ASTRAZENECA AB et al v. ANDRX LABS, LLC et al

Filing 107

MEMORANDUM OPINION filed. Signed by Judge Mary L. Cooper on 1/11/2017. (km)

NOT FOR PUBLICATION
UNITED STATES DISTRICT COURT
DISTRICT OF NEW JERSEY
:
ASTRAZENECA AB, AKEIEBOLAGET : CIVIL ACTION NO. 14-8030 (MLC)(DEA)
HÄSSLE, ASTRAZENECA LP, and
:
ZENECA INC.
:
:
Plaintiffs,
:
:
v.
:
:
ANDRX LABS, LLC, ANDRX
:
CORPORATION, and ACTAVIS, INC., :
:
Defendants.
:
__________________________________ :
:
ASTRAZENECA AB, AKEIEBOLAGET : CIVIL ACTION NO. 15-1057 (MLC)(DEA)
HÄSSLE, ASTRAZENECA LP, and
:
ZENECA INC.
:
:
Plaintiffs,
:
:
v.
:
MEMORANDUM OPINION
:
PERRIGO COMPANY PLC, PERRIGO :
COMPANY, and L. PERRIGO
:
COMPANY,
:
:
Defendants.
:
__________________________________ :
Cooper, District Judge
OUTLINE
PRELIMINARY STATEMENT
I.

BACKGROUND

A.
B.
C.
II.

Procedural background
The asserted ‘085 and ‘070 Patents and claims
Disputed claim terms

DISCUSSION
A.
B.

Legal Standard
Application
1.

Summary of intrinsic evidence
a.
b.
c.
d.
e.

Relevant claim terms
Common ‘085 and ‘070 Patent written description
Prosecution history of ‘085 Patent
Prosecution history of ‘070 Patent
Related prior art patent family

2.

Definition of person of ordinary skill in the art

3.

Summary of extrinsic evidence
a.
b.
c.

4.

Court findings and claim constructions
a.
b.
c.

III.

Plaintiff’s expert testimony and exhibits
Andrx’s expert testimony and exhibits
Perrigo’s expert testimony and exhibits

“the magnesium salt of S-omeprazole trihydrate”
“characterized by the following major peaks in its X-ray
diffractogram”
“represented by FIG. 1”

CONCLUSION

PRELIMINARY STATEMENT
This is a claim construction opinion in two related but not consolidated HatchWaxman cases. The plaintiff group of companies, AstraZeneca AB, Aktiebolaget Hässle,
AstraZeneca LP, KBI Inc., and KBI-E Inc. (collectively, “AstraZeneca” or “plaintiff”),
2

has sued defendants Andrx Labs, LLC, Andrx Corporation, and Actavis, Inc.
(collectively, “Andrx”) in Civil Action No. 14-8030. Plaintiff has sued defendants
Perrigo Company, PLC, Perrigo Company, L. Perrigo Company, and Paddock
Laboratories, LLC (collectively, “Perrigo”) in Civil Action No. 15-1057.
Both cases arise under 35 U.S.C. § 271(e)(2). AstraZeneca alleges in each case
that a drug product that the named defendants propose to market infringes claims of U.S.
Patent Nos. 6,369,085 (“the ‘085 Patent”) and 7,411,070 (“the ‘070 Patent”) (together,
“the patents-in-suit”).
The claim construction proceedings in both of these cases (and in a related case
against Mylan Laboratories that recently settled, Civil Action No. 12-1378) were
conducted jointly. This procedure was used because both cases involve the same two
patents-in-suit, and the same three claim terms are disputed in each case. Nevertheless,
each defendant group is a potential marketplace competitor and has its own claim
construction positions, which we will address in the course of this opinion. Citations to
the record of docketed filings in the body of this opinion will refer to Civil Action No.
14-8030. Citations to filings in the other two case dockets will be placed in the footnotes,
for reasons explained here in the margin.1

1

We begin with a procedural note. Due to the joint nature of these proceedings in two
non-consolidated cases, this opinion is filed on the docket of each case. Those are, as stated, the
case against Andrx (“the Andrx action”), Civil Action No. 14-8030; and the case against Perrigo
(“the Perrigo action”), Civil Action No. 15-1057. There is a certain amount of overlap in the
claim construction filings of the parties in each case, as would be expected, although of course
there are also individual filings pertinent only to each respective case. The procedural picture is
further complicated by the fact that filings in the now-settled Civil Action No. 12-1378, against
Mylan (“the Mylan action”), were largely duplicative of filings in the other two cases as well.

3

AstraZeneca sells Nexium®, one of the drugs known as “proton pump inhibitors,”
or “PPIs,” that are used in the treatment of disorders caused by stomach acid reflux. The
FDA approved Nexium® in 2001. Although many of the patents that protected it are
currently expired, the ‘085 and ‘070 Patents will not expire until May 25, 2018. (Dkt. 1
at 4-5.) The patents-in-suit claim the magnesium salt of S-omeprazole trihydrate, the
form of the compound that AstraZeneca uses as the active pharmaceutical ingredient
(“API”) in Nexium®. (Id. at 4.)
Andrx has submitted to the FDA an Abbreviated New Drug Application
(“ANDA”) seeking approval for a generic version of Nexium®. (Id. at 7.) The Andrx
ANDA includes “Paragraph IV certifications,” asserting that the patents-in-suit are
invalid, unenforceable, or will not be infringed by Andrx’s ANDA product. (Id.)
Separately and independently, Perrigo has submitted to the FDA an ANDA seeking

All three groups of defendants participated and argued at the claim construction hearing, each
defendant referring to the docket filings in its individual case.
This claim construction opinion thoroughly addresses the positions of all of the litigants
in the two current cases, the Andrx action and the Perrigo action, with occasional reference to the
now-settled Mylan action where relevant. However, in an effort to make clear and consistent the
record citations here, the Court will use the following citation protocol.
All record citations in the body of the opinion are to the docket of the Andrx case, Civil
Action No. 14-8030. Record citations to the Perrigo case, Civil Action No. 15-1057, are placed
in the margin as footnotes, and designated “Perrigo action.” Likewise, citations to the settled
Mylan case, Civil Action No. 12-1378, are in the footnotes and designated “Mylan action.”
Document entry numbers filed on the Electronic Case Filing System (“ECF”) are cited as
“dkt.” Page pincites refer to the ECF pagination, e.g., “dkt. 46-1 at 33-35.”
Thus, ECF citations in the body of the opinion are simply “dkt.,” referring to the docket
in the Andrx case, Civil Action No. 14-8030. ECF citations to the Andrx case in the footnotes
also use the format “dkt.” ECF citations to the Perrigo case and the Mylan case in the footnotes
use the format, e.g., “Perrigo action, dkt. 52-1 at 12-15.”

4

approval for its generic version of Nexium®, described as esomeprazole magnesium
delayed-released capsules, 20 mg, including a Paragraph IV certification against the same
two patents-in-suit.2
The ‘085 Patent, issued in 2002, is entitled “Form of S-Omeprazole.”3 The ‘070
Patent, based on a divisional application related to the ‘085 Patent, has the same name
and was issued in 2008. The ‘070 Patent is subject to a terminal disclaimer whereby both
patents will expire on May 25, 2018. (Dkt. 1 at 4-5.) All parties agree that the priority
date for published prior art is May 30, 1997, based on a Swedish priority application date.
(Dkt. 81.)
This opinion construes a claim term common to both patents-in-suit, a claim term
in the ‘070 Patent, and a claim term in the ‘085 Patent. The primary issue concerns
construction of the common claim term, which constitutes the entirety of claim 1 of the
‘070 Patent. It is “[t]he magnesium salt of S-omeprazole trihydrate.” (See n. 10, infra.)
The same claim term appears, with various limitation language, in all other claims of both
patents. The Court has reviewed the written submissions of the parties, and conducted a
two-day claim construction hearing based on those submissions and the oral arguments of
the parties.

2

(Perrigo action, dkt. 1 at 5.)

The ‘085 Patent is Exhibit A to the Complaint in both the Andrx action and the Perrigo
action. (See, e.g., dkt. 1-2 at 2-14.) Likewise, the ‘070 Patent is Exhibit B to each Complaint.
(See, e.g., dkt. 1-3 at 2-12.) Copies of both patents-in-suit are also attached as exhibits to various
Markman papers filed in each case. We will simply cite those patents by page, drawing sheet, or
column and line number.
3

5

The Court concludes, for the reasons stated herein, that an appropriate
construction of the term “the magnesium salt of S-omeprazole trihydrate” is “the
magnesium salt of S-omeprazole in crystalline form containing three molecules of water
of crystallization per molecule of magnesium salt of S-omeprazole.” The other two claim
terms in dispute are also construed as explained in this opinion.
I.

BACKGROUND
A.

Procedural background

There is a long history of patent litigation relating to AstraZeneca’s successful
Nexium® products containing the active pharmaceutical ingredient omeprazole. We will
not attempt to recount that history here in detail. Some aspects of that history, of which
this Court and the parties are well aware, are pertinent to the present claim construction
issues. Here is a brief summary of those aspects of the procedural background.
AstraZeneca’s ‘085 and ‘070 Patents are part of one family of patents (including
some that have expired) that have covered Nexium® products down through the years
since initial FDA approval in 2001. Those patents include a patent issued as recently as
2013. We will refer to that patent family as the ‘085 Patent Family, and we have
docketed a chart identifying that group of patents. (Dkt. 106.) The Nexium® product
line has been the subject of numerous challenges by entities seeking FDA approval to sell
generic versions in the U.S. market.4

4

There is another pharmaceutical product line named Vivomo®, also formerly marketed
by AstraZeneca and now marketed by a successor entity, that contains the same omeprazole API
as used in Nexium® and has been covered also by the same group of patents, as well as other
patents directed to those pharmaceutical compositions. The Vivomo® products are the subject

6

The litigation stemming from those generic challenges has been extensive in this
Court, beginning in 2005 when the first Hatch-Waxman case pertaining to Nexium® was
brought here.5 That body of Nexium® litigation has also involved an earlier-established
AstraZeneca patent family, also pertaining to omeprazole, that included now-expired U.S.
Patent No. 5,714,504 (the “‘504” Patent). We will refer to that patent family as the ‘504
Patent Family, and we have also docketed a chart identifying that group of patents. (Dkt.
105.) The first Hatch-Waxman case in this line, filed in 2005, for example, had as its
patents-in-suit the ‘504 Patent, the ‘085 Patent, and another patent in the ‘504 Patent
Family. See AstraZeneca AB, et al. v. Dr. Reddy’s Laboratories, Ltd., et al., Civil Action
No. 05-5553, dkt. 246, 2010 WL 1981790 (D.N.J. May 18, 2010).
Claim construction opinions have been rendered in some of these cases, pertaining
to the claim terms at issue in this matter and other related claim terms in these two Patent
Families. See Sections II.B.1.e and II.B.4, infra. While not binding on the parties in the
current matter, those opinions are of record and cannot be overlooked by this Court in
ruling on the issues presented here.
The landscape of Nexium® Hatch-Waxman litigation in this Court includes a long
line of closed cases.6 It also includes numerous more recent and currently-pending cases,

of another group of Hatch-Waxman cases in this Court, not listed here. See, e.g., Horizon
Pharma, Inc., et al. v. Dr. Reddy’s Laboratories Inc., et al., Civil Action No. 11-2317.
5

The Civil Action numbers for the original group of Nexium® cases in this Court, now
all completed, were No. 05-5553, No. 06-1057, No. 08-328, and No. 08-4993. Those actions
were against various defendants including Ranbaxy, Teva, and Dr. Reddy’s Laboratories.
6

As shown in n. 5, supra, the original Nexium®-related ANDA cases in this Court were
filed between 2005 and 2008. This line of litigation has continued unabated since that time.

7

including the Andrx and Perrigo actions that are the subject of this opinion. In fact, new
cases in this group continue to be filed.
A listing of the currently-pending Nexium® Hatch-Waxman cases in this Court is
set forth in the margin.7 The parties in several of those other cases have filed stipulations
agreeing to be bound by the claim construction rulings rendered in this opinion, subject to
any appellate review. See, e.g., Civil No. 14-4782, dkt. 59; Civil No. 15-6025, dkt. 56;
and Civil No. 15-7415, dkt. 28.
B.

The asserted ‘085 and ‘070 Patents and claims

The ‘085 Patent was filed on June 8, 1998, and was issued to AstraZeneca AB as
assignee on April 9, 2002. (‘085 Pat., pg. 1.) It is entitled “FORM OF SOMEPRAZOLE.” (Id.) Plaintiff states that the ‘085 Patent will expire on May 25, 2018,
with pediatric exclusivity until November 25, 2018. (Dkt. 1 at 4.) The ‘085 Patent
contains twelve claims and five drawing sheets. Claim 1 is independent. Claims 2-12
depend from claim 1, directly or indirectly. The claims of the ‘085 Patent are reproduced
in the margin.8

Some of the more recently-closed cases include the following Civil Actions (listed by docket
number and lead defendant name): No. 12-1378 (Mylan); No. 13-1669 (Actavis); No. 13-7298
(Aurobindo); No. 13-7299 (Kremers); No. 14-7263 (Actavis); No. 14-7870 (Actavis); No. 156609 (Alkem).
7

The Civil Action numbers for the currently pending Nexium® ANDA cases here, in
addition to the Andrx action and the Perrigo action captioned in this opinion (likewise listed by
lead defendant name), are No. 13-4854 (Wockhardt); No. 14-4782 (Zydus); No. 15-6025 (Hec);
No. 15-6092 (Lupin); No. 15-7415 (Zydus); No. 15-8267 (Dr. Reddy’s); No. 16-1682
(Macleods); No. 16-2442 (Hetero); No. 16-4414 (Aurobindo); and 16-7330 (Glenmark).
8

Claim 1. The magnesium salt of S-omeprazole trihydrate, wherein the compound is
characterized by the following major peaks in its X-ray diffractogram:

8

Claim 2. The magnesium salt of S-omeprazole trihydrate according to claim 1, wherein
the compound is in a highly crystalline form.
Claim 3. The magnesium salt of S-omeprazole trihydrate according to claim 1, wherein
the compound is in a stable form.
Claim 4. A process for the preparation of the magnesium salt of S-omeprazole trihydrate
according to any of claims 1, 2 or 3 which comprises treating a magnesium salt of S-omeprazole
of any other form with water.
Claim 5. A process for the preparation of the magnesium salt of S-omeprazole trihydrate
according to any of claims 1, 2 or 3 which comprises the following steps:
a) mixing a potassium salt of S-omeprazole with an organic solvent;
b) converting the potassium salt of S-omeprazole into a corresponding
magnesium salt of S-omeprazole by treating the potassium salt with a
magnesium source;
c) precipitating the magnesium salt of S-omeprazole by addition of a nonsolvent;
d) isolating the obtained magnesium salt of S-omeprazole;
e) treating the obtained magnesium salt of S-omeprazole with water, and
f) isolating and drying the obtained magnesium salt of S-omeprazole trihydrate.
Claim 6. The process according to claim 5, wherein the organic solvent of step a) is
methanol.
Claim 7. The process according to claim 5, wherein the non-solvent of step c) is acetone.
Claim 8. The process according to claim 5 wherein steps a) to e) are replaced by the
following single step: converting the potassium salt of S-omeprazole into a corresponding
magnesium salt of S-omeprazole by treating the potassium salt with a magnesium source in
water.
Claim 9. The process according to claim 5, wherein the magnesium source is
magnesium sulfate.
Claim 10. The process according to claim 8, wherein the magnesium source is
magnesium sulfate.

9

With respect to Andrx, AstraZeneca asserts infringement of claims 1-4 and 12 of
the ‘085 Patent. (Dkt. 38 at 5-6.) With respect to Perrigo, AstraZeneca asserts
infringement of at least claims 1-4 and 11-12 of the ‘085 Patent.9
The ‘070 Patent, also entitled “FORM OF S-OMEPRAZOLE,” was filed on
September 25, 2003, and was issued to plaintiff AstraZeneca AB as assignee on August
12, 2008. (‘070 Pat., p.1.) The ‘070 Patent was terminally disclaimed over the ‘085
Patent, making its expiration date the same as that of the ‘085 Patent. (See Id.) The ‘070
Patent contains four claims and five drawing sheets. Claim 1 is independent. Claims 2-4
depend from claim 1, directly or indirectly. The claims of the ‘070 Patent are reproduced
in the margin.10

Claim 11. A pharmaceutical composition comprising the magnesium salt of Someprazole trihydrate according to any of claims 1, 2 or 3 as active ingredient and a
pharmaceutically acceptable carrier.
Claim 12. A method of treating a gastric acid related condition which method comprises
administering to a subject suffering from said condition a therapeutically effective amount of the
magnesium salt of S-omeprazole trihydrate according to any of claims 1, 2 or 3. (‘085 Pat., col.
10, l. 15 – col. 12, l. 5.)
9

(Perrigo action, dkt. 44 at 7.)

10

Claim 1. The magnesium salt of S-omeprazole trihydrate.
Claim 2. The magnesium salt of S-omeprazole trihydrate according to claim 1
represented by FIG. 1.
Claim 3. A process for the preparation of the magnesium salt of S-omeprazole trihydrate
according to claim 1 which comprises treating a magnesium salt of S-omeprazole of any other
form with water.
Claim 4. A process for the preparation of the magnesium salt of S-omeprazole trihydrate
according to claim 2 which comprises treating a magnesium salt of S-omeprazole of any other
form with water. (‘070 Pat., col. 10, ll. 51-62.)

10

AstraZeneca asserts infringement of claims 1-4 of the ‘070 Patent against both
Andrx and Perrigo. (See dkt. 38 at 5-6.)11
The ‘085 and ‘070 Patents are two of several patents originating from an
international application filed May 25, 1998, including United States Patents No.
6,747,155, No. 6,677,455, No. 7,745,466, No. 8,076,361, and No. 8,466,175. (See dkt.
106.) All of the above-listed patents are directed to various forms of S-omeprazole and
methods, processes, or compositions employing those forms. They all share essentially
the same specification language.
Omeprazole is the generic name for the compound 5-methoxy-2-[[(4-methoxy3,5-dimethyl-2-pyridinyl)methyl]sulfinyl]-1H-benzimidazole. (‘070 Pat., col. 1, ll. 2527.) It is a proton pump inhibitor and is useful as an antiulcer agent. (Id., col. 1, ll. 2931.) It is a sulfoxide and chiral compound with the sulfur atom being the stereogenic
center, meaning omeprazole is a racemic mixture of its two single enantiomers, the Rand S-enantiomers of omeprazole. (Id., col. 1, ll. 35-38.) The Background section of the
‘070 Patent acknowledges that certain salts of single enantiomers of omeprazole,
including magnesium salts of R- and S-omeprazole, were previously disclosed in the
prior art. (Id., col. 1, ll. 25-58.)
The Field of Invention and Abstract sections of the ‘085 and ‘070 Patents, which
are nearly identical, state that the invention relates to a novel form of S-omeprazole, and
more specifically to a novel form of the magnesium salt of the S-enantiomer of

11

(See Perrigo action, dkt. 44 at 7.)

11

omeprazole trihydrate. (Id., col. 1, ll. 12-17.) These sections also state that the invention
relates to processes for preparing such a novel form and the pharmaceutical compositions
containing it, as well as intermediates used in the process, and their preparation. (Id., col.
1, ll. 17-21.) The claim construction dispute in this action centers upon the magnesium
salt of the S-enantiomer of omeprazole trihydrate.
C.

Disputed claim terms

The parties have identified four claim terms requiring construction in this matter:
(1) “highly crystalline form;” (2) “the magnesium salt of S-omeprazole trihydrate;” (3)
“characterized by the following major peaks in its X-ray diffractogram;” and (4)
represented by FIG. 1.” (See dkt. 36 at 2-3.)12 As discussed in more detail below, three
of these four claim terms are disputed by the parties.
1.

“highly crystalline form”

The claim term “highly crystalline form” appears only in dependent claim 2 of the
‘085 Patent. See n. 8, supra. However, that term is incorporated by reference into
dependent claims 4-12 of the ‘085 Patent. See id. Claims 1 and 2 of the ‘085 Patent state
as follows:
1.
The magnesium salt of S-omeprazole trihydrate,
wherein the compound is characterized by the following major
peaks in its X-ray diffractogram:

12

(See Perrigo action, dkt. 37 at 2-3.)

12

2.
The magnesium salt of S-omeprazole trihydrate
according to claim 1, wherein the compound is in a highly
crystalline form.
(‘085 Pat., col. 10, ll. 15-37 (emphasis added).)
For the purposes of this case, the parties agree that this term means “having a
higher crystallinity than any other form of magnesium salt of S-omeprazole disclosed in
the prior art.” (Dkt. 36 at 2.)13
2.

“the magnesium salt of S-omeprazole trihydrate”

The term “the magnesium salt of S-omeprazole trihydrate” is applicable to claims
1-12 of the ‘085 Patent and claims 1-4 of the ‘070 Patent. The term appears in claims 1-5
and 11-12 of the ‘085 Patent and claims 1-4 of the ‘070 Patent. See nn. 8, 10, supra.
Independent claim 1 of the ‘085 Patent, in which the term first appears, is reproduced
above in Section I.C.1. The term is incorporated by reference into dependent claims 6-10
of the ‘085 Patent. See n. 8, supra. Claim 1 of the ‘070 Patent, in which the term first

13

(Perrigo action, dkt. 37 at 2.)

13

appears, states in its entirety: “The magnesium salt of S-omeprazole trihydrate.” (‘070
Pat., col. 10, l. 52.) The parties’ respective constructions are as follows:
Term Identified
for Construction
the magnesium salt
of S-omeprazole
trihydrate

AstraZeneca’s
Proposed
Construction
a trihydrate of a
magnesium salt of
S-omeprazole
containing three
molecules of water
per molecule of
magnesium salt of
S-omeprazole

Andrx’s Proposed
Construction
the magnesium salt
of S-omeprazole in
crystalline form
containing three
molecules of water
of crystallization
per molecule of
magnesium salt of
S-omeprazole

Perrigo’s
Proposed
Construction
crystals of the
magnesium salt of
S-omeprazole
trihydrate,
substantially free
from magnesium
salts of Romeprazole and
other forms of
magnesium salts of
S-omeprazole

(See dkt. 36 at 3.)14,15
AstraZeneca’s proposed construction of this term is identical to the construction
Judge Pisano arrived at in AstraZeneca AB v. Dr. Reddy’s Laboratories Inc., No. 112317 (JAP), 2013 WL 1847639 (D.N.J. May 1, 2013). In that case, AstraZeneca offered
the following construction: the “magnesium salt of S-omeprazole having approximately
three molecules of bound water per molecule of S-omeprazole magnesium.” Id. at *8.16

14

(Perrigo action, dkt. 37 at 3.)

Mylan’s proposed construction for this term was similar to Andrx’s proposed
construction and was “the substantially pure crystalline magnesium salt of S-omeprazole having
exactly three waters of hydration.” (Mylan action, dkt. 218 at 2.)
15

16

The construction offered by AstraZeneca was submitted shortly before the Markman
hearing in that case, and differed from a prior proposed construction offered by AstraZeneca.
See AstraZeneca AB v. Dr. Reddy’s Laboratories Inc., No. 11-2317 (JAP), 2013 WL 1847639 at
n.1 (D.N.J. May 1, 2013). AstraZeneca’s initial construction for the term broke the term into two
separate parts: “magnesium salt of” and “S-omeprazole trihydrate.” For “magnesium salt of,”
AstraZeneca proposed the following meaning: “a compound formed between positively-charged

14

The defendants, in that case, proposed this construction: “a trihydrate of a magnesium
salt of S-omeprazole containing three molecules of water per molecule of magnesium salt
of S-omeprazole in a unit cell of the crystal lattice that is substantially free from
magnesium salts of R-omeprazole and other prior art forms of magnesium salts of Someprazole including S-omeprazole dihydrate and amorphous forms.” Id. Relying on a
comparison of the claims, the court found that the term was “intended to have a broad
construction” and refused to import the limitations sought by defendants. Id. However,
the court did not adopt AstraZeneca’s proposed construction either, noting AstraZeneca’s
proposal was flawed in that it sought to define the trihydrate as having “approximately”
three molecules of water, a definition that ran afoul of the plain and ordinary meaning of
the term “trihydrate.” Id. Thus, the court construed the term to mean “a trihydrate of a
magnesium salt of S-omeprazole containing three molecules of water per molecule of
magnesium salt of S-omeprazole.” Id.
3.

“characterized by the following major peaks in its X-ray
diffractogram”

The term “characterized by the following major peaks in its X-ray diffractogram”
appears only in claim 1 of the ‘085 Patent, which is reproduced above in Section I.C.1.

Magnesium (Mg) cations and negatively-charged esomeprazole anions.” AstraZeneca AB v. Dr.
Reddy’s Laboratories Inc., No. 11-2317, dkt. 50-1 at 4. For “S-omeprazole trihydrate,”
AstraZeneca proposed the following meaning: “(S)-omeprazole having a structure that has a
theoretical ratio of three molecules of bound water per molecule of ((S)-omeprazole)2
magnesium, but which does not necessarily contain exactly three molecules of water, whose
structure may be determined by analytical methods identified in the patent and known to those of
ordinary skill. In the ‘085 patent the structure is determined by examining XRD.” Id. at 4-5.

15

The term is incorporated by reference into dependent claims 2-12 of the ‘085 Patent. See
n. 8, supra. The parties’ respective constructions are as follows:
AstraZeneca’s
Proposed
Construction
characterized by the identifiable by
following major
reference to an Xpeaks in its X-ray
ray diffractogram
diffractogram
that includes the
major peaks below

Andrx’s Proposed
Construction

Term Identified
for Construction

(Dkt. 36 at 3.)17,18

Perrigo’s
Proposed
Construction
having each of the
referenced major
peaks in its X-ray
powder
diffractogram
within normal
experimental error

exhibiting, within
the range of normal
experimental error,
each of the
following as major
peaks in its X-ray
diffractogram, each
such peak falling
within the range for
relative intensity
recited for the peak,
such ranges being
defined at Column
5, lines 28-39
The relative intensity range definitions provided at column 5, lines 28-

39 of the ‘085 Patent, as referenced in Andrx’ proposed construction, appear as follows:

(‘085 Pat., col. 5, ll. 32-39.)

17

(Perrigo action, dkt. 37 at 3.)

Mylan’s proposed construction for this term was similar to Perrigo’s proposed
construction and was “the magnesium salt of the S-omeprazole trihydrate has an x-ray diffraction
diffractogram having each peak as set forth in the table in claim 1 within normal experimental
error.” (Mylan action, dkt. 218 at 2.)
18

16

AstraZeneca’s proposed construction of this term is identical to the construction
AstraZeneca proposed in AstraZeneca AB v. Dr. Reddy’s Laboratories Inc., which was
adopted verbatim by that court. The defendants’ proposed construction for the term was
“having all of the referenced major peaks in its X-ray diffractogram,” a construction
similar to that proposed by both Andrx and Perrigo in this matter. See AstraZeneca AB,
No. 11-2317, 2013 WL 1847639 at *8-9. In adopting AstraZeneca’s proposed
construction, the Court found that the defendants’ proposed construction was too rigid
and that it failed to account for experimental error:
The Court concludes that Defendants' construction, which
would require an exact match, is too rigid. The claim language
requires only that the S-omeprazole trihydrate be
“characterized” by the peaks in the table, not necessarily that it
have a perfect one-to-one relationship. Even Defendants'
expert concedes that although the X-ray diffraction of a
compound will have the same “general appearance,” the
positions of the peaks may differ somewhat because of slight
experimental errors. Plaintiffs' construction accounts for such
differences, while Defendants' would not.
Id. at *9.
4.

“represented by FIG. 1”

The term “represented by FIG. 1” appears only in claim 2 of the ‘070 Patent.
Claim 2 of the ‘070 Patent states: “The magnesium salt of S-omeprazole trihydrate
according to claim 1 represented by FIG. 1.” (‘070 Pat., col. 10, ll. 53-54.) The term is
incorporated by reference into dependent claim 4 of the ‘070 Patent. See n. 10, supra.
The parties’ respective constructions are as follows:

17

Term Identified
for Construction
represented by FIG.
1

AstraZeneca’s
Proposed
Construction
represented by
Figure 1 of the ‘070
patent

Andrx’s Proposed
Construction
exhibiting an x-ray
powder
diffractogram that
is the same as
Figure 1 of the ‘070
patent, within
normal
experimental error

Perrigo’s
Proposed
Construction
having an X-ray
powder
diffractogram that
is the same as
Figure 1 of the ‘070
patent within
normal
experimental error

(Dkt. 36 at 3.)19,20 Figure 1 of the ‘070 Patent, which is described as an “X-ray powder
diffractogram of the magnesium salt of S-omeprazole trihydrate prepared according to
the present invention,” (‘070 Pat., col. 1, ll. 62-64), is reproduced below:

19

(Perrigo action, dkt. 37 at 3.)

Mylan’s proposed construction for this term was identical to Andrx’s proposed
construction. (See Mylan action, dkt. 218 at 2.)
20

18

(‘070 Pat., Fig. 1.) Figure 1 of the ’070 Patent is identical to Figure 1 of the ‘085 Patent.
(Compare ‘070 Pat., Fig. 1 with ‘085 Pat., Fig. 1.) The written description explains that
the data in Example 1, Table 1 is a reading of the X-ray powder diffractogram depicted in
Figure 1. (See ‘070 Pat., col. 5, ll. 26 – col. 6, ll. 10.)
AstraZeneca’s proposed construction of this term is identical to the construction
AstraZeneca proposed in AstraZeneca AB v. Dr. Reddy’s Laboratories Inc., which was
adopted verbatim by that court. The defendants’ proposed construction for the term was
“having an X-ray diffractogram the same as FIG. 1,” a construction similar to that
proposed by both Andrx and Perrigo. See AstraZeneca AB, No. 11-2317, 2013 WL
1847639 at *9. The court found the defendants’ proposed construction to be too rigid for
the same reasons it expressed with respect to defendants’ proposed construction for the
term “characterized by the following major peaks in its X-ray diffractogram.” See id.
The written submissions of the parties on these claim construction issues are
described in the margin.21 In addition, the Court conducted a Markman hearing. (Dkt.
82, Markman Hr’g Tr. Vol. I; dkt. 83, Markman Hr’g Tr. Vol. II.)

21

In the Andrx action: dkt. 36 to dkt. 36-3, Joint Claim Construction & Prehearing
Statement; dkt. 38, Andrx’s Opening Br.; dkt. 38-1 to dkt. 38-2, Edwards I Decl.; dkt. 38-3,
Zaworotko I Decl.; dkt. 39, Plaintiff’s Opening Br.; dkt. 39-1 to dkt. 39-10, Stole I Decl.; dkt. 40
to dkt. 40-17, Byrn I Decl.; dkt. 50, Andrx’s Responsive Br.; dkt. 47-1, Edwards II Decl.; dkt.
47-2, Zaworotko II Decl.; dkt. 48, Plaintiff’s Responsive Br.; dkt. 48-1 to dkt. 48-16, Stole II
Decl.; dkt. 49 to dkt. 49-8, Byrn II Decl.; dkt. 54, Andrx’s Reply Br.; dkt. 54-1, Zaworotko III
Decl.; dkt 80 to dkt. 80-3, Plaintiff’s Ltr. 5/24/16 with attachments; dkt. 86-1, Gannon Decl. with
exhibits. Plaintiff also submitted supplemental materials under seal, pertaining to the Andrx
ANDA product. (Dkt. 77, dkt. 77-1, dkt. 84 to dkt. 84-3, all SEALED.) Andrx objected to these
materials as irrelevant to the pending Markman proceedings. (Dkt. 89.) The Court agrees with
Andrx on this point. Nevertheless, we have reviewed those sealed submissions from plaintiff,
and we do not find that the contents would change the claim construction rulings set forth here.

19

II.

DISCUSSION
A.

Legal standard

“It is a ‘bedrock principle’ of patent law that ‘the claims of a patent define the
invention to which the patentee is entitled the right to exclude.’” Phillips v. AWH Corp.,
415 F.3d 1303, 1312 (Fed. Cir. 2005) (en banc) (quoting Innova/Pure Water Inc. v. Safari
Water Filtration Sys., Inc., 381 F.3d 1111, 1115 (Fed. Cir. 2004)). Claim construction
determines the correct claim scope, and is a determination exclusively for the court as a
matter of law. Markman v. Westview Instruments, Inc., 52 F.3d 967, 978-79 (Fed. Cir.
1995) (en banc), aff’d, 517 U.S. 370 (1996).

The focus in construing disputed terms in

claim language “is on the objective test of what one of ordinary skill in the art at the time
of the invention would have understood the term[s] to mean.” Id. at 986.

In the Perrigo action: Perrigo action, dkt. 37 to dkt. 37-3, Joint Claim Construction and
Prehearing Statement; Perrigo action, dkt. 44, Perrigo’s Opening Br.; Perrigo action, dkt. 44-1 to
dkt. 44-2, Buckton I Decl.; Perrigo action, dkt. 44-3 to dkt. 44-22, Shrestha I Decl.; Perrigo
action, dkt. 45, Plaintiff’s Opening Br.; Perrigo action, dkt. 45-1 to dkt. 45-10, Stole I Decl.;
Perrigo action, dkt. 46 to dkt. 46-17, Byrn I Decl.; Perrigo action, dkt. 50, Perrigo’s Responsive
Br.; Perrigo action, dkt. 50-1, Buckton II Decl.; Perrigo action, dkt. 50-2 to dkt. 50-5, Shrestha II
Decl.; Perrigo action, dkt. 51, Plaintiff’s Responsive Br.; Perrigo action, dkt. 51-1 to dkt. 51-16,
Stole II Decl.; Perrigo action, dkt. 52 to dkt. 52-8, Byrn II Decl.; Perrigo action, dkt. 70, Perrigo
Ltr. 5/6/16; Perrigo action, dkt. 72 to dkt. 72-1, Flax Decl.; Perrigo action, dkt 76 to dkt. 76-3,
Plaintiff’s Ltr. 5/24/16 with attachments; Perrigo action, dkt. 83, Perrigo Joint Reply Ltr.
6/13/16.
In the Mylan action: Mylan action, dkt. 218 to 218-2, Joint Claim Construction and
Prehearing Statement; Mylan action, dkt. 230, Plaintiff’s Opening Br.; Mylan action, dkt. 230-1
to dkt. 230-10, Stole I Decl.; Mylan action, dkt. 231 to dkt. 231-17, Byrn I Decl.; Mylan action,
dkt. 232, Mylan’s Opening Br.; Mylan action, dkt. 232-1 to dkt. 232-15, Beel I Decl.; Mylan
action, dkt. 232-16 to dkt. 232-17, Atwood I Decl.; Mylan action, dkt. 232-18 to dkt. 232-19,
Ruffolo Decl.; Mylan action, dkt. 244 to dkt. 244-1, Mylan’s Responsive Br.; Mylan action, dkt.
245, Plaintiff’s Responsive Br.; Mylan action, 245-1 to 245-16, Stole II Decl., Mylan action, dkt.
246 to dkt. 246-8, Byrn II Decl.; Mylan action, dkt. 247 to 247-5, Beel II Decl.; Mylan action,
dkt. 247-6, Atwood II Decl.; Mylan action, dkt. 247-7, Ruffolo II Decl.; Mylan action, dkt. 269,
Mylan’s Reply Br.; Mylan action, dkt. 269-1 to dkt. 269-2, Atwood III Decl.

20

To determine the meaning of the claims, courts start by considering the intrinsic
evidence. Phillips, 415 F.3d at 1313; C.R. Bard, Inc. v. U.S. Surgical Corp., 388 F.3d
858, 861 (Fed. Cir. 2004); Bell Atl. Network Servs., Inc. v. Covad Commc’ns Group,
Inc., 262 F.3d 1258, 1267 (Fed. Cir. 2001). The intrinsic evidence includes the claims
themselves, the specification, and the prosecution history. Phillips, 415 F.3d at 1314;
C.R. Bard, Inc., 388 F.3d at 861.
The claims themselves provide substantial guidance in determining the meaning of
particular claim terms. Phillips, 415 F.3d at 1314. First, the context in which a term is
used in the asserted claim can be very instructive. Id. Other asserted or unasserted
claims can aid in determining the claim’s meaning because claim terms are normally used
consistently throughout a patent. Id. Differences among claims can also assist in
understanding a term’s meaning. Id. For example, when a dependent claim adds a
limitation, there is a presumption that the independent claim does not include that
limitation. Id. at 1314-15.
“[C]laims ‘must be read in view of the specification of which they are a part.’” Id.
at 1315 (quoting Markman, 52 F.3d at 979). “[T]he specification ‘is always highly
relevant to the claim construction analysis. Usually, it is dispositive; it is the single best
guide to the meaning of a disputed term.’” Id. (quoting Vitronics Corp. v. Conceptronic,
Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996)). This is true because a patentee may define his
own terms, give a claim term a different meaning than the term would otherwise possess,
or disclaim or disavow the claim scope. Id. at 1316. In these circumstances, the
inventor’s lexicography governs. Id. The specification may also resolve the meaning of
21

ambiguous claim terms “where the ordinary and accustomed meaning of the words used
in the claims lack sufficient clarity to permit the scope of the claim to be ascertained from
the words alone.” Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1325 (Fed. Cir.
2002). But, “[a]lthough the specification may aid the court in interpreting the meaning of
disputed claim language, particular embodiments and examples appearing in the
specification will not generally be read into the claims.” Comark Commc’ns, Inc. v.
Harris Corp., 156 F.3d 1182, 1187 (Fed. Cir. 1998) (quoting Constant v. Advanced
Micro-Devices, Inc., 848 F.2d 1560, 1571 (Fed. Cir. 1988)); accord Phillips, 415 F.3d at
1323.
The prosecution history is another tool to supply the proper context for claim
construction. It “can often inform the meaning of the claim language by demonstrating
how the inventor understood the invention and whether the inventor limited the invention
in the course of prosecution, making the claim scope narrower that it would otherwise
be.” Phillips, 415 F.3d at 1317.
“Extrinsic evidence consists of all evidence external to the patent and prosecution
history, including expert and inventor testimony, dictionaries, and learned treatises.”
Markman, 52 F.3d at 980. Although extrinsic evidence can be useful, it is “less
significant than the intrinsic record in determining ‘the legally operative meaning of
claim language.’” Phillips, 415 F.3d at 1317 (quoting C.R. Bard, Inc., 388 F.3d at 862).
Dictionaries and treatises may aid a court in understanding the underlying
technology and the manner in which one skilled in the art might use claim terms, but
dictionaries and treatises may provide definitions that are too broad or may not be
22

indicative of how the term is used in the patent. Id. at 1318. Similarly, expert testimony
may aid a court in understanding the underlying technology and determining the
particular meaning of a term in the pertinent field, but an expert’s conclusory,
unsupported assertions as to a term’s definition are entirely unhelpful to a court. Id.
Generally, extrinsic evidence is “less reliable than the patent and its prosecution
history in determining how to read claim terms.” Id. The Supreme Court recently
explained the role of extrinsic evidence in claim construction:
In some cases, however, the district court will need to look
beyond the patent's intrinsic evidence and to consult extrinsic
evidence in order to understand, for example, the background
science or the meaning of a term in the relevant art during the
relevant time period. See, e.g., Seymour v. Osborne, 11 Wall.
516, 546, 20 L.Ed. 33 (1871) (a patent may be “so interspersed
with technical terms and terms of art that the testimony of
scientific witnesses is indispensable to a correct understanding
of its meaning”). In cases where those subsidiary facts are in
dispute, courts will need to make subsidiary factual findings
about that extrinsic evidence. These are the “evidentiary
underpinnings” of claim construction that we discussed in
Markman, and this subsidiary factfinding must be reviewed for
clear error on appeal.
Teva Pharm. USA, Inc. v. Sandoz, Inc., 135 S. Ct. 831, 841 (2015).
As is sometimes the case, the claim terms requiring construction may have been
the subject of prior claim construction proceedings by other district courts, or even the
same court. While rulings of the Federal Circuit on issues of claim construction for a
given patent are binding on later district courts analyzing the same patent, interpretations
of the same or other district courts of the same terms in the patent or patent family are
generally not binding. See Shire Dev. LLC v. Amneal Pharms. LLC, No. 15-2865, 2016

23

WL 4119940, at *2 (D.N.J. Aug. 2, 2016); see also Ravo v. Tyco Healthcare Group LP,
No. 11-1637, 2013 WL 3326657, at *6 (W.D. Pa. Mar. 13, 2013). However, the
interpretations of the same or other district courts are generally considered to be highly
relevant and persuasive authority. Id.
Overall, in construing the claims, “[t]he judge’s task is not to decide which of the
adversaries is correct. Instead, the judge must independently assess the claims, the
specification, and if necessary the prosecution history and relevant extrinsic evidence,
and declare the meaning of the claims.” Exxon Chem. Patents, Inc. v. Lubrizol Corp., 64
F.3d 1553, 1556 (Fed. Cir. 1995).
B.

Application
1.

Summary of intrinsic evidence

In this section, we will summarize the intrinsic evidence of record, i.e., the
language of the claims, the common written description, and the prosecution history of
the ‘085 and ‘070 Patents.
a.

Relevant claim terms

As discussed above in Section I.C, supra, both of the disputed claim terms of the
‘085 Patent appear for the first time in independent claim 1. Claim 1 of the ‘085 Patent,
with the disputed claim terms highlighted, states:
Claim 1.
The magnesium salt of S-omeprazole trihydrate,
wherein the compound is characterized by the following major peaks in
its X-ray diffractogram:

24

(‘085 Pat., col. 10, ll. 15-34.) AstraZeneca admits that claim 1 of the ‘085 Patent is
limited to a specific form of the magnesium salt of S-omeprazole magnesium
characterized by certain X-ray diffraction “d-values,” which are indicative of crystalline
material. (Dkt. 39 at 8-9.) Dependent claims 2-12 recite either further limitations to
claim 1, a process for preparing the claimed product, a pharmaceutical composition
comprising the claimed product, or a method for treating a gastric acid related condition
with the claimed product. (‘085 Pat., col. 10, l. 2 – col. 12, l. 6.)22 The claim term
“characterized by the following major peaks in its X-ray diffractogram” appears only in
claim 1 of the ‘085 Patent.
As discussed above in Section I.C, supra, both of the disputed claim terms of the
‘070 Patent appear for the first time in claims 1 and 2. Claims 1 and 2 of the ‘070 Patent,
with the disputed claim terms highlighted, state:
Claim 1.

The magnesium salt of S-omeprazole trihydrate.

As discussed above, claims 2-5, and 11-12 of the ‘085 Patent also include the claim
term “the magnesium salt of S-omeprazole trihydrate.” See n. 8, supra.
22

25

Claim 2.
The magnesium salt of S-omeprazole trihydrate
according to claim 1 represented by FIG. 1.
(‘070 Pat., col. 10, ll. 51-54.) Dependent claims 3 and 4 each describe a process for
preparing the magnesium salt of S-omeprazole trihydrate. (See ‘070 Pat., col. 10, ll. 5562.)23 The claim term “represented by FIG. 1” appears only in claim 2 of the ‘070 Patent.
Claim 4 depends from claim 2, and incorporates the disputed term “represented by
FIG.1.”
The claim term “the magnesium salt of S-omeprazole trihydrate” appears in each
of the relevant claims of both the ‘085 and ‘070 Patents; but it stands alone in claim 1 of
the ‘070 Patent without further limitation. This is where most of the disagreement
between the parties lies. AstraZeneca argues that, under the principles of claim
differentiation, the later-issued claim 1 of the ‘070 Patent was issued without any
restrictions as to form, and thus should not be limited to a crystalline form – as claim 1 of
the ‘085 Patent admittedly is. (See dkt. 39 at 8-10.) Andrx and Perrigo both argue that
claim differentiation does not require such a result because claim 1 of the ‘070 Patent can
be broader than its dependent claim 2 and even claim 1 of the ‘085 Patent, while still
being limited to a crystalline form. (See dkt. 38 at 18-19.)24

Claims 3 and 4 also include the claim term “the magnesium salt of S-omeprazole
trihydrate.” See n. 10, supra.
23

24

(See Perrigo action, dkt. 50 at 22-23.)

26

b.

Common ‘085 and ‘070 Patent written description

The ‘085 and ‘070 Patents share a common written description.25 The written
description varies only in the description of related applications. (Compare ‘085 Pat., col.
1, ll. 3-4, with ‘070 Pat., col. 1, ll. 3-8.) The patent figures of the ‘085 and ‘070 Patents,
which are referenced in the written description, are identical.
In the Background section, the written description begins by describing the state of
the art at the time of the invention. The inventors acknowledge that omeprazole and
therapeutically acceptable salts thereof, including specific alkaline salts, have been
disclosed in the prior art. (‘070 Pat., col. 1, ll. 25-34.) The inventors also acknowledge
that the (+)-enantiomer of the magnesium salt of omeprazole and (-)-enantiomer of the
magnesium salt of omeprazole were known, and found to have an R and S configuration,
respectively. (Id., col. 1, ll. 35-49.) Certain salts of single enantiomers of omeprazole
(i.e., salts of S-omeprazole and R-omeprazole) were also known to have “improved
pharmacokinetic and metabolic properties” that provide an “improved therapeutic profile
such as a lower degree of interindividual variation.” (Id., col. 1, ll. 50-54.) Processes for
the preparation of single enantiomers of omeprazole and certain salts of single
enantiomers of omeprazole were also known in the prior art, as were suitable tablet
dosage forms of magnesium salts of the single enantiomers of omeprazole. (Id., col. 1, ll.
55-58.)

25

When describing the contents of the common written description, we cite only the
written description of the ‘070 Patent for convenience. We may also refer to the common written
description as only “the written description” for convenience.

27

The inventors describe their invention as a novel form of S-omeprazole,
specifically “a novel form of the magnesium salt of the S-enantiomer of omeprazole
trihydrate.” (Id., col. 1, ll. 11-16.) Their invention also includes processes for preparing
the novel form as well as pharmaceutical compositions containing it. (Id., col. 1, ll. 1618.)
The portions of the written description that the parties and their respective experts
argue shape the construction of the claim terms appear in column two of the ‘070 Patent,
and are quoted here in full:
It has surprisingly been found that the magnesium salt of Someprazole occurs in a number of structurally different forms.
It is an object of the present invention to provide a substantially
pure magnesium salt of S-omeprazole trihydrate, hereinafter
referred to as the compound of the invention. This trihydrate
can be obtained as a well defined compound. The present
invention also provides a process to obtain and a method of
differentiating the magnesium salt of S-omeprazole trihydrate
from other forms of magnesium salts of S-omeprazole.
The compound of the invention is advantageous because it is
more stable than the corresponding magnesium salt
compounds in prior art and is therefore easier to handle and
store. The compound of the invention is also easier to
characterize because it exists in a well defined state.
Additionally, the compound of the invention is easier to
synthesize in a reproducible manner and thereby easier to
handle in a full scale production.
The magnesium salt of S-omeprazole trihydrate obtained
according to the present invention is substantially free from
magnesium salts of R-omeprazole. The magnesium salt of Someprazole trihydrate obtained according to the present
invention is also substantially free from other forms of
magnesium salts of S-omeprazole, such as the corresponding
magnesium salt compounds described in prior art, and
28

dihydrates used in the preparation of the trihydrate compound
according to the present invention.
The compound of the invention is characterized by the
positions and intensities of the major peaks in the X-ray
powder diffractogram, but may also be characterized by
conventional FT-IR spectroscopy. These characteristics are not
exhibited by any other form of magnesium salt of Someprazole and accordingly, the magnesium salt of Someprazole trihydrate is easily distinguishable from any other
crystal form of the magnesium salt of S-omeprazole disclosed
in prior art. The compound of the invention is characterized by
being highly crystalline, i.e. having a higher crystallinity than
any other form of magnesium salt of S-omeprazole disclosed
in the prior art. With the expression “any other form” is meant
anhydrates, hydrates, solvates, and polymorphs or amorphous
forms thereof disclosed in the prior art. Examples of any other
forms of magnesium salt of S-omeprazole includes [sic], but
are not limited to, anhydrates, monohydrates, dihydrates,
sesquihydrates, trihydrates, alcoholates, such as methanolates
and ethanolates, and polymorphs or amorphous forms thereof.
The compound of the invention may also be characterized by
its unit cell.
(‘070 Pat., col. 2, ll. 15-64.) Generally, plaintiff argues that these paragraphs describe
examples and embodiments and are not intended to limit the scope of the claims of either
patent. Defendants, on the other hand, view some of these statements as describing the
claimed invention, and argue they are thus limiting.
The parties also repeatedly refer to the following language found at column 3,
lines 54-60:
The compound of the invention, i.e. the magnesium salt of Someprazole trihydrate, prepared according to the present
invention may be analyzed by XRPD, a technique which is
known per se. The amount of water in the magnesium salt of
S-omeprazole trihydrate is determined by thermogravimetric
analysis, a technique which is known per se.
29

(‘070 Pat., col. 3, ll. 54-60.)
The written description provides multiple examples that “illustrate the preparation
of the compound of the invention, according to different process routes and including
new intermediates.” (See ‘070 Pat., col. 5, l. 7 – col. 10, l. 50.) Examples 1, 4 and 7 are
most relevant here.26 The portion of the written description that pertains to Example 1 is
reproduced below:
Example 1
S-5-methoxy-2-[[(4-methoxy-3,5-dimethyl-2-pyridinyl)methyl]sulfinyl]-1H-benzimidazole Magnesium Salt
Trihydrate
Water (157 kg) was added to the wet crystals of the magnesium
salt of S-omeprazole, prepared according to Example 4, below.
The mixture was heated to 38° C. with stirring and left for 3
hours. The crystals were filtered off and dried in vacuo. Yield:
31.6 kg.
X-ray powder diffraction analysis was performed on a sample
of the crystals prepared above according to standard methods,
which can be found in e.g. Kitaigorodsky, A. I. (1973),
Molecular Crystals and Molecules, Academic Press, New
York; Bunn, C. W. (1948), Chemical Crystallography,
Clarendon Press, London; or Klug, H. P. & Alexander, L. E.
(1974), X-Ray Diffraction Procedures, John Wiley and Sons,
New York. The analysis gave the diffractogram depicted in
FIG. 1. The main peaks, with positions and relative intensities,
have been extracted from the diffractogram in FIG. 1 and is
given below in table 1. The relative intensities are less reliable

26

The additional examples are directed to processes for the preparation of other
compounds, (see ‘070 Pat., col. 6, l. 11 – col. 7, l. 18; col. 7, l. 33 – col. 9, l. 27), however those
additional compounds need not be discussed in depth here. The additional compounds described
are: the potassium salt of S-omeprazole (examples 2 and 3), and the magnesium salt of Someprazole dihydrate (examples 5 and 6).

30

and instead of numerical values the following definitions are
used.

Some additional very weak peaks found in the diffractogram
have been omitted from table 1.

(‘070 Pat., col. 5, l. 15 – col. 6, l. 10.)
As noted above, an X-ray powder diffraction analysis was performed on a sample
of the crystals prepared according to Example 1. (‘070 Pat., col. 5, ll. 26-33.) The
written description explains that this X-ray powder diffraction analysis “gave the
diffractogram depicted in FIG. 1 [reproduced below].” (‘070 Pat., col. 5, ll. 33-34.)

31

(‘070 Pat., Fig. 1.)
During the Markman hearing, the parties explained that Figure 1 and Table 1
depicted the same test data, even though each of the 13 peaks set forth in Table 1 are not
visible to the naked eye in Figure 1. (Dkt. 82 at 95-97.)
The product yielded by Example 7 was also analyzed using X-ray powder
diffraction. (‘070 Pat., col. 9, ll. 48-50.) The written description indicates that the X-ray
powder diffraction result for this example “complies with FIG. 1 and Table 1.” (‘070 Pat.,
col. 9, ll. 48-50.) However, there is no further elaboration in the written description
regarding the language “complies with.” The portion of the written description that
pertains to Example 7 is reproduced below:
Example 7

32

S-5-methoxy-2-[[(4-methoxy-3,5-dimethyl-2-pyridinyl)methyl]sulfinyl]-1H-benzimidazole Magnesium Salt
Trihydrate
22.0 g (29,1 mmol) of S-5-methoxy-2-[[(4-methoxy-3,5dimethyl-2-pyridinyl)
methyl]sulfinyl]-1H-benzimidazole
potassium salt was dissolved in 40 mL of water. The solution
was seeded with 0.11 g (0,1 mmol) S-5-methoxy-2-[[(4methoxy-3,5-dimethyl-2-pyridinyl)-methyl]sulfinyl]-1Hbenzimidazole magnesium salt trihydrate. 22 mL (69,6 mmol)
of MgSO4 (aq) was added under a 3 h period. The slurry was
filtered off and the precipitate was elutriated in water for
approximately 30 minutes and the crystals were filtered off and
dried (35° C., vacuum).
Yield: 9.15 g (11,6 mmol; 80%). The substance had a purity
(HPLC): 99.8 area %, Mg content: 3.40% (w/w) and ee: 99.8%.
The product was analyzed using X-ray powder diffraction and
the result complies with FIG. 1 and Table 1.
(‘070 Pat., col. 9, ll. 30-50.)
Example 4 teaches a process for the preparation of the magnesium salt of Someprazole. (See ‘070 Pat., col. 7, ll. 19-32.) Example 1 notes that the magnesium salt
of S-omeprazole prepared according to Example 4 is used as a precursor in Example 1.
(Id. at col. 5, ll. 13-15.)
The written description concludes with a description of a prior art method of
preparing a single enantiomer of omeprazole magnesium salt, identified as “Reference
Example A.” (See ‘070 Pat., col. 9, ll. 14 – col. 10, ll. 13.) As noted, the method used is
in accordance with the method described in Example A in International Patent
Publication WO 96/01623. That method discusses increasing the optical purity of the
magnesium salt via crystallization. (See ‘070 Pat., col. 9, ll. 31-47.)

33

As noted above, the parties each interpret the written description somewhat
differently and argue that it supports their differing constructions. With respect to the
claim term “the magnesium salt of S-omeprazole trihydrate,” AstraZeneca argues that
although the written description refers to crystalline and highly crystalline embodiments,
it also describes “the magnesium salt of S-omeprazole trihydrate” generally, with no
reference to crystallinity. (See dkt. 39 at 10-11.) AstraZeneca also points to the various
testing methods for characterizing a trihydrate disclosed in the written description,
arguing these methods do not require the trihydrate to be crystalline. (See id.)
AstraZeneca further argues that the written description provides no lexicography or
disavowal to alter the term’s plain and ordinary meaning. (See dkt. 48 at 16-23.)
Andrx argues that the written description shows that the claimed trihydrate is
crystalline and that the waters making it a trihydrate must be waters of crystallization.
Specifically, Andrx points to the multiple statements regarding the “compound of the
invention” found at column two of the ‘070 Patent. (See dkt. 38 at 13-14; dkt. 47 at 56.)27
Perrigo argues that statements found in column two of the written description also
define the term “the magnesium salt of S-omeprazole trihydrate” as a “bulk product” and
unequivocally disavow compositions containing substantial quantities of R-omeprazole

27

On this point, Perrigo and Mylan are in agreement with Andrx. (See Perrigo action,
dkt. 44 at 16-20; dkt. 50 at 19; Mylan action, dkt. 232 at 16-18; dkt. 244 at 8-12.)

34

magnesium and/or other physical forms of S-omeprazole magnesium, relative to any
trihydrate that may be present.28
AstraZeneca characterizes these statements identified by the defendants as
descriptions of claimed and unclaimed embodiments and examples. (See dkt. 48 at 1722.)29
Andrx also argues that the only examples provided in the written description for
making the claimed trihydrate (Examples 1 and 7) state that the product was obtained as
crystals, not any amorphous form. (See dkt. 38 at 14.)30 Thus, Andrx argues that forms
other than crystalline forms were not contemplated by the inventors. AstraZeneca
acknowledges the limited disclosure, but argues that defendants’ positions call for the
improper importation of limitations from the written description into the claims. (See
dkt. 48 at 22-23.)
With respect to the term “characterized by the following major peaks in its X-ray
diffractogram,” AstraZeneca argues that the written description supports its construction
because nothing in that written description requires the presence of every single peak

28

(See Perrigo action, dkt. 44 at 16-20; dkt. 50 at 12-13, 15-16). Mylan argued that
additional statements found in column two of the written description disavowed compounds that
were not substantially pure. (See Mylan action, dkt. 232 at 16; dkt. 244 at 12-13.)
29

Perrigo disagrees that the language AstraZeneca refers to as descriptions of claimed
and unclaimed embodiments, and argues that it should be seen as unequivocally limiting the
invention. (See Perrigo action, dkt. 50 at 17-18.)
30

Mylan addresses the same examples, but their argument is slightly different, they
consider the two crystalline examples to constitute a disclaimer of non-crystalline compounds.
(See Mylan action, dkt. 232 at 18-19.)

35

identified in Figure 1 or Table 1. (See dkt. 39 at 14-17.) AstraZeneca also argues that
the written description readily accounts for experimental error (an element of each of the
defendants’ proposed constructions) and inaccuracies in the claimed table of peaks. (See
id. at 16.) Andrx and Perrigo both argue that the description of Example 1 explains that
claim 1 of the ‘085 Patent is directed to a particular trihydrate form that exhibits exactly
the 13 peaks recited in claim 1’s table, and Andrx further argues that each peak must fall
into the specified relative intensity range. (See dkt. 38 at 22-23.)31
With respect to the claim term “represented by Fig. 1,” AstraZeneca argues that
the written description does not require the stringent limitation “the same as” as proposed
by each of the defendants – pointing to the “complies with” language in Example 7. (See
dkt. 39 at 17-18.)32 Perrigo and Andrx both argue that the written description fails to
explain what is meant by “complies with” and that there is no XRPD data or a
diffractogram for the product of Example 7, and that as a result, it should be interpreted
as the “same as Figure 1.” (See dkt. 38 at 28-29.)33

31

(See Perrigo action, dkt. 44 at 29-30; dkt. 50 at 26-27.)

Although AstraZeneca’s argument relies on the “complies with” language found in the
written description, Perrigo argues that AstraZeneca does not cite to any intrinsic evidence for its
construction of this term. (See Perrigo action, dkt. 50 at 31.)
32

33

(See Perrigo action, dkt. 44 at 32.)

36

c.

Prosecution history of the ‘085 Patent

The relevant portions of the prosecution history of the ‘085 Patent have been
supplied to the Court in the parties’ Markman submissions. (Dkt. 38-1 at 30-88; dkt. 397; dkt. 48-11 to dkt. 48-13.)34
U.S. Patent Application No. 09/077719 (the “‘719 Application”), which issued as
the ‘085 Patent, was filed on June 8, 1998 by inventors Hanna Cotton, Anders
Kronstrӧm, Anders Mattson, and Eva Mӧller (collectively “the Applicants”). (‘085 Pat.
at pg. 1.) The ‘719 Application is the National Stage Entry of International Application
No. PCT/SE98/00974, filed May 25, 1998, which claims priority to Swedish Patent
Application No. 9702065 filed May 30, 1997. (Id.) The ‘085 Patent was issued on April
9, 2002, so its prosecution process took approximately four years.
The docketed portions of the ‘085 Patent history begin in 1998 with the filing of
the ‘719 Application. The ‘719 Application, as originally filed, included 16 claims.
(Dkt. 38-1 at 53-57.) Claims 1-3 as originally filed are provided in the margin.35 Claim 1

34

(See also Perrigo action, dkt. 44-6 to dkt. 44-12; id., dkt. 45-7; id., dkt. 50-5; id., dkt.
51-11 to dkt. 51-13; Mylan action, dkt. 230-7; id., dkt. 232-2 at 3-41; id., dkt. 245-11 to dkt. 24513.) Some submissions in the Andrx, Perrigo, and Mylan actions were duplicative, whereas
some of the submissions were not. The Court will discuss the prosecution history of the ‘085
Patent chronologically, utilizing many of the non-duplicative submissions of the parties, in an
effort to paint a complete picture of the ‘085 Patent’s prosecution history.
35

Claim 1.

The magnesium salt of S-omeprazole trihydrate.

Claim 2.
The magnesium salt of S-omeprazole trihydrate according
to claim 1, characterized by being highly crystalline.
Claim 3.
The magnesium salt of S-omeprazole trihydrate according
to claim 1 characterized by the following major peaks in its X-ray powder
diffractogram:

37

of the ‘719 Application, as originally filed, is identical to issued claim 1 of the ‘070
Patent. This topic will be discussed in more detail in Section II.B.1.d, infra.
On November 27, 2000, in a first Office action, the examiner rejected claims 1-4,
14, 16 and 1736 of the ‘719 Application under 35 U.S.C. § 103(a) (obviousness) as being
unpatentable over:
1. U.S. Patent No. 4,738,974 (the “‘974 Patent”), example 5 (dihydrate form of
esomeprazole).
2. U.S. Patent No. 5,693,818 (the “‘818 Patent”), examples 3-7.
3. International Patent Application Publication WO 95/01977 (“WO 95/01977”),
page 3, lines 30-31 (for claim 4, see page 5, lines 7-8).
4. U.S. Patent No. 5,690,960 (the “‘960 Patent”), examples 2, 4.

(Dkt. 38-1 at 53).
36

Claims 5-13 and 18 were withdrawn from further consideration, as Applicants elected
to prosecute claims 1-4, 14, and 16-17 in response to a Restriction/Election Requirement. (See
dkt. 38-1 at 70-72.)

38

5. U.S. Patent No. 5,900,424 (the “‘424 Patent”).
(Dkt. 38-1 at 68-73.) The examiner also rejected the claims “optionally” in view of
Evans, An Introduction to Crystal Chemistry (1964) (“Evans”) or Japanese Chemical
Society, Experimental Chemical Seminar, Vol. 18 (1958) page 505 [sic] (“Japanese
Chemical Society”). (Id. at 71.) The examiner asserted that “[a]ll references (except
reference 1 [the ‘974 Patent – disclosing the dihydrate form]) disclose some form of
magnesium salt of omeprazole without clearly specify[ing] water content.” (Id.) The
examiner asserted that “[t]he difference between the claims and the references’ disclosure
is that the claimed compound is S-form and trihydrate.” (Id.) The examiner asserted that
because the prior art references (except the ‘974 Patent) are “silent regarding the water
content of the salt form, one having ordinary skilled [sic] in the art would have expected
the [prior] art compound in the process of preparing inherently has water molecule
attached.” (Id. at 72.) The examiner went on to state:
It is recognized in the art that organic solid material may
include guest molecules to form crystals. However, the guest
inclusion is mechanical in nature (see Evans p. 396), thus the
instant claims are drawn to alternative form of a known
compound. The alternative form is the innate nature for such
a compound. Furhtermore [sic], the Japanese reference teaches
that spontaneous resolution occurs in recrytlization [sic]. It is
expected that if the crystalization [sic] solvent is water, then
optical pure hydrate would result.
(Id.)
The examiner also rejected the pending claims for obvious-type double patenting
in view of the ‘424 patent, the ‘974 Patent, and U.S. Patent No. 5,714,504 (the “‘504
Patent”), and optionally in view of Evans or Japanese Chemical Society. (Id. at 72-73.)
39

The examiner took the position that the claims, although not identical, were not
patentably distinct from each other because alternative forms of a known substance
would be obvious. (See id. at 73.)
On April 26, 2001, the Applicants responded to the first Office action. (Dkt. 38-1
at 75-88; dkt. 39-7 at 2-5.) In their response, the Applicants provided the following
recitation with respect to the claimed invention:
The present invention is directed to a new and
advantageous form of the magnesium salt of the (-)-enantiomer
of omeprazole. As defined in the specification at page 3, line
3-9, the compound of the claimed invention is referred to as the
magnesium salt of S-omeprazole trihydrate (claim 1).
However, as used throughout this communication, the
expressions “the (-)-enantiomer of omeprazole” and “Someprazole” will be used interchangeably. The claimed
magnesium salt of S-omeprazole trihydrate is substantially free
from magnesium salts of R-omeprazole and other prior art
forms of magnesium salts of S-omeprazole, e.g., S-omeprazole
magnesium dihydrate.
At the time of the invention, it was not known that
crystalline magnesium salt of the (-)-enantiomer of omeprazole
occurred in structurally different forms.
The claimed
magnesium salt of S-omeprazole trihydrate is highly
crystalline (claim 2) and is uniquely characterized by an X-ray
powder diffractogram (claim 3) which distinguishes the
claimed compound from any other form of the magnesium salt
of S-omeprazole. As demonstrated by the comparative data set
forth in the accompanying Declaration of Frans W. Langkilde
(the “Declaration”), the claimed compound is more stable than
the corresponding magnesium salt compounds of the prior art
(claim 17).
Applicants have surprisingly discovered that the
claimed magnesium salt of S-omeprazole trihydrate is obtained
by treating a magnesium salt of S-omeprazole of any other
form, e.g., anhydrate, monohydrate, dihydrate, etc., with water
(claim 4). Notwithstanding the different methods . . . by which
40

the claimed compound is produced, Examples 1 and 7 of the
specification show that the product of the different methods is
characterized by the same X-ray powder diffractogram. In this
regard, the Examiner’s attention is directed to the following
support: Example 1, Table 1 at page 9; Example 7 at page 17,
lines 4-5; and Figure 1. In contrast and disadvantageously, the
same degree of reproducibility is not obtained with the prior art
dihydrate compound as shown in Examples 5 and 6 (See
Figures 3 and 4).
The demonstrated reproducibility and improved
stability characterizing the claimed magnesium salt of Someprazole trihydrate renders the claimed compound
particularly suitable as the active ingredient in full scale
manufacturing processes of pharmaceutical formulations for
treating gastric acid related conditions.
(Dkt. 38-1 at 79-80.) The Applicants also submitted a declaration of Dr. Frans W.
Langkilde37 (the “Langkilde Declaration”). (Dkt. 39-7 at 2-5.)
The Langkilde Declaration provided the following description of the claimed
invention:
Specifically, the invention is directed to a new and
advantageous form of the magnesium salt of S-omeprazole. As
stated in the specification at page 3, line 3-9, the compound of
the claimed invention is referred to as the magnesium salt of Someprazole trihydrate. The claimed magnesium salt of Someprazole trihydrate is substantially free from magnesium
salts of R-omeprazole and other prior art forms of magnesium
salts of S-omeprazole.
(Dkt. 39-7 at 3.) The Langkilde Declaration also described a side-by-side comparison of
certain physical chemical properties of “the claimed magnesium salt of S-omeprazole

Dr. Frans W. Langkilde, Ph.D. is not an inventor of the ‘085 or ‘070 Patents. At the
time of his declaration, Dr. Langkilde was employed by AstraZeneca as Group Leader,
Analytical Development, Pharmaceutical and Analytical Research and Development. (Dkt. 39-7
at 3.)
37

41

trihydrate and S-omeprazole magnesium dihydrate of the prior art.” The Langkilde
Declaration concluded that the magnesium salt of S-omeprazole trihydrate was different
than the prior art S-omeprazole magnesium salt dihydrate:
The magnesium salt of S-omeprazole trihydrate is different
when compared to the prior art S-omeprazole magnesium salt
dihydrate. The solid state properties defining these forms are
different. The claimed trihydrate is the more stable substance
and, after storage of six (6) months or longer, is chemically
more pure than the prior art dihydrate form. Furthermore, at
the time the claimed invention was made, there was no
suggestion that the magnesium salt of S-omeprazole existed in
a trihydrate form. It was indeed surprising, therefore, to obtain
the claimed compound and determine that the claimed
magnesium salt of S-omeprazole trihydrate is more stable than
the corresponding prior art compounds, thereby rendering the
claimed trihydrate form more advantageous for use in
pharmaceutical formulations.
(Id. at 4.)
The Applicants asserted that the Langkilde Declaration “demonstate[d] that the
claimed compound is characterized by a superior and unexpected improvement in
stability.” (Dkt. 38-1 at 80.) Specifically, the Applicants stated that “[t]he comparative
data [provided in the Langkilde Declaration] shows that the claimed magnesium salt of Someprazole trihydrate is more stable after 6 and 12 months than the prior art magnesium
salt of S-omeprazole dihydrate. (Id.) The Applicants argued that the “improved physical
and chemical properties that characterize the claimed compound” were not disclosed in
the prior art and that the demonstrated improvement in stability represents a greater than
expected result. (Id. at 80-81.) The Applicants also stated, with reference to the

42

Langkilde Declaration, that “the magnesium salt of S-omeprazole trihydrate is different
when compared to the prior art magnesium salt of S-omeprazole dihydrate.” (Id. at 81.)
With respect to the obviousness rejection under 35 U.S.C. § 103(a), the Applicants
responded by arguing that none of the cited references (i.e., the ‘974 Patent, the ‘818
Patent, WO 95/01977, the ‘960 Patent, the ‘424 Patent, Evans, and Japanese Chemical
Society) “suggest that there could be other crystalline forms of the magnesium salt of Someprazole” or “suggest that the claimed compound could be obtained by treating a
magnesium salt of S-omeprazole of any other form with water.” (Id.) The Applicants
addressed and distinguished each of the cited references in turn. (See id. at 82-86.) The
Applicants also argued that “it was unpredictable at the time the claimed invention was
made whether the prior art compounds, when treated in an aqueous environment, would
inherently have added water as a guest molecule in the crystals as a result of such a
treatment.” (Id. at 82.) The Applicants added that the “cited references fail to suggest the
reproducibility and unexpected improvement in stability which characterize the claimed
magnesium salt of S-omeprazole trihydrate as demonstrated by the Examples and
[Langkilde] Declaration.” (Id.)
With respect to the obvious-type double patenting rejection, the Applicants
differentiated the ‘504 Patent and reiterated that the claimed compound was characterized
by “superior and unexpected” improvements not suggested by the prior art:
The ‘504 patent is a continuation-in-part of the cited ‘818
patent. The ‘504 patent claims the magnesium salt of the (-)enantiomer of omeprazole. However, the ‘504 patent does not
suggest that the magnesium salt of the (-)-enantiomer of
omeprazole could exist in another more advantageous form,
43

e.g., a trihydrate. Nor does the ‘504 patent suggest that the
claimed magnesium salt of S-omeprazole trihydrate is more
stable than the corresponding magnesium salt compounds of
the prior art. Moreover, the Examples and [Langkilde]
Declaration show that the claimed magnesium salt of Someprazole trihydrate is reproducible and more stable after 6
and 12 months than the prior art magnesium S-omeprazole
compound.
(Id. at 86-87.)
On July 5, 2001, in a second Office action, the examiner again rejected claims 1,
2, 4, 14, 16 and 17. (See dkt. 48-11 at 2-6.) Specifically, the examiner rejected claims 1,
2, 4, 14, 16-17 under 35 U.S.C. § 103 as being “unpatentable over the art of record for the
reasons of record.” (Id. at 4.) With respect to the Langkilde Declaration, the examiner
noted that it was carefully considered, but in the examiner’s point of view it only applied
to the “hydrate having the specific X-ray powder diffraction value as listed in claim 3.”
(Id.) The examiner suggested amending claim 1 to include the content of claim 3 to
overcome the 35 U.S.C. § 103 rejection. (Id.) The examiner also noted that the obvioustype double patenting rejection would be overcome by such amendment. (Id.)
In that same second Office action, the examiner also provisionally rejected claims
1, 2, 4, 14, 16-17 as obvious under 35 U.S.C. § 103(a), based upon a presumption of
future patenting of co-pending Application No. 09/690,044 (the “‘044 Application”).
(Id.)38 A provisional obvious-type double patenting rejection over the ‘044 Application
was also issued. (Id. at 5.)

As discussed in more detail in Section II.B.1.e, infra, the ‘044 Application is a part of a
related prior art patent family. The ‘044 Application issued as U.S. Patent No. 6,875,872.
38

44

On September 18, 2001, the Applicants responded to the second Office action.
(See dkt. 48-12 at 2-9.) The Applicants amended claim 1 to incorporate the embodiment
of claim 3, as suggested by the examiner.39 (Id. at 2, 5.) The Applicants did not dispute
the examiner’s assertion that the Langkilde Declaration was “only relevant to the claimed
compound having the X-ray powder diffractogram values as recited in claim 3.” (See Id.
at 6.) With respect to the rejections pertaining to the ‘044 Application, the Applicants
argued that those rejections were mooted by the amendment of claim 1. (Id.)
On November 16, 2001, claims 1-2, 4-9, 14, 16-17 and 19 (renumbered as 1-12)
were allowed. (See dkt. 48-13 at 2.) The examiner did not provide a statement of
reasons for allowance in the Notice of Allowance, (see id. at 2); however, it is clear that
the ‘719 Application was allowed based on the amendment of claim 1 to incorporate the
embodiment of claim 3. On April 9, 2002, the ‘719 Application issued as the ‘085
Patent. (‘085 Pat., pg. 1.)
As with the written description, the parties interpret the prosecution history of the
‘085 Patent differently, arguing that it supports each of their respective positions.
AstraZeneca argues that the prosecution history makes clear that at the time of the
invention, the trihydrate form of the magnesium salt of S-omeprazole was unknown – as
evidenced by the absence of any prior art rejections over such a form. (See dkt. 39 at 11.)
Perrigo disagrees, interpreting the November 27, 2000 Office action as including a prior

39

The Applicants also amended claims 4, 5, 9, 14 and 16 to correct the recitation of
claims from which they depended, canceled claims 3 and 18, and added new claim 19. (Id. at 35.)

45

art rejection and an acknowledgement by the examiner that the magnesium salt of Someprazole in the prior art contained at least some trihydrate.40 Perrigo argues that the
presence of some trihydrate in the prior art required the Applicants to respond to the first
Office action and define “the magnesium salt of S-omeprazole trihydrate” as a bulk
product, substantially free from R-omeprazole magnesium and other forms of Someprazole magnesium.41 Perrigo also argues that the Applicants’ remarks and
arguments in response to the first Office action characterized the claimed trihydrate as
crystalline.42 Andrx likewise argues that the Applicants’ remarks and arguments in
response to the first Office action presume the crystallinity of the trihydrate. (See dkt. 38
at 14-15.)43
With respect to the claim term “characterized by the following major peaks in its
X-ray diffractogram,” Andrx argues that the prosecution history of the ‘085 Patent
supports its proposed construction. Specifically, Andrx points to a statement by the
Applicants in response to the first Office action, noting that Examples 1 and 7 show that
the product is characterized by the same X-ray powder diffractogram. (See dkt. 38 at 23.)

40

(See Perrigo action, dkt. 44 at 22, 24-27; dkt. 50 at 14-15.)

41

(See Perrigo action, dkt. 50 at 15.)

42

(See Perrigo action, dkt. 50 at 19-20.)

43

Mylan devotes substantial attention to this first Office action response as well. Mylan
agrees with Andrx and Perrigo regarding the Applicants’ emphasis of the crystallinity
requirement. (See Mylan action, dkt. 232 at 19-20; dkt. 244 at 13-14.) Mylan also argues that
the Applicants’ remarks and arguments in that Office action response support its construction
that the claimed compound be “substantially pure.” (See Mylan action, dkt. 232 at 19-20; dkt.
244 at 13-14.)

46

Andrx argues that this statement amounts to an understanding by the Applicants that the
term “complies with” and “characterized by” are interchangeable. (Id.) Perrigo argues
that the Applicants’ response to the first Office action relied on the XRPD to distinguish
the claimed invention from the prior art, and as a result, the term at issue should be
required to include each of the major peaks.44, 45
d.

Prosecution history of the ‘070 Patent

The relevant portions of the prosecution history of the ‘070 Patent have been
supplied to the Court in the parties’ Markman submissions. (Dkt. 38-1 at 90-168; dkt.
39-8 to dkt. 39-10; dkt. 48-14 to dkt. 48-16.)46
U.S. Patent Application No. 10/672,936 (the “‘936 Application”) was filed on
September 25, 2003 and issued as the ‘070 Patent on August 12, 2008. (‘070 Pat. at pg.
1.) The ‘936 Application is a continuation application of U.S. Patent Application No.
10/076,711, filed February 14, 2002, which is a divisional application of the ‘719
Application.
The docketed portions of the ‘070 Patent history begin in 2003 with the
filing of the ‘936 Application. A Preliminary Amendment was filed concurrently with

44

(See Perrigo action, dkt. 44 at 30; dkt. 50 at 27-28.) Perrigo also puts forth essentially
the same argument and relies on the prosecution history of the ‘085 Patent for its construction of
the claim term “Represented by FIG. 1” in the ‘070 Patent. (See Perrigo action, dkt. 44 at 32-33.)
Mylan’s argument was consistent with Perrigo’s on this point. (See also Mylan action,
dkt. 244 at 30-32.)
45

46

(See also Perrigo action, dkt. 44-13 to dkt. 44-22; id., dkt. 45-8 to dkt. 45-10; id., dkt.
51-14 to dkt. 51-16; Mylan action, dkt. 230-8 to 230-10; id., dkt. 232-2 at 43-96; id., dkt. 245-14
to dkt. 245-16; id., dkt. 247-5.)

47

the ‘936 Application, leaving five claims for prosecution.47 Those five claims are
provided in the margin.48 The Applicants49 included the following remarks regarding the
preliminary amendment to the claims and their relationship to the ‘719 Application:
The claimed matter of the subject application is directed
to those embodiments which were cancelled in the grandparent
application ‘719 application [sic] for the sake of advancing the
‘719 application to allowance. Specifically, during the
examination of the ‘719 application, the claims were limited to
the embodiment or [sic: of] original claim 3. The ‘719
application thereafter matured into the ‘085 patent.
Applicants submit that the limitation of the claimed
subject matter of the ‘719 application to the embodiment of
original claim 3 was not done in acquiescence of any objection
or rejection relating to patentability of the canceled claims or
deleted embodiments. Rather, the claims of the ‘719
application were so amended to advance the ‘719 application
to allowance so that Applicants could enjoy the benefits,

47

48

(Perrigo action, dkt. 44-13 at 2-6.)
Claim 1.

The magnesium salt of S-omeprazole trihydrate.

Claim 2.
The magnesium salt of S-omeprazole trihydrate according
to claim 1, wherein the compound is in a highly crystalline form.
Claim 14.
A pharmaceutical composition comprising the magnesium
salt of S-omeprazole trihydrate according to claim 1 or 2 as active ingredient in
association with a pharmaceutically acceptable carrier.
Claim 17.
The pharmaceutical composition according to claim 14
further comprising one or more therapeutically active ingredients.
Claim 18.
A method of treating a gastric acid related condition which
method comprises administering to a subject suffering from said condition a
therapeutically effective amount of the magnesium salt of S-omeprazole
trihydrate according to claim 1 or 2.
(Perrigo action, id.)
The “Applicants” of the ‘936 Application were identical to the ‘719 Application, and
were inventors Hanna Cotton, Anders Kronstrӧm, Anders Mattson, and Eva Mӧller.
49

48

without delay, conferred by a U.S. patent for allowable subject
matter.
Therefore, by this continuation application, Applicants
wish to resume examination of the claimed subject matter of
the ‘719 application that was previously canceled or deleted.
Applicants submit that they are entitled to patent protection as
broadly supported by the application as originally filed. As
such, the pending claims of this continuation application are
directed to the magnesium salt of S-omeprazole trihydrate,
pharmaceutical compositions comprising the magnesium salt
of S-omeprazole trihydrate, a pharmaceutically acceptable
carrier and optionally one or more additional active ingredients
and methods of treating gastric related condition [sic] by
administering the claimed compounds.50
On July 16, 2004, in a first Office action, the examiner rejected claims 1, 2, 14, 17
and 18 under 35 U.S.C. § 103 (obviousness) as being unpatentable over International
Patent Application Publication WO 94/27988 to Lindberg (“Lindberg”).51 Specifically,
the examiner asserted that Lindberg disclosed the magnesium salt of S-omeprazole and
“improved pharmacokinetic and metabolic properties of pure enantiomers of
omeprazole,” but did not disclose a trihydrate form.52,53 The examiner concluded that “in
absence of some unexpected superior efficacy of instant trihydrate over the prior art
known magnesium salt of S-omeprazole, it would have been obvious to one skilled in the

50

(Perrigo action, dkt. 44-13 at 5-6.)

51

(Perrigo action, dkt. 44-14 at 3-6.)

52

(Id. at 5.)

53

As discussed in Section II.B.1.b, supra, the inventors acknowledged that the
magnesium salt of S-omeprazole was known in the prior art, as evidenced by the identification
and discussion of the Lindberg reference in the Background section of the written description.
See ‘070 Pat., col. 1, ll. 35-55.

49

art to prepare hydrates of magnesium salt of S-omeprazole without affecting its utility for
treating ulcer.”54 The examiner also rejected claims 1, 2, 14, 17 and 18 under 35 U.S.C. §
103 as being unpatentable over International Patent Application Publication WO
96/01623 to Bergstrand (“Bergstrand”) for reasons similar to that of Lindberg.55 The
examiner also rejected claims 1, 2, 14, 17 and 18 under 35 U.S.C. § 101 (statutory double
patenting) as claiming the same invention as that of claims 1, 2, 11 and 12 of the ‘085
Patent.56
On December 15, 2004, the Applicants responded to the first Office action.57 In
their response, the Applicants explained that: “The present invention is directed to the
magnesium salt of the (-)-enantiomer of omeprazole. As defined in the written
description at page 3, line 3-9, the compound of the clamed invention is referred to as the
magnesium salt of S-omeprazole trihydrate.”58 With respect to the obviousness
rejections, the Applicants responded by arguing that the claimed invention was nonobvious in view of Lindberg and Bergstrand because neither Lindberg or Bergstrand
suggested that another form of magnesium salt of S-omeprazole was possible.59 With

54

(Perrigo action, dkt. 44-14 at 5-6.)

55

(Id.) The Bergstrand reference was also identified and discussed by the inventors in
the Background section of the written description. (See ‘070 Pat., col. 1, ll. 55-58.)
56

(Id. at 6-7.)

57

(See Mylan action, dkt. 232-2 at 51-55.)

58

(Id. at 52.)

59

(Id. at 54.)

50

respect to the double patenting rejection under 35 U.S.C. § 101, the Applicants responded
that the scope of the claims of the ‘085 patent was different from the scope of the claims
of the ‘936 Application, as evidenced by the express language of the claims.60
On April 29, 2005, in a second Office action, the examiner maintained the
rejections of claims 1, 2, 14, 17 and 18 under 35 U.S.C. §§ 101 and 103.61 With respect
to those rejections, the examiner provided the following comments:
In regard to obviousness rejections, the examiner does not
agree with the applicants arguments also since the utility is
identical. The applicants did not provide any unexpected data
or superior activity of the instant trihydrate or crystalline form
as compared to known magnesium salt of S-omeprazole of the
two cited references. What is the evidence that the trihydrate
or crystalline form is maintained in the pharmaceutical
composition or following in vivo administration of magnesium
salt of S-omeprazole? In regard to double patenting rejection,
the examiner does not agree with the applicants arguments that
the scope is different. The applicants did not provide any
reasons or provided any evidence to show how the scope is
different? Are the major peaks in X-ray diffractogram
different in the instant magnesium salt of S-omeprazole as
compared to the cited patent?62
On October 28, 2005, the Applicants responded to the second Office action.63
With respect to the statutory double patenting rejection under 35 U.S.C. § 101, the
Applicants argued that “[c]laims 1, 2, 14 and 18 of the [‘719 Application] and claims 1,

60

(Id.)

61

(See Perrigo action, dkt. 44-15 at 2-6.)

62

(Id. at 4.)

63

(See Perrigo action, dkt. 44-16 at 2-9.)

51

2, 11 and 12 of the ‘085 patent are related as genus and species, respectively” and that as
a result there is no identity of invention between them.64 With respect to the obviousness
rejection, the Applicants argued that although the dihydrate of S-omeprazole magnesium
salt was known at the time of the claimed invention, the prior art did not suggest the
claimed trihydrate form of the magnesium salt of S-omeprazole.65 The Applicants also
argued that neither Lindberg nor Bergstrand suggested “the claimed magnesium salt of Someprazole trihydrate is more stable than the corresponding magnesium salt compounds
of the prior art” and that the “demonstrated improvement in stability represents a greater
than expected result.”66
On November 21, 2005, in a third Office action, the examiner again rejected
claims 1, 2, 14, 17, and 18, maintaining the obviousness and double patenting
rejections.67 With regard to the obviousness rejection, the examiner argued that the
utility was identical and the Applicants had failed to provide any data of superior activity
or physical properties that would distinguish the trihydrate from the dihydrate form. 68
With regard to the double patenting rejection, the examiner maintained that the pending

64

(Id. at 5-6.)

65

(Id. at 6-8.)

66

(Id. at 8-9.)

67

(See Perrigo action, dkt. 44-17 at 2-4.)

68

(Id.)

52

claims and cited ’085 Patent were the same, and asserted that the Applicants did not
provide any differentiation.69
On May 19, 2006, the Applicants responded to the third Office action, canceling
claims 14, 17 and 18 so that the only claims remaining were claims 1 and 2.70 The
Applicant summarized the content of an examiner interview, indicating that the examiner
clarified that the 35 U.S.C. § 103 (obviousness) rejection applied only to claims 14 and
17 (now moot, due to the cancelation of those claims).71 With respect to the statutory
double patenting rejection under 35 U.S.C. § 101, the Applicants argued that the act of
rewriting original claim 3 of the ‘085 patent in independent form during the prosecution
of the ‘085 Patent is an acknowledgement that the claim scope of pending claim 1 and
claim 1 of the ‘085 patent must be different and have a species and genus relationship.72
On August 10, 2006, in a fourth Office action, the Examiner maintained the
statutory double patenting rejection of claims 1 and 2, again asserting that claims 1 and 2
were identical to claim 1 of the ‘085 Patent. (Dkt. 39-8 at 2-5.) The Applicants appealed
the examiner’s decision. (See dkt. 39-9 at 2-11.) Applicants’ Appeal Brief provided the
following summary of the claimed subject matter:
Esomeprazole (S-omeprazole) is the (-)-enantiomer of 5methyoxy-2-[[(4-methoxy-3,5-dimethyl-2-pyridinyl)69

(Id.)

70

(See Mylan action, dkt. 232-2 at 69-73.)

71

(Id. at 71.)

72

(Id. at 71-72.)

53

methyl]sulfinyl]-1H-benzimidazole (specification at p. 1, lines
4-5).
Claim 1 is the only independent claim. As defined by claim 1,
the claimed invention is directed to the magnesium salt of Someprazole trihydrate (specification at p. 1, lines 6-7). At the
time the claimed invention was made, it was not known that
the magnesium salt of S-omeprazole occurs in a number of
structurally different forms (specification at p. 3, lines 3-4).
The compound of the invention is advantageous because it is
more stable than the corresponding magnesium salt
compounds of S-omeprazole in the prior art and, therefore, it
is easier to handle and store (specification at p. 3, lines 11-13).
(Id. at 4.) With respect to the statutory double patenting rejection under 35 U.S.C. § 101,
the Applicants again argued that a comparison of the prosecution history of the ‘085
Patent confirmed that pending claims 1 and 2 were different in scope than claims 1 and 2
of the ‘085 Patent. (Id. at 4-10.) The Applicants concluded that argument as follows:
The quoted statements from the prosecution history of the ‘085
patent evidence the PTO’s position and how the Examiner
understood the claimed invention. Specifically, the Examiner
understood there to be a difference in scope between original
claim 1 directed to the magnesium salt of S-omeprazole
trihydrate and original claim 3 referring back to and further
limiting claim 1. To advance the ‘719 application to allowance
and to enjoy the benefits conferred by a U.S. patent for
allowable subject matter, claim 1 of the ‘719 application was
amended to recite the X-ray powder diffraction values of claim
3 which was then canceled. The § 103 and obviousness-type
double patenting rejections were withdrawn and the ‘085
patent was granted.
Since it was the PTO’s position that original claim 3 of the ‘719
application referred back to and further limited the subject
matter of claim 1 of the ‘719 application in satisfaction of 35
U.S.C. § 112, ¶4, and the related regulation 37 C.F.R. § 1.75(c),
then for the same reason claim 1 of the ‘085 patent must be
limited or narrower in scope than claim 1 of the subject
application. The PTO cannot now change its position since
54

Applicants previously relied on the PTO and amended the
claims accordingly.
(Id. at 9.)
The appeal portion of the prosecution history abruptly ceased following the filing
of the appeal. The next event that appears in the docketed submissions is an August 14,
2007 Office action, the fifth Office action issued by the examiner. In that fifth Office
action, the examiner withdrew the statutory double patenting rejection under 35 U.S.C. §
101,73 but the examiner issued an obvious-type double patenting rejection of claims 1 and
2. (See dkt. 39-10 at 2-6.) The examiner also rejected claim 2 under 35 U.S.C. § 112 as
being indefinite because the meaning of the term “highly crystalline form” was allegedly
unclear. (Id. at 4.)74

73

The examiner noted that the Final Rejection was withdrawn in view of an appeals
conference in response to Applicants’ appeal brief, but the examiner did not provide details or a
summary of the appeals conference. (See dkt. 39-10 at 4.)
74

Specifically, the examiner stated:
In claim 2, the term --- highly crystalline form --- is indefinite since
its meaning is not clear. According to the specification on page 4,
first paragraph, this term is defined as --- having a higher
crystallinity than any other form of magnesium salt of S-omeprazole
disclosed in the prior art ----. This does not tell anything about the
crystallinity of instant trihydrate form since there is no comparative
data present as compared to prior art known forms including
trihydrate form (see line 5 page 4). Does the instant trihydrate form
has [sic] higher crystallinity as compared to the trihydrate form of
Pat 6,669,085?

(Dkt. 39-10 at 2-6.)
55

On September 21, 2007, the Applicants responded to the fifth Office action.75 The
Applicants added claims 20 and 21, which are provided in the margin.76 The Applicants
canceled claim 2, mooting the rejection under 35 U.S.C. § 112.77 To overcome the
obvious-type double patenting rejection, the Applicants filed a terminal disclaimer over
the ‘085 Patent.78
On December 6, 2007, the Applicants filed a supplemental amendment, amending
claim 20 to correct antecedent basis, canceling claim 21, and adding new claims 22 and
23, which are provided in the margin.79
On December 7, 2007, the examiner issued a notice of allowance, allowing claims
1, 20, and 21. (See dkt. 48-16 at 2-3.) It appears that the examiner issued this notice of
allowance prior to reviewing the contents of the supplemental amendment. On March 4,

75

(See Perrigo action, dkt. 44-19 at 2-6.)

76

Claim 20. The magnesium salt of S-omeprazole trihydrate according to claim 1
represented by the Figure 1.
Claim 21. A process for the preparation of the magnesium salt of S-omeprazole
trihydrate according to claim 1 or 20 which comprises treating a magnesium salt of Someprazole of any other form with water. (Perrigo action, dkt. 44-19 at 3.)
77

(Perrigo action, dkt. 44-19 at 3-4.)

78

(Id. at 4, 6.)

79

Claim 22. A process for the preparation of the magnesium salt of S-omeprazole
trihydrate according to claim 1 which comprises treating a magnesium salt of S-omeprazole of
any other form with water.
Claim 23. A process for the preparation of the magnesium salt of S-omeprazole
trihydrate according to claim 20 which comprises treating a magnesium salt of S-omeprazole of
any other form with water. (Perrigo action, dkt. 44-20 at 2-4.)

56

2008, the examiner issued a supplemental notice of allowability allowing claims 1, 20, 22
and 23 in response to the December 6, 2007 supplemental amendment.80 On August 12,
2008, the ‘936 Application issued as the ‘070 Patent. (‘070 Pat., pg. 1.)
Again, the parties each argue that the prosecution history of the ‘070 Patent
supports their respective positions. AstraZeneca argues that claim 1 of the ‘085 Patent
and claim 1 of the ‘070 Patent must have a different scope because AstraZeneca was able
to overcome the statutory double patenting rejection, arguing that claim 1 of the ‘070
Patent was different and broader than claim 1 of the ‘085 Patent. (See dkt. 39 at 11-12.)
Perrigo acknowledges this, but argues that the withdrawal of this rejection does not
warrant interpreting claim 1 of the ‘070 Patent beyond that allowed by the written
description or its prosecution history.81 AstraZeneca also argues that nothing in the
prosecution history of the ‘070 Patent constitutes a disavowal or disclaimer of the scope
of the term “the magnesium salt of S-omeprazole trihydrate” in claim 1 of the ‘070
Patent, and therefore AstraZeneca argues that any statements made in the prosecution
history of the ‘085 Patent should not limit the ‘070 Patent claims. (See dkt. 48 at 23-25.)
Andrx argues that the Applicants’ remarks and arguments in response to the
October 28, 2005 Office action were premised on the notion that the waters defining the
trihydrate were indeed waters of crystallization that conferred a unique structure to the
claimed trihydrate. (See dkt. 38 at 15.)

80

(See Perrigo action, dkt. 44-22 at 2-5.)

81

(See Perrigo action, dkt. 50 at 22-23.)

57

e.

Related prior art patent family

As discussed in Section II.B.1.c, supra, the ‘719 Application (which became the
‘085 Patent) was subject to an obvious-type double patenting rejection in view of
multiple patents, including the ‘504 Patent.82 The ‘504 Patent is assigned to Astrazeneca
AB, the same assignee as the ‘085 and ‘070 Patents. The ‘504 Patent is an expired patent
also covering the Nexium® product. The ‘504 Patent is a continuation-in-part of U.S.
Patent Application No. 08/256,174 filed June 28, 1994, which issued as U.S. Patent No.
5,693,818, which was the national stage entry of International Patent Application No.
PCT/SE94/00509, filed May 27, 1994, which claimed priority to Swedish Patent
Application No. 9301830-7. This family of patents83 also includes the following issued,
but expired patents: U.S. Patent Nos. 5,877,192 (the “‘192 Patent”), 6,875,872 (the “‘872
Patent”)84, and 6,143,771 (the “‘771 Patent”). (See dkt. 105.)
The ‘504 Patent claimed pharmaceutical formulations (and treatment methods)
comprising solid, chemically pure alkaline salts of the S-enantiomer of omeprazole and a
pharmaceutically acceptable carrier. This patent is further limited to “optically pure”
alkaline salts of the S-enantiomer of omeprazole (claim 2); specific alkaline salts, such as

A copy of the ‘504 patent was submitted by Mylan and can be found at Mylan action,
dkt. 232-10.
82

83

A diagram of this patent family tree illustrating the relationships among these patents
was prepared by the Court for convenience and may be found at dkt. 105.
A copy of the ‘872 patent was submitted by Mylan and can be found at Mylan action,
dkt. 232-11.
84

58

magnesium or sodium (claims 3 and 5); and salts in a “substantially crystalline form”
(claim 4). Claims 6 through 10 are directed to methods of using the claimed formulation.
The ‘192 Patent claimed methods for treating gastric acid related diseases with the
S-enantiomer of omeprazole to achieve certain biological benefits (claims 1 and 2), and a
method for the production of an S-enantiomer of omeprazole-containing medicament for
treating gastric diseases (claim 12). Other claims of the ‘192 Patent recite additional
improved properties of the S-enantiomer of omeprazole (claims 3-6 and 13-18), specific
administration methods and dosage amounts (claims 7-11 and 13-22), and an
enantiomeric purity limitation (claim 23).
The ‘872 Patent claimed the magnesium salts of the S-enantiomer of omeprazole
exhibiting high or very high optical purities of “at least about”: 94% (claim 1), 98.4%
(claim 4), 99.8% (claim 7), and 99.9% (claim 10), measured by enantiomeric excess
(commonly abbreviated as “ee”). Claims 2, 5, 8, and 11 are parallel to claims 1, 4, 7, and
10, only the “about” language is removed. Claims 3, 6, and 9 limit the independent
claims to those containing the magnesium salt of the S-enantiomer of omeprazole “in
crystalline form.”
The ‘771 Patent claimed pharmaceutical formulations comprising an optically
pure solid state sodium salt of the S-enantiomer of omeprazole as the active ingredient
(claims 1, 2) and methods of treating gastrointestinal inflammatory diseases and
conditions with the same (claims 6, 7, 11, 12). Claim 3 limits the sodium salt of the Senantiomer of omeprazole to a substantially crystalline form. Claims 4, 5, and 8-10
further limit the pharmaceutical formulation.
59

The Court notes that the ‘504, ‘192 and ‘872 Patents have been the subject of
extensive prior litigation in this Court and have been the subject of multiple claim
construction opinions. See, e.g., AstraZeneca AB v. Dr. Reddy’s Laboratories, Ltd., No.
05-5553, 2010 WL 1981790 (D.N.J. May 18, 2010) (construing terms of the ‘504, ‘192,
and ‘872 Patents); AstraZeneca AB v. Dr. Reddy’s Laboratories, Inc., Nos. 11-2317, 114275, 11-6348, 2013 WL 1847639 (D.N.J. May 1, 2013) (construing terms of the ‘504
and ‘872 Patents, among others); Astrazeneca AB v. Hanmi USA, Inc., No. 11-760, 2012
WL 6203602 (D.N.J. Dec. 12, 2012) (construing terms of the ‘504 and ‘192 Patents),
aff’d 554 Fed. Appx. 912 (Fed. Cir. 2013).
3.

Summary of extrinsic evidence

This subsection summarizes the expert positions and exhibits on claim
construction issues.85 Our analysis will follow in Section II.B.4. The organization of this
subsection is explained in the margin.86

85

To the degree expert witnesses addressed intrinsic evidence and stated their view as to
how a POSA would interpret the intrinsic evidence, it appears that the Court’s analysis of those
portions could be subject to clear error review, rather than de novo review. See Teva Pharm.
USA, Inc. v. Sandoz, Inc., 135 S. Ct. 831, 837-38 (2015). We express no view on that issue.
86

This subsection, Section II.B.3, summarizes the extrinsic claim construction evidence.
That evidence consists of filed submissions, including declarations and deposition testimony of
expert witnesses. Some portions of the docketed deposition testimony were filed by the
opposing party. (See, e.g., dkt. 47-1 at 4-25 (excerpts of deposition of plaintiff’s expert, filed by
defendant Andrx); dkt. 48-3 at 2-34 (excerpts of deposition of defendant Perrigo’s expert, filed
by plaintiff); dkt. 48-6 at 2-23 (excerpts of deposition of defendant Andrx’s expert, filed by
plaintiff).)
Plaintiff identifies one expert witness, Dr. Stephen Byrn. In each action, plaintiff filed
two declarations of Dr. Byrn. (Dkt. 40; dkt. 49; Perrigo action, dkt. 46; dkt. 52; Mylan action,
dkt. 231; dkt. 246.) Dr. Byrn’s first declaration in each case was nearly identical. (See dkt. 40;

60

a.

Plaintiff’s expert testimony and exhibits

Dr. Byrn was presented by plaintiff as an expert in the field of solid state
chemistry, including the analysis of the form, stability, solvates, and polymorphs of
drugs. His qualifications are summarized in the margin.87 Dr. Byrn provided testimony
on the claim terms “the magnesium salt of S-omeprazole trihydrate,” “characterized by
the following major peaks in its X-ray diffractogram,” and “represented by FIG. 1.”
i.

“The magnesium salt of S-omeprazole trihydrate”

To review, AstraZeneca proposes that the claim term “the magnesium salt of Someprazole trihydrate” be construed as follows: “a trihydrate of a magnesium salt of Someprazole containing three molecules of water per molecule of magnesium salt of S-

Perrigo action, dkt. 46; Mylan action, dkt. 231). Dr. Byrn’s second declaration in each case was
also nearly identical. (See dkt. 49; Perrigo action, dkt. 52; Mylan, action, dkt. 246)
Andrx identifies one expert witness, Dr. Michael Zaworotko. In the Andrx action, Andrx
filed three declarations of Dr. Zaworotko. (Dkt. 38-3 at 1-29; dkt. 47-2; dkt. 54-1 at 1-6.)
Perrigo identifies one expert witness, Dr. Graham Buckton. In the Perrigo action, Perrigo
filed two declarations of Dr. Buckton. (Perrigo action, dkt. 44-1; dkt. 50-1.)
Mylan identified two expert witnesses, Dr. Jerry Atwood and Dr. Robert R. Ruffolo. In
the Mylan action, Mylan filed three declarations of Dr. Atwood (Mylan action, dkt. 232-16; dkt.
247-6; dkt. 269-1), and two declarations of Dr. Ruffolo. (Id., dkt. 232-18; id., dkt. 247-7.)
The organization of this subsection is as follows: Section II.B.3.a summarizes the
evidence from plaintiff’s expert, Dr. Byrn. Section II.B.3.b summarizes the evidence from
Andrx’s expert, Dr. Zaworotko. Section II.B.3.c summarizes the evidence from Perrigo’s expert,
Dr. Buckton. A summary of the evidence from Mylan’s experts is not provided, but when
appropriate, reference is made to such evidence in the margin.
87

Dr. Byrn has qualifications as a university professor in various positions in the areas of
pharmaceutical and solid state chemistry since 1972. (Dkt. 40 at 2.) He has published two books
and over 170 articles on the subject of solid state chemistry. (Id.) He has also served as a
consultant to various entities within the pharmaceutical community. (Id. at 2-3.)

61

omeprazole.” (Dkt. 36 at 3.) Dr. Byrn declared that the POSA would understand the
phrase “the magnesium salt of S-omeprazole trihydrate,” as used in the claims of the ‘085
and ‘070 Patents, to have three characteristics: “a magnesium salt; of the S-omeprazole
enantiomer; and a trihydrate.” (Dkt. 40 at 6.) There appears to be no disagreement
among the experts regarding the term “a magnesium salt,” or the term “S-omeprazole,”
which Dr. Byrn explained means the S- or (-)-enantiomer of omeprazole. The
disagreement in this matter concerns the term “trihydrate” and its meaning to the POSA
at the time of the invention.
In Dr. Byrn’s opinion, the POSA would understand the term “trihydrate,” as
describing a compound having three molecules of bound water per every molecule of the
magnesium salt of S-omeprazole. (Id.) His definitions and understandings differ from
that of the other experts in that Dr. Byrn does not limit the trihydrate to a crystalline
form. Specifically, Dr. Byrn opined that a POSA would not understand the term “the
magnesium salt of S-omeprazole trihydrate” to be limited to a crystalline form. In the
context of claim 1 of the ‘070 Patent, he opined that the POSA would understand that
term to “cover esomeprazole magnesium trihydrate of any degree of crystallinity, and to
cover esomeprazole magnesium trihydrate in any amount, even if it is mixed with a
different form of esomeprazole magnesium.” (Id. at 7-8.)

62

It is Dr. Byrn’s opinion that a hydrate need not be crystalline, it must merely have
some structure that “can bind water in a regular way.” (Id.)88 He explained that the term
hydrate “differentiates a compound from other substances in which the water present is
loosely associated or only present on the surface.” (Id.) To support his opinion that the
term “hydrate” is not limited to a crystalline form, Dr. Byrn declared that the existence of
amorphous hydrates, in general, were well known at the time of the invention. (Id.)89 As
pointed out by Dr. Zaworotko, and discussed in more detail in Section II.B.3.b, infra, the
references cited by Dr. Byrn for this proposition are after the May 30, 1997 priority date

88

Dr. Byrn does acknowledge that hydrates may be crystalline, i.e., water may be
“bound” or incorporated into the crystal lattice of a drug substance’s unit cell. This example
originates from a paper he co-authored, Pharmaceutical Solids: A Strategic Approach to
Regulatory Considerations, 12 Pharmaceutical Research 945, 946 (1995) (dkt. 40-8). That paper
provides a description of polymorphs, hydrates (solvates), desolvated solvates, and amorphous
forms:
“Polymorphs exist when the drug substance crystallizes in different
crystal packing arrangements all of which have the same elemental
composition (Note that hydrates can exist in polymorphs). Hydrates
exist when the drug substance incorporates water in the crystal
lattice in either stoichiometric or non-stoichiometric amounts.
Desolvated solvates are produced when a solvate is desolvated
(either knowingly or unknowingly) and the crystal retains the
structure of the solvate. Amorphous forms exist when a solid with
no long range order and thus no crystallinity is produced.”
(Dkt. 40-8 at 3.)
89

The references cited by Dr. Byrn to support his opinion on this point were: Megarry, et
al., Amorphous trehalose dihydrate by cryogenic milling, 346 Carbohydrate Research 1061-64
(2011) (“Megarry”) (dkt. 40-9); U.S. Patent No. 7,638,634, titled “Amorphous Esomeprazole
Hydrate” (“the ‘634 Patent”) (dkt. 40-10); U.S. Patent No. 7,244,842, titled “Amorphous
Hydrate of a Cephalosporin Antibiotic” (“the ‘842 Patent”) (dkt. 40-11); and International Patent
Publication No. WO 04/020436, titled “Amorphous Hydrates of Esomeprazole Magnesium and
Process for the Preparation Thereof (“WO 04/02036”) (dkt. 40-12).

63

of the ‘085 and ‘070 Patents, and thus are not evidence of what was known at the time of
the invention. (See dkt. 47-2 at 4-6.) Dr. Byrn acknowledged this in his second
declaration, and cited six additional references which he asserted provide examples of
amorphous hydrates at the time of the invention. (See dkt. 49 at 6-7.)90 Dr. Zaworotko
again took issue with these additional references and opined that they are irrelevant to the
scientific field to which the ‘085 and ‘070 Patents pertain. See Section II.B.3.b, infra.
Dr. Byrn also cited two dictionary definitions for the term “trihydrate” to support his
opinion that the compound need not be limited to a crystalline form. (Id. at 8.)91 Neither
of those definitions include a definitional crystallinity requirement.
Beyond amorphous hydrates in general, Dr. Byrn also stated that amorphous forms
of the magnesium salt of S-omeprazole were known, citing International Patent

90

The additional references cited by Dr. Byrn were: Peter S. Belton, An 1H Pulsed N.m.r.
Study of Some Amorphous Silicate Hydrates, 29 J. Chem. Tech. Biotechnol. 19-25 (1979)
(“Belton”) (dkt. 49-2); Delzeit, et al., Infrared Spectra of HCl Complexed/Ionized in Amorphous
Hydrates and at Ice Surfaces in the 15-90 K Range, 97 J. Phys. Chem. 10312-10318 (1993)
(“Delzeit”) (dkt. 49-3); H. Neil McMurray, Hydrothermal Modification of Electrocatalytic and
Corrosion Properties in Nanosize Particles of Ruthenium Dioxide Hydrate, 4(8) J. Mater. Chem.,
1283-1287 (1994) (“McMurray”) (dkt. 49-4); Koch, et al., Low-Temperature Photochemistry of
Submicrometer Nitric Acid and Ammonium Nitrate Layers, 100 J. Phys. Chem. 11402-11407
(1996) (“Koch”) (dkt. 49-5); Kalceff, et al., Cathodoluminescence Microanalysis of Natural
Hydrated Amorphous SiO2; Opal, 24 Phys. Chem. Minerals 131-138 (1997) (“Kalceff”) (dkt. 496); U.S. Patent No. 5,395,602, titled “Method for the Production of Sodium Perborate Hydrate
Granulates (“the ‘602 Patent”) (dkt. 49-7).
The 2002 Merriam-Webster’s Medical Desk Dictionary defines “trihydrate” as “a
chemical compound with three molecules of water.” (Dkt. 40-14 at 5.)
91

The 1989 Oxford English Dictionary defines “trihydrate” as “[a] compound containing
three molecules of water combined with an element or radical or with another compound.” (Dkt.
40-15 at 4.)

64

Publication No. WO2004/020436. (See dkt. 40 at 7.) Andrx took issue with this
assertion, noting that the language “amorphous hydrates of S-omeprazole magnesium,”
appearing in this publication was error, acknowledged by both the USPTO and applicants
by way of certificate of correction filed in U.S. Patent No. 7,612,098. (See dkt. 86-1 at 45,10-22.) Specifically, the term “hydrate of” was replaced with “hydrous” throughout the
‘098 Patent. (See id. at 21-22.)
Dr. Byrn also opined that a POSA would understand that water can be present in a
regular arrangement (i.e., short range order) in solid magnesium compounds, without the
formation of a crystalline structure, by forming a hydrophilic layer that incorporates
water molecules. (Dkt. 40 at 7.)92 He also stated that processes for making a crystalline
material amorphous, such as milling, grinding, or granulating, were within the knowledge
of a POSA. (Id.) He explained that the POSA would understand these techniques could
be applied to the magnesium salt of S-omeprazole trihydrate to make it amorphous. (Id.)
Dr. Byrn cited intrinsic evidence of record (i.e., the written description, the
prosecution histories of the ‘085 and ‘070 Patents, and the Langkilde Declaration)
supporting his position. He explained that, based on his review of the prosecution
histories, the novelty of the inventions of both the ‘085 and ‘070 Patents was in the
discovery of the trihydrate form of the magnesium salt of S-omeprazole, which was not

92

As evidence, he cited T. Egami & S.J.L. Billinge, Underneath the Bragg Peaks:
Structural Analysis of Complex Materials, 12 (R.W. Cahn ed., 2003.)

65

limited to a specific state (i.e., crystalline or amorphous). (Id. at 8.)93 He acknowledged
a POSA would understand claim 1 of the ‘085 Patent, which is defined by a particular list
of X-ray powder diffraction peaks, to be narrower than claim 1 of the ‘070 Patent, but he
asserted that claim 1 of the ‘085 Patent is still not limited to form or amount. (Id. at 89.)94 He interpreted the written description’s statements regarding the “invention” and
the “compound of the invention” found at column 2 of the ‘070 Patent, discussed in
Section II.B.1.b, supra, as non-limiting. It was Dr. Byrn’s position that any language in
the patent referring to the invention being crystalline or highly crystalline would not be
understood by a POSA as inherent limitations in the claims because the POSA would
understand these statements as describing different example compounds and preferred
embodiments made pursuant to the specific examples set forth in the written description.
(See id. at 9-10.)
When discussing the opposing construction positions of the parties, Dr. Byrn again
emphasized that, in his opinion, a POSA would not understand non-crystalline forms of
the magnesium salt of S-omeprazole trihydrate to be excluded from claim 1 of the ‘070

For example, he pointed to the following statement in the Langkilde Declaration: “at
the time the claimed invention was made, there was no suggestion that the magnesium salt of Someprazole existed in a trihydrate form.” (Dkt. 40 at 8, citing dkt. 40-4 at 4.)
93

It is not clear to the Court what Dr. Byrn means by the term “form” in this context.
The Court notes that AstraZeneca admitted, in its written submissions, that claim 1 of the ‘085
patent is limited to a crystalline form. (See dkt. 39 at 8 (“The tabulated peaks of claim 1 are
indicative of crystalline material. So, Defendants’ attempt to read a separate crystallinity
limitation into ‘trihydrate’ would be redundant in the claims of the ‘085 patent, which are already
limited to crystalline forms.”).)
94

66

Patent. (See dkt. 49 at 5-8.) He explained that there is no sharp distinction between
crystalline and amorphous states:
A person of ordinary skill in the art would understand that “the
magnesium salt of S-omeprazole trihydrate” could be
crystalline or amorphous and could have a mixture of
crystalline and amorphous features. At the time of the
invention, it was well-understood by those skilled in the art that
solid state materials are rarely 100% amorphous or 100%
crystalline. They exist on what could be described as a
continuum, ranging from completely amorphous material
exhibiting complete disorder to very crystalline material
exhibiting three-dimensional long range order with few
ordering defects. Within this continuum, material can exist in
an intermediate state or mesophase. Mesophases lack three
dimensional long range order but can exhibit X-ray diffraction
peaks. There is no definitive dividing line between the
crystalline and amorphous states.
(Dkt. 49 at 5.)95

He asserted that the Haleblian text, cited by defendants’ experts,

acknowledged this point.96 He concluded that a POSA would understand that material with
Dr. Atwood disagreed with Dr. Byrn’s discussion regarding crystalline and amorphous
materials existing on a continuum. (See Mylan action, dkt. 269 at 6-7.) He cautioned that
describing a “continuum” is not the same as describing a mixture of crystalline and amorphous
solids in a composition of matter. (Id. at 6.) He asserted that the discussion of an “intermediate
state” or “mesophase” was out of context – as the POSA would understand mesophase as
referring to an intermediate state of matter between liquid and a solid crystal, not an amorphous
solid and a solid crystal. (Id. at 7.)
95

96

The portion of the Haleblian text Dr. Byrn refers to discusses the differences between
X-ray patterns given by amorphous and crystal forms, and is reproduced here:
The most common amorphous solid is glass. Its atoms are put
together in a nonuniform array compared to its crystalline form.
...
Randall and coworkers (24-28) reported that when a substance is
capable of existing in both amorphous and crystalline forms, then
the X-ray pattern given by the amorphous form may be regards as a
very diffuse version of the crystal pattern. There is, in fact, no sharp
distinction between crystalline and amorphous states. If, starting
with a coarsely crystalline solid, the size of the crystals could be

67

less than 100% crystallinity could be covered by claims 1 and 2 of the ‘085 Patent. (Id. at
5.)
Dr. Byrn disagreed with the other experts that the POSA would have understood
the term hydrate to denote a crystalline form. (Id. at 6.) He explained that his prior
publications, cited by the other experts on this issue (see n. 113, infra), which admittedly
describe hydrates as crystalline, do not state that hydrates can only be crystalline. (Id.)
He also stated that the specification itself contemplates amorphous hydrates. (Id. at 7.) 97
He further stated that the Khankari text, see n. 116, infra, cited by defendants’ experts,
reflects the POSA’s understanding of the existence of amorphous hydrates and
techniques for creating the same. (Id. at 8.)
Dr. Byrn also opined that the POSA would not understand the term “the
magnesium salt of S-omeprazole trihydrate” to include any purity limitations, such as

reduced by stages and an X-ray diffraction photograph could be
taken at each stage, the photographs would become diffuse when the
crystal size fell below about 10-5 cm. With reduction of crystal size,
the reflections become increasingly diffuse until the limit is reached
at 10-7–10-8 cm, the region of atomic dimensions, where the word
crystal, with its implication of precise pattern repetition, ceases to
be appropriate.
Haleblian, Characterization of Habits and Crystalline Modifications of Solids and Their
Pharmaceutical Applications, 64 J. Pharm. Sci. 6, 1269-88 (1975). (Dkt. 38-2 at 83.)
Dr. Byrn referred to the language found in the ‘070 Patent at col. 2, ll. 55-62. This
language has been the subject of intense disagreement among the experts. For example, Dr.
Atwood opined that this language would be understood to reference the existence of an
amorphous form of the magnesium salt of S-omeprazole, but not amorphous hydrates. (See
Mylan action, dkt. 232-16 at 39-40; dkt. 269-1 at 10-11.)
97

68

those proposed by Perrigo or Mylan. (See id. at 8-10.)98 He repeatedly emphasized the
fact that no purity limitations appear in either claim 1 of the ‘085 Patent or claim 1 of the
‘070 Patent. Dr. Byrn asserted that the written description discusses three types of purity:
chemical purity, enantiomeric purity, and phase purity, which he previously stated
described examples or preferred embodiments. (See id.) He stated that the term
“substantially pure,” as used in the written description, would be understood by the
POSA as referring to chemical purity, which he explained “signifies the extent to which
the substance is free from detectable impurities or contaminants.” (Id. at 9.) He stated
that the term “substantially free,” as used in the written description, would be understood
by the POSA as referring to enantiomeric purity (i.e., “substantially free from magnesium
salts of R-omeprazole”) and phase purity (i.e., substantially free from other forms of
magnesium S-omeprazole”). (Id. at 9-10.) Based on this description, he asserted that the
purity limitations proposed by Perrigo (enantiomeric purity and phase purity) and Mylan
(chemical purity) were different from each other, injecting further confusion into the

98

Both Mylan and Perrigo proposed constructions that include a purity limitation.
Perrigo’s proposed construction requires the “crystals of the magnesium salt of S-omeprazole
trihydrate” to be “substantially free from magnesium salts of R-omeprazole and other forms of
magnesium salts of S-omeprazole.” See nn. 14 and 15, supra and accompanying text. Mylan’s
proposed construction requires the “crystalline magnesium salt of S-omeprazole having exactly
three waters of hydration” to be “substantially pure.” See n. 15, supra. Dr. Ruffalo explained
that the term “substantially pure,” as used in the context of Mylan’s proposed construction, is
also considered to be “polymorphically pure” S-omeprazole magnesium trihydrate to the extent
that no other physical forms are present (i.e. hydrates other than the trihydrate, solvates,
amorphous materials, magnesium salts of R-omeprazole and other prior art forms, etc.). (See
Mylan action, dkt. 232-18 at 14.)

69

defendants’ proposed constructions. (Id. at 10.)99 He further opined that the purity
language found in the written description was in relation to the process aspects of the
patent (e.g., examples 1 and 7), not the compound itself. (Id. at 9.)
ii.

“characterized by the following major peaks in its
X-ray diffractogram” and “represented by “FIG.
1”

AstraZeneca proposes that the claim term “characterized by the following major
peaks in its X-ray diffractogram” be construed as follows: “identifiable by reference to an
X-ray diffractogram that includes the major peaks below.” (Dkt. 36 at 3.) AstraZeneca
proposes that the claim term “represented by FIG. 1” be construed as follows:
“represented by Figure 1 of the ‘070 patent.” (Id.)
Dr. Byrn’s position was that the claimed compound would be identifiable or
distinguishable by the presence of a certain unique X-ray diffraction peak or collection of
peaks, without exhibiting every single peak listed in claim 1 of the ‘085 Patent or shown
in Figure 1 of the ‘070 Patent. The level of sameness between the major listed peaks in
claim 1 or Figure 1 required to identify the claimed compound is at the heart of the
dispute with respect to these claim terms.
With respect to the term “characterized by the following major peaks in its X-ray
diffractogram” in claim 1 of the ‘085 Patent, Dr. Byrn declared that a POSA would

99

Dr. Atwood disagreed with this distinction. (See Mylan action, dkt. 269-1 at 12.) He
asserted that Mylan’s proposed construction would include all three types of purity and that the
specification supports this interpretation based on the language “free from.” (See id.)
Additionally, Dr. Ruffalo’s second declaration was devoted almost exclusively to rebutting Dr.
Byrn’s opinions regarding the term “substantially pure.” (See Mylan action, dkt. 247-7.)

70

understand the term “characterized” to be an inclusive or open-ended term that does not
exclude additional, unrecited elements. (Dkt. 40 at 10.)100 For this reason, he explained
that a POSA would understand that the claimed compound would be distinguishable by
the presence of the listed major peaks in claim 1 of the ‘085 Patent, although additional,
unrecited peaks may also be present. (Id. at 10-11.) He further declared that a POSA
would also understand the term “characterized” to allow for experimental error. He
explained that, with experimental error, not every peak listed in claim 1 of the ‘085
Patent would necessarily be present in every diffractogram. (Id. at 11.) Therefore, the
claimed compound would be distinguishable to the POSA by the presence of certain
unique peaks or collections of peaks, even if every peak listed in claim 1 is not displayed.
(Id. at 11.)
Dr. Byrn explained that a POSA would understand the term “d-value/Å,” as it
appears in claim 1 of the ‘085 Patent, to be a measurement indicating the spacing
between lattice planes in a crystal (“d-spacings”) for the peaks in an X-ray diffractogram.
(Id. at 11.) He asserted that the accuracy in the measurement of d-spacings is variable
and that one should look for a range of d-spacings when identifying a compound in
relation to a reference diffractogram. (Id.) Dr. Byrn also stated that the term “relative

For this definition of “characterized,” Dr. Byrn cited section 2111.03 of the 2014
edition of the MANUAL OF PATENT EXAMINING PROCEDURE, which provides an identical
definition for that term, as follows: “The transitional term “comprising”, which is synonymous
with “including,” “containing,” or “characterized by,” is inclusive or open-ended and does not
exclude additional, unrecited elements or method steps.” (Dkt. 40-16 at 7.) On the other hand,
Dr. Atwood opined that the POSA would not be familiar with this reference, but he did not
disagree with the notion that “characterized” is an inclusive or open-ended term that does not
exclude additional, unrecited elements. (See Mylan action, dkt. 247-6 at 10.)
100

71

intensity,” as used in the table in claim 1 of the ‘085 Patent, is understood by a POSA to
mean the ratio of the intensity of a peak (i.e., the relevant peak size) with a particular dvalue to the intensity of the strongest peak in the diffraction pattern. (Id.) He explained
that relative intensities may vary considerably between a sample and a reference of the
same substance.101 Dr. Byrn also indicated that a POSA would understand that the
appearance of an X-ray diffractogram could be influenced by multiple factors, such as
experimental conditions, that could result in a situation where not every listed peak would
be present. (Id. at 11-12.)102 For these reasons, he concluded that every single peak listed
in claim 1 of the ‘085 Patent would not necessarily need to be present, or present with the
listed relative intensity, for the POSA to identify the compound of claim 1 of the ‘085
Patent. (Id.)
With respect to the term “represented by FIG. 1” in claim 1 of the ‘070 Patent, Dr.
Byrn likewise declared that a POSA would understand the term “represented” to be a
flexible term, not requiring exact identity with Figure 1, because X-ray powder
diffraction is an experimental technique that is subject to experimental variation and
multiple external factors. (Id. at 12.)

101

Dr. Atwood agreed that relative intensities can vary to some extent. (See Mylan
action, dkt. 247-6 at 10.)
102

Dr. Atwood disagreed with Dr. Byrn on this point. He asserted that experimental error
will not cause peaks to be missing from a diffractogram. (See Mylan action, dkt. 247-6 at 10.)
Rather, he stated that experimental error means that peak positions will vary to a certain extent,
but will not be missing. (Id.)

72

When discussing the opposing construction positions of the parties for the terms
“characterized by the following major peaks in its X-ray diffractogram” and “represented
by FIG. 1,” Dr. Byrn opined that the phrase “normal experimental error” proposed by the
opposing parties was unclear and undefined, and that it may not adequately account for
factors that may affect an X-ray diffractogram or how X-ray diffraction is performed and
analyzed by a POSA. (Dkt. 49 at 10-11.) Dr. Byrn explained that other materials present
in a dosage form or composition may generate obscuring or interfering peaks:
[W]here a compound is a component of a dosage form or
pharmaceutical composition, other materials present in the
dosage form or composition, such as crystalline excipients,
may generate their own X-ray diffraction peaks, which can
obscure or interfere with the peaks of the compound in
question. Many common pharmaceutical excipients generate
their own X-ray diffraction patterns and will accordingly have
this effect. As a result, in a multicomponent system, such as a
dosage form or pharmaceutical composition, a [POSA] will
consider the X-ray diffractogram as a whole, taking into
account interfering peaks from other components when
identifying or characterizing a particular compound.
(Id. at 11.)103 Thus, Dr. Byrn’s position is that the construction proposed by AstraZeneca
would necessarily account for the experimental error that is inherent in x-ray powder
diffraction testing.

103

Dr. Atwood agreed that other materials present in a dosage form, such as crystalline
excipients, may exhibit their own XRPD pattern that may interfere with the recited peaks. (See
Mylan action, dkt. 269-1 at 13.) However, he asserted that the POSA would run the individual
excipients in order to identify those peaks attributable to the excipients. (Id.)

73

b.

Andrx’s expert testimony and exhibits

Dr. Zaworotko was presented by Andrx as an expert in the field of solid state
chemistry. His qualifications are summarized in the margin.104 Dr. Zaworotko’s first
declaration (dkt. 38-3) includes a thorough recitation of the scientific background
pertinent to the claim construction, including discussion of crystalline and amorphous
solid forms, solvates and hydrates, characterization of solid state forms, and coordination
chemistry of the magnesium salt of S-omeprazole. (See dkt. 38-3 at 10-20.)105 His
second declaration (dkt. 47-2) includes additional background information on the stages
of drug discovery and development. (Dkt. 47-2 at 2.) We will provide a brief summary
of the scientific background as set forth by Dr. Zaworotko here.
Dr. Zaworotko explained that crystals are a type of solid composed of molecules,
atoms, or ions that interact with each other (via chemical bonds) to form an ordered
arrangement that repeats in three dimensions. (Dkt. 38-3 at 10.) There are multiple
general classes of crystalline solids, including molecular crystals, metallic crystals, ionic
crystals and network crystals. (Id. at 11.) Dr. Zaworotko explained that most drug

104

Dr. Zaworotko has qualifications as a university professor in various positions in the
areas of chemistry and crystal engineering. (Dkt. 38-3 at 3-4.) He has published approximately
330 articles and is a named inventor on approximately twenty issued or pending patents,
pertaining mostly to crystallization, X-ray crystallography, crystal engineering, crystal packing,
polymorphism, and related topics. (Id. at 5.) He has also served as a consultant to various
entities within the pharmaceutical community. (Id. at 6.)
Mylan’s expert, Dr. Atwood, also provided a thorough recitation of the scientific
background, covering topics such as crystallization, polymorphism, solvates and hydrates,
amorphous solids, and characterization by X-ray powder diffraction. (See Mylan action, dkt.
232-16 at 10-20.)
105

74

substances or active pharmaceutical ingredients (“APIs”) are molecular crystals formed
from organic molecules and ions. (Id.) The arrangement of molecules defines a given
crystal form. The “unit cell” is the basic building block of a crystalline solid. (Id.) Dr.
Zaworotko also explained that a solid can assume an amorphous form, which he
describes as “not crystalline.” (Id.) He explained that amorphous solids do not have a
“unit cell,” are composed of randomly oriented molecules, ions, or atoms with no longrange order, and are less stable than a crystalline form of the same compound. (Id. at 1112.)
Dr. Zaworotko stated that compounds existing in more than one crystal structure
are “polymorphic,” which means the compound can crystallize into more than one
distinct crystal structure through different crystal packing motifs. (Id. at 12-13.)
Dr. Zaworotko defined a “solvate” as a compound that “crystallizes from solution
into a crystalline solid that traps solvent molecules within the crystal lattice.” (Id. at 13.)
He explained that solvates can be stoichiometric (having a fixed ratio of solvent to
compound) or non-stoichiometric (having a variable ratio of solvent to compound). (Id.)
If the solvent is water, the crystalline solid is known as a “hydrate.” (Id.)
Dr. Zaworotko stated that in the field of solid state chemistry, a hydrate cannot be
amorphous. He explained that in a stoichiometric hydrate, water trapped in the crystal
lattice can contribute to the dimensions of the resulting “unit cell,” and that there may be
different stoichiometric hydrates of the same compound. (Id. at 14-15.) On this point, he
stated that prior to 1997, a standardized nomenclature of stoichiometric hydrates existed.
(Id. at 14-15.) Dr. Zaworotko explained that a hydrate with one water of crystallization
75

for every two host molecules was known as a “hemihydrate.” (Id. at 14.) A hydrate with
one water of crystallization for each host molecule was known as a “monohydrate.” (Id.
at 15.) A hydrate with two waters of crystallization for each host molecule was known as
a “dihydrate.” (Id.) A hydrate with three waters of crystallization for each host molecule
was known as a “trihydrate.” (Id.) He also explained that solvates and hydrates may be
polymorphic, which he defined as having “different unit cell dimensions and/or packing
motifs . . . even though the ratio of solvent or water in the lattice to host is the same in
each polymorph.” (Id. at 15.)
With respect to the characterization of solid state forms, Dr. Zaworotko explained
that numerous methods, such as single crystal X-ray diffraction (SCXRD) and X-ray
powder diffraction (“XRPD”) are known to identify crystals and determine attributes of a
crystal structure. Dr. Zaworotko described XRPD as the “gold standard” for
identification of the crystal form of an API, since it can produce a pattern of peaks that
act as a signature for a particular crystalline form. (Id. at 16.) He explained that with
XRPD, X-rays are directed at the sample (which he stated may be a “bulk crystalline” or
“microcrystalline” sample) at varying angles, and that if the sample is crystalline, it will
diffract X-rays at specific angles due to its crystalline structure, providing a unique
pattern of peaks. (Id.) He also noted that sample-specific factors may influence readings,
and that XRPD is subject to some experimental error. (Id. at 17.) He further explained
that amorphous materials do not diffract X-rays at specific angles, and instead scatter Xrays to create a diffuse pattern. (See id. at 17-18.)

76

Dr. Zaworotko explained that the magnesium salt of S-omeprazole belongs to a
class of compounds known as coordination compounds, which are compounds comprised
of metal cations bonded with one or more molecules or ions called ligands. (Id. at 18.)
The geometry of coordination compounds enables a phenomenon known as structural
isomerism in which molecules with an identical chemical formula have bonded together
in different orders. (See id. at 19.) Dr. Zaworotko explained that structural isomers with
differing structures (but the same chemical formula) exhibit different crystal structures
and properties. (Id.) He concluded that S-omeprazole anions and water molecules can
coordinate to magnesium cations in multiple modes and geometric arrangements due to
structural isomerism. (Id. at 20.)
In his second declaration, Dr. Zaworotko explained that the drug discovery and
development process generally takes place in three distinct stages: a “molecules” stage, a
“materials” stage, and a “medicines” stage. He stated that the education and training of a
POSA and the nomenclature in each stage was not always consistent; thus, distinguishing
among these stages is important. (Dkt. 47-2 at 2.) He describes the “molecules” stage as
including “medicinal chemistry for the discovery of new chemical entities along with
biological and pharmacological screening activity.” (Id.) He describes the “materials”
stage as the pre-formulation stage as including “discovery of a drug substance . . .
suitable for use as a material in a drug product.” (Id.) He describes the “medicines”
stage as the formulation stage which includes the combining of the drug substance with
inactive ingredients, excipients in order to form the drug product. (Id.) He stated that the

77

‘085 and ‘070 Patents relate to the “materials” stage of drug discovery and development,
and that it is important to consider the ‘085 and ’070 Patents in that context. (Id.)
i.

“The magnesium salt of S-omeprazole trihydrate”

To review, Andrx proposes that the claim term “the magnesium salt of Someprazole trihydrate” be construed as follows: “the magnesium salt of S-omeprazole in
crystalline form containing three molecules of water of crystallization per molecule of
magnesium salt of S-omeprazole.” (Dkt. 36 at 3.) Dr. Zaworotko opined that the POSA
would have understood the term “the magnesium salt of S-omeprazole trihydrate” to refer
to “the magnesium salt of S-omeprazole in crystalline form containing three molecules of
water of crystallization per molecule of magnesium salt of S-omeprazole,” which he
stated was the plain and ordinary meaning to a POSA at the time of the invention. (Dkt.
38-3 at 21.) He explained that the term “trihydrate” refers to a crystalline form that has
three water molecules, entrapped in its crystal lattice, for every molecule of compound.
(Id.) With respect to the three water molecules, he declared that:
[t]he POSA referred to those water molecules as “waters of
crystallization” or “waters of hydration.”
The POSA
understood that the term “trihydrate” refers to a stoichiometric
hydrate and thus that the waters of crystallization in the
claimed trihydrate contribute to the dimensions of the unit cell.
The POSA thus understood that waters of crystallization
contribute to the structure unique to the hydrate (or the
particular polymorph of the hydrate). Thus, the POSA
understood that waters of crystallization were the waters that
define the hydration level (e.g., mono-, di-, tri-, etc.) of hydrate.

78

(Id. at 22.)106
Looking at the intrinsic evidence for the ‘085 and ‘070 Patents, Dr. Zaworotko
stated that both the written description and the prosecution histories of the ‘085 and ‘070
Patents support his opinion that the claimed trihydrate is crystalline. In his opinion,
statements made in the written description indicate that the trihydrate is crystalline and
were used to distinguish it from the prior art on that basis. (Id. at 22-23.)107 Other
statements, he noted, are premised on the trihydrate being crystalline. He pointed to the
discussion of XRPD and the unit cell at column 2, lines 43-51 and 63-64 of the ’070
Patent. (Id. at 22-23.)
Further, he stated that the methods of making the trihydrate in Examples 1 and 7
would indicate to the POSA that the resulting trihydrate is crystalline. (Id. at 23.)108 He
explained that Examples 1 and 7 involve crystallization from water followed by drying

106

Dr. Zaworotko also stated that the construction offered by Perrigo and Mylan for this
term were consistent with Andrx’s proposed construction, as all three require the hydrate to be
crystalline. (Id. at 26.) Mylan’s experts, Dr. Atwood and Dr. Ruffalo described the magnesium
salt of S-omeprazole as crystalline slightly differently as “having exactly three waters of
hydration.” (See Mylan action, dkt. 232-16 at 38; dkt. 232-18 at 14.)
For example, he pointed to the statement at column 2, lines 51-53 of the ‘070 Patent:
“[t]he compound of the invention is characterized by being highly crystalline, i.e., having a
higher crystallinity than any other form of magnesium salt of S-omeprazole disclosed in the prior
art.” (Dkt. 38-3 at 22-23.)
107

108

Dr. Ruffalo and Dr. Atwood agree that Examples 1 and 7 indicate to the POSA that
the resulting trihydrate is crystalline and none of the examples teach a process for making an
amorphous form of S-omeprazole magnesium trihydrate. (See Mylan action dkt. 232-18 at 1718; dkt. 232-16 at 40-41.)

79

under vacuum, a process intended to remove excess water, which he describes as “water
that is not integrated into the unit cell.” (Id.)
With respect to the prosecution history, Dr. Zaworotko opined that a POSA would
understand many of the Applicants’ arguments made in response to Office action
rejections to be premised on the concept that the claimed trihydrate is crystalline.
Specifically, Dr. Zaworotko pointed to statements describing the invention made in
response to the November 27, 2000 Office action during the prosecution of the ‘085
Patent and in response to the April 29, 2005 Office action during the prosecution of the
‘070 Patent. (Id. at 24.)109,110 Dr. Zaworotko also highlighted the Applicants’ discussion
of “guest” molecules in their response to the November 27, 2000 Office action as
evidencing an understanding by the Applicants that the waters in a trihydrate must be
waters of crystallization. (Id.) He explained that the Applicants argued it was
unpredictable to add water as a guest molecule in response to the examiner’s assertion
that it was known that organic solid material may include “guest” molecules for the
purpose of forming crystals. (Id.)
Looking at the extrinsic evidence, Dr. Zaworotko stated that in the field of solid
state chemistry, hydrates were long recognized as crystalline forms defined by the
number of waters of crystallization they contain in their crystal lattice. On this point, he

109

See Section II.B.1, supra.

110

Dr. Atwood and Dr. Ruffalo agreed and referred to the same statements as well as
others on this point. (See Mylan action, dkt. 232-16 at 42-46l dkt. 232-18 at 20-30.)

80

referred to several texts.111 Two of these texts were authored by plaintiff’s expert, Dr.
Byrn. See n. 111, supra.112 In Dr. Zaworotko’s opinion, Dr. Byrn’s prior writings
distinguished solvates and hydrates from amorphous forms on the basis of crystallinity. 113

111

Dr. Zaworotko cited these texts as supporting the statement that hydrates were known
in the art to be crystalline forms defined by the number of waters of crystallization: Brittain,
Methods for the Characterization of Polymorphs and Solvates, in Polymorphism in
Pharmaceutical Solids, Vol. 95 (Marcel Dekker, New York 1999) (“Brittain”) (dkt. 38-2 at 254); Byrn, et al., Solid-State Pharmaceutical Chemistry, Chem. Mater. 1994, 6, 1148-58 (1994)
(“Byrn 1994”) (dkt. 38-2 at 56-66); Byrn et al., Pharmaceutical Solids: A Strategic Approach to
Regulatory Considerations, Pharm. Res. Vol. 12, No. 7, 945-54 (1995) (“Byrn 1995”) (dkt. 38-2
at 68-78); Haleblian, Characterization of Habits and Crystalline Modifications of Solids and
Their Pharmaceutical Applications, 64 J. Pharm. Sci. 6, 1269-88 (1975) (“Haleblian”) (dkt. 38-2
at 80-99); Khankari & Grant, Pharmaceutical hydrates, Thermochimica Acta, 248 (1995)
(“Khankari”) (dkt. 38-2 at 101-116); and Wells, Structural Inorganic Chemistry (3d 1962)
(“Wells”) (dkt. 38-2 at 118-120).
112

Dr. Atwood also cited the Byrn 1995 reference, the Haleblian reference, and a third
reference, Byrn, Solid-State Chemistry of Drugs, Academic Press (1982) (Mylan action, dkt.
232-7), to demonstrate the understanding of the term “hydrate” and “solvate” to a POSA.
113

The Byrn 1994 article states:

The most common solid forms that may be found for a given drug substance are as
follows: crystalline polymorphs, forms having the same chemical composition but
different crystal structure and therefore different densities, melting points,
solubilities, and many other important properties; solvates, forms containing
solvent molecules within the crystal structure, giving rise to unique differences in
solubility, response to atmospheric moisture, loss of solvent, etc. Sometimes a drug
product may be a desolvated solvate, formed when solvent is removed from a
specific solvate while the crystal structure is essentially retained—again, many
important properties are unique to such a form; finally, amorphous solid forms that
have no long-range molecular order (i.e., no crystallinity) and which tend to be
more soluble, more prone to moisture uptake, and less chemically stable than their
crystalline counterparts (pharmaceutical processing operations may produce solids
of low crystallinity intermediate between that of a crystalline solid and an
amorphous solid).
Byrn, et al., Solid-State Pharmaceutical Chemistry, Chem. Mater. 1994, 6, 1148-58 (1994). (Dkt.
38-2 at 56 (emphasis in original).)
The Byrn 1995 article states:

81

Polymorphs exist when the drug substance crystallizes in different crystal packing
arrangements all of which have the same elemental composition (Note that hydrates
can exist in polymorphs). Hydrates exist when the drug substance incorporates
water in the crystal lattice in either stoichiometric or non-stoichiometric amounts.
Desolvated solvates are produced when a solvate is desolvated (either knowingly
or unknowingly) and the crystal retains the structure of the solvate. Amorphous
forms exist when a solid with no long range order and thus no crystallinity is
produced.
Byrn et al., Pharmaceutical Solids: A Strategic Approach to Regulatory Considerations, Pharm.
Res. Vol. 12, No. 7, 945-54 (1995). (Dkt. 38-2 at 69.)

82

He further stated that the additional cited literature on the subject, namely Britain,114
Haleblian,115 Khankari,116 and Wells,117 also reflect hydrates being known in the art as
crystalline forms. (Id. at 25.)

114

The Brittain text states:

Compounds may be polymorphs (forms having the same chemical composition but
different crystal structures), solvates (forms containing solvent molecules within
the crystal structure, desolvated solvates (forms when the solvent is removed from
a specific solvate while still retaining the original crystal structure), or amorphous
(solid forms that have no long-range molecular order).
Brittain, Methods for the Characterization of Polymorphs and Solvates, in Polymorphism in
Pharmaceutical Solids, Vol. 95 (Marcel Dekker, New York 1999). (Dkt. 38-2 at 4 (italics in
original).)
The Haleblian text states: “Solvates are molecular complexes that have incorporated
the crystallizing solvent molecule in their lattice. When the solvent incorporated in the solvate is
water, it is called a hydrate.” Haleblian, Characterization of Habits and Crystalline
Modifications of Solids and Their Pharmaceutical Applications, 64 J. Pharm. Sci. 6, 1269-88
(1975). (Dkt. 38-2 at 87.)
115

116

The Khankari text states:

With some crystalline solids, solvent in the surrounding medium may become
incorporated into the crystal lattice of the compound in stoichiometric proportions.
These molecular adducts are termed solvates. Hydrates are formed when water is
the solvent of crystallization. In hydrates water occupies definite positions in the
crystal lattice, usually by forming hydrogen bond(s) and/or coordinate covalent
bond(s) with the anhydrate drug molecules.
Khankari & Grant, Pharmaceutical hydrates, Thermochimica Acta, 248 (1995). (Dkt. 38-2 at 102.)
117

The Wells text states:

One of the simplest ways of purifying a compound is to recrystallize it from a
suitable solvent. The crystals separating from the solution may consist of the pure
compound or they may contain ‘solvent of crystallization’. For most salts water is
a convenient solvent, so that crystals containing water—hydrates—have been
known from the earliest days of chemistry, and many inorganic compounds are
normally obtained as hydrates.

83

Dr. Zaworotko disagreed with many points and issues raised by Dr. Byrn. First,
he disagreed with Dr. Byrn that a POSA would interpret the written description as
acknowledging the existence of an amorphous hydrate. (Dkt. 47-2 at 3.)118 His
interpretation of this portion of the written description was that the inventors were
distinguishing their invention from prior art forms, which included an amorphous form –
not that hydrates or solvates were or could be amorphous. (Id.)119
As noted in Section II.B.3.a, supra, Dr. Zaworotko challenged Dr. Byrn’s reliance
on four post-priority date references (see n. 89, supra) for the proposition that the
existence of amorphous hydrates were well known by POSAs at the relevant time. In
addition to challenging the references based on their post-priority date status, Dr.
Zaworotko challenged them substantively. Specifically, Dr. Zaworotko opined that
Megarry (2011), appeared to be the first reported instance of using a process called
cryomilling to obtain an amorphous hydrate. (Id. at 4.) However, he noted that Megarry
(2011) did not relate to S-omeprazole magnesium, and that it is unknown how Someprazole magnesium responds to cryomilling. (Id. at 4, 10.) He explained that the
‘634 Patent does not discuss a stoichiometric hydrate or suggest a stoichiometric hydrate
could be amorphous, and that its disclosure was not sufficient to allow a POSA to

Wells, Structural Inorganic Chemistry (3d 1962). (Dkt. 38-2 at 120.)
Specifically, Dr. Zaworotko refers to the language found in the ‘070 Patent at col. 2,
ll. 55-62, that as noted supra has been the subject of intense disagreement among the experts.
118

119

Dr. Atwood and Dr. Ruffalo have similar positions regarding this portion of the
written description. (See Mylan action, dkt. 232-16 at 39; dkt. 232-18 at 16.)

84

conclude a hydrate was obtained, noting the ‘634 Patent failed to report on measured
moisture content. (Id.) He explained that the ‘842 Patent relates to cephalosporin, not Someprazole. (Id.) He stated that the examples purporting to describe an amorphous
hydrate of cephalosporin failed to explain how water content was measured and did not
provide sufficient information to determine whether water was absorbed into the crystal
lattice as water of crystallization or by adhering to the surface of the solid. (Id.) With
respect to WO2004/020436, Dr. Zaworotko opined that it fails to provide sufficient
information for a POSA to conclude that an amorphous hydrate was obtained or even
possible because it was not clear whether water is absorbed into the crystal lattice or
simply adheres to the surface of the solid described in that reference. (Id. at 6.)
In addition, and as noted in Section II.B.3.a, supra, Dr. Zaworotko also took issue
with Dr. Byrn’s reliance on the six additional references (see n. 90, supra) offered in Dr.
Byrn’s second declaration for the same proposition, that amorphous hydrates were known
and discussed in the literature at the relevant time. (See dkt. 54-1.) Dr. Zaworotko
opined that these references were irrelevant to the scientific field to which the ‘085 and
‘070 Patents pertain, i.e., the discovery of solid forms of drug substances suitable for use
in drug products, namely Type A molecular solids. (See id. at 2.)120 Dr. Zaworotko
explained that there are four distinct categories of solid substances: molecular solids,
ionic solids, metallic solids, and network solids. (Id. at 2.) He further explained that

120

Dr. Atwood also took issue with these additional references, asserting that each was
irrelevant to the field to which the ‘085 and ‘070 Patents pertain. (See Mylan action, dkt. 269-1
at 8-10.)

85

there are two types of molecular solids, Type A121 and Type B,122 and that solid forms of
drug substances are Type A molecular solids. (Id.) He stated that this was important
because the nomenclature for molecular solids, and specifically the term “hydrate,” is not
necessarily interchangeable among the four classes of solid substances, or even between
Type A and Type B molecular solids. (Id. at 4.) He reviewed each of the six additional
references and concluded that they referred to Type B molecular solids (i.e., polymers or
nanoparticles) or network solids, not Type A molecular solids. (Id.) For this reason, he
concluded that each of the six additional references were not relevant to the ‘085 or ‘070
Patents. (See id. at 4-6.)
Dr. Zaworotko also disagreed with Dr. Byrn’s articulation of what constitutes a
hydrate, stating that Dr. Byrn’s definition, characterizing a hydrate as having “some
structure that can bind water in a regular way,” is overbroad, unsupported, and in conflict
with Dr. Byrn’s own publications both before and after the priority dates of the ‘085 and

121

Dr. Zaworotko provided the following description of Type A molecular solids:
Type A [molecular solids] occur when the compound that forms a
molecular solid is relatively pure and of small or medium molecular
weight. Type A [molecular solids] tend to be homogeneous and
crystalline, as crystalline solids are energetically favored. Almost
all organic molecules, organic salts and coordination complexes
exist as Type A molecular solids.

(Dkt. 54-1 at 3.)
Dr. Zaworotko provided the following description of Type B molecular solids: “Type
B [molecular solids] occur when the compound forming a molecular solid is impure with high
molecular weight, such as a polymer. Type B tend to be amorphous. Polymers or nanoparticles,
containing an array of different sized molecules or particles, tend to form Type B molecular
solids.” (Dkt. 54-1 at 3.)
122

86

‘070 Patents. (Id. at 6.) He stated that any implication that the magnesium salt of Someprazole can form a hydrophilic layer that incorporates water molecules is speculative
and not a typical feature for such compounds. (Id. at 7.) Dr. Zaworotko also stated that
Dr. Byrn’s overbroad definition of the term “trihydrate” conflicts with the International
Union of Pure and Applied Chemistry’s (“IUPAC”) Principles of Chemical
Nomenclature because it would include aqua complexes,123 i.e., compounds in which
none of the three waters are absorbed into the crystal lattice, but instead are all in
coordination bond with the magnesium ion. (See id. at 7-9.) Dr. Zaworotko similarly
found the dictionary definitions for the term “trihydrate” cited by Dr. Byrn (see n. 91,
supra) to be overly broad and inapposite to the field of the patents-in-suit. (Dkt. 47-2 at
9-10.)
Finally, Dr. Zaworotko clarified the scope of Andrx’s proposed construction. He
explained that it would not limit claim 1 of the ‘070 Patent to a particular polymorphic
form, and would in fact “cover all crystalline polymorphs of S-omeprazole magnesium
trihydrate, and thus, be broader than claim 1 of the ‘085 patent, which is limited to a
particular crystalline polymorph of that trihydrate.” (Id. at 11.)124

123

In coordination nomenclature, a coordination entity is composed of a central atom or
atoms to which are attached other atoms or groups of atoms, which are termed ligands. Leigh,
Principles of Nomenclature: A Guide to IUPAC Recommendations (1998) (“Leigh”) (dkt. 47-1
at 173.) The number of central atoms in a single coordination entity is denoted by the nuclearity:
mononuclear, dinuclear, trinucleaer, etc. (Id. at 175.) For mononuclear coordination
compounds, the central atom is listed first, followed by the formally anionic ligands, neutral
ligands, and polydentate ligands. (Id. at 176.) Water molecules as ligands take the name aqua.
(Id. at 177.)
Dr. Ruffolo provided a similar argument with respect to claim 1 and 2 of the ‘070
Patent. He stated that claim 1 of the ‘070 patent would encompass all crystalline forms of S124

87

ii.

“characterized by the following major peaks in its
X-ray diffractogram” and “represented by “FIG.
1”

Andrx proposes that the claim term “characterized by the following major peaks in
its X-ray diffractogram” be construed as follows: “exhibiting, within the range of normal
experimental error, each of the following as major peaks in its X-ray diffractogram, each
such peak falling within the range for relative intensity recited for that peak such ranges
being defined at Column 5, lines 29-39 [of the ‘085 Patent].” (Dkt. 36 at 3.) Andrx
proposes that the claim term “represented by FIG. 1” be construed as follows: “exhibiting
an x-ray powder diffractogram that is the same as Figure 1 of the ‘070 patent within
normal experimental error.” (Id.)
As noted above, the level of sameness between the major listed peaks in claim 1 of
the ‘085 Patent or Figure 1 of the ‘070 Patent required to identify the claimed compound
is at the heart of the dispute with respect to these claim terms. Dr. Zaworotko advocated
for constructions that would require all of the peaks to be present and explicitly account
for experimental error.
With respect to the claim term “characterized by the following major peaks in its
X-ray diffractogram” of claim 1 of the ‘085 Patent, Dr. Zaworotko’s position was that
this term should be construed “to require all 13 peaks recited in the table set forth in the
claim, including their respective relative intensities, within normal experimental error.”
(Dkt. 38-3 at 27.) He declared that the POSA would understand that claim 1 of the ‘085

omeprazole magnesium trihydrate, and that claim 2 defines a specific crystalline structure that
was characterized using XRPD. (See Mylan action, dkt 232-18 at 31.)

88

Patent is directed to a particular trihydrate that, when analyzed using XRPD, would
exhibit the 13 peaks recited in the table, and that those peaks must fall into the specified
relative intensity ranges. (Id. at 28.) He explained that a POSA looking to identify (or
distinguish) one sample from another or looking to compare the XRPD of a sample to a
reference XRPD would compare the pattern in its entirety, and not focus on matching or
mismatching one peak, or even several peaks to the exclusion of others. (Id. at 28.) He
further opined that the POSA would understand the pattern of claim 1 of the ‘085 Patent
to be compared as the set of 13 peaks, each with a relative intensity value falling into a
particular recited range due to the ‘085 Patent’s presentation of the 13 peaks as intensity
ranges, rather than particular numerical values. (Id. at 28-29.)125
Dr. Zaworotko disagreed with Dr. Byrn that claim 1 of the ‘085 Patent could be
construed as requiring fewer than all recited peaks. (See id. at 29.) Turning to the
written description, he declared that the POSA would not interpret the statement found at
the conclusion of Exhibit 7, “[t]he product was analyzed using X-ray powder diffraction
and the result complied with FIG.1 and Table 1” (see ‘070 Pat., col. 9, ll. 48-50), to allow
for such imprecision. (Id.) In his opinion, the POSA would conclude the XRPD data for
the product of Example 7 was similar enough to the data for Example 1, with respect to
all the major peaks and relative intensities of those peaks. (Id.)

The explanatory note to the table shown in claim 1 of the ‘085 Patent does translate
those verbal intensity ranges into numerical ranges, but not precise numbers, e.g., vs (very
strong) is 25-100% relative intensity. (See ‘085 Pat., col. 5 ll. 33-39.)
125

89

Dr. Zaworotko also stated that the POSA would understand that XRPD is subject
to some experimental error, which would be accounted for by a POSA in any comparison
of XRPD diffractograms or data sets. (Id. at 17-18, 29-30.) However, he did not define
experimental error, describe the nature of its effect on an XRPD, or how it would be
accounted for.
With respect to the term “represented by FIG. 1,” Dr. Zaworotko declared that the
POSA would understand that term to mean that “the claimed trihydrate exhibits the same
[XRPD] as Figure 1, taking into account normal experimental error.” (Id. at 30.) He
explained that the POSA would subject a given crystal form to XRPD analysis and
compare the resulting patterns to determine whether that crystal form is “represented by”
Figure 1. (Id.) He again discounted the “complies with” language found in Example 7,
stating it should not be viewed as broadening the term “represented by” beyond the
construction proposed by Andrx. (Id. at 31.)
c.

Perrigo’s expert testimony and exhibits

Dr. Buckton was presented by Perrigo as an expert in the field of pharmaceutical
research and development including selection and characterization of physical forms,
formulation development, pharmaceutical processing and drug delivery systems. His
qualifications are summarized in the margin.126

126

Dr. Buckton has qualifications as a university professor in various positions in the area
of pharmaceutics. (Dkt. 44-1 at 2-3.) He has also published more than 180 publications. (Id. at
3.)

90

Like Dr. Zaworotko, Dr. Buckton provided a thorough recitation of the scientific
background pertinent to the claim construction in his first declaration.127 Dr. Buckton’s
explanations were in accord with those provided by Dr. Zaworotko (see Section II.B.3.b,
supra), so there is no need for the Court to repeat them at length here. The Court notes
that Dr. Buckton offered explanations for the following concepts and topics: the
characteristics and properties of crystalline and non-crystalline (or amorphous) materials,
polymorphism, stoichiometric hydrates, and X-ray powder diffraction.
i.
`

“The magnesium salt of S-omeprazole trihydrate”

Perrigo proposes that the claim term “the magnesium salt of S-omeprazole

trihydrate” be construed as follows: “crystals of the magnesium salt of S-omeprazole
trihydrate, substantially free from magnesium salts of R-omeprazole and other forms of
magnesium salts of S-omeprazole.” (Dkt. 36 at 3.)
Dr. Buckton declared that a POSA would understand the term “the magnesium salt
of S-omeprazole trihydrate” to mean:
crystals of S-omeprazole magnesium trihydrate in the context
of a bulk product, and that the bulk product is substantially free
from: magnesium salts of R-omeprazole, and other physical
forms of magnesium salts of S-omeprazole, such as the
corresponding magnesium salt compounds described in prior
art, which include dihydrates used in the preparation of the
trihydrate, and other forms such as anhydrates, hydrates,
solvates, and polymorphs or amorphous forms thereof
disclosed in the prior art, such as monohydrates, dihydrates,

127

(See Perrigo action, dkt. 44-1 at 6-11.)

91

sesquihydrates, trihydrates, and alcoholates (e.g., methanolates
and ethanolates).128
Dr. Buckton’s opinion is based exclusively on intrinsic evidence, i.e., the written
description and the prosecution histories of the ‘085 and ’070 Patents. Dr. Buckton did
not cite to any scientific literature in his opinion.
As noted above, Dr. Buckton agreed with Dr. Zaworotko that the term “the
magnesium salt of S-omeprazole trihydrate” would be understood by a POSA as limited
to a crystalline form. Dr. Buckton stated he was not aware of any publication
characterizing a hydrate as amorphous at the time of the invention.129 However, and as
discussed in more detail below, Dr. Buckton’s proposed construction included additional
limitations that relied on the magnesium salt of S-omeprazole trihydrate being in “bulk
product” form.130 He stated that the Applicants limited “the magnesium salt of Someprazole trihydrate” to products for which the S-omeprazole magnesium at the bulk
level is trihydrate, “substantially free” of other forms. He explained that it is necessary to
consider the entirety of S-omeprazole magnesium present in a product to determine
whether that product is (or contains) the claimed magnesium salt of S-omeprazole
trihydrate:
[I]f the entirety of S-omeprazole magnesium for a given
product were to consist of trihydrate crystals plus an
128

(Perrigo action, dkt. 44-1 at 13-15.)

129

(Perrigo action, dkt. 50-1 at 10-11).

At his deposition, Dr. Buckton admitted that the term “bulk product” does not appear
in the claims or the written description. (Perrigo action, dkt. 51-3 at 4-5.)
130

92

insubstantial quantity of another form or forms (e.g., dihydrate,
amorphous, etc.), the S-omeprazole magnesium would be “the
magnesium salt of S-omeprazole trihydrate” as claimed in the
patents. On the other hand, if the entirety of S-omeprazole
magnesium for a given product were to consist of trihydrate
crystals plus a substantial quantity of another form or forms
(e.g., dihydrate, amorphous, etc.), the S-omeprazole
magnesium would not be “the magnesium salt of S-omeprazole
trihydrate.”131
Dr. Buckton stated that his position was supported by the intrinsic evidence. He
pointed to the written description, which he opined makes clear to the POSA that “the
magnesium salt of S-omeprazole trihydrate” refers to the entirety of a bulk product in
solid form, in which substantially all of the S-omeprazole magnesium exists in the
trihydrate form, not another solid form or forms, such as anhydrous, amorphous,
dihydrate, etc.132 Dr. Buckton explained that, in his opinion, the “substantially free”
language found in the written description is meaningful only in the context of a bulk
product.133 He stated that additional statements, such as the invention being
“substantially pure,” “exist[ing] in a well-defined state,” and being “easily
distinguishable from any other crystal form of the magnesium salt of S-omeprazole
disclosed in the prior art” are further evidence that the term refers to a bulk product.134
He went on to opine that the stated advantages of the invention, i.e., “being more stable,”

131

(Perrigo action, dkt. 44-1 at 17.)

132

(Id. at 15.)

133

(Id.)

134

(Id. at 15-16.)

93

“easier to characterize,” “easier to handle and store,” etc., would be understood by the
POSA as referring to a bulk product.135
At his deposition, Dr. Buckton was asked to define the bounds of the term
“substantially free.” He could not, and refused to quantify exactly where the term
“substantially free” begins and ends.136 He simply used substitutes for the term
“substantially free,” describing the amount as “the vast majority” or “majority.” See e.g.,
n. 136. Dr. Buckton disagreed with Dr. Byrn that the “substantially free” language found

135

(Id. at 16.)

136

Dr. Buckton was asked, in a deposition excerpt filed by both AstraZeneca and Perrigo,
what the term “is substantially free from” meant, as used in paragraph 44 of his first Declaration.
He replied:
Substantially free means the vast majority of the omeprazole in say
the tablet, if that’s what we’re looking at, the vast majority of it is
present as the magnesium salt of S-omeprazole trihydrate. There
conceptionally could be some other form there but that would be in
the minority, and that’s what I’m taking as the meaning now.
Whether that is a definite term that you could decide on, we can talk
about, but that’s what I’m taking as now.
(Perrigo action, dkt. 50-3 at 11-12.) He was then asked if the term “substantially free from” would
be satisfied by a material that contains 51 percent of S-omeprazole magnesium trihydrate and 49
percent of something else. In response, he refused to provide a specific cut off quantity:
I don’t have a number so I can’t go one by one by one along the way
to get to a number. But what I could tell you categorically is if the
S-omeprazole trihydrate is in the minority and is a very small part
of what is present there, it cannot be that the total S-omeprazole in
that product is S-omeprazole trihydrate.
(Id. at 12.) When asked why he could not give an exact number, Dr. Buckton explained that he
did not know if the term was definite to determine “whether you’re in it or not.” (Id.) Dr. Buckton
also did not believe the POSA would be able to determine an exact number. (See Id. at 13-14.)

94

in the written description referred only to examples or embodiments of the invention.137
He declared that the POSA would understand the nature of the “substantially free”
language as limiting based on the written description and prosecution history.138
With respect to the written description, Dr. Buckton again pointed to the language
found at column 2 of the ‘070 Patent, and asserted that the characteristics and advantages
of the “compound of the invention,” always applied to the trihydrate form and would be
consistent with the description of a bulk product substantially free of R-omeprazole
magnesium and other forms of S-omeprazole magnesium.139 He explained that the stated
characteristics and advantages would not be realized in a product that was dominated by a
mixture of R-omeprazole magnesium or other forms of S-omeprazole magnesium.140
With respect to the prosecution history, Dr. Buckton relied on the November 27,
2000 Office action and the response thereto in the ‘085 Patent’s prosecution history.141
Dr. Buckton viewed the examiner’s statements and rejection of originally filed claim 1 of
‘719 Application in the November 27, 2000 Office action as concluding that Someprazole magnesium in the prior art contained at least some trihydrate.142 Thus, he

137

(Perrigo action, dkt. 50-1 at 3)

138

(Id. at 4.)

139

(Id. at 4-5, 7-9.)

140

(Id. at 9.)

141

(See id.)

142

(See id. at 5.)

95

viewed the Applicants’ response and the Langkilde Declaration as an attempt to
distinguish the invention from the prior art by arguing the claimed trihydrate form was
substantially free of R-omeprazole and other prior art forms of S-omeprazole
magnesium.143
Dr. Buckton also disagreed with the claim differentiation arguments put forth by
plaintiff. He stated that any differences between claims would not change the POSA’s
understanding that “the magnesium salt of S-omeprazole trihydrate” is limited to a bulk
product substantially free of R-omeprazole magnesium and other forms of S-omeprazole
magnesium.144 With respect to the “highly crystalline form” limitation in claim 2 of the
‘085 Patent, he reasoned that all material covered by claim 1 and 2 of the ‘085 is
necessarily “highly crystalline” according to the written description.145, 146

143

(See id. at 5-6.)

144

(Id. at 10.)

145

(Id.)

Dr. Atwood’s view differed on this point to some degree. In his opinion, the POSA
would understand that claim 1 of the ‘085 patent must be crystalline because it exhibits the
recited peaks as a result of its long range order. (See Mylan action, dkt. 269-1 at 3-5.) The
difference between the claims, he stated, was that claim 2 limits the crystalline forms in claim 1
to those with a “higher degree of long range order.” (Id. at 4.) He explained that the degree of
long range order in a crystalline solid reveals itself in the XRPD pattern by the sharpness of the
peaks. (Id.) Thus, he concluded that claim 2 would exhibit the same peaks as claim 1, but those
peaks would appear to be sharper due to the greater degree of long range order, i.e., higher
crystallinity. (Id. at 4-5.)
146

96

ii.

“characterized by the following major peaks in its
X-ray diffractogram” and “represented by FIG. 1”

Perrigo proposes that the claim term “characterized by the following major peaks
in its X-ray diffractogram” be construed as follows: “having each of the referenced major
peaks in its X-ray powder diffractogram within normal experimental error.” (Dkt. 36 at
3.) Andrx proposes that the claim term “represented by FIG. 1” be construed as follows:
“exhibiting an x-ray powder diffractogram that is the same as Figure 1 of the ‘070 patent
within normal experimental error.” (Id.)
Dr. Buckton opined that the claim term “characterized by the following major
peaks in its X-ray diffractogram” means that “the product in question exhibits all of the
referenced major peaks in its powder X-ray diffractogram within normal experimental
error.”147,148 Dr. Buckton opined that the claim term “represented by FIG. 1” means that
“the product in question exhibits the same powder X-ray diffractogram shown in Figure 1
within normal experimental error.”149,150

147

(Perrigo action, dkt. 44-1 at 18.)

Mylan’s construction for this term was similar to Perrigo’s, in that Mylan’s proposed
construction required each peak to be present, within normal experimental error. Mylan’s
experts, Dr. Atwood and Dr. Ruffalo, asserted such a construction was supported by the intrinsic
evidence (i.e., the context of the claims, statements found in the written description, and
statements made during the prosecution history of the ‘085 Patent) as well as extrinsic evidence
(i.e., the use of XRPD to characterize crystalline solid states in the prior art, and prior uses of the
word “characterized” in this context.) (See dkt. 232-16 at 53-56; dkt. 232-18, at 35-36.) Dr.
Atwood also described AstraZeneca’s proposed construction for this term as “incomplete,”
asserting it does not require the compound to be characterized by XRPD. (See dkt. 232-16 at 5657.)
148

149

(Id.)

Mylan’s construction for this term was almost identical to Perrigo’s, in that Mylan’s
proposed construction required the X-ray powder diffraction pattern of the claimed compound to
150

97

Dr. Buckton explained at his deposition that “normal experimental error” in the
context of his proposed construction means a shift in the positions of peaks, which may
be caused by different instruments.151 He further explained that there may be variation in
peak intensity between different samples analyzed by the instrument, and that it was
possible for weak peaks to be absent as a result of experimental error, concepts he
described as being understood by a POSA.152
4.

Court findings and claim constructions
a.

“the magnesium salt of S-omeprazole trihydrate”

The Court has considered the parties’ respective arguments, the patent claims, the
specification, the prosecution history, and the relevant extrinsic evidence. The Court
adopts Andrx’s proposed construction and construes the claim term “the magnesium salt
of S-omeprazole trihydrate” as used in the specified claims of the ‘085 and ‘070 Patents,
to mean “the magnesium salt of S-omeprazole in crystalline form containing three
molecules of water of crystallization per molecule of magnesium salt of S-omeprazole.”

be the same as Figure 1 of the ‘070 Patent. Mylan’s experts asserted such a construction was
supported by the intrinsic evidence (i.e., statements found in the written description and the
specificity used in the Langkilde Declaration when describing a comparison to the X-ray
diffractogram) and the extrinsic evidence discussed in the footnote immediately above. (See dkt.
232-16 at 58-60; dkt. 232-18 at 36-37.)
151

(Perrigo action, dkt. 50-3 at 16.)

152

(See id. at 17-18.)

98

In arriving at our construction, we understand that we do not write on a blank
slate.153 We are cognizant that we depart from Judge Pisano’s construction in some
aspects, and agree with it in others. As discussed above, the major disagreement in this
matter concerns the term “trihydrate” and its meaning to the POSA at the time of the
invention. For the following reasons, the Court finds that the POSA would understand
the term “trihydrate” to indicate a crystalline form. The Court declines to adopt the
additional purity limitations sought by Perrigo and Mylan because the Court finds no
basis to do so.
We being our analysis by focusing on the additional purity limitations sought by
Perrigo and Mylan in their proposed constructions, namely “substantially free from
magnesium salts of R-omeprazole and other forms of magnesium salts of S-omeprazole”
(Perrigo) and “substantially pure” (Mylan). Neither phrase appears in the language of
any claim of the ‘085 or ‘070 Patents. The language appears in the specification and the
prosecution history.

153

As discussed in Section I.C.2, supra, Judge Pisano previously construed this term to
mean “a trihydrate of a magnesium salt of S-omeprazole containing three molecules of water per
molecule of magnesium salt of S-omeprazole,” the same construction AstraZeneca advocates for
here. He rejected the construction proposed by the defendants, “a trihydrate of a magnesium salt
of S-omeprazole containing three molecules of water per molecule of magnesium salt of Someprazole in a unit cell of the crystal lattice that is substantially free from magnesium salts of
R-omeprazole and other prior art forms of magnesium salts of S-omeprazole including Someprazole dihydrate and amorphous forms,” finding that there was no basis to import the
limitations sought by the defendants in that case into the claim language.

99

Perrigo and Mylan respectively argue that these phrases in the written description
and prosecution history amount to disavowal and disclaimer.154 The Court does not
agree. A disclaimer or disavowal of claim scope must be clear and unmistakable,
requiring “words or expressions of manifest exclusions or restriction” in the intrinsic
record. Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1327 (Fed. Cir. 2002).
The language Mylan refers to in the written description is far from a clear and
unmistakable disavowal. (See ‘070 Pat., col. 2, ll. 18-20 (“It is an object of the present
invention to provide substantially pure magnesium salt of S-omeprazole trihydrate. . . .”
(emphasis added)).) Setting forth “objects of the present invention” in the written
description is “a common practice in patent drafting.” Pacing Techs., LLC v. Garmin
Int’l, Inc., 778 F.3d 1021, 1025 (Fed. Cir. 2015). “The characterization of features as ‘an
object’ or ‘another object,’ or even as a ‘principal object,’ will not always rise to the level
of disclaimer.” Id. There is no statement that the invention (i.e., the magnesium salt of
S-omeprazole trihydrate) is required to or must accomplish this stated “object.” As
AstraZeneca points out, the drafter may have been referring to the processes for obtaining
the compound, not the compound itself. (See dkt. 48 at 20-21.) There is no clear and
unmistakable disavowal that would require this limitation to be imported into the
meaning of this claim term.
The language Perrigo identifies in the written description is similarly insufficient
to find disclaimer:

154

(See, e.g., Perrigo action, dkt. 48 at 13-15, 20, 26-27.)

100

The magnesium salt of S-omeprazole trihydrate obtained
according to the present invention is substantially free from
magnesium salts of R-omeprazole. The magnesium salt of
S-omeprazole trihydrate obtained according to the present
invention is also substantially free from other forms of
magnesium salts of S-omeprazole, such as the
corresponding magnesium salt compounds described in the
prior art, and dehydrates used in the preparation of the
trihydrate compound according to the present invention.
(‘070 Pat., col. 2, ll. 34-43 (emphasis added).) Nothing in these statements constitutes a
mandatory claim limitation that must be read into the claims. The Court does not read
this written description as clearly and unmistakably requiring that the invention, i.e., the
magnesium salt of S-omeprazole trihydrate, include these limitations. Again, the drafter
may have been referring to the processes for obtaining the compound. Claim 1 of the
‘085 Patent and Claim 1 of the ‘070 Patent are compound claims. Each claim, on its face,
does not include such limitations.
In addition, the Court does not read the prosecution history as clearly and
unmistakably requiring the importation of these limitations. In determining whether a
clear and unambiguous disclaimer attaches to particular claim language, it is important to
consider the statements made by the applicant both in the context of the entire
prosecution history and in the then-pending claims. See Ecolab, Inc. v. FMC Corp., 569
F.3d 1335, 1342 (Fed. Cir. 2009) (“Even if an isolated statement appears to disclaim
subject matter, the prosecution history as a whole may demonstrate that the patentee
committed no clear and unmistakable disclaimer.”). In the context of the overall
prosecution histories of the ‘085 and ‘070 Patents, the isolated statements identified by
Perrigo do not meet the high standard for prosecution disclaimer to attach. The Court
101

declines to import these limitations into the claims based on the written description or the
prosecution history.
The Court also disagrees with Perrigo and Dr. Buckton that the written description
defines the term “the magnesium salt of S-omeprazole trihydrate” as a “bulk product.”
That term does not appear anywhere in the written description or the claims, as Dr.
Buckton readily admits. Cf. SmithKline Beecham Corp. v. Apotex Corp., 403 F.3d 1331,
1338-1339 (Fed. Cir. 2005) (finding claim not limited to commercially significant
quantities.) The Applicants never discussed the invention in terms of amount or quantity,
other than providing a “yield” in Examples 1 and 7. (See ‘070 Pat., col. 5, ll. 21-24; col.
9, ll. 45-48.) Thus, the Court will not import any such limitation into the claims at issue.
We now focus on the term trihydrate and its meaning to the POSA at the time of
the invention. We begin our analysis by focusing on the intrinsic evidence, starting with
the claims themselves. The term “the magnesium salt of S-omeprazole trihydrate”
appears in multiple claims. It appears in claim 1 of the ‘085 Patent with additional
language that limits that claim to a specific crystalline form embodiment. (See dkt. 39 at
8.) It appears alone in claim 1 of the later issued ‘070 Patent, without this language or
any further limitation. It appears in claim 2 of the ‘070 Patent, with reference to the Xray diffractogram shown in Figure 1 of that patent, which is also indicative of a specific
crystalline form embodiment. AstraZeneca argues that claim 1 of the ‘085 Patent and
claim 1 of the ‘070 Patent have a different scope. AstraZeneca also argues that claims 1
and 2 of the ‘070 Patent have a different scope. Based on the difference in scope,
AstraZeneca contends that the doctrine of claim differentiation counsels against a
102

crystallinity limitation associated with the term trihydrate. (See dkt. 39 at 8-10; dkt. 48 at
18-19, 28.)
Claim differentiation “only creates a presumption that each claim in a patent has a
different scope; it is not a ‘hard and fast’ rule of construction.” Bristol-Myers Squibb Co.
v. Ben Venue Labs., Inc., 246 F.3d 1368, 1376 (Fed. Cir. 2001) (citing Comark
Commc'ns, Inc. v. Harris Corp., 156 F.3d 1182, 1187 (Fed. Cir. 1998)). “However, the
doctrine of claim differentiation cannot broaden claims beyond their correct scope,
determined in light of the specification and the prosecution history and any relevant
extrinsic evidence.” Multiform Desiccants, Inc. v. Medzam, Ltd., 133 F.3d 1473, 1480
(Fed. Cir. 1998).
The Court finds that claim 1 of the ‘070 Patent is broader than dependent claim 2
of the ‘070 Patent. Claim 1 of the ‘070 Patent is also broader than claim 1 of the ‘085
Patent.155 A comparison of these claims alone reveals as much. Both claim 1 of the ‘085
Patent and claim 2 of the ‘070 Patent include limitations beyond that of claim 1 of the
‘070 Patent. However, the doctrine of claim differentiation cannot broaden the term “the
magnesium salt of S-omeprazole trihydrate” beyond its correct scope. Recognizing that
claim 1 of the ‘070 Patent must be broader than claim 2 of the ‘070 Patent, as well as

155

The Federal Circuit has acknowledged that two claims with different terminology can
define the exact same subject matter. See Tandon Corp. v. U.S. Int'l Trade Comm’n, 831 F.2d
1017, 1023 (Fed. Cir. 1987). As discussed in Section II.B.1.b, supra, the parties have explained
that Figure 1 and Table 1 (which is incorporated into claim 1 of the ‘085 Patent) depict the same
test data. Although claim 1 of the ‘085 Patent and claim 2 of the ‘070 Patent use different
terminology, these claims define the same subject matter, a specific embodiment of a compound
that when analyzed would produce the X-ray diffractogram shown in Figure 1 and the major
peaks listed in Table 1.

103

claim 1 of the ‘085 Patent, the task confronting the Court is to determine the correct
scope of that term, specifically the meaning of the term trihydrate. AstraZeneca contends
that the term trihydrate should be given its customary and ordinary meaning, which it
argues is a chemical compound with three molecules of water. (See dkt. 39 at 12.) All of
the defendants contend that a trihydrate was understood by the POSA to mean a
crystalline form.
The defendants argue that certain statements in the written description suggest the
trihydrate is crystalline. (See, e.g., dkt. 38 at 13-14.) For example, the written
description states that “[t]he compound of the invention is characterized by being highly
crystalline, i.e., having a higher crystallinity than any other form of magnesium salt of Someprazole disclosed in the prior art.” (‘070 Pat., col. 2, ll. 52-55.) It also states that
“[t]he compound of the invention may also be characterized by its unit cell.” (Id., col. 2,
ll. 63-64.) AstraZeneca argues that these statements do not suggest the trihydrate must
always be crystalline, and instead refer to characteristics of various embodiments. (See
dkt. 39 at 10-11.) The Court agrees with AstraZeneca on this point. These statements, as
well as others, emphasize that the compound may exhibit favorable characteristics.
Similar statements state this exactly: “[t]he compound of the invention is advantageous
because it is more stable than the corresponding magnesium salt compounds in the prior
art and is therefore easier to handle and store.” (‘070 Pat., col. 2, ll. 26-28.) “A
description of characteristics does not redefine a compound with an established and
unambiguous structural definition.” SmithKline Beecham Corp. v. Apotex Corp., 403
F.3d 1331, 1339 (Fed. Cir. 2005). Moreover, where a patent includes a long list of
104

different characteristics or embodiments that correspond to features positively recited in
one or more claims, it seems unlikely that the Applicants intended for each claim to be
limited to all of these characteristics. See ‘085 Pat., claims 2, 3.
The defendants also emphasize that the only two examples provided in the
specification of the trihydrate form yield a crystalline product, suggesting that only a
crystalline form was contemplated. (See, e.g., dkt. 38 at 14.) This may be true, but the
Federal Circuit has repeatedly held that it is “not enough that the only embodiments, or
all of the embodiments, contain a particular limitation” to limit claims beyond their plain
meaning. Thorner v. Sony Computer Entm't Am. LLC, 669 F.3d 1362, 1366 (Fed Cir.
2012); GE Lighting Sols., LLC v. AgiLight, Inc., 750 F.3d 1304, 1309 (Fed. Cir. 2014).
The Court declines to limit this term based on the statements made in the specification.
The written description does not provide a definition for the term trihydrate.
However, AstraZeneca argues that it does provide an explanation that hydrates can exist
in crystalline as well as amorphous forms. (See dkt. 48 at 13-14.) Specifically,
AstraZeneca points to the following language:
The compound of the invention is characterized by being
highly crystalline, i.e., having a higher crystallinity than any
other form of magnesium salt of S-omeprazole disclosed in the
prior art. With the expression “any other form” is meant
anhydrates, hydrates, solvates, and polymorphs or amorphous
forms thereof disclosed in the prior art. Examples of any other
forms of magnesium salt of S-omeprazole includes, but are not
limited
to
anhydrates,
monohydrates,
dehydrates,
sesquihydrates, trihydrates, alcoholates, such as methanolates
and ethanolates, and polymorphs or amorphous forms thereof.

105

(‘070 Pat., col. 2, ll. 44-62.) The Court disagrees. This language is far from clear. It
does not unequivocally suggest such a teaching or the understanding of the Applicants.
The written description itself fails to describe the bounds of the term trihydrate.
Turning to the prosecution history, the defendants point to various excerpts from
the file histories of both patents and contend they support the notion that the trihydrate
was understood to be crystalline. (See dkt. 38 at 14-15.)156 Most citations are to the file
history of the ‘085 Patent. Many of the statements cited by the defendants are the same
or similar to the statements found in the written specification. In fact, some are direct
quotes from the specification. AstraZeneca interprets the prosecution history as clear
evidence that the ‘070 Patent has a broader scope than the crystalline trihydrate of claim
1 of the ‘085 Patent. (See dkt. 39 at 11-12; dkt. 48 at 23-26.) AstraZeneca further argues
that statements made in the ‘070 prosecution history do not support the defendants’ views
regarding crystallinity and that the defendants put too much emphasis on the prosecution
history of the ‘085 Patent.
As explained above, we do not disagree with AstraZeneca with respect to the
difference in scope of claim 1 of the ‘070 Patent and claim 1 of the ‘085 Patent. The
USPTO acknowledged a difference in scope when it withdrew the statutory double
patenting rejection during the prosecution of the ‘070 Patent. See n. 73 and
accompanying text, supra. While some statements in the prosecution history may be
viewed as suggesting the trihydrate is crystalline, other statements do not. With respect

156

(See Perrigo action, dkt. 50 at 19-20.)

106

to the ‘070 Patent’s file history, there is a lack of any conclusive statements regarding
crystallinity.
After reviewing the specification and prosecution histories in detail, the question
of whether the term “trihydrate” was understood by the POSA to connote a crystalline
form is left unanswered. As a result, the Court must turn to the extrinsic evidence.
After thoroughly reviewing the extrinsic evidence, the Court finds the testimony
and exhibits of Andrx’s expert, Dr. Zaworotko most persuasive. He opined that the term
“trihydrate” was understood by the POSA to refer to a crystalline form that has three
water molecules, entrapped in its crystal lattice, for every molecule of compound. Each
of the texts he cited, (see n. 111, supra), were consistent with the definition he provided
and evidenced a common understanding of the term at the relevant time. Two of these
texts were authored by AstraZeneca’s expert, Dr. Byrn. Dr. Byrn’s writings from 1994
and 1995 made clear distinctions between hydrates and amorphous forms on the basis of
crystallinity. See n. 113, supra. Based on these references, the Court is persuaded that
the POSA would have understood the term “trihydrate” to mean a crystalline form.
The references advanced to the contrary by Dr. Byrn were not persuasive. The
dictionary definitions relied on by Dr. Byrn are too broad. See n. 91, supra. Both
definitions explain that there were three waters, but do not provide any detail regarding
the form in which those waters take. Thus, the dictionary definitions fail to properly
define the term “trihydrate” as it would be understood by the POSA looking at these
pharmaceutical patents-in-suit. The references Dr. Byrn initially cited to support the
notion that amorphous hydrates were known at the time of the invention were well after
107

the May 30, 1997 priority date, and thus were not evidence of what was known at the
time of the invention. (See n. 89, supra.) The six additional references he cited fail to
conclusively establish that amorphous hydrates were known at the relevant time. (See n.
90, supra.)
Additionally, the Court does not agree with Dr. Byrn’s assertions that solid state
materials exist on a continuum of amorphous and crystalline material. See n. 95, supra.
His prior writings directly contradict this position. See n. 113, supra. His prior writings
also directly contradict his characterization of a hydrate as having some structure that
“can bind water in a regular way.” See n. 88 and accompanying text, supra.
As discussed above, the doctrine of claim differentiation cannot broaden the claim
beyond its correct scope. Here, it cannot transform the meaning of a “trihydrate” to
encompass amorphous forms. Yet, claim 1 of the ‘070 Patent must be broader than claim
2 of the ‘070 Patent (and claim 1 of the ‘085 Patent). The Court finds that it is. As
explained by Dr. Zaworotko, compounds existing in more than one crystal structure are
“polymorphic,” which means the compound can crystallize into more than one distinct
crystal structure through different crystal packing motifs. Thus, claim 1 of the ‘070
Patent may cover all crystalline polymorphs of S-omeprazole magnesium trihydrate. Dr.
Byrn’s writings are consistent on this point. See n. 88, supra. Claim 1 of the ‘070 Patent
is broader than both claim 1 of the ‘085 Patent and claim 2 of the ‘070 Patent, which are
both limited to a particular crystalline polymorph of that trihydrate. Construing the term
trihydrate to connote a crystalline form does not violate the doctrine of claim
differentiation.
108

After independently assessing the claims, the specification, the prosecution
history, and the relevant extrinsic evidence, the Court finds Andrx’s construction to be
most accurate in declaring the meaning of this claim term as of the priority date in 1997.
In adopting Andrx’s proposed construction for this term, the Court must be
cognizant of the crystalline definition of prior related cases.157 Judge Pisano previously
construed the term “in crystalline form” found in claims 3, 6, 9, and 12 of the related ‘872
Patent to mean “at least some of the magnesium salt of esomeprazole present is in a solid
with a repeating pattern of atoms or molecules of the constituent chemical species,”
adopting AstraZeneca’s proposed construction. AstraZeneca AB v. Dr. Reddy’s
Laboratories, Ltd., No. 05-5553, 2010 WL 1981790, at *3 (D.N.J. May 18, 2010). The
defendants’ construction, “a solid in which the constituent molecules are arranged in an
orderly, repeating pattern or lattice in all three spatial dimensions,” was rejected. Id. The
disagreement between the parties in that case was the degree of crystallinity the term
required, i.e., the compound having “some” degree of crystallinity versus the compound
being entirely crystalline. Id. The Court resolved the dispute by relying on the intrinsic
and extrinsic evidence, which suggested that a reference to “crystalline” material does not
require that the material be completely or even mostly crystalline. See id. This issue is
not before us. We were not asked to define the bounds of crystallinity as closely as Judge

See, e.g., AstraZeneca AB v. Dr. Reddy’s Laboratories, Ltd., No. 05-5553, 2010 WL
1981790 (D.N.J. May 18, 2010) (construing the term “in crystalline form” of the ‘872 Patent and
the term “substantially crystalline form” of the ‘504 Patent); AstraZeneca AB v. Dr. Reddy’s
Laboratories, Inc., Nos. 11-2317, 11-4275, 11-6348, 2013 WL 1847639 (D.N.J. May 1, 2013)
((construing the term “in crystalline form” of the ‘872 Patent and the term “substantially
crystalline form” of the ‘504 Patent).
157

109

Pisano did. The distinction the experts made in this case, and the one we have
determined, was whether the term trihydrate was understood by the POSA to mean a
crystalline form, or was broad enough to include amorphous forms. If we must define the
bounds of crystallinity, we leave it for another day.
b.

“characterized by the following major peaks in its X-ray
diffractogram”

The Court adopts Perrigo’s proposed construction and construes the claim term
“characterized by the following major peaks in its X-ray diffractogram,” as used in the
specified claims of the ‘085 Patent, to mean “having each of the referenced major peaks
in its X-ray powder diffractogram within normal experimental error.”
The Court rejects AstraZeneca’s proposed construction because it improperly
conflates claim construction and infringement, two separate steps in the infringement
analysis. See Conroy v. Reebok Int’l, Ltd., 14 F.3d 1570, 1572 (Fed. Cir. 1994) (“The
determination of infringement requires a two-step analysis: (1) a proper construction of
the claim to determine its scope and meaning, and (2) a comparison of the properly
construed claim to the accused device or process.” (citation omitted)). The focus at this
stage of the proceedings is determining the scope and meaning of the claim terms, not
what evidence, tests, or techniques may be used to prove infringement. To be clear, the
Court does not express any view regarding how “normal experimental error” will affect
the infringement analysis, whether a potentially infringing sample must be compared
using XRPD to prove infringement, or whether infringement may be shown by another
technique such as Fourier transform-infrared (“FT-IR”) spectroscopy, Raman
110

spectroscopy, etc. (See ‘070 Pat., col. 2, ll. 44-47.) Those issues shall be addressed by
the parties during the infringement portion of this case. For now, the Court’s task is to
determine the scope and meaning of this claim term.
Claim 1 of the ‘085 Patent claims a specific embodiment of a compound, “the
magnesium salt of S-omeprazole trihydrate,” and claims that specific embodiment by
listing 13 major peaks of its X-ray diffractogram and their relative intensities. Those 13
major peaks are significant. For example, AstraZeneca explains that claim 1 is limited to
a crystalline form because “[t]he tabulated peaks of claim 1 are indicative of crystalline
material.” (Dkt. 39 at 8.) AstraZeneca does not point to less than all 13 peaks in making
this statement, but refers to the peaks collectively. (See id.)
The prosecution history of the ‘085 Patent reveals a similar usage of the 13 major
peaks. For example, prior to amending claim 1 to include the limitations of claim 3, the
Applicants stated in an Office action response that “[t]he claimed magnesium salt of Someprazole trihydrate is highly crystalline (claim 2) and is uniquely characterized by an
X-ray powder diffractogram (claim 3) which distinguishes the claimed compound from
any other form of the magnesium salt of S-omeprazole.” (Dkt. 38-1 at 79-80 (emphasis
in original).) Again, this statement does not refer to less than all 13 major peaks, it
addresses them collectively. More importantly, this statement recognizes the significance
of those peaks, to uniquely define the claimed compound in order to distinguish it from
other forms, which is precisely what AstraZeneca has done in its papers. (See, e.g. dkt.
39 at 8.) The written description is consistent on this point. It explains that the
compound of the invention may be characterized by the positions and intensities of the
111

major peaks and may be distinguished based on those characteristics. (See ‘070 Pat., col.
2, ll. 43-51.)
The parties acknowledge that the plain language of claim 1 indicates that the
compound is characterized by the listed 13 major peaks. The Applicants elected to claim
the compound of claim 1 by listing the 13 major peaks of its XRPD. The written
description reveals that the XRPD depicted in FIG. 1 included “some additional very
weak peaks,” however the Applicants chose to omit those additional very weak peaks in
Table 1, (see ‘070 Pat., col. 5, ll. 49-50) and also in claim 1. The additional very weak
peaks could have been shown in Table 1 and claimed. They were not. Based on
AstraZeneca’s arguments, some of the 13 claimed major peaks shown in Table 1 and
claimed could have been omitted. However, they were not. Further, the Applicants
could have elected to characterize the compound by conventional FT-IR spectroscopy
and claim it accordingly, however they did not. The Applicants had a variety of options,
but chose to claim the compound in a certain way, and they must live with that choice.
See Ethicon Endo-Surgery, Inc. v. U.S. Surgical Corp., 93 F.3d 1572, 1583 (Fed. Cir.
1996) (“[The Patentee] need not have included this limitation in its claims. Having done
so, it must live with the language it chose.”)
Although the Court construes claim 1 of the ‘085 Patent to have each of the 13
major peaks recited in the claim, the Court is mindful that construing the claims to
require an exact match is too rigid. The parties and their experts all agree that there will
be some range of normal experimental error in an X-ray powder diffractogram. For
example, both Dr. Byrn and Dr. Atwood opined that relative intensities can vary to some
112

extent. The written description itself states that the “relative intensities are less reliable.”
(‘070 Pat., col. 5 ll. 36-38.) For this reason, the Court rejects Andrx’s proposed
construction that would require each peak to fall within the exact range for relative
intensity recited for that peak.
This Court’s construction is not inconsistent with the construction Judge Pisano
arrived at in AstraZeneca AB v. Dr. Reddy’s Laboratories Inc., No. 11-2317 (JAP), 2013
WL 1847639 (D.N.J. May 1, 2013). As discussed in Section I.C.3, supra, Judge Pisano
rejected the defendants’ proposed construction (“having all of the referenced major peaks
in its X-ray diffractogram”) as too rigid, noting it failed to account for experimental error.
However, Judge Pisano did not construe the term to include less than all of the recited 13
major peaks. Instead, he held that a perfect one-to-one relationship was not required
because “the positions for the peaks may differ somewhat because of slight experimental
errors.” This Court’s construction, accounting for normal experimental error,
satisfactorily addresses Judge Pisano’s concerns.
The Court rejects AstraZeneca’s proposed construction because it focuses on the
question of infringement, and not the meaning and scope of the claim. Much of
AstraZeneca’s briefing with respect to this claim term focuses on how the POSA would
compare a diffractogram for a tested compound to a reference diffractogram to determine
whether there is a match for purposes of infringement. (See dkt. 39 at 14-16 (“[A] person
of ordinary skill in the art would recognize that while the S-omeprazole trihydrate at issue
should be ‘identifiable by reference’ to a diffractogram including the listed peaks, an
infringing magnesium salt of S-omeprazole trihydrate need not necessarily contain all of
113

the listed peaks.”); dkt. 48 at 32-33 (“Peaks from other components in [a drug] product,
such as excipients, may interfere with the peaks attributable to the compound”).)
Whether peaks from other excipients in a drug product would interfere with the peaks
attributable to the compound is irrelevant at this juncture. The Court is not focused on a
tested compound. Rather, the concern is defining the proper meaning and scope of this
claim term.
After independently assessing the claims, the specification, the prosecution
history, and the relevant extrinsic evidence, the Court finds Perrigo’s construction to be
most accurate in declaring the meaning of this claim term.
c.

“represented by FIG. 1”

The Court adopts Perrigo’s proposed construction and construes the claim term
“represented by FIG. 1,” as used in the specified claims of the ‘070 Patent, to mean
“having an X-ray powder diffractogram that is the same as Figure 1 of the ‘070 Patent
within normal experimental error.”
Claim 2 references and incorporates Figure 1 of the ‘070 patent, which shows the
pattern resulting from an XRPD analysis. See ‘070 Pat., Fig. 1. As discussed in Section
II.B.3, supra, the level of sameness between Figure 1 required to identify the claimed
compound is at the heart of the dispute with respect to this claim term.
The written description does not define “represented by.” In fact, that term
appears for the first and only time in the ‘070 Patent in claim 2. AstraZeneca argues that
no construction is required and that the “plain meaning” of this term should govern, (see
dkt. 39 at 17-18), but does not provide a clear basis for determining what the “plain
114

meaning” of this term was. AstraZeneca asks this Court to reject each of the defendants’
proposed constructions as too rigid, arguing that they require an exact match with the Xray powder diffractogram of Figure 1 to identify the claimed compound. AstraZeneca
argues that the “complies with” language describing the analysis of the product yielded
by Example 7, (see ‘070 Pat., col. 9, ll. 49-50), “makes it clear that ‘represented by’ does
not mean ‘the same as.’” (Dkt. 39 at 18.)
Contrary to AstraZeneca’s assertion, the defendants’ proposed constructions do
not require an exact match with the X-ray powder diffractogram of Figure 1 to identify
the claimed compound. Each of the defendants’ proposed constructions, including the
one adopted by this Court, accounts for normal experimental error. Construing this claim
term to require an exact match, disregarding variances due to normal experimental error,
would unduly narrow the scope of the claim.
AstraZeneca’s argument pertaining to the “complies with” language reveals why
AstraZeneca’s proposed construction must be rejected. AstraZeneca argues that the
“complies with” language demonstrates that an exact match is not required. However,
AstraZeneca fails to suggest any level of sameness to satisfactorily identify the
compound. The written description does not explain what is meant by “complies with.”
The entire sentence including that language states that “[t]he product [of Example 7] was
analyzed using X-ray powder diffraction and the result complies with FIG. 1 and Table
1.” (‘070 Pat., col. 9, ll. 49-50.) The written description does not report any XRPD data
or provide a diffractogram for the product of Example 7. There is simply no way to
determine the level of sameness between Figure 1 and the results of Example 7’s X-ray
115

powder diffraction analysis because the latter was omitted by the Applicants.
AstraZeneca infers that they are different in some way, but there is nothing to suggest
that they were not identical. Accepting AstraZeneca’s construction would create
ambiguity and leave the disagreement between the parties over this claim term
unresolved.
The Applicants chose to claim the compound by referencing Figure 1, and they
must live with that choice. See Ethicon Endo-Surgery, Inc., 93 F.3d at 1583. Figure 1
shows the pattern resulting from an XRPD analysis. The Applicants claimed a specific
embodiment of a compound that when analyzed would produce a specific pattern. As all
the parties and experts agree, that pattern would be subject to some variance within
normal experimental error. As a result, the Court finds Perrigo’s construction to be most
accurate in declaring the meaning of this claim term.
Again, this Court’s construction is not inconsistent with Judge Pisano’s
construction of this term. See AstraZeneca AB, 2013 WL 1847639, at *9. As discussed
in Section I.C.4, supra, Judge Pisano rejected the defendants’ proposed construction
(“having an X-ray diffractogram the same as FIG. 1,”) as too rigid. He viewed the
defendants’ proposed construction as requiring an X-ray diffractogram that is perfectly
identical to Figure 1. Id. This Court’s construction does not require perfect identity and
accounts for normal experimental error.

116

III.

CONCLUSION
For the reasons stated above, the Court hereby construes the claim terms as

follows:
the claim term “the magnesium salt of S-omeprazole trihydrate,” as used in the
specified claims of the ‘085 and ’070 Patents, means “the magnesium salt of Someprazole in crystalline form containing three molecules of water of crystallization per
molecule of magnesium salt of S-omeprazole;”
the claim term “characterized by the following major peaks in its X-ray
diffractogram,” as used in the specified claims of the ‘085 Patent, means “having each of
the referenced major peaks in its X-ray powder diffractogram within normal experimental
error;” and
the claim term “represented by FIG. 1,” as used in the specified claims of the ‘070
Patent, means “having an X-ray powder diffractogram that is the same as Figure 1 of the
‘070 patent within normal experimental error.”
The Court will issue an appropriate order.

s/ Mary L. Cooper
MARY L. COOPER
United States District Judge
Dated: January 11, 2017