Date: 19990423
Docket: T-2870-96
BETWEEN:
APOTEX INC.
Plaintiff
(Defendant by Counterclaim)
- and -
SYNTEX PHARMACEUTICALS INTERNATIONAL LIMITED
and HOFFMANN-LA ROCHE LIMITED
Defendants
(Plaintiffs by Counterclaim)
REASONS FOR JUDGMENT
REED, J.:
[1] The plaintiff seeks a declaration that Canadian Letters Patent No. 1,204,671 (the 3671 patent) is invalid and that the manufacture and sale of certain naproxen controlled release tablets made by the plaintiff do not infringe the patent. The defendants, in a counterclaim, seek a declaration that the 3671 patent has been infringed by the plaintiff.
[2] The patent carries the title "Controlled Release Naproxen and Sodium Naproxen Tablets". It is a composition or formulation patent, claiming an invention in a novel pharmaceutical composition, the main ingredients of which are naproxen, or naproxen sodium, and hydroxypropylmethylcellulose. Naproxen is a well known pain killer used, for example, in treating rheumatoid arthritis. Hydroxypropylmethylcellulose ("HPMC") is a hydrophilic gum, a water soluble cellulose polymer used, inter alia, for the preparation of sustained release pharmaceutical products.
Earlier Proceedings
[3] The present action is not the first proceeding between the parties concerning controlled release naproxen tablets. There has been extensive litigation arising under the Patented Medicines (Notice of Compliance Regulations).1 The overall scheme of those Regulations and their effect has now been set out in many decisions and I do not find it necessary to repeat the description here.2 For present purposes, the following summary suffices: (1) a patent owner, in this case Syntex, files a list of the patents it owns with the Minister of National Health and Welfare, identifying the drugs to which the patents relate; (2) when another person applies to the Minister for a Notice of Compliance to allow it to market its version of the drug, it must file a document (a Notice of Allegation) stating why the marketing of its drug will not infringe the relevant patent; (3) if the patent owner is not satisfied with that explanation it may commence an application for an order of prohibition to prevent the Minister from issuing a Notice of Compliance.
[4] In June of 1993, Apotex filed a Notice of Allegation (the 1993 NOA) with respect to an application it was making to the Minister for a Notice of Compliance for its sustained release naproxen tablets. That Notice contained two assertions. The first was that "no claim for the medicine itself and no claim for the use of the medicine" would be infringed by Apotex making and selling its naproxen controlled release tablets, because the 3671 patent, as a composition (formulation) patent, did not claim the medicine itself or the use of the medicine. The second assertion was an undertaking by Apotex that any tablets it produced and sold would not fall within the claims of the 3671 patent:
| ... Apotex Inc. hereby undertakes that any tablets produced and sold by Apotex will not fall within the scope of the claims of patent 1204671, so that no claim would be infringed. |
[5] In June of 1993, the Notice of Compliance Regulations had only recently been promulgated and the exact manner in which they would operate was not known. Among the unknowns were: (1) the meaning of "no claim for the medicine itself and no claim for the use of the medicine" in Regulations 4(2)(b) and 5(1)(b)(iv), and (2) the amount of factual detail that should be included in a Notice of Allegation.
[6] By the time Syntex's application for an order to prohibit the issuance of a Notice of Compliance to Apotex was heard on January 24, 1996, (file number T-1893-93) the jurisprudence had made it clear that composition (formulation) claims were claims for the medicine itself and therefore fell within the scope of the Regulations: Hoffmann-La Roche Ltd. v. Canada (Minister of National Health and Welfare) (1995), 62 C.P.R. (3d) 58 (F.C.T.D.), affirmed 67 C.P.R. (3d) 25 (F.C.A.). Thus, Apotex's first reason for asserting non-infringement in the 1993 NOA was no longer seriously in issue.
[7] It had also been decided that the procedure for obtaining an order of prohibition was to be summary in nature, by way of an application, not involving a trial, discovery, or the production of documents, and that a notice of allegation must contain sufficient factual information to justify the allegation of non-infringement that was made therein: Syntex (U.S.A.) Inc. v. Novopharm Ltd. (1996), 65 C.P.R. (3d) 499 (F.C.T.D.). Thus, Apotex's second reason for asserting non-infringement in its 1993 NOA was not likely to be accepted, since there was no factual basis given for it - it was merely an undertaking.
[8] Accordingly, when Syntex's application for an order of prohibition was heard, Apotex argued that the undertaking in the 1993 NOA should be interpreted as a factual statement that the Apotex tablets did not have the same weight percent of HPMC as that of the patented composition (4-9 weight percent). If this was not accepted, counsel argued that Apotex should be granted leave to adduce further factual information concerning the formulation of Apotex's tablets. These arguments were rejected on the basis that the undertaking was not a statement of fact and the request to adduce further evidence was made at too late a time (that is, only at the end of the argument). In addition, the Court commented that even if one were to interpret the undertaking as a factual statement as was being suggested, that is, as an assertion that the Apotex formulation did not contain 4-9 weight percent of HPMC, the allegation of non-infringement would remain unjustified since the weight percent identified in the patent was "about" 4-9 percent.
[9] On March 22, 1996, two days after an order of prohibition issued in the T-1893-93 proceeding, preventing the issuance of a Notice of Compliance to Apotex, Apotex filed and served a second Notice of Allegation (the 1996 NOA). This Notice also alleged that the 3671 patent would not be infringed by the making and selling of Apotex's tablets. It contained as a justification for that allegation a factual description of the formulation that Apotex had sought to adduce in evidence in T-1898-93. It also contained two further grounds for alleging non-infringement: (1) the use of the word "about" in the claims rendered the 3671 patent ambiguous, and therefore invalid for non-compliance with subsection 34(2) of the Patent Act ; (2) the patent was invalid because the invention claimed was not an inventive step but an obvious result derived from routine experimentation.
[10] The 1996 NOA gave rise to another application (file number T-998-96), this time by Syntex and Hoffmann-La Roche (hereinafter "Roche"),3 for an order of prohibition to prevent the Minister from issuing a Notice of Compliance in reliance on the 1996 NOA. Before that application could be heard on the merits, Roche filed, on August 28, 1996, a motion for an order declaring the 1996 NOA to be void and of no effect, or in the alternative, a stay of the proceedings until after the appeal respecting the 1993 NOA had been finally determined. The motion also sought an extension of the time limit during which the Minister is prohibited from issuing a Notice of Compliance, set out in section 7(1)(e) of the Regulations, corresponding to the length of any stay that might be granted. This motion was heard in the fall of 1996 together with several others (T-1712-95 and T-421-96), all relating to the validity of second notices of allegation.
[11] At about the same time, the Federal Court of Appeal upheld the trial division decision relating to the 1993 NOA, (file number A-264-96).4 The argument that the undertaking in the 1993 NOA could only mean that the tablets "would be composed in one or more respect outside the specific range of weight percentages" claimed in the patent was rejected, as was the contention that the undertaking constituted a factual statement.
[12] On December 23, 1996, Apotex filed and served the Statement of Claim in the present action. A decision was subsequently rendered on the motions that had been heard regarding the 1996 NOA.5 The Court held that it had no authority to declare notices of allegation void since those documents were not pleadings; they were administrative documents filed with the Minister.6 A stay of the T-998-96 proceedings, however, was granted. At the same time, an extension of the 30 month time period provided for by Regulation 7(1)(e) was not included in the order. It was assumed that the earlier order of prohibition (that granted in T-1898-93) would continue to operate to prevent the issuance of a notice of compliance for Apotex's controlled release naproxen tablets.
[13] An appeal of the decision refusing to declare the 1996 NOA void was filed by Roche. A cross-appeal of the stay that had been granted was filed by Apotex.
[14] The fact that no extension of the 30 month time period, prescribed by section 7(1)(e) of the Regulations, was included in the T-998-96 stay order apparently led Apotex, in the fall of 1998, to prompt the Minister to issue a Notice of Compliance allowing Apotex to market its controlled release tablets. The Minister did so in November 1998. Roche, in response, called the Minister's attention to the text of the Court's reasons in T-998-96, where it was stated that the order of prohibition that had issued in T-1898-93 would continue to prevent the issuance of a Notice of Compliance to Apotex. Consequently, the Minister informed the parties that he had decided that he had had no jurisdiction to issue the November 1998 Notice of Compliance, and he considered it to be void.
[15] This skirmish prompted Roche to bring an application for an order that the Notice of Compliance that had been issued in reliance on the 1996 NOA was void, and that the Minister had had no jurisdiction to issue a Notice of Compliance (file number T-2309-98). An order to that effect was granted by Mr. Justice Evans on April 7, 1999. In his reasons, Mr. Justice Evans noted that the 1993 NOA was not identical to the 1996 NOA since the patent's validity had been put in issue in the latter. At the same time, he noted that Apotex had chosen not to pursue its cross-appeal of the stay granted in T-998-96, and thus it should be taken to have accepted that decision. The appeal and cross-appeal of the T-998-96 decision (file number A-23-97) were ordered discontinued by the Chief Justice on March 25, 1999, as both parties had advised him, in response to a Notice of Status Review, that they were prepared to discontinue their respective appeals conditional on the other doing likewise.
[16] Therefore, there now exists an order of prohibition dated March 20, 1996 requiring the Minister not to issue a Notice of Compliance in reliance on the 1993 NOA, an order of the Court staying the proceedings in T-998-96 with respect to the 1996 NOA, and an order declaring that the Minister has no jurisdiction to issue a Notice of Compliance in reliance on the 1996 NOA. This last must, I think, be interpreted as meaning that jurisdiction does not exist as long as the proceeding in T-998-96 remains stayed.
Standing, Res Judicata
[17] In this context, Roche argues that at the date of the filing of the Statement of Claim in the present action, December 23, 1996, Apotex had no standing to commence such an action.
[18] Subsections 60(1) and (2) of the Patent Act provide:
| 60. (1) A patent or any claim in a patent may be declared invalid or void by the Federal Court at the instance of the Attorney General of Canada or at the instance of any interested person. |
| (2) Where any person has reasonable cause to believe that any process used or proposed to be used or any article made, used or sold or proposed to be made, used or sold by him might be alleged by any patentee to constitute an infringement of an exclusive property or privilege granted thereby, he may bring an action in the Federal Court against the patentee for a declaration that the process or article does not or would not constitute an infringement of the exclusive property or privilege. [Underlining added.] |
[19] It is argued that since an order of prohibition was in existence that prevented the Minister from issuing a Notice of Compliance to Apotex, Apotex could not have entered the market with its sustained release naproxen tablets. Accordingly, it is argued that Apotex, at that time, could not satisfy the requirements of subsection 60(2) and, thus, did not have sufficient interest to support an action for declarations that the 3671 patent was invalid and that the Apotex formulation was non-infringing.
[20] I am not persuaded that this argument can prevail. A person who has filed a submission for a Notice of Compliance seeking regulatory approval for a drug product, the issuance of which is being prevented because of a patent listed under section 4 of the Regulations, is an interested person for the purposes of subsection 60(1) of the Patent Act.
[21] That person is attempting to get into the market with its product but is blocked from doing so because of the assumption, implicit in the regime established by the Regulations, that the listed patent will be infringed by such action. The plaintiff in this case is a person who proposes to make, use or sell an article that not only "might be alleged by any patentee to constitute an infringement" (emphasis added) of its patent, as described in subsection 60(2), but where such an allegation has actually been made. That Apotex is an interested person for the purposes of section 60 is evidenced by the fact that the defendants responded to the within Statement of Claim by filing a Counterclaim. The defendants' argument that the plaintiff lacks standing cannot prevail.
[22] The defendants argue that the question whether Apotex's formulation infringes the 3671 patent is an issue that is res judicata as a result of the earlier proceedings. It is argued that in the T-1898-93 proceeding an allegation of non-infringement was made with respect to the same formulation that is in issue in this action, and the Court found the allegation not to be justified. It is argued that the plaintiff had the opportunity in that case to put the relevant evidence before the Court but did not do so, and it should not be allowed now to relitigate the issue.
[23] In addition, Mr. Justice Stone in A-264-93, the appeal affirming the decision in T-1898-93, when rendering the decision in that proceeding put forward an interpretation of the 3671 patent at pages 4-6:
| While the Cross-Appellant alleged and undertook that its tablets would not fall within the scope of the Patent, I cannot see that this could only mean that those tablets "would be composed in one or more respects outside the specific range of weight percentages" claimed in the Patent, as the Cross-Appellant contends. Those weight percentages appear in Claim 1 of the Patent which reads in part: |
| 1. A controlled release tablet for once-daily oral administration of about 500-1200 mg of naproxen or naproxen sodium, said tablet comprising a homogeneous matrix comprising: |
| about 4-9 weight percent of hydroxypropylmethylcellulose having a number average molecular weight in the range from about 80,000 to about 130,000, |
| about 81-96 weight percent of naproxen or naproxen sodium, |
| 0.1 to about 2 weight percent of a pharmaceutically acceptable lubricating agent, and |
| 0 to about 8 weight percent of other pharmaceutically acceptable excipients. |
| A definition of the word "about" appears in the disclosure at page 17, lines 13-14. It reads: |
| As used herein, the term "about", when used in reference to number average molecular weight, indicates that the actual number average molecular weight may vary up to 10% above or below the stated value. |
| As used herein, the term "about", when used with reference to the amount of naproxen, naproxen sodium, lubricating agent or other optionally included pharmaceutically acceptable excipients, incorporated in each tablet, is intended to indicate that the actual amount of that ingredient may vary up to 10% above or below the stated amounts. |
| However, this definition is limited to the use of the word "about" "in reference to number average molecular weight" and to the amount of the various identified ingredients to be incorporated in each tablet. The word is not defined in reference to the "weight percent of hydroxypropyl-methylcellulose". |
| Nor am I persuaded that the disclosure supplies a definition of the word "about" in the context in question. The Cross-Appellant relies on the following statement at page 11, line 29 to page 12, line 1 as amounting to such a definition: |
| The hydroxypropylmethylcellulose must have a number average molecular weight in the range of from about 80,000 to about 130,000, preferably from about 120,000 to about 130,000. When the polymer has a number average molecular weight of 120,000-130,000, it constitutes preferably about 4-6 weight percent of a naproxen controlled release tablet ... |
| I do not read this portion of the disclosure as eliminating the approximation of "weight percent" contemplated by the word "about" in Claim 1. The indication in the disclosure is that when the hydroxypropylmethylcellulose has a number average molecular weight of 120,000-130,000 "it constitutes preferably about 4-6 weight percent of the naproxen tablet". The claim, however, is not in itself limited to the 120,000-130,000 range but to a wider range ""from about 80,000 to about 130,000". |
[24] I will consider first Mr. Justice Stone's comments, I think it is only necessary to note that they do not differ from the evidence that has been given in this case. All the experts accept that a definition of the word 3about3 in relation to the 4-9 weight percent of HPMC is not provided for in the patent. It is the meaning of the word 3about3 in that context that is in issue. Even if Mr. Justice Stone's comments could be interpreted as creating some sort of res judicata , there is no repudiation of his comments in the present proceeding.
[25] With respect to the broader question, the effect of the plaintiff having had an opportunity, at an earlier time, to have had the question of infringement based on the formulation now in issue determined, the jurisprudence discussing the differences between an application for an order of prohibition under the Notice of Compliance Regulations and an action for patent infringement is relevant. The Federal Court of Appeal and the Supreme Court of Canada have made it very clear that a determination as to whether or not a Notice of Allegation is justified is a separate and different proceeding from a finding of infringement or invalidity in a patent action. In Merck Frosst Canada Inc. v. Canada (Minister of National Health and Welfare) (1994), 55 C.P.R. (3d) 302 at 320, the Federal Court of Appeal stated:
| Those proceedings, after all, are instituted by the patentee and seek a prohibition against the Minister; since they take the form of a summary application for judicial review, it is impossible to conceive of them giving rise to a counterclaim by the respondent seeking such a declaration. Patent invalidity, like patent infringement, cannot be litigated in this kind of proceeding. I can only think that the draftsperson had in mind the possibility of there being parallel proceedings instituted by the second person which might give rise to such a declaration and be binding on the parties. [Underlining added.] |
[26] In Pharmacia Inc. v. Canada (Minister of National Health and Welfare) (1994), 58 C.P.R. (3d) 209 at 217, Mr. Justice Strayer writing for the Federal Court of Appeal stated:
| It will be noted that the regulations nowhere create or abolish any rights of action between the parties; instead they confer a right on the patentee to bring an application for prohibition against the Minister of National Health and Welfare. That is, the regulations pertain to public law, not private rights of action. ... If the Governor in Council had intended by these regulations to provide for a final determination of the issues of validity or infringement, a determination which would be binding on all private parties and preclude future litigation of the same issues, it surely would have said so. This court is not prepared to accept that patentees and generic companies alike have been forced to make the sole assertion of their private rights through the summary procedure of a judicial review application. As the regulations direct that such issues as may be adjudicated at this time must be addressed through such a process, this is a fairly clear indication that these issues must be of a limited or preliminary nature. If a full trial of validity or infringement issues is required this can be obtained in the usual way by commencing an action. [Underlining added.] |
The Supreme Court recently approved this approach and adopted it in Eli Lilly & Co. v. Novopharm Ltd., [1998] 2 S.C.R. 129 at 184-185.
[27] Counsel for the defendants argues that the comments in these decisions have the character of obiter dicta because those cases did not deal directly with a situation in which an issue had been litigated in one proceeding, and then was sought to be relitigated in another. He argues that there are good policy reasons why that should not be allowed to occur - it is wasteful of time and resources, including judicial resources, and it can potentially lead to conflicting decisions being rendered by different judges on the same facts.
[28] Counsel for the plaintiff argues that there are good reasons why the two procedures have been held to be separate and independent proceedings. For example, the notice of allegation is not a pleading, so it cannot be amended after it has been filed, even if additional facts or prior art relevant to the matter are discovered.7 Also, there is no opportunity of discovery or requirement for document production by the parties. While an application can be turned into an action and oral evidence can be heard, this can only happen with leave of the Court.
[29] As I understand the jurisprudence, one of the reasons the Federal Court of Appeal has held that the two proceedings are separate and distinct is because an application for an order of prohibition pursuant to the Regulations is a summary proceeding. Also, the issue in the two proceedings is different, since in the one case it is the justification of an allegation made to the Minister that is the subject of the proceeding, while the other determines private rights as between the parties. I do not consider the comments in the above jurisprudence to be merely obiter dicta, and I note that the law does not always endorse a policy of no duplication of proceedings. There are many circumstances in which hearings or trials de novo are allowed, or where an administrative decision-making process and private litigation deal with the same underlying factual situation without a determination in the former precluding an independent determination in the latter.
[30] In addition, in the present case, there has never been a judicial determination as to whether or not Apotex's formulation infringes the 3671 patent. There was no evidence of the Apotex formulation considered by the Court in the T-1898-93 proceeding. As noted, what was in issue was an undertaking that counsel was seeking to convert into a statement of fact concerning the content of the formulation. While there was evidence of the Apotex formulation before the Court in the T-998-96 proceeding, no decision on the merits was ever made. There has never been a consideration of the patent's validity.
[31] In light of the above, I cannot accept the defendants' argument that the issue of non-infringement is res judicata.
Evidence
[32] Before dealing with the merits of this action, I will set out some comments on the evidence. As is usual in these cases, experts appeared before the Court taking diametrically opposite positions with respect to the conclusions that should be drawn from the evidence. The plaintiff called Dr. Langer and Dr. Robinson. The defendant called Dr. Banker. It is this evidence together with that given by Dr. Sherman, the Chairman and Chief Executive Officer of Apotex, that the Court must evaluate. I do not propose to refer to all of the evidence but will set out below my conclusions based on the evidence that I found to be most credible, reliable, and pertinent. I have not ignored the evidence that I do not specifically mention. I will, however, make one general observation. It is difficult for expert witnesses to remain objective. Some appear to find it easier to do so than others. I must note that among the expert witnesses in this case, Dr. Banker had the most difficulty maintaining an objective stance. He often avoided answering the questions put to him by making long speeches that were not directly responsive to the questions he had been asked. When compared to the evidence of the other experts, his evidence has less credibility.
Interpretation of the Patent
(1) The Invention as Described in the Patent
[33] The summary of the invention found in the patent reads:
| The present invention is a controlled release tablet for once-daily oral administration of 500-1200 mg of naproxen or naproxen sodium which is formed from a matrix comprising: |
| about 4-9 weight percent of hydroxypropyl-methylcellulose having a number average molecular weight in the range of from about 80,000 to about 130,000, |
| about 81-96 weight percent of naproxen or naproxen sodium, |
| 0.1 to about 2 weight percent of a pharmaceutically acceptable lubricating agent, and |
| 0 to about 8 weight percent of other pharmaceutically acceptable excipients. [Underlining added.] |
[34] The patent states that the inventors developed a new controlled release oral dosage tablet form of naproxen or naproxen sodium, which provides sustained therapeutic plasma drug levels for at least 24 hours, requiring a small amount of HPMC (4-9 weight percent), making possible a once-daily naproxen or naproxen sodium dosage form tablet without excessive bulk.8
(2) The Relevant Law
[35] The law is not in dispute.9 The construction of a patent is a question for the court, reading the patent through the eyes of a person skilled in the art to which the patent relates. This may require evidence concerning the meaning of technical terms and the state of the common general knowledge of those skilled in the art. Construction of a patent precedes a consideration of its validity or infringement and the specification must be read as a whole. The Court should strive to give a purposive interpretation to the claims, "being neither benevolent nor harsh, but rather seeking a construction which is reasonable and fair to both patentee and public."10 The starting point for patent construction is the claims, since they alone define the monopoly that has been granted to the patentee.
[36] There are two quotations from the jurisprudence that I think are particularly relevant to the construction of the patent presently before the Court. One is the words of Mr. Justice Pratte in Eli Lilly & Co. v. O'Hara Manufacturing Ltd. (1989), 26 C.P.R. (3d) 1 at 7 (F.C.A.):
| ... A court must interpret the claims: it cannot redraft them. When an inventor has clearly stated in the claims that he considered a requirement as essential to his invention, a court cannot decide otherwise for the sole reason that he was mistaken. |
[37] The other is the comments of the Court in Catnic Components Ltd. v. Hill & Smith Ltd. (1980), 7 F.S.R. 60 at 65-6, quoted with approval in Beecham Can. Ltd. v. Procter & Gamble Co. (1982), 61 C.P.R. (2d) 1 at 10 (F.C.A.):
| The question in each case is: whether persons with practical knowledge and experience of the kind of work in which the invention was intended to be used, would understand that strict compliance with a particular descriptive word or phrase appearing in a claim was intended by the patentee to be an essential requirement of the invention ... [Underlining added.] |
(3) A Person Skilled in the Art
[38] The claims are to be construed from the perspective of the average person skilled in the art to which the patent relates. The average person skilled in the particular art has been described as being a skilled addressee of the patent specification:
| For it is to the skilled addressee " a skilled man reasonably well versed in the art " that the specification is deemed to be addressed, and it is by the standards of the common general knowledge of such a man that one tests whether the invention was obvious or not.11 |
[39] I adopt also the comments of John Bochnovic in his article, "Invention/Inventive Step/Obviousness", in G.F. Henderson, ed., Patent Law of Canada (1944) at 47-48:
| Care must be taken not to underestimate the ability of the unimaginative skilled technician who has become the standard bearer of the question applied in Canada today. This artificial creature must be adapted to the technology and background of individual patents and individual cases. The technician today is more likely to be a scientist or engineer, particularly where sophisticated technology is involved. The suggestion that the skilled technician should be unimaginative should not strip that technician of the ability to pursue reasonable and logical inquiries. |
[40] A person skilled in the art of the 3671 patent would be a person who has had three to five years experience working in the pharmaceutical field as a drug formulator, having come to that employment with a first level science degree. Most people working in the field have a Bachelor of Pharmacy degree, but other types of scientific academic training, such as a B.Sc. in chemistry, chemical engineering or biology could also provide the requisite academic background.
(4) What Has Been Claimed ?
[41] I turn then to claim 1. All the other claims are dependent on claim 1. That claim asserts a monopoly in:
| 1. A controlled release tablet for once-daily oral administration of about 500 - 1200 mg of naproxen or naproxen sodium, said tablet comprising a homogenous matrix comprising: |
| about 4 - 9 weight percent of hydroxypropylmethylcellulose having a number average molecular weight in the range of from about 80,000 to about 130,000, |
| about 81 - 96 weight percent of naproxen or naproxen sodium, |
| 0.1 to about 2 weight percent of a pharmaceutically acceptable lubricating agent, and |
| 0 to about 8 weight percent of other pharmaceutically acceptable excipients. |
[42] The claim appears quite clear. Two aspects, however, are important for the present litigation: one is the meaning of the phrase "about 4-9" weight percent as used in reference to HPMC; the other is the importance or lack thereof of the number average molecular weight of the HPMC.
| a) "About" |
[43] The word "about" is defined in the disclosure as " 10% when used in reference to number average molecular weight, and the amount of naproxen, naproxen sodium, lubricating agent or excipients.12 A defined meaning is not given for the word 3about3 when it is used in the phrase "about 4-9 weight percent" of HPMC. The plaintiff's experts opined that in that context the word should be interpreted as also meaning " 10%. This would then encompass a weight percentage range for the HPMC of 3.6 to 9.9. The defendants' expert, Dr. Banker, suggested that the word 3about3 signified a much wider range, a range that would include weight percentages as low as 3. At the same time, he also interpreted the word 3about3 as applying only to the first number of the range, that is to the number 4 but not the number 9.
[44] The unconvincing nature of Dr. Banker's opinion appears immediately on the articulation of his reasoning. He states that the number 4 would be understood by those skilled in the art to encompass the range 3.5 to 4.5, and, that it is to this range that the word "about" should be applied. By doing so he then concludes that "about 4" includes 3 and 5. The evidence does not establish that those skilled in the art would apply such reasoning or draw such a conclusion. What is more, there is nothing in the patent to indicate that whole numbers are not intended to be read and understood as whole numbers. When the patentee thought it necessary to express amounts using decimals, it did so, for example, in claim 1 in describing the amount of lubricating agent that would be used.
[45] The word 3about3 carries a dictionary definition of "near" or "close". Absent a specific definition to the contrary, the ordinary meaning of the word applies. This would allow, in my view, a flexibility no greater than half way to the next number. In the case of the 4-9 weight percent range for HPMC, I read the approximation allowed by the word "about" as extending no more widely than from 3.5 to 9.5. A percentage below 3.5 or above 9.5 is not near or close to 4 or 9, respectively, but to 3 and 10. One would expect that the patentee would have described the permissible range as from about 3-10 had a range closer to 3 than 4 and closer to 10 than 9 been intended.
[46] Counsel for the plaintiff argues that it is possible that what was meant by "about 4-9 weight percent" is "somewhere within the 4-9 range". There is some merit in this position but I am not persuaded that it can prevail given the clear indication that the word "about" is not used in the rest of the patent in that manner when it is used to describe a range of numerical values.
[47] With respect to the opinion that the word 3about3 only modifies the lower number of the range, this is based on the different usages found in the patent. For example, when describing the weight percentage of HPMC the phrase "about 4-9 weight percent . . ." is used, but when describing the number average molecular weight the phrase "about 180,000 to about 130,000" is used. It is argued that in the former case only the lower number of the range is modified by the word "about", while in the latter both the upper and the lower numbers are so modified. I do not accept that such a distinction can be drawn. The different usage signifies nothing more than two different stylistic ways of expressing the same thought. For example, while claim 1 expresses the molecular weight range of HPMC as being "from about 80,000 to about 130,000", claim 3 describes the range with which it is concerned as "about 120,000-130,000". Claim 4 refers to the amount of colorant as "0 to about 1 weight percent ...", while claim 6 speaks of "about 0.01-0.05 weight percent ..." On page 10, line 35, of the patent, the amount of naproxen or naproxen sodium to be incorporated into a tablet is described as "between about 500 and about 1200 mg", in claim 1 it is described as "about 500-1200 mg". A difference in meaning is not intended by these different modes of expression.
b) Number Average Molecular Weight - Essential ?
[48] The defendants assert that the description in claim 1 of the HPMC, as "having a number average molecular weight in the range of from about 80,000 to about 130,000" is an unessential part of the claim. Indeed, it is argued that all that is necessary to obtain the functional result arising from the invention is the teaching found in the first four lines:
| 1. A controlled release tablet for once-daily oral administration of about 500 - 1200 mg of naproxen or naproxen sodium, said tablet comprising a homogenous matrix comprising: |
| about 4 - 9 weight percent of hydroxypropylmethylcellulose ... |
[49] A straight forward reading of claim 1, as well as the text of the other claims, and the descriptive part of the specification (the disclosure) do not support this conclusion. For example, claims 11 and 20 are addressed to specific molecular weight ranges, and would make no sense if number average molecular weight was considered by the patentee to be a non-essential part of the claims:
| 11. A controlled release tablet of claim 1 wherein the hydroxypropyl-methylcellulose has a number average molecular weight of about 120,000-130,000.13 |
. . .
| 20. A controlled release tablet of claim 1 wherein the hydroxypropyl-methylcellulose has a number average molecular weight of about 85,000-95,000.14 |
[50] In the descriptive part of the specification, at page 11, line 30, the patentee states:
| The hydroxypropylmethylcellulose must have a number average molecular weight in the range of about 80,000 to about 130,000 ... [Emphasis added]. |
[51] The defendant's expert, Dr. Banker, suggests that the disclosure makes it clear that a much wider range than 80,000-130,000 is intended to be claimed by the patentee. Dr. Banker reads the patent as claiming a preferred embodiment in the 80,000-130,000 range. He does not read the claims, and particularly claim 1, as being defined by that range. The text of the disclosure on which he relies reads:
| The hydroxypropylmethylcellulose utilized in the present invention is a water soluble cellulose ether, and is commercially available in various grades under the tradenames mentioned above in the BACKGROUND OF THE INVENTION. The physicochemical properties of these polymers vary over a wide range. Preferred embodiments of this invention utilize premium grade polymers of a single viscosity type having number average molecular weights in the range of about 80,000-130,000. |
| The number average molecular weight of the hydroxypropylmethylcellulose which is used in the tablet matrix substantially influences the release profile which is obtained. The number average molecular weight (mn) is the sum of the individual molecular weights of a representative sample population of molecules divided by the number of molecules in that sample, and is calculated from the limiting osmotic pressure of the solvent as the concentration of the hydroxypropylmethylcellulose approaches zero. The hydroxypropylmethylcellulose must have a number average molecular weight in the range of from about 80,000 to about 130,000, preferably from about 120,000 to about 130,000 when the polymer has a number average molecular weight of 120,000-130,000, it constitutes preferably about 4-6 weight percent of a naproxen controlled release tablet, or about 6-8 weight percent of a naproxen sodium controlled release tablet. A second preferred rate of number average molecular weight is about 85,000 to about 95,000. When the polymer has a number average molecular within this range, it constitutes preferably about 7-9 weight percent of the controlled release naproxen or naproxen sodium tablet. [Underlining added.] |
[52] One is entitled, when a patent is written in English, to read the patent in the light of the ordinary principles of English composition. From the opening sentences of the two paragraphs set out above, a reader would expect the subject matter of the first paragraph to be "grades" of HPMC, and that of the second to be the "number average molecular weight" of the HPMC. Thus, the most reasonable reading of the paragraphs is that the preferred embodiment referred to in the first paragraph is one in which the HPMC is a "premium grade . . . of a single viscosity type." The second paragraph continues the description of the preferred embodiment by reference to the number average molecular weight ranges of the HPMC to be chosen (120,000-130,000 and 85,000-95,000). Those number average molecular weight ranges correspond to two commercially available HPMCs, sold by the Dow Chemical Co. (Methocel K4M Premium and Methocel K15M Premium). Thus, the text of the specification does not indicate that the 80,000-130,000 number average molecular weight range requirement (as expanded by the " 10% definition to 72,000-143,000) is a preferred embodiment. Rather, the text supports a reading of claim 1 that requires an HPMC having a number average molecular weight between 80,000 and 130,000 (" 10%).
[53] The range of molecular weight HPMCs available at the relevant time (January 1, 1983) extended beyond the 72,000-143,000 range. The patent itself discloses this on page 2:
| Hydroxypropylmethylcelluloses are commercially available in various grades, under several tradenames ... Commercial designations of the various hydroxypropylmethylcelluloses reflect their individual viscosity types ... The viscosities range from 15 cps to 30,000 cps and represent number average molecular weights of from about 10,000 to over 150,000. Each of the various grades under a given tradename is a hydroxypropylmethylcellulose of a single viscosity type, e.g. 50 cps, 100 cps, 4,000cps, 15,000 cps etc. [Underlining added.] |
[54] A Dow Chemical Co. handbook of 1974 shows a Methocel K product with a viscosity of 19,000 and a number average molecular weight of 144,000 (calculated in accordance with instructions given in the handbook from a table set out for another product, Methocel A). In 1982, Dow was advertising six grades of HPMC, one carrying the designation XD-30018.00 (K-100 M Premium), having a viscosity of 100,000 cps.
[55] There is a relationship between number average molecular weight and viscosity such that the higher the former, the greater the latter. The viscosity of HPMC is important for the sustaining function it plays in pharmaceutical preparations because on ingestion the HPMC operates to form a pseudo-gel layer around the tablet. The greater the viscosity (which as noted is related to a higher number average molecular weight), the less HPMC is required to perform the sustaining function.
[56] Dr. Banker suggests that what now carries the tradename Methocel K100M was not included in the claim because at the relevant date it was an experimental product, it is identified as such in the Dow brochure. He also notes that the USP standards allow the various grades of HPMC to vary within a molecular weight range and still be identified by the relevant grade designation. He gave evidence that the determination of number average molecular weight varies depending upon the test method used and that the weight distribution of various grades overlap.
[57] Despite the fact that K100M was experimental in 1982, the evidence does not establish that it was not suitable for human consumption. The evidence also does not establish that it was left out because it was unknown at the relevant time. The fact that the USP standard allows for some variability in the average number molecular weight of HPMC identified under a given designation does not mean that the patentee intended to include all grades as coming within the invention claimed. Also, the variability that is allowed by those standards does not mean that someone relying on the patent is absolved from the requirement of checking the number average molecular weight of the product he or she is using to see if it comes within the patent or not. The fact that different test methods for determining number average molecular weight give different results is not relevant since the patentee described the method that was being relied upon (calculation from osmotic pressure measurements) and this is the same method used by Dow and is the method the plaintiff relies upon for assessing the number average molecular weight of the HPMC it uses. The overlapping weight distribution is also not significant for present purposes because one is talking about number average molecular weights.
c) Conclusion
[58] The conclusion that arises is that the invention claimed by the patentee contains as an essential element the requirement that the HPMC have a number average molecular weight in the 80,000-130,000 (" 10%) range and that the percentage weight ratio of HPMC to the tablet as a whole fall within a range that at the lower end is closer to 4% than 3%, and at the upper end is closer to 9% than to 10%. While tablets composed of a higher number average molecular weight and comprising a lower weight percent of the tablet may be functionally equivalent to what has been claimed, the patentee, for whatever reason, chose to so limit its claim. To construe claim 1 as the defendants suggest would be to rewrite the claim.
Non-Infringement
[59] Apotex's Statement of Claim described its formulation of naproxen controlled release tablets as comprising less than 3.6% by weight HPMC and as using an HPMC that had a number average molecular weight exceeding 143,000. On August 26, 1997, Apotex, in response to a Court order, disclosed more precise information concerning its formulation. It disclosed that its formulation contains 3.08 weight percent of HPMC and that the HPMC has a number average molecular weight of 190,000. At trial the latter number was corrected; the HPMC Apotex is using has a number average molecular weight of 245,000-250,000.
[60] Apotex uses Dow's K100M product. The K100M product is listed in the USP as having a number average molecular weight of from about 220,000 to 270,000. The plaintiff's experts determined the number average molecular weight of the product purchased by Apotex from Dow to be 250,000. Apotex uses 24 mg of the K100M per 780 mg tablet of naproxen. Thus, the weight percentage of the former to the latter is 3.08 percent.
[61] Apotex is entitled to a declaration that its formulation does not infringe the 3671 patent.
Validity - Obviousness
[62] For a patent to be valid, the claimed invention must be new, useful and inventive. These are three different characteristics. The theoretical bargain between the patentee and the state, which leads to the issuance of a patent, is that the patentee by his or her disclosure is adding something of a substantial character to the existing stock of human knowledge. Thus, if the claimed invention is not inventive, that is, if it was obvious in light of the prior art and the common general knowledge of those skilled in the art at the relevant time, the patent is not valid.15 The test for finding that a patented invention lacks inventive ingenuity is a very difficult test to satisfy since a "scintilla of invention" will do.16 The test is often expressed as being whether the unimaginative person skilled in the art would "have come directly and without difficulty to the solution taught by the patent."17 One starts with the statutory presumption of validity arising from what is now section 43 of the Patent Act, the approach to be taken as explained in Diversified Products Corp. v. Tye-Sil Corp. (1991), 35 C.P.R. (3d) 350 at 357-359 (F.C.A.).
[63] At the relevant date, HPMC was commercially available in various grades and from several sources. This is disclosed in the patent. As noted, one source was the Dow Chemical Company, U.S.A. In 1982, Dow published an advertising brochure entitled Formulating Sustained Release Pharmaceutical Products with METHOCEL. This brochure is referred to in the patent as part of the prior art. METHOCEL is the tradename under which Dow sold and sells a number of cellulose ethers of which HPMC is one. The brochure, after some introductory comments, contains a section entitled "Why Sustained Release Pharmaceuticals?" The same advantages of such products as the patent describes are set out therein. The brochure continues with descriptions under the following four headings: Types of Oral Controlled Release Dosage Forms, Formulation with Methocel, Other Factors, and References (i.e., literature cited). One of the types of oral dosage forms described is matrix systems. Interestingly, both the patent and the brochure identify an important aspect of making tablets by this method in very similar terms. The brochure reads:
| The matrix may be tableted by direct compression, with compression granulations, or with conventional wet granulations. The important factor is that the METHOCEL product must be uniformly dispersed throughout the matrix to achieve uniform sustained action. [Underlining added.] |
The patent, at page 10, lines 24-27 states:
| An important aspect of the present invention is the fact that the hydroxypropylmethylcellulose is uniformly dispersed throughout the matrix to achieve uniform drug release. The matrix may be made by any pharmaceutically acceptable technique which achieves uniform blending, including dry blending, conventional wet granulation, compression granulation ... [Underlining added.] |
[64] The important feature of the brochure, for present purposes, however, is that it describes the mechanism by which sustained release is achieved using HPMC, and it teaches a reader how to formulate sustained release tablets using HPMC in a matrix. It teaches the use of various concentrations or viscosity grades depending upon the dissolution rate desired for a drug. It teaches that varying the amount of HPMC used varies the release rate of the drug. It teaches that varying the number average molecular weight (and thus the viscosity) of the HPMC controls the release rate of the drug (examples are found at pages 5, 8 and 10 of the brochure). HPMC at the relevant time was one of the most popular and widely used mechanisms to control the release of the active ingredient in tablet formulations.
[65] In addition, there was a great deal of information available at the time concerning naproxen. An article written in 1972 by Runkel, Kraft, Boost et al., entitled Naproxen Oral Absorption Characteristics18 was put in evidence. An article written in 1979 by Brogden, Heel, Speight and Avery entitled Naproxen up to Date: A Review of its Pharmacological Properties and Therapeutic Efficacy and Use in Rheumatic Diseases and Pain States19 was put in evidence. These articles describe, amongst other factors, the half life of naproxen. The Runkel article reported that the drug had a plasma half-life of approximately 14 hours. The Brogden article states that "Naproxen's half-life of 12 to 15 hours makes it suitable for twice, or possibly once, daily adminstration." The pharmacokinetic and physiochemical properties of naproxen were well known.
[66] Drs. Langer and Robinson described the type of experimentation that is part of the routine work of a drug formulator. Dr. Robinson explained that a skilled formulator in making a sustained release tablet would typically begin with 10 per cent of the polymer that had been chosen for the prospective formulation and then adjust the weight percent or the grade of the polymer (number average molecular weight being related to viscosity) in order to develop the desired formulation.
[67] Counsel for the defendants refers to the statement in the patent that in vitro results do not necessarily reflect the results one will get in vivo. Dr. Banker also gave evidence in this regard. Dr. Langer's evidence is more persuasive. He noted that while it is true to say one could not predict in vivo results exactly, a formulator having knowledge of the properties of HPMC and the pharmacology of naproxen would have a fairly reasonable expectation about what would occur.
[68] Dr. Banker's position was that the prior art teaches away from using HPMC with naproxen, because a daily dose requires 750 mg of the drug and the 1982 Dow brochure suggests that a formulator start with 10% of the polymer when deciding how much Methocel to use. He states that the large size of the tablet that would result from adding 10% to 750 mg (for a total of 825 mg) would lead formulators not to try the combination. The patent speaks of the amount of HPMC that is used as being "surprisingly small". Dr. Banker gave evidence concerning the general practice in the industry regarding when development work would be done with respect to various formulation and when it would not. The text of the Dow brochure that allegedly teaches away from using HPMC to make a once-daily dosage of naproxen reads:
| Starting at the 10% level with METHOCEL K15M Premium for both soluble or insoluble drugs is suggested. From this point, slower dissolution can be achieved by changing (increasing) the level of METHOCEL or by using a higher molecular weight product like Experimental Polymer XD-30018.00 (K100 M Premium). |
[69] Dr. Banker's evidence of industry practice is general and speculative. There is a lack of specific evidence concerning the circumstances under which the formulation claimed in the patent was developed. Representatives of the defendants were not called as witnesses. The inventors were not called. I do not interpret the evidence as containing any convincing reason why a formulator would not have used HPMC to develop the formulation in question. The literature indicated that a once-daily dosage was in some cases effective, without any sustaining release composition. In addition, I cannot conclude that the Dow brochure teaches away from using less than 10%. It merely suggests 10% as a starting point. The evidence does not establish that there would have been a lack of motivation to use HPMC in combination with naproxen to make a controlled release dosage form because of the size of the resulting tablet. It does establish that size is always a consideration in tablet or capsule formulation because the product has to be swallowed.
[70] Evidence that there were few sustained release matrix type products on the market at the time is not evidence that carries much weight. All of the prior art is relevant to the question of obviousness, not merely which products have actually been brought to market.
[71] I conclude that a person skilled in the art on the basis of the prior art and the existing common knowledge of the time would directly and without difficulty arrive at the formulation claimed in the 3671 patent. There was no inventive step in the formulation. The patent is therefore invalid.
Claims Respecting Naproxen Sodium
[72] One last consideration must be addressed. The defendants argue that even if the claims of the patent that relate to naproxen are invalid, those claiming an invention with respect to naproxen sodium remain valid.
[73] Counsel for the defendants refers to section 58 of the Patent Act:
58. When, in any action or proceeding respecting a patent that contains two or more claims, one or more of those claims is or are held to be valid but another or others is or are held to be invalid or void, effect shall be given to the patent as if it contained only the valid claim or claims.
58. Lorsque, dans une action ou procédure relative à un brevet qui renferme deux ou plusieurs revendications, une ou plusieurs de ces revendications sont tenues pour valides, mais qu'une autre ou d'autres sont tenues pour invalides ou nulles, il est donné effet au brevet tout comme s'il ne renfermait que la ou les revendications valides.
[74] Claim 25 of the patent reads:
| The controlled release tablet of claim 1 which contains naproxen sodium. |
[75] In the present case, the claims relate to one invention. They are intertwined; they cannot be separated insofar as the finding of obviousness set out above is concerned. I have not been persuaded that section 58 is relevant to the present decision.
Conclusion
[76] For the reasons set out above, the plaintiff is entitled to a declaration that the 3671 patent is invalid, and that, in any event, the Apotex formulation that has been put in issue in this action would not infringe that patent.
Judge
OTTAWA, ONTARIO
April 23, 1999
__________________1 SOR/93-133, promulgated March 12, 1993.
2 See, for example, Merck Frosst Canada Inc. v. Canada (Minister of National Health and Welfare) (file T-1502-96, December 18, 1998).
3 The two companies had by this date amalgamated.
4 Hoffmann-La Roche Ltd. v. Canada (Minister of National Health and Welfare)(1996), 70 C.P.R. (3d) 1.
5 AB Hassle v. Minister of National Health & Welfare (1997), 71 C.P.R. (3d) 129 (F.C.T.D.).
6 Ibid at 138-139 quoting from Pharmacia Inc. v. Canada (Minister of National Health and Welfare)(1994), 58 C.P.R. (3d) 204 at 209 (F.C.A.).
7 Counsel refers to Bayer AG v. Apotex Inc. (1998), 84 C.P.R. (3d) 23 (F.C.T.D.), as an example of this situation.
8 The exact text of the patent is:
The present invention is directed to a new controlled release oral dosage formulation for naproxen or naproxen sodium which provides sustained therapeutic plasma drug levels for at least 24 hours, and requires a surprisingly small amount, 4-9 weight percent, of hydroxypropyl-methylcellulose. The low level of matrix material required by the present invention makes possible a once-daily naproxen or naproxen sodium dosage form without excessive bulk, having weight and size characteristics which make it well-adapted for practical and acceptable patient administration.
9 A useful summary is found in W.L.Hayhurst, "The Art of Claiming and Reading a Claim", in G.F. Henderson, ed., Patent Law of Canada (1994) at 186-199.
10 Consolboard Inc. v. MacMillan Bloedel (Saskatchewan) Ltd. (1981), 56 C.P.R. (2d) 145 at 157 (S.C.C.).
11 General Tire & Rubber Co. v. Firestone Tyre and Rubber Co., [1972] R.P.C. 457 at 482 (C.A.).
An eloquent description by Mr. Justice Hugessen is found in Beloit Canada Ltd. v. Valmet Oy (1986), 8 C.P.R. (3d) 289 at 294.
12 The exact text of the patent is:
As used therein, the term 3about3, when used in reference to number average molecular weight, indicates that the actual number average molecular weight may vary up to 10% above or below the stated value.
As used therein, the term 3about3, when used with reference to the amount of naproxen, naproxen sodium, lubricating agent or other optionally included pharmaceutically acceptable excipients, incorporated in each tablet, is intended to indicate that the actual amount of that ingredient may vary up to 10% above or below the stated amounts.
13 Claim 12, dependent thereon, reads:
12. The controlled release tablet of claim 11 which comprises:
about 5 weight percent of the hydroxypropyl-methylcellulose,
about 85-96 weight percent naproxen or naproxen sodium,
0.1 to about 2 weight percent lubricating agent, and
0 to about 8 weight percent other pharmaceutically acceptable excipients.
14 Claim 21, dependent thereon, reads:
21. The controlled release tablet of claim 20 which comprises:
about 7-9 weight percent of the hydroxypropyl-methylcellulose,
about 81-93 weight percent naproxen or naproxen sodium,
0.1 to about 2 weight percent lubricating agent, and contd...
... contd
0 to about 8 weight percent other pharmaceutically acceptable excipients.
15 Beecham Canada Ltd. v. Proctor & Gamble Co. (1982) 61 C.P.R. (2d) 1 at 27-28 (F.C.A.); Beloit Canada Ltd. v. Valmet Oy (1986), 8 C.P.R. (3d) 289 at 294 (F.C.A.).
16 Diversified Products Corp. v. Tye-Sil Corp. (1991), 35 C.P.R. (3d) 350 at 366 (F.C.A.).
17 Ibid; see also Beloit Canada Ltd. v. Valmet Oy (1986), 8 C.P.R. (3d) 289 at 294 (F.C.A.).
18 (1972) 7 Chem. Pharm. Bull. 1457.
19 Evaluations on New Drugs (ADIS Press Australasia Pty Ltd., 1979).