Printer Friendly

A shift in the Canadian law of obviousness: a comment on Sanofi-Synthelabo Canada Inc. v. Apotex Inc.

Abstract

The Supreme Court of Canada's 2008 decision in Sanofi-Synthelabo Canada Inc. v. Apotex Inc. is significant for patent law generally because it modified the legal tests for two of the three fundamental requirements for patentability: novelty and non-obviousness. This comment focuses on the changes the Court made to the law of obviousness.

This comment argues that Sanofi represents a subtle, but important shift in the law of obviousness, and will trace four of Sanofi's contributions to the jurisprudence. First, the Court emphasizes that the test must be functional and flexible. Second, the Court adopts a formal framework--the "Windsurfing questions"--drawn from the UK law of obviousness. Although the Windsurfing questions largely deal with the same issues that were present in the Beloit test, the new framework permits the obviousness inquiry to be divided into discrete issues as opposed to the Beloit test, which condensed these issues into a single question. Third, the Court adopts an "obvious to try" approach that brings Canadian law closer to the UK and US law of obviousness. Fourth, the Sanofi decision appears to change our conception of the person having ordinary skills in the art (PHOSITA). A natural extension of a functional approach to obviousness is the construction of a PHOSITA with an ordinary level of creativity, as opposed to one with no creativity.

This comment argues that this shift is a positive development because it is commensurate with a purposive approach to patent law. The purpose of the obviousness test is to filter out those inventions that do not appreciably add to the store of public knowledge from those that are truly inventive. The emphasis on a functional and flexible approach should assist courts in undertaking a purposive analysis of obviousness, as opposed to placing undue emphasis on a particular phrase or legal formulation. Moreover, the flexibility in the new test should permit the law of obviousness to be better calibrated according to industry-specific differences in innovation.

The comment concludes with a discussion of the interaction between selection patents and obviousness. Aside from reiterating that selection patents are valid in principle, the Sanofi decision does little to clarify this interaction. Nevertheless, this comment argues that the decision of the Federal Court in BMS v. Apotex, released three months after Sanofi, is a coherent interpretation of the Supreme Court's pronouncements on selection patents. Specifically, this approach involves treating selection patents similarly to other patents and using the framework of the Sanofi test to address the issues that selection patents raise.

Resume

La decision de la Cour Supreme du Canada, en 2008, dans l'arret Sanofi-Synthelabo Canada Inc. c. Apotex Inc. est importante pour le droit des brevets parce qu'elle a modifie le test pour deux des trois exigences fondamentales de la brevetabilite : la nouveaute et la non-evidence. Ce commentaire portera sur les changements faits a la loi de la non-evidence.

Ce commentaire soutiendra que Sanofi represente un changement de la loi de la non-evidence, a la fois subtil et important, et il tracera quatre contributions de Sanofi a la jurisprudence. Premierement, la Cour a souligne que le test devrait etre fonctionnel et flexible. Deuxiemement, la Cour adopte un cadre formel--les Questions << Windsurfing >> --developpees dans la loi de la non-evidence du Royaume-Uni. Malgre que les questions << Windsurfing >> impliquent largement les memes questions qui etaient presentes dans le test Beloit, le nouveau cadre permet que l'enquete quant a la non-evidence soit divise en causes distinctes, ce qui est different du test dans Beloit qui a condense tout les causes en une seule question. Troisiemement, le Cour adopte une notion << d'essai evident >> qui rapproche la loi canadienne, quant a la non-evidence, a celle du Royaume Uni et des Etats-Unis. Quatriemement, la decision dans Sanofi semble changer notre conception de la personne possedant des habiletes ordinaires dans l'art (PPHOA). Une extension naturelle d'une approche fonctionnelle quant a la non-evidence est la construction d'un PPHOA avec un niveau ordinaire de creativite au lieu d'un PPHOA sans creativite.

Ce commentaire soutiendra que ce changement constitue une evolution positive parce qu'il est proportionnel a une interpretation utile du droit des brevets. Le but du test de la non-evidence est de filtrer les inventions qui n'ajoutent pas appreciablement a la connaissance du public de celles qui sont reellement inventives. L 'accent sur un test qui est fonctionnel et flexible aidera les cours a entreprendre une analyse utile de la non-evidence, au lieu de trop mettre l' accent sur une phrase ou une formulation particuliere. De plus, la flexibilite du nouveau test devrait permettre a la loi de la non-evidence de mieux se calibrer aux differences en innovation parmi les differentes industries.

Ce commentaire conclura avec une discussion de l'interaction entre les brevets de selection et la non-evidence. Mis a part la repetition que les brevets de selection sont en principe valables, la decision dans Sanofi ne clarifie guere cette interaction. Neanmoins, ce commentaire soutiendra que la decision de la Cour Federale dans BMS v. Apotex, annoncee trois mois apres celle dans Sanofi, est une interpretation coherente des declarations de la Cour Supreme sur les brevets de selection. En particulier, cette methode consiste a traiter les brevets de selection de la meme facon que les autres brevets, et a utiliser le cadre du test dans Sanofi pour aborder les questions concernant les brevets de selection.
  I INTRODUCTION
 II BACKGROUND: PATENT LAW AND THE OBVIOUSNESS STANDARD

The Purpose of the Patent Act and the Patent Bargain
Requirements for a Valid Patent
Law of Obviousness--Relationship to the Patent Bargain
Canadian Obviousness Standard Prior to Sanofi

III SANOFI v. APOTEX

Facts and Lower Court Decisions
The Supreme Court of Canada's Decision

 IV SHIFT IN THE LAW OF OBVIOUSNESS

Emphasis on Functional Approach
Formal Structure of Restated Windsurfing Questions
Obvious to Try Approach
    Obvious to Try Factors
    Pfizer v. Apotex: Sanofi Interpreted (Part 1)
Does Sanofi Change Our Conception of the PHOSITA?

  V SANOFI TEST, POLICY LEVERS, AND THE PATENT BARGAIN
 VI SELECTION PATENTS AND OBVIOUSNESS

Unresolved Issues
BMS v. Apotex: Sanofi Interpreted (Part 2)

VII CONCLUSION


I INTRODUCTION

On November 6th, 2008, the Supreme Court of Canada released its decision in Sanofi-Synthelabo Canada Inc. v. Apotex Inc. (1) The case involved a dispute between a brand drug manufacturer, Sanofi-Synthelabo Canada Inc. ("Sanofi"), and a generic drug manufacturer, Apotex Inc. ("Apotex"). At issue was the validity of Sanofi's patent covering the compound clopidogrel bisulfate ("clopidogrel"), which is marketed as a blood clot inhibitor in Canada by Sanofi under the trade name Plavix. Although the Court rejected all of Apotex's grounds of appeal, finding the patent to be valid, the Sanofi decision is significant for patent law generally because it modified the legal tests for two of the three fundamental requirements for patentability: novelty and non-obviousness. (The third, utility, was not at issue in the case.) This comment will focus on the changes the Court made to the law of obviousness. (2)

The patent system is commonly understood as a notional bargain; in return for disclosure of the invention to the public, the inventor is granted a time-limited monopoly to exploit the invention--a patent. The law of obviousness precludes the grant of patents for inventions that are obvious extensions of existing public knowledge. Prior to the decision in Sanofi, the leading test for obviousness from Beloit Canada Ltd. v. Valmet Oy held that an invention was obvious if a person having ordinary skills in the art ("PHOSITA") (3) "would ... have come directly and without difficulty to the solution taught by the patent". (4) In Sanofi, the Court explicitly reformulated the law of obviousness by adopting a more flexible standard that incorporates an "obvious to try" approach, which means that an invention may be obvious if it was obvious to the PHOSITA to try the solution taught by the patent.

Despite the Court's attempt to put forth a definitive statement on the law of obviousness, Sanofi has given rise to surprisingly divergent interpretations. At one extreme, Sanofi has been interpreted as radically lowering the bar for invalidating patents on the grounds of obviousness, such that any invention is obvious if it was "worth a try". (5) At the other extreme, Sanofi has been described as a mere "refinement" of the Beloit test, (6) and some commentators have argued that there was no substantive change in the law. (7) Part of this variation in interpretation may be due to the highly polarized nature of pharmaceutical litigation. The pharmaceutical industry forms an important part of the Canadian economy, and pharmaceutical drugs are one of the fastest growing sectors of health care expenditures. (8) Patent law plays a critical role in drug development and marketing. Generic drug manufacturers uniformly seek to lower the bar for invalidating patents as obvious, and brand manufacturers uniformly seek to strengthen patent protection. Part of the variation in interpretation, however, may be due to genuine confusion relating to the vagueness of the phrase "obvious to try" and to the state of the law prior to Sanofi. Accordingly, the first aim of this comment is to explore the impact of Sanofi on the law of obviousness. The second aim of this comment is to argue that this shift is a positive development because it is commensurate with a purposive approach to patent law.

This comment will argue that Sanofi represents a subtle, but important shift in the law of obviousness. It is neither a radical reorientation, nor a mere restatement of the law. The decision shifts the law of obviousness in at least four ways. First, the Court emphasizes that the test must be functional and flexible. Second, the Court adopts a formal framework--the "Windsurfing questions"--drawn from the UK law of obviousness. (9) Although the Windsurfing questions largely deal with the same issues that were present in the Beloit test, the new framework permits the obviousness inquiry to be divided into discrete issues as opposed to the Beloit test which condensed these issues into a single question. Third, the Court adopts an "obvious to try" approach that brings Canadian law closer to the UK and US law of obviousness. This was the most striking change, given the Supreme Court's previous rejection of an "obvious to try" approach. (10) Nonetheless, the Federal Court of Appeal's decision in Pfizer Canada Inc. v. Apotex Inc., released two months after Sanofi, illustrates that the phrase "obvious to try", by itself, reveals little about the substance of the test, and that the new test is not a radical change. (11) Fourth, the Sanofi decision appears to change our conception of the PHOSITA.

I argue that this shift in the law is a positive development because it is commensurate with a purposive approach to patent law. The purpose of the obviousness test is to filter out those inventions that do not appreciably add to the store of public knowledge from those that are truly inventive. According to the patent bargain, only the latter are deserving of patent protection. The emphasis on a functional approach should assist courts in undertaking a purposive analysis of obviousness, as opposed to placing undue emphasis on a particular phrase or legal formulation. Moreover, the flexibility in the new test should permit the law of obviousness to be better calibrated according to industry-specific differences in innovation.

As a separate matter, the decision is of particular significance for the pharmaceutical industry because of the Court's pronouncements on selection patents. Selection patents refer to the factual scenario where a first patent (the "genus" or "originating" patent) is filed that contains a range or class of solutions, and then a second patent (the "selection" patent) is filed whose invention is the selection of a subset of solutions that are perceived to have some advantage over the general class. (12) Ordinarily, the two patents are filed by the same person. While by no means unique to pharmaceuticals, (13) selection patenting is a very common approach in that sector. (14) In the pharmaceutical context, the genus patent typically covers a broad class of molecules that appear to be useful for treatment, and then the selection patent selects from among the genus those molecules that have been determined to have particular benefits. The Patent Act does not make any reference to selection patents, and one of the issues in Sanofi was what analytical framework should be applied to selection patents.

I will argue that Sanofi actually reveals very little about selection patents. The Court merely reiterated that selection patents are valid in principle, and the decision is ambiguous or even silent on two important issues. First, the Court approved a specific test for selection patents, but failed to clarify the interplay between this test and the conventional requirements for a patent. Second, the Court did not address what kind of "improvement" a selection patent needs to demonstrate. Nevertheless, I will argue that the decision of the Federal Court in BMS v. Apotex, (15) released three months after Sanofi, is a coherent interpretation of the Supreme Court's pronouncements on selection patents. Specifically, this approach involves treating selection patents similarly to other patents and using the framework of the Sanofi test to address the issues that selection patents raise.

The comment will begin in Part II with an overview of patent law and the law of obviousness prior to Sanofi. In Part III, I will summarize the facts and decisions in Sanofi. In Part IV, I will describe how Sanofi has shifted the law of obviousness. In Part V, I will argue that this shift is a welcome development because it is commensurate with a purposive approach to patent law. Finally, in Part VI, I will discuss the interaction between selection patents and the law of obviousness.

II BACKGROUND: PATENT LAW AND THE OBVIOUSNESS STANDARD

The Purpose of the Patent Act and the Patent Bargain

Patent law is widely understood to rest on utilitarian considerations. (16) In particular, patent rights are an extension of a system whose purpose is to advance research and development and to encourage broader economic activity. (17) The basic theory is simple--inventions pose a public goods problem. The costs necessary to create an invention, such as those of research and development, are borne by the inventor. Once an invention is disclosed, however, owing to the non-rivalrous and non-excludable aspects of information, it is often simple for competitors to imitate or copy the invention, and the benefits of the invention will accrue to the public. Inventors may have difficulty capturing sufficient benefits in order to cover their costs or to make a profit, and inventions will tend to be produced at a socially sub-optimal level.

The patent system attempts to solve this problem with a "bargain" between the inventor and the public: in return for disclosure of the invention to the public, the inventor is granted a time-limited monopoly to exploit the invention--otherwise known as the "patent". (18) The patent bargain is a notional bargain because patent law, similarly to copyright law, is wholly statutory in nature. (19) Accordingly, it is inappropriate to think of individual patent grants as being actual bargains between the state and the inventor. The requirements for a valid patent and the scope of the monopoly grant are fixed by legislation and case law concerning patents, hence principles of contract law and real and personal property law are not relevant. (20) Therefore, it is more accurate to describe the patent system as a regulatory scheme that creates incentives for innovation. Nonetheless, I will retain the term "bargain" because it provides a useful shorthand for referring to this theoretical underpinning. Although the patent bargain is not codified in the Patent Act (the "Act"), the Supreme Court of Canada has recognized that this concept underlies the entire history of patent grants. (21)

Requirements for a Valid Patent

The requirements for a valid patent grant are specified in the Act. A patent application is divided into two sections: the disclosure and the claims. The disclosure is information that the inventor includes in order to explain how the invention works. The disclosure often describes a physical embodiment of the invention. The claims are used to define the scope of the legal monopoly. (22) When infringement is alleged, the infringer's device or product is compared to the scope of the written claims. The process of construing the scope of the written claims is referred to as claims construction.

Assuming appropriate subject-matter, there are three substantive requirements for a valid patent: (1) utility, (2) novelty, and (3) non-obviousness. (23) First, the patent must be directed to a practical use, and it must do the job the inventor claims that it does. The invention does not need to provide actual social benefits--a functioning invention for a commercially useless purpose meets the utility requirement. Second, the invention must be new. An invention which has been previously disclosed is said to have been "anticipated". Inventions will only be anticipated if they were previously disclosed in a single source. Third, the patent cannot be an obvious extension of existing public knowledge. Justice Hugessen's comments in Beloit are often used to illustrate the distinction between anticipation and obviousness:
   [O]bviousness is an attack on a patent based on its lack of
   inventiveness. The attacker says, in effect, "Any fool could have
   done that." Anticipation, or lack of novelty, on the other hand, in
   effect assumes that there has been an invention but asserts that it
   has been disclosed to the public prior to the application for the
   patent. The charge is: "Your invention, though clever, was already
   known." (24)


In patent prosecution--that is, the patent application process--this final requirement is usually referred to as non-obviousness or inventiveness. Conversely, in patent litigation, the legal test for validity is usually referred to as the obviousness standard. There is a direct, inverse relationship between the non-obviousness requirement and the obviousness standard. (25) I will refer to the obviousness standard as opposed to the non-obviousness requirement, as this term is more commonly used in the literature and in case law.

Patent validity is determined at two different stages. The first stage is patent prosecution. If the patent office approves the application, then the patent is presumptively valid. (26) A subset of patents are subjected to a second stage of scrutiny through litigation as the Act permits defendants to raise invalidity as a defence to an infringement action. (27) A two-stage process which relies on litigation is usually considered to be more efficient than a single thorough determination of validity at the patent office. (28) In Canada, patent prosecution is a dialogue between the applicant and the patent office. It is not an adversarial process--interested third parties may submit evidence, but there is no hearing at which they may participate. (29) The commercial potential of a patent is often not known at the time of filing, and most patents have no commercial value, (30) so few parties have an incentive to participate at this first stage. Consequently, it is likely that many invalid patents continue to enjoy monopoly protection because there is little incentive to challenge them. (31) Therefore, patent litigation can also be considered a public good because it results in the removal of unjustified monopolies. In this sense, it is a reverse of the conventional problem that is addressed through the incentive of patent grants--the party challenging the validity of the patent bears all of the cost, but the benefit of defeating an invalid patent accrues to the public. (32)

Law of Obviousness--Relationship to the Patent Bargain

The three legal requirements noted above--utility, novelty, and non-obviousness--along with other elements of the patent system that affect the length and scope of the patent grant, are the means that operationalize the underlying concept of the patent bargain.

Economic analysis suggests that the obviousness standard has a considerable impact on societal levels of innovation. Innovation can be stifled by either too little or too much patent protection. If the bar for invalidating patents on the grounds that they are obvious is set too low, and too few patents are found valid, then there may be insufficient incentives to undertake costly research and development. Conversely, if the bar is set too high, and too many patents are ruled valid, then patentees may be able to "fence in" broad areas of research and development, and this may have a chilling effect on the research of competitors. (33) Furthermore, if the obviousness standard is too high, then there may be greater transaction costs to the system, including a patent's value for purposes of licensing, assignment, and settlement; whether parties will litigate patents and how frequently; and how many patents a firm will obtain on the same or similar technology in order to avoid litigation. (34) Accordingly, the obviousness standard has been described as a policy lever by which patent law minimizes inefficient transfers of wealth under conditions where a patentee obtains a right to exclude others from making or using their invention, yet does not appreciably add to the store of public knowledge. (35)

If one accepts, as the Supreme Court of Canada has, that the patent system is a statutory scheme that provides incentives for innovation, then the elements of patent doctrine ought to be commensurate with the underlying economic considerations. Specifically, the legal test for obviousness should function as a screen that filters out inventions that would have happened anyway--that is, in, the absence of the incentive of patent rights--from inventive discoveries.

Canadian Obviousness Standard Prior to Sanofi

Section 28.3 of the Act specifies that the invention must not be obvious; the word "obvious", however, is not defined in the Act, and courts continue to apply the common law definition. (36) Prior to the enactment of s. 28.3 in 1989, courts inferred the non-obviousness requirement from the use of the word "invention". (37) The classic formulation of the obviousness test was the "Cripps Question", named after the English barrister who posed this question in the context of a chemical patent in the 1920s:
   Was it for all practical purposes obvious to any skilled chemist in
   the state of chemical knowledge existing at the date of the patent
   which consists of the chemical literature available ... and his
   general chemical knowledge, that he could manufacture [the
   invention at issue] ...? (38)


Although there has been considerable development since the Cripps Question was first posed, (39) it continues to illustrate two core concepts of the obviousness standard. First, it is an objective inquiry. The court is required to examine the issue of obviousness through the lens of the PHOSITA. The PHOSITA is able to look to both existing literature--the "prior art"--and the "common knowledge" of persons working in the field. The PHOSITA is an artificial creation; actual experts at trial can provide evidence regarding the knowledge of the PHOSITA and how the PHOSITA would interpret prior art, but they are not supposed to testify about the ultimate issue of whether a PHOSITA would have found the invention obvious. (40) Second, the obviousness inquiry is a retrospective exercise. (41) Accordingly, hindsight bias is always a danger.

When restated in the abstract ("Would it have been obvious to the PHOSITA, in light of the prior art and general knowledge, that he or she could have come to the invention, had he or she sought to achieve the desired result?"), it is clear that the Cripps Question does not provide a great deal of guidance regarding what it means for something to be obvious. In an effort to clarify the standard, courts in the UK, the US and Canada have come up with various other formulations--an invention is obvious if it was "obvious to try"; "worth a try"; if it involved "routine testing"; or if it could be arrived at ".directly and without difficulty".

In considering the merits of these phrases, it is useful to heed the warning of the House of Lords:
   We agree, however, with what was said by Diplock, L.J. (as he then
   was) and Willmer, L.J. in the Johns-Manville case ... deprecating
   "coining" phrases which may later be suggested to be of general
   application. "Obvious" is, after all, a much-used word and it does
   not seem to us that there is any need to go beyond the primary
   dictionary meaning of "very plain". (42)


This warning was recently echoed by Hughes J. who noted that "[i]t is not useful to use such phrases as they tend to work their way into expressions of law or statements of expert witnesses". (43) For example, the phrase "obvious to try" does not provide much guidance unless it also refers to the PHOSITA's expectation of success. A solution could be "obvious to try" in light of a number of considerations, including: it might work--that is, it is "worth a try"; there is a reasonable likelihood of success; or there is a high degree of likelihood of success.

Prior to the decision in Sanofi, the leading test for obviousness was that articulated in 1986 by Hugessen J.A. in Beloit v. Valmet: (44)
   The test for obviousness is not to ask what competent inventors did
   or would have done to solve the problem. Inventors are by
   definition inventive. The classical touchstone for obviousness is
   the technician skilled in the art but having no scintilla of
   inventiveness or imagination; a paragon of deduction and dexterity,
   wholly devoid of intuition; a triumph of the left hemisphere over
   the right. The question to be asked is whether this mythical
   creature (the man in the Clapham omnibus of patent law) would, in
   the light of the state of the art and of common general knowledge
   as at the claimed date of invention, have come directly and without
   difficulty to the solution taught by the patent. It is a very
   difficult test to satisfy.


There are two distinct aspects to the Beloit test worth noting. First, the PHOSITA is attributed to have the unlikely combination of: (1) complete knowledge of the prior art; and (2) zero creative ability. Second, given this knowledge, the PHOSITA would have to come "directly and without difficulty" to the solution taught by the patent. This language has been often been interpreted stringently such that any experimentation or testing vitiates a finding of obviousness. (45) In other words, the Beloit test is incompatible with any "obvious to try" approach.

In addition, inventions need not be major breakthroughs, as the Federal Court of Appeal confirmed that a mere "scintilla of invention" is sufficient. (46) Small, incremental improvements can be the proper subject-matter for patents, provided that the improvements are not obvious.

Although Beloit continued to be cited as the leading precedent, there was a substantial body of Canadian case law on obviousness that developed in the more than two decades since Beloit was decided by the Federal Court of Appeal. There was considerable confusion in the pre-Sanofi case law regarding how the Beloit obviousness test ought to be applied, and indeed whether it still accurately reflected the Canadian obviousness standard. For example, in a recent pre-Sanofi survey, Ron Bouchard argues that that the case law dealing with pharmaceuticals does not evince an unequivocal and predictable obviousness test. (47) Instead, Bouchard argues that there are two main lines of cases--the first line applying the Beloit test stringently, and the second line applying a lower standard in which experimentation does not necessarily vitiate a finding of obviousness. Furthermore, Bouchard argues that there is even a third line of cases which purported to apply the Beloit test but actually permitted testing.

In Novopharm Ltd. v. Janssen-Ortho Inc., the Federal Court of Appeal recently approved a list of factors ("Janssen-Ortho factors") that can be used as a further guide to applying the Beloit test. (48) This list included "principal factors": (1) the invention; (2) the PHOSITA; (3) the PHOSITA's knowledge; (4) the climate in the relevant field at the time the alleged invention was made; (5) the motivation in existence at the time the alleged invention was made to solve a recognized problem; and (6) the time and effort involved in the invention. It also included "secondary factors" which may be helpful: (7) commercial success; and (8) meritorious awards. Some of these factors relate to elements that were already considered in the Beloit test: the invention, the PHOSITA, and the PHOSITA's knowledge. Other factors appeared to extend beyond the traditional Beloit considerations. The "climate" in the relevant field is important because the "state of the art includes not only knowledge and information but also attitudes, trends, prejudices and expectations". (49) If there was a specific problem that everyone in the field was trying to solve--namely, there is a general motivation or "long felt need"--then it may be more likely that the solution required inventive ingenuity. The length of time and expense involved in the invention may be indicators of inventive ingenuity, but they are not determinative because the exercise is an objective determination. Nevertheless, the Federal Court of Appeal was clear that this list of factors was "a useful tool" rather than a legal test. (50)

As will be discussed below in Part IV, some of the variation in the interpretation of Sanofi may be due to confusion about the state of the law prior to the decision. The existence of this ambiguity further underscores the importance of the Supreme Court of Canada's pronouncements on obviousness in Sanofi.

III SANOFI v. APOTEX

Facts and Lower Court Decisions

Sanofi were the holders of Canadian patent 1,194,875 ("the genus patent"). The application for the genus patent was filed in 1983, the patent was issued in 1985, and it expired in 2002. Sanofi was also the holder of a second patent--Canadian patent 1,336,777 (the "selection patent"). The application for the selection patent was filed in 1988, it was issued in 1995, and unless it was found to be invalid, it would expire in 2012. The selection patent claims a single compound, clopidogrel bisulfate, which is marketed by Sanofi under the trade name Plavix as an anti-coagulant that inhibits platelet aggregation activity in the blood.

The genus patent disclosed a class of compounds useful in inhibiting platelet aggregation activity in the blood which is important in treating coronary artery, peripheral vascular and cerebral vascular diseases. The genus included over 250,000 possible compounds that were claimed to be useful for this purpose. Twenty-one of these compounds were isolated, tested, and included in the specification of the genus patent; one of these compounds was the racemic mixture of clopidogrel. A racemic mixture is a substance containing equal amounts of two compounds with different 3-dimensional structures, called enantiomers. (51) The two enantiomers, the d-enantiomer and the 1-enantiomer, (52) are mirror images of each other. The common analogy used to illustrate the relationship between enantiomers is the relationship between a left and a right hand. The racemic mixture of clopidogrel is the easiest to synthesize; separating the enantiomers requires further effort. Clopidogrel is the d-enantiomer of the racemic mixture, and it has beneficial properties over both the racemic mixture and the l-enantiomer. Clopidogrel is both more efficacious (greater platelet aggregation inhibiting activity) and less toxic than the 1-enantiomer and the racemic mixture. (53) The selection patent also specified that it was the bisulfate salt of clopidogrel that displayed the beneficial properties.

In Sanofi, the generic manufacturer, Apotex, alleged that Sanofi's selection patent was invalid on account of anticipation, obviousness, and double patenting. (54) The prohibition against double patenting (sometimes referred to as "evergreening") exists because there may only be one patent covering an invention. (55) There are two branches to this prohibition. The first branch, "same invention" double patenting, arises if the claims of the second patent are "identical or coterminous" with the first patent. (56) The second branch, "obviousness" double patenting, is less literal and it arises if the claims of the second patent are not "patentably distinct" from those of the earlier patent. (57)

The Federal Court rejected all three invalidity arguments raised by Apotex and granted an order of prohibition that precluded Apotex from marketing generic clopidogrel. (58) The applications judge, Shore J., framed the inventive aspect of the selection patent as the discovery of the beneficial properties resulting from the combination of the d-enantiomer (clopidogrel) with the bisulfate salt.

First, Shore J. held that the anticipation argument failed because the genus patent only disclosed the existence of enantiomers generally; there was no specific disclosure of enantiomers of the racemic mixture of clopidogrel, nor was there disclosure of how to separate the enantiomers. Therefore, a PHOSITA, following the genus patent, would only ever arrive at the racemic mixture.

Second, Shore J. applied the Beloit test and held that the invention was not obvious. The evidence indicated that there were five well-known separation techniques that could be used to obtain the d-enantiomer. Nevertheless, Shore J. found that a PHOSITA had no way of knowing, based on the genus patent and common general knowledge, which of the five common techniques would have worked. Moreover, he found that a PHOSITA had no way of knowing what the properties of the d-enantiomer and its salts would be without testing the individual combinations; therefore, the beneficial properties of clopidogrel were not known before the racemic mixture was separated into its enantiomers. Given that each separation technique, and each salt combination, was merely "worth a try", a PHOSITA would not have been led "directly and without difficulty" to the solution taught. Therefore, the invention was not obvious.

Finally, Shore J. dismissed the double patenting argument as adding nothing to Apotex's anticipation and obviousness arguments. (59) Justice Shore held that "obviousness" double patenting merely means that the claims must be novel and non-obvious, and that it really only comes into play when a party seeks to rely on internal work that is not public in order to prevent the issuance of two patents for essentially the same invention. In this case, Apotex relied on the same prior art for double patenting as it relied upon for anticipation and obviousness.

The Federal Court of Appeal dismissed Apotex's appeal, upholding Shore J.'s findings with respect to anticipation, obviousness, and double patenting. (60) The Court also held that the process of separating the enantiomers, selecting the proper salt, and testing the resulting compounds was not routine testing. Rather, the difficulty and unpredictability involved in this process demonstrated that the invention was not obvious.

Although Shore J. did not frame any of the issues at first instance in terms of selection patents, the Federal Court of Appeal did explicitly address selection patents. The Federal Court of Appeal defined a selection patent as "one which claims an advantage for a compound within a previously disclosed class of compounds which has not been disclosed in the prior patent", and held that selection patents are valid. (61) The Court also held that a selection patent framework was a helpful background against which to assess Apotex's allegations of invalidity, and that Shore J. had implicitly engaged in just such an analysis. (62) Accordingly, on appeal to the Supreme Court of Canada, Apotex added the argument that the Federal Court of Appeal had, over the course of several recent cases, created a "doctrine of selection patents" that was inconsistent with the general tenets of patent law. (63)

The Supreme Court of Canada's Decision

The Supreme Court of Canada, in a unanimous judgment authored by Rothstein J., dismissed Apotex's appeal. The Court first considered whether selection patents were valid in principle. The Court cited Maugham J. in the English case of I.G. Farbenindustrie for the proposition that a selection patent "does not in its nature differ from any other patent". (64) The Court approved of Maugham J.'s three conditions that must be satisfied for a selection patent to be valid: (1) the selected compound(s) must have an advantage over the general class; (2) all of the selected compounds must possess this advantage; and (3) the selection must be in respect of a quality of a special character peculiar to the selected group--that is, if research showed that a larger number of unselected compounds possessed the same advantage, the quality of the compound claimed in the selection patent would not be of a special character. Given that selection patents do not differ from other patents, Maugham J.'s three-part test was only a "useful starting point", and the validity of Sanofi's selection patent had to be considered on the grounds of anticipation, obviousness and double patenting.

On the issue of anticipation, the Court overruled its previous pronouncements on anticipation, (65) instead adopting the test set out by the House of.Lords in Synthon v. SKB. (66) The Synthon test distinguishes between two requirements for anticipation that were not previously considered separately: (1) prior disclosure and (2) enablement. Essentially, the new test requires that a prior patent (or other document) both disclose the invention in the patent at issue and enable the PHOSITA to perform or work the invention without undue burden. Although this is a potentially interesting change in Canadian law, the obviousness standard is considered distinctly from the anticipation standard, and anticipation will not be further considered in this comment. (67) Applied to the facts, it was clear that Shore J. found that there was no prior disclosure of the specific beneficial properties of clopidogrel. (68) This was sufficient to dismiss anticipation as a ground of appeal.

With respect to obviousness, the Court rejected the rigid manner in which the Beloit formulation has been applied by Canadian courts:
   I do not think that Hugessen J.A. in Beloit intended that the
   rather colourful description of obviousness that he coined be
   applied in an acontextual manner applicable to all classes of
   claims. (69)


By contrast, Rothstein J. noted that both the UK and the US have adopted more flexible tests in which an "obvious to try" approach may be appropriate, and that this convergence "suggests that the restrictiveness with which the Beloit test has been interpreted in Canada should be re-examined". (70)

The Court then laid out a four-step approach to obviousness derived from the UK cases of Windsurfing International Inc. v. Tabur Marine (Great Britain) Ltd and Pozzoli SpA v. BDMO SA. (71) Rothstein J. restated the four Windsurfing questions as:

1. Identify the notional PHOSITA and determine their relevant general knowledge;

2. Identify the inventive concept of the claim in question;

3. Identify what differences exist between the "state of the art" and the inventive concept of the claim;

4. Viewed without any knowledge of the alleged invention as claimed, would those differences have been obvious to the PHOSITA or do they require any degree of inventiveness? (72)

Under this fourth step, the Court indicated that an "obvious to try" test may be appropriate in "areas of endeavour where advances are often won by experimentation". (73) The pharmaceutical industry was explicitly used as the example of this type of area. For cases where an "obvious to try" test is appropriate at the fourth step of the obviousness inquiry, the Court gave a non-exhaustive list of factors that should be taken into consideration:

a) Is it self-evident that what is be.ing tried ought to work, and is the number of predictable solutions finite?;

b) What is the extent, nature and amount of effort required to achieve the invention? Is the testing routine or is it prolonged and arduous?;

c) Is there a motive provided in the prior art to find the solution the patent addresses?;

d) (Perhaps) consider the evidence of the actual course of conduct which resulted in the making of the invention. (74)

Applied to the facts in Sanofi, the Court held that the appropriate PHOSITA was a trained pharmachemist. (75) The five methods of separating enantiomers were well-known, but the PHOSITA would have no knowledge of the relative advantage of the d-enantiomer (clopidogrel). At the second step, the inventive concept of the selection patent was the selection of a compound with greater therapeutic effect and less toxicity than the other compounds of the genus patent, and the disclosure of methods for obtaining that compound. (76) At the third step, there was nothing in the prior art that distinguished the racemic mixture of clopidogrel from the other 20 compounds made and tested in the genus patent, and there was nothing regarding the disclosure of advantages of the d-enantiomer. (77)

At the fourth step, the Court held that an "obvious to try" test was warranted in this case. (78) Justice Rothstein noted that this was where the analysis of the applications judge ended, but held that the factual record was sufficiently developed that some additional findings of fact could be made by the Supreme Court, and that this was preferable to remitting the matter. (79)

With respect to whether what is being tried ought to work, the Court held that the PHOSITA would have known how to separate enantiomers, and that there were a finite number of techniques. The PHOSITA would have known that enantiomers, in general, may have different therapeutic properties, but they would not have known that those properties would be different for clopidogrel and the l-enantiomer. Therefore, it was not self-evident that what was being tried ought to work. (80)

With respect to the extent, nature and amount of effort, the evidence was that the PHOSITA would eventually find the right technique, but there were potentially five different methods to try. In order to give some sense of how much effort would have been involved, the Court noted that it took a Sanofi chemist, who was already familiar with the genus patent, five months of laboratory work to find the solution claimed in the selection patent. (81) As will be discussed below in the section entitled "Obvious to Try Factors", the Court did not discuss whether it considered this to be a small or large amount of effort, although one presumes from the result of the appeal that this was implicitly considered to be a substantial effort.

With respect to motive, the Court held that selection from the genus patent might be expected, but that there was nothing in the prior art that suggested a specific motivation to try the solution laid out in the selection patent. (82)

With respect to the actual course of conduct, the Court quoted the affidavit of a Sanofi chemist for the proposition that Sanofi decided to investigate the enantiomers after it had "spent millions of dollars and several years developing [the racemic mixture of clopidogrel] up to the point of preliminary human clinic trials". Once Sanofi discovered the advantages of clopidogrel, it abandoned the work on the racemic mixture. (83)

After looking at the various factors, the Court held that the invention in the selection patent was not obvious to try and dismissed Apotex's appeal on the grounds of obviousness.

Finally, the Court considered Apotex's appeal on the grounds of double patenting--the rule that two patents cannot be issued for the same invention. With respect to selection patents, the Court held that there was no double patenting provided that (1) the claims of the selection patent were not identical to the claims of the genus patent; and (2) the selection patent claimed a compound that was "patentably distinct" from the compounds claimed in the genus patent. The Court clarified that the latter requirement meant that the selection patent must "define in clear terms the nature of the characteristic which the patentee alleges to be possessed by the selection for which he claims a monopoly". (84) The Court dismissed Apotex's appeal on these grounds because the claims of the selection patent and the genus patent were not identical, and the selection patent clearly indicated the characteristics possessed by clopidogrel.

IV SHIFT IN THE LAW OF OBVIOUSNESS

As a preliminary comment, it is interesting that in laying out new tests for anticipation and obviousness the Supreme Court went well beyond the scope of what was necessary to decide the case. The issues raised by the facts of Sanofi are similar to those raised in many pharmaceutical patent cases. The conclusions with respect to anticipation, obviousness and double patenting were the same at all three levels. There are at least three non-mutually exclusive explanations.

The first explanation involves conflicting application of the Beloit test in lower court decisions as discussed above in the section entitled "Canadian Obviousness Standards Prior to Sanofi". A definitive statement of the law of obviousness by the Supreme Court should remedy this level of confusion in the lower courts.

A second explanation involves the scope of the grounds of Apotex's appeal. Apotex presented the Federal Court of Appeal decision as a clash between the doctrine of selection patents and the conventional grounds of validity for a patent, and the Supreme Court may have felt it necessary to clarify that selection patents are valid in principle. (85) Moreover, Apotex implicated the public interest in patent litigation by arguing that the doctrine of selection patents was inconsistent with the patent bargain, and the Supreme Court may have been attracted by these policy issues. (86)

A third explanation involves the facts of Sanofi. It might strike a neutral observer as intuitively strange that Sanofi was able to get a genus patent covering over 250,000 compounds, then get a selection patent for a single compound based on the fact that only one enantiomer of the originally tested racemic mixture was actually therapeutically effective. If clopidogrel's "mirror image" does not work, then what does this say about the likely utility of the other quarter of a million compounds covered in the genus patent? In other words, although the selection patent demonstrated an improvement, it also demonstrated that Sanofi had not accurately characterized the therapeutic effectiveness of the range of chemicals in the genus patent. The concern is that the grant of selection patents can reward under-disclosure in the genus patents. (87) The Supreme Court clarified that selection patents are not the problem in this scenario (if indeed, any problem even exists). (88) The Supreme Court alluded to the fact that the genus patent might have been overbroad, but this was not a live issue as the genus patent expired in 2002. (89)

Regardless of the Supreme Court's reasons for making such a definitive statement in Sanofi, I will discuss four aspects, of the law of obviousness that were altered by the decision: (1) the emphasis on a functional approach; 12) the adoption of a formal framework as embodied in the Windsurfing questions; (3) the adoption of an "obvious to try" approach; and (4) a potential change in our conception of the PHOSITA.

Emphasis on Functional Approach

The emphasis on a functional and flexible approach signals a shift in the law of obviousness. As noted above, Rothstein J. explicitly rejects the rigid formulation of the Beloit test on the grounds that it is insensitive to the contextual nature of the obviousness inquiry. (90) Instead, the court prefers the flexible approach to obviousness that has been followed in the UK and the US. (91) Given this holding, it is difficult to accept Campbell J.'s view in Bridgeview Manufacturing that Sanofi was a mere "refinement" of Beloit. (92) The emphasis on a functional and flexible approach is a fundamental reorientation of how courts are to go about the obviousness inquiry. It is no longer sufficient merely to look to the wording of the test; rather, courts must remain cognizant of the purpose of the obviousness inquiry.

Formal Structure of Restated Windsurfing Questions

The basic shift in approach is complemented by the adoption of a formal framework (the restated "Windsurfing Questions") that permits the obviousness inquiry to be divided into discrete issues. This framework does not introduce any new elements into the obviousness inquiry, as the same issues were present in the Beloit test. The Beloit test, however, condensed these issues into a single question--would the PHOSITA have come directly and without difficulty to invention X in light of knowledge Y? By contrast, the restated Windsurfing Questions require precise determinations of (1) the PHOSITA and their relevant knowledge; (2) the inventive concept; and (3) how the inventive concept differs from the state of the art; before getting to the ultimate issue of (4) was it obvious?

Prior to Sanofi, it would have been possible for a court to have applied the Beloit test in such a step-wise analysis, but there was no reason for preferring a step-wise analysis over a condensed analysis. Although courts are not necessarily required by Sanofi to engage in this step-wise process--because of the flexible and functional nature of the test--the Supreme Court has indicated that this process should be the default framework for the obviousness analysis. This guidance should bring needed consistency to the application of the obviousness test. From a substantive perspective, commentators have suggested that this formal framework may require patent owners and defendants to each take a more precise position on the court record as to how broadly or narrowly to define the inventive concept. (93) The usefulness of the formal framework in facilitating a functional approach to obviousness is illustrated in the decision of the Federal Court in BMS v. Apotex, discussed below in the section entitled "BMS v. Apotex: Sanofi Interpreted (Part 2)".

Obvious to Try Approach

The incorporation of an "obvious to try" approach at the fourth Windsurfing question is the most striking change in Sanofi, given the Supreme Court's previous rejection of an "obvious to try" approach. (94) Nevertheless, the extent to which this approach is a substantive change in the law of obviousness will depend upon (1) the elements of an "obvious to try" approach; and (2) the extent to which these elements were already present in Canadian law prior to Sanofi. On the latter issue, as discussed above in the section entitled "Canadian Obviousness Standards Prior to Sanofi", the law of obviousness had developed subsequent to Beloit, and some of the variation in the interpretation of Sanofi may be due to confusion about the state of the law prior to the decision. In particular, in 2007, the Federal Court of Appeal had approved the Janssen-Ortho factors as useful indicia of obviousness. (95)

A curious aspect of the Supreme Court's decision in Sanofi is that it does not even refer to Janssen-Ortho. One would have thought that the Janssen-Ortho decision would have been a particularly important precedent for two reasons. First, as described above in the section entitled "Canadian Obviousness Standards Prior to Sanofi", in Janssen-Ortho the Federal Court of Appeal discussed general principles related to the obviousness analysis. The Federal Court of Appeal stated that the obviousness analysis must be functional, and it put forth a non- exhaustive list of factors to be flexibly applied depending on the facts of the particular case. (96) Second, the facts of Janssen-Ortho shared similarities with Sanofi. The patent at issue in Janssen-Ortho was a selection patent that claimed the antimicrobial drug levofloxacin. Levofloxacin was the enantiomer of a racemic mixture, ofloxacin, that was claimed in a previous patent. Accordingly, one of the main issues in the case was whether the invention of levofloxacin was obvious in light of the previous disclosure of ofloxacin and the prior art related to the separation of enantiomers. (97)

The most plausible explanation for the Supreme Court's failure to refer to Janssen-Ortho is the confusion in the state of the law of obviousness prior to Sanofi. In the Federal Court of Appeal decision in Sanofi, NoEl J.A. cited the trial level decision in Janssen-Ortho regarding the application of the obviousness test to the separation of enantiomers, but he placed little weight on the case as the appeal before the Federal Court of Appeal had not yet been heard. (98) The Federal Court of Appeal's decision in Janssen-Ortho was released in June 2007, ten months before the Sanofi appeal was heard by the Supreme Court. The Federal Court of Appeal's decision in Janssen-Ortho was cited in the factums of both Sanofi and Apotex, but only for relatively minor points. (99) Neither party referred to the Janssen-Ortho factors as forming part of the general standard for obviousness. Instead, both parties characterized Beloit as continuing to represent the obviousness standard in Canada. (100)

In the remainder of this section, I will demonstrate that a comparison of the obvious to try factors from Sanofi with the Janssen-Ortho factors reveals at least three differences. Then, I will consider the Federal Court of Appeal's interpretation, from the case of Pfizer v. Apotex, of the relationship between the law of obviousness as set out in Sanofi, Beloit, and Janssen-Ortho. I will observe that, in Pfizer, the Federal Court of Appeal held that it was possible, through a combination of Beloit plus the Janssen-Ortho factors, to engage in the kind of functional approach to obviousness espoused by the Supreme Court in Sanofi.

Obvious to Try Factors

The first factor in the obvious to try analysis, whether "it is more or less self-evident that what is being tried ought to work", suggests that the predictability of the outcome of the testing is critical. It should be noted that in using the phrase "finite number of identified predictable solutions", the Court probably intended to convey the meaning "relatively few in number" as opposed to the meaning "determinable in number". (101) The Court stated that the "[re]ere possibility that something might turn up is not enough", indicating that the PHOSITA must not only think it was obvious to try the solution, but they must also have a reasonably high expectation that the solution will be successful. (102) This approach to testing is not all that different from the second line of cases discussed in Bouchard's review, as discussed under the heading "Canadian Obviousness Standards Prior to Sanofi" that permitted some experimentation without necessarily vitiating a finding of obviousness. Although predictability was not set out as a separate factor in Janssen-Ortho, the predictability of the properties of the enantiomers of ofloxacin was relevant to the obviousness analysis. (103)

The second factor in the "obvious to try" analysis considers the extent, nature and amount of effort that were required to achieve the invention. It is interesting that the Court has juxtaposed the two extremes as "routine" against "prolonged and arduous". It appears that "routine" is being used to refer to the level of effort (relatively minimal effort) rather than the kind of effort (non-inventive or non-imaginative). This interpretation is supported by the Court's application of this factor to the facts of Sanofi. In Sanofi, the evidence was that there were five identified methods to try, and a skilled person would eventually find the right technique. This could be considered "routine" in the sense of not requiring any imaginative spark, but the Court infers that the testing required was prolonged and arduous because it took a Sanofi chemist five months of testing to achieve the invention. (104) On this point, it is surprising that the Court does not explicitly consider whether this would be considered a prolonged and arduous amount of work by the standards of the pharmaceutical industry. Nonetheless, this second factor is very similar to the Janssen-Ortho consideration of the length of time and expense involved in coming up with the invention.

The third factor in the "obvious to try" analysis is whether the prior art provides a motive for the solution--that is, whether the PHOSITA would be prompted by the prior art to look for a particular solution to a problem. It is clear from the manner in which this factor was applied to the facts of Sanofi that the Court is referring to whether there was a motive to pursue the specific solution taught by the patent,l[degrees]s The existence of a specific motive increases the likelihood that a solution was obvious. The Court is not referring to a general motive to look for a solution to an identified problem. This is the first difference from Janssen-Ortho, as it was this more general meaning of motive that was included in the list of factors. (106) Moreover, under Janssen-Ortho, the existence of a more general motive actually pushes in the opposite direction--an invention is less likely to be obvious if everyone was looking for a solution but no one else had found it. (107) These disparate meanings of the word "motive" have the potential to cause considerable confusion. (108)

The meaning of "motive" from Sanofi is similar to the meaning of the word in the Teaching-Suggestion-Motivation (TSM) test from US patent law. This test, which was developed by the US Court of Appeals for the Federal Circuit, limited a finding of obviousness to those situations where the prior art contained a specific teaching, suggestion or motivation that would have led a PHOSITA to combine elements of pre-existing inventions. In KSR v. Teleflex, the US Supreme Court rejected a rigid application of the TSM test as inconsistent with the US law of obviousness:
   [I]t can be important to identify a reason that would have prompted
   a person of ordinary skill in the relevant field to combine the
   elements in the way the claimed new invention does. This is so
   because inventions in most, if not all, instances rely upon
   building blocks long since uncovered, and claimed discoveries
   almost of necessity will be combinations of what, in some sense, is
   already known.
   Helpful insights, however, need not become rigid and mandatory
   formulas; and when it is so applied, the TSM test is incompatible
   with our precedents. (109)


The US experience with the TSM test should provide a cautionary lesson for Canadian courts--although the existence or absence of a specific motive is relevant evidence, this factor should not be a precondition for a finding of obviousness.

The fourth factor which can be considered in the "obvious to try" analysis is the actual course of conduct leading to the invention. Although this factor appears as a supplementary, rather than an enumerated, consideration in the analysis, the Court in Sanofi seemed to place considerable reliance on the actual history of the invention. (110) The obviousness inquiry is an objective exercise--that is, would the invention have been obvious to the notional PHOSITA? Explicit consideration of the actual course of conduct appears to import a subjective element--in other words, was the invention obvious to the patentee before they began their research? On the other hand, one might argue that the actual course of conduct was already being implicitly considered during other aspects of the test. As noted above, the actual amount of time it took Sanofi to separate the enantiomers was also considered under the second factor of the "obvious to try" analysis. Aside from the subjective element and the potential for confusion with the second factor, consideration of the actual course of conduct can have the undesirable effect of penalizing inventors who are efficient, insightful, or lucky. (111) Consideration of the actual course of events as an explicit factor, rather than simply as evidence relating to the kind of work required to come up with the invention, is a second way in which the Sanofi framework differs from the Janssen-Ortho factors.

Finally, the third difference is that there is no mention in Sanofi of the "secondary indicia" from Janssen-Ortho. Commercial success or meritorious awards could not be considered under the heading "actual course of conduct leading to invention" because these indicia occur after the invention has been made. It is possible that such secondary indicia could be considered in the obvious to try analysis because the list of factors in Sanofi was "non-exhaustive"; it remains to be seen, however, whether lower courts will be sufficiently bold to consider additional factors.

Pfizer v. Apotex: Sanofi Interpreted (Part 1)

In Pfizer, the Federal Court of Appeal held that it was possible, through a combination of Beloit and the Janssen-Ortho factors, to engage in the kind of functional approach to obviousness espoused by the Supreme Court in Sanofi. Approximately two months after the Supreme Court of Canada's decision in Sanofi, the Federal Court of Appeal released its first decision on the new law of obviousness in Pfizer Canada Inc. v. Apotex Inc. (112) In Pfizer, Apotex challenged the validity of Pfizer's patent claiming the drug sildenafil, which is marketed by Pfizer in Canada under the brand name Viagra, for the treatment of erectile dysfunction. (113) Apotex amended its appeal to argue that Sanofi had modified the Canadian obviousness standard, and that the Federal Court decision needed to be overturned because the judge had applied the Beloit test. Apotex's first ground of appeal was that Sanofi had incorporated a "worth a try" approach to obviousness. Apotex's second ground of appeal was that Sanofi brought the law in Canada in line with that applicable in the UK, making UK precedents highly relevant. In particular, Apotex pointed to previous UK decisions dealing with the equivalent UK patent as providing a "blueprint" for the application of the obviousness standard. The Pfizer decision is useful for interpreting Sanofi because it: (1) provides insight into how the Sanofi "obvious to try" approach will be interpreted by lower courts; (2) is consistent with the idea of Sanofi being a subtle shift rather than a radical reorientation in the obviousness standard; and (3) indicates that the phrase "obvious to try" stands for a general approach rather than a particular legal test.

Justice Noel rejected Apotex's contention that Sanofi had incorporated a "worth a try" approach to obviousness, pointing out that Rothstein J. had used this phrase only once in the judgment, and that the test adopted was correctly identified as "obvious to try". (114) Moreover, Noel J.A. held that the essential component of the Sanofi test was whether it was "more or less self-evident that it would work",ns It is potentially troubling that Noel J.A. appears to condense the flexible and functional test laid out in Sanofi into a single phrase; using "more or less self-evident that it would work" as a shorthand for the obviousness standard is no more useful than previous expressions such as "directly and without difficulty" or "obvious to try". On the other hand, the Federal Court of Appeal was only asked whether the specific decision on appeal, which was released prior to Sanofi, was consistent with the new obviousness standard. In the subsequent BMS case, Hughes J. also held that "more or less self-evident that it would work" was the essence of the obvious to try approach, but he only applied this standard after thoroughly answering the Windsurfing questions. Pfizer and BMS suggest that lower courts will apply the full framework as laid out in Sanofi, although the phrase "more or less self-evident that it would work" may become the dominant feature of the "obvious to try" analysis in the fourth Windsurfing question.

Most importantly for the present discussion, Noel J.A. held that the approach "in the lower court decision, to apply the Beloit test in conjunction with the Janssen-Ortho factors, was consistent with the new Sanofi test. While the Federal Court did not use the phrase "obvious to try", by finding that the prior art could only have led a PHOSITA to speculate about the solution taught by the patent, the Federal Court "drew the line precisely where the Supreme Court drew it in [Sanofi] when it held that 'the mere possibility that something might turn up is not enough"'. (116) This holding is consistent with Sanofi being a subtle shift in the law of obviousness rather than a radical reorientation. The point is not that a combination of Beloit plus the Janssen-Ortho factors is equivalent to the Sanofi standard. Instead, Pfizer stands for the proposition that it was possible, prior to Sanofi, to engage in the kind of functional approach to obviousness espoused by the Supreme Court. (117) Given this holding, it seems likely that the Janssen-Ortho factors continue to form part of the Canadian law of obviousness, provided that they are filtered through the Sanofi framework. (118)

The Federal Court of Appeal's findings regarding UK law are consistent with the idea that it is the substance of the obviousness inquiry rather than the label "obvious to try" that matters. In Sanofi, Rothstein J. explicitly cites the convergence of the UK and the US law of obviousness as a reason for modifying the Canadian law. (119) This receptivity suggests that UK and US patent judgments may be cited more frequently by parties and courts involved in Canadian patent litigation. (120) Yet, Noel J.A. dismissed Apotex's second ground of appeal--that previous UK decisions dealing with the equivalent UK patent provided a "blueprint" for the application of the obviousness standard--on the basis that the test applied in the UK litigation was different in substance from the Sanofi test. (121) The UK cases dealing with the equivalent patent were decided in 2001 and 2002, and they applied an earlier UK standard of obviousness. Specifically, they pre-dated Jacob L.J.'s 2005 statement in Saint-Gobain PAM SA v. Fusion Provida Ltd., quoted by Rothstein J. in Sanofi, that used the phrase "more or less self-evident that what is being tested ought to work". Therefore, although these cases applied an "obvious to try" approach, this example indicates that this phrase alone reveals very little about the legal test. (122) Canadian courts are not bound by the outcome of patent litigation in other countries, but foreign decisions will be relevant for their persuasive value, and this value will depend on the degree of substantive similarity between the law of obviousness in Canada and in the foreign jurisdiction. (123)

Does Sanofi Change Our Conception of the PHOSITA?

Although not explicitly discussed by the Court, Sanofi appears to change how we ought to think about the PHOSITA. The PHOSITA is intended to be the notional expert through whom the Court can consider the question of obviousness. The PHOSITA thus serves two functions related to obviousness. The first is a "timing" function--the PHOSITA is positioned in the past in order to guard against hindsight bias. The second is a "skills" function--obviousness is considered by a PHOSITA who has specialized skills, including knowledge of the prior art.

Rebecca Eisenberg argues that US courts have often erred by conflating the timing and the skills functions, with the result that the PHOSITA becomes merely a guard against hindsight bias. (124) This "marginalization" of the PHOSITA occurs when courts focus on the question of defining the state of the art (obvious compared to what?) and ignore the question of whose assessment of obviousness matters (obvious to whom?). It is worth noting that Eisenberg's criticism was directed in particular against the decisions of the US Court of Appeals of the Federal Circuit, and that it was written prior to the KSR decision of the US Supreme Court. There are some differences between the US and Canada regarding the construction of the PHOSITA's knowledge that might call into question the relevance Of Eisenberg's criticism in Canada. For example, in the US, there is a tendency to construct the PHOSITA's knowledge based primarily on the written record of prior art. (125) By contrast, in Canada, the evidence of expert witnesses may play a greater role in constructing the PHOSITA's knowledge. (126) Despite these differences, the Beloit characterization--"the man in the Clapham omnibus of patent law" who is a skilled technician but has "no scintilla of inventiveness"--produces the same result that was criticized by Eisenberg: the role of the PHOSITA is marginalized to a mere guard against hindsight bias. (127)

The strict interpretation of the PHOSITA from Beloit had already received both academic and judicial criticism prior to Sanofi. In the 1995 case of Bayer AG v. Apotex, Lederman J. of the Ontario Court of Justice felt bound to apply the Beloit test, but criticized the extent to which it marginalized the PHOSITA:
   [A]lthough one would normally imagine that this mythical person's
   laboratory is filled with mythical test tubes and Petri dishes and
   that his or her daily life is spent in experimentation, for the
   purposes of this legal exercise, no research of any kind can,be
   contemplated. So, although it may have been logical to an actual
   skilled person at the time, based on the state of the art, to
   conduct certain testing, that is not open to the mythical skilled
   technician. The mythical researcher cannot have an inquiring or
   thinking mind which ultimately would lead him or her to the answer
   but rather he or she is expected to instantly and spontaneously
   exclaim, without more, "I already know the answer and it is
   obvious." Nor is it appropriate to say that there were significant
   telltales which pointed the way for the mythical expert.... (128)


Remarking on the seemingly paradoxical position that the PHOSITA was supposed to be a well versed, yet totally unimaginative, scientist, Lederman J. quoted a Canadian patent law text for the idea that "[t]he suggestion that the skilled technician should be unimaginative should not strip that technician of the ability to pursue reasonable and logical inquiries". (129) Finally, Lederman J. noted that the Beloit characterization strictly limited the abilities of a Canadian PHOSITA as compared to the PHOSITA under UK law. (130)

The strict interpretation of the PHOSITA from Beloit had already been criticized by the Supreme Court in the context of claims construction. In Whirlpool Corp. v. Camco Inc., Binnie J. acknowledged that the PHOSITA is reasonably diligent in keeping up with advances in the field, and that the PHOSITA's common knowledge undergoes continuous evolution and growth. (131) Although this was stated in the context of claims construction, the Federal Court of Appeal in Janssen-Ortho cited the holding in Whirlpool as relevant to the characterization of the PHOSITA for the purpose of obviousness, noting that knowledge is not limited to printed form. (132) Similarly, the Janssen-Ortho factor of "climate in the relevant field" might also be interpreted as relating to the capacity of the PHOSITA, m that [t]he general state of the art includes not only knowledge and information but also attitudes, trends, prejudices and expectations." (133) Nevertheless, it seems likely that the extent to which Janssen-Ortho altered the Beloit characterization of the PHOSITA was minimal given that the Federal Court of Appeal viewed the Janssen-Ortho factors as merely a guide to implementing the Beloit test.

Justice Rothstein's comment that too much emphasis has been placed on the colourful description of obviousness in Beloit (134) can be interpreted as a critique of the Beloit characterization of the PHOSITA. If the obviousness test is to be applied in a functional manner, then it is essential to think about the PHOSITA in relation to the function that this concept serves. The PHOSITA, as the lens through which the court addresses obviousness, has both a timing and a skills function. Assuming that the PHOSITA has "no scintilla of inventiveness" does not serve the purpose of the obviousness inquiry because this characteristic is inconsistent with the skills function of the PHOSITA. This view reduces the PHOSITA from an expert to an automaton. For example, in the context of pharmaceuticals, Ron Bouchard argues that the Beloit vision of the PHOSITA bears no resemblance to the contemporary realities of drug development:
   [T]o say that the PHOSITA can possess no scintilla of inventiveness
   is to say that a scientist cannot be a scientist. To impose a
   standard of no scintilla on a pharmaceutical scientist working in
   one of the largest, most sophisticated, profitable and competitive
   industries in the world is either completely irrational or simply
   far too low a rational standard when one constructs the nature of
   the PHOSITA in the relevant context. (135)


The obviousness inquiry is better served by viewing the PHOSITA as someone possessing ordinary creativity, relative to the level of skill that the PHOSITA is construed to have, rather than no creativity. If creativity is one of the principal attributes of the technicians or researchers in a particular area of endeavour, then a PHOSITA in those circumstances should be deemed to be relatively creative. Conversely, if the area of endeavour is not characterized by highly creative practitioners, then the PHOSITA should be given a lower level of creativity. The Supreme Court did not explicitly endorse this view of the PHOSITA in Sanofi, but I argue that this view is a natural extension of a functional approach to obviousness because the PHOSITA requires an ordinary level of creativity in order to perform its skills function.

V SANOFI TEST, POLICY LEVERS, AND THE PATENT BARGAIN

The shift in the legal test for obviousness introduced by Sanofi is a welcome development because it is commensurate with a purposive approach to patent law. As discussed above in the section on "Law of Obviousness--Relationship to the Patent Bargain", the purpose of the obviousness test is to filter out those inventions that do not appreciably add to the store of public knowledge from those that are truly inventive. According to the patent bargain, only the latter are deserving of patent protection. The emphasis on a functional approach should assist courts in undertaking a purposive analysis of obviousness, as opposed to placing undue emphasis on a particular phrase or legal formulation.

The framework provided by the Windsurfing questions is useful because it requires that courts devote attention to the elements of the obviousness inquiry prior to answering the ultimate question. The first question involves defining the lens through which obviousness should be approached--that is, the PHOSITA-and the prior art. The second question involves defining the inventive concept as defined by the relevant claim(s), an inquiry that is chiefly internal to the patent itself. The third question involves explicitly comparing the patent to the preexisting store of knowledge. This is the core of the obviousness inquiry--the issue is the relative advancement of knowledge rather than whether the invention was complicated, impressive, or the like. Once these constitutive elements have been determined, then the ultimate issue of obviousness can be addressed at the fourth question.

Compared to the Beloit test, the new obviousness test is better aligned with the concept of the patent bargain because it directs courts to filter out obvious inventions in a manner that is sensitive to differences among fields of research. The Patent Act is facially neutral with respect to different technologies. Nonetheless, legal and economic research has demonstrated that different industries vary greatly in how they approach innovation, the cost of innovation, and the importance of innovation to continued growth. (136) There are numerous reasons why enshrining technology-specific measures in a patent statute is a bad idea: the fact that the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPs) prohibits member states from discriminating in the grant of patents based on the type of technology at issue; (137) the difficulty of drafting industry-specific statutes so that they accommodate technological change; the difficulty of defining boundaries between industries; and public choice theory arguments that each new amendment represents an opportunity for counterproductive special interest lobbying. (138) In any case, the legislature has not chosen technology-specific standards. Yet, by adopting an obviousness requirement in the statute without defining the standard, it is arguable that Parliament has left the job of tailoring the obviousness standard to the courts. This is reasonable given that obviousness is a heavily fact-dependent and contextual issue.

Burk and Lemley have argued, in the US context, that patent law gives courts the discretion to take account of differences among technologies through the use of "policy levers". (139) They distinguish between policy levers that operate on a "macro" level, which expressly treat different industries differently, and those that operate on a "micro" level, which treat different inventions differently, without express regard to industry, but in ways that have disproportionate impacts on different industries. The Court in Sanofi incorporated several such policy levers in the new obviousness framework.

In reference to the fourth Windsurfing question, the Court explicitly held that whether or not an "obvious to try" approach is appropriate depends upon the methods through which advances are typically achieved in a given area of endeavour. If experimentation is commonplace in a given field, such as pharmaceuticals, then an invention may be obvious despite the fact that some experimentation was required in order to find the solution. This would appear to be a "macro" policy lever. Commentators have questioned, however, to what extent the "obvious to try" analysis will be discretionary, given that experimentation is likely used in all fields of endeavour. (140) In the Bridgeview Manufacturing case, released two months after Sanofi, Campbell J. refused to apply the "obvious to try" approach to a patent claiming a bale processor, a piece of agricultural machinery:
   [T]he concept of "obvious to try" is not in play in the present
   case because it lends itself to a situation where advances are won
   by experimentation, and the novel concept of the invention claimed
   in the Patent was not developed in this way. (141)


I would argue that Campbell J. asked the wrong question, as the question set out in Sanofi is not whether experimentation was used to develop the particular invention at issue, but rather whether experimentation is used in this field of endeavour. Therefore, the appropriate question would have been whether experimentation was typically used in mechanical or agricultural engineering, not whether experimentation was actually used to produce this particular bale processor. This raises the interesting question of what Rothstein J. meant by "experimentation", which unfortunately is beyond the scope of this paper. (142)

If an "obvious to try" approach is warranted, then the specific factors can be used to calibrate how much experimentation is permissible before an invention will be non-obvious. This is necessarily an artificial line-drawing exercise, but then so is the entire obviousness inquiry. The first factor, whether "it is more or less self-evident that what is being tried ought to work", could be interpreted as either a "micro" lever--if courts look at the predictability of the specific kinds of experimentation--or a "macro" lever--if courts consider the predictability of experimentation to vary according to the field of research. The second factor, the extent, nature and amount of effort that were required to achieve the invention, could be used as a policy lever to take account of differences in research and development costs among industries. Nevertheless, this factor was not applied in this manner in Sanofi, as the Court made no attempt to determine whether the amount of effort required was substantial relative to everyday practice in pharmaceutical research and development. In addition, this factor could potentially be used as a policy lever to affect the extent to which we want to reward labour in the research and development process per se, as opposed to labour as evidence of inventiveness. (143)

It has frequently been recognized that the PHOSITA is a powerful policy lever for taking account of differences in innovation among industries and over time. (144) Under the Sanofi test, the PHOSITA can consider experimentation and the likely results of any such experimentation under the "obvious to try" approach, provided that the area of endeavour is one in Which a PHOSITA would actually engage in such testing. The PHOSITA's skills are not limited to factual knowledge but also extend to understanding of processes and problem-solving approaches. How strong of a policy lever the PHOSITA is depends on what skills they are given. Ron Bouchard has argued that because of the nature of modern drug development, a PHOSITA in the pharmaceutical context should be attributed to have a range of skills including science, law, economics and policy. (145) It remains to be seen how courts will approach the PHOSITA after Sanofi, but it seems likely that parties will devote more attention to the construction of the PHOSITA at the first Windsurfing question.

Although the Sanofi test is commensurate with the underlying considerations in the patent bargain, it is worth clarifying that the approach advocated by the Supreme Court is an evolution of existing patent doctrine. This evolution can be contrasted with more radical proposals that seek to base legal tests for patentability on direct economic estimation of the incentives involved in the patent regime. For example, Tun-Jen Chiang has proposed a cost-benefit approach to obviousness that dispenses with traditional patent doctrine, instead focusing on direct estimation of the societal costs and benefits associated with granting a patent for a particular invention. (146) Consideration of Chiang's specific proposal is beyond the scope of this note. (147)

VI SELECTION PATENTS AND OBVIOUSNESS

Unresolved Issues

In Sanofi, the Supreme Court confirmed that selection patents are valid in principle, but otherwise the decision does very little to help us understand how selection patents should be analyzed. I will note two specific issues related to the relationship between obviousness and selection patents which the Supreme Court failed to address: (1) the interplay between the three-part test for selection patents and the conventional requirements for patentability; and (2) what kind of improvement a selection patent needs to demonstrate.

The ruling that selection patents are valid in principle was hardly surprising given that improvements of existing solutions have always been viewed as inventions worthy of a patent. (148) In this respect, the Supreme Court noted that selection patents are not different from any other patent. (149) Nonetheless, the Supreme Court then adopted the I.G. Farbenindustrie three-part test for validity of a selection patent: (1) the selected compound(s) must have an advantage over the general class; (2) all of the selected compounds must possess this advantage; and (3) the selection must be in respect of a quality of a special character peculiar to the selected group. The Supreme Court stated that this test was a "useful starting point", but otherwise failed to clarify how this test relates to the conventional requirements for a patent--utility, novelty, and non-obviousness. Is it a separate, threshold analysis that selection patents must pass prior to meeting the conventional considerations? This would seem to run counter to the idea that selection patents are no different from other patents, and that selection patents are not referred to in the Patent Act. Alternatively, the three-part test could be a means of calibrating the conventional requirements when a selection patent exists on the facts.

Second, the Supreme Court did not clarify what kind of improvement a selection patent needs to demonstrate. In order to be non-obvious, does the selection need to exhibit a qualitative improvement over the class disclosed in the genus patent, or is a quantitative improvement sufficient? Apotex argued that selection patents should only be permissible if they disclose a qualitative improvement--that is, a new use for a subset of compounds. The Sanofi decision might be seen as implicitly rejecting this argument, and holding that a quantitative improvement is sufficient. Clopidogrel was claimed for the same use--inhibiting platelet aggregation activity in the blood--as the compounds in the genus patent. The Supreme Court, however, did not give any indication regarding the level of quantitative improvement necessary in order to justify a new patent. On a strict interpretation, the "mere scintilla of invention" concept from Diversified Products could be interpreted to mean that any quantitative improvement is sufficient, for instance, a one percent improvement in efficacy of a drug would be sufficient. (150) Moreover, in economic terms, the qualitative vs. quantitative dichotomy may not accurately reflect the benefits to society of a given invention. For example, a quantitative improvement in the treatment of a highly prevalent, serious disease might yield greater benefits to society than the addition of a new treatment for a rare, mild disease. Therefore, it could be argued that the Supreme Court's silence on this issue was a positive feature rather than a failure, given the contextual nature of the obviousness inquiry. Nevertheless, given the prevalence of this fact scenario in pharmaceutical patent litigation, the Court might have offered some basic guidance on this issue while making clear to limit its remarks to the pharmaceutical context.

BMS v. Apotex: Sanofi Interpreted (Part 2)

Three months after the Supreme Court's decision in Sanofi, Hughes J. of the Federal Court undertook the first detailed application of the new obviousness test at the trial level in BMS. (151) Although Pfizer was decided prior to BMS, in Pfizer the Federal Court of Appeal was considering a lower court decision that pre-dated Sanofi. As discussed above, the court in Pfizer was only required to decide whether the obviousness analysis in the lower court decision was consistent with Sanofi, as opposed to actually applying the new framework set out in Sanofi. I have separated the discussion of Pfizer and BMS in this comment because I am using the two cases for different purposes. I am using Pfizer primarily to illustrate how the obviousness test set out in Sanofi relates to the law of obviousness prior to Sanofi. By contrast, I am using BMS primarily to show how the law of obviousness can be applied to selection patents.

In BMS, Apotex challenged the validity of BMS's patent claiming the monohydrate form of cefepime dihydrochloride, which is an antibiotic marketed by BMS in Canada as Maxipime. (152) In BMS, Hughes J. found BMS's patent to be invalid on the grounds of obviousness. (153) The argument that Sanofi has completely displaced the Beloit test is supported by Hughes J.'s application of the Sanofi framework without even referring to Beloit. The BMS decision is also useful because it provides a coherent interpretation of Sanofi that resolves the two issues discussed in the previous section. The facts in BMS fit a selection patent scenario because the prior art contained a Greek patent application (that had been filed by a predecessor of BMS) that was nearly identical to the patent at issue except that it did not disclose the improved stability of the monohydrate form relative to the anhydrate form of cefipime dihydrochloride.

On the issue of the interaction of the selection patent test with the conventional requirements for patentability, Hughes J. notes that the Court in Sanofi found selection patents to be no different than other patents, and he criticizes the usefulness of the label "selection patents":
   There is a danger in giving names to certain patents and then,
   having given such names, arguing whether the patent meets criteria
   for such names as may have been discussed in one or another piece
   of jurisprudence. (154)


Instead, he interprets the three-part test as a means of calibrating the conventional requirements:
   [The three-part test] was a useful starting point for Rothstein
   J.... What is really being said is, if a patent claims to have
   "selected" a particular member from a previously disclosed larger
   group, then to be an unobvious selection, the selected member or
   group should have previously undisclosed (i.e. not anticipated) and
   unexpected (i.e. not obvious) advantages over what was previously
   disclosed. (155)


Consistent with this point, Hughes J.'s answer to the second unresolved issue is that there is no particular threshold level of improvement that a selection patent needs to demonstrate. Instead, the degree of improvement that is necessary to justify a patent will be contextual, and it must be addressed through the framework of the obviousness inquiry.

When considering obviousness, Hughes J. methodically addressed the Windsurfing Questions, and dealt with the degree of improvement at the third and fourth questions. The prior art did not distinguish between the various hydrated forms of the selected salts, instead teaching that all the selected salts have excellent stability. Hughes J. used the meagre scientific data that was disclosed in BMS's patent to evaluate the difference presented by the inventive concept, and subsequently offered the following analogy on why the improvement was obvious:
   Is it unexpected that given a class of materials all having
   excellent stability, that one member of the class, when tested
   once, would exhibit superiority over "another? All members of a
   professional basketball team are accomplished players, it is not
   unexpected that on a particular day one player will perform better
   than one of the other players. That does not mean that such a
   player is surprisingly or unexpectedly better than all or most of
   the rest of the team. (156)


By way of distinguishing BMS from Sanofi, Hughes J. compared the improvement in BMS's patent to the improvement disclosed in Sanofi's patent claiming clopidogrel, and held that there was nothing in BMS's patent that was surprising or unexpected. (157) It remains to be seen whether Hughes J.'s interpretation of obviousness and selection patents will be adopted by other courts. (158)

VII CONCLUSION

This comment argues that Sanofi is a subtle, but important, shift in the law of obviousness. The test set out in Sanofi goes well beyond a mere "refinement" of Beloit. (159) Instead, the Supreme Court made at least four substantive changes in the law of obviousness: (1) emphasizing a functional approach; (2) adopting a formal framework as embodied in the Windsurfing questions; (3) adopting an "obvious to try" approach that brings Canadian law closer to the UK and US law of obviousness; and (4) changing the conception of the PHOSITA, because a natural extension of a functional approach to obviousness is the construction of a PHOSITA with an ordinary level of creativity, as opposed to one with no creativity. Even if one assumes (contrary to my opinion) that many of these substantive elements existed in the Canadian law of obviousness prior to Sanofi, the "codification" of these elements serves the important purpose of resolving ambiguity in the law.

This shift is a welcome development because the flexible and functional nature of the new test will assist courts in filtering out those inventions that do not appreciably add to the store of public knowledge from those that are truly inventive. The degree to which the new test will take account of differences among industries in research and development practices will depend on how courts employ the various policy levers that are built into the Sanofi framework. In particular, courts' construction of the role of the PHOSITA will be critical. Finally, Sanofi itself tells us relatively little about the interaction between selection patents and obviousness, but the Sanofi framework, as applied in BMS, seems to provide courts with the methodology to address the issues that selection patents raise. As a cautionary note, courts will need to apply the multiple steps of the Sanofi framework without losing sight of the Supreme Court's admonition that the test is purposive, and the whole point of the exercise is to get at the core question: is it obvious?

* The author gratefully acknowledges the assistance of the Senior Board Notes, Comments and Reviews Editors, in particular Lead Researcher Rebecca Rodal, and the Senior Editors of the University of Toronto Faculty of Law Review. The author also thanks Alexander Stack and Ariel Katz.

(1) 2008 scc 61, 69 C.P.R. (4th) 251 [Sanofi].

(2) It is not necessary to consider the two requirements together because the legal tests are independent. This comment focuses on obviousness, which is, in my view, the more interesting and challenging legal issue, despite the recent attention that anticipation has received in Canadian legal literature--see infra note 67 for references.

(3) The PHOSITA character is alternatively referred to as the "person of skill in the art" (PSITA), "skilled reader", or the "technician skilled in the art". I use the term PHOSITA for the sake of convenience, but I regard all of these terms as interchangeable.

(4) (1986), 8 C.P.R. (3d) 289 (F.C.A.) at 293 [Beloit].

(5) Apotex's argument in Pfizer Canada Inc. v. Apotex Inc., 2009 FCA 8, 72 C.P.R. (4th) 141 [Pfizer].

(6) Bridgeview Manufacturing Inc. v. 921409 Alberta Ltd., 2009 FC 50 [Bridgeview].

(7) See e.g. Michael E. Charles, "Supreme Court of Canada aligns Canadian patent law with the UK and US" (7 November 2008), online: <http://www.bereskinparr.com/ENG/News/ NewsletterArticles/update-Now7-2008.html>. The article simultaneously states that the "obvious to try" approach in the US & UK has invalidated many patents but that "[i]n substance there is not much change [after Sanofi]".

(8) Steve Morgan, "Peeling the Onion: What Drives Pharmaceutical Expenditures in Canada" (Paper prepared for the Institute for Research on Public Policy (IRPP) conference Toward a National Strategy on Drug Insurance: Challenges and Priorities, 23 September 2002), online: <http://www.chspr.ubc.ca/node/204> (at $15 billion per year, pharmaceuticals are the second largest cost component of the Canadian health care system); Steve Morgan, Jae Kennedy, Katherine Boothe et al., "Toward an Understanding of High Performance Pharmaceutical Policy Systems: A 'Triple-A' Framework and Example Analysis" (2009) 2 The Open Health Service and Policy Journal 1 at 7 (from 1995-2005, per capita spending on pharmaceuticals in Canada grew by 5.8%; over this same period, per capita health care expenditure in Canada grew by 3.2%; this is similar to the OECD average over this same period of 5.0% and 3.6% respectively); Organization for Economic Co-operation and Development, "OECD Health Data 2008--Frequently Requested Data", online: <http://www.oecd.org/document/16/0,3343, en_2649_34631_2085200_1_1_l_l,00.html> (in 1995, pharmaceuticals comprised 13.8% of total health care spending in Canada; by 2005, this share had increased to 17.2%; in 2007, it was 17.5%).

(9) As discussed below, the four-step approach was first outlined in Windsurfing International Inc. v. Tabur Marine (Great Britain) Ltd. (1984), [1985] R.P.C. 59 (Eng. C.A.) [Windsurfing], and subsequently modified in Pozzoli SpA v. BDMO SA, [2007] F.S.R. 37, [20071 EWCA Civ 588 (Eng. C.A.) [Pozzoli]. Justice Rothstein later restated the four questions in Sanofi, but I will refer to the approach as the Windsurfing questions for the sake of convenience and in accordance with the nomenclature in the literature.

(10) Farbwerke Hoechst A.G. v. Halocarbon (Ontario) Ltd., [1979] 2 S.C.R. 929 at 943-44 [Hoechst].

(11) Pfzer, supra note 5.

(12) Sanofi, supra note 1.

(13) Ibid. at para. 99; extra-curial remarks of Jacob L.J. in "Patents and Pharmaceuticals--a Paper given on 29th November at the Presentation of the Directorate-General of Competition's Preliminary Report of the Pharma-sector inquiry", online: <http://ec.europa.eu/competition/sectors/ pharmaceuticals/inquiry/jacob.pdf>.

(14) Alfred A. Macchione et al., "Patentees score a victory in the Supreme Court of Canada: validity of selection patents upheld" (7 November 2008), online: <http://www.mccarthytetrault.ca/article_detail.aspx?id=4221>. The article states that "[a]ccording to the evidence in the [Sanofi] case, about eight out of every 10 pharmaceuticals currently on the market involve selection patents".

(15) Bristol-Myers Squibb Canada Co. v. Apotex Inc., 2009 FC 137, 74 C.P.R. (4th) 85 [BMS].

(16) Dan Burk and Mark Lemley, "Policy levers in patent law" (2003) 89 Va. L. Rev. 1575 at 1580 [Burk and Lemley]; Bronwyn H. Hall, "Patents and Patent Policy" (2007) 23 Oxford Rev. of Econ. Pol'y 568 at 568-71 [Hall]. The literature also contains non-instrumentalist accounts, but a consideration of alternative justifications for patents is beyond the scope of this comment. See e.g. A. Samuel Oddi, "Un-Unified Economic Theories of Patents-The Not-Quite-Holy Grail" (1996) 71 Notre Dame L. Rev. 267 at 275-77; Seana Valentine Shiffrin, "Lockean Arguments for Private Intellectual Property" in Stephen R. Munzer, ed., New Essays in the Legal and Political Theory of Property (Cambridge: Cambridge University Press, 2001) 138 at 152; Adam Mossoff, "Rethinking the Development of Patents: An Intellectual History, 1550-1800" (2001) 52 Hastings L.J. 1255 at 1257 (arguing that "English lawyers and jurists drew upon natural-law conceptions of the social contract and the moral significance of labor, and, in this way, the natural law philosophers shaped much of the initial common-law definition of patent rights").

(17) Free World Trust v. Electro Sante Inc., [2000] 2 S.C.R. 1024, 2000 SCC 66 at para. 42 [Free World Trust].

(18) Patent Act, R.S.C. 1985, c. P-4, s. 44. Patents are granted for a 20 year period, but note that the effective monopoly period is less than 20 years because compensation for infringement is available only from when the specification is published, which may be as long as 18 months after filing.

(19) Sanofi, supra note 1 at para. 12; Synthon BV v. Smithkline Beecham pie (2005), [2006] 1 All E.R. 685, [2005] UKHL 59 at paras. 57-58 [Syntbon]; Farbwerke Hoechst A.G. vormals Meister Lucius & Bruning v. Canada (Commissioner of Patents) (1963), [1964] S.C.R. 49 at 57 (no common law right to a patent]; Com'po Co. v. Blue Crest Music Inc., [1980] 1 S.C.R. 357 at 372-73 (clarifying that copyright is entirely statutory).

(20) David Vaver, Intellectual Property Law: Copyright, Patents, Trade-marks (Toronto: Irwin Law, 1997) at 12-13.

(21) Patent Act, supra note 18; Free Worm Trust, supra note 17 at para. 13.

(22) A patent application is only supposed to disclose a single invention, but the claims are drafted so as to cover all aspects of the invention and to describe the invention at varying levels of generality.

(23) Patent Act, supra note 18, s. 2. This provision describes appropriate subject-matter for patents. It defines an invention as "any new and useful art, process, machine, manufacture or composition of matter, or any new and useful improvement [of one of these]". It is also the source of the requirements of utility and novelty. The source of the non-obviousness requirement is described below in Part II.

(24) Beloit, supra note 4 at para. 12.

(25) For example, a low non-obviousness requirement (which will make it easier to obtain patents) will correspond to a high obviousness standard (which will make it difficult to invalidate patents on the grounds of obviousness).

(26) Patent Act, supra note 18, s. 43(2).

(27) Ibid., s. 59.

(28) Mark A. Lemley, "Rational Ignorance at the Patent Office" (2001) 95 NW. U.L. Rev. 1495 [Lemley].

(29) By contrast, Europe and Japan have opposition procedures whereby third parties can challenge issued patents without resorting to litigation.

(30) Mark Lemley and Carl Shapiro, "Probabilistic Patents" (2005) 19 J. Econ. Persp. 75: "Between 55 and 67 percent of issued U.S. patents lapse for failure to pay maintenance fees before the end of their term, which indicates that these patents are of little value to their owners. The distribution of value of patents appears to be highly skewed, with the top 1 percent of patents more than a thousand times as valuable as the median patent."

(31) Jean O. Lanjouw and Mark Schankermann, "Characteristics of Patent Litigation: A Window on Competition" (2001) 32 RAND J. Econ. 129; Lemley, supra note 28 (in the US, only 1.5% of all patents are litigated, and only 0.1% are litigated to trial); Joshua Lerner, "Patenting in the Shadow of Competitors" (1995) 38 J.L. & Econ. 463 (litigation rates vary by industry and reach as high as 6% in biotechnology); John R. Allison and Mark A. Lemley, "Empirical Analysis of the Validity of Litigated Patents" (1998) 26 A.I.P.L.A.Q.J. 185 (of those patents litigated to a final determination, 46% have been held invalid); Kimberly A. Moore, "Judges, Juries and Patent Cases--An Empirical Peek Inside the Black Box" (2000) 99 Mich. L. Rev. 365; Ron Bouchard, "KSR v. Teleflex Part 1: Impact of U.S. Supreme Court Patent Law on Canadian Intellectual Property and Regulatory Rights Landscape" (2007) 15 Health L.J. 221 [Bouchard, "KSR"] (cites both Edward Hore, "A Comparison of US and Canadian Laws as They Affect Generic Pharmaceutical Drug Entry" (1992) 55 Food & Drug L.J. 373 and Andrew A. Caffrey & Jonathan M. Rotter, "Consumer Protection, Patents and Procedure: Generic Drug Market Entry and the Need to Reform the Hatch-Waxman Act" (2004) 9 Va. J.L. & Tech. 1 for data demonstrating that 75% of pharmaceutical patents litigated on their merits under respective US and Canadian "linkage regulations"--the regimes that link patent protection of marketed pharmaceuticals with the drug approval process based on safety and efficacy--are either invalid or not infringed).

(32) Joseph Farrell and Robert P. Merges, "Incentives to Challenge and Defend Patents: Why Litigation Won't Reliably Fix Patent Office Errors and Why Administrative Patent Review Might Help" (2004) 19 Berkeley Tech. L.J. 943.

(33) Nancy T. Gallini, "The Economics of Patents: Lessons from Recent U.S. Patent Reform" (2002) 16 J. Econ. Persp. 131 at 136.

(34) Glynn Lunney, "E-Obviousness" (2001) 7 Mich. Telecomm. & Tech. L. Rev. 363 at 374.

(35) Burk and Lemley, supra note 16; Boucbard, "KSR", supra note 31.

(36) Patent Act, supra note 18, s. 28.3. The section basically codifies the Cripps Question, as discussed.

(37) Sanofi, supra note I at para 51, citing Vaver, supra note 20 at 136.

(38) Sharp and Dohme Inc. v. Boots Pure Drug Co. Ltd., (1928) 45 R.P.C. 153 at 173; approved by the Supreme Court of Canada in Hoechst, supra note 10 at 155-57.

(39) See e.g. Judy A. Errat, "A Study on the Patent Law Standard of Non-obviousness" (1996) I.P. Pol'y Directorate, online: Industry Canada <http://www.ic.gc.ca/eic/site/ippd-dppi.nsf/ eng/ip00108.html> (review of the historical development of Canadian obviousness standards) [Errat]; Ron A. Bouchard, "Should Scientific Research in the Lead-up to Invention Vitiate Obviousness under the Patented Medicines (Notice of Compliance) Regulations: To Test or Not To Test?" (2007) 6 C.J.L.T. 1 [Bouchard, "To Test"].

(40) Note that this is a distinction which is not always followed in practice.

(41) Section 28.1(1) of the Act defines the claim date. For the purpose here, it can be understood to mean the time at which the patent application was filed (the "file date"). This is actually a bit more complicated because the claim date can be earlier than the filing date if reference is made to a related patent filed in another country. Furthermore, s. 28(3) of the Act permits inventors to disclose their invention for one year prior to the filing date without vitiating their patent due to obviousness, but this point is not relevant to the discussion here.

(42) General Tire & Rubber Co. v. Firestone Tyre & Rubber Co., Ltd., [1972] R.P.C. 457 at 497 [references omitted].

(43) Janssen-Ortho Inc. v. Novopharm Ltd. (2006), 57 C.P.R. (4th) 58 (F.C.) at para. 113 [Janssen-Ortho, FC], aff'd (2007) 2007 FCA 217, 59 C.P.R. (4th) 116 [Janssen-Ortho, FCA].

(44) Beloit, supra note 4 at para. 17.

(45) Bouchard, "To Test", supra note 39.

(46) Diversified Products Corp. v. Tye-Sil Corp., (1991) 35 C.P.R. (3d) 350 (F.C.A.) at para. 37 (quoting Tomlin J. (as he then was) in Samuel Parkes & Co. v. Cocker Bros. (1929), 46 R.P.C. 241 at 248 (C.A.)) [Diversified Products].

(47) Bouchard, "To Test", supra note 39.

(48) Janssen-Ortho, FCA, supra note 43 at para. 25, adopting the factors set out by Hughes J. in Janssen-Ortho, FC, supra note 43. Although these are sometimes referred to as the "Hughes factors", I will use the term "Janssen-Ortho factors" for two reasons. First, this term reflects that the Federal Court of Appeal adopted a slightly modified version of the list set out by Hughes J. Second, this term will distinguish between the Janssen-Ortho case and the decision of Hughes J. in BMS v. Apotex that is discussed at length in this paper.

(49) Ibid.

(50) Ibid. at para. 27.

(51) Also referred to as optical isomers.

(52) The dextro-rotatory isomer and the levo-rotatory isomer, so named because they rotate plane-polarized light in opposite directions.

(53) Essentially, the d-enantiomer (clopidogrel) causes all of the benefits of the racemic mixture; whereas, the l-enantiomer causes all of the toxicity. Such "absolute stereo-selectivity" is unusual because usually one enantiomer will be more biologically active than the other--i.e., one will be both more efficacious and more toxic than the other. In addition, the salts of the d-enantiomer, such as clopidogrel bisulfate, have a better therapeutic index than the salts of the racemic mixture.

(54) The case arose under the Patented Medicines (Notice of Compliance) Regulations SOR/93-133 (the "NOC Regulations"). These are the Canadian "linkage regulations" that prevent the entry of generic drugs into the market unless the generic manufacturer has demonstrated that they have addressed all relevant patents. For the purposes of this article, it is sufficient to note that courts apply the same tests of patent validity in NOC Proceedings as in infringement actions grounded in the Patent Act; see e.g. Sanofi, supra note 1 at paras. 15, 17.

(55) Whirlpool Corp. v. Camco Inc., [2000] 2 S.C.R. 1067, 2000 SCC 67 at para. 63 [Whirlpool].

(56) Ibid. at para. 65.

(57) Ibid. at para. 66. In my opinion, this definition is problematic because it is circular. A discussion of "obviousness-type" double patenting is beyond the scope of this comment. It is sufficient to note that it is often unclear how it differs from the normal obviousness inquiry--Shore J. provides one opinion on this issue. The Supreme Court in Sanofi does not provide further clarity on this issue, using the term "patentably distinct" without defining its meaning.

(58) (2005), 271 F.T.R. 159, 2005 FC 390 [Sanofi, FC]. In accordance with the NOC Regulations, the order technically prohibits the Minister of Health from issuing to Apotex the necessary authorization to market generic clopidogrel, rather than restraining Apotex directly.

(59) Ibid. at para. 87.

(60) (2006), 282 D.L.R. (4th) 179, 2006 FCA 421 [Sanofi, FCA].

(61) Ibid. at paras. 16-18 (in support of this position, the Court relied on a previous decision of the Federal Court of Appeal in Pfizer Canada Inc. v. Canada (Minister of Health), 2006 FCA 214, 52 C.P.R. (4th) 241).

(62) Ibid. at paras. 19-20.

(63) Sanofi-Synthelabo Canada Inc. v. Apotex Inc., 2008 SCC 61, 69 C.ER. (4th) 251 (Factum of the Appellant at paras. 8-11) [Apotex Factum]. In particular, Apotex argued that the "doctrine of selection patents" was fundamentally inconsistent with the prohibition against double patenting, and that it rewarded under-disclosure in genus patents in a manner that was inconsistent with the patent bargain.

(64) I.G. Farbenindustrie A.G.'s Patents, Re (1930), 47 R.P.C. 289 (Eng. Ch. Div.) at 322 [I.G. Farbenindustrie].

(65) Beloit, supra note 4 at para 29 (under the previous leading test, a patent was only anticipated if a prior publication contained "so clear a direction that a [PHOSITA] reading and following it would in every case and without possibility of error be lead to the claimed invention"); Free World Trust, supra note 17 (Supreme Court of Canada adopting the Beloit anticipation test).

(66) Synthon, supra note 19.

(67) For recent (pre-Sanofi) Canadian commentary on the law of anticipation see H. Sam Frost, "The Unique Problem of Inventions Which Are Fully Enabled and Fully Described, but Not Fully Understood (Merrell Dow's Terfenadine Revisited)" (2007) 20 I.P.J. 369; Tina Loucaides and Carol Yip, "Inherent Anticipation: Where Is Canada Heading?" (2006) 20 I.P.J. 91.

(68) Sanofi, supra note 1 at para. 41.

(69) Ibid. at para. 62.

(70) Ibid. at para. 60.

(71) Windsurfing, supra note 9; Pozzoli, supra note 9.

(72) Sanofi, supra note 1 at para. 67.

(73) Ibid. at para. 68.

(74) See ibid. at paras. 59, 64: in setting out these factors, Rothstein J. acknowledges the influence of the UK cases Generics (UK) Ltd. v. H. Lundbeck A/S, [2008] R.P.C. 19, [2008] EWCA Civ 311 and Conor Medsystems Inc. v. Angiotech Pharmaceuticals Inc., [2007] R.P.C. 20, [2007] EWCA Civ 5 at para. 45, rev'd [2008] R.P.C. 28, [2008] UKHL 49; the specific set of factors and their formulation is, however, distinct to Sanofi.

(75) Sanofi, ibid. at paras. 74-75.

(76) Ibid. at paras. 76-78.

(77) Ibid. at paras. 79-80.

(78) Ibid. at para. 81.

(79) Ibid. at paras. 72, 82.

(80) Ibid. at paras. 84-85.

(81) Ibid. at paras. 86-89.

(82) Ibid. at para. 90.

(83) Ibid. at para. 91.

(84) Ibid. at para. 114.

(85) Apotex Factum, supra note 63 at paras. 7, 28, 93.

(86) Ibid. at paras. 64-70.

(87) Under-disclosure in a genus patent would permit a patentee to effectively extend the duration of the monopoly by subsequently obtaining a selection patent. Under-disclosure could occur through the deliberate withholding of information--i.e., inclusion of only a subset of research data in the disclosure--or simply because of lack of incentives to conduct extensive research prior to obtaining the genus patent.

(88) In my opinion, the facts of Sanofi raise important questions about the doctrine of sound prediction. In Canada, to sustain its validity, a patent must disclose either an actually achieved result (i.e., prove that it does what it claims) or a basis for sound prediction of the result (i.e., show that it is likely to do what it claims). This discussion, however, is outside of the scope of this comment.

(89) Sanofi, supra note 1 at para. 107.

(90) Ibid. at para. 62.

(91) Ibid. at para. 60.

(92) Supra note 4 at para. 48; Sanofi, ibid. at para. 49 (it is noteworthy that Rothstein J. describes the new approach to anticipation as "a refinement of the approach set out in Beloit", but he does not use similar language to describe the relationship between the new obviousness framework and Beloit approach to obviousness).

(93) Noel Courage and Melanie Szweras, "What You Need to Know About the New Canadian Supreme Court Decision" (10 December 2008), online: <http:llwww.bereskinparr.com/ENG/ News/NewsletterArticles/update-Dec-9-2008.html>.

(94) Hoechst, supra note 10.

(95) Janssen-Ortho, FCA, supra note 43. As listed above, the "principal factors" are: (1) the invention; (2) the PHOSITA; (3) the PHOSITA's knowledge; (4) the climate in the relevant field at the time the alleged invention was made; (5) the motivation in existence at the time the alleged invention was made to solve a recognized problem; and (6) the time and effort involved in the invention. The "secondary factors" which may be helpful are: (7) commercial success; and (8) meritorious awards.

(96) Ibid. at paras. 24-28.

(97) The Janssen-Ortho case is procedurally complicated. The generic pharmaceutical manufacturer, Novopharm, was successful in the NOC proceeding because the claims for levofloxacin, its salts and their formulations were found obvious (Janssen-Ortho Inc. v. Novopharm Ltd., 2004 FC 1631, 35 C.P.R. (4th) 353, aff'd on other grounds (2005) 39 C.P.R. (4th) 197, leave to appeal to S.C.C. refused, [2005] 1 S.C.R. 776). Nonetheless, NOC proceedings do not constitute res judicata with respect to patent infringement actions: Novartis AG v. Apotex Inc. (2002), 22 C.P.R. (4th) 450 (F.C.A.) at para. 9. In a subsequent infringement action involving the same patent and the same parties, Hughes J. of the Federal Court, relying on a more fulsome record, came to the opposite conclusion--the patent was not obvious: JanssenOrtho, FC, supra note 43. The decision of Hughes J. was upheld on appeal to the FCA: Janssen-Ortho, FCA, ibid.

(98) Janssen-Ortho, FCA, ibid. at paras. 39-40.

(99) Apotex Factum, supra note 63 (Apotex cited Janssen-Ortho for the propositions that (1) Beloit is still the dominant test of obviousness (at para. 101) and (2) the obviousness analysis should not consist of rigid preventative rules that deny factfinders recourse to common sense (at para. 103)); Sanofi-Synthelabo Canada Inc. v. Apotex Inc., 2008 SCC 61, 69 C.P.R. (4th) 251 (Factum of the Respondent at paras. 16, 21) [Sanofi Factum]. (Sanofi cited Janssen-Ortho for the proposition that selection patents have been upheld by the Federal Court and the Federal Court of Appeal in other recent cases).

(100) Apotex Factum, ibid. at para. 98 (Apotex argued that a new obviousness standard should be adopted instead of Beloit); Sanofi Factum, ibid. (Sanofi argued that general principles of obviousness from Beloit, Hoechst and Diversified Products continued to be good law (at paras. 53-59), and that there was no problem with applying these principles to selection patents (at paras. 60-78)).

(101) After all, the genus patent disclosed roughly 250,000 different compounds, but 250,000 is still a finite number of solutions.

(102) Sanofi, supra note 1 at para. 66.

(103) Janssen-Ortho, FCA, supra note 43 at para. 43.

(104) Sanofi, supra note 1 at para. 89.

(105) See the text associated with supra note 82.

(106) Janssen-Ortho, FCA, supra note 43 at para. 40: "The interests and focus of other researchers in the field during the same period were not on ofloxacin at all, indicating that others skilled in the art were not motivated to work on the same problem that was occupying Dr. Hayakawa. Justice Hughes found as a fact that, during the relevant period, Daiichi was motivated to obtain levofloxacin from ofloxacin, and no one else was."

(107) Ibid.

(108) For an illustration of the problem, see e.g. the Consultation Document entitled "Practice Notice on Obviousness" released by the Canadian Intellectual Property Office (CIPO) dated 14 May 2009, online: <http://www.cipo.ic.gc.ca/eic/site/cipointernet-internetopic.nsf/eng/ wr01841.html>. CIPO's interpretation is problematic in that it fails to recognize that there are two disparate meanings of motivation, instead suggesting at 3-4 that the existence of a Janssen-Ortho-type general motive is sufficient to meet the Sanofi motive criterion [CIPO Practice Notice].

(109) KSR International v. Teleflex Inc., 127 S. Ct. 1727 (2007) at 1741[KSR].

(110) Sanofi, supra note 1 at para. 92: "The course of conduct and the time involved throughout demonstrate that the advantage of the dextro-rotatory isomer was not quickly or easily predictable. Had the dextro-rotatory isomer been 'obvious to try', it is difficult to believe that Sanofi would not have opted for it before unnecessary time and investment were spent on the racemic mixture."

(111) Janssen-Ortho, FCA, supra note 43 at para 25: "The length of time and expense involved in the invention may be indicators of inventive ingenuity, but they are not determinative because an invention may be the result of a lucky hit, or the uninventive application of routine techniques, however time consuming and expensive they may be."

(112) Pfizer, supra note 5.

(113) The Pfizer case was a NOC Proceeding. See supra note 54 for a brief description of NOC Proceedings.

(114) Pfizer, supra note 5 at para 28.

(115) Ibid. at para. 29.

(116) Ibid. at para. 37.

(117) The idea that it was possible to engage in a functional approach prior to Sanofi is supported by a quote from Janssen-Ortho, FCA, supra note 43 at para 27: "[The list of factors] is not a list of legal rules to be slavishly followed; nor is it an exhaustive list of the relevant factors. The task of the trial judge in each case is to determine, on the basis of the evidence, sound judgment and reason, the weight (if any) to be given to the listed factors and any additional factors that may be presented."

(118) Lundbeck Canada Inc. v. Canada (Minister of Health), 2009 FC 146, 73 C.P.R. (4th) 69 at para. 57 (suggesting the Janssen-Ortho factors may be relevant to the "obvious to try" inquiry given that the factors laid out by Rothstein J. were explicitly described as non-exhaustive) [Lundbeck]; CIPO Practice Notice supra note 108 at 3 ("The Office considers the guidance affirmed by the Federal Court of Appeal in [Janssen-Ortho] to remain relevant in view of the guidance of the Supreme Court in [Sanofi]."). Given that Janssen-Ortho is still good law (which follows from the holding in Pfizer) and the list of factors from Sanofi is non-exhaustive, it appears that there is nothing precluding courts from using the secondary factors from Janssen-Ortho (commercial success and meritorious awards).

(119) See the text associated with supra note 70.

(120) The citation of foreign patent decisions, in particular those from the UK, is not a new phenomenon in Canadian patent law. The contention here is that this phenomenon will become increasingly common.

(121) Pfizer, supra note 5 at paras. 42-45.

(122) More generally, it remains unclear after Pfizer to what extent UK jurisprudence on obviousness is relevant, and the only definitive conclusion is that Canadian courts ought not to feel bound by the outcome of patent litigation in other countries.

(123) See e.g. Lundbeck, supra note 118 at paras. 12-17, which reviews decisions on equivalent patents from UK, US, Australia, Germany, and notes that the rationale of these decisions may be persuasive.

(124) Rebecca Eisenberg, "Obvious to whom? Evaluating inventions from the perspective of the PHOSITA" (2004) 19 Berkley Tech. L.J. 885. Note that this criticism does not dispute that hindsight bias is a legitimate concern, instead it suggests that courts have been unduly focused on this issue.

(125) Ibid. at 897-99. It is worth noting, however, that this trend might change after KSR, supra note 109.

(126) See e.g. Sanofi, FC, supra note 58 at paras. 32-39, 70, 72, and 80-82 (Justice Shore's conclusions regarding the PHOSITA's knowledge rely heavily on the evidence presented by Sanofi's expert witness and Apotex's five expert witnesses).

(127) Ron A. Bouchard, "Living separate and apart is never easy: Inventive capacity of the PHOSITA as the tie that binds obviousness and inventiveness" (2007) 4:1 U. Ottawa L. & Tech.J. 1 [Bouchard, "PHOSITA"].

(128) Bayer Aktiengesellschaft v Apotex Inc. (1995), 60 C.P.R. (3d) 58 at para. 58; aff'd (1998), 16 C.ER. (4th) 417 (Ont. C.A.), leave to appeal to S.C.C. denied [1998] SCCA No 563 [Bayer[.

(129) Ibid. at para. 56, quoting John Bochnovic, "Invention/Inventive Step/Obviousness" in Gordon F. Henderson, ed., Patent Law of Canada (Scarborough: Carswell, 1994) at 48.

(130) Ibid. at para. 58.

(131) Whirlpool, supra note 55 at para. 74

(132) Janssen-Ortho, FCA, supra note 43.

(133) Ibid.

(134) Sanofi, supra note 1 at para. 62.

(135) Bouchard, "PHOSITA", supra note 127 at 29.

(136) For a review, see Burk and Lemley, supra note 16.

(137) Agreement on Trade-Related Aspects of Intellectual Property Rights, 15 April 1994, 33 I.L.M. 81, art. 27(1).

(138) Burk and Lemley, supra note 16 at 1634-38.

(139) Ibid.; contra R. Polk Wagner "Of Patents and Path Dependency: A Comment on Burk and Lemley" (2003) 18 Berkeley Tech. L.J. 1341, which criticizes the policy levers approach, in particular when it is applied at the "macro" level, on the grounds that inter alia it places policy choices in the hands of judges rather than the legislature, and it involves the application of categorical rules in a manner that is impermissible under traditional patent doctrine.

(140) Andrew Bernstein, "What's Obvious about (or after) Sanofi?" (Presentation to the Toronto Chapter of the Licensing Executives Society, 22 January 2009).

(141) Bridgeview, supra note 6 at para. 49; see also Uview Ultraviolet Systems Inc. v. Brasscorp Ltd., 2009 FC 58, 73 C.P.R. (4th) 161 at para. 189 ("obvious to try" test held to be not appropriate in the field of automotive air conditioning, but no discussion of why it was not appropriate).

(142) Does experimentation refer broadly to the use of the scientific method, or does it refer to particular experimental settings or materials? For example, do computer simulations constitute "experiments"?

(143) I am still operating under a utilitarian justification for patents. Therefore, I am not suggesting that labour should be rewarded because it is a justification for property rights--i.e., the Lockean view of property. Instead, I am referring to situations in which it may be desirable to incentivize research that is mundane and labour intensive if such research methods are the only methods available.

(144) Burk and Lemley, supra note 16 at 1648-51 (importance of PHOSITA in US jurisprudence); Errat, supra note 39; E. Richard Gold and Karen Lynne Durell, "Innovating the Skilled Reader: Tailoring Patent Law to New Technologies" (2005) 19 I.P.J. 189 (discussing how the PHOSITA can be a policy tool to set the appropriate scope of patents through claims construction).

(145) Bouchard, "PHOSITA", supra note 127.

(146) Tuno-Jen Chiang, "A Cost-Benefit Approach to Patent Obviousness" (2008) 82 St. John's L. Rev. 39.

(147) Chiang's proposal is useful for illuminating the challenges involved in applying a patentability test based on direct estimation of costs and benefits, and for illustrating that such a system is potentially feasible. Nevertheless, among the many challenges involved in applying Chiang's proposal would be the estimation of the timing of independent invention. This variable appears in the equations defining both the costs and the benefits associated with a patent, and Chiang's suggestion for how this variable might be estimated involves many of the same considerations that are part of the current obviousness inquiry. Ibid. at 65-66, 79.

(148) Improvements are listed as valid subject-matter in s. 2 of the Act, supra note 18. Moreover, s. 32 of the Act specifies that you can obtain a patent for an improvement to a patented invention, but that this does not give you the right to infringe the original invention.

(149) As noted above, the Court cited Maugham J. in I.G. Farbenindustrie, supra note 64 at 322 for the proposition that a selection patent "does not in its nature differ from any other patent".

(150) In some circumstances, this might make sense, as a 1% improvement in efficacy can be clinically relevant. In practice, it is questionable whether such a fine-grained improvement could be detected based on the quality of data from preliminary testing that is typically included in patent applications.

(151) BMS, supra note 15; Bridgeview, supra note 6, was released three weeks before BMS, but I do not consider it to be a detailed application of the new obviousness test because, as described above, it considered Sanofi to merely be a refinement of Beloit.

(152) The BMS case was a NOC Proceeding. See supra note 54 for a brief description of NOC Proceedings.

(153) Hughes J. also considers the anticipation test set out in Sanofi, but this is outside of the scope of this comment.

(154) BMS, supra note 15 at para. 180.

(155) Ibid. at para. 183.

(156) Ibid. at para. 160.

(157) Ibid. at paras. 170-71.

(158) For an alternative approach, see Lundbeck, supra note 118 at paras. 7, 44: in this post-Sanofi case, the three-part selection patent test is interpreted as an additional hurdle to the conventional requirements, but the issue is not explored in depth as Harrington J. concludes that the patent at issue is not actually a selection patent.

(159) See text associated with supra note 92; Eli Lilly Canada Inc. v Novopharm Ltd., 2009 FC 235, 73 C.P.R. (4th) 253 (Hughes J. describes Sanofi as a "restatement" of the Canadian law of obviousness); CIPO Practice Notice, supra note 108 (drafting of new practice notice is consistent with the idea that obviousness analysis has changed after Sanofi).

ANDREW MOESER, B.Sc. (McGill), M.Sc. (McGill), J.D. (Toronto)
COPYRIGHT 2009 University of Toronto, Faculty of Law
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Moeser, Andrew
Publication:University of Toronto Faculty of Law Review
Date:Mar 22, 2009
Words:20456
Previous Article:Law, Mystery, and the Humanities.
Next Article:Law and Capitalism: What Corporate Crises Reveal About Legal Systems and Economic Development Around the World.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters