OVERVIEW
The
Proceeding and its Background
[1]
Fournier
Pharma Inc. markets LIPIDIL EZ in 48 mg and 145 mg tablets. The active
pharmaceutical drug in LIPIDIL EZ is fenofibrate which reduces LDL (bad)
cholesterol and increases HDL (good) cholesterol in patients.
[2]
Sandoz
Canada Inc. (Sandoz) has sought approval from the Minister of Health (Minister)
to market Sandoz Fenofibrate E (the Sandoz Tablet), its generic version of
LIPIDIL EZ. In its Abbreviated New Drug Submission (ANDS) filed with the
Minister, Sandoz referenced LIPIDIL EZ “for the purpose of demonstrating
bioequivalence or bioavailability characteristics” of the Sandoz Tablet. Three
patents are listed on the register in respect of LIPIDIL EZ: Canadian Patent
2,219,475 (the ‘475 Patent), Canadian Patent 2,372,576 (the ‘576 Patent), and
Canadian Patent 2,487,054 (the ‘054 Patent).
[3]
The
Patented Medicines (Notice of Compliance) Regulations, SOR/93-133, as
amended (PMNOC Regulations) provide that the person seeking a Notice of
Compliance (NOC) (Sandoz in this case), referred to in the legislation as the
“second person,” must serve a Notice of Allegation (NOA) on the person who
filed the new drug submission that is referenced in its ANDS (Fournier Pharma
Inc. in this case), referred to in the legislation as the “first person.”
[4]
Sandoz
served a separate NOA in respect of each of the above referenced patents.
Fournier Pharma Inc. commenced three separate applications in this Court each
seeking an order, pursuant to subsection 6(1) of the PMNOC Regulations,
prohibiting the Minister from issuing a NOC to Sandoz in connection with its
orally administered 48 mg and 145 mg tablets containing fenofibrate, until
after the expiration of the relevant Canadian patents. Fournier Laboratories
Ireland Ltd., a co-owner of the ‘054 Patent, joined with Fournier Pharma Inc.
as applicants in this application (collectively Fournier). The three
applications, and the patents at issue are as follows:
a. Court
File T-991-10 filed June 24, 2010, in relation to the ‘576 Patent;
b. Court
File T-1054-10 filed June 30, 2010, in relation to the ‘475 Patent; and
c. Court
File T-1184-10 filed July 22, 2010, in relation to the ‘054 Patent.
[5]
Court
File T-1054-10 was discontinued by Fournier on January 25, 2012. The
application in Court File T-1184-10 was heard the week commencing March 26,
2012, and Court File T-991-10 was heard the following week. Each application
was heard separately; however, as noted below, an Order was issued prior to the
hearings directing that some of the evidence filed in one proceeding could be
referenced by the parties in the other proceeding.
[6]
Unless
Fournier is granted an order of prohibition, the Minister is prohibited from
issuing a NOC to Sandoz until twenty-four (24) months after this proceeding
commenced, i.e. until July 22, 2012, unless prior to that date the Court
declares that the ‘054 Patent is invalid or if one of the other conditions in
paragraph 7(2)(b) of the PMNOC Regulations apply.
[7]
Sandoz
alleges that the Sandoz Tablet does not infringe the ‘054 Patent because the size
of the fenofibrate particles in the Sandoz Tablet do not fall within the fenofibrate
particle size range set out in the claims of the ‘054 Patent. It further
submits that to the extent of any infringement, the ‘054 Patent is invalid.
[8]
The
respondent Alkermes Pharma Ireland Limited (Alkermes), a co-owner of the ‘054
Patent, filed a record in this proceeding but no Memorandum of Fact and Law,
and made no oral submissions. Its counsel attended the hearing.
Burden
of Proof
[9]
Subsection
43(2) of the Patent Act, RSC 1985, c P-4 provides that an issued patent
is presumed to be valid “in the absence of any evidence to the contrary.” In a
proceeding under the PMNOC Regulations if there is evidence in the
record that, if accepted, is capable of establishing the invalidity of the
patent, then the burden is on the applicant to establish on a balance of
probabilities (the civil standard) that the allegations of invalidity are not
justified: Abbott Laboratories v Canada (Minister of Health), 2007 FCA
153. Such evidence is in the record in this proceeding; accordingly, the issue
for the Court is whether Fournier has proved on the balance of probabilities
that all of the allegations raised by Sandoz are not justified.
The
Drug and the ‘054 Patent
[10]
Fenofibrate
is known to reduce bad cholesterol in the blood and the risk of a heart attack;
however fenofibrate comes with two central problems. First, it is almost
insoluble in water. As a result, the patient must take large doses if an
effective quantity of fenofibrate is to make its way into the blood stream. Second,
the pharmacokinetics (the absorption and distribution) of fenofibrate vary on
whether the patient takes it when in a fed or in a fasted state. It is better
absorbed when taken in a fed state and particularly when taken with fatty
foods. Dr. Mayersohn described as “diabolical” the fact that the patient is
instructed to take fenofibrate with fatty food when the patient, on a low
cholesterol diet, is otherwise instructed to avoid fatty food. The ‘054 Patent
is directed to the problem of bioequivalence in the fed and fasted states.
[11]
The
disclosure of the ‘054 Patent, entitled “Nanoparticulate Fibrate Formulations”
describes these challenges and asserts that “[t]he present invention satisfies
these needs.” It explains that:
Because
fibrates, including fenofibrate, are so insoluble in water, significant
bioavailability can be problematic. In addition, conventional fibrate, including
fenofibrate, formulations exhibit dramatically different effects depending upon
the fed or fasted state of the patient. Finally, conventional fibrate,
including fenofibrate, formulations require relatively large doses to achieve
the desired therapeutic effects. There is a need in the art for
nanoparticulate fibrate formulations which overcome these and other problems
associated with prior conventional microcrystalline fibrate formulations. The
present invention satisfies these needs.
The
‘054 Patent
[12]
Fournier
asserts that the Sandoz Tablet infringes two separate sets of claims: “(a)
claims 10, 27- 28, and 38 as they depend on claims 1-3; and (b) claim 20 which
it alleges specifically covers Sandoz’s 145 mg tablet.” The relevant claims
from the ‘054 Patent are reproduced in Appendix A.
THE EVIDENCE
[13]
The
filing of evidence was partially reversed: Fournier and Alkermes filed their
evidence on infringement and their factual evidence on invalidity first.
Sandoz then filed all of its evidence on the application and Fournier and
Alkermes then filed their responding expert evidence on invalidity.
[14]
Fournier
filed two affidavits, both sworn by its proposed expert, Dr. Fernando Muzzio:
the first dealing with the issue of infringement and the second dealing with
issues relating to invalidity. Alkermes filed two affidavits: the first sworn
by Dr. Stephen B. Ruddy, one of the inventors named in the ‘054 Patent, and the
other by Dr. Cory J. Berkland, who performed dissolution tests on the Sandoz
Tablet. Sandoz filed affidavits sworn by its proposed four experts: Dr.
Grégoire Leclair, Dr. Abu Serajuddin, Dr. Isadore Kanfer and Dr. Michael
Mayersohn. It also filed an affidavit of Deborah Zak, a law clerk, attaching
as exhibits the prior art referenced by Sandoz in its NOA and an affidavit of
Christoph Heinemann, a law clerk, attesting that samples of the Sandoz Tablet
were delivered to Fournier and Alkermes. As is discussed below, Sandoz urged
the Court to draw an adverse inference from the fact that Fournier filed no test
data on the Sandoz Tablets that had been provided to it.
Person
of Ordinary Skill in the Art (POSITA)
[15]
The
identification of the POSITA identifies the person or group of persons to whom
the patent is addressed. Because the patent is addressed to the POSITA, that
person may assist the Court which is unlikely to have familiarity with the
subject matter of the patent. However, that assistance is limited, as was
observed by Justice Hughes in Merck & Co v Pharamscience Inc, 2010
FC 510, at para 70:
Experts may assist in two ways;
first they may inform the Court as to the knowledge that a person skilled in
the art would have had at the relevant time, so as to bring that knowledge to
bear reading both the description and the claims; second, an expert may assist
in explaining any technical terms not within the experience expected of a
Court.
[16]
Fournier’s
expert, Dr. Fernando Muzzio, attests that the POSITA “would have a PhD in
pharmaceutics along with 1-3 years of experience in the area of formulation of
pharmaceutical products. Alternatively, the POSITA could have a Masters degree
in pharmaceutics with 5-7 years of experience in the area of formulation of
pharmaceutical products.” That description of the relevant POISTA does not
vary significantly from that offered by Sandoz’s experts and I accept it as a
statement of the qualifications of the POSITA. It is the same description
accepted to be the POSITA in T-991-10.
Expert
Evidence Filed By Fournier
Dr.
Fernando Muzzio (On Infringement)
[17]
Dr.
Muzzio has a B.Sc. in Chemical Engineering from the University of Mar del
Plata, Argentina, and a Ph.D. from the University of Massachusetts in Chemical
Engineering. He is a Professor II of Chemical Engineering at Rutgers University and teaches a course called Pharmaceutical Unit Operations which includes
lectures on size reduction of particles and milling. He directs a staff of
approximately 130 persons at the National Science Foundation Engineering
Research Center (NSFERC) where he oversees projects relating, in part, to
nanoparticle stabilization of pharmaceutical products. As part of NSFERC’s
industrial mentor program, he interacts and collaborates closely with an
additional 120 representatives. He is also the Director of the National
Science Engineering and Research training program in Nanopharmaceutical
Engineering at Rutgers University. Additionally, he teaches courses for
industry and regulatory agencies such as the U.S. Food and Drug Administration
and he has authored approximately 200 peer-reviewed scientific papers and book
chapters on subjects which include powder mixing fundamentals and mixing and
segregation in tumbling blenders.
[18]
In
Dr. Muzzio’s affidavit on infringement, he says that Sandoz’s allegation that
the Sandoz Tablet would not infringe any claims of the ‘054 Patent is
incorrect. He provides his reading of the claims in the ‘054 Patent and notes
that Sandoz’s only allegation of non-infringement with respect to claims 1-3
and 24-31 is:
The dispersion used to prepare the Sandoz tablets
prior to formulation in the solid dosage form (the tablet) comprise particles
which are not fenofibrate per se and/or which particles are not within
the claimed particle size ranges. Furthermore, the Sandoz Tablets once
formulated and in final dosage forms do not comprise particles of fenofibrate per
se in the claimed particle size range.
[19]
Dr.
Muzzio reviews the manufacturing procedure of the Sandoz Tablet and refers to
various figures and tables in Sandoz’s ANDS relating to the particle size of
the fenofibrate and says that he has no doubt that the particle size of the fenofibrate
in the Sandoz Tablet are within the ‘054 Patent’s claimed ranges. Finally, he
notes that there are no manufacturing steps [ …] [ …]
[ …] such that the Sandoz Tablets,
once formulated and in final dosage form, do not comprise particles of
fenofibrate per se within the claimed particle size distribution range.
[20]
Dr.
Muzzio responds to Sandoz’s allegation that claims 13-16 are not infringed
because the procedure to measure Tmax is not defined in the ‘054
Patent by pointing out that Tmax could be measured as part of a
procedure well known as of May 2004.
[21]
Dr.
Muzzio responds to Sandoz’s allegation that claims 34-38 would not be infringed
because the Sandoz Tablet does not have the dissolution required by providing a
table illustrating the fenofibrate dissolution in the Sandoz Tablet and noting
that it does fall within all of the ranges in claims 34-38.
[22]
[
…]
[ …]
[ …]
[ …] .
Dr.
Fernando Muzzio (On Validity)
[23]
Dr.
Muzzio opines that the inventive concept of the ‘054 Patent is that: (a) it has
essentially the same bioavailability in the fed and fasted states; (b) the D50
particle size of fenofibrate is less than about 200 or 150 nm; (c) it has a
rapid dissolution rate; (d) it redisperses to a particle size no larger than 2 μm;
and (e) it permits reduced dosing.
[24]
Dr.
Muzzio says that none of the prior art referenced by Sandoz discloses or
enables the POSITA to come to the inventive concept of the ‘054 Patent. Moreover,
he says that the inventive concept of the ‘054 Patent would not have been
obvious to the POSITA. In his opinion, the POSITA was unaware that it was
possible to achieve essentially the same bioavailability in the fed and fasted
state with fenofibrate compositions. The prior art included many attempts to
reduce the food effect, but no one achieved what the ‘054 Patent claimed and
achieved, the elimination of the food effect. He notes examples in the ‘054
Patent which, in his opinion, compel the conclusion that the 145 mg fenofibrate
formulation of the invention would be bioequivalent to the prior art 200 mg
formulation.
[25]
Dr.
Muzzio states that the different aspects of the inventive concept are what
render the ‘054 Patent useful. He disagrees with Sandoz’s allegation that the
‘054 Patent lacks sound prediction. He says that the particle size was
predicted through Examples 5 and 6 in the ‘054 Patent; guidance was given for
the choice of surface stabilizers as the ‘054 Patent stated the preferred stabilizers
and referenced the Handbook of Pharmaceutical Excipients; and the
reduced dosage could be predicted through Example 8.
[26]
Dr.
Muzzio disagrees with the Sandoz experts who opined that the inventors of the
‘054 Patent were not entitled to claim any fenofibrate composition apart from
the precise composition in Example 5 of the ‘054 Patent. Dr. Muzzio highlights
and relies on one of the statements in the description which reads:
The following examples are given to illustrate the
present invention. It should be understood, however, that the invention is not
to be limited to the specific conditions or details described in these
examples.
According to Dr. Muzzio, the POSITA
would have understood that the claimed invention was not the formulation of
Example 5, but rather that fenofibrate compositions with the claimed particle
sizes can be made and will achieve bioequivalence in the fed and fasted
states.
[27]
Dr.
Muzzio writes that the ‘054 Patent provided a sound line of reasoning to
predict that other surface stabilizers not found in the stated example would
likewise be effective. He notes that the dissolution profile was not overly
broad and was demonstrated in Example 8. Similarly, he notes that the particle
sizes were shown in Tables 1, 4, 5 and 7.
[28]
Dr.
Muzzio addresses and disagrees with Dr. Kanfer’s statement that the boundaries
of the claims are undefined. In Dr. Muzzio’s opinion, the POSITA would have
understood that the invention was limited to phospholipid-free compositions,
for oral administration, with bioequivalence in fed and fasted states. No
undue experimentation would be required to come to formulations covered by the
invention described in the ‘054 Patent.
Expert
Evidence Filed By Sandoz
Dr.
Grégoire Leclair
[29]
Dr.
Leclair has a Ph.D. in Pharmaceutical Sciences from Université de Montréal,
where he is currently employed as an Assistant Professor. His current research
includes nanoparticulate compositions for active ingredients which have poor
solubility in water.
[30]
Dr.
Leclair notes that the only examples in the ‘054 Patent which are shown to
exhibit bioequivalence in the fed and fasted states are Examples 5 and 6 which
relate to tablets containing 160 mg nanoparticulate fenofibrate. In his
opinion, the POSITA would not understand any of the other ranges of particle
size to be bioequivalent in the fed versus fasted states and therefore, in his
view, the particle ranges claimed in the ‘054 Patent are too broad. Similarly,
he says that the redispersion of the particle size is too broad because the
claims do not contain a limitation. The claims in the ‘054 Patent merely state
that redispersed particle sizes can be determined with reference to another U.S. patent.
[31]
Dr.
Leclair says that the lack of limitations relating to surface stabilizers in
the claims of the ‘054 Patent also renders the claims of the ‘054 Patent too
broad. He notes that all the preferred formulations set out in the ‘054 Patent
use as surface stabilizers docusate sodium (DOSS) and hypromellose, and he says
that the POSITA would know that the inventors could not predict which other
stabilizers would provide compositions that are bioequivalent in the fed versus
fasted states; because, as the ‘054 Patent indicates, not all stabilizers will
work.
[32]
Additionally,
Dr. Leclair advances that there is no sound line of reasoning given as to why
particle sizes under 2 μm would work, nor as to why other surface
stabilizers could be effective. Dr. Leclair also notes that the ‘054 Patent
does not describe how to measure the particle size in the composition.
[33]
Finally,
Dr. Leclair says that Dr. Muzzio’s assertion that the fenofibrate in the Sandoz
Tablet has the same particle size as in the suspension used in the process to
make the tablet is not correct. In his opinion, aggregation will inevitably
occur [ …]
[
…] . Dr. Leclair describes a procedure he
designed in order to determine the minimum particle size of the fenofibrate in
the Sandoz Tablet and states that Dr. Muzzio has wrongly assumed that there is
no agglomeration in the Sandoz Tablet. Dr. Leclair’s results show that the
mean size of the fenofibrate in the Sandoz Tablet is greater than [ …] ,
with a D50 greater than [ …] and a D90 greater than [ …] . Moreover, [ …]
[ …]
[ …] .
[34]
Lastly,
Dr. Leclair disagrees with Dr. Muzzio that the inventors of the ‘054 Patent
solved the food effect problem by making very small particles. Dr. Leclair
states that the inventors did not demonstrate any comparison to other particle
sizes. Also, he challenges Dr. Muzzio’s statement that: “The inventors report
that the nanoparticlulate fenofibrate tablet of the invention was as effective
at a dosage of 160 mg as the 200 mg dosage available in the prior art.” Dr.
Leclair notes that the inventors state that:
Finally, as shown by the data in Table 16, below,
administration of a 160 mg nanoparticulate fenofibrate tablet in a fasted state
is not bioequivalent to administration of a 200 mg conventional
microcrystalline fenofibrate capsule (TRICOR®) in a fed state. This is because
CI 90% for the two treatments is outside 0.80 to 1.25 for AUC and Cmax
[emphasis by Dr. Leclair].
Dr.
Abu Serajuddin
[35]
Dr.
Serajuddin obtained his Bachelor in Pharmacy at Dhaka University in 1967, his
Masters of Science in pharmaceutics from Colombia University in 1976 and his
Ph.D. in Industrial Pharmacy from St. John’s University in 1982. His thesis
related to solubility and dissolution of poorly soluble drugs. Since September
2008 he has been a Professor of Industrial Pharmacy at St. John’s University. Prior to joining St. John’s University, Dr. Serajuddin worked for more than 30
years in the pharmaceutical industry in a variety of positions relating to the
development of drug formulations.
[36]
Dr.
Serajuddin explains the general knowledge of the POSITA at the relevant time
and states that the advantages described in the ‘054 Patent as set out below
“described the usefulness of the invention of the ‘054 Patent promised by the
inventors:”
(1) smaller tablet or other solid dosage form size;
(2) smaller doses of drug required to obtain the same pharmacological effect;
(3) increased bioavailability; (4) substantially similar pharmacokinetic
profiles of the nanoparticulate fibrate, preferably fenofibrate, compositions
when administered in the fed versus fasted state; (5) improved pharmacokinetic
profiles; (6) bioequivalence of the nanoparticulate fibrate, preferably
fenofibrate, compositions when administered in the fed versus fasted state; (7)
an increased rate of dissolution for the nanoparticulate fibrate, preferably
fenofibrate, compositions; (8) bioadhesive fibrate, preferable fenofibrate,
compositions; and (9) the nanoparticulate fibrate, preferably fenofibrate,
compositions can be used in conjunction with other active agents useful in
treating dyslipidemia, hyperlipidemia, hypercholesterolemia, cardiovascular
disorders, or related conditions.
[37]
Dr.
Serajuddin reviews the claims of the ‘054 Patent and finds that they are
broader than the invention itself. According to him, all the advantages
“address what the composition of the invention does or is supposed to do but
not what the composition is.” He also notes that notwithstanding the
difficulty in selecting the surface stabilizer to be used, the inventors refer
to a broad group of surface stabilizers that could be used in the compositions
and then refer to thousands of possible surface stabilizers and combinations.
According to him, the POSITA would know that there was nothing inventive in
this or any of the other characteristics of the invention. Additionally, he
states that the promised utilities cannot be predicted for the claimed
compositions. In his opinion, the POSITA would not have been able to predict
that the composition was physically stable and would be bioequivalent in the
fed versus fasted states.
[38]
Dr.
Serajuddin also believes that the POSITA would require experimentation to
practice what is found in the invention. In this regard, Dr. Serajuddin
targets the very broad range of stabilizers. He says that to select those that
would work as promised would require more than routine testing.
[39]
Dr.
Serajuddin reviews prior publications and states that from each of them the
POSITA would have been able to prepare the compositions covered by the claims
he identified in the ‘054 Patent.
[40]
Dr.
Serajuddin disagrees with the conclusion of Dr. Muzzio that the Sandoz Tablet
would infringe the ‘054 Patent. Specifically, he disagrees with Dr. Muzzio’s
statement that a tablet taken orally will redisperse in gastrointestinal fluids
to the same size as the original particles used to prepare the formulation.
Dr. Serajuddin refers to United States Patent 6,375,986 (US ‘986 Patent) which
found redispersion to the particle size in the original suspension to be a
problem. He states:
US ‘986 clearly shows that Dr. Muzzio’s statement is
incorrect. In US ‘986, Elan states that it has discovered a combination of
surfactants and surface stabilizers (for example DOSS, SLS and PVP) that
resulted in an improvement in the particle size of the redispersed materials.
However, the mean particle sizes of the redispersion was still more than 20
times larger than the mean particle size of the original suspension …
Dr.
Isadore Kanfer
[41]
Dr.
Kanfer has a B.Sc. (Pharm) as well as a B.Sc. (Hons) and a Ph.D. (Pharm) from Rhodes University, South Africa. He is currently Professor and Dean Emeritus in the
Faculty of Pharmacy of Rhodes University where he teaches undergraduate,
Masters and Ph.D. courses relating to biopharmaceutical analysis,
bioavailability, dissolution, bioequivalence, drug interaction, drug absorption
and drug regulation.
[42]
Dr.
Kanfer reviews the ‘054 Patent and echoes Dr. Serajuddin in stating that the
advantages listed in the description are the promises of the invention. Dr.
Kanfer adds that the POSITA would also understand that the inventors were
promising the nanoparticulate compositions to be stable. He provides graphs
and tables to explain that none of the promises in the ‘054 Patent were
demonstrated or soundly predictable.
[43]
Dr.
Kanfer says that the claims in the ‘054 Patent are too broad and cover
formulations that would not work. He emphasizes that there is no direction as
to which surface stabilizers would work and, as such, bioequivalence in the fed
versus fasted states cannot be predicted; undue experimentation would be
required.
Dr.
Michael Mayersohn
[44]
Dr.
Mayersohn holds a B.Sc. in Pharmacy from Columbia University and was awarded a
Ph.D. in Pharmaceutics/Pharmacokinetics from State University of New York in
1971. Since 1983 he has been a Full Professor of Pharmaceutical Sciences at
the University of Arizona. He has conducted research projects which involved
examination and characterization of oral bioavailability and pharmacokinetic
properties of drugs and their metabolites in animals and humans.
[45]
Dr.
Mayersohn compares the ‘054 Patent to the prior art. In his view, Canadian Patent
2,423,335 (the ‘335 Patent) and WO Patent 02/24193 (the WO ‘193 Patent)
describe and enable fenofibrate formulations having a particle size of most
preferably 100 nm to 2 µm and although those prior patents prefer the use of
phospholipid stabilizers, they also describe and enable the use of surface
stabilizers that are not phospholipids. Additionally, he says that both can be
administered without regard to food. Even if the ‘054 Patent was not
anticipated, everything in the prior art rendered the inventive concept
obvious.
[46]
According
to Dr. Mayersohn, the inventive concept of claims 1-3 are: (a) a stable
nanoparticulate fenofibrate composition; (b) a particular particle size for the
fenofibrate; (c) at least one surface stabilizer; (d) bioequivalence; and (e)
redispersion in a biorelevant media. He says that the POSITA would not
understand phospholipid-free to be part of the inventive concept since the ‘054
Patent lists useful stabilizers which include phospholipids. Alternatively, being
phospholipid-free would have been obvious to the POSITA.
[47]
Dr.
Mayersohn reviews the prior art and states that the ‘335 Patent, WO ‘193
Patent, United States Patent 5,145,684 (US ‘684 Patent), United States Patent
5,510,118 (US ‘118 Patent), WO 01/21154 (WO ‘154 Patent), United States Patent
6,177,103 (US ‘103 Patent), and United States Patent 6,368,620 (US ‘620 Patent)
all describe and enable the invention of the ‘054 Patent. Moreover, Dr.
Mayersohn says, these prior patents render the inventive concept of the ‘054
Patent obvious.
Additional
Expert Evidence
[48]
One
issue of significance in this application is the particle size of the
fenofibrate particles in the Sandoz Tablet. It is also an issue of
significance in T-991-10. A few days prior to the hearing of these applications,
Fournier brought a motion seeking an order allowing the cross-referencing of
certain affidavits and cross-examination transcripts as between Court Files
T-1184-10 and T-991-10 on the basis that the evidence filed in these two
applications by Sandoz, relating to the particle size of the fenofibrate in the
Sandoz Tablet, was contradictory and that the interests of justice demanded
that all of this evidence be before the applications judge in both matters.
[49]
Sandoz
resisted the motion, in part, on the basis that there is no conflict in the
evidence, and in part of the basis that it would suffer prejudice because it
has not had an opportunity to file expert evidence to assist the Court is
assessing whether there truly is a conflict in the evidence.
[50]
I
granted the motion on the basis that:
[T]he interests of justice in having all of the
relevant evidence before the Court and avoiding the possibility of conflicting
findings of fact outweigh any prejudice of the sort Sandoz asserts it will
suffer. Further, it is not clear that expert evidence of the type it says it
would file is necessary or helpful to the Court in assessing what are
statements of fact by the various expert witnesses.
[51]
The
issue as to whether there was any actual conflict in the evidence was to be
determined as part of the decision on the merits of each application, if
necessary.
[52]
The
portions of the Order relevant to this application are paragraphs 1 and 2 which
read as follows:
1. The portions of the affidavit of Dr. Muzzio dated February 21,
2011, filed in T-991-10 on particle size, as well as the portions of his
cross-examination transcript relating to particle size, are incorporated into
the Applicants’ Record in T-1184-10.
2. The portions of the affidavits of Drs. Bogardus and Fairhurst in
T-991-10 on particle size, as well as the portions of their cross-examination
transcripts relating to particle size, are incorporated into the Applicants’
Record of T-1184-10.
[53]
The
evidence set out in these documents and the relevance, if any, in this
application is discussed below.
Challenges
to the Opinions of the Experts
[54]
Both
Fournier and Sandoz raised complaints about the expert(s) proposed by the other
and made representations as to whether that evidence ought to be accepted and,
if so, the weight it ought to be given.
[55]
Fournier,
at paragraph 12 of its Memorandum of Fact and Law, asked the Court to approach
with caution the evidence of the experts put forward by Sandoz:
Sandoz put forward a 413 paragraph affidavit from
Dr. Mayersohn (a pharmacist), a 384 paragraph affidavit from Dr. Serajuddin (a
pharmacist), a 361 paragraph affidavit from Dr. Kanfer (a pharmacist), and a
241 paragraph affidavit from Dr. LeClair (a pharmacist). Sandoz’s experts were
not always forthcoming in their cross-examinations, were difficult, refused to
answer important questions, and accepted direction from Sandoz’s counsel to
avoid answering questions. Not only were these witnesses advocates for
Sandoz’s position, they clung to unreasonable positions with uncommon
truculence, and in some cases gave evidence incapable of belief. Worse, Sandoz
has put forward witnesses in this case who have given opinions directly
opposite the position it has taken in the Notice of Allegation of T-991-10. If
any of the issues in this application fall to be determined on the basis of any
assessment of credibility, it is respectfully submitted that the evidence of
Sandoz’s experts should be approached with caution, given the above [references
omitted].
[56]
Sandoz,
for its part, at paragraphs 12 and 13 of its Memorandum of Fact and Law,
challenged the evidence of Fournier’s expert, Dr. Muzzio:
Fournier’s only expert Dr. Muzzio admits that he
does not have the credentials of the person to which the ‘054 Patent is
addressed by his own definition.
Fournier has not challenged the qualifications of
Sandoz’s experts or their expertise to provide the opinions in their affidavits
and none of their opinions were shaken on cross-examination. Sandoz challenges
the qualifications’ of Fournier’s expert Dr. Muzzio. Fournier did not permit
Sandoz to complete his cross-examination [references omitted].
[57]
Shortly
before this application was heard, Sandoz brought a motion seeking to strike
from the Application Record in this proceeding the affidavits of Dr. Muzzio on
the ground that he was unresponsive and long winded, and because the respondent
had been prevented from completing the cross-examination, despite being given
additional time. I dismissed that motion and provided the following short
endorsement setting out my reasons for so doing:
I accept that Dr. Muzzio’s responses to the
questions asked were neither brief nor direct. He was, as was suggested by
counsel, a difficult witness. However, Sandoz cross-examined this witness for
four days and made no request at the end of that period for further
cross-examination. It cannot now, a few days prior to the hearing on the
merits, seek as a remedy the striking of that evidence when it failed to seek
any additional time when it could have been provided.
It is also submitted that “Dr. Muzzio breached his
signed Code of Conduct by relying on testing he had not done in his lab to
support his opinion regarding sound prediction without referring to it in the
Muzzio Validity Affidavit.” I am not satisfied on the balance of probabilities
that the testing complained of was in the mind of the witness and formed a part
of the basis for the opinion set out in the affidavit he provided. Rather, it
appears more likely that this “testing” came to mind as he was being
cross-examined and he mentioned it as a further basis to support his opinion.
I find no breach of the Code of Conduct. Had I found a breach, given its
nature, I would not have struck his evidence but rather would have determined
that the violation would go to the weight to be given his evidence where there
was conflicting evidence.
[58]
It
will be noted that the complaint raised by Fournier as to the challenges it
faced in cross-examining the experts put forward by Sandoz mirror the complaint
raised by Sandoz with respect to Fournier’s witness and which has been dealt
with by the Court. For the reasons given in rejecting Sandoz’s motion to
strike, I likewise give little weight to Fournier’s complaint. To paraphrase
that endorsement, and with reference to Fournier’s complaint, I say this:
Sandoz’s witnesses on some occasions were difficult; however, Fournier
completed its cross-examination of them and made no request for further
cross-examination or to have any objected questions answered.
[59]
I
find inappropriate and unfair the submission that the experts were advocates
for Sandoz’s position. Each witness provided an affidavit as required by Rule
52.2 of the Federal Courts Rules, SOR/98-106, attesting that he or she
had read and agreed to be bound by the Code of Conduct for Expert Witnesses.
Paragraphs one and two of that Code provide as follows:
1. An
expert witness named to provide a report for use as evidence, or to testify in
a proceeding, has an overriding duty to assist the Court impartially on matters
relevant to his or her area of expertise.
2. This
duty overrides any duty to a party to the proceeding, including the person
retaining the expert witness. An expert is to be independent and objective. An
expert is not an advocate for a party [emphasis added].
[60]
Fournier’s
challenges to the impartiality of these witnesses are set out in footnotes 21
and 22 of its Memorandum of Fact and Law. These footnotes include the
following two serious allegations: a witness’s “claim to have calculations –
undisclosed contrary to the Code of Conduct – that yielded a 10% difference is
reported to be 61%,” and similarities between the wording used in portions of
the affidavits of Dr. Serajuddin and Dr. Mayersohn.
[61]
I
reject the submission that a Sandoz witness, contrary to the Code, failed to
make a disclosure as has been suggested. Fournier cites as support for this
allegation the cross-examination of Dr. Mayersohn taken October 28, 2011, Q
664-669 and 754-770. I see nothing in those passages that makes any reference
to a claim to have calculations that yielded a 10% difference. Dr. Mayersohn
at Question 665 said that he “pulled that [figure] out of the air” and at
Question 750 said that it was a “guesstimation.” I find that Fournier’s
characterization of the evidence of Dr. Mayersohn’s evidence in this respect to
be inaccurate and its submission that he breached the Code of Conduct to be
without foundation.
[62]
I
considered an allegation that a witness copied another’s words in his affidavit
in Janssen-Ortho Inc v Novopharm Limited, 2010 FC 42 and commented that
“fairness and reason dictate that when a party proposes to make a submission
that the words contained in a sworn affidavit are those of another and do
not express the views of the affiant, that proposition ought to be squarely
put to the affiant in order that he has an opportunity to respond [emphasis
added].” Having reviewed the cross-examination of Dr. Serajuddin, I have
concluded that Fournier did just that.
[63]
Dr.
Serajuddin, in response to a direct question, answered that he did not copy any
portion of his affidavit from Dr. Mayersohn’s affidavit, that he had no
discussion with him, had no copy of his affidavit, and provided no copy of his
own affidavit to Dr. Mayersohn. He describes the process that was followed in
preparing his affidavit. He prepared text and then counsel for Sandoz would
question him on it and he would “try to explain it” and that “there would then
be maybe a proposal to put it slightly differently [and] I would look at it and
then maybe agree.” This manner of developing expert affidavit evidence
intended to assist the Court is not offensive or inappropriate. I would be
surprised if counsel for Fournier did not follow much the same process when
having its experts prepare and finalize their affidavits.
[64]
At
the end, what is important when considering an expert’s affidavit is two-fold.
First, that the affidavit is helpful to the Court. This requires that it be
written in a manner that is understandable by judges who are not experts in the
area. Drafting in this manner is something that many experts, especially it
seems those in the scientific community, are ill-equipped to do without
assistance. An affidavit written in technical language understandable only by
another expert is of little assistance to a court. Second and most important
is the requirement that the words set out in the affidavit express the views of
the affiant. Dr. Serajuddin on cross-examination stated: “[I]t was my
opinions and that is why I signed the affidavit.” Fournier makes no real
challenge to that assertion.
[65]
For
these reasons, I reject the submission of Fournier that the evidence of
Sandoz’s experts ought to be approached with caution. This evidence, like all
evidence, must be weighed and considered within the context of the case as a
whole, the other evidence, and the submissions made by the parties.
[66]
There
is merit in the submission of Sandoz that Dr. Muzzio does not have the
credentials of the POSITA by his own definition. He does not profess to have
these credentials; rather, in his affidavit sworn February 23, 2011, he states
with respect to the POSITA: “I have worked with many individuals with these
credentials and can speak to their understanding of this area of science and
their capabilities.” Some of the evidence of Dr. Muzzio might best be
described as “scientific hearsay” insofar as it is the evidence of a scientific
witness without the required background offering an opinion on what scientists
with the relevant background would understand a patent that is directed to them
to mean. Taken to the extreme, it is akin to a physicist saying that he has
worked with biologists and chemists and therefore can speak to their
understanding of a document that is directed to biologists or chemists. The evidence
of that physicist would be given little weight.
[67]
The
situation of Dr. Muzzio is not quite akin to the extreme example offered
above. Dr. Muzzio does have education and expertise that is relevant and
therefore I am not prepared to simply strike his affidavit. For example, as a
chemical engineer, he is able to speak to the meaning of some of the terms used
in the ‘054 Patent and assist the Court in understanding basic chemistry and
formulations relevant to the invention. However, whenever his evidence is in
conflict with that of the experts put forward by Sandoz, or is not supported by
another expert or scientific evidence on a matter properly within the role of
an expert, I prefer their evidence as Fournier made no challenge to their
credentials or the fact that they were experts and met the definition of the
POSITA.
THE
INTERPRETATION OF THE ‘054 PATENT
[68]
What
is the invention captured by the ‘054 Patent? The Supreme Court has instructed
that patent claims are to be purposively interpreted, the key to which is “the
identification by the court, with the assistance of the skilled reader, of the
particular words or phrases in the claims that describe what the inventor
considered to be the “essential” elements of his invention [emphasis
added]:” Whirlpool Corp v Camco Inc, 2000 SCC 67, at para 45. Claims
construction is a question of law: Bristol-Myers Squibb Co v Apotex Inc,
2007 FCA 379, at para 27. The purposive approach to construction ought to be
used when construing the patent as a whole. The aim is to determine the
inventor’s view as to what it is that he or she has invented and how that
invention is to be made or the process described followed; how the invention is
to be used or the result obtained from the invented process; and what the invention
does.
[69]
What
then are the essential elements of the ‘054 Patent? This is where the parties
are in fundamental disagreement. The answer to this question goes a long way
to resolving the numerous other issues between these parties.
Bioequivalence
and Pharmacokinetic Profile
[70]
Each
of the three independent claims (claims 1-3) stipulates that the composition
“exhibits bioequivalence upon administration to a human subject in a fed state
as compared to administration to a human subject in a fasted state.” These
claims define how bioequivalence is to be “established” namely by “(a) a 90%
Confidence Interval for AUC which is between 80% and 125%, and (b) a 90%
Confidence Interval for Cmax, which is between 80% and 125%.” This
is the pharmacokinetic profile (PK Profile) of the invention. Fournier says
that this PK Profile is an essential element; Sandoz says it is not.
[71]
The
witnesses agree that there are two different guidelines that one may use when
speaking of bioequivalence in the fed and fasted state: the US Food and Drug
Administration (USFDA) guideline, and the European Medicines Agency (EMEA)
guideline.
[72]
Dr.
Kanfer informs the Court that under the USFDA “requirements, bioequivalence is
established by a 90% [Confidence Interval] for AUC and Cmax of
80%-125% [whereas] under the EMEA requirements, bioequivalence is established
by a 90% [Confidence Interval] for AUC of 80%-125% and a [Confidence Interval]
for Cmax of 70%-143%.” As he notes, “the EMEA requirement permits a
wider limit in the Cmax between two products to establish
bioequivalence.” What is also clear is that a composition that is
bioequivalent as defined by the USFDA guideline will also be bioequivalent as
defined by the EMEA guideline, but that a composition that is bioequivalent as
defined by the EMEA guideline will not necessarily also be bioequivalent as
defined by the USFDA guideline.
[73]
Sandoz
says that the PK Profile in claims 1-3 is not an essential element of the
invention for two main reasons. First, it says that it is ambiguous because
reference is made elsewhere in the disclosure to the EMEA guideline, and the PK
Profile in claims 1-3 “is inconsistent with dependant claims 4-7 which include
compositions that do not fall within the USFDA Confidence Intervals.” Second,
Sandoz says that the Cmax in the PK Profile is not essential for
bioequivalence.
[74]
In
relation to Sandoz’s first allegation, I note that a composition that is
marketed in Canada as bioequivalent must meet the USFDA guideline. The PK
Profile set out in claims 1-3 reflects the USFDA guideline. I am unable to
accept that the PK Profile is arbitrary since it reflects exactly the USFDA
bioequivalence requirement. Given that the inventors of the ‘054 Patent were
seeking bioequivalence, given that Canada requires that bioequivalence be
established using the USFDA guideline, and given that the ‘054 Patent is a
Canadian patent for an invention to be marketed in Canada, I find that the PK
Profile as set out in claims 1-3 is a minimum requirement and an essential feature
of the invention.
[75]
Both
Dr. Kanfer and Dr. Mayersohn point to the AUC difference set out in claims 4-7
and note that claims 4-7 include compositions that do not meet the USFDA
guidelines. Accordingly, they say, the PK Profile cannot be an essential element.
That view seemed to be confirmed by Dr. Mayersohn who swore that “an Ordinary
Skilled Worker would not deem strict compliance with this element to be
important to the function of the invention.” However, on cross-examination he
admitted that strict compliance with the USFDA guidelines, although not the
EMEA guidelines, was a requirement of the claims and a person skilled in the
art would read the claims of the ‘054 Patent as requiring such:
1161 Q. Okay,
so you have to meet the US/Canada requirements?
A.
Definitely.
1162 Q. Got
it, and that is bioequivalence as determined by the 90 percent confidence
interval for AUC and Cmax of 80 to 125?
A.
Correct.
1163 Q. And
that is how the skilled person would understand claims 1 to 3 of the ‘054
Patent?
A.
The only reason I am hesitating, Counsel, is I want to be sure that I am being
consistent with what I believe to be the claimed invention, which is paragraph
17, and that includes bioequivalence, so the answer is yes.
[76]
Even
if I had accepted Sandoz’s reasoning, then at best I would say that claims 4-7
make it unclear or ambiguous as to what PK Profile constitutes bioequivalence
in the inventors’ minds. From there, I would agree with Fournier that the
disclosure at pages 62-64 of the ‘054 Patent makes it clear that meeting the
USFDA guideline is what makes the composition bioequivalent in the fed and
fasted state and that this is an essential element of the invention. The
inventors at those pages discuss the data contained in Table 14, the pharmacokinetic
parameters of the invention disclosed in testing and examine that data with
specific reference to the UDFDA guideline and no other and state as follows:
Accordingly, pursuant to [USFDA] regulatory
guidelines, administration of a nanoparticulate fenofibrate tablet in a fasted
state is bioequivalent to administration of a nanoparticulate fenofibrate
tablet in a fed state. Thus, the invention encompasses a fibrate
composition wherein administration of the composition to a subject in a fasted
state is bioequivalent to administration of the composition to a subject in a
fed state [emphasis added].
[77]
Further,
at page 63 of the ‘054 Patent, when comparing the 160 mg nanoparticulate tablet
with the 200 mg TRICOR tablet (the existing product) the inventors state that
Table 16 shows that when their product is administered in a fasted state it is
not bioequivalent to TRICOR in the fed state “because [Confidence Interval of]
90% for the two treatments is outside 0.80 to 1.25 for AUC and Cmax
[emphasis in the original]” meaning it is outside the USFDA guideline.
Critically, as was noted by Fournier, the Cmax range of 1.08 to
1.304 in that example meets the EMEA guideline as that criterion goes up to
1.43 whereas the USFDA guideline ends at 1.25. In my view, this makes it
abundantly clear that the inventors consider that bioequivalence is met only
if the USFDA guideline has been met.
[78]
Sandoz’s
second allegation, that the Cmax is not essential, relies on the
Dalton Rule. This rule, according to Dr. Mayersohn’s affidavit at paragraph 98
teaches that “if the arithmetic mean for AUC and Cmax differs by
less than 10% or less than 5%, an Ordinary Skilled Worker would understand that
the 90% [Confidence Interval] of 80-125% for AUC and Cmax,
respectively, would be met.”
[79]
This
proposition was put to another of Sandoz’s experts, Dr. Kanfer, who responded
as follows:
407. Q. If you
came to the FDA with evidence of a difference in means of less than 10 percent
and no other evidence of bioequivalence, they would say, "That doesn't cut
it"?
A. They would say,
"Calculate the confidence intervals."
408. Q. You can't
calculate the confidence intervals just from the difference in the means, can
you?
A. You need to look at the study, you need to see
that the study is powered correctly.
409. Q. Right.
A. It is far more complicated than just looking at a
simple difference between means or ratio.
[80]
This
exchange explains that the arithmetic mean alone cannot give you the confidence
intervals required for USFDA approval. As such, if the invention does not
provide a Cmax confidence interval of 80%-125%, it will not
necessarily obtain approval. This, in my view, is sufficient to reject
Sandoz’s second allegation.
[81]
It
is my finding that bioequivalence established by a 90% Confidence Interval for
AUC and Cmax of 80%-125% is essential for the invention and claims
1-3.
At
Least One Surface Stabilizer and Phospholipid-Free
[82]
Each
of the independent claims 1-3 specifies that the composition is
phospholipid-free. Fournier says that this is an essential element of the
invention. Sandoz asserts that it is not and that it has been included in the
claims in order to avoid WO 02/24193, a prior art reference containing
phospholipids.
[83]
The
‘054 Patent requires “at least one surface stabilizer” and both parties agree
that it is an essential element of the invention. Both parties also agree that
it is an essential element of the invention that it be a “stable fenofibrate
composition.” WO 03/013474, a prior art reference relied on by Sandoz, at page
7 of its disclosure states that “fenofibrate suspensions stabilized using only
phospholipids are reported as not stable.” Accordingly, if the essential
element, the “at least one surface stabilizer” was a phospholipid, then the
invention would not work as it would not be stable. For this reason, I agree
with Fournier that, as the ‘054 Patent is drafted, it is an essential element
of the invention that it be phospholipid-free.
Redisperses
in a Biorelevant Media
[84]
Claim
1 uses the word “disperses” whereas claim 2 and claim 3 use the word
“redisperses.” Nothing turns on this slight variation in terminology. What is
meant is that the component parts of the composition disseminate in a
“biorelevant media.” Dr. Serajuddin informs the Court that the particle size
of the redispersed fenofibrate would be less than 2µm:
Claim 1 refers to “disperses” and claims 2 and 3
refer to “redisperses”. The particle size on redispersion would be restricted
by the meaning of “stable” pharmaceutical composition. As such, the particle
size would be limited to less than 2 µm.
Dr. Muzzio agrees, stating: “ [I]t
is my opinion that the POSITA reading the ‘054 Patent as a whole as of May 2004
would understand that the inventive concept of the claims requires compositions
that dispersed or redisperse in a biorelevant media to a particle size of less
than 2 microns (2000 nm).”
[85]
Dr.
Serajuddin further informs the Court that the POSITA would know that there are
two biorelevant media for the fenofibrate composition: gastric fluid and intestinal
fluid. Dr. Kanfer informs the Court that the POSITA would know the recipes for
formulating simulations of these fluids.
[86]
Is
it an essential element that the composition redisperses to a particle size of
less than about 2 µm? I am of the view that it is. Dr. Muzzio was
cross-examined quite extensively regarding the size of the redispersed
fenofibrate particles. He informs us, as does the prior art, that
bioavailability of fenofibrate is improved as the particle size is reduced.
Sandoz, in its Memorandum of Fact and Law captures this as being common general
knowledge of the POSITA:
The rate of dissolution of poorly water soluble
drugs could be increased by decreasing the particle size and thereby increasing
the surface area of drug particles which was expected to increase the
bioavailability and reduce or eliminate the food effect for drugs like
fenofibrate.
Moreover, the Court takes guidance
from the observation of Dr. Serajuddin that the composition must be “stable”
and, in his opinion, this means that the particle size would be less than 2 µm.
[87]
As
such, for the invention to work, the particles must remain small when
redispersing in a biorelevant media and the composition must be stable. For
these criteria to be met, the ‘054 Patent teaches that the particles must
redisperse to under 2 µm. That is therefore an essential element of the
claims.
Particle
Size of the Fenofibrate in the Composition
[88]
The
three independent claims, claims 1-3, each speak to the size of the fenofibrate
particles in the “composition for oral administration” which is to say in the
tablet.
[89]
Dr.
Muzzio asserts, and I accept, that a large part of the inventors’ solution to
the problem in the prior art of differences between administration of
fenofibrate in the fed versus the fasted state was the use of “very small
particles of fenofibrate.” The particle size distribution is set out in the
three independent claims of the ‘054 Patent.
[90]
Claim
1 asserts that the composition will be comprised of fenofibrate particles
having a D50 of less than about 500 nm; i.e. 50% of the particles of
fenofibrate will have a particle size of about 500 nm or less. Claim 2 asserts
that the composition will be comprised of fenofibrate particles having a D90 of
less than about 700 nm; i.e. 90% of the particles of fenofibrate will have a
particle size of about 700 nm or less. Claim 3 asserts that the composition
will be comprised of fenofibrate particles having mean particle size of less
than about 500 nm; i.e. the average particle size is about 500 nm.
[91]
Claim
28, which is one of the dependant claims asserted by Fournier in this
application further limits the particle size to a D50, D90, or mean particle
size of less than about 150 nm.
[92]
In
my view, the fenofibrate particle size is an essential element of the ‘054
Patent. This is not seriously in dispute. The issue of infringement rests on
a determination of whether or not Fournier has established on the balance of
probabilities that the particle size of the fenofibrate in the Sandoz Tablet
infringes the particle size distribution range set out in the ‘054 Patent.
Claim
20
[93]
Claim
20 covers a “composition of any one of claims 1 to 19 comprising a dosage of
about 145 mg of fenofibrate” with two additional limitations: “(a) the dosage
is therapeutically effective; and (b) the composition is bioequivalent to a 160
mg micronized fenofibrate tablet or 200 mg micronized fenofibrate
capsule, wherein bioequivalency, when administered in a human, is established
by a 90% Confidence Interval of between 0.80 and 1.25 for both Cmax
and AUC [emphasis added].”
[94]
The
issue between the parties on the proper construction of claim 20 is whether the
claim requires that the composition be bioequivalent to both the 160 mg and the
200 mg composition or only to one or the other. This dispute is relevant to a
utility submission made by Sandoz in that it is not disputed that the Fournier
product is not bioequivalent to the 160 mg composition but only to the 200 mg
composition.
[95]
Fournier
argues, and I agree, that this submission by Sandoz is one created by its
counsel and not by its experts whose interpretation of the ‘054 Patent does not
clearly state that the POSITA would understand that the composition must be
bioequivalent to both the 160 and 200 mg composition. Nonetheless, it is an
issue requiring the Court’s attention because it is an issue of patent
construction which is an issue of law for the Court and not for the experts.
[96]
Sandoz
relies on the decisions of Justice von Finkenstein in Abbott Laboratories v Canada (Minister of Health), 2005 FC 1095 and Justice Phelan in Abbott Laboratories v Canada (Minister of Health), 2005 FC 1332; however I find them to be of no assistance.
First, they both deal with a patent different from that at hand. Second, the
claim being construed in those cases used the connecting word “and” and not
“or” as in this case.
[97]
In
my view, the plain and unambiguous meaning of claim 20 is that the composition
must be bioequivalent either to the 160 mg composition or to the
200 mg composition, but need not be bioequivalent to both. This view is
consistent with that of Dr. Serajuddin who swore at paragraph 186 of his
affidavit that “Claim 20 relates to bioequivalency of the 145 mg dosage to the
160 mg or 200 mg micronized fenofibrate capsule (i.e. no food effect).”
Conclusion
on Construction of the ‘054 Patent
[98]
The
purposive approach to interpretation focused on “the particular words or
phrases in the claims that describe what the inventor considered to be the
‘essential’ elements of his invention” as directed by the Supreme Court in Whirlpool
leads me to conclude that the proper interpretation of this patent is the
following.
[99]
The
first sets of claims, claims 10, 27-28 and 38, as they depend on claims 1-3,
has the following essential elements:
a. The
composition exhibits bioequivalence a fed versus a fasted state as defined by
the USFDA when administered to a human (as specified in claims 1-3);
b. The
composition is phospholipid-free (as specified in claims 1-3);
c. The
particle size of the fenofibrate particles on dispersion or redispersion in a
biorelevant media (either gastric fluid or intestinal fluid) is less than 2 µm
(as specified in claims 1-3);
d. The D50, D90, or
mean particle size of the fenofibrate in the tablet is less than about 500 nm,
700 nm, or 500 nm (as specified in claims 1-3) and any one of those criteria is
less than 150 nm (as specified in claim 28);
e. The
difference in AUC of the composition when administered to a human in the fed
versus the fasted state is less than about 5% (as specified in claim 10); and
f. At
least 80% of the composition is dissolved within 10 minutes or less (as
specified in claim 38).
[100] The
second set of claims, being claim 20, has the following essential elements:
a. The
essential elements of any one of claims 1 to 19 in a dosage of about 145 mg;
b. The
dosage is therapeutically effective; and
c. The
composition is bioequivalent as defined by the USFDA to either (a) a 160 mg
micronized fenofibrate tablet, or (b) a 200 mg micronized fenofibrate tablet.
[101] With
this construction of the ‘054 Patent, I turn to the substantive allegations of
infringement and invalidity addressed by Fournier.
INFRINGEMENT
[102] If
the size of the fenofibrate particles in the Sandoz Tablet falls within the
particle distribution ranges specified in the ‘054 Patent, then the Sandoz
Tablet infringes the ‘054 Patent. All of the submissions of the parties as to
the particle size in the Sandoz Tablet rely on evidence of the particle size in
either an earlier state (the dispersion) or a later state (the redispersion).
There is no direct specific evidence of the particle size of the fenofibrate in
the Sandoz Tablet; no one actually measured the particles in those tablets.
Dr. Muzzio in cross-examination advises that Scanning Electron
Microscopy-Energy Dispersive Spectroscopy would have determined the particle
size in the Sandoz Tablet but it was not done:
The particle size in the Sandoz tablets is what it
would have been measured if anybody had bothered to use the right method, which
is SEM with EDS, where you can actually see the particles and then you can
answer your question the size is this. … The right method was to use EDS.
Nobody did it, so the exact answer to your question is not there
[emphasis added].
[103] Fournier’s
evidence was described by its counsel as a mixture of direct and circumstantial
evidence; however, Fournier relies primarily on the evidence of Dr. Muzzio. In
this respect it must be noted that when cross-examined and specifically asked
the size of the fenofibrate particles in the Sandoz Tablets are, Dr. Muzzio
responded: “I can give you an educated estimate and that educated
estimate is that its going to be similar to or slightly larger than the [
…] that they start with except for the caveat that there could be some
moderate amount of growth [emphasis added].” Later he says: “I expect
there is a very good chance that indeed the value would be under [ …] [emphasis
added].” The qualifications expressed in these responses are why I do not
accept Dr. Muzzio’s opinion on the particle size of the fenofibrate in the
Sandoz Tablet absent supporting and corroborative evidence.
[104] The
evidence before the Court on the particle size in the Sandoz Tablet requires a
general understanding of the manufacturing and dissolution process of the
compositions. Dr. Muzzio, in his affidavit sworn February 23, 2011, having
reviewed Sandoz’s manufacturing process, states that “Sandoz’s process closely
matches every step in the process described in the ‘054 Patent.” There is no
evidence to the contrary.
[105] Example
5 in the ‘054 Patent describes the process in making the invention in the
following steps:
1. A
fenofibrate nanoparticulate dispersion is prepared by combining fenofibrate and
other components including surface stabilizers and then milling the dispersion
until the fenofibrate is reduced to nanoparticulate size.
1.
A
granulated feed dispersion is prepared by combining the fenofibrate
nanoparticulate dispersion with some additional components, including surface
stabilizers.
2.
The
granulated feed dispersion is then sprayed onto very small particles of lactose
which is then dried to a powder.
3.
This
powder of dried fenofibrate covered lactose particles is formed into tablets.
4.
When
the tablet is taken by a patient the fenofibrate particles in the tablet
redisperse.
In summary, the process is
comprised of four stages: (1) the creation of a dispersion that contains
nanoparticles of fenofibrate, (2) spraying the dispersion onto lactose and then
drying it, (3) tableting the dried powdered fenofibrate, and (4) the
redispersion of the fenofibrate in the patient.
[106] The
parties are in agreement that it is the particle size of the fenofibrate in the
tablet that is relevant when assessing whether or not there is infringement.
Sandoz’s allegation of non-infringement is presumed to be true. Accordingly,
Fournier must disprove that allegation on a balance of probabilities. Fournier
submits that the evidence shows that the particle size distribution of the
fenofibrate in the Sandoz Tablets falls within the particle size distribution range
set out in the ‘054 Patent. Sandoz, on the other hand, submits that the size of
the fenofibrate in its tablets is above [ …] , well outside the ‘054 Patent’s
scope.
[107] Fournier’s
expert, Dr. Muzzio, opines, based on data from Sandoz’s ANDS, that the particle
size of the fenofibrate in the Sandoz Tablet infringes the ‘054 Patent. His
opinion is based on data obtained from Sandoz’s ANDS showing the particle size
of the fenofibrate in four different dispersion samples [ …] . [ …]
[ …]
[ …] . If
these particle size distributions are found in the Sandoz Tablet as well, then
the Sandoz Tablet infringes the ‘054 Patent.
[108] To
determine the particle size of the fenofibrate in the tablet form from that in
the dispersion form, Fournier relies on the evidence of Dr. Muzzio, together
with portions of the evidence of Dr. Kanfer and Dr. Fairhurst.
[109] Dr.
Muzzio attests that there is nothing in the manufacturing process used by
Sandoz that would appreciably increase the size of the fenofibrate from that
stage in the process to the tablet stage.
I have reviewed Sandoz’s manufacturing process and
it is my opinion that the particle size of the fenofibrate measured by Sandoz
after 11 hours in the suspension will be essentially the same as the
particle size of the fenofibrate in the Sandoz Tablets. [ …]
[ …]
[ …]
[ …]
[110] Fournier
submits that Dr. Muzzio’s opinion that the particle size in the dispersion
would not change when the composition is tabletted is supported, in part, by
the evidence of Dr. Kanfer.
[111] Dr.
Kanfer agreed in cross-examination that the ‘054 Patent teaches that a
stabilizer is added to the fenofibrate in the milling process to ensure that
the fenofibrate retains its small size throughout the manufacturing process up
to the spraying and drying of the dispersion onto the lactose. Dr. Kanfer’s
evidence below supports a conclusion that the size of the fenofibrate dried on
the lactose is the same as its size in the dispersion that results after
milling:
694 Q. Thank
you. In example 5 [of the ‘054 Patent], a dispersion of particles with a mean
particle size of 169 nanometres was prepared and sprayed onto some lactose; is
that correct?
A. Yes.
695 Q. And then
that material was tabletted?
A. From an intermediate, it was
granulated.
696 Q. Right.
A. And then tabletted.
697 Q. So what
goes into the tablet is the particles from the dispersion that are sprayed onto
the lactose where they dry; is that the idea?
A. Yes.
698 Q. And one of
the reasons why you want a stable dispersion is so that the particles remain in
their original small size through the spraying process; is that correct?
A. Yes, when the...yes.
699 Q. So you
have a stable dispersion where the particles don't agglomerate and you spray
that onto the lactose where it dries...
A. M'hmm.
700 Q. ...in
order to get those nano-sized particles onto the lactose?
A. To keep them at that size.
701 Q. To keep
them at that size on the lactose?
A. Right.
[emphasis added]
[112] At
best, Dr. Kanfer’s evidence only partially supports Dr. Muzzio’s opinion; there
is no change in the size of the fenofibrate after the milling process through
the spraying and drying process provided the dispersion is stable.
[113] Fournier
also relies on the evidence of Dr. Fairhurst, an expert put forward by Sandoz in
T-991-10 and which evidence, as outlined previously, was ordered to form part
of the record in this proceeding.
[114] At
paragraph 83(9) of his affidavit sworn June 11, 2011, Dr. Fairhurst attests to
the following:
[ …]
[ …]
[ …]
[ …]
[ …]
[ …] [emphasis
added].
[115] When
cross-examined on this statement he affirms that the particle size of the
fenofibrate in the Sandoz Tablet is less than [ …] :
Q. And so what you’re saying in [paragraph 83(9)]
is that when a patient takes the Sandoz tablet, and we’re here concerned about
dissolution and bioavailability, that 90 percent of the particles in the
tablet would be less than [ …]?
A. Yes.
Q. And that’s your opinion in this case?
A. From the data shown to me, yes.
[emphasis added]
[116] Dr.
Fairhurst’s evidence as to the D90 of the fenofibrate in the Sandoz Tablet
coincides with [ …]
[ …] . As such, Dr.
Fairhurst’s evidence appears to confirm that the particle size of the
fenofibrate in the Sandoz Tablet does not vary materially from the size in the
dispersion; this evidence supports Dr. Muzzio’s testimony. However, like Dr.
Muzzio, Dr. Fairhurst provides no independent scientific basis as support for
the proposition that the fenofibrate particles will not agglomerate in the
period beginning from the end of the milling process and ending after the
tableting.
[117] Evidence
contrary to Dr. Muzzio’s position is offered by Dr. Serajuddin and Dr. Leclair.
[118] Dr.
Serajuddin says “when the nanoparticles are in close contact with each other in
the dried state they will agglomerate because the stabilizer diffuses away from
the high energy zones between the two particles.” As support for that
statement at paragraph 333 of his affidavit he relies on an article by L.
Pelonen et al. entitled "Pharmaceutical nanocrystals by nanomilling:
critical process parameters, particle fracturing and stabilization methods."
He quotes the following passage under the heading “Amount of Stabilizer” in the
article:
It is important to remember that although the steric
stabilization of polymer chains may be sufficient in aqueous suspensions, in
the dry form it may no longer effectively maintain steric repulsion.
Also, in a suspension flocculated particles may lead to irreversible
aggregation as a function of time. When particles are close to each, it is
possible that the stabilizer is slowly diffused away from the high
energy zones between two separated particles, and the stabilizing effect is
lost [emphasis added].
[119] This
authority is not nearly as definitive as Dr. Serajuddin’s testimony. It does
not say that nanoparticles in a dried state will agglomerate. Rather,
it points out, when assessing the amount of stabilizer to be used, that there
are differences between the aqueous and dry state and what may be a sufficient
amount to maintain steric stabilization in the aqueous state may not be
sufficient in the dry state. The authors remind the formulator that when
particles are close to each other it is possible the stabilizer will slowly
diffuse away from the high energy zones resulting, in time, in the
agglomeration of the particles. Presumably this possibility must also be taken
into account when determining the amount of stabilizer to use. However, while
the scientific basis for Dr. Serajuddin’s opinion is not as definitive as he
suggests, it does support the proposition that the particle size may increase
in the period after creating the dispersion and creating a tablet.
[120] Dr.
Serajuddin also says that the tableting process itself will result in the
particles agglomerating:
In addition, during the tableting step, the
fenofibrate particles will be further compressed such that additional contact
between particles will occur resulting in agglomeration. Also, the tableting
process involves applying very high pressure to the ingredients to form a solid
tablet. This pressure results in fenofibrate particles being brought into
contact and forming bridges between them.
[121] He
cites as support for his opinion US ‘986 Patent. That patent teaches that a solid
spray granulated nanoparticulate composition comprising a poorly water soluble
active agent (such as fenofibrate), at least one surface stabilizer, and DOSS, redisperses
to a particle size of the active agent that is larger than its particle size in
the dispersion prior to incorporation into a solid dose form. Dr. Serajuddin
in cross-examination, Applicant’s Record vol. 4 page 845, explains that Sample
E in Tables 5 and 6 shows that the particle size from dispersion to
redispersion increased at least 10 fold:
Q. The range of multipliers supported by U.S. 986 is
10 to 20 times; is that your opinion?
A. If you look at formulation E in Table 5 and then
you look at formulation E in Table 6 unsonicated, you will see that the numbers
are 2186, 2163, 2544 and 2000 2203, so those are the mean values. So that’s
how I considered it has to be at least ten times higher when you do the
dispersing test.
I note that Table 6 shows that if
Sample E is sonicated for one minute the particle size drops substantially
(297, 357, 436, 291, and 292) but it still increases by close to double the
original size in the suspension.
[122] In
spite of vigorous cross-examination, Dr. Serajuddin was firm in stating that
there would be no agglomeration of the particles in the dispersion prior to the
spraying, although there could be agglomeration in the spray drying process.
Importantly, he is equally firm that there would be no agglomeration in the
redispersion provided there was stabilizer in the composition:
Q. Would you agree with me that the summary
agglomeration could happen in the spray drying process?
A. Yes.
Q. And some re-agglomeration could happen in the
dispersion?
A. Not in the dispersion. It could happen in the
spray dry process because once you take some spray dry material and you put it
in a solvent, there is no way of re-agglomeration. If they're agglomerated,
they're agglomerated during spray dry, not in the dispersion.
…
Q. If there is an adequate amount of surface
stabilizer and no steps are taken to prevent re-agglomeration, then there could
be some re-agglomeration of the particles in the re-dispersion media over time,
correct?
A. There is no chance of re-agglomeration in a
dispersion medium because this formulations, if you look at them, if you look
at Table 5, to begin with, they are very fine particles and they have not said
that these are not stabilized formulations. They double up formulations here.
Then they put them in a tablet and once they're dispersed from the tablet, they
gave very fine particles. So there is no chance there. When you're dispersing
something -- if I have a tablet or granules and it dispersed, it means that
it's coming out from that agglomerated state or solid state and it's
dispersing. Then there is no other force to put them back later in a dispersion
medium. So there cannot be any re-agglomeration later on.
[123] In
summary, it is his evidence, based on the US ‘986 Patent, that the particle
size of the fenofibrate in the redispersion will reflect the particle size in
the solid form and, based on a demonstrated increase of at least ten-fold, Dr.
Serajuddin is of the opinion that the D50 particle size of the fenofibrate in
the Sandoz Tablet would be expected to be at least [ …], well outside the
particle size distribution range of the ‘054 Patent.
[124] Sandoz
also proffered the evidence of Dr. Leclair who also asserts that the particle
size of the fenofibrate in the tablet is larger than that in the dispersion.
At paragraph 25 of his affidavit he states:
Manufacturing processes employed to produce a tablet
of a drug nanosuspension include spray drying or spray granulating and
tableting. These processes will cause aggregation of the primary particles.
The particle size of a dispersion of drug particles which undergoes spray
drying or spray granulation will increase.
[125] Based
on a redispersion test he conducted he opines that the D50 particle size would
be expected to be at least [ …], well outside the particle distribution ranges
in the ‘054 Patent.
[126] I
give no weight whatever to the redispersion test Dr. Leclair conducted. He is
not an expert in such testing. This was the first he has done. After placing
the Sandoz Tablets in the relevant medium, without the addition of any
surfactant or stabilizer, he proceeded to stir the mixture for about 45 minutes
before filtering it to determine the particle size. The evidence is clear that
absent a stabilizing agent the fenofibrate particles will agglomerate.
Accordingly, there is no reasonable scientific basis to find that the particle
size he determined through this test equals the particle size in the solid form.
[127] However,
like Dr. Serajuddin, and relying in part on the US ‘986 Patent, Dr. Leclair
opines that the particle size in the solid form will be greater than in the
dispersion and that tableting will increase particle size due to the pressures
brought to bear in that process.
[128] This
summarizes the evidence of the particle size of the fenofibrate in the Sandoz
Tablet.
[129] I
cannot conclude that Fournier has established, on the balance of probabilities,
that the particle size of the fenofibrate in the dispersion that is disclosed
in the Sandoz ANDS reflects the size of the fenofibrate particles in the Sandoz
Tablet. The evidence of Dr. Muzzio and Dr. Fairhurst is not supported by any
reference to scientific literature or study. I prefer the evidence of Dr.
Serajuddin and Dr. Leclair, which is supported by US ‘986 Patent, and which
shows that a solid spray granulated nanoparticulate composition or tablet
redisperses to a particle size of the active agent that is larger than its
particle size in the dispersion prior to incorporation into a solid dose form.
[130] We
know from Sandoz’s disclosure that the particle size in their dispersion has a
D50 of [ …] and a D90 of [ …].
We also know that those particles may agglomerate. The mean particle size of
similar dispersions has been found to agglomerate to a factor of at least 10 to
20 times the original particle size. As such, the scientific evidence is that
the particle size increases and it may be of a magnitude sufficient to remove
the Sandoz Tablet particle size from the particle size distribution ranges of
the ‘054 Patent. Accordingly, I find that Fournier has not shown on the
balance of probabilities that Sandoz’s allegation of non-infringement is not
justified.
[131] Before
leaving this discussion, I wish to briefly address a submission made by
Sandoz. Sandoz provided a number of the Sandoz Tablets to Fournier, presumably
so it could test the product. No test results were filed in this application
and Sandoz asks the Court to draw an inference that it has been done but not
disclosed because it would not support its position. Its position is that the
Court should apply the common law presumption that “where a party fails to lead
evidence of a fact that it is in a better position to establish, the Court will
infer that the facts are adverse to that party’s interest.”
[132] The
difficulty with Sandoz’s submission is that Fournier was not in any better
position than Sandoz to establish the particle size from tests done directly on
the tablets of the sort suggested by Dr. Muzzio. Admittedly, the burden of
proof lies with Fournier and not Sandoz; however, it was fully within Sandoz’s
ability to do the testing and establish conclusively whether its tablets did or
did not infringe the ‘054 Patent. Fournier submits that if any adverse
inference is to be drawn, it should be drawn against Sandoz. I disagree. No
adverse inference against either party is warranted in these circumstances.
[133] My
finding on non-infringement is sufficient to dispose of this application
because Fournier cannot be granted the prohibition order it seeks unless, on
the balance of probabilities, it disproves all of the allegations of Sandoz
that, if unchallenged, would allow the Minister to issue a NOC. However, for
completeness, I will briefly address the allegations of invalidity raised by
Sandoz and addressed by Fournier.
ANTICIPATION
[134] Anticipation
requires that there be a single prior art reference that discloses all of the
essential elements of the invention at issue: Eli Lilly & Co v Apotex
Inc, 2009 FC 991, at para 410. Anticipation may be found when an essential
element of the invention is not specifically disclosed in the prior art if,
when the prior art is performed, the patent under review will be infringed. In
that circumstance, the prior art anticipates the invention even if it was not
apparent to anyone at the time of the prior art. However, in that
circumstance, “the infringement must be more than merely a possible or even
likely consequence of performing the invention disclosed by the prior
disclosure. It must be necessarily entailed:” Synthon BV v
Smithkline Beecham plc, [2005] UKHL 59, at para 23.
[135] When
determining whether the prior art discloses the invention, experimentation is
not permissible. As was noted by the Supreme Court in Sanofi-Synthelabo
Canada Inc v Apotex Inc, 2008 SCC 61 [Sanofi-Synthelabo] at paras 25
and 27:
When considering the role of the person skilled in
the art in respect of disclosure, the skilled person is "taken to be
trying to understand what the author of the description [in the prior patent]
meant" (para. 32). At this stage, there is no room for trial and error or
experimentation by the skilled person. He is simply reading the prior patent
for the purposes of understanding it.
…
Once the subject matter of the invention
is disclosed by the prior patent, the person skilled in the art is assumed to
be willing to make trial and error experiments to get it to work. While trial
and error experimentation is permitted at the enablement stage, it is not at
the disclosure stage. For purposes of enablement, the question is no longer
what the skilled person would think the disclosure of the prior patent meant,
but whether he or she would be able to work the invention.
[136] I
have found that bioequivalence, as defined in the ‘054 Patent, is an essential
element of that invention. Not one of the numerous pieces of prior art cited
by Sandoz specifically mentions or describes bioequivalence, let alone
bioequivalence as defined in the ‘054 Patent. If one of the prior art
references does anticipate the ‘054 Patent then it must be on the basis that if
performed, it must necessary result in the infringement of the ‘054 Patent, i.e.
it must result in such specified bioequivalence. There is no evidence that any
of the prior art references do so and therefore I find that the ‘054 Patent was
not disclosed in the prior art.
[137] If I
had found that the ‘054 Patent was disclosed in the prior art, then I would
have had to ask whether that disclosure was enabling. Does the prior patent
give enough information to allow the subsequently planned invention to be
performed without undue burden? Routine trials are acceptable and will not be
considered undue burden, but experiments or trials and errors are not to be
prolonged even in fields where trials and experiments are generally carried
out. Prolonged or arduous trial and error is not considered routine.
[138] Generally,
I accept all of the submissions made by Fournier on enablement. In particular,
I accept its submission that the evidence of Dr. Kanfer proves that there is no
enablement in the prior art even if there was disclosure. Dr. Kanfer, at
paragraph 301 of his affidavit states, with reference to the ‘054 Patent, that
“an Ordinary Skilled Worker would not be able to formulate compositions which
are covered by the scope of claim 1, i.e. that are different than the
formulation in the Examples, without undue experimentation [emphasis
added].” I agree with Fournier’s counsel: “The skilled person trying to work
the [‘054 Patent], knows of the prior art and still Dr. Kanfer says they
couldn't work the patent and make the claimed inventions even with the prior
art and even with the [‘054 Patent] and it's flatly inconsistent with the
allegation that the prior art is enabling.”
[139] Accordingly,
I find that Fournier has established on the balance of probabilities that
Sandoz’s allegation of anticipation is not justified.
OBVIOUSNESS
[140] Obviousness
is to be determined in accordance with the four-part analysis outlined by the
Supreme Court in Sanofi-Synthelabo at para 67:
(1) Identify the notional
"person skilled in the art" and the relevant common general knowledge
of that person;
(2) Identify the inventive
concept of the claim in question or if that cannot readily be done, construe
it;
(3) Identify what, if any,
differences exist between the matter cited as forming part of the "state
of the art" and the inventive concept of the claim or the claim as construed;
(4) Viewed without any knowledge
of the alleged invention as claimed, do those differences constitute steps
which would have been obvious to the person skilled in the art or do they
require any degree of invention?
The Supreme Court set out additional
factors to consider within the fourth step, the “obvious to try” step, at
paragraphs 69-70:
a. Is it more or less
self-evident that what is being tried ought to work? Are there a finite number
of identified predictable solutions known to persons skilled in the art?
b. What is the extent, nature and
amount of effort required to achieve the invention? Are routine trials carried
out or is the experimentation prolonged and arduous, such that the trials would
not be considered routine?
c. Is there a motive provided in
the prior art to find the solution the patent addresses?
d. The actual course of conduct
which culminated in the making of the invention.
[141] The
POSITA has previously been identified as someone with a Ph.D. in pharmaceutics
with 1-3 years of experience in the area of formulation of pharmaceutical
products or as someone with a Masters degree in pharmaceutics with 5-7 years of
experience.
[142] The
promise is the use the inventor claims for his invention and it is the
statement of that use which led to the inventor being granted a monopoly for
his invention. Justice O’Reilly put it the following way in Hoffman-La
Roche v Apotex Inc, 2011 FC 875, at para 20:
[I]n construing a patent, the Court must
identify the claimed invention - the purportedly new and useful thing.
In analyzing the question whether the inventors have met the requirement of
utility, the Court will consider whether the inventors have disclosed a
"new article, a better article, a cheaper article, or affords the public a
useful choice" (Consolboard Inc v MacMillan Bloedel
(Sask) Ltd, [1981] 1 S.C.R. 504 at 521) [emphasis added].
[143] In my
view, the invention reflected in the ‘054 Patent promises to solve the problems
associated with fenofibrate and the previous fenofibrate compositions. This is
clear from the disclosure wherein it is written:
Because
fibrates, including fenofibrate, are so insoluble in water, significant
bioavailability can be problematic. In addition, conventional fibrate,
including fenofibrate, formulations exhibit dramatically different effects
depending upon the fed or fasted state of the patient. Finally, conventional
fibrate, including fenofibrate, formulations require relatively large doses to
achieve the desired therapeutic effects. There is a need in the art for nanoparticulate
fibrate formulations which overcome these and other problems associated with
prior conventional microcrystalline fibrate formulations. The present
invention satisfies these needs [emphasis added].
[144] Additionally,
the explicit claims and the ample bioequivalence testing in the disclosure make
it clear that the inventors were promising bioequivalence in the fed and fasted
state as defined in the ‘054 Patent.
[145] In
short, the promise of the invention is a fenofibrate composition that is
bioavailable, does not require large doses to achieve the therapeutic effect,
and provides effectiveness independent of whether the patient is in a fed or
fasted state as defined by the USFDA guideline.
[146] Fournier
submits that while it was known that increasing dissolution would increase
bioavailability and known that dissolution is increased by decreasing the
particle size, it was not part of the common knowledge of the POSITA that this
would result in the food effect being reduced or eliminated. That view is supported
by Sandoz’s expert Dr. Kanfer who, in cross-examination, testifies that one
would not have known that reducing particle size would eliminate the food
effect:
724. Q.
Having regard to the common general knowledge and the prior art that you
reviewed, would you agree with the following proposition: an ordinary skilled
worker would have expected, for formulations that incorporate nano particles of
fenofibrate that re-disperse, by which I mean do not significantly aggregate in
biorelevant media, would provide improved bioavailability and as such would
reduce or eliminate the food effect?
A. An ordinary skilled
worker would require additional information.
725. Q.
So you would disagree with that statement?
A. I disagree.
726. Q.
Is that a statement that a reasonable formulator could believe in? Or do you
think it is unreasonable?
A. A reasonable
formulator would accept that, as well. That he or she would look at the
formulation and then would have to do a lot of experimentation to make sure
that there is nothing predictable in terms of all the other possibilities. So
they would understand that, "This is what works, but it may not work for a
whole bunch of other things."
727. Q.
So you are saying a reasonable formulator would agree with you?
A. Yes.
728. Q.
And a reasonable formulator would disagree with the statement I asked you
about? No reasonable formulator could hold that view?
A. That it would
expect to show increased or bioequivalence?
729. Q.
Would provide improved bioavailability and as such would reduce or eliminate
the food effect?
A. I think you need to
separate that, as well. Because there are components here which relate to
nano-sizing and micronizing. And we know that that is one way of improving the
solution and probably improving bioavailability. But as to the end result of
showing bioequivalence, that is unknown.
730. Q.
What I am asking is whether you agree with the statement...I will read it
again: Having regard to the common general knowledge in the prior art, an
ordinary skilled worker would have expected for formulations that incorporated
nano particles of fenofibrate that re-dispersed...i.e., did not significantly
aggregate, in biorelevant media, that such formulations would provide improved
bioavailability and as such would reduce or eliminate the food effect?
A. I agree with half
of that statement.
731. Q.
But you don't agree with the second half?
A. Right.
732. Q.
And I am asking you whether a reasonable formulator could come to the view that
both halfs of that statement were true?
A. He would also agree
with half of it.
733. Q.
No reasonable formulator could agree with both halfs?
A. No.
[147] As a
result, I find that it was not obvious to a POSITA that reducing the particle
size of the fenofibrate would achieve bioequivalence as defined by the USFDA
guideline. Fournier has shown, on the balance of probabilities, that the
allegations of obviousness by Sandoz, are not justified.
CLAIMS BROADER /
INUTILITY / SOUND PREDICTION
[148] Sandoz
submits that the ‘054 Patent is overly broad because it is not limited to the
one formulation, set out as an example in the patent, which was proven by the
inventors to work. This allegation brings up the doctrine of sound prediction
as Justice Mactavish observed in Aventis Pharma Inc v Apotex Inc, 2005 FC 1283, at para 156:
The law is clear that a patentee is not to be
limited to specific compounds that he or she has actually made and tested prior
to filing for patent protection. A patentee is able to claim more broadly, so
as to cover a class of compounds, as long as the claim is based upon a sound
prediction.
[149] The
‘054 Patent tells the POSITA that there are combinations and concentrations of
surface stabilizers that do not work and shows one that does work. Moreover,
and importantly, the ‘054 Patent contains functional limitations and sets out
the tests to be used to ascertain which formulations meet and which do not meet
those limitations. I accept the submission of Fournier that the evidence shows
that those tests are within the routine capabilities of the POSITA and
accordingly the allegations of overbreadth and of lack of sound prediction and
inutility are not justified.
INSUFFICIENCY OF
DISCLOSURE
[150] Sandoz
alleges that the POSITA would have to conduct tests beyond mere routine experimentation
to obtain formulations covering the scope of the claims in the ‘054 Patent. In
support, it refers to Dr. Ruddy’s affidavit which explains the difficulties
that were overcome by the inventors of the ‘054 Patent. Specifically, Sandoz
says: “The ‘054 Patent does not describe that the concentration in the NCD were
critical, or that SLS, DOSS and sucrose were necessary.”
[151] Sandoz’s
allegation must fail. Nowhere did Dr. Ruddy say that those problems require
more than routine experimentation. Moreover, Dr. Ruddy was describing the
experimentation required before the existence of the invention in the ‘054
Patent. The efforts of the inventors to achieve that invention for the first
time are not relevant to the current issue. Unlike the inventors of the ‘054
Patent, the POSITA has the ‘054 Patent at its disposal when trying to perform
the invention. I accept Fournier’s proposition that the POSITA would become
aware of the cited difficulties and could use routine experimentation to
overcome them. I do not agree with Sandoz that a degree of inventiveness was
required to practice what is claimed.
CONCLUSION
[152] As Fournier has not established, on the balance of probabilities,
that all of Sandoz’s allegations are not justified, this application must be
dismissed. Sandoz is entitled to its reasonable costs. If the parties are
unable to agree on an amount, they may advise the Court and further directions
will issue.
JUDGMENT
THIS
COURT’S JUDGMENT is that:
1.
The
application is dismissed; and
2.
Sandoz
Canada Inc. is entitled to costs in accordance with these Reasons.
"Russel
W. Zinn"
APPENDIX A
Relevant Claims of the
‘054 Patent
1. A stable fenofibrate composition for oral
administration comprising:
(a) particles of fenofibrate having a D50 particle
size of less than about 500 nm, and
(b) at least one surface stabilizer,
wherein the composition exhibits bioequivalence upon
administration to a human subject in a fed state as compared to administration
to a human subject in a fasted state;
wherein:
(i) bioequivalency is established by:
(a) a 90% Confidence Interval for AUC which is
between 80% and 125%, and
(b) a 90% Confidence Interval for Cmax,
which is between 80% and 125%;
(ii) the composition disperses in a biorelevant
media; and
(iii) the composition is phospholipid-free.
2. A stable fenofibrate composition for oral
administration comprising:
(a) particles of fenofibrate having a D90 particle
size of less than about 700 nm, and
(b) at least one surface stabilizer,
wherein the composition exhibits bioequivalence upon
administration to a human subject in a fed state as compared to administration
to a human subject in a fasted state;
wherein:
(i) bioquivalency is established by:
(a) a 90% Confidence Interval for AUC which is
between 80% and 125%, and
(b) a 90% Confidence Interval for Cmax,
which is between 80% and 125%;
(ii) the composition redisperses in a biorelevant
media; and
(iii) the composition is phospholipid-free.
3. A stable fenofibrate composition for oral
administration comprising:
(a) particles of fenofibrate having a mean particle
size of less than about 500 nm, and
(b) at least one surface stabilizer,
wherein the composition exhibits bioequivalence upon
administration to a human subject in a fed state as compared to administration
to a human subject in a fasted state;
wherein:
(i) bioequivalency is established by:
(a) a 90% Confidence Interval for AUC which is
between 80% and 125%, and
(b) a 90% Confidence Interval for Cmax,
which is between 80% and 125%;
(ii) the composition redisperses in a biorelevant
media; and
(iii) the composition is phospholipid-free.
10. The composition of any one of claims 1-3,
wherein the difference in AUC of the fenofibrate composition, when administered
to a human subject in the fed versus the fasted state, is less than about 5%.
20. The composition of any one of claims 1 to 19
comprising a dosage of about 145 mg of fenofibrate, wherein:
(a) the dosage is therapeutically effective; and
(b) the composition is bioequivalent to a 160 mg
micronized fenofibrate tablet or 200 mg micronized fenofibrate capsule, wherein
bioequivalency, when administered to a human, is established by a 90%
Confidence Interval of between 0.80 and 1.25 for both Cmax and AUC.
27. The composition of any one of claims 1 to 22,
wherein the D50, D90, or mean particle size of the particles of fenofibrate is
less than about 200 nm.
28. The composition of any one of claims 1 to 22,
wherein the D50, D90, or mean particle size of the particles of fenofibrate is
less than about 150 nm.
36. The composition of any one of claims 1-35,
wherein
(a) within about 5 minutes at least about 40% of the
composition is dissolved;
(b) within about 10 minutes at least about 60% of
the composition is dissolved; or
(c) within about 20 minutes at least about 90% of
the composition is dissolved,
wherein dissolution is measured in a discriminating
aqueous media comprising sodium lauryl sulfate at 0.025 M, and wherein the
rotating blade method is used to measure dissolution.
38. The composition of claim 36, wherein within 10
minutes at least 80% of the composition is dissolved.
Email this Content