The Medicines Company v. Mylan Inc. et al

Filing 590

MEMORANDUM Opinion and Order Signed by the Honorable Amy J. St. Eve on 10/27/2014:Mailed notice(kef, )

IN THE UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF ILLINOIS
EASTERN DIVISION
THE MEDICINES COMPANY,

)
)
)
)
)
)
)
)
)
)
)

Plaintiff,
v.
MYLAN INC., MYLAN
PHARMACEUTICALS INC., and
BIONICHE PHARMA USA, LLC,
Defendants.

No. 11-cv-1285

MEMORANDUM OPINION AND ORDER
AMY J. ST. EVE, District Court Judge:
In this patent infringement action, The Medicines Company (“TMC”) asserts that
Defendants’ proposed generic bivalirudin drug product infringes United States Patent No.
7,582,727 (the “’727 patent”). Specifically, TMC claims that Defendants Mylan, Inc., Mylan
Pharmaceuticals Inc. and Bioniche Pharma USA, LLC (collectively, “Mylan”) proposed generic
bivalirudin drug infringes claims 1-3, 7-10, and 17 of the ‘727 patent. (See R.1, Compl.; PTX
397, Parties’ Statement of Uncontested Facts, ¶ 38.) Mylan disputes that its proposed drug
infringes the ‘727 patent and asserts that the ‘727 patent is invalid on grounds of anticipation,
obviousness, and non-enablement and unenforceable for inequitable conduct. The Court held a
bench trial on TMC’s infringement claim and Mylan’s counterclaims regarding the ‘727 patent.1
Having considered the evidence and the parties’ arguments, the Court finds as follows: (1)
Mylan failed to prove by clear and convincing evidence that claims 1-3, 7-10, and 17 of the ‘727
1

The Court previously granted Mylan summary judgment of non-infringement regarding the
other patent-in-suit, United States Patent No. 7,598,343 (the “’343 patent”). (See R.309.) Accordingly,
only the parties’ claims and counterclaims related to the ‘727 patent were at issue at trial.

1

patent are invalid or unenforceable; and (2) Mylan’s proposed bivalirudin drug product infringes
claims 1-3, 7-10, and 17 of the ‘727 patent.
BACKGROUND
I.

The ‘727 Patent
The ‘727 patent was filed with the United States Patent and Trademark Office (“PTO”)

as U.S. Application No. 12/180,553 on July 27, 2008 and issued as the ’727 patent on September
1, 2009. (PTX 1, ‘727 patent; see PTX 397, ¶ 6.) The ‘727 patent is entitled “Pharmaceutical
Formulations of Bivalirudin and Processes of Making the Same.” (Id.) Dr. Gary Musso and Dr.
Gopal Krishna are the named inventors of the ‘727 patent, but TMC owns all rights, title, and
interest to the ‘727 patent.2 (PTX 397, ¶¶ 8, 9.)
The ‘727 patent generally pertains to pharmaceutical formulations of bivalirudin, a
reversible thrombin inhibitor used to temporarily prevent blood clotting during catherization
procedures. (See id., at col. 1:21-56; PTX 397, ¶ 6.) The section of the ‘727 patent entitled
“Background of the Invention” states: “[o]ne class of anticoagulants is direct thrombin inhibitors
that disrupt the activity of thrombin, an important protein in the coagulation cascade.” (Id., col.
1:49-51.) The ‘727 further states: “[i]n particular, bivalirudin (ANGIOMAX®), which directly
inhibits thrombin by specifically binding to both its catalytic site and to the anion-binding
exosite, is regarded as a highly effective anticoagulant for use during catherization [sic]
procedures.” (Id., col. 1:52-56.) The medical and therapeutic applications of bivalirudin, make
it “essential that the bivalirudin formulation maintains a high level of purity.” (Id., col. 2:1-3.)
The compounding process for the bivalirudin formulation may generate impurities, such as Asp9bivalirudin (from deamidation of asparagine at position 9 of bivalirudin to aspartic acid) and
2

As the Federal Circuit has noted, “[i]nventions are created by individuals, not corporations.”
MBO Labs., Inc. v. Becton, Dickinson, & Co., 474 F.3d 1323, 1326, n. 1 (Fed. Cir. 2007). For simplicity,
however, the Court refers to “TMC” as shorthand for applicants.

2

D-Phe

12

impurities (from isomerization of L-phenylalanine at position 12 of bivalirudin to the

D-isomer).

(Id. at col. 2:1-13.) The ‘727 patent also relates to development of processes for

synthesizing bivalirudin that minimize the generation of impurities and pharmaceutical batches
produced by such processes. (Id. at col. 2:19-22.) By way of background, this compounding
process involves three basic steps. (See DTX 1080, Compounding Instructions-Bivalirudin
Formulation, at MEDMYL4509463, MEDMYL450971.) First, the active pharmaceutical
ingredient (“API”), bivalirudin, is dissolved into a mannitol solution to form a bivalirudin
solution. Second, the resulting bivalirudin solution is mixed with a pH-adjusting solution, such
as sodium hydroxide, to raise the pH of the bivalirudin API to an acceptable level. Third, the
mannitol solution and the pH-adjusting solution are removed from the mixture to form the final
drug product.3
The asserted claims describe pharmaceutical batches of bivalirudin. Specifically, claim 1
of the ‘727 patent teaches “pharmaceutical batches” of bivalirudin having a maximum Asp9
impurity level of 0.6%. (PTX 1, col. 25:56-64.) Dependent claims 2 and 3 contain even stricter
limitations on Asp9 impurities, reducing the maximum allowable level of Asp9 impurities to
0.4% and 0.3%, respectively. (See id. at col. 25:65–26:56.) Dependent claims 7-10 and 17
contain all the limitations of claim 1 and additional limitations regarding the maximum level of
D-Phe

12

-bivalirudin impurities (claim 7), the type of pharmaceutically acceptable carrier

contained in the final drug product (claims 8-10), and the base used to adjust the pH of the final
drug product (claim 17). (See id. at col. 27-28.)

3

Throughout this Opinion, the Court discusses the compounding process used to manufacture
bivalirudin final drug products. Although, as the Court previously ruled, the ‘727 patent does not contain
“efficient mixing” process limitations, this compounding process is relevant to the development of the
invention in the ‘727 patent and various other aspects of the parties’ infringement and invalidity
arguments.

3

The asserted claims of the ‘727 patent read as follows:
1. Pharmaceutical batches of a drug product comprising bivalirudin (SEQ ID No:
1) and a pharmaceutically acceptable carrier for use as an anticoagulant in
a subject in need thereof, wherein the batches have a pH adjusted by a
base, said pH is about 5-6 when reconstituted in an aqueous solution for
injection, and wherein the batches have a maximum impurity level of
Asp9-bivliarudin of about 0.6% as measured by HPLC [high-performance
liquid chromatography].
2. The pharmaceutical batches of claim 1, wherein the maximum impurity level of
Asp9-bivalirudin does not exceed about 0.4% as measured by HPLC.
3. The pharmaceutical batches of claim 2, wherein the maximum impurity level of
Asp9-bivalirudin does not exceed about 0.3% as measured by HPLC.
7. The pharmaceutical batches of claim 1, wherein the batches have a maximum level of
12
D-Phe -bivalirudin that does not exceed about 2.5% as measured by HPLC.
8. The pharmaceutical batches of claim 1, wherein the pharmaceutically acceptable
carrier comprises one or more of a bulking agent or a stabilizing agent.
9. The pharmaceutical batches of claim 8, wherein the bulking agent is a sugar.
10. The pharmaceutical batches of claim 9, wherein the sugar is mannitol.
17. The pharmaceutical batches of claim 1, wherein the base is sodium hydroxide.
(PTX 1, col. 25:55–28:9.) During claim construction, the Court construed the term
“pharmaceutical batches” as follows:
“Pharmaceutical batches” may include a single batch, wherein the single batch is
representative of all commercial batches (see generally, Manual of Policies and
Procedures, Center for Drug Evaluation and Research, MAPP 5225.1, Guidance
on the Packaging of Test Batches at 1) made by a compounding process, and
wherein the levels of, for example, Asp9-bivalirudin, total impurities, and largest
unknown impurity, and the reconstitution time represent levels for all potential
batches made by said process. “Batches” may also include all batches prepared
by a same compounding process.
(See R.119, Claim Construction Order.)
The ‘727 patent specification describes several experiments Dr. Musso and Dr. Krishna
ran and provides the resulting Asp9 impurity levels for the batches produced in each experiment.
Examples 4 and 5 in the specification provide a comparison of an old “inefficient” mixing
process (Example 4) and a new “efficient” mixing process (Example 5), and Tables 6 and 7
4

report the impurity levels and reconstitution times for Original Angiomax® batches produced
using inefficient mixing (Table 6) and Improved Angiomax® batches produced using efficient
mixing (Table 7). (See PTX 397, ¶¶ 13-17.)
Comparison of the results in Table 6 (“inefficient mixing”) and Table 7 (“efficient
mixing”) reveals that the percentage of Asp9-bivalirudin impurities in the 24 reported batches of
Improved Angiomax® (shown in Table 7, recreated below) has a lower mean, standard
deviation, and a lower maximum level when compared to the 87 reported batches of Original
Angiomax® (shown in Table 6, recreated below):
TABLE 6
Characteristics of the batches generated by the compounding process that features
rapid addition of a pH-adjusting solution and inefficient mixing rates.
No. of batches

Mean ± SD

Maximum

Asp9-bivalirudin (%)

87

0.5 ± 0.4

3.6

Total Impurities (%)

63

1.4 ± 0.5

3.0

Largest unknown impurity (%)

86

0.3 ± 0.1

0.5

Reconstitution Time (seconds)

85

30 ± 12

72

(See PTX 1, col. 22:10-20.)

5

TABLE 7
Characteristics of the batches generated by the compounding process that features
addition of a pH-adjusting solution at a constant rate with efficient mixing.
No. of batches

Mean ± SD

Maximum

Asp9-bivalirudin (%)

24

0.3 ± 0.1

0.6

Total Impurities (%)

24

1.0 ± 0.4

2.0

Largest unknown impurity (%)

24

0.2 ± 0.1

0.3

Reconstitution Time (seconds)

24

18 ± 6

42

(See id. at col. 23:1-13.) In addition to the 24 reported batches of Improved Angiomax®, the
patent specification discloses that Table 7 did not include the results for one batch because “the
method used to generate the batch was not compliant with the protocol established for this
study.” (See id. at col. 23:14-16.)
II.

The Present Litigation
TMC is the owner of New Drug Application (“NDA”) No. 20-873, which was approved

by the U.S. Food and Drug Administration (“FDA”), on December 15, 2000, for the manufacture
and sale of a bivalirudin drug product for intravenous injection. (PTX 397, ¶ 19.) TMC’s
bivalirudin dug product is marketed in the United States under the tradename Angiomax®. (Id.,
¶ 20.) Bivalirudin is the active ingredient in TMC’s Angiomax® drug, which is indicated for use
as an anticoagulant during coronary angioplasty and stenting procedures. (Id., ¶¶ 20-24.)
Mylan submitted Abbreviated New Drug Application (“ANDA”) No. 202471 to the
FDA, seeking approval to engage in the commercial manufacture, use, sale, offer for sale, and/or
importation of a generic equivalent to Angiomax® prior to the expiration of the ‘727 patent.
(See PTX 133; PTX 397, ¶¶ 30-31, 34.) The product specification for Mylan’s proposed
bivalirudin product seeks approval for a bivalirudin drug with Asp9 impurities of up to 2.0%
6

(1.0% α-Asp9 and 1.0% β-Asp9). Mylan submitted an exhibit batch of its proposed bivalirudin
product to the FDA in support of its ANDA. Mylan’s exhibit batch, CMB10001, had total Asp9
impurities of 0.194%. In November 2011, the FDA sent a deficiency letter to Mylan regarding,
among other things, the Asp9 impurity levels provided in Mylan’s specification. Specifically, the
FDA wrote, “[y]our drug product impurity release specification is wide, and not supported by the
exhibit batch test results.” (See PTX 162.) To date, Mylan has not submitted a response to the
FDA’s deficiency letter. An internal draft response prepared in January 2012, however,
proposed changing the maximum Asp9 impurity level to 1.0% (0.5% α-Asp9 and 0.5% β-Asp9).
(See PTX 164.8.)
On February 23, 2011, TMC filed this patent infringement action under the
Hatch-Waxman Act, alleging that the manufacture, sale, and offer for sale of Mylan’s proposed
bivalirudin drug will infringe the ‘727 patent and United States Patent No. 7,598,343 (the “’343
patent”). (R.1.) Before trial, the Court granted summary judgment of non-infringement in favor
of Mylan with respect to the ‘343 patent. (See R.309.) The Court also granted Mylan summary
judgment on TMC’s willfull infringement claim. (Id.) The parties proceeded to trial on TMC’s
infringement claims and Mylan’s counterclaims regarding the ‘727 patent.
The bench trial on the parties’ claims regarding the ‘727 patent lasted approximately six
full days. During the trial, the parties admitted numerous exhibits, including the ‘727 patent file,
Mylan’s ANDA, and various e-mails and batch testing documents. Additionally, the following
witnesses testified at trial:


Anthony Flammia, Vice President of New Business Ventures for TMC



Gary Musso, Co-inventor of the ‘727 patent



Julie Simon, formerly the Senior Director of Business Development for Mylan

7



Daniel Hoch, an employee of Protocol Link, Inc., a consulting firm that assisted
Mylan in preparing its bivalirudin ANDA



Martina O’Sullivan (via videotaped deposition), Director of Regulatory Affairs for
Mylan and Bioniche



Leena Selvaraj (via videotaped deposition), formerly the Director of External
Research and Development and Project Management for Bioniche and Mylan



Malini Sen (via videotaped deposition), Senior Manager of Quality Assurance at
Biocon, the drug manufacturer Mylan contracted with to produce Mylan’s ANDA
exhibit batch.



Alexander Klibanov, TMC’s expert in the fields of chemistry and in pharmaceutical
formulations containing peptides, including their stability



Wayne Talton, Vice President of Global Regulatory Affairs for Mylan



Ian McKeague, Mylan’s expert in statistics



David Auslander, Mylan’s expert in the field of pharmaceutical formulation and
process development



Rajeshwar Motheram, formerly TMC’s Manager of Technical Operations



Gopal Krishna, Co-inventor of the ‘727 patent



Alan Salzberg, TMC’s expert in statistics



Nancy Linck, Mylan’s expert in patent prosecution and PTO procedures



Sandra Kuzmich, TMC’s patent prosecution attorney

The parties also submitted deposition designations and videotape clips for the following
witnesses: Angela Green; Daniel Robins; James Leary; Lisa-Sue Wood; Pamela Savoy; Russell
Garman; Steve McKinnon; and Timothy Smith. During the course of the trial, the Court
carefully evaluated the demeanor and credibility of each witness including, where applicable, the
witnesses’ body language, tone of voice, facial expressions, mannerisms, and other indicative
factors. Based on the evidence presented at trial, the Court makes the detailed findings of fact
and conclusions of law set forth below.

8

DISCUSSION
I.

Development of the Invention
A.

Original Angiomax®

TMC is the owner of NDA No. 20-873, which the FDA approved on December 15, 2000
for the manufacture and sale of a bivalirudin drug product for intravenous injection. (PTX 397, ¶
19.) TMC’s bivalirudin product is marketed under the trade name Angiomax® and among other
things, is indicated for use as an anticoagulant in patients with unstable angina undergoing
percutaneous transluminal coronary angioplasty. (Id. ¶ 20.) Anticoagulants are substances that
prevent blood from clotting and are commonly used during coronary procedures. (Id. ¶¶ 23-24.)
Bivalirudin is the API in TMC’s Angiomax® product that exhibits anticoagulant activity.
(Id. ¶¶ 21, 22.) Bivalirudin is a synthetic peptide made up of twenty amino acid residues that are
linked in a specific sequence to form a peptide. (Id. ¶¶ 22, 25.) The bivalirudin peptide
sequence in Angiomax® is: D-Phe-Pro-Arg-Pro-Gly-Gly-Gly-Gly-Asn-Gly-Asp-Phe-Glu-GluIle-Pro-Glu-Glu-Tyr-Leu. (PTX 397, ¶ 26 (emphasis added).) The ninth amino acid of the
chain, shown above in italics, is asparagine (abbreviated as Asn9 or Asn9). (Id. ¶ 27.) Under
certain conditions, an asparagine (Asn) may decompose (undergo deamidation) to form an
aspartic acid (Asp). (Id. ¶ 28) The impurity “Asp9-bivalirudin,” may form in Angiomax® (or a
generic equivalent) when the Asn9 residue deamidates to an Asp9 residue. (Id. ¶ 29.)
Before the inventions of the ‘727 and ‘343 patents, the compounding process used to
manufacture TMC’s Angiomax® drug resulted in variable and sometimes high levels of Asp9
impurities in the final drug product. At least two batches of Angiomax® manufactured using the
original compounding process (“Original Angiomax®”) failed to meet the specification requiring
a 1.5% maximum Asp9 impurity level applicable to Original Angiomax® batches. Lot 716184,

9

which TMC’s drug manufacturer Ben Venue Laboratories (“Ben Venue” or “BVL”) produced on
June 8, 2005, had an Asp9 impurity level of 3.6%. (See PTX 223.) Ben Venue investigated the
failure of Lot 716184 and determined that laboratory error did not cause the failure. In an
attempt to avoid future lot failures, Ben Venue and TMC reviewed and implemented changes to
its manufacturing instructions, e.g., instructions for adding the pH-adjusting solution to the
bivalirudin solution in aliquots. (Tr. 61:22-62:12 (Flammia); Tr. 207:9-208:3 (Musso); PTX
223.15; DTX 1211; DTX 1207; PTX 218 at MEDMYL4399135-36.) Less than a year later,
however, Ben Venue produced a second lot with Asp9 impurity levels that again exceeded the
1.5% specification. (See PTX 218; Tr. 1399:23-1402:22 (Klibanov); Tr. 207:9-2008:25 (Musso);
Tr. 62:17-64:22 (Flammia).) The second failed lot, Lot 896002, had an Asp9 impurity level of
2.3%. (See id.) Ben Venue could not definitively determine the root cause of either lot failure.
Following this second lot failure, TMC decided to become involved with Ben Venue’s
investigation into the underlying cause of the high Asp9 impurity levels for the failed lot. The
co-inventors of the ‘727 patent, Dr. Gopal Krishna and Dr. Gary Musso, led the investigation for
TMC. Dr. Krishna was a TMC employee at the time, and TMC hired Dr. Musso as a consultant
to assist in the investigation.
On July 14, 2006, Dr. Musso and two TMC employees, Anthony Flammia and Angie
Green, met with several Ben Venue employees involved in the production of Original
Angiomax® to discuss the lot failures. Dr. Krishna was on vacation and did not attend the July
14, 2006 meeting. Discussions at the meeting addressed the taffy or marshmallow-like
precipitate formed during the compounding process and suggestions on how to improve the
process were made. First, variations in the heights of the two stirrers in the tank were proposed
so that one stirrer mixes the upper part of the solution and the other mixes the lower part of the

10

solution. (See PTX 7.) The second proposal involved increasing the stirring speed prior to the
addition of the base. (See id.) Finally, a proposal regarding addition of the base through a dip
tube was made, so that the base dispersed at the bottom of the tank where mixing is still possible.
(See id.)
After the operators left the meeting, senior Ben Venue staff and TMC discounted the
suggestions. The first suggestion—to vary the stirrer heights—was already part of the Original
Angiomax® manufacturing instructions, and Ben Venue had used the third suggestion—adding
the base through a dip tube—in manufacturing the first failed lot in 2005. With respect to the
second suggestion—increasing the stirring speed before addition of the base—Dr. Musso
testified that Ben Venue staff expressed concern that this may lead to foaming.4
B.

Improved Angiomax®

Following the July 14, 2006 meeting, Dr. Musso and Dr. Krishna observed the
compounding process used to manufacture Original Angiomax® and developed a set of
experiments designed to identify improvements to reduce the formation of Asp9 impurities
during the compounding process. Dr. Musso and Dr. Krishna then supervised and directed the
Ben Venue employees who performed the experiments. Ultimately, through these experiments,
Dr. Musso and Dr. Krishna developed an “efficient mixing” process that resulted in lower and
more consistent Asp9 impurity levels in the bivalirudin final drug product.
Under the direction of Dr. Musso and Dr. Krishna, Ben Venue manufactured 26 batches
of Angiomax® using the improved compounding process (“Improved Angiomax®”). Twentyfour of those batches had a maximum Asp9 impurity level of about 0.6% or less, but the

4

Dr. Musso testified that Ben Venue staff informed him that, in a prior batch of bivalirudin, the
bivalirudin API foamed when stirred in a tank. The Court accepted Dr. Musso’s testimony for nonhearsay purposes only, not for the truth of the matter.

11

remaining two batches had an Asp9 impurity level of 12.4% (Lot 116050) and 1.5% (Lot
1344985). Ben Venue conducted investigations into both failed lots. With respect to Lot
116050, Ben Venue concluded after its investigation that operator error had caused the batch
failure. With respect to Lot 1344985, however, Ben Venue found no operator error in the
manufacturing process and closed its investigation without identifying the root cause of the
failure. TMC and the co-inventors challenged Ben Venue’s conclusion that the high Asp9
impurity level in Lot 1344985 was not due to operator error. Ultimately, Ben Venue agreed to
destroy the lot and provide a credit to TMC’s account for the batch failure. To date, however,
neither TMC nor Ben Venue has definitively identified the root cause of the failure.
II.

Invalidity
The Court first turns to Mylan’s counterclaims that the ‘727 patent is invalid due to

anticipation, obviousness, and non-enablement. Because patents are presumed valid, the party
seeking to invalidate a patent—in this case, Mylan—bears the burden of proving the patent’s
invalidity by clear and convincing evidence. See Microsoft Corp. v. i4i Ltd. P’ship, --- U.S. ---,
131 S. Ct. 2238, 2242, 180 L. Ed. 2d 131 (2011) (citing 35 U.S.C. § 282). The presumption of
validity creates “ ‘a heavy burden of persuasion,’ requiring proof of the defense by clear and
convincing evidence.... [T]he presumption encompassed not only an allocation of the burden of
proof but also an imposition of a heightened standard of proof.” Id., 131 S.Ct. at 2246; see also
AbbVie Deutschland GmbH & Co., KG v. Janssen Biotech, Inc., 759 F.3d 1285, 1297-98 (Fed.
Cir. 2014) (“[P]atents are presumed to be valid, and overcoming this presumption requires clear
and convincing evidence.”). New evidence supporting an invalidity defense that the PTO did not
consider before issuing the patent-in-suit may “carry more weight” in an infringement action
than evidence that the PTO previously considered. See Microsoft Corp., 131 S. Ct. at 2251. The

12

introduction of new evidence not previously considered by the PTO, however, does not lessen
the burden of the party seeking invalidity to prove that the patent is invalid by clear and
convincing evidence. Id. at 2250-51.
A.

Anticipation Under 35 U.S.C. § 102

Mylan alleges that claims 1-3, 7-10, and 17 of the ‘727 patent are invalid as anticipated
under 35 U.S.C. § 102(b).5 (See R.553, Mylan’s Post-Trial Brief, at 2-11.)
1.

Legal Standard

Section 102(b) provides in relevant part: “A person shall be entitled to a patent unless ...
the invention was ... in public use or on sale in this country, more than one year prior to the date
of the application for patent in the United States ....” 35 U.S.C. § 102(b); see also ResQNet.com,
Inc. v. Lansa, Inc., 594 F.3d 860, 866 (Fed. Cir. 2010) (“An offer for sale, sale, or public use, if
more than one year before the patent application was filed, will bar patenting of the product.”).
“The on-sale bar applies when two conditions are satisfied before the critical date [i.e. more than
one year before the patent was filed]: (1) the claimed invention must be the subject of a
commercial offer for sale; and (2) the invention must be ready for patenting.” Hamilton Beach
Brands, Inc. v. Sunbeam Prods., Inc., 726 F.3d 1370, 1374 (Fed. Cir. 2013) (citing Pfaff v. Wells
Elecs., Inc., 525 U.S. 55, 67, 119 S. Ct. 304, 142 L. Ed. 2d 261 (1998)).
In determining whether claims are invalid due to an on-sale bar, courts should determine
“whether the subject of the barring activity met each of the limitations of the claim, and thus was
an embodiment of the claimed invention.” Netscape Commc’ns Corp. v. Konrad, 295 F.3d 1315,
1323 (Fed. Cir. 2002) (citations omitted). In addition, “[o]nly an offer which rises to the level of

5

The America Invents Act (“AIA”), Pub. L. No. 112–29, took effect on September 16, 2012.
Because the application for the ‘727 patent was filed before that date, the Court refers to the pre-AIA
version of § 102. See Allergan, Inc. v. Apotex Inc., 754 F.3d 952, 958, n. 1 (Fed. Cir. 2014).

13

a commercial offer for sale, one which the other party could make into a binding contract by
simple acceptance (assuming consideration), constitutes an offer for sale under [section] 102(b).”
Id (citations omitted). “To determine if the offer is sufficiently definite, one must examine the
language of the proposal in accordance with the principles of general contract law.” Id. at 132324.
2.

Analysis

The critical date for the ‘727 patent is July 27, 2007. (See PTX 1.) Mylan relies on five
Original Angiomax® batches sold in the United States between 2002 and 2004, prior to the
critical date, for its § 102(b) argument. Mylan, for example, submitted evidence alleging TMC
sold Lot No. 273786 in the United States to Amerisource Bergen Drug Corp. in Louisville,
Kentucky (sold on 12/17/02), and to Morris & Dickson Co., LLC in Shreveport, Louisiana (sold
on 12/17/02).6 (See DTX 1658 at MEDMYL4580643-44; see also JTX 3028 (McKinnon Dep.
126:13-127:10).) Mylan argues that the sales of these batches satisfy the requirements for an
on-sale bar, namely, that TMC made the sales before the critical date, and: (1) the sales
constitute the claimed invention as a subject of the commercial offer for sale; and (2) the claimed
invention was ready for patenting as of that date. TMC argues that the batches to which Mylan
refers represent only a subset of the multiple batches of Original Angiomax® and that the Court
would have to consider all the batches in its analysis of whether the subject of the commercial
offer for sale was in fact the claimed invention.

6

Mylan relies on four additional sales that occurred between 2002 and 2004, for Lot Nos.
273787, 335834, 339257, and 515495 in the United States of Original Angiomax® lots in the sales data
report from ICS. (See DTX 1658 at MEDMYL4580643-46, 4580658; see R.553 at 5-6.)

14

a.

TMC Sold Original Angiomax® Batches Prior to the
Critical Date

As an initial matter, TMC does not dispute that the sale of Original Angiomax® batches
constitutes a commercial offer for sale and the evidence at trial established the same. During
prosecution of Application No. 12/180,553, which issued as the ‘727 patent, TMC admitted the
Original Angiomax® batches referenced in Table 6 were sold, stating “[t]he drug products
generated from the old compounding process, which were sold/marketed/offered for sale for
more than one (1) year prior to the filing date of the accompanying application, comprised a
maximum Asp9-bivalirudin level of about 3.6%, a maximum reconstitution time of about 72
seconds, and a maximum amount of total impurities of about 3.0%.” (PTX 3.145 (emphasis
added); PTX 3.708-709; see also R.560, TMC’s Responsive Post-Trial Br., at 17 (“[T]he file
history demonstrates that the PTO was made aware that these Original Angiomax® batches were
sold . . . .”); see also R.418, Pretrial Order, Exhibit M, Mylan’s Proposed Findings of Fact and
Conclusions of Law on Invalidity and Unenforceability and TMC’s Responses, TMC’s Response
to FF.9, stating “[i]n 2001, TMC only sold pharmaceutical batches of Original Angiomax®.”)
The evidence at trial also demonstrated that TMC sold Original Angiomax®, through its
distributor—Integrated Commercialization Solutions (“ICS”)—to various customers in the
United States prior to July 27, 2007. (See DTX 1658; DTX 1775; Tr. 150:24-154:9 (Flammia);
JTX3028 (McKinnon Dep. 126:13:5-14:17.). The Court, therefore, finds that the Original
Angiomax® batches constitute a commercial offer for sale.
b.

Original Angiomax® Batches Sold Are Not the Claimed
Invention

The Court now turns to whether the commercial offers for sale were of the claimed
invention. At their core, Mylan’s and TMC’s arguments regarding the batches of Original

15

Angiomax® focus on whether a single batch can be used to meet the “pharmaceutical batches”
requirement of the ‘727 patent claims.
During claim construction, the Court construed the term “pharmaceutical batches” as
follows:
“Pharmaceutical batches” may include a single batch, wherein the single batch is
representative of all commercial batches (see generally, Manual of Policies and
Procedures, Center for Drug Evaluation and Research, MAPP 5225.1, Guidance
on the Packaging of Test Batches at 1) made by a compounding process, and
wherein the levels of, for example, Asp9-bivalirudin, total impurities, and largest
unknown impurity, and the reconstitution time represent levels for all potential
batches made by said process. “Batches” may also include all batches prepared
by a same compounding process.
(See R.119.) Although the parties agreed to this construction during the claim construction
hearing, they now disagree on its meaning. TMC argues that the Court’s construction consists of
two parts: (1) the first part, regarding a “single batch,” refers only to an ANDA test batch, as
evidenced by the parenthetical citation to MAPP 5225.1 regarding the packaging of test batches,
and (2) the second part, regarding “all batches,” refers to prior art like Original Angiomax® in
which TMC produced multiple batches. Mylan, on the other hand, argues that the first part of
the Court’s construction regarding a “single batch” may apply to single batches other than
ANDA test batches and that the Court should consider whether individual batches of Original
Angiomax® on sale before July 27, 2007 anticipate the asserted claims.
The Court need not resolve this dispute, however, because even if the Court accepts
Mylan’s interpretation, Mylan failed to establish that the individual Original Angiomax® batches
on which it relies meet “each of the limitations of the claim, and thus [are] an embodiment of the
claimed invention.” See Netscape Commc’ns, 295 F.3d at 1323 (citations omitted). Namely,
Mylan failed to prove that any of the cited Original Angiomax® batches are “representative” as
required by the claim. Namely, that the cited Original Angiomax® batches are: (1)
16

“representative of all commercial batches . . . made by a compounding process,” and (2)
“wherein the levels of . . . Asp9-bivalirudin … represent levels for all potential batches made by
said process.” (PTX 1, claim 1; see also R.119.)
i.

TMC Batches 339257 and 515495 Are Not
“Representative” of All Commercial Batches

Mylan contends that two of its five cited batches—batches 339257 and 515495—satisfy
the “single batch” prong of the Court’s “pharmaceutical batches” construction because the
batches are “representative.” Specifically, Mylan argues that TMC’s selection of batches 339257
and 515495 as “stability batches” for use in FDA-required testing on the stability, or shelf-life, of
its Original Angiomax® drug product demonstrates their “representative” nature. While TMC
does not dispute that the batches are from “a compounding process”, it argues that the batches
are not “representative” as required by the ‘727 patent claims. TMC contends that any
“representative” nature of the single stability batches identified by Mylan pertains only to their
stability, not to the general nature of the batch (which would include being representative of
properties other than stability, such as level of impurities.) Both parties, however, seem to
ignore the fact that the use of “representative” in the claims of the ‘727 patent has a specialized
meaning.
In order for a single batch to be a “pharmaceutical batch” of the ‘727 patent claims, the
single batch must be “representative.” The “representative” concept shows up in two portions of
the claim language. A single batch is “representative” if it is: “[1] representative of all
commercial batches . . . made by a compounding process, and [2] wherein the levels of, for
example, Asp9-bivalirudin, total impurities, and largest unknown impurity, and the reconstitution
time represent levels for all potential batches made by said process.” (See PTX 1, claim 1
(emphasis added); see also R.119.)
17

Mylan claims that TMC’s chemistry expert’s (Dr. Klibanov’s) reliance on FDA
compliance documents that expressly state that lots placed on stability must be “representative of
the marketed product,” further demonstrates the representative nature of TMC’s stability batches.
(See PTX 108.24 (emphasis added); Tr. 1458:5-15 (Klibanov).) The FDA compliance
documents use of “representative,” however, does not contemplate the use of that term in the
‘727 patent. As such, the explicit recitation of “representative” in the FDA compliance
document does not bear on whether TMC’s stability batches meet the “pharmaceutical batches”
limitation as used in the ‘727 patent claims.7 See Rembrandt Vision Techs., L.P. v. Johnson &
Johnson Vision Care, Inc., 725 F.3d 1377, 1379, 1382-83 (Fed. Cir. 2013) (citing Fed. R. Evid.
403) (affirming the district court’s exclusion of plaintiff’s circumstantial evidence as potentially
confusing and non-probative of infringement where the evidence described the accused
infringing product as “soft” and “soft gas permeable contact lens” was a claim term the court
construed as requiring the lens to have a Short D Hardness less than five).
Mylan also argues that the ICH Harmonized Stability Guidelines support its position that
TMC’s stability batches are “representative.”8 This argument, however, views the

7

As Mylan’s argument stems from a dispute of the claim term “pharmaceutical batches,” the
Court’s consideration of the ICH Harmonization Stability Guidelines and the FDA compliance documents
as evidence of the meaning of the ‘representative” nature of the “pharmaceutical batches” equates to
reliance on extrinsic evidence, which although it can shed useful light on the relevant art, is not as reliable
as the intrinsic evidence of the language of the claim itself. See Phillips v. AWH Corp., 415 F.3d 1303,
1317 (Fed. Cir. 2005) (en banc) (citations omitted); see also 01Communique Lab., Inc. v. LogMeIn, Inc.,
687 F.3d 1292, 1295-96 (Fed. Cir. 2012) (“To ascertain the scope and meaning of the asserted claims, we
look to the words of the claims themselves, the specification, the prosecution history, and, if necessary,
any relevant extrinsic evidence.”) (quoting Chicago Bd. Options Exch., Inc. v. Int’l Sec. Exch., LLC, 677
F.3d 1361, 1366 (Fed. Cir. 2012) (emphasis added)); Kara Tech. Inc. v. Stamps.com Inc., 582 F.3d 1341,
1348 (Fed. Cir. 2009) (“extrinsic sources like expert testimony cannot overcome more persuasive intrinsic
evidence”). Here, looking at the intrinsic evidence of the claim language, it is clear that “representative”
as defined in the ‘727 patent claims requires limitations that the Original Angiomax® batches do not
meet.
8

Mylan also argues that the “stability” of a final drug product references the rate at which
impurities form while the product is packaged on the shelf, citing the ICH Harmonization Stability

18

“representative” limitation of the claim with a myopic focus on the “representative of all
commercial batches” portion of the limitation. TMC’s Annual Product Review for Original
Angiomax® identified the stability batches, stating:
The bivalirudin drug product stability program is performed by BVL, on behalf of
MDCO. Per agreement with MDCO, BVL conducts an active bivalirudin drug
product stability program with specified lot samples stored at temperatures and
conditions in keeping, testing, and reporting within the ICH Harmonization
Stability Guidelines and in keeping with FDA and EMEA commitments.
(DTX 1698 at MEDMYL4115027 (emphasis added); see DTX 1744 (FDA Guidance adopting
the ICH Guidelines).) The ICH Harmonized Stability Guidelines § 1.3 states:
The purpose of stability testing is to provide evidence on how the quality of a
drug substance or drug product varies with time under the influence of a variety of
environmental factors such as temperature, humidity, and light, and to establish a
retest period for the drug substance or a shelf life for the drug product and
recommended storage conditions.
(DTX 1087 at MEDMYL4054113.) The ICH Harmonized Stability Guidelines § 2.2.3 regarding
the selection of stability batches state that “[t]he manufacturing process used for primary batches
should simulate that to be applied to production batches and should provide product of the same
quality and meeting the same specification as that intended for marketing.” (See DTX 1087 at
MEDMYL4054114 (emphasis added).)
In an effort to address the “representative” claim limitation, Mylan relies on the ICH
Harmonized Stability Guidelines’ reference to stability batches as being the same as that
intended for marketing. TMC’s Vice President of Pharmaceutical Development, Dr. Rajeshwar
Motheram, however, testified at trial that TMC generally followed ICH guidelines in choosing
stability lots and selected stability lots that were “representative” of the stability of their
commercial batches (as opposed to representative of the batches generally). (See Tr. 1101:6-10,
Guidelines (DTX 1087) at MEDMYL4054113. The cited reference, however, does not support this
assertion as it does not address the rate at which impurities form.

19

1104:2-10.) While the ICH Harmonized Stability Guidelines may address use of a batch for
stability testing that mimics “that intended for marketing,” the ‘727 patent claims require more
from a pharmaceutical batch than it simply being “representative of all commercial batches.”
Namely, in order for a single batch to be “representative” in the ‘727 patent claims, it must also
have “levels of, for example, Asp9-bivalirudin, . . . [that] represent levels for all potential batches
made by said process.”
ii.

TMC Batches 339257 and 515495 Do Not “Represent”
the Asp9-Bivalirudin Levels For All Potential Batches

Mylan also has not proven by clear and convincing evidence that the two asserted
stability batches are “representative” in the sense that “the levels of, for example,
Asp9-bivalirudin, . . . represent levels for all potential batches made by said process.” (PTX 1,
claim 1.) Mylan asserts that Original Angiomax® batches 273786, 273787, 335834, 339257,
and 515495, which TMC sold through its distributor ICS in 2002-2004 (see DTX 1658), render
the asserted claims invalid under § 102(b). Two of the batches have Asp9 impurities of 0.5% or
less with three of the batches (Lot Nos. 273786, 273787, and 335834) having Asp9 impurities of
0.3% or less.9 Several Original Angiomax® batches made by “said process” (i.e., the
compounding process of the cited batches), however, have Asp9 impurity levels above 0.6%. Lot
Number 896002, for example, has Asp9 levels of 2.3% (PTX 218.2) and Lot No. 716184 has
Asp9 levels of 3.6% (PTX 223.2). Both levels exceeded an Asp9 level of about 0.6%. Even
Mylan’s expert, Dr. Auslander, admitted that Original Angiomax®, considered as a whole, fails
to meet the Asp9 impurity level limitations in the asserted claims. (See Tr. 1022:2-23.)

9

Indeed, the first 12 batches of prior art Angiomax® made using TMC’s “inefficient” mixing
9
process each had Asp levels at or below 0.6%. Batch number 273786, manufactured in about May 2002,
9
9
had an Asp level of 0.0%. Five of these first dozen batches had Asp levels equal to or below 0.3%.
9
Three of these batches had Asp levels less than or equal to 0.2%.

20

Because the claims require the representative batch to not only be representative of all
commercial batches, but also represent levels for all potential batches made by said process, the
“stability” batches—Lot Nos. 339257 and 515495—do not meet the “representative” claim
limitation. TMC’s action in selecting and designating batches 339257 and 515495 as stability
batches is insufficient to establish that the Asp9 impurity levels in those batches represent the
impurity levels for all potential Original Angiomax® batches made by the same process, as
required by the claim. This is especially true in light of the other evidence at trial which showed
just the opposite—i.e., that some batches of Original Angiomax® contained Asp9 impurities well
above the impurity levels observed in batches 339257 and 515495.
Accordingly, Mylan has failed to prove that the Original Angiomax® batches alleged by
it to meet the on-sale bar under § 102(b) meet each and every limitation of the claim,10 and thus
Mylan has not established that the claimed invention was the subject of a commercial offer for
sale by clear and convincing evidence and therefore cannot establish invalidity of the ‘727 patent
claims due to an on-sale bar. See Netscape Commc’ns, 295 F.3d at 1323 (explaining that in
making a determination as to whether claims are invalid due to an on-sale bar, the court should
determine “whether the subject of the barring activity met each of the limitations of the claim,
and thus was an embodiment of the claimed invention.”). Because Mylan failed to satisfy the
first element of the on-sale bar, the Court does not address the second element, namely, whether
the invention of the ‘727 patent was ready for patenting. See Hamilton Beach, 726 F.3d at 1374.
10

Regarding the “maximum impurity level of Asp9-bivalirudin that does not exceed about 0.6%”
claim limitation, several Original Angiomax® batches contained Asp9 impurities greater than the
limitations in the asserted claims of about 0.6% (claim 1), 0.4% (claim 2), or 0.3% (claim 3). Assuming
that a single batch (other than an ANDA batch) meets the “pharmaceutical batches” term of the claim, the
Court would only look to the impurity level of the single batch itself regarding the “maximum impurity
level” claim limitation. Because TMC’s stability batches 339257 and 515495 do not meet the
“representative” limitation of the ‘727 patent claims, the fact that they each have a 0.5% Asp9-bivalirudin
impurity level would in isolation meet the “maximum impurity level” limitation is moot. See Netscape
Commc’ns, 295 F.3d at 1323.

21

B.

Obviousness

Mylan next contends that Claims 1-3, 7-10, and 17 of the ‘727 patent are invalid as
obvious under 35 U.S.C. § 103.11 (See R.553 at 12-24.)
1.

Legal Standard

Section 103 forbids issuance of a patent when “the differences between the subject matter
sought to be patented and the prior art are such that the subject matter as a whole would have
been obvious at the time the invention was made to a person having ordinary skill.” See U.S.C. §
103(a) (2006); see also KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 406, 127 S. Ct. 1727, 167 L.
Ed. 2d 705 (2007). The party seeking to invalidate a patent as obvious must prove by clear and
convincing evidence that “a skilled artisan would have been motivated to combine the teachings
of the prior art references to achieve the claimed invention, and that the skilled artisan would
have had a reasonable expectation of success from doing so.” Bristol-Myers Squib Co. v. Teva
Pharms. USA, Inc., 752 F.3d 967, 973 (Fed. Cir. 2014) (quoting Procter & Gamble Co. v. Teva
Pharms. USA, Inc., 566 F.3d 989, 994 (Fed. Cir. 2009)); see also Amgen, Inc. v. F. Hoffman–La
Roche Ltd., 580 F.3d 1340, 1362 (Fed. Cir. 2009) (“An obviousness determination requires that a
skilled artisan would have perceived a reasonable expectation of success in making the invention
in light of the prior art.”).
Obviousness is a question of law based on underlying factual findings: (1) the scope and
content of the prior art; (2) the differences between the claims and the prior art; (3) the level of
ordinary skill in the art; and (4) objective considerations of nonobviousness. See InTouch
Techs., Inc. v. CGO Comm’cns, Inc., 751 F.3d 1327, 1347 (Fed. Cir. 2014) (citing Graham v.
11

The AIA Pub.L. No. 112–29 took effect on September 16, 2012. Because the application for
the ‘727 patent was filed before that date, the Court refers to the pre-AIA version of § 103. See Allergan,
Inc. v. Apotex Inc., 754 F.3d 952, 958, n. 1 (Fed. Cir. 2014); Q.I. Press Controls, B.V. v. Lee, 752 F.3d
1371, 1377, n. 3 (Fed. Cir. 2014).

22

John Deere, 383 U.S. 1, 17-18, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966)). If, after assessing these
factors, a court concludes that the claimed invention was obvious, the claim is invalid under
§ 103. See KSR, 550 U.S. at 407. The Supreme Court has warned, however, that, while an
analysis of any teaching, suggestion, or motivation to combine known elements is useful to an
obviousness analysis, the overall obviousness inquiry must be expansive and flexible. See KSR,
550 U.S. at 415, 419; see also In re Cyclobenzaprine Hydrochloride Extended-Release Capsule
Patent Litig., 676 F.3d 1063, 1068-69 (Fed. Cir. 2012).
2.

Analysis

Mylan identified the references below as alleged prior art, along with the testimony of its
expert, Dr. Auslander, on the state of the pharmaceutical industry formulation and mixing arts, in
support of its obviousness contentions:


The Original Angiomax® product and processes;



The 2005 Original Angiomax® product insert (DTX 1498 at MEDMYL00019161919); and/or U.S. Patent No. 5,196,404 (PTX 277);



Dr. Musso’s July 15, 2006 Trip Summary Report (“Musso Trip Summary”) (PTX
007; DTX 1246);



KL Amsberry, et al., Compatibility and Stability of Bivalirudin in IV Admixtures,
AAPS Pharm. Sci. (1999) (“Amsberry Abstract”) (DTX 1498 at MEDMYL00019161919);



R. Bischoff & H. Kolbe, Deamidation of Asparagine and Glutamine Residues in
Proteins and Peptides: Structural Determinants and Analytical Methodology, 662 J.
Chromatography 261-78 (1994) (“Bischoff”) (DTX 1021); and



The 2005 European Medicines Agency (“EMEA”) publication (DTX 1498 at
MEDMYL0001239-1270).

(See R.553 at 12.)

23

a. The Scope and Content of the Prior Art
In order for prior art to be used in combination to determine obviousness under 35 U.S.C.
§ 103, the alleged prior art must first qualify as prior art under § 102(a), (b), (e), (f), or (g). See
OddzOn Prods., Inc. v. Just Toys, Inc., 122 F.3d 1396, 1402 (1997); see also Panduit Corp. v.
Dennison Mfg. Co., 810 F.2d 1561, 1568 (Fed. Cir. 1987) (“Before answering Graham’s
‘content’ inquiry [under the § 103 obviousness analysis], it must be known whether a patent or
publication is in the prior art under 35 U.S.C. § 102 – a legal question.”). In this case, the parties
do not dispute that the 2005 Original Angiomax® product insert,12 U.S. Patent No. 5,196,404,13
Amsberry Abstract,14 Bischoff,15 and the 2005 EMEA publication16 qualify as prior art under §
102(b). (See R.553 at 12; R.560 at 6-8.) The parties do, however, dispute whether the Original
Angiomax® product and the process used to produce it qualify as prior art under § 102(b). (See
id.) The parties also dispute whether the Musso Trip Summary qualifies as prior art under §
102(f). (See id.) The Court first addresses the teachings of the agreed prior art followed by those
references in dispute.

12

The 2005 Original Angiomax® product insert is dated December 6, 2005 and was available to
the public when TMC sold Original Angiomax® in 2005 and is therefore prior art under § 102(b).
13

The ‘404 patent issued on March 23, 1993 and is therefore prior art under § 102(b).

14

The Amsberry Abstract was published in The AAPS Journal in 1999 and is prior art under §

102(b).
15

The Bischoff article was published in the Journal of Chromatography B in 1994 and is prior art
under § 102(b).
16

Although Mylan’s reference is to a “2005 EMEA publication,” the document is the same as that
submitted during prosecution of the ‘727 patent (see DTX 1498 at MEDMYL0001239-1270) and referred
to as the EMEA publication, dated September 1, 2004 (see id. at MEDMYL0001289-90). Under either
date, the EMEA publication, which addresses Angiox—the European version of Angiomax® (see Tr.
147:20-22 (Flammia))—is prior art under § 102(b).

24

i.

The Undisputed Prior Art

The person of ordinary skill in the art17 knew that Original Angiomax® was a product
manufactured by Ben Venue, distributed by ICS, and marketed by TMC, indicated for use as an
anticoagulant and contained bivalirudin as its API. (See DTX 1498 at MEDMYL00019161919.) The “Description” of the Original Angiomax® product states:
Angiomax® (bivalirudin) is a specific and reversible direct thrombin inhibitor.
The active substance is a synthetic, 20 amino acid peptide. The chemical name id
D-phenylalanyl-L-prolyl-L-arginyl-L-prolyl-glycyl-glycyl-glycyl-glycyl-Lasparagyl-glycyl-L-aspartyl-L-phenylalanyl-L-glutamyl-L-glutamyl-L-isoleucylL-prolyl-L-glutamyl-L-glutamyl-L-tyrosyl-L-leucinetrifluoroacetate (salt) hydrate
(Figure 1). The molecular weight of Angiomax is 2180 daltons (anhydrous free
base peptide). Angiomax is supplied in single-use vials as a white lyophilized
cake, which is sterile. Each vial contains 250 mg bivalirudin, 125 mg mannitol,
and sodium hydroxide to adjust the pH to 5-6 (equivalent to approximately 12.5
mg sodium). When reconstituted with Sterile Water for injection the product
yields a clear to opalescent, colorless to slightly yellow solution, pH 5-6.
(Id. at MEDMYL0001916.)
Based on Original Angiomax®, the person of ordinary skill knew that thrombin
inhibitors may be formulated using conventional methods to prepare pharmaceutically useful
compositions with pharmaceutically acceptable carriers and with suitable fillers or bulking
agents, such as mannitol. (DTX 1498 at MEDMYL0001916; id. at MEDMYL0001855; see also
id. at MEDMYL0001239.) Regarding the pH of the reconstituted bivalirudin product (for
injection), the person of ordinary skill also knew that the pH could be adjusted by a base to
achieve a desired pH range of 5-6. (DTX 1498 at MEDMYL0001916.)
In addition, the person of ordinary skill knew that bivalirudin was composed of 20 amino
acids, and that the ninth position of the 20-mer peptide chain was an asparagine (Asn). (Id. at
17

Although the parties presented definitions for the person of ordinary skill that differ, the parties
agreed that the differences were essentially irrelevant and neither party argued that the differences had
any effect on the ultimate determination of obviousness. (Tr. 830:3-8; 1490:20-1492:6; see R.553; R.555;
R.558; R.560.)

25

MEDMYL0001916; see also id. at MEDMYL0001240; PTX 277, U.S. Patent No. 5,196,404
(disclosing structure of bivalirudin as a thrombin inhibitor).) The person of ordinary skill also
knew that Asn deamidates to Asp, and more specifically that “[a] known degradation product of
bivalirudin involves the deamidation of asparagine in position 9 to asp9-bivalirudin.” (Amsberry
Abstract, DTX 1498 at MEDMYL0001675, 1677; see also Bischoff, DTX 102118 (“[n]onenzymatic deamidation of asparagine (Asn) … residues in peptides and proteins is a welldocumented phenomenon which may occur under physiological conditions ...).)
ii.

The Disputed Prior Art
a)

Original Angiomax® Process is Prior Art
Under § 102(b)

Mylan argues that sales of the Original Angiomax® product prior to the critical date
render the process used to make it a public process. (See R.553 at 16.) Mylan further argues that
because the process was public, knowledge of the high Asp9 impurity levels and of TMC’s
internal efforts to avoid high Asp9 levels by adjusting the compounding process were also public.
(Id., at 17 (referencing TMC’s internal documents amending its compounding instructions for
Original Angiomax® relating to equal addition of base).) Finally, Mylan contends that the
person of ordinary skill would have been motivated to combine the attempt made by TMC to
adjust its compounding process with the other prior art to arrive at the claimed invention. (Id. at
17.)
Mylan relies on Torpharm, Inc. v. Ranbaxy Pharms., Inc., 336 F.3d 1322, 1326 (Fed. Cir.
2003) and Dippin’ Dots, Inc. v. Mosey, 476 F.3d 1337, 1344-45 (Fed. Cir. 2007) to assert that
18

DTX 1021 was admitted not for the truth of the matter asserted, but rather to establish the state
of the prior art. See Tr. 837:21-838:3; see also Abbott Labs. v. Diamedix Corp., 969 F.Supp. 1064, 1067,
n. 1 (N.D. Ill. 1997) (relying on a reference as non-hearsay when being offered as a verbal act and not to
prove the truth of the matter asserted and recognizing its relevance as its very existence and the effect its
existence had on the knowledge base of those in the field of the art).

26

because TMC sold Original Angiomax® product one year before the ‘727 patent was filed, the
process by which Original Angiomax® was produced is public and, hence prior art under §
102(b). In both cases, the claims-at-issue were process claims and the Federal Circuit found
because the product made by the claimed process was sold more than one year before the filing
of the patent-in-suit, that sale rendered the process prior art under § 102(b), regardless of whether
it was disclosed to the public or not. See Torpharm, 336 F.3d at 1326; Dippin’ Dots, 476 F.3d
1337, 1344-45 (Fed. Cir. 2007) (emphasis added) (stating “[t]he public sale of goods produced
by a process more than one year before a patent is filed places that process in the § 102(b) prior
art”); see also D.L. Auld. Co., 716 F.2d 1144, 1147-48 (Fed. Cir. 1983) (holding that “a party’s
placing of a product of the method invention on sale more than a year before that party’s
application filing date must act as a forfeiture of any right to the grant of a valid patent on the
method to that party if circumvention of the policy animating § 102(b) is to be avoided in respect
of patents on method inventions.”).
The Court agrees with Mylan that the Original Angiomax® product and the process by
which it was made are prior art under § 102(b). The Torpharm case supports the Original
Angiomax® product as “a reference under section 103 against the claimed invention” because it
was sold before the ‘727 patent’s critical date. See id., 336 F.3d at 1327 (explaining that the
obviousness inquiry requires the district court to determine whether the Form 1 ranitidine of
uncertain genesis sold in 1992, in conjunction with the prior art, rendered obvious TorPharm’s
claimed process of crystallizing improved Form 1 ranitidine from a three- or four-carbon alcohol
solvent.”). Because the Court has found that Original Angiomax® was sold before the ‘727
patent’s critical date (see supra Discussion,II.A.2.a.), the Original Angiomax® product becomes
a reference under § 103 against the claimed invention. The Dippin Dots case supports the

27

premise that because the Original Angiomax® product was sold before the ‘727 patent’s critical
date, not only is the Original Angiomax® product prior art, but the process by which Original
Angiomax® product was made is also prior art under the § 102(b) on-sale bar and “is also prior
art for the purposes of obviousness under § 103.” See id., 476 F.3d at 1344.
These cases, however, do not reach as far as Mylan contends. Namely, these cases do not
stand for the proposition that everything TMC internally knew about the Original Angiomax®
product and the process used to make it amounts to prior art under § 102(b). The claims at issue
are directed to pharmaceutical batches of a product made by a compounding process that differs
from the process used to make the Original Angiomax® product. In other words, the claims at
issue in this case are not the process claims for production of Original Angiomax® like the
claims presented in Torpharm. Nor are the claims at issue in this case the exact process for
production of Original Angiomax®, with additional steps recited, like the claims presented in
Dippin’ Dots. Rather, the claims at issue here are directed to pharmaceutical batches of
bivalirudin with maximum impurity levels of Asp9-bivalirudin impurity of 0.5%. As such, the
Court finds consistent with Torpharm and Dippin’ Dots, that the Original Angiomax® product
and the process used to produce Original Angiomax® qualify as prior art for the purposes of
obviousness under § 103. The Court does not find, however, that all of TMC’s internal efforts to
avoid high Asp9 levels by adjusting the Angiomax® compounding process also become prior art,
absent a showing that this knowledge was disclosed to the public. (See infra Discussion,
II.B.2.a.ii.c).)
The Court finds, therefore, that the person of ordinary skill in the art being aware of the
Original Angiomax® product, knew bivalirudin to be a 20 amino acid peptide with an asparagine
residue at the ninth position, and knew it was used as an anticoagulant. The Original

28

Angiomax® product also taught that bivalirudin drug product was formulated as a dry,
lyophilized cake and when reconstituted, the pH was adjusted with base (sodium hydroxide) to
reach a pH of 5-6. The Original Angiomax® product was also known to be produced by a
compounding process. The compounding process creates the final pharmaceutical product and
involves three basic steps. First, the bivalirudin API is dissolved into a mannitol solution to form
a bivalirudin solution. Second, the resulting bivalirudin solution is mixed with a pH-adjusting
solution, such as sodium hydroxide, to raise the pH of the bivalirudin API to an acceptable level.
Third, the mannitol solution and the pH-adjusting solution are removed from the mixture to form
the final drug product. More specifically, the original manufacturing process for commercial
batches of Original Angiomax® was disclosed in the ‘727 patent as Example 4. (See PTX 1,
col.21:44-col.22:27; Tr. 211:8-11; 235:6-14; 269:23 (Musso).)
b)

Portions of the Musso Trip Summary Are
Prior Art Under § 102(f)

Section 102(f) is a derivation provision which provides that a person shall be entitled to a
patent unless “he did not himself invent the subject matter sought to be patented.” 35 U.S.C. §
102(f). Thus, § 102(f) “requires that the patentee be the actual inventor of the subject matter
patented.” See New England Braiding Co., Inc. v. A.W. Chesterton Co., 970 F.2d 878, 883 (Fed.
Cir. 1992).
Mylan argues that “the July 14, 2006 communication from BVL employees to named
inventor Musso” is prior art under 35 U.S.C. § 102(f). (See R.553 at 17.) The Musso Trip
Summary is an internal report prepared by Dr. Musso on July 15, 2006, summarizing a meeting
that occurred at Ben Venue on July 14, 2006 “to follow-up on a recent failure of a bivalirudin
batch due to high Asp9 levels (Limit NMT 1.5%).” (PTX 7.) TMC responds that in order to find
the Musso Trip Summary prior art under § 102(f), Mylan must prove by clear and convincing
29

evidence “both prior conception of the invention by another and communication of the
conception to the patentee . . . .” (See R.560 at 6 (emphasis original).) TMC further argues that
“[t]he communication must be sufficient to enable one of ordinary skill in the art to make the
patented invention.” (Id.) TMC’s recited requirements, however, apply when a party is
attempting to prove invalidity by derivation under § 102(f), not when a party is attempting to use
the disclosure as prior art under § 102(f) in an obviousness analysis. See Creative Compounds,
LLC v. Starmark Labs., 651 F.3d 1303, 1313 (Fed. Cir. 2011) (citing Eaton Corp. v. Rockwell
Int’l Corp., 323 F.3d 1332, 1344 (Fed. Cir. 2003) (explaining that to prove derivation under §
102(f), a patent challenger must prove both prior conception of the invention by another and
communication of that conception to the patentee by clear and convincing evidence); Galderma
Labs, LP v. Paddock Labs, Inc., No. 2:09cv002Y, 2011 WL1119700, at *7 (N.D. Tex. March 28,
2011); see also Gambro Lundia AB v. Baxter Healthcare Corp., 110 F.3d 1573, 1577 (Fed. Cir.
1997) (explaining the requirements of a showing of prior conception and an enabling disclosure
for an anticipation argument of derivation under § 102(f) and emphasizing there is not a
determination of obviousness in a § 102(f) analysis).
Mylan relies on OddzOn Products, Inc. v. Just Toys, Inc., and its holding that “subject
matter derived from another not only is itself unpatentable to the party who derived it under §
102(f), but, when combined with other prior art, may make a resulting obvious invention
unpatentable to that party under a combination of §§ 102(f) and 103.” 122 F.3d 1396, 1403-04
(Fed. Cir. 1997). In OddzOn, the Federal Circuit found that the district court did not err by
considering confidential disclosures known to the inventor to be prior art under the combination
of §§ 102(f) and 103. See id. at 1404; see also IGT v. Bally Gaming Int’l., Inc., 610 F. Supp. 2d
288, 321 n.24 (D. Del. 2009) (referencing OddzOn’s holding that “§ 102(f) non-public subject

30

matter can be used to reject a claim of invention by another in possession of the § 102(f) subject
matter under a combination of § 102(f) and § 103”). The confidential disclosures in OddzOn
were two design references provided to the inventor by another.19 122 F.3d at 1403-04.
The July 15, 2006 Musso Trip Summary, however, is not the same as the confidential
disclosures in OddzOn, because it is not a disclosure provided to Dr. Musso from another.
Instead, Dr. Musso prepared the report himself to memorialize the meeting with Ben Venue, and
dated it the following day. (PTX 7; DTX 1246; Tr. 199:21-201:7 (Musso).) Mylan’s request
that the “July 14, 2006 communication from BVL employees to named inventor Musso is prior
art under § 102(f),” when the report is actually dated as prepared by Dr. Musso on July 15, 2006
is not a mistake on Mylan’s part, but rather reflects the complex nature of Mylan’s request. The
Musso Trip Summary contains Dr. Musso’s recollection of discussions he had on July 14, 2006
with Ben Venue employees working with the bivalirudin batches for Original Angiomax® and
TMC, discussions that addressed the recent failure of bivalirudin batches due to high Asp9 levels
(i.e. levels beyond the 1.5% allowed for in the product specification). (See PTX 7.1.) As such,
the report contains both summaries of information Ben Venue employees allegedly provided to
Dr. Musso on July 14, 2006, as well as Dr. Musso’s own thoughts, reactions, and proposals
experienced during and after that meeting. Mylan specifically advocates for consideration of the
“July 14, 2006 communication from Ben Venue employees to named inventor Musso” as prior
art under § 102(f), and not consideration of the Musso Trip Summary in its entirety created on
July 15, 2006.

19

Although the patent in OddzOn was a design patent, the obviousness inquiry “applies with
equal force.” See Litton Systems, Inc., v. Whirlpool Corp., 728 F.2d 1423, 1441 (Fed. Cir. 1984)
(overruled on other grounds, Braun Inc. v. Dynamics Corp. of America, 975 F.2d 815 (Fed. Cir. 1992))
(“Accordingly, 35 U.S.C. § 103 (and all the case law interpreting that statute) applies with equal force to a
determination of the obviousness of either a design or a utility patent.”)

31

Employees of both TMC and Ben Venue, along with the inventor Dr. Musso, attended
the July 14, 2006 meeting. (PTX 7; DTX 1245; Tr. 67:19-68:18 (Flammia); Tr. 200:6-202:19
(Musso); Smith Dep. 71:06-71:18, June 10, 2014.) Dr. Musso testified that the Musso Trip
Summary was an accurate memorialization of what happened during the meeting. (See Tr.
200:6-201:11 (Musso).) Dr. Musso further testified that Ben Venue operators shared their
experiences regarding the compounding process, and referred to the section of three bullet points
in the report, shown below, which states:
The operators shared their experience based comments which included:
 Vary the stirrer heights so that one mixed the upper and one mixed the lower
level of the tank
 Increase the stir speed prior to base addition
 Add base through a dip tube so that it is dispersed in the bottom of the tank
where mixing is still possible (the marshmallow solid was suggested to be
more on the top portion of the formulation).
(PTX 7.2; Tr. 202:17-204:7 (Musso).) Dr. Musso also testified about the operators insights
regarding the compounding process, and referred to the section of five bullet points in the report
as referring to the Original Angiomax® compounding process, also described in Example 4 of
the ‘727 patent. (Tr. 261:263:8 (Musso).)20 The portion of the Musso Trip Summary that Dr.

20

TMC employee, Angie Green, worked on the commercial side of Angiomax® and was a part
of the Ben Venue and TMC team as a liaison to facilitate communication between the companies.
(Deposition Video, Angie Green, 44:22-45:17; 47:25-48:25.) Ms. Green testified that she attended the
meeting with Ben Venue on July 14, 2006 and kept minutes from that meeting. (See DTX 1245.) The
“Comments” section of those minutes reflects various comments made by employees of Ben Venue and
TMC (including Musso) throughout the meeting. (Id.) The minutes indicate the Ben Venue provided Dr.
Musso with information, but the conclusions regarding what the problem may be are attributed to
inventor Gary Musso, “BVL gave Gary the sequence of the cleaning process. Gary does not believe this
is the problem.” In addition, numerous suggestions are attributed to inventor, Musso, e.g., a suggestion of
“testing ph at bottom of mixers,” (DTX 1245, at MEDMYL0007884); “3 different mixing times,” (id.);
“using glass for container in order to see solution at the bottom” (id.). The numerous contributions noted
by inventor Musso as well as by other TMC employees highlight the challenge in parsing out what
portions of the Musso Trip Summary correlate to statements communicated from Ben Venue or TMC
during that meeting.

32

Musso testified contained the comments from the Ben Venue operators is shown below, and
states:
The two operators provided some interesting insights such as:
 There is no safety issue with base addition, so stirring could be rapid in the
base addition
 They had no problem adding all the base at once (both operators appeared
physically strong)
 They do not recall shutting off the stirrers during base addition
 They also commented that with the low speed stir, the mixture is like a
“marshmallow” after ¾ of the base addition and the stirrer is ineffective at
low speed.
 Under low speed mixing, the formulation is generally not stirred at the end
of the addition as the solid mass is too thick. In a discussion of portionwise addtion [sic] the operators also indicated that the last base addition
just sat on top of the marshmallow and did not mix – similar to the
comments related to viscosity discussion held previously.
(PTX 7.2; Tr. 261:263:8 (Musso).) The evidence revealed that at some point later in the
meeting when the participants were still discussing the issues, Ben Venue operators were no
longer present. (Tr. 68:9-10 (Flammia); Tr. 264:10 (Musso).) Timothy Smith, Director of
Product and Process Development at Ben Venue at the relevant time, also testified that he
attended the meeting with TMC on July 14, 2006. (Smith Dep. 71:06-71:18; see Gezzar Dep.
89:14-89:25, January 18, 2013.) Mr. Smith testified that Ben Venue operators were asked to join
the meeting, identifying one of the operators by name. (Smith Dep. 73:08-73:19.) In response to
questions from Mylan’s counsel regarding the Musso Trip Summary, Mr. Smith testified that he
“presumed” what the document was referring to, but provided no testimony regarding whether
the Ben Venue operators provided the recited information to Musso and TMC or whether the
meeting participants discussed the recited information, and Mylan never asked. (See Smith Dep.
74:04-79:11.) Mylan’s scientific expert, Dr. Auslander, provided a conclusory opinion regarding
the Musso Trip Summary, but later conceded that without having attended the meeting, it was
not possible for him to tell who had led the discussion, what dialogue took place, or the dynamic
33

of the discussion (e.g., who asked who what), and that it would be speculation to attempt to
know. (Tr. 1013:6-1017:13 (Auslander).) In fact, Dr. Musso testified that the cause of the high
Asp9 failure was not identified at the meeting. (Tr. 201:19-21.) Dr. Musso further stated that he
and Dr. Krishna “came up with the ideas on how to fix and make the improved product. Ben
Venue did not contribute to those inventions.” (Tr. 205:19-21.)
Based on this record, and in particular on Dr. Musso’s testimony and the Musso Trip
Summary itself (a document created by Dr. Musso very close to the date of communication with
Ben Venue), Mylan has established by clear and convincing evidence that the two bulleted
portions from the Musso Trip Summary cited above qualify as prior art under § 102(f) in a § 103
obviousness analysis.21 Mylan has failed, however, to establish by clear and convincing
evidence whether any additional portions of the Musso Trip Summary qualify as prior art
communications from Ben Venue to TMC and Musso under § 102(f) for the § 103 obviousness
inquiry. The Court, therefore, is unable to rely on the disclosures in the summary, in whole, as
prior art.
c)

TMC’s Efforts Associated With
Production of Original Angiomax®
Identified in Internal Documentation Are
Not Prior Art Under § 102(b)

Mylan contends that the efforts TMC made associated with Original Angiomax®
production identified in internal TMC documentation are also prior art and were known to the
person of ordinary skill. (R.553 at 16-17.) The Court disagrees.

21

Although the second bullet point seems to memorialize not only what Ben Venue operators
communicated to Dr. Musso, but also includes additional information that appears to be reflective of Dr.
Musso’s own thoughts or impressions (“both operators appeared physically strong”), the remaining
portions of the report are clearly understood as attributable to comments made by Ben Venue operators.
(See PTX 7.2.)

34

Again relying on the Federal Circuit’s decision in Torpharm, Mylan asserts that TMC’s
2005 effort to avoid high Asp9 levels by adjusting the Angiomax® compounding process
constitutes prior art. (Id. at 17.) Specifically, Mylan argues that TMC’s amendment to its
compounding instructions in 2006, reflected only in TMC’s internal documents, is prior art. (Id.
at 17.) Mylan references the confidential copies of the Master Production Record for
Bivalidurdin, 50 mg/ml from Ben Venue to TMC. (See DTX 1211.) The Torpharm and Dippin’
Dots cases do not, however, stand for the proposition that because the Original Angiomax®
product and the process used to make the sold batches of Original Angiomax® product are prior
art, all TMC’s internal and confidential knowledge about that product and its process also
become prior art. The Court has already found that the process as disclosed in Example 4 of the
‘727 patent is prior art based on the sales of Original Angiomax®. (See supra Discussion,
II.B.2.a.ii.a).) This discussion indicates the process change that was implemented in the 2006
amendment. In particular, the addition of the pH-adjusting solution to the bivlirudin solution as
“either all at once, or rapidly in multiple portions, while the bivalirudin solution was mixed by
two paddle mixers located at the top and bottom of the bivalirudin solution.” (PTX 1, col. 5962.)
The fact that this process change is included in the prior art disclosure of the bivalirudin
compounding process, however, does not equate to a finding that all of TMC’s internal efforts
surrounding that product and its process or problems with that process and the evolution of the
process also become prior art. Indeed, TMC and Mylan both agree and the evidence at trial
demonstrated that the two failed Original Angiomax® batches—Lot Nos. 716184 and 896002
with Asp9 levels above the 1.5% specification—were not sold and, thus, were not publicly
available. (See Tr. 59:4-9 (Flammia); see R.553 at 9 (“When Original Angiomax batches were

35

not representative of this approved process … such batches were not representative and
destroyed …”).) The batches of Original Angiomax® that TMC sold to the public were those
batches that followed the original compounding process and met the regulatory requirements of
the specification. These batches would, therefore, if tested by the person of ordinary skill, not
demonstrate any problems with levels of Asp9 that fell outside of the product and process
specification’s maximum level of 1.5% Asp9 impurity.
The internal efforts TMC made to adjust its compounding process over time and the
internal, confidential communications surrounding those adjustments do not become prior art
based on Torpharm and Dippin’ Dots, and Mylan has not cited any case law that supports such a
proposition. Mylan also failed to offer any evidence surrounding public disclosure of TMC’s
internal documents addressing the problem of high Asp9 levels in Original Angiomax® or the
steps TMC undertook to combat that problem. See In re Lister, 53 F.3d 1307, 1311 (Fed. Cir.
2009) (citing In re Klopfenstein, 380 F.3d 1345, 1350 (Fed. Cir. 2004)) (explaining the public
accessibility of a reference “is determined on a case-by-case basis based on the facts and
circumstances surrounding the reference's disclosure to members of the public.”). The Court,
therefore, finds no evidentiary support that those documents were “sufficiently accessible to the
public interested in the art” and as such, cannot rely on TMC’s internal documents as prior art
under § 102(b). See In re Lister, 583 F.3d at 1311 (citations omitted) (explaining that “[i]n order
for a printed publication to qualify as prior art within the meaning of § 102, a reference ‘must
have been sufficiently accessible to the public interested in the art.’”).
b. Differences Between the Claimed Invention and the Prior Art
In order to determine whether the claimed invention of the ‘727 patent was obvious to the
person of ordinary skill at the time the invention was filed (July 27, 2006), the Court must

36

determine the differences between the claimed invention and the prior art. See InTouch Techs.,
Inc., 751 F.3d at 1347. The ‘727 patent’s claimed invention is directed to “pharmaceutical
batches” of “bivalirudin” and a “pharmaceutically acceptable carrier for use as an anticoagulant”
wherein the batches have a pH adjusted by a base (pH of 5-6) and wherein the batches “have a
maximum Asp9 impurity level of 0.6%.” (PTX 1, col. 25:56-64.)
The differences between the prior art and the claims of the ‘727 patent are narrow when
compared to Original Angiomax® as prior art. The prior art Original Angiomax® meets a
number of the limitations of claim 1 of the ‘727 patent. (See R.553 at 15; (Tr. 1431:10-1432:22
(Klibanov).) The person of ordinary skill knew that Original Angiomax® contained bivalirudin
formulated with a pharmaceutically acceptable carrier (e.g., including a bulking agent such as
mannitol) for use as an anticoagulant. (DTX 1498 at MEDMYL0001916; id. at
MEDMYL0001855; see also id. at MEDMYL0001239.) The person of ordinary skill also knew
that the pH of the reconstituted bivalirudin product (for injection) could be adjusted by a base to
achieve a desired pH range of 5-6. (DTX 1498 at MEDMYL0001916.) Regarding bivalirudin
and its impurities, the person of ordinary skill knew that bivalirudin was composed of 20 amino
acids, and that the ninth position of the 20-mer peptide chain was asparagine. (Id. at
MEDMYL0001916; see also id. at MEDMYL0001240; PTX 277, U.S. Patent No. 5,196,404
(disclosing structure of bivalirudin as a thrombin inhibitor).) Further, the person of ordinary skill
knew that asparagine (Asn) deamidates to aspartic acid (Asp), and more specifically that “[a]
known degradation product of bivalirudin involves the deamidation of asparagine in position 9 to
asp9-bivalirudin.” (Amsberry Abstract, DTX 1498 at MEDMYL0001675, 1677; see also
Bischoff, DTX 1021 (“[n]on-enzymatic deamidation of asparagine (Asn) … residues in peptides

37

and proteins is a well-documented phenomenon which may occur under physiological conditions
...).)
i.

The Prior Art Does Not Disclose All Elements of the
Claimed Invention

The prior art does not, however, disclose the remaining limitations of claim 1 requiring
that the “pharmaceutical batches” and the “maximum impurity level of Asp9-bivalirudin does not
exceed about 0.6%” claim limitations. (See R.553 at 15; (Tr. 1431:10-1432:22 (Klibanov).)
In an attempt to overcome this absence, Mylan first relies on the story of the invention.
(See R.553, 13-15.) The story of the Improved Angiomax® invention, as recounted by Mylan,
however, repeatedly relies on evidence that does not qualify as prior art under § 102 – namely,
internal TMC documents and knowledge that indicate the problems with the Asp9-bivalirudin
levels. In particular, Mylan relies on these documents for disclosure of the problem with Asp9
levels in the prior art. Without the support provided by these documents as prior art, the
obviousness analysis requiring the information they contain fails. (See infra Part II.B.2.b.ii.)
Second, Mylan essentially repeats its anticipation argument and relies on internal TMC
documents showing that 79 of 87 prior art Angiomax® batches had Asp9 levels at or below about
0.6% and that TMC sold dozens of these batches prior to the critical date for the ‘727 patent.
The Court must, however, consider the claimed subject matter as a whole and, in doing so, it
cannot focus only on the Asp9 levels in isolation. See 35 U.S.C. § 103; see also KSR, 550 U.S. at
406 (holding that a party asserting obviousness must show “the subject matter [of the asserted
claims] as a whole would have been obvious”) (emphasis added). To ignore the “pharmaceutical
batches” limitation of the ‘727 patent claims would, therefore, be improper in the obviousness
analysis. As noted above, in relation to Mylan’s anticipation argument and the parties’ dispute
regarding the “pharmaceutical batches” limitation, Mylan failed to establish that the Original
38

Angiomax® batches on which it relies meet each and every limitation of the asserted claims.
(See supra Part II.A.2.) In particular, Mylan failed to prove that any of the cited Original
Angiomax® batches are: (1) “representative of all commercial batches . . . made by a
compounding process,” and (2) have “levels of . . . Asp9-bivalirudin … [that] represent levels for
all potential batches made by said process.” (PTX 1, claim 1; see also R.119.) As such, the
Court finds that the evidence of prior art provided in this case, does not meet all the limitations
of the claimed invention.
ii.

The Prior Art Did Not Provide a Motivation for a
Person Skilled in the Art to Combine the Prior Art
Elements – the Asp9 Problem Was Not Known or
Predictable

In addition, Mylan asserts that the above limitations were obvious to the person of
ordinary skill in the art, a person who understood the generation of the Asp9 impurity in a
bivalirudin compounding process. (See R.553 at 17.) Mylan relies on the Amsberry Abstract
and Bischoff for teaching the person of ordinary skill about generation of the Asp9 impurity in
the bivalirudin compounding process. Further, Mylan argues that the person of ordinary skill
would apply this knowledge to bivalirudin because he or she: (1) knew that the bivalirudin
peptide contained an asparagine residue (Asn) preceded and followed by glycine residues (Gly);
(2) knew that the bivalirudin compounding process included a pH-adjustment step wherein the
bivalirudin solution is exposed to sodium hydroxide, a strong base (see DTX 1080, at
MEDMYL4509447-75 (compounding instructions for mannitol solution (Part I), sodium
hydroxide/mannitol solution (Part II), and bivalirudin formulation (Part III)); PTX 223); and (3)
knew the specific Asp9 levels of prior art Original Angiomax® as they had access to the product
and would have been motivated to measure the Asp9 levels of prior art Original Angiomax®.
(See R.553 at 17-18.)
39

The Court agrees that the prior art teaches the bivalirudin peptide sequence and teaches
that asparagine residues, generally, can be susceptible to deamidation under basic conditions.
The prior art also teaches that a known degradation product of bivalirudin involves deamidation
of asparagine in position 9 to asp9-bivalirudin. The Court also agrees that the prior art teaches
the presence of a pH-adjustment step of the Original Angiomax® compounding process.
Mylan’s argument fails, however, from its reliance on the presence of a problem with the levels
of Asp9-bivalirudin impurity and, hence has no basis for a motivation to measure the Asp9 levels
of prior art Original Angiomax®. This motivation presumes the person of ordinary skill has
knowledge of the problem of high and variant Asp9 levels with prior art Original Angiomax®.
The prior art does not specifically teach that asparagine deamidation or high and variant Asp9
levels were a problem with the bivalirudin peptide. Mylan did not establish that the specification
parameters for Original Angiomax® were available as prior art,22 nor did it establish that the
person of ordinary skill would be aware of any problems with the Original Angiomax® batches
falling out of specification.
Mylan’s argument relies on the inferences to link the prior art disclosures that create a
drawn out hypothetical scenario that does not render the claims obvious by clear and convincing
evidence. First, Mylan’s argument relies on the person of ordinary skill in the art to have known
to produce batches of bivalirudin drug product following a version of the prior art process
disclosed in Example 4 of the ‘727 patent and in doing so, to have recreated the Asp9-bivalirudin
impurity level problem experienced by TMC and Ben Venue. Second, Mylan’s argument relies
on that person of ordinary skill understanding that because the known degradation product of the

22

Mylan’s expert testified that the term “out of specification” was a colloquial term that “refers to
a specification that’s been generally filed with the government, with the FDA, or it could be an internal
specification.” (Tr. 798:25-799:11.)

40

Asp9-bivalirudin impurity was produced by exposure to basic conditions, that the pH-adjustment
step in the compounding process was the culprit of the problem (meanwhile ignoring the prior art
that indicates other problems could occur and affect the rate of deamidation). Mylan’s argument
then relies on the person of ordinary skill to have known that changes to the compounding
process, namely to the mixing conditions, would result in a more effective process and a lower
Asp9-bivalirudin impurity level. In short, Mylan’s argument is fraught with unsupported
assumptions, which the Court addresses below.
While it is true that the Amsberry Abstract teaches that Asp9-bivalirudin was a known
degradation product of bivalirudin. Amsberry also indicates that under the conditions examined,
“no significant change in the concentration of asp9-bivalidurin” was observed and further
concludes that “[b]ivaliridin is stable to deamidation during the normal course of IV
administration in PTCA and AMI clinical trials.” (DTX 1498 at MEDMYL0001677.) This
disclosure would not lead the person of ordinary skill in the art to believe that bivalirudin had
problems with high and variant Asp9 levels, either in production or in practice.23 The general
disclosure of Bischoff is no more convincing. Bischoff does teach the person of ordinary skill
that generally, a deamidation of asparagine can occur (at neutral and basic pH values) in peptides
and proteins containing asparagine residues, but the prior art also teaches that other residues
contained in bivalirudin, e.g., aspartic acid at position 11, were susceptible to changes in pH and
to reactions such as isomerization. (See DTX 1023; see also PTX 1, claim 1 (“… wherein the
23

Mylan’s reliance on Dr. Klibanov is unpersuasive and misconstrues his testimony. Mylan
alleges Dr. Klibanov “readily admits that the susceptibility of asparagine to deamidation when it is
followed by a glycine was ‘widely known.’” (R.553 at 18.) In response to Mylan’s questions regarding
the adverse effects of a high level of the Asp9 impurity, Dr. Klibanov testified, however, that “[i]t’s
widely known in the industry that when you produce the aspartic acid residue [from deamidation of
asparagine], which is followed by glycine, that particular peptide bond [Asp-Gly] will be susceptible to
hydrolysis.” (Tr. 1428:12-1429:13.) Dr. Klibanov was therefore testifying about the susceptible nature of
an Asp-Gly peptide bond that has already formed (after deamidation of the Asn residue), not the
susceptibility of the occurrence of the deamidation reaction, itself.

41

batches have a pH adjusted by a base, said pH is about 5-6 when reconstituted in an aqueous
solution for injection).) The bivalirudin peptide contains both asparagine and aspartic acid
residues. (See PTX 397, ¶ 26.) Mylan provided no testimony to justify ignoring Aswad’s
disclosure for the potential problems of pH changes and aspartic acid isomerization, but
accepting Bischoff’s disclosure for asparagine deamidation. Therefore, while the person of
ordinary skill may have been aware that asparagine residues, followed by glycine, can be
susceptible to deamidation, they would also have been aware that other residues, like aspartic
acid, were susceptible to isomerization and would not have, therefore, been led by the prior art to
focus on the deamidation of asparagine as a problem with bivalirudin.
Mylan’s reliance on an internal memorandum dated in 1994 from Biogen, Inc. to TMC,
admittedly not prior art, (DTX 1020) is unpersuasive. The evidence showed that even after
receiving the information from Biogen, Inc., TMC still experienced high variable Asp9 levels.
(Tr. 1397:9 – 1399:10 (Klibanov).) In addition, the batches made by Biogen, Inc. did not all
report Asp9 levels, as such, it is impossible to determine if the production of batches from
Biogen, Inc.’s compounding process changes would have resulted in “pharmaceutical batches”
as required by the claims of the ‘727 patent. (Id.) Mylan admits that the Biogen solution
“produc[ed] batches that appeared to satisfy all the requirements of the asserted claims. (R.553
at 23.) Because Mylan did not prove by clear and convincing evidence that the Biogen batches
meet each and every limitation of the claimed invention and were made within a comparatively
short period of time, however, they do not suffice as evidence of obviousness. See Geo M.
Martin Co. v. Alliance Machine Sys. Int’l. LLC, 618 F.3d 1294, 1305 (Fed. Cir. 2010) (citing
Concrete Appliances Co. v. Gomery, 269 U.S. 177, 184, 46 S.Ct. 42, 70 L.Ed. 222 (1925)
(finding a simultaneous invention made one year prior to be strong evidence of what constitutes

42

the level of ordinary skill in the art as it was “[i]ndependently made, simultaneous inventions,
made ‘within a comparatively short space of time,” and therefore “persuasive evidence the
claimed apparatus ‘was the product only of ordinary mechanical or engineering skill.’”).
Without the internal documentation of the prior specification, the problem, and the entire
Musso Trip Summary available as prior art, Mylan has little to no evidentiary support for its
obviousness argument. Mylan failed to establish that the person of ordinary skill was aware of
TMC’s problems with the Original Angiomax® process – more specifically, the problem with
the Asp9 levels falling out of specification. Further, Mylan failed to establish that the prior art
taught the person of ordinary skill what the product specification for Original Angiomax was, as
Mylan consistently refers to confidential, internal TMC and Ben Venue documents for this
premise. (See e.g., DTX 1080; DTX 1076; DTX 1240; DTX 1216; DTX 1263; DTX 1366; DTX
1406; 1645.)24 Even if the specification limit of 1.5% Asp-bivalirudin impurity level for
Original Angiomax® had been established as known in the prior art, however, it is
counter-intuitive that the person of ordinary skill knowing that an FDA approved product had a
specification of 1.5% maximum for use as an injectable would find that limit to be a problem,
and Mylan failed to prove otherwise. (See Tr. 1429:25-1430:3 (Klibanov) (“Q. And, in fact,
TMC believed it was safe to sell Angiomax to doctors to inject into patients for those years with
Asp9 levels of 1.5 percent, correct? A. That’s my understanding.”)
The person of ordinary skill would have had access to the prior art batches of Original
Angiomax® sold and had the ability to test those batches, including the ability to test the Asp9
24

Indeed, during cross examination of Dr. Klibanov, Mylan’s questioning indicated that even
generic filers do not know what the Asp9-bivalirudin levels are for the Angiomax® products. (See Tr.
1434:8-10 (“Q. Right. Because, of course, the generic filers don’t know what the Asp9 level is of
Angiomax set by the FDA, right? A. I’m not sure about that.”); see also DTX 1526, Portion of TMC’s
NDA for Angiomax, dated 2/23/2011, and marked “Confidential.”) Mylan also admits that DTX 1080
and DTX 1216 are internal reports, both produced under the “Highly Confidential-Outside Counsel Only
designation and that neither is a patent or publication. (See R.553 at 39.)

43

levels of those batches. Those sold batches, however, did not contain problematic Asp9 levels.
Mylan argues the person of ordinary skill would have been aware that the bivalirudin
compounding process included a pH-adjustment step that exposed the bivalirudin solution to a
strong base, sodium hydroxide, and paired this knowledge with the prior art’s teaching of
susceptibility of the Asn residue in the ninth position of bivalirudin to deamidation. Even
knowing this information, however, when asked if the person of ordinary skill would predict a
deamidation problem for bivalirudin, Mylan’s own expert testified that he would “say that a
scientist or a person of ordinary skill in the art [would] say that’s a very plausible or possible
pathway and would do those experiments.” (Tr. 972:24-975:3.) Plausible and possible,
however, do not rise to the level of clear and convincing. See Buildex Inc. v. Kason Industries,
Inc., 849 F.2d 1461, 1463 (Fed. Cir. 1988) (citing Colorado v. New Mexico, 467 U.S. 310, 316,
104 S.Ct. 2433, 2437-38, 81 L.Ed.2d 247 (1984)) (“Although not susceptible to precise
definition, ‘clear and convincing’ evidence has been described as evidence which produces in the
mind of the trier of fact ‘an abiding conviction that the truth of [the] factual contentions are
highly probable.’”); see also United States v. Boos, 329 F.3d 907, 911 (7th Cir. 2003) (“highly
probable” is the Supreme Court’s definition of clear and convincing standard of proof).
Given that the problem of high and variant Asp9 levels remained unknown, as TMC
points out, a finding of obviousness is precluded. The Court recognizes that “[o]ne of the ways
in which a patent’s subject matter can be proved obvious is by noting that there existed at the
time of invention a known problem for which there was an obvious solution encompassed by the
patent’s claims.” KSR, 550 U.S. at 419–20. When a problem is “known in the field of endeavor
at the time of invention and addressed by the patent,” that problem “can provide a reason for
combining the elements in the manner claimed.” Id. at 420. In this case, however, Mylan has not

44

presented evidence that the person of ordinary skill knew of the problem with high and variant
Asp9 levels. The evidence at trial provided only conclusory and factually unsupported testimony
of a general motivation in the drug formulation business to minimize the level of impurities and
that the person of ordinary skill in the art would have been motivated to minimize or eliminate
the level of Asp9 impurity in the bivalirudin drug product. (Tr. 839:13-19 (Auslander); (Tr.
366:22-23 (Musso) (“one always seeks to control and limit impurities …”).) Mylan did not
provide facts to explain how the person of ordinary skill would know to combine the prior art
addressing the presence of Asp9 impurities in Asn-containing peptides generally with the
bivalirudin Original Angiomax® product specifically, especially in light of the evidence
presented at trial that there are other scientific considerations that must be taken into account
with peptides, such as bivalirudin. (See infra Part II.B.2.b.iii.) The inventor of the ‘727 patent
testified that addressing the high degree of variability in Asp9 impurity levels was the inventors’
motivation, a motivation that was not known to the person of ordinary skill. (See Tr. 328:3-23
(Musso).)
Without the problem of variability or high Asp9 levels known in the prior art, Mylan has
failed to establish by clear and convincing evidence a reason to combine the elements in the
manner claimed. See KSR, 550 U.S. at 420; see also Leo Pharm. Prods., Ltd. v. Rea, 726 F.3d
1346, 1356-57 (Fed. Cir. 2013) (“The problem was not known, the possible approaches to
solving the problem were not known or finite, and the solution was not predictable. Therefore
the claimed invention would not have been obvious to try to one of ordinary skill in the art.
Indeed ordinary artisans would not have thought to try at all because they would not have
recognized the problem.”). To ignore prior art teaching such as Aswad, and to focus on the
Amsberry Abstract and Bischoff, without also having knowledge of the problem with Asp9 levels

45

in Original Angiomax® or evidence that it would be predictable, becomes an improper analysis
driven by hindsight. See Innogenetics, N.V. v. Abbott Labs., 512 F.3d 1363, 1374 n.3 (Fed. Cir.
2008) (finding it impermissible to use “hindsight reconstruction of references to reach the
claimed invention without any explanation as to how or why the references would be combined
to produce the claimed invention.”); see also KSR, 550 U.S. at 421 (“A factfinder should be
aware … of the distortion caused by hindsight bias and must be cautious of arguments reliant
upon ex post reasoning.”). Thus, the Court finds that the prior art did not provide a motivation
for a person skilled in the art to combine the prior art elements.
iii.

Mylan Failed to Prove That a Person of Ordinary Skill
Would Not Have Had a Reasonable Likelihood of
Success in Creating the Claimed Invention

The Federal Circuit has observed that “in the medical arts ‘potential solutions are less
likely to be genuinely predictable,’ as compared with other arts such as the mechanical devices in
KSR.” Eisai Co. v. Dr. Reddy’s Labs., Ltd., 533 F.3d 1353, 1359 (Fed. Cir. 2008). In terms of
deamidation of a peptide sequence, Mylan’s expert testified that the person of ordinary skill in
the art would not be able to predict the rate (or existence) of deamidation based on a peptide
sequence without experimentation or pre-formulation analysis, but rather would be “very
encouraged to do this work” if the peptide sequence contained an amide bond (e.g., asparagine
residue). (Tr. 968:4-973:7 (Auslander) (asked if one could predict a deamidation problem in
bivalirudin without experimentation, Dr. Auslander responded: “to predict the chemistry of a
molecule that hasn’t been handled before is unrealistic ….”).) Mylan’s expert acknowledged
that his analysis focused on deamidation because it was allegedly a known problem at the time25

25

The basis for Dr. Auslander’s knowledge of deamidation as a problem, however, was internal
documents from Biogen that Mylan acknowledges are not prior art. (See Tr. 941:14-24 (“Q. Well why
did you focus on deamidation? A. Based upon the Biogen experiences, it was a known sensitivity of

46

and that he did not consider other possible reactions. (Tr. 941:2-12 (Auslander).) A peptide
sequence, like that of bivalirudin, however, is not only composed of asparagine residues with
amide bonds, and the person of ordinary skill would have also been aware of the susceptibilities
surrounding other amino acid residues to undergo interaction during a compounding process e.g.,
hydrolysis of peptide bonds, isomerization, racemization, oxidation, aggregation, precipitation,
adsorption. The evidence at trial established other types of degradation reactions can occur with
a peptide, including isomerization of aspartic acid residues (present at position 11 of bivalirudin),
(Tr. 937:12-946:18 (Auslander); see also DTX 1021, Bishchoff, at 276 (“The present review is
mainly concerned with deamidation of Asn residues and with isomerization at Asp-Gly sites as
one of the more common modifications found in both natural and recombinant proteins.”).) The
evidence at trial also demonstrated that other concerns existed surrounding potential changes in
the compounding process that affect the peptide, e.g., susceptibility of peptides to pH changes or
high-shear mixing. (DTX 1023; PTX 224; Tr. 997:24-98:21 (Auslander); 1394:6-15
(Klibanov).)
The evidence of the unpredictable nature of the chemical response of bivalirudin in a
compounding process to produce “pharmaceutical batches” of bivalirudin with “a maximum
impurity level of Asp9-bivalirudin that does not exceed about 0.6% as measured by HPLC,”
provides the person of ordinary skill, in light of the prior art, no reasonable expectation of
success. Moreover, Mylan did not present evidence to overcome the “difficult hurdle” of
unpredictability to establish its burden of a clear and convincing standard that a person of
ordinary skill would have a reasonable expectation of success to achieve “pharmaceutical
batches” as claimed in the ‘727 patent. See also Eisai Co. Ltd., 533 F.3d at 1359 (“To the extent
bivalirudin . . . .”).) In fact, at one point, Dr. Auslander stated “I don’t know how much public
information was available about bivalirudin.” (Tr. 942:23-24.)

47

an art is unpredictable, as the chemical arts often are, KSR’s focus on these ‘identified,
predictable solutions’ may present a difficult hurdle because potential solutions are less likely to
be genuinely predictable.”)
c. Objective Indicia of Nonobviousness
In an obviousness analysis, courts must consider objective evidence of nonobviousness in
the form of secondary considerations, as it “play[s] a critical role in the obviousness analysis.”
Leo Pharm. Prods, 726 F.3d at 1358. The Federal Circuit has repeatedly emphasized that
objective indicia “may often be the most probative and cogent evidence of nonobviousness in the
record.” Mintz v. Dietz & Watson, Inc., 679 F.3d 1372, 1378 (Fed. Cir. 2012) (quoting Ortho–
McNeil Pharm. Inc. v. Mylan Labs., Inc., 520 F.3d 1358, 1365 (Fed. Cir. 2008)). These indicia
“provide objective evidence of how the patented device is viewed in the marketplace, by those
directly interested in the product” at the time of the invention. See Mintz, 679 F.3d at 1378
(quoting Demaco Corp. v. F. Von Langsdorff Licensing Ltd., 851 F.2d 1387, 1391 (Fed. Cir.
1988)). In doing so, they help “inoculate the obviousness analysis against hindsight.” Id.; see
also Graham, 383 U.S. at 36 (recognizing that the objective indicia of obviousness “guard
against slipping into use of hindsight” and help “resist the temptation to read into the prior art the
teachings of the invention in issue”). Here, TMC points to failure of others, Mylan’s alleged
copying of TMC’s product, and unexpected results, as objective indicia of nonobviousness.
(R.560 at 14-16; see also R.416 at 39-48.)
i.

Failure of Others

First, TMC argues that the evidence at trial demonstrated the failure of others to solve the
problem of high Asp9-bivalirudin levels. (See R.560 at 13-14.) Evidence that others tried but
failed to develop a claimed invention “may carry significant weight in an obviousness inquiry.”

48

In re Cyclobenzaprine, 676 F.3d at 1081; see also Galderma Labs., L.P.v. Tolmar, Inc., 737 F.3d
731, 736 (Fed. Cir. 2013). TMC contends that the evidence showed that Ben Venue could not
solve the problem with the high Asp9-bivalirudin levels despite its ten plus years of experience in
formulating bivalirudin. (Tr. 59:21-61:1; 62:17-24 (Flammia); PX223.) Ben Venue investigated
the lot failure and, with TMC, reviewed and implemented manufacturing process changes, e.g.,
instructions for adding the pH-adjusting solution to the bivalirudin solution in aliquots. (Tr.
61:22-62:12 (Flammia); Tr. 207:9-208:3 (Musso); PTX 223.15; DTX 1211.) This attempt,
however, was unsuccessful as a lot Ben Venue manufactured after implementing the changes
failed due to an Asp9-bivalirudin level of 2.3%. (PTX 218; Tr. 1399:23-1402:22 (Klibanov); Tr
207:9-2008:25 (Musso); Tr. 62:17-64:22 (Flammia).)
Mylan argues that the Court should afford this evidence little weight in the obviousness
context because at the time Ben Venue was working with bivalirudin, it was TMC that owned
the Angiomax® NDA and the compounding instructions from the master batch record used to
make the bivalirudin product required TMC’s approval. (Tr. 81:4-8 (Flammia); see also R.418
at 10, TMC’s Response to FF.18, stating “[a]dmit that TMC ‘has ultimate approval authority
concerning the content of, and modifications to, the [Master Batch Record]”.) The Court agrees
and does not place substantial weight on any lack of success that Ben Venue had in changing a
process that TMC had joint control over.
ii.

Copying of TMC’s product

Second, TMC argues that Mylan and other generic drug companies copied the Improved
Angiomax® drug product, with many stipulating to infringement. (See R.560 at 14-15.) In
some cases, evidence that the defendant or other competitors have copied a product embodying
the claimed invention can provide objective evidence of nonobviousness. See Wm. Wrigley Jr.

49

Co. v. Cadbury Adams USA LLC, 683 F.3d 1356, 1364 (Fed. Cir. 2012.) Courts, however, have
questioned whether copying can serve as evidence of nonobviousness in the ANDA context. See
Bayer Healthcare Pharms., Inc. v. Watson Pharms., Inc., 713 F.3d 1369, 1377 (Fed. Cir. 2013);
Purdue Pharma Prods. L.P. v. Par Pharm., Inc., 377 Fed App’x 978, 983 (Fed. Cir. 2010)
(unpublished). The Hatch-Waxman Act requires a generic drug manufacturer to demonstrate
that its generic formulation has the same active ingredients, route of administration, dosage form,
strength, and bioequivalency as an approved drug. See 35 U.S.C. § 355(j)(2)(A)(ii)-(iv). As the
Court has previously noted, the claim of copying has little force in the Hatch-Waxman context.
Bayer, 713 F.3d at 1377 (“copying in the ANDA context is not probative of nonobviousness
because a showing of bioequivalence is required for FDA approval”).
According to TMC, Mylan’s copying of Improved Angiomax® far exceeded its legal
obligations under the Hatch-Waxman Act by copying the inactive ingredients and impurity
profile of the Improved Angiomax® product. (See R.560 at 16; see also R.488 at 7; Tr.
1404:5-13.) Mylan responds that its use of inactive ingredients such as mannitol and sodium
hydroxide mirrors the use of the same excipients from the Original Angiomax® product and is
therefore nothing more than the use of prior art excipients. The evidence of copying presented
here – that other ANDA applicants have copied TMC’s formulation with some stipulating to
infringement – is afforded little weight by the Court because of the FDA’s bioequivalence
mandate for approval of a generic drug application.
iii.

Unexpected Results

Lastly, TMC argues that the inventors of the ‘727 patent were unexpectedly surprised
“that bivalirudin is stable to high shear mixing conditions.” (See R.560 at 16.) TMC
interweaves its teaching away argument with its alleged unexpected results that because the prior

50

art taught away from high shear mixing with peptides, it was surprising to find that high shear
mixing did not negatively impact bivalirudin. (See also Tr. 1402:23-1403:4.) The claims of the
‘727 patent are not limited to high speed mixing, and as such, this argument is unpersuasive.
The results that TMC obtained with the Improved Angiomax® product (less than 0.6%
Asp9-bivalirudin levels) are not unexpected results probative of nonobviousness. TMC argues
that the achieved lowering of Asp9-bivalirudin impurity levels were unexpected in light of the
fact that the person of ordinary skill in the art would have been discouraged from using high
speed mixing. This is nothing more than a reframing of TMC’s teaching away argument and as
an unexpected result indicative of nonobviousness is afforded little weight.
3.

Conclusion

After consideration of the parties’ arguments26 and the evidence presented at trial, the
Court finds that Mylan has failed to prove the asserted claims of the ‘727 patent invalid due to
obviousness. After a thorough examination of the evidence and analysis of the established
teachings of the prior art and the elements of the claimed invention, the Court finds there is no
motivation for a person of ordinary skill to combine the teachings of the prior art and arrive at
the claimed invention with a reasonable expectation of success. The Court therefore finds that
26

Mylan also argues that the changes TMC made to reduce the Asp9 maximum are nothing more
than a routine optimization of its Original Angiomax® product and process and the changes made to the
prior art compounding process are “few and routine.” (See R.553 at 19-21.) First, Mylan’s argument
ignores the fact that the claims of the ‘727 patent are not limited to a compounding process. Second,
Mylan makes conclusory attorney arguments that are undeveloped and unsupported by the evidence at
trial. In alleged support of its arguments, Mylan cites its expert’s testimony that provides no substantive
opinion about the differences between the changes, but rather notes only that such differences exist. (Tr.
823:15-16 (Auslander).) Even with the benefit of the evidence presented at trial, Mylan failed to establish
that TMC’s optimization of its prior art process and the resulting batches produced by the improved
process constitute routine experimentation. This is especially true where the prior art was not proven to
include the prior specification for Original Angiomax, nor the knowledge of the problems with the out of
specification Asp9-bivalirudin impurity levels. As such, Mylan has failed to prove by clear and
convincing evidence that the alleged differences between the prior art and improved compounding
processes, to achieve a reduced level of Asp9-bivalirudin impurities in “pharmaceutical batches” as
claimed in the ‘727 patent, were obvious as routine optimization.

51

Mylan has failed to prove obviousness of claims 1-3, 7-10, 27 and 17 of the ‘727 patent by clear
and convincing evidence.
C.

Enablement

Mylan also alleges that claims 1-3, 7-10, and 17 of the ‘727 patent are invalid for lack of
enablement under 35 U.S.C. § 112.28 (See R.553 at 25-29.)
1.

Legal Standard

Under 35 U.S.C. § 112, a patent specification must include
a written description of the invention, and of the manner and process of making
and using it, in such full, clear, concise, and exact terms as to enable any person
skilled in the art to which it pertains, or with which it is most nearly connected to
make and use the same, and shall set forth the best mode contemplated by the
inventor of carrying out his invention.
35 U.S.C. § 112, ¶ 1 (emphasis added). “To be enabling, the specification of a patent must teach
those skilled in the art how to make and use the full scope of the claimed invention without
undue experimentation.” MagSil Corp. v. Hitachi Global Storage Techs., Inc., 687 F.3d 1377,
1380 (Fed. Cir. 2012). Enablement is determined as of the effective filing date of the patent’s
application. See ALZA Corp. v. Andrx Pharms., LLC, 603 F.3d 935, 940 (Fed. Cir. 2010). The
Federal Circuit has provided the framework for the enablement analysis:
To prove that a claim is invalid for lack of enablement, a challenger must show by
clear and convincing evidence that a person of ordinary skill in the art would not
be able to practice the claimed invention without “undue experimentation.” . . .
27

Because the Court has found claim 1 valid and nonobvious, Mylan’s arguments regarding claim
7 are moot as this claim contains the novel limitations set forth in claim 1 from which it depends and is
therefore nonobvious for the reasons presented herein. This is true, even if the additional element
provided by claim 7, namely, the D-Phe12 limitation, is taught by the prior art and imparts no additional
novelty. For these reasons, Mylan has failed to meet its burden in proving claim 7 of the ‘727 patent
obvious.
28

Paragraph 1 of 35 U.S.C. § 112 was replaced with newly designated § 112(a) by § 4(c) of the
AIA, Pub. L. No. 112–29, and AIA § 4(e) makes those changes applicable “to any patent application that
is filed on or after” September 16, 2012. Because the application resulting in the ‘727 patent was filed
before that date, the Court will refer to the pre-AIA version of § 112. See Alcon, 745 F.3d 1180, 1183, n.
1 (Fed. Cir. 2014).

52

After the challenger has put forward evidence that some experimentation is
needed to practice the patented claim, the factors set forth in Wands then provide
the factual considerations that a court may consider when determining whether
the amount of that experimentation is either “undue” or sufficiently routine such
that an ordinarily skilled artisan would reasonably be expected to carry it out.
Alcon Research Ltd. v. Barr Labs, Inc., 745 F.3d 1180, 1188 (Fed. Cir. 2014) (citing In re
Wands, 858 F.2d 731, 736-37 (Fed. Cir. 1988).) The Court may consider the following factors
when determining if a disclosure requires undue experimentation:
(1) the quantity of experimentation necessary, (2) the amount of direction or
guidance presented, (3) the presence or absence of working examples, (4) the
nature of the invention, (5) the state of the prior art, (6) the relative skill of those
in the art, (7) the predictability or unpredictability of the art, and (8) the breadth of
the claims.
Wands, 858 F.2d at 737.
2.

Analysis

Mylan advances two arguments in support of its contention that the specification does not
teach those skilled in the art how to make the full scope of the claimed invention. (See R.553 at
25-28.) First, Mylan argues that the patent specification does not enable the asserted claims
because, as shown by failed Lot 1344985, even the co-inventors of the ‘727 patent, Dr. Musso
and Dr. Krishna, could not produce the claimed invention (i.e., the maximum 0.6% Asp9bivalirudin impurity level of claims 1, 7-10, and 17). Second, Mylan argues that even if asserted
claims 1, 7-10, and 17 are enabled, claims 2 and 3, which teach pharmaceutical batches having
maximum Asp9 impurity levels of 0.4% and 0.3%, respectively, are not enabled because some
Improved Angiomax® batches actually had Asp9 impurity levels greater than the allowed
maximums. The Court addresses each set of claims and arguments in turn.

53

a. Claims 1, 7-10, and 17
The Court finds that Mylan has failed to show by clear and convincing evidence that
claims 1, 7-10, and 17 are invalid due to lack of enablement. Mylan’s non-enablement argument
with respect to these claims hinges upon Mylan establishing that Lot 1344985 was made
according to a compounding process and without experiencing any operator error or
investigation concerns. Lot 1344985 was one of two failed batches (the other being Lot
1116050) that Ben Venue manufactured purportedly using the Improved Angiomax®
compounding process described in Example 5 of the ‘727 specification. Lot 1116050 had an
Asp9 impurity level of 12.4%, and Lot 1344985 had an Asp9 impurity level of 1.5%. The
inventors disclosed the Asp9 impurity results for Lot 1116050 to the PTO, explaining that the
data presented regarding Improved Angiomax® batches did not include the results for Lot
1116050 because “the method used to generate the batch was not compliant with the protocol
established for [the] study.” (See PTX 1, col. 23:14-16.) The inventors, however, did not
disclose the Asp9 impurity results of Lot 1344985 to the PTO or explain why they had omitted
those results from the data regarding Improved Angiomax® batches.
Mylan essentially argues inoperability of the ‘727 patent claims, asserting that the 1.5%
Asp9 impurity level of Lot 1344985 establishes not only that the patentees were not able to make
the claimed invention, but that the ‘727 patent fails to teach those skilled in the art to produce the
claimed invention (i.e., pharmaceutical batches of bivalirudin with a maximum Asp9 impurity
level of about 0.6%). Mylan, however, failed to prove that Ben Venue did not properly follow
the compounding process in manufacturing Lot 1344985.
Mylan relies heavily on the fact that Ben Venue’s investigation into the high Asp9
impurity level in Lot 1344985 found that no operator error had occurred during the compounding

54

process. The evidence at trial, however, does not establish this fact under a clear and convincing
standard. Indeed, the evidence showed persuasive reasons to question the results of Ben Venue’s
investigation. This evidence substantially decreases the weight of Ben Venue’s investigation.
First, around the time period that Ben Venue manufactured Lot 1344985 in May 2008,
the FDA had issued a warning letter to Ben Venue for, among other reasons, not closing its
quality investigations timely and not complying with scientific standards in performing those
investigations. The FDA issued the warning letter to Ben Venue in May 2007, and, in June
2008, Ben Venue recognized its need to “implement[] . . . improved systems” in a
Comprehensive Corrective Action Plan. (See DTX 1407.)
Second, Ben Venue had failed to follow the manufacturing instructions for another failed
batch, Lot 116050, prior to manufacturing Lot 1344985. Ben Venue’s initial investigation into
Lot 116050—like its initial investigation into Lot 1344985—found that the high Asp9 impurity
level was not the result of operator error in the compounding process. Only after Ben Venue
reopened its investigation into Lot 116050 at TMC’s urging did Ben Venue identify operator
error as responsible for the batch failure. Lot 116050 is the only Improved Angiomax® batch
other than Lot 1344985 to have Asp9 impurity levels above about 0.6%. Given all the evidence,
Ben Venue’s failure to follow the manufacturing process with respect to Lot 116050 suggests
that Ben Venue also may have failed to follow the manufacturing process with respect to Lot
1344985. Ben Venue’s failure to identify operator error in its initial investigation into Lot
116050 gives further reason to question the similar finding of no operator error in Ben Venue’s
initial investigation into Lot 1344985. Ultimately, Ben Venue agreed to destroy Lot 1344985
and issue a credit to TMC’s account for the cost of manufacturing Lot 1344985, although it is
unclear if Ben Venue agreed to this resolution because it determined that the batch failure was

55

the result of operator error or for other reasons, such as maintaining a good business relationship
with TMC.
Third, the evidence at trial showed an unidentified peak in the chromatogram for Lot
1344985 similar to an unidentified peak in the chromatogram for Lot 1116050. Lot 1344985 and
Lot 1116050 are the only Improved Angiomax® batches that showed this unidentified peak, and
they are the only batches that had Asp9 impurity levels greater than about 0.6%. This evidence
further suggests that Ben Venue failed to follow the manufacturing process for Lot 1344985 as it
had for Lot 1116050.
Mylan failed to submit sufficient evidence to support an affirmative finding that Lot
1344985 was made by the improved compounding process and did not encounter operator error
or some other issue related to Ben Venue’s incompliance with regard to the timeliness and
standards of its quality investigations (see DTX 1407, Ben Venue’s Comprehensive Corrective
Action Plan related to receipt of FDA warning letter). At best, the evidence at trial raised
questions regarding whether Ben Venue followed the Improved Angiomax® compounding
process in manufacturing Lot 1344985. Further, Mylan failed to present sufficient evidence to
meet its clear and convincing burden on this issue. Put simply, Mylan failed to prove by clear
and convincing evidence that the Improved Angiomax® process, rather than another reason, e.g.,
operator error, caused the high Asp9 impurity level in Lot 1344985.
Even if the Court had found that Mylan established that the invention of claims 1, 7-10,
and 17 was inoperable based on the high Asp9 impurity level in Lot 1344985, that showing alone
would not establish invalidity due to lack of enablement. Mylan still needed to show that the
person of ordinary skill would not be able to practice the claimed invention without “undue
experimentation.” See Alcon, 745 F.3d at 1190-91 (reversing the district court’s finding of

56

invalidity for lack of enablement because defendant failed to provide evidence that established
the person of ordinary skill would need to engage in “undue experimentation” in order to
practice the asserted claims). Mylan has the burden to establish that TMC’s patent lacked
enabling disclosures, yet it “failed to make the threshold showing that any experimentation is
necessary to practice the claimed methods.” See id. at 1189. Putting aside the fact that Mylan’s
briefing was silent as to the legal requirement under Wands for “undue experimentation”, Mylan
failed to present evidence at trial that would satisfy a requirement for any experimentation with
bivalirudin, let alone undue experimentation under the clear and convincing standard.
Indeed, Mylan’s own expert, Dr. Auslander, admitted the focus of all his testimony was
deamidation, rather than other possible reactions (e.g., hydrolysis, racemization, oxidation,
aggregation, precipitation) that can occur with aspartate peptides, such as bivalirudin. (Tr.
941:2-13.) Dr. Auslander further testified on cross examination that in focusing on deamidation,
he considered the effects of pH and oxidation, but did not consider the effects of temperature,
ionic strength, and excipients (e.g., mannitol), and was not able to provide conclusive answers as
to what those effects may be other than that they “may” have one. (Tr. 956:17-966:20.) In
Alcon, the defendant’s expert testified that “[v]arious parameters including pH, buffer, buffer
concentration, preservatives, chelating agents, and other excipients may affect the chemical
stability of prostaglandins in ophthalmic formulations.” 745 F.3d at 1189 (citations omitted)
(emphasis original). The Federal Circuit found the “unsubstantiated conclusory statement”
insufficient to establish evidence of undue experimentation. Id. In particular, the Federal Circuit
reasoned that while adjusting variables may be relevant to optimizing the claimed invention, the
defendants offered no evidence that any experimentation, let alone undue experimentation, with
those variables would be necessary in order to practice the claimed invention. See id. Similarly,

57

Mylan provided no evidence that any experimentation, let alone undue experimentation, would
be required with variables its own expert admitted could have an effect on asparate peptides.
Therefore, Mylan failed to show that the embodiments of claims 1, 7-10 and 17 would be
inoperable and failed to provide evidence that a person of ordinary skill in the art would have
been unable to practice the asserted claims without resorting to any experimentation, let alone
undue experimentation. See id. at 1191. Mylan had the burden of proof to show that TMC’s
patents lacked enabling disclosures, but failed to carry that burden. As such, claims 1, 7-10, and
17 are not invalid based on nonenablement.
b. Claims 2 and 3
Mylan also failed to satisfy its burden on its non-enablement counterclaim regarding
claims 2 and 3, which teach pharmaceutical batches having a maximum Asp9 impurity level of
0.4% and 0.3%, respectively. Mylan argues that the ‘727 patent specification fails to disclose
any method that even purports to produce batches of bivalirudin having Asp9 impurity levels not
exceeding “about 0.4%” or “about 0.3%.” The only arguably enabling disclosure, Mylan notes,
is Example 5 which reports a maximum Asp9 impurity level of 0.6% (not 0.4% or 0.3%). TMC
responds that “Mylan ignores Example 3 of the ‘727 patent which discloses different processes
that resulted in batches having maximum values of Asp9 after sodium hydroxide addition of
0.34% for one experiment (PTX 1, col. 21, Table 5a) and 0.40% (Table 5b) in another
experiment.” (R.560 at 21.)
The Court, however, does not need to resolve this dispute because even if the Court
found that Mylan proved that the specification did not teach the full scope of the claimed
invention, the inquiry does not end there. Mylan must also establish that the person of ordinary
skill would not be able to practice the claimed invention without “undue experimentation.” See

58

Alcon, 745 F.3d at 1190-91; see also Johns Hopkins Univ. V. CellPro, Inc., 152 F.3d 1342, 1360
(Fed. Cir. 1998) (“[I]t is imperative when attempting to prove lack of enablement to show that
one of ordinary skill in the art would be unable to [practice] the claimed invention without undue
experimentation”). As addressed above, Mylan has the burden of proof to show that TMC’s
patent lacked enabling disclosures and Mylan has “failed to make the threshold showing that any
experimentation is necessary to practice the claimed methods.” See Alcon, 745 F.3d at 1189.
Because the evidence is insufficient under the clear and convincing standard to satisfy a
requirement of “undue experimentation”, the Court concludes that claims 2 and 3 are not invalid
based on nonenablement.
III.

Unenforceability for Inequitable Conduct
Mylan alleges that the ‘727 patent is unenforceable due to inequitable conduct. (See

R.553 at 29-34.)
A.

Legal Standard

“Inequitable conduct is an equitable defense to patent infringement that, if proved, bars
enforcement of a patent.” Therasense, Inc. v. Becton, Dickinson & Co., 649 F.3d 1276, 1285
(Fed. Cir. 2011) (en banc). To prevail on an inequitable conduct defense, the accused infringer
must show by clear and convincing evidence (1) that “the applicant misrepresented or omitted
material information” on patentability and (2) the applicant had “the specific intent to deceive the
PTO.” Id. at 1287; see also American Calcar, Inc. v. American Honda Motor Co., Inc., No.
2013-1061, __ F.3d __, 2014 WL 4783613, at *3 (Fed. Cir. Sept. 26, 2014) (quoting Ohio
Willow Wood Co. v. Alps S., LLC, 735 F.3d 1333, 1344 (Fed. Cir. 2013)). Materiality and intent
must be separately established. See Therasense, 649 F.3d at 1290; see also In re Rosuvastatin
Calcium Patent Litig., 703 F.3d 511, 519 (2012).

59

A patent challenger can establish materiality by showing that the PTO would not have
issued the claim but for the applicant’s misrepresentation or omission. Therasense, 649 F.3d at
1291; see also American Calcar, 2014 WL 4783613, at *3. Because the Court assesses but-for
materiality from the PTO’s perspective, the Court applies the preponderance of the evidence
standard in assessing whether the withheld or allegedly misrepresented information would have
blocked patentability. Therasense, at 1291-92. Absent but-for materiality, a patent challenger
also can prove presumptive materiality by showing that TMC “has engaged in affirmative acts of
egregious misconduct, such as the filing of an unmistakably false affidavit.” Id. at 1292; see also
Ohio Willow Wood, 735 F.3d at 1345.
A patent challenger must also prove that the patentee acted with the specific intent to
deceive the PTO. Therasense, 649 F.3d at 1290. Because it is rare to find direct evidence of
intent to deceive, courts may “infer [deceptive] intent from indirect and circumstantial evidence.”
Id; see also American Calcar, 2014 WL 4783613, at *5. To meet the clear and convincing
standard, however, the specific intent to deceive must be “the single most reasonable inference
able to be drawn from the evidence.” Therasense, 649 F.3d at 1290 (quoting Star Scientific, Inc.
v. R.J. Reynolds Tobacco Co., 537 F.3d 1357, 1366 (Fed. Cir. 2008)); see also Therasense, 649
F.3d at 1290 (emphasis added) (quoting Kingsdown Med. Consultants, Ltd. v. Hollister Inc., 863
F.2d 867, 873 (Fed. Cir. 1988)) (explaining that to prove inequitable conduct, the circumstantial
evidence “‘must be sufficient to require a finding of deceitful intent in light of all the
circumstances.’”).
B.

Analysis

Mylan contends Drs. Musso and Krishna engaged in inequitable conduct by (1) failing to
disclose and misrepresenting to the PTO the Asp9 impurity level in Lot 1344985,

60

(2) misrepresenting the number of Original Angiomax® batches that contained an Asp9 impurity
level greater than about 0.6% in Table 6 of the ‘727 patent, (3) failing to fully disclose the
process used to produce Original Angiomax® batches, and (4) failing to disclose the
contributions of Ben Venue employees as inventors. (R.553 at 29.) The Court addresses each
argument in turn.
1.

Failure to Disclose the Asp9 Impurity Level of Lot 1344985

Mylan first argues that the PTO would not have issued the ‘727 patent due to
non-enablement if the inventors had disclosed the Asp9 impurity level in Lot 1344985. As
explained above, Mylan failed to prove its non-enablement argument based on Lot 1344985 by
clear and convincing evidence. Mylan also failed to prove its inequitable conduct argument
based on Lot 1344985 because Mylan failed to establish that the inventors intended to deceive
the PTO by not disclosing the Asp9 impurity level for Lot 1344985.
Both Dr. Musso and Dr. Krishna credibly testified at trial regarding their reasons for not
disclosing the results of Lot 1344985. Dr. Musso testified that he did not believe Ben Venue had
manufactured Lot 1344985 according to the Improved Angiomax® process, and, therefore, did
not believe he had a duty to disclose it to the PTO. (Tr. 507:9-18.) Dr. Musso based his belief
on a number of factors. First, he took into account the ongoing FDA investigations at Ben
Venue. (See id. at 389:19-20, 393:9-14, 412:2-9.) Dr. Musso knew of Ben Venue’s past
manufacturing problems, including the operator error identified in manufacturing Lot 1116050.
(See id. at 423:20-25.) Second, Dr. Musso knew that Ben Venue’s and TMC’s investigation into
Lot 1344985 was ongoing during the ‘727 patent’s prosecution. (See id. at 435:23-436:10.) It
was Dr. Musso’s belief that because Lot 1344985 was still under investigation and he believed
the lot failure was due to operator error, he did not need to disclose Lot 1344985 to the PTO.

61

(See id. at 413:14-20.) Third, Dr. Musso expressed his distrust of the results of Ben Venue’s
internal investigations into failed lots. (See id. at 426:23-427:4.) Based on his experience, he did
not find Ben Venue’s preliminary “no root cause” finding regarding failure of Lot 1344985
reliable. (Id.) Dr. Musso testified that, in light of these considerations, he “never considered”
whether he should disclose Lot 1344985 to the PTO. (See id. at 436:18-21.)
Dr. Krishna gave similar testimony regarding Lot 1344985. Like Dr. Musso, Dr. Krishna
did not believe he had a duty to disclose the Asp9 level of Lot 1344985 to the PTO. (Tr. 1206:59, 1206:24-1207:7.) Dr. Krishna testified that he based his determination on the fact that, from
TMC’s standpoint, the investigation into Lot 1344985 was ongoing. (See id. at 1183:18-22,
1201:14-19.) Furthermore, Dr. Krishna believed that Ben Venue had not followed the Improved
Angiomax® process in manufacturing Lot 1344985. (See id. at 1187:6-11, 1220:22-1221:3.)
Dr. Krishna believed that had Ben Venue followed the Improved Angiomax® process, Lot
1344985 would have had an Asp9 level below about 0.6%. (See id. at 1220:22-1221:3.) For
these reasons, Dr. Krishna believed that the results for Lot 1344985 were not relevant to the
claimed subject matter. (See id. at 1206:5-9, 1206:24-1207:7.) Having observed Dr. Musso’s
and Dr. Krishna’s demeanor on the stand and ability to withstand cross examination as well as
the evidence surrounding Ben Venue, the Court finds the co-inventors’ testimony to be credible.
Mylan called Dr. Linck, its expert in patent prosecution and PTO procedures, to testify on
the issue of inequitable conduct and Lot 1344985. Dr. Linck noted that Dr. Musso became
aware of the 1.5% Asp9 level in Lot 1344985 prior to signing a petition to the PTO claiming a
bivalirudin drug product with a “maximum impurity level of Asp9-bivalirudin not exceeding
about 0.6%.” (Tr. 1306:18-23.) Dr. Linck testified that she “saw no evidence supporting the
allegations that Lot 1344985 was not made by the so-called efficient process.” (Id. at 1314:14-

62

16.) Dr. Linck also questioned the inventors’ stated reasons for not disclosing Lot 1344985
when they had disclosed Lot 1116050. (Id. at 1318:20-25.)
While the inventors’ disclosure of failed Lot 1116050 due to operator error raises
questions regarding why they did not also disclose failed Lot 1344985 even if they believed the
failure was due to operator error, the evidence at trial did not establish that intent to deceive the
PTO is “the single most reasonable inference” the Court can draw. See Therasense, 649 F.3d at
1290. The credibility of Dr. Musso’s and Dr. Krishna’s testimony at trial and the evidence
corroborating their testimony regarding problems with Ben Venue’s manufacturing process and
investigations into failed batches weighs against a finding of intent to deceive the PTO. In sum,
Mylan failed to present evidence “sufficient to require a finding of deceitful intent in light of all
the circumstances. See Therasense, 649 F.3d at 1290 (emphasis in original). Mylan’s
inequitable conduct theory based on Lot 1344985 therefore fails.
2.

Misrepresentation of the Data in Table 6

Mylan next argues that the inventors misrepresented the number of Original Angiomax®
batches that contained an Asp9 impurity level greater than about 0.6% in Table 6 of the ‘727
patent. Table 6 discloses the mean, standard deviation, and maximum Asp9 impurity level for
the 87 Original Angiomax® batches analyzed:

63

(PTX 1, col. 22:10-20.) Mylan’s statistical expert, Dr. Ian McKeague, however, opined that the
Table 6 is misleading because applying a normal distribution to the data suggests that only 60%
of Original Angiomax® batches had Asp9 impurity levels of less than 0.6% (see Tr. 647:7-649:2)
when, in actuality, 87% of Original Angiomax® batches had Asp9 impurity levels below 0.6%.
Mylan argues that if the inventors had disclosed to the PTO that nearly 90% of Original
Angiomax® batches met the ‘727 claim limitation regarding Asp9 impurity levels, the PTO
would not have issued the ‘727 patent.
Mylan’s argument is unpersuasive. First, Dr. McKeague’s assumption that a normal
distribution applied to the data in Table 6 is highly questionable. TMC’s statistics expert, Dr.
Alan Salzberg, calculated that given that the mean value (0.5%) is close to zero and the
maximum value (3.6%) is “nearly eight standard deviations away from the mean,” (see Tr.
1271:22-1272:8 (Dr. Salzberg testimony)), the chance that the data followed a normal
distribution is less than 1 in 2.5 trillion. (See id. at 1272:19-1273:20.) Accordingly, Dr. Salzberg
opined that Dr. McKeague’s assumption of a normal distribution was inappropriate. (See id. at
1273:21-24.) Although Dr. McKeague challenged Dr. Salzberg’s 1 in 2.5 trillion calculation, Dr.
McKeague did not provide an alternative calculation to show that the data supported his
assumption of a normal distribution. Instead, he appears to have based his assumption on the
form in which Table 6 presented the data, rather than the data itself. (See id. at 647:17-25
(assuming a normal distribution because (1) Table 6 reported the mean and standard deviation,
which characterize a normal distribution, and (2) the “±” sign before the standard deviation
indicates a symmetrical distribution).)
Furthermore, while Dr. McKeague acknowledged during cross-examination that
alternatives to a normal distribution for the data in Table 6 exist, he offered no persuasive reason

64

why he failed to analyze whether one of those alternatives may better fit the data than a normal
distribution. (See id. at 721:16-724:3.) Dr. McKeague admitted that given the mean and
standard deviation, a normal distribution is “not a perfect choice,” but rather “a quick back-of- –
back-of-the-envelope choice.” (See id. at 704:13-705:2.) The Court puts more weight on Dr.
Salzberg’s testimony than Dr. McKeague’s based both on the substance of his testimony and his
demeanor on the stand. Mylan, in sum, failed to prove that the inventors withheld or
misrepresented information regarding the number of Original Angiomax® batches that had an
Asp9 impurity level below about 0.6% that, if accurately disclosed, would have led the PTO to
reject the asserted claims. See Ohio Willow Wood, 735 F.3d at 1345.
Mylan also failed to prove by clear and convincing evidence that the inventors intended
to deceive the PTO about the Asp9 impurity levels of Original Angiomax® batches. See
Therasense, 649 F.3d at 1290. Dr. Musso and Dr. Krishna credibly testified that they believed
the data in Table 6 fully disclosed the characteristics of the 87 Original Angiomax® batches
analyzed, and they had never even considered whether to disclose the Asp9 impurity levels of all
87 batches. (See, e.g., Tr. 329:15-330:20; 435:9-22; 1243:18-1244:19.) This testimony was
unrebutted at trial. Mylan, therefore, failed to prove either prong of its inequitable conduct
theory based on the data in Table 6.
3.

Failure to Disclose the Original Angiomax® Compounding Process

Mylan argues that the inventors deceived the PTO by failing to disclose how similar the
Original Angiomax® compounding process described in Example 4 of the patent specification
was to the Improved Angiomax® compounding process described in Example 5. Specifically,
Mylan argues that the inventors should have disclosed that (1) the Original Angiomax®
compounding process involved initially operating the mixers at a rate between 400 and 800

65

RPMs and then increasing the rate to 900 and 1300 RPMs upon completion of the base addition
step, and (2) in making Original Angiomax®, the base solution “was sometimes added by a third
method—via a nitrogen pressurized hose.” (See Mylan Post-Trial Br. at 33.) Mylan’s arguments
are unpersuasive.
First, Mylan failed to prove that the increase in the mixing rate after the completion of the
base-addition step was material to the asserted claims and the ‘727 patent would not have issued
if the inventors had disclosed it. See Ohio Willow Wood, 735 F.3d at 1345. The comparison of
the Example 4 and Example 5 processes in the patent specification focused on the rate of mixing
during the base-addition step—i.e., the time at which the precipitate formed—not the mixing rate
after addition of the base solution. (Compare PTX 1, col. 21:59-64 with id., col. 22:47-52.) The
patent specification accurately disclosed the mixing rates used during the base-addition steps in
the two processes. (See Tr. 316:2-18.) Furthermore, the patent specification also omitted the
fact that the mixing rate increased after addition of the base solution in the Example 5 process,
like in the Example 4 process. (See id. at 318:9-14 (“[A]lso in Example 5, the increase in speed
after base addition is also increased and not described.”).)
Second, Mylan failed to prove that the fact that, in the original compounding process, the
base solution sometimes was added via a pressurized hose was material to the asserted claims.
The comparison of the Example 4 and Example 5 processes focused on the rate at which the base
solution was added, not the mechanism for adding the base. In Example 4, the base solution was
added “either all at once, or rapidly in multiple portions,” whereas in Example 5, the base
solution was added “at a controlled rate of 2 L/min.” (Compare PTX 1, col. 21:60 with id., col.
22:49.) The evidence at trial confirmed that the rate at which the base solution was added in the
Example 4 process—even when a nitrogen pressurized hose was used—was significantly more

66

rapid than the rate at which the base solution was added in the Example 5 process. Adding the
solution through the nitrogen pressurized hose in the Example 4 process took a period of “much
less than five minutes” (see Tr. 315:3-18; see also PTX 223), whereas, in the Example 5 process,
the base solution was added over a twenty-minute period. (See Tr. 750:22-751:7.) Mylan,
therefore, failed to prove that the fact that the Example 4 process sometimes involved the use of
a pressurized nitrogen hose was material to the asserted claims and the ‘727 patent would not
have issued if the inventors had disclosed it. See Ohio Willow Wood, 735 F.3d at 1345.
Third, Mylan failed to prove that the inventors intended to deceive the PTO about the
similarities between the Example 4 and Example 5 processes. As discussed, to meet the clear
and convincing standard for proving deceptive intent, “the specific intent to deceive must be ‘the
single most reasonable inference able to be drawn from the evidence.’” Therasense, 649 F.3d at
1290. That is not the case here. Dr. Musso credibly testified that he believed the patent
specification disclosed the important aspects of the Example 4 and Example 5 processes to the
PTO. (See Tr. 319:10-320:11.) Dr. Musso explained that the patent specification described the
mixing rate during the base-addition step, rather than following the base-addition step, because
the base-addition step is the important point in the process in which the marshmallow-like
precipitate formed. (See Tr. 318:9-319:4.) He further explained that adding the base solution
through a pressurized nitrogen hose came within the umbrella of adding the solution “all at
once,” as disclosed in the description of the Example 4 process. (See, e.g., Tr. 314:2-315:24.)
This evidence supports an inference that the inventors omitted the aspects of the Original
Angiomax® compounding process at issue because they did not believe the information to be
material to the asserted claims, not because they intended to deceive the PTO. Accordingly,

67

Mylan failed to prove the specific intent element of its inequitable conduct theory. See
Therasense, 649 F.3d at 1290.
4.

Failure to Disclose Ben Venue Employees’ Contributions as Inventors

Finally, Mylan argues that the inventors failed to disclose the contributions of Ben
Venue’s operators as co-inventors. In patent law, the “inventor” is the person or persons who
conceived of the invention. See Ohio Willow Wood, 735 F.3d at 1350. Conception is the
“formation in the mind of the inventor, of a definite and permanent idea of the complete and
operative invention, as it is hereafter to be applied in practice.” Invitrogen Corp. v. Clontech
Labs., Inc., 429 F.3d 1052, 1063 (Fed. Cir. 2005). In the case of co-inventors, “each joint
inventor must generally contribute to the conception of the invention.” Ethicon, Inc. v. U.S.
Surgical Corp., 135 F.3d 1456, 1460 (Fed. Cir. 1998). “One who simply provides the inventor
with well-known principles or explains the state of the art without ever having ‘a firm and
definite idea’ of the claimed combination as a whole does not qualify as a joint inventor.” Id.
Inventorship can be material to patentability. See Leviton Mfg., Co. v. Universal Sec.
Instruments, Inc., 606 F.3d 1353, 1360 (Fed. Cir. 2010).
During the July 14, 2006 meeting attended by Dr. Musso and several TMC and Ben
Venue employees, Ben Venue operators familiar with the Original Angiomax® compounding
process joined the meeting and engaged in some discussions regarding how to minimize the
formation of Asp9 impurities during the process. Specifically, Mylan argues that the operators
suggested: (1) varying the stirrer heights so that one stirrer mixes the upper part of the solution
and one mixes the lower part; (2) increasing the mixing speed prior to base addition; and (3)
adding the base solution through a dip tube so that it dispersed at the bottom of the tank where
mixing is still possible. (See PTX 7.) Mylan contends that Dr. Musso incorporated these

68

concepts without attribution into the Improved Angiomax® compounding process described in
Example 5, and, in doing so, materially misled the PTO regarding the inventorship of the patent.
(See R.553, Mylan Post-Trial Br. at 33-34.) The Court disagrees.
As discussed above (see Part II.B.2.a.ii.b)), Mylan established that some portions of the
report, authored by Dr. Musso, are attributable to Ben Venue employees. Employees of both
TMC and Ben Venue, along with the inventor Dr. Musso, attended the July 14, 2006 meeting.
(PTX 7; DTX 1245; Tr. 67:19-68:18 (Flammia); Tr. 200:6-202:19 (Musso); Smith Dep.
71:06-71:18, June 10, 2014.) The evidence Mylan presented did not support a finding that the
information in the Musso Trip Summary in its entirety came from Ben Venue operators. Indeed,
the evidence revealed that at some point later in the meeting when the participants were still
discussing the issues, Ben Venue operators were no longer present. (Tr. 68:9-10 (Flammia); Tr.
264:10 (Musso).) In addition, attendees of the meeting who testified at trial, namely Timothy
Smith – Ben Venue employee – and Angie Green – TMC employee, did not recall who
specifically provided the suggestions and more importantly, the respective roles of inventor
Musso and TMC in the discussion and in making the suggestions. Dr. Musso also testified that
the cause of the high Asp9 failure was not identified at the meeting and that he and Dr. Krishna
“came up with the ideas on how to fix and make the improved product. Ben Venue did not
contribute to those inventions.” (Tr. 201:19-21; 205:19-21.) Even treating the suggestions and
insights from the Ben Venue operators as prior art, the Court already found the claims valid and
therefore, the “but-for” materiality of the operators’ disclosures to Dr. Musso is not established
here. Mylan provided no additional evidence that despite this lack of “but-for” materiality, TMC
“engaged in affirmative acts of egregious misconduct.” See Therasense, 649 F.3d at 1291-92
(assessment of but-for materiality in an inequitable conduct defense is assessed by a

69

preponderance of the evidence standard to determine whether the withheld or allegedly
misrepresented information would have blocked patentability); see also Ohio Willow Wood, 735
F.3d at 1345.
Furthermore, Mylan failed to establish that TMC and Musso intended to deceive the PTO
about the contents of the Musso Trip Summary from the July 14, 2006 meeting. Dr. Musso
credibly testified that it was he and Dr. Krishna who “came up with the ideas on how to fix and
make the improved product,” not Ben Venue. (See Tr. 201:19-21; 205:19-21.) Mylan’s
allegations of TMC’s pattern of withholding information to establish specific intent fails. (See
R.553 at 34-35.) The Court does not find that Dr. Musso’s intent to deceive the PTO is the
single most reasonable inference drawn from the evidence. See Therasense, 649 F.3d at 1290.
Instead, the evidence at the very least supports a reasonable inference that Dr. Musso worked
collaboratively with a team of scientists at TMC and Ben Venue on the production of bivalirudin
batches and that he and Dr. Krishna identified the underlying problems responsible for
generation of the high and variable Asp9 levels, and discovered a solution to that problem.
Hence, Mylan failed to establish by clear and convincing evidence that the inventors of
the ‘727 patent (1) misrepresented or omitted material information on patentability, and (2) had a
specific intent to deceive the PTO.
IV.

Infringement
A. Legal Standard
A determination of infringement requires a two-step analysis. First, the asserted claims

are interpreted by the courts to determine their scope and meaning. See Presidio Components,
Inc. v. American Tech. Ceramics Corp., 702 F.3d 1351, 1358 (Fed. Cir. 2012). Second, the
fact-finder compares the properly construed claims to the allegedly infringing product. See id.

70

To establish infringement, TMC must prove by a preponderance of the evidence that Mylan’s
proposed drug product contains every limitation in the asserted claims. See Trebro Mfg., Inc. v.
Firefly Equip., LLC, 748 F.3d 1159, 1166 (Fed. Cir. 2014); Meyer Intellectual Props. Ltd. v.
Bodum, Inc., 690 F.3d 1354, 1370 (Fed. Cir. 2012).
For claims of infringement brought under 35 U.S.C. § 271(e)(2), the infringement inquiry
is whether the proposed commercial ANDA product will infringe if manufactured or sold in the
United States, i.e., whether it will contain each and every limitation of the asserted patent claims.
Bayer AG v. Elan Pharm. Research Corp., 212 F.3d 1241, 1249 (Fed. Cir. 2000). Because the
ANDA may not yet be approved and a product may not yet be marketed, this inquiry is
necessarily a hypothetical one focusing on “what the ANDA applicant will likely market if its
application is approved, an act that has not yet occurred.” Bayer AG, 212 F.3d at 1248 (quoting
Glaxo, Inc. v. Novopharm, Ltd., 110 F.3d 1562, 1569 (Fed. Cir. 1997)). “[T]his hypothetical
inquiry is properly grounded in the ANDA application and the extensive materials typically
submitted in its support.” Id. (quoting Glaxo, 110 F.3d at 1569). “[I]f a product that an ANDA
applicant is asking the FDA to approve for sale falls within the scope of an issued patent, a
judgment of infringement must necessarily ensue. Sunovion Pharms., Inc. v. Teva Pharms. USA,
Inc., 731 F.3d 1271, 1278 (Fed. Cir. 2013) (citing Abbott Labs. v. TorPharm, Inc., 300 F.3d
1367, 1373 (Fed. Cir. 2002)). In other words, “[w]hat a generic applicant asks for and receives
approval to market, if within the scope of a valid claim, is an infringement.” Sunovion, 731 F.3d
at 1279.

71

B. Analysis
TMC claims that Mylan’s proposed drug product infringes claims 1-3, 7-10, and 17 of the
‘727 patent under 35 U.S.C. § 271(e)(2)(A).29 (See R.555, TMC’s Op. Post-Trial Br.) The
parties do not dispute that Mylan’s proposed drug meets many limitations of claim 1—i.e., that it
constitutes a drug product comprising bivalirudin and a pharmaceutically acceptable carrier, it is
indicated for use as an anticoagulant, and it has a pH adjusted by a base and said pH is about 5-6
when reconstituted in an aqueous solution for injection. (See Tr. 553:8-561:14 (Dr. Klibanov
testimony and exhibits cited therein); R.558, Mylan’s Post Tr. Reply Br. (contesting only the two
disputed terms recited above); see also e.g.,PTX 133.203-204; PTX 133.42-44, PTX 133.47.)
The parties also do not dispute that Mylan’s proposed drug meets the additional limitations in
dependent claims 7-10 and 17—i.e., that it has a maximum level of D-Phe12-bivalirudin that does
not exceed about 2.5% as measured by HPLC (claim 7); that the pharmaceutically acceptable
carrier in the drug is a bulking agent, more specifically, the sugar mannitol (claims 8-10); and
that the base used to adjust the drug’s pH level is sodium hydroxide (claim 17). (See Tr. 566:1568:12 (Dr. Klibanov testimony and exhibits cited therein); R.558 at 8-18 (contesting only the
two disputed terms recited above) see also e.g., PTX 133.203-204; PTX 133.42-44, PTX
133.47.) The dispute regarding infringement of the asserted claims centers around two claim
limitations “pharmaceutical batches” and “wherein the batches have a maximum impurity level
of Asp9-bivalirudin of about 0.6%.”30 The Court finds that Mylan’s proposed ANDA product
infringes asserted claims 1-3, 7-10, and 17 of the ‘727 patent for the reasons stated below.
29

TMC did not advance arguments at trial regarding infringement under the doctrine of
equivalents, therefore, TMC’s infringement allegations for the asserted claims are treated as allegations of
literal infringement. (See R.309 at 22-23; R.366; R.485.)
9

30

For asserted claims 2 and 3, which depend from claim 1, the maximum impurity level of Asp bivalirudin is about 0.4% and 0.2%, respectively.

72

TMC’s primary argument for infringement is that Mylan’s proposed drug product
infringes the asserted claims as a matter of law under the Federal Circuit’s recent opinion in
Sunovion. (R.555 at 11-14.) To the extent Sunovion is not dispositive – TMC argues that
Mylan’s representations in its ANDA and its ANDA exhibit batch demonstrate that its proposed
drug product will infringe the asserted claims. (Id. at 14-30.) The Court agrees that Sunovion is
dispositive on the infringement issue and, under Sunovion, Mylan’s proposed ANDA drug
product infringes the asserted claims as a matter of law.
1. The Federal Circuit’s Decision in Sunovion
In Sunovion, another Hatch-Waxman case, the plaintiff owned the rights to a patent
directed to a pharmaceutical composition containing the drug eszopiclone, the active ingredient
in Lunesta®. The patent claimed a pharmaceutical composition containing eszopiclone
“essentially free of its levorotatory isomer,” which the district court construed as meaning “less
than 0.25% of [the] levorotatory isomer.” See Sunovion, 731 F.3d at 1273, 1274 (alteration in
original). Defendant submitted its ANDA seeking FDA approval to manufacture, use, and sell a
generic version of Lunesta® before the expiration of the patent-in-suit. See id. The defendant
originally sought approval for levorotatory isomer levels “[n]ot less than 0.3% and [n]ot more
than 1.0%,” but after the FDA required the defendant to “tighten” the proposed levorotatory
isomer levels, defendant amended the levorotatory isomer levels in its ANDA to “not more than
0.6% (i.e., 0.0-0.6%)”. See id. at 1274-75. The district court granted the defendant’s motion for
summary judgment of noninfringement despite the fact that the amended ANDA specification
allowed the proposed drug product to have levorotatory isomer levels within the claimed range
of 0.0 – 0.25%. Id. at 1275. The district court relied on the defendant’s internal manufacturing
guidelines that the levels “would likely be ‘outside the infringing range of less than 0.25% …’”

73

Id. The district court also relied on a “certification” submitted as a declaration from the
defendant’s employee vowing to only market eszopiclone with levorotatory isomer levels above
the claimed range, namely between 0.3-0.6%. Id.
The Federal Circuit reversed the district court’s grant of summary judgment of
noninfringement and instead found defendant’s ANDA infringed as a matter of law. Id. at 1280.
Because the defendant’s ANDA allowed the defendant to market a drug product within the
patent-in-suit’s claimed levorotatory isomer range, the Federal Circuit held that the proposed
drug product infringed as a matter of law. See id. at 1280. The Federal Circuit instructed that
“[w]hat a generic applicant asks for and receives approval to market, if within the scope of a
valid claim, is an infringement.” Id. at 1279 (citing Abbott, 300 F.3d at 1373). The HatchWaxman framework establishes that the act of filing an ANDA, in and of itself, constitutes a
technical infringement for jurisdictional purposes under 35 U.S.C. § 271(e)(2)(A). Id. at 1278.
“But the ultimate infringement question is determined by traditional patent law principles and, if
a product that an ANDA applicant is asking the FDA to approve for sale falls within the scope of
an issued patent, a judgment of infringement must necessarily ensue.” Id. Also, the Federal
Circuit was not persuaded by the defendants’ certification to the court that its product would
remain outside the claimed range. See id. at 1279-80. What defendant “has asked the FDA to
approve as a regulatory matter is the subject matter that determines whether infringement will
occur,” and any filing in court “does not overcome the basic fact that [the defendant] has asked
the FDA to approve, and hopes to receive from the FDA, approval to market a product within the
scope of the issued claims.” Id. at 1278.

74

2. Mylan’s Proposed ANDA Drug Product Infringes as a Matter of Law
Like the defendant in Sunovion, Mylan seeks FDA approval for a drug product that falls
within the scope of the asserted ‘727 patent claims. TMC has alleged that Mylan’s ANDA
infringes the ‘727 patent under § 271(e)(2)(A). (R.1, Compl. ¶ 25.) Mylan’s ANDA
specification allows it to market drug product with Asp9 impurity levels from 0.0%-2.0% (see
PTX 133.1785; PTX 159.2), a range that includes the ‘727 patent’s claimed ranges of 0.0-0.6%
(claim 1), 0.0-0.4% (claim 2) and 0.0-0.3% (claim 3).31 (See PTX 1, col. 25:62–26:56; PTX 133
at MYL0001785.) Mylan’s ANDA specification references the drug product as produced from a
batch or batches. (See, e.g., PTX 133.1852; PTX 133.205-208 (Quality Overall Summary
referring to “batch records” in the manufacturing steps and “the manufactured batch.”).)
Because Mylan’s ANDA specification seeks approval for a bivalirudin drug product made from
pharmaceutical batches allowed to have a maximum impurity level for Asp9 within the scope of
the ‘727 patent’s issued claims, Mylan infringes as a matter of law. See Sunovion, 731 F.3d. at
1278, 1280.
Mylan’s attempts to distinguish this case from Sunovion fail. First, Mylan argues
Sunovion is distinguishable because the claims in Sunovion were directed to individual
compounds below the defined maximum impurity level and whether they would infringe the
patent, whereas in the present case, the claims of the ‘727 patent are directed to all batches
containing Asp9 impurities below 0.6% and whether they infringe TMC’s ‘727 patent. Mylan’s
argument is unavailing. Mylan’s ANDA specification allows it to produce all batches having
less than 0.6% Asp9 impurities. (PTX 133.1785; PTX 159.2; see also Tr. 609:19-23 (Wayne
31

9

The same reasoning that applies to determining whether Mylan’s proposed drug meets the

Asp impurity limitation in claim 1 also applies to determining whether Mylan’s proposed drug meets the
Asp9 impurity limitations in claims 2 and 3. For the sake of simplicity, the Court will refer only to the
Asp9 impurity range in claim 1 (i.e., 0.0-0.6%) for the remainder of the infringement analysis.
75

Talton, Mylan’s Vice President of Global Regulatory Affairs) (“Q. There is nothing in Mylan’s
ANDA that prevents, from a regulatory standpoint, selling a product that has an Asp9 level below
.6 for every batch, correct? A. That would be an acceptable result.”) As the Federal Circuit
stated in Sunovion, “[w]hat a generic applicant asks for and receives approval to market, if
within the scope of a valid claim, is an infringement.” Id. (citing Abbott, 300 F.3d at 1373).
What Mylan seeks approval for (a bivalirudin drug product made from pharmaceutical batches
with Asp9 impurity levels ranging from 0.0-2.0%) allows it to make product that falls within the
scope of the ‘727 patent’s asserted claims (a bivalirudin drug product made from pharmaceutical
batches with Asp9 impurity levels ranging from 0.0-0.6%).32 Mylan, therefore, infringes the ‘727
patent as a matter of law. See id. at 1280.
Second, Mylan argues that its compounding process likely will lead to some batches
having Asp9 impurity levels above 0.6%.33 It is not, however, Mylan’s batch production
likelihoods that govern the infringement analysis under § 271(e)(2)(A). Rather it is what Mylan

32

Both TMC and Mylan cite the Delaware district court’s judgment in the related case of The
Medicines Co. v. Hospira, Inc., No. 09-750-RGA, 2014 WL 1292802 (D. Del. Mar. 31, 2014). As Mylan
notes, the Delaware court found that Hospira’s ANDA product did not infringe the ‘727 patent due to
Hospira’s use of inefficient mixing. The Delaware court’s finding of noninfringement was due to the fact
that, unlike in the present case, that court construed the ‘727 patent claims to include an “efficient
mixing” limitation (id. at *2, n.4), and the Delaware court found that Hospira’s proposed ANDA product
did not meet that limitation (id. at *5-*8). As Mylan also has noted, in this case, the Court has held that
“efficient mixing” is not a limitation of the ‘727 patent claims. The Delaware court did, however, make
an express finding that based on Sunovion, Hospira’s ANDA product infringed the “maximum impurity
9
level of Asp -bivalirudin that does not exceed about 0.6%” limitation. Id. at *5.
33

Mylan also argues that the Court decided the infringement issue as it relates to whether the
ANDA specification is dispositive in the Court’s summary judgment opinion. This argument fails as the
issue on summary judgment was whether viewing the evidence in the light most favorable to TMC, the
Court found that TMC offered sufficient evidence to establish a genuine issue of material fact, not
whether Mylan’s proposed ANDA product specification infringed the asserted claims as a matter of law
under Sunovion. See HK Systems, Inc. v. Eaton Corp., 553 F.3d 1086 (7th Cir. 2009) (explaining that
although the standard for granting judgment as a matter of law and granting summary judgment is the
same, the record compiled in a trial is bound to differ from the record on which a motion for summary
judgment was based); see also Aycock Eng’g, inc. v. Airflite, Inc., 560 F.3d 1350, 1356 (Fed. Cir. 2009)
(explaining that law of the case doctrine does not apply to a denial of summary judgment).

76

has asked the FDA to approve as a regulatory matter that is the subject matter for a determination
of whether infringement will occur. See Sunovion, 731 F.3d at 1278. The Federal Circuit
rejected a similar argument in Sunovion by finding infringement despite both the defendant’s
promise to only market product outside of the patent’s claimed impurity range and the likelihood
that internal manufacturing guidelines would produce drug product outside of the claimed 0.00.25% levorotatory isomer range. See id. at 1275, 1278 (explaining that even a certification to
the court to market only product outside the claimed range does not overcome infringement
when the ANDA specification is “within the scope of the issued claims.”). Allowing such an
“unconventional and unenforceable ‘guarantee’ ” to the court when Mylan is asking for and may
receive FDA approval to market a product within the scope of TMC’s patent, “would be
incompatible with the basic principles of patent law.” See id. at 1279. “If it had no intent to
infringe, [Mylan] should not have requested, or should not accept, approval to market a product
within the scope of the [‘727 patent claims].” See id. at 1279.
Lastly, Mylan argues that Sunovion is distinguishable because the methods of Mylan’s
ANDA, namely, “inefficient mixing,” produce variable and randomly high Asp9 results
exceeding the claimed maximum and that Mylan has committed to using the “inefficient mixing”
process. The claims of the ‘727 patent, however, are not limited to “efficient mixing” either
explicitly or as construed by the Court. (See R.119, Claim Construction Order; R.309, Summary
Judgment Order, at 35-39; R.558 at 22, n.11.) Therefore, Mylan’s use of an “inefficient mixing”
method does not rescue its ANDA specification from falling within the scope of the claims.

77

In the present case, Mylan’s ANDA test batch, produced using Mylan’s proposed
manufacturing process, did in fact have an Asp9 impurity level below 0.6%. (PTX 159.2.)34
Indeed, Mylan’s ANDA allows it to produce all batches below 0.6% Asp9 impurity levels. See
Sunovion, 731 F.3d at 1279. Sunovion renders it irrelevant that some batches may end up outside
the ‘727 patent’s claimed maximum impurity level. See id.; see also The Medicines Co. v.
Hospira, Inc., No. 09-750-RGA, 2014 WL 1292802, at *4-5 (D. Del. Mar. 31, 2014) (finding the
limitation of the “Asp9 impurity level below 0.6%” of the ‘727 patent claims met by defendant’s
ANDA specification which sought approval for drug product with Asp9 impurity levels up to
1.0%, noting “the fact that the ANDA application includes Asp9-bivalirudin levels above 0.6%,
and at some point Hospira might make a batch with levels above 0.6%, does not negate a finding
of infringement”).
Finally, the supplemental cases Mylan submitted to the Court do not alter the Court’s
decision that Mylan has infringed as a matter of law under Sunovion. In both cases—Ferring
B.V. v. Watson Labs., Inc.-Florida, 764 F.3d 1382, (Fed. Cir. 2014) (hereinafter, Apotex) and
Ferring B.V. v. Watson Labs., Inc.-Florida, 764 F.3d 1401 (Fed. Cir. 2014) (hereinafter,
Watson)—the ANDA at issue was silent with respect to the claim limitations of the patents-insuit. See Apotex, 764 F.3d at 1387; Watson, 764 F.3d at 1409. In Apotex, the ANDA
specification provided a dissolution release rate for the proposed drug in which “NLT [not less
than] 80% of the labeled amount of tranexamic acid dissolved in 60 minutes,” whereas the claim
limitations in the patents-in-suit pertained to the dissolution release rates only at 15 minutes, 45
minutes, 90 minutes, and 120 minutes. See Apotex, 764 F.3d at 1385-86. Because the ANDA at

34

The Court refers to the test batch as merely an example of Mylan’s proposed ANDA product,
as the specific impurity levels of this test batch are not necessary for the infringement analysis under
Sunovion.

78

issue was silent with respect to the claim limitations of the patents-in-suit, which did not specify
the dissolution rate at 60 minutes, the Federal Circuit held that Sunovion did not apply. Id. at
1387-88. Similarly, in Watson, the ANDA at issue was silent with respect to the claim
limitations of the patents-in-suit concerning the drug’s dissolution release rate in water. See
Watson, 764 F.3d at 1409. Accordingly, the Federal Circuit held that the ANDA specification
alone did not resolve the question of infringement, and Sunovion therefore did not apply. See id.
at 1408-09.
Here, unlike in Apotex and Watson, Mylan’s ANDA specification speaks directly to the
disputed claim limitations in the ‘727 patent and therefore resolves the question of infringement.
As explained above, Mylan’s ANDA specification, if approved, will allow it to market a drug
with Asp9 impurity levels from 0.0%-2.0%, a range that includes the ‘727 patent’s claimed range
of 0.0-0.6%. Because Mylan’s ANDA specification clearly describes a product that meets the
limitations of the asserted claims, Sunovion is directly on point and establishes that Mylan’s
proposed drug product infringes claims 1-3, 7-10 and 17 of the ’727 patent as a matter of law.
See Sunovion, 731 F.3d. at 1278, 1280.

79

CONCLUSION
For the reasons explained above, the Court finds that: (1) Mylan failed to prove by clear
and convincing evidence that claims 1-3, 7-10, and 17 of the ‘727 patent are invalid or
unenforceable; and (2) Mylan’s proposed bivalirudin drug product infringes claims 1, 7-10, and
17 of the ‘727 patent as a matter of law under Sunovion. Accordingly, the Court enters judgment
in favor of TMC on its infringement claim regarding the ‘727 patent and on Mylan’s
counterclaims regarding claims 1-3, 7-10, and 17.

DATED: October 27, 2014

ENTERED

_________________________________
AMY J. ST. EVE
United States District Court Judge

80