Merck Sharp & Dohme Corp.

Patent Trials and Appeals Board

May 10, 2021

IPR2020-01060 (P.T.A.B. May. 10, 2021)

Trials@uspto.gov Paper 91
571-272-7822 Entered: May 7, 2021
UNITED STATES PATENT AND TRADEMARK OFFICE

____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

MYLAN PHARMACEUTICALS INC.,
Petitioner,

v.

MERCK SHARP & DOHME CORP.,
Patent Owner.
____________

IPR2020-000401
Patent 7,326,708 B2
____________

Before SHERIDAN K. SNEDDEN, ROBERT A. POLLOCK, and
TIMOTHY G. MAJORS, Administrative Patent Judges.

MAJORS, Administrative Patent Judge.

JUDGMENT
Final Written Decision
Determining No Challenged Claims Unpatentable
35 U.S.C. § 318(a)
Denying Patent Owner’s Motion to Exclude
37 C.F.R. § 42.64

1 Dr. Reddy’s Laboratories, Inc. and Dr. Reddy’s Laboratories, Ltd. were
joined as parties to this proceeding via Motion for Joinder in IPR2020-
01060; and Sun Pharmaceuticals Industries Ltd. was joined as a party to this
proceeding via Motion for Joinder in IPR2020-01072.
IPR2020-00040
Patent 7,326,708 B2
2
I. INTRODUCTION
Mylan Pharmaceuticals Inc. (“Petitioner” or “Mylan”),2 on October
30, 2019, filed a Petition to institute inter partes review of claims 1–4, 17,
19, and 21–23 of U.S. Patent No. 7,326,708 B2 (Ex. 1001, “the ’708
patent”). Paper 1 (“Pet.” or “Petition”). On May 12, 2020, based on the
preliminary record, we instituted inter partes review of the challenged
claims on all asserted grounds. Paper 21 (“Inst. Dec.”).
After institution, Patent Owner Merck Sharp & Dohme Corp. (“Patent
Owner” or “Merck”) filed a Response. Paper 41 (“PO Resp.”). Petitioner
filed a Reply. Paper 65 (“Reply”). Patent Owner filed a Sur-reply. Paper
74 (“Sur-reply”). Also before us is Patent Owner’s Motion to Exclude (see
Papers 81, 85). We held an oral hearing on February 11, 2021, and the
transcript is on file. Paper 90 (“Tr.”).
As a brief overview, the claims here relate to a compound called
“sitagliptin” and, specifically, to particular dihydrogenphosphate (“DHP”)
salt forms of it that have a 1-to-1 ratio, or stoichiometry, between the
relevant phosphate anion and the corresponding sitagliptin cation. Pet. 1–2;
PO Resp. 1 (discussing “1:1 sitagliptin DHP”); Ex. 1001, 2:44–65, 15:64–
16:15 (claim 1). Sitagliptin is among a class of compounds known as
dipeptidyl peptidase-IV inhibitors, which can inhibit an enzyme implicated
in the etiology of non-insulin dependent diabetes mellitus (i.e., Type 2
diabetes). Id. at 1:3–36. Indeed, Merck developed and sells its drug

2 Petitioner identifies itself, Mylan Inc., and Mylan N.V. as the real parties-
in-interest. Pet. 6.
IPR2020-00040
Patent 7,326,708 B2
3
product, Januvia, which is indicated for treatment of Type 2 diabetes and
includes a 1:1 sitagliptin DHP salt. PO Resp. 1, 25–26; Ex. 2003 ¶ 2.3
The dispute in this case focuses, in large part, on whether an earlier-
filed international patent application, which Merck also owns, expressly or
inherently discloses the 1:1 sitagliptin DHP salt claimed in the ’708 patent.4
At institution, and despite our determination that this prior art included no
explicit disclosure of a phosphate salt of sitagliptin having the 1:1
stoichiometry, we nevertheless instituted trial based, inter alia, on testimony
from Petitioner’s expert that sitagliptin can only be mono-protonated and
reacting sitagliptin with phosphoric acid forms the 1:1 DHP salt “every
time” and is, thus, inherent. Inst. Dec. 52–53 (noting preliminary record
“suggest[s] the 1:1 salt is the necessary byproduct of contacting phosphoric
acid and sitagliptin”). Because it is undisputed that the prior art does not
expressly disclose the specific 1:1 DHP salt of sitagliptin,5 and the evidence
through trial now shows that sitagliptin can form phosphate salts in non-1:1
ratios without necessarily forming the 1:1 salt (i.e., no inherency), Merck
argues that Petitioner’s anticipation challenge fails. PO Resp. 6–19.

3 Merck has indicated that “the crystalline monohydrate form of the DHP
salt . . . is the solid form of sitagliptin used today in Merck’s products.”
Paper 10, 4–5.
4 The published international patent application (WO 03/004498) and a U.S.
counterpart patent (US 6,699,871; also asserted here as anticipating art)
contain materially “identical” disclosures. See Pet. 33; Tr. 7:8–13.
5 See Tr. 15:7–19 (Petitioner’s counsel agreeing that “there’s no express
disclosure of a 1:1 DHP salt of sitagliptin in the WO [’498] reference or the
’871 reference”); Ex. 2103 ¶ 67.
IPR2020-00040
Patent 7,326,708 B2
4
If anticipation fails, Petitioner is left with obviousness. But, in
Merck’s telling, the obviousness challenge fares no better because Merck’s
inventors reduced to practice the subject matter of almost all the challenged
claims before the key prior art published, thus disqualifying that art as a
§ 102(a) reference; and, even if that art still qualifies under § 102(e),
Merck’s common ownership of the art eliminates it from the obviousness
analysis under § 103(c)(1).6 PO Resp. 22–28. For the two dependent claims
for which Merck does not argue an earlier reduction to practice, Merck
contends those claims are not obvious because, among other things, that
claimed subject matter was highly unpredictable and Petitioner failed to
show a reason why it would have been made by an ordinarily skilled person
with a reasonable expectation of success. Id. at 38–59.
We address in detail the parties’ arguments on anticipation and
obviousness in the sections below. On this trial record, however, we find
Petitioner has failed to show by a preponderance of the evidence that claims
1–4, 17, 19, and 21–21 are unpatentable. Petitioner has, thus, not met its
burden and proved unpatentability of the challenged claims. 35 U.S.C.
§ 316(e). Our reasoning is detailed in Section II below.
We also deny Patent Owner’s Motion to Exclude. Infra Section III.

6 Under the pre-AIA § 103(c)(1) exception, subject matter developed by
“another person” that qualifies as prior art under § 102(e) can be eliminated
from use in an obviousness analysis if that subject matter and the claimed
invention are commonly owned or under obligation of assignment to the
same person or entity at the time of the invention. 35 U.S.C. § 103(c)(1).
IPR2020-00040
Patent 7,326,708 B2
5
Related Patents and Proceedings
“[T]here are no related United States patents or pending applications”
and “this is the first IPR directed to the ’708 patent.” Pet. 7, 67.
Petitioner identifies several related cases before the courts including,
without limitation: Merck Sharp & Dohme Corp. v. Mylan Pharm. Inc. et
al., 1:19:-cv-00101 (N.D. W. Va.); Merck Sharp & Dohme Corp. v. Mylan
Pharm. Inc. et al., 1:19-cv-01489 (D. Del.); and Merck Sharp & Dohme
Corp. v. Sandoz, Inc., 1:19-cv-00312 (D. Del.). Pet. 6–7 (listing cases).
Patent Owner states that it “filed Hatch-Waxman suits alleging infringement
of the ’708 patent, among others, against fourteen generic drug companies
including Mylan, Teva, Apotex, Par, Sun, and Sandoz.” Paper 10, 10. The
litigation against the generic drug companies “has been consolidated for
pretrial proceedings in a multidistrict litigation (‘MDL’)” before the district
court in Delaware. Id. (identifying In re Sitagliptin Phosphate (’708 &
’921) Patent Litig., C.A. No. 19-md-2902-RGA (D. Del.)).
There are also related matters filed with the Board. After institution,
other petitioners filed substantially identical petitions challenging claims of
the ’708 patent and requested joinder with Mylan in this proceeding. See
IPR2020-01045 (“Teva” matter); IPR2020-01060 (“Dr. Reddy’s” matter);
IPR2020-01072 (“Sun” matter). We instituted trial in those other matters
and joined the petitioners as parties here. IPR2020-00040, Papers 44–46.
The Dr. Reddy’s and Sun parties remain joined. The Teva parties (Teva
Pharmaceuticals USA, Inc. and Watson Laboratories, Inc.) have settled with
Merck and IPR2020-01045 is terminated. IPR2020-01045, Paper 25. The
Teva parties are no longer joined. IPR2020-00040, Paper 73, 2–3.
IPR2020-00040
Patent 7,326,708 B2
6
Asserted Grounds of Unpatentability
Petitioner asserts six grounds of unpatentability (Pet. 12) as set forth
in the table below:
Claim(s) Challenged 35 U.S.C. § Basis
1–3, 17, 19, 21–23 102(a),
102(e)(2)7

WO ’4988
1–3, 17, 19, 21–23 102(e)(2) ’871 patent9
3, 17, 19, 21–23 103 WO ’498
1–3, 17, 19, 21–23 103 WO ’498, Bastin10
4 103 WO ’498, Bastin, Brittain11
4 103 WO ’498, Brittain

7 The Leahy-Smith America Invents Act, Pub. L. No. 112-29, 125 Stat. 284
(2011) (“AIA”), amended 35 U.S.C. §§ 102 and 103. The ’708 patent’s
claims have an effective filing date before the effective date of those
amendments so we refer to the pre-AIA versions of §§ 102 and 103 here.
8 Edmondson et al., WO 03/004498 A1, published Jan. 16, 2003 (Ex. 1004,
“WO ’498”). WO ’498 published from Application No. PCT/US02/21349,
filed July 5, 2002, which claims priority to US Provisional Application No.
60/303,474, filed July 6, 2001 (Ex. 1012).
9 Edmondson et al., US 6,699,871 B2, issued Mar. 2, 2004 (Ex. 1007, “the
’871 patent”). The ’871 patent issued from an application filed July 5, 2002,
and claims priority to US Provisional Application No. 60/303,474, filed July
6, 2001 (Ex. 1012).
10 Richard J. Bastin et al., Salt Selection and Optimisation Procedures for
Pharmaceutical New Chemical Entities, 4 ORGANIC PROCESS RESEARCH &
DEVELOPMENT 427–435, 2000 (Ex. 1006, “Bastin”).
11 Polymorphism in Pharmaceutical Solids, Harry G. Brittain ed., 1999
(Ex. 1005, “Brittain”).
IPR2020-00040
Patent 7,326,708 B2
7
Petitioner relies on the Declaration of Mukund Chorghade, Ph.D.
(Ex. 1002) and Dr. Chorghade’s Reply Declaration (Ex. 1035), among other
evidence.
Patent Owner relies on the Declaration of Allan S. Myerson, Ph.D.
(Ex. 2101), the Declaration of Adam J. Matzger, Ph.D. (Ex. 2103), and
testimony from several current or former Merck employees (including many
of the ’708 patent’s eight named inventors), among other evidence. See,
e.g., Exs. 2002 (Vydra Decl.), 2003 (Wenslow Decl.), 2004 (Ferlita Decl.),
2005 (Diddle Decl.), 2109 (Herman Decl.), 2124 (Cypes Decl.), 2127
(Hansen Decl.), and 2140 (Shultz Decl.).
The ’708 Patent
The ’708 patent is titled “PHOSPHORIC ACID SALT OF A
DIPEPTIDYL PEPTIDASE-IV INHIBITOR.” Ex. 1001, code (54). The
’708 patent claims priority to non-provisional and provisional patent
applications filed, respectively, on June 23, 2004, and June 24, 2003. Id. at
codes (21), (22), (60). The patent issued February 5, 2008. Id. at code (45).
According to the ’708 patent, “[t]he present invention relates to a
particular salt of a dipeptidyl peptidase-IV inhibitor,” and specifically, the
dihydrogenphosphate (“DHP”) salt of 4-oxo-4-[3-(trifluoromethyl)-5,6-
dihydro[1,2,4]triazolo[4,3-a]pyrazin-7(8H)-yl]-1-(2,4,5-
trifluorophenyl)butan-2-amine. Id. at 1:13–17. The chemical,4-oxo-4-[3-
(trifluoromethyl)-5,6-dihydro[1,2,4]triazolo[4,3-a]pyrazin-7(8H)-yl]-1-
(2,4,5-trifluorophenyl)butan-2-amine, is also known as “sitagliptin.” See
IPR2020-00040
Patent 7,326,708 B2
8
Ex. 2003 ¶ 2; Pet. 1 n.1.12 The structural formula for the DHP salt of
sitagliptin is shown below as formula (I):

Ex. 1001, 2:44–63. This formula reflects a salt with one phosphate anion
associated with one sitagliptin amine cation (with a stereogenic carbon at *).
Id. at 3:46–52 (“[T]he dihydrogenphosphate salt of the present invention is
comprised of one molar equivalent of mono-protonated [sitagliptin] . . . and
one molar equivalent of the dihydrogenphosphate (biphosphate) anion.”).
The ’708 patent states that this salt is “useful for the treatment and
prevention of diseases and conditions for which an inhibitor of dipeptidyl
peptidase-IV is indicated, in particular Type 2 diabetes.” Id. at 1:19–22.
In a section related to background of the invention, the ’708 patent
identifies WO 03/004498 (i.e., WO ’498), which is “assigned to Merck &

12 Petitioner notes, without dispute, that sitagliptin is also the compound with
the chemical name: 7-[(3R)-3-Amino-4-(2,4,5-trifluorophenyl)butanoyl]-3-
(trifluoromethyl)-5,6,7,8-tetrahydro-1,2,4-triazolo[4,3-α]pyrazine. Pet. 1
n.1; Ex. 1004, 47 (Example 7). In citing asserted references and technical
publications in this Decision, we generally use the page numbers added to
the exhibit not the original pagination, except that, for US patents, we use
the column and line number format or other indicia.
IPR2020-00040
Patent 7,326,708 B2
9
Co.” Id. at 1:49–50. The ’708 patent states that WO ’498 “describes a class
of beta-amino tetrahydrotriazolo[4,3-a]pyrazines, which are potent inhibitors
of DP-IV and therefore useful for the treatment of Type 2 diabetes.” Id. at
1:50–52. According to the ’708 patent, WO ’498 “[s]pecifically disclose[s]”
the 4-oxo-4-[3-(trifluoromethyl)-5,6-dihydro[1,2,4]triazolo[4,3-a]pyrazine-
7(8H)-yl]-1-(2,4,5-trifluorophenyl)butan-2-amine. Id. at 1:53–55. The
patent states that “[p]harmaceutically acceptable salts of this compound are
generically encompassed within the scope of WO 03/004498.” Id. at 1:53–
57. “However,” the ’708 patent states, “there is no specific disclosure in the
above reference [(WO ’498)] of the newly discovered monobasic
dihydrogenphosphate salt . . . of structural formula I.” Id. at 1:58–62.
The ’708 patent further notes the following about WO ’498:
The crystalline dihydrogenphosphate salt of the present
invention exhibits pharmaceutic advantages over the free base
and the previously disclosed hydrochloride salt (WO 03/004498)
in the preparation of a pharmaceutical drug product containing
the pharmaceutically active ingredient. In particular, the
enhanced chemical and physical stability of the crystalline
dihydrogenphosphate salt monohydrate constitute advantageous
properties in the preparation of solid pharmaceutical dosage
forms containing the pharmacologically active ingredient.
Id. at 4:19–28.
Challenged Claims
The ’708 patent includes twenty-four claims. Petitioner challenges
claims 1–4, 17, 19, and 21–23. Claims 1, 2, and 4 are illustrative and read as
follows:
IPR2020-00040
Patent 7,326,708 B2
10
1. A dihydrogenphosphate salt of 4-oxo-4-[3-(trifluoromethyl)-
5,6-dihydro[1,2,4]triazolo[4,3-a]pyrazine-7(8H)-yl]-1-(2,4,5-
trifluorophenyl)butan-2-amine of structural formula I:

or a hydrate thereof.
2. The salt of claim 1 of structural formula II having the (R)-
configuration at the chiral center marked with an *

4. The salt of claim 2 characterized as being a crystalline
monohydrate.

Ex. 1001, 15:64–16:30, 16:48–49. Each of the other challenged claims
depends (directly or indirectly) on claims 1 or 2. See, e.g., id. at 17:29–32
(claim 19: method of treating type 2 diabetes with the salt of claim 2),
17:37–18:5 (claim 21: process for preparing the salt of claim 2).
IPR2020-00040
Patent 7,326,708 B2
11
II. ANALYSIS
Principles of Law
“In an IPR, the petitioner has the burden from the onset to show with
particularity why the patent it challenges is unpatentable.” Harmonic Inc. v.
Avid. Tech., Inc., 815 F.3d 1356, 1363 (Fed. Cir. 2016) (citing 35 U.S.C.
§ 312(a)(3)).
To show anticipation under 35 U.S.C. § 102, each and every claim
element, arranged as in the claim, must be found in a single prior art
reference. Net MoneyIN, Inc. v. VeriSign, Inc., 545 F.3d 1359 (Fed. Cir.
2008). The prior art need not, however, use the same words as the claims.
In re Gleave, 560 F.3d 1331, 1334 (Fed. Cir. 2009). The anticipation
inquiry takes into account the literal teachings of the prior art reference, and
inferences the ordinarily skilled person would draw from it. Eli Lilly and
Co. v. Los Angeles Biomedical Res. Inst. at Harbor-UCLA Med. Ctr., 849
F.3d 1073, 1074–75 (Fed. Cir. 2017). Indeed, “a reference can anticipate a
claim even if it does not expressly spell out all the limitations arranged or
combined as in the claim, if a person of skill in the art, reading the reference,
would at once envisage the claimed arrangement or combination.”
Kennametal, Inc. v. Ingersoll Cutting Tool Co., 780 F.3d 1376, 1381 (Fed.
Cir. 2015) (internal quotation marks omitted).
As to obviousness, a claim is unpatentable under 35 U.S.C. § 103 if
the differences between the subject matter sought to be patented and the
prior art are such that the subject matter as a whole would have been obvious
at the time the invention was made to a person having ordinary skill in the
art. KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 406 (2007). The question
IPR2020-00040
Patent 7,326,708 B2
12
of obviousness is resolved on the basis of underlying factual determinations
including: (1) the scope and content of the prior art; (2) any differences
between the claimed subject matter and the prior art; (3) the level of ordinary
skill in the art; and (4) objective evidence of nonobviousness when
presented. Graham v. John Deere Co., 383 U.S. 1, 17–18 (1966). A party
who petitions the Board for a determination of unpatentability based on
obviousness must show that “a skilled artisan would have been motivated to
combine the teachings of the prior art references to achieve the claimed
invention, and that the skilled artisan would have had a reasonable
expectation of success in doing so.” In re Magnum Oil Tools Int’l, Ltd., 829
F.3d 1364, 1381 (Fed. Cir. 2016) (quotations and citations omitted).
Person of Ordinary Skill in the Art
In determining the level of skill in the art, we consider the problems
encountered in the art, the art’s solutions to those problems, the rapidity with
which innovations are made, the sophistication of the technology, and the
educational level of active workers in the field. Custom Accessories, Inc. v.
Jeffrey-Allan Indus., Inc., 807 F.2d 955, 962 (Fed. Cir. 1986).
Petitioner contends a person of ordinary skill in the art (or “POSA”) at
the time of the invention would have had:
(i) a Ph.D. in chemistry, biochemistry, medical chemistry,
pharmacy, pharmaceutics, or a related field, and at least two
years of relevant experience in drug development including an
understanding of salt selection in drug development; (ii) a
master’s degree in the same fields and at least five years of the
same relevant experience; or (iii) a bachelor’s degree in the same
fields and at least seven years of the same relevant experience.
IPR2020-00040
Patent 7,326,708 B2
13
Pet. 11; Ex. 1002 ¶¶ 45–46. Patent Owner does not oppose this definition.13
We find Petitioner’s proposed definition is consistent generally with
the cited prior art, and we apply it for the purposes of this Decision. See
Okajima v. Bourdeau, 261 F.3d 1350, 1355 (Fed. Cir. 2001) (explaining that
specific findings regarding ordinary skill level are not required “where the
prior art itself reflects an appropriate level and a need for testimony is not
shown”) (quoting Litton Indus. Prods., Inc. v. Solid State Sys. Corp., 755
F.2d 158, 163 (Fed. Cir. 1985)).
Claim Construction
We interpret a claim “using the same claim construction standard that
would be used to construe the claim in a civil action under 35 U.S.C.
282(b).” 37 C.F.R. § 42.100(b) (2019). Under this standard, we construe
the claim “in accordance with the ordinary and customary meaning of such
claim as understood by one of ordinary skill in the art and the prosecution
history pertaining to the patent.” Id. “[W]e need only construe terms ‘that
are in controversy, and only to the extent necessary to resolve the
controversy.’” Nidec Motor Corp. v. Zhongshan Broad Ocean Motor Co.

13 Merck’s experts, Drs. Myerson and Matzger, phrase the definition
differently—limiting the fields of study to “chemistry, chemical engineering
or a related field,” and requiring two or more years (depending degree
attained) “working with pharmaceutical solids, including polymorphic
forms.” Ex. 2101 ¶ 39; Ex. 2103 ¶ 40. Neither party contends, however,
that any disputed matter turns on acceptance of one definition or the other.
Merck’s experts state that their opinions do not change under either
definition. Ex. 2101 ¶ 40; Ex. 2103 ¶ 41.
IPR2020-00040
Patent 7,326,708 B2
14
Ltd., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (quoting Vivid Tech., Inc. v. Am.
Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999)).
At institution, neither party requested any express claim construction.
Pet. 10 (“Petitioner submits that no further construction [beyond the claims
as written] is necessary”); Ex. 1002 ¶¶ 57–59.
Claim 21 is a dependent claim and recites a process for preparing the
1:1 (R)-sitagliptin DHP of claim 2. Ex. 1001, 17:37–18:5. Petitioner, in its
Reply, argues that claim 21’s “contacting” limitation “means an actual
physical interaction between molecules.” Reply at 11–12. Petitioner’s
position is that the claim phrase “contacting one equivalent of [sitagliptin] in
an organic solvent or aqueous organic solvent with about one equivalent of
phosphoric acid at a temperature in the range of about 25–100°C” does not
limit the molar amounts of the acidic and basic reagents provided in the
solvent, but rather how discrete molecules actually interact to form the final
salt. Id.; Ex. 1001, 17:38–18:5; Tr. 23:23–25:9. In Petitioner’s view, “the
sitagliptin base molecule is only able to interact once with a phosphoric acid
molecule.” Reply 12. And because a 1:1 salt allegedly always forms when
an acid (e.g., hydrochloric or phosphoric) is reacted with sitagliptin, the
“contacting” step would be met even if the reaction conditions were to
include using a large excess (i.e., a substantially non-equivalent molar
amount) of acid molecules relative to the base, such as the excess
hydrochloric acid in WO ’498’s Example 7.14 Reply 12–13.

14 Put differently, Petitioner’s position is that it does not matter what molar
amounts of the acid and base are used (e.g., 100 moles of acid vs. 1 mole of
IPR2020-00040
Patent 7,326,708 B2
15
Patent Owner responds that Petitioner is abrogating claim 21’s
limitations. According to Patent Owner, Petitioner’s interpretation and
application of the art to claim 21 renders irrelevant: (i) the starting materials
(insofar as the cited art starts with BOC-protected sitagliptin, not sitagliptin
free base); and (ii) the claim’s “one equivalent” language (insofar as the
cited example in the art uses a substantial molar excess of the acid), so long
as a salt with 1:1 stoichiometry might somewhere form in a downstream
process. Sur-reply 13–14; see also Ex. 2103 ¶ 232 (describing WO ’498’s
Example 7 as “using ~1000 fold excess of hydrochloric acid”).
Given the present record, we are unpersuaded that a further
construction of claim 21 is required. Even if we agreed with Petitioner’s
interpretation, Petitioner has not proved that the cited art expressly or
inherently discloses preparing the 1:1 DHP salt of sitagliptin (discussed infra
for anticipation challenges). Alternatively, if claim 21 required use of one
molar equivalent of sitagliptin and about (i.e., “approximately”) one molar

base would still be encompassed); key to Petitioner’s argument is the final
result of a 1:1 salt, thus allegedly proving the reagents will only interact in a
1:1 ratio. We explained at the oral hearing that we found Petitioner’s
interpretation problematic because it cited no intrinsic evidence in support
and also appeared to read the term “about” out of the claim. Tr. 25:2–27:6
(asking how it’s possible to have interaction at a molecular level as
suggested by Petitioner (seemingly requiring whole numbers ratios, like 1:1
or 1:2, in which the molecules might interact and bond) between one
equivalent of a molecule and about one equivalent of another). Petitioner
had no satisfactory response to those concerns. Id.; see also id. at 75:14–21
(Petitioner’s counsel later suggested that even if the “contacting” step
required use of equal molar amounts, using such amounts would be an
obvious change “that you would indeed use”).
IPR2020-00040
Patent 7,326,708 B2
16
equivalent of acid in the solvent (see Ex. 1001, 6:29–55 (disclosing a
process for preparing the claimed 1:1 sitagliptin DHP)), Petitioner has not
shown that WO ’498 describes that reaction. Again, it is undisputed that the
cited example (Example 7 of WO ’498) uses a substantial molar excess of
the acid (HCl) to the base. On obviousness and as explained below (see
Section II(F)), we conclude that Patent Owner’s reduction-to-practice
evidence is dispositive for several claims, including claim 21, and that issue
does not turn on whether we accept or reject Petitioner’s interpretation
(which encompasses, but is not limited to, using equimolar amounts). PO
Resp. 25 (explaining, citing undisputed evidence, how the inventors made
1:1 sitagliptin DHP with “equimolar” amounts of sitagliptin and phosphoric
acid under conditions encompassed by claim 21).
Anticipation by WO ’498
Petitioner asserts that claims 1–3, 17, 19, and 21–23 are unpatentable
as anticipated by WO ’498. Pet. 12–31. We provide an overview of
WO ’498, and then turn to analysis of the alleged anticipation.
1. Overview of WO ’498 (Exhibit 1004)
WO ’498 “is directed to compounds which are inhibitors of the
dipeptidyl peptidase-IV enzyme (‘DP-IV inhibitors’) and which are useful in
the treatment or prevention of diseases in which the dipeptidyl peptidase-IV
enzyme is involved, such as diabetes and particularly type 2 diabetes.”
Ex. 1004, 1. WO ’498 is further “directed to pharmaceutical compositions
comprising these compounds and the use of these compounds and
compositions” for treatment or prevention of the above-noted diseases. Id.
IPR2020-00040
Patent 7,326,708 B2
17
WO ’498 discloses several examples of the compounds of its
invention. Id. at 38–47 (describing Examples 1–7); see also id. at 48–49
(identifying other example compounds, numbered examples 8–33). The salt
of Example 7 is shown in the chemical structure below.

Id. at 47:1–5; see also id. at 47:6–26 (describing steps for preparing the
compound and salt of Example 7). The structure in Example 7 of WO ’498
“depicts the hydrochloride salt of sitagliptin in its (R)-configuration.”
Ex. 1002 ¶ 67.
WO ’498 does not describe or exemplify any specific phosphate salt
of sitagliptin (or any phosphate of the other compounds). More generically,
however, WO ’498 claims (R)-sitagliptin and several other compounds, and
pharmaceutically acceptable salts thereof. WO ’498 claims:
IPR2020-00040
Patent 7,326,708 B2
18
15. A compound which is selected from the group consisting of:
***

***
and pharmaceutically acceptable salts thereof.
Ex. 1004, 55:16–61:5 (claim 15 depicts thirty-three DP-IV inhibitor
compounds, one of which is (R)-sitagliptin (shown in the excerpt above),
and, after depicting all those compounds, recites “and pharmaceutically
acceptable salts thereof”).
WO ’498 elsewhere describes “‘pharmaceutically acceptable salts’
[as] refer[ring] to salts prepared from pharmaceutically acceptable non-toxic
bases or acids including inorganic or organic bases and inorganic or organic
acids.” Id. at 10:27–29. “When the compound of the present invention is
basic,” WO ’498 indicates the “salts may be prepared from pharmaceutically
acceptable non-toxic acids, including inorganic and organic acids,” and
WO ’498 identifies twenty-six illustrative acids. Id. at 11:8–14 (“Such acids
include acetic, benzenesulfonic, benzoic, . . . hydrochloric, . . . phosphoric,
succinic, sulfuric, tartaric, p-toluenesulfonic acid, and the like.”). Among
the twenty-six acids named, WO ’498 discloses that “[p]articularly preferred
are citric, hydrobromic, hydrochloric, maleic, phosphoric, sulfuric, fumaric,
IPR2020-00040
Patent 7,326,708 B2
19
and tartaric acids.” Id. at 11:14–15; see also id. at 11:16–17 (“It will be
understood, as used herein, references to the compounds of Formula I are
meant to also include the pharmaceutically acceptable salts.”).
2. Analysis of Alleged Anticipation
a) The Parties’ Arguments
Petitioner contends that WO ’498 discloses (R)-sitagliptin and its
pharmaceutically acceptable salts. Pet. 16–17 (citing Ex. 1004, 55–60
(“Claim 15 (7th compound)”). Petitioner further argues that WO ’498’s
disclosure of acceptable salts would include at least the eight “[p]articularly
preferred” acids, one of which is phosphoric acid, as a non-toxic acid for use
in forming salts with compounds like sitagliptin. Id. at 16–18; see Ex. 1004,
10:8–15. Thus, Petitioner argues, “WO ’498 teaches the phosphoric acid
salt of sitagliptin” and a specific example is not needed. Id. at 16–18.
According to Petitioner, this case involves anticipation by disclosure
of “lists,” not necessarily anticipation where the art discloses a genus and a
species within that genus are claimed. Pet. 20–22 (citing, e.g., Wm. Wrigley
Jr. Co. v. Cadbury Adams USA LLC, 683 F.3d 1356, 1361 (Fed. Cir. 2012));
see also id. at 23–24 (citing In re Gleave, 560 F.3d 1331, 1331 (Fed. Cir.
2009) (“For purposes of whether they are anticipatory, lists and genera are
often treated differently under our case law.”)).15 More specifically,

15 Whether a disclosed genus anticipates a species ordinarily turns on
whether the prior art “expressly spelled out a definite and limited class of
compounds that enabled a person of ordinary skill in the art to at once
envisage each member of this limited class.” Eli Lilly & Co. v. Zenith
Goldline Pharm., Inc., 471 F.3d 1369, 1376 (Fed. Cir. 2006); Atofina v.
IPR2020-00040
Patent 7,326,708 B2
20
Petitioner contends “WO ’498 provides two closed lists,” where “[t]he
primary list (i.e., Claim 15) provides 33 compounds” and the “secondary”
list “identifies by name the eight” preferred acids. Id. at 22. Because the
first list includes (R)-sitagliptin and the second list includes phosphoric acid,
Petitioner contends “neither list leaves anything to the imagination” and
WO ’498 anticipates claims 1 and 2. Id. at 22–24. Petitioner asserts that
WO ’498’s disclosures even arguably “collapse to form a single
comprehensive list” of all the compounds and salts. Id. at 23–24; Ex. 1002
¶ 80. And, because lists are involved, Petitioner argues “the number of
[listed items] is irrelevant.” Id. at 24 (citing In re Gleave, 560 F.3d at 1333,
1338 (explaining that the art “expressly lists every possible fifteen-base-long
oligodeoxynucleotide sequence in IGFBP-2,” and that the “list include[d]
more than 1400 sequences”)).
Petitioner does not, in the Petition, address the 1:1 stoichiometry
limitation of the claimed salt head-on. To the extent addressed at all,
Petitioner’s discussion appears in parts of a footnote spanning several pages.
Pet. 18–20 n.8. There, Petitioner states that claim 1’s “dihydrogenphosphate
salt” of sitagliptin, “is nothing more than another name for the (monobasic)
phosphoric acid salt of sitagliptin.” Id. Petitioner cites the ’708 patent’s
disclosure on reacting one equivalent of sitagliptin with approximately one

Great Lakes Chem. Corp., 441 F.3d 991, 999 (Fed. Cir. 2006) (“It is well
established that the disclosure of a genus in the prior art is not necessarily a
disclosure of every species that is a member of that genus. . . . There may be
many species encompassed within a genus that are not disclosed by a mere
disclosure of the genus. On the other hand, a very small genus can be a
disclosure of each species within the genus.”).
IPR2020-00040
Patent 7,326,708 B2
21
equivalent of phosphoric acid. Id. (citing Ex. 1001, 6:29–55). And
Petitioner cites Dr. Chorghade’s testimony that sitagliptin can only be mono-
protonated at sitagliptin’s primary amine so that the DHP salt forms “every
time.” Id. (citing Ex. 1002 ¶ 76); see also Ex. 1002 ¶ 77 (noting, with no
further analysis, that Example 7 of WO ’498 is a 1:1 hydrochloride salt). At
institution, we observed that Petitioner’s position was “light” on analysis for
the claimed stoichiometry. Inst. Dec. 53. But we instituted trial, as there
was some preliminary support for the position that a 1:1 stoichiometry was
inherent, and because, at that time, Patent Owner offered in rebuttal only
attorney argument about different protonation states and stoichiometric
ratios. Id. at 53–54.
In its Response at trial, Patent Owner contends that Petitioner “cannot
and does not assert that these references [(WO ’498 or the related ’871
patent)] expressly disclose” 1:1 sitagliptin DHP. PO Resp. 6–7. And, Patent
Owner argues, Petitioner’s critical assumptions necessary to prevail on
anticipation have now been proved incorrect. Id. Thus, whether Petitioner’s
theory is based on lists or “envisaging” the claimed subject matter, or on
“inherency,” Patent Owner contends the challenge fails. Id. at 7–19.
According to Patent Owner, the unrebutted evidence shows that the
salt formation here is unpredictable, it involves “trial and error” processes,
and that, absent sufficient detail, a POSA has no way of knowing whether
any particular salt within a genus of potential salts will form at all. Id. at 8–
10 (citing, e.g., Ex. 2103 ¶¶ 163–71, 86–92; Ex. 2127 ¶ 15; Ex. 2042, 5 (“No
predictive procedure to determine whether a particular acidic or basic drug
would form a salt with a particular counter-ion has been reported in the
IPR2020-00040
Patent 7,326,708 B2
22
literature”); Ex. 2051, 116:22–117:3 (Dr. Chorghade admitting salt
formation lacks predictability and is a “trial and error process”)). Patent
Owner points out that WO ’498 does not expressly describe any phosphate
salts of any compound, much less any phosphate salt of sitagliptin—only a
hydrochloride salt is shown. Id. at 7 (citing Ex. 2103 ¶¶ 72–74; Ex. 1004,
46:1–26 (Example 7)). Also, Patent Owner contends, Petitioner has
improperly narrowed the combinations in WO ’498 when, in fact, there are
millions of compounds and hypothetical salts encompassed by WO ’498
broad teachings (e.g., compounds based on Formula I), none of which were
reviewed by Dr. Chorghade beyond the “33 compounds” in claim 15 and the
preferred “8 counterions.” Id. at 12–13 (citing Ex. 2103 ¶¶ 77–85).
On this record, Patent Owner argues that the “envisaging” doctrine
does not help Petitioner. PO Resp. 8. According to Patent Owner,
envisaging the claimed subject matter might apply “[i]f the genus resulting
from combining the two lists is ‘of such a defined and limited class that one
of ordinary skill in the art could at once envisage each member of the
genus.’” Id. (quoting Wrigley, 683 F.3d at 1361) (internal quotation marks
and citation omitted). But that is not so here, Patent Owner argues, because
the genus is not limited as Petitioner says and “[t]he POSA reading WO ’498
cannot envisage—because she does not know—the genus of salts that will
form.” Id. at 8–9; Ex. 2051 (Chorghade dep.) 63:22–64:3 (“Q: You
wouldn’t know without performing research whether a phosphate salt of
sitagliptin would form without doing that work. Right? A. That is correct”).
Patent Owner emphasizes another allegedly important distinction. To
the extent Petitioner’s anticipation theory requires combining reagents to
IPR2020-00040
Patent 7,326,708 B2
23
perform an unpredictable chemical reaction, Patent Owner contends there is
no similarity to other cases like Wrigley, where a “simple combination . . .
[of] two chewing gum components” was at issue. PO Resp. 11. According
to Patent Owner, the district court in Shire recognized such a distinction in
rejecting the argument that a POSA would at once envisage the claimed
compound from combining and chemically reacting two reagents (L-lysine
and d-amphetamine) allegedly among lists in a prior art reference. Id.
(citing Shire LLC v. Amneal Pharm., LLC, 2014 WL 2861430 * 15 (D.N.J.
June 23, 2014), aff’d in part, rev’d in part and remanded, 802 F.3d 1301
(Fed. Cir. 2015)).16
Patent Owner further argues that Petitioner is foreclosed from using
the “at once envisage” doctrine to fill in limitations missing in WO ’498.
PO Resp. 18. As noted by Patent Owner, “[t]he ‘at once envisage’ test ‘does
not stand for the proposition that a reference missing a limitation can
anticipate a claim if a skilled artisan viewing the reference would at once
envisage the missing limitation.’” Id. (quoting Nidec Motor Corp. v.
Zhongshan Broad Ocean Motor Co., 851 F.3d 1270, 1274–75 (Fed. Cir.
2017) (internal quotation marks omitted)). Because “1:1 sitagliptin DHP” is
not described and that salt’s “1:1 stoichiometry limitation” is “indisputably

16 The district court in Shire, in granting summary judgment of no
anticipation, distinguished Wrigley, finding that the claimed compound
“does not appear to be something that you get just by mixing a little L-lysine
and a little d-amphetamine together” and “[t]here is no dispute that some
chemical reactions have to take place.” Shire, 2014 WL 2861430 at * 15.
The Federal Circuit later reversed and remanded on the issue of induced
infringement. Shire, 802 F.3d at 1310–11.
IPR2020-00040
Patent 7,326,708 B2
24
missing from WO ’498,” Patent Owner contends “the Board cannot ‘fill in
missing limitations simply because a skilled artisan would immediately
envision them.’” Id. at 18–19 (quoting Nidec, 851 F.3d at 1274–75); see
also Sur-reply 1–4 (addressing further the “envisage” theory).
On inherency, Patent Owner argues that undisputed evidence shows
sitagliptin and phosphoric acid also react to form non-1:1 phosphate salts.
Id. at 14–19. So, Patent Owner contends, Petitioner’s and Dr. Chorghade’s
assertions that a 1:1 DHP sitagliptin salt forms “every time” are
demonstrably “wrong.” Id. at 15 (citing Ex. 2013 ¶¶ 99–102, 113–118);
Ex. 1002 ¶ 76. In support, Patent Owner identifies non-1:1 phosphate salts
of sitagliptin reported in the literature—salts Dr. Chorghade did not
consider. PO Resp. 15 (citing Ex. 2220). And Patent Owner’s expert
“Dr. Matzger has actually made other non-1:1 salts.” Id. (citing Ex. 2103
¶¶ 123–176). As Dr. Matzger explains, sitagliptin includes multiple sites
that can accept protons, and phosphoric acid is “triprotic”—meaning it can
donate up to three protons in an acid-base reaction. Id. at 16 (citing
Ex. 2103 ¶¶ 103–112). Indeed, as Patent Owner contends, Dr. Matzger
formed a 3:2 sitagliptin phosphate salt, as well as a 2:1 salt. Id. at 17 (citing
Ex. 2103 ¶¶ 123–38, 148–176).17 According to Patent Owner, other

17 Dr. Matzger testifies, inter alia, that “my experiments have definitively
shown that non-1:1 salts can indeed form in chemical reactions between
sitagliptin and phosphoric acid. . . . [I]t is my opinion that a 1:1 DHP salt of
sitagliptin is not the inevitable result of a chemical reaction between those
species.” Ex. 2103 ¶ 118. As also explained by Dr. Matzger, Exhibit 2220
(WO 2012/166420 (“WO ’420”)) depicts, among other things, a salt with
IPR2020-00040
Patent 7,326,708 B2
25
chemistry professors, including Drs. Jerry Atwood and Leonard Chyall,
respectively from University of Missouri and Purdue University, also
formed non-1:1 salts of sitagliptin and phosphoric acid using conventional
salt-screening protocols. Id. (“Dr. Matzger reproduced one of the non-1:1
sitagliptin phosphate salts of WO ’420 [Ex. 2220] (a salt Dr. Chyall had also
produced), and determined—via multiple techniques—it was a 3:2 salt.”)
(citing, e.g., Exs. 2224, 2225, 2227, 2192; Ex. 2103 ¶¶ 123–135, 148–160).
Patent Owner contends the existence of these non-1:1 salts “precludes a
finding that the 1:1 salt is inherently anticipated.” PO Resp. 17–18; see also
Sur-reply 5–12 (addressing Petitioner’s “inherency” theory).
In Reply, Petitioner restates its view that the claimed salt is found in a
combination of two lists in WO ’498—sitagliptin “among a narrow list of 33
exemplified compounds” and phosphoric acid in a list of eight preferred
acids. Reply 2–5. Petitioner contends that there is no dispute that 1:1 salts
would be encompassed by the claims of WO ’498. Id. at 3. And, in
furtherance of its envisaging theory, Petitioner contends that Patent Owner’s
expert, Dr. Matzger, testified that “you can imagine that the [1:1 phosphoric
acid salt] would exist.” Id. (quoting, in part, Ex. 1025, 146:13–147:5)
(brackets added by Petitioner)). According to Petitioner, Patent Owner’s
emphasis on unpredictability is flawed because the issue is anticipation, not
obviousness, and also the claimed 1:1 sitagliptin DHP was “made easily” in

two molecules of sitagliptin to one molecule of the phosphate anion (a 2:1
salt), as well as a 1:2 salt, with one sitagliptin molecule associated with two
phosphate anions. Ex. 2103 ¶¶ 121–122 (citing Ex. 2220, 8 (Compound 3),
16 (Compound 7)).
IPR2020-00040
Patent 7,326,708 B2
26
a salt-screen by one of the inventors who had little work experience at that
time. Id. at 5.
Petitioner’s Reply raises two primary points on inherency. First, that,
in methanol-based experiments run years ago by Dr. Chyall, the reaction of
phosphoric acid and sitagliptin produced 1:1 sitagliptin DHP—results that
Merck allegedly does not dispute. Reply 5–7 (citing Ex. 222518). Petitioner
contends that because those experiments used “methanol/ambient
conditions” they are consistent with the process in WO ’498’s Example 7.
Id. at 6–8 (citing Ex. 2225 ¶¶ 22–52, 72; Ex. 1035 ¶¶ 22–26, 63–67).
Second, Petitioner argues that Merck’s testing evidence misses the mark
because it does not resemble WO ’498’s Example 7. Id. at 9 (citing
testimony that Dr. Matzger was “not trying to reproduce anything in the
’498” (Ex. 1025, 87:21–88:17)). Insofar as Dr. Matzger’s testing used an
isopropanol/water solvent, not methanol, Petitioner contends that the test
results are less relevant (or irrelevant) to whether the claimed 1:1 sitagliptin
DHP is inherent based on WO ’498. Id. at 8–10; see also id. at 3 (arguing
“Merck turns to non-prior art evidence (WO420) using different conditions
than WO498, to contend that other possible stoichiometries may exist”).

18 Exhibit 2225 is Dr. Leonard J. Chyall’s Declaration (dated Aug. 3, 2010)
apparently prepared for an opposition proceeding in Israel, concerning a
foreign counterpart to the ’708 patent. See Reply 6; Ex. 2225 ¶ 1. Merck
submitted Exhibit 2225, along with other exhibits containing Dr. Chyall’s
prior testimony, when it filed its Patent Owner Response. See PO Resp. 17
(citing Exhibits 2225, 2192, 2224, and 2227). Petitioner and Dr. Chorghade
relied, in the Reply, on some of Dr. Chyall’s experiments, particularly those
in Exhibit 2225 using methanol. Patent Owner moves to exclude in part
Exhibit 2225, which we address in Section III.
IPR2020-00040
Patent 7,326,708 B2
27
b) Whether WO ’498 Discloses 1:1 Sitagliptin DHP
(i) Petitioner’s Lists & “Envisage” Theories
All the challenged claims require a 1:1 sitagliptin DHP salt, as recited
in claims 1 or 2. Ex. 1001, 15:64–16:30 (claim 2 is the (R)-enantiomer of
1:1 sitagliptin DHP). We focus on those claims because whether WO ’498
discloses, expressly or inherently, that phosphate salt of sitagliptin with its
1:1 stoichiometry is decisive. Vergegaal Bros., Inc. v. Union Oil Co. of
Cal., 814 F.2d 628, 631 (Fed. Cir. 1987) (“A claim is anticipated only if
each and every element as set forth in the claim is found, either expressly or
inherently described, in a single prior art reference.”).
It is undisputed that a 1:1 sitagliptin DHP salt is not expressly
disclosed in WO ’498. See Tr. 15:7–19; Ex. 2103 ¶ 67. No phosphate salts
of sitagliptin, or any of the other thirty-two exemplary DP-IV compounds,
are shown in WO ’498, nor are there details given about making them.
Ex. 2103 ¶¶ 74, 81. Petitioner’s declarant, Dr. Chorghade conceded these
points on cross-examination. Dr. Chorghade agreed, for example, that he
was “not aware of any prior process where sitagliptin and phosphoric acid
are being reacted to form a phosphate salt” and that “WO’498 does not
disclose a process in which sitagliptin and phosphoric acid have been
reactive.” Ex. 2283, 50:18–51:6; see also id. at 15:2–12.
What WO ’498 does describe in some detail are hydrochloride salts,
made by reacting a different acid (hydrochloric acid) with several of the
example DP-IV inhibitor compounds. See, e.g., Ex. 1004, 38–47 (Examples
1–7). One of those compounds is sitagliptin. Id. at 47; see also id. at 56
(bottom-depicted compound).
IPR2020-00040
Patent 7,326,708 B2
28
It is true that, in Example 7, WO ’498 describes a process for making
a hydrochloride salt of sitagliptin, which appears to form in a 1-to-1 ratio.
Id. at 47. On this record, however, we credit Dr. Matzger’s testimony that a
POSA would not simply conclude that whatever applies for hydrochloric
acid and sitagliptin also applies to the many different acids that are
disclosed—and phosphoric acid, in particular. See, e.g., Ex. 2103 ¶¶ 103–
113. First, hydrochloric acid and phosphoric acid are different for at least
the reason that hydrochloric acid has only one proton to donate, whereas
phosphoric acid has three. Id. ¶¶ 33–34 (citing references), 113 n. 39
(“Since HCl is monoprotic, the 1:1 stoichiometry of Example 7 is irrelevant
to the question whether a polyprotic acid could form non-1:1 salts with
multiple sitagliptin molecules.”). Second, persuasive evidence shows that
sitagliptin (and structurally-similar analogues) can, indeed, accept multiple
protons. Ex. 2103 ¶ 122 (showing a 1:2 sitagliptin phosphate); id. ¶¶ 103–
105 (discussing Examples 1–5 of WO ’498, in which one molecule of the
analogue is protonated by two HCl ions), 108 (testifying on other potential
ratios with polyprotic acids); Ex. 1004, 38 (Example 1); Ex. 2051
(Chorghade dep.), 152:21–153:10 (agreeing other example compounds are
sitagliptin analogues but declining to concede that they have a non-1:1
stoichiometry). This is consistent with Patent Owner’s position and
Dr. Matzger’s testimony that sitagliptin and analogous compounds have, not
just one, but multiple sites that are capable of accepting protons when
forming salts. Ex. 2103 ¶¶ 103–104 (showing proton-accepting sites at the
primary amine and triazole ring).
IPR2020-00040
Patent 7,326,708 B2
29
Petitioner’s “list(s)” and “envisage” theories do not make up for the
absence of express disclosure of the claimed 1:1 sitagliptin DHP salt on this
record. Even accepting that WO ’498 includes a “list” of 33 example active
compounds and also a “list” of eight preferred acids as possible counterions
to form potential salts with such compounds, there is no “list” that identifies
expressly all the phosphate salts in any, much less all, the potential
stoichiometric ratios. Petitioner does not persuasively argue otherwise. And
that fact distinguishes this case from others, where the relevant subject
matter was listed expressly. In re Gleave, 560 F.3d at 1333, 1338 (holding
that the reference “expressly lists” every possible one of the relevant
sequences).
Petitioner did not attempt to quantify the breadth of compounds and
hypothetical salts encompassed by WO ’498’s disclosures. Nor does
Petitioner rebut Patent Owner’s evidence that does. See, e.g., Ex. 2103
¶¶ 73, 97; Ex. 2051 (Chorghade dep.), 57:2–12 (testifying that he did not
attempt to quantify the compounds), 69:11–16 (agreeing, however, that
sitagliptin is one of “at least 1,000 compounds” in WO ’498). Dr. Matzger
testifies persuasively that, even if it were appropriate to limit the genus to
the 33 example compounds (not the thousands or millions within Formula I),
the potential combinations with the 8 preferred acids (rather than the open-
ended list of twenty-six acids), and then accounting for varied stoichiometry,
“there would theoretically be approximately 957 salts” that might exist.
Ex. 2103 ¶ 97 (describing this as a “conservative[] estimate”). Again, this
evidence is uncontested. Whether the disclosure cited by Petitioner is
characterized as one list or two, we find that there is no express listing of all
IPR2020-00040
Patent 7,326,708 B2
30
these 950-plus hypothetical salts, much less the specific 1:1 sitagliptin DHP
as claimed, sufficient to anticipate claims 1 and 2.
With no express disclosure of all limitations of the 1:1 sitagliptin DHP
salt in WO ’498, Petitioner cannot fill gaps by arguing a POSA would
“envisage” what is missing.19 See Nidec, 851 F.3d 1270, 1274–75; accord
Galderma Labs, L.P. v. Teva Pharm. USA, Inc., 799 F. App’x 838, 845 (Fed.
Cir. 2020) (holding “the district court was not permitted to ‘fill in missing
limitations simply because a skilled artisan would immediately envision
them.’”). On that point, at the oral hearing, Petitioner sought to pivot the
inquiry from envisaging to enablement—whether a 1:1 sitagliptin DHP salt
is enabled by WO ’498. Tr. 13:24–15:3. But, as we pointed out then,
Petitioner’s argument conflates anticipation and enablement. Id. at Tr. 15:7–
16:9. “Whether a prior art reference is enabled is a separate question from
whether it discloses, expressly or inherently, the claimed limitations at
issue.” See Galderma, 799 F. App’x at 844 (explaining that “Teva makes
the impermissible leap from enablement to disclosure” and “[w]hat a POSA
‘envisages,’ . . . is undoubtedly a question of disclosure, not enablement”).20

19 This gap-filling is precisely what Petitioner attempts to do. See, e.g.,
Tr. 12:9–11 (Petitioner, after discussing the two alleged lists, arguing “[s]o
the only thing we have to show is whether or not a POSA would envision
the 1:1 stoichiometry of the 1:1 DHP stoichiometry and whether it would be
able to make it.”).
20 Although Petitioner spent little effort developing argument around
Merck’s listing of the ’871 patent in the Orange Book for Januvia in its
papers, Petitioner brought up the Orange Book over ten times at oral
hearing. See, e.g., Tr. 14:1–14; see Pet. 5, 18 n.8, 66 (Orange Book
IPR2020-00040
Patent 7,326,708 B2
31
Separately, even if the law did not foreclose Petitioner’s “envisaging”
theory, Petitioner’s evidence on it is weak. Dr. Chorghade does not opine
that a POSA would “at once envisage” the 1:1 sitagliptin DHP salt. See
generally Ex. 1002. His initial testimony was, instead, that a 1:1 ratio results
“every time” sitagliptin and phosphoric acid are reacted (i.e., inherency).
Ex. 1002 ¶ 76; Pet. 18 n.8. Now, Dr. Chorghade states simply that he
“agree[s] with Dr. Matzger that looking at Example 7, when replacing
phosphoric acid for HCL, a POSA ‘can imagine that the [1:1 phosphoric
acid salt] would exist.’” Ex. 1035 ¶ 9 (quoting Ex. 1025 (Matzger dep.),
147:3–4) (brackets added by Dr. Chorghade); Reply 3 (same). But, a closer
look at Dr. Matzger’s testimony, in its more complete context, tells a

(Ex. 1009) mentioned once in an “Overview” section, once in a footnote,
and one other time in an effort to preempt a possible nexus showing because
the Orange Book lists multiple patents); Reply (no mentions). We warned
the parties that oral hearing is not the place for new argument. Paper 80, 3
(citing Dell Inc. v. Acceleron, LLC, 884 F.3d 1364, 1369 (Fed. Cir. 2018)).
Nevertheless, as best understood, Petitioner sees the listing of the ’871
patent in the Orange Book as evidence of or a concession by Merck that the
’871 patent (and equally WO ’498) claims, describes, and enables the 1:1
sitagliptin DHP salt. There appears to be no dispute by these parties that the
salt would be encompassed by broad claims of those references. And while
there is some tension between Merck’s Orange Book listing and certain
arguments made by Merck here (see Tr. 48:9–50:9), for reasons explained in
this Decision, we find on this record that there is no express or inherent
anticipating disclosure of the 1:1 sitagliptin DHP salt in WO ’498 (or the
’871 patent). Whether Merck’s listing of the ’871 patent in the Orange Book
was appropriate, and whether claims in the reference are so broad that they
fail the written description or enablement requirements of § 112, are not
issues for us to resolve in this proceeding.
IPR2020-00040
Patent 7,326,708 B2
32
different story—one that does not persuade us a POSA would have at once
envisaged the claimed salt:
[Q] [W]hen you read the phrase or interpret the phrase ‘at
once envisage,’ how did you interpret that in connection with
your analysis?
A. That you can clearly see all the possibilities based on
the disclosure.
Q. So, in example 7, where it disclosed a 1:1 HCl salt, can
you not clearly see the 1:1 phosphoric acid also being there as a
result of the HCl salt?
A. So by looking at that, you can imagine that it would
exist but that’s different than believing it would exist and that it
could be produced. So an organic chemist looking at that could
exchange the counterion in their mind, but that’s different than
understanding what could be produced, what would be possible
to make.
Q. So you don’t dispute that it actually comes to your
mind, the 1:1 phosphoric acid salt based on example 7.
A. Looking at the general disclosure, no. It does not come
to your mind.
Ex. 1025, 146:13–147:17 (objections and counsel names omitted). The
testimony is, thus, more nuanced than Petitioner’s and Dr. Chorghade’s
limited quotation would suggest. Dr. Matzger does not testify that a POSA
would immediately envisage all the possibilities. Id. Dr. Matzger further
clarified that the claimed 1:1 sitagliptin DHP “does not” come to mind based
on WO ’498. Id. And he distinguished a POSA hypothetically imagining
something versus understanding from the disclosure that a particular salt
would exist or could be made. Id. Paired with Dr. Matzger’s unequivocal
and repeated testimony elsewhere (see, e.g., Ex. 2013 ¶¶ 91–98) that a
POSA would not and could not at once envisage all the possible salts, nor
IPR2020-00040
Patent 7,326,708 B2
33
the claimed 1:1 sitagliptin DHP salt, the isolated “imagine” testimony does
not carry the weight Petitioner places on it.
Petitioner’s envisaging theory fails for yet another reason. Petitioner
does not show that a POSA would “at once envisage each member of th[e]
limited class” of phosphate salts allegedly disclosed in WO ’498. See Eli
Lilly, 471 F.3d at 1376. Petitioner’s expert did not even “at once envisage”
each member of the sub-class of different sitagliptin phosphate salts that can
be formed (accepting, for sake of argument, the position that combining the
identified non-toxic acids with sitagliptin in WO ’498 would constitute a
disclosure of all pharmaceutically acceptable salts of sitagliptin). Quite the
opposite, Dr. Chorghade testified that sitagliptin forms the 1:1 salt with
phosphoric acid “every time.” Ex. 1002 ¶ 76. The evidence, however,
shows that non-1:1 phosphate salts of sitagliptin, such as 1:2, 2:1, and 3:2
salts, do exist and can be made by conventional techniques—with none of
those techniques exempted from WO ’498’s broad disclosure. See, e.g.,
Ex. 2103 ¶¶ 122–135, 148–160. In short, the envisaging theory is also
undermined by Petitioner’s own (initial) position, and its own expert’s
testimony, that the 1:1 salt is the only possible sitagliptin phosphate salt.
In reaching our decision, we need not embrace the distinction about
chemical reactions that Patent Owner asks us to draw from the district
court’s decision Shire. PO Resp. 10–11. From a lay- or less-technical
perspective, combining ingredients in chewing gum (as in Wrigley) may be
easier to envisage compared to envisaging compounds that may (or may not)
form through chemical reactions. Yet the envisaging inquiry and what
would be understood and inferred from the prior art do not turn on a lay-
IPR2020-00040
Patent 7,326,708 B2
34
perspective; they are from the POSA’s vantage point. Acoustic Tech., Inc. v.
Itron Networked Solutions, Inc., 949 F.3d 1366, 1373 (Fed. Cir. 2020)
(explaining that “[e]xpert testimony may shed light on what a skilled artisan
would reasonably understand or infer from a prior art reference”).
Where the level of technical skill is reasonably high, as here, a POSA
might well “envisage” a complex product notwithstanding that chemical
reactions are required to make it. To the extent Patent Owner suggests
“envisaging” is inapplicable if chemical reactions are involved, we decline
to adopt such a rule. On the other hand, the art’s predictability, at least in
some way, colors the lens through which the POSA reads and understands
the art.21 And here, the evidence is undisputed that making salts like those
disclosed in the ’708 patent and prior art is an unpredictable endeavor. See,
e.g., Ex. 2103 ¶¶ 163–71; Ex. 2042, 5 (describing, as of 2002 in the
“Handbook of Pharmaceutical Salts,” that “[n]o predictive procedure . . . has
been reported in the literature”); Ex. 2051, 116:22–117:3. As best we can, in
discerning whether WO ’498 discloses the claimed 1:1 sitagliptin DHP salt,
this Decision considers the art through the ordinarily skilled person’s eyes
based on the relevant evidence without either improperly inflating or
diminishing their abilities.22

21 Cf. In re Baxter Travenol Labs., 952 F.2d 388, 390 (Fed. Cir. 1991)
(holding “extrinsic evidence may be . . . used to explain, but not expand, the
meaning of a reference”).
22 That one of the inventors allegedly “made easily” the claimed salt, as
asserted by Petitioner (Reply 5), does not mean that the reaction product was
known or predictable. Whether a certain salt-screening protocol might be
IPR2020-00040
Patent 7,326,708 B2
35
For the above reasons, we find that there is no express disclosure of
the claimed 1:1 sitagliptin DHP salt in WO ’498 sufficient to anticipate
claims 1 or 2. The other challenged claims (claims 3, 17, 19, and 21–23)
depend from claims 1 or 2, so our finding applies to those claims as well.
(ii) Inherency
“An element may be inherently disclosed only if it is necessarily
present, not merely probably or possibly present, in the prior art.”
Guangdong Alison Hi-Tech Co. v. Int’l. Trade Comm’n., 936 F.3d 1353,
1364 (Fed. Cir. 2019). “[A] limitation or the entire invention is inherent and
in the public domain if it is the natural result flowing from the explicit
disclosure of the prior art.’” Schering Corp. v. Geneva Pharm., Inc., 339
F.3d 1373, 1379 (Fed. Cir. 2003) (internal quotation and citation omitted).
Petitioner did not, in the Petition, provide any experimental results or
data to show that sitagliptin can only be mono-protonated and will form the
1:1 DHP salt every time when reacted with phosphoric acid. See Ex. 2051
(Chorghade dep.) 168:11–169:14 (agreeing the declaration (Ex. 1002)
provides no literature or data supporting his opinion that sitagliptin can only
be mono-protonated). Neither did Petitioner’s expert, even throughout trial,
conduct testing to confirm that his earlier opinion is correct.
What Petitioner and Dr. Chorghade do rely on now, however, is
testing performed by Dr. Chyall many years ago. Reply 1, 5–8. According
to Petitioner, Dr. Chyall’s testing (described in Ex. 2225) involved “12

easy to run is different from predictability of the salts that may (or may not)
form upon running it; and, as discussed, the preponderance of the evidence
is that screening for actual salts is an unpredictable, trial-and-error process.
IPR2020-00040
Patent 7,326,708 B2
36
methanol-based experiments adhering to the WO498’s Example 7” and
“each experiment resulted in the 1:1 sitagliptin DHP salt.” Reply 1 (citing
Ex. 2225 ¶¶ 24, 48, 52). Petitioner highlights, in particular, one experiment
in which Dr. Chyall reacted sitagliptin and phosphoric acid (in a 1-to-5.01
molar ratio) in methanol at ambient temperature, and made 1:1 sitagliptin
DHP. Id. (citing Ex. 2225 ¶¶ 24, 27) (quoting portion of a table related to
Sample ID “234584” and “Notebook No. 4063-19-01”).23
Patent Owner, as an initial matter, contends that this is an untimely
new anticipation argument from Petitioner, which the Board should not
consider. Sur-reply 6 (citing 37 C.F.R. § 42.23(b)). That is, the notion that
a POSA would have adhered to WO ’498’s Example 7 to make a phosphate
salt is not in the Petition and, thus, is a new inherency theory. Id. In any
event, Patent Owner contends the theory is wrong as Dr. Chyall was not
trying to reproduce WO ’498 or any example from it, nor did he adhere to
Example 7 because there are at least eight changes between Dr. Chyall’s
testing and Example 7—not the least of which is that Example 7 uses
hydrochloric acid, not phosphoric acid. Id. at 6–8 (listing eight differences)
(“WO498 does not disclose any of these modifications to Example 7, so it
cannot support inherency as a matter of law”).

23 Exhibit 2225 includes a table (Table 1) listing twelve salt formation
experiments, and indicates three variables were modified between
experiments: API (i.e., sitagliptin)/H3PO4 molar ratios for the starting
materials (e.g., 3:1, 1:5.01, etc.); Reaction solvent (methanol or
methanol/water); and Temperature (ambient, 0°C, 65°C). Ex. 2225 ¶ 24.
IPR2020-00040
Patent 7,326,708 B2
37
As discussed below, whether Petitioner’s reliance on Dr. Chyall’s
testing and a modified Example 7 is new or not, on the merits, Petitioner has
not proved inherency by a preponderance of the evidence on this record.
Example 7 of WO ’498 is explicitly a process for preparing a
hydrochloride salt. Ex. 1004, 47. One could run that process 10,000 times
and it would never produce any phosphate salt of sitagliptin. Ex. 2283
(Chorghade dep.), 52:17–22 (following Example 7 without modification will
only produce the hydrochloride salt). Dr. Chyall did not, thus, “adher[e]” to
Example 7 because that example must necessarily be changed to produce
any phosphate salt. Reply 1. And, there is no process explicitly described in
WO ’498 for making phosphate salts of sitagliptin (or any compound).
Ex. 2283, 50:18–51:6; Eli Lilly & Co. v. Barr Labs., Inc., 251 F.3d 955, 970
(Fed. Cir. 2001) (explaining that inherency must “flow[] from the
reference’s explicitly explicated limitations”); Schering, 339 F.3d at 1379
(same). Nor are we directed to any explicit disclosure in WO ’498 that
instructs the POSA to apply the process used with hydrochloric acid (e.g., as
in Example 7) with the numerous other preferred or non-preferred acids
identified. It may be a possible or even likely option that a POSA could
change Example 7 to accommodate a use of phosphoric acid, but inherency
does not turn on probabilities. “The inherent result must inevitably result
from the disclosed steps; ‘[i]nherency . . . may not be established by
probabilities or possibilities.’” In re Montgomery, 677 F.3d 1375, 1379–80
(Fed. Cir. 2012) (citations omitted, alterations in original).
Petitioner also admits that Dr. Chyall’s testing involved more than
merely swapping hydrochloric acid for phosphoric acid in the Example 7
IPR2020-00040
Patent 7,326,708 B2
38
process24 (even assuming Dr. Chyall was trying to follow Example 7 of
WO ’498, and there is no evidence that he was). Patent Owner shows that
there were at least half-a-dozen other changes from Example 7, including
using “dropwise” acid addition, different reactant concentrations, different
reaction times, and a “slurry crystallization.” Sur-reply 8 (citing e.g.,
Ex. 2283, 43:12–45:10, 48:15–49:7). That Dr. Chorghade offers possible
explanations for the changes or opines that they should make little difference
in the end result (Ex. 1035 ¶¶ 26–42) does not alter the fact that what
Dr. Chyall did was not the same process described in Example 7 (even
beyond the change from one acid to another).
We agree that Dr. Chyall’s testing is, in some respects, closer to
WO ’498’s Example 7 than some of the testing conditions presented by
Patent Owner. Similar to Example 7, Dr. Chyall used methanol as the
solvent and ambient temperature in some of the cited experiments. Compare
Ex. 1004, 47, with Ex. 2225 ¶ 24. In contrast, Dr. Matzger used, for
example, an isopropanol solvent system and heat (70°C). Ex. 1025, 80:1–9,
92:2–6. But, when anticipation is the issue, close is not enough. Connell v.
Sears, Roebuck & Co., 722 F.2d 1542, 1548 (Fed. Cir. 1983) (explaining
that “[a] prior art disclosure that ‘almost’” discloses all the elements
arranged exactly as in the claim, “may render the claim invalid under § 103,
[but] it does not ‘anticipate’”) (internal citation omitted). Anticipation is
still a doctrine based on “strict identity.” Trintec Indus., Inc. v. Top-U.S.A.

24 See Tr. 19:6–20 (Q: “It wasn’t just swapping out HCL for phosphoric
acid, there were changes in the protocol [of Dr. Chyall] even from what’s
explicitly described in Example 7.” Petitioner’s Counsel: “Of course.”).
IPR2020-00040
Patent 7,326,708 B2
39
Corp., 295 F.3d 1292, 1296 (Fed. Cir. 2002). We conclude Petitioner’s
theory, inasmuch as it relies on a protocol with several admitted changes
from what is actually described in an effort to show inherent anticipation,
even if such protocol resembles WO ’498’s Example 7 in some ways, strays
into the realm of obviousness.
For the above reasons, we agree with Patent Owner that Dr. Chyall’s
experiments are no more the actual, asserted “prior art” than Dr. Matzger’s.
Sur-reply 6. So, while Dr. Chyall’s approach in Exhibit 2225 might be one
option a POSA might take if screening for potential phosphate salts of
sitagliptin, there are other undisputedly reasonable and conventional
approaches a POSA could pursue—including the experimental protocols run
by Dr. Matzger. See, e.g., Ex. 2103 ¶¶ 123–176. Such protocols include
screening salts in an isopropanol/water solvent, and Dr. Matzger’s results
show production of 2:1 and 3:2 phosphate salts of sitagliptin in that solvent
system. See, e.g., id. ¶¶ 123–135 and 148–160 (discussing results on 3:2
salt), 136–138 and 161–178 (discussing results on 2:1 salt). Petitioner does
not challenge that Dr. Matzger made these non-1:1 salts. Dr. Matzger also
explains persuasively that although his experiments show the creation of
several non-1:1 salts, “there is no evidence of any 1:1 salt” in the reaction
product he created as evidenced by the absence of certain peaks in X-ray
powder diffraction measurements he observed. Ex. 2103 ¶ 167.
Petitioner counters that the non-1:1 sitagliptin phosphate salts cited
(and made) by Patent Owner are reported in non-prior art evidence. Reply 3
(citing “WO420” (i.e., Ex. 2220)). That argument is, however, misplaced.
Non-prior art evidence can be used to show that certain subject matter is or
IPR2020-00040
Patent 7,326,708 B2
40
is not inherent. Hospira, Inc. v. Fresenius Kabi USA, LLC, 946 F.3d 1322,
1329–30 (Fed. Cir. 2020) (agreeing a court can use extrinsic, non-prior art
evidence in the inherency analysis); see Monsanto Tech. LLC v. E.I. DuPont
de Nemours & Co., 878 F.3d 1336, 1345 (Fed. Cir. 2018) (explaining that
evidence on inherency “need not antedate the critical date of the patent”).
As for Petitioner’s contention (Reply 3) that Dr. Matzger used
different conditions than WO ’498’s Example 7, so did Dr. Chyall in the
testing that Petitioner relies upon, as explained above. And Petitioner does
not seriously contest that Dr. Matzger’s salt-screening conditions were
anything but conventional. Tr. 74:6–19 (Petitioner’s counsel agreeing that
an isopropanol/water salt screen like Dr. Matzger used was a conventional
process). Indeed, the preponderance of the evidence shows just that. See,
e.g., Ex. 2103 ¶¶ 125, 128 (testifying that Drs. Atwood, Chyall, and Matzger
all made non-1:1 phosphate salts of sitagliptin “using conventional
methods”); Ex. 1025 (Matzger dep.), 219:1–7 (testifying a POSA would run
a salt screen, varying solvents, stoichiometry of the reagents, and
temperature); Ex. 2283 (Chorghade dep.), 30:12–31:5 (testifying a POSA
would have run a salt screen and known that they could have used
isopropanol, ethanol, or methanol as solvents). Even for the methanol
solvent that Petitioner treats as sacrosanct and not to be changed, Patent
Owner cites evidence that non-1:1 sitagliptin salts also form in methanol.
Sur-reply 10 (citing Ex. 2220, 32 (Example 2, showing a 2:1 sitagliptin
phosphate salt in a methanol system)); Ex. 2283 (Chorghade dep), 87:8–22
(testifying he has “no opinions” on Exhibit 2220 (WO ’420)); Ex. 1025,
IPR2020-00040
Patent 7,326,708 B2
41
221:21–228:5 (testifying about Exhibits 2273 and 2274 and non-1:1
sitagliptin sulfates forming in methanol).
In sum, the evidence—confirmed experimentally and reported in the
technical literature—undeniably shows that non-1:1 sitagliptin phosphate
salts do exist and that the claimed 1:1 sitagliptin DHP does not form “every
time” sitagliptin and phosphoric acid are reacted. Petitioner’s criticisms that
such evidence applies conditions that do not follow Example 7 of WO ’498
as closely as the conditions of Petitioner’s relied-upon testing are misplaced
because that testing too is changed from what WO ’498 explicitly describes.
We also find that the experimental results cited by Patent Owner are not
mere aberrations based on irrelevant conditions. To the contrary, the
evidence shows that the conditions under which non-1:1 salts form are
conventional salt-screening techniques that a POSA, acting reasonably,
could have pursued. We, therefore, find Petitioner’s inherent anticipation
theory unpersuasive.
Petitioner has not proved that WO ’498 expressly or inherently
discloses the subject matter of claims 1 and 2. Thus, we find on this record
that claims 1 and 2 are not anticipated by WO ’498. The other challenged
claims (claims 3, 17, 19, and 21–23) depend from claims 1 or 2, so our
finding applies to those claims as well.
Anticipation by the ’871 patent
Petitioner contends that claims 1–3, 17, 19, and 21–23 are anticipated
by the ’871 patent. Pet. 31–38. According to Petitioner, the ’871 patent is
prior art under 35 U.S.C. § 102(e)(2). Id. at 31–32.
IPR2020-00040
Patent 7,326,708 B2
42
Petitioner states that “[t]he specifications of the ’871 patent and
WO ’498 are identical in all relevant material respects.”25 Id. at 33. Indeed,
Petitioner’s challenge based on the ’871 patent parallels the anticipation
challenge based on WO ’498. Id. at 33–38. Petitioner agreed that, like the
parties, the Board should treat WO ’498 and the ’871 patent “for all intents
and purposes as having the same disclosures.” Tr. 7:8–13; Reply 1 n.2
(noting that Petitioner’s “Reply discusses WO498 and the ’871 patent
together as WO498”).
Petitioner’s anticipation challenge based on the ’871 patent fails for
the same reasons as addressed above for WO ’498. See supra Section II(D).
We, thus, similarly find that Petitioner has not proved by a preponderance of
the evidence that the ’871 patent anticipates claims 1–3, 17, 19, and 21–23.
Obviousness
Petitioner asserts that claims 1–3, 17, 19, and 21–23 would have been
obvious over WO ’498, or WO ’498 in combination with Bastin. Pet. 39–46
(Ground 3),26 46–59 (Ground 4). Petitioner asserts that claim 4 would have

25 WO ’498 and the ’871 patent each include an identical claim 15 (depicting
33 compounds); the ’871 patent also includes one claim that depicts only
sitagliptin and further recites “or a pharmaceutically acceptable salt thereof.”
Ex. 1007, 36:50–40:48 (claim 15), 41:1–14 (claim 17).
26 The Petition does not identify claims 1 and 2 as being allegedly obvious
over WO ’498 alone. See Pet. 38–39 (starting obviousness analysis with
claim 3). That allegation is implicit in Petitioner’s allegation that claims
depending from claims 1 and 2 would have been obvious. To satisfy
limitations of claims 1 and 2 appearing in the dependent claims, however,
Petitioner’s WO ’498 obviousness challenge relies on Petitioner’s
IPR2020-00040
Patent 7,326,708 B2
43
been obvious over the combination of WO ’498, Bastin, and Brittain (id. at
59–62 (Ground 5)), or just WO ’498 and Brittain (id. at 62–63 (Ground 6)).
1. Antedation and Application of § 103(c)(1) Exception
All of Petitioner’s obviousness grounds rely on WO ’498. See
generally Pet. 38–62. Before reaching the merits of Petitioner’s challenge,
we address a threshold issue. Patent Owner argues that it reduced to practice
the subject matter of claims 1, 2, 17, 19, and 21–23 before WO ’498
published. PO Resp. 22–28; Sur-reply 21–22. If Patent Owner prevails on
that issue, WO ’498 is not § 102(a) prior art, and it is merely a § 102(e)
reference. Because it is, however, undisputed that the inventions claimed in
the ’708 patent and the subject matter of WO ’498 were commonly owned
by Merck, or under obligation of assignment to Merck, at the time of
invention, the exception under pre-AIA § 103(c)(1) would apply. 35 U.S.C.
§ 103(c)(1). Therefore, if we agree with Patent Owner on the antedation
issue, WO ’498 cannot be used to show that claims 1, 2, 17, 19, and 21–23
would have been obvious. That would then leave only claims 3 and 4
subject to Petitioner’s obviousness challenge.
In an inter partes review, the burden of persuasion is on petitioner to
prove unpatentability by a preponderance of the evidence, and that burden
never shifts to patent owner. Dynamic Drinkware, LLC v. Nat’l Graphics,
Inc., 800 F.3d 1375, 1378 (Fed. Cir. 2015). The petitioner has the initial
burden of production to show that an asserted reference qualifies as prior art

“predicate” anticipation arguments. Pet. 39–40 (incorporating by reference
the reasons advanced for Ground 1 (anticipation by WO ’498)).
IPR2020-00040
Patent 7,326,708 B2
44
under 35 U.S.C. § 102. Id. at 1378–79. Once a petitioner meets the initial
burden, a burden of production shifts to the patent owner to argue or produce
evidence that the asserted reference does not render the claims unpatentable,
or that the reference is not prior art (e.g., because patent owner actually
reduced to practice the invention before the applicable date of the asserted
art). Id. at 1379–80 (citing Tech. Licensing Corp. v. Videotek, Inc., 545 F.3d
1316, 1327 (Fed. Cir. 2008)).
Patent Owner argues that it actually reduced to practice the subject
matter of claims 1, 2, 17, 19, and 21–23 before WO ’498’s January 16, 2003,
publication date. PO Resp. 22–23; In re Steed, 802 F.3d 1311, 1318 (Fed.
Cir. 2018) (explaining that actual reduction to practice requires an inventor
have “constructed an embodiment or performed a process that met all the
limitations” of the invention, and “determined that the invention would work
for its intended purpose”). According to Patent Owner, the testimony of
inventors Vicky Vydra and Dr. Karl Hansen, and from witness Dr. Rebecca
Leigh Schultz, which is further corroborated by documentary evidence,
shows the creation of a 1:1 sitagliptin DHP salt in December 2001, with
experimental confirmation of the salt’s form and stoichiometry documented
in early 2002. Id. at 23–25 (citing, for example, Ex. 2002 ¶¶ 9–20; Ex. 2103
¶¶ 46–53; Ex. 2127 ¶¶ 11–14, 16–22 (testifying, inter alia, on the confirmed
stoichiometry using HPLC); Ex. 2140 ¶¶ 30–31; Ex. 2135 (delivery sheet for
1:1 sitagliptin DHP, indicating a testing date of Mar. 25, 2002, and sent date
of Apr. 22, 2002)). This evidence, Patent Owner contends, shows successful
reduction to practice of the subject matter of claims 1 and 2. Id.; see also id.
at 25–26 (citing additional and persuasive testimony and documentary
IPR2020-00040
Patent 7,326,708 B2
45
evidence (unrebutted by Petitioner) supporting reduction to practice of the
subject matter of claims 17, 19, and 21–23).
This proof of reduction to practice, Patent Owner argues, “means that
WO ’498 qualifies as prior art only under §102(e).” Id. at 27. And, citing
the exception under pre-AIA § 103(c)(1) and documents showing Patent
Owner’s common ownership of the claimed subject matter and WO ’498,
Patent Owner argues WO ’498 is eliminated as a § 102(e) reference from the
obviousness inquiry for claims 1, 2, 17, 19, and 21–23. Id. at 27–28 (citing
agreements requiring assignment from the inventors to Merck, such as in
Ex. 2005 (Diddle Decl.), 7–8 (Edmondson agreement), 28–29 (Cypes
agreement), 44–45 (Vydra agreement)).27
Petitioner does not dispute that Merck reduced 1:1 sitagliptin DHP
salts to practice before WO ’498 published, nor that Merck commonly
owned the relevant subject matter. Petitioner, instead, raises a relatively
discrete counterargument. Reply 15–18.28 According to Petitioner, absent
Merck proving an earlier reduction to practice of hydrates of the 1:1

27 Patent Owner argues these employment “agreements required the
inventors to assign the claimed inventions to Merck, which they in fact did.”
PO Resp. 27 (citing supporting evidence including Exs. 2009 and 2010
(certified assignments)).
28 Before institution, Petitioner argued that the ’708 patent “plainly admits”
that WO ’498 is prior art by listing it in the “Background of the Invention.”
Pet. 2. We explained previously why we rejected that position and noted, in
any event, that it was not controlling on whether the § 103(c)(1) exception
applies. Inst. Dec. 41. Petitioner did not contest our explanation or raise
that issue again during trial and, thus, Petitioner appears to have abandoned
argument based on the alleged “admitted” prior art status of WO ’498. To
the extent not abandoned, we adopt and incorporate our prior analysis. Id.
IPR2020-00040
Patent 7,326,708 B2
46
sitagliptin DHP salt, Merck’s antedation argument fails. Id. at 17–18.
Petitioner alleges the following: (i) the ’708 patent’s claims do not exclude
hydrates, (ii) WO ’498 discloses, as a general comment, that “[s]alts in the
solid form may exist in more than one crystal structure, and may also be in
the form of hydrates,” and (iii) although Merck previously made multiple
anhydrous forms of 1:1 sitagliptin DHP, it did not make the crystalline
monohydrate of 1:1 sitagliptin DHP until March 26, 2003—about two
months after WO ’498’s publication. Id. at 16–18 (citing Ex. 1004, 10:32–
34; Ex. 2124 ¶¶ 9 (testifying that, as of January 2003, “no hydrates of the 1:1
DHP salt” had been identified), 14 (describing crystalline monohydrate
isolation on March 26, 2003)).
In more plain terms, Petitioner argues that WO ’498 shows more of
the claimed invention than Merck had actually reduced to practice at the
relevant time. Reply 15–18. Hence, Petitioner argues, we should reject
Merck’s antedation argument because Merck has not proved “priority with
respect to so much of the claimed invention as the [prior art] happens to
show.” Id. at 15–16 (quoting In re Clarke, 356 F.2d 987, 991 (CCPA 1966)
(brackets added by Petitioner)).29
It is true, on this record, that Merck had made no hydrates of 1:1
sitagliptin DHP as of the January 16, 2003, publication date of WO ’498.

29 See also In re Mantell, 454 F.2d 1398, 1401 (CCPA 1972) (affirming the
Board’s decision that antedation evidence was insufficient for certain
generic claims where “affidavit evidence shows reduction to practice of
much less than the reference shows”).
IPR2020-00040
Patent 7,326,708 B2
47
Ex. 2124 ¶ 9. We also agree with Petitioner’s position that the challenged
claims do not exclude hydrates. In fact, claim 1 includes the clause “or a
hydrate thereof.”30 Ex. 1001, 16:14–15. But, in resolving the antedation
issue here, we are guided by Clarke, which explains that “it is not the entire
scope of the claims which is determinative . . . . Rather, it is how much the
reference shows of the claimed invention that is crucial to the requirement.”
In re Clarke, 365 F.2d at 991 (emphasis added). The dispositive question is,
thus, whether the evidence shows Merck’s reduction to practice of as much
of the claimed invention as is shown in WO ’498. Id.
As discussed above, WO ’498 does not expressly show any phosphate
salts of any disclosed compound, much less a phosphate salt of any of the
thirty-three example compounds, a phosphate salt of sitagliptin, or the
particular 1:1 sitagliptin DHP of the challenged claims. Supra Section II(D).
WO ’498 discloses sitagliptin among many compounds, and salts thereof.
Ex. 1004, 55–60. WO ’498 actually describes a 1:1 hydrochloride salt of
sitagliptin—notably, this salt is not a hydrate. Id. at 47. And, elsewhere,
WO ’498 lists several potential salt-forming acids, one of which is
phosphoric acid among a sub-list of eight preferred acids. Id. at 11:8–15.
We find that Merck’s undisputed, detailed, and repeated actual reduction to
practice of the claimed 1:1 (R)-sitagliptin DHP salts shows more, not less,

30 Claim 2, which is directed to the (R)-configuration of 1:1 sitagliptin DHP
does not include the clause “or a hydrate thereof.” Ex. 1001, 16:16–30.
Claim 4, however, depends from claim 2 and expressly recites “a crystalline
monohydrate” of the 1:1 (R)-sitagliptin DHP—claim 2 must, therefore, also
encompass hydrates. Id. at 16:47–49.
IPR2020-00040
Patent 7,326,708 B2
48
than what is shown in WO ’498 for the claimed subject matter. PO Resp.
23–26; Ex. 2103 ¶¶ 46–62; Ex. 2002 ¶¶ 9–20; Ex. 2103 ¶¶ 46–53; Ex. 2127
¶¶ 11–14, 16–22; Ex. 2140 ¶¶ 30–31.
Even assuming WO ’498’s disclosure were sufficient to show the
claimed 1:1 sitagliptin DHP, there is no hydrate of that salt shown anywhere
in the reference. Indeed, as Petitioner’s declarant concedes, WO ’498
provides no polymorphic data for sitagliptin and none of the WO ’498
compounds are disclosed as being synthesized as hydrates. Ex. 2051,
197:22–198:9. Unless WO ’498 shows a 1:1 sitagliptin DHP hydrate,
Merck need not have reduced such a hydrate to practice to antedate
WO ’498’s publication date. In re Clarke, 365 F.2d at 991.
Petitioner emphasizes the sole mention of “hydrates” in WO ’498,
which reads, in full, “[s]alts in the solid form may exist in more than one
crystal structure, and may also be in the form of hydrates.” Ex. 1004,
10:32–34; Reply 16–17. From this, Dr. Chorghade opines that the POSA
“would have expected that (R)-sitagliptin phosphate of Claim 2 exists as a
crystalline hydrate.” Ex. 1002 ¶ 169. That conclusory opinion, however,
does not hold up to scrutiny.
First, none of the salts described in WO ’498—for sitagliptin or
otherwise—are hydrates, as Dr. Chorghade admits. Ex. 2051, 197:22–198:9.
Second, there are at least thousands, likely millions, of salts embraced by
hypothetical combinations of DP-IV compounds and potential counterions in
WO ’498. Ex. 2103 ¶¶ 73, 77. The solitary “hydrates” sentence is not
specific to sitagliptin (or any particular compound), and we credit
Dr. Myerson’s testimony that a POSA would not have understood the
IPR2020-00040
Patent 7,326,708 B2
49
sentence as applying to sitagliptin phosphate salts in particular. Ex. 2101
¶¶ 102, 136. Third, the sentence states that crystalline forms (e.g., hydrates)
“may” exist. Ex. 1004, 10:32–34. WO ’498 does not show that any such
hydrates do exist for any disclosed or hypothetical salt (much less a 1:1
sitagliptin DHP), nor is there any guidance in WO ’498 on making them.
Ex. 2101 ¶¶ 93, 124, 150 (expert opinion supporting this interpretation); Sur-
reply 17. And fourth, as discussed in more detail below, the substantial
weight of evidence shows that whether any particular salt is even capable of
forming as a hydrate is highly unpredictable. Indeed, Dr. Chorghade, when
asked if he could “predict with reasonable certainty” whether a hydrate
might exist, answered that he “couldn’t predict with any degree of
certainty.” Ex. 2051, 257:18–258:11 (emphasis added). We, thus, find more
credible the testimony of Merck’s expert, Dr. Myerson, that “the POSA
would not have understood that [one] statement [in WO ’498] as disclosing
that every compound, or every one of the disclosed or exemplified
compounds of WO ’498, can exist as a hydrate.” Ex. 2101 ¶ 150.
Petitioner suggests, in a footnote, that WO ’498 is presumptively
enabling for the hydrated forms of the various salts. Reply 18 n.9. Even if
that presumption were appropriate here, we would find it overcome based on
a complete absence of examples or guidance in WO ’498 about making such
hydrates, and the substantial unpredictability concerning whether a hydrate
of any specific salt will even form, as discussed below.
Patent Owner and Dr. Myerson explain persuasively that a POSA
would not have known or predicted whether particular salts, especially the
1:1 sitagliptin DHP, could crystallize at all, form a hydrate, or form a
IPR2020-00040
Patent 7,326,708 B2
50
monohydrate. Ex. 2101 ¶¶ 58–63 (“The unpredictability of crystalline forms
applies equally in the context of solvates and hydrates.”) (citing supporting
evidence), 146–150, 163–164; see PO Resp. 46–47, 50–53. The literature of
record firmly supports them. For example, in one of Teva’s patents, it notes
that “[t]he existence and physical properties of polymorphs, hydrates and
solvates is unpredictable.” Ex. 2048, 3:3–5. In a publication from 2007
(naming Dr. Chorghade as editor), the authors note that “it is still fair to
presume that no method exists to predict polymorphs with a considerable
degree of certainty,” and that discovery and prediction of such forms is filled
with “trial and error, and serendipities.” Ex. 2045, 215; see also 2047, 10
(observing that it is “not yet possible to predict when materials will
crystallize” and that polymorphs “are all too frequently discovered by
serendipity”); Ex. 2046, 18 (“Predicting the formation of solvates or
hydrates . . . is complex and difficult.”).31 And, on cross examination,
Dr. Chorghade agreed that the art was unpredictable on precisely this point:
“Q. Do you agree that it is unpredictable which set of experimental
conditions will yield a particular crystal form? . . . [A] It is not predictable

31 We agree with Patent Owner that the unpredictability of forming
crystalline polymorphs, including hydrates, has been repeatedly recognized
in other matters by the Board and the courts. See, e.g., Lupin Ltd. v. Janssen
Sciences Ireland UC, IPR2015-01030, Paper 17 at 20 (PTAB Oct. 16, 2015)
(finding “it would have been known that preparing hydrates generally was
unpredictable”); PO Resp. 41–42 (citing cases); Sanofi-Synthelabo v.
Apotex, Inc., 550 F.3d 1075, 1089 (Fed. Cir. 2018) (“The experts of both
parties agreed that whether a pharmaceutically suitable crystalline salt will
form from a particular acid-base combination is unpredictable.”). Our
Decision here is, however, based on the evidentiary record in this case.
IPR2020-00040
Patent 7,326,708 B2
51
with any degree of certainty.” Ex. 2051, 269:11–20; see also id. at 245:21–
246:8 (admitting “[i]t was presumably true in 2003 and earlier” that it
“wasn’t possible to predict when materials will crystallize”).
Consistent with this evidence, Merck also provides additional
testimony and documentation showing that its actual discovery of a hydrate
of 1:1 sitagliptin DHP—a crystalline monohydrate—was unforeseen and
arose only after substantial work with other salt forms. For example, one of
the inventors, Dr. Wenslow, testifies that even after “over a year of
development on the 1:1 DHP salt had taken place, including extensive
experiments with the 1:1 DHP salt in both water and aqueous solvent
mixtures, . . . the DPP-IV project team had not observed a monohydrate until
its surprising and unexpected creation very late in the development cycle.”
Ex. 2116 ¶ 36 (citing Ex. 2119 (project team memorandum), 1–5, 10–11).
Another inventor, Stephen Cypes, testifies on the “unexpected” discovery of
the hydrated form while trying to find more efficient synthetic routes to
directly crystallize a pure anhydrous form, and using an isoamyl
alcohol/water solvent. Ex. 2124 (Cypes Decl.) ¶¶ 9–15; PO Resp. 5–6; see
also Ex. 2101 ¶ 94 (opinion of Dr. Myerson that isoamyl alcohol/water is an
unusual solvent system, that he is not aware of its use being reported in the
art, and that a “POSA would not have arrived at the particular experimental
conditions necessary . . . absent undue experimentation”).32

32 Petitioner comments that, irrespective to any non-enablement of a
monohydrate, claim 1 of the ’708 patent and WO ’498 encompasses
hydrates, not just monohydrates. Reply 18 n.9. But the evidence on
IPR2020-00040
Patent 7,326,708 B2
52
Altogether, we find the preponderance of the evidence shows that
Merck reduced to practice at least as much (indeed, more) of the claimed
subject matter versus what is shown in WO ’498. We disagree that, to
antedate WO ’498’s publication, Merck needed to prove a prior reduction to
practice of the 1:1 sitagliptin DHP hydrate. Accordingly, WO ’498 is only a
§ 102(e) reference. Based on Merck’s undisputed ownership (or ownership
right by assignment obligation) at all relevant times of the claimed subject
matter and WO ’498, that reference is excluded from consideration for
obviousness purposes under pre-AIA § 103(c)(1) for claims 1, 2, 17, 19, and
21–23. Without WO ’498, Petitioner’s § 103 challenge to those claims fails.
That leaves claims 3 and 4, for which Patent Owner does not assert a prior
reduction to practice, subject to the obviousness challenge.

unpredictability cited above is not limited to monohydrates; it extends to
crystalline forms and hydrates generally. Also, as to other hypothetical
hydrates of 1:1 sitagliptin DHP, it is not apparent on this record that any
exist. Merck cites evidence that, despite years of research, the ’708 patent’s
monohydrate is the only such form known. Ex. 2101 ¶ 205 (citing support);
Ex. 2122, 34 (describing “Single crystal form”); Ex. 2124 ¶¶ 9, 15; Sur-reply
25. As to the alleged enablement through use of isopropanol as a solvent
(Reply 24–25), although some evidence indicates a 1:1 sitagliptin DHP
monohydrate can form in isopropanol/water (Ex. 2101 ¶ 33 (citing Ex. 1001,
12:61–13:21)) there is also experimental evidence to the contrary, which
underscores the unpredictable nature of salt formation. See, e.g., Ex. 2103
¶¶ 123–78 (Matzger testing creating 3:2 and 2:1 sitagliptin phosphate salts in
isopropanol/water); Sur-reply 21 (“[I]n isopropanol/water, Dr. Matzger
obtained two crystalline salts outside the scope of any claims.”).
IPR2020-00040
Patent 7,326,708 B2
53
2. Claim 3
Claim 3 recites “[t]he salt of claim 1 of structural formula III having
the (S)-configuration at the chiral center marked with an *” and further
depicts the chemical structure of that (S)-form of the 1:1 DHP salt.
Ex. 1001, 16:31–45. Patent Owner does not argue that it reduced to practice
the subject matter of claim 3 before WO ’498 published. WO ’498 is,
therefore, available as a § 102(a) reference for purposes of analyzing the
alleged obviousness of claim 3.
Petitioner argues claim 3 would have been obvious over WO ’498
alone (Pet. 38–41), or over a combination of WO ’498 and Bastin (Pet. 56).
We gave an overview of WO ’498 above (see Section II(D)(1)). We provide
an overview of Bastin below, then turn to the obviousness analysis.
a) Overview of Bastin (Exhibit 1006)
Bastin is a publication from 2000 related to “[s]election of an
appropriate salt form for a new chemical entity,” giving a chemist “the
opportunity to modify the characteristics of the potential drug substance and
to permit the development of dosage forms with good bioavailability,
stability, manufacturability, and patient compliance.” Ex. 1006, 1 (Abstr.).
According to Bastin, “[w]here possible, a range of salts should be prepared
for each new substance and their properties compared during a suitable
preformulation program.” Id.
Bastin teaches that “the choice of salt is governed largely by the
acidity or basicity of the ionisable group,” but that “safety” and “drug
indications” should also be considered. Id. at 2. According to Bastin, “[t]he
vast majority of salts are developed to enhance aqueous solubility of drug
IPR2020-00040
Patent 7,326,708 B2
54
substances,” and “[f]or weakly basic drugs substances, salts of an inorganic
acid (e.g., hydrochloride, sulphate, or phosphate) . . . could be considered.”
Id. (Table 1 (listing “common pharmaceutical salts” including, among
others, hydrochloride, hydrobromide, sulfate, nitrate, and phosphate)).
Bastin teaches that “[h]ydrochloride salts have often been the first choice for
weakly basic drugs . . . [but] the potential disadvantages of hydrochloride
salts may include unacceptably high acidity in formulations (e.g., parenteral
products), the risk of corrosion, less than optimal solubility . . . and the
potential for poor stability.” Id.
b) Analysis of Alleged Obviousness
The challenge to claim 3 as obvious over WO ’498 alone fails because
it is based on Petitioner’s “predicate” anticipation challenge to claim 1.
Pet. 40–41. For the reasons above (supra Section II(D)(2)), we reject
Petitioner’s challenge to claim 1. Petitioner has, thus, failed to succeed on
its “predicate” showing. Pet. 40–41.
We turn next to the challenge to claim 3 as obvious over WO ’498 and
Bastin. Pet. 56; see also id. at 46–56 (claim 1 discussion); Reply 21–23.
Petitioner contends that, “[s]ince WO ’498 discloses (R)-sitagliptin
and its ‘pharmaceutically acceptable salts,’ the skilled artisan would have
been motivated to optimize the salt.” Pet. 51. Petitioner contends a POSA
would, in making such salts, have used non-toxic acids because sitagliptin is
basic. Id. at 51–52; Ex. 1002 ¶ 147. According to Petitioner, because
WO ’498 and Bastin identify phosphoric acid for use as a possible salt-
forming acid with basic drugs, and considering Bastin’s salt selection and
optimization procedures, the POSA would have formed claim 1’s 1:1
IPR2020-00040
Patent 7,326,708 B2
55
sitagliptin DHP. Pet. 53–54. Also, Petitioner argues, “Bastin would have
given the POSA motivation to make alternatives to the hydrochloride salt
due to its potential problems” and “the combined teachings . . . would
motivate the POSA to use the following list of three salts as alternatives to
the hydrochloride salt exemplified in WO ’498: hydrobromide, sulfate, and
phosphate.” Id. at 53; Ex. 1002 ¶¶ 148–152 (testifying about overlapping
acids). At a minimum, Petitioner contends “it would be obvious to try”
alternatives, and “making the phosphoric acid salt . . . would have been a
matter of routine experimentation.” Pet. 55–56.
For the (S)-enantiomer limitation, Petitioner contends that WO ’498
describes (R)-sitagliptin, but “WO ’498 taught that there are only two
configurations ((R) and (S)) and how to synthesize each.” Pet. 56 (citing
Ex. 1004, 10:3–26, 12:11–14); Ex. 1002 ¶¶ 155–156.
On this record, we find Petitioner’s challenge to claim 3 unpersuasive.
The Petition does not identify where the recited 1:1 ratio is necessarily
satisfied upon the combination of WO ’498 and Bastin. Petitioner’s reliance
on Bastin goes, not to any stoichiometry, but to its disclosure of phosphoric
acid, some desired attributes of pharmaceutical salts like better stability, and
potential disadvantages to using hydrochloric acid. Pet. 53–56; Ex. 1006.
Petitioner, however, cites no specific screening or optimization protocol in
Bastin that, combined with WO ’498, would have led to a 1:1 sitagliptin
DHP, or the 1:1 (S)-sitagliptin DHP of claim 3. Again, the Petition relied on
the opinion of Dr. Chorghade that the only possible salt arising from the
combination of phosphoric acid and (R)-sitagliptin in WO ’498 is one with
IPR2020-00040
Patent 7,326,708 B2
56
the recited 1:1 stoichiometry. Pet. 18 n.8; Ex. 1002 ¶ 76. That is incorrect
as discussed above. Supra Section II(D)(2).
We are also unpersuaded a POSA would have been motivated to make
the 1:1 phosphate salt of (S)-sitagliptin, in particular, and that a POSA
would have reasonably believed that salt form would have solved any
problem with WO ’498’s hydrochloride salt of (R)-sitagliptin. To the
contrary, we agree with Patent Owner that there were numerous reasons why
a POSA would not have made all the choices that would been needed to
arrive at the claimed subject matter. Ex. 2103 (Matzger Decl.) ¶¶ 31–34
(testifying persuasively on the differences between hydrochloric acid and
phosphoric acid and why they are not interchangeable), ¶¶ 214–218 (no
shortcomings with the hydrochloride salt of sitagliptin described), ¶¶ 227–
229 (testifying, citing supporting literature, that phosphates were known to
reduce solubility and stability versus hydrochloride salts); Ex. 2051
(Chorghade dep.) 193:21:–194:2 (admitting nothing in WO ’498 suggests
any problem with the hydrochloride salt of sitagliptin); Sur-reply 23–24.
Notwithstanding the above, assuming Petitioner’s claim 3 challenge is
proposing that it would have been obvious to substitute phosphoric acid for
hydrochloric acid, and to run the process of Example 7 but with all the
changes to it made by Dr. Chyall, and further assuming a POSA would have
made such modifications, that would still not produce the claimed subject
IPR2020-00040
Patent 7,326,708 B2
57
matter.33 That would, at best, make a 1:1 (R)-sitagliptin DHP. Supra
Section II(D)(2); Reply 1 (citing Ex. 2225 ¶¶ 24, 48, 52).
Claim 3 is, however, to the (S)-enantiomer of 1:1 sitagliptin DHP.
See Sanofi-Synthelabo, 550 F.3d at 1086 (“The determination of obviousness
is made with respect to the subject matter as a whole, not separate pieces of
the claim.”). Neither WO ’498 nor Bastin show any (S)-configuration nor
any racemate (mixture of configurations) of any sitagliptin salt. Ex. 2051
(Chorghade dep.) 176:18–177:10 (“Q. The WO ’498 does not identify the
S-enantiomer – A. No, it doesn’t”). And, significantly, Petitioner “does not
explain why the POSA would have been motivated to specifically make the
(S)-configuration.” PO Resp. 38; Ex. 2103 ¶ 255 (“Dr. Chorghade has failed
to explain why the POSA would have been motivated to make the (S)-
configuration.”); see In re Stepan, 868 F.3d 1342, 1345–46 n.1 (Fed. Cir.
2017) (“Whether a rejection is based on combining disclosures from
multiple references, combining multiple embodiments from a single
reference, or selecting from large lists of elements in a single reference,
there must be a motivation to make the combination and a reasonable
expectation that such a combination would be successful, otherwise a skilled
artisan would not arrive at the claimed combination.”).

33 As discussed above, this strains the arguments made for claim 3 in the
Petition, which provided no specific details about how Example 7 should be
changed (e.g., details of the screening protocol, such as in Dr. Chyall’s
testing), nor any experimental proof that a 1:1 stoichiometry arises “every
time.” Ex. 1002 ¶ 76.
IPR2020-00040
Patent 7,326,708 B2
58
Petitioner advances no expected or even theoretical benefit to making
an (S)-enantiomer of 1:1 sitagliptin DHP. Sur-reply 23 (“Mylan fails to
identify any reason why the POSA would have formed a salt of (S)-
sitagliptin.”). And we find WO ’498’s general disclosure on diastereomers,
which again appears in reference that putatively encompasses millions of
potential compounds and salts, does not show a motivation to make, with a
reasonable expectation of success, 1:1 (S)-sitagliptin DHP specifically as
claimed—especially where the evidence (discussed herein) also shows that
forming such salts is highly unpredictable. See, e.g., Ex. 2042, 5 (“No
predictive procedure . . . reported in the literature”); Ex. 2051, 116:22–117:3
(salt formation is an unpredictable “trial and error process”).
For the above reasons, we find that Petitioner has not shown by a
preponderance of the evidence that claim 3 would have been obvious.
3. Claim 4
Claim 4 recites the “salt of claim 2 characterized in being a crystalline
monohydrate.” Ex. 1001, 16:47–48. As noted above, Patent Owner does
not argue that it reduced to practice the subject matter of claim 4 before
WO ’498 published. WO ’498 is, thus, available as a § 102(a) reference for
purposes of analyzing the alleged obviousness of claim 4.
Petitioner asserts that claim 4 would have been obvious over the
combination of WO ’498, Bastin, and Brittain. Pet. 59–62 (Ground 5).
Petitioner also asserts that claim 4 would have been obvious over WO ’498
and Brittain. Pet. 62–63 (Ground 6). WO ’498 and Bastin are discussed
above. We provide an overview of Brittain below, followed by our analysis.
IPR2020-00040
Patent 7,326,708 B2
59
a) Overview of Brittain (Ex. 1005)
Brittain is a chapter from a book titled “Polymorphism in
Pharmaceutical Solids,” published in 1999. Ex. 1005, 1–4; Pet. 59. Brittain
teaches that “it is estimated that approximately one-third of the
pharmaceutical actives are capable of forming crystalline hydrates” and,
based on data, “[t]his shows the expected trend in which monohydrates are
most frequently encountered, and . . . the frequency decreases almost
exponentially as the hydration number increases.” Ex. 1005, 6.
b) Analysis of Alleged Obviousness
Ground 6, obviousness over WO ’498 and Brittain, relies on a
“predicate finding” that claim 2 is anticipated by WO ’498. Pet. 62. We
disagree that claim 2 is anticipated by WO ’498 as explained above. Supra
Section II(D). Ground 6 fails for that reason, but it is also deficient for other
reasons discussed below on Ground 5.
For Ground 5, Petitioner first alleges that “Claim 2’s (R)-sitagliptin
phosphate would have been obvious over WO ’498 and Bastin.” Pet. 61.
Citing WO ’498’s disclosure that “[s]alts in the solid form may exist in more
than one crystal structure, and may also be in the form of hydrates,”
Petitioner contends the POSA “would have expected that (R)-sitagliptin . . .
exists as a crystalline hydrate.” Id. (citing Ex. 1004, 10:32–34). Then,
pointing to Brittain and Dr. Chorghade’s opinion, Petitioner contends that
crystalline monohydrates are “frequently encountered” with pharmaceutical
substances. Id. (citing Ex. 1005, 6 (disclosing “approximately one-third of
the pharmaceutical actives are capable of forming crystalline hydrates” and
an “expected trend in which monohydrates are most frequently
IPR2020-00040
Patent 7,326,708 B2
60
encountered”); Ex. 1002 ¶ 170). Combined with WO ’498’s alleged
teaching “that hydrates exist of the compounds disclosed therein,” Petitioner
argues the 1:1 (R)-sitagliptin DHP crystalline monohydrate of claim 4
“would have been obvious to a POSA.” Id. at 61–62.
Patent Owner responds that Petitioner’s “cursory treatment [of
claim 4] bears almost no resemblance to the required analysis of
obviousness.” PO Resp. 39. According to Patent Owner, courts (and the
Board) consistently uphold as nonobvious claims to specific crystal forms,
including monohydrates, even where the salt is disclosed expressly in the art.
Id.; see also at 41–44 (citing cases recognizing unpredictability with
crystalline forms). In any event, Patent Owner argues that Petitioner must
show a motivation to make the crystalline monohydrate of claim 4 with a
reasonable expectation of success, which Petitioner fails to do. Id. at 39–59.
Indeed, Patent Owner argues, “Mylan fails to articulate any reason why the
POSA would have been motivated to pursue a crystalline monohydrate form
of 1:1 sitagliptin DHP,” and the evidence of record gives reasons that would
have discouraged the POSA from doing so. Id. at 54–59 (citing known
problems with hydrates, such as instability and low solubility; Ex. 1005, 4–
6). Patent Owner reiterates that none of the salts shown in WO ’498 are
hydrates, nor is guidance provided on how to make them. Id. at 40–41.
And, Patent Owner emphasizes, “[c]rystal formation, including the structure,
properties, and way to make any hydrate forms was (and is) unpredictable.”
Id. at 41 (citing, e.g., Ex. 2101 ¶¶ 50–64, 88–91; Ex. 2047, 10). Given the
state of the art, and the recognized unpredictability of crystalline salt
formation, Patent Owner argues the POSA would not have reasonably
IPR2020-00040
Patent 7,326,708 B2
61
expected to obtain the crystalline monohydrate of 1:1 sitagliptin DHP. Id. at
44–53 (citing, e.g., testimony from Dr. Chorghade that a POSA “couldn’t
predict with any degree of certainty” hydrate formation (Ex. 2051, 257:18–
258:11), and agreeing it “is rather difficult to predict how many molecules of
water or solvent can be incorporated” (Ex. 2051, 238:8–18)).
Beyond the above, Patent Owner also argues that the unexpectedly
favorable properties of claim 4’s monohydrate are objective evidence of
nonobviousness. Id. at 60–64. Such properties include thermal stability, and
formulation benefits such as reduced sticking and discoloration. Id. at 61–64
(citing, for example, Ex. 2101 ¶¶ 189–201 (analyzing thermal stability of the
claimed monohydrate versus hydrated sitagliptin HCl and tartrate)).
We find Petitioner has failed to show by a preponderance of the
evidence that claim 4 would have been obvious. We explain below.
As noted by Patent Owner, the Petition provides no rationale to
explain why the ordinarily skilled person would have been motivated to
make the claimed crystalline monohydrate of 1:1 sitagliptin DHP. PO Resp.
54. Even a generic notion that a POSA might have screened for different
polymorphs and hydrates (which Petitioner does not articulate here) would
be wanting for sufficient reasoning and evidentiary support. In re
Armodafinil Patent Litig. Inc., 939 F. Supp. 2d 456, 500 (D. Del. 2013)
(holding that “more than a general motivation to find new crystal forms” is
required). Petitioner’s expert, Dr. Chorghade, offered no motivation
(persuasive or otherwise) in his declaration and admitted that he gave no
opinion on why a POSA would have preferred a hydrate of sitagliptin.
Ex. 2051, 282:18–283:4 (“I don’t think I specifically categorized that they
IPR2020-00040
Patent 7,326,708 B2
62
would have preferred the hydrate.”); Ex. 1002 ¶¶ 168–170. And, when
Petitioner was asked at the oral hearing to identify where, in the Petition, it
had given any reason for the skilled artisan to have made the monohydrate of
claim 4, Petitioner was unable to do so. Tr. 32:23–34:16.
In Reply, Petitioner attempts to fill this hole by citing vaguely to
pages of testimony from Merck’s expert, Dr. Myerson. Reply 23 (citing
Ex. 1031, 70:1–72:21). But Dr. Myerson’s testimony that a POSA may
generally prefer “crystalline substances . . . over amorphous forms” does not
explain why the specific hydrate of claim 4 would have been pursued or
preferred. Ex. 1031, 72:13–21 (testifying that crystalline forms may be
preferred for stability, but amorphous forms are sometimes preferred for
solubility); Ex. 2051 (Chorghade dep.), 274:14–275:17 (admitting that
amorphous forms “can have advantages relative to crystalline materials”).
Nor does Petitioner address the numerous downsides of hydrates that are
reported in the literature and cited by Patent Owner. PO Resp. 56 (“[T]he
POSA would have had a number of reasons for avoiding hydrates.”).
Patent Owner cites undisputed evidence that hydrates are known to
exhibit lower solubility (and consequent lower bioavailability), as well as
physical instability due to dehydration, among other problems. Id. at 59;
Ex. 2160, 6; Ex. 2162, 8–9; Ex. 2163, 6; Ex. 2101¶¶ 56–57, 128–132. Even
Petitioner’s relied-upon Brittain reference catalogs many “problems” and
“potential issues around hydrates in the development process.” Ex. 1005, 5–
6 (listing “lower solubility,” “variable potencies,” and other “negative
consequences” like “tablet capping, chemical instability, discoloration, and
more”). Petitioner pinpoints Brittain’s teaching that “monohydrates are [the]
IPR2020-00040
Patent 7,326,708 B2
63
most frequently encountered” of the crystalline hydrate stoichiometries, but
never grapples persuasively with the problems that accompany hydrates and
how that would (or would not) have impacted the POSA’s motivation to
make the crystalline monohydrate of 1:1 sitagliptin DHP. Pet. 63; Reply
23–24. On the other hand, Dr. Myerson, a professor of chemical
engineering from MIT with over 40 years’ experience, deals with these
issues directly. He testifies persuasively that, considering all the evidence,
including the vague disclosure about crystalline forms in WO ’498 balanced
against the widely-reported problems with hydrates, a POSA would have
sought to avoid hydrates and would not have been motivated to make a
crystalline monohydrate of 1:1 sitagliptin DHP. Ex. 2103 ¶¶ 127–138.
Petitioner’s failure to provide a persuasive motivation for making the
crystalline monohydrate of claim 4 would, alone, justify rejecting its
obviousness challenge. That, however, is not the only problem with
Petitioner’s challenge. On the full trial record, Petitioner does not persuade
us that the ordinarily skilled person would have made the 1:1 sitagliptin
DHP crystalline monohydrate with a reasonable expectation of success.
As discussed above, the preponderance of the evidence shows that
forming crystalline salts, including hydrates, was and is highly
unpredictable. Dr. Myerson testifies persuasively that “it is unpredictable
whether any particular compound will be able to be crystallized” and that no
means existed to reasonably predict formation of hydrates—much less the
specific 1:1 sitagliptin DHP crystalline monohydrate of claim 4. See, e.g.,
Ex. 2101 ¶¶ 146–149. Petitioner’s expert, Dr. Chorghade, concedes that
determining whether hydrates will form is difficult and fraught with
IPR2020-00040
Patent 7,326,708 B2
64
unpredictability. See, e.g., Ex. 2051, 200:1–6, 238:8–18, 245:21–246:6
(admitting it “wasn’t possible to predict when materials will crystallize”),
257:18–258 (admitting he “couldn’t predict with any degree of certainty”
hydrate formation). And the bulk of the technical literature on this record
confirms the same. See, e.g., Ex. 2047, 10 (“[I]t is not yet possible to predict
when materials will crystallize”); Ex. 2050, 31 n.2 (“It is well known among
practising chemists that it is often difficult to crystallize a newly synthesized
compound.”); Ex. 2049, 2 (“[I]t is still not possible to predict with any
reasonable level of confidence the crystal structure of an organic material,
much less the existence of polymorphism.”).34 Against this evidence, we
find unpersuasive Dr. Chorghade’s conclusory opinion that the POSA
“would have expected that (R)-sitagliptin phosphate . . . exists as crystalline
hydrate.” Ex. 1002 ¶ 169.
Petitioner’s probability-based argument, relying on Brittain, is also
over-stated and unpersuasive. PO Resp. 48. We agree with Patent Owner
that Brittain undermines Petitioner’s argument insofar as the reference
discloses that only “one-third of the pharmaceutical actives [APIs] are
capable of forming crystalline hydrates,” and that only about half of the
hydrates in a public database were monohydrates. Id. (quoting 1005, 6). In

34 See also Ex. 2046, 18 (“Predicting the formation of solvates and hydrates
. . . is complex and difficult. . . . There may be too many possibilities so that
no computer programs are currently available for predicting the crystal
structure of hydrates and solvates.”); Ex. 2048, 3:3–5 (“The existence and
physical properties of polymorphs, hydrates and solvates is unpredictable”);
Ex. 2171, 7:32–37 (“There are no rules exist that allow prediction of
whether a compound will exist as solvated forms of an organic solvent.”).
IPR2020-00040
Patent 7,326,708 B2
65
other words, contrary to what Petitioner’s argument suggests about the
prevalence of monohydrates, only about one-sixth of APIs (active
compounds that have been extensively studied) could form a monohydrate
according to Brittain. Id.; Ex. 2101 ¶ 157. Dr. Chorghade, on cross
examination, agreed with this interpretation of Brittain’s data. Ex. 2051,
204:18–205:12. Petitioner’s probabilistic argument, especially given that
the probability is low, is problematic and unpersuasive in showing a POSA’s
reasonable expectation of success in forming the particular crystalline
monohydrate of claim 4. See Grunenthal GmbH v. Alkem Labs. Limited,
919 F.3d 1333, 1341–44 (Fed. Cir. 2019) (affirming finding of no reasonable
expectation of success in producing the “Form A” polymorph of tapentadol
hydrochloride, despite disclosure of the “Form B” polymorph in prior art
and evidence that about 30–35% of the compounds are polymorphic); see
also Lupin, IPR2015-01030, Paper 17, 20–21 (finding similar probability
argument did not show a reasonable expectation of success in formulating a
hydrate of darunavir); PO Resp. 48–49; Sur-reply 17–18.
In Reply, Petitioner argues that the crystalline monohydrate can be
made using an isopropanol/water solvent. Reply 24–25 (citing testimony of
Dr. Matzger that use of isopropanol/water as a screening solvent is “old
chemistry” (Ex. 2103 ¶ 149; Ex. 1025, 232:14–19)). Putting aside that this
is a new theory based on a hypothetical use of an isopropanol solvent system
that we could rightly refuse to consider in determining the alleged
obviousness of claim 4 (37 C.F.R. § 42.23(b)), it does not justify a finding in
Petitioner’s favor. Sur-reply 18.
IPR2020-00040
Patent 7,326,708 B2
66
There are numerous solvents and reaction parameters that a POSA
could choose in screening for possible crystalline salts (see, e.g., Ex. 2101
¶¶ 79, 90–91, 163–164, 166), but
to have a reasonable expectation of success, one must be
motivated to do more than merely vary all parameters or try each
of numerous possible choices until one possibly arrived at a
successful result, where the prior art gave either no indication of
which parameters were critical or no direction as to which of
many choices is likely to be successful.
Grunenthal, 919 F.3d at 1342 (internal quotation marks omitted); accord
Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348, 1365 (Fed. Cir. 2007).
Petitioner’s new focus on isopropanol/water appears to be rooted in
hindsight35 and also fails because the record shows its use, even assuming it
was an obvious solvent choice, does not necessarily result in the claimed
crystalline monohydrate of the 1:1 sitagliptin DHP. As discussed above,
actual testing shows that reacting sitagliptin and phosphoric acid in an
isopropanol/water solvent produces non-1:1 salts. Supra Section II(D)(2)
(discussing Dr. Matzger’s experiments); see also Ex. 2127 (Hansen Decl.)
¶ 34 (no hydrate despite use of aqueous IPA (isopropanol), citing Ex. 2131,
18)). And, if isopropanol/water was chosen, Petitioner still fails to explain
persuasively why the skilled artisan would have reasonably expected the
resulting salt to be the claimed crystalline monohydrate.

35 Petitioner’s focus on isopropanol/water likely arose because the
’708 patent describes a preparation of the monohydrate in that solvent.
Reply 24–25 (citing Ex. 1001, 8:1–13:21, and noting that “inherency” can be
based on the patent). A patent’s own disclosure can evidence inherency, but
Petitioner’s use here looks like hindsight to support the alleged obviousness.
IPR2020-00040
Patent 7,326,708 B2
67
For the above reasons, we disagree that a POSA would have
reasonably expected to successfully produce the subject matter of claim 4
from the combination of WO ’498, Bastin, and Brittain.
Finally, although we need not reach the alleged unexpected properties
of the claimed crystalline monohydrate, such unexpected results are further
evidence undermining Petitioner’s challenge to claim 4. See Hamilton
Beach Brands, Inc. v. f’real Foods, LLC, 908 F.3d 1328, 1343 (Fed. Cir.
2018) (holding no need to reach objective indicia of nonobviousness where
petitioner had not made a showing necessary to prevail on threshold
obviousness issues). Petitioner does not contest that Patent Owner’s relied-
upon results were surprising or unexpected, and Dr. Myerson’s supporting
testimony, which we credit, indicates that they were. Ex. 2101 ¶¶ 186–214.
Rather, Petitioner argues, “1:1 sitagliptin DHP is disclosed in WO498 . . .
and other sitagliptin phosphoric acid salts are disclosed in WO498,” which
Petitioner characterizes as “the closest prior art.” Reply 25. So
characterized, Petitioner argues Merck’s results are insufficient because
Merck did not compare the results against the “closest” salts, and instead
used “more distant salts e.g., HCl, tartrate, etc.” Id. Petitioner also contends
that “Merck only provides data tied to a single crystalline monohydrate
form” of 1:1 sitagliptin DHP, and therefore the results are not
“commensurate with the scope” of the claims. Id. at 25, 27.
We disagree with Petitioner’s above-noted criticisms of the
unexpected results evidence. Id. at 25–27. The only salt of sitagliptin
actually identified in WO ’498 is the sitagliptin hydrochloride of Example 7.
Ex. 1004, 47. No phosphate salt of sitagliptin is specifically identified, and
IPR2020-00040
Patent 7,326,708 B2
68
no hydrates of any salt are described. Ex. 2101 ¶ 136. Merck’s evidence, in
making comparisons against, for example, hydrated forms of sitagliptin
hydrochloride provides a more robust and persuasive comparison than it was
even required to make. Id. ¶¶ 189–201 (analyzing stability results where
water in the claimed monohydrate’s lattice does not dehydrate until over
100°C versus dehydration at about 20°C for hydrated sitagliptin HCL and
tartrate). “Unexpected results are shown in comparison to what was known,
not what was unknown.” Millennium Pharma., Inc. v. Sandoz Inc., 862 F.3d
1356, 1368 (Fed. Cir. 2017) (holding patent owner, in presenting unexpected
results, “was not required to create the glycerol ester, when the product had
not been created in the prior art”).
Moreover, even assuming a 1:1 sitagliptin DHP salt was identified in
WO ’498, Merck provides evidence, for example, that anhydrous forms of
that salt undergo undesirable form conversion while the crystalline
monohydrate of claim 4 unexpectedly does not. Ex. 2101 ¶¶ 202–207; see
also id. ¶ 210 (unpredictable reduction in “sticking” of the monohydrate
versus the anhydrous form). Inasmuch as Petitioner’s position would require
Merck to have further compared the claimed subject matter to itself, that is
not required based on the prior art asserted here. See Millennium, 862 F.3d
at 1368 (citing, with approval, In re Geiger, 815 F.2d 686, 690 (Fed. Cir.
1987) (Newman J., concurring) (“The applicant is not required to create
prior art, nor prove that his invention would have been obvious if the prior
art were different than it actually was.”)).
We also conclude that Petitioner’s criticism about Merck testing only
one form of the crystalline monohydrate of 1:1 sitagliptin DHP is
IPR2020-00040
Patent 7,326,708 B2
69
unavailing. The testing must be “reasonably commensurate” to claim 4’s
scope, and there is no requirement to test “every embodiment.” In re Kao,
639 F.3d 1057, 1068 (Fed. Cir. 2011). The undisputed evidence here is that,
despite Merck’s years of research and efforts to identify other forms, “the
crystalline monohydrate in the ’708 patent is the only monohydrate of 1:1
sitagliptin known to exist.” Sur-reply 26; Ex. 2101 ¶ 205; Ex. 2122, 34;
Ex. 2124 ¶¶ 9, 15. On this record, Merck’s testing was reasonable and
demonstrated unexpected, advantageous properties of the crystalline
monohydrate of claim 4.
For the reasons above, Petitioner has not proved by a preponderance
of the evidence that claim 4 would have been obvious.
III. MOTION TO EXCLUDE
Patent Owner moves to exclude three exhibits (or portions thereof).
See generally Paper 81. Petitioner opposes the motion. Paper 83. Patent
Owner moves to exclude Exhibit 2225, the declaration of Dr. Leonard J.
Chyall dated August 3, 2010, “to the extent Mylan relies on it.” Paper 81, 3.
Patent Owner moves to exclude Exhibit 1030, Dr. Chyall’s lab notebook, in
its entirety. Id. And lastly, Patent Owner moves to exclude portions of
Exhibit 1035, Dr. Chorghade’s Reply Declaration, “to the extent it addresses
Exhibit 1030 or Exhibit 2225.” Id. at 3–4. Patent Owner has the burden of
proof to show that it is entitled to the requested relief. 37 C.F.R. § 42.20(c).
We deny Patent Owner’s motion for the reasons explained below.
IPR2020-00040
Patent 7,326,708 B2
70
Exhibit 2225 (Chyall Declaration)
Merck argues Exhibit 2225 is inadmissible for two reasons. First,
Merck argues the Board should exclude portions of Exhibit 2225 cited by
Mylan because they are hearsay. Paper 81, 3–6. Second, Merck argues
Exhibit 2225 is inadmissible expert testimony (citing Fed. R. Evid. (“FRE”)
702 and Daubert v. Merrill Dow Pharm., Inc., 509 U.S. 579 (1993)
(“Daubert”)). We disagree Exhibit 2225 should be excluded on these bases.
We overrule Merck’s hearsay objection because Merck introduced
this exhibit and, despite its argument to the contrary, it too relied on
Dr. Chyall’s testimony (in Exhibit 2225 and several other exhibits) for the
truth of the matters asserted.
In its Response, Merck cited generally to “EX2225” (and other Chyall
testimony from prior proceedings in Israel, such as in Exhibits 2192,36 2224,
and 2227) without qualification, limitation, or objection. See, e.g., PO Resp.
17; Paper 83, 1–4. Merck cited this testimony to support its argument that
“Dr. Leonard Chyall, of Purdue University, repeatedly reproduced the same
material [i.e., non-1:1 salts of sitagliptin and phosphoric acid].” Id. Also,
Merck’s expert, Dr. Matzger, cited repeatedly to Exhibit 2225, including
some of the same portions relied on by Petitioner and Dr. Chorghade. For
example, Dr. Matzger, pointing to Dr. Chyall’s methanol-based experiments,

36 Exhibit 2192 (Dr. Chyall’s Second Declaration, dated March 7, 2012), for
example, also appears to include some of the same details of Dr. Chyall’s
testimony in Exhibit 2225 on the topic of whether it is possible to prepare
non-1:1 phosphoric acid salts of sitagliptin; Mylan cites and Dr. Chorghade
relies on Exhibit 2192 as well, but Merck does not move for its exclusion.
IPR2020-00040
Patent 7,326,708 B2
71
writes: “Dr. Chyall, Petitioner Teva’s expert in the Israeli opposition . . .
used an excess of sitagliptin base in multiple of his experiments as well as
tested the influence of excess acid.” Ex. 2103 ¶ 126 (citing Ex. 2225 ¶¶ 23–
25); see also id. ¶¶ 130–131 (citing details of Dr. Chyall’s solubility study
and those results in Exhibit 2225 ¶ 69). This was plainly reliance on
Dr. Chyall’s testimony for its truth, and Merck’s argument that it was cited
merely for non-hearsay purposes and because it allegedly motivated
Dr. Matzger’s own experiments is not persuasive. Paper 85, 2.37
Having relied on Dr. Chyall’s testimony for factual details about his
salt-forming experiments and the actual results of some of those
experiments, as Merck does, it’s hardly surprising Petitioner responded by
citing experimental results in Merck’s own exhibit that allegedly support
Petitioner’s argument—experiments showing formation of 1:1 phosphate
salts of sitagliptin when reacted in methanol under certain conditions.
Merck’s argument that it “could not have foreseen” Mylan’s use of Exhibit
2225 in this way is, thus, unpersuasive. Paper 85, 3. Once Merck opted to
use Dr. Chyall’s testimony as a sword, Petitioner was entitled to respond in
kind. Cf. Caterpillar, Inc. v. Wirtgen Am., Inc., IPR2018-01091, Paper 49 at
71 (PTAB Nov. 27, 2019) (“We will not endorse Patent Owner’s attempt to
use the transcript as a sword for its purposes, and our rules as a shield to

37 For similar reasons, Patent Owner’s citation to Fed. R. Evid. 105 is
inapposite. Paper 85, 1. We disagree that Patent Owner’s reliance on
Chyall’s testimony is not for its truth. Accordingly, this is not a situation
where there is a permissible non-hearsay use by one party and an
impermissible hearsay use by another.
IPR2020-00040
Patent 7,326,708 B2
72
prevent Petitioner from using the same transcript to rebut Patent Owner’s
contentions.”); Paper 83, 1.
As for Merck’s contention that Dr. Chyall was originally an expert of
joinder petitioner Teva, that reliance on Dr. Chyall was withdrawn after
joinder was granted, and that Dr. Chyall was not cross-examined in this case,
we recognize that Merck sees this as unfair to it. Paper 81, 5–6. On the
other hand, Merck opened the door to Petitioner citing Dr. Chyall’s
experiments where 1:1 salts were made when Merck cited those very
experiments (without noting their results) along with other Chyall
experiments where he produced non-1:1 phosphate salts. PO Resp. 17.
Moreover, early in this case, Merck suggested to the Board that it may
pursue Dr. Chyall’s deposition. Ex. 1018, 13:15–14:2. The Board signaled
it would promptly handle any disputed request for additional discovery, if
raised. Paper 44, 9–10. Yet Merck apparently never requested a deposition
of Dr. Chyall, nor sought the Board’s further help in resolving any discovery
dispute. Paper 83, 12–13. In short, Merck cannot escape that its own
actions (and inactions) played a significant role in placing Merck in the
position it finds itself now with respect to Exhibit 2225.
Merck also requests exclusion of Exhibit 2225 under FRE 702. Paper
81, 7–10. According to Merck, Dr. Chyall’s declaration never mentions
WO ’498, nor did Dr. Chyall seek to reproduce any examples from that
reference. Inasmuch as Dr. Chyall’s declaration reflects an attempt to make
1:1, or non-1:1, salt forms of sitagliptin, Merck contends his testimony was
specific to issues in the proceedings in Israel and “bears no connection to
U.S. inherent anticipation law.” Id. at 8–9. For at least those reasons, Patent
IPR2020-00040
Patent 7,326,708 B2
73
Owner contends the testimony has no scientific connection to the facts of
this case and is unhelpful. Id. at 7–10.
We agree with Petitioner that Merck’s arguments go the weight of this
evidence, not admissibility. Paper 83, 13–15. To the extent there are
differences between Dr. Chyall’s experiments and the examples in the prior
art, or other alleged shortcomings with Dr. Chyall’s work, that potentially
diminish the scientific connection and relevance to the particular issues here,
Merck has pointed those things out. See, e.g., Sur-reply 5–7. And we have
considered and weighed such matters in reaching our conclusions.
Similarly, to the extent Merck’s expert relied on Dr. Chyall’s work, but also
conducted his own experiments in this proceeding to confirm formation of
non-1:1 salts, whereas Dr. Chorghade relied on Dr. Chyall’s experiments
without running his own, that too goes to the weight of the respective
testimony. Paper 81, 10. We, thus, overrule Patent Owner’s FRE 702 and
Daubert objection to Exhibit 2225.
Exhibit 1030 (Chyall Lab Notebook)
Merck moves to exclude Exhibit 1030 for lack of authentication under
FRE 901, violation of 37 C.F.R. § 42.65, and as hearsay. Paper 81, 11–12.
We deny Merck’s motion to exclude Exhibit 1030. There is a
sufficient showing that this exhibit is an authentic copy of Dr. Chyall’s lab
notebook, as evidenced by the declaration of Noam Blei (Ex. 1036 ¶¶ 1–3),
and based on indicia on the document itself (each page titled “Notebook
4063”) in combination with Dr. Chyall’s declarations (e.g., Exhibits 2221
and 2225, cross-referencing to “Notebook 4063”); FRE 901(b)(3)-(4)
(authentication provided by comparison of expert or trier of fact, or by
IPR2020-00040
Patent 7,326,708 B2
74
distinctive characteristics of the evidence taken together with all the
circumstances). Merck did not timely object to Exhibit 1030 under Rule
42.65, so the objection was waived.38 Paper 68, 2; 37 C.F.R. § 42.64(b)(1).
Finally, even if we agreed that the notebook is hearsay, we conclude it was
permissible, under FRE 703, for Dr. Chorghade to have relied upon the data
therein in forming his opinions. That Patent Owner’s own expert relies on
similar lab notebooks from Dr. Chyall is sufficient to show that the data
therein are the type that experts in this field reasonably rely on in forming
technical opinions. See, e.g., Ex. 2103 ¶ 131 (citing and relying on Ex. 2226
(“Chyall Lab Notebook 4031”)).
Exhibit 1035 (Chorghade Reply Declaration)
Merck moves to exclude Exhibit 1035 because “large portions . . . rely
on and/or parrot the inadmissible experiments and conclusions of Dr. Chyall
in Exhibits 2225 and 1030.” Paper 81, 12. We disagree with Merck for
reasons already explained above. Under the circumstances here, it was
reasonable for Dr. Chorghade to rely on Dr. Chyall’s testing and testimony.
Merck’s contention that the facts and data in Exhibits 2225 and 1030 are not
the sort that experts in this field would reasonably rely upon is undermined
by its own expert’s testimony. Id. at 13; see Ex. 2103 ¶ 126 (citing and
relying on Ex. 2225), ¶ 131 (citing and relying on Ex. 2225 and Ex. 2226
(“Chyall Lab Notebook 4031”)). That Dr. Chorghade may lack “personal

38 Even if the objection had been timely, the declarations of Drs. Chyall and
Chorghade include sufficient information to explain the cited testing and
comply with Rule 42.65, which does not specify that the affiant must have
performed the testing first hand. 37 C.F.R. § 42.65.
IPR2020-00040
Patent 7,326,708 B2
75
knowledge” about that data because he did not generate it himself goes to
the weight of his opinions, not admissibility. Paper 81, 13. And, for similar
reasons to those above, we disagree that Dr. Chorghade’s Reply Declaration,
to the extent it relies on Dr. Chyall’s allegedly “unhelpful” testing or
testimony, should be excluded under FRE 702. Paper 81, 10, 14–15.
For the above reasons, Merck’s Motion to Exclude is denied.
IV. CONSTITUTIONALITY
Patent Owner states: “[T]his IPR should be dismissed as
unconstitutional because administrative patent judges are principal officers
requiring presidential appointment and Senate confirmation.” Sur-reply 26–
27. We decline to consider Patent Owner’s constitutional challenge, as the
issue has been addressed by intervening authority in Arthrex, Inc. v. Smith &
Nephew, Inc., 941 F.3d 1320, 1335 (Fed. Cir. 2019), cert. granted sub nom.
United States v. Arthrex, Inc., 2020 WL 6037206 (Oct. 13, 2020).
V. CONCLUSION
Based on the information in this record, we conclude Petitioner has
not shown by a preponderance of the evidence that claims 1–4, 17, 19, and
21–23 of the ’708 patent are unpatentable for anticipation or obviousness.
Claims
35
U.S.C. §
Reference(s)/Basis
Claims
Shown
Unpatentable
Claims
Not shown
Unpatentable
1–3, 17,
19, 21–
23
102(a),
(e)(2)
WO ’498 1–3, 17, 19,
21–23
1–3, 17,
19, 21–
23
102(a),
(e)(2)
’871 patent 1–3, 17, 19,
21–23
IPR2020-00040
Patent 7,326,708 B2
76

VI. ORDER
In consideration of the foregoing, it is hereby:
ORDERED that Petitioner has not shown by a preponderance of the
evidence that claims 1–4, 17, 19, and 21–23 are unpatentable;
FURTHER ORDERED that Patent Owner’s Motion to Exclude is
denied; and
FURTHER ORDERED that, because this is a Final Written Decision,
parties to the proceeding seeking judicial review of the decision must
comply with the notice and service requirements of 37 C.F.R. § 90.2.
3, 17, 19,
21–23
103 WO ’498 3, 17, 19, 21–
23
1–3, 17,
19, 21–
23
103 WO ’498, Bastin 1–3, 17, 19,
21–23
4 103 WO ’498, Bastin,
Brittain
4
4 103 WO ’498, Brittain 4
Overall
Outcome
1–4, 17, 19,
21–23
IPR2020-00040
Patent 7,326,708 B2
77
FOR PETITIONER:

Jitendra Malik
Alissa Pacchioli
Heike Radeke
Christopher West
Lance Soderstrom
KATTEN MUCHIN ROSEMAN LLP
jitty.malik@katten.com
alissa.pacchioli@katten.com
heike.radeke@kattenlaw.com
christopher.west@katten.com
lance.soderstrom@kattenlaw.com

Russell W. Faegenburg
Tedd W. Van Buskirk
Michael H. Teschner
LERNER, DAVID, LITTENBERG, KRUMHOLZ & MENTLIK, LLP
rfaegenburg.ipr@ldlkm.com
tvanbuskirk@lernerdavid.com
mteschner.ipr@ldlkm.com

Jovial Wong
WINSTON & STRAWN LLP
jwong@winston.com

FOR PATENT OWNER:
Stanley E. Fisher
Jessamyn S. Berniker
Shaun P. Mahaffy
Anthony H. Sheh
WILLIAMS & CONNOLLY LLP
sfisher@wc.com
jberniker@wc.com
smahaffy@wc.com
asheh@wc.com