Apotex Inc. v. Lundbeck Canada Inc.

Apotex Inc. v. Lundbeck Canada Inc.
Court (s) Database
Federal Court of Appeal Decisions
Date
2010-11-25
Neutral citation
2010 FCA 320
File numbers
A-129-09, A-135-09, A-139-09
Decision Content
Federal Court of Appeal
Cour d'appel fédérale
Date: 20101125
Dockets: A-129-09
A-135-09
A-139-09
Citation: 2010 FCA 320
CORAM: NOËL J.A.
PELLETIER J.A.
TRUDEL J.A.
Docket: A-129-09
BETWEEN:
APOTEX INC.
Appellant
and
LUNDBECK CANADA INC.
Respondent
and
THE MINISTER OF HEALTH
Respondent
and
H. LUNDBECK A/S
Respondent
Docket: A-135-09
BETWEEN:
MYLAN PHARMACEUTICALS ULC
(formerly Genpharm ULC)
Appellant
and
LUNDBECK CANADA INC.
Respondent
and
THE MINISTER OF HEALTH
Respondent
and
H. LUNDBECK A/S
Respondent
Docket: A-139-09
BETWEEN:
COBALT PHARMACEUTICALS INC.
Appellant
and
LUNDBECK CANADA INC., H. LUNDBECK A/S
and THE MINISTER OF HEALTH
Respondents
Heard at Ottawa, Ontario, on September 14, 2010.
Judgment delivered at Ottawa, Ontario, on November 25, 2010.
REASONS FOR JUDGMENT BY: NOËL J.A.
CONCURRED IN BY: PELLETIER J.A.
TRUDEL J.A.
Federal Court of Appeal
Cour d'appel fédérale
Date: 20101125
Dockets: A-129-09
A-135-09
A-139-09
Citation: 2010 FCA 320
CORAM: NOËL J.A.
PELLETIER J.A.
TRUDEL J.A.
Docket: A-129-09
BETWEEN:
APOTEX INC.
Appellant
and
LUNDBECK CANADA INC.
Respondent
and
THE MINISTER OF HEALTH
Respondent
and
H. LUNDBECK A/S
Respondent
Docket: A-135-09
BETWEEN:
MYLAN PHARMACEUTICALS ULC
(formerly Genpharm ULC)
Appellant
and
LUNDBECK CANADA INC.
Respondent
and
THE MINISTER OF HEALTH
Respondent
and
H. LUNDBECK A/S
Respondent
Docket: A-139-09
BETWEEN:
COBALT PHARMACEUTICALS INC.
Appellant
and
LUNDBECK CANADA INC., H. LUNDBECK A/S
and THE MINISTER OF HEALTH
Respondents
REASONS FOR JUDGMENT
NOËL J.A.
[1] These are three appeals from judgments rendered by Harrington J. of the Federal Court (the Applications Judge) granting the applications brought by Lundbeck Canada Inc. (the respondent or Lundbeck) to prohibit the Minister of Health (the Minister) from issuing a Notice of Compliance (NOC) to Apotex Inc. (Apotex), Mylan Pharmaceuticals ULC, formerly Genpharm ULC (Genpharm) and Cobalt Pharmaceuticals Inc. (Cobalt) (collectively the appellants) pursuant to section 6 of the Patented Medicines (Notice of Compliance) Regulations, SOR/93-133, in respect of each of the appellants’ generic version of the drug containing escitalopram for use as an antidepressant, until after the expiration of Canadian Patent No. 1,339,452 (the ‘452 patent).
[2] The applications before the Federal Court were heard consecutively over a three-week period from December 1 to December 18, 2008. Although they were never joined, the Applications Judge opted to dispose of the applications on the basis of a single set of reasons. The following reasons dispose of the three appeals.
[3] The issue in the three appeals is whether the Applications Judge properly held that the appellants’ respective allegations of invalidity regarding the ‘452 patent were not justified and that accordingly they had failed to show that their generic version of escitalopram would not infringe this patent. For the reasons which follow, I am of the view that the appeals should be dismissed.
BACKGROUND
[4] Apotex, Genpharm and Cobalt filed Notice of Allegations (NOAs) on April 20, January 23 and June 18, 2007 respectively, making a number of allegations, some common and some specific to the particular applicant. At the core of each NOA is the allegation that the ‘452 patent is an invalid selection patent and that the alleged invention was both obvious and anticipated. Also raised are allegations that the ‘452 patent lacks utility, fails to soundly predict the invention (Apotex), provides insufficient disclosures (Apotex, Genpharm and Cobalt) and is ambiguous (Apotex and Genpharm).
[5] The ‘452 patent, entitled “Enantiomers of Citalopram and Derivatives Thereof”, was applied for in June 1989 by the respondent, based on a United Kingdom priority date of June 1988. The patent was granted in 1997 and expires in 2014. It claims escitalopram as a useful antidepressant and describes two methods of obtaining it.
[6] The claims at issue are claims 1 and 3, as well as claim 5, insofar as it is dependent on claim 3:
- 1 -
A compound selected from substantially pure (+)-1-(3-Dimethylaminopropyl)-1-(4’-fluorophenyl)-1, 3-dihydroisoben-zofuran-5-carbonitrile and non-toxic acid addition salts thereof.
[…]
- 3 -
A pharmaceutical composition in unit dosage form useful as an antidepressant comprising a pharmaceutically-acceptable diluent or adjuvant and, as an active ingredient, an effective amount of a compound as defined in [c]laim 1.
[…]
- 5 -
A pharmaceutical composition in unit dosage form, useful as an antidepressant according to claim 3 or 4, wherein the active ingredient is present in an amount from 0.1 to 100 milligram per unit dose.
[7] It is also useful to set out claim 2:
A compound of [c]laim 1 being the pamoic acid salt of (+)-1-(3-dimethylaminopropyl)-1-(4’-fluorophenyl-1, 3-dihydroisobenzofuran-5-carbonitrile.
[8] The ‘452 patent notes that citalopram was disclosed in the now expired U.S. patent number 4,136,193 (U.S. patent ‘193). U.S. patent ‘193 disclosed a formula which might produce a few hundred compounds and specifically claimed citalopram as a useful antidepressant.
[9] The ‘452 patent also indicates that a precursor of citalopram, a diol, was disclosed in U.S. patent number 4,650,884 (U.S. patent ‘884), filed in August 1985 and entitled “Novel Intermediate and Methods for Its Preparation”.
[10] The Applications Judge provided a useful summary of the relevant chemistry with which the parties do not take issue (Reasons, paras. 22 to 28). It can be seen from this summary that the chemical compound at issue in this case, escitalopram, is one of the two enantiomers of citalopram, a racemate. Escitalopram is also known as (+) citalopram and S-citalopram.
[11] Carbon-centered molecules, like the compound at issue, have a three-dimensional structure. If that carbon atom is bonded to four different atoms or groups of atoms, as is the case here, the molecule is described as having an asymmetric centre. These chemical compounds are identical save that they exist in two space-occupying forms called “enantiomers,” which are non-superimposable mirror images of one another. Such asymmetric molecules are called chiral, coming from the Greek word for hand, as a left hand and a right hand are mirror images of each other but are not superimposable. When a molecule with these characteristics is synthesized, both enantiomers are produced in equal proportions. This mixture is called a racemate or a racemic mixture.
[12] When a drug is a racemate, although the two enantiomers of the drug have the same molecular formula, they can interact differently within the human body. Just like the key and lock analogy, the racemate and each of its enantiomers can dock in different ways with biomolecules in the body; the consequence being that they can have pharmacological properties of their own.
[13] Because of their identical chemical formula, two unrelated nomenclatures are used to identify enantiomers. The first nomenclature is based on the direction in which the enantiomer directs the plane of polarized light. If the plane is turned clockwise, the enantiomer is identified as (+), d or dextro-rotary; if the plane is turned counter-clockwise, it is identified as (-), l or levo-rotary. Escitalopram is the (+) enantiomer of citalopram: it thus directs the plane of light in a clockwise direction. Because a racemate is a mixture of two enantiomers that rotate polarized light in opposite directions, it is designated as (+/-).
[14] The second nomenclature is the Cahn-Ingold-Prelog convention which specifies absolute configuration. The substituents around the chiral centre are “sized” according to their atomic numbers. If the sequence from the largest to the smallest flows in a clockwise direction, the enantiomer is assigned the R or rectus designation. Otherwise it is assigned the S or sinister designation. A racemate is designated (R, S). Escitalopram is the S-enantiomer of citalopram.
[15] The above summary is drawn from the expert evidence of Professor Davies who was called by Lundbeck in the three proceedings and Dr. Newton who was called by both Genpharm and Cobalt in their respective proceedings (Reasons, para. 28). Seven other experts were called: Professor Clark testified on behalf of Lundbeck in all three proceedings; Dr. Keana, Dr. McClelland and Professor Ward appeared on behalf of Apotex; Professor Chong and Dr. Collicott on behalf of Genpharm; and Dr. Kissinger on behalf of Cobalt.
[16] Each expert advanced views on racemates, the methods to resolve them and the degree of difficulty which this can present (Reasons, paras. 63 to 72). In the end, the Applications Judge came to the view that the opinions of Professors Davies and Clark who testified on behalf of Lundbeck in all three proceedings were to be preferred.
RELEVANT LEGAL PROVISIONS
[17] Given the date on which the ‘452 patent was applied for, the Patent Act, R.S.C. 1985, c. P-4 (the Patent Act) as it read prior to October 1, 1989, applies. The term “invention” is defined in section 2 as follows:
[…] any new and useful art, process, machine, manufacture or composition of matter, or any new and useful improvement in any art, process, machine, manufacture or composition of matter;
Toute réalisation, tout procédé, toute machine, fabrication ou composition de matières, ainsi que tout perfectionnement de l'un d'eux, présentant le caractère de la nouveauté et de l'utilité.
[18] Subsection 27(1) regarding disclosure reads:
Subject to this section, any inventor or legal representative of an inventor of an invention that was
(a) not known or used by any other person before he invented it,
(b) not described in any patent or in any publication printed in Canada or in any other country more than two years before presentation of the petition hereunder mentioned, and
(c) not in public use or on sale in Canada for more than two years prior to his application in Canada,
may, on presentation to the Commissioner of a petition setting out the facts, in this Act termed the filing of the application, and on compliance with all other requirements of this Act, obtain a patent granting to him an exclusive property in the invention.
Sous réserve des autres dispositions du présent article, l’auteur de toute invention ou le représentant légal de l’auteur d’une invention peut, sur présentation au commissaire d’une compétition exposant les faits, appelée dans la présente loi le « dépôt de la demande », et en se conformant à toutes les autres prescriptions de la présente loi, obtenir un brevet qui lui accorde l’exclusive propriété d’une invention qui n’était pas :
a) connue ou utilisée par une autre personne avant que lui-même l’ait faite;
b) décrite dans un brevet ou dans une publication imprimée au Canada ou dans tout autre pays plus de deux ans avant la présentation de la pétition ci-après mentionnée;
c) en usage public ou en vente au Canada plus de deux ans avant le dépôt de sa demande au Canada.
[19] Subsection 34(1) concerning the specification reads as follows:
An application shall in the specification of the invention
(a) correctly and fully describe the invention and its operation or use as contemplated by the inventor;
(b) set out clearly the various steps in a process, or the method of constructing, making, compounding or using a machine, manufacture or composition of matter, in such full, clear, concise and exact terms as to enable any person skilled in the art or science to which it pertains, or with which it is mostly closely connected, to make, construct, compound or use it;
…
Dans le mémoire descriptif, le demandeur :
a) décrit d’une façon exacte et complète l’invention et son application ou exploitation, telles que les a conçues l’inventeur;
b) expose clairement les diverses phases d’un procédé, ou le mode de construction, de confection, de composition ou d’utilisation d’une machine, d’un objet manufacturé ou d’un composé de matières, dans des termes complets, clairs, concis et exacts qui permettent à toute personne versée dans l’art ou la science dont relève l’invention, ou dans l’art ou la science qui s’en rapproche le plus, de confectionner, construire, composer ou utiliser l’objet de l’invention;
…
THE FEDERAL COURT DECISION
[20] Although seized with distinct applications, the Applications Judge opted to issue one set of reasons. He explained that counsel for the appellants were invited to attend all three hearings, that memoranda of fact and law in all three applications were exchanged and that the commonality of the applications greatly surpassed their distinctiveness (Reasons, para. 20). He added that the relevant distinctions would be made in the course of his reasons (Reasons, para. 21).
[21] The Applications Judge noted at the beginning of his analysis that patent construction was at the heart of the dispute and outlined the applicable principles as set out by the Supreme Court in Free World Trust v. Électro Santé Inc., 2000 SCC 66, [2000] 2 S.C.R. 1024 [Free World Trust] and Whirlpool Corp. v. Camco Inc., 2000 SCC 67, [2000] 2 S.C.R. 1067.
[22] According to the Applications Judge, the invention disclosed in the ‘452 patent relates to the two novel enantiomers of citalopram including their pharmaceutically acceptable salts and their use as an antidepressant. The ‘452 patent also notes that previous attempts at resolving citalopram, which the patent states were disclosed in the U.S. patent ‘193, had failed and that it was discovered that a precursor of citalopram, a diol disclosed in U.S. patent ‘884, could be resolved into its enantiomers and then converted to the enantiomers of citalopram in a stereoselective way. The ‘452 patent describes two methods to obtain escitalopram (Reasons, para. 41).
[23] With respect to the selection patent issue, the question was whether escitalopram had a special or unexpected advantage over citalopram (Reasons, para. 37). The Applications Judge found that escitalopram was at best 1.6 times more potent than citalopram. This was not sufficiently unexpected to serve as the foundation for a selection patent “[s]ince it was well within the realm of possibility that more, and indeed sometimes all, of the desired biological activity of a racemate might rest within one enantiomer rather than in the other” (Reasons, para. 43). He thus concluded that if the ‘452 patent was a selection patent, it was invalid.
[24] However, the Applications Judge held that the ‘452 patent was not a selection patent. He found instead that it was a patent for a new substance: substantially pure escitalopram. He reached that conclusion based on the fact that this particular compound was not disclosed let alone claimed in either U.S. patents ‘193 or ‘884 (Reasons, para. 42). In coming to this conclusion, the Applications Judge dismissed the argument that a patent for a racemate automatically discloses and claims the two enantiomers (Reasons, para. 47).
[25] With respect to anticipation, the Applications Judge after reviewing the evidence held that the documents put into evidence by the appellants including the prior patents did not disclose escitalopram as a useful antidepressant and could not therefore form the basis for an allegation of anticipation by prior disclosure. Although it was known to the skilled addressee that within citalopram were two enantiomers, and that it might not be a surprise that one might be more potent than the other, one would not know the qualities of the two enantiomers without separating and testing them (Reasons, paras. 50 to 52).
[26] Turning to obviousness, the Applications Judge applied the four-step approach identified by the Supreme Court in Apotex Inc. v. Sanofi-Synthelabo Canada Inc., 2008 SCC 61, [2008] 3 S.C.R. 265 [Sanofi]. First, he found that the skilled addressee was a “team centered around a medicinal chemist who had access to and makes use of others with different skill sets such as analytical chemists and psychiatrists” and that “theoretical knowledge of, and practical experience in, the methods of resolving racemate were essential” (Reasons, paras. 36 and 53 to 58).
[27] Dealing with the second step, the Applications Judge found that the construction of the relevant claims posed no difficulty, i.e. claim 1 was for substantially pure escitalopram and non-toxic additional salts thereof, claim 3 was for a chemical composition in unit dosage form useful as an antidepressant, and claim 5 was for a unit dosage form wherein the active ingredient ranged from 0.1 to 100 milligrams per unit dose. He noted that the ‘452 patent did not claim that escitalopram was better than citalopram (Reasons, para. 59).
[28] As to the difference between the prior art and the inventive concept underlying the ‘452 patent, the Applications Judge found that the prior art disclosed citalopram to be useful as an antidepressant, but that it neither disclosed its two enantiomers nor predicted that either would be useful as an antidepressant. He held that the inventive step was the resolution of citalopram in sufficient quantity to permit the testing disclosed in the patent; without it, it was impossible to determine the usefulness of the enantiomers (Reasons, para. 60).
[29] Turning to the fourth step – i.e. whether the claimed invention was obvious to the skilled addressee – the Applications Judge after reviewing the expert testimony considered the methods for resolving citalopram available in 1988, the route taken by Lundbeck to resolve citalopram and the allegation that the prior art disclosed the resolution of citalopram (Reasons, paras. 63 to 74). He first noted that motivation is not instructive in this case (Reasons, paras. 79 to 83). Concerning the resolution of citalopram, or of its precursor (the diol which is a racemate), the Applications Judge described the two methods available to resolve citalopram at the claim date: the classical method of fractional crystallization and chiral high pressure liquid chromatography (HPLC) (Reasons, paras. 84 to 88). The Applications Judge later referred to Lundbeck’s eight-year quest to resolve citalopram (Reasons, paras. 90 to 102). He found that it was not obvious to try to resolve citalopram and that in any event, it was certainly not obvious that what was being tried would work (Reasons, para. 103).
[30] The Applications Judge went on to hold that an inventive step was required to resolve citalopram with the result that the allegation of obviousness was not made out (Reasons, para. 124).
[31] The Applications Judge then addressed Genpharm’s contention that escitalopram was anticipated on the theory that the body resolves citalopram by itself into the two enantiomers. According to Genpharm, the instruction to ingest citalopram in U.S. patents ‘193 and ‘884 results in the production of escitalopram in the body. Genpharm made this proposition on the basis of the decision of the House of Lords in Merrell Dow Pharmaceuticals Inc. v. HN Norton and Co. Ltd., [1995] UKHL 14, [1996] RPC 76 [Merrell Dow] (Reasons, paras. 125 to 128).
[32] The Applications Judge distinguished Merrell Dow and held that Genpharm’s contention that the body resolves citalopram into substantially pure escitalopram was based on conjecture (Reasons, para. 129).
[33] The Applications Judge then dealt with Genpharm’s allegation that claim 1 was ambiguous because it did not define what it meant by “substantially pure escitalopram”. The Applications Judge found no ambiguity since the examples given showed purity in excess of 99%; one expert asserted that “substantially pure” would mean at least 95% since a standard method only detects impurities if they are present at a level of at least 5% (Reasons, para. 130). The Applications Judge also dismissed Apotex’s suggestion that the +/- nomenclature used is ambiguous because different solvents may rotate light in a different way thus altering the +/- designation. He pointed out that the solvents to be used were fully described in the patent (Reasons, para. 131).
[34] The Applications Judge also dismissed Apotex’s argument that the ‘452 patent did not offer a sound prediction of utility because it was based on studies conducted on rodents. According to the Applications Judge the testing done on rodents, which was the same as had been done for citalopram, soundly predicted that escitalopram would be useful as an antidepressant in humans. The Applications Judge concluded that “[u]sefulness was promised, usefulness was predicted and usefulness was delivered” (Reasons, para. 134).
[35] Finally, the Applications Judge dismissed Apotex’s contention that escitalopram lacked utility on the basis that the pamoic salt of escitalopram in claim 2 was toxic. Relying on Burton Parsons Chemicals, Inc. v. Hewlett-Packard (Canada) Ltd., [1976] 1 S.C.R. 555, [Burton Parsons], he held that the skilled addressee would not use a toxic salt (Reasons, para. 139).
ALLEGED ERRORS
Issues common to all three appellants
The ‘452 patent is an invalid selection patent
[36] All three appellants submit that the ‘452 patent is an invalid selection patent. They argue that the ‘452 patent is a selection because escitalopram was disclosed in U.S. patent ‘193 which claimed citalopram and its use as an antidepressant. Although expressed in a different way, both Apotex and Genpharm submit that the Applications Judge misapprehended and misapplied the law of selection patents by requiring a prior enabling disclosure or claim for escitalopram in order to characterize the ‘452 patent as a selection patent.
[37] The appellants submit, relying primarily on the decision of the Supreme Court in Sanofi, that courts have treated patents for enantiomers as selection patents even though the enantiomers were not anticipated. As well, they refer to the decision of this Court in Pfizer Canada Inc. v. Canada (Minister of Health), 2008 FCA 108 [Pfizer] to show that patents claiming an enantiomer of a previously disclosed racemate have been held to be selection patents. The decisions of the Federal Court in Janssen-Ortho Inc. v. Novapharm Ltd., 2004 FC 1631 [Janssen-Ortho I] and Janssen-Ortho Inc. v. Novapharm Ltd., 2006 FC 1234 [Janssen-Ortho II] are also referred to.
[38] Genpharm for its part emphasizes the fact that the ‘452 patent describes escitalopram as a “surprising discovery”. According to Genpharm, the “surprise” is the “special quality” that would validate escitalopram as a selection from citalopram.
[39] Apotex argues that the Applications Judge erred in construing the claims contained in U.S. patent ‘193 from the perspective of a patent lawyer. By observing that it would be unwise to draft U.S. patent ‘193 so as to include untested compounds because to do so would leave the patent open to an overclaim attack, the Applications Judge took the perspective of the patent draftperson instead of the person of ordinary skill in the art and made inquiries which are not permissible when construing the patent.
[40] Apotex contends that by describing the subject matter of U.S. patent ‘193 in terms of a chemical formula without reference to optical information, the draftsperson intended to encompass all compounds having the same chemical formula: the racemate, R-citalopram and S-citalopram. It follows that U.S. patent ‘193 discloses and claims escitalopram.
Anticipation
[41] All three appellants argue that escitalopram was anticipated by U.S. patent ‘193. Apotex alleges that escitalopram is formed “always”, “inevitably” and “with no possibility of error” upon making citalopram and that the Applications Judge found as such when he noted that anyone making escitalopram would infringe on U.S. patent ‘193 (Reasons, para. 83). Genpharm also relies on this passage.
[42] Cobalt submits that the Applications Judge’s finding that U.S. patent ‘193 did not disclose the enantiomers of citalopram is inconsistent with his finding that an undergraduate student in organic chemistry would have known that citalopram was made up of two enantiomers. Since the person skilled in the art would be reading with a mind willing to understand the patent, Cobalt alleges that U.S. patents ‘193 and ‘884 disclosed and enabled both enantiomers of citalopram.
[43] Genpharm contends that escitalopram was anticipated because both U.S. patents ‘193 and ‘884 provided instruction to ingest citalopram which, if followed, would lead to the production of substantially pure escitalopram in the body. Genpharm submits that the Applications Judge erred in rejecting the evidence adduced on this point as “outright conjecture” without analyzing it.
Issues common to Genpharm and Cobalt
Obviousness
[44] With respect to obviousness, Genpharm submits that there was no difference between the state of the art and the ‘452 patent. In particular, Genpharm contends that since U.S. patents ‘193 and ‘884 disclosed to the person skilled in the art citalopram and its two enantiomers and, as their usefulness as an antidepressant was known, the ‘452 patent contributes nothing to the existing body of knowledge. In any event, it was more or less evident that what was being tried ought to work.
[45] Genpharm further alleges that the Applications Judge erred in assuming that 100 milligrams of escitalopram were required for testing. With respect to the resolution of citalopram using chiral HPLC, Genpharm submits that the Applications Judge made a palpable and overriding error in concluding that analytical columns could not have produced sufficient material for detection and for the biological testing required by the ‘452 patent. Genpharm also submits that it would have been obvious in 1988 to use the intermediate diol, disclosed in U.S. patent ‘884, to obtain substantially pure escitalopram.
[46] With respect to the Applications Judge’s finding that the experts were operating with hindsight, Genpharm submits that the Applications Judge failed to distinguish between inevitable and impermissible hindsight. Genpharm further alleges that, due to the U.S. Food and Drug Administration 1987 Guidelines (1987 FDA Guidelines), drug companies were motivated to resolve racemates so as to obtain information on the properties of the enantiomers. Finally, Genpharm submits that Dr. Bøgesø, the co-inventor of citalopram, had built-in biases which prevented him from quickly resolving citalopram; the person skilled in the art would not have such biases.
[47] Cobalt also submits that it was obvious for Lundbeck to resolve citalopram because citalopram and its enantiomers were known to the person skilled in the art as was their use as an antidepressant. It would therefore have been self-evident to Lundbeck to pursue escitalopram and it had a fair expectation of success.
Procedural fairness and inadequacy of reasons
[48] Genpharm and Cobalt allege that the Applications Judge erred in using evidence that was not led in their respective proceeding and in issuing one set of reasons for three different proceedings.
[49] Genpharm contends that the Applications Judge’s reasons do not adequately identify the specific findings with respect to Genpharm regarding anticipation and obviousness. According to Genpharm, this makes “meaningful review impossible and prevents [it] from properly arguing its appeal” (Genpharm’s memorandum of fact and law, para. 48).
[50] Cobalt further argues that since Lundbeck had access to the documents filed in all three proceedings and made submissions comparing the evidence, Lundbeck had a tactical advantage that was prejudicial to it.
Issues specific to Apotex
Sound prediction
[51] Apotex argues that the utility of escitalopram was not soundly predicted. In particular, there was “nothing in the ‘452 patent to correlate the ability of citalopram to cause changes in mouse motor function and rat brain-cell effects with its clinical antidepressant effect in humans” (Apotex’s memorandum of fact and law, para. 68). Apotex contends that without this correlative information, there can be no factual basis for sound prediction.
[52] Apotex also contends that the ‘452 patent lacks utility because it covers a toxic pamoate salt which Lundbeck admitted is toxic in a 2004 patent application. Further, Apotex submits that the Applications Judge erred in construing claim 1 of the ‘452 patent as excluding such salt because it would be obvious to a person skilled in the art not to use toxic salt. According to Apotex, the Applications Judge’s task was to determine how the person skilled in the art would interpret the words of claim 1 – “non-toxic acid addition salts” – at the date of issue of the ‘452 patent. Apotex submits that the Applications Judge erred in his construction of claim 1.
Ambiguity
[53] Apotex submits that the ‘452 patent does not teach the person skilled in the art how to use the solvents to obtain escitalopram. Since there was no such indication, it cannot be said that the patent was defined in “full, clear, concise and exact” terms (section 34 of the Patent Act).
Insufficiency of disclosure
[54] Finally, Apotex submits that the disclosure in the ‘452 patent is insufficient for laying a false trail regarding human administration of escitalopram and that the Applications Judge “erroneously understood Apotex’s allegation to be made pursuant to section 53 of the Patent Act” rather than section 34 (Apotex’s memorandum of fact and law, para. 90). Apotex contends that had the proper inquiry been made, the allegation of insufficiency would have been justified.
ANALYSIS AND DECISION
[55] Before embarking on the analysis, it is useful to recall that questions of law are to be ascertained on a standard of correctness and that factual findings cannot be reversed absent a palpable and overriding error (Housen v. Nikolaisen, 2002 SCC 33, [2002] 2 S.C.R. 235). The identification of the legal requirements for the existence of a selection patent as well as the construction of the relevant claims in the patents considered by the Applications Judge give rise to questions of law. The remaining issues are for the most part factual.
[56] The appellants essentially reiterate on appeal the arguments which they made before the Applications Judge. The reasons which follow address most of these issues. With respect to those that are not addressed, I endorse the reasons of the Applications Judge in disposing of them.
Selection patent
[57] The selection patent issue, which is central to each of the appellants’ case, presents itself in simple terms: the appellants maintain that the ‘452 patent is a selection patent and as the Applications Judge found that the selected compound had no special advantage, the patent is invalid.
[58] The Applications Judge held that the ‘452 patent was not a selection patent because, in his view, escitalopram was an original compound which was not selected from a previously patented compound. He therefore conducted his analysis on the basis that the ‘452 patent was an ordinary patent for an original compound and as he found this compound to be both novel and useful (see the definition of “invention” in section 2 of the Patent Act), he held the patent to be valid. At the same time, he made it clear that escitalopram’s claimed usefulness was no greater than citalopram’s and that therefore the invention, as claimed, had no special advantage over citalopram.
[59] The question whether the ‘452 patent is a “selection patent” depends on the legal meaning to be given to these words. The term “selection patent” is not found in the Patent Act. However, the Supreme Court in Sanofi held that a system of genus and selection patents is acceptable in principle under the Patent Act, on the line of authority stemming from in I. G. Farbenindustrie A. G.’s Patents (1930), 47 R.P.C. 289 (Ch. D.) [Farbenindustrie]. The Supreme Court refers to that case to circumscribe what is to be considered as a selection patent (Sanofi, para. 9):
The locus classicus describing selection patents is the decision of Maugham J. in [Farbenindustrie]. At p. 321, he explained that in the field of chemical patents (which would of course include pharmaceutical compounds), there are often two “sharply divided classes”. The first class of patents, which he called originating patents, is based on an originating invention, namely, the discovery of a new reaction or a new compound. The second class comprises patents based on a selection of compounds from those described in general terms and claimed in the originating patent. Maugham J. cautioned that the selected compounds cannot have been made before, or the selection patent “would fail for want of novelty”. But if the selected compound is “novel” and “possess[es] a special property of an unexpected character”, the required “inventive” step would be satisfied (p. 321). At p. 322, Maugham J. stated that a selection patent “does not in its nature differ from any other patent”.
[My emphasis]
[60] In accepting that a system of genus and selection patents was acceptable under Canadian law, the Supreme Court explained that its application was consistent with the Patent Act (Sanofi, para. 31):
Section 27(1) of the Act requires as a condition for obtaining a patent that the invention was not “known or used” and was not “described” in any patent or any publication more than two years before the patent application was filed. In the context of genus and selection patents, in E. I. Du Pont de Nemours & Co. (Witsiepe’s) Application, [1982] F.S.R. 303 (H.L.), Lord Wilberforce stated, at p. 311:
It is the absence of the discovery of the special advantages, as well as the fact of non-making, that makes it possible for such persons to make an invention related to a member of the class.
The compound made for the selection patent was only soundly predicted at the time of the genus patent. It was not made and its special advantages were not known. It is for those reasons that a patent should not be denied to the inventor who made and discovered the special advantages of the selection compound for the first time.
[61] It is apparent from the foregoing that a selection patent must be preceded by a prior patent – referred to as a genus or originating patent – which, in the words of Maugham J. in Farbenindustrie, describes in general terms and claims compounds from which a selection is made. That the selection is made from compounds generally described and claimed in a prior patent does not necessarily mean that the selected compound is anticipated (Sanofi, para. 19). So long as the selected compound is new – in that it has never been made – and has a special advantage that was not previously known and that is peculiar to it, patent protection may be available (Sanofi, paras. 10 and 31). However, a definitive conclusion cannot be reached absent a complete analysis (Eli Lilly Canada Inc. v. Novopharm Limited, 2010 FCA 197, paras. 27 to 33 [Eli Lilly]). In this respect, it is worth repeating that a selection patent does not differ from any other patent (Sanofi, para. 9).
[62] Against this background, the first question which has to be answered is whether U.S. patents ‘193 and ‘884, together or singly, describe in general terms and claim compounds from which escitalopram was selected. In this respect, the recent decision of this Court in Eli Lilly on which the parties made supplemental submissions is of limited assistance since it was accepted that the compound in issue in that case had been selected from a previously patented class of compounds (Eli Lilly, para. 7).
[63] The Applications Judge answered this question in the negative. In particular, he found that U.S. patent ‘193 claims citalopram and that this claim did not encompass escitalopram. Apotex made the point that escitalopram was nevertheless within the claim because the subject matter of that patent was described in terms of a chemical formula without optical information distinguishing the racemate from the enantiomers.
[64] However, the Applications Judge held that the person skilled in the art would have read the formula, as of the claim date, as referring to the compound produced by the formula, i.e. the racemate, and nothing else. He came to this conclusion because no stereochemical information was provided. Given this, he held that the skilled addressee would not have read U.S. patent ‘193 as claiming anything other than the racemate (see Professor Davies’ affidavit at para. 85, A-129-09 Appeal Book, Vol. 4, p. 934; A-135-09 Appeal Book, Vol. 4, p. 1099; A-139-09 Appeal Book, Vol. 4, p. 1337).
[65] In so holding, the Applications Judge acknowledged that citalopram’s chemical structure – and the chemical formula reflecting it – reveals the existence of the enantiomers. However, he rejected the argument that this in itself was sufficient to read U.S. patent ‘193 as claiming the enantiomers. In particular, he rejected the submission that Sanofi is authority for the proposition that a claim for a racemate is ipso facto a claim for its two enantiomers (Reasons, para. 47).
[66] I can detect no error in this regard. Contrary to what is asserted, the Supreme Court in Sanofi did not hold that a claim for a racemate automatically includes a claim for its enantiomers. The conclusion in Sanofi that the genus patent also claimed the enantiomers is based on claims 1 and 14 thereof which specifically claimed the racemate and the two enantiomers (Sanofi, paras. 101 and 103).
[67] The appellants made the further argument that the decision of this Court in Pfizer shows that a patent claiming an enantiomer of a previously disclosed racemate can be viewed as a selection patent. No doubt that is so. However, this depends on the particularities of the patents in issue. In Pfizer, the patent labeled as a selection patent (the ‘546 patent) states that the enantiomer in question was among compounds previously claimed in the prior (i.e. genus) patent (Pfizer, para. 47). No such statement appears in the ‘452 patent.
[68] With respect to the decision of the Federal Court in Janssen-Ortho II, it is significant that Hughes J., who heard the infringement action, did not deal with the patent in issue in that case as a selection patent even though the relevant claim was for an enantiomer of a previously disclosed racemate. In so doing, he declined to follow the approach of Mosley J. in Janssen-Ortho I who treated the same patent as a selection patent in the context of an earlier NOC proceeding. However, Mosley J. held the patent to be a selection patent only after finding that the knowledge required to separate the two enantiomers was common to the person skilled in the art (Janssen-Ortho I, para. 53), a finding that was not shared by Hughes J. on the evidence in the infringement action (Janssen-Ortho II, para. 104).
[69] Apotex further submits that the Applications Judge erred in making the existence of a selection patent dependent on specific disclosure or claim of the selected compound in a prior patent. The decision of the Federal Court, Trial Division in Pfizer Canada Inc. v. Apotex Inc. (1997), 77 C.P.R. (3d) 547, page 556 is relied upon. With respect, the Applications Judge did no such thing. A selection patent, by definition, is directed at a compound which comes within those generally described and claimed in a prior patent. What the Applications Judge found is that escitalopram did not come within such a description because it was not amongst those previously described and claimed.
[70] In construing the claims of U.S. patent ‘193, the Applications Judge noted that it was known at the relevant time that some enantiomers are toxic and that escitalopram’s utility could not be ascertained without first resolving citalopram, which had yet to be done. After pointing out that one who overclaims stands to loose everything, he concluded that the skilled person would not have read the relevant claims as