilled worker must have had all this knowledge plus all the practical skills in applying the techniques to a multitude of genes.

Having said that, I note that all of the documents cited by Sambrook on the knockout technology were limited to the mouse model and I note that at the priority date the knockout technology had not been tested in animals other than mice. Both parties conceded in their evidence that working with pig cells was difficult and despite the statement on page 26, lines 22-23 of the specification that ES cell lines had been derived for both mice and pigs, the evidence suggests that porcine ES cells were unavailable at the time of filing. Given the conceded difficulties with working with pig ES cells, my view is that the techniques involved in the knockout technology were well known as at the priority date but that they were limited to the mouse model system in so far as the production of knockout cells or animals generated therefrom.
Was there an invention in applying the known general knockout techniques to the specific gal T gene?
Having accepted that the techniques involved in knockout technology (using homologous recombination) were well known in a general sense at least in mice, and that the citations directed the skilled worker to produce a knockout animal for the gal T gene, the key question then is whether the techniques were so routine that it was non-inventive at the priority date to apply the technique in a specific instance to produce a transgenic animal with the gal T gene knocked out.
The opponent argued that it must have been straight forward to produce a knockout animal as the applicant themselves were able to quickly generate a knockout mouse once they had decided to embark on the project. However, even if this is true, the applicant's relatively quick success is not necessarily indicative of a lack of inventive step. As noted in Wellcome Foundation Ltd v V.R. Laboratories (Aust.) Pty Ltd 148 CLR 262 (at 285), the level of research and experiment made by a person claiming to be an inventor will not of itself prove (or, as in the current case, disprove) that an invention had been made:
"Invention will depend on the nature of the result ultimately claimed, whether product or process, viewed against the background of common general knowledge."

The applicant may have been extremely fortunate in the decisions they made and, with the benefit of hindsight, such decisions could appear obvious. This was certainly Professor Whitelaw's opinion of the applicant's work in evidence in answer (paragraph 135):
"I find that at each step of the opposed application, the inventors have been extremely thorough and clever in constructing a complicated construct using FRT to delete a gene in a knockout mouse. In total, this is not a course that any skilled worker would, at the priority date, have taken, there are too many unknowns and too many decisions to take, too many years of experiments that have a high likelihood, or probability of failure."
The applicant listed a multitude of decisions and choices (Cox's evidence in answer, paragraph 26) that they alleged they had to make to realise the invention, including:
1.cloning of the entire coding region of the porcine gal T gene. In undertaking this choice the inventors had to overcome obstacles including:
i.difficulties aligning the known murine and bovine sequences because exon 3 was absent in the bovine gene;
ii.after realignment, only one island of 20 base pairs showed the required homology;
iii.similar difficulties in alignment in the 3' region;
iv.a need to use a reduced annealing temperature for the 3' primer.
2.the selection of homologous recombination strategy including:
i.choice of exon to target;
ii.identification of a suitable restriction site (mouse)
iii.insertion of a Sal I site into porcine
iv.choice of a selectable marker:
v.NEO chosen as initial marker
vi.HYG chosen as second marker to permit selection of homozygotes
vii.Decision to excise NEO (and HYG) after homologous recombination.
The specification particularly highlighted the steps of cloning the porcine gal T gene and characterising the gal T cDNA sequences from murine, porcine and bovine genomes as being important. The applicant reiterated this in their submissions on fair basis. These steps were fundamental to the success of the project as they enabled the applicant to compare the murine, bovine and porcine sequences to determine the locations of the exons boundaries for the gal T gene and identify the areas of high homology in the genes to target for homologous recombination.
The opponent argued that the steps taken at each stage of the applicant's strategy were all part of the common general knowledge and did not involve an inventive step. While they conceded that there were many choices involved in applying the technology to the gal T gene, their view was that the knockout techniques were "robust" enough in 1993 for the skilled worker to apply the technology without requiring an inventive step.

I accept that all the individual steps taken by the applicant were part of the common general knowledge but they were also part of a myriad of other possibilities which the applicant might also have taken but which may not have been similarly successful. There were many steps between the conception of the goal of generating a knockout animal and its successful completion and at each step towards the goal, there were critical choices for the applicant to make without having a real idea of which one might ultimately lead to the success of the project. In that sense, the opponent's argument that the successful path taken by the applicant through the maze of possibilities is obvious is actually one of hindsight.

The opponent's argument also appears to be one of "obvious to try", a test which was recently ruled in the High Court to be an impermissible test for obviousness [see Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59 (12 December 2002)]. Of course, as noted in the High Court decision citing the US Court of Appeals decision in re O'Farrell 853 F 2d 894 (1988), it is not easy to decide whether a case falls foul of the "obvious to try" test because:
"Any invention that would in fact been obvious under §103 would also have been, in a sense, obvious to try. The question is: when is an invention that was obvious to try nevertheless non-obvious?"
I note that the US judge (Rich) then provided some guidance in relation to his question by referring to two kinds of errors which occur when applying the standard of "obvious to try":
"In some cases, what would have been "obvious to try" would have been to vary all parameters or try each of numerous possible choices until one possibly arrived at a successful result, where the prior art gave either no indication of which parameters were critical or no direction as to which of many possible choices is likely to be successful.......in others, what was "obvious to try" was to explore a new technology or general approach that seemed to be a promising field of experimentation, where the prior art gave only general guidance as to the particular form of the claimed invention or how to achieve it."
Applying Judge Rich's principles to the facts in the current case, I note that the prior art gave only general guidance as to how to achieve a knockout animal for gal T. There was no suggestion that all cloning strategies could have successfully isolated the porcine gene and insufficient guidance in the prior art to assist the applicant in devising which particular strategy would work for the cloning of the porcine gal T gene and the identification of the particular exons to target for homologous recombination. There was also nothing to indicate which parameters were critical and no real direction as to which of many possible choices was likely to be successful. In that sense, while it might have been obvious to try and aim for the goal of producing a knockout animal for the gal T, the steps required to successfully achieve this goal were non-obvious. All the claims are therefore inventive in light of the citations provided by the opponent.
Section 40 - fair basis
Having concluded that the applicant's invention is based on them making certain critical choices amongst a maze of possibilities, the question then becomes how much of the "maze" they can now claim. The applicant argued that having achieved their goal of generating one type of knockout construct/animal for the gal T gene, they were then entitled to claim any and all knockout construct/animals for the gal T gene. They referred to the Genetics Institute Inc v Kirin-Amgen Inc (1998) 41 IPR 325 decision to support their premise suggesting that claims could be expressed in general terms if:
i.There was a beneficial property common to the class [of proteins];
ii.The patent application disclosed a principle of general application; and
iii.The exemplified [coding sequence] enabled other products of the class to be produced.
I note that in Genetics Institute Inc v Kirin-Amgen Inc (no.3) [supra], Heerey J found that in providing a coding sequence, the patent application disclosed a (new) "principle of general application" which was the coding sequence for erythropoietin. Heerey's view was that this sequence enabled reliable probes to be produced which could then be used to generate other products of the class of erythropoietins.

In the current case, the applicant has disclosed sequence information from which they were able to determine the locations of the exons boundaries for the gal T gene and identify the areas of high homology in the genes to target for homologous recombination. In using this information to generate a knockout mouse, the opposed specification has demonstrated this as a "general principle". Further, the disclosure in the specification of one successful strategy from "the maze" of possibilities provides a sound base from which other strategies can now be designed. This provides an enabling disclosure for knockout constructs based on the exon principles disclosed in the specification regardless of the way they were made.

Having said that, I believe that the "principle of general application" does not extend to any and all knockout constructs of the gal T gene. The opposed specification teaches that not all sites in the gene are suitable targets for generating a knockout construct and a critical part of the applicant's invention was to identify potential areas to target of the gal T gene to produce a knockout construct. The specification specifically identifies 4 particular regions (exons 4, 7, 8 and 9) as potential targets for homologous recombination and does not indicate any other regions which might be useful in a knockout construct. In my view, this only provides a "real and reasonably clear" disclosure (as per CCOM Pty Ltd v Jiejing Pty Ltd 28 IPR 481) of knockout constructs involving exons 4, 7, 8 and 9. Claims 1, 3-5, 7-22 which encompass knockout constructs outside of these regions have no support in the specification and are therefore not fairly based.

I also note that the applicant only provided knockout constructs of the mouse gene but my view is that they can extrapolate these results to other species. Although there is inherent unpredictability in the field, there is a known principle that protein sequences are conserved between related species and the applicant is entitled to make a reasonable prediction based on that principle (see, for example, Genetics Institute, Inc v Kirin-Amgen, Inc (No. 3) [supra]). The evidence demonstrates that there is cross species homology of gal T genes and hence, the applicant can predict that the disruptions to particular exons could affect functionality of the gene in all species.

However, while I accept that there is sufficient information to make knockout constructs of the gal T gene from a range of species based on the disclosures with the mouse gene, I note that the applicant has also claimed methods of generating (all) non-human mammals lacking a functional gal T gene using the knockout constructs. The applicant argued that there was support for such claims in example 16 which provided a prophetic example of how a skilled worker might apply the technique to pig cells and in the fact that an important part of their inventive concept was the realisation that they could make and successfully breed a mammal with a disrupted non-functional gal T gene.

I am not convinced that the success in the model mouse animal can be similarly extended to methods which generate other animals/cells, particularly porcine, as currently claimed in claims 9, 12-15, 18-22 . Lord Hoffman in Biogen vMedeva [1997] RPC 1 acknowledged that there was more than one way in which the breadth of a claim might exceed the technical contribution to the art embodied in the invention, including by:

"claiming results which it did not enable, such as making a wide class of products when it enabled only one of those products and disclosed no principle which could enable others to be made. Or it might claim every way of achieving a result when it enabled only one way and it was possible to envisage other ways of achieving that result which made no use of the invention."

Success in the mouse model does not disclose a principle which would enable other types of knockout animals to be produced. The methods claimed are a broad "wish list" of the steps that are needed to achieve a successful knockout animal. However, they do not indicate which parameters are critical in achieving this success and there is insufficient direction in the specification to assist the skilled worker in devising a particular strategy which would work in an animal other than mouse. As the evidence on obviousness demonstrates, there are a multitude of choices to put the invention into practice in a particular animal model and the specification has not provided any real guidance about which choices are likely to be successful.

In addition, the common knowledge in the knockout technology was limited to the mouse model. Both parties conceded in their evidence that working with pig cells was difficult and despite the claims on page 26, lines 22-23 of the specification that ES cell lines had been derived for both mice and pigs, the evidence suggests that porcine ES cells were unavailable at the time of filing. The detailed methods for the isolation and culturing of ES cells (example 10, pages 63 - 75) therefore do not appear to directly translate to pigs and it appeared that prophetic example 16 has not been worked in practice.

In my view, the applicant, in producing particular knockout constructs and knockout mice, has achieved one step on the road to successfully producing other types of knockout animals. Each subsequent step on the road would involve a similar myriad of possibilities that the applicant faced in achieving their first step. There is no basis for applicant to claim the entire field based on their taking that first step. There is thus insufficient support in the specification to claim methods of producing knockout animals other than mice and claims 9, 12-15, 18-22 which are directed to methods to produce knockout cells or animals other than mice are not fairly based.
CONCLUSION
I have found that the invention as claimed in claims 1-22 is novel and inventive in light of the citations provided by the opponent but that claims 1, 3-5, 7-22 which omit the features of particular exons and claims 9, 12-15, 18-22 which are directed to methods producing knockout cells or animals other than mice are not fairly based. I believe that there is patentable subject matter in the specification and that the deficiencies noted above can be resolved by amendment. I therefore allow the patent applicant 60 days from the date of this decision in which to file appropriate amendments.
COSTS
Costs normally follow the event. In this case, I have found that the opponent has failed on all grounds they pressed but nevertheless there is still a major problem with the fair basis of the claims. In these circumstances, I do not think it appropriate to award costs.
Karen Ayers
Delegate of the Commissioner of Patents
06 November 2003
Patent attorneys for the applicant : Peter Maxwell and Associates, Sydney
Patent attorneys for the opponent : Blake Dawson Waldron Patent Services, Melbourne