The Government of the United States of America, as Represented by the Secretary, Department of Health and Human Services v the University of Queensland and CSL Limited
[2005] APO 33
•13 July 2005
ABSTRACTS OF DECISIONS
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Application : No. 683220 in the name of The Government of the United States of America, as Represented by the Secretary, Department of Health and Human Services
Title: Self-Assembling Recombinant Papillomavius Capsid Proteins
Action: Opposition under section 59 by The University of Queensland and CSL Limited.
Decision: Issued 13 July 2005.
Abstract
The opponent did not establish any section 40 deficiencies with the claims nor that any of the claims lacked novelty or inventive step in light of any of the citations raised in their evidence.
While this concludes my findings in relation to the issues raised by the opponent, there is also a related application (688759) which is also currently under opposition by the same opponent. This related application contains similar subject matter as the opposed specification. The current specification (683220) was introduced into the related opposition under regulation 5.11 as a whole of contents novelty citation. The applicant of the related application (688759) in addressing this issue has challenged the priority date of the current application (683220).
If the current application is not entitled to its earliest priority date, then there is a serious question about the novelty of their claims in light of the related application with respect to HPV16.
Novelty issue in either 688759 or 683220 cannot be finalised until their respective priority dates can be determined. All parties in both oppositions were invited to provide submissions and a second separate decision will issue limited to this issue.
PATENTS ACT 1990
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Re:Patent Application No. 683220 by The Government of the United States of America, as Represented by the Secretary, Department of Health and Human Services; and opposition thereto under section 59 by The University of Queensland and CSL Limited.
BACKGROUND
Patent application 683220 was filed under the provisions of the PCT on 3 September 1993 in the name of The Government of the United States of America, as Represented by the Secretary, Department of Health and Human Services (hereafter "NIH") claiming priority from two US basic applications (941371 and 32869) filed on 3 September 1992 and 16 March 1993 respectively. The inventors were listed as Douglas R. Lowy, John T. Schiller and Reinhard Kirnbauer.
The Australian application was advertised accepted on 6 November 1997 and on 5 February 1998 an opposition was filed in the name of The University of Queensland and CSL Limited. The main evidence stages were completed by 17 December 2001. The opponent made two subsequent applications for leave to serve further evidence. The applicant objected to the second request but at a hearing on the objection (in August 2003) agreed to a small amount of material being included provided that they would be given the opportunity to respond if it appeared at the substantive hearing that this additional material was key to the decision.
Further evidence was finally completed on 11 December 2003 and the substantive hearing was held in Canberra on 19 July 2004. The applicant was represented by Mr David Catterns SC assisted by Glenn McGowan of counsel and Dr Mark Olive patent attorney of F.B. Rice & Co., Sydney. Nancy Ways Vensko, registered US patent attorney also attended on behalf of the applicant. The opponent was represented by Mr David Yates SC assisted by Mr Robin Kelly, patent attorney of Fisher Adams Kelly, Brisbane. Mr John Cox of CSL also attended on behalf of the opponent.
SPECIFICATION
The specification relates to papillomaviruses. These viruses infect the epithelia of a wide variety of species of animals (including humans), generally inducing benign and fibro-epithelial tumours or warts. However, in some cases, lesions can undergo malignant transformation and there is a strong association between cervical cancer and genital warts caused by certain human papillomavirus (HPV) genotypes particularly HPV16.
A vaccine which induces neutralising antibodies to particular papillomavirus genotypes could potentially prevent infection by the virus and possibly lower the rate of cervical cancer in women. However it has not been possible to generate in vitro the large stocks of infectious virus required to develop such vaccines. In particular, cultured cells containing papillomavirus only express oncoproteins and other non-structural proteins. Expression of the structural viral proteins, L1 and L2 (and subsequent assembly of infectious virus) occurs only in terminally differentiated layers of infected epithelial tissues.
Previous workers (notably Pilacinski et al 1984) had expressed the bovine papillomavirus (BPV) L1 and L2 genes in bacteria but found that neutralising antisera generated with such constructs had low titres (apparently because neither protein had the structural features of the native virion). Later workers (in particular Zhou et al, 1991) produced papillomavirus capsids in vitro by cloning HPV L1 and L2 genes into a vaccinia virus vector and infecting CV-1 mammalian cells. The expression products from this system were able to self-assemble into viral-like particles (VLPs) but the workers did not report any neutralising activity for these particles.
The present invention, according to the specification, is the discovery that the gene encoding the L1 major capsid protein of BPV or HPV16 has the intrinsic capacity to self-assemble into empty capsomer structures that closely resemble the intact virion under electron microscropic analysis or EMK (see example 4). The VLPs from BPV1 are also able to generate high levels of neutralising antibodies which can protect against against BPV1 infection (as shown in examples 5 and 6 and table 1).
The specification ends with 28 claims. Apart from the omnibus claims, the only independent claim is claim 1 which defines:
1. A genetic construct comprising a papillomavirus L1 conformational coding sequence encoding a papillomavirus L1 capsid protein and a recombinant vector, wherein said construct is capable of directing expression in a host cell of a capsomer structure having conformational epitopes, which are capable of inducing high titre neutralizing antibodies against native papillomaviruses, by self- assembly of said capsomer structure comprising said papillomavirus L1 capsid protein, with the proviso that said recombinant vector is a non-mammalian cell vector and said cell is a non-mammalian host cell".
The only other claims specifically referred to by the opponent at the hearing were claims 9-11 which are as follows:
9. The construct of claim 8, wherein said human papillomavirus L1 gene is a wild-type HPV16 L1 gene.
10. The construct of claim 1, wherein said sequence encodes the amino acid sequence of SEQ ID NO:2.
11. The construct of claim 1, wherein said sequence encodes the nucleotide sequence of SEQ ID NO:2.
Claim 9 is dependent on claim 8 and for completeness, I have also outlined this below:
8.The construct of claim 8, wherein said sequence is derived from a human papillomavirus L1 gene, a bovine papillomavirus.
KEY AREAS OF DISPUTE
Despite the large amount of evidence in this case, the parties confined their arguments to a small number of key points with helpful submissions provided both at (and after) the hearing. This was extremely useful as it consolidated the evidence and provided much needed focus for the hearing.
The opponent only pursued three grounds of objection to the grant of a patent:
i.All the claims lack novelty in light of Australian patent application 23666/92 (published as WO 93/02184 on 23 February 1993) and now granted as patent 651727 in the name of the University of Queensland and CSL limited. Jian Zhou and Ian Frazer are listed as the inventors of this citation which will be referred to hereafter as the "Frazer citation";
ii.Claim 9-11 are not fairly based because the claims were not limited to a particular wild type HPV16 sequence.
iii.The claims as a whole are unclear because the term "conformational coding sequence" is unclear.
I will discuss each of these objections in turn.
DECISION
Novelty
The test for novelty is the "reverse infringement test" as set out in Meyers Taylor Pty Ltd v Viccar Industries ( 1977) 137 CLR 228 at page 235 where Aickin J stated:
"The basic test for anticipation or want of novelty is the same as that for infringement and generally one can properly ask oneself whether the alleged invention would if the patent were valid, constitute an infringement."
Infringement is said to occur where "each and every one of the essential features of that claim have been taken" (Rodi and Wienenberger AG v. Henry Showell Ltd (1969) RPC 367).
A prior disclosure will only invalidate a claim if, after having read it, the skilled addressee would rather than could have produced all the essential integers of the claim. Thus, in General Tire & Rubber Co v Firestone Tyre & Rubber Co Ltd, (1972) RPC 457, the court noted at page 485:
"To anticipate the patentees claim, the prior publication must contain clear and unmistakable directions to do what the patentee claims to have invented ... A signpost, 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 destination before the patentee."
Frazer citation
The Frazer citation (outlined above) was published after the earliest priority date of the current specification. However, it is an Australian patent application with an earlier priority date (19 July 1991) than the currently opposed application. Therefore the citation is part of the "whole of contents" novelty prior art base [Patents Act 1990 section 7(1)(c)] and can be considered for novelty (but not inventive step) purposes.
The Frazer citation discloses a general method of making constructs of recombinant DNA molecules encoding the papillomavirus L1 or L1 and L2 genes and transfecting a suitable host cell to produce viral-like particles (VLPs) which could be useful as a diagnostic agent as well as forming a component of a vaccine. The citation taught that the L1 or L1 and L2 gene products were able to self-assemble into VLPs. The citation noted that the L1 protein alone is sufficient to form VLPs for bovine papillomavirus type 1 (BPV1), HPV11 and HPV6. However, HPV16 required both the L1 and L2 proteins (page 6, lines 15 et seq).
The Frazer citation analyses in detail VLPs of both HPV16 (example 1) and BPV1 (example 3). The BPV1 VLPs were found to closely resemble the native virion (page 32, lines 34-36). This was also true for the VLPs of HPV6 and HPV11 (page 30, lines 2-6). In contrast, the citation notes (page 15, lines 13-23) that HPV16 VLPs were smaller than the native virion (the diameter of the HPV16 VLP was 35nm - 40nm compared with a diameter of 50nm for the native virion).
The main examples in the Frazer citation use the vaccinia vector/mammalian cell expression system to produce VLPs, but the specification suggests that other expression systems could also be used including a baculovirus/insect cell system (page 6, line 28 et seq) and a yeast cell system (page 7, lines 1-3). One example (page 19, lines 23-30) uses the baculovirus/insect system to express the HPV16 L1 or L2 ORF. The example was only used to generate L1 or L2 recombinant proteins for use in the succeeding Western blotting assay (not capsomers) and the presence of VLPs was therefore not determined.
Did Frazer generally disclose immunologically correct HPV VLPs?
The applicant argued that the claimed VLPs in the opposed specification had two "limitations by result" which characterised the invention and distinguished the claims from the prior art:
1.the capsomer structure of the VLP had conformational epitopes;
2.the capsomer structure of the VLP was capable of generating high titre neutralising antibodies.
According to the applicant, the Frazer citation did not verify whether their VLPs had conformational epitopes nor that they could be used to generate neutralising antibodies. The citation therefore did not "direct, recommend or suggest with sufficient clarity" the claimed invention to deprive the claims of their novelty (Bristol-MyersSquibb Company v FH Faulding & Co 46 IPR 55, C. Van der Lely NV v Bamfords Ltd [1963] RPC 61 and Ramset Fastener's (Aust) Pty Ltd vAdvanced Building Systems Pty Ltd (1999) 44 IPR 481).
For the purposes of this decision, I’ll refer to a VLP which contains both the claimed limitations as “an immunologically correct” VLP. The opponent suggested that the Frazer citation disclosed the use of their VLPs as a vaccine (page 8, line 22-29). Because an effective vaccine would necessarily be immunologically correct, the opponent argued that it could be directly inferred from the citation that the VLPs described therein had those properties. According to the opponent, there is no difference in principle between an "express statement of fact and material from which the skilled worker would clearly infer the existence of that fact" for a novelty-destroying disclosure (referring to C. Van der Lely NV v Bamfords Ltd [supra]; Hoechst CelaneseCorp v BP Chemicals [1998] FSR 586 at 599-601; Ramset Fasteners (Aust) Pty Ltd v Advanced Building Systems Pty Ltd [Supra]).
However, the Frazer citation does not disclose an effective vaccine, it merely speculates that the VLPs might be useful as a vaccine. Of course, this would only be true if the Frazer VLPs were immunologically correct but there is no evidence of this in the citation. While Frazer's results indicated that the purified HPV16 VLPs were immunogenic, this does not mean that high titre neutralising antibodies would necessarily be produced in the immune response. Further, although the prior art cited in Frazer had indicated that antibodies to BPV structural proteins had virus-neutralising activity (see Pilancinski et al 1986, Ciba Found. Symp. 120: 136-156), there is no evidence that the same neutralising antibodies would be produced from the structural protein when in the form of a VLP let alone a high titre neutralising antibody (noting that the neutralising activity in Pilancinski was low in titre). I therefore do not accept the opponent's argument.
The opponent also argued that the Frazer VLPs could still have the same properties as the currently claimed VLPs even though this could not be directly inferred from the citation. The opponent noted that the only method disclosed in the Frazer citation for generating VLPs involves the self-assembly of L1 (or L1/L2) gene products in a suitable expression system. This single option provides the skilled worker clear directions to use that particular method to generate VLPs. If those VLPs inherently have conformational and neutralising epitopes, then the citation is an anticipation because, as noted in General Tire & Rubber Co v Firestone Tyre & Rubber Co Ltd [1972] RPC 457 at 485-486,:
"if in carrying out the directions contained in a prior inventor's publication will inevitably result in something being made or done which, if the claim of the opposed specification were a claim of a valid patent, would constitute and infringement of that claim, then that claim would have been anticipated".
The opponent never tried to replicate Frazer's work and directly verify in their evidence whether the VLPs produced in accordance with the instructions of the Frazer patent were immunologically correct. However, they suggested that later workers (including, they argued, the opposed specification) had used the Frazer methodology to generate VLPs from papillomaviruses. According to the opponent, these later workers had verified that the VLPs could induce neutralising antibodies, and their assumption was that the earlier Frazer work must therefore have also generated such VLPs. As a result, the opponent argued that the only advance proffered by the opposed specification over the Frazer citation was the verification of certain properties of the end-product VLPs but it is well established patent law that discovering new properties of a known substance will not confer novelty to the already known substance.
I note that the opposed specification (as did later workers) used a baculovirus-insect host expression system rather than the Frazer vaccinia-mammalian host expression system so the opponent's claim that later workers used the same methodology was overstated. However, I also accept that the "cross reactive" feature was also not something that the applicant has deliberately added to their VLPs. The applicant merely verified that the VLPs produced by the method disclosed in the opposed specification inherently have the "limitations by result" referred to above by the applicant. The key question in this opposition (and the related opposition on application 688759) is whether the Frazer methodology was the same as the method disclosed in the currently opposed specification and automatically (or inevitably) generated the same entity.
The applicant argued in their evidence that Frazer stressed the importance of linear epitopes rather than conformational epitopes in example 2 by analysing the linear antigenic regions of HPV16 L1 and mapping linear B epitopes. According to the applicant, the skilled artisan in following the teaching by Frazer would therefore inevitably look for VLPs having linear epitopes and be taught away from conformational epitopes.
However, this argument is clearly incorrect. There is no explicit teaching in Frazer to avoid conformational epitopes and no other technical reason for the skilled worker to use linear epitopes in preference to conformational epitopes. The evidence suggests that both conformational and linear epitopes can exist in the one protein antigen and a neutralising antibody could potentially be generated from either type of epitope. The latter type is much easier to identify because its amino acid sequences is contiguous. The skilled worker is likely to choose the easier type of epitope to study first when investigating the nature of the epitope and this is consistent with Frazer's approach.
In any case, the VLPs in the Frazer citation are generated by a particular self-assembly method. Prior knowledge about the nature of the epitopes will not affect the end-product of such a method. Thus, even if Frazer taught that linear epitopes were important as the applicant argued above, the skilled worker would not be able to use this information to alter the VLPs produced by the Frazer method. This can be distinguished from the case of Van der Lely NV v Bamfords Ltd [supra] which was referred to by the applicant. In that case, the disclosure was a photograph and it was imperative that the photograph clearly disclosed all the essential features of the invention for the skilled worker to produce a workable machine.
The applicant also argued that Frazer's failure to confirm the presence of conformational epitopes on the VLPs of the citation left open the possibility that these were not present. They supported this by referring to parts of their evidence which showed that not all PV vaccines contain native conformational epitopes (eg Lancaster, exhibit WDL6) and that not all VLPs having the appearance of native virions produced conformational epitopes that induced high titre neutralising antibodies (Shah September 12 2000 at paragraph 31 and exhibit KVS34). The applicant's evidence (see Lowy's declaration 27 October 2000 at paragraph 37) also suggests that visualising VLPs is not sufficient to determine whether they have the proper conformation as empty capsids often have different conformation than virions (see Kajgaya et al (1991) PNAS volume 88 page 4646 - exhibit DRL-11).
I accept that the applicant's evidence establishes that that there is problem predicting whether immunologically correct VLPs could be generated for a particular virus family. However, having determined that VLPs can be generated from a particular family, there is nothing in the evidence to suggest that there is the same level of unpredictability when it comes to extrapolating success within a particular virus family. Thus, the opposed specification having established that VLPs from a particular type of papillomavirus were immunologically correct could then confidently predict that successful VLPs could be generated from a full range of papillomaviruses using a wide range of expression systems. These types of predictions appear prima facie inconsistent with the applicant's "unpredictability" argument in relation to VLPs within a particular family. As a consequence, the applicant's "unpredictability" evidence only appears relevant to whether success in producing VLPs from papillomavirus could be inferred from the Frazer citation but not to whether the papillomavirus VLPs produced using the Frazer methodology were inherently able to induce neutralising antibodies (although this was not verified at the time).
I note that Frazer used a similar method to produce papillomavirus VLPs (apart from HPV16) which closely resembled the native virions. While the opposed specification used a baculovirus-insect host expression system rather than the Frazer vaccinia-mammalian host expression system (as did later workers who verified that the VLPs were immunologically correct), there was no suggestion in the evidence that the different expressions systems would have produced a different product or that a specific expression system was critical to success. In fact, the opposed specification concedes that any suitable expresssion system could be used to produce VLPs, including Frazer's vaccinia/mammalian cell system (see pages 10, line 33 - page 11, lines 1). The applicant also conceded at the hearing there was no selective difference in their baculovirus expression system compared with the Frazer vaccinia virus expression system.
Given these concessions, the lack of any evidence establishing that the expression system is critical, the otherwise close similarity of the methodologies of Frazer and the opposed specification and that the Frazer VLPs (for papillomavirus other than HPV16) were visually correct using electron microscopy, I am convinced that the papillomavirus VLPs produced by Frazer (except for HPV16) were the same as those produced using the methodology of the current specification.
Did Frazer disclose immunologically correct VLPs from HPV16?
Much of the evidence in this opposition concerned whether the HPV16 VLPs produced by Frazer were (or were not) defective and thus, whether they were also likely to contain conformational epitopes.
The Frazer citation reported difficulties generating VLPs from L1 of HPV16 and further noted that L1/L2 HPV16 VLPs were slightly smaller than the native virions. Originally, the opponent had thought the problem resided with the source of the L1 HPV16 DNA used by Frazer. In their evidence, the opponent suggested that Frazer has used the prototype (or Gissman) clone that had been isolated from a particular cervical carcinoma (Seedorf et al 1985). This differed from "wild type" HPV16 by a single point mutation gene at amino acid position 202 (His to Asp). The opposed specification found that when the L1 wild type HPV16 was used in their system, L1 alone was able to self assemble as a VLP with an efficiency comparable to the BPV L1 capsid protein (page 8, lines 24-31) and further that the capsids were the same size as native virions (page 13, lines 1-3).
Some time after the hearing (which itself was 12 years after the Frazer publication), the opponent resiled from their original position. They argued instead that Frazer's problem with producing VLPs of HPV16 was not because they used the prototype clone but rather because they used the incorrect codon to initiate transcription.
In the end, the exact cause of the problem is not critical to this decision. Whatever the explanation for Frazer's failure, the citation clearly indicates that there was a problem generating HPV16 VLPs. The opponent argued that the VLPs produced by Frazer contained a regular array of capsomers and although they were smaller, they had a similar appearance to native PV virions. Their assumption appeared to be that the HPV16 VLPs were "similar enough" to be able to generate neutralising antibodies.
.
I accept there is a possibility that Frazer's HPV16 VLPs were immunologically correct. The applicant's evidence was not convincing enough to prove the converse. However, a possibility that the HPV16 had the correct conformation is insufficient to prove a case of novelty. The opponent has to establish that it is "more likely than not" that the Frazer HPV16 VLPs were correct. In this regard, in contrast with the VLPs from other papillomaviruses, Frazer reports problems with generating the HPV16 VLPs. The citation also concedes that there may be differences in the appearance of the HPV16 VLPs. All of this raises serious questions about whether the Frazer VLPs have the same properties as those generated by later workers where the same problems have not occurred. The fact that the opponent is postulating different reasons why Frazer might have failed to generate HPV16 VLPs some 12 years after the event raises further questions about the reliability of the citation.
In these circumstances, the burden of proof falls squarely on the opponent to establish that the HPV16 VLPs were immunologically correct. An assumption that the HPV16 VLPs would necessarily have the correct conformation is not enough. Even visualisation comparison is insufficient given the applicant's evidence that this does not necessarily mean that the VLPs are immunologically correct. Given that Frazer had difficulties in producing the VLPs, and that the reasons for this difficulty are still not understood, it is not possible to determine what product was produced by the Frazer method and assess whether it was immunologically correct. My view is therefore that the opponent has failed to establish that the Frazer HPV16 VLPs were immunologically correct.
Are the claims novel in light of the Frazer disclosure?
I have concluded above that the Frazer has disclosed immunologically correct VLPs from papillomaviruses except for HPV16. The question now is whether in light of this finding, Frazer deprives any of the claims of their novelty.
Claims 1-8, 12-16, 23, 24 and 27 define genetic constructs capable of producing immunologically correct VLPs from the general class of papillomavirus in a non-mammalian expression system, host cells containing such constructs, and methods of generating capsomers using those constructs as well as the capsomers themselves. The Frazer citation disclosed each of these features although it did not explicitly teach that a non-mammalian expression system should be used. In the general description, it taught that any expression systems could be used including a baculovirus/insect cell system (page 6, line 28 et seq) and a yeast cell system (page 7, lines 1-3). However, there are no instructions to use a non-mammalian system over any of the other expression systems and its preferred method used a vaccinia/mammalian cell expression system.
My view is that a skilled worker reading the Frazer citation would be at least as likely (and probably more likely) to use a mammalian system expression system rather than a non-mammalian one. Therefore the citation does not provide "clear and unmistakable directions" to one of the features of the claims. As noted in General Tire & Rubber Co v Firestone Tyre & Rubber Co Ltd, (1972) RPC 457(at page 486):
"If ... the prior publication contains a direction which is capable of being carried out in a manner which would infringe the patentee's claim, but would be at least be as likely to be carried out in a way that would not do so, the patentee's claim will not be anticipated."
Having decided that the Frazer citation does not provide “clear and unmistakeable directions” to use a non-mammalian expression system, the key question then becomes whether that feature is an essential feature of the claims. The applicant advised at the hearing that they did not intend to argue a "selective advantage" based on the non-mammalian cell vector. As a consequence, they chose not to rely on this feature to distinguish the current claims from the citation. I do not accept this as a concession that the non-mammalian system is a "non-essential feature" of above claims. In fact, the specification clearly highlights an advantage in using the non-mammalian cells over mammalian cells which was to avoid the problems of occult viruses (see page 11, lines 1-3). This means that the feature of a non-mammalian expression system is an essential feature of the claims in which it appears. While there may be an argument about whether such an advantage was be obvious to the skilled worker (and therefore not a "selective advantage" as conceded by the applicant), this is an inventive step rather than novelty issue and is not relevant in the current case because the citation was published too late to be part of the prior art base for inventive step.
Having accepted that the non-mammalian expression system is an essential feature and that the Frazer citation fails to provide “clear and unmistakeable directions” to use a non-mammalian expression system, my view is that claims 1-8, 12-16, 23, 24 and 27 are novel in light of the Frazer citation.
Claims 17-22, 25-26, and 28 define diagnostics or vaccines using the papillomavirus VLPs. I have concluded above that it could not be inferred from the citation at the relevant date that the VLPs Frazer produced were immunologically correct. My view is that this is “a new and unsuspected property” of the known papillomavirus VLPs. It is well established patent law in such circumstances that a subsequent inventor is entitled to claim the new use (but not the known substance itself). It follows that claims 17-22, 25-26, and 28 which define the use of the VLP in a diagnostic or a vaccine are novel compared with the Frazer citation.
Claims 9-11 are specifically directed to HPV16 constructs. I have found above that the opponent has not established that Frazer citation discloses immunologically correct VLPs from HPV16. Claims limited to HPV16 are therefore novel in light of the prior art cited by the opponent.
Inventive Step
The Frazer group also published some of their work in the non-patent literature prior to the earliest priority date of the opposed specification in Zhou et al (1991) and (1992):
Zhou, J et al, (1991) Expression of vaccinia recombinant HPV16 L1 and L2 ORF proteins in epithelial cells is sufficient for assembly of HPV virion-like particles J. Virology 185:251-257 (Zhou 1).
Zhou, J et al, (1992) Human Papillomavirus Type 16 virions produced by a recombinant Vaccinia virus Virology 189(2) 592-599 (Zhou 2).
Therefore, in contrast to the Frazer citation, the Zhou articles are part of the prior art base for both novelty and inventive step. In fact, Zhou 1 was referred to as part of the admitted prior art of the opposed specification. In consolidating their arguments, the opponent reasoned that the key dispute between the parties was one of novelty rather than inventive step. They therefore did not pursue the ground of inventive step at the hearing.
For completeness, I reviewed the Zhou documents in my consideration of inventive step. I agree that they do not deprive any of the claims of their novelty or inventive step because they are restricted to Frazer's HPV16 work. Given the admitted problems with HPV16, the onus lay with the opponent to establish that those VLPs were immunologically correct. Their evidence was not sufficient to discharge this onus and reverse the burden of proof. Hence, the Zhou citations by themselves do not deprive any of the claims of their novelty. Further, it was not obvious at the relevant date what had caused the problems with HPV16. Hence, there was nothing to direct to worker on a particular course which might resolve those problems. As a consequence, the Zhou citations do not deprive any of the claims of their inventive step.
Regulation 5.11 - Patent application 688759
I note that there is another patent application (688759) relating to similar subject matter as the current specification which has also been opposed by the same opponent. The evidence in that case indicates that there were at least four separate groups of scientists working on VLPs in HPV at around the priority date of the current application - the Frazer, Hagensee, Rose and Kirnbauer (NIH) groups. Three of the groups went on to file patent applications in Australia with close priority and publication dates. I have summarised the details of these applications in the table below:
Number Inventive Group earliest priority date dates of other priority documents filing date publication dates 651727 Frazer (opponent) 19 July 1991 nil 20 July 1992 4 Feb 1993 683220 NIH or Lowy (Kirnbauer) -current applicant 3 Sept 1992 16 Mar 1993 3 Sept 1993 17 Mar 1994 688759 Rochester (Rose) 9 Mar 1993 7 Mar 1994 8 Mar 1994 15 Sept 1994
The opponent has opposed both the current application by NIH and the related application by Rochester. Neither Rochester nor NIH has opposed any of the other patent applications and the Frazer application has now been sealed as a patent in Australia. All three patent applications disclosed the generation of papillomavirus VLPs by expressing constructs of the papillomavirus L1 or L1 and L2 genes in a suitable host cell. In contrast to the Frazer patent application, both the NIH and Rochester applications reported that they were able to successfully generate HPV16 VLPs from L1 alone. They were also both able to confirm that their VLPs contained confirmational epitopes.
The opponent primarily relied on their own work (ie patent 651727) to argue a lack of novelty in both oppositions rather than raise the NIH or Rochester applications. However, in the related opposition, the current application became important to the novelty considerations because a publication date was questioned of a key non-patent literature citation authored by the inventors of the current application. It was therefore introduced into the related opposition proceedings under regulation 5.11. Both parties were provided with an opportunity after the hearing to provide evidence/submissions to address this new document.
Rochester did not attempt to distinguish their claims from the disclosure in the current (NIH) application (683220) in their submissions after the hearing but rather argued that the current application (683220) was not entitled to its earliest priority date. According to Rochester, the first priority document of the current application (filed on 3 September 1992) did not disclose which HPV16 clone should be used as the source material. They argued that the earliest possible priority date for 683220 in respect of HPV16 was therefore 16 March 1993 which is after the earliest priority date of the 688759 (9 March 1993). Of course, if this is correct, then potentially 688759 may deprive that current application of its novelty rather than the other way around, at least in regard to HPV16 (given that Rochester did not challenge the priority date of claims to papillomavirus strains other than HPV16).
As a result, it is clearly critical to determine the respective priority dates of 683220 and 688759 in relation to HPV16. Given that this issue affects both oppositions, my view is that all parties on both oppositions should have an opportunity to provide evidence or make representations on the following questions:
· What is the first valid priority claim to immunologically correct HPV16 VLPs in both 683220 and 688759?
· If the current application was found to lose its priority date, would the other application (688759) deprive the current claims of novelty with respect to HPV16?
The parties have 21 days of the date of this decision to provide comments on these specific questions. All parties will also be given an opportunity to respond to any submissions provided by the other parties. Once the submissions are complete, I will issue a common decision for both oppositions in relation to the specific questions raised above.
I advise that if the Rochester application is found to have an earlier priority date than the current application in relation to HPV16, I intend to raise the Rochester application under regulation 5.11 in relation to the current (NIH) opposition. Please note that the response to the question posed in the second dot point above will be the parties’ opportunity to give evidence or make representations in respect of the Rochester application as required under regulation 5.11(2)(c).
I also advise that in the event, an appeal is filed on the current decision or on the related case with regard to the other substantive matters determined in this decision, I still envisage continuing with determining the priority date of both applications in relation to HPV16 and the consequential issue of novelty. Once a decision is issued on the priority date consideration, my understanding is that any appeal on this decision can be joined to the earlier appeal and the Court will be able to consider all issues relating to validity of the patent application.
Section 40
fair basis of claims 9-11
The opponent suggested that the applicant's invention lay in the discovery that a particular wild type HPV16 L1 sequence will self assemble into a VLP with a previously unknown efficiency which was comparable to that of BPV1. The opponent argued that the claims should be limited to this specific genetic sequence for HPV16 L1 and not all wild type sequences which may not achieve the promise of the invention (as per AMP Inc vUtilux Pty Ltd (1971) 45 ALR 123 at 131). As a consequence, the opponent argued that claims 9-11 which encompassed sequences other than the specific gene sequence were not fairly based.
As I understand their argument, the opponent is suggesting that the only novel conferring feature is a particular sequence of HPV16 and the claims should therefore be limited to that sequence. However, this appears to be a disguised novelty objection rather than a legitimate fair basis argument. As noted in the recent High Court decision Lockwood Security Products v Doric Products Pty Ltd (2004) 62 IPR 461 (see paragraph 46 et seq), the "inventiveness" or "meritoriousness" of the technical contribution made by the specification are issues for novelty or inventive step and not fair basis. The High Court in that decision found that the test for fair basis instead requires an analysis of the specification to determine whether the invention as claimed is consistent with the invention as described in the specification (as properly construed).
The specification clearly describes a broad range of papillomavirus VLPs generated from different L1 sequences. It suggests there is an element of predictability which enables success of one papillomavirus gene sequence to be extrapolated to other sequences. In my view, this provides a “real and reasonably clear disclosure” of the invention claimed. There was no evidence from the opponent that there was any difficulty in generating VLPs from any number of wild type gene sequences. In that regard, the Gissman clone was the one notable exception rather than the rule. I therefore find that the claims and specification define and describe the same invention and that the opponent has failed to establish that claims 9-11 lack fair basis.
clarity of the term "conformational"
The opponent argued that the expression "L1 conformational coding sequence" is unclear in claim 1 because it is not clear what is intended to be qualified by the adjective "conformational". In particular, the opponent suggested it is not evident what distinguishes a coding sequence which is conformational form one which is "non-conformational" and that therefore the term should be deleted.
I note that, in a technical sense, coding sequences are generally not described as "conformational" as the opponent has suggested. However, despite this, none of the experts in this opposition had any difficulty in understanding the scope of the claims containing the term [including Frazer who raised this problem in his evidence (paragraph 43 Frazer 17 July 2001)]. Reading the phrase within the context of the claim as a whole, it is clear that a conformational coding sequence is one which encodes a protein containing a conformational epitope and I have no difficulty in understanding the claims or determining their scope. In my view, the claims are clear.
CONCLUSION
The opponent did not establish any section 40 deficiencies with the claims nor that any of the claims lacked novelty or inventive step in light of the Frazer citation.
While this concludes my findings in relation to the issues raised by the opponent, I note that there is also a related application (688759) which is also currently under opposition by the same opponent. This related application contains similar subject matter as the opposed specification. The current specification (683220) was introduced into the related opposition under regulation 5.11 as a whole of contents novelty citation. The applicant of the related application (688759) in addressing this issue has challenged the priority date of the current application (683220).
Of course, if the current application is not entitled to its earliest priority date, then there is a serious question about the novelty of the current claims which relate to HPV16 in light of 688759. That patent application was not raised by the opponent in the current opposition and regulation 5.11 will need to be invoked to admit the document. Until this issue is finally resolved, I am unable to complete my consideration of the novelty of the claims in respect to HPV16. I expect to issue a second decision in due course after both parties have been given an opportunity to comment on the issue of priority date of the claims to HPV16 in 688759 and 683220 and in light of the determination of priority date, whether one of these applications is a novelty citation for the other.
COSTS
As there is still an outstanding issue of novelty to be resolved, this is necessarily an interim decision. I will therefore defer my consideration of costs until all substantive issues have been completed.
Karen Ayers
Delegate of the Commissioner of PatentsPatent attorneys for the applicant : F.B. Rice & Co., Sydney
Patent attorneys for the opponent : Fisher Adams Kelly, Brisbane
0
2
0