Date: 20090601
Docket: A-472-08
Citation: 2009 FCA 183
CORAM: LINDEN J.A.
SEXTON
J.A.
SHARLOW
J.A.
BETWEEN:
THE MINISTER OF HEALTH and
THE ATTORNEY GENERAL OF CANADA
Appellants
(Respondents)
and
PHARMASCIENCE INC.
Respondent
(Applicant)
REASONS FOR JUDGMENT
SEXTON J.A.
[1]
This is an
appeal by the Minister of Health (“the Minister”) and the Attorney General of
Canada from an order of Justice Simpson of the Federal Court, allowing
Pharmascience’s application for judicial review of the Minister’s decision
requiring Pharmascience to address certain patents in respect of a supplementary
abbreviated new drug submission (SANDS) under subsection 5(1) of the Patented
Medicines (Notice of Compliance) Regulations, SOR/93-133 (“the NOC
Regulations”).
[2]
For the
reasons that follow, I conclude that the appeal should be dismissed. The
Minister erred in law by failing to perform the patent-specific analysis
mandated by the Supreme Court’s decision in AstraZeneca v. Canada (Minister
of Health), 2006 SCC 49, [2006] 2 S.C.R. 560. The applications judge
therefore undertook this analysis herself, and in my view her conclusion was
reasonable and should be upheld.
RELEVANT
LEGISLATION
[3]
Subsection
5(1) of the NOC Regulations (as it read at the material time) requires a
second person filing a submission for a notice of compliance (NOC) to address
all patents listed on the register in respect of a NOC issued to a first
person, where the second person compares its drug with, or makes reference to,
the drug for which that NOC was issued (emphasis mine):
|
5.
(1) Where
a person files or has filed a submission for a notice of compliance in
respect of a drug and compares that drug with, or makes reference to, another
drug for the purpose of demonstrating bioequivalence on the basis of
pharmaceutical and, where applicable, bioavailability characteristics and
that other drug has been marketed in Canada pursuant to a notice of
compliance issued to a first person and in respect of which a patent list has
been submitted, the person shall, in the submission, with respect to each
patent on the register in respect of the other drug,
(a) state that the
second person accepts that the notice of compliance will not issue until the
patent expires; or
(b) allege that
(i) the statement made by
the first person under paragraph 4(4)(c) is false,
(ii) the patent has
expired,
(iii) the patent is not
valid, or
(iv) no claim for the medicine
itself and no claim for the use of the medicine would be infringed by the
making, constructing, using or selling by that person of the drug for which
the submission for the notice of compliance is filed.
|
5. (1) Lorsqu’une personne dépose
ou a déposé une demande d’avis de conformité pour une drogue et la compare,
ou fait référence, à une autre drogue pour en démontrer la bioéquivalence
d’après les caractéristiques pharmaceutiques et, le cas échéant, les
caractéristiques en matière de biodisponibilité, cette autre drogue ayant été
commercialisée au Canada aux termes d’un avis de conformité délivré à la
première personne et à l’égard de laquelle une liste de brevets a été
soumise, elle doit inclure dans la demande, à l’égard de chaque brevet
inscrit au registre qui se rapporte à cette autre drogue :
a) soit une déclaration portant qu’elle accepte que l’avis de
conformité ne sera pas délivré avant l’expiration du brevet;
b) soit une allégation portant que, selon le cas :
(i) la
déclaration présentée par la première personne aux termes de l’alinéa 4(4)c)
est fausse,
(ii) le
brevet est expiré,
(iii) le
brevet n’est pas valide,
(iv) aucune
revendication pour le médicament en soi ni aucune revendication pour
l’utilisation du médicament ne seraient contrefaites advenant l’utilisation,
la fabrication, la construction ou la vente par elle de la drogue faisant
l’objet de la demande d’avis de conformité.
|
FACTS
[4]
The facts
of this appeal are not in dispute and were set out in detail by the applications
judge in paragraphs 2 to 30 of her reasons. The respondent is attempting to
bring to market a generic version of ramipril, an ACE inhibitor. Ramipril is
marketed by Sanofi-Aventis Canada Inc. (“Sanofi”) in Canada under the brand name ALTACE. As of July
10, 2000, Sanofi had been granted 4 NOCs for ALTACE, against which it had
listed three patents. At that time, ALTACE was approved for the treatment of
hypertension.
[5]
On July
10, 2000, the respondent purchased ALTACE capsules in four strengths (1.25 mg,
2.5 mg, 5 mg, and 10 mg) for use as Canadian reference products. On September
4, 2001, the Minister received the respondent’s abbreviated new drug submission
(ANDS). This ANDS was based on the asserted bioequivalence of each strength of
the respondent’s ramipril capsules with the equivalent strength of ALTACE. The
respondent submitted bioavailability data demonstrating the bioequivalence of
its 10 mg ramipril capsule with the 10 mg ALTACE capsule. It then sought a waiver
of the requirement to submit bioavailability data in respect of the other
strengths under the Minister’s Proportional Formulations Policy (“the
proportionality policy”). The respondent only sought approval for the use of
its ramipril capsules for the treatment of hypertension.
[6]
The ANDS
attached the July 10, 2000 invoice for the purchase of the drug samples, and a
clinical report on the bioequivalence of the 10 mg capsules stating that the
ALTACE samples were received by the lab for testing on December 15, 2000.
[7]
On
February 24, 2003, the respondent withdrew its ANDS in respect of the 1.25 mg
capsules due to a lack of stability data. On August 27, 2003, the Minister
found that the respondent was entitled to a NOC for the other strengths (2.5
mg, 5 mg, and 10 mg) subject only to compliance with the NOC Regulations.
[8]
On
December 30, 2005, the respondent filed a SANDS in respect of its 1.25 mg
ramipril capsules. It again sought a waiver of the requirement to submit
bioavailability data under the proportionality policy, proposing to demonstrate
the proportionality of its 1.25 mg capsule with its 10 mg capsule.
[9]
In the period
between the respondent’s ANDS and its SANDS, on November 6, 2003, Sanofi was
granted a further NOC (“the sixth NOC”), against which it listed Canadian
Patents Nos. 2,382,387 and 2,382,549 (the 387 and 549 patents, respectively).
These patents teach a new use of ramipril, namely, for treatment following a
heart attack. The respondent is not seeking approval of its ramipril capsules
for this new indication.
[10]
Nonetheless,
on April 12, 2007, the Minister issued a decision requiring the respondent to
address the 387 and 549 patents under subsection 5(1) of the NOC Regulations
in connection with its SANDS. On May 17, 2007, the Minister, applying the
proportionality policy, found that the respondent’s 1.25 mg ramipril capsule is
bioequivalent to the 1.25 mg ALTACE capsule. Thus, the respondent is entitled
to a NOC for the last strength, subject only to compliance with the
Regulations.
[11]
The Minister
denied the respondent’s request for reconsideration of its April 12 decision on
June 8, 2007, and the respondent commenced an application for judicial review.
DECISION BELOW
[12]
The
applications judge allowed the respondent’s application on the basis of the
principles set out by the Supreme Court in AstraZeneca. She found that AstraZeneca
stands for the proposition that “a generic company need only address patents
listed against NOC’s [sic] filed at the time it purchases the comparator drug
it selects for the purposes of its ANDS” (2008 FC 922 at para. 32).
[13]
Since the
respondent is not seeking approval of its ramipril capsules for treatment
post-heart attack, the applications judge also concluded that it “has not, in
fact, made use of the patented inventions taught by the 387 and 549 patents”
(at para. 31). Accordingly, she held that the respondent should not be required
to address these patents under the NOC Regulations.
ISSUE AND STANDARD OF REVIEW
[14]
The only
issue on this appeal is whether the applications judge erred in holding that
the respondent should not be required to address the 387 and 549 patents. In my
view, AstraZeneca satisfactorily determined that the standard of review
for an interpretation of the NOC Regulations is correctness (at para.
25). The factual findings made by the applications judge in applying the
correct legal test are entitled to deference and must stand if they were
reasonably open to her on the record.
ANALYSIS
[15]
Both
parties discussed the Supreme Court’s decision in AstraZeneca at length
in their submissions. In that case, the issue before the court was whether
Apotex should be required to address two patents listed against NOCs granted to
AstraZeneca for omeprazole after Apotex had filed its ANDS. The circumstances
were somewhat unique in that AstraZeneca had never marketed any drug in Canada incorporating the inventions
taught by the two after-listed patents. Writing for the unanimous court,
Justice Binnie emphasized that the NOC Regulations were enacted to
prevent abuse of the “early working exception” provided by subsection 55.2(1)
of the Patent Act, R.S.C. 1985, c. P-4, and should be interpreted with
that purpose in mind (at paras. 15-16).
[16]
He
described the Regulations as part of a balancing exercise, whereby generic drug
companies are given the right to early work an innovator’s patented inventions
to satisfy the requirements under the Food and Drugs Act, R.S.C. 1985,
c. F-27 and Regulations (C.R.C. 1978, c. 870) for a NOC. However, if the
innovator initiates an application for an order prohibiting the Minister from
issuing a NOC to the generic, a 24 month stay of ministerial action arises
automatically. The Supreme Court had previously described this stay as
“draconian” in Merck Frosst Canada Inc. v. Canada (Minister of National
Health and Welfare), [1998] 2 S.C.R. 193 at para. 33.
[17]
Justice
Binnie concluded that in principle, a generic need only address patents
relevant to the NOC that gave rise to the comparator drug (AstraZeneca at
para. 39, emphasis mine):
In my view, s.
5(1) requires a patent-specific analysis, i.e. the generic manufacturer is
only required to address the cluster of patents listed against submissions
relevant to the NOC that gave rise to the comparator drug, in this case the
1989 version of Losec 20.
The comparator drug in that case was the drug that Apotex
had actually purchased and analyzed for the purposes of demonstrating the
bioequivalence of its own omeprazole tablets (at paras. 34-37).
[18]
The
dispute in the instant case is how “the comparator drug” should be identified. The
appellants submit that legally, ALTACE is not simply a physical compound.
Rather, ALTACE is defined as the sum total of all submissions approved for it
as of a given date—that is, the initial new drug submission in respect of which
its first NOC was issued, and all subsequent supplemental new drug submissions
giving rise to later NOCs. Therefore, the appellants argue that the comparator
drug in this case is the version of ALTACE available to be copied as of the
date the respondent submitted its SANDS—namely, the version for which Sanofi
received its sixth NOC, against which it listed the 387 and 549 patents.
[19]
The
respondent, on the other hand, submits that the comparator drug is the drug it
actually purchased in July 2000 and analyzed for the purpose of demonstrating
the bioequivalence of its ramipril capsules. Although it withdrew its initial
ANDS in respect of the 1.25 mg capsules and later submitted an SNDS, it argues
that it still relied only on the drug it purchased in 2000 to demonstrate
bioequivalence with ALTACE.
[20]
Despite
the very able argument of counsel for the appellants, I am unable to agree with
their interpretation of AstraZeneca. In that case, Justice Binnie wrote
(at para. 28):
I
accept the linguistic point made by Noël J.A. in the Federal Court of Appeal
that the words "in respect of which" in s. 5(1) refer to "the
other drug", i.e. the Canadian reference product, and not to a particular
patent list or amended patent list. However, it seems to me that the
"other drug" is the drug to which the generic manufacturer makes
reference "for the purpose of demonstrating bioequivalence".
AstraZeneca also emphasized that “it is the actual
drug, from which samples can be taken and used for comparative purposes
that is relevant to the application of subsection 5(1) of the NOC
Regulations” (at para. 34, quoting from the judgment of Justice Noel of
this court, 2005 FCA 189, [2006] 1 F.C.R. 297 at para. 46, emphasis Justice
Binnie’s).
[21]
In Ferring
Inc. v. Canada (Minister of Health), 2007 FC 300, [2008] 1 F.C.R. 19, aff’d
2007 FCA 276, 370 N.R. 263, Justice Hughes acknowledged the importance of the
date the generic purchased the comparator drug to the patent specific analysis
required by AstraZeneca. However, in
obiter dicta, he expressed concern that the purchase
date of the samples would only be known to the generic itself. As an
evidentiary matter, he suggested that the Minister have regard to the date the
generic filed its ANDS, as this would be “logically the last date upon which
the comparator drug could have been obtained by the generic”. He was also of
the view that the Minister should require a generic to address after-listed
patents only if the generic had made use of changes made to the comparator drug
since the date of purchase, for the purposes of demonstrating bioequivalence
(at para. 65).
[22]
Thus, the
jurisprudence is clear that the patent specific analysis requires a generic to
address only those patents in respect of which it takes advantage of the early
working exception in the Patent Act for the purposes of demonstrating
bioequivalence and obtaining a NOC.
[23]
The
appellants’ position is that the generic should be required to address any
patent that is notionally available for early working (i.e. any patent listed
prior to an ANDS or SANDS) and that the Minister should not have to determine
on the evidence whether a patent was in fact early worked. Thus, they say that
in cases where there is a dispute, the proper forum for that dispute is a NOC
proceeding in which the innovator may participate. The appellants seek to
distinguish AstraZeneca on the basis that in that case a drug had never
become available incorporating the teachings of the after-listed patents, and
that unlike in this case, early working was a factual impossibility.
[24]
I cannot
accept this argument, which would require a generic to address patents as fast
as an innovator could list them, with no regard to whether the generic has
taken advantage of the early working exception. As Justice Binnie wrote in AstraZeneca
(at paras. 21 and 23, underlining mine):
There
is no linkage between the 037 and 470 patents and the submissions that lead to
the Losec 20 product copied by Apotex. Those
after-acquired patents were listed in relation to a SNDS dated January 22, 1999
by AstraZeneca for a new medical use for Losec 20
(treatment of H. Pylori), a use for which the Apotex
product is not approved, and to an administrative
SNDS submitted by AstraZeneca dated July 12, 2000, which submission has nothing
at all to do with the technology incorporated in Losec 20.
…
…On
this view a "first person" could carry on "evergreening"
its product indefinitely by the addition of new patents of marginal
significance which would trigger an indefinite series of 24-month statutory
freezes even though such subsequently listed patents are not the subject of
"early working" by the generic manufacturer, and from which (as in
the circumstances here) the generic manufacturer derives no advantage.
[25]
As the
applications judge noted, it is the Minister’s responsibility to conduct the
patent specific analysis and “to identify the precise patents which are
relevant to a generic manufacturer’s early working of a copycat product” (at
para. 35, citing AstraZeneca at para. 22). In my view, the date the
comparator drug was purchased is the starting point. The Minister must then
evaluate the evidence before him to determine whether the generic has taken
advantage of the teachings of any after-listed patents. Evidence including
invoices for the purchase of drug samples, clinical studies stating when drug
samples were received for testing, and the generic’s product monograph listing
the proposed uses of the drug, is generally before the Minister when an ANDS or
a SANDS is filed and can be marshalled for this purpose. In cases where the
evidence is unclear or there is an absence of reliable evidence, we agree with
the applications judge that the Minister may use the filing date of an ANDS (or
a SANDS, where appropriate) as a fallback position.
[26]
The
Minister failed to conduct this analysis. Thus, it was open to the applications
judge to do so. In this case, there was evidence before the Minister (in the
form of the July 2000 invoice) as to when the respondent purchased the
comparator drug. It submitted bioavailability data that relied exclusively on
those samples. While it withdrew its ANDS in respect of the 1.25 mg capsules
and later submitted a SANDS, it did so only due to an initial lack of stability
data for those capsules and not for the purpose of conducting new bioequivalence
studies or modifying its drug to incorporate any new technologies.
[27]
As the
respondent points out, a SANDS is not a stand-alone submission. The Minister
must take into account the relationship between a particular SANDS and an earlier-filed
ANDS when conducting the patent specific analysis.
[28]
Most
significantly in this case, the respondent has never sought approval for the
new use of ramipril capsules for treatment post-heart attack, as taught by the
after-listed patents. The appellants argue that this fact goes to the ultimate
question of infringement and not to whether the respondent should have to
address the new patents. However, the Supreme Court saw fit to consider this
factor in AstraZeneca, when it noted at least twice that Apotex was not seeking
to take advantage of a new indication taught by the after-listed patents in
that case (at paras. 21 and 42).
[29]
The
applications judge found as a fact that the respondent has not made use of the
patented inventions taught by the 387 and 549 patents (reasons for order at
para. 31), and is not “on this occasion within the mischief aimed at by the NOC
Regulations” (AstraZeneca at para. 38). Indeed, the appellants were
not able to refer this court to any evidence in the record contrary to this
finding. Accordingly, this conclusion was reasonably open to the applications
judge, and there is no basis for this court to interfere.
CONCLUSION
[30]
I
therefore conclude that the applications judge was correct to conclude that the
respondent should only be required to address the patents listed against the
NOC giving rise to the comparator drug it actually purchased and analyzed for
the purposes of demonstrating bioequivalence, in the circumstances of this case.
The Minister erred in requiring it to address the 387 and 549 patents.
[31]
For these
reasons, I would dismiss the appeal with costs.
"J.
Edgar Sexton"
"I
agree
A.M. Linden J.A."
"I
agree
K. Sharlow J.A."
Email this Content