Bionomics Limited v McGill University (Corrected Version No 2)
[2007] APO 6
•29 January 2007
CORRECTED VERSION – No. 2
ABSTRACTS OF DECISIONS
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Application : No. 18465/01 in the name of McGill University
Title: Loci for idiopathic generalised epilepsy, mutations thereof and method using same to assess, diagnose, prognose or treat epilepsy
Action: Request under Section 36 by Bionomics Limited
Decision: Issued 29 January 2007.
Abstract
Professor Berkovic found to be a co-inventor of the invention described and claimed in the current application and by virtue of assignment Bionomics Limited has co-entitlement to the invention.
The preliminary linkage analysis (in which Berkovic played a key role by identifying family A which was used in the studies) significantly narrowed the search for the gene to a particular locus which enabled the later isolation of the IGE gene. Even through Berkovic did not contribute to the later phases of the identification of the gene, his preliminary work with identifying family A had a “material effect on the invention” and entitles him to co-inventorship.
In contrast, Drs Mulley and Wallace’s contribution was after the final conception of the invention and means that they are not co-inventors.
PATENTS ACT 1990
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Re:Patent Application No. 18465/01 by McGill University and an application under section 36 for a declaration of an eligible person by Bionomics Limited
BACKGROUND
1. Australian patent application 18465/01 was filed in the name of McGill University (hereafter “McGill”) under the provisions of the PCT on 24 November 2000 claiming priority from a US basic application (60/167623) filed on 26 November 1999. The named inventors are Guy Rouleau, Ronald LaFreniere, Daniel Rochefort, Patrick Cossette and David Ragsdale. The PCT application has an international publication date of 31 May 2001 and entered the National Phase in Australia on 29 May 2002.
2. On 20 November 2004, Bionomics Limited (hereafter “Bionomics”) filed a request under section 36 of the Patents Act for a declaration under section 36(1) that they were an eligible person as defined by section 15 of the Act. Both parties provided evidence in support of their claim to entitlement. Each party was then provided an opportunity to respond to the other party’s claim. A round of further evidence followed and all stages of evidence were finally concluded on 14 July 2006 after which the matter was set for hearing in Canberra on 25 September 2006.
3. Mr Barry Hess SC of counsel instructed by Dr Clarissa Wynn and Dr Ken Finney patent attorneys from Cullen & Co, Brisbane represented the patent applicant and Mr Ben Fitzpatrick of counsel instructed by Mr Mark O’Donnell patent attorney from Madderns, Adelaide represented the section 36 requestor.
SPECIFICATION
4. According to the specification, epilepsy is one of the most common neurological conditions in humans affecting 1-2% of the general population. The disease is characterized by paroxysmal abnormal electrical discharges in the brain, which lead to transient cerebral dysfunction in the form of a seizure. Epilepsy can be “partial” (when the epileptic discharge is limited to part of one brain hemisphere) or “generalized” (when it involves both cerebral hemispheres at the onset).
5. Epilepsy can also occur in symptomatic, cryptogenic or idiopathic forms. In both symptomatic and cryptogenic forms of epilepsies, there is thought to be an underlying cause for the seizures. In symptomatic forms, the cause would be known (a brain injury, CNS infection or metabolic disorder) and in cryptogenic forms an underlying cause is suspected but unknown (hidden). In contrast, in idiopathic epileptic syndromes, there is no evidence of evidence of any cerebral dysfunction other than the seizures themselves and the neurological examination is normal. In these forms of the disease (particularly the generalized idiopathic epilepsy form or “IGE”), genetic factors are thought to be important in predisposing an individual to seizures. Unfortunately the most common genetic epilepsies patterns are not explained by segregation of a single autosomal gene and the general belief is that seizure susceptibility probably reflects complex interactions of multiple factors affecting neuronal excitability[1] [2]. This complexity means that it is very difficult to identify (and therefore isolate) genes which predispose or induce epilepsy.
[1] Andermann (1982) Multifactorial inheritance of generalized and focal epilepsy in: Anderson VE et al (eds) Genetic Basis of the Epilepsies. New York, Raven Press pp355-374.
[2] Ottman et al (1989) Seizure Risks in offspring of parents with generalised versus partial epilepsy Epilepsia 30: 157-161
6. Despite this, the specification notes that there are some specific epileptic syndromes which aggregate in some families according to more definable “monogenic abnormalities” and which therefore provide a “unique opportunity” to rapidly map genes that play a role in determining predisposition to seizures (see page 32, line 27- page 33, line 2). Previous genetic approaches have identified a total of six loci [3] involving three genes [4] in specific IGE syndromes. Interestingly all three genes are membrane ion channels, including a mutation in the Na+ channel (SCN) identified by Wallace et al [5].
[3] Greenberg et al (1988) Juvenile myoclonic epilepsy (JME) may be linked to the BF and HLA loci on human chromosome 6 Am. J. Hum. Genet. 31: 185-192.[4] Bievert et al (1998) A potassium channel mutation in neonatal human epilepsy Science 279: 403-406[5] Wallace et al (1998) Febrile Seizures and generalised epilepsy associated with a mutation in the Na+-channel (1 sub-unit gene SCN1B) Nat. Genet. 19:366-370
7. The applicant undertook genetic linkage studies using 190 markers on 41 members (including 21 affected individuals) of a large Australian family (family A) segregating an autosomal dominant form of IGE (also associated with the GEFS syndrome). The applicant mapped the IGE gene to a particular locus (ch 2q23-q31) on chromosome 2 in a region flanked by markers D2S156 and D2S311.
8. Using the known association of epilepsy with membrane ion channels, the applicant performed searches of public data bases for similar genes which mapped to the ch 2q23-q31 locus. These searches revealed a number of interesting genes including a cluster of alpha sub-unit Na+ channels (SCNA) which the applicant hypothesized might be the most likely candidate gene for epilepsy. To test this, the applicant compared DNA sequences of the SCNA genes from a number of affected and non-affected individuals finding a number of clinically significant mutations within the SCN1A, SCN2A and SCN3A genes which they thought were associated with the disease.
9. Functional studies confirming their hypothesis were presented in example 6 (at page 54). In this example, a particular mutation in SCN1A (D188V) introduced into rat clones significantly affected sodium channel function. Neither the particular mutation nor the functional studies were disclosed in the priority documents. The example acknowledges that the pathogenic mutation was found in the large Australian family (family A). It also notes that this amino acid is conserved in all sodium channels thus far identified in all species and observes that this particular substitution has not been found after 700 chromosome assessments (involving 150 epileptic patients and 200 control patients). The specification however notes that their results are consistent with work another group (published after the earliest priority date) who also found mutations in SCN1A (Thr875Met and Arg1648His) associated with epilepsy.[6]
[6] Escagy et al (2000) Nature Genetics 24(4): 343-345
10. The specification ends with 66 claims of which 14 claims are independent (claims 1, 8, 18, 19, 20, 38, 44, 45, 53, 58, 61-64). These claims variously define:
(a) methods for determining an individual’s predisposition to or presence of epilepsy by determining the genotype of their SCN1A gene (claims 1-17, 61-63);
(b) screening methods using the SCN1A gene to identify test compounds useful in the treatment of epilepsy (claims 18-37);
(c) purified human SCN1A nucleic acid molecules having particular nucleic acid sequences (claims 38-49); vectors/cells containing such molecules.(claims 50-52);
(d) an alpha sub-unit of a SCN1A sodium channel with a specific amino acid sequence (claims 53-56) and an antibody specific to an epitope of that alpha subunit (claim 57);
(e) a compound having a therapeutic effect on epilepsy identified by screening test compounds with a SCN1A gene or protein (claims 58-60); and
(f) a kit for determining a predisposition for epilepsy comprising particular oligonucleotide probes which hybridise to a human SCN1A protein (claims 64-66).
11. The most relevant claims for this ownership dispute are claims 1, 19, 38 and 58 which read as follows:
1. A method of determining an individual’s predisposition to epilepsy and/or development of epilepsy, as well as predicting this individual’s response to medication, said method comprising the step of determining the genotype of a SCN1A gene of the individual or of a DNA variant, equivalent, or mutation which shows a linkage disequilibrium therewith, thereby determining an individual’s predisposition to epilepsy, development of epilepsy and/or response to medication.
19.A method of using specific alleles of the SCN1A gene or a variant, equivalent, or mutation thereof which shows linkage disequilibrium therewith, to set up a screening assay for identifying compounds that modulate sodium channel functions and are useful for treating epilepsy or other neurological disorders.
38.A purified human SCN1A nucleic acid molecule comprising a polynucleotide selected from the group consisting of:
(a)a polynucleotide sequence selected from amongst SEQ ID Nos: 189-192;
(b)a polynucleotide sequence encoding a SCN1A polypeptide selected from amongst SEQ ID Nos: 1-2;
(c)a polynucleotide sequence selected from amongst SEQ ID Nos: 5-26 and 28-32;
(d)a polynucleotide sequence encoding a SCN1A polypeptide selected from amongst SEQ ID Nos: 3 or SEQ ID 4;
(e)a nucleotide sequence fully complementary to any of the SCN1A nucleotide sequences in (b) or (d);
(f)a nucleotide sequence fully complementary to any of the nucleotide sequences in (a) or (c);
(g)a nucleotide sequence at least 95% identical to any of the nucleotide sequences in (b) or (d) or (e);
(h)a nucleotide sequence at least 95% identical to any of the nucleotide sequences in (a) or (c) or (f); and
(i)a nucleotide sequence which hybridizes under high stringency conditions to any of the nucleotide sequences in at least 95% identical to any of the nucleotide sequences in (a) or (c), (f) or (h).
58. A compound having therapeutic effect on epilepsy or other neurological disorders, identified by a method comprising:
(a)providing a screening assay comprising a measurable biological activity of SCN1A protein or gene;
(b)contacting said screening assay with a test compound;
(c)detecting if said test compound modulates the biological activity of SCN1A protein or gene; and
(d)selecting a test compound of (c) which modulates the activity of SCN1A as measured by the screening assay in (a);
wherein a test compound which modulates said biological activity is a compound with said therapeutic effect.
SECTION 36 APPLICATION
12. Bionomics alleged that in mid-1992, Professor Samuel Berkovic (on behalf of Melbourne University) and Dr Guy Rouleau (on behalf of McGill University) formed a collaboration to identify the genetic loci of the mutation responsible for epilepsy in family A. As part of that collaboration, Professor Berkovic provided DNA samples from Family A which showed an autosomal pattern of inheritance for generalized idiopathic epilepsy (GFS+). According to Bionomics, these DNA samples were critical to the identification of the functional mutations in the SCN1A genes associated with eplilepsy.
13. Bionomics also alleged that Drs Robyn Heather Wallace and John Charles Mulley from the Women’s and Children’s Hospital, Adelaide provided critical information regarding a clinically significant mutation in the SCNA gene which McGill used to isolate the functional mutation associated with epilepsy in family A.
14. Bionomics argued that as a result of these inventive contributions, Berkovic, Wallace and Mulley should be named as co-inventors of the current application. Bionomics should then be named as a co-applicant of the current application by virtue of an agreement between Bionomics, the University of Melbourne and the Women’s and Children’s Hospital, Adelaide.
EVIDENCE
15. Bionomic’s evidence consisted of statutory declarations by Herbert Herzog (Herzog 1 26 April 2005 with exhibits HH-1 to HH-11), Francis David Placanica (dated 18 April 2005) and Samuel Frank Berkovic (Berkovic 1 dated 18 April 2005 with exhibits SFB-1 to SFB-15) filed in Evidence in support and Samuel Frank Berkovic (Berkovic 2 dated 28 June 2005 with exhibits SFB-16 to SFB-23) filed in evidence in response. They also filed further evidence by Samuel Frank Berkovic (Berkovic 3 dated 10 April 2006), Herbert Herzog (Herzog 2 dated 24 March 2006 with exhibits HH-13 and HH-13), Steve Wilton (dated 24 March 2006 with exhibits SW-1 and SW-2) and Mark Timothy O’Donnell (dated 1 June 2006 with exhibits MTO-1 and MTO-2).
16. McGill’s evidence consisted of declarations by Guy A. Rouleau (Rouleau 1 dated 15 April 2005), Ronald G. Lafreniere (dated 15 April 2005), Daniel Rochefort (dated 15 April 2005), David Ragsdale (dated 15 April 2005), Patrick Cossette (dated 15 April 2005) and Ian de Jonge (dated 18 April 2005 with exhibits IDJ-1 to IDJ-8) filed in evidence in support) and Lorne Andrew Michael Clarke (dated 13 October 2005 with exhibits LC-1 to LC-6), Thong Trinh (dated 13 October 2005 with exhibits TNT-1 to TNT-3) and Guy A Rouleau (Rouleau 2 dated 16 October 2005 with exhibits GR-1 to GR-13) filed in evidence in response. They also provided another declaration by Guy A Rouleau (Rouleau 3 dated 26 July 2006) in response to the section 36 requestor’s further evidence.
CHRONOLOGY OF EVENTS
17. From the evidence, the key events in the chronology of the dispute are as follows:
October 1990 Professor Berkovic who was at that time working at Austin Health, Melbourne (hereafter the “Melbourne Group”), contacted Dr Guy Rouleau at McGill to see if Rouleau was interested in collaborating with the Melbourne Group on a genetic study of a large multi-generational family (family A) showing an autosomal dominant pattern of inheritance of epilepsy (IGE).
Feb 92After a series of letters between the parties[7], the terms of the collaboration were finalized[8]
[7] See Berkovic 18 April 2005 and exhibits SFB-04, SFB-05, SFB-06 and SFB-07.
[8] See Berkovic 18 April 2005 and exbibit SFB-08.
Aug 92-Dec 93 The Melbourne group sent McGill cultured lymphocytes prepared from 60+ members of family A[9].
[9] See Berkovic 18 April 2005 and exbibit SFB-09.
8 March 1996 Dr Lopes-Cendes (from McGill) wrote to Berkovic informing him that the McGill Group had found a likely chromosomal locus (ie chromosome 2q23-31) for the gene underlying the epilepsy disease in Family A but asked him to keep the information confidential.[10]
[10] See Berkovic 18 April 2005 and exbibit SFB-11.
3 April 1996 Berkovic writes to Rouleau requesting confidential mapping information concerning the locus 2q23-31 to send to Drs Wallace and Mulley from the Women’s and Children’s Hospital in Adelaide (hereafter the “Adelaide group”) for them to test a Tasmanian family for any linkage to the family A gene[11].
[11] See Berkovic 18 April 2005, paragraph 5.1, and SFB-11.
19 April 1996 Letter from Professor Berkovic to Dr Rouleau commenting on further proposed work and acknowledging a “happy conclusion” to the genetic linkage part of the collaboration[12].
[12] See Berkovic 18 April 2005 paragraph 4.8 and exhibit SFB-11.
21 June 1996 Lopes-Cendes provides confidential mapping information including flanking markers D2S511 and D2S124 to Berkovic to pass onto to the Adelaide Group.[13]
[13] See Berkovic 18 April 2005 and exhibit SFB-14.
Spring 1996Berkovic sends genetic samples of Family A to the Adelaide group with a view to identify the underlying gene and mutation for GEFS+ in family A.[14]
[14] See Berkovic 18 April 2005, paragraph 5.
7-10 Dec1996 Publication of McGill/Melbourne collaborative work at a conference[15].
[15] Lopes-Cendes I, Scheffer IE, Berkovic SF, Rousseau M, Andermann E and Rouleau GA (1996) “Mapping a locus for idiopathic generalized epilepsy in a large multiplex family.” Epilepsia 37 Suppl 5:127 (see HH-04).
The abstract from the conference proceedings only indicates that the locus for IGE has been mapped chromosome 2 but does not identify either the locus or the flanking markers. Rouleau however claims that the locus was orally disclosed at the conference [16]
[16] See Rouleau declaration of 15 April 2005 (para 17) and declaration of 26 July 2006 (para 3.2).
Nov 98-April 99 McGill identifies a number of clinically significant mutations in the SCN genes of 2q23-24[17].
[17] See Lafreniere declaration at paragraph 8, Rochefort declaration at paragraph 5).
26 Nov 1999McGill filed US basic application 60/167 623 [18].
[18] See Berkovic 18 April 2005, and exhibit SFB-21.
11 May 2000 The Adelaide group identifies a clinically significant mutation at the 2q23-31 locus in family A (D188V in SCN1A).[19]
[19] See Berkovic 18 April 2005 and exbibit SFB-15.
17 May 2000 Professor Berkovic phones Rouleau to tell him of the Adelaide discovery but does not identify where or what the mutation is.[20]
[20] See Berkovic 18 April 2005 at paragraph 4.9.
28 July 2000 Rouleau emails Berkovic and advises him that McGill had just found a D188V mutation in family A which they had previously missed.[21]
[21]See Berkovic 18 April 2005 and exhibit SFB-13.
24 Nov 2000 McGill filed their PCT application claiming priority from the earlier US basic document. The D188V mutation in SCN1A is disclosed in the PCT specification. The application has an international publication date of 31 May 2001.
12 Dec 2000 Bionomics filed their own basic patent application (independent of McGill) in December 2000 followed by a PCT application (PCT/AU01/01648) on December 20 2001. Lafreniere at paragraphs 14-15 notes that there were 3 mutations within SCN1A identified in the Bionomics work:
Asp 188 Val (D 188 V)
Val 1353 Leu (V 1353 L) and
Ile 1656 Met. (I 1656 M).
The only mutation in common with the McGill University patent application is D 188 V from family A.
9 Feb 2001 Berkovic’s last contact with Rouleau where Dr Rouleau declined Professor Berkovic’s invitation to be named as a co-author in a paper being prepared by Mulley and Wallace. Berkovic considered that the collaboration between the Melbourne and McGill groups as being concluded at that point.
RELEVANT LAW
18. Section 36 of the Patents Act 1990 provides:
36 (1) Where:
(a)a patent application has been made and, in the case of a complete application, the patent request and complete specification have not been accepted; and
(b)the Commissioner is satisfied, an application made by a person in accordance with the regulations, that the nominated person is not an eligible person in relation to an invention disclosed in a specification filed in relation to the application but that some other person is an eligible person in relation to the invention as so disclosed;
the Commissioner may (whether or not the patent application lapses or is withdrawn) declare in writing that the other person is such an eligible person.
19. An eligible person under section 36 is one who is entitled to be granted a patent under section 15 of the Patents Act [1990]. Under section 15, a person could have direct entitlement to a patent through their own inventorship or through deriving entitlement from the inventor. An eligible person under section 36 is one who is entitled to be granted a patent under section 15 of the Patents Act [1990]. Under section 15, a person could have direct entitlement to a patent through their own inventorship or through deriving entitlement from the inventor.
20. In Upham v Commissioner of Patents [1998] AATA 853 (at paragraphs 19 and 20), the AAT discussed a number of Australian Patent Office decisions relating to inventorship including Harris vCSIRO (1993) 26 IPR 369, Costa v GR & IE Daking Pty Ltd 29 IPR 241, CSIRO v Gilbert & Hazlewood (1995) 31 IPR 67 and Falkenhagen et al v Polemate 1995 APO 32. The AAT observed that a common theme from these decisions is that a person has entitlement to an invention if that person’s contribution (either solely or jointly with others) had a “material effect on the invention” and a secondary issue is whether that person’s contribution involved a key inventive step. The key authority for this approach is the US decision of Mueller Brass Co v Reading Industries 17 USPQ 361 which noted at page 372:
“The exact parameters of what constitutes joint inventor ship are quite difficult to define. It is one of the muddiest concepts in the muddy metaphysics of the patent law. On the one hand, it is reasonably clear that a person who has merely followed instructions of another in performing experiments is not a co-inventor of the object to which those experiments are directed. To claim inventorship is to claim at least some role in the final conception of that which is sought to be patented. Perhaps one need not be able to point to a specific component as one’s sole idea, but one must be able to say that without his contribution to the final conception, it would have been less – less efficient, less simple, less economical, less something of benefit. This Court has found no case in which co-inventorship status was not deemed in some way, at least presumptively, to have beneficially affected the final concept of the claimed invention, and if such a case exists, it would be so anomalous as to warrant little attention.”
21. The US patent law terms of “final conception” and “final concept” which appear in Mueller Brass v Reading Industries (supra) does not mean that a co-inventor need necessarily be involved in the final form of the invention. As noted in Monsanto Co. v Kamp 154 USPQ 259, a co-inventor’s role could occur at any stage in the development of the invention:
“Each needs to perform but a part of the task if an invention emerges from all of the steps taken together. It is not necessary that the entire inventive concept should occur to each of the joint inventors, or that the two should physically work together. One may take a step at one time, the other an approach at different times. One may do more of the experimental work while the other makes suggestions from time to time. The fact that each inventors plays a different role and that the contribution of one may not be as great as that of another, does not detract from the fact that the invention is joint, if each makes some original contribution, though partial, to the final solution of the problem.”
22. The delegate in Primmcoy v Barry Charles Teer [2003] APO 37 (at paragraph 26) found that a person who took an initial step leading from the problem to the ultimate solution would still be an inventor even if another person was needed to finalize the invention. In coming to that conclusion, the delegate noted that this approach does not generally apply to someone who merely points out the state of the art or explains well known principles and instead:
“The step must be one that one that materially contributes to the invention ultimately described and the question sometimes put is whether the invention would have occurred without the involvement of the party seeking to claim entitlement – see Harris v CSIRO (1993) 26 IPR 469 and Costa v GR & IE Daking Pty Ltd (1994) 29 IPR 241.”
DECISION
Role of Professor Berkovic in the development of the invention
23. Professor Berkovic (from Melbourne) identified a large Australian family (family A) with an unusual genetic predisposition for epilepsy. He determined through his clinical investigations that this family had an unusual single autosomal dominant gene pattern of inheritance for epilepsy. The specification noted (page 32, line 27– page 33, line 2) that such families provided a “unique opportunity” to rapidly map genes that play a role in determining pre-disposition to seizures. Family A had the added advantage that it was a large multi-generational family which could provide a rich source of data for the genetic linkage studies. This could potentially provide a more precise location for the IGE gene making it easier to identify and isolate the IGE gene. Dr Rouleau (from McGill) noted in his letter of 7 November 1990[22] that large families were “definitely greatly superior to many smaller families for doing linkage” and both parties referred to the family as “remarkable” [23].
[22] See Berkovic 18 April 2005 and exhibit SFB-05
[23] Berkovic’s letter of 26 October 1990 (SFB-04) and Rouleau’s letter of 7 November 1990 (SFB-05).
24. Berkovic established a collaboration with Dr Rouleau of McGill to undertake “genetic studies” of family A. Berkovic provided both his clinical data and DNA samples from family A as part of the collaboration. McGill used this material to undertake preliminary genetic linkage studies in which they compared the patterns of inheritance of epilepsy in that family with known chromosomal markers. By determining the markers most closely linked to the IGE gene, the gene was located between the flanking markers D2S511 and D2S124 at a particular locus (ch2q23-q31) on chromosome 2.
25. This locus was still large (29 cM) and potentially contained hundreds of genes. It was therefore not feasible to sequence the entire region for mutations in family A related to IGE. Instead, McGill had observed that prior art genes associated with epilepsy had all coded for membrane ion channels. They therefore screened the ch2q23-q31 locus for membrane ion channels and identified a number of possible candidate epilepsy genes including a SCN (alpha sub-unit of Na+ channels) cluster of genes. McGill analyzed DNA from 83 unrelated samples looking for significant mutations in the candidate genes associated with epilepsy. This work was all performed at McGill with no intellectual input from Berkovic. Only one of the DNA samples came from the family A[24]. All the other samples were from Canadian samples, primarily Quebec and were collected by either Rouleau or Dr Patrick Cossette. Between November 1998 and April 1999, McGill identified a number of clinically significant mutations in the SCN genes in the Canadian samples:[25]
[24] See Lafreniere at paragraph 6
[25] See Lafreniere declaration at paragraph 8, Rochefort declaration at paragraph 5
gene mutation Date SCN1A E 1238 D 30 November 1998 SCN2A L 768 V August 1999 SCN2A K 908 R 15 October 1999 SCN3A N 43 DEL 13 April 1999
26. In his evidence, Lafreniere declared that after identifying these sequence variants, the McGill group realized that rare variants in the SCN1A, SCN2A and SCNA3A genes were associated with epilepsy and on that basis filed their priority patent application on 26 November 1999. Significantly, the McGill group did not identify the SCN1A mutation in family A (D188V) until after the priority document was filed (in July 2000). Nevertheless, McGill used this mutation in the functional studies described in the complete specification (see example 6) to verify that the mutation could affect sodium channel activity.
Was the contribution of Professor Berkovic material to the invention?
27. The DNA samples from family A provided by Professor Berkovic were not used to isolate the IGE gene but they were used in the preliminary genetic linkage studies which broadly located the gene to the ch2q23-q31 locus. As the specification acknowledges (see page 28, line 26- page 32, line 7), this was an important step in McGill’s overall strategy to identify the IGE gene because it enabled them to significantly narrow their search to a workable number of candidate IGE genes.
28. McGill argued that Berkovic’s role was very early in the eventual isolation of the gene before any of the difficult cloning work. They suggested that the collaboration between the parties was limited to the genetic linkage studies and once this work had been completed and published, they were free to use the preliminary genetic linkage work in their cloning experiments without restraint. In that sense (according to McGill), the linkage and cloning stages of the isolation of the IGE gene were completely separate inventive projects and not simply different phases of the one project.
29. However both parties knew at the beginning of the collaboration that family A was a rich source of data which might eventually lead to the identification of the IGE gene. Berkovic wrote to Rouleau on 6 April 1992 [26] proposing a number of conditions for the collaboration including the following item (1):
“All papers containing data on linkage analysis and possible later gene cloning will include the Melbourne Group as authors. The first author on such papers should be the scientist responsible for the major part of the work and who will also presumably do the majority of the writing for the manuscript. The second, third and last authorship positions will be shared amongst two members of the Melbourne Group and either yourself or another major contributor from Montreal…… ” [my emphasis]
[26] See Berkovic 18 April 2005 and exhibit SFB-08
30. In response[27], Rouleau agreed with most of Berkovic’s terms but noted with respect to item (1) that:
“I agree with item 1 with respect to the linkage phase of the project. However the cloning efforts will likely involve additional families and constitute an enormous effort from my lab. You, plus colleagues in Melbourne will certainly continue to be co-authors, but I don’t think you should control the second, third and last authorship positions.” [my emphasis]
[27] See Berkovic 18 April 2005 and exhibit SFB-08
31. Rouleau refers to the linkage “phase” of a single project to isolate the IGE gene. Although Rouleau noted that McGill would have a greater role in the later (cloning) phase, he conceded that the Melbourne Group should still retain co-authorship on McGill’s papers. Rouleau had therefore understood that the overall goal for the collaboration was to isolate the IGE gene. In fact, as the section 36 requestor noted, the only practical purpose for identifying the IGE locus is to localize the gene because this would allow a better understanding of epilepsy and potentially lead to better treatment options.
32. The evidence shows that there were on-going problems with the collaboration between McGill and Berkovic. Eventually the collaboration broke down with both parties separately pursuing the cloning work of isolating mutant IGE genes. However by this time, the collaboration had completed the preliminary genetic linkage work which was a critical part of the later cloning strategies of both parties. As discussed above, the law in Australia is that a person who took an initial step leading from the problem to the ultimate solution would still be an inventor even if another person was needed to finalize the invention. The collaboration set up by McGill and Melbourne had a specific and defined goal (to identify the IGE gene) and Berkovic’s work in identifying family A was a critical preliminary step in the achieving this goal. Professor Berkovic was therefore prima facie important in the overall development of the invention even though he himself had no role in the cloning phase.
33. McGill noted that the collaborative genetic linkage work was at least partly published at a conference in 1996[28] and that two different research groups had also published the general IGE locus shortly before the priority date of the specification in 1999[29]. McGill argued that as the IGE locus was published before the priority date of the current application, the DNA samples provided from family A added no information to the prior art. According to McGill, the DNA samples of family A did not contribute anything to the invention now claimed and offered McGill no “private advantage” compared with other workers in the field.
[28] Lopes-Cendes I, Scheffer IE, Berkovic SF, Rousseau M, Andermann E and Rouleau GA (1996) “Mapping a locus for idiopathic generalized epilepsy in a large multiplex family.” Epilepsia 37 Suppl 5:127 an abstract from conference proceedings 7-10 December 1996 (HH-04)
[29] Beaulac et al (September 8 1999) A second locus for familial generalized epilepsy with febrile seizures plus (GEGS+) on chromosome 2q21-q33 Am J Hum Genet (1999) Oct; 65(4) 1078-8534. However the issue of entitlement is a fundamentally different concept from novelty and inventive step. It is therefore irrelevant whether the inventive contribution has been published by other workers or whether the applicant has any “private advantage” over other workers. As noted above, the key test for inventorship is whether a person’s contribution (either solely or jointly with others) had a “material effect on the invention” and a secondary issue is whether that person’s contribution involved a key inventive step [see Upham v Commissioner of Patents [1998] AATA 853 (at paragraphs 19 and 20)].
35. McGill did not obtain the genetic linkage data from the prior art but from DNA samples provided by Berkovic as part of a collaborative project to isolate the IGE gene. Berkovic’s preliminary work was a key reason for McGill’s later success in isolating the IGE gene and therefore his work had a “material effect on the invention” and entitles him to co-inventorship of the current application.
Role of Drs Mulley and Wallace in the development of the invention.
36. McGill identified a clinically significant mutation of family A ( D188V in SCN1A) after filing of their priority document. Bionomics claim that D188V is a key component of the invention described because it is the only functional mutation identified in any of the SCN genes at locus 2q23-31. It is also the mutation that McGill uses in their functional studies in examples 6 and 7 of their complete specification.
37. Bionomics alleged that McGill only found the D188V mutation after a telephone conversation in May 2000 in which Berkovic told Rouleau that Drs Wallace and Mulley from the Adelaide group had succeeded in identifying a clinically significant mutation in family A. Bionomics claim that the input from Mulley and Wallace led to McGill re-screening family A DNA samples for mutations. Because McGill identifies the D188V soon after the critical telephone conversation in July 2000), the work of Mulley and Wallace was material to identifying the D188V mutation.
38. McGill argued that Rouleau independently located the mutation without inventive input from the Adelaide Group. They do not deny a conversation occurred between Rouleau and Berkovic in May 2000 but dispute that this conversation led Rouleau to review his family A sequence data and locate the relevant mutation. Rouleau states in his declaration of 26 July 2006 that family A had been sequenced twice without finding a mutation. His group found the mutation after a third round of sequencing. Rouleau believed that the Escagy publication in April 2000[30] might have motivated the third round.
[30] Escagy et al Nature Genetics 24(4): 343-5 (January 2000)
39. There is a remarkable coincidence in the timing of the telephone conversation in May 2000 and the subsequent identification of the D188V mutation in July 2000 given that there were two previous unsuccessful attempts to identify the mutation in family A. This together with Rouleau communicating to Berkovic by email as soon as he confirmed the identification of a clinically significant mutation [31] suggests that there may be a link between the two events. Rouleau denies this but his alternative explanation was vague, inconsistent and uncorroborated. Given that the identification of the mutation was a significant advance, I would have expected Rouleau to have a clearer recollection of the circumstances which led to the re-screening of family A. I therefore have difficulty believing Rouleau’s version of events and I find that the telephone conversation in May 2000 was highly likely to have triggered the re-screening of family A as Bionomics alleged.
[31] See Rouleau 26 July 2006 paragraphs 7.5-7.6
Was the contribution of Mulley and Wallace material to the invention?
40. Although I found that the work of the Adelaide Group led McGill to the identification of the D188V mutation, I do not believe that this also entitles Mulley and Wallace to co-inventorship status. Before the priority date, McGill had identified a number of significant mutations in the SCN genes at the ch2q23-q31 locus. Based on this, McGill filed a priority patent application which concluded that mutations in the SCN genes were associated with epilepsy. The D188V mutation may have been identified later and is the only functional mutation identified but it is still clearly within the scope of that invention broadly disclosed in the priority document and does not represent a different invention. In my view, the D188V mutation is merely a preferred embodiment of the broad invention described in the priority document.
41. Since Mulley and Wallace’s contribution occurred after the filing of the priority document when the invention was finally conceived, they cannot be said to have had a “material effect” on the invention and in my view do not have any entitlement to the invention described in the complete specification.
CONCLUSION
42. The genetic linkage studies and the identification of the gene were two phases or steps in the one project. As noted above, the preliminary linkage analysis (in which Berkovic played a key role by identifying family A which was used in the studies) significantly narrowed the search for the gene to a particular locus which enabled the later isolation of the IGE gene. Even through Berkovic did not contribute to the later phases of the identification of the gene, his preliminary work with identifying family A had a “material effect on the invention” and entitles him to co-inventorship. In contrast, Drs Mulley and Wallace’s contribution was after the final conception of the invention and means that they are not co-inventors.
43. As a consequence, I conclude Professor Berkovic is a co-inventor of the invention described and claimed in the current application and by virtue of assignment Bionomics Limited therefore has co-entitlement to the invention. Pursuant to s.36(1), I declare Bionomics and McGill University are eligible persons in relation to invention disclosed in the specification of patent application 18465/01.
COSTS
44. In matters before the Commissioner, costs generally follow the event. As the section 36 requestor was successful in establishing co-ownership of the invention, I award costs against the patent
applicant, McGill University.
Karen Ayers
Delegate of the Commissioner of Patents
29 January 2007Patent attorneys for the applicant: Cullen & Co, Brisbane
Patent attorneys for the section 36 requestor: Madderns Adelaide.
Leppert et al (1989) Benign familial neonatal convulsions linked to genetic markers on chromosome 20 Nature 337: 647-648
Lewis et al (1993) Heterogeneity in benign familial neonatal convulsions: identification of a new locus on chromosome 8q Am. J. Hum. Genet. 53: 670-675
Elmslie et al (1997) Genetic Mapping of a major susceptibility locus for juvenile myoclonic epilepsy on chromosome 15q Hum. Mol. Genet. 6: 1329 - 1334
Guipponi et al (1997) Linkage mapping of benign familial infantile seizures (BFIS) to chromosome 19q Hum. Mol. Genet. 6: 473-477
Wallace et al (1998) Febrile Seizures and generalised epilepsy associated with a mutation in the Na+-channel (1 sub-unit gene SCN1B) Nat. Genet. 19:366-370
Singh et al (1998) A novel potassium channel gene, KCNQ2, is mutated in an inherited epilepsy of newborns Nat. Genet. 18: 25-29
Wallace et al (1998) Febrile Seizures and generalised epilepsy associated with a mutation in the Na+-channel (1 sub-unit gene SCN1B) Nat. Genet. 19:366-370
Moulard B et al “Identification of a new locus for generalized epilepsy with febrile seizures plus (GEFS+) on chromosome 2q24-q33. Am. J. Hum. Genet. (1999) Nov; 65(5):1396-400 published October 6 1999
1
4
0