Novopharm Ltd. v. AstraZeneca AB, 2003 FC 1212

Date: 20031017

Dockets: T-1470-99

T-1471-99

Citation: 2003 FC 1212

Ottawa, Ontario, Friday, the 17^th day of October 2003

PRESENT:        The Honourable Madam Justice Dawson

BETWEEN:

                                                            NOVOPHARM LIMITED

                                                                                                                                                       Applicant

                                                                              - and -

ASTRAZENECA AB and

THE REGISTRAR OF TRADE-MARKS

                                                                                                                                               Respondents

                                               REASONS FOR ORDER AND ORDER

DAWSON J.

[1]                 These are appeals brought by Novopharm Limited ("Novopharm") pursuant to section 56 of the Trade-marks Act, R.S.C. 1985, c. T-13 ("Act") from two decisions of the Registrar of Trade-marks ("Registrar") made on June 14, 1999. In each decision the Registrar rejected Novopharm's opposition to an application made by AstraZeneca AB, or its predecessor (together referred to as "Astra") to register a trade-mark in relation to tablets containing felodipine. Felodipine is a pharmaceutical preparation for use in the treatment of hypertension. One trade-mark application was in relation to tablets that are pink, round and biconvex in shape (5 mg dosage) and the other was in relation to tablets that are red-brown in colour, round and biconvex in shape (10 mg dosage).

[2]                 In each case the Registrar rejected Novopharm's opposition on the basis that Novopharm's Statement of Opposition was insufficiently detailed to enable Astra to respond to Novopharm's assertion that Astra's trade-marks are not distinctive. Alternatively, the Registrar found that "there is no quantitative evidence, and insufficient qualitative evidence, of sales or use of any pharmaceutical tablets which have a colour and shape combination resembling the applied for mark".

[3]                 On appeal to this Court, the Registrar's decisions were initially upheld on the primary ground. See: (2001), 13 C.P.R. (4^th) 61. On further appeal to the Court of Appeal, the appeals were allowed and the matters were remitted to the Trial Division for determination on the merits. See: (2002), 21 C.P.R. (4^th) 289. In its reasons remitting the matter to the Trial Division, the Court of Appeal found that:

(i)          The Registrar committed a reviewable error in failing to consider the evidence in assessing whether the applicant knew the case it had to meet.

(ii)         The evidence filed with the Registrar cured any insufficiency in the pleadings.

(iii)        With respect to the Registrar's secondary basis for rejecting the opposition, new evidence was filed in the Trial Division. This included qualitative evidence about the names, manufacturers and uses of products and quantitative evidence of the number of prescriptions dispensed by affiant pharmacists and the annual sales of specific pink and red-brown products. This evidence would have affected the decisions of the Registrar so that the review of the Registrar's secondary basis for rejecting Novopharm's oppositions should be a review on the standard of correctness.

[4]                 Additional memoranda of argument were filed by the parties on the re-hearing in the Trial Division. While Astra acknowledged in its written argument that the Court of Appeal had determined that Novopharm had met its evidential burden, in oral argument Astra argued in the red-brown case that if Novopharm is permitted only to rely upon tablets pleaded in its statement of opposition, there was no evidence with respect to sales of the one drug named in the statement of opposition (Apo-Indapamide). It was therefore said that the Court of Appeal failed to consider "whether the quantitative evidence was sufficient". In reply, Novopharm tendered a copy of the factum filed by Astra on the appeal to the Court of Appeal. In its factum Astra advanced to the Court of Appeal the same argument now advanced that Novopharm did not meet its evidential burden in the red-brown case. In that circumstance, the conclusion of the Court of Appeal that Novopharm met its evidential burden with respect to the red-brown case is dispositive of the issue.

[5]                 It follows that what is to be determined in this proceeding is whether Astra has met its burden to establish that the proposed trade-marks were distinctive as of the date of opposition. This turns upon the factual question as to whether as of the date of opposition, tablets marketed in an appearance similar to Astra's 5 mg and 10 mg tablets render Astra's marks non-distinctive and thereby preclude registration of the trade-mark.

[6]                 The term "distinctive" is defined in section 2 of the Act in the following terms:

[7]                 As the Court of Appeal wrote in AstraZeneca AB v. Novopharm Ltd., 2003 FCA 57 at paragraph 16:

[...] A mark actually distinguishes by acquiring distinctiveness through use, resulting in distinctiveness in fact. A mark that is "adapted so to distinguish" is one that does not depend upon use for its distinctiveness because it is inherently distinctive. A coined or invented word mark falls into this category: Standard Coil Products (Canada) Ltd. v. Standard Radio Corp., [1971] F.C. 106 (T.D.), at 115; The Molson Companies Limited v. Carling O'Keefe Breweries of Canada Limited, [1982] 1 F.C. 175 (T.D.), at 278-79.

[8]                 Principles to be applied when considering this issue are:

1.          The trade-mark applicant must satisfy the tripartite test enunciated in Phillip Morris v. Imperial Tobacco Ltd. (1985), 7 C.P.R. (3^d) 254 (F.C.T.D.) at page 270. See: AstraZeneca v. Novopharm, supra at paragraph 19. The third part of the tripartite test requires that the association between the mark and the product enables the owner of the mark to distinguish his product from that of others.

2.          Colour alone has not been viewed as being inherently distinctive. See: AstraZeneca v. Novopharm, at paragraph 18.

3.          Proof of actual distinguishment is not an easy burden to discharge. See: AstraZeneca v. Novopharm, at paragraph 20.

4.          Where the active ingredient in the pharmaceutical product is not claimed as the trade-mark, and the trade-mark sought to be registered is the colour and shape of the tablet, the applicant must show that the colour and shape distinguishes the tablet from the tablets of other manufacturers. See: AstraZeneca v. Novopharm, at paragraph 22.

5.          It is incumbent on the trade-mark applicant to show that physicians, pharmacists or patients can and do use the proposed trade-mark in choosing whether to prescribe, dispense or request the product. See: Novopharm Ltd. v. Astra Aktiebolag (2000), 6 C.P.R. (4^th) 16 (F.C.T.D.); aff'd (2001) 15 C.P.R. (4^th) 327 (F.C.A.).

6.          It is not fatal to an application that consumers may also use means other than the mark for identifying the product with a single source. As Mr. Justice Evans, as he then was, wrote in Novopharm Ltd. v. Bayer Inc. (1999), 3 C.P.R. (4^th) 305 at paragraph 79; aff'd (2000) 9 C.P.R. (4^th) 304 (F.C.A.):

[...] Thus, while pharmacists rely mainly on the brand name and other identifying indicia on the stock bottles and packaging containing the product, or the inscription on the tablets, which is not part of the mark, if there is evidence that to any significant degree they also recognized the product by its appearance (excluding the markings on the tablet because they are not part of the mark), this may be sufficient to establish the distinctiveness of the mark.

[9]                 In the present case, Astra argues that the colour and shape of its tablets actually distinguish the wares from the wares of other manufacturers (that is, that the proposed mark has acquired distinctiveness), and also that the colour and shape are adapted so as to distinguish the wares (that is, that the proposed marks are inherently distinctive).

[10]            Because the marks were applied for on the basis of proposed use, and because there is no evidence of any use of the pink tablets as at the date of opposition, it is not open to Astra to assert that the proposed trade-mark in respect of the pink tablets has acquired distinctiveness. The application therefore in respect of the pink tablets rests upon whether this proposed mark is inherently distinctive.

[11]            Applying the principles set out above and carefully considering the evidence before me, I am unable to conclude that the colour and shape of Astra's 5 mg and 10 mg felodipine tablets are inherently distinctive, or that the red-brown mark has acquired distinctiveness so as to distinguish the wares from the wares of others.

(i) Inherent Distinctiveness

[12]            In arguing that the colour and shape of the proposed marks are inherently distinctive, Astra argues that its marks are arbitrary and that arbitrary marks are inherently distinctive. The marks are said to the arbitrary in that the colour and shape are not dependent upon the active ingredient contained in the tablets, and that the colour and shape were chosen for marketing reasons and to be distinctive.

[13]            A similar argument was rejected by the Court of Appeal in AstraZeneca v. Novopharm, supra in respect of Astra's proposed trade-mark for 2.5 mg yellow tablets containing felodipine. While the question whether a mark is inherently distinctive is a question of fact, I am similarly satisfied in this case that the evidence fails to establish that the proposed marks are inherently distinctive. The red-brown and pink colours and the round biconvex shape are neither unique nor unusual. There is nothing novel or singular about the appearance of the tablets in the sense that, like an invented word, the mark is inherently distinctive. It is insufficient for Astra to show that the trade-mark is independent of the active ingredient and was chosen for marketing reasons. By itself that evidence does not establish that the mark is distinctive by its very nature.

[14]            I therefore find on the evidence that Astra has not established on a balance of probabilities that the colour and shape of its tablets has the effect of distinguishing Astra's tablets from the tablets of other manufacturers.

(ii) Acquired Distinctiveness

[15]            At the outset, it is necessary to consider whether the trade-mark must distinguish Astra's 10 mg felodipine from:

(i)          the felodipine of its competitors that are interchangeable with Astra's felodipine;

(ii)         all pharmaceuticals in the same therapeutic class, that is all tablets used to treat hypertension; or

(iii)        all pharmaceutically active ingredients available, even non-competing ones.

[16]            Astra argues that the relevant market is limited to tablets containing felodipine that are interchangeable with Astra's felodipine.

[17]            This issue was considered by the Court of Appeal in AstraZeneca v. Novopharm, supra where the Court of Appeal rejected Astra's submission that the relevant market place was felodipine tablets. At paragraph 22, Mr. Justice Stone, for the Court, wrote that:

Nor would the evidence appear to establish that the combination of colour and shape of the appellant's tablets had the effect of "actually distinguishing" the appellant's wares from those of others. Counsel points out that as the appellant's tablets were the only hypertensive prescription drug in the Canadian market place that contained the active ingredient "felodipine", it readily distinguishes that drug from other prescription drugs because none of the others relied upon contained that active ingredient. There was thus no possibility of some other drug being substituted for the PLENDIL 2.5 mg tablet. Indeed, "felodipine" is identified in the trade-mark application as the "wares" in association with which the trade-mark had been used in Canada since 1994. The appellant maintains from this that both the Registrar and Kelen J. erred in this respect by expanding the relevant market to all round and yellow tablets for the treatment of hypertension rather than restricting it to "felodipine" wares. Indeed, the respondent adduced some evidence of other non anti-hypertensive yellow and round tablets in the Canadian pharmaceutical market, and asserts that the relevant comparison market is all pharmaceutical pills including other yellow and round anti-hypertensive tablets. However, it is to be noted that the active ingredient as such is not claimed by the appellant as the trade-mark. The trade-mark sought to be registered is the colour and shape, or appearance, of the 2.5 mg tablets that happens to contain the active ingredient. In order to bring its application within this branch of the "distinctive" test in section 2, the appellant had, therefore, to show that through use over time the colour and shape of its tablets actually distinguishes them from tablets of other manufacturers.

[18]            Astra argues that a different conclusion should be reached in this case because distinctiveness is an issue of fact and because the analysis of the Court of Appeal "was flawed, in confusing the 'wares' with the 'trade-mark'". I am not prepared to depart from the conclusion of the Court of Appeal. While distinctiveness is essentially an issue of fact, the conclusion of the Court of Appeal in the above quoted paragraph, in my view, is not simply a conclusion of fact.

Where the active ingredient as such is not claimed in a trade-mark, the Court of Appeal has held that the applicant must show that through use over time the colour and shape of its tablet actually distinguishes it from other manufacturers' tablets. This conclusion is, in my view, binding upon me.

[19]            Turning to the evidence, no evidence was provided from patients, and Astra conceded that physicians do not pay much, if any, attention to the colour and shape of pills when dispensing medication. Astra, however, relied upon evidence that pharmacists use the shape and colour of the red-brown pills when choosing whether to dispense Astra's 10 mg felodipine tablets. Specifically, Astra relied upon evidence that pharmacists know the colour and shape of a product, and that they rely upon colour, shape and/or size to ensure that the correct brand is dispensed. If a pharmacist saw, for example, a green tablet he or she would know the tablet could not be Astra's felodipine. While Astra's felodipine tablets are dispensed in a box which contains the tablets in blister packs, a pharmacist must verify the number of tablets which are dispensed. The evidence is that, in so doing, the pharmacist "can't help but see what the colour and the shape of the Plendil tablets are".

[20]            However, the evidence filed in this Court establishes, in my view, that at the relevant time there were many red-brown pills in the market. A pharmacist confirmed on his re-examination that Astra's 10 mg felodipine is the same colour as LOSEC 20 mg, such that he told the difference between the two pills by their markings. Other red brown (or dark pink) pills on the market include:

Adalat - XL - (dark pink and round - launched in 1992) - sales for August 1993/July 1994 - $34.5 million

Novo-Ferrosulfate ECT - (dark pink and round - launched in 1966) sales for August 1993/July 1994 - $200,000

Premarin .625 mg - (red-brown and roundish - launched in 1942) - sales for August 1993/July 1994 - $10.7 million

Voltaren SR 100 mg - (red-brown and round - launched in 1985) - sales for August 1993/July 1994 - $21.7 million

Zestril 10 mg and 20 mg (dark pink and round - launched 1990) - sales for August 1993/July 1994 - $14.7 million.

[21]            While Astra objected to reference to this evidence, I am satisfied that this evidence is properly before the Court. Novopharm's statement of opposition was express that its opposition was not limited to the tablets of three specific manufacturers listed in the statement of opposition. On the basis of the evidence filed by Novopharm, Astra at all times knew the case it had to meet.

[22]            Further, I am satisfied on the evidence that pharmacists do not dispense felodipine to a significant degree on the basis of colour and/or shape. The appearance of colour and shape is simply one item, a secondary check, which a pharmacist will consider. I accept the evidence that pharmacists primarily identify tablets by their Drug Identification Number, their markings, and the labels on the packaging. I accept the submission of Novopharm that it is insufficient that a pharmacist would know that Astra's felodipine tablets are not green. The proper question is what does a red-brown pill mean to a pharmacist?

[23]            In sum, I find that the evidence fails to establish on the balance of probabilities that the shape and colour of the red-brown tablet actually distinguishes Astra's tablets from the tablets of other manufacturers.

CONCLUSION AND COSTS

[24]            Having concluded that Astra has failed to establish that any of physicians, pharmacists or patients can and do use the proposed trade-marks in choosing whether to prescribe, dispense or request Astra's felodipine 5 mg or 10 mg tablets it follows that the appeals will be allowed.

[25]            It further follows that it is not necessary to deal with the argument advanced by Novopharm in the red-brown case that Astra cannot rely upon a 1995 written license to establish that Hoechst-Roussel Canada Inc. is a licensee of the proposed trade-mark in respect of Hoechst's product RENEDIL. However, assuming, without deciding, that the issue is properly raised by Novopharm, I am satisfied for the reasons argued by Astra that Astra's evidence establishes that any felodipine tablets sold in Canada by Hoechst were manufactured by Astra pursuant to a trade-mark license agreement.

[26]            Novopharm is entitled to its costs in the Trial Division. If not agreed, they shall be assessed on the basis of the high-end of column III to Tariff B of the Federal Court Rules, 1998.

ORDER

[27]            IT IS HEREBY ORDERED THAT:

1.          The appeals from the decisions of the Registrar are hereby allowed and the oppositions granted.

2.          Trade-mark application numbers 699,917 and 699,918 are denied registration.

3.          The respondent AstraZeneca shall pay to the applicant its costs for all proceedings in the Trial Division. If not agreed, such costs shall be assessed on the basis of the high-end of column III to Tariff B of the Federal Court Rules, 1998.

                 "Eleanor R. Dawson"         

Judge

FEDERAL COURT

TRIAL DIVISION

NAMES OF COUNSEL AND SOLICITORS OF RECORD

DOCKETS:                  T-1470-99 and T-1471-99

STYLE OF CAUSE:          Novophram Limited v. AstraZeneca AB and

The Registrar of Trade-marks

PLACE OF HEARING:        Toronto, Ontario

DATE OF HEARING:         Monday, June 9, 2003

REASONS FOR ORDER AND ORDER OF

THE HONOURABLE MADAM JUSTICE DAWSON

DATED:                    October 17, 2003

APPEARANCES:

Warren Sprigings           APPLICANT

Lilly Sormaz

Gunars A. Gaikis            RESPONDENT (AstraZeneca AB)

Nancy P. Pei

SOLICITORS OF RECORD:

Hitchman & Sprigings       APPLICANT

Barristers and Solicitors

Toronto, Ontario

Smart & Biggar

Barristers and Solicitors

Toronto, Ontario            RESPONDENT (AstraZeneca AB)

Morris Rosenberg

Deputy Attorney General of Canada             RESPONDENT