Docket: T-1125-12
Citation:
2014 FC 566
Ottawa, Ontario, June 13, 2014
PRESENT: The
Honourable Mr. Justice O'Reilly
|
BETWEEN:
|
|
ALLERGAN INC AND ALLERGAN, INC
|
|
Applicants
|
|
and
|
|
THE MINISTER OF HEALTH AND
COBALT PHARMACEUTICALS COMPANY
|
|
Respondents
|
JUDGMENT AND REASONS
I.
Overview
[1]
The applicants, Allergan, ask me to issue an
order prohibiting the Minister of Health from issuing a notice of compliance
(NOC) to Cobalt Pharmaceuticals Company, relying on s 6(1) of the Patented
Medicines (Notice of Compliance) Regulations, SOR/93-133 (see Annex A for
all enactments cited). The NOC would enable Cobalt to market a generic version
of a product protected by Allergan’s Canadian Patent No 2,585,691 (the ‘691
patent). The ‘691 patent
relates to eye drops used in the treatment of glaucoma and ocular hypertension.
Allergan calls its product “LUMIGAN RC”. In these reasons, for ease of
reference, I will call this product “new Lumigan”. The ‘691 patent issued in
2009 and will not expire until 2026.
[2]
Cobalt has alleged that Allergan’s patent is
invalid on the grounds of obviousness, lack of utility, and anticipation and,
therefore, that Cobalt should be entitled to an NOC for an equivalent generic
product. Allergan maintains that Cobalt’s allegations are unjustified and that
the Minister should be prohibited from granting Cobalt its NOC.
[3]
I agree with Allergan that Cobalt’s allegations
of obviousness, lack of utility, and anticipation are unjustified. Therefore, I
must allow Allergan’s application.
[4]
There are three issues:
1.
Is the subject matter of the ‘691 patent
obvious?
2.
Could the stated utility of the ‘691 patent be
soundly predicted?
3.
Was the ‘691 patent anticipated?
II.
The ‘691 Patent
[5]
Allergan previously owned a patent for
formulations of ocular products with the same purpose as, and similar
ingredients to, new Lumigan (Canadian Patent No 2,144,967 – the ‘967 patent).
The ‘967 patent, which expired in 2013, covered a wide range of concentrations
of active ingredients and preservatives and included within that range a
product containing .03% bimatoprost (Bp) and 50 parts
per million (ppm) benzalkonium chloride (BAK) as the preservative. Allergan
marketed this product as “LUMIGAN” and I will refer to it as “old Lumigan”.
[6]
Old Lumigan caused some dose-dependent side
effects, particularly an eye irritation called “hyperemia.” Allergan sought to
create a product with fewer adverse reactions. It arrived at the formulation
for new Lumigan, which contained only a third of the Bp contained in old
Lumigan, but used four times as much BAK. Based on its studies of this revised
formulation, Allergan believed that patients using new Lumigan would likely
enjoy a comparable reduction of intraocular pressure (IOP), but with fewer side
effects than with old Lumigan.
[7]
The ‘691 patent is entitled “Enhanced
Bimatoprost Ophthalmic Solution”. It makes reference to the prior art and
includes data relating to the testing of various concentrations of Bp (and
other active ingredients) and BAK (and other preservatives). In particular, it
mentions old Lumigan containing .03% Bp and 50 ppm BAK, as well as other
products using BAK as a preservative in concentrations of 150 to 200 ppm.
[8]
The ‘691 patent relates to a formulation for eye
drops primarily for use in treating glaucoma. The formulation contains a single
active ingredient, Bp, in a concentration of between .005% and .02%. The other
main ingredient is BAK at concentrations from 100 to 200 ppm. While BAK acts
primarily as a preservative, it has other properties, too, as will be discussed
below. Claim 16 of the ‘691 patent falls within this formulation. It claims a
combination of .01% Bp and 200 ppm of BAK. Claim 19 claims the use of this
formulation in the treatment of glaucoma or intraocular hypertension. These are
the only claims in issue here.
[9]
The patent lists various embodiments of the
invention ranging from .01% to .02% Bp, all with 200 ppm BAK. In terms of data,
the patent contains two graphs (Examples 2 and 4). The first shows the results
of an in vivo test in rabbits of .03% Bp with 50 ppm BAK (ie, old
Lumigan) compared with .03% Bp with 200 ppm BAK. Other data were provided but
they are not relevant here. The results showed that use of BAK at 200 ppm
increased by 57% the concentration of Bp in the aqueous humour of the rabbits’
eyes.
[10]
The second graph provides the results of an in
vitro study in rabbit corneal epithelial cell layers of two concentrations
of Bp (.015% and .03%) combined with various concentrations of BAK (from 50 ppm
to 200 ppm). None of the combinations corresponds with new Lumigan (ie,
.01% Bp with 200 ppm BAK). However, .015 Bp was tested both with 50 ppm and
200 ppm BAK. The results showed that the use of the higher amount of BAK
achieved greater penetration of Bp across the corneal cells.
[11]
The patent also describes a treatment scenario
in which a product with a fairly low concentration of Bp (.015%) and a
relatively high amount of BAK (125 ppm) – which is obviously not new Lumigan –
would achieve a greater lowering of IOP and less hyperemia than old Lumigan. No
such test was ever conducted.
III.
Construction of the ‘691 Patent
[12]
Generally speaking, the patent must be construed
through the eyes of the person skilled in the relevant art. Here, the parties
agree that this notional person would have the skills of a formulator
experienced in creating optical formulations and those of an ophthalmologist
experienced in treating glaucoma.
[13]
In my view, the claims in issue are clear. Claim
16 relates to a product with lower Bp (.01%) and higher BAK (200 ppm) than old
Lumigan (which was comprised of .03% Bp and 50 ppm BAK, respectively). Claim 19
relates to the use of that product for treating glaucoma in humans. Since they
are unambiguous, I need not resort to the patent’s specification to construe
them. However, the specification helps explain the inventive concept of the
patent and describes the utility of the invention, which forms part of the
analysis of obviousness and sound prediction below.
IV.
Issue One – Is the subject matter of the ‘691
patent obvious?
[14]
The parties agree that Supreme Court of Canada
laid out the test for obviousness in Apotex v Sanofi-Synthelabo Canada Inc,
2008 SCC 61, at para 67 [Sanofi]. I will address each element of the
test, as follows.
A.
Identify the skilled person to whom the patent
is addressed
[15]
As mentioned, the parties accept that the
relevant person would have the skills of an optical formulator and an
ophthalmologist experienced in treating glaucoma.
[16]
Both parties provided expert opinions on the
scope of the ‘691 patent. Allergan relied on the opinions of Dr Valentino
Stella and Dr Harry Quigley. Cobalt relied on Dr Paul Laskar’s opinion. The
experts’ credentials are summarized in Annex B.
B.
Identify the state of the art
[17]
The relevant date for this assessment is that on
which the ‘691 patent was filed – March 16, 2005. At that time, the experts all
agree that Bp was a compound known to be useful for lowering IOP, and BAK was a
well-known preservative. Bp was the active ingredient (at .03%) in old Lumigan,
which had 50 ppm BAK.
[18]
The skilled person would likely have viewed a
reduction of Bp by one third to .01% in new Lumigan to cause a corresponding
diminution of efficacy in lowering IOP. Still, .01% Bp would likely still have
some efficacy. A 2001 study showed that .03% Bp was more effective than .01%
Bp.
[19]
However, he or she would also have regarded a
reduction of Bp as causing fewer side effects. Bp caused dose-dependent eye
hyperemia. Therefore, reducing the dose of Bp would correspondingly reduce
hyperemia.
[20]
The skilled person would likely have been concerned
about the increase of BAK to 200 ppm because of its cytotoxicity. While optical
formulations containing that amount of BAK were known, tolerated by patients,
and on the market, a skilled person would likely have regarded a substantial
increase in the concentration of BAK as a negative factor. Since 50 ppm of BAK
was sufficient to achieve a preservative effect in old Lumigan, there was no
obvious reason to increase its concentration.
C.
Identify the inventive concept of the claims
[21]
The parties differ on the meaning of “inventive
concept”. Allergan argues that the inventive concept is different from the
claimed invention which, in turn, is different from the stated utility of the
claimed invention. On the question of obviousness, then, the inventive concept
of the ‘691 patent, according to Allergan, is the creation of a new formulation
(new Lumigan) that achieves a comparable effect to that of old Lumigan, but
with less Bp.
[22]
However, on the question of construction of the claims,
Allergan argues that there is no specific indication (or “promise”) in the
patent that the claimed invention will have any particular effect in lowering
IOP. The claimed invention is simply an ocular formulation with .01% Bp and 200
ppm BAK.
[23]
Cobalt maintains that the patent should be given
only one interpretation for all purposes – infringement, obviousness, and
utility. Therefore, as Allergan suggests, the claimed invention is a formulation
with .01% Bp with 200 ppm BAK and that, too, is the definition of the invention
that should be applied to all grounds of alleged invalidity, including
obviousness. Accordingly, Cobalt argues that altering the amounts of Bp and BAK
was an obvious modification of old Lumigan.
[24]
In my view, the interpretation of the patent
should be as consistent as possible across the various issues in play. A
patentee should not, for example, be able to “read up the
invention for obviousness and read it down for utility” (Hoffmann-La
Roche v Apotex, 2011 FC 875, at para 22). To do so would be unfairly
advantageous for a patent holder who might wish to assert that its invention
was an unforeseeable innovation (and, therefore, not obvious) and, at the same
time, contend that the invention’s useful properties could be readily inferred
(and, therefore, soundly predictable).
[25]
However, that does not mean that construction of
the claims necessarily determines the inventive concept for purposes of the
obviousness analysis. Claims construction is an important preliminary exercise
particularly where infringement is in issue. One has to know what the
parameters of the claimed invention are in order to determine whether another
party has encroached on them. Identifying the inventive concept is a separate
exercise forming part of the analysis of whether the patent claims something
that is truly inventive, and therefore not obvious. Where the patent relates to
a bare chemical formula, the court must refer to the claim and the
specification to determine what the claim’s inventive concept is (Sanofi,
above, at 67, 77). I believe the same should be true, as here, where the claim
relates to a bare list of ingredients.
[26]
In my view, reading the ‘691 patent as a whole,
the inventive concept of Claims 16 and 19 is a formulation with reduced Bp but
with comparable efficacy to old Lumigan, achieved by increasing BAK.
[27]
As discussed above, the data in the patent show
that BAK had penetration enhancing properties, in addition to its known
preservative qualities. Even though there was no data for the precise
formulation identified in Claim 16, a skilled reader of the patent would see
that raising BAK to 200 ppm caused a significant increase in the penetration of
Bp. The closest data showed that a formulation with .015 Bp and 200 ppm BAK
achieved substantially greater permeability across rabbit cell layers than a
.015% Bp and 50 ppm BAK formulation. In fact, the results for .015% Bp and 200
ppm BAK were higher than for .03% Bp and 50 ppm BAK (ie, old Lumigan).
Similarly, increasing BAK from 50 ppm to 200 ppm in a .03% Bp formulation
caused a substantial rise in the concentration of Bp in rabbit eyes.
[28]
From this data, it appears to me that a
conclusion about the efficacy of a .01% Bp-200 ppm BAK solution would involve
no more than a simple extrapolation from the data regarding other combinations.
Since .015% Bp with 200 ppm BAK achieved greater effect than old Lumigan,
reducing Bp further to .01% would likely result in values more or less
comparable to old Lumigan. In other words, new Lumigan would likely cause
roughly equivalent IOP lowering to old Lumigan.
[29]
As will be seen below in the discussion of sound
prediction, this interpretation of the data in the ‘691 patent is supported by
expert opinion.
D.
Identify any differences between the state of
the art and the inventive concept
[30]
Cobalt argues that combinations of Bp and BAK
for use in treating glaucoma were included in US Patent No 5,688,819. In
addition, old Lumigan itself represented such a combination. Further, it was
known that Bp could be effective at low concentrations (even as low as .003%),
so a skilled person would expect that using Bp at .01% would lower IOP, at
least to some extent. Accordingly, there is no difference between the inventive
concept of the claims in issue and the state of the art.
[31]
I have already found the inventive concept to be
a formulation with a fairly low concentration of Bp (.01%) that has a
comparable effect in lowering IOP to old Lumigan (at .03%), due to the
penetration enhancing effect of 200 ppm BAK. Nothing in the state of the art
suggested such a possibility.
[32]
As mentioned, according to the state of the art,
to reduce Bp would be to reduce efficacy. Increasing BAK could be contrary to
the accepted wisdom that BAK had its own side effects and, therefore, that its
use should be minimized or eliminated, not increased. Further, there was nothing
in the literature that pointed to BAK being a penetration enhancer for a
compound like Bp. Dr Stella noted that studies showed that BAK might act as a
penetration enhancer for some compounds but not for a relatively lipophilic
molecule like Bp.
E.
Do those differences constitute steps that would
have been obvious to the skilled person?
[33]
Cobalt submits that the inventive concept of the
claims in issue corresponds with the state of the art and, therefore, the
claimed formulation was obvious. Accordingly, in Cobalt’s submission a skilled
person would have known that the combination of .01% and 200 ppm BAK was
possible and would cause lowering of IOP.
[34]
In my view, as described above, there are
significant differences between the inventive concept of the claims in issue
and the state of the art at the relevant time. Those differences would not have
been obvious to the skilled person. The formulation of new Lumigan with
efficacy comparable to old Lumigan required experimentation (as seen in the patent
itself) and inventive steps, such as the trial of various combinations of Bp
and BAK (and other ingredients).
[35]
The evidence shows that Allergan conducted
numerous tests before arriving at the claimed invention. The company spent
millions of dollars and devoted thousands of person-hours to the project. The
result was a commercially successful product. Clearly, it was not self-evident
that the claimed formulation would work as it did. In my view, that formulation
was not obvious to try.
[36]
Accordingly, Cobalt’s allegation that the
subject matter of the ‘691 patent was obvious is not justified.
V.
Issue Two – Could the stated utility of the ‘691
patent be soundly predicted?
[37]
On my reading of the patent, the stated utility
of the claims in issue is that new Lumigan would have a comparable effect to
old Lumigan, with less Bp (and, therefore, fewer side effects). That utility had
not actually been demonstrated as the data in the patent did not relate
directly to the claimed formulation - .01% Bp with 200 ppm BAK. Therefore, the
question here is whether it could have been soundly predicted that new Lumigan
was likely to cause a comparable lowering of IOP compared to old Lumigan, given
its substantial reduction of Bp and significant increase in BAK.
[38]
Cobalt maintains that the data set out in the
‘691 patent do not provide a basis for a sound prediction of the stated utility,
namely, a comparable effect in lowering IOP to that of old Lumigan. In
addition, the patent does not set out the line of reasoning that would link
that data to the stated utility. For example, the patent makes no connection
between the two primary data sets (the in vitro and in vivo
studies) that would enable a skilled person to arrive at a prediction about the
utility of the invention.
[39]
A patent will be valid based on a sound
prediction of utility where it sets out a factual basis for the prediction and
a sound line of reasoning connecting the facts and the predicted utility (Apotex
Inc v Wellcome Foundation Ltd, 2002 SCC 77, at para 70). In my view, both a
factual basis and a sound line of reasoning are set out in the ‘691 patent.
[40]
The factual basis consists of the data described
above. The sound line of reasoning is implicit in the data itself and would be
apparent to the skilled reader; it did not have to be explicitly laid out.
[41]
The data clearly showed that BAK had a
penetration enhancement effect, so that a relatively low dose of Bp (.01%)
combined with a relatively high dose of BAK (200 ppm), would reduce IOP
comparably to old Lumigan. The skilled person would be able to interpret the
data easily to arrive at that prediction.
[42]
On this question, I am persuaded by the
reasoning of Dr Stella, who stated:
Taken together, the
experiments disclosed in the ‘691 Patent also taught the skilled formulator
that the concentration of bimatoprost could be reduced from the previously
marketed concentration of 0.03% and still yield a formulation that achieved a
similar aqueous humor concentration of bimatoprost. The in vitro data
showed that the apparent permeability across rabbit corneal epithelial cell
layers increased with increasing concentrations of BAK from 0.005% (50 ppm) to
0.02% (200 ppm) for a formulation with 0.015% bimatoprost. From Figure 2 of the
‘691 Patent, in comparing the second and fourth bars (from the left), the apparent
permeability increased by approximately 3x when 200 ppm BAK was present instead
of 50 ppm BAK. This provides useful, relative data that the skilled formulator
would use to predict that a lower concentration of bimatoprost would be
sufficient to achieve a similar aqueous humor concentration.
Specifically, the
composition of claim 16 contains 0.01% bimatoprost and 0.02% (200 ppm) BAK.
Because the concentration of bimatoprost is reduced the skilled formulator
would ordinarily expect that the aqueous humor concentration of bimatoprost
would be similarly reduced. However, because the effect of increasing BAK on
increasing permeability has been demonstrated with the in vivo data from
Example 2, and supported by the in vitro data from Example 4, the
skilled formulator would have soundly predicted that the four-fold increase in
BAK concentration (from LUMIGAN® 0.03%) would permit a 2/3 reduction in
bimatoprost concentration (from LUMIGAN® 0.03%) and still achieve a similar
aqueous humor concentration.
Therefore, based on
the disclosure of the ‘691 Patent, the skilled formulator would have had a
factual basis for soundly predicting that the formulation claimed in Claim 16
would achieve superior corneal penetration and thus yield comparable aqueous
humor levels as the old LUMIGAN® 0.03% formulation in humans.
[43]
For Cobalt, Dr Laskar explained that the data in
the patent would not permit a skilled person to predict the utility of the
claims. His concerns related to the fact that measurements were taken only at a
single time point (60 minutes); the data did not take account of possible
trans-sclera migration of Bp in rabbit eyes; the data in Example 2 did not
permit any inferences about .01% Bp since only .03% Bp was tested; Example 4
did not test .01% Bp either; the in vitro test can give exaggerated
results; the data contain large margins of error, so some of the differences
between concentrations of Bp and BAK are probably not statistically
significant; and the FDA study showed that new Lumigan was actually inferior to
old Lumigan.
[44]
I am satisfied that all of Dr Laskar’s concerns
were answered by Allergan’s experts, as follows:
• The
purpose of the tests in the ‘691 patent was to compare concentrations of Bp
with various amounts of BAK. Since most of the Bp would be absorbed in an hour,
the 60-minute time point was appropriate. (Dr Stella)
• The
rabbit model is reliable and commonly used for studying ophthalmic drugs. Bp
can migrate across the sclera in rabbit eyes making the rabbit model
particularly suitable for Bp studies. (Dr Stella and Dr Quigley)
• While the
studies in the ‘691 patent did not use the .01% Bp formulation, they permitted
skilled readers to make conclusions about the penetration enhancing effects of
BAK, from which a reasonable extrapolation could be made regarding .01% Bp
formulations. (Dr Quigley)
• There are
no indications of any problems with the in vitro study (such as cell
damage) that could rise to skewed results. In fact, the results are consistent
with those obtained in the in vivo study. (Dr Stella)
• While
some of the differences between concentrations of Bp and BAK may not be
statistically significant, overall, the data show a linear relationship between
BAK amounts and the permeability of Bp. (Dr Stella). On cross-examination, Dr
Laskar agreed that the relationship between BAK concentration and permeability
was likely significant.
• The FDA
study found that new Lumigan was not equivalent to old Lumigan. Old Lumigan was
slightly better at lowering IOP, but new Lumigan caused fewer adverse events.
The study actually shows that new Lumigan was comparable to old Lumigan in
treating glaucoma, but had fewer side effects. This is consistent with Dr
Quigley’s clinical experience.
[45]
Accordingly, I am satisfied that the stated
utility of the claims in issue was soundly predicted based on the factual
evidence and the line of reasoning set out in the ‘691 patent. Cobalt’s
allegation to the contrary is not justified.
VI.
Issue Three – Was the ‘691 patent anticipated by
the ‘819 patent?
[46]
In its Notice of Allegation (NOA), Cobalt
alleged that the invention claimed in the ‘691 patent was anticipated by the
earlier US patent (the ‘819). It also alleged that the ‘691 patent was an
invalid selection patent. Cobalt did not pursue the latter issue in its
memorandum of fact and law before this Court. In respect of the former issue,
Cobalt relied on its written submissions and presented no oral arguments.
[47]
Cobalt contends that the ‘819 patent included a
formulation equivalent to that covered by Claim 16 of the ‘691 patent and
disclosed that it could be used to treat glaucoma, as in Claim 19 of the ‘691.
It also raised the so-called “Gillette defence” – that is, that it simply sought
to market a product disclosed in the prior art (the ‘819 patent).
[48]
Cobalt’s position is not supported by case law
on anticipation. Anticipation requires proof of disclosure of the claimed
invention sufficient to enable others to work it (Sanofi, above, at
30-37). Here, while it is true that the formulation claimed in the ‘691 patent
falls within the scope of the ‘819 patent, there is nothing in the ‘819 that
would enable a skilled person to arrive at the invention set out in Claims 16
and 19 of the ‘691 patent. As discussed above, nothing in the prior art,
including the ‘819 patent, would have alerted a skilled person to the
possibility of developing a formulation with .01% Bp and high BAK to deliver a
comparable IOP lowering effect to a .03% Bp formulation.
[49]
Accordingly, I cannot conclude that Cobalt’s
allegation that the ‘819 patent anticipates the invention claimed in the ‘691
patent is justified, or that Cobalt is simply relying on the prior art.
VII.
Conclusion and Disposition
[50]
Allergan has met its burden of showing that
Cobalt’s allegations of invalidity are unjustified. Therefore, I will grant an
order prohibiting the Minister of Health from issuing an NOC to Cobalt for its
generic version of LUMIGAN RC® until the expiry of the ‘691 patent.
Email this Content