Dockets: A-194-14
A-94-14
Citation:
2014 FCA 250
|
CORAM:
|
NOËL C.J.
TRUDEL J.A.
BOIVIN J.A.
|
|
BETWEEN:
|
|
APOTEX INC.
|
|
Appellant
|
|
and
|
|
PFIZER CANADA INC. and G.D. SEARLE & CO.
|
|
Respondents
|
|
AND BETWEEN:
|
|
MYLAN PHARMACEUTICALS ULC
|
|
Appellant
|
|
and
|
|
PFIZER CANADA INC., G.D. SEARLE & CO.
|
|
Respondents
|
REASONS
FOR JUDGMENT
NOËL C.J.
[1]
These are two appeals brought by Mylan
Pharmaceuticals ULC (A-94-14) and Apotex Inc. (A-194-14) (Mylan and Apotex, or
the appellants) from decisions of the Federal Court (2014 FC 38, the Mylan
decision, and 2014 FC 314, the Apotex decision), wherein Harrington J.
(the Federal Court judge) allowed applications brought by Pfizer Canada Inc.
and G.D. Searle & Co. (together, the respondent) and issued orders
prohibiting the Minister of Health (the Minister) from issuing notices of
compliance (NOC) in respect of celecoxib. These prohibition orders will cease
to have effect on November 14, when Canadian Patent No. 2,177,576 (the ‘576
Patent) which conveys a monopoly over this compound expires.
[2]
The main issue in both appeals turns on whether
the Federal Court judge properly held that the patent in issue did not promise
certain specified results thereby declining to hold that the appellants’
allegations of invalidity were justified by reason of the patent’s alleged
failure to procure these results.
[3]
The two decisions under appeal were rendered by
the Federal Court judge in separate reasons which however share common lines of
reasoning. For that reason, the two appeals were heard together. The reasons
which follow dispose of both appeals.
BACKGROUND
[4]
For over a century now, inflammation in humans
and certain animals has been treated using a particular class of
pharmaceuticals known as non-steroidal anti-inflammatory drugs (NSAIDs). These
drugs reduce inflammation by inhibiting a certain enzyme called cyclooxygenase
(COX).
[5]
In the 1970s, researchers began to notice that
NSAIDs can have dangerous side effects in the long-run, particularly in the
gastrointestinal (GI) tract, where bleeding, ulcers and perforations can take
place. The reason for this was determined to be that COX plays an important
role in maintaining many tissues, particularly in the GI tract. From that point
onwards, North American regulators required that NSAIDs be sold with a warning
label addressing these risks.
[6]
In the 1990s, researchers discovered that there
exist two different COX enzymes. It was hypothesized that, while COX-1 plays a
general maintenance role in many tissues, the body produces COX-2 in response
to injury, and it is this second enzyme that causes inflammation. The “COX-2
hypothesis” developed by NSAIDs researchers was that, if one could develop a
“COX-2 selective” drug that either inhibited COX-2 exclusively or at least
inhibited COX-2 significantly more than it did COX-1, then that drug could
reduce inflammation without producing the sort of side effects associated with
the set of NSAIDs on the market at the time.
[7]
It was in this context that the respondent
developed a new class of NSAID compounds, including celecoxib. The ‘576 Patent
was granted to the respondent with an effective filing date of November 14,
1994. The Minister subsequently added celecoxib to the Register of Patented Medicines
maintained under the Patented Medicines (Notice of Compliance) Regulations,
SOR/93-133 (the Regulations).
[8]
The claimed compounds, including celecoxib, were
new compounds as of the filing date. At issue are claims 4 and 8 to 13. Claim 4
claims celecoxib; claim 8 claims a therapeutically-effective amount of
celecoxib; claims 9 to 13 claim the use of celecoxib to treat inflammation and
other specified conditions or disorders (Mylan appeal book, vol. 1, pp. 304 to
305). Also at issue is claim 16 which is directed at the prevention of
colorectal cancer.
[9]
In the “Description of the Invention” section of
the ‘576 Patent’s specification, the inventor described the issue of side
effects in the class of NSAIDs on the market around the filing date, and wrote “(t)he compounds are useful as anti-inflammatory agents, such
as for the treatment of arthritis, with the additional benefit of having
significantly less harmful side effects”. In the next paragraph of that
section, the inventor stated that the invention “preferably includes” compounds
selectively inhibiting COX-2 over COX-1 and that
“(s)uch preferred selectivity may indicate an ability to reduce the incidence
of common NSAID-induced side effects” (Mylan appeal book, vol. 1, p. 119).
[10]
In 2007, the respondent was successful in
preventing the Minister from issuing another party an NOC in respect of
celecoxib. Novopharm Ltd. had made in respect of the ‘576 Patent allegations of
insufficiency, lack of utility, obviousness and abandonment. Though the Federal
Court (Hughes J.) ruled that the respondent had demonstrated that the first two
allegations were not justified, it held that the respondent had failed to make
that demonstration with respect to the last two. The respondent’s application
was accordingly dismissed (G.D. Searle & Co. v. Novopharm Ltd., 2007
FC 81, [2008] 1 F.C.R. 477 [Novopharm FC]). The decision was reversed on
appeal, where this Court upheld the conclusions on insufficiency and lack of
utility, but found that the respondent had also demonstrated that the
allegations of obviousness and abandonment were not justified (G.D. Searle
& Co. v. Novopharm Ltd., 2007 FCA 173, [2008] 1 F.C.R. 529 [Novopharm
FCA]).
[11]
Five years later, the respondent was again
called upon to defend its monopoly over celecoxib further to notices of
allegation (NOA) filed by Mylan and Apotex. Each of the two ensuing prohibition
applications filed by the respondent were heard by the Federal Court judge, who
allowed both in decisions issued on January 28, 2014 (the Mylan decision)
and on April 15, 2014 (the Apotex decision). These are the decisions now
under appeal.
THE MYLAN DECISION
[12]
Before the Federal Court judge, Mylan argued
that because the ‘576 Patent promised reduced side effects in humans and celecoxib
did not procure such reduced side effects, the ‘576 Patent was invalid for lack
of utility.
[13]
Mylan supported this contention by reference to
the language in the specification itself, particularly where it describes the
COX-2 hypothesis and where it states “(t)he compounds
are useful as anti-inflammatory agents … with the additional benefit of having
significantly less harmful side effects” (Mylan appeal book, vol. 1, pp.
118 to 119).
[14]
Mylan also relied on Hughes J.’s construction of
the ‘576 Patent in Novopharm FC. In describing the patent in general, he
stated “(a)fter some discussion, counsel for the
applicants conceded that both the anti-inflammatory properties and lesser side
effects were necessary to the utility of the claimed invention” (Mylan decision at para. 74, quoting Novopharm FC at
para. 14). In construing claim 4 of the patent specifically, he further stated
(Mylan decision at para. 74, quoting Novopharm FC at para. 27):
[No use of (celecoxib) is stated in that
claim but,] as conceded by counsel for the applicants, the utility of that
compound is set out in the specification as being the duality of treatment of
inflammation and reduction of unwanted side effects such as ulcers of the
gastrointestinal system.
[15]
Relying on the above passages, Mylan argued
before the Federal Court judge that because the respondent had conceded in Novopharm
FC that reduced side effects were necessary to establish celecoxib’s
utility, its attempt to argue against this very premise before him amounted to
an abuse of process.
[16]
The respondent resisted these allegations on
three alternative grounds, arguing that the ‘576 Patent did not promise reduced
side effects; that, if it had so promised, the promise did not extend to humans
and that, if the promise had so extended, celecoxib was in fact proven to have
reduced side effects in humans. The respondent placed great emphasis on the
equivocal nature of the specification statement that celecoxib “may” reduce
side effects.
[17]
The Federal Court judge accepted the
respondent’s first argument, ruling that the ‘576 Patent did not promise
reduced side effects. As to the effect of the prior decision in Novopharm FC,
he rejected Mylan’s contention that this decision bound him to construe the
patent so as to promise the utility of reduced side effects in humans. Citing
paragraphs 102 and 103 of that decision, he observed that Hughes J. found
demonstrated utility in that case (Mylan decision at para. 75). However,
it was not clear that reduced side effects in humans had been demonstrated (Mylan
decision at para. 77). The Federal Court judge added that he would have
been bound by Hughes J.’s decision had it “turned on” patent construction, as
this is a question of law (Mylan decision at para. 78). However, he went
on to find that as utility, whether demonstrated or predicted is a question of
fact, he was not bound by Hughes J.’s decision.
[18]
Turning to the construction of the patent, the
Federal Court judge accepted the respondent’s more limited construction of the
patent on two principal grounds. First, he found that the word “may” as it
appeared in the specification represented a clear indication that the patent
made no promise of reduced side effects. Whether read within the context of
standard statutory interpretation principles or from the perspective of a
skilled addressee, the word “may” could not be taken to imply anything more
than a possibility of reduced side effects (Mylan decision at para. 65).
[19]
Second, the Federal Court judge observed that
the claims themselves were devoid of any mention of reduced side effects.
Citing Federal Court jurisprudence, he held that uses which do not appear in
the claims specification ought to be considered as mere statements of
advantage, absent clear and unequivocal language promising such uses (Mylan decision
at para. 70, citing Fournier Pharma Inc. v. Canada (Health), 2012 FC 741,
[2012] F.C.J. No. 901 at para. 126 [Fournier]). He found further support
for the distinction between promises and statements of advantage or potential
use in the concurring opinions issued by this Court in Sanofi-Aventis v.
Apotex Inc. 2013 FCA 186, [2013] F.C.J. No. 856 (Leave to Appeal to SCC granted
on January 30, 2014, 35562) [Plavix FCA] (Mylan decision at
paras. 68 and 69).
THE APOTEX DECISION
[20]
Before the Federal Court judge, Apotex made two
submissions very similar to those made by Mylan. First, it argued that the
patent promised reduced side effects in humans and that such utility could now
be proven not to have been achieved. Second, it argued that, because the
respondent conceded before Hughes J. in Novopharm FC that utility
necessarily included reduced side effects, it would constitute an abuse
of process for the respondent to dispute that premise before the Federal Court
judge.
[21]
Apotex advanced a number of additional arguments.
First, it took the position that utility in respect of celecoxib’s use as an
anti-inflammatory in humans was neither demonstrated nor soundly predicted.
Though Apotex conceded that celecoxib had since been proven to be so useful, it
asserted that the respondent could neither demonstrate such utility nor provide
a sound basis for predicting it, as of the filing date.
[22]
Second, Apotex argued on the basis of language
in the specification as well as that in claim 16 that the patent had promised
the utility of preventing colorectal cancer and that this utility was neither
demonstrated nor soundly predicted as of the filing date.
[23]
Finally, Apotex attacked the ‘576 Patent for
insufficiency of disclosure. The essence of this argument was that the
respondent buried its “true invention”, namely the use of celecoxib for
treating inflammation, among a smattering of other compounds and claims which
it knew in fact to be useless or unfounded. For instance, one particular
compound claimed in claim 5 was known to be toxic at the filing date, and was
therefore useless.
[24]
The Federal Court judge rejected these
contentions. On the question of utility in treating inflammation, he accepted
the respondent’s argument that rats could be considered to constitute
“subjects” and that, to the extent that the patent had promised to treat
inflammation in a subject, this promise had been demonstrated to have been met
(Apotex decision at paras. 28 and 29, citing Plavix FCA and Mylan
Pharmaceuticals ULC v. Pfizer Canada Inc., 2012 FCA 103, 2012 F.C.J. No.
386 [Donepezil FCA]).
[25]
On the question of utility in treating side
effects in humans, the Federal Court judge rejected Apotex’ argument on the
basis of a revised version of his reasons in the Mylan decision. He
conceded the “inappropriate” nature of some of his justifications in that
decision for ruling that the ‘576 Patent did not promise reduced side effects
in humans, namely his discussion at paragraph 44 of the principle that what is
not claimed is generally disclaimed (Apotex decision at paras. 30 and
36).
[26]
The Federal Court judge concluded, however, that
his earlier ruling was nevertheless correct, and reiterated that the side
effects statements were excluded from the specification’s claims, that the law
generally presumes such statements to be aimed at advantages (as opposed to
promises) and that the word “may” as it appears in the specification reflected
a critical degree of equivocation.
[27]
Further, he rejected Apotex’ argument that the
less equivocal side effects statement (“…with the additional
benefit of … significantly less harmful side effects”) referred to one
set of side effects (i.e. harmful side effects) while the more equivocal statement (“… may indicate an ability to reduce the incidence of common
… side effects”) referred to another (i.e. common side effects). Finally, he rejected Apotex’s
invitation to apply an English case constructing the European celecoxib patent
such that its utility included reduced side effects. In essence, he reasoned
that English patent law varies from Canadian patent law in a number of areas,
including questions of utility and, more specifically, promise.
[28]
The Federal Court judge concluded that part of
his analysis by highlighting what was demonstrated by the evidence before him (Apotex
decision at para. 41):
What the in vitro tests demonstrated was that the compounds
tested were COX II selective. That might have led the inventors to hope that
eventually it would be established that this COX II selectivity equated with
reduced side effects. Perhaps, they could have made a promise, but they did
not. Consequently, it is not necessary to decide whether or not the tests as
set out in the record establish that Celebrex® has fewer side effects. Pfizer
does not have to meet a promise it never made.
[29]
Concerning the claim regarding prevention of
colorectal cancer, the Federal Court judge concluded that Apotex had provided
evidence of the claim’s invalidity, but agreed with the respondent that, under
section 58 of the Patent Act, R.S.C., 1985, c. P-4 (the Act), this
claim could be severed from the rest, and that the remaining claims could
support the prohibition order sought by the respondent.
[30]
The Federal Court judge also rejected Apotex’
submissions on insufficiency. Although the compound claimed in claim 5 was indeed
known to be toxic at high doses, it was nevertheless an effective
anti-inflammatory, and the respondent could not be seen to have promised that
the compound would receive regulatory approval.
[31]
With respect to the question whether the
respondent had “obscured” its true invention, the Federal Court judge
distinguished the case at bar from Teva Canada Ltd. v. Pfizer Canada Inc.,
2012 SCC 60, [2012] 3 S.C.R. 625 [Teva], where the Supreme Court of
Canada found that the patentee, in claiming two different compounds while
knowing that only one was effective, had done exactly that. The Federal Court
judge explained that in the case at bar, each of the three compounds claimed
had been demonstrated to reduce inflammation (Apotex decision at para.
59). Though celecoxib may have been the central compound in the respondent’s
commercial designs, such designs need not be disclosed, and the ‘576 Patent’s
“true invention” remained a class of compounds including celecoxib, rather than
celecoxib alone.
[32]
The Federal Court judge finally rejected Apotex’
claim of abuse of process, observing that the construed utility in Novopharm
FC had been demonstrated on the basis of proven COX-2 selectivity and not
on the basis of the reduction of side effects in humans (Apotex decision
at para. 62). He further explained that patent construction is a matter of law,
and the court is not bound by a party’s concession or admission on such
questions (Apotex decision at para. 61).
POSITION OF THE APPELLANTS
[33]
Before this Court, Apotex advances four distinct
arguments, namely lack of utility in treating inflammation in humans, lack of
utility in reducing side effects, lack of utility in preventing colorectal
cancer and insufficient disclosure. Though it frames matters slightly
differently, Mylan’s submissions all go to the second of these four arguments. The
following is a joint summary of the submissions made by the appellants.
[34]
In arguing lack of utility in treating
inflammation, Apotex argues that the ‘576 Patent promised to treat inflammation
in humans and that, as of the filing date, it could only be demonstrated that
celecoxib could reduce inflammation in rats.
[35]
Both in stating the utility of its invention and
in claiming its monopoly, the ‘576 Patent speaks of treatment in “a subject”.
According to Apotex, this subject must be understood to include humans, as the
disorders in respect of whose treatment the invented compounds are described
and claimed are all suffered by humans and some of them are suffered
exclusively by humans. No person of skill in the art (skilled person), asserts
Apotex, would take the view that the respondent intended to monopolize the use
of the claimed compounds to treat a given set of disorders in a group of
“subjects” incapable of suffering the disorders in question (Apotex memorandum
at paras. 64 and 65). Furthermore, a patent which lays claim to a particular
use necessarily includes a promise in respect of that use (Apotex memorandum at
para. 67, citing Bauer Hockey Corp. v. Easton Sports Canada Inc., 2010
FC 361, [2010] F.J.C. No. 431 at para. 289 [Bauer FC], aff’d 2011
FCA 83, [2011] F.C.J. No. 331 and Apotex Inc. v. Wellcome Foundation Ltd.,
2002 SCC 77, [2002] 4 S.C.R. 153 at para. 92 [AZT]).
[36]
Apotex further argues that the Federal Court judge
erred in citing two cases from this Court to justify rejecting the
interpretation advocated by Apotex. First, his reliance on Plavix FCA was
misplaced, as the patent in issue in that case contained no claims of
treatment, let alone treatment of human disorders (Apotex memorandum at para.
70). In Donepezil FCA, neither the claims nor the testing mentioned
humans, but the promised utility was construed to include treatment of humans
on the basis that the specification and claims mentioned treatment of a human
disorder (Apotex memorandum at para. 71).
[37]
Because the Federal Court judge held that the
respondent had neither demonstrated nor been able to soundly predict treatment
of inflammation in humans, a finding of invalidity for lack of utility would
result automatically if this Court were to conclude that the word “subject” in
the ‘576 Patent extends to humans.
[38]
In arguing lack of utility in reducing side
effects, the appellants both take the position that Novopharm FC was
binding on the Federal Court judge so that he had to find a promise of reduced
side effects in humans. If the Federal Court judge wanted to depart from this
earlier construction, he had to justify that departure, whether on the basis of
an error in the earlier construction or because of distinct evidence (Mylan
memorandum at para. 43, citing Apotex Inc. v. Allergan Inc., 2012 FCA
308, [2012] F.C.J. No. 1467 [Allergan] at paras. 48 and 51; Apotex
memorandum at para. 97). As the Federal Court judge provided no such
justification, he erred in not following Hughes J.’s construction.
[39]
Apotex further asserts that an innovator cannot
relitigate an issue “with additional evidence it chose
not to adduce” in earlier proceedings to which it was a party (Apotex
memorandum at para. 88 citing Sanofi-Aventis Inc. v. Novopharm Ltd., 2007
FCA 163, [2008] 1 F.C.R. 174 at para. 50 [Ramipril FCA]). Nor
can an innovator accept and reject the same position in different proceedings
under the Regulations in respect of the same patent (Apotex memorandum
at para. 89, citing Apotex Inc. v. AstraZeneca Canada Inc., 2012 FC 559,
[2012] F.C.J. No. 621 [Omeprazole FC] at paras. 137 and 138, citing Johnson
v. Agnew, [1980] A.C. 367 (HL)).
[40]
Finally, Mylan submits that, in limiting its NOA
to the issue of utility, it expressly relied on the construction of the ‘576
Patent rendered in Novopharm FC and left undisturbed in Novopharm FCA
(Mylan memorandum at paras. 37 and 41).
[41]
Turning to the Federal Court judge’s own
construction of the ‘576 Patent, Mylan cites two decisions of this Court in
which a promise was found to extend to reduced side effects (Mylan memorandum
at paras. 55 and 56, citing Eli Lilly Canada Inc. v. Novopharm Limited,
2010 FCA 197, [2012] 1 F.C.R. 349 [Olanzapine] at paras. 27 and 99 and Apotex
Inc. v. Pfizer Canada Inc., 2011 FCA 236, [2011] F.C.J. No. 1234 [Latanoprost])
and two decisions in which a promise was found not to so extend (Mylan
memorandum at paras. 58 and 59, citing Plavix FCA at para. 67 and Mylan
Pharmaceuticals ULC. v. AstraZeneca Canada Inc., 2012 FCA 109,
[2012] F.C.J. No. 422 [Anastrozole] at paras. 6, 22, 29 and 30). Mylan
argues that the ‘576 Patent is “qualitatively more
similar” to the patents in the first set of cases (Mylan memorandum at
para. 60).
[42]
Mylan also argues that, in relying on the
testimony of the respondent’s expert, Dr. Young, to support his construction,
the Federal Court judge failed to follow this Court’s guidance in Olanzapine.
While that case made it clear that the judge is required to analyze the
expert testimony, the Federal Court judge merely “parlayed
and ‘broadly agreed’” with the perspective offered by Dr. Young (Mylan
memorandum at paras. 75 and 76). Had the Federal Court judge truly analyzed
this testimony, he would have noticed multiple statements supporting the view
that, at the filing date, the chief goal of NSAID researchers was to develop a
COX-2 selective drug with reduced side effects (Mylan memorandum at para. 77).
[43]
Mylan further argues for the first time on
appeal that, even if the Court finds that the promise of the ‘576 Patent
excludes side effect superiority, it must find that it at least included COX-2
selectivity distinctly higher than existing NSAIDs (Mylan memorandum at para.
85, citing the Apotex decision at paras. 38 to 42, 59 and 62).
[44]
Finally, consistent with its argument that the
‘576 Patent did promise reduced side effects, Mylan contends that celecoxib can
be shown not to have achieved this result. Specifically, Mylan points to the
refusal by the North American regulators to conclude that celecoxib possesses a
side effect advantage over pre-existing NSAIDs (Mylan memorandum at paras. 98
to 108) and alleges that the respondent itself cannot provide studies
supporting any side effect superiority unless it presents them in a misleading
fashion (Mylan memorandum at paras. 109 to 116). Mylan contends that if the
Court construes the ‘576 Patent as promising an enhanced degree of COX-2
selectivity, celecoxib can be shown to be no more than slightly more selective
than pre-existing NSAIDs (Mylan memorandum at paras. 117 to 122).
[45]
For its part, Apotex makes two unique submissions
attacking the Federal Court judge’s conclusion that the ‘576 Patent’s promise did
not extend to reduced side effects.
[46]
First, Apotex reiterates its proposed
distinction between the harmful side effects to which the
specification’s less qualified statement on side effects would have been
directed and those common side effects at which the specification’s more
qualified statement would have been directed. Furthermore, it attacks the
Federal Court judge’s rejection of this argument. Although the Federal Court
Judge stated that none of the experts saw this distinction, patent construction
is a matter for the Court alone to decide (Apotex memorandum at para. 83,
citing Plavix FCA at para. 33).
[47]
Second, Apotex advances a new distinction
between these two statements on side effects, suggesting that the less
qualified statement set out an explicit promise that the ‘576 Patent’s COX-2
selective compounds would have less side effects than other NSAIDs existing at
the time (Apotex memorandum at para. 76). The more qualified statement was not
directed at comparing the invention with other NSAIDSs, but rather at comparing
different subsets of the compounds disclosed in the specification. Thus, while
the first statement went to the utility of the invention, the second one went
merely to issues of relative safety among the invented compounds. The word
“may” in the latter only reflects uncertainty as to whether relative COX-2
selectivity among those compounds would result in relative side effect
superiority (Apotex memorandum at paras. 78 and 79).
[48]
In arguing lack of utility in preventing
colorectal cancer, Apotex argues that the Federal Court judge improperly
considered its argument on this point as going to insufficiency of disclosure,
when Apotex had argued the point in respect of utility (Apotex memorandum at
para. 73). The patent explicitly promised utility “for the prevention of
colorectal cancer”, however, and monopolized this use of the disclosed
compounds through claim 16 (Apotex memorandum at para. 72). Where a promise is
made in a patent, that promise is “overarching and
inherent to the invention and thus all of the claims” (Apotex memorandum
at para. 54, citing Apotex Inc. v. Merck & Co., [1995] 2 F.C. 723
(FCA) at para. 33, Merck & Co. Inc. v. Apotex Inc., 2006 FC 524 at
paras. 122 to 125 and Sanofi-Aventis Canada Inc. v. Apotex Inc., 2009 FC
676 [Sanofi]at paras 119 to 124 and 138, aff’d 2011 FCA 300 at
para. 3). Where this promise cannot be met the entire patent is rendered invalid
(Apotex memorandum at para. 55, citing AZT at para. 92, Plavix FCA at para. 54, Pfizer Canada Inc.v. Pharmascience Inc., 2008 FC
500 at para. 95, New Process Screw v. PL Robertson Manufacturing (1961),
39 C.P.R. 31 [New Process Screw] at paras. 27 to 28, 31, and 38 to 39
(CT) and Turner v. Winter (1787), 99 ER 1274 at 1276 (KB)). As the
Federal Court judge found that the compounds of the invention are not useful in
preventing colorectal cancer, this lack of utility goes to the validity of the
entire ‘576 Patent and the respondent’s application must be dismissed (Apotex
memorandum at para. 104).
[49]
In arguing insufficiency of disclosure, Apotex
submits that the ‘576 Patent “played games” with
its reader, in that the specification concealed the “true invention” of
celecoxib among two other compounds (those claimed in claims 5 and 6) known at
the filing date to lack utility (Apotex memorandum at para. 113). The reader
would have had to complete more work than the reader in Teva in order to
discover the true invention of the ‘576 Patent, and the amount of work required
in Teva was too great to support a finding of sufficient disclosure
(Apotex memorandum at para. 114).
[50]
Apotex criticizes the Federal Court judge’s
specific finding that “the fact that tests had revealed
high doses of the compound in [c]laim 5 were toxic in rats does not detract
from the fact that [c]laim 5 works as an anti-inflammatory” (Apotex
memorandum at para. 116, citing the Apotex decision at para. 44). This
statement reveals two errors, submits Apotex. First, the Federal Court judge
speaks of “high doses”, but the actual evidence was not qualified in this
respect (Apotex memorandum at para. 117). Second, it is nonsensical to suggest
that toxicity in rats would not detract from utility in treating
inflammation in rats (Apotex memorandum at para. 117).
[51]
Finally, Apotex criticizes the Federal Court judge’s
conclusion that the compound claimed in claim 6 “worked” because it was “the basis of a treatment of arthritis in dogs”
(Apotex memorandum at para. 120, citing the Apotex decision at para.
59). That, as it turned out after the filing date, the compound could be put to
such use is irrelevant to the sufficiency of the disclosure of the invention as
of that date (Apotex memorandum at para. 121).
POSITION OF THE RESPONDENT
[52]
The respondent argues that each of the Federal
Court judge’s decisions withstands the appellants’ attacks.
[53]
On the question of utility in treating
inflammation in humans, the respondent submits that Apotex’ argument conflates
the scope of a patent’s claims with the extent of any utility it may have
promised. Once the proper principles on patent construction are applied, one
can see that no “promise of (the) specific result”
of treatment of inflammation in humans appears in the ‘576 Patent (memorandum of
the respondent in response to Apotex at para. 35, citing Plavix FCA at
paras. 49 and 50). In any event, the respondent submits that the evidence is
clear that treatment in humans was soundly predicted (memorandum of the
respondent in response to Apotex at paras. 61 to 67). Although Apotex insists
that the Federal Court judge found otherwise at paragraph 17 of the Apotex
decision, the respondent submits that he was merely describing the position
taken by Apotex.
[54]
With respect to abuse of process and related
doctrines, the respondent submits that Novopharm FC was not binding on
the Federal Court judge and, even if it had been, it would not have led to the
outcome argued before this Court by the appellants, as Hughes J. found utility
to have been demonstrated through experiments on rats (memorandum of the respondent
in response to Apotex at paras. 80 to 83).
[55]
The respondent adds that a change in the law is
an important exception to abuse of process and comity. In this respect, the
principles for construing the promise of a patent had yet to be articulated
when Novopharm FC was decided seven years ago (memorandum of the respondent
in response to Apotex at para. 82). Given the change in the law, culminating in
Plavix FCA, the Federal Court judge properly refused to follow Novopharm
FC (memorandum of the respondent in response to Apotex at para. 83, citing R.
v. Bernard [1988], 2 S.C.R. 833 at 849 and 855 and R. v. Chaulk [1990],
3 S.C.R. 1303 at 1352).
[56]
With respect to the claim relating to colorectal
cancer, the respondent argues that different claims can contain different
promises, and that, contrary to Apotex’ claim, not every promise need be
construed as “overarching and inherent” to a patent’s invention and each of its
claims (memorandum of the respondent in response to Apotex at para. 41, citing Pfizer
Canada Inc. v. Apotex Inc., 2007 FC 26 at paras. 42 to 43, Teva Canada
Ltd. v. Novartis AG, 2013 FC 141 [Imatinib]at paras. 174 to 180, Pfizer
Canada Inc. v. Mylan Pharmaceuticals ULC, 2011 FC 547 at paras 191 to 193
and s. 58 of the Act). Furthermore, if the ‘576 Patent made any
promise of utility in preventing colorectal cancer, it would be limited to claim
16 as none of the other claims can sensibly be construed to contain such a
promise (memorandum of the respondent in response to Apotex at para. 77). As
for the statement in the specification that the compounds would be useful for
the prevention of colorectal cancer, this was the disclosure that would have
enabled claim 16, and not a promise that could be imported into each of the
patent’s claims (Ibidem).
[57]
With respect to Apotex’ arguments of
insufficiency, the respondent submits that these arguments never appeared in
Apotex’ NOA, and can therefore not be advanced before this Court (memorandum of
the respondent in response to Apotex at paras. 87 and 99 to 100).
ANALYSIS AND DECISION
Standard of Review
[58]
Broadly speaking, these appeals concern three
separate allegations of lack of utility and one allegation of insufficient
disclosure. The allegations pertaining to utility raise distinct issues of patent
construction and demonstrated or predicted utility.
[59]
The parties are in agreement that patent
construction gives rise to a question of law which stands to be assessed on a standard
of correctness (Plavix FCA at para. 33; Housen v. Nikolaisen, 2002
SCC 33, [2002] 2 S.C.R. 235 at para. 8 [Housen]).
[60]
However, whether utility has been made out, by
being demonstrated or predicted, is a question of fact to be reviewed only for
palpable and overriding error (Novopharm Limited v. Pfizer Canada Inc.,
2010 FCA 242, [2012] 2 F.C.R. 69 [Pfizer] at paras. 91-93; Housen at
para. 10). Finally, sufficiency of disclosure, because it gives rise to a mixed
question of fact and law, is reviewed only for palpable and overriding error
absent an extricable error of law (Housen at paras. 36 to 37).
[61]
In addition to alleging that the Federal Court
judge erred in his construction of the ‘576 Patent, the appellants also argue
that he breached the principles of comity and stare decisis in failing
to follow the construction reached by Hughes J. in Novopharm FC and
affirmed by this Court in Novopharm FCA. Pursuing the same theme, the
appellants argue that the Federal Court judge improperly countenanced an abuse
of process on the part of the respondent by allowing it to plead that reduced
harmful side effects were not promised whereas it had conceded that point in Novopharm
FC.
[62]
The scope and application of the doctrine of stare
decisis is a question of law for which the standard of review is
correctness (Air Canada Pilots Association v. Kelly, 2012 FCA 209,
[2013] 1 F.C.R. 308 at para. 40 [Kelly]). As to abuse of process, the
decision of the Federal Court judge allowing the respondent to argue that
reduced side effects were not promised is discretionary in nature, and cannot
be overturned absent an error of law or principle, or a wrongful exercise of
discretion with respect to the factors considered or not considered (Ramipril
FCA at para. 13; Elders Grain Co. v. Ralph Misener (Ship), 2005 FCA
139, [2005] 3 F.C.R. 367 at para. 13).
Plan of Analysis
[63]
I propose to address these issues under the
following headings: Utility in Treating Inflammation in Humans; Utility in Reducing
Side Effects; Utility in Preventing Colorectal Cancer; Insufficiency of Disclosure;
and Abuse of Process, Stare Decisis, and Comity. Consideration of these
topics requires a brief discussion about the legal approach set out in Plavix
FCA and specifically whether it amounts to new law, as the respondent
contends.
[64]
Under the Act, an invention must be
useful in order to deserve protection (s. 2). The courts, however, have long
held that the minimum requirements for utility under the Act are fairly
forgiving. First, the inventor need not expressly set out the utility of the
invention in the patent (Consolboard Inc. v. MacMillan Bloedel (Saskatchewan) Ltd., [1981] 1 S.C.R. 504 at 525 and 526 [Consolboard]. It is
merely required that, where the inventor is called upon to prove the utility of
the invention, utility can be shown to be demonstrated or soundly predicted as
of the patent’s filing date (AZT). Second, the threshold that must be proven to establish utility is
generally quite low, described as being no more than a “scintilla
of utility” (Olanzapine).
[65]
The promise doctrine represents an exception to
the above minimum statutory requirements. Though an inventor need not describe
any particular utility for the invention, an inventor who explicitly promises a
specific result will be held to that promise when called upon to prove utility
(Plavix FCA at paras. 48 and 49). That the invention may well have
satisfied the scintilla threshold is of no assistance in establishing utility where
a promise, if it be made, cannot be met (Plavix FCA at para. 54).
[66]
The promise doctrine will hold an inventor to an
elevated standard only where a clear and unambiguous promise has been made. Where
the validity of a patent is challenged on the basis of an alleged unfulfilled
promise, the patent will be construed in favour of the patentee where it can
reasonably be read by the skilled person as excluding this promise. This approach
can be traced back to the earliest mentions of the promise doctrine. In Consolboard,
the source of the promise doctrine in Canadian law, the Supreme Court of Canada
reiterated the longstanding principle that (Consolboard at 521, citing Western
Electric Company, Incorporated, and Northern Electric Company v. Baldwin
International Radio of Canada, [1934] S.C.R. 574 at 570):
… where the language of the specification,
upon a reasonable view of it, can be so read as to afford the inventor
protection for that which he has actually in good faith invented, the court, as
a rule, will endeavour to give effect to that construction.
[67]
This rule in favour of saving an invention
rather than invalidating it in case of ambiguity has been consistently applied
by this Court. While the principle is sometimes invoked by reference to the
original language found in Consolboard (Anastrozole at paras. 17
and 19) affirming AstraZeneca Canada Inc. v. Mylan Pharmaceuticals ULC,
2011 FC 1023, [2011] F.C.J. No. 1262 at para. 88), it is at other times given
effect through the requirement that promises be “explicit” (see Olanzapine
at para. 76, Eli Lilly and Company v. Teva Canada Limited, 2011 FCA 220,
[2011] F.C.J. No. 1028 at paras. 18 to 21 [Atomexetine], Plavix FCA
at para. 49). Drawing an analogy with the threshold test applicable to
selection patents, the Court in Plavix FCA expressed the need for
explicitness by saying that a promise must be supported by language “… at least as clear and unambiguous as that used to
establish the advantages of the selection over the compounds of a genus patent”
(Plavix FCA at para. 66). It follows that it is not enough to merely label
a promise as “explicit” if it can only be supported on the basis of equivocal
inferences and ambiguous indications (Plavix FCA at paras. 64-66).
[68]
It is apparent from the foregoing that Plavix
FCA merely applies a long established legal approach to a new set of facts.
Contrary to what the respondent asserts, it does not create new law.
[69]
I now turn to the question whether the Federal
Court judge correctly construed the ‘576 Patent in holding that no promise was
made that celecoxib would be useful in treating inflammation in humans.
Utility in Treating Inflammation
in Humans
[70]
Apotex raised the bar by arguing not only that
the ‘576 Patent promised utility in treating inflammation, but that this
promise extended to humans. The crux of Apotex’ argument is that, where a
patent “lays claim” to a particular use, the patent cannot conceivably be read
as not including a promise for that very use (Apotex memorandum at para. 67).
The only authority cited in support of this categorical proposition is Bauer
FC. In that case, Gauthier J., sitting as a trial judge, stated (at para.
289):
It is settled law that results or advantages
included in the claims must be met. Similarly, in the context of selection
patents where the advantages described are really the basis upon which the
patentee is given the right to monopolize a substance or product already
covered in a prior patent as part of a larger group of substances or products,
the inventor will be held to its promise (Ratiopharm Inc. v. Pfizer Ltd.,
2009 FC 711, 76 C.P.R. (4th) 241, 350 F.T.R. 250 (Pfizer (2009)).
[71]
This passage does not support the broad
proposition advanced by Apotex. In my view, Gauthier J. (as she then was) was
merely stating that, when a result or advantage is asserted in a patent’s
claims, it will generally be seen as a promise of utility. This is entirely
consistent with Zinn J.’s warning in Fournier at paragraph 126 that
statements going to utility are particularly vulnerable to being read as
promises when they are expressed in a patent’s claims. Apotex has failed to establish
how any of the ‘576 Patent’s claims can be shown to describe use in humans as a
particular advantage of the claimed compounds.
[72]
When the ‘576 Patent is read in light of the
approach set out in Plavix FCA, it becomes clear that no explicit
promise of treatment in humans was made. Apotex itself recognizes that the
claims speak only of “subjects”, and nothing outside the claims could be said
to represent the sort of unequivocal language contemplated by the reasoning in Plavix
FCA. In my view, the Federal Court judge correctly held that the promise of
the patent did not extend to humans.
[73]
The above determination renders irrelevant the
question whether, as Apotex alleges, the Federal Court judge found as a fact that
treatment in humans was demonstrated or soundly predicted. However, I would
agree with the respondent that, on a fair reading of the relevant passage (Apotex
decision at para. 17), it seems clear that he was merely describing the
position taken by Apotex in the matter before him.
Utility in Reducing Side Effects
[74]
In the Mylan decision, the Federal Court
judge began his analysis of the promise of the ‘576 Patent by carefully
canvassing the expert evidence before him (Mylan decision at paras.
52-59). He then construed the promise on the basis of the language of the patent,
analyzing it from the perspective of the skilled person based on the approach
set out in Plavix FCA (Mylan decision at paras. 60-71). He
reaffirmed this analysis in the Apotex decision (Apotex decision
at paras. 30 and 35). He also rejected Apotex’ attempts to distinguish in the
disclosure’s language between harmful side effects and common side effects,
finding that none of the experts read the patent in this manner (Apotex decision
at paras. 32 and 33).
[75]
Mylan seeks to distinguish the case at bar from
Plavix FCA and Anastrozole, arguing that the ‘576 Patent is “qualitatively more similar” to those in Olanzapine
and Latanoprost (Mylan memorandum at para. 60). However, Mylan provides
no arguments in support of this assertion.
[76]
Mylan’s argument that the Federal Court judge
failed to analyze expert testimony, merely “parlay(ing)
and broadly agree(ing)” with Dr. Young (the respondent’s expert) is
unconvincing. The only support for this attack are a number of statements by
the expert in question confirming that NSAID researchers were, at the filing
date, primarily interested in achieving reduced side effects via the COX-selectivity
mechanism. As the Federal Court judge points out however, not every research
goal will form the basis for a later patent’s promised utility and these
statements all spoke merely to goals (Mylan decision at para. 68, citing
Plavix FCA).
[77]
The appellants submit that the Federal Court judge
erred in holding that a promise must be explicitly stated in a patent’s claims.
They argue that a promise can also appear in the specification, provided that
the language is clear and explicit (Apotex memorandum at para. 85, citing Fournier).
No doubt that is so. However, I read the reasons of the Federal Court judge as merely
acknowledging the principle that statements outside of the claim should not be
presumed to be promises (Mylan decision at para. 70; Apotex
decision at para. 36). I can detect no error in this regard.
[78]
Mylan also argues that, even if the promise of
the ‘576 Patent is found to have excluded reduced side effects, it must still
be construed to have included elevated COX-2 selectivity “that is distinctly higher” than existing NSAIDs because
the Federal Court judge “appears to have recognized as
much” in the Apotex decision (Mylan memorandum at para. 85,
citing the Apotex decision at paras. 38 to 42, 59 and 62).
[79]
There are a number of problems with this
submission. In procedural terms, it appears not to have been advanced in
Mylan’s NOA. Also, it seems that the Federal Court judge would have been functus
when he made those comments, if correcting his earlier decision was in fact
what he was attempting to do. First and foremost, however, the passages cited
by Mylan simply do not support this interpretation. In the first set of
paragraphs, the Federal Court judge is simply summarizing an English case
relied upon by Apotex, and declining to follow it due to variation between
English and Canadian patent law (Apotex decision at paras. 38 to 40).
Though the remaining paragraphs do address COX-2 selectivity, the Federal Court
judge merely refers to the same absolute level of selectivity found to be
demonstrated by Hughes J. (Apotex decision at paras. 41 to 42, 59
and 62). There is no indication whatsoever that he found some particular level
of relatively enhanced selectivity to be promised.
[80]
The two arguments advanced solely by Apotex on
lack of utility in reducing side effects must also be rejected.
[81]
First, Apotex reiterates before us the argument
presented to and rejected by the Federal Court judge that the specification clearly
promised a reduction of harmful side effects, and merely equivocated
when referring to common side effects. The principal basis for attacking
the Federal Court judge’s rejection of this view is that, in so holding, he abdicated
his role by blindly adopting the opinion of the respondent’s expert (Apotex
memorandum at para. 83).
[82]
I would first note that nothing in the reasons
suggests that the Federal Court judge viewed the experts’ perspectives as
controlling his own. Rather, he simply relied on their views to support his own
(Apotex decision at para. 33). Furthermore, his view that none of the
experts adopted the interpretation advanced by Apotex is well-founded. Though
Apotex makes multiple references to expert testimony purportedly supportive of
the separate side-effect reading, none of this is in fact helpful in
establishing its case. For instance, in support of the view that the less
equivocal statement on side effects would have been read by the skilled person as
referring only to harmful side effects, Apotex points to a reference by
Dr. Flower to “severe side effects”. In the passage in question, however, Dr.
Flower merely opines that “significantly less harmful
side effects” would be understood to include such severe side effects
(Apotex’ appeal book, vol. 19, p. 5391). This is among Apotex’ stronger
references.
[83]
Furthermore, Apotex argues that the less
equivocal side effects statement promised reduced side effects in its invented
compounds relative to pre-existing compounds, while the more equivocal side
effects statement expressed doubt merely as to whether relative COX-2
selectivity among the invented compounds might correlate to relatively reduced
side effects (Apotex memorandum at paras. 78 and 79).
[84]
This argument must also fail as Apotex makes no
attempt to show that its interpretation is supported by any of the expert evidence.
This submission illustrates the sort of “unchaperoned
romp through the disclosure” that this Court admonished in Donepezil
FCA (at para. 57).
Utility in Preventing Colorectal
Cancer
[85]
In their written and oral submissions, the
parties dedicated little attention to this question. The central point at issue
is whether, as a matter of law, any given promise made in a patent must be
construed as overarching to the invention and all of the patent’s claims. As I
understand the argument, a patent containing a failed promise cannot be saved
by severance pursuant to section 58 of the Patent Act.
[86]
In effect, Apotex argues that any given promise
must be construed as overarching, and that, because the respondent has failed
to rebut Apotex’ allegation that the promise of preventing colorectal cancer
was neither demonstrated nor soundly predicted as of the filing date, the
validity of the patent as a whole is undermined. The respondent for its part argues
that not every promise need be construed as overarching, and that any promise
of preventing colorectal cancer was limited to claim 16 of the ‘576 Patent.
[87]
A review of the jurisprudence reveals a lack of
support for the proposition advanced by Apotex. Not one case cited by Apotex
stands for the proposition that a promise, once made and shown not to have been
met, must be construed as invalidating the invention as a whole. Of the eight
cases cited at paragraphs 54 and 55 of Apotex’ memorandum of fact and law, only
two could be read as addressing the extent to which a given promise may extend
to various claims within the patent containing it (Sanofi and New
Process Screw). Furthermore, each of these cases illustrates, at most, that
a promise can be construed so as to extend to each of a patent’s claims
(see Sanofi at paras. 119 to 124 or New Process Screw at 45 to 46).
In each case, the court did no more than construe the promise and made no
general statement of law on the matter. The respondent’s proposition, namely
that some promises are properly construed so as to touch only a subset of
claims, is therefore not inconsistent with the cases cited by Apotex, which
merely feature promises that were not so narrow.
[88]
The respondent provides compelling support for
its alternative proposition by citing examples where a promised utility is more
narrowly construed. Of particular relevance to the case at bar is a decision by
Snider J. wherein she specifically distinguished claims for a compound from
claims for its uses, and held that the latter are “directed
at the use of [the claimed compounds] for specified maladies and their utility
should be assessed on that basis” (Imatinib at para. 177). The
issue is one of patent construction and the respondent’s proposition in my view
represents the correct approach.
[89]
Though I accept that some promises may impose
utility requirements across each of a patent’s claim, Apotex has offered no
reason to depart from the Federal Court judge’s determination that the promise
of colorectal cancer prevention, if any, can be severed. That his decision to
apply s. 58 of the Act appeared in the sufficiency of disclosure section
of his reasons does not detract from the force of his reasons on this point.
Insufficiency of Disclosure
[90]
Apotex’ central argument is that the respondent
concealed its “true invention” among a group of compounds that it knew not to
be useful. In particular, the compound claimed in claim 5 was known to be toxic
and the compound claimed in claim 6 was only found after the filing date to be
useful in dogs.
[91]
I agree with the respondent that this argument is
an afterthought triggered by the decision of the Supreme Court in Teva,
which Apotex’ NOA does not announce. Apotex’ insufficiency allegation as set
out in its NOA is based solely on comparator issues (Apotex’ appeal book, vol.
1, p. 98). The allegation is that the respondent failed to clearly identify
which drugs its invention would improve upon and which specific side effects
would have to be examined to see the improvement.
[92]
This Court has long held that subsection 5(3) of
the Regulations prevents any second person from resisting a prohibition
application by advancing any legal or factual basis not stated in its NOA (AB
Hassle v. Canada (Minister of National Health and Welfare), [2000] F.C.J.
No. 855 at paras. 21 to 24; Proctor & Gamble Pharmaceuticals
Canada, Inc v Canada (Minister of Health), 2002 FCA 290, [2003] 1 F.C. 402
at paras. 21 to 24). Critically, none of the insufficiency arguments made
before this Court were included in Apotex’ NOA, and none of the insufficiency
arguments in its NOA are argued here. Apotex has therefore failed to provide
this Court any procedurally legitimate arguments in support of its insufficiency
allegation.
Abuse of Process, Stare Decisis, and Comity
[93]
Though the arguments and analysis pertaining to
the doctrines of abuse of process, stare decisis, and comity have
been intermingled by the parties and the Federal Court judge, it bears
emphasizing that these are distinct doctrines that merit separate analysis (see
Allergan at para. 39). I first turn to abuse of process.
Abuse of Process
[94]
Before the Federal Court judge, the appellants
argued that, in disputing whether the ‘576 Patent promised reduced side
effects, the respondent was abusing the process of the court, as it had
conceded before Hughes J. in Novopharm FC that reduced side effects were
necessary to the utility of the claimed invention.
[95]
The Federal Court judge dismissed this argument
in both cases. In the Mylan decision, he ruled that “a concession made by (the respondent) in one NOC proceeding
(was) not an admission binding upon it in another” (Mylan decision
at para. 78). In the Apotex decision, he came to the same conclusion,
citing Apotex’ failure to provide “a single case in
which a ‘concession’ or ‘admission’ in one in personam case applie(d) in
another” (Apotex decision at para. 61).
[96]
As discussed above, the Federal Court judge’s
decision to allow the respondent to make the argument that no premise was made
as to side effects was discretionary in nature, and can only be overturned by
this Court if he proceeded on the basis of an improper principle, or engaged in
a wrongful exercise of discretion.
[97]
Before this Court, it is principally Apotex which
seeks to overturn the decision of the Federal Court judge on the basis of his
alleged failure to give effect to the doctrine of abuse of process. Though
Mylan raises the argument in its Notice of Appeal, its memorandum of fact and
law does not advance the argument in any detail. Mylan rather argues that the Federal
Court judge focused too narrowly on this issue without giving due consideration
to the issue of stare decisis (Mylan memorandum at para. 44).
[98]
For its part, Apotex relies on three distinct,
yet related arguments in attacking the Federal Court judge’s decision on abuse
of process. First, Apotex argues that a patentee cannot “elevate the ‘inventive concept’ to support non-obviousness
and then read down the promised utility” (Apotex memorandum at para. 93,
citing Hoffman-La Roche v. Apotex, 2011 FC 875, 104 C.P.4. (4th)
233 [Mycophenolate FC], Allergan, Plavix FCA, and Olanzapine).
Second, Apotex argues that an innovator such as the respondent cannot “approbate and reprobate by taking fundamentally inconsistent
positions in different proceedings under the Regulations in respect of the same
patent” (Apotex memorandum at para. 89, citing Omeprazole FC at
paras. 137 and 138). Third, Apotex argues that an innovator cannot “relitigat(e)…an issue already decided in a proceeding to
which it was a party with the aid of additional evidence it chose not to adduce
in the earlier proceedings” (Apotex memorandum at para. 88, citing Ramipril
FCA at para. 50).
[99]
Apotex’ first argument fails to confront the
Federal Court judge’s reasons in rejecting it, as the proposition as framed
would not apply across proceedings. Indeed, the only cited case which actually
applies this proposition does so within the context of a single proceeding (see
Mycophenolate FC). The Federal Court judge noted that he had not been
provided with any authority for the application in one in personam case
of a concession made in another. None of the cases referred to by Apotex in
support of the inventive concept proposition overcome this objection. Apotex’ first
argument therefore demonstrates no error in the Federal Court judge’s exercise
of discretion.
[100] Apotex’ second argument moves beyond this limitation. The contention
is that an innovator cannot switch positions across NOC proceedings on the same
patent. The only authority it cites for this proposition, however, is Omeprazole
FC, a case which, in my view, cannot be read in so broad a manner.
[101] In that case, Apotex successfully argued abuse of process in
the course of seeking damages against AstraZeneca pursuant to section 8 of the Regulations.
Years earlier, Apotex had sought an NOC to market omeprazole capsules, which
required it to serve on AstraZeneca a Notice of Allegation addressing the
latter’s Canadian Patent No. 2,133,762. In response, AstraZeneca applied to
prohibit the issuance of an NOC to Apotex until the patent’s expiry. Upon O’Keefe
J.’s dismissal of AstraZeneca’s prohibition application (AstraZeneca AB v.
Apotex Inc., 2004 FC 313, 33 C.P.R. (4th) 97), Apotex sought damages under
section 8 (Omeprazole FC).
[102] When AstraZeneca argued in this latter proceeding that Apotex did
not constitute a “second person” within the meaning of the Regulations, Hughes
J. found an abuse of process. The reason for this was that AstraZeneca’s
initial application, and the resulting stay against the issuance of an NOC to
Apotex, was based on the premise that Apotex was a second person within the
meaning of the Regulations (Omeprazole FC at para. 138). The
innovator should not be permitted to take one position to avail itself of the stay,
and then resile from that position when the generic seeks compensation for
having been subjected to that very stay.
[103] In ascertaining the scope of the principle applied in Omeprazole
FC, regard must be had to the twin nature of a failed prohibition
application and a subsequent claim for section 8 damages. Though the
proceedings are technically separate, the second represents the generic’s
remedy for the losses it incurred as a result of the initiation of the first.
The proceedings are between the same parties and jointly dispose of the same
underlying dispute. Indeed, it is not without significance that Hughes J. also
barred AstraZeneca from taking this position on the alternative ground of issue
estoppel, a doctrine which applies only when the same issue has been finally
decided between the same parties or their privies (Omeprazole FC at
paras. 130 to 135).
[104] I accept that there may be instances where a concession made in one
proceeding under the Regulations may be construed as binding upon the
conceding party in a later proceeding involving a different party. However, there
is no existing precedent on point, and the Federal Court judge has not been
shown to have committed any error in principle in exercising his discretion and
deciding not to apply Omeprazole FC in favour of the appellants.
[105] Apotex’ third argument is based on Ramipril FCA. In that
case, this Court found an abuse of process where an innovator sought to
relitigate in one NOC proceeding an issue it had lost in a prior NOC proceeding.
Specifically, the innovator was barred from relying on evidence that it could
have adduced in the prior proceeding, but did not. The trial judge in that case
had allowed the generic’s motion to strike the innovator’s prohibition
application, on the basis that, due to the determination made in the prior
proceeding, it was plain and obvious that the innovator’s application had no
chance of success (Ramipril FCA at para. 29).
[106] On appeal, this Court intervened, as the determination in question
had been one of fact, and therefore would not have bound the decision-maker in
the second proceeding (Ramipril FCA at paras. 30 and 31). However, this
Court still found an abuse, as the innovator sought to rely on an enhanced
evidentiary record that could have been relied upon in the earlier proceeding (Ramipril
FCA at para. 47). Were an innovator not required to put its “best foot
forward” in response to an initial set of allegations, it would threaten finality
and consistency in judicial decision-making, and therefore undermine the
credibility of the adjudicative process (Ramipril FCA at paras. 35, 36
and 47).
[107] The present case is factually different. First, the respondent in
the current proceeding was successful in the dispute disposed of in the earlier
proceeding. This Court is therefore being invited to apply a judgment that
seeks to achieve finality and consistency so as to reach a different conclusion
than the one reached in Novopharm FCA. Second, the principal abuse that
Apotex complains of in the case at bar is not an attempt to rely upon an
enhanced evidentiary record, but rather the respondent’s refusal to abide by an
earlier concession. Nothing in Ramipril FCA or any of the
subsequent decisions cited by Apotex in support of this third argument provide
any support for the notion that shifting positions across proceedings is an
abuse.
[108] In the present case, the Federal Court judge exercised his
discretion to allow the respondent to take a position consistent with the law
rather than to compel it to abide by a prior concession which based on the conclusion that he reached, did not
conform with the law. I can detect no error in this exercise of
discretion.
Stare Decisis and Comity
[109] Turning to the doctrine of stare decisis, the appellants
argued before the Federal Court judge that he was bound by the construction of
the ‘576 Patent reached by Hughes J. in Novopharm FC and affirmed by
this Court in Novopharm FCA. The Federal Court judge was thus required
as a matter of law to conclude that reduced side effects were necessary to the
utility of the claimed invention.
[110] The Federal Court judge dismissed this argument in both cases. In
the Mylan decision, he held that Hughes J.’s ruling in Novopharm FC
would have bound him by stare decisis had it “turned
on the construction of the patent”, as this is a question of law (Mylan
decision at para. 78, citing Apotex Inc. v. Pfizer Ireland
Pharmaceuticals, 2012 FC 1339). Yet because demonstrated utility is a
matter of fact, he ruled, he was not bound (Mylan decision at para. 78).
In the Apotex decision, he appears to have disposed of the stare
decisis argument on the basis that, even if he had applied the Novopharm
FC construction, this construction does not extend to reduced side effects
in humans (Apotex decision at para. 62)
[111] Before this Court, it is principally Mylan which advances as a
ground of appeal the allegation that the Federal Court judge failed to abide by
the doctrine of stare decisis. Though Apotex does raise the matter
briefly in its memorandum of fact and law, its arguments supporting the
application of Hughes J.’s construction are otherwise entirely based on the
doctrine of abuse of process (Apotex memorandum at para. 97). Significantly, stare
decisis is not a ground alleged in Apotex’ Notice of Appeal.
[112] Mylan frames its argument on the doctrine of stare decisis from
two distinct angles. Though a court can be bound by the legal conclusions of a
higher court, it must also take into account legal conclusions reached by
judges at its own level – i. e. comity or “horizontal stare decisis”
(Mylan memorandum at para. 38). Mylan acknowledges that the doctrine of comity
is of limited application in proceedings under the Regulations, where
differences in allegations and evidence may call for varying conclusions (Mylan
memorandum at para. 40). It maintains, however, that a judge of the
Federal Court may only depart from a prior construction of a given patent by
another judge of the same court where he or she “is
convinced that the departure is necessary and can articulate cogent reasons for
doing so” (Mylan memorandum at para. 43, citing Allergan at
paras. 48 and 51).
[113] The Federal Court judge was therefore required as a matter of comity
to follow the construction of the ‘576 Patent set out in Novopharm FC unless
he could articulate a legitimate reason not to do so. Following Novopharm
FCA, he was also bound to this construction by vertical operation of stare
decisis given that the construction remained undisturbed by this Court on
appeal.
[114] The doctrine of stare decisis requires that “courts render decisions which are consistent with the
previous decisions of higher courts” (Kelly at para. 54, citing Canada (Attorney General) v. Bedford, 2012 ONCA 186, 346 D.L.R. (4th) 385 [Bedford
ONCA] at para. 56). The doctrine is not unlimited in its scope, however, as
not every statement in a given judgment will be binding on lower courts. In
determining the authoritative force of any given passage, one must essentially
ask “What did the case decide”? (R. v. Henry,
2005 SCC 76, [2005] 3 S.C.R. 609 at para. 57 [Henry]). At the very
least, an appellate judgment will stand as authority for its own ratio
decidendi, or the “reasoning that was necessary for
the court to reach a result on the issues that were presented to it for a
decision” (Kelly at paras. 54 and 55, citing Bedford ONCA).
The remaining obiter dicta will vary in authoritative force, ranging
from guidance to merely helpful commentary (Henry at para. 57).
[115] In contrast, the doctrine of comity or horizontal stare decisis
is not binding. Mylan cites Allergan for the proposition that a Federal Court
judge may only with certain justifications adopt a patent construction at odds
with a colleague’s prior construction. This decision does not go so far.
Rather, this Court highlighted the uncertainty that is created when two judges
of the same court reach distinct results on the same question of law without
explanation. It remains that, as shown by Allergan, the only thing that an
appellate court can do when this happens is to eliminate the uncertainty by
settling the question of law (Allergan at para.53). There is no legal
sanction for a judge’s failure to abide by comity.
[116] It follows that, although I agree that the justification offered by
the Federal Court judge in the Mylan decision falls short (see paragraph
110 above), nothing turns on this in this appeal.
[117] As for Novopharm FCA, the central question is “what did the
case decide?” Novopharm FCA does not decide that reduced side effects
were necessary to the utility of the invention claimed in the ‘576 Patent.
Rather, in that case, this Court was seized of two narrow issues, being the
correct claim date for the ‘576 Patent and the question whether one of the
appellants could avail itself of a statutory provision excluding certain public
disclosures from consideration in deciding obviousness. Whether the ‘576
Patent’s utility included reduced side effects is not a question to which this
Court turned its mind.
[118] The Federal Court judge was not persuaded by the Novopharm FC construction,
and nothing in the Novopharm FCA ruling compelled him to decide
differently.
DISPOSITION
[119] For the above reasons, I would dismiss both appeals with costs in
each case.
“Marc
Noël”
“I agree
Johanne Trudel J.A.”
“I agree
Richard Boivin J.A.”
Email this Content