Purdue Pharma L.P. et al v. Alvogen Pine Brook, Inc.

Filing 178

ORDER Construing the Terms of U.S. Patent Nos. 6,733,783, 8,361,499, 8,551.520, 8,647,667, 9,023,401, 8,309,060, 8,529,948, 8,808,740, 9,056,052, 9,060,940, 9,084,816, 9,095,614, 9,096,615, 9,198,863, and 9,205,056. Signed by Judge Gregory M. Sleet on 5/10/2017. (mdb)

IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF DELAWARE

PURDUE PHARMA L.P .,
PURDUE PHARMACEUTICALS L.P.,
THE P.F. LABORATORIES, INC., and
GRUNENTHAL GmbH
Plaintiffs,
v.

ALVOGEN PINE BROOK, LLC, and
ACTAVIS LABORATORIES FL, INC.,
Defendants.

)
)
)
)
)
)
)
)
)
)
)
)
)
)
)

C.A. No. 15-687-GMS
(Consolidated)

~~~~~~~~~~~)

ORDER CONSTRUING THE TERMS OF U.S. PATENT NOS. 6,733,783, 8,361,499,
8,551.520, 8,647,667, 9,023,401, 8,309,060, 8,529,948, 8,808,740, 9,056,052, 9,060,940,
9,084,816, 9,095,614, 9,096,615, 9,198,863, AND 9,205,056
After considering the submissions of the parties and hearing oral argument on the matter,
IT IS HEREBY ORDERED, ADJUDGED, and DECREED that, as used in the asserted claims of
U.S. Patent Nos. 6,733,783 ("the '783 patent"), 8,361,499 ("the '499 patent"), 8,551,520 ("the
'520 patent"), 8,647,667 ("the '667 patent"), 9,023,401 ("the '401 patent), 8,309,060 (''the '060
patent"), 8,529,948 ("the '948 patent"), 8,808,740 ("the '740 patent"), 9,056,052 ("the '052
patent"), 9,060,940 ("the '940 patent"), 9,084,816 ("the '816 patent"), 9,095,614 ("the '614
patent"), 9,095,615 ("the '615 patent"), 9,198,863 ("the '863 patent"), and 9,205,056 ("the '056
patent"):

1.

The term "controlled release material," as used in the '783, '499, '520, '667, '401,
'052, and '940 patents, is construed to mean "a material other than the active

ingredient that causes the release of the drug (e.g., hydrocodone) at such a rate that
blood concentrations are maintained within the therapeutic range but below toxic
concentrations over a period of time of about 12 hours or longer." 1
2.

The term "matrix," as used in the '783, '499, '060, '520, '667, '052, '056, '740,
'816, '614, and '615 patents, is construed to mean "pharmaceutical preparation that
incorporates a dispersed/embedded active ingredient/agent". 2

1

After the Markman hearing, the parties reached agreement on the construction of "controlled
release material." (D.I. 120 at 1).
2
The parties' dispute centers on whether the term "matrix" should be considered a pharmaceutical
preparation or a discrete solid structure. (D.I. 101 at 6). The court finds that the term should be construed
as a "pharmaceutical preparation that incorporates a dispersed/embedded active ingredient/agent" because
that construction comports with the description of the term in the asserted patents.
First, there is no evidence that the "matrix" needs to be a discrete structure. To the contrary, during
the prosecution of the '499, '520, '667, '052, and '056 patents, applicants submitted a definition of matrix
which defined it as a "substance containing something: a substance in which something is embedded or
enclosed." Matrix Definition, Microsoft Encarta 2006. The '783 patent also describes the invention as "a
tablet comprising a matrix of drug and controlled release material, and optionally other pharmaceutically
desirable ingredients (e.g., diluents, binders, colorants, lubricants, etc.)." The patent and the prosecution
history make no mention that the matrix must be separate from other pharmaceutical ingredients.
Additionally, Defendants offer no explanation of what they mean by "discrete" except to say that "discrete
just means it has to be distinct. It has to occupy some sort of contiguous identifiable space." Markman
Hr'g Tr. 41:4-5. The court finds that Defendants' construction imports unwarranted and unnecessary
confusion into the claims. Accordingly, the court finds no reason to import the term "discrete" into the
construction of matrix.
To further support inclusion of the word "discrete," Defendants allege that a POSA must be able to
identify individual matrices because claim 17 of the '401 patent discloses "a plurality of pharmaceutically
acceptable matrices." '401 patent, col. 26 11. 30-31. According to Defendants, the matrix must occupy
some discrete, identifiable space because that is the only way to determine whether a dosage form has
multiple matrices; therefore, satisfying the limitation of claim 17. Markman Hr'g Tr. 41:5. Defendants
posed a question to the court during the Markman Hearing: "[h]ow is one to know if a dosage form with an
active ingredient and say half a dozen excipients has one matrix or two matrices or six matrices if there
[does not] ... have to be any special distinction between them." Id. 41 :24-42:2. The court is not persuaded
by Defendants' argument because the claims do not require a certain number of matrices. Defendants admit
that "examples of dosage forms ... include a core, and then another layer, or a shell," implying multiple
matrices. Tr. 42:6-7. Just because examples of dosage forms require the matrices to be discrete, that does
not mean that every dosage form has to have discrete matrices. Accordingly, the court finds Plaintiff's
position persuasive-the term "matrices" is just the plural of the term "matrix."
Second, outside of preferred embodiments discussed in the '783 patent, there is no evidence
supporting the inclusion of "solid" in the construction of "matrix." Defendants argue that the specification
solely provides examples of a solid dosage forms, which would inherently require solid matrices. (D.I. 102
at 4). That does not convince the court that it must adopt Defendants' construction, however. The Federal
Circuit has repeatedly cautioned against importing limitations from the specification into the claims. See
ComarkCommc'ns, Inc. v. Harris Corp., 156 F.3d 1182, 1186 (Fed. Cir. 1998); Phillips v. AWHCorp., 415

2

3.

The term "matrices," as used in the '401 patent, is construed to be the plural version
of "matrix"-"pharmaceutical preparations that incorporate a dispersed/embedded
active ingredient/agent."3

4.

The term "controlled release matrix material" in the '740 patent is construed to
mean "a material other than the active ingredient that is in a matrix and causes the
release of the drug (e.g., hydrocodone) at such a rate that blood (e.g., plasma)
concentrations are maintained within the therapeutic range but below toxic
concentrations over a period of time of about 12 hours or longer. " 4

F.3d 1303, 1323 (Fed. Cir. 2005); Computer Docking Station Corp. v. Dell, Inc., 519 F.3d 1366, 1374 (Fed.
Cir. 2008). The court also sees a distinction between a solid dosage form and a solid matrix. The patents
with the term matrix in their claim language do refer to solid dosages forms, but the court does not take that
to mean that a solid matrix is also required. Defendants admit that the '060 patent refers to matrix materials
that could exist in a semi-solid form, such as wax. Markman Hr'g Tr. 39: 14-40:2. They contend, however,
that the '060 patent requires a breaking strength of 500 N, so it does not make sense to have a dosage form
with a very high breaking strength if the dosage form has a liquid matrix. Id No scientific evidence was
presented on that point, though. The fact that the patent mentions a semi-solid used in forming the matrix
is enough to persuade the court it cannot adopt Defendants' construction. There is simply nothing in the
asserted patents that convinces the court that the "matrix" must be a solid structure.
3
The reasoning for this construction is identical to reasoning for the construction of "matrix"
because "matrices" was construed as the plural of "matrix."
4 Plaintiffs' and Defendants' proposed constructions are nearly identical, differing by only five
words-"is in a matrix and." Plaintiffs contend that Defendants' inclusion of the phrase "in a matrix"
removes the option of forming the matrix out of the claimed material itself. (D.I. 14 7 at I). As evidence
to support their argument, Plaintiffs point to the fact that polyethylene oxide can be both a matrix material
and a controlled release material. '740 patent, col. II II. 48-52, col. 14, 1.65-col.15 I. 29. The court,
though, does not understand how polyethylene oxide's status as a controlled release material and a matrix
material in the '740 patent means that "controlled release matrix material" cannot be construed as "in a
matrix."
In the '740 patent, regardless of whether polyethylene oxide is a matrix material or a controlled
release material, it is still "in" the matrix. The patent provides a "non-limiting list of suitable controlled
release materials," '740 patent, col. 14 ll.65-67, and then explains that "[c]ertain embodiments utilize
mixtures of two or more of the foregoing controlled release materials in the matrix of the core." Id. col. 15
11.11-12. When the patent describes polyethylene oxide as a matrix material it states that in certain
embodiments, the first and second matrix material "comprise[] polyethylene oxide." '740 patent, col. 6 I.
3. Further, the patent states that "polyethylene oxide is contained in both components,'' where "both
components" refers to the first and second matrix materials. Accordingly, requiring the controlled release
matrix material to be "in a matrix" does not remove the option of forming the matrix out of the claimed
material itself, as evidenced by the patents example of forming the matrix out of polyethylene oxide. The
court finds that Defendants' proposed construction of "controlled release matrix material" is consistent with
the usage of that phrase and related phrases throughout the '740 patent.

3

5.

The term "controlled release matrix material" in the '060 patent is construed to
mean "a material other than the active ingredient in which the active ingredient is
embedded that also

serv~s

to control the release of the active ingredient from the

dosage form." 5
6.

The term "steady state," as used in the '667 patent, is construed to mean "the amount
of the drug reaching the system is approximately the same as the amount of the drug
leaving the system. " 6

5

Again, Plaintiffs' and Defendants' proposed constructions for "controlled release matrix material"
in the '060 patent are almost identical. Defendants contend that the controlled release matrix material must
"cause the release of the active ingredient from the dosage form at controlled rates." (D.I. 102 at 2). The
court does not see an appreciable difference between "cause the release of the active ingredient from the
dosage form at controlled rates" and "serve to control the release of the active ingredient." The court adopts
Plaintiffs' proposed construction because the patent provides more support for it than "cause the release."
See '060 patent, col. 16 11. 19-20 ("Controlled release from the dosage form according to the invention is
preferably achieved by embedding the active ingredient in a matrix. The auxiliary substances acting as
matrix materials control active ingredient release.") (emphasis added).
The court also declines to adopt Defendants' proposed construction because of the phrase "at
controlled rates." The court is unsure of the meaning of "controlled rates." The court agrees with Plaintiffs
that the phrase implies a rate comparison to an immediate release dosage form when the patent and the
prosecution history do not necessitate such a comparison. (D.I. 147 at 2). In the prosecution history, the
applicants told the Examiner that "in dosage forms according to the subject invention the high molecular
weight polyalkylene oxide serves as a controlled-release matrix thereby retarding the release profile of the
drug." (D.I. 116, Ex. 31, Response to Dec. 28. 2006 Office Action at 12). Based on the inventor's statement
to the Examiner, it seems that no comparison to an immediate release dosage form is necessary to explain
the purpose of controlled-release matrix materials-to retard the release of the active pharmaceutical
ingredient from the dosage form. For those reasons, the court adopts Plaintiffs' proposed construction.
6
For the term "steady-state," the patentees acted as their own lexicographers. The '667 patent
specification explicitly defines "steady state" as when "the amount of the drug reaching the system is
approximately the same as the amount of the drug leaving the system." Col. 4 II. 47-49. Defendants'
proposed construction tracks the definition in the specification verbatim.
Plaintiffs main argument is that Defendants ignore the sentence that comes after the specification's
definition of steady-states-"Thus, at 'steady-state,' the patient's body eliminates the drug at approximately
the same rate that the drug becomes available to the patient's system through absorption into the blood."
'667, col. 411. 49-52. Plaintiffs contend that "the blood stream (i.e. blood serum) is the [relevant] "system"
for "steady-state." (D.I. 101 at 15). Plaintiffs thus maintain that the proper construction of "steady state"
is "[a] state in which the amount of drug reaching blood serum is approximately the same as the amount of
drug leaving blood serum." Id. The court does not find Plaintiffs' argument persuasive. The court sees no
reason to import the limitation "blood serum" into the patentees' explicit definition in the specification.

4

7.

The term "a C24/Cmax ratio," as used in the '783, '499, '520, '667, and '940 patents,
is construed to mean "the ratio of the plasma concentration of the drug at 24 hours
after administration to the highest plasma concentration of the drug attained within
the dosing interval. " 7

8.

The term "A C24/Cmax hydrocodone ratio," as used in the '863 and '056 patents, "the

ratio of the plasma concentration of the drug at 24 hours after administration to the
highest plasma concentration of the drug attained within the dosing interval. " 8
9.

The term "a relatively flat serum plasma profile," as used in the '499 patent, is
construed to mean "the plasma level of the drug provides a "C24/Cmax ratio of about
0.55 to about 1.0."9

7

Again, the patentees acted as their own lexicographer, defining "C24/Cmax ratio" as "the ratio of
the plasma concentration of the drug at 24 hours after administration to the highest plasma concentration
of the drug attained within the dosing interval." '783 patent, col. 4 11. 4-7. Defendants' proposed
construction restates the specification's definition of the term. Plaintiffs' proposed construction seeks to
limit the dosing interval to 24 hours. The court sees no reason to unnecessarily narrow the patentees'
definition to a 24-hour dosage interval. Plaintiffs' cite to one preferred embodiment where a 24-hour dosage
interval is mentioned. Id col. 3 II. 11. Because the patentees acted as their own lexicographers, and because
the patent only mentions a 24-hour dosing interval in a preferred embodiment, the court construes "C24/Cmax
ratio" in accordance with its definition in the specification of the '783 patent.
8
The court finds that there is no reason to construe this term differently than a "C24/Cmax ratio."
Plaintiffs argue that the term should be construed to mean "a ratio of blood plasma concentration of
hydrocodone 24 hours after administration of the dosage form to the highest plasma concentration of
hydrocodone attained within the dosing interval (here 24 hours)." (D.I. 101 at 15-16). For the reasons
previously stated, the court is not going to import the limitation of a 24-hour dosing interval into the claim
language when the patent does not necessitate such a limitation. Additionally, the court finds it redundant
and unnecessary to substitute "hydrocodone" for "drug." The patent's definition of a "C24/Cmax ratio" is
"the ratio of the plasma concentration of the drug at 24 hours after administration to the highest plasma
concentration of the drug attained within the dosing interval." When the '863 and '056 patents describe a
"C24/Cmax ratio hydrocodone ratio," it is clear that said "drug" in the "C24/Cmax ratio" definition is
hydrocodone. Accordingly, the court finds no reason to construe a "C24/Cmax hydrocodone ratio" differently
from a "C24/Cmax ratio."
9
The court adopts Plaintiffs proposed construction for "a relatively flat serum plasma profile."
Plaintiffs contend that "[l]imiting the construction to a single data point, as Defendants propose, is
incorrect." (D.I. 101 at 16). Plaintiffs argue that, in the context of the claim as a whole, the concept of a
"flat profile" requires more than just a single data point on a graph." Id To support their construction,
Plaintiffs cite to references that the Examiner considered during prosecution. Id at 17. One such reference
was a patent by the same inventors where they described "a flat plasma profile" as "a plasma concentration
ofhydrocodone after attaining a maximum concentration during the dosing interval stays within about 60%

5

10.

The term "obtaining a serum profile," as used in the '940 patent, is construed to
mean "establishing a blood plasma serum profile in a patient." 10

to about 100% of the maximum plasma concentration for the remainder of the dosing interval." Id; U.S.
Patent No. 7,943,174, col. 33 1.66--col.34 l. 4. The court does not understand how the inventors description
of "a flat plasma profile" in the '174 patent is at all inconsistent with the definition they propose here. The
inventors' description of a flat plasma profile in the '174 patent is essentially equivalent to saying that the
C24/Cmax ratio is about 0.60 to 1.0-a ratio very close to the one in the '499 patent. Additionally, the patent
discloses "[a] relatively flat serum plasma profile, meaning that the plasma level of the opioid provides a
C24/Cmax ratio of about 0.55 to about 1.0. " '499 patent, col.1 11. 45-53. Such a statement provides enough
evidence that the adopted construction is the intended meaning of the claim term.
10 The parties' disagreement centers on the word "obtaining." Defendants contend that "obtaining"
requires a separate active method step-"taking a sample or assaying serum samples." (D.I. 112 at 6). To
support that construction, Defendants cite to case law for the proposition that "[w]here a gerund phrase,
such as 'obtaining a serum profile,' begins a subparagraph in a method claim, as it does here, that term is
properly read as a distinct step requiring action in the method." (D.I. 112 at 6). Defendants also cite case
law for the proposition that patent law somehow prohibits single-step method claims. The court does not
think that the cases Defendant cites require that conclusion any time a gerund phrase is used in a claim.
The meaning of the gerund phrase is context dependent.
The court in Combined Sys., Inc. v. Def Tech. Corp. ofAm., 350 F.3d 1207 (Fed. Cir. 2003), found
that the term "forming" required "deliberate and systematic creation of folds. In the comi's analysis, it
explained that the district court was instructed by not only the gerund form of the term, but also by the
American Heritage Dictionary definition of the term and the term's use in the specification. Id at 1210.
Here, while the American Heritage Dictionary and the gerund form of the word might counsel against
Plaintiffs proposed construction, the specification and the claim itself make clear that "obtaining a serum
plasma profile" is actually a passive consequence of administering the drug.
First, claim 20 of the '940 patent discloses "[a] method of providing effective pain relief to a patient
in need thereof comprising: obtaining a serum profile," having a specific plasma level, "by administering
orally on a once-a-day basis a controlled release formulation of the hydrocodone." From that language, it
appears that the desired serum profile is "obtained" by administering the drug. The specification supports
the courts understanding.
According to the patentees, the novelty of the invention-improving the efficiency and quality of
pain management in patients experiencing moderate pain-follows from providing the controlled release
opioid formulation. See '940 patent, col. 111. 38-40. Once the drug is administered, because of the release
profile, there is "an early onset of therapeutic effect," and then "after rising to a maximum concentration
during the dosage interval,'' the controlled release opioid formulation "provide[s] a relatively flat serum
plasma profile, meaning that the plasma level of the opioid provides a C24/Cmax ratio of about 0.55 to about
1.0." Id ll. 50-58. The word "assaying," in Defendants' construction is never used in the patent and the
specification, neither implicitly nor explicitly, requires regular blood sampling. While the court
understands how "obtaining" could imply an active assaying step in other contexts, the word, as used in the
context of this patent, seems to rather clearly imply a passive occmTence that is the result of administering
the controlled release formulation. For that reason, the court adopts Plaintiffs proposed construction.

6

11.

The term "thermoformed dosage form," as used in the '060 patent, is construed to
include the use of subsequent heat. 11

12.

The term "breaking strength of at least 500 N," as used in the '060 patent, is
construed to mean "only separates into two or more pieces when exposed to a force
of at least 500 N." 12

11

There is agreement between the parties that "thermoforming" requires the application of heat and
pressure. The dispute lies in whether the '060 patent allows for the subsequent application of heat to a
dosage form. The court believes it does.
Defendants propose a construction of "thermoforming" that excludes dosage forms made using
pressure and subsequent exposure to heat. The asserted claim, claim 1 of the '060 patent, does not mention
a heating step. Instead, it just discloses a "thermoformed dosage form comprising one or more active
ingredients with abuse potential." Col. 21 II. 6-7. Claim 25 describes "[a] process for the production of a
dosage form according to claim 1, comprising" mixing a number of components and then "press-forming
the resultant mixture ... to yield the dosage form with preceding, simultaneous, or subsequent exposure to
heat." '060 patent, col. 23 II. 1-8. There are only two steps in claim 25: (1) mixing the components; and
(2) press-forming the resultant mixture. There is no implication that the mixture is thermoformed and then,
at some later point, exposed to heat. The claim explicitly links "yield[ing] the dosage form" with the
"preceding, simultaneous, or subsequent exposure to heat."
The specification is also replete with examples of subsequent applications of heat. The use of heat,
subsequent to the application of pressure, is mentioned four separate times in the specification. '060 patent,
col. 1111. 19, 27, 33, 44. Defendants attempt to undermine the disclosure of subsequent heating by pointing
to one line in the specification: "[i]n direct tableting with subsequent exposure to heat, the formed tablets
are briefly heated at least to the softening temperature (glass transition temperature, melting temperature,
sintering temperature) of component (C) and cooled again." Defendants contend that the reference to
"cooled again" implies that the dosage form was exposed to heat, then cooled, then heated again, and then
"cooled again." In that scenario, Defendants argue, the dosage form is "thermoformed" using preceding or
simultaneous heat, and then later, only after thermoforming is complete, is the dosage form exposed to heat
again. "[C]ool[ing] again" is only discussed in the context of "direct tableting." '060 patent, col. 1111. 2840. Direct tableting is explicitly labeled as one example by which the mixture can be formed into tablets.
Id. For that reason, the specification's disclosure of a formed tablet being "cooled again" does not alter the
court's conclusion that a thermoformed dosage form can be created with pressure and the subsequent
application of heat.
12
Plaintiffs contend that Defendants' proposed construction ignores the fact that "plastic
deformation (i.e., change in shape) is not breaking." (D.I. 111 at 19). Interestingly, there is no dispute over
that point-Defendants agree that plastic deformation is not breaking. Markman Hr'g Tr. 154:2-8. The
examples in the patent specification also confirm that tablets are not considered broken when they suffer
plastic deformation. See '060 patent, col. 19 11. 25-26 ("The tablets did not break when exposed to a force
of 500 N. The tablets did suffer a little plastic deformation."). The only dispute is over whether or not the
construction should explicitly include plastic deformation. Construing the phrase at issue to include the
clarification that plastic deformation is not breaking would be superfluous and confusing. Accordingly, the
court adopts Defendants' proposed construction.

7

13.

The term "viscosity-increasing agent," as used in the '060 patent, is construed to
mean "at least one substance that increases the thickness of the dosage form by
forming a gel when exposed to a liquid, said substance being different from the
synthetic or natural polymer (C) of claim 1 of the '060 patent." 13

14.

The term "necessary minimum quantity of an aqueous liquid," as used in the '060
patent, is construed to mean "an aqueous liquid in a necessary minimum quantity." 14

13

The only dispute between the parties with regard to this term is whether the viscosity-increasing
agent of claim 9 must be distinct from hardening polymer (C) of claim 1. (D.I. 102 at 11 ). Multiple other
com1s have resolved this same dispute. See Endo Pharm. Inc. v. Amneal Pharm., LLC, No. 12 CIV. 8060
(TPG), 2015 WL 9459823, at *24 (S.D.N.Y. Aug. 18, 2015) (finding that the viscosity increasing agent is
distinct from the hardening polymer of claim 1); In re OxyContin Antitrust Litig., No. 04 MD. 1603 SHS,
2014 WL 2198590, at *9 (S.D.N.Y. May 27, 2014) (same); Purdue Pharma L.P. v. Amneal Pharm., LLC,
No. l 15CV01152RGASRF, 2017 WL 634939, at *6 (D. Del. Feb. 16, 2017) (same). The court agrees with
the other courts that have considered the issue. The "viscosity-increasing agent" in claim 9 must be a
different substance than the "synthetic or natural polymer (C)" of claim 1.
Plaintiffs contend that the claims and the intrinsic record do not exclude polymer (C) from serving
as the viscosity-increasing agent. The court disagrees. Claim 9 discloses "[a] dosage form according to
claim 1, which additionally comprises at least one of the following components a)-f): ... (b) at least one
viscosity-increasing agent." '060 patent, col. 21 11. 37-42. The phrase "additionally comprises" leads the
court to believe that the viscosity-increasing agent must be a component distinct from the components
disclosed in claim 1. The specification confirms the court's reading of the plain language ofclaim 9.
The specification explains that dosage forms according to the invention cannot be crushed into a
powder, effectively nullifying the possibility of intravenous or nasal abuse. '060 patent, col. 6 11. 24-26.
Polymer (C)-a component listed in claim 1-is included to increase the hardness of the dosage form and
ensure that it cannot be easily pulverized or crushed. Id. col. 6 11. 20-23. The specification goes on to
explain that in order to prevent any abuse if a dosage form is somehow pulverized, the dosage form may
"contain further agents which complicate or prevent abuse as auxiliary substances (B)." Id. 11. 28-34. The
viscosity-increasing agent is listed as an auxiliary substance that can be included in the dosage form. Id.
11. 40-43. The fact that the auxiliary substance (B) is different from polymer (C) in claim 1, and the fact
that the specification classifies auxiliary substances as "further agents," confirm the court's findingthat the
viscosity-increasing agent is distinct from polymer (C).
14
Plaintiffs argue that the meaning of a "necessary minimum quantity of an aqueous solution" is
derived from a test in the specification used to verify the suitability of a certain viscosity-increasing agent.
That test provides:
In order to verify whether a viscosity-increasing agent is suitable as
component (b) for use in the dosage form according to the invention, the
active ingredient is mixed with the viscosity-increasing agent and
suspended in 10 ml of water at a temperature of25°C. If this results in the
formation of a gel which fulfills the above-stated conditions, the
corresponding viscosity-increasing agent is suitable for preventing or
averting abuse of the dosage forms of the invention.
'060 patent, col. 8 11. 5 5-62. The meaning of "a necessary minimum quantity of an aqueous liquid" cannot
be derived from the specification test because of key differences between it and the gelling test explained

8

15.

The term "combining said [opioid/hydrocodone] or pharmaceutically acceptable
salt thereof with each of said low molecular weight polyethylene oxide and said
high molecular weight polyethylene oxide to form at least one blend," as used in

the '816, '614, and '615 patents, is construed to mean "forming at least one blend
comprising said opioid or pharmaceutically acceptable salt thereof, said low
molecular weight polyethylene oxide, and said high molecular weight polyethylene
oxide." 15

in claim 9. Claim 9 requires that a gel form from the mixture of the "viscosity-increasing agent," a
"necessary minimum quantity of an aqueous liquid," and "the extract obtained from the dosage form." '060
col. 21 11. 41-56. As recognized in OxyContin, the claim 9 test "applies to the dosage form as a whole,"
whereas the specification test "is designed to assess whether a particular viscosity-increasing agent is
suitable for use in the invention." 2014 WL 2198590, at *13.
Another key difference between the two tests is that the specification test calls for the mixture of
the active-ingredient and the viscosity-increasing agent. The claim 9 test, by contrast, calls for a mixture
of the viscosity-increasing agent and "the extract obtained from the dosage form." '060 patent, col. 211.
43. It is clear from claim 1 that the dosage form comprises components other than the active ingredient.
Id. col. 21 11. 6-14. Those additional components could affect gel formation. Accordingly, the example 3
test cannot inform the court's construction of "an aqueous liquid in a necessary minimum quantity."
Because the court finds that the patent does not provide a meaning for the claimed phrase, the court
adopts Defendants' proposed plain meaning construction. The possible indefiniteness of the claim phrase
is a matter that the parties can explore at trial.
15
The parties disagree over the ingredients that must be mixed into a single blend and when that
mixing must take place. Plaintiffs contend that all three ingredients--opioid, low molecular weight
polyethylene oxide, and high molecular weight polyethylene oxide-need not be mixed into a single blend.
(D.I. 101 at 14). Defendants argue that the claim phrase's plain meaning, when read alongside the rest of
claim 1 of the '615, '614, 'and '816 patents, compels one conclusion: at least one of the blends in the tablet
contains the recited opioid, the low molecular weight polyethylene oxide, and the high molecular weight
polyetheylene oxide. (D.I. 112 at 13). Plaintiffs point to claim 21 of the '615 patent to demonstrate that it
would be rendered meaningless if the court adopts Defendants' proposed construction. Id. While the court
certainly understands Plaintiffs' argument, it cannot rewrite the language of claim 1. Even considering
claim 1's awkward phrasing, there appears to be no way around the requirement that, as a baseline, the
tablet must include an opioid, a high molecular weight polyethylene oxide, a low molecular weight
polyethylene oxide, and an additive. Under Plaintiffs' construction, there is a possibility that if the tablet
comprises only one blend, that blend can have either a low weight polyethylene oxide or a high weight
polyethylene oxide, plus the additive. Such a tablet would fall outside of the scope of claim 1. The court's
analysis of the parties' arguments is laid out below.
Claim 1 of the '615 patent discloses a "cured shaped tablet" comprising "an opioid ... , at least
one low molecular weight polyethylene oxide ... , at least one high molecular weight polyethylene oxide .
. . ; and at least one additive." '615 patent, col. 166 ll. 13-23. Claim 1 also discloses the process by which
the tablet is prepared, which requires:

9

16.

The term "a gelling agent comprising polyethylene oxide in an effective amount

to impart viscosity of at least about 10 cP when the dosage form is subjected to
tampering by dissolution in from about 0.5 to about 10 ml of an aqueous liquid,"
as used in the '948 patent, is construed to mean "dissolving the dosage form in from
about 0. 5 to about 10 ml of an aqueous liquid, causing the resulting mixture to have
a viscosity of at least about 10 cP." 16

[C]ombining said opioid or pharmaceutically acceptable salt thereof with
each of said low molecular weight polyethylene oxide and said high
molecular weight polyethylene oxide to form at least one blend, wherein
said at least one blend further comprises at least one additive.
Id. ll. 26-31 (emphasis added). As Defendants point out, the plain and ordinary meaning of the language
in claim 1 of the '615 patent requires all three ingredients to be present in the blend. (D.I. 102 at 13). The
court finds that combining the opioid with "each of' the different-weight polyethylene oxides-where the
claim uses the conjunctive "and"-means that the opioid must be combined with both the high and low
molecular weight polyethylene oxides. Plaintiffs do not seriously dispute that the plain language of claim
1 of the patents appears to require all three ingredients in a single blend. Instead, they instruct the court to
consider claim 21. (D.I. 101at14).
Claim 21 of the '615 patent then requires a "cured shaped tablet" according to claim 1, where "said
low molecular weight polyethylene oxide comprises a first blend and said high molecular weight
polyethylene oxide comprises a second blend, wherein at least one said blend further comprises said at least
one additive." Col. 168 ll. 16-20. Plaintiffs contend that, if the court adopts Defendants' proposed
construction, claim 21 is broader than claim 1, in violation of a cannon of claim construction. (D.I. 111 at
13). The court does not agree with Plaintiffs' contention. Claim 1 states the requirements when the cured
shaped tablet comprises one blend. Claim 1 explicitly allows for the possibility of more than one blend,
however. '615 patent, col. 166 I. 29. Claim 21 discloses a cured shaped tablet of claim 1 that comprises
more than one blend, making claim 21 narrower than claim 1. Id. col 168 II. 16-19. Claim 21 makes clear
that when the tablet has more than one blend, those additional blends can be different from each otherone blend can have a mixture of opioid and a low molecular weight polyethylene oxide, and the other blend
can have a mixture of opioid and high molecular weight polyethylene oxide. Claim 21 does nothing to
negate the requirement that at least one blend comprises the opioid, a high molecular weight polyethylene
oxide, and a low molecular weight polyethylene oxide.
Plaintiffs also contend that claim 23 of the '615 patent supports their construction because it
discloses "[a] cured tablet as defined in claim 21, wherein at least one blend is compressed by direct
compression." Id. col. 168 II. 26-27. Similar to claim 21, claim 23 only states a requirement for one of the
blends. The court does not read claim 23 as contradicting the language of claim 1. Claim 1 discloses
"applying each said blend to form a shaped tablet." Id. col. 1661. 32. Claim 23 is an example of applying
one of the blends, via direct compression, to form a shaped tablet.
16
The court adopts Defendants' plain meaning construction, postponing the resolution of the
indefiniteness issue to trial. Plaintiffs seek to limit the claim phrase to a specific embodiment in the
specification-the syringe method of example 3. The syringe method of example 3-where it is hard to
fill an insulin syringe with a solution of 10 cP-cannot serve to limit the scope of the types of viscosity
tests encompassed by the '948 patent because there is absolutely nothing in the patent that warrants such a

10

17.

The term "viscosity of at least about 60 cP," as used in the '948 patent, is construed
to mean "dissolving the dosage form in from about 0.5 to about I 0 ml of an aqueous
liquid, causing the resulting mixture to have a viscosity of at least about 60 cP." 17

18.

The terms "based on rheological measurements, ... molecular weight" and

"molecular weight ... based on rheological measurements," as used in the '816,
'614, and '615 patents, are construed to mean "a measurement of the mass of a
molecule based on rheological measurements." 18

limitation. See Liebel-Flarsheim Co. v. Medrad, Inc., 358 F.3d 898, 904 (Fed. Cir. 2004) ("[I]t is improper
to read a limitation from the specification into the claims.").
The prosecution history of the '948 patent does not persuade the court to adopt Plaintiffs'
construction. Plaintiffs argue that during the prosecution of the'948 patent, they pointed the Examiner to
example 3 in the specification to "support the significance of the viscosity of at least 10 cP." (D .I. 115, Ex.
20 at 7). The applicants' statements only demonstrated that example 3 was one way to support the
significance of the viscosity of at least 10 cP. It appears that the Examiner agreed during the interview that
the Kao reference is totally "silent as to the dosage forms ... achieving a viscosity of at least 10 cp when
tampered." Id The court agrees with Defendants that such a statement cannot amount to the necessary
clear disavowal that would be required to limit the viscosity test of the '948 patent to xample 3. See DigitalVending Servs. Int'!, LLC v. Univ. of Phoenix, Inc., 672 F.3d 1270, 1276 (Fed. Cir. 2012) (explaining the
importance of not limiting the claim scope based on statements made during prosecution unless there is a
clear disavowal or different definition). The court finds that the statements the applicants made during
prosecution do nothing to limit the construction of the claim phrase at issue.
17
For the same reasons as the previous limitation, the court adopts Defendants' plain meaning
construction and reserves the issue of indefiniteness for trial.
18
The court in this instance declines to adopt either party's construction. Regardless of whether
the wording of the claim is "based on rheological measurements, ... molecular weight" or "molecular
weight ... based on rheological measurements" the construction remains the same. The specification of
the McKenna patents (the '816, '614 and '615 patents) defines the claim phrase at issue. In the part of the
specification that lays out the definitions of terms used therein, the specification states:
For the purpose of this invention the approximate molecular weight is
based on rheological measurements. Polyethylene oxide is considered to
have an approximate molecular weight of 1,000,000 when a 2% (by wt)
aqueous solution of said polyethylene oxide using a Brookfield viscometer
Model RVF, spindle No. 1, at 10 rpm, at 25° C. shows a viscosity range of
400 to 800 mPa s (c).
'614 patent, col. 7 l. 61-col. 8 l. 1. The specification also goes on to define the meaning of other weights
of polyethylene oxide. The specification's definition of "rheological measurements" is not found in a
preferred embodiment or example, but instead, in the section where definitions of other terms are laid out.
Because the court finds that the patentees essentially acted as their own lexicographers with regard to the
meaning of "molecular weight ... based on rheological measurements," the court declines to construe the
term any more precisely than the patent warrants. In their briefing, Defendants took issue with the fact that
Plaintiffs offered no construction for the phrase "rheological measurement.'' (D .I. 102 at 19). The co mt

11

19.

The term "average molecular weight," as used in the '863 and the '740 patents, is
construed to mean "the average of the molecular weights." 19

20.

The term "weight-average molecular weight," as used in the '863 and '948 patents,
is construed to mean "the weighted average of the molecular weights." 20

21.

The phrase "wherein the polymer (C) has a molecular weight of at least 0.5
million according to rheological measure_ments," as used in the '060 patent, is

finds that the meaning of that phrase is defined by way of the specific test outlined in the specification.
Accordingly, the court sees no need to be any more specific in its construction of the disputed phrases.
Lastly, Plaintiffs' proposed that the phrases be construed to mean "a measurement of the mass of a
molecule [based on] correlation of molecular weight to viscosity according to rheological measurements."
The specification states that the "polyethylene oxide is considered to have an approximate molecular when
the "Brookfield viscometer Model RVF, spindle No. 1, at 10 rpm, at 25° C. shows a [specific] viscosity
range." The court declines to infer a correlation between molecular weight and viscosity from that
disclosure. '614 patent, col. 7 I. 61-col. 8 I. L
19
Plaintiffs contend that when the patent says "average molecular weight" it actually means
"weight-average molecular weight. (D.I. 102 at 16). Plaintiffs thus advance the same construction for both
terms: "Calculating molecular weight of a polymer sample when taking into account the molecular size of
each polymer chain rather than just the number of molecules." (D.I. 101 at 10). Plaintiffs outline a number
of reasons why Defendant's construction is incorrect, but they never cite evidence as to why their proposed
construction is correct. In their opening brief, Plaintiffs state that "number average is a simpler calculation,
but the weight-average molecular weight formula accounts for longer polymer chains having more mass."
Id. at 9. They provide no evidence to support that fact. Plaintiffs also claim that the patentees' intent to
convey "weight-average molecular weight" when using the term "average molecular weight" is evidenced
by their use of viscosity measurements to approximate molecular weight. (D.I. 101 at 9). According to
plaintiffs, viscosity measurements and weight-average molecular weight measurements account for the
different length of polymer chains. Id. While all of that may be true, the court received rio intrinsic or
extrinsic evidence-in briefing or at the Markman hearing-to support Plaintiffs' proposed construction.
The court cannot determine the meaning of weight-average and average molecular weight based on attorney
argument alone.
Further, the patents do not just use the terms "average molecular weight" and "weight-average
molecular weight." The '863 and the '948 patents mention both weight-average and number-average
molecular weights. '948 patent, col. 24 11. 55-57 (number-average), col. 23 11. 35-37 (weight-average);
'863 patent, col. 9 11. 12-15 (number-average), col. 16 11.34-37 (weight-average). The '740 mentions
neither weight-average nor number-average molecular weight, and, instead, explains approximate
molecular weight as a number determined by a viscosity range. '740 patent, col. 15 1.30-col.16 I. 3.
Plaintiffs admit that "there are different ways to characterize molecular weight for polymers, the tWo most
well-known being weight-average and number-average." (D.I. 101 at 9) (internal quotation marks omitted).
Defendants' Expert, Dr. Ryu, also stated in his declaration that "there are different kinds of 'average
molecular weights' that can be used to characterize a polydisperse polymer such as PEO." (D.I. 105 at 7).
Accordingly, the court adopts Defendants' proposed constructions, which follows from the plain and
ordinary meaning of both terms, and reserves ruling on indefiniteness until trial.
20
The explanation for the construction of "average molecular weight," supra note 19, also applies
to "weight-average molecular weight."

12

construed to mean "wherein the mass of the molecules of polymer (C) is at least 0.5
million according to rheological measurements."21
22.

The phrase "a shell encasing the core," as used in the '740 patent, is construed to
mean "the outer layer of the dosage form surrounding the core." 22

21

Plaintiffs' proposed construction changed only the word "molecular weight" in the claim phrase,
making it "weight-average molecular weight." Defendants claim that the phrase is indefinite, but to the
extent it can be construed, it should be construed as "[w]herein the mass of the molecules of polymer (C)
is at least 0.5 million according to measurements based on the change in form and the flow of matter." The
court declines to adopt either party's proposed construction. The court reserves ruling on the indefiniteness
issue until trial.
The '060 patent never mentions "weight-average molecular weight." Plaintiffs point to the
prosecution history of the parent application, now U.S. Patent No. 8,114,383, as evidence that when the
applicant's used the word "molecular weight" they meant "weight-average molecular weight." It is true
that the applicants described component (C) in Example 3 as having an "average molecular weight Mw of
100,000 g/mol." (D.I. 116, Ex. 31, Aug. 25, 2008 Bartholomaus Deel. at 5). There does not appear to be
a disagreement between Plaintiffs and Defendants that Mw is a symbol commonly used in scientific
literature to denote weight-average molecular weight. See (D.I. 101at12); (D.I. 105 at 7). Still, the court
is not persuaded that "average molecular weight" should be limited to weight-average molecular weight in
the '060 patent. The court does not find it proper to limit the scope of "average molecular weight" when
the only evidence in support of such a limitation comes from the prosecution history of a related patent.
There is nothing in the specification or claims of the '060 patent that counsels in favor oflimiting the scope
of "average molecular weight."
The court also declines to limit the term "rheological measurement" to "measurements based on
the change in form and the flow of matter." Again, the '060 patent does not support that construction. The
'060 patent indicates what it means by "rheological measurements." The specification of the '060 patent
discloses:
[P]olyethylene oxides, with a molecular weight of at least 0.5 million, preferably of up to
15 million, determined by rheological measurements, are very particularly preferred.
These polymers have a viscosity at 25' C. if 4500 to 17600 cP, measured on a 5 wt. %
aqueous solution using a model RVF Brookfield viscometer.
'060 patent, col. 5 1.66-col.6 1. 4. Accordingly, the court construes "rheological measurements" to have its
plain and ordinary meaning to a person skilled in the art, in the context of the entire patent. Eon Corp. IP
Holdings v. Silver Spring Networks, 815 F.3d 1314, 1320 (Fed. Cir. 2016) (explaining that the plain and
ordinary meaning "of a claim term is its meaning to the ordinary artisan after reading the entire patent").
22
The dispute as to this issue can be distilled to one question: can the "shell" be an internal layer
of the dosage form? Though Plaintiffs proposed a construction for "core," Defendants did not propose a
counter-construction. There does not appear to be a genuine dispute over the meaning of "core." Markman
Hr'g Tr. 106:6-9. As for the term "shell," Plaintiffs contend that it is "a layer or surface enclosing a space
or surrounding an object." By contrast, Defendants argue that when the inventor used the word "shell" he
meant to refer to only the very outer layer of the dosage form comprising the opioid and polyethylene oxide.
For the reasons that follow, the court adopts Defendants' construction.
The specification uses two different definitions to describe the layer that "encase[es] the core" and
comprises "a second portion of the opioid analgesic," '740 patent, col. 4711. 38-40, depending on the layer's
location. When that layer is the outermost layer, the specification refers to it as "the shell." See id. col. 28

13

Dated: May j_Q_, 2017

II. 17--41. When that layer is the interior layer, the specification refers to it as the "dry coat." See id col.
32 I. 43-col. 33 I. 48. It is clear from the patent that the inventors wanted to refer to the layers differently
despite the fact that they have close to the same composition. The necessary implication is that when the
dosage form includes a cosmetic coat, the layer previously described as "the shell" becomes the "dry coat"
because it is no longer the outermost layer.
Plaintiffs argue that construing the "shell" to be the "outer layer of the dosage form,'' somehow
makes it so that the cosmetic layer becomes the "shell" in examples 7-12 and 14-20. Markman Hr'g Tr.
115:14-19. The court disagrees. Defendants admitted during the Markman Hearing that examples 7-20
and 14-20 do not disclose a shell. Id 116:3-14. There is no dispute that the cosmetic coat disclosed in
examples 7-12 and 14-20 cannot be the "shell" discussed in the asserted claims. It simply does not meet
the claim requirements for a "shell." '740 patent, col. 4711. 38--40. Further, there seems to be no issue with
the fact the examples 7-12 and 14-20 do not disclose a shell because claims 91-95 do not require a shell.
Id. col. 52 I. 3 8-col. 54 I. 19. Therefore, examples 7-12 and 14-20 do not fall outside of the scope of all
of the claims. The court thus adopts Defendants' proposed construction for "a shell encasing the core."

14