Mars, Incorporated v Hill's Pet Nutrition, Inc
[2014] APO 67
•21 October 2014
IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
Mars, Incorporated v Hill’s Pet Nutrition, Inc. [2014] APO 67
Patent Application: 2005280008
Title:Genome-based diet design
Patent Applicant: Hill’s Pet Nutrition, Inc.
Opponent: Mars, Incorporated
Delegate: Dr B Akhurst
Decision Date: 21 October 2014
Hearing Date: 22 July 2014, in Canberra
Catchwords: PATENTS - section 59 - manner of manufacture - is a computer-implemented method of for determining nutrition for an animal a theoretical scheme or abstract plan - artificially created state of affairs - the claims define a manner of manufacture - novelty - lack of novelty not established - inventive step - lack of inventive step not established
Representation: Patent applicant: Ben Fitzpatrick of Counsel, instructed by Stephen Gledhill and Sylvan Browne, patent attorneys of FB Rice, Melbourne.
Opponent: Mark Summerfield, John Golding and Amanda Johns, patent attorneys of Watermark, Sydney.
IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
Patent Application: 2005280008
Title:Genome-based diet design
Patent Applicant: Hill’s Pet Nutrition, Inc.
Date of Decision: 21 October 2014
DECISION
The opposition fails. Costs awarded according to Schedule 8 against Mars.
REASONS FOR DECISION
Background
Patent application 2005280008 was filed by Colgate-Palmolive Company on 30 August 2005 via the PCT, claiming priority from basic document US 60/605,573 filed on 30 August 2004. The applicant was amended to Hill’s Pet Nutrition, Inc. (Hills) on 17 April 2007.
After examination, the application was advertised as accepted on 16 June 2011. Mars Incorporated (Mars) served a notice of opposition to grant of a patent on 16 September 2011, followed by a statement of grounds and particulars on 16 December 2011. Evidence in support was completed on 21 December 2012, evidence in answer on 21 November 2013 and evidence in reply on 21 February 2014.
The matter was heard on 22 July 2014 in Canberra. Summaries of submissions were filed on 10 July 2014 and Hills on 17 July 2014.
The Evidence
The evidence in support consisted of declarations by:
·Dr Neale Fretwell (NF#1) dated 11 October 2012 with exhibits NF-1 to NF-3
·Mr Tim McCallum (TM#1) dated 16 October 2012 with exhibits TM-1 to TM-2
·Dr Neale Fretwell (NF#2) dated 20 December 2012 with exhibits NF-4 to NF-10
·Mr Tim McCallum (TM#2) dated 17 December 2012 with exhibits TM-3 to TM-7
Evidence in answer consisted of declarations by:
·Dr Shubiao Wu (SW) dated 21 November 2013 with exhibits SW-1 to SW-7
·Dr Wendy Brown (WB) dated 21 November 2013 with exhibits WB-1 to WB-9
Evidence in reply consisted of declarations by:
·Dr Neale Fretwell (NF#3) dated 19 February 2014
·Tim McCallum (TM#3) dated 21 February 2014
Grounds of opposition
At the hearing, the grounds of opposition pressed by the opponent were manner of manufacture, novelty and inventive step.
Onus of Proof
The request for examination in this case was filed on 17 December 2007. Therefore, the substantive amendments to the Patents Act brought about by the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 do not apply to the present application, including subsection 60(3A) which allows the Commissioner to refuse a patent application if satisfied on the balance of probabilities that a ground of opposition has been made out. Instead, the onus of proof in this opposition proceeding lies with the opponent, who must establish that it is clear that a valid patent cannot be granted (F.Hoffman-La Roche AG v New England Biolabs Inc [2000] FCA 283 at [29], [67]; 50 IPR 305; Commissioner of Patents v Sherman [2008] FCAFC 182 at [18], [22]; 79 IPR 426).
For the same reason, the substantive amendments to grounds of opposition that came into effect on 15 April 2013 do not apply to the present opposition. Instead the law that applies to the grounds of opposition in this case is that which stood immediately prior to 15 April 2013.
The specification
The complete specification is not to be read in the abstract. It must be construed through the eyes of the person skilled in the art, in the light of the common general knowledge and the art before the priority date (Kimberley Clark Australia Pty Ltd v Arico Trading International Pty Ltd [2001] HCA 8 at [24]; 207 CLR 1 at 16). Further principles to be applied in construing a patent specification are well settled in law (Flexible Steel Lacing Company v Beltreco Ltd [2000] FCA 890 at [70] - [81]; (2001) 49 IPR 331 at 347 [70] - [81]; Pfizer Overseas Pharmaceuticals v Eli Lilly and Company [2005] FCAFC 224 at [247] - [250]; 68 IPR 1 at 52-54).
The specification is titled ‘Genome-based diet design’. At para 2 of the description, the field of the invention relates to methods of designing and preparing animal foods based on an animal’s genome. The animal food industry has traditionally developed and marketed food based on different animal attributes or phenotypes (e.g. size, age, body condition), and in some cases for specific breeds or phenotypically defined breed types including the American Kennel Club (AKC) breed groups (para 4).
At para 21, the invention provides a new approach to enhancing animal nutrition and health care based on the genotype of an animal. Specifically, a plurality of genome-based breed clusters of an animal species are identified that exhibit a nutritionally appropriate degree of genetic similarity within clusters and more marked genetic diversity between breed clusters. Once genome-based breed clusters have been established, phenotypic information for each breed cluster can be analysed, and specific foods developed which are tailored to the particular nutritional needs of each cluster, and optionally other factors such as age, or to treat a disease in the animal (para 22).
At para 9b, the specification further discloses methods for determining nutrition for an animal comprising determining a genome-based breed cluster to which the animal belongs, and selecting a food for the animal based at least in part on the nutritional needs of the breed cluster. Optionally, the method comprises preparation of the food. In the claimed embodiments, the method is implemented by means of a computer.
The claims and their construction
The specification contains 26 claims. Claims 1 and 16 are independent. The description provides a comprehensive dictionary for many of the terms in the claims and I have referred to it, where appropriate, in construing the claims.
Claim 1 is as follows:
1. A computer-implemented method for determining nutrition to be provided for an animal comprising:
receiving input data relating to the animal;
identifying a genome-based breed cluster to which the animal belongs based on the input data and a first data set relating a plurality of breed clusters to genome-related attributes of each breed cluster; and
selecting a food for the animal having a nutritional formula matched at least in part to nutritional needs of animals in the breed cluster.
genome-based breed clusters & breed clusters
The meaning of the phrase “genome-based breed cluster” and “breed cluster” in the claims is not clear. In this situation, it is permissible to resort to the body of the specification to define or clarify the meaning of the words used in the claims (Interlego AG v Toltoys Pty Ltd [1973] HCA 1 at [14]; (1973) 130 CLR 461 at 479). The description provides a definition of the term:
“[0021] Unlike previous efforts to provide genotype-specific foods, eg. breed-specific foods, methods provided herein utilize a more comprehensive genomic profile of an animal species, coupled with rigorous statistical analysis, to define breed clusters that exhibit a nutritionally appropriate degree of genetic similarity within clusters and more marked genetic diversity between clusters. Breed clusters so defined are described herein as “genome-based” breed clusters. … Except where the context demands otherwise, the term “breed cluster” herein means a genome-based breed cluster, as opposed to a cluster of breeds grouped according to criteria other than genotype. Thus traditional classifications of animal breeds based on phenotypic criteria such as AKCs classification of canine breeds into seven groups (sporting, hound, working, terrier, toy, non-sporting and herding groups) do not meet the definition of “breed clusters” as understood herein.”
Applying this definition, the term “genome-based breed clusters” refers to animal breeds clustered (i.e. grouped) according to genotypic criteria.
At para 40 of the description, genome-based breed clusters may be identified by known methods of phylogenetic analysis and suitable methods are listed. At paras 45-49 it is stated that statistical analysis of genotypes is well known in the art and the invention extends to the use of any clustering algorithm and/or software to analyse genotype data, including a variety of strategies are listed. The specification exemplifies an analysis and identification of genome-based breed clusters by reference to the publication Parker et al. (2004) Science 304: 1160-1164 (Parker) (paras 40-44). Parker describes the use of genetic markers combined with phylogenetic analysis and genetic clustering to define four groups of genetically-related breeds referred to as antiquity, muscular, herding and hunting breeds. To differentiate these four genome-based breed clusters from AKC breed groupings with similar names, the specification renames them Clusters I to IV, respectively (paras 40-44).
It should be noted that para 21 of the opposed specification states that, unless the context demands otherwise, references to “breed clusters” means a genome-based breed cluster, as opposed to a cluster of breeds grouped according to criteria other than genotype. Further definition is provided at para 28 of the of the description, which states:
““Breed clusters” herein will be understood to embrace clusters of animal subpopulations whether or not these have arisen by human-directed breeding.”
In effect, while genome-based breed clusters must be derived using genotypic criteria, once established, they may correspond to traditionally recognised breed groupings. This is evident in Parker’s four breed clusters. Cluster IV (Parker’s hunting breeds) corresponds to the AKC sporting group (para 44 of the description), while Clusters I-III do not correspond to traditional breed groupings. My construction is consistent with that of Dr Brown and Dr Wu (WB at [38], [63]; SW at [30], [43]). Although Mars’ experts interpreted para 28 to mean that traditional breed groupings can be considered a “breed cluster” for the purposes of the invention, both acknowledged their conclusion was inconsistent with the definition of breed clusters in para 21 (NF#2 at [8]; TM#2 at [9]).
Hills submitted that the requirement in paragraph 21 for a comprehensive genomic profile and rigorous statistical analysis in deriving the breed clusters means the use of bioinformatics on the scale described in Parker. However, that construction is inconsistent with the description of the invention. Firstly, the reference in para 21 (reproduced above) to “a more comprehensive genomic profile” does not necessitate a comprehensive genome-wide genetic analysis. Construed in the context it appears it merely requires a more comprehensive profile than “previous efforts”. This construction is consistent with the definition of “genome” at para 37 of the description:
“The term “genome” generally means all the genetic material of an organism, but as used herein the term “genome” refers to the total genetic constitution or any fraction thereof sufficiently large to be amenable to analysis for the purpose of determining degree of genetic similarity or difference between organisms or populations of organisms” (emphasis added).
Furthermore, at para 45 it is stated that statistical analysis is well known in the art and any clustering algorithm and/or software can be used to analyse the genotype data. Suitable methods are identified in paragraphs 45-49 of the description, and include less complex methods than that used by Parker. On this basis, I reject Hills submissions that the requirement for “rigorous” statistical analysis necessitates the use of the more technically complex analyses. In its plain meaning, it simply requires the use of an appropriate method of statistical analysis in order to achieve valid estimates of statistical significance in the model being tested.
input data relating to an animal
There was no dispute that the input data must include sufficient information to allocate a particular animal to the genome-based breed cluster to which it belongs. This encompasses the use of genotypic and phenotypic information (TM#2 at [8]; TM#3 at [24]; NF#2 at [16], [27]; NF#3 at [35]; WB at [40]; SW at [30]). The input data may include one or more of breed, breed inheritance and genetic markers that individually or collectively are indicative of a breed cluster or, for an animal of mixed breed, a best-fit breed cluster (para 124). “Breed inheritance” is defined at para 73 as “the breed ancestry of an animal, namely the one or more breeds that have contributed to an animals genome”.
a first data set relating a plurality of breed clusters to genome-related attributes of each breed cluster
The first data set of claim 1 may include a list of breeds for each breed cluster, and the genome-related attributes can include one or more of breed, breed inheritance and genetic markers (para 124).
selecting a food for an animal having a nutritional formula matched at least in part to nutritional needs of animals in the breed cluster
The term “at least in part” is not clear. Resort to the description reveals at para 22 that the invention provides methods for formulating food for an animal based on the nutritional needs of the breed cluster to which it belongs. However, in some embodiments, the method includes formulating the food based in part on the animal’s age, or to address a disease the animal has (para 22). Therefore, the effect of the qualification “at least in part” is that while the nutritional formula of the selected food must be matched to some degree to the nutritional needs of the breed cluster to which the animal belongs, the formulation may also be matched in part to the nutritional requirements of the individual animal.
Independent claim 16 is as follows:
16. A computer-aided system comprising:
computerised interface means to receive input data relating to an animal;
a first data set relating a plurality of breed clusters to genome-related attributes of each breed cluster; and
a first algorithm capable, while drawing on the first data set, of
(i) processing the input data to determine one or more genome-related attributes of the animal to define a breed cluster to which the animal can be allocated, and
(ii) designing a nutritional formula appropriate to nutritional needs of the breed cluster.
There was no dispute, and I agree, that claim 16 defines a system comprising the components required to implement the method of claim 1. Therefore, the input data and first data set take the same meaning as in claim 1. The first algorithm is generically defined by the words of claim 16.
Having regard to the dictionary in para 21 of the description, construed in the context they appear, the “breed clusters” in claim 16 are genome-based breed clusters as opposed to a cluster of breeds grouped according to criteria other than genotype. Regarding “designing” an appropriate nutritional formula, resort to the description at para 20 indicates term is “used broadly, and can include selection of a food from a pre-existing set of options, and/or creation de novo of a food in a form of a nutritional formula or recipe”.
Manner of Manufacture
Mars opposed the claims of the patent on the basis that they do not define a manner of manufacture.
Section 18(1)(a) requires that an invention must be a manner of manufacture within the meaning of section 6 of the Statute of Monopolies. For determining whether an invention is a manner of manufacture, the right question is:
“Is this a proper subject of letters patent according to the principles which have been developed for the application of s.6 of the Statute of Monopolies?”
(National Research Development Corporation (NRDC) [1959] HCA 67 at [14], [25]; 102 CLR 252 at 269)
The threshold requirement for a patentable invention is that it must not be apparent on the face of the specification that the claimed invention lacks the necessary quality of inventiveness for it to be a proper subject of letters patent under the Statute of Monopolies (NV Philips Gloeilampenfabrieken v Mirabella International Pty Ltd [1995] HCA 15 at [9]; (1995) 183 CLR 655). In this regard, the specification states at para 4 that traditionally animal food has been developed and marketed based on phenotype (e.g. size, age, body condition, breed) or phenotypically-defined breed types. At para 21, the invention provides a new approach to enhancing animal nutrition and health care based on the genotype of an animal, and genotypically-defined (genome-based) breed groupings. While the means to implement the method of claim 1 are described as known in the art, it is not apparent on the face of the specification that the combination of features in claim 1 is obvious, particularly the use of genome-based breed clusters as the basis for selecting appropriate animal nutrition. Furthermore, although the claimed invention is exemplified using the canine breed clusters pre-published by Parker, there is no indication in the specification that it was known to use these breed groupings in the context of canine nutrition.
In order to be held patentable, a claimed process must result in an artificially created state of affairs that offers a material advantage in the field of economic endeavour (NRDC [1959] HCA 67 at [22], [25]; 102 CLR at 275, 277). An artificially created state of affairs should be understood in the sense of a concrete, tangible, physical, or observable effect (Grant v Commissioner of Patents [2006] FCAFC 120 at [30]; 234 ALR 230 at 237). The process must offer a material advantage in the sense that it belongs to a useful art as distinct from a fine art and that its value to the country is in the field of economic endeavour (NRDC [1959] HCA 67 at [22]; 102 CLR at 275).
It was not in dispute that the animal food industry is a field of economic endeavour. Mars’ opposition under this ground was primarily on the basis that the claimed invention does not achieve an end result that is an artificially created state of affairs.
Referring to Tables 6 and 19 in the specification in respect of Cluster IV (Parker’s hunting breed group which corresponds to the AKC sporting group), Mars submitted that there is no new artificially created state of affairs produced by looking up existing information. Table 6, titled “Phenotypic analysis of canine breeds of Cluster IV”, lists the breeds identified by Parker as belonging to the hunting breed group together with the phenotypic characteristics (size, shedding, trainability and activity level) of each of these breeds based on AKC information. Table 19 “Food composition for a canine of Cluster IV” lists ingredients and their relative proportions in the food composition.
Although Mars refers to these tables as existing information, Mars has adduced no evidence that the particular combination of information in Tables 6 and 19 was existing information before the priority date. In any event, I did not understand Mars’ submission to be that the combination of information in Tables 6 and 19, which may potentially exist for the AKC sporting group, was previously known in the context of genotypically-defined breed clusters, and nor is this apparent on the face of the specification.
Mars further contended that were it not for references to computer-implementation, claim 1 is directed to a theoretical scheme or abstract plan that can be summarised as “selecting food for an animal based on the genetic characteristics of the animal”. Responding, Hills identified the artificially created state of affairs as the nutritional formulation or food with an improved formula that is more appropriate for an animal or alternatively a change in the state of the computer.
Although much of the parties’ submissions centred on the effect achieved by computer implementation, in my view manner of manufacture in this opposition does not turn on this point.
Underpinning the method of claim 1 is the requirement for genome-based (i.e. genotypically-defined) breed clusters to be established and the nutritional needs of animals in each breed cluster to be defined. Thereafter, knowledge of the genome-based breed cluster to which a particular animal belongs allows selection of a food for the animal based on the nutritional requirements of the relevant cluster. Claim 1 does not require the animal to consume the food, thus the effect of the method does not extend to a physical effect on the animal. Therefore, consistent with the parties’ submissions, it is reasonable to conclude that the effect achieved by the method of claim 1 is the selection of a nutritionally appropriate food for an animal.
In my view, the method of claim 1 is analogous to diagnostic or screening methods that, when carried out in respect of an animal or human subject, provide information relevant to improving or maintaining the subject’s health and wellbeing. Diagnostic and screening methods represent more than a theoretical scheme or abstract plan and the effect achieved by such methods has traditionally been considered to lie in the useful arts and be proper subject matter for a patent.
Since the nature of the opposed claims is closely aligned to subject matter that is traditionally considered patentable, I find the method of claim 1 and its dependent claims define a manner of manufacture. It was not in dispute that claim 16 defines a system capable of performing the method of claim 1 and applying the same reasoning, this claim and dependent claims also define a manner of manufacture.
Mars has not established that the claims are not for a manner of manufacture.
Priority date
Mars challenged the priority date of the opposed claims. If the claim to priority from US 60/605,573 (US’573) is invalid, then Mars’ patent specification WO 2004/113570 A2 published on 29 December 2004 is a ‘whole of contents’ citation and does not form part of the prior art base for consideration of inventive step (see the definition of prior art base in Schedule 1 of the Act).
Subsection 43(1) provides that each claim of a specification must have a priority date. Subsection 43(2) and reg 3.12(1), as in force immediately before 15 April 2013, provide that the priority date of a claim is the earlier of the date of filing of the specification or, where the claim is fairly based on matter disclosed in one or more priority documents, the date of filing of the priority document in which the matter was first disclosed.
In Sigma Pharmaceuticals (Australia) Pty Ltd v Wyeth (Sigma) [2011] FCAFC 132, at [65], the Full Court confirmed that the principles that apply to considerations of ‘external fair basis’ are the same as those for determining fair basis under section 40(3). The test for determining whether a claim is fairly based on a priority document is determined by the application of the principles set out in Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (Lockwood) [2004] HCA 58, 217 CLR 274, at [69] (Sigma [2011] FCAFC 132 at [64]). Fair basis requires a real and reasonably clear disclosure of what is then claimed. The invention must be broadly, that is in a general sense, described in the body of the specification and not travel beyond the matter disclosed (Lockwood at [2004] HCA 58 at [69], [83]; 217 CLR at [69], [83]). It is wrong to employ an overly meticulous verbal analysis or seek to isolate “essential features” of an alleged invention and ask whether they correspond to the essential features of the claim in question (Lockwood [2004] HCA 58 at [68]). Fair basis requires consideration of whether, as a matter of drafting, the claims can be said to reflect what is disclosed or stated in the specification (Sigma [2011] FCAFC 132 at [64]).
The provisional specification must be read as a whole (Anaesthetic Supplies Pty Limited v Rescare Limited [1994] FCA 1065 at [86]; 122 ALR 141), through the eyes of the person skilled in the art.
Mars acknowledged that US’573 relates to formulating animal foods, disclosing methods involving genotyping a plurality of breeds then clustering the breeds, analysing phenotypic information for each of the clusters, and formulating the food for each cluster based on the phenotypic information. However, Mars submitted that the opposed application discloses and claims extensive additional information and the degree of progression from US’573 is not allowable. Specifically, Mars submitted that the priority document contains no real and reasonably clear disclosure of the following features:
·a “genome-based breed cluster” as it is defined in the opposed specification;
·a method for determining nutrition to be provided for an animal by identifying the breed cluster to which it belongs and selecting a food for the animal matched at least in part to the nutritional needs of the animals in the breed cluster; or
·the computer-implemented method of claim 1 or computer-aided system of claim 16, and in particular of “a first data set relating a plurality of breed clusters to genome-related attributes of each breed cluster” or any “algorithm capable, while drawing on the first data set” of processing the input data to define a breed cluster to which the animal can be allocated.
Responding, Hills drew my attention to a number of passages in US’573 which it submitted provided the necessary disclosure.
Mars does not rely on expert evidence to establish the priority claim is invalid. I will consider each of Mars points above in turn.
genome-based breed cluster
US’573 describes the invention as generally relating to companion animal nutrition formulations based on the animal’s genome (para 1). At para 4, under the heading “Summary of the invention”, the invention is a new approach to formulating animal nutrition and health care based on the genotype of the animal, which involves genotyping a plurality of breeds of the animal then clustering the breeds. At para 9, Figure 1 is described as illustrating the process of determining genetic clusters. Further references to “genetic clusters”, “genotype clusters” or “genotypic clusters” can be found throughout the specification (paras 22-23, 25-26, 28 and 31). At para 19, US’573 states:
“The present invention defines an independent classification of dogs based on genetic variation and using this classification designs health and nutrition programmes for clusters of dog breeds that are clustered based on their genetic similarities.”
US’573 does not explicitly use the phrase “genome-based breed clusters”. However, I consider that the person skilled in the art would understand the passages referring to genetic, genotype or genotypic clusters, and references to clustering based on genetic similarities, to provide a real and reasonably clear disclosure of the genome-based breed clusters as these are defined in the opposed specification.
a method for determining nutrition for an animal involving identifying the breed cluster to which it belongs and selecting a food matched at least in part to animals in the breed cluster
US’573 describes formulating an appropriate food for an animal on the basis of its genotype. Once genotypically-defined breed clusters have been identified, specific foods are developed on the basis of the phenotypic information associated with each cluster such as disease prevalence and physical attributes (paras 4-6, 22-24, 28). US’573 exemplifies the invention by providing food compositions for feeding to a dog identified as belonging to the antiquity, muscular, herding or hunting breed clusters identified by Parker (paras 19, 21-22, 26-27).
At para 31, consumer communication apparatus may be necessary “so the consumer understands which genotypic cluster his/her dog resides in”. In a similar context, an embodiment is disclosed in which a consumer inputs information including the breed type and age of an animal into a computer kiosk after which the computer identifies the correct food formulation for the consumer’s companion animal (para 31). Graphics or colours may be used for the same purpose (para 31). I consider this paragraph to disclose identifying the breed cluster an animal belongs to in order to select an appropriate food formulation.
Noting that opposed claim 1 requires the nutritional formula to be matched only “at least in part” to the nutritional needs of the breed cluster, Mars submitted that the priority document does not disclose formulation of food for animals in breed clusters that is not based on analysis of phenotypic information. However, US’573 describes methods of formulating food that take into account the age of an animal and disease (paras 4, 24, 28). The animal’s age at least is not phenotypic information relevant to the breed cluster, it is information specific to the animal. This is illustrated at paras 24 and 28 of US’573, where consideration of age involves breaking the clusters into subgroups. Therefore, at least where the age of an animal is taken into account, the food is matched only in part to the nutritional requirements of the breed cluster.
In summary, in the passages indicated in this above, I consider US’573 to provide a real and reasonably clear disclosure of a method for determining nutrition for an animal involving identifying the breed cluster it belongs to and selecting a food matched at least in part to the nutritional needs of the animals in the breed cluster.
the computer-implemented method of claim 1 or computer-aided system of claim 16, and in particular of “a first data set relating a plurality of breed clusters to genome-related attributes of each breed cluster” or any “algorithm capable, while drawing on the first data set” of processing the input data to define a breed cluster to which the animal can be allocated
a first data set relating a plurality of breed clusters to genome-related attributes of each breed cluster
Tables 3-6 and 7-10 in US’573 list four genotypically-defined (i.e. genome-based) breed clusters and the phenotypic attributes and common diseases of each breed in the cluster. From this information the average phenotypic characteristics and disease traits of the breed clusters are summarised in Table 11. Although not explicitly described as “a first data set”, Table 11 of US’573 provides a real and reasonably clear disclosure of a relevant data set relating a plurality of breed clusters to the genome-related attributes of each breed cluster.
an algorithm capable, while drawing on the first data set, of processing the input data to define a breed cluster to which the animal can be allocated
As indicated above, US’573 discloses animal breeds clustered according to genotypic criteria. Once the clusters are established, information regarding the breed of a companion animal allows that animal to be assigned to the breed cluster to which it belongs. Paragraph 31 of US’573 describes the use of consumer communication apparatus and point of sale displays. In one embodiment, the user inputs information such as the breed type and age into a touch screen in a computer kiosk, after which the computer identifies the correct food formulation for the consumer’s companion animal. In some embodiments, the point of sale display may be a small microprocessor that would ask for the breed and the age then the microprocessor would output the proper formulation for the consumer’s companion animal. Graphics may be displayed in, among other things, computers and microprocessors to indicate a particular genotype cluster “so that identification of the results are easily found”.
Although paragraph 31 does not refer explicitly to an algorithm, construed in the context of the specification as a whole, inputting information such as the dog’s breed in a computer kiosk to identify the genotypic cluster a particular dog belongs to, and thus the correct food formulation, provides a generic or implicit disclosure of the algorithm required to perform this process.
computer-implementation of the method
A reference to computer implementation of the method can be found at para 29:
“A clustering algorithm yields genotypes and these genotypes may be clustered. Genotypes are related to other characteristics such as disease propensity, prevention and/or treatment of diseases, age, ingredients, and neutral requirements. In various embodiments of the present invention, the components, such as genotype clusters, age groups, ingredients, nutritional requirements, and disease propensity maybe put into databases and in various embodiments algorithms maybe run so that a formula matrix may be created.”
Construction of food formulations based on the requirements of the genotype cluster and the age, disease prevalence and/or physical attributes of the animal by means of a matrix is described at paras 29-31 and Figure 2 of US’573.
I consider that the passages indicated above broadly, in a general sense, disclose computer implementation and computer-aided systems and a relevant data set, in addition to providing a generic disclosure of an algorithm capable, while drawing on the data set, of processing the input data to define a breed cluster to which the animal can be allocated.
In summary, I find that US’573 provides a real and reasonably clear disclosure of the invention claimed in the opposed application. Mars has not established that the claims are not entitled to claim priority from US’573.
Novelty
It is well established that the general test for anticipation is the reverse infringement test. The classic formulation of this test is that given by Aicken J in Meyers Taylor Pty Ltd v Vicarr Industries Ltd [1977] HCA 19 at [20]; (1977) 137 CLR 228 at 235:
“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 anticipation would, if the patent were valid, constitute an infringement”
This test is satisfied if the alleged anticipation discloses all of the essential features of the invention as claimed (Nicaro Holdings Pty Ltd v Martin Engineering Co [1990] FCA 40 at [19]; (1990) 91 ALR 513 at 517). To meet this requirement, the prior art must contain “clear and unmistakable directions” to the claimed invention (Novozymes A/S v Danisco A/S [2013] FCAFC 6 at [145], referring to General Tire & Rubber Co v Firestone Tyre & Rubber Co Ltd (1971) 1A IPR 121 at 138).
Mars cited the following documents for novelty purposes:
WO 2004/113570 A2 (Mars, Incorporated) 29 December 2004 (WO’570)
WO 2002/102172 A1 (Nestlé Purina Petcare Company) 27 December 2002 (WO’172)
WO 2003/029912 A2 (Hemopet) 10 April 2003 (WO’912)
US 6537213 (Dodds, W. J.) 25 March 2003 (US’213)
Hills submitted that none of these documents disclose the genome-based breed clusters on which the methods of the opposed claims depend. I have found above that the genome-based breed clusters of the opposed specification require breeds of an animal species to be grouped (clustered) according to genotypic criteria. If the cited documents do not teach this feature, they cannot anticipate the claimed invention.
WO 2004/113570
Mars opposed claims 1-4, 6-13, 16-25 for lack of novelty over WO’570.
WO’570 was published on 29 December 2004, after the earliest priority date of the opposed claims, but having an earlier priority date than the opposed claims. Consequently, this is a “whole of contents” citation that is prior art for novelty purposes only.
WO’570 is titled “Genotype test”. On page 1, the invention takes advantage of the breed structure of the dog population to provide a genetic test for, among other things, determining the nutritional requirements of a dog. On page 6, detecting the presence or absence of SNP markers in the dog’s genome allows identification of its breed ancestry, i.e. the breeds that have contributed to the dog’s genome, also described as its genetic breed inheritance.
Mars submitted that the references in WO’570 to the “breed inheritance” and “genetic breed inheritance” of a dog were equivalent to the breed clusters of the opposed application. In contrast, Hills argued that WO’570 uses SNPs to assign a given dog to a particular, traditionally recognised breed and uses the term “genetic breed inheritance” as a definition of the ancestry of a dog.
Consistent with the parties submissions, Mars’ experts understand WO’570 to anticipate claims of the oppose application (NF#2 at [49]; [53]; [53]-[59], [66]), while Hills’ experts understand it to describe the use of genetic analysis to identify the traditionally-recognised breed(s) a particular dog belongs to in order to design a nutritional diet (SW at [36]; WB [47]). Dr Wu and Dr Brown agree that WO’570 does not disclose the construction of breed clusters as defined in the opposed application (SW at [36], [46]; WB [46], [47]).
In his supporting evidence, Dr Fretwell construed the terms “genetic breed inheritance”, “breed inheritance” and “canine breed inheritance” in WO’570 as comparable to the genome-based breed clusters and breed clusters in the opposed specification (NF#2 at [61]). However, in his reply evidence Dr Fretwell acknowledges as “perhaps correct” Dr Brown’s opinion that “genetic breed inheritance” is used in WO’570 as a definition of the ancestry of a dog indicative of traditionally recognised breeds (NF#3 at [47]). Notwithstanding, Dr Fretwell continued:
“In my opinion, the information of [WO’570] (genetic markers in the form of SNPs) allows identification of the ‘breed inheritance’ of a dog and to then place the dog in a breed, but equally in a ‘genetic breed inheritance’ group too. I state this because the Parker et al … document (and others) describes how to determine a commonality cluster based on similar genetics. The dog can be assigned to a ‘(traditional) breed cluster’ as well as to a ‘(genetic) breed cluster’ according to Parker et al. It is then straight forward, using known nutritional requirements of the breed cluster, and having regard to other factors such as disease susceptibility etc., to come up with a diet that is nutritionally appropriate for the ‘(genetic) breed cluster’.” (NF#3 at [49])
I understand Dr Fretwell to be saying that the SNPs taught by WO’570 could be used as input data to allocate a dog to either a traditional breed cluster or to a genome-based breed cluster as described by Parker. Dr Fretwell’s evidence continued at [90]:
“While … [WO’570] does not use the term ‘genome-based breed clusters’, the methodology employed in providing a method for assessing a nutritional requirement, disease susceptibility or behavioural characteristic of a dog indeed identifies the genetic breed inheritance of a dog based on genotypic markers (SNPs) of the dog. I refer to pages 12-15 of the WO document in support of my view that breed cluster grouping, based on genetic information, is also performed in the WO document.” (Emphasis in original)
Consistent with Hill’s expert evidence (SW at [36], WB at [47]), on pages 12-15, WO’570 describes the use of one or more SNPs to determine a dog’s genetic inheritance from one or more of the traditional breeds or breed categories. It does not disclose the construction of the genome-based breed clusters of the opposed application.
Mr McCallum had difficulty understanding the scope of the term ‘breed clusters’ in the opposed application (TM#2 at [9]-[10], [17], TM#3 at [33]). He construed them as encompassing “any cluster of animals that are related by virtue of their genetic inheritance” (TM#2 at [17]), which would include traditional breed groupings such as those of the AKC. Furthermore, Mr McCallum does not clearly distinguish the methods of deriving the genome-based breed clusters described in the opposed application from the opposed claims that require an animal to be assigned to such a cluster (TM#2 at [17], TM#3 at [33], [66]). Since the proper construction of “genome-based breed clusters” is critical for novelty purposes, I have accorded Mr McCallum’s evidence in this regard less weight that the other experts.
WO’570 does not disclose the identification of genome-based breed clusters within the meaning of the opposed application. I find the claims novel in light of WO’570.
WO 2002/102172
Mars opposed claims 1-3, 6, 8-13, 16-20 and 25 for lack of novelty in light of WO’172.
WO’172 is titled “Methods and apparatus for customising pet food”. On page 1, the invention relates generally to methods for manufacturing pet foods and particularly to a process for manufacturing a pet food customised to the health and nutrition requirements of an individual pet.
Mars described WO’172 as disclosing methods for manufacturing a pet food comprising “obtaining a pet profile and receiving through an electronic interface a user input data that includes an individual pet profile for the pet as well as obtaining a biological analysis and processing the received information to create a pet food formula specific to the pet”, and associated systems. Mars submitted that the pet profile information can include phenotypic information and genetic information.
Hills did not dispute that WO’172 discloses a computerised method of creating customised pet food, which involves the use of biological information for a given pet. However, it argued that the document discloses breeds, breed categories or breed groups according to the traditionally recognised breed categorisations, such as the AKC, rather than genome-based breed clusters.
None of the experts understood WO’172 to disclose the use of genome-based breed clusters in the design of pet food formulations (NF#2 at [90], NF#3 at [57]; TM#2 at [79], [84]; SW at [38]; WB at [58]). At best Mars’ expert evidence is to the effect that such information was available before 2004 and could have been used with the methods of WO’172 (NF#2 at [90]; NF3# at [57]; TM#3 at [54]).
Mars has not established that WO’172 contains clear and unmistakable directions to the methods of the opposed claims. The claims are novel in light of WO’172.
WO 2003/029912 & US 6537213
Mars opposed claims 1-26 for lack of novelty in light of the separate disclosures of WO’912 and US’213.
These documents, by the same inventor, each describe methods and systems for managing the nutrition and/or modulating disease of non-livestock animals. The methods/systems use various datasets relating to the animal species or a group of species, which include phenotypic and genotypic data, and laboratory data, and the use to determine a nutritional regimen for the management of the animal (WO’912 paras 2, 4-10, 14, 15-16; WO’213 at column 1, line 12 to column 3, line 38; column 4 lines 50-64 column 6, lines 21-33; NF#2 at [108]-[109]; TM#2 at [86]-[89]; WB at [59]-[60]; SW at [39]-[40]). At the hearing, Mars argued that the person skilled in the art would use both the traditional and genotypic datasets taught by WO’912 or US’213 from which they could derive different clusters from those of the AKC, and thus arrive at the claimed invention.
WO 2003/029912
WO’912 is titled “Animal healthcare, well-being and nutrition”. The technical field of the invention is described on page 1 as a method, system and apparatus for the management of comprehensive and cumulative genetic and health assessment databases in relation to animals worldwide. On page 4, the invention is the first to store and/or present phenotypic and genotypic information as a comprehensive and cumulative assessment of individual animal subjects, families of subjects, breeds of subjects, or species of animals in a computerised format which is available through computer networking to authorised remote users.
At para 18, in a preferred embodiment, the database for at least one of the species or the group is updated over time to obtain cumulative data. The term ‘group’ is defined at para 38 as having many different characteristics and can include a specific breed of an animal, or its purpose (e.g. companion pets, performance animals, etc.). A group can involve animals living in a particular geographical area or climatic conditions or urban or rural areas, or animals of various ages, intact or neutered or for reproduction. The description summarises the term ‘group’ as being used in a very broad sense, which can apply to any group the user wishes to enquire of the database, e.g. any selected subset of the healthy or diseased or disordered animals within the entire database.
At paras 36-37, the genotypic information on the database relates to the genetic constitution of the animals and includes: pedigree, family history or heritable physical characteristics of the animal or genetic screening tests, DNA testing genomic mapping, or assessment of the specific phenotypic characteristic(s) resulting from expression of the phenotype for known or suspected congenital and heritable traits, which may include laboratory test results.
None of the experts understood WO’912 to specifically disclose the genome-based breed clusters of the opposed application (WB at [59]; SW at [39], [47]; NF#3 at [57]; TM#3 at [54]). Notwithstanding, given the broad definition in WO’912 of the term “group”, Dr Fretwell understands it to encompass the “genome-based breed clusters” of the opposed application (NF#2 at [110]-[111]). Dr Fretwell understands references to a group of a species to be equivalent to a “breed cluster” (NF#2 at [111]). Responding to Drs Brown and Wu’s evidence, Dr Fretwell understands WO’912 to describe:
“… various genomic bases for designing nutritional products, including developing databases of information for an animal. The documents are relevant to me in that they provide information for developing nutritionally relevant food products or diets for an animal having regard to genomic information.” (NF#3 at [57])
Dr Fretwell continues at NF#3 [100]:
“… at pages 15 to 16 and 19 there is described the use of a variety of diagnostic tests to provide both phenotypic and genotypic descriptors. The genotypic tests include [a number of tests that provide genetic information]. …. these techniques … provide both qualitative and quantitative genetic data that could be used to define a ‘breed cluster’ and be used to derive allele frequencies for different breeds and enable genetic breed clusters to be derived using the software available at the Priority Date. [WO’912] describes the use of both phenotypic and genotypic descriptors to construct “breed clusters””.
The passages that Dr Fretwell refers to describe diagnostic test panels including standard laboratory test panels for general health and endocrine and immunologic function, infectious diseases and other diagnostic tests, some of which provide genetic information. I understand from Dr Fretwell’s evidence that the genetic information could be used to derive the genome-based breed clusters of the opposed application, using software that, while not disclosed in WO’912, was available before the priority date. Such a disclosure falls short of providing clear and unmistakable directions to the invention of the opposed claims.
Similarly, Mr McCallum understands WO’912 to use terminology that includes both phenotypic data and genomic data, which information would and could be used for developing specific dietary products for an animal (TM#3 at [31], [54]). This evidence does not establish that the document provides clear and unmistakable directions to do so.
Mars has not established that WO’912 anticipates any claim.
US 6537213
US’213 is titled “Animal health care, well-being and nutrition”. In column 1, the invention is concerned with animal health diagnosis and in particular the testing, diagnosis and prediction of diseases and disorders of animal companions such as dogs and cats. In a preferred embodiment, the invention comprises combining genetic data with health assessment data to predict health, disease and disorder probabilities and longevity of selected animals (column 6, lines 26-33). The genetic database is either a specific file of a selected animal or a generalized animal database relating to group characteristics, and is cross-relatable with the phenotypic database of particular selected subject animals (column 7, para 3).
None of the experts understood US’213 to specifically describe the use of genome-based breed clusters as defined in the opposed specification (WB at [60]; SW at [40], [86]; NF#3 at [57]; TM#3 at [54]). However, in his second declaration at [111], Dr Fretwell identified claim 32 of US’213 as relevant to opposed claim 1. Claim 32 is as follows:
“A method of managing the nutrition of a non-livestock pet animal subject comprising:
a) obtaining a phenotypic and genotypic databases relating to at least one of:
i. the species of the animal subject,
ii. a selected group of the species;
b) obtaining data relating to the subject, the data including laboratory test data relating to the subject;
c) correlating the databases of a) with the data of b) by a computer to determine a relationship between the databases of a) and the data of b); and
d) determining, based on c), a regimen for the management of the subject.
As in WO’912, the term “group” in part a)ii of this claim is defined in a very broad sense and applies to any group the user wishes to enquire of the database, e.g. any selected subset of the healthy or diseased or disordered animals within the entire database (column 22, line 67 to column 23, line 4). As for WO’912, Dr Fretwell understood the selected group of US’213 to include the genome-based breed clusters of the opposed claims (NF#2 at [110]-[111]). Consistent with his submissions for WO’912, Dr Fretwell understood US’213 to discuss various genomic bases for designing nutritional products, and considered it relevant to the provision of information for developing nutritionally relevant food products or diets for an animal having regard to genomic information. Mr McCallum’s evidence regarding US’213 is identical to that he provides for WO’912 (TM#3 at [31], [54]).
For the reasons provided above for WO’912, I find that US’213 does not provide clear and unmistakable direction to the subject matter of the opposed claims. The claims are novel in light of US’213.
In summary, Mars has not discharged the onus of establishing that any claim lacks novelty.
Inventive step
Under the provisions of subsections 7(2) and 7(3) of the Patents Act 1990, an invention is taken to involve an inventive step when compared with the prior art base unless it would have been obvious to a person skilled in the art. ‘Obvious’ means ‘very plain’ (Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) [2007] HCA 21 at [51] - [52]; (2007) 72 IPR 447 at 461 [51] - [52]). The invention must be obvious in the light of the common general knowledge as it existed in the patent area before the priority date, either on its own or together with information in a document, or combination of documents, that the person skilled in the art could, before the priority date of the relevant claim, be reasonably expected to have ascertained, understood and regarded as relevant and, where necessary, combined.
The test for obviousness was provided by Justice Aicken in Wellcome Foundation Ltd v VR Laboratories (Aust) Pty Ltd [1981] HCA 12 at [45]; 148 CLR 262 at 286 as follows:
“The test is whether the hypothetical addressee faced with the same problem would have taken as a matter of routine whatever steps might have led from the prior art to the invention, whether they be the steps of the inventor or not.”
The High Court in Aktiebolaget Hässle v Alphapharm Pty Ltd (‘Alphapharm’) [2002] HCA 59 at [51]- [53]; 212 CLR 411 at [51]-[53] approved this approach, in addition to that taken in Olin Mathieson Chemical Corporation v Biorex Laboratories Ltd [1970] RPC 157 at 187 in which Graham J had posed the question:
“Would the notional research group at the relevant date in all the circumstances directly be led as a matter of course to try [the claimed invention] in the expectation that it might well produce a useful [desired result]?”
Where the invention lies in a combination of features, the question is whether the combination, not each individual feature, is obvious when compared to the prior art base (Alphapharm [2002] HCA 59 at [41]; 212 CLR 411 at [41]; Minnesota Mining & Manufacturing Co v Beiersdorf (Australia) Ltd (Minnesota Mining) [1980] HCA 9 at [116]; (1980) 144 CLR 253 at 293).
The usual approach to determining inventive step is the problem-solution approach. Once the problem has been formulated and the common general knowledge and the prior art base has been determined, the question of whether the claimed solution is obvious must be addressed.
The problem
In determining the problem or ‘starting point’ for considering inventive step, I am mindful of the words of the majority of the Full Court in AstraZeneca AB v Apotex Pty Ltd [2014] FCAFC 99 at [202]-[203] as follows:
“If the problem addressed by a patent specification is itself common general knowledge, or if knowledge of the problem is s 7(3) information, then such knowledge or information will be attributed to the hypothetical person skilled in the art for the purpose of assessing obviousness. But if the problem cannot be attributed to the hypothetical person skilled in the art in either of these ways then it is not permissible to attribute a knowledge of the problem on the basis of the inventor’s “starting point” such as might be gleaned from a reading of the complete specification as a whole.”
Reflecting a statement in paragraph 8 of the description, Mars submitted that the problem addressed by the application is to develop new methods for designing nutrition and health programs including formulating foods for animals based on genomic information. Mars submitted that knowledge of this problem was common general knowledge in the art or section 7(3) information in view of Swanson.
The expert evidence establishes that the principle set out in para 8 of the description, that new methods of formulating foods for animals based on genomic information would represent a useful advance in the art, was common general knowledge in the art (NF1 at [39]; TM#1 at [33]; WB at [26]). It follows that this information can be attributed to the person skilled in the art as the problem or starting point for assessing obviousness.
The person skilled in the art
In general, the skilled person or addressee is the person who works in the art or science with which the invention is connected. He or she is a person, or team, likely to have a practical interest in the subject matter of the invention. While the skilled person may be assumed to be well versed in the relevant art, such a person must be taken to be non-inventive. (Root Quality Pty Ltd v Root Control Technologies Pty Ltd [2000] FCA 980 at [71]-[72]; 49 IPR 225 at [71]-[72] referring to Catnic Components Ltd v Hill &Smith Ltd [1982] RPC 183 at 242 and General Tire [1972] RPC 457 at 485; Minnesota Mining [1980] HCA 9 at [115]; 144 CLR 253 at 293).
The experts
After completing a BSc. in biological sciences and a PhD in human genetics, Dr Fretwell joined Mars Waltham Centre for Pet Nutrition in the UK in November 1999 and has since been involved in applying molecular genetics to the field of pet nutrition, first as a researcher and later as Technology Development Manager (NF#1 at [10]-[14], [19], [27]).
Mr McCallum has a BSc. in chemical engineering. Before the priority date he worked for Mars and Mars subsidiary companies in Australia (from 1997 to mid-2000) and Europe (mid-2000 to 2003), developing pet foods including those targeted at treating specific conditions. (TM#1 at [10]-[13]; TM#2 at [13]).
Before the priority date Dr Brown, a nutritionist, was a research-based academic in the Animal Science division of the University of New England in Armidale, Australia (WB at [4], [53]). Dr Brown completed a MSc. in 2004 and a PhD in 2008 in the field of canine nutrition (Exhibit WB-2). For her Masters project Dr Brown investigated dietary means for reducing the risk of urinary stones in dogs and the impact of diet on canine dental health (WB at [5] and Exhibit WB-2).
Dr Wu holds a BSc. in botany, and a MSc. and PhD in plant reproductive biology. Consistent with his research positions and publications before August 2004, Dr Wu describes his expertise as predominantly in the field of genetics, genomics and molecular biology in animals and plants (SW at [3]-[4] and Exhibit SW-2).
The hypothetical person skilled in the art or team
Mars submitted that the relevant person skilled in the art or team would include an animal nutritionist, process engineers, chemical engineers, including those with experience in pet food design, and an animal geneticist. Hills did not explicitly identify the person skilled in the art but submitted that the team did not include a geneticist or a person with experience in bioinformatics.
The expert evidence is consistent with the person skilled in the art or team including an animal nutritionist, process engineers and chemical engineers, including those with experience in pet food design (NF#1 at [16], [21], [25]-[27]; TM#1 at [21]-[24]; WB at [23]; TM#3 at [14]). Different people would be involved at different times in the process of product development (NF#1 at [17]).
Mr McCallum confirmed that Mars had an active pet (including dog) food research and development presence in Australia, including at least two animal nutritionists (TM#3 at [14]). Dr Fretwell states that the work linking genetics with nutrition was being carried out in the US, UK and Europe, and Australian pet nutritionists could follow the overseas developments in sector-relevant published literature and journals as did he (NF#3 at [20]). This is consistent with Dr Brown’s statement at [13] that during 2004 she actively reviewed the published literature relating to animal nutrition, including information in the field of nutrigenomics.
Before the priority date, the relevant team at Mars included a geneticist (NF#1 at [25]-[27], [29], [39], [53]; NF#3 at [50]; TM#3 at [13]). It remains to be determined whether the relevant person skilled in the art or team in Australia would have done so. In my view they would, albeit that it may have been achieved by international collaboration.
The evidence establishes that before the priority date, a company involved in the pet food industry operated at a global level and collaborated across borders in developing new food products (TM#1 at [14]; NF#1 at [19]; TM#1 at [19]). While working at a Mars subsidiary company in Australia between 1998 and July 2000, Mr McCallum was responsible for developing new pet food products for both the Australian and Japanese markets (TM#1 at [13]). From January 2003, Mr McCallum was working in the UK responsible for a research and development team of eight people based in three countries - the UK, Germany and France (TM#1 at [19]).
Dr Fretwell, a geneticist, states his personal knowledge base extended to bioinformatics before the priority date (NF#1 at [26]). However, he identifies bioinformatics as a skill of a person involved in (which I understand to mean practically applying) nutrigenomics, rather than a geneticist (NF#1 at [26]). Therefore, I cannot conclude that all geneticists have bioinformatics skills. Nor has it been established that the hypothetical person skilled in the art would possess bioinformatics skills by other means. Although an animal nutritionist and a geneticist may follow the literature for nutrigenomics, they would not necessarily gain skills in bioinformatics in this way.
Given the evidence as a whole, I see no reason to conclude that the relevant person skilled in the art in Australia would be substantially different to overseas. On this basis, I find that the hypothetical person skilled in the art in Australia would possess the skills of a geneticist and a team member, such as the geneticist or nutritionist, would have an interest in nutrigenomics.
The experts and the hypothetical person skilled in the art
As a chemical engineer working in the field of pet food nutrition before the priority date Mr McCallum is representative of the person skilled in the art. As an animal nutritionist with research interests in canine nutrition before the priority date Dr Brown’s evidence is relevant to the opposition. Dr Wu does not appear to have worked in the field of animal nutrition before the priority date (SW at [1]-[6], [10] and Exhibit SW-2) and this may affect the weight accorded to his evidence.
Dr Fretwell worked in the pet food industry developing nutritional food formulations before the priority date and as such his evidence is relevant to the opposition. However, I accept Hills’ submission that as a co-inventor on a number of patent applications (including WO 2004/113570 cited in this opposition), Dr Fretwell does not represent the non-inventive person skilled in the art. This may affect the weight I give to aspects of his evidence.
The common general knowledge in the art
A definition of common general knowledge was provided by Aitken J in Minnesota Mining [1980] HCA 9 at [115]; 144 CLR 253 at 292:
“The notion of common general knowledge itself involves the use of that which is known or used by those in the relevant trade. It forms the background knowledge and experience which is available to all in the trade in considering the making of new products, or the making of improvements in old, and it must be treated as being used by an individual as a general body of knowledge.”
Information cannot be treated as part of the common general knowledge in the absence of evidence of its general acceptance and assimilation by persons skilled in the relevant art (Alphapharm [2002] HCA 59 at [31]; 212 CLR 411 at [31]).
Mr McCallum, Dr Brown and Dr Fretwell agree that the ‘Background’ section in the opposed specification provides a good summary of what was well known to workers in the pet food industry in 2004 (TM#1 at [33]; WB at [26]; NF1 at [39]). Relevantly, para 3 of the description states that selective breeding by humans has led to a degree of reproductive isolation in dogs, which has been formalised since the mid-19th century by the advent of breeding in clubs according to breeding standards documented by the AKC and European and Japanese Kennel Clubs.
Animal nutritional needs have been known for centuries and a large pet food industry has developed which, before the priority date, developed and marketed nutrition products for different phenotypic characteristics of an animal such as the size, body condition or growth rate, or for one or more specific breeds or phenotypically-defined breed groups such as those of the AKC (paras 4-5 of the description and NF#1 at [37]-[38], [42], [44]; TM#1 at [31]-[32], [34]; WB at [30]).
Paragraph 7 of the description states that healthy nutrition is important for pet care, but many owners do not understand their pet’s nutritional requirements, and have difficulty determining if it is receiving a balanced and healthy diet. Paragraph 8 states that with recent innovations in health and medicine based on information from genome projects, genetics is becoming an important component in determining health and nutrition programs. New methods for designing nutrition and health programs including formulating foods for animals based on genomic information would represent a useful advance in the art (para 8).
Scientific journal articles
Mars asserted that the sort of information the person skilled in the art would have accessed during the course of their work was common general knowledge in the art before the priority date, in particular the contents of two scientific journal articles cited in this opposition (Parker and Swanson identified below). However, the expert evidence is directed to whether the person skilled in the art had read and understood these documents before the priority date (TM#3 at [29]; WB at [28], [45]; NF#2 at [66]; NF#3 at [77]). The evidence does not establish that the information in these documents was generally accepted and assimilated by the relevant person skilled in the art before the priority date such that it was common general knowledge.
Prior art base
The prior art relied on by Mars to establish obviousness is as follows:
WO 2004/113570 A2 (Mars, Incorporated) 29 December 2004
Parker, H. G. et al. (21 May 2004) Science 304: 1160-1164 (Parker)
Swanson, K. S. et al. (2003) Journal of Nutrition 133: 3033-3040 (Swanson)
I have found above that the priority date of the opposed application is 30 August 2004. Therefore, WO 2004/113570 is not part of the prior art base for consideration of inventive step.
Mars relied on the combined disclosure of Parker and Swanson to establish obviousness of the claimed invention in light of both the common general knowledge in the art and section 7(3) information.
I have found above that the information in Parker and Swanson is not common general knowledge in the art and consequently this aspect of the opposition fails. For Parker and Swanson to be section 7(3) information, it must be established that before the priority date the person skilled in the art could be reasonably expected to have ascertained, understood and regarded these documents as relevant.
Mars submitted that before the priority date the person skilled in the art would look to Swanson, which provides a clear link between genetic information and pet food formulation. Then, in seeking relevant genetic information, the person skilled in the art would understand the role of Parker as a source of such information for dogs. I understand Mars’ submission to be that the significance of Parker would only become apparent after Swanson had been ascertained, understood and regarded as relevant. Further searching to find relevant information necessary to supplement information in a document would be common in many fields of technology, and I see no reason why section 7(3) would not support such a process.
Ascertained, understood and regarded as relevant
Swanson was published in by the American Society of Nutritional Sciences in the Journal of Nutrition. It is titled “Nutritional Genomics: Implications for Companion Animals”. Parker is also published in a prestigious journal. Both are in the English language, so where ascertained they would have been understood. It is reasonable to expect that Swanson would have been ascertained by the person skilled in the art researching the problem. I will address the further requirements of section 7(3) in my reasoning below.
Hills submitted that in Swanson the starting point of investigation is a consideration of how a particular diet may alter gene expression in an animal. Hills considered the document to teach that an animal’s diet can be altered in order to achieve a more beneficial gene expression pattern in that animal. However, this narrow interpretation of Swanson as limited to functional genomics is not supported by a review of the document and the evidence.
Consistent with its title, Swanson is a review of nutritional genomics as applied to companion animals. It summarises the state of the art at the time in the application of genome maps (i.e. purely genotypic information) to veterinary medicine, including nutritionally-related diseases (Abstract; page 3033 right column; pages 3034-3036 under the heading “Current use of genome maps” and the first para under “Nutritional genomics” on page 3036). Swanson predicts that functional genomics (investigation of RNA and protein expression, protein localization and protein-protein interactions) will emerge as important areas of study that will play a vital role in the future of pet foods (Abstract, pages 3036-3038 under the heading “Nutritional Genomics” commencing at second para). The Abstract concludes “Nutritional genomics and proteomics will definitely play a vital role in the future of pet foods.”
Dr Fretwell understands Swanson to describe “the concept and potential for the application of genetic information, as it related to genetic diseases, along with known breed characteristics, to the design and development of specialised diets/pet food formulations”, i.e. the concept of nutrigenomics (NF#1 at [52]). Dr Brown and Dr Wu’s evidence is not inconsistent with this (WB at [28], [54]; SW at [22], [37]).
Dr Fretwell considers Swanson relevant in that it provides a clear statement of the future of nutrigenomics in devising animal pet food products (NF#3 at [31]). Mr McCallum understands the document to relate to nutrigenomics and on this basis I understand he considers it relevant (TM#3 at [29]). Dr Brown describes the information in Swanson as predominantly conceptual and speculative (WB at [28], [55]-[56]). However, her evidence regarding Swanson does not take account of the problem I have identified above and consequently I give it less weight. I find Swanson would be considered relevant.
There was no dispute that Swanson does not discuss genetic clustering, or the use of genetic clustering analyses to develop a nutritional diet for an animal. The disclosure of Swanson is consistent with the statement in the specification that new methods of formulating foods based on genomic information would be an advance in the art. However, while it predicts genomics as important to the future of pet food (NF#3 at [16]), as a review article it does not provide practical instructions or experimental examples (WB at [28], [54]; SW at [22], [37]). In this situation, it is reasonable to conclude that the person skilled in the art would seek relevant genomics information, in order to apply genomics to nutrition as suggested by Swanson.
There was no dispute that Parker teaches an alternative way of grouping animals using genetic information (NF#1 at [51]; SW at [35]; WB at [45]; TM#3 at [28]) and as such it would have been ascertained. Dr Fretwell believes that people working in the relevant field at the time would have been aware of the implications of Parker’s work for the design and manufacture of customised dog food products (NF#2 at [66]), and would equally view both Swanson and Parker as relevant in linking the genomic history of dogs with nutrigenomics (NF#3 at [31]). I understand from Dr Fretwell’s evidence that he understood and regarded Parker as relevant and would have combined the information with Swanson. However, I must consider Dr Fretwell’s capacity for invention.
I infer from her evidence that Dr Brown did not think Parker relevant because it does not refer to canine nutrition (WB at [28], [45]). However, her opinion appears influenced by her view that it is unlikely that before the priority date anyone would have considered relating genomic data to diet design (WB at [30]). Given the problem itself provides this context, as does Swanson, I accord her evidence less weight. Dr Wu states Swanson does not fully describe the concept of nutrigenomics in the absence of discussion of the effect of diet on particular genes (SW at [22]). His evidence does not challenge Dr Fretwell’s conclusion, which is based on a broader view of the information.
In the context of the problem, I find that Parker would be ascertained, understood and regarded as relevant by the person skilled in the art who would have combined its information with that of Swanson.
Was the invention obvious?
I have found above that a key feature of the method and system of independent claims 1 and 16 is that of genome-based breed clusters, i.e. animal breeds clustered (grouped) according to genotypic criteria.
Whether the claimed invention is obvious in light of the combined disclosure of Swanson and Parker turns on the answers to two questions. The first is whether, before the priority date, the person skilled in the art seeking new methods of formulating foods for animals based on genomic information would directly be led as a matter of course to use Parker’s four breed clusters as an alternative to traditional breed groupings, with a reasonable expectation of that this would produce a useful, desired result. The second is whether that same skilled person would directly be led as a matter of course to replicate the type of analysis described in Parker to generate genome-based breed clusters and in that manner proceed to the claimed invention.
The answer to the first question is no. The evidence establishes that the person skilled in the art before the priority date would not directly be led as a matter of course to apply Parker’s four breed clusters to the formulation of pet food in the expectation that it might well produce a useful, desired result.
The opposed specification exemplifies its invention using Parker’s four breed clusters, which are designated Clusters I-IV in the opposed specification (paras 40 and 43-44). In reviewing the specification in his second declaration at [10], in effect, Dr Fretwell considers the merits of the genome-based breed clusters disclosed by Parker as applied to the formulation of pet foods. Dr Fretwell states that the four clusters do not show strong trends with respect to nutritional differences or similarities that are common to the breeds within the cluster, but markedly different to other clusters. In particular, he notes that Clusters I and II contain breeds of all three dog size categories; that there is no consistent trend difference in activity level, shedding or trainability between the four clusters; and that there is no pattern in the stated levels of disease predispositions. As such, he considers Clusters I-IV (i.e. Parker’s breed clusters) are not reliable for assessing nutritional needs. He concludes that consideration by breed alone would be far more reliable.
The evidence does not establish that the answer to the second question is yes. I have found above that the hypothetical person skilled in the art in Australia possessed the skills of a geneticist, and had an interest in nutrigenomics, but did not necessarily possess the practical skills in bioinformatics that would be required to replicate the type of analysis described by Parker to generate genome-based breed clusters.
Mars has not established that the claimed invention lacks an inventive step in light of the combined disclosure of Swanson and Parker.
CONCLUSION
The opposition fails.
COSTS
Hills submitted that costs should follow the event. Mars agreed to the usual practice.
It is normal in actions before the Commissioner that costs should follow the event. However, subregulation 5.20(6) permits the conduct of the parties in filing their summary of submissions to be taken into account in awarding costs. Subregulation 5.20(4) specifies the time that each party must file a summary of submissions.
Both parties filed their summaries of submissions two days later than required by 5.20(4) and no explanations for the delay were provided. Hills has been successful in this opposition and I would ordinarily award costs in its favour. Since Mars’ late filing would have contributed to Hills’ delay, and Hill’s filed its summary within 5 working days of Mars, it is appropriate to award costs according to Schedule 8 against Mars.
Barbara Akhurst
Delegate of the Commissioner of Patents
0
21
0