Bayer AG v. Apotex Inc., 2004 FC 177

                                                                                                                                Date: 20040202

                                                                                                                             Docket: T-877-02

                                                                                                                     Citation: 2004 FC 177

BETWEEN:

                                                 BAYER AG AND BAYER INC.

                                                                                                                                          Applicants

                                                                          and

                                                                APOTEX INC.

                                                                          and

                        THE MINISTER OF NATIONAL HEALTH AND WELFARE

                                                                                                                                       Respondents

                                                    REASONS FOR ORDER

CAMPBELL J.

[1]                 Since 1987, the Applicant Bayer AG has held Canadian Patent 1,218,067 ("the Patent" or "the '067 patent") for the antibacterial medicine ciprofloxacin hydrochloride ("ciprofloxacin") when made by the processes claimed in the Patent or by their obvious chemical equivalents. Ciprofloxacin is sold by the Applicant Bayer Inc. (formerly Miles Canada Inc.) in Canada under a Notice of Compliance (NOC) granted by the Minister of National Health and Welfare (the "Minister") under s.C.08.004 of the Food and Drug Regulations, C.R.C. 1985, c. 870 ("Food and Drug Regulations"). The '067 patent was issued February 17, 1987, on an application filed in August 1982, and expires February 17, 2004. Since 1993, the Respondent Apotex Inc. ("Apotex") has filed several Notices of Allegation (NOA) claiming invalidity or non-infringement of the Patent. The NOA at issue in the present case is Apotex' eighth NOA with respect to ciprofloxacin, and its second non-infringement allegation regarding the '067 patent. Apotex' eighth NOA alleges that no claim for the medicine itself and no claim for use of the medicine would be infringed by the manufacture or sale of ciprofloxacin when prepared by Apotex' proposed process.

[2]                 As a preliminary matter, the Applicants Bayer AG and Bayer Inc. ("Bayer") seek an order declaring that Apotex' NOA is invalid for non-compliance with s.5 of the Patent Medicines (Notice of Compliance) Regulations, SOR/93-133 as amended (the "NOC Regulations") on the basis that Apotex failed to file a submission for a NOC with respect to the Apotex process for ciprofloxacin prior to serving its non-infringement NOA on Bayer. Bayer also argues that Apotex' NOA does not "relate to" its new drug submission and, as a result, Apotex has failed to comply with s.5(3)(c)(i). Bayer therefore asserts that the Court does not have jurisdiction to render a decision on the merits of Apotex' non-infringement NOA.

[3]                 In the alternative, Bayer seeks an order pursuant to s.6(1) of the NOC Regulations prohibiting the Minister from issuing a NOC under s.C.08.004 of the Food and Drug Regulations to Apotex in respect of ciprofloxacin tablets of 100 mg, 250 mg, 500 mg, and 750 mg dosage strengths for oral administration, until after the expiration of the '067 patent. In these proceedings, Bayer seeks to demonstrate that step 2 of the six-step Apotex process for producing ciprofloxacin is an obvious chemical equivalent of the process claimed in claims 8(1) and 14(10) of the '067 patent and that Apotex' allegation of non-infringement is, therefore, not justified.

[4]                 The legislative scheme and the detailed procedures relevant to these proceedings have been explained in many cases and need not be repeated here (see for example Hoffman LaRoche v. Canada (1996), 67 C.P.R. (3d) 484 (F.C.T.D.), affd (1996), 70 C.P.R. (3d) 206 (F.C.A.)). It is important to note, however, that the burden of proof rests with Bayer to demonstrate on a balance of probabilities that Apotex' allegation of non-infringement is not justified. In order to discharge that burden, Bayer must disprove some or all of the allegations in the NOA, which are presumed to be true, and which, if left unchallenged, would allow the Minister to issue a NOC.

[5]                 The two main questions to be answered in the present application are as follows:

I. The jurisdictional question: Does the Court lack jurisdiction to render a decision on the merits on the basis that Apotex failed to comply with s.5 of the NOC Regulations?

II. The non-infringement question: Is step 2 of the six-step Apotex process for producing ciprofloxacin an obvious chemical equivalent of the process claimed in claims 8(1) and 14(10) of the '067 patent? If the answer to this question is "no", Apotex' allegation of non-infringement is justified.

I. The Jurisdictional Question

[6]                 In the course of oral argument, it was agreed that the abuse of process and "sham submission" arguments raised by Bayer in its written submissions be considered merged into the following jurisdictional argument.

[7]                 Bayer argues that Apotex has failed to comply with s.5 of the NOC Regulations, and as a result, Apotex' NOA is a nullity, leaving the Court with no jurisdiction to render a decision on the merits.

[8]                 First, Bayer contends that the modification of Apotex' Abbreviated New Drug Submission by way of a "notifiable change" is a "submission for a NOC" within the meaning of s.5 of the NOC Regulations, and consequently, under s.5(3)(c)(i), Apotex was obliged to file its notifiable change prior to serving its non-infringement NOA; Bayer argues that Apotex failed to comply with this timing requirement.

[9]                 Second, Bayer advances an alternative argument in the event of a finding that a notifiable change update is not a "submission for a NOC", being that the "submission for a NOC", within the meaning of s.5, is Apotex' underlying Abbreviated New Drug Submission. Bayer argues that, pursuant to s.5(3)(c)(i), where a second person, in this case Apotex, has filed a submission for a NOC, and the person alleges non-infringement, that person is required to serve on the first person, in this case Bayer, a NOA "relating to" the submission for a NOC filed with the Minister. Bayer states that Apotex'Abbreviated New Drug Submission contains a previously-prohibited process, and does not contain the current Apotex process which is the subject of Apotex' NOA in the present proceedings. Bayer, therefore, argues that Apotex' NOA does not "relate to" its Abbreviated New Drug Submission and, as a result, Apotex has failed to comply with s.5(3)(c)(i).

A. Has Apotex failed to comply with the timing requirements in s.5(3)(c)(i)?

[10]            Under s.5(1) of the NOC Regulations, where a second person, in the present case Apotex, files or has filed a "submission for a NOC" in respect of a drug, and compares that drug with, or makes reference to, another drug for the purpose of demonstrating bioequivalence, and that other drug has been marketed in Canada pursuant to a NOC and in respect of which a patent list has been submitted, the second person must either state that it accepts that a NOC will not issue until the patent expires or make an allegation of non-infringement (or one of the other allegations set out in s.5(1)(b)).

[11]            Section 5(3)(c)(i) provides that a submission for a NOC must be filed with the Minister before, or at the same time as, a non-infringement NOA is served. Subparagraph 5(3)(c)(i) of the NOC Regulations, reads in part:

[12]            With respect to these requirements, certain features of the present case are not in dispute. Apotex originally filed its Abbreviated New Drug Submission for ciprofloxacin in 1993 and, therefore, this submission was before the Minister at the time when the NOA at issue in these proceedings was served. The parties agree that on April 25, 2002, Apotex filed information with the Minister with respect to a "notifiable change" containing the Apotex process, which process is the subject of the non-infringement NOA in the present proceedings. The parties further agree that Apotex' notifiable change updates the Abbreviated New Drug Submission for ciprofloxacin previously filed by Apotex. The Minister acknowledged receipt of the information with respect to the notifiable change on April 26, 2002. Apotex sent its NOA with respect to its process to Bayer by registered mail on April 22, 2002, and under s.9(2) of the NOC Regulations, the NOA is deemed to have been served on Bayer five days after mailing, namely, April 27, 2002.

[13]            With respect to this fact pattern, Bayer argues that Apotex' Abbreviated New Drug Submission, as amended by the notifiable change in 2002, is a "submission for a NOC" within the meaning of s.5 of the NOC Regulations, and, therefore, the timing requirements in s.5(3)(c)(i) were engaged by the filing of the notifiable change in 2002. Accordingly, Bayer argues that s.5(3)(c)(i) required Apotex to have submitted its notifiable change prior to, or at the same time as, serving the non-infringement NOA on Bayer.

[14]            Apotex responds that a notifiable change is not a "submission for a NOC" within the meaning of s.5 of the NOC Regulations, and, therefore, the requirement in s.5(3)(c)(i) that a submission for a NOC be filed with the Minister before, or at the same time as, the NOA is served does not apply to Apotex' notifiable change.

[15]            An important fact which drives Bayer's argument is that the information filed as the subject matter of the notifiable change was at first found to be deficient by Health Canada, and a number of months went by before it was found to be acceptable; on April 26, 2002, Apotex attempted to file the notifiable change relating to its process, but the subject matter of it was not found to be acceptable until June 28, 2002. Thus, Bayer contends that Apotex' filing did not become a "submission" until June 28 2002, and, therefore, was filed after the service of the NOA on Bayer on April 27, 2002. Therefore, Bayer argues that Apotex failed to comply with the timing requirements set out in s.5(3)(c)(i).

[16]            In order to determine whether Apotex has complied with the timing requirements set out in s.5(3)(c)(i), it is first necessary to determine the legal force and effect of a "notifiable change", and then to determine whether the modification of an abbreviated new drug submission by way of notifiable change is a "submission for a NOC" within the meaning of s.5 of the NOC Regulations.

[17]            Bayer relies on the Management of Drug Submissions Policy, Therapeutic Products Directorate ("Management of Drug Submissions Policy") to argue that if a notifiable change is deficient upon filing, it does not constitute a "submission for a NOC" until all appropriate documents are filed and are accepted for review by the Minister.

[18]            The stated purpose of the Management of Drug Submissions Policy is to outline the way in which the Therapeutic Products Directorate will manage information and material submitted by sponsors in accordance with the Food and Drugs Act, R.S. 1985, c. F-27 and the Food and Drug Regulations. The policy applies "to all submission types", including notifiable changes.

[19]            The Management of Drug Submissions Policy reads in part as follows (note that "notifiable change" is abbreviated as "NC"):

2.1            Acceptable Original Information and Material

If original information and material is found to be acceptable on screening, it will be accepted for review and considered to be a submission. All submission types will be considered workload from the date of acceptance.

For NDSs, SNDs, ANDs, SANDSs, NCs, and those DINAs with a submission fee greater than $10,000 the sponsor will be notified of the acceptability by mail.

The date of acceptance for INDs will be considered the date of receipt in the Therapeutic Products Directorate, provided the information and material is accepted on screening. The days allocated to the screening of INDs are considered part of the days allocated for the submission review.

2.2            Unacceptable Original Information and Material

If deficiencies are identified during screening of original information and material the sponsor will be issued a Screening Deficiency Notice identifying the deficiencies. The sponsor will be required to submit all of the requested information and material identified in the Screening Deficiency Notice, within 45 calendar days from the date of request. As a general rule, submissions containing interim analyses of pivotal trial data or safety studies will be considered deficient on screening (unless otherwise specified in 3.2.2).

If the sponsor fails to provide all requested information within 45 calendar days, or the submitted information is incomplete, deficient or contains unsolicited information, the original information and material will be rejected and returned to the sponsor at the sponsor's expense. A Rejection Letter will be issued by the Therapeutic Products Directorate. If the sponsor wishes to resubmit the information and material at a future time, it will be processed as new information and material, and will be assigned a new control number.

After receipt of the information requested in the Screening Deficiency Notice, a new screening period commences (with a new performance target), and the requested material and information will be screened for completeness. The original information and material will be considered a submission when all requested information is found to be acceptable. The sponsor will be notified of the acceptability as delineated in 2.1.

DINAs may be rejected during screening for several reasons, without the issuance of a Screening Deficiency Notice. e.g. if an NDS is required, if a proposed ingredient is a prohibited substance, or if a monograph attestation is found not to reflect the submission content. If the sponsor wishes to resubmit the information at a future time, it will be processed as new information and material and will be assigned a new control number. [Emphasis added]

(Application Record (A.R.), v. V, 9F, pp. 865-866)

[20]            Thus, on the basis of the emphasised statements in the passage just quoted, Bayer argues that there is no "submission" until Health Canada gives its approval to the filings made. Bayer argues that the Management of Drug Submissions Policy and the NOC Regulations are linked and, therefore, the definition of "submission" in the Management of Drug Submissions Policy and the definition of "submission for a NOC" in s.5 of the NOC Regulations are the same, and both definitions include a notifiable change; thus, "a notifiable change" is a "submission" that modifies a pre-existing submission for a notice of compliance".

[21]            The term "submission for a notice of compliance" is not defined in the NOC Regulations, however s.2 does define "notice of compliance" as "a notice under section C.08.004 of the Food and Drug Regulations."

[22]            Section C.08.004 of the Food and Drug Regulations provides that a NOC shall be issued after a review of a "new drug submission", an "abbreviated new drug submission", or a "supplement" to a new drug submission or to an abbreviated new drug submission, and a determination that the submission or supplement complies with other requirements in the Food and Drug Regulations. As Apotex argues, a notifiable change is not mentioned in the Food and Drug Regulations, and in particular, in Division 8, which pertains to "new drugs"; that is, a NOC is only issued in respect of a new drug submission, an abbreviated new drug submission, or a supplement to either of those.

[23]            Subsection C.08.002(1) of the Food and Drug Regulations provides that no person shall sell or advertise a new drug unless the manufacturer of the new drug has filed with the Minister a new drug submission or an abbreviated new drug submission that is satisfactory to the Minister and has been issued a NOC in respect of the new drug submission or abbreviated new drug submission.

[24]            Section C.08.003 of the Food and Drug Regulations addresses the situation where a NOC has been issued to a manufacturer, but where the manufacturer must file with the Minister a supplement to the new drug submission or a supplement to the abbreviated new drug submission, and a notice of compliance must be issued in response to the supplementary submission before the new drug can be sold (See Apotex Inc. v. Canada (Minister of Health) (1999), 87 C.P.R. (3d) 271 (F.C.T.D.) affd (2001), 11 C.P.R. (4^th) 538 (F.C.A.)).

[25]            The expression "notifiable change" does not appear in the Food and Drug Regulations, nor is it found in the NOC Regulations. The only mention of "notifiable change" important to the present application is in the policy guidelines of the Management of Drug Submissions Policy. In my opinion, this distinction means that a notifiable change is merely a notice-giving document required for administrative purposes by Health Canada. Therefore, I find that a notifiable change has no force and effect in law.

[26]            Further support for Apotex' argument that the definition of a "submission for a NOC" within the meaning of s.5 of the NOC Regulations does not include a notifiable change is found in a line of authority concluding that the expression "submission for a NOC" as used in s.4 and s.5 of the NOC Regulations means, only, a new drug submission, an abbreviated new drug submission, and a supplement to a new drug submission or to an abbreviated new drug submission.

[27]            In Apotex Inc. v. Canada (Minister of Health) (1999), 87 C.P.R. (3d) 271 (F.C.T.D.) affd (2001), 11 C.P.R. (4^th) 538 (F.C.A.) at para. 36, Justice McGillis defined the term "submission for a NOC" as follows:

In considering the expression "submission for a NOC" in sections 4 and 5 of the Patented Medicines (Notice of Compliance) Regulations, the mechanisms which trigger the issuance of a notice of compliance are, by virtue of the definition of "notice of compliance" in section 2, those specified in section C.08.004 of the Food and Drug Regulations, namely a new drug submission, an abbreviated new drug submission, and a supplement to a new drug submission or to an abbreviated new drug submission. In the circumstances, the expression "submission for a NOC" in sections 4 and 5 of the Patented Medicines (Notice of Compliance) Regulations, means a new drug submission, an abbreviated new drug submission, and a supplement to a new drug submission or to an abbreviated new drug submission.    

[28]            While the main issue before Justice McGillis was whether a supplement to a new drug submission constitutes a "submission for a NOC" within the meaning of s.4, she analysed, in a comprehensive manner, the legislative scheme contained in the Food and Drug Regulations and the NOC Regulations and arrived at the conclusion that the definition of the expression "submission for a NOC" is the same for both s.4 and s.5 of the NOC Regulations.

[29]            Justice McGillis was quoted with approval by Justice Russell in Glaxo Smith Kline Inc. v. Apotex Inc. 2003 FC 1055 at para. 46, as follows:

In the circumstances, the expression "submission for a Notice of Compliance", as used in Sections 4 and 5 of the Regulations means a new drug submission, an abbreviated new drug submission and a supplement to a new drug submission or to an abbreviated new drug submission as was clearly established in Apotex Inc. v. Canada (Minister of Health) (1999), 87 C.P.R. (3d) 271 at 287 (F.C.T.D.), affirmed (2001), 11 C.P.R. (4^th) 538 (F.C.A.)

[30]            In Bristol Myers Squibb Co. v. Canada (Attorney General) (2003), 24 C.P.R. (4^th) 417 (F.C.A.) at para. 15, the Federal Court of Appeal cited with approval another decision of Justice McGillis (Merck & Co. v. Canada (Attorney General) (1999), 176 F.T.R. 21 at para. 59 (T.D.), affd (2000), 5 C.P.R. (4^th) 138 (F.C.A.)) and held that, for the purposes of s.5(1.1) of the NOC Regulations, "submission" means a new drug submission, an abbreviated new drug submission, or a supplement to either.

[31]            In Merck & Co. v. Canada (Attorney General), supra, Justice McGillis considered whether s.5(1) of the NOC Regulations requires a generic manufacturer to provide an allegation and, in particular, whether the generic manufacturer Nu-Pharm had filed a submission for a NOC comparing its drug product to a drug product marketed in Canada by a first person pursuant to a NOC for which a patent list had been filed. Justice McGillis held at para. 59 that the expression "submission for a notice of compliance" in s.5 of the NOC Regulations "means a new drug submission, an abbreviated new drug submission, and a supplement to either of those submissions."

[32]            Given this line of authority, together with my finding that a notifiable change has no force and effect in law, I find that a notifiable change is not a "submission for a NOC" within the meaning of s.5 of the NOC Regulations and cannot trigger s.5(3)(c)(i). Thus, I find that the requirement in s.5(3)(c)(i) that a submission for a NOC be issued prior to, or at the same time as, the service of a non-infringement NOA does not apply to a notifiable change.

[33]            The evidence is clear that Apotex originally filed its Abbreviated New Drug Submission for ciprofloxacin in 1993 and subsequently filed its NOA respecting the present application on April 27, 2002. On the basis of the analysis just concluded, I find that this Abbreviated New Drug Submission is the "submission for a NOC" contemplated by s.5, and it is this new drug submission which fulfils the requirement that a "submission for a NOC" be on file with the Minister before, or at the same time as, the service of a non-infringement NOA.

[34]            Therefore, my answer to the question posed in this section of these reasons is "no".

B. Has Apotex failed to comply with s.5(3)(c)(i) because its NOA does not "relate to" its submission for a NOC?

[35]            Unders.5(3)(c)(i), where a second person, in this case Apotex, has filed a submission for a NOC, and the person alleges non-infringement, that person is required to serve on the first person, in this case Bayer, a NOA "relating to"the submission for a NOC filed with the Minister. Bayer correctly states that Apotex' Abbreviated New Drug Submission contains a previously-prohibited process, and does not contain the current Apotex process which is the subject of Apotex' NOA in the present proceedings. Therefore, Bayer submits that Apotex' Abbreviated New Drug Submission is not "related to"its NOA as the Abbreviated New Drug Submission does not contain the entire basis for Apotex' non-infringement allegation. Bayer contends that as Apotex' NOA does not "relate to" its New Drug Submission, Apotex has failed to comply with s.5(3)(c)(i).

[36]            This issue has been addressed in AB Hassle v. RhoxalPharma Inc. (2002), 21 C.P.R. (4^th) 298, wherein the Applicants argued that RhoxalPharma had failed to comply with s.5(3)(c)(i) of the NOC Regulations because it did not have a submission filed with the Minister that "related to" its non-infringement NOA. The Applicants had contended that it was incumbent on the Court to ensure that RhoxalPharma's new drug submission, upon which the prohibition proceeding was based, fully corresponded to the NOA before the Court. Justice Gibson rejected the Applicant's argument on this point and held as follows at para. 31:

...I favour the position advocated by counsel for RhoxalPharma. If those who framed the relevant amendments to the Regulations on which the Applicants rely had intended to place an obligation on this Court to satisfy itself of the concordance between an underlying new drug submission and a relevant notice of allegation, they could have done so much more clearly and directly. It is worthy of note that the amended Regulations place no onus on a second person such as RhoxalPharma to file its new drug submission with this Court and that the authority of the Court to order production of relevant portions of a new drug submission is discretionary, not mandatory where a first person such as the Applicants here seeks such production. I regard the interpretation of the amendments to the Regulations that counsel for the Applicants urges on the Court as imaginative, but unduly strained.    

[37]            Although Bayer submits that I am not bound by Justice Gibson's reasoning and I should find to the contrary, I can find no serious reason to do so. Indeed, I agree with Justice Gibson's finding.

[38]            Therefore, my answer to the question posed in this section of these reasons is "no".

[39]            Accordingly, I find that Apotex has complied with s.5 of the NOC Regulations, and this Court has jurisdiction to render a decision on the merits.

II. The Non-Infringement Question

A. Principles of patent construction

[40]            While the purpose of these proceedings is not to decide whether the '067 patent has been infringed, the approach that has been adopted in similar cases considers established patent infringement considerations (see Janssen Pharmaceutica Inc. v. Apotex Inc. (2000), 5 C.P. R. (4^th) 53 at para. 59 (F.C.T.D.), affd [2002] 1 F.C. 393 (F.C.A.)).

[41]            The first objective in deciding whether a patent has been infringed is to construe the claims of the patent to determine what exactly lies within the scope of the monopoly granted by the patent. Once that has been determined, the second objective is to consider whether, for example, an impugned pharmaceutical process falls within the scope of the claims (for the principle see: Mobil Oil Corp. et al. v. Hercules Canada Inc. (1995), 63 C.P.R. (3^rd) 473 at 489 (F.C.A.)). Determining the scope of a patent involves giving it a purposive construction (Catnic Components Ltd. v. Hill & Smith Ltd., [1989] R.P.C. 183 at 242-243 (H.L.)). In Free World Trust v. Électro Santé Inc. , [2000] 2 S.C.R. 1024, Justice Binnie explored the principles of patent construction at paras. 14-15 as follows:

Patent claims are frequently analogized to "fences" and "boundaries", giving the "fields" of the monopoly a comfortable pretense of bright line demarcation....

...

In reality, the "fences" often consist of complex layers of definitions of different elements (or "components" or "features" or "integers") of differing complexity, substitutability and ingenuity. A matrix of descriptive words and phrases defines the monopoly, warns the public and ensnares the infringer. In some instances, the precise elements of the "fence" may be crucial or "essential" to the working of the invention as claimed; in others the inventor may contemplate, and the reader skilled in the art appreciate, that variants could easily be used or substituted without making any material difference to the working of the invention. The interpretive task of the court in claims construction is to separate the one from the other, to distinguish the essential from the inessential, and to give to the "field" framed by the former the legal protection to which the holder of a valid patent is entitled.                         [Emphasis added]

[42]            The significance of distinguishing essential from non-essential elements is that the substitution or omission of an essential element of a patent invention will defeat an allegation of infringement, whereas the substitution or omission of a non-essential element will not necessarily foreclose a patentee's claim of infringement (see Janssen, supra at paras. 46-47).

[43]            The claims of a patent are to be read from the perspective of a person skilled in the art to whom the patent is addressed, and who applies his or her knowledge in the field to which the patent relates (Whirlpool Corp. v. Camco Inc., [2000] 2 S.C.R. 1067 at paras. 48, 52-53). In the present proceedings, a person skilled in the art is someone who has advanced knowledge in the field of organic chemistry.

B. The evidence on construction of the '067 patent

[44]            The '067 patent is a product by process patent, as was required by s.39(1) of the Patent Act, R.S.C. 1985, c. P-4 (the "Patent Act") when the Patent was issued in 1987. At the relevant time, s.39(1) read as follows:

[45]            As stated in the '067 patent, the Patent relates to 1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-7-piperazino-quinoline-3-carboxylic acids, to a process for their production, and to their use in feed additives and as antibacterial agents; "1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-7-piperazino-quinoline-3-carboxylic acid" is the chemical name for ciprofloxacin.

[46]            The last of three claimed steps in the '067 patent is the reaction of piperazine and "7-chloro-1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-quinoline-3-carboxylic acid"; the latter compound is abbreviated to "6-FQA". This third step is embodied in claims 1, 8, 10 and 14, which are the only claims at issue in the present proceedings. Claims 1, 8, 10 and 14 are best understood by reference to the written and diagrammatic descriptions in the '067 patent cited in Appendix I to these reasons.

[47]            Claims 1 and 10 disclose the process itself, while claims 8 and 14 are product by process claims for the compound ciprofloxacin as prepared by the process defined in the process claims, or an "obvious chemical equivalent thereof". Claim 8 of the Patent claims ciprofloxacin (and three other compounds), as made by the process of claim 1 or an obvious chemical equivalent of this process ("claim 8(1)"). Where R is a hydrogen atom, the compound of this formula is ciprofloxacin. Claim 14 is a claim specifically for ciprofloxacin, when made by the process of claim 10 (or two other processes) or an obvious chemical equivalent of this process ("claim 14(10)"). For the purpose of making ciprofloxacin, claim 8(1) and claim 14(10) are effectively the same and are referred to as the "Bayer process"; in argument, counsel for Bayer focussed primarily on claim 14(10).

[48]            Both Bayer and Apotex adduce the evidence of expert witnesses with respect to construction of the '067 patent, and the nature of both the Bayer and the Apotex processes for producing ciprofloxacin. The professional qualifications of these experts are detailed in Appendix II. On behalf of Bayer, Dr. Overman filed two Affidavits: one sworn August 5, 2002 (A.R. v. 1, p. 12), and another sworn December 11, 2002 (A.R. v. 1, p. 151). Dr. Wuest filed an Affidavit sworn September 5, 2002 (A.R. v. IV, p. 602). On behalf of Apotex, Dr. McClelland filed two Affidavits: one sworn October 18, 2002 (A.R. v. VIII, p. 1751), and another sworn January 10, 2003 (A.R. v. IX, p. 2074). Finally, Dr. Meth-Cohn filed two Affidavits: one sworn October 18, 2002 (A.R. v. X, p. 2264), and another sworn January 21, 2003 (A.R. v. XI, p. 2525).

1. The interaction between piperazine and 6-FQA

[49]            Bayer argues that the Patent has wide scope. Dr. Overman, in the August 5, 2002

Affidavit at para. 15, and Dr. Wuest, in the September 5, 2002 Affidavit at para. 14, share the opinion that claims 8(1) and 14(10) describe and claim a process by which hydrogen from the piperazine compound, and chlorine from the 6-FQA, are eliminated thereby forming a bond between the piperazine compound and 6-FQA. Apotex argues that the process in the Patent requires a fully built, or intact piperazine and a fully built, or intact 6-FQA.

[50]            In cross examination, Bayer's expert Dr. Overman agreed that the relevant claims describe piperazine being added to an otherwise fully-built molecule to make ciprofloxacin:

173      Q. Dr. Overman, with respect to the '067 patent, I wanted to ask you whether you would agree with me that in each instance, either in the disclosure or in the claims where ciprofloxacin is made as illustrated or described, piperazine is added to an otherwise fully-built molecule to make ciprofloxacin?

A. Are you referring specifically and only to ciprofloxacin or in the '067 patent, ciprofloxacin is among the compounds that are referred to in the patent?

174      Q. Let's begin with ciprofloxacin.

A. I believe it is true.

(A.R. v. XIII, Tab 25, p. 2884, questions 173-174)

He further answered that an intact piperazine is used in the process claimed in the '067 patent:

192      Q. Still with the '067 patent, can you tell me whether there is any discussion or illustration of the building of the piperazine moiety or whether in all instances the piperazine moiety that is employed is intact?

A. The piperazine moiety that is employed, the piperazine or substituted piperazine is intact at the time of use.

(A.R. v. XIII, Tab 25, p. 2885, question 192)

[51]            Therefore, on Dr. Overman's evidence, I find that both the piperazine and the 6-FQA must be fully built.

[52]            A second dispute exists as to whether piperazine is essential to Bayer's process claims. On this point, in cross-examination, Dr. Wuest analyzed "example 4", which is an example in the disclosure of the '067 patent, to make the point that piperazine is not essential:

109      Q. Can you just tell me whether any of these examples illustrate the making of the compounds which are the subject of the invention, the compounds of Formula I, and if so, whether they illustrate that by utilization of compound II reacted with piperazine or piperazine derivatives?

A. My reading of the examples is that all of them refer to compounds of the invention, that is of general type I, and they all involve reactions of 6-FQA with a piperazine or piperazine derivative, except possibly Example 4 which I will have to look at more carefully because I am not quite sure what it talks about just yet.

110      Q. Okay

- A Short Pause

A. Example 4 relates to compounds of type I, but I do not see a reaction in which 6-FQA is made to react with a piperazine.

111      Q. Am I not correct that Example 4 in its starting materials starts with ciprofloxacin as one of the reactants?

A. Yes, it does.

112      Q. This is an example that actually starts with a compound that is within the compounds of the invention and then is manipulated in a way to alter it by altering one of the substituents on the piperazine group. Is that right.

A. That is true, although as far as I can tell in my reading of Example 4, it does not say how the ciprofloxacin was obtained.

113      Q. Right.

A. It could have been by a different route altogether.

114      Q. It is silent on the route by which the starting material for Example 4, one of the starting materials, cipro, was prepared.

A. Exactly.

115      Q. Then having looked through the entire disclosure, am I right that the only discussion and the only illustrations of the making of the compounds of Formula I, the process by which that is illustrated and discussed in each instance involves a 6-FQA compound or derivative reacted with piperazine or a piperazine derivative?

MR. BELLMORE: How are you defining "derivative" when you use that term in relation to either piperazine and 6-FQA?

MR. RADOMSKI: Compound II is COOR1, so it is whatever is permitted for R1 as part of compound II and for piperazine or piperazine derivative. Those are the words that are used by the disclosure on page 2 at the top. That is all I meant by my choice of the words -

THE WITNESS: I have become confused by what your question is. Would you mind repeating it, please?

116      Q. Not at all. Having now gone through the disclosure, my question is, am I right that the only process or processes that are discussed and illustrated are those which involve compound II, 6-FQA or its derivative, and piperazine or a piperazine derivative?

A. If you are limiting yourself to what is explicitly disclosed in the '067, I would agree with you. However, it is my understanding that someone of normal skill in the art would recognize that if this particular reaction can be carried out with piperazine, then conceivably it could also be carried out with ammonia itself by the same type of substitution reaction, and the NH2 group that results from that could be converted by conventional chemistry into a piperazine. That is one of the things that would occur to a reader of '067 of normal skill in the art because we do not know exactly how the ciprofloxacin in Example 4 was made. It seems to me that that leaves open the possibility that some alternative route was actually used for its construction.

...

120      Q. What you describe for me, though, takes 6-FQA or one of its derivative compounds with chlorine at the 7 position and then reacts that with ammonia and then requires further manipulation in order to convert the 7 position into a piperazine having had ammonia at that position in place of the chlorine. Is that correct?

A. I think that is a possibility suggested immediately to someone of normal skill in the art who reads '067.

(A.R. v. XII, Tab 24, pp. 2676-77, questions 109-116 and 120)

It appears from this evidence that Dr. Wuest did not know how ciprofloxacin is formed through the example, but he postulated one alternative whereby the same type of reaction could be carried out with ammonia, instead of piperazine, which could then be converted into piperazine by conventional chemistry.

[53]            In oral argument, counsel for Bayer referred to this ammonia reaction to show that the '067 patent covers not only piperazine reacting with 6-FQA, but also something that becomes piperazine. Counsel for Apotex, however, pointed out that Dr. Wuest did not assert that either Bayer's or Apotex' process uses chemistry similar to this "ammonia reaction." Apotex argued that while Dr. Wuest claimed that this "ammonia reaction" is based upon Example 4 of the '067 patent, this example does not make ciprofloxacin, let alone mention ammonia, and is, therefore, not relevant to the present case (Exhibit R-3, "Response to Bayer Hand-ups", p. 7).

[54]            I give weight to Apotex' argument and find that piperazine is essential.

2. The nature of the nucleophilic substitution reaction

[55]            Dr. Wuest states that the reaction of 6-FQA with piperazine in the Bayer process is a nucleophilic substitution reaction (September 5, 2002 Affidavit at para. 25, and cross-examination, A.R. v. XII, Tab. 24, p. 2680, question 149). On cross-examination, Dr. Wuest responded to questioning regarding the nucleophilic substitution reaction as follows:

146     Q. That nucleophilic substitution, as per what Mr. Belmore said, is more particularly described as a nucleophilic aromatic substitution. Would you agree with that?

A. It is described in the paragraph you referred me to in Exhibit J to nucleophilic substitution.

147      Q. Right, and I am asking you whether that is more particularly described, the substitution that occurs, 6-FQA plus piperazine, as a nucleophilic aromatic substitution?

A. It could be done that way because the leaving group is attached to an aromatic ring.

148       Q. I don't know what you mean by "it could be done that way", but that is in fact what happens; right? The leaving group is attached to an aromatic ring and, hence, the reaction is more particularly described as a nucleophilic aromatic substitution. Is that right?

A. It could be done that way.

149      Q. I don't know what you mean by "it could be done that way". I am asking you whether that descriptor, "nucleophilic aromatic substitution", correctly describes what occurs when you react 6-FQA with piperazine. Is that a nucleophilic aromatic substitution?

A. It is a aromatic substitution because a group that serves as a leaving group is displaced in a reaction with a nucleophile and a new bond is formed to the nucleophile and the bond to the leaving group is broken.

If you want to look at the details, then I agree with you that because the leaving group is attached to an aromatic ring, you could make a more specific reference to a nucleophilic aromatic substitution to distinguish this from, for example a nucleophilic aliphatic substitution. But both of these reactions are nucleophilic substitutions.

150      Q. Let's leave the aliphatic to the side. You could not describe the reaction of 6-FQA and piperazine as a nucleophilic aliphatic substitution. Is that correct?

A. No, you could not.

151      Q. So let's leave that to the side. I think we are in agreement that a correct description of the reaction of 6-FQA and piperazine to a person skilled in the art would be that it is a nucleophilic aromatic substitution. Quite aside from how else it might be described, that is an accurate description in terms of what the reaction is. Is that right?

A. It is an accurate description although I am not quite sure what you mean by a correct description. I think it depends on what context you are referring to.

My understanding is that we are talking about methods for constructing molecules and not engaging in a detailed mechanistic analysis of how these processes occur at the most intimate levels of detail.

152      Q. I only meant by "correct" that if a person skilled in the art were to say to James Wuest, "The reaction of 6-FQA with piperazine is a nucleophilic aromatic substitution," that would be a correct description.

A. That would be correct as would nucleophilic substitution.

153      Q. The words "nucleophilic aromatic substitution" are sometimes described by the letters SNAR?

A. That can be used, yes.

[Emphasis added] (A.R. v. XII, Tab 25, p. 2680, questions 146-153)

[56]            Dr. Overman agreed with Dr. Wuest that the reaction is a nucleophilic aromatic substitution reaction as follows:

201      Q. In the reaction of 6-FQA and piperazine to make ciprofloxacin, I want to understand what takes place. I have said before and I really mean this, I am not a chemist. I have a colleague who can assist me.                                                

Is it correct to say that one of the nitrogens on piperazine acts as a nucleophile and attacks the carbon chlorine of 6-FQA?

A. Yes. By carbon chlorine, if you are meaning the carbon at C-7 as the chlorine substituent -

202       Q. That's right.

A. That reaction is nucleophilic substitution.

203      Q. So am I right? One of the nitrogens from the piperazine acts as the nucleophile and attacks the carbon 7 position of the 6-FQA?

A. That's correct.

204      Q. Then that nitrogen bonds to the carbon 7?

A. Yes.

205      Q. Would it be correct to call that reaction a nucleophilic aromatic substitution?

A. It is a nucleophilic substitution of the specific class that is occurring at an aromatic carbon. So the subclass that is more specific would be a nucleophilic aromatic substitution.

206      Q. In terms of a nucleophilic substitution, am I correct that that is a description for a broad set of subclasses?

A. Nucleophilic substitution?

207      Q. Yes.

A. Yes.

208      Q. Within that descriptor of nucleophilic substitution, one of the subclasses is the nucleophilic aromatic substitution you have told me about and there are a number of others.

A. There is basically one other.

209      Q. What is the basically one other?

A. Nucleophilic aliphatic substitution.

[Emphasis added] (A.R. v. XIII, Tab 25, p. 2886, questions 201-209)

[57]            Apotex' witness, Dr. Meth-Cohn, agrees with Dr. Overman and Dr. Wuest that the Bayer process claimed in the '067 patent uses a nucleophilic aromatic substitution reaction:

The '067 Patent discloses in addition to the compound ciprofloxacin, the following which is relevant to the production of ciprofloxacin:...a one-step process for producing ciprofloxacin by way of a nucleophilic aromatic substitution reaction between piperazine and 7-chloro-1-cyclopropyl-6-fluro-1, 4-dihydro-4-oxo-quinoline-3-carboxylic acid or an ester therof (for simplicity the acid and esters have been referred to as 6-FQA)...

...

The person of ordinary skill in the art, reading these claims on February 17, 1987, would immediately identify this reaction as a 'nucleophilic aromatic substitution reaction. Such reactions were well known both prior to the discovery of ciprofloxacin and in 1987.

This person would also immediately understand that the sole purpose of the reaction depicted in these claims is to prepare ciprofloxacin by means of introducing the piperazine group in the 7-position of the already formed 1-cyclopropylquinolone.

The scope of these claims does not include any of the other subject-matter disclosed within the '067 Patent related to the production of ciprofloxacin. It does not include the process by which one would prepare the 6-FQA intermediate, the novel intermediate 2,4-dichloro-5-flurobenzoyl chloride nor the process by which to prepare 2,3-dichloro-5-fluorobenzoyl chloride. As such, the person of ordinary skill in the art would have understood that such subject-matter was not intended to be covered within these claims.

(October 18, 2002 Affidavit at paras. 42, and 50-53)

[58]            Given the agreement in the evidence, I find that the '067 patent claims a nucleophilic aromatic substitution reaction.

3. The presence of an activating carbonyl group at the C-4 position

[59]            With respect to patent construction, Dr. Overman, in the August 5, 2002 Affidavit at para. 16, and Dr. Wuest, in the September 5, 2002 Affidavit at para. 23, both "generalize" the Bayer process by, for example, diagrammatically omitting the carbonyl group at the C-4 position and replacing it with a "squiggly line" to show that the reaction could be carried out with something other than a carbonyl group at the C-4 position. Dr. Overman and Dr. Wuest also use this technique of drawing squiggly lines to "generalize" the Apotex process. In the October 18, 2002 Affidavit at para. 53, Dr. McClelland is critical of this approach by Dr. Overman and Dr. Wuest, stating that:

This is a considerable oversimplification of the chemistry that occurs in the two reactions. The person skilled in the art is taught in their very first studies of organic synthesis that other groups (the wiggly lines) can have a profound role on the fate of reactions, and that a large number of reactions falling under the two depictions do not work (i.e. where there are actual groups present, not wiggly lines)...

[60]            Apotex contends that the cross-examination of Dr. Overman demonstrates how the "squiggly lines" technique used by Bayer's witnesses to "generalize" obscures the fact that the nucleophilic aromatic substitution reaction proceeds as a result of, among other things, the presence of an activating carbonyl group at the C-4 position of the 6-FQA:

245      Q. ...Could I ask you to go back to paragraph 16 and let me just ask you this question: Looking at the reaction step and looking at the two squigglies that are attached to the nitrogen in the right intermediate of the two - not the piperazine, but the other intermediate that is being reacted with the piperazine. Do you see that?

A. Yes.

246      Q. If I were to put an oxygen at each squiggly, so that I would have an NO2, a nitro group, would that reaction work?

A. It would depend on what the other squiggly was that is on the -

247      Q. A hydrogen, let's say.

A. If the other is a hydrogen, if the piperazine is carbon 4 in the final product, so if carbon 4 then had a hydrogen and a carbon 3, the substituent was a nitro -

248      Q. Right.

A. This would not be the product that would be formed.

249      Q.      In fact, what you would get in that circumstance would be that the piperazine would displace the fluorine, not the chlorine?

A. That's correct.

250      Q. That would be because there would be an activating group, the NO2 opposite the fluorine?

A. In that specific case, the activating group is opposite the fluorine.

251      Q. Whereas in the case of 6-FQA being reacted with piperazine, the chlorine is displaced because it is opposite the carbonyl activating group. Is that right?

A. It is opposite the carbonyl activating group and the fluorine is opposite the electron donating amino group. It is a combination of those two.

252      Q. That is why the chlorine gets displaced by the piperazine in the Bayer step?

A. Any reaction of this type is going to depend on the detailed nature of the substituent. We could have either fluorine or chlorine displaced depending on how we substitute the aromatic ring. In this case, the chlorine would be displaced.                   [Emphasis added]

(A.R., v. XIII, Tab 25, p. 2889, questions 245-252)

[61]            In substantial agreement with Dr. Overman, Dr. Wuest had this to say about the presence of an activating carbonyl group at the C-4 position of the 6-FQA:

167      Q. Am I correct that in compound II, for purposes of the reaction with piperazine, the C-7 position is activated by the carbonyl which is found in the right ring at the top in position 4?

A. I think you could look at it that way. You have placed the burden of activation on the carbonyl group and it may be that the whole pyridone carboxylic acid ring has an effect on that as well.

But I would agree that the presence of the carbonyl group plays a role in activation. [Emphasis added]

(A.R. v. XII, Tab 25, pp. 2681-2682, questions 158-167)

[62]            Thus, on the basis of the evidence just emphasized, I find that the presence of a carbonyl group at the C-4 position of the 6-FQA is an essential of the '067 patent.

C. The arguments and conclusion on construction of the '067 patent

[63]            Bayer contends that the use of the word "comprises" rather than terms such as "consists of" or "is" in claim 10, in addition to the use of the term "obvious chemical equivalent" in claim 14, requires a broad interpretation of claim 14(10) (Merck Frosst Canada Inc. v. Canada (Minister of National Health and Welfare) (1998), 84 C.P.R. (3d) 492 at para. 52, 160 F.T.R. 161 (F.C.T.D), affd 8 C.P.R. (4^th) 48, 259 N.R. 88 (F.C.A.)).

[64]            Bayer focuses on claim 14(10), and argues that the following elements are essential features of the claim:

To a person of ordinary skill at February, 1987 with a mind willing to understand and uphold a really useful invention, this claim would be construed as:

Ciprofloxacin when prepared by a process which comprises the introduction, by a bimolecular reaction using nucleophilic substitution, of a piperazinyl moiety (i.e. piperazine or something that becomes piperazine) onto a substituted aniline [that could be 6-FQA or something that becomes 6-FQA] at what becomes position 7 of the ciprofloxacin molecule (or an obvious chemical equivalent of this process).

(Exhibit A-5, Essential Features of Claim 14(10), p. 1)

[65]            Apotex asserts that claims construction should not take place at too high a level of generality, and urges the adoption of the following elements as essential features of claims 8(1) and 14(10):

...a single step SNAr [nucleophilic aromatic substitution] reaction between an intact piperazine and an intact 6-FQA, to make the nitrogen C-7 bond necessary to make ciprofloxacin. The essential elements therefore are as follows: An intact piperazine having the nucleophile, the nitrogen; secondly, an intact 6-FQA or at least 6-fluoro-1-cyclopropyl quinoline with a C4 activating group for the C7 carbon; third, an SNAr reaction; fourth, the nitrogen C7 bond is made via a nitrogen external to 6-FQA which acts as the nucleophile and causes the leaving group to leave from the C7 aromatic carbon. And lastly, that reaction makes ciprofloxacin.

(Transcript of proceedings, v. 4, December 11, 2003, p. 511, lines 13-26)

[66]            Given the finely detailed evidence tendered with respect to the construction of the Patent, and given the conclusions reached on the three substantive topics just addressed, I agree that the proper construction of the Patent is that argued by Apotex. However, in recognition of the express wording of s.39(1) of the Patent Act, as well as the wording in the Patent itself, I also add the words "or an obvious chemical equivalent" to each of the five essentials expressed.

D. The law on obvious chemical equivalents   

[67]            The key focus of the present infringement inquiry is whether step 2 of the Apotex process is an obvious chemical equivalent of the Bayer process, and is, therefore, within the scope of claims 8(1) and 14(10) of the '067 patent. Justice Binnie in Free World Trust v. Électro Santé Inc., supra at para. 55, explains the Supreme Court's notion of equivalence:

It would be unfair to allow a patent monopoly to be breached with impunity by a copycat device that simply switched bells and whistles to escape the literal claims of the patent. Thus the elements of the invention are identified as either essential elements (where substitution of another element or omission takes the device outside the monopoly), or non-essential elements (where substitution or omission is not necessarily fatal to an allegation of infringement). For an element to be considered non-essential and thus substitutable, it must be shown either (i) that on a purposive construction of the words of the claim it was clearly not intended to be essential, or (ii) that at the date of publication of the patent, the skilled addressees would have appreciated that a particular element could be substituted without affecting the working of the invention, i.e., had the skilled worker at that time been told of both the element specified in the claim and the variant and "asked whether the variant would obviously work in the same way", the answer would be yes: Improver Corp. v. Remington, supra at p. 192. In this context, I think "work in the same way" should be taken for our purposes as meaning that the variant (or component) would perform substantially the same function in substantially the same way to obtain substantially the same result. [Emphasis added]

(See also Janssen Pharmaceutica Inc. v. Apotex Inc. (2000), 5 C.P. R. (4^th) 53 at para. 59 (F.C.T.D.), affd [2002] 1 F.C. 393 (F.C.A.) at paras. 46-47).

E. Step 2 of the Apotex process and obvious chemical equivalence

[68]            While the Apotex process comprises six steps, the focus of the inquiry is on step 2 as it is this step which, in Bayer's submission, infringes claims 8(1) and 14(10) by being an obvious chemical equivalent of the Bayer process. Bayer contends that step 2 of the Apotex process produces a precursor to ciprofloxacin which is then easily converted to ciprofloxacin through the conventional chemistry of steps 3-6.

[69]            Step 2 of the Apotex process involves the building and installation of a protected piperazine onto an aniline compound. Specifically, step 2 proceeds by way of two successive nucleophilic aliphatic substitution reactions ("SN2"). In these SN2 reactions, one of the two C-Cl bonds on the initial compound (A-2) and one of the N-H bonds on the aniline (A-3) are broken and an intermolecular reaction occurs, releasing HCl or hydrochloric acid. Then, the second C-Cl and N-H bonds are broken in an intramolecular reaction to close what will become the piperazine ring, releasing a second mole of hydrochloric acid (Respondent's Memorandum of Fact and Law, para. 18).

[70]            The expert witnesses are in agreement that the Bayer process uses a nucleophilic aromatic substitution reaction, while the reaction in step 2 of the Apotex process is nucleophilic aliphatic substitution. However, the dispute is whether, with respect to the reaction for making ciprofloxacin, the nucleophilic aliphatic substitution reaction in step 2 of the Apotex process is an obvious chemical equivalent of the nucleophilic aromatic substitution reaction employed in Bayer's claim 8(1) and 14(10) process.

1. Bayer's argument and evidence

[71]            Bayer argues that both the nucleophilic aromatic substitution reaction and the nucleophilic aliphatic substitution reaction perform the same function, in substantially the same manner, and are obvious chemical equivalents of each other.

[72]            Bayer submits that step 2 of the Apotex process "builds" the piperazine ring, while the Bayer process "adds" piperazine, but both processes take place via a nucleophilic substitution reaction that produces hydrochloric acid as a by-product. According to Bayer, the only difference between the two reactions is that the building process in step 2 of the Apotex process uses nucleophilic aliphatic substitution, whereas the Bayer process uses nucleophilic aromatic substitution. Bayer contends that while Apotex' expert witnesses emphasize the differences in the two substitution reactions as being meaningful, Bayer's expert witness Dr. Overman gave evidence that there are only two types of nucleophilic substitution reactions, and the use of one over the other depends on the strategy employed to perform the reaction (Applicants' Memorandum of Fact and Law, para. 38). The evidence upon which this argument is based is as follows.

[73]            On behalf of Bayer, Dr. Overman states that in February 1987 a person of ordinary skill would conclude that step 2 of the Apotex process is an obvious chemical equivalent of the relevant Bayer claims. He states that both processes involve the elimination of HCl with the resultant formulation of a piperazinyl moiety covalently bonded to a benzene ring. In his opinion, the only difference between the processes is the positioning of the hydrogen and chlorine atoms that will be eliminated during the reaction. In step 2 of the Apotex process, the chlorine and hydrogen atoms are reversed with respect to their positioning on each reactant, which therefore necessitates that an aniline reactant be reacted with a protected bis(2-chloroethyl) amine compound. Dr. Overman holds the view that the relative placement on each reactant of the chlorine and hydrogen atoms is not critical since these atoms are to be eliminated during the reaction (August 5, 2002 Affidavit at paras. 23-26; December 11, 2002 Affidavit at paras. 52-53).

[74]            In cross-examination, Dr. Overman further stated that both the Apotex and Bayer processes are bimolecular processes which share a rate-determining intermolecular step, although the Apotex process would further undergo an additional intra-molecular step. Dr. Overman defined a rate-determining step as the step of a sequence that has the slowest rate and which, as a result, determines the rate of the overall process (A.R. v. XIII, Tab 25, pp. 2891 and 2912, questions 85-86 and 171).

[75]            Dr. Wuest shares Dr. Overman's opinion that the Apotex process is an obvious chemical equivalent of the Bayer process. In the September 5, 2002 Affidavit at paras. 19-22, Dr. Wuest states:

I believe that the Apotex Process involves additional complexity, with no apparent benefit, when compared with the Bayer Process for producing ciprofloxacin. For example, in Step 2 of the Apotex Process, a protected piperazine moiety is introduced into the molecule and carried through the process as such until deprotection in Step 6, whereas in the Bayer Process, unprotected piperazine is used, and no deprotection is required....

I see no apparent advantage (eg. improved yield, improved purity, etc.) in adding such complexity to the process. For example, introduction of a piperazinyl moiety might be done using piperazine itself, which is conveniently available through a number of commercial channels. I do not see any particular advantage in constructing the piperazinyl moiety as illustrated in Step 2 of the Apotex Process.

At para. 25 of the Affidavit, Dr. Wuest describes the similarities between the Bayer and

Apotex processes:

In my opinion, a person of ordinary skill as of February 1987 would understand and conclude that the Bayer Process and Step 2 of the Apotex Process are similar because they both introduce the piperazinyl moiety into the structure. Specifically, the Bayer Process and Step 2 of the Apotex Process both involve elimination of HCl with resultant formation of a piperazinyl moiety covalently bonded to a benzene ring at the same position. Moreover, both processes occur by the mechanism of nucleophilic substitution. In both cases, a nitrogen atom serves as the nucleophile, and a chlorine atom is replaced by substitution. In comparing the two schemes shown above, the relative placement on each reactant of moieties to be eliminated (i.e., the chlorine and hydrogen atoms) during the reaction is not critical.

A person of ordinary skill, as of February 1987, would understand that a piperazinyl moiety could be incorporated into the chemical structure of a compound using a substitution reaction like the one used in Step 2 of the Apotex Process and starting with N,N-bis(2-chloroethyl)acetamide, the same compound that appears in Steps 1 and 2 of the Apotex Process. See, for example, Spanish patent number 539,112...publicly available prior to Februrary 1987....

[76]            In response to statements made by Apotex' expert witness, Dr. McClelland, with respect to the differences between the two nucleophilic substitution reactions, Dr. Overman, in the December 11, 2002 Affidavit at para. 52, states that the two reactions are obvious chemical equivalents and that the differences between the two are merely "mechanistic detail":

...[A] person of ordinary skill in the art is highly skilled and would have more knowledge than the simple textbook references relied on by Dr. McClelland. This person of ordinary skill in the art would readily understand that the two substitution reactions, while being different from one another in mechanistic details, are in fact very similar reactions, and may be considered to be equivalents of each other for the addition of the piperazine moiety to the ciprofloxacin molecule.

[77]            While earlier in these reasons a portion of Dr. Wuest's cross-examination was referred to for the purposes of patent construction, it is helpful to re-state it here for the purpose of making a finding with respect to "obvious chemical equivalence":

149      Q. ...I am asking you whether that descriptor, "nucleophilic aromatic substitution", correctly describes what occurs when you react 6-FQA with piperazine. Is that a nucleophilic aromatic substitution?

A. It is a nucleophilic substitution because a group that serves as a leaving group is displaced in a reaction with a nucleophile and a new bond is formed to the nucleophile and the bond to the leaving group is broken.

If you want to look at the details, then I agree with you that because the leaving group is attached to an aromatic ring, you could make a more specific reference to nucleophilic aromatic substitution to distinguish this from, for example, a nucleophilic aliphatic substitution. Both of these reactions are nucleophilic substitutions.

150       Q. Let's leave the aliphatic to the side. You could not describe the reaction of 6-FQA and piperazine as a nucleophilic aliphatic substitution. Is that correct?

A. No, you could not.

151      Q. So let's leave that to the side. I think we are in agreement that a correct description of the reaction of 6-FQA and piperazine to a person skilled in the art would be that it is a nucleophilic aromatic substitution. Quite aside from how else it might be described, that is an accurate description in terms of what the reaction is. Is that right?

A. It is an accurate description although I am not quite sure what you mean by a correct description. I think it depends on what context you are referring to.

My understanding is that we are talking about methods for constructing molecules and not engaging in a detailed mechanistic analysis of how those processes occur at the most intimate levels of detail.

152      Q. I only mean by "correct" that if a person skilled in the art were to say James Wuest, "The reaction of 6-FQA with piperazine is a nucleophilic aromatic substitution," that would be a correct description.

A. That would be correct as would nucleophilic substitution.                                           [Emphasis added]

(A.R. v. XII, Tab 24, p. 2680 at pp. 70-72, questions 149-152)

2. Apotex' argument and evidence

[78]            On the issue of obvious chemical equivalents, Apotex argues that step 2 of the Apotex' process for making ciprofloxacin is not an obvious chemical equivalent of the Bayer process. Apotex submits that the Bayer process does not "introduce a piperazine moiety to a substituted aniline" as was stated by Dr. Wuest, but rather an already intact piperazine is reacted with 6-FQA, a quinolone, through a nucleophilic aromatic substitution reaction (SNAr) to form ciprofloxacin. Apotex states that the Bayer process involves a reaction in which an N-H bond on the piperazine and a Cl-C bond at the C-7 carbon on 6-FQA are broken, a new bond is formed between the N on the piperazine and the C-7 carbon on 6-FQA and one mole unit of HCl is released, and that the presence of the carbonyl group (c=0) at C-4 on 6-FQA activates the C-7 carbon to effect this reaction.

[79]            Apotex stresses that its process uses none of the reactants, none of the reactions and none of the mechanisms of reaction described in the '067 patent, and that the two reactions function differently, at a different position, and form different bonds. Apotex argues that, unlike the Bayer process, step 2 of the Apotex process occurs at a nitrogen that is already attached to aniline, not a nitrogen on the piperazine. Unlike the Bayer process, which proceeds due to a carbonyl group at C-4 on 6-FQA which activates the quinolone at C-7, the aniline in step 2 of the Apotex process does not have an "activating group" because it uses different chemistry (ie. the nucleophilic aliphatic substitution reactions). Apotex cites the evidence of Dr. McClelland in the October 18, 2002 Affidavit at paras. 53-54, which establishes that, if Apotex attempted to use the nucleophilic aromatic substitution reaction claimed in the Bayer process instead of the nucleophilic aliphatic substitution reaction, step 2 of the Apotex process would not work.

[80]            Apotex further contends that while it would be possible to postulate different "leaving groups", such as fluoro, bromo or tosylate, at the C-7 carbon to be removed during the nucleophilic aromatic substitution reaction or the use of protecting groups on the piperazine, the skilled person would understand that the nucleophilic aromatic substitution reaction must occur between piperazine and a 6-fluoro-1-cyclopropyl quinolone having a C-7 leaving group and an activating group at the C-4 position. Apotex submits that, given the key importance placed on the 6-FQA intermediate in the '067 patent, a skilled person would understand that he or she could not venture far if at all from 6-FQA in employing this process. The evidence upon which this argument is based is as follows.

[81]            On behalf of Apotex, Dr. McClelland states in his October 18, 2002 Affidavit at para. 51 that step 2 of the Apotex process is not an obvious chemical equivalent of the Bayer Process. At para. 48, he states that the Bayer process functions by substituting a piperazine at the 7-position of a 4-quinolone to make ciprofloxacin, while the Apotex process functions by making the piperazine on a 2-fluoro-4-nitrobenzene, followed by four further steps to make ciprofloxacin. At paras. 43-48 and 51, he compares the Apotex and Bayer processes in the following manner (note that another name for the nucleophilic aliphatic substitution reaction is nucleophilic alkyl substitution.):

I compare here the single step in the Apotex process where the piperazine is introduced, step 2, with the single step of the "claim 8 and claim 14" process. It is my opinion that on making this comparison the person skilled in the art would conclude that the Apotex step does not perform substantially the same function in substantially the same way. Apotex do [sic] not substitute a piperazine onto a 4-quinolone. They do not even substitute a piperazine. The reaction in step 2 of the Apotex process actually makes the piperazine ring.

...The '067 process functions by using a reagent with an intact piperazine ring, which substitutes at C7 of the fully-formed 4-quinolone. The Apotex process starts with a compound where the nitrogen that becomes one of the ring nitrogens of the piperazine is already attached to the benzene ring. The reaction that occurs actually makes the piperazine ring, by making two new bonds within that ring. Neither the starting material nor the product of step 2 of the Apotex process are quinolones.

...

The person skilled in the art would also recognize that the reactions performing the function of introducing the piperazine are not equivalent. The reaction of claims 1 and 10 is a nucleophilic aromatic substitution, symbolized as SNAr. The reaction at step 2 of the Apotex process is a nucleophilic alkyl substitution and is symbolized as SN2 (actually done twice). These are different reactions and are treated as such by standard references - see Exhibit 10, in particular the discussion at the bottom of the page 499; "Although this reaction (nucleophilic aromatic substitution) appears to be similar to an SN1 or SN2 reaction, it is quite different".

While both reactions fall under the category "nucleophilic substitution", they function in different ways. This has an important consequence with respect to the timing of the introduction of the piperazine within the overall synthetic sequence.

In particular, the piperazine group cannot be introduced at the stage of the Apotex process using the reaction of claims 1 and 10. I illustrate this below. To apply claims 1 and 10 at the early stage of the Apotex process requires a reaction where piperazine (or a protected piperazine) displaces a leaving group such as chlorine that is located on the carbon adjacent to the florine on the benzene ring. Such a reaction does not work (as shown in Exhibit 11 to this affidavit). That the piperazine can be introduced at this stage with nucleophilic alkyl substitution is yet another reason that the person skilled in the art would conclude that the Apotex process does not perform substantially the same function in substantially the same way. In order to work at a stage where the patented process does not, the Apotex process must function in a different manner.

...

It is therefore my conclusion that the person of ordinary skill in the art would not conclude that the variant process, the Apotex process, was obviously working in the same way as the claim 1 and claim 10 process. In addition the person skilled in the art would conclude that even the step making the piperazine in the Apotex process did not perform substantially the same function in substantially the same way. The Apotex process made the piperazine; the claim 1 and claim 10 process employs an intact piperazine. The two reactions are different. This difference means that the Apotex process can function at an early stage, where the reaction of claim 1 and claim 10 fails. For these reasons it is my opinion that the Apotex process is not the obvious chemical equivalent of the claim 1 and 10 process.

[82]            In response to the statements made in the Affidavits of Dr. Overman and Dr. Wuest, Dr. McClelland states in the October 18, 2002 Affidavit at paras 54-55 as follows:

...Dr. Wuest and Dr. Overman are also ignoring the fact that the two reactions function in different manners. The claim 8 and 14 process is a nucleophilic aromatic substitution while step 2 is a nucleophilic alkyl substitution. While these are both termed nucleophilic substitutions, they are different reactions, and function in different ways. An important consequence of this different functioning is that the claim 8 and claim 14 process cannot be performed at the stage of the Apotex process.

Dr. Wuest (W25) [Dr. Wuest's Affidavit at para. 25] states that "the relative placement on each reactant of moieties to be eliminated (i.e. the chlorine and hydrogen atoms) during the reaction is not critical". A very similar statement appears in O25 [Dr. Overman's Affidavit at para. 25]. I disagree. The relative placement is critical to the Apotex process. If one were to switch to the placement of the process claimed within claim 1 and claim 10, the process would not work.

[83]            With respect to the differences between the aromatic and aliphatic substitution reactions, Dr. McClelland has this to say in the October 18, 2002 Affidavit at para 73:

The reaction shown in step 2 [of the Apotex process] involves an "aliphatic" (a term used to describe the class of compounds containing alkyl chains, being distinct from aromatic compounds) nucleophilic substitution reaction, whereby an aniline derivative (the nucleophile) reacts with an alkyl halide. The alkyl halide is in fact a bis-(chloroalkyl)amine derivative ("bis" implying two units) which reacts twice with the aniline to form the piperazine ring. It should be noted that this aliphatic nucleophilic substitution process is a totally different type of reaction to the earlier discussed aromatic nucleophilic substitution reaction. For example, no "activation" is required in the case of the aliphatic nucleophilic substitution reaction.

[84]            Dr. McClelland also states in the January 10, 2003 Affidavit at para. 104 as follows:

I disagree with Dr. Overman that the person of ordinary skill would readily understand that nucleophilic aromatic substitution and nucleophilic alkyl substitution are "very similar reactions". The person skilled in the art would recognize these different reactions. The fact that they have different mechanisms provides the warning flag. The means that the two reactions have very different requirements as to the nature of the groups that can and cannot be present in the two reagents for the reaction to succeed. This has the consequences that the two reactions cannot be employed interchangeably.... Dr. Overman fails to address this point in either of his affidavits.

[85]            In the October 18, 2002 Affidavit at para. 61, Dr. Meth-Cohn contemplates the types of obvious equivalents that would be covered by the '067 patent claims:

...[T]he person of ordinary skill in the art would understand these claims much as they are written, as claiming the compound ciprofloxacin when it is prepared by reacting piperazine with the 6-FQA intermediate. With respect to the types of obvious chemical equivalents the claims make reference to, such equivalents would include different leaving groups on the 7-position (for example fluoro, bromo or tosylate) of the 6-FQA or the use of protecting groups on either the piperazine or 3-carboxyl group (-CO2R) on the 6-FQA intermediate (for example by use of an acetyl-group (CH3CO) on the piperazine or an ethyl ester of the acid (CO2CH2CH3)). But in each and every case, the person of ordinary skill in the art would understand that the reaction with piperazine would occur with a 6-fluoro-1-Cyclopropyl quinolone. In light of the key importance placed on the 6-FQA intermediate, this person would understand that he could not venture far, if at all, from this compound.

3. Conclusion

[86]            Dr. McClelland is adamant that the difference between a nucleophilic aromatic substitution reaction and a nucleophilic aliphatic substitution reaction is significant. Dr. Overman effectively agrees that there is a difference, but wishes to diminish its significance by saying that it lies merely in the "mechanistic detail". I put weight on the agreement in the evidence that there is a difference, and for the following reasons, I find that the difference matters.

[87]            While both step 2 of the Apotex process and the Bayer process use nucleophilic substitution reactions, the evidence shows that even standard textbook references treat nucleophilic aromatic substitution and nucleophilic aliphatic substitution reactions as different reactions. More importantly, there is evidence that the nucleophilic aromatic substitution reaction used in the Bayer process and the nucleophilic aliphatic substitution reaction used in the Apotex process function in substantially different ways. Step 2 of the Apotex process involves an nucleophilic aliphatic substitution reaction; "aliphatic" is a term used to describe the class of compounds containing alkyl chains, which are distinct from aromatic compounds. In this nucleophilic aliphatic substitution reaction, an aniline derivative (the nucleophile) reacts with an alkyl halide. The alkyl halide reacts twice with the aniline to form the piperazine ring. A person skilled in the art understands that this is a totally different reaction to the nucleophilic aromatic substitution reaction used in the Bayer process. The two reactions have very different requirements as to the nature of the groups that can and cannot be present in the two reagents for the reaction to succeed. The presence of the carbonyl group at position C-4 on the 6-FQA in the Bayer process, for example, activates the C-7 carbon to effect a nucleophilic aromatic substitution reaction such that, if the carbonyl group were not present, the reaction would not proceed. In contrast, the evidence is that no activation is required in the case of the nucleophilic aliphatic substitution reaction.

[88]            For the reasons just stated, in my opinion, Bayer has not shown that the nucleophilic aliphatic substitution reaction performs substantially the same function in substantially the same way to obtain substantially the same result as the nucleophilic aromatic substitution reaction of claims 8(1) and 14(10) in the '067 patent. Thus, by

applying the notion of equivalence set out in Free World, I find that the evidence establishes that the nucleophilis aliphatic substitution reaction used in stept 2 of the Apotex process is not an obvious chemical equivalent of the aromatic subsitution reaction used in the Bayer process of claims 8(1) and 14(10) of the '067 patent.

[89]            Accordingly, I find that Bayer has failed to demonstrate that Apotex' allegation of non-infringement is not justified.

"D.R. Campbell"

     Judge

OTTAWA

1.              A process for preparing a 1-cyclopropyl-6-fluoro-1,4­dihydro-4-oxo-7-piperazino-quinoline-3-carboxylic acid of the general formula

(a) reacting 7-chloro-l-cyclopropyl-6-fluoro-1,4-dihydro 4-oxo-quinoline-3-carboxylic acid of the formula

with a piperazine or a piperazine derivative of the general formula

Bayer AG and Bayer Inc. v. Apotex Inc.

Docket: T-877-02

Bayer AG and Bayer Inc.v. Apotex Inc.

Docket: T-877-02

8.              A 1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-7-piperazino-

quinoline-3-carboxylic acid of formula I as defined in claim 1 or a pharmaceutically acceptable salt or hydrate thereof when prepared by a process according to claim 1

or an obvious chemical equivalent thereof.

...

10.            A process for preparing 1-cyclopropyl-6-fluoro-1,4

dihydro-4-oxo-7-piperazino-quinoline-3-carboxylic acid which comprises reacting

7-chloro-1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-quinoline-3-carboxylic acid with piperazine.

11.          A process according to claim 10 wherein the 7-chloro-

1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-quinoline-3-carboxylic acid is obtained

by cyclising ethyl 2-(2,4-dichloro-5-fluoro-benzoyl)-3-cyclopropylamino-acrylate by

reaction with sodium hydride in anhydrous dioxane.

12.            A process according to claim 11 wherein the ethyl 2-

(2,4-dichloro-5-fluoro-benzoyl)-3-cyclopropylamino-acrylate is obtained by reacting

ethyl 2-(2,4-dichloro-5-fluoro-benzoyl)-3-ethoxy-acrylate with cyclopropylamine.

...

14.            The compound 1-cyclopropyl-6-fluoro-1,4-dihydro-4-

oxo-7-piperazino-quinoline-3-carboxylic acid when prepared-by a process

according to claim 10, 11 or 12 or an obvious chemical equivalent thereof.

                                                                APPENDIX II

Expert Witnesses for Bayer

Dr. Larry Overman

Dr. Overman completed his doctoral dissertation in 1969 at the University of Wisconsin, and joined the faculty at the University of California, Irvine in 1971, where he is currently a Distinguished Professor of Chemistry. His research focus is the development of new reactions and strategies in organic synthesis of natural products and exploratory drug candidates, especially with respect to new cyclization reactions for the preparation of heterocyclic compounds. For many years his research focussed on the development of new cyclization reactions for the preparation of heterocyclic compounds.

Dr. James Wuest

Dr. Wuest obtained his Ph.D. in Organic Chemistry in 1973 from Harvard. He has taught chemistry at Harvard and is currently a Professor of Chemistry at the University of Montreal. He researches and publishes extensively in the field of synthetic organic chemistry, in particular the design, synthesis, structure, and reactions of organic compounds. He has also done a substantial amount of work in the area of Heterocyclic Chemistry.

Expert Witnesses for Apotex

Dr. Robert McClelland

Dr. McClelland received his Ph.D. in Chemistry in 1969 from the University of Toronto where he has been a member of the faculty since 1973. He is an expert in mechanistic organic and bioorganic chemistry, especially in the area of nucleophilic substitution and addition reactions, as well as in the area of medicinal chemistry, particularly the properties of heterocyclic drugs and the synthesis of new analogs thereof.   

Dr. Otto Meth-Cohn

Dr. Meth-Cohn obtained his Ph.D. in 1961 and a D.Sc in 1976 from the University of Salford, England, and is currently a Professor of Chemistry at the Institute of Pharmacy and Chemistry at the University of Sunderland. He has been engaged in heterocyclic chemistry research for over 40 years and has published extensively in this field, as well as in the area of the synthesis of quinolines.

                                                                                     FEDERAL COURT

                                               NAMES OF COUNSEL AND SOLICITORS OF RECORD

DOCKET:                                                        T-877-02                                     

STYLE OF CAUSE:                                     Bayer AG and Bayer Inc. v. Apotex Inc.

and The Minister of National Health & Welfare

PLACE AND DATE                                     Vancouver, BC - December 8, 9, 10, 15, 16, 2003

OF HEARING:                                               and

                                                                           Toronto, ON       - January 26, 2004

REASONS FOR ORDER

AND ORDER BY:                                        THE HONOURABLE JUSTICE D. R. CAMPBELL

DATED:                                                            Monday, February 02, 2004

APPEARANCES:

Neil R. Belmore/Ken Clark                          FOR APPLICANTS

H.B. Radomski/Richard Naiberg              FOR RESPONDENTS

Rick Tuzi

SOLICITORS OF RECORD:

Gowling Lafleur Henderson LLP              FOR APPLICANTS

Toronto, ON

Goodmans LLP, Toronto, ON                   FOR RESPONDENTS

Ivor M. Hughes, Thornhill, ON

                                 FEDERAL COURT

Date: 20040202

Docket: T-877-02

BETWEEN:

                      BAYER AG AND BAYER INC.

                                                                             Applicants

                                            and

                                   APOTEX INC.

                                            and

THE MINISTER OF NATIONAL HEALTH & WELFARE

                                                                          Respondents

                         REASONS FOR ORDER and ORDER