GlaxoSmithKline Inc. v. Canada (Attorney General), 2005 FCA 197

Date: 20050524

Docket: A-686-04

Citation: 2005 FCA 197

CORAM:        DESJARDINS J.A.

NADON J.A.

PELLETIER J.A.

BETWEEN:

                                                      GLAXOSMITHKLINE INC.

                                                                                                                                            Appellant

                                                                             v.

                                             ATTORNEY GENERAL OF CANADA

AND THE MINISTER OF HEALTH

                                                                                                                                      Respondents

                                           Heard at Ottawa, Ontario, on March 8, 2005.

                                  Judgment delivered at Ottawa, Ontario, on May 24, 2005.

REASONS FOR JUDGMENT BY:                                                                       DESJARDINS J.A.

CONCURRED IN BY:                                                                                                   NADON J.A.

CONCURRING REASONS BY:                                                                              PELLETIER J.A.

Date: 20050524

Docket: A-686-04

Citation: 2005 FCA 197

CORAM:        DESJARDINS J.A.

NADON J.A.

PELLETIER J.A.

BETWEEN:

                                                      GLAXOSMITHKLINE INC.

                                                                                                                                            Appellant

                                                                           and

                                             ATTORNEY GENERAL OF CANADA

AND THE MINISTER OF HEALTH

                                                                                                                                      Respondents

                                                    REASONS FOR JUDGMENT

DESJARDINS J.A.

[1]                This is an appeal of a decision of the Federal Court (GlaxoSmithKline Inc. v. Canada (Attorney General) (2004), 37 C.P.R. (4^th) 336; von Finckenstein J.), which dismissed the appellant's application for judicial review of the refusal of the Minister of Health to list Canadian Letter Patent No. 1,298,479 (the '479 patent) and 2,031,393 (the '393 patent) on the Register of Patents maintained by the Minister pursuant to the Patented Medicines (Notice of Compliance) Regulations S.O.R./93-133 (the NOC Regulations).

[2]                The specification to patent '479 indicates the following (AB Vol. 1, p. 33):

This "invention relates to a system able to release a contained active substance at a practically constant controlled rate into a fluid able to cause swelling of the deposit-core containing the active substance".

The active substance to be released can consist of pesticides, herbicides, fertilizers, room deodorants, medicaments or generally any type of substance which requires to be released at a controlled rate into a aqueous fluid.

[my emphasis]

[3]                Claim 1 reads in part:

1. A system for the controlled-rate release of active substances, consisting of: a deposit-core comprising the active substance and having defined geometric form, and a support-platform applied to said deposit-core, characterized in that said deposit core contains, mixed with the active substance a) at least a polymeric material having high degree of swelling on contact with aqueous liquids, selected from the group consisting of:

...

[my emphasis]

[4]                In patent '393, claim 1 reads in part:

Tablets with controlled-rate release of the active substances, consisting of:

a core of defined geometrical form containing the active substance, polymer substances which swell on contact with aqueous liquids, substances with gelling properties, and a support applied to said core partially covering its surface, wherein said support consists of polymer substances which are slowly soluble and/or slowly gellable in aqueous liquids, and plasticizing substances, the amount of the polymer substances being 30-90% by wt. of the support composition and that of the plasticizing substances being 2-15% by wt.; ...

...

[my emphasis]

[5]                Do patents '479 and '393 contain "a claim for the medicine itself or a claim for the use of the medicine" as prescribed by paragraph 4(2)(b) of the NOC Regulations?

[6]                The motions judge indicated at paragraph 14 of his reasons:

Within the claims of either patent, there is, however, no explicit claim to either the medicine paroxetine hydrochloride or its use.

[7]                In the end, he concluded that neither patent qualified for registration under paragraph 4(2)(b) of the NOC Regulations. In doing so, he came to the same result as the Minister.

[8]                There was in evidence an affidavit by Dr. James W. McGinity, a pharmaceutical scientist, who indicated the following (AB Vol. I, p. 318-319):

31.    As with the '393 Patent discussed above, the inventors consistently use the term "active substance" throughout the specification for the '479 Patent. Indeed, the inventors define "active substance" broadly include "pesticides, herbicides, fertilizers, room deodorants, medicaments and generally any type of substance which requires to be released at a controlled rate into a aqueous fluid" at the second paragraph on page 1 of the specification. Clearly, paroxetine hydrochloride is a medicament.

32.    No further limitation is placed on the term "active substance" by the inventors. In particular, non-limiting examples of controlled-release tablets of certain medicines (as discussed above) are used, thereby suggesting that other medicines appropriate for formulation in controlled-release tablets are also within the scope of the claimed invention.

33.    Since GSK markets controlled-release tablets of paroxetine hydrochloride, it is clear that one skilled in the art would understand that the medicine paroxetine hydrochloride is an active substance suitable for formulation in a controlled-release tablet such as that claimed in the claims of the '479 Patent.

...

[my emphasis]

Conclusion

35.    Both the '393 Patent and the '479 Patent include within the scope of their respective claims controlled-release tablets of the medicine paroxetine hydrochloride. ...

...

[9]                The appellant relies on the decision of the Supreme Court of Canada in Burton Parsons Chemicals Inc. et al. v. Hewlett-Packard (Canada) Ltd. et al (1974), 17 C.P.R. 297 for the proposition that the patents at issue constitute a claim "for the medicine itself" as required by the NOC Regulations.

[10]            In Burton Parsons Chemicals Inc., the validity of the description of the invention (then subsection 36(1) of the Patent Act, R.S.C. 1970, c. P-4, now paragraph 27(3)(b)), was challenged. The patent in suit was for a conductive cream to be used in making electrocardiograms, electroencephalograms, and the like. It also covered the process of making such a cream. According to Pigeon J., the patented substance was composed of an aqueous emulsion containing sufficient highly ionizable salt to provide good electrical conductivity. In practical terms, this meant about 5% common salt. Claim 17 read as follows:

17. An electrocardiograph cream for use with skin contact electrodes and compatible with normal skin, comprising a stable aqueous emulsion that is anionic, cationic or non-ionic and containing sufficient highly ionizable salt to provide good electrical conductivity.

An expert for Burton Parsons, Dr. Shansky, recognized that some compositions would be unsuitable for normal skin since they could cause irritation.

[11]            Pigeon J.'s analysis was then the following:

...

In the present case, the invention relates to a mixture and a process for making it. This mixture is of no fixed composition. A great many different substances can be used, hundreds if not thousands, said Shansky. The essential is to combine a highly ionizable salt with an aqueous emulsion. As a result of this combination, the wetting action of the emulsion on the skin makes it possible to use the salt in a low concentration (from 1 to 10%). If the patent is to have a practical value, it must cover all the emulsions and salts which can yield the desirable result, namely, all "emulsions with the outer phase or the continuous phase being water" and all salts that are highly ionizable enough to carry an electric current with low resistivity on the skin excluding only such substances as are not compatible with normal human skin. The evidence makes it clear that this was obvious to any person skilled in the art because the characteristics of suitable emulsions and of suitable salts were well known. Only the combination was new.

...

[my emphasis]

[12]            The invention therefore related to a mixture and a process for making it. The mixture was of no fixed composition. The essential was to combine a highly ionizable salt with an aqueous emulsion compatible with normal skin. The evidence showed , however, that the characteristics of suitable emulsions and of suitable salts were well known to any person skilled in the art, so that the claims of the patent were held not to be broader than the invention.

[13]            In the case at bar, the claims in the patents give no guidance as "for the medicine itself" (a defined term found in section 2 of the NOC Regulations), except through the use of an "active substance". I find these words too imprecise to satisfy the requirement of paragraph 4(2)(b) of the NOC Regulations. Expert evidence cannot fill in when the terms of the claims give insufficient guidance.

[14]            Subsection 27(5) of the Patent Act does not assist the appellant since there is no "claim which defines the subject-matter of an invention in the alternative". Subsection 27(5) of the Patent Act reads:

[15]            I would dismiss this appeal, with costs to the respondents.

        (s) "Alice Desjardins"                                                                                                                                            J.A.

"I agree.

     M. Nadon J.A."

PELLETIER J.A. (concurring reasons)

INTRODUCTION

[16]            GlaxoSmithKline Inc. (GSK) appeals the decision of von Finckenstein J. of the Federal Court dismissing its application for judicial review of the refusal of the Minister of Health to list Canadian Letters Patent No. 1,298,479 (the '479 patent) and 2,031,393 (the '393 patent) on the Register of Patents maintained by the Minister pursuant to the Patented Medicines (Notice of Compliance) Regulations, S.O.R. /93-133, as amended (the NOC Regulations). GSK sought to have the patents listed in relation to its product PAXIL CR, paroxetine hydrochloride 12.5 mg and 25 mg controlled-release tablets.

FACTS AND PROCEDURAL HISTORY

[17]            This Court has held in the past that the NOC Regulations protect only the substances which are administered to patients. They do not protect devices, such as dermal patches, by which a substance is administered to a patient. This case falls on the boundary between these two types of cases in that the system by which a medicine is administered to a patient in a controlled release is itself ingested by the patient. In those circumstances, the question for the Court is whether the patents which protect the controlled-release system contain a claim for the medicine which is released.

[18]            The '393 and '479 patents both deal with the controlled-rate release of an active substance in an aqueous fluid. The '479 patent describes a system of controlled release of an active substance while the '393 patent deals with a tablet in which an active substance is released in a controlled fashion into an aqueous fluid. The system described in the '479 patent is similar to that taught in the '393 patent.

[19]            PAXIL CR is a novel dosage form of a well-known drug, paroxetine hydrochloride, in which the drug is released into the patient's system in a controlled fashion using the systems described in the '393 and '479 patents. The Minister refused to list the '393 and '479 patents on the Patent Register in relation to PAXIL CR because, in his view, they did not contain a claim for the medicine itself, or for the use of the medicine, as required by paragraph 4(2)(b) of the NOC Regulations.

[20]            GSK's claim to list the '393 and '479 patents in relation to PAXIL CR rests upon the affidavit evidence of Dr. McGinity, who paraphrased Claim 1 of the '393 patent as follows:

... I note that Claim 1 claims tablets with controlled-rate release of the active substance, consisting of a geomatrix core containing the active substance, swellable polymers, substances with gelling properties and a support applied to the core, which support is made of slowly soluble or slowly gellable substances as well as plasticizing substances.

[Appeal Book, p. 315.]

[21]            Dr. McGinity paraphrased claim 1 of the '479 patent as well:

... I note that Claim 1 claims a controlled-release system including the active substance mixed together with highly swellable polymers to form a deposit-core as well as a support platform comprising polymeric material applied to the deposit-core...

[Appeal Book, p. 318.]

[22]            Dr. McGinity then compared the two patents:

In this regard, the invention claimed in the '479 Patent is similar to the invention claimed in the '393 Patent. They both claim the active substance mixed with other substances in a core, which core is then coated...

[Appeal Book, p. 318.]

[23]            Since both patents involve the release of an active substance, Dr. McGinity went on to consider whether paroxetine hydrochloride was included in the scope of "active substance":

As with the '393 Patent discussed above, the inventors consistently use the term "active substance" throughout the specification for the '479 Patent. Indeed, the inventors define "active substance" broadly to include "pesticides, herbicides, fertilizers, room deodorants, medicaments and generally any type of substance which requires to be released at a controlled rate into a aqueous fluid" at the second paragraph on page 1 of the specification. Clearly, paroxetine hydrochloride is a medicament.

No further limitation is placed on the term "active substance" by the inventors. In particular, non-limiting examples of controlled-release tablets of certain medicines (as discussed above) are used, thereby suggesting that other medicines appropriate for formulation in controlled-release tablets are also within the scope of the claimed invention.

Since GSK markets controlled-release tablets of paroxetine hydrochloride, it is clear that one skilled in the art would understand that the medicine paroxetine hydrochloride is an active substance suitable for formulation in a controlled-release tablet such as that claimed in the claims of the '479 Patent.

...

Both the '393 Patent and the '479 Patent include within the scope of their respective claims controlled-release tablets of the medicine paroxetine hydrochloride...

[24]            Dr. McGinity's evidence established that all of the elements required to effect controlled release are chemical compounds which are combined with the active substance in such a way that when the system is placed in an aqueous fluid, the active substance is released into the fluid in a controlled fashion.

[25]            GSK argued before this Court, as it did before the Minister and the applications judge, that these facts justify the listing of the '393 and '479 patents in relation to PAXIL CR. In order to be included on the register, a patent must include a claim for the medicine itself. It says that if the medicine is an essential element of the invention, as disclosed by the construction of the claims of the patent, then the patent contains a claim for the medicine itself. The evidence of Dr. McGinity establishes that paroxetine hydrochloride is an active substance within the meaning of the claims of the patent. Therefore it is GSK's position that the patents contain a claim for paroxetine hydrochloride.

[26]            GSK relies upon the decision of the Supreme Court of Canada in [1976] 1 S.C.R. 555">Burton Parsons Chemicals, Inc. v. Hewlett-Packard Canada Ltd., [1976] 1 S.C.R. 555 (Burton Parsons Chemicals Inc.) in support of its position. Counsel argued that this decision stands for the proposition that where a patent is drafted so as to claim a class of elements, the patent must be read as claiming every element of that class. According to counsel, this interpretation was carried forward into the Patent Act by amendments which introduced subsection 27(5) into the Act:

[27]            Both the Minister and the applications judge rejected GSK's arguments. The Minister found that the patents in question contained claims "to a system or tablet which could be used in the administration of a variety of active substances". The Minister concluded that " ...there is no claim within either patent to paroxetine hydrochloride or its use..." (Appeal Book, p. 92). In other words, the patents relate to systems for the administration of a drug as opposed to the drug which is administered.

[28]            The applications judge noted that the patent referred only to "active substances" and not to paroxetine hydrochloride. While one skilled in the art, like Dr. McGinity, could find that "active substances" include within their scope paroxetine hydrochloride, such a person could not "equate active substance with paroxetine hydrochloride." In other words, "active substances" include paroxetine hydrochloride, but are not limited to paroxetine hydrochloride. The applications judge noted that according to the terms of the patent itself, the active ingredient could just as easily be a herbicide or a fungicide as a medicament. In the result, he found that the patents did not include a claim for a medicine or the use of a medicine.

ANALYSIS

[29]            Both parties agreed that the standard of review is that of correctness, as set out by this Court in Eli Lilly Canada Inc. v. Canada (Minister of Health) (C.A.), [2003] 3 F.C. 140, 2003 FCA 24.

[30]            While this case calls upon the Court to interpret the '393 and '479 patents, the critical issue is the meaning of the phrase "a claim for the medicine itself or a claim for the use of the medicine". To that extent, whether the expression "active substance" includes paroxetine hydrochloride is not determinative of this case.

[31]               The expressions "a claim for the medicine itself" and "medicine" are defined in the NOC Regulations:

[32]            A "claim for the medicine itself" is a claim for the medicine when prepared or produced by the methods or processes of manufacture particularly described and claimed. Do the patents in issue describe a method or process of manufacture of paroxetine hydrochloride? It is clear that a pharmaceutical scientist who was unfamiliar with the chemical composition of paroxetine hydrochloride could not derive the methods or processes of manufacture of paroxetine hydrochloride from the '393 and '479 patents. On the other hand, a pharmaceutical scientist who was already familiar with the composition of paroxetine hydrochloride could study the '393 and '479 patents and learn from them the means of preparing a novel dosage form of paroxetine hydrochloride which made it available to a patient in a controlled-release fashion.

[33]            There is an ambiguity in the language of paragraph 4(2)(b) of the NOC Regulations. Does the expression "a claim in the patent for the medicine itself when prepared or produced by the methods or processes of manufacture particularly described and claimed" refer to the manufacture of the medicinal ingredient only, or does it refer to the fabrication of any combination of the medicinal ingredient and other substances? On the facts of this case, the question is whether the patents must teach a method or process for the manufacture of paroxetine hydrochloride itself, or is it sufficient if they teach a method or process for the manufacture of a controlled-release version of paroxetine hydrochloride?

[34]            In Hoffman-La Roche Ltd. v. Canada (Minister of National Health and Welfare) (1995), 97 F.T.R. 288 aff'd (1995), 67 C.P.R. (3d) 25 (C.A.) (Hoffman-La Roche Ltd.), this Court decided that a claim to a medicine included a formulation claim, i.e., a claim in which the medicine is combined with other substances. In that case, four patents on the Register were challenged on the ground that they did not contain a claim to the medicine itself but rather to the medicine in combination with other compositions. Two of the patents in issue were in relation to a hemihydrate, polymorphic and crystalline form of a known composition, flunisolide, which forms yielded a pharmaceutically acceptable formulation for nasal sprays. As in this case, the patents taught a combination of a known medicine with other elements to produce a new drug. Unlike this case, there were specific references to the medicinal element in the patents in issue.

[35]            This case more closely resembles Pfizer Canada Inc. v. Canada (Attorney General), (2004) 251 F.T.R. 195, 2004 F.C. 370 (Pfizer Canada Inc.), where the patent in issue taught a method of constructing a tablet which allowed for controlled release of the drug formulation to the patient after a certain time interval had passed following ingestion. Claim 1 of the patent began by saying: "A dosage form for use in a method of administering a drug to the gastrointestinal tract of a warm-blooded animal, which form is characterised by: - ..." followed by a description of the six components of the dosage form, including "...a particular dose amount of a drug for producing therapeutic effect". Other claims listed twenty-seven drugs which could be administered using the patented system, including the drug verapamil. The Minister refused to list the patent in relation to Pfizer's drug Chronovera, whose active ingredient was verapamil hydrochloride, because he determined that the patent did not contain a claim for the drug verapamil hydrochloride itself, or for the use of the drug.

[36]            After carefully reviewing the parties' submissions, the legislation and the terms of the patent itself, Mosley J. concluded:

44. In my opinion, the '376 patent does not claim protection for the "medicine" found in Pfizer's CHRONOVERA tablet. Rather, I construe the patent as being for a delivery system for the administration of any one of the 27 listed drugs, including verapamil hydrochloride. The "essential elements" of this invention are the features of the dosage form that allow for the time-varied delivery of various drugs. The delivery system is protected, not the listed drugs.

[37]            Counsel for GSK sought to distinguish Pfizer Canada Inc. by placing it within the line of cases dealing with medical devices such as Glaxo Group Ltd. v. Novopharm Ltd. (1999), 244 N.R. 199 (F.C.A.) (inhalers) (Glaxo Group Ltd.), Novartis Pharmaceuticals Canada Inc. v. Canada (Minister of Health) (2003) F.C.A. 299, [2003] F.C.J. No. 1065 (Estracomb patches) (Novartis Pharmaceuticals Canada Inc.) and Eli Lilly Canada Inc. v. Canada (Minister of Health), (2002) 225 F.T.R. 110, 2002 FCT 1248 (Compudose implants). According to counsel, the characteristic which divides these cases from the formulation cases is that in the "device" cases, the medicinal ingredient is not mixed with other ingredients, whereas in the formulation cases, the active ingredient is mixed with other ingredients so as to produce the desired result.

[38]            This distinction is drawn from the trial level decision in Glaxo Group Ltd. v. Novopharm Ltd. (1998), 144 F.T.R. 252 (Glaxo Group Ltd.) where Tremblay-Lamer J. compared the inhaler in her case to the nasal spray in Hoffman-La Roche Ltd.:

[11] In my opinion, the Applicants' interpretation of Hoffmann-La Roche is incorrect. The case specifically deals with compositions of active and inactive ingredients, such as coatings and inert carriers, which are physically mixed together and which are ingested into the body as a single composition. It was obviously not intended to include a mechanical device as an inactive ingredient.

...

[13] Therefore, in my view, the Regulations do not extend the meaning of the word "substance" to include mechanical devices made of metal and plastic which cannot be ingested into the body. Consequently, I find that the Respondent's allegation is justified.

[39]            The reasoning in Glaxo Group Ltd. can also be found in Novartis Pharmaceuticals Canada Inc. where the Court distinguished the patches in issue in that case from the nasal spray in issue in Hoffman-La Roche Ltd. on the basis that, because the active ingredients and the other elements of the patches were not mixed together, they did not constitute a formulation. See paragraph 20.

[40]            The following observation in Pfizer Canada Inc. suggests that Mosley J. had the device cases in mind in coming to the conclusion he did:

[51] ... the active ingredients in this case, the verapamil hydrochloride, is not mixed together with various inactive ingredients or excipients but rather the design of the tablet structure, as set out in the claim 1 of the patent, allows for "discrete and separate components" of drug free layers and the dose of the drug itself".

[41]            On the strength of this reasoning, and the decision of this Court in Hoffman-La Roche Ltd., GSK asks us to conclude that since the active ingredient paroxetine hydrochloride is mixed with the other compounds described in the claims, and since the entire preparation is ingested by the patient, the patent protects a tablet which is administered to the patient, and not a device which administers a medicine. The failure to specifically identify paroxetine hydrochloride in the claims is a gap which can be filled by the knowledge of the person skilled in the art who, according to Dr. McGinity, would recognize that "active substance" includes paroxetine hydrochloride. See Burton Parsons Chemicals Inc. Following this reasoning, the '393 and the '479 patents contain a claim for the medicine itself and are eligible for listing on the register of patents.

[42]            The difficulty with this reasoning is that it ignores the nature of the '393 and '479 patents. It is clear that these patents are designed to protect the system by which a great number of compounds, be they pesticide, herbicide, medicament, or room deodorizer, can be released into an aqueous fluid in a controlled manner. The "active substances" referred to in the patents are nothing more than the payload carried by the delivery system protected by the patents.

[43]            If one reviews the "medical devices" cases referred to above, one notes that the theme which runs through them all is the dichotomy between the delivery system and its payload. The attempts to define "claim for the use of the medicine itself" on the basis of whether the ingredients are mixed, or the presence of physical devices, all point to a more fundamental distinction between a delivery system and that which is delivered by that system. The distinction articulated in Glaxo Group Ltd. (C.A.) between devices for the administration of medicaments and the medicaments which are themselves administered is another way of expressing the difference between delivery system and payload. But, as this case shows, the distinction is more difficult to make when a tablet is both the thing administered and that which administers the drug. The distinction between delivery system and payload bridges both types of tests by focussing on the substance of the patent. Does the patent protect the delivery system or does it protect the payload?

[44]            If the patent protects the delivery system, then it does not contain a claim for the medicine itself, or the use of the medicine, even if it contains a reference to the medicine as payload. For example, in Janssen-Ortho Inc. v. Canada (Minister of Health), (2003) 229 F.T.R. 268, 2003 FCT 286, confirmed by [2004] F.C.J. No. 242 (F.C.A.), leave to appeal to the SCC denied, August 26, 2004, the patent protecting a patch type device specifically refers to the medicine fentanyl but in the context of "a 'transdermal system' or 'device' designed for the transdermal administration of fentanyl...". Heneghan J. found that it was not eligible for listing, in spite of the specific reference to a medicine. She found that the patent described a patch and that the patent contained no claim for the medicine itself. I take the latter ground to be a recognition that a claim for a medicine as payload is not a claim for the medicine itself.

[45]            In the end result, I would dismiss this appeal, with costs to the respondents.

                                                                      (s) " J.D. Denis Pelletier"     

                                                                                                                                                                        J.A.

                          FEDERAL COURT OF APPEAL

    NAMES OF COUNSEL AND SOLICITORS OF RECORD

DOCKET:                  A-686-04

Appeal from an Order of the Honourable justice von Finckenstein dated December 10, 2004,

in file T-834-04

STYLE OF CAUSE: GLAXOSMITHKLINE INC. v. ATTORNEY GENERAL OF CANADA AND THE MINISTER OF HEALTH

PLACE OF HEARING:                                 Ottawa, Ontario

DATE OF HEARING:                                   March 8, 2005

REASONS FOR JUDGMENT:                    DESJARDINS J.A.      

CONCURRED IN BY:                                  NADON J.A.

CONCURRING REASONS BY:                  PELLETIER J.A.         

DATED:                     May 24, 2005               

APPEARANCES:

SOLICITORS OF RECORD: