Date: 19980205
Docket: T-1714-95
BETWEEN:
PFIZER CANADA INC. and
PFIZER CORPORATION
Applicants
- and -
APOTEX INC. and
THE MINISTER OF NATIONAL HEALTH AND WELFARE
Respondents
REASONS FOR ORDER
REED, J.:
[1] The applicants (sometimes referred to collectively as "Pfizer") seek an order to prevent the Minister of National Health and Welfare from issuing notices of compliance to the respondent Apotex Inc. ("Apotex") for tablets and capsules of fluconazole. This application is brought pursuant to subsection 6(1) of the Patented Medicines (Notice of Compliance) Regulations, SOR/93-133. Pursuant to those regulations, Pfizer Canada Inc., filed a patent list in respect of each of the 50 mg, 100 mg and 200 mg tablets and the 150 mg capsules of fluconazole for which it had obtained a notice of compliance. That list contains Canadian Letters Patent No. 1,187,076 ("the 3076 patent"). An injunction is sought to prevent the issuance of notices of compliance to Apotex for the fluconazole it intends to market until the expiry of the 3076 patent.
[2] The regulatory scheme established under the amendments to the Patent Act, R.S.C. 1985, c. P-4, as amended S.C. 1993, c. 2, and by the Patented Medicines (Notice of Compliance) Regulations has been described in several decisions. See, for example, Bayer A.G. et al. v. Canada (Minister of Health and Welfare) (1993), 51 C.P.R. (3d) 329 (F.C.A.), and Eli Lilly and Co. v. Novopharm Ltd. (1995), 60 C.P.R. (3d) 417 (F.C.T.D.). I refer to those cases for the descriptions found therein and will not repeat the description here.
[3] Pursuant to that regulatory scheme Apotex served a Notice of Allegation on Pfizer dated June 23, 1995, stating that it would not infringe the 3076 patent by making, constructing, using or selling fluconazole. Pfizer seeks to show that that Notice of Allegation is not justified on two grounds: (1) there has been no linkage established between the process of manufacture that Apotex says will be used to make the fluconazole and the Apotex New Drug Submission ("NDS"); (2) the process that has been disclosed is a process that infringes the patent.
Linkage Arguments
[4] The Notice of Allegation states that no claim for "the medicine itself or for the use of the medicine" will be infringed. In section 2 of the Regulations a "claim for the medicine itself" is defined as including a claim in the patent for the medicine itself when prepared or produced by the methods or processes of manufacture particularly described and claimed, or by their obvious chemical equivalents. The allegation of non-infringement rests on the assertion that the process to be used by Apotex, in the manufacture of its fluconazole, is a non-infringing process. Paragraph 5(3)(a) of the Regulations requires that a detailed statement of the legal and factual basis for an allegation be provided.
[5] No details of that process were contained in the June 23, 1995 letter but an undertaking to provide the details, once a confidentiality order was in place, was given. A confidentiality order was obtained on September 11, 1995. On September 15, 1995, Dr. Sherman filed an affidavit disclosing the process that he says will be used.
Dr. Sherman was cross-examined on June 28, 1996. In the course of that cross-examination, he was asked questions about the content of the Apotex NDS. A Direction to Attend had been served on him, instructing him to bring with him the information relating to the process for making fluconazole that was in the Apotex NDS.1
[6] The requested documents were not produced by Dr. Sherman. Counsel for Apotex took the position that there was nothing in Pfizer's application for an order preventing the issuance of a notice of compliance that made the documents in the NDS relevant. In the course of the cross-examination Dr. Sherman expressed the opinion that he did not think, in any event, that the process for producing the raw material fluconazole was required as part of a NDS:
| ... in this country the New Drug Submission relates to the dosage form not the raw material and in fact if one has a Notice of Compliance one doesn't even need a new Notice of Compliance to do a different process. ...2 |
And again, the following exchange is recorded:
| Q. But you also disclosed either in the New Drug Submission or by reference to a drug master file what the process is to be used, correct? |
| A. Not necessarily because the rules were changed recently whereby a process is not considered to be integral to a submission. One does not need, for example, a new Notice of Compliance to change a process.3 |
[7] Prior to these comments, an exchange between counsel for Pfizer and counsel for Apotex took place. It underlies part of Pfizer's argument so it is necessary to set out that exchange below:
| MR. RADOMSKI [counsel for Apotex ]: ... Let me be clear. There is nothing in your originating Notice of Motion wherein the grounds are set out where you challenge the allegation, where there is any suggestion that the New Drug Submission is in any way pertinent. You have alleged two grounds. One is Section 55.1 of the Patent Act by way of infringement and then you have alleged a second ground. There is nothing else in your originating Notice of Motion that makes the New Drug Submission at all relevant. |
| MR. MACKLIN [counsel for Pfizer]: Well, that may or may not be the case. |
| MR. RADOMSKI: Well, that's my position. |
| MR. MACKLIN: I understand that. In any event you do not have any of the documents that you were asked to produce today under this Direction to Attend? |
| MR. RADOMSKI: That's correct. |
| MR. MACKLIN: You have the ability however to produce them if required to do so? |
| MR. RADOMSKI: of course.4 |
| . . . . |
[8] Pfizer subsequently obtained leave to amend their Originating Notice of Motion. It was amended on November 28, 1996 to include the assertion that the Apotex's allegation was not justified because the process to be used by Apotex must be in its NDS and that process had not been produced to Pfizer.5
[9] On February 21, 1997, Pfizer obtained an order that Dr. Sherman re-attend for cross-examination and that he bring with him the chemistry and manufacturing portion of Apotex's NDS and, in particular, the part dealing with the chemical process by which Apotex intended to manufacture fluconazole. Pfizer sought both the material that was on file with the Minister when Apotex served its Notice of Allegation on June 23, 1995, and that which was on file 46 days after service of the Notice of Allegation, on August 8, 1995.
[10] On May 7, 1997, Dr. Sherman re-attended for cross-examination. When he was asked whether he had brought the documents that were being sought, he stated that those documents did not exist6. They did not exist because the Apotex NDS for fluconazole was a cross-reference submission:
| Our submission has nothing in it on the Chemistry and Manufacturing because it is a submission which says, please give us an approval if, and when, you approve the product of another manufacturer.7 |
[11] Prior to this explanation, Dr. Sherman, in response to questions put to him, had stated:
| A: If we get approval on the basis of this allegation [that is, the process disclosed by Dr. Sherman on September 15, 1995], and we will make the product using this material, we will only do so, of course, if it is also approved by the Health Protection Branch. ... assuming we get approval, satisfy the regulations through this allegation, we will be making the product using the process as given to you. We will also need approval from the Health Protection Branch, and that approval, obviously, will only come if the Health Protection Branch is satisfied on the basis of whatever submission has been given to them by the people we are relying upon. ... |
| . . . . |
| Q: Have you submitted the process which is Exhibit I to your Affidavit in your New Drug Submission? |
| A: Our New Drug Submission does not have the process in it. It is not relevant to the Health Protection Branch. ... |
| . . . . |
| Q: If there is another process in [the cross-referenced NDS] ... that is approvable, would they get approval, Dr. Sherman? |
| A: They might get approval for a different process, but if they don't get approval for this process, we won't be able to sell the product because this is the process we are going to use ... I have given you that undertaking. |
| Q: Once again, you don't know, though, what process is in the New Drug Submission to which you are cross-referencing? |
| A: I don't even know if it is necessary for a process to be in the submission. This is a submission for tablets. It is not a submission for raw material. It may not be relevant to the Health Protection Branch.8 |
[12] Counsel for Pfizer then brought a motion seeking an order requiring Apotex to produce the NDS process documents, whether they be in its own NDS, or in the cross-referenced NDS. In the alternative, an order was sought requiring the respondent Minister to produce those documents. If the documents were not produced an order to strike from the record the affidavits filed in support of Apotex's position was requested. That motion came before Madame Justice Tremblay-Lamer on June 10, 1997. It was withdrawn on that day, without prejudice to Pfizer's right to raise any or all of the matters "if they so choose, as preliminary matters at the hearing of the application for prohibition".
[13] If I recall counsel for Pfizer's argument correctly, he indicated that Madame Justice Tremblay-Lamer had been of the opinion that the subject matter of the motion was properly a matter for the judge hearing the application on the merits and, accordingly, had suggested that the motion be withdrawn to allow the issue to be dealt with by that judge. The Registrar's minutes of the June 10, 1997 hearing, however, do not reflect that position. They indicate that a withdrawal was encouraged to allow the parties an opportunity to resolve the dispute without involvement of the Court. Implicit in a without prejudice withdrawal is the right to bring the motion forward again at a later time. The motion was not brought forward and counsel did not make a preliminary motion to me for an order requiring production of the documents. Counsel simply argued that because the documents had not been produced, the evidence that had been filed by Apotex should be struck from the record. I am not prepared to accept that position in the face of Pfizer having withdrawn, at an earlier date, the motion seeking production of the documents from Apotex or, alternatively, from the Minister.
[14] I also note that it is not correct to characterize the non-production of the documents as a breach of Mr. Justice Noël's order of February 21, 1997. While it is regrettable that the non-existence of the documents in Apotex's NDS was not disclosed earlier, and was not made an alternative basis for the refusals of June 28, 1996, it cannot be a breach of the February 21, 1997 order to fail to produce something that does not exist.
[15] Counsel for Pfizer calls attention to Mr. Radomski's response set out in paragraph 7 above. He characterizes that as being an admission that the documents exist and are being withheld. Mr. Radomski informed the Court that his "of course" response was meant to signify that he and his client could produce whatever was in the Apotex NDS. He had not focused on whether there were in fact process documents in that submission. I accept that explanation.
[16] I turn then to Pfizer's argument that the non-disclosure by Apotex of the documentation that has been sought renders its Notice of Allegation unjustified. Counsel for Apotex responds to this argument in the following terms: (1) there is no requirement that the process be in the Apotex NDS because that submission is a cross-reference submission; (2) there is no legislative requirement that the process to make the raw material fluconazole, as opposed to the manufacturing and chemical information concerning the dosage form, be part of a NDS; (3) the Minister may as a matter of practice, as an exercise of his discretion when assessing the safety of a substance, require the disclosure of such information - and he probably does; (4) the Regulations do not require that the NDS be produced in order to justify a Notice of Allegation; (5) the cross-referenced NDS will itself be required to meet all the requirements of the Regulations; and (6) in any event, the process that will be used has been produced to Pfizer and forms part of the record on this application.
[17] I find these arguments persuasive. I do not read the relevant legislative provisions as requiring the inclusion in the NDS of the process to make the raw ingredient, in this case fluconazole. This follows from the wording of section C.08.001.(a) of the Food and Drugs Act, R.S.C. 1985, c. F-27, which defines a new drug as a substance that has not been sold as a drug in Canada for sufficient time and in sufficient quantity to establish its safety and effectiveness.9 Whether the Minister will require the production of information relating to the process used to create the raw material, in this case fluconazole, is a matter for the Minister's discretion.
[18] In addition, the cross-referenced submission will itself be subject to the regulatory scheme established in the Patented Medicines (Notice of Compliance) Regulations. And, most importantly, the process that will be used has been disclosed.
[19] Counsel for Pfizer indicated that he had planned to raise a number of other issues to support a finding that the Notice of Allegation was not justified but that these had been settled, at least as far as this Court is concerned, by the recent Federal Court of Appeal decision in Eli Lilly & Company v. Apotex (A-339-97, September 29, 1997). At the same time, counsel indicated that he wanted to make it clear that he was reserving his right to argue those issues at a higher level, should it become appropriate, at a later date, to do so.
[20] In summary the Apotex Notice of Allegation has not been shown to be unjustified as a result of the non-disclosure of the process documents that Pfizer alleges exist and to which the arguments set out above relate. The process that will be used has been disclosed.
Apotex Process
[21] The process disclosed by Dr. Sherman, in Exhibit 1 to his affidavit, is described in terms of the structural formulae of the successive transformations involved in building the fluconazole molecule. This is sometimes referred to as a flow chart or pathway. That exhibit reads:
[22] The arabic numbers signify compounds; the Roman numerals signify process steps.
[23] In the diagrammatic shorthand that organic chemists use to communicate, they sometimes depict the individual reaction steps or intermediates within what is one overall reaction, one operation; they sometimes do not. For example, in the process step identified above as step II there are at least three and perhaps as many as six individual reaction steps.10 At the same time, the process steps labelled III and IV are part of one reaction, one operation.11
[24] The evidence differs as to the significance of process step IV. It is described as a geometric isomer separation. The result of process step III is the creation of a mixture of two geometric isomers (depicted as arabic 4 + 5). This is depicted in Dr. Sherman's flow chart by what I will call the "squiggly" bond line between the central carbon and the carbon in the C3 position.12 As a result of process step IV the two isomers are separated. The two separated isomers (arabic 4 and 5) are recognizable as such on Dr. Sherman's exhibit by the different angles at which the nitrogen of the triazole is bonded to the end carbon.
[25] Dr. Hendrickson, who was given no information except the 3076 patent and the exhibit to Dr. Sherman's affidavit, stated that the different isomers did not matter, he would not have written the flow chart the way it was written in the Exhibit to Dr. Sherman's affidavit. Dr. McClelland who had more detailed information concerning the Apotex process explained that while the two isomers are the same, in that they are composed of the same constitutents, they have different physical and chemical properties because the constitutent parts are arranged differently. He asserted that "there can be quite significant quantitative differences in terms of reactivity", and this was in fact the case with the isomers in question. His opinion was that one isomer was preferred over the other to make fluconazole: "the epoxidation step is very different in compound 4 and compound 5 ... compound 4 can also be converted to the epoxide but the conditions are different".13
[26] In any event, the process depicted in the affidavit can be described as: (process step I) subjecting compound 1 (a fluoroaromatic) to a Friedel-Crafts alkylation14 to form compound 2 (a 2-arylpropane - an arylalkane); (process step II) brominating compound 2 three times to form compound 3 (a 2-aryl-1,2,3-tribromopropane - a tribromoalkane); (process step III) reacting compound 3, in the presence of a base, with triazole moieties to produce a mixture of two geometric isomers, compounds 4 and 5 (bis-triazolealkenes); and (process step IV being part of the same operation as process III) separation of the geometric isomers to isolate compound 5; (process step V) converting compound 5 by an alkene epoxidation to form compound 6 (a bis-triazoleepoxide); (process step VI) the reduction of compound 6 by reacting it with a reducing agent (a hydride) to form fluconazole. It is the process steps from III onward that Pfizer identifies as particularly relevant for comparison purposes to the patent claims.
[27] Two components of the diagrams must be described before further reasons are given. There is reference in the evidence and in these reasons to a three carbon chain. It is depicted in the structural diagram above as . The lower vertical bond line is connected to the fluroaromatic. The two end bonds have, in many of the diagrams, a bromine or a triazole attached, the triazole being depicted by the structure. The three carbon chain is also found with some hydrogens, in Dr. McClelland's affidavit, as -CH2-C-CH2-. The carbons can be described as being located reading from left to right in the third, second or first position: C3-C2-C1 positions.
[28] There is frequent reference in the evidence and in these reasons to the epoxide ring. The epoxide ring is the triangular structure that is depicted in Exhibit 1 to Dr. Sherman's affidavit as . It is a two carbon, one oxygen structure and can be depicted as " C " C ", carbon being an atom with four bonds and oxygen having two.
[29] The individual reaction steps in steps III and IV of the Apotex process can be depicted, assuming one ignores the isomer separation step and details concerning when the bromines attached to the end carbons leave the molecule, (the mechanism by which the triazoles are added) as follows:
[30] The reaction sequence of process steps III and IV can be described as: elimination of the middle bromine on the three carbon chain; creation of a carbon - carbon double bond (an olefin) and the formation of a 1,3-dibromopropene; reaction with the first triazole (by way of an allylic nucleophilic displacement) so that the triazole attaches to one end of the three carbon chain and the bromine on the other side leaves the molecule; (4) the double bond then shifts (in the diagram above from between positions C1 and C2 to between positions C2 and C3) and the triazole attaches to the other terminal carbon; the carbon - carbon double bond shifts back to its original position and the bromine on that side of the three carbon chain, as a result of the second allylic nucleophilic displacement, has left the molecule. This results in the creation of the
geometric isomers compounds 4 and 5. Dr. McClelland and Dr. Hendrickson depicted these individual reaction steps as follows:
[31] As is obvious from the above diagram, not only do organic chemists differ with respect to the level of detail they depict from time to time in the flow charts they write, they also differ with respect to the particular type of diagrammatic shorthand they use at any given time. On the flow chart that Dr. Sherman produced there are no carbons or hydrogens expressly noted. These are omitted as a matter of convention, their presence being understood by the knowledgeable reader. They exist, however, at the corners of the various geometric shapes and arrangements. Sometimes descriptions are set out by reference to molecular formula rather than structural formula. Sometimes the diagrammatic shorthand is a combination of both.
[32] The creation of the fluconazole molecule in the Apotex process is completed by two additional separate reaction operations, after that depicted by process steps III and IV. Step V is the reaction of compound 5 (a bis-triazolealkene) with a peracid, for example, hydrogen peroxide. This introduces the oxygen and an epoxide ring is created:
[33] And lastly, step VI involves reacting compound 6 with lithium aluminum hydride, and then the addition of an aqueous acid of some sort to protonate the anion of fluconazole into fluconazole:15
Process step VI is a reduction reaction and it is a two step sequence even though the diagram above does not disclose that level of detail. The reaction results in the introduction of a hydrogen at the terminal carbon of the epoxide ring and the ring opens creating the hydroxyl (OH) necessary for completion of the fluconazole molecule.
[34] While I have identified the separate reaction operations that occur in the Apotex process, I do not want to be misunderstood as concluding that the similarity or equivalence of chemical processes depends upon whether there is one continuous reaction operation as opposed to several such steps.
Pfizer Process and Patent Claims
[35] I turn then to the Pfizer patent. The 3076 Patent has eleven claims. All of the claims depend upon claim 1. That claim reads:
| 1. A process for preparing a compound of the formula: - |
| or a pharmaceutically acceptable salt thereof, |
| which comprises |
| (a) reacting 1,2,4-triazole or an acid addition salt thereof with a compound of the formula: - |
| or with an acid addition salt thereof; |
| (b) reacting 1,2,4-triazole or an acid addition salt thereof with a compound of formula IV |
| wherein Q is a facile leaving group; or |
| (c) reacting 1,2,4-triazole or an acid addition salt thereof with a compound of the formula V: - |
| wherein Ha1 is a halogen atom; |
| and if required converting the compound of formula I into a pharmaceutically acceptable salt thereof. |
[36] The full text of the patent claims is set out as an appendix to these reasons. In summary, claims 2 to 5 are process claims, dependent on claim 1, that specify the nature of some of the reagents involved in the processes of claim 1. Claims 6 and 7 are process claims dependent on claim 1(a). Claim 8 is a process claim dependent on claim 1(b). Claim 9 is a process claim dependent on claim 1(b) or claim 8. Claim 10 is a product-by-process claim that claims fluconazole when prepared by a process of claim 1, 2, or 3, or an obvious chemical equivalent. Claim 11 is a product-by-process claim that claims fluconazole when prepared by a process of claims 6, 7, or 8, or an obvious chemical equivalent.
[37] Dr. McClelland describes the overall route of claim 1(a) of the 3076 patent as involving the reaction of a fluoroaromatic and acid chloride (claim 7, example 2(A)) to form an á-halo substituted ketone (claim 7). This ketone is reacted with 1,2,4-triazole (claim 7, example 2(B)) to form an á-triazole substituted aromatic ketone of formula III which is converted to an epoxide of formula II (claim 6, example 2(c)) and then fluconazole (claim 1(a), example 2(D)). The process steps of the route of claim 1(a) of the 3076 patent involves a Friedel-Crafts acylation, a nucleophilic displacement of a leaving group, an epoxide formation via a sulfonium ylide and a nucleophilic ring opening of the epoxide.
[38] He described the route of claim 1(b) as involving the reaction of a dihaloketone, with an organometallic reagent (claim 8, example 1(i)) to form a dihaloalcohol of formula IV which is reacted with two 1,2,4-triazole molecules (claim 1(b), example 1(ii)) to form fluconazole. The route of claim 1(b) involves an organometallic addition to a ketone and the reaction of that compound with triazole (a nucleophilic displacement or substitution reaction).16
[39] The route of claim 1(c) according to Dr. McClelland is similar to (b) except that an intermediate halomethyl substituted epoxide (compound V) is isolated prior to the reaction with the two 1,2,4-triazole molecules (claim 1(c)).
[40] Below is a flow chart, which was included by Dr. McClelland in his affidavit, that depicts the overall construction of fluconazole as claimed in the 3076 patent.
[41] Dr. McClelland's flow chart was prepared by reference to the examples set out in the patent and therefore has a "Q", in Box A under process 1(c), bonded to the third carbon. "Q" is used in the patent to signify "a facile leaving group". Compounds that fall into this category include, for example, bromine, chlorine, iodine. Also, because the chart was prepared by reference to the patent, one finds chlorine (Cl) depicted in the positions in which bromine (Br) is shown in other structural formula diagrams that are part of the record. In comparing the Box As on this flow chart, then, with other charts that are in evidence, it is not inappropriate to equate the "Q" and the chlorines (Cl) to the bromine (Br) found on those other diagrams.
[42] The Box A at the end of each column describes the subject matter of claims 1(a), 1(b) and 1(c) respectively. These overlap: claim 1(a) depicts the last part of the process claimed in claim 1(b). Claim 1(b) does not depict the individual reaction steps or intermediates that actually occur in the process step claimed by that claim. Claim 1(c) is essentially the same as claim 1(b) except that it starts after the epoxide has been formed but before a triazole is added.
[43] The actual steps in the reaction that occur can be depicted by going from compound IV (in process claim 1(b)) to compound V in process 1(c) to compound II in claim 1(a) to compound I (fluconazole). This is depicted in Dr. McClelland's affidavit as:
[44] Using a diagram of a similar type to that used in Exhibit 1 to Dr. Sherman's affidavit, the process can be depicted as:
[45] I have depicted the process with alternate configurations for the third transformation because the evidence is somewhat unclear as to whether the first triazole attaches in every case through the opening of the epoxide ring rather than by displacing a bromine. Dr. Fallis' evidence was that when chlorine is used as the leaving group the first triazole attaches by breaking open the epoxide ring.17 In the case of bromine, however, the result could be a mixture.18 Dr. McClelland gave the clearest evidence that in both cases the attachment was by the opening of the epoxide ring.19 In any event, the second triazole attaches in every case by the opening of the epoxide ring.
[46] The full depiction of the process described in the Pfizer patent, including claims 6, 7 and 8, as found in Drs. Hendrickson's and McClelland's affidavits, is set out as Appendix II to these reasons.
[47] The reaction sequence for the process operation claimed by 1(b), then, is one in which the hydrogen in the (OH) hydroxyl group leaves as does one bromine, and an epoxide ring structure is formed with the carbon at one end of the three carbon chain. Then, the triazole reacts with either the terminal carbon that is part of that epoxide ring or it reacts with the bromine on the other side of the three carbon chain. In the former case the epoxide ring structure opens to allow the attachment of the triazole (depicted above at C1), and the epoxide structure then reforms in conjunction with the carbon at the other end of the three carbon chain in the C3 position, the bromine on that side having left the molecule. The second triazole attaches in either case by breaking open the epoxide ring and simultaneously creating the hydroxyl (OH) to complete the fluconazole molecule.
[48] In any case the process described in claims 1(a), 1(b) and 1(c) introduce the second triazole to the molecule in every case by a nucleophilic ring opening. In other words, the final reaction step (which can be isolated) and which is the step claimed in claim 1(a), involves the ring opening to provide for the attachment of the second triazole and the simultaneous formation of the hydroxyl to complete the fluconazole molecule.
[49] Dr. Hendrickson described claim 1(a) as the reaction of an epoxide of formula II with 1,2,4-triazole which involves the opening of the epoxide ring during a bimolecular nucleophilic substitution, i.e, SN2 reaction. The following diagrammatic representation depicts the mechanism involved:
Expert Affidavits
[50] Expert affidavits by Dr. Hendrickson and Dr. McClelland were filed on behalf of the respondent Apotex. Expert affidavits by Dr. Crawford and Dr. Fallis were filed on behalf of the applicants. As can be expected they take diametrically opposite positions.
The affidavits of Dr. Crawford and Dr. Fallis, filed on behalf of Pfizer, assert that the Apotex pathway (process) is within the process claimed in the 3076 patent and that the two pathways (processes) are obvious chemical equivalents. The affidavits of Dr. Hendrickson and Dr. McClelland, filed on behalf of Apotex, assert that the Apotex process is not within the process described in the patent claims and that the two are not obvious chemical equivalents.
[51] Dr. Crawford's evidence is not convincing. He is Director of Process Research and Development with Pfizer Inc. His answers on cross-examination demonstrate that he was not an objective affiant. He seemed to take every opportunity to assert that the two processes were chemical equivalents without realizing that it is the reason he asserts that there is equivalence that will be the focus of the Court's attention, rather than the number of times it is repeated. It appears from his cross-examination that he is of the view that the more often he can repeat his opinion, the firmer it becomes established. This is not the case.
[52] Dr. Crawford's over-zealousness in attempting to promote the Pfizer position is particularly well illustrated in one portion of his cross-examination. He was attempting to support an analysis that the Apotex process was merely a re-ordering of the steps of the Pfizer process and therefore the former must be the chemical equivalent of the latter. In his over enthusiasm he described step VI of the Apotex process as being a reintroduction of the epoxy or the oxygen at a later stage than that found in the Pfizer pathway.20 There was no reintroduction of the epoxy or the oxygen since neither had existed earlier in that process. Dr. Crawford corrected himself when questioned further by counsel for the respondent.
[53] Dr. Fallis' cross-examination exhibits some of this same tone, in the applicant's favour, as does Dr. McClelland's, in the respondent's favour, but neither of these show the same lack of objectivity as found in Dr. Crawford's cross-examination.
[54] Both Dr. Crawford's and Dr. Fallis' affidavits contain statements that lead a reader to question the well-foundedness of their positions. For example, in Dr. Fallis' affidavit, he asserts that because the two pathways lead to the same result, that is the creation of fluconazole, it is his opinion that the Apotex pathway is an obvious chemical and functional equivalent of claim 1 of the 3076 patent, and particularly of claim 1(b).21
An expert affidavit that cites the fact that two processes lead to the making of the same compound as support for a conclusion that the two processes are chemically equivalent, when by definition they must both lead to that result, does not instill a reader with a great deal of confidence in the opinion proffered. In using the phrase "by definition" I mean that if they did not result in the same compound, there would be no subject matter for this litigation.
[55] Dr. Crawford cites as one reason for finding chemical equivalence, the fact that both processes (in the parts Pfizer identified as relevant for comparison) have starting compounds in which all but one of the building blocks of the fluconazole are the same. The fact that all but one of the building blocks of the molecule are the same, when the process step that is the subject of the patent claim is the last step in the building of that molecule, is not persuasive reasoning.
[56] Dr. Fallis identifies, as reasons for criticizing Dr. McClelland's opinion, the fact that Dr. McClelland does not mention that the two pathways both lead to the making of fluconazole, and that both pathways use two equivalents of triazole. These statements are made in the context of an analysis of two processes that must lead to the making of fluconazole (or there would be no litigation), and to a compound, fluconazole, that must have two triazoles as part of its structure. These are not convincing criticisms.
[57] I find Drs. Hendrickson's and McClelland's evidence more persuasive than that of Drs. Crawford and Fallis.
Patent Interpretation - Significance of Epoxide Ring
[58] Prior to January 1, 1994, subsection 41(1) of the Patent Act prevented the issuance of patents for substances used as medicines except insofar as they were produced by a particular process of manufacture:
41. (1) In the case of inventions relating to substances prepared or produced by chemical processes and intended for food or medicine, the specification shall not include claims for the substance itself, except when prepared or produced by the methods or processes of manufacture particularly described and claimed or by their obvious chemical equivalents.
41. (1) Lorsqu'il s'agit d'inventions couvrant des substances préparées ou produites par des procédés chimiques et destinées à l'alimentation ou à la médication, le mémoire descriptif ne doit pas comprendre les revendications pour la substance même, excepté lorsque la substance est préparée ou produite par les modes ou procédés de fabrication décrits en detail et revendiqués, ou par leurs équivalents chimiques manifestes.
(emphasis added)
[59] The 3076 patent is stated to relate to "a novel bis-triazole derivative which has antifungal activity and is useful in the treatment of fungal infections in animals, including humans". Thus it is an invention that, at the time the patent was issued, was governed by subsection 41(1). The patentee could of course claim the making of the substance by a number of processes and presumably would have sought to include in its patent as many as it knew about at the time the patent was sought.
[60] There is no dispute about the approach to be taken in construing patent claims. They are to be given a purposive construction rather than a purely literal one. A purposive construction involves a realistic construction of the language of the claim through the eyes, and with the learning of a person skilled in the art. One should not engage in an over close parsing of the words: Beecham Canada Ltd. v. Proctor and Gamble Co. (1982), 61 C.P.R. (2d) 1 at 10 (F.C.A.); Hoffman-LaRoche v. Minister of National Health and Welfare (1996), 70 C.P.R. (3d) 206 at 215 - 217; AT & T Technologies Inc. v. Mitel Corp. (1989), 26 C.P.R. (3d) 238 (F.C.T.D.).
[61] The patent claim, as noted above, on which all the claims depend and which is particularly relevant to this application is claim 1. Applying the accepted approach to patent construction, referred to above, the conclusion is that the essence of all three processes claimed in claim 1 is the formulation of fluconazole by a process that adds the second triazole to the pre-existing molecular structure by an epoxide ring opening with the simultaneous completion of the fluconazole molecule by formation of the hydroxyl.
[62] This conclusion is drawn from the "wording" of the claims. I use wording in quotation marks because these claims are not written solely in words. They are written partly in what I have called the diagrammatic shorthand of organic chemists. In both claims 1(a) and 1(c) the epoxide ring structure is an obvious element - it is specifically depicted on the face of the claim. This is not so in claim 1(b) but the expert evidence shows that while the organic chemistry shorthand used to describe claim 1(b) does not expressly depict the epoxide ring, it is nevertheless a necessary element of the process claimed in 1(b).
[63] While it is not appropriate to refer to the disclosure part of a patent to extend or limit the scope of a claim, it is appropriate to refer to it to explain the meaning of a term (particularly a technical term) in a claim.22 The disclosure, at pages 7b and 8 of the patent, supports the conclusions of Drs. Hendrickson and McClelland that the functioning of the epoxide ring is an essential element of the process claimed in claim 1(b) as well as those in claims 1(a) and 1(c).
[64] Dr. Hendrickson and Dr. McClelland read claim 1, indeed the whole patent, as having as an essential element the epoxide ring functioning as set out above. Dr. Hendrickson states:
| the function of the epoxide in the Pfizer patent is central to the whole idea of the patent ... it is there to provide the reactivity to add triazole ...23 |
[65] Counsel for the applicants referred often and vigorously to Mr. Justice Richard's decision in Pfizer Canada Inc. v. Apotex (T-2389-94, August 18, 1997). Mr. Justice Richard held that the essence of the invention to which the patent relates was fluconazole and its therapeutic properties, not the process by which it was created. Thus, it was held that the process by which fluconazole was made was not an essential element of the patent:
| But for section 41(1), the patentee would have been entitled to a product per se claim to the compound fluconazole when made by any process. The types of reactions included in the claims were general reactions known at the time of the Pfizer Patent and there was nothing inventive in the idea of using these reactions. These were only included to comply with section 41(1) in order that the claims of the patent include process limitations. Therefore, these features are not essential features. This conclusion is supported by the fact that the underlying invention is fluconazole and not the claimed process. |
[66] I have concerns about an argument that asks a judge to adopt findings of another judge in another case as determinative of an issue. The evidence that was before that other judge is not evidence in the second case. Mr. Justice Richard found that in the Apotex process which he had under consideration, the allegation was not justified because there was "functionally similar and parallel behaviour" (at page 18) between the Pfizer and ACIC pathways. "The cyclic sulfate functional group in the ACIC Pathway is an obvious and very well known chemical and functional equivalent of the epoxide group in the Pfizer Pathway of claim 1(a)" (at page 19).
[67] I understand counsel's argument to be that Mr. Justice Richard's reasons should be interpreted as saying that the process for making fluconazole is a non-essential element of the claims and therefore the making of that compound by a different process cannot be an infringement of the patent. I do not read Mr. Justice Richard's reasons in that way. Such an interpretation would mean that all of claim 1, for example, the claim on which all the others depend, was an unessential element. In addition, it would mean that a patent for the medicine itself, i.e., fluconazole, would exist, and that result was precluded by the Patent Act as it then read.
[68] In any event, on the basis of the evidence before me an essential element of all the subsections of claim 1 is the functioning of the epoxide ring. This element does not form part of the Apotex process. A finding that claim 1 has as an essential element the opening of an epoxide ring to add the triazole(s) and concomitantly complete the molecule with the formation of the hydroxyl, does not end the analysis that must be undertaken. As Mr. Justice Richard indicated, one must proceed, as a result of claim 10 and, in this case, claims 11 and 6 as well to ask whether the respondent's process is an obvious chemical equivalent to the Pfizer process. Obvious chemical equivalents are within the claims of the patent. Even if they had not been specifically claimed they would still be brought within the protected field by operation of section 41 of the Patent Act.
Equivalence - Jurisprudence
[69] A question that arises is whether the "equivalence" claimed in claims 10 and 11 is merely an articulation of the results that would flow in any event from a purposive construction of the patent. President Thorson in McPhar Engineering v. Sharpe Investments24 refers to some of the history of the doctrine of equivalence. He concluded that the doctrine continued to be applicable to patent interpretation even after 1883 when the inclusion of claims in a patent specification became necessary.25
[70] He describes the doctrine as "a particular application of the general doctrine that a patent may be infringed by taking the substance of the invention".26 He refers to the decision in Graver Tank & Mfg. Co., Inc., et al. v. Linde Air Products Co.27
[71] Mr. Justice Jackson, speaking for the Court in Graver Tank, described the purpose of the doctrine as being to prevent "the unscrupulous copyist ... [making] unimportant and insubstantial changes and substitutions in the patent which, though adding nothing, would be enough to take the copied matter outside the claim".28 After noting that the assessment of equivalence is a matter of asking about an alleged infringing device, whether "it performs substantially the same function in substantially the same way to obtain the same result",29 he continues:
| ... In its early development, the doctrine was usually applied in cases involving devices where there was equivalence in mechanical components. Subsequently, however, the same principles were also applied to compositions, where there was equivalence between chemical ingredients. Today the doctrine is applied to mechanical or chemical equivalents in compositions or devices. See discussions and cases collected in 3 Walker on Patents (Deller's ed. 1937) "" 489-492 Ellis; Patent Claims (1949) "" 59-60. |
| What constitutes equivalency must be determined against the context of the patent, the prior art, and the particular circumstances of the case. Equivalence, in the patent law, is not the prisoner of a formula and is not an absolute to be considered in a vacuum. It does not require complete identity for every purpose and in every respect. In determining equivalents, things equal to the same thing may not be equal to each other and, by the same token, things for most purposes different may sometimes be equivalents. Consideration must be given to the purpose for which an ingredient is used in a patent, the qualities it has when combined with the other ingredients, and the function which it is intended to perform. An important factor is whether persons reasonably skilled in the art would have known of the interchangeability of an ingredient not contained in the patent with one that was.30 |
[72] That jurisprudence, then, would seem to equate a claim of chemical equivalence with a purposive construction of the claims to which it relates.
[73] I find, that I aIso must ask whether section 41 of the Patent Act adds any broader protection to that existing as a result of the obvious chemical equivalence claims found in claims 10 and 11 of the patent. It may be that that phrase was made part of the Patent Act to ensure that a purposive and not a literal interpretation was applied to the process claims on which the patent protection depended. In Eli Lilly and Co. v. Novopharm Ltd. (1995), 60 C.P.R. (3d) 417, Mr. Justice Richard expressed the view that the phrase "obvious chemical equivalent" in section 41 afforded broader protection than merely a purposive interpretation approach covering minor variants (at pages 437 - 438). For present purposes, I will accept that this is the case. The analysis becomes, nevertheless, a factual one: is the Apotex process the obvious chemical equivalent of the process(es) claimed in the patent.
Chemical Equivalence - Pfizer and Apotex
[74] Pfizer argues that the Apotex process is an obvious chemical equivalent: (1) the only difference in the starting compound is that there is a bromine atom attached to the central carbon of the three carbon chain instead of an hydroxyl; (2) the same role is played by the carbon - carbon double bond (olefin), which is created after the elimination of the central bromine, as that played by the ring opening of the epoxide - it operates so as to allow the triazoles to attach; (3) the processes are similar in that first one triazole is added and then the other in sequence, and in both instances a halogen (bromine being a halogen) is replaced by a triazole; (4) there is an epoxide structure in both pathways and it operates to form the hydroxyl. Thus, it is argued that there is functional equivalency in the steps, with some reordering of them, and Apotex's process is an obvious chemical and functional equivalent.
[75] Dr. Fallis characterizes the similarity as being merely a reordering of the process steps, which he states in the Pfizer process are (1) epoxidation; (2) substitution of one bromine by a triazole; (3) substitution of the second bromine by a triazole contemporaneous with (4) an epoxide ring opening to complete the hydroxyl and thus the molecule. He characterizes the Apotex process as being (1) substitution of one bromine by a triazole; (2) substitution of the other bromine by a second triazole; (3) creation of the epoxide by epoxidation; (4) ring opening to form the hydroxyl and complete the molecule.
[76] With respect to the starting compounds, as noted, bromine was chosen by Pfizer for the purpose of comparing Pfizer's process to the Apotex process. Thus the starting compound for claim 1(b), used for comparative purposes for this application, is diabromoalcohol. The starting compound in step III of the Apotex process is a tribromoalkane. An apparently small difference in chemical compounds can signify a very important difference in chemical properties, and in the behaviour of the compound. Dr. Hendrickson's evidence was that the difference between a bromine and a hydroxyl group attached to the central carbon is a major difference.31 The resulting compounds do not operate chemically in the same way.
[77] With respect to the allegation that the carbon-carbon double bond (olefin) plays the same role as the epoxide ring in that it allows the triazoles to attach, I accept the evidence that while this may be true at a general and very superficial level, the two do not operate in a chemically equivalent fashion. The mechanisms for attaching the triazoles are quite different because one proceeds by way of an allylic nucleophilic displacement while the other proceeds by nucleophilic ring opening (displacement).
[78] Counsel for Pfizer argues that this analysis is too mechanistic, that it focuses on the details of the chemical reactions rather than the overall functions they serve. The description of the reaction mechanisms may, however, be relevant depending upon the processes that are being compared. They are one aspect of the total context relevant to assessing chemical equivalence. Equally, in some circumstances precursors to the processes a party seeks to compare may be relevant, in others they may not.32 In the present case, the description of the mechanisms is useful in assessing chemical equivalence.
[79] The fact that in both processes, first one triazole is added and then the other, does not demonstrate chemical equivalence. Of great significance is the fact that in the Pfizer process the attachment of the triazoles leads directly to the completion of the fluconazole molecule. This is not the case in the Apotex process.
[80] With respect to the fourth argument, while there is an epoxide structure in both pathways and they both lead to the creation of the hydroxyl and the completion of the molecule, the function of the epoxide in the Pfizer patent is very different from that in the Apotex pathway. In the Pfizer process the epoxide ring provides the reactivity to add the triazole, whereas in the Apotex process the epoxide is used by way of a two step process to convert a double bond into a hydroxyl group, after the triazole attachments have been made.33 As Dr. Hendrickson noted, there is certainly an epoxide to be found in each pathway, but the functions of the epoxide are quite different in the two processes.34 In one the ring opening is nucleophilic and in the other it is reductive. In one a triazole is introduced and in the other a hydrogen is introduced.35 Also, the introduction of the epoxide is by completely different chemistry in the two processes. In the Pfizer patent the epoxides are created by displacement reactions catalyzed by a base. In the Apotex process the epoxide is formed by epoxidation using ... an oxide agent of peracid - "there is chemically quite a big difference ...".36
[81] I turn then to Dr. Fallis' characterization of the Apotex process as merely a reordering of the Pfizer steps because each involves the creation of an epoxide ring, the substitution of a bromine by a triazole (twice) and the opening of an epoxide ring to complete the molecule. These are descriptions of the process steps at a high level of generality. Such a level of generality can disguise chemical difference. For example, describing a process as a "substitution" reaction or as a nucleophilic substitution reaction, as though nothing more detailed is required, obscures the chemical process. Dr. Crawford, for example, conceded that a nucleophilic substitution reaction is a very, very common reaction applying to many, many reactions in chemistry.37 Dr. McClelland spoke about the term "nucleophilic displacement" and pointed out that despite the fact that the triazole was introduced in both cases by reactions that are formally nucleophilic displacements "there is no way the two reactions could be considered equivalent to each other".38
[82] Dr. Hendrickson's evidence was that with the exception of the fluoroaromatic compound, the two processes proceed via different precursors of different structure and different functional class; the reactions that operate on the precursors to fluconazole are all different, employing different reagents and belonging to different reaction classes. Although the end result is fluconazole, the individual steps in the Apotex process all act and function in a different way from the individual steps in the 3076 patent.
[83] In the present case on the basis of the evidence I conclude that the two processes are not chemical equivalents.
Obvious Chemical Equivalence
[84] The above reasons describe the lack of chemical equivalency without any discussion of whether the two processes are obvious chemical equivalents. Having found that they are not chemical equivalents, it is not, strictly speaking, necessary to consider obviousness.
[85] Nevertheless, I note that in the course of the cross-examination of Dr. Sherman, on June 28, 1996, he produced Canadian letter's Patent No. 2,106,032 (the "3032 patent"). It issued on November 7, 1995, after Apotex's expert affidavits had been filed.
The following aspects of the Apotex pathway are claimed as inventions in the 3032 patent: (1) compounds 4 and 5 are the subject of claims 69 and 70; (2) the process of manufacture of compounds 4 and 5 from compound 3 is the subject of claims 54 and 55; (3) the process of manufacture of compound 3 from compound 2 is the subject of claims 52 and 53, (4) the process of manufacture of compounds 4 and 5 from compounds 1, 2 and 3 is the subject matter of claim 14.39
[86] Apotex also advises that the two final steps of the Apotex process are presently the subject of a patent application by Apotex. Apotex states that these processes are not disclosed in the prior art: "[t]hese processes were developed by overcoming hurdles with respect to the effects of steric hindrance and undesired opening of the epoxide of compound 6". Therefore it is argued that these process steps cannot be obvious chemical equivalents of the process steps claimed in Pfizer's 3076 patent.
[87] Counsel for the applicant argued that the 3032 patent should not be accepted as relevant for these proceedings: it had not been introduced under cover of an expert affidavit; the Court could not be expected to interpret it without expert opinion evidence; in any event, it relates to the intermediates and the processes prior to those that are relevant for present purposes.
[88] The question to be addressed for the purposes of this application, is not the inventiveness of the Pfizer process nor the inventiveness of the Apotex process but whether the two are "obvious chemical equivalents". Lack of obviousness is of course one of the requirements for a valid patent. One would hope that a patent examiner would not approve the issue of a patent for an "invention" that was obvious. However, this can occur. The most that can be drawn from the existence of the 3032 patent is that there is some reason to think that novel intermediates and processes exist in the part of the Apotex process covered by that patent.
[89] The most convincing evidence regarding lack of obviousness, however, is that of Drs. Hendrickson and McClelland. They attest that the two processes are not obvious chemical equivalents.40
Conclusion
[90] The respondent, Apotex, states in its Notice of Allegation that the process it will use to produce fluconazole is a non-infringing process. The applicant has not proven that that allegation is not justified.
[91] For the reasons given the application for an order of prohibition will be dismissed.
Judge
OTTAWA, ONTARIO
February 5, 1998
A P P E N D I X I
| THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS: |
| 1. A process for preparing a compound of the formula:- |
| or a pharmaceutically acceptable salt thereof, |
| which comprises |
| (a) reacting 1,2,4-triazole or an acid addition salt thereof with a compound of the formula:- |
| or with an acid addition salt thereof; |
| (b) reacting 1,2,4-triazole or an acid addition salt thereof with a compound of formula IV |
| wherein Q is a facile leaving group; or |
| (c) reacting 1,2,4-triazole or an acid addition salt thereof with a compound of the formula V:- |
| wherein Hal is a halogen atom; |
| and if required converting the compound of formula I into a pharmaceutically acceptable salt thereof. |
| 2. A process according to claim 1 wherein the 1,2,4-triazole is added in the form of an acid additional salt. |
| 3. A process according to claim 1, wherein the compound of formula II, is added in the form of an acid addition salt. |
| 4. A process according to claim 1, 2, or 3 wherein the reaction is carried out in the presence of a base. |
| 5. A process according to claim 1, 2 or 3 wherein the reaction is carried out in the present of potassium carbonate or sodium hydride. |
| 6. A process according to claim 1(a) wherein the compound of formula II is obtained by reacting a ketone of formula III |
| with dimethyloxosulphonium methylide prepared from trimethylsulphoxonium iodide and cetrimide/sodium hydroxide. |
| 7. A process according to claim 6 wherein the ketone of formula III is obtained by reacting 1, 3-difluorobenzene with chloroacetyl chloride in the presence of aluminum trichloride, followed by reaction with 1,2,4-triazole in the presence of a base. |
| 8. A process according to claim 1(b) wherein Q is a halogen and the compound of formula IV is obtained by reacting 1, 3-dihalo-acetone with 2, 4-difluorophenylmagnesium iodide, 2, 4-difluorophenylmagnesium bromide or 2, 4-difluorophenyllithium. |
| 9. A process according to claim 1(b) or 8 wherein Q is chlorine or bromine. |
| 10. The compound of formula I as defined in claim 1 or a pharmaceutically acceptable salt thereof when prepared by a process according to claim 1, 2 or 3 or an obvious chemical equivalent thereof. |
| 11. The compound of formula I as defined in claim 1, or a pharmaceutically acceptable salt thereof, when prepared by a process according to claim 6, 7, or 8 or an obvious chemical equivalent thereof. |
A P P E N D I X I I
__________________1 The relevant portion of the document instructs him to bring: a copy of the Chemistry and Manufacturing section of Apotex's New Drug Submission ... including any Drug Master Files referred to, relied upon or incorporated into the said New Drug Submission, which identifies the manufacturer and the process by which the said fluconazole is to be made; (any)information contained in Apotex's NDS relating to the process or processes for the manufacture of fluconazole and the date at which the process or processes were filed with the Health Protection Branch as part of the Apotex NDS. (underlining added) Applicants' Application Record, p. 100.
5 The amendment to the Originating Notice of Motion reads, in part:
... Apotex has failed to comply with the Regulations by not providing a detailed statement of the legal and factual basis for the allegation at the time of service of the notice of allegation. In the absence of any such information it should be presumed that Apotex's product formulation, processes and methods used to make its fluconazole will infringe the claims of the Patent.
The process as disclosed by Apotex and alleged by it in its Notice of Allegation to be used to produce fluconazole is not the same process identified in the New Drug Submission of Apotex to the Minister of health for a Notice of Compliance at the date of its
contd ... ... contd
Notice of Allegation, and as such the Notice of Allegation is false and misleading and is a nullity for failure to comply with the Patented Medicines (Notice of Compliance) Regulations.
6 Applicants' Application Record, p. 188, Q. 124".
9 Subsection C.08.022.(2) sets out the information that must be included in a NDS: (a) a description of the new drug and a statement of its proper name or its common name if there is no proper name; (b) a statement of the brand name of the new drug or the identifying name or code proposed for the new drug; (c) a list of the ingredients of the new drug, stated quantitatively, and the specifications for each of those ingredients; (d) a description of the plant and equipment to be used in the manufacture, preparation and packaging of the new drug; (e) details of the method of manufacture and the controls to be used in the manufacture, preparation and packaging of the new drug;
10 Applicants' Application Record, p. 402.
12 An explanation of the C3 , C2 and C1 positions is found infra p. 15.
13 Applicants' Application Record, pp. 409 - 413.
19 See generally, Id., pp. 429 - 435. If it were necessary to choose, I would prefer the evidence of Dr. McClelland.
22 TRW Inc. v. Walbar of Canada Inc. (1991), 132 N.R. 161 (F.C.A.) leave to appeal denied by S.C.C. 139 N.R. 400.
23 Applicants' Application Record, pp. 294 - 295.
24 (1960), 35 C.P.R. 105 (Ex. Ct.) at 157".
27 (1950), 85 U.S.P.Q. 328 (U.S. Sup. Ct.).
32 For example, in Bayer A.G. v. Canada (Minister of National Health and Welfare) (1996), 65 C.P.R. (3d) 203 at 209, Mr. Justice MacKay found it important to compare more than just the final steps in the two processes there under consideration.
33 Applicants' Application Record, p. 295.