Regents of the University of Minnesota

Patent Trials and Appeals Board

May 21, 2021

IPR2017-01712 (P.T.A.B. May. 21, 2021)

Trials@uspto.gov Paper 67
571-272-7822 Date: May 21, 2021

UNITED STATES PATENT AND TRADEMARK OFFICE

____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

GILEAD SCIENCES, INC.,
Petitioner,

v.

REGENTS OF THE UNIVERSITY OF MINNESOTA,
Patent Owner.
____________

IPR2017-01712
Patent 8,815,830 B2
____________

Before ERICA A. FRANKLIN, ZHENYU YANG, and
CHRISTOPHER G. PAULRAJ, Administrative Patent Judges.

YANG, Administrative Patent Judge.

JUDGMENT
Final Written Decision
Determining All Challenged Claims Unpatentable
35 U.S.C. § 318(a)

IPR2017-01712
Patent 8,815,830 B2

2
INTRODUCTION
Gilead Sciences, Inc. (“Petitioner”) filed a Petition (Paper 1 (“Pet.”)),
seeking an inter partes review of claims 1–9, 11–21, and 23–28 of U.S.
Patent No. 8,815,830 B2 (Ex. 1001, “the ’830 patent”). We instituted trial to
review the challenged claims. Paper 46 (“Dec.”). Thereafter, Regents of the
University of Minnesota (“Patent Owner”) filed a Response to the Petition
(Paper 54, “PO Resp.”), Petitioner filed a Reply (Paper 57), and Patent
Owner filed a Sur-Reply (Paper 58). An oral hearing for this proceeding was
held on February 3, 2021, and a transcript of that hearing is of record. See
Paper 66 (“Tr.”).
The Board has jurisdiction under 35 U.S.C. § 6 and issues this final
written decision pursuant to 35 U.S.C. § 318(a) and 37 C.F.R. § 42.73. For
the reasons provided below, and based on the evidence and arguments
presented in this proceeding, we conclude Petitioner has established by a
preponderance of the evidence that claims 1–9, 11–21, and 23–28 of the
’830 patent are unpatentable.
Related Matters
According to the parties, Patent Owner asserted the ’830 patent
against Petitioner in Regents of the University of Minnesota v. Gilead
Sciences, Inc., No. 16-cv-02915 (D. Minn.). Pet. x; Paper 3, 1. The case was
later transferred to the U.S. District Court for Northern District of California
and was docketed as Regents of the University of Minnesota v. Gilead
Sciences, Inc., No. 3:17-cv-06056 (N.D. Cal.). Paper 22, 1; Paper 23, 1.
Petitioner also filed three other petitions (IPR2017-01753,
IPR2017-02004, IPR2017-02005), all challenging the claims of the
IPR2017-01712
Patent 8,815,830 B2

3
’830 patent. Paper 23, 1. We previously denied institution in those
proceedings. IPR2017-01753, Paper 42; IPR2017-02004, Paper 38;
IPR2017-02005, Paper 40.
Case History
Petitioner filed the Petition on July 7, 2017. Paper 5. With our
authorization, the parties briefed the issue of whether the doctrine of
sovereign immunity applies in this proceeding such that we should grant
Patent Owner’s Motion to Dismiss the Petition. Papers 14, 15, 16.
While the Motion to Dismiss in this case was pending, the Board
denied Patent Owner’s motions to dismiss based on sovereign immunity in
several other inter partes review proceedings. LSI Corp. v. Regents of the
Univ. of Minn., IPR2017-01068, Paper 19 (PTAB Dec. 19, 2017); Ericsson
Inc. v. Regents of the Univ. of Minn., IPR2017-01186, -01197, -01213,
-01214, -01200, -01219 (PTAB Dec. 19, 2017). On February 12, 2018,
Patent Owner filed a Notice of Appeal, seeking immediate appellate review
of those decisions. See, e.g., IPR2017-01186, Paper 22.
Under such circumstances, and at the request of Patent Owner, we
suspended this proceeding in view of the appellate adjudication of the state-
sovereign-immunity issue. Papers 17, 22, 25, 28. Petitioner then sought, and
was granted, leave to intervene in those appeals. Paper 26, 2.
On June 14, 2019, the Federal Circuit affirmed the Board’s decision
denying Patent Owner’s motion to dismiss in those proceedings. Regents of
the Univ. of Minn. v. LSI Corp., 926 F.3d 1327, 1330 (Fed. Cir. 2019)
(holding state sovereign immunity does not apply to IPR proceedings).
On January 13, 2020, the Supreme Court denied Patent Owner’s petition for
IPR2017-01712
Patent 8,815,830 B2

4
writ of certiorari. Regents of the Univ. of Minn. v. LSI Corp., 140 S. Ct. 908
(Jan. 13, 2020) (No. 19-337). The next day, we denied Patent Owner’s
Motion to Dismiss and lifted the stay order in this proceeding. Paper 32.
The ’830 Patent
The ’830 patent issued from application No. 14/229,292 (“the
’292 application”), filed on March 28, 2014, which is a continuation of
application No. 13/753,252 (hereinafter “NP4”), filed on January 29, 2013,
which is a continuation of application No. 11/721,325 (hereinafter “NP3”),
filed on June 8, 2007, which is a national stage application of
PCT/US2005/044442 (hereinafter “NP2”), filed on December 8, 2005,
which claims priority to provisional application No. 60/634,677 (hereinafter
“P1”), filed on December 9, 2004. Ex. 1001, codes (21), (22), (60), (63),
1:7–15; Pet. 29.
The ’830 patent relates to nucleosides with antiviral and anticancer
activity, specifically nucleotide phosphoramidate prodrugs that are
potentially good substrates for human histidine triad nucleotide-binding
protein 1 (“hHINT1”). Ex. 1001, 2:13–47. According to the ’830 patent,
“[i]nspection of the active site of hHINT1 has revealed that hydrogen
bonding, ion pairing or polar interactions at the 2'- and 3'-positions
preferentially interact with the active site residue Asp-43, which is consistent
with the reduced ability of 2'-deoxy nucleoside phosphoramidates to serve as
substrates.” Id. at 2:36–42. In addition, the ’830 patent discloses that
compounds containing an electropositive group at the 2'-position “are
especially good substrates for hHINT1.” Id. at 2:44–48.
IPR2017-01712
Patent 8,815,830 B2

5
The ’830 patent acknowledges that “U.S. Pat. No. 6,475,985 reports
certain specific nucleoside phosphoramidate analogs having anticancer
and/or antiviral properties.” Id. at 1:61–63. It states that there were other,
continued interests “in phosphoramidate nucleoside analogs due to their
demonstrated utility as prodrugs of antiviral and anticancer nucleoside
monophosphates, or pronucleotides.” Id. at 1:63–66. The ’830 patent states
that despite the prior-art studies on this topic, there was still “a need for
chemotherapeutic agents with antiviral and[/]or anticancer properties.” Id.
at 2:6–8. According to the ’830 patent, its invention provides such
“compounds that act as antiviral and[/]or anticancer agents.” Id. at 2:48–49.
Illustrative Claim
Claim 1 is independent and is reproduced below:
1. A compound of formula I:

wherein:
R1 is guanine, cytosine, thymine, 3-deazaadenine, or uracil,
optionally substituted by 1, 2, or 3 U; wherein each U is
independently halo, hydroxy, (C1-C6)alkyl, (C3-C6)cycloalkyl,
(C1-C6)alkoxy, (C3-C6)cycloalkyloxy, (C1-C6)alkanoyl,
(C1-C6)alkanoyloxy, trifluoromethyl, hydroxy(C1-C6)alkyl,
––(CH2)1-4P(=O)(ORw)2, aryl, aryl(C1-C6)alkyl, or NRxRy;
R2 is halo;
IPR2017-01712
Patent 8,815,830 B2

6
R6 and R7 are independently H or (C1-C6)alkyl;
R3 is hydroxy;
R4 is hydrogen, (C1-C6)alkyl, (C3-C6)cycloalkyl, aryl,
aryl(C1-C6)alkyl, or 2-cyanoethyl;
R5 is an amino acid;
X is oxy, thio, or methylene;
each Rw is independently hydrogen or (C1-C6)alkyl;
Rx and Ry are each independently hydrogen, (C1-C6)alkyl,
(C3-C6)cycloalkyl, phenyl, benzyl, phenethyl, or
(C1-C6)alkanoyl; or Rx and Ry together with the nitrogen to which
they are attached are pyrrolidino, piperidino or morpholino;
wherein any (C1-C6)alkyl of R1, R4-R7, Rw, Rx, and Ry is
optionally substituted with one or more halo, hydroxy,
(C1-C6)alkoxy, (C3-C6)cycloalkyloxy, (C1-C6)alkanoyl,
(C1-C6)alkanoyloxy, trifluoromethyl, azido, cyano, oxo (=O),
(C1-C6)alkyl, (C3-C6)cycloalkyl, (C3-C6)cycloalkyl(C1-C6)alkyl,
(C1-C6)alkyl-S––(C1-C6)alkyl-, aryl, heteroaryl,
alkyl(C1-C6)alkyl, or heteroaryl(C1-C6)alkyl, or NRajRak;
wherein each Raj and Rak is independently hydrogen,
(C1-C6)alkyl, (C3-C6)cycloalkyl, phenyl, benzyl, or phenethyl;
and wherein any aryl or heteroaryl may optionally be substituted
with one or more substituents selected from the group consisting
of halo, hydroxy, (C1-C6)alkyl, (C3-C6)cycloalkyl,
(C1-C6)alkoxy, (C3-C6)cycloalkyloxy, (C1-C6)alkanoyl,
(C1-C6)alkanoyloxy, trifluoromethyl, trifluoromethoxy, nitro,
cyano, and amino;
or a pharmaceutically acceptable salt thereof.
IPR2017-01712
Patent 8,815,830 B2

7
Instituted Ground of Unpatentability
We instituted trial to determine whether claims 1–9, 11–21, and 23–28
of the ’830 patent are unpatentable based on the following single ground:
Claims Challenged 35 U.S.C. § References
1–9, 11–21, 23–28 102 Sofia1
Petitioner relies on the Declarations of Gerardus Josephus Petrus
Henricus Boons, Ph.D. Exs. 1011,2 1039. Patent Owner relies on the
Declaration of Andrea Brancale, Ph.D. Ex. 2058.
ANALYSIS
Legal Standards
In an inter partes review, the burden of persuasion is on the petitioner
to prove unpatentability by a preponderance of the evidence, and that burden
never shifts to the patentee. Dynamic Drinkware, LLC v. Nat’l Graphics,
Inc., 800 F.3d 1375, 1378 (Fed. Cir. 2015).
A patent claim is entitled to the benefit of the filing date of an earlier
filed, related application if it meets the requirements of 35 U.S.C. § 120.
Hollmer v. Harari, 681 F.3d 1351, 1355 (Fed. Cir. 2012). “Section 120
places the burden on the patent owner to provide a clear, unbroken chain of

1 Sofia et al., U.S. Patent Appl. Pub. No. 2010/0016251, published
January 21, 2010 (Ex. 1004).
2 Exhibit 1011 was originally executed by Victor E. Marquez, Ph.D. Later,
we granted Petitioner’s request to substitute the Marquez Declaration in
support of the Petition (Ex. 1011), with the Boons Declaration. Paper 48.
Dr. Boons states that he adopts as his own the substantive statements and
opinions in the Marquez Declaration. Ex. 1036 ¶ 29.
IPR2017-01712
Patent 8,815,830 B2

8
priority.” Droplets, Inc. v. E*TRADE Bank, 887 F.3d 1309, 1317 (Fed.
Cir. 2018) (citing Medtronic CoreValve, LLC v. Edwards Lifesciences
Corp., 741 F.3d 1359, 1366 (Fed. Cir. 2014)).
To claim “the benefit of the filing date of an earlier application under
35 U.S.C. § 120, each application in the chain leading back to the earlier
application must comply with the written description requirement of
35 U.S.C. § 112.” Lockwood v. American Airlines, Inc., 107 F.3d 1565, 1571
(Fed. Cir. 1997); see also Novozymes A/S v. DuPont Nutrition Biosciences
APS, 723 F.3d 1336, 1344 (Fed. Cir. 2013) (“[C]laims added during
prosecution must find support sufficient to satisfy § 112 in the written
description of the original priority application.”).
“[T]o satisfy the written description requirement, the disclosure as
originally filed does not have to provide in haec verba support for the
claimed subject matter at issue.” Purdue Pharma L.P. v. Faulding, Inc.,
230 F.3d 1320, 1323 (Fed. Cir. 2000). It, however, must convey with
reasonable clarity to those skilled in the art that the inventor was in
possession of the invention. Id. “Put another way, one skilled in the art,
reading the original disclosure, must immediately discern the limitation at
issue in the claims.” Id.
“The purpose of the written description requirement is to prevent an
applicant from later asserting that he invented that which he did not.”
Amgen, Inc. v. Hoechst Marion Roussel, Inc., 314 F.3d 1313, 1330
(Fed. Cir. 2003); see also Droplets, 887 F.3d at 1316 (emphasizing
that § 120 embodies an important public policy and requires “strict
adherence to its requirements”). Accordingly, “[e]ntitlement to a filing date
IPR2017-01712
Patent 8,815,830 B2

9
does not extend to subject matter which is not disclosed, but would be
obvious over what is expressly disclosed.” Lockwood, 107 F.3d at 1571–72;
see also Ariad Pharms., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1352
(Fed. Cir. 2010) (en banc) (“[A] description that merely renders the
invention obvious does not satisfy the requirement.”). Likewise, a “mere
wish or plan” for obtaining the claimed invention does not satisfy the written
description requirement. Regents of the Univ. of Cal. v. Eli Lilly & Co., 119
F.3d 1559, 1566 (Fed. Cir. 1997).
The test for written description support “requires an objective inquiry
into the four corners of the specification from the perspective of a person of
ordinary skill in the art.” Ariad, 598 F.3d at 1351. A sufficient description of
a genus “requires the disclosure of either a representative number of species
falling within the scope of the genus or structural features common to the
members of the genus so that one of skill in the art can visualize or
recognize the members of the genus.” Id. at 1350 (internal quotation marks
omitted). “[A]n adequate written description requires a precise definition,
such as by structure, formula, chemical name, physical properties, or other
properties, of species falling within the genus sufficient to distinguish the
genus from other materials.” Id.
The primary consideration in written description analysis is factual
and must be assessed on a case-by-case basis. Vas-Cath Inc. v. Mahurkar,
935 F.2d 1555, 1562 (Fed. Cir. 1991); see also Smith v. Horne,
450 F.2d 1401, 1404 (CCPA 1971) (“The question as to whether an
application forms a proper support for a claim to a composition which is not
specifically disclosed, but which falls among compositions suggested by
IPR2017-01712
Patent 8,815,830 B2

10
general language in the application is one which must be determined largely
by the particular circumstances of each case.”). The Federal Circuit has
warned that each case involving the issue of written description “must be
decided on its own facts. Thus, the precedential value of cases in this area is
extremely limited.” Vas-Cath, 935 F.2d at 1562 (quoting In re Driscoll,
562 F.2d 1245, 1250 (CCPA 1977)).
Claim Construction
In an inter partes review based on a petition filed before
November 13, 2018, the Board interprets a claim term in an unexpired patent
according to its broadest reasonable construction in light of the specification
of the patent in which it appears. 37 C.F.R. § 42.100(b) (2016);3 Cuozzo
Speed Techs., LLC v. Lee, 136 S. Ct. 2131, 2144–46 (2016). Under that
standard, and absent any special definitions, we assign claim terms their
ordinary and customary meaning, as would be understood by one of ordinary
skill in the art at the time of the invention, in the context of the entire patent
disclosure. In re Translogic Tech., Inc., 504 F.3d 1249, 1257
(Fed. Cir. 2007). Any special definitions for claim terms must be set forth
with reasonable clarity, deliberateness, and precision. In re Paulsen,
30 F.3d 1475, 1480 (Fed. Cir. 1994).

3 The rule changing the Board’s claim construction standard to the same
standard used by district courts does not apply here, as the Petition was filed
before the effective date of the final rule, November 13, 2018. See Changes
to the Claim Construction Standard for Interpreting Claims in Trial
Proceedings Before the Patent Trial and Appeal Board, 83 Fed. Reg. 51,340
(Oct. 11, 2018) (amending 37 C.F.R. § 42.100(b) effective November 13,
2018).
IPR2017-01712
Patent 8,815,830 B2

11
Claim terms need only be construed to the extent necessary to resolve
the controversy. Wellman, Inc. v. Eastman Chem. Co., 642 F.3d 1355, 1361
(Fed. Cir. 2011). On this record and for purposes of this Decision, we see no
need to construe any term expressly.
Anticipation by Sofia
Petitioner argues that Sofia anticipates claims 1–9, 11–21, and 23–28.
Pet. 16–27.
Sofia discloses phosphoramidate prodrugs of nucleoside derivatives
for the treatment of viral infections. Ex. 1004, Abstract. Example 25 of Sofia
is (S)-2-{[(2R,3R,4R,5R)-5-(2,4-Dioxo-3,4-dihydro-2H-pyrimidin-1-yl)-4-
fluoro-3-hydroxy-4-methyl-tetrahydro-furan-2-ylmethoxy]-
phosphorylamino}-propionic acid isopropyl ester. Id. at 58. It has the
following structure:

The figure above shows the structure of example 25 of Sofia. Id. at 49.
According to Petitioner, example 25 of Sofia contains, in the nomenclature
of the ’830 patent, oxy at X, uracil at R1, fluoro at R2, hydroxy at R3, phenyl
at R4, the isopropyl ester of L-alanine, an amino acid at R5, methyl at R6, and
hydrogen at R7. Pet. 18 (citing Ex. 1011 ¶ 143). Thus, Petitioner asserts that
IPR2017-01712
Patent 8,815,830 B2

12
example 25 of Sofia meets every limitation of challenged claim 1. Id.
at 18–19. Petitioner also points to evidence to support its argument that
example 25 of Sofia meets every additional limitation of challenged
claims 2–9, 11–21, and 23–28. Id. at 20–27.
Patent Owner does not dispute, and after reviewing the record, we
agree with the parties, that Sofia discloses every limitation of each asserted
claim.
This, however, does not end our inquiry, because the outcome of this
case also turns on the priority date of the challenged claims, and thus, the
prior-art status of Sofia. As explained below, we agree with Petitioner that
the challenged claims are not entitled to a priority date earlier than
March 28, 2014. Therefore, Sofia qualifies as prior art.
Priority Date
According to Petitioner, none of the priority applications (P1, NP2,
NP3, and NP4) provides written description support for the challenged
claims. Pet. 27–73. Thus, Petitioner contends that the priority date of the
challenged claims is no earlier than March 28, 2014, the filing date of the
application that issued as the ’830 patent. Id. at 27. Patent Owner argues that
Petitioner fails to establish NP2 lacks written description support for the
challenged claims. PO Resp. 21–75. As explained below, based on the
evidence of record, we find Petitioner’s arguments more persuasive.
As a preliminary matter, we note that (1) NP2 incorporates P1 by
reference, and contains additional disclosure (see Ex. 1007, 1:9–11,
26:22–27:7); (2) NP3 contains the same disclosure as NP2 (compare
Ex. 1006, with Ex. 1007); and (3) NP4 was filed after the publication date of
IPR2017-01712
Patent 8,815,830 B2

13
Sofia (see Ex. 1001, code (63)). Thus, in this Decision, as Patent Owner
does in its Response, we focus our analysis on NP2, and through the
incorporation by reference, P1. We also focus, as the parties do, on
substituents at R2, R3, R5, R6, and R7. See Pet. 35–39; PO Resp. 26.
Petitioner’s Arguments
NP2 discloses a genus of nucleoside phosphoramidates with a
skeleton, specified by formula I, identical to challenged claim 1’s skeleton.
Ex. 1007, 3:18–19. Petitioner presents Table 3 in the Petition, reproduced
below, to compare the substituents at each relevant position between P1/NP2
and challenged claim 1:
IPR2017-01712
Patent 8,815,830 B2

14

IPR2017-01712
Patent 8,815,830 B2

15
IPR2017-01712
Patent 8,815,830 B2

16
IPR2017-01712
Patent 8,815,830 B2

17
IPR2017-01712
Patent 8,815,830 B2

18

Petitioner presents Table 3 to compare the substituents at each relevant
position between P1/NP2 and challenged claim 1. Pet. 35–39; see also
Ex. 1007, 3:16–4:33.
According to Petitioner, the table above shows the comparison of
substituents between disclosures in P1/NP2 and the challenged claim 1.
Pet. 35, 55 (“Differences between the disclosure in NP3/NP2 and ’830
patent claim 1 is the same as shown above in Table 3.”).
Petitioner argues that “a combinatorial calculation of the number of
substituents encompassed by R2, R3, R5, R6, and R7 in NP3/NP2 formula I
(the R groups narrowed in ’830 patent claim 1) results in 7,441,875 classes
of R group substituents from which the later claimed ’830 patent claim 1
subgenus is derived.” Id. at 56 (citing Ex. 1011 ¶ 96, emphasis omitted), see
also id. at 39–40 (citing Ex. 1011 ¶¶ 54–60, asserting the same regarding the
substituents disclosed in P1). According to Petitioner:
The ’830 patent subgenus selects one substituent (halo) out of 15
possibilities for R2, one substituent (hydroxy) out of 15
possibilities for R3, one substituent (amino acid) out of three
possibilities for R5, two substituents [H or (C1-C6)alkyl] out of
15 possibilities for R6, and two substituents [H or (C1-C6)alkyl]
out of 15 possibilities for R7. This is akin to selecting seven trees
IPR2017-01712
Patent 8,815,830 B2

19
from a forest of 7,441,875 trees, with no guidance in P1 [or NP2]
pointing the POSA to those seven particular trees.
Pet. 40 (citing Ex. 1011 ¶ 61, emphases omitted), 56–57 (citing Ex. 1011
¶ 97).
Petitioner asserts that, although challenged claim 1 narrows the scope
of substituents disclosed in P1 and NP2, there are no blaze marks in either
P1 or NP2 “to guide a POSA to select those specific classes of R group
substituents from the myriad possibilities in the broad genera disclosed” in
P1 or NP2. Id. at 56 (citing Ex. 1011 ¶ 95), see also id. at 39 (citing
Ex. 1011 ¶ 53). For example, Petitioner points out that neither application
“disclose[s] preferred definitions for each R group substituent in Formula I,”
or “teach[es] any improved properties, such as improved antiviral or
anticancer activity for compounds with the R groups selected in the ’830
patent claim 1 subgenus.” Id. at 40.
Instead, Petitioner argues that P1 and NP2 actually direct a POSA to
select other classes of R group substituents because the only compounds
“useful in the methods of the invention” have hydroxy—not halo, as in
challenged claim 1—at R2, and the only synthetic scheme for making the
compounds in P1 and NP2 is one that results in hydroxy at R2. Id. at 41, 43
(citing Ex. 1008, Figs. 1, 3), 57, 59 (citing Ex. 1007, Figs 1, 3); see also
Ex. 1007, 7:9–14; Ex. 1008, 7:9–14.
In addition, P1 and NP2 disclose that compounds containing an
electropositive group at the 2'-position “are especially good substrates for
hHINT1.” Ex. 1007, 3:11–14; Ex. 1008, 3:9–11. Petitioner argues that
halogens, such as fluorine, are electronegative, not electropositive. Pet. 43–
44, 59–60 (citing Ex. 1011 ¶ 103). According to Petitioner, “fluorine is the
IPR2017-01712
Patent 8,815,830 B2

20
most electronegative element in the periodic table.” Id. at 44 (citing
Ex. 1011 ¶¶ 66–67; Ex. 1021, emphasis omitted). Thus, Petitioner concludes
the teachings in P1 and NP2 that “electropositive groups at the 2'-position
are favored for phosphoramidate hydrolysis” would lead a POSA in a
different direction from selecting halo at R2. Id. (citing Ex. 1011 ¶ 68),
60 (citing Ex. 1011 ¶ 104).
Petitioner further contends that neither P1 nor NP2 discloses “any
representative species or subgenus that falls within the scope of ’830 patent
claim 1.” Id. at 42, 57. Nor do they contain any examples of compounds that
fall within the scope of that subgenus. Id. at 42–43, 57. In P1, all
“representative compounds” have hydroxy at R2. Id. at 43 (citing Ex. 1008,
Figs. 1, 3); see also Ex. 1008, 7:9–13 (stating compounds in Figures 1 and 3
are “representative compounds” and compounds “useful in the methods of
the invention”). NP2 discloses 24 additional “[r]epresentative compounds.”
Ex. 1007, 26:22–27:7. According to Petitioner, all but one of these
compounds have hydroxy at R2. Pet. 59. The only one left contains
hydrogen, again, not halo, at R2. Id. It also has hydroxy, and not hydrogen or
(C1-C6)alkyl, as required in challenged claim 1, at R6. Id. Thus, Petitioner
concludes all the representative compounds in P1 and NP2 are outside the
scope of challenged claim 1. Id. at 43 (citing Ex. 1011 ¶¶ 63–65), 59 (citing
Ex. 1011 ¶¶ 101–102).
Petitioner acknowledges that both P1 and NP2 contain numerous
multiple dependent claims. Id. at 44, 60. According to Petitioner, however,
there is nothing in P1 or NP2 singling out the claimed classes of R group
substituents as preferred, and there are “no blaze marks directing a POSA to
IPR2017-01712
Patent 8,815,830 B2

21
select” the R group substituents falling within the scope of challenged
claim 1 subgenus from the many R groups recited in other claims.
Id. at 44–45 (citing Ex. 1008, 23–24; Ex. 1011 ¶¶ 69–73), 60–62 (citing
Ex. 1007, 40–41; Ex. 1011 ¶¶ 105–109).
For P1, Petitioner points out the relevant claims “are all multiple
dependent claims which themselves depend from large numbers of other
multiple dependent claims.” Id. at 45. For example, claim 47 recites “[t]he
compound of any one of claims 1–46, wherein R7 is hydrogen or
(C1-C6)alkyl.” Id. at 45–46 (citing Ex. 1008, 27, emphasis omitted).
Petitioner argues “[t]here are no blaze marks in this claim or any of the
claims of P1 specifically pointing to the substituents recited in ’830 patent
claim 1.” Id. at 46 (citing Ex. 1011 ¶ 74). Thus, according to Petitioner:
[E]ven if claim 47 were to be understood by a POSA as
demonstrating a “preference” for R7 as hydrogen or (C1-C6)alkyl,
claim 47 still represents 70,875 classes of R group substituents
for R2, R3, R5, and R6 with no blaze marks in P1 to direct the
POSA to the classes of R group substituents later claimed in the
’830 patent claim 1 subgenus.
Id. at 46 (citing Ex. 1011 ¶ 74).
To reach challenged claim 1, Petitioner argues, it would require
hindsight to pick, among 46 claim possibilities in claim 47, four claims
(claims 13, 21, 33, and 45), without any direction in P1’s disclosure. Id.
Given the multiple classes of substituents for each of R2, R3 and R5, “and the
lack of blaze marks in P1 to direct a POSA to select R2 as halo, R3 as
hydroxy, and R5 as amino acid (as specifically claimed in the ’830 patent
claim 1 subgenus) from the 7,441,875 other possible combinations,”
IPR2017-01712
Patent 8,815,830 B2

22
Petitioner concludes that the applicant “was not in possession of the ’830
patent claim 1 subgenus as of the filing date of P1.” Id. at 47.
For NP2, Petitioner points out that it differs from P1 in the
organization of the multiple dependent claims. Id. at 61. This, according to
Petitioner, “precludes combinations that would result in the classes of R
group substituents later claimed in the ’830 patent.” Id. at 65 (emphasis
omitted). Thus, Petitioner concludes “even in hindsight[,] a POSA would not
be able to combine the classes of R group substituents later claimed in the
’830 patent claim 1 subgenus from the disclosures and claims of . . . NP2.”
Id. (citing Ex. 1011 ¶ 119).
For similar reasons, Petitioner also contends that neither P1 nor NP2
provides written description support for the challenged dependent claims.
Id. at 50–53 (citing Ex. 1011 ¶¶ 84–90), 66–69 (Ex. 1011 ¶¶ 122–131).
According to Petitioner, “there are no ‘preferences’ or other teachings” in P1
or NP2 that “would provide direction to a POSA, with no foreknowledge of
the later-claimed ’830 patent subgenus, to select the narrowed R2, R3, R5, R6,
and R7 groups claimed in the ’830 patent dependent claims.” Id. at 50 (citing
Ex. 1011 ¶ 84), 66 (citing Ex. 1011 ¶ 122).
In sum, Petitioner concludes
Given the enormous size of the genera of compounds disclosed
in [P1 and] NP2, the lack of blaze marks that would direct a
POSA to the specifically narrowed R groups claimed in the ’830
patent, combined with the teachings in [P1 and] NP2 that would
direct a POSA to select something other than the later claimed R
groups, a POSA would have understood that Applicant was not
in possession of the subject matter of the ’830 patent claims.
Id. at 69 (citing Ex. 1011 ¶ 132), see also id. at 53–54 (citing Ex. 1011 ¶ 91).
IPR2017-01712
Patent 8,815,830 B2

23
Patent Owner’s Arguments
Patent Owner argues that Petitioner overstates the number of possible
combinations in NP2’s broadest genus because it ignores (1) the Polar
Proviso,4 and (2) “‘hydrogen or (C1-C6)alkyl’ was of special interest at R6
and R7.” PO Resp. 4, 28–29 (citing Ex. 2058 ¶¶ 360–362, 376–385,
390–393). Patent Owner also challenges the factorial approach to counting
the number of subgenera as “hopelessly exaggerated.” Sur-Reply 8 (citing
Snitzer v. Etzel, 465 F.2d 899, 903 (CCPA 1972), emphasis omitted).
Instead, Patent Owner adopts a multiplication approach. PO Resp. 45–46
(citing Ex. 2058 ¶¶ 287–289).
Patent Owner also faults Petitioner for ignoring “P1’s disclosure when
analyzing NP2’s, despite P1 being incorporated by reference in NP2.” Id. at
4 (emphasis omitted), 55–56. According to Patent Owner, P1’s dependent
claims and NP2’s specific values are reinforcing blaze marks, and must be
considered together, and Petitioner’s failure to do so is fatal. Id. at 55–56.
Patent Owner further contends that “[a] P1 dependent claim and
corresponding NP2 ‘specific value’ identif[y] the exact substituent(s)
Challenged Claim 1 requires at every challenged R position.” Id. at 5, 30.
Thus, Patent Owner argues, “the only plausible basis for challenging written

4 Patent Owner points out that, in NP2, “[w]hile Formula I recites fifteen
possibilities for each of R2 and R6, the 2′-position selections are limited by
the following: ‘provided that one of R2 and R6 is hydroxy, halo,
(C1-C6)alkoxy, (C3-C6)cycloalkyloxy, trifluoromethyl, cyano, or NRadRae.’”
PO Resp. 3–4. Because “[a]ll seven classes following ‘provided that’ are
polar,” Patent Owner refers to this restriction on R2/R6 selections as “Polar
Proviso.” Id. at 4.
IPR2017-01712
Patent 8,815,830 B2

24
description would be to allege that NP2 does not demonstrate possession of
a genus incorporating them together.” Id. at 30. According to Patent Owner,
“[t]here are at least four ways in which blaze marks demonstrate possession
of Challenged Claim 1’s combination.” Id. at 31.
First, Patent Owner asserts that “P1’s claim 47 recites Challenged
Claim 1’s combination of substituents.” Id. at 33. According to Patent
Owner:
A P1 multiple dependent claim—which NP2 incorporates by
reference—claims Challenged Claim 1’s subgenus: claim 47
(“R7 is hydrogen or (C1-C6)alkyl”) depends from claim 45 (“R6
is hydrogen or (C1-C6)alkyl”) depends from claim 33 (“R5 is an
amino acid”) depends from claim 21 (“R3 is hydroxy”) depends
from claim 13 (“R2 is halo”) depends from claim 2 (R1 meets
Challenged Claim 1’s list of options) depends from claim 1 (R4
and X meet Challenged Claim 1’s list of options).
Id. at 6, see also id. at 31–33 (the same).
Second, Patent Owner argues “Challenged Claim 1’s subgenus was
one of a limited number of options identified by NP2’s specific values for
narrowing NP2’s genus at the challenged R positions.” Id. at 31. Patent
Owner emphasizes there is “a clear interrelationship between the R2/R6
selections at the 2'-position and between the R3/R7 selections at the 3'-
position.” Id. According to Patent Owner, “[t]hese inter-relationships limit
the number of combinations identified as of special interest.” Id., see also id.
at 7 (the same).
Patent Owner argues that “NP2’s specific values and P1’s
corresponding dependent claims concerning the 2' and 3'-positions (R2/R6
and R3/R7) identify just fifty-six subgenera of special interest: seven 2'-
position permutations multiplied by eight 3'-position permutations.” Id. at 44
IPR2017-01712
Patent 8,815,830 B2

25
(citing Ex. 2058 ¶ 263), see also id. at 6–7, 35–43 (analyzing substituents at
the 2′- and 3′-positions). Adding the “only three options at R5” in Formula I,
Patent Owner concludes “Challenged Claim 1’s subgenus is among only 168
combinations of class selections at R2-R3 and R5-R7 (seven at R2/R6 times
eight at R3/R7 times three at R5) that NP2’s specific values and P1’s claims
identify as of special interest,” not 7,441,875, as Petitioner alleges.5 Id.
at 44–45.
Third, Patent Owner points out that “[a]ll of NP2’s compounds
(NP2’s example compounds 100–123 and the compound of P1’s claim 49)
have the same 3'-position selections (hydroxy at R3 and hydrogen at R7)
satisfying Challenged Claim 1’s requirements.” Id. at 49 (citing Ex. 1007,
14:23–25, 26:22–27:10; Ex. 2058 ¶¶ 297–308). In Patent Owner’s view,
NP2’s examples blaze an undeniable trail to this single
combination of choices as of special interest at the 3′-position.
When that single combination of R3/R7 selections is considered
with the seven (or at most ten) choices NP2’s specific values and
P1’s corresponding dependent claims specify at R2/R6 . . . and
the three choices Formula I permits at R5, NP2 provides blaze
marks to twenty-one (or at most thirty) subgenera of special
interest.
Id. (citing Ex. 2058 ¶ 315). Patent Owner asserts that “POSAs would have
readily envisaged all these subgenera, including Challenged Claim 1’s
subgenus.” Id. (citing Ex. 2058 ¶ 316).

5 Alternatively, Patent Owner asserts “[e]ven if one counts options reciting
species of classes in Formula I,” the total is still only 300: “ten each at the 2'
and 3'-positions,” and three at R5. PO Resp. 44–46 (citing Ex. 2058 ¶¶ 264,
288, emphasis omitted).
IPR2017-01712
Patent 8,815,830 B2

26
Fourth, according to Patent Owner, of NP2’s 25 compounds
(compounds 100–123 and one recited in P1’s claim 49), seven (compounds
105–108, 119–120, and 122) “satisfy all Challenged Claim 1’s requirements
except the Polar Proviso is met at R2 by hydroxy rather than halo,” and six
additional examples (compounds 100–103 and 110–111) satisfy all disputed
elements except R2. Id. at 50–51 (citing Ex. 1007, 26:26–27:5; Ex. 2058
¶¶ 322–323, emphasis omitted). Thus, Patent Owner argues that “over half
of NP2’s examples would satisfy the contested R2/R3/R5/R6/R7 genus but for
having hydroxy rather than halo at R2.” Id. at 51 (citing Ex. 2058 ¶ 324,
emphasis omitted). Because halo is one of the Polar Proviso’s six
alternatives to hydroxy (id. (citing Ex. 1007, 3:29–4:1)), Patent Owner
contends that
POSAs would have recognized [the inventor]’s special interest
in six subgenera within Formula I, corresponding to the R3 and
R5-R7 selections in the majority of NP2’s examples combined
with each of the six Polar Proviso classes recited for R2 in NP2’s
“specific value[s]” (and in P1’s dependent claims). One of these
six subgenera corresponds exactly to Challenged Claim 1.
Id. at 53 (citing Ex. 2058 ¶¶ 329–331, emphasis omitted).
According to Patent Owner, these four blaze marks “are consistent
and reinforcing—collectively compelling the conclusion that NP2 supports
the Challenged Claims.” Id. at 5, 53.
Discussion
After reviewing the entire record and weighing evidence offered by
both parties, we find NP2, which incorporates P1 by reference, does not
provide sufficient written support for the challenged claims. We focus our
analysis on claim 1.
IPR2017-01712
Patent 8,815,830 B2

27
As an initial matter, we find no ipsis verbis disclosure of the subgenus
of challenged claim 1, even though Patent Owner contends that P1’s
claim 47 encompasses challenged claim 1’s exact subgenus. See PO Resp. 5,
32–33. P1’s claim 47 specifies that “R7 is hydrogen or (C1-C6)alkyl.”
Ex. 1008, 27:1–2. Claim 47, however, is a multiple dependent claim that
depends from “any one of claims 1–46.” Id. To reach the combination of R2,
R3, R5, and R6 recited in challenged claim 1, an ordinarily skilled artisan
would have to select the substituents defined in claims 13 (“R2 is halo”),
claim 21 (“R3 is hydroxyl”), claim 33 (“R5 is an amino acid”), and claim 45
(“R6 is hydrogen or (C1-C6)alkyl”) from the 46 claims, most of which are
themselves multiple dependent claims. See Pet. 45–47; PO Resp. 5, 32–33.
Such picking and choosing from a “laundry list” disclosure does not
constitute an ipsis verbis disclosure of every subgenus, including the
subgenus of challenged claim 1. See Fujikawa v. Wattanasin, 93 F.3d 1559,
1571 (Fed. Cir. 1996) (“[J]ust because a moiety is listed as one possible
choice for one position does not mean there is ipsis verbis support for every
species or sub-genus that chooses that moiety.”).
Ipsis verbis disclosure, however, is not necessary to satisfy the written
description requirement of section 112. Id. at 1570. Instead, when analyzing
the written description support for a claimed species, or as here, a claimed
subgenus, in description of a broader genus, we look for blaze marks that
single out particular trees in a forest. In re Ruschig, 379 F.2d 990, 994–95
(CCPA 1967). These blaze marks must be clear because “it is easy to bypass
a tree in the forest, even one that lies close to the trail.” Fujikawa, 93 F.3d at
1571. In this case, we find the point at which one must leave the trail to find
IPR2017-01712
Patent 8,815,830 B2

28
the tree is not well marked in P1 and NP2. Thus, P1 and NP2 do not provide
sufficient written description support for the subgenus of challenged claim 1.
Before delving into the facts of this case, we discuss the parties’
dispute over the law on written description. Petitioner argues:
A broad generic description of an invention does not provide
written description support to that genus unless the description:
(1) discloses a representative number of species falling within the
scope of the genus or (2) includes preferences that narrow the
genus to a sufficiently small disclosure so that a POSA can
easily “visualize or recognize” the members of the genus.
Pet. 31 (citing Ariad, 598 F.3d at 1349–50, emphasis added).
In Patent Owner’s view, the emphasized language above misstates the
law on written description. PO Resp. 22. According to Patent Owner, Ariad
neither limits blaze marks to preferences, nor suggest a “sufficiently small
disclosure” test. Id. at 22–23 (emphasis omitted). Instead, Patent Owner
argues “blaze marks evidencing a subgenus was of ‘special interest’—even
if not ‘preferred’—demonstrates possession of that subgenus and provides
written description support.” Id. at 21 (citing Fujikawa, 93 F.3d at 1571).
Although we agree with Patent Owner that blaze marks do not require
identifying the species- or subgenus-at-issue as preferred, such as in
preferred embodiments, we do not discern material difference between
“preferences,” as stated by Petitioner, and “special interest,” as emphasized
by Patent Owner.
Indeed, Smith v. Horne, a binding authority that Patent Owner relies
on (PO Resp. 51, 52, 59, 62 (citing Smith, 450 F.2d at 1404)), explains that
“[t]he indication or lack of indication of a preference for the composition, in
the application disclosure, is an important factor to be considered in making
IPR2017-01712
Patent 8,815,830 B2

29
the determination [of written description support], since anyone attempting
to carry out the disclosure of an application would logically begin with the
preferred examples given.” 450 F.2d at 1404. The Federal Circuit later
reiterated the same in Heymes v. Takaya, another binding authority in which
the court affirmed a BPAI decision relied upon by Patent Owner.6
867 F.2d 616 (Fed. Cir. 1989).
At bottom, we agree with Petitioner that regardless of the label we
apply, when analyzing written description support, “the task is to search for
blaze marks that guide a skilled artisan to the claimed subgenera.” Reply 3–4
(citing Fujikawa, 93 F.3d at 1571); see also Smith, 450 F.2d at 1404 (“The
determining factor is whether the application would fairly suggest to the
skilled worker in the art the particular composition claimed, or whether the
desirability of that composition could be as[c]ertained only by extensive
experimentation.”).
As for Patent Owner’s argument that “there are at least four ways in
which blaze marks demonstrate possession” of the subgenus of challenged
claim 1 (PO Resp. 5–9, 31–53), we are not persuaded.
First, we are not persuaded that P1’s claim 47, viewed as a part of
NP2 because P1 is incorporated into NP2 by reference, constitutes an
adequate blaze mark. As explained above, claim 47 is a multiple dependent
claim that depends from “any one of claims 1–46.” Ex. 1008, 27:1–2. To
reach the subgenus of challenged claim 1, an ordinarily skilled artisan would

6 Patent Owner cites Heymes v. Takaya, 6 U.S.P.Q.2d 1448 (BPAI Feb. 10,
1988) passim, and recognizes that the BPAI decision was affirmed in 867
F.2d 616 (Fed. Cir. 1989). See PO Resp. vi.
IPR2017-01712
Patent 8,815,830 B2

30
have to select claims 13, 21, 33, and 45 from the numerous multiple
dependent claims. PO Resp. 5, 32–33. Petitioner presents the following
figure to illustrate this point:

The figure above is Petitioner’s presentation of “an enormous number
of possibilities” when “picking and choosing from P1’s claims,” “even
starting with claim 47.” Pet. 47. We agree with Petitioner regarding the
magnitude of those possibilities.
P1 itself does not contain any disclosure to guide an ordinarily skilled
artisan to choose a specific R2, R3, or R5 at each position, let alone the
claimed combination. NP2 incorporates P1 by reference, and also lists
“specific values” for each R position. For example, for R2, NP2 discloses:
IPR2017-01712
Patent 8,815,830 B2

31
A specific value for R2 is hydroxy.
A specific value for R2 is halo.
A specific value for R2 is fluoro.
A specific value for R2 is (C1-C6)alkoxy.
A specific value for R2 is methoxy.
A specific value for R2 is trifluoromethyl.
A specific value for R2 is cyano.
A specific value for R2 is amino, mthylamino, dimethylamino,
ethylamino, or dimethylamino.
Ex. 1007, 12:8–17. These specific values for R2 duplicate those recited in
claims 12–20 in P1. See Ex. 1008, 23:6–24. Similarly, NP2 discloses
specific values for R3, R5, R6, and R7 that duplicate those recited in claims
21–29 and 33–48 in P1. Compare Ex. 1008, 23:25–24:17, 24:25–27:7, with
Ex. 1007, 12:18–27, 13:4–14:22.
Patent Owner argues “NP2’s specific values and P1’s dependent
claims are mutually reinforcing.” PO Resp. 34. Even if that is the case,
NP2’s specific values only reinforce P1’s claims that there are multiple
choices at each relevant R position. They do not guide an ordinarily skilled
artisan to pick claims 13, 21, 33, and 45 of P1 to combine with claim 47 in
order to narrow the substituents to the combination of challenged claim 1.
As Petitioner correctly points out, “[a] POSA could just as likely combine
claim 17 (where R2 is methoxy), claim 27 (where R3 is azido), and claim 35
(where R5 is a peptide), in combination with claim 47, or any other
combination of claims.” Pet. 46 (citing Ex. 1011 ¶ 75). It would require
hindsight to pick, among 46 claim possibilities—many of which are multiple
dependent claims themselves—claims 13 (“R2 is halo”), 21 (“R3 is
hydroxy”), 33 (“R5 is an amino acid”), and 45 (“R6 is hydrogen or
(C1-C6)alkyl”) to reach the subgenus of challenged claim 1.
IPR2017-01712
Patent 8,815,830 B2

32
That, of course, is assuming that an ordinarily skilled artisan would
not have started with claim 48, another multiple dependent claim, which
recites “R3 is hydrogen or alkyl” and limits R7 to a list of fourteen categories
of substituents. Ex. 1008, 27:4–7. P1 and NP2 do not guide against starting
with claim 48. In fact, NP2 discloses the same R3 and R7 substituents as in
P1’s claim 48, and labels it, in contrast to the “specific value(s)” when
listing other substituents, a “specific compound,” which suggests a special
interest in this combination. See Ex. 1007, 14:19–22. In any event, were
claim 48 the starting point, then no matter how it is combined with the
claims it depends from, an ordinarily skilled artisan would not have reached
the subgenus of challenged claim 1.
Second, we also are not persuaded that “Challenged Claim 1’s
subgenus is among only 168 combinations of class selections at R2-R3 and
R5-R7 . . . that NP2’s specific values and P1’s claims identify as of special
interest.” See PO Resp. 45 (citing Ex. 2058 ¶¶ 287–289). For example, in
counting the choices for R7, Patent Owner treats the “fourteen classes” of
substituents in P1’s claim 48 and NP2’s corresponding specific compound as
a single option. Id. at 40. Similarly, for R6, Patent Owner treats the “various
classes”—in fact, also fourteen classes—of substituents in P1’s claim 46 and
the corresponding specific value in NP2 as a single option. Id. at 35, see also
id. at 37 (treating “R2 is amino, methylanino, dimethylamino, ethylamino, or
dimethylamino” as a single option), 41 (treating “R3 is amino, methylamino,
IPR2017-01712
Patent 8,815,830 B2

33
dimethylamino, ethylamino, or dimethylamino” as a single option). Thus,
Patent Owner undercounts the combinations of substituents at R2/R3/R6/R7.7
Third, we further are not persuaded that a single combination of
selections at the 3'-position is a sufficient blaze mark for written description
support of the subgenus of challenged claim 1. Patent Owner emphasizes
that “[a]ll of NP2’s compounds . . . have the same 3'-position selections
(hydroxy at R3 and hydrogen at R7) satisfying Challenged Claim 1’s
requirements.” Id. at 49 (citing Ex. 1007, 14:23–25, 26:22–27:10; Ex. 2058
¶¶ 297–308). This, according to Patent Owner, should limit the selection of
R3/R7 combination to only one. Id.

7 Patent Owner also undercounts the number of substituents at R5. See PO
Resp. 44 (“Formula I includes only three options at R5.”). P1 and NP2 both
disclose R5 as “an amino acid, a peptide, or NRaRb.” Ex. 1007, 4:9; Ex.
1008, 4:9. They both, however, define “each Ra and Rb is independently
hydrogen, (C1-C6)alkyl, (C3-C6)cycloalkyl, aryl, or aryl(C1-C6)alkyl; or Ra
and Rb together with the nitrogen to which they are attached form a
pyrrolidino, piperidino or morpholino.” Ex. 1007, 4:15–17; Ex. 1008,
4:15–17. P1 further includes claims, and NP2 further includes corresponding
specific values, directed to NRaRb. Ex. 1007, 13:6–14:12; Ex. 1008,
24:29–26:18 (claims 35–44). In our view, thus, NRaRb represents more than
a single option at R5. Nonetheless, the Petition, as well as the supporting
expert declaration, fail to take this into consideration. See Pet. 47 (Figure 1
listing only amino acid and peptide as options for R5); Ex. 1011 ¶ 58 (expert
testifying R5 “only has three classes of R group substituents”). It is not until
the Reply that Petitioner recognizes “all possibilities for the variables Ra and
Rb.” Reply 17. That is simply too late. Thus, for purpose of this Decision, we
consider there are only three options for R5. See Patent Trial and Appeal
Board Consolidated Trial Practice Guide November 2019, 73 (“Petitioner
may not submit new evidence or argument in reply that it could have
presented earlier.”).
IPR2017-01712
Patent 8,815,830 B2

34
Patent Owner, however, neglects to explain why an ordinarily skilled
artisan would focus on this R3/R7 combination while ignoring the fact that
P1 and NP2 exemplifies hydroxy, not halo, at R2. Indeed, as Petitioner
points out, all “representative compounds” and compounds “useful in the
methods of the invention” in P1 and figures of NP2, as well as twenty-three
out of twenty-four additional “[r]epresentative compounds” in NP2, have
hydroxy at R2.8 Pet. 43 (citing Ex. 1008, Figs. 1, 3), 57–59 (citing Ex. 1007,
26–27, Figs. 1, 3); Ex. 1007, 7:9–13; Ex. 1008, 7:9–13. Thus, we agree with
Petitioner that Patent Owner “inconsistently cherry-picks only a portion of R
groups in NP2’s compounds to create an imaginary example found nowhere
in the written description.” Reply 21.
Fourth, we are not persuaded that the fact that thirteen out of twenty
example compounds in NP2 “satisfy Challenged Claim 1’s requirements for
all challenged R positions except R2 and meet the Polar Proviso at R2 but
with hydroxy rather than halo” serves as an adequate blaze mark for
challenged claim 1’s subgenus, which requires halo at R2. See PO Resp. 32
(emphasis omitted), see also id. at 9–10, 50–53 (the same). Indeed, written
description analysis requires “[t]aking each claim . . . as an integrated whole
rather than as a collection of independent limitations.” Novozymes, 723 F.3d
at 1349. On this point, we agree with Petitioner that
[Patent Owner’s] efforts to make NP2’s examples into blaze
marks is pure hindsight. On the one hand, [Patent Owner]

8 The only one left contains hydrogen, again, not halo, at R2. Pet. 59 (citing
Ex. 1011 ¶ 101); see also Ex. 1038, 36:3–14 (Patent Owner’s expert
testifying that “[a]t the R2 position, there is variation between two groups,
hydrogen or hydroxyl”).
IPR2017-01712
Patent 8,815,830 B2

35
suggests a skilled artisan would not look beyond the specific
substituents contained in NP2’s examples at every position other
than R2/R6. POR, 51; Boons [Ex. 1039] ¶¶73-78. On the other
hand, [Patent Owner] broadens R2 and R6 beyond the substituents
reflected in each example (hydrogen and hydroxy) in order to
manufacture imaginary possibilities where R2 is halo. POR, 51.
There is no reasonable justification for interpreting the examples
as teaching the selections made at some positions (R3 and R7) but
not at others (R2 and R6); either NP2’s examples blaze a trail to
selections actually made, or they blaze no trail at all. Whatever
the case, NP2’s examples do not lead a skilled artisan to the
subgenera claimed in the ’830 patent. Boons [Ex. 1039]
¶¶ 103-105.
Reply 22.
Patent Owner’s other arguments, some of which have more merit than
others, do not change the outcome of this case. For example, Patent Owner
faults Petitioner for ignoring the Polar Proviso. PO Resp. 3–4. In response,
Petitioner points out that the proviso itself does not mention polarity, despite
the label Patent Owner coins, and that NP2 mentions polarity only once,
despite the special interest Patent Owner alleges. Reply 10–11 (citing Ex.
1007, 3:5–9). According to Petitioner, regardless of what “polar” means, the
proviso leaves out as many polarity substituents as it includes. Id. at 11–12
(citing Ex. 1039 ¶¶ 36–42).
Patent Owner does not dispute these arguments. Instead, Patent
Owner accuses Petitioner of “Dwell[ing] on Labels While Neglecting
Substance.” Sur-Reply 22. Patent Owner explains that “all the Polar Proviso
classes are capable of ‘polar interactions,’” and “certain classes in Formula I
excluded from the Polar Proviso” are not polar. Id. at 23. Patent Owner
IPR2017-01712
Patent 8,815,830 B2

36
further argues that “[i]t is immaterial that other classes excluded from the
Polar Proviso may theoretically permit polar interactions.” Id.
We do not need to resolve the parties’ dispute on this issue, other than
observing that NP2 discloses polar interactions at the 2'- and 3'-positions,
together with hydrogen bonding and ion pairing at these positions,
“preferentially interact with the active site residue Asp-43, which is
consistent with the reduced ability of 2’-deoxy nucleoside phosphoramidates
to serve as substrates.” See Ex. 1007, 3:5–9.
We, however, agree with Patent Owner that, because the Polar
Proviso is not considered, the Petition overstates the number of possible
combinations in NP2’s broadest genus. PO Resp. 4, 28–29. In fact,
Petitioner recognizes this and has revised the number of possible
combinations with different R substituents from 7,441,875 in the Petition to
4,266,675 in the Reply to account for the proviso. Reply 25–26 (citing
Ex. 1039 ¶¶ 58–66).
We also question the propriety of Petitioner’s factorial approach in
calculating the number of subgenera in formula I. See Pet. 39–40 (citing
Ex. 1011 ¶¶ 54–60); see also Ex. 1011 ¶¶ 56–57 (testifying that “for R2
alone, there are 32,767 possible ways to combine the 15 classes of R group
substituents,” because “there is no teaching in P1 [or NP2] that discusses any
preference for only choosing one of these classes, as opposed to choosing a
combination of two classes, or three classes, etc.”). Neither Petitioner nor its
expert has provided sufficient evidence or argument to show P1 or NP2
teaches more than one class of substituent at each R position. Thus, we find
Petitioner’s number of subgenera higher than warranted.
IPR2017-01712
Patent 8,815,830 B2

37
At the same time, as explained above, we find the number of
subgenera Patent Owner put forth, 168, to be significantly lower than the
proper count. But, even if we were to accept that NP2 identifies 168
combinations of substituents at the relevant R positions, we still find
insufficient blaze marks in NP2 to guide an ordinarily skilled artisan to the
subgenus of challenged claim 1. We reach this conclusion in view of
evidentiary record in this case, and guidance from the Federal Circuit and its
predecessor court, including that found in Ruschig, Fujikawa, and
Novozymes.
In Ruschig, the applicants challenged the decision by the Patent
Office’s Board of Appeals, finding chlorpropamide, the compound of the
claim-at-issue, “is not named or identified by formula and it can find support
only as choices made between the several variables involved.” 379 F.2d at
991–92. The Board concluded “[t]his is not regarded as adequate support for
a specific compound never named or otherwise exemplified in the
specification as filed,” and the Court of Customs and Patent Appeals
affirmed. Id. at 992.
The court noted “the general disclosure of the application
encompasses something like half a million possible compounds. It also
discloses a number of specific compounds.” Id. at 993. The appellants
argued an original claim was “in effect one of the ‘guides,’” because “there
are ‘approximately’ 48 compounds within the scope of that claim[,] all of
which are ‘readily determinable by skilled chemists.’” Id. at 994.
Rejecting this argument, the court stated:
IPR2017-01712
Patent 8,815,830 B2

38
Specific claims to single compounds require reasonably specific
supporting disclosure and while . . . naming [the compounds] is
not essential, something more than the disclosure of a class of
1000, or 100, or even 48, compounds is required. Surely, given
time, a chemist could name (especially with the aid of a
computer) all of the half million compounds within the scope of
the broadest claim, which claim is supported by the broad
disclosure. This does not constitute support for each compound
individually when separately claimed.
Id.
The appellants further contended that “[o]ther ‘guides’ allegedly
lead[] to chlorpropamide.” Id. According to the appellants, the specification
disclosed eleven processes for making the many compounds of the
invention, five of which employ an alkylamine at R(2) in the general
formula. Id. The court observed in that formula, “there is also the variable R
which may be hydrogen, chlorine, bromine, methyl, or methoxy and R(1)
which may be either chlorine or bromine.” Id. Together with some other
variables, the court concluded “[t]his makes for more than a few unidentified
possibilities not determined by the use of alkylamine alone.” Id.
To lead to the claimed compound, “R must be hydrogen and R(1)
must be chlorine and the alkylamine, R(2), must be propylamine.” Id. The
appellants argued that the “guide” became “more crystallized by the
recitation of the alkylamines which can be employed in the four or five
reactions described as using them.” Id. at 995. The appellants emphasized
n-propylamine as one of those amines recited. Id.
Rejecting this argument, the court first noted the specification listed
nineteen primary amines that may be used. The court explained that listing
propylamine with eighteen others does not add anything to “the initial
IPR2017-01712
Patent 8,815,830 B2

39
statement that one may use an alkyl amine containing from 2 to 6 carbon
atoms,” because “[p]ropylamine is such an amine but one is not led to it in
preference to the others merely by listing them all and identifying it, with
the others, by name.” Id. (emphasis added).
In the instant proceeding, formula I in NP2 encompasses a large
number of subgenera. In fact, Patent Owner acknowledges that even P1’s
claim 47 encompasses many subgenera. Tr. 29:1–3. Patent Owner argues P1
and NP2 identify 168 combinations of special interest. PO Resp. 45. But, to
provide adequate written description support for challenged claim 1’s single
subgenus, “reasonably specific supporting disclosure,” or, as the Ruschig
court put it, “something more than the disclosure of a class of 1000, or 100,
or even 48, [subgenera] is required.” 379 F.2d at 994. A class of 168
subgenera, even if that were the correct number, is squarely within the
Rushig’s illustrative range and is insufficient to provide reasonably specific
support for a single subgenus.
Patent Owner argues that halo is one of seven (or ten) alternatives at
R2, and hydroxy is one of eight (or ten) alternatives at R3. PO Resp. 35–46.
But merely listing them together with the others, without more, would not
guide an ordinarily skilled artisan to each specific substituent, let alone the
claimed combination.
In Fujikawa, the application disclosed compounds of a certain generic
structure with four variable groups (R, R0, R1, and R2), each of which could
be independently chosen from a list of functional groups. 93 F.3d at 1570.
The subgenus-at-issue “is directed to compounds of the above structure in
IPR2017-01712
Patent 8,815,830 B2

40
which R is cyclopropyl and R0 is 4–fluorophenyl.”9 Id. The Board of Patent
Appeals and Interferences found insufficient blaze marks for the subgenus.
Id. at 1570–71. The Federal Circuit affirmed. Id. at 1571.
There, the appellant pointed out that “with respect to practically every
position on the compound,” the claimed subgenus recited at least one of the
preferred choices. Id. With respect to position R, the appellant argued that
one of ordinary skill would have “substitute[d] cyclopropyl for isopropyl
because the two substituents are isosteric.” Id. The court rejected this
argument, noting that “[a]lthough, in hindsight, the substitution of
cyclopropyl for isopropyl might seem simple and foreseeable, [the]
disclosure provides no indication that position R would be a better candidate
for substitution than any other.” Id. Thus, the court concluded that the
disclosure would not lead an ordinarily skilled artisan to the subgenus-at-
issue. Id.
The facts in the instant proceeding are similar to those in Fujikawa.
Here, all but one of the exemplary compounds in P1 and NP2 contain
hydroxy, not halo, at R2. Like in Fujikawa, although “in hindsight,” the
substitution of halo for hydroxy “might seem simple and foreseeable,”
neither P1 nor NP2 provides any suggestion that halo would be a better
candidate than hydroxy, or any other substituents identified for R2 in P1’s
claims and NP2’s specific values.

9 There was no dispute over the substituents at R1 and R2. Fujikawa, 93 F.3d
at 1570.
IPR2017-01712
Patent 8,815,830 B2

41
Extending Ruschig’s metaphor, “it is easy to bypass a tree in the
forest, even one that lies close to the trail, unless the point at which one must
leave the trail to find the tree is well marked.” Fujikawa, 93 F.3d at 1571.
Here, although the specific values “do blaze a trail through the forest,” the
trail “runs close by” the tree, but “does not direct one to the . . . tree in
particular, and does not teach the point at which one should leave the trail to
find it.” Id.
Similarly, in Novozymes, the claim-at-issue recited three limitations,
each of which “is expressly stated in the disclosure” of a 2000 application.
723 F.3d at 1348. The Federal Circuit found “the supporting disclosure of
the 2000 application provides only generalized guidance,” but “contains no
disclosure of any variant that actually satisfies” the claim. Id. at 1346, 1348.
The court stated:
Taking each claim—as we must—as an integrated whole rather
than as a collection of independent limitations, one searches the
2000 application in vain for the disclosure of even a single
species that falls within the claims or for any “blaze marks” that
would lead an ordinarily skilled investigator toward such a
species among a slew of competing possibilities.
Id. at 1349.
We have the same situation here. It is undisputed that P1 and NP2
expressly disclose the substituents at each R position as claimed; yet, they do
not disclose challenged claim 1’s subgenus, or provide sufficient blaze
marks that would lead an ordinarily skilled artisan, among a myriad
possibilities, to such a subgenus. Similar to the appellant in Novozymes,
Patent Owner here works backward from a knowledge of challenged claim 1
by hindsight, and “seeks to derive written description support from an
IPR2017-01712
Patent 8,815,830 B2

42
amalgam of disclosures plucked selectively from” P1 and NP2. See id.
“[V]iewing the matter from the proper vantage point ‘of one with no
foreknowledge of the specific [subgenus],’” and taking challenged claim 1
as a whole rather than as the sum of its individual limitations, we find the
particular subgenus of challenged claim 1 lacks meaningful support in P1
and NP2. See id.
In sum, the disclosures of P1 and NP2 do not provide sufficient blaze
marks to guide an ordinarily skilled artisan through the forest of disclosed
possibilities toward the subgenus of challenged claim 1. Thus, NP2, with P1
incorporated by reference, does not provide adequate written description
support for challenged claim 1.
Regarding the challenged dependent claims, Petitioner argues
although they narrow the broad scope of the classes of R group substituents
in independent claim 1, “similar to P1, there are no ‘preferences’ or other
teachings in NP3/NP2 that would provide direction to a POSA, with no
foreknowledge of the later-claimed ’830 patent subgenus, to select the
narrowed R2, R3, R5, R6, and R7 groups” claimed therein. Pet. 66 (citing
Ex. 1011 ¶ 122), see also id. at 50 (citing Ex. 1011 ¶ 50) (discussing P1).
According to Petitioner, “[f]or similar reasons that ’830 patent claim 1 does
not have written description support in NP3/NP2, the dependent claims of
the ’830 patent also fail to have written description support in NP3/NP2.” Id.
at 66, see also id. at 50 (discussing P1).
Patent Owner contends that the additional limitations of the dependent
claims use “language drawn verbatim from additional P1 dependent claims
IPR2017-01712
Patent 8,815,830 B2

43
and corresponding NP2 ‘specific values,’” or are supported by NP2’s figures
or examples. PO Resp. 74–75. We are not persuaded.
We reiterate that in analyzing written description support, we must
take each claim “as an integrated whole rather than as a collection of
independent limitations.” Novozymes, 723 F.3d at 1349. The challenged
dependent claims require picking and choosing multiple substituents, not
just the one further limited in those claims. Thus, we agree with Petitioner,
and determine, for similar reasons explained above, NP2, with P1
incorporated by reference, does not provide adequate written description
support for challenged claims 2–9, 11–21, and 23–28.
In sum, because NP2, with P1 incorporated by reference, does not
provide sufficient written description support for the challenged claims, the
priority date of those claims is no earlier than March 28, 2014, the filing date
of the application that issued as the ’830 patent. Thus, Sofia qualifies as
prior art and anticipates those claims.
CONCLUSION10
After reviewing the entire record and weighing evidence offered by
both parties, we determine that Petitioner has demonstrated by a

10 Should Patent Owner wish to pursue amendment of the challenged claims
in a reissue or reexamination proceeding subsequent to the issuance of this
decision, we draw Patent Owner’s attention to the April 2019 Notice
Regarding Options for Amendments by Patent Owner Through Reissue or
Reexamination During a Pending AIA Trial Proceeding. See 84 Fed. Reg.
16,654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue application
or a request for reexamination of the challenged patent, we remind Patent
IPR2017-01712
Patent 8,815,830 B2

44
preponderance of the evidence that Sofia anticipates claims 1–9, 11–21, and
23–28 of the ’830 patent.
In summary:
ORDER
In consideration of the foregoing, it is hereby:
ORDERED that claims 1–9, 11–21, and 23–28 of the ’830 patent are
held unpatentable;
FURTHER ORDERED that, because this is a Final Written Decision,
parties to this proceeding seeking judicial review of our Decision must
comply with the notice and service requirements of 37 C.F.R. § 90.2.

Owner of its continuing obligation to notify the Board of any such related
matters in updated mandatory notices. See 37 C.F.R. § 42.8(a)(3), (b)(2).
Claims

35
U.S.C. §
Reference Claims
Shown
Unpatentable
Claims Not
Shown
Unpatentable
1–9, 11–21,
23–28
102 Sofia 1–9, 11–21,
23–28

Overall
Outcome
1–9, 11–21,
23–28

IPR2017-01712
Patent 8,815,830 B2

45
FOR PETITIONER:

Alicia Russo
Frederick Millett
Robert S. Schwartz
Stephen K. Yam
VENABLE LLP
arusso@venable.com
Gilead830ipr@venable.com
RSchwartz@Venable.com

FOR PATENT OWNER:

Edward Gates
Richard Giunta
Gerald Hrycyszyn
WOLF, GREENFIELD & SACKS, P.C.
egates-ptab@wolfgreenfield.com
Rgiunta-ptab@wolfgreenfield.com
Ghrycyszyn-ptab@wolfgreenfield.com