Novartis AG v. Torrent Pharmaceuticals Ltd.

United States Court of Appeals

for the Federal Circuit

______________________

NOVARTIS AG, MITSUBISHI PHARMA CORP.,

Appellants

v.

TORRENT PHARMACEUTICALS LIMITED,

APOTEX INC., MYLAN PHARMACEUTICALS INC.,

Appellees

______________________

2016-1352

______________________

Appeal from the United States Patent and Trademark

Office, Patent Trial and Appeal Board in Nos. IPR2014-

00784, IPR2015-00518.

______________________

Decided: April 12, 2017

______________________

ROBERT TRENCHARD, Gibson, Dunn & Crutcher LLP,

New York, NY, argued for appellants. Appellant Novartis

AG also represented by JANE M. LOVE; MICHAEL A. VALEK,

Dallas, TX; ALEXANDER N. HARRIS, San Francisco, CA.

JOSEPH M. O'MALLEY, JR., Paul Hastings LLP, New

York, NY, for appellant Mitsubishi Pharma Corp. Also

represented by ERIC WILLIAM DITTMANN.

2 NOVARTIS AG v. TORRENT PHARMACEUTICALS

TERESA STANEK REA, Crowell & Moring, LLP, Wash-

ington, DC, argued for appellees. Appellee Apotex Inc.

also represented by VINCENT JOHN GALLUZZO; JONATHAN

M. LINDSAY, Irvine, CA.

MICHAEL K. LEVY, Andrews Kurth Kenyon LLP, New

York, NY, for appellee Torrent Pharmaceuticals Limited.

SHANNON BLOODWORTH, Perkins Coie, LLP, Washing-

ton, DC, for appellee Mylan Pharmaceuticals Inc. Also

represented by BRANDON MICHAEL WHITE; DAN L.

BAGATELL, Hanover, NH.

______________________

Before TARANTO, CHEN, and STOLL, Circuit Judges.

CHEN, Circuit Judge.

This is an appeal from the Final Written Decision of

the United States Patent and Trademark Office, Patent

Trial and Appeal Board (Board) in two consolidated inter

partes review (IPR) proceedings of U.S. Patent No.

8,324,283 (the ’283 patent), owned by Novartis AG and

Mistubishi Tanabe Pharma Corp. (collectively, Novartis).

The Board instituted IPRs on all claims of the ’283 patent

based on petitions filed by Torrent Pharmaceuticals

Limited, Apotex, Inc. and Mylan Pharmaceuticals Inc.

(collectively, Petitioners). After reviewing the claims,

receiving extensive briefing, and hearing oral argument,

the Board found all original claims of the ’283 patent and

Novartis’ proposed substitute claims unpatentable as

obvious. See Torrent Pharm. Ltd. v. Novartis AG, Nos.

IPR2014-00784, IPR2015-00518, 2015 WL 5719630

(PTAB Sept. 24, 2015) (Final Written Decision). Novartis

raises a series of challenges to the Board’s analysis of the

evidence and ultimate determination of unpatentability.

For the reasons stated below, we affirm.

NOVARTIS AG v. TORRENT PHARMACEUTICALS 3

BACKGROUND

I.

The ’283 patent relates to a solid pharmaceutical

composition suitable for oral administration, comprising a

sphingosine-1 phosphate (S1P) receptor agonist and a

sugar alcohol, which the patent explains is useful for the

treatment of certain autoimmune diseases such as multi-

ple sclerosis. ’283 patent, col. 1, lines 11–14, 33–35; col.

12, lines 19–49. According to the specification, S1P

receptor agonists generally exhibit properties that make

formulations suitable for oral administration of a solid

composition difficult to create. However, “solid composi-

tions comprising a sugar alcohol provide formulations

which are particularly well suited to the oral administra-

tion of S1P receptor agonists.” See id. at col. 1, lines 36–

39. They also “provide a convenient means of systemic

administration of S1P receptor agonists, do not suffer

from the disadvantages of liquid formulations for injection

or oral use, and have good physiocochemical and storage

properties.” Id. at col. 1, lines 39–43. In such a composi-

tion, the S1P receptor agonist is the active ingredient and

the sugar alcohol acts as an excipient—the substance

formulated alongside the active ingredient as a diluent,

carrier, filler and/or bulking agent for the composition.

See id. at col. 9, lines 53–54.

The ’283 patent states that there are multiple known

S1P receptor agonists appropriate for use in the claimed

invention, set forth in the specification as formulas I–XIII.

Id. at col. 1, line 51 to col. 8, line 4. The ’283 patent also

states that a “particularly preferred S1P receptor agonist

of formula I is FTY720, i.e., 2-amino-2-[2-(4-octylphenyl)

ethyl]propane-1,3-diol in free form or in a pharmaceutical-

ly acceptable salt form . . . .” Id. at col. 8, lines 23–26.

FTY720 is also known as fingolimod. The ’283 patent

further discloses that the specific sugar alcohol used in

the claimed composition “may suitably be mannitol,”

4 NOVARTIS AG v. TORRENT PHARMACEUTICALS

because of its non-hygroscopic properties (i.e., it is not

likely to absorb moisture, which is beneficial in manufac-

turing solid oral pills). Id. at col. 9, lines 53–54.

Claims 1 and 19 of the ’283 patent are the only inde-

pendent claims and are illustrative of the claimed subject

matter:

1. A solid pharmaceutical composition suitable for

oral administration, comprising:

(a) a S1P receptor agonist which is select-

ed from 2-amino-2-[2-(4-octylpheny

l)ethyl]propane-1,3-diol, 2-amino-2-[4-(3-

benzyloxyphenoxy)-2-chloropheny

l]propyl-1,3-propane-diol, 2-amino-2-[4-(3-

benzyloxyphenylthio)-2-

chlorophenyl]propyl-1,3-propane-diol, or 2-

amino-2-[4-(3-benzyloxyphenylthio)-2-

chlorophenyl]-2-ethyl-1,3-propane-diol,

and its phosphates or a pharmaceutically

acceptable salt thereof; and

(b) a sugar alcohol.

19. A solid pharmaceutical composition suitable

for oral administration, comprising mannitol and

2-amino-2-[2-(4-octylphenyl)ethyl]propane-1,3-diol

or a pharmaceutically acceptable salt thereof.

Id. at col. 17, lines 2–11; col. 18, lines 7–10. Thus, claim 1

is directed towards a solid oral composition comprised of

the combination of one of a handful of S1P receptor ago-

nists and any sugar alcohol, whereas claim 19 is directed

towards the specific combination of fingolimod and man-

nitol in a solid oral composition.

The dependent claims are directed towards various re-

finements of the composition, including for example, the

addition of a lubricant:

NOVARTIS AG v. TORRENT PHARMACEUTICALS 5

20. A composition according to claim 19, further

comprising a lubricant.

Id. at col. 18, lines 11–12. Other claims are directed

towards adjusting the respective amount of ingredients:

22. A composition according to claim 19, wherein

the compound 2-amino-2-[2-(4-

octylphenyl)ethyl]propane-1,3-diol, or a pharma-

ceutically acceptable salt thereof, is present in an

amount of 0.5 to 5% by weight, based on the total

weight of the composition.

23. A composition according to claim 19, wherein

mannitol is present in an amount of 90 to 99.5%

by weight, based on the total weight of the compo-

sition.

Id. at col. 18, lines 15–22.

While the application leading to the ’283 patent was

pending at the Patent Office, Novartis applied to the U.S.

Food and Drug Administration (FDA) for approval to sell

a fingolimod-mannitol pill to treat multiple sclerosis

under the “Gilenya” brand name. The FDA approved

Gilenya for the treatment of multiple sclerosis in 2010.

II.

On May 27, 2014, Torrent filed a petition to institute

an inter partes review of claims 1–32 of the ’283 patent.

Torrent’s petition presented three separate patentability

challenges:

1. claims 1–32 are unpatentable as obvious over

the combination of U.S. Patent No. 6,004,565

(Chiba) and Pharmaceutics: The Science of Dosage

Form Design (Aulton); and

2. claims 1–4, 7, 8, 19, 22 and 32 are unpatenta-

ble as anticipated by U.S. Patent No. 6,277,888

(Sakai); and

6 NOVARTIS AG v. TORRENT PHARMACEUTICALS

3. claims 1–32 are unpatentable as obvious over

Chiba and Sakai.

Chiba teaches the use of immunosuppressive com-

pounds with fingolimod as the preferred species. J.A.

18442. 1 Chiba also teaches that these immunosuppres-

sive compounds are useful for treating “autoimmune

diseases such as . . . multiple sclerosis,” among other

diseases and conditions. J.A. 18443. Chiba goes on to

disclose oral administration of fingolimod, including

“admix[ing] with [a] carrier, excipient, diluent, and so on

and formulat[ion] into . . . capsules [or] tablets . . . for

administering to patients.” J.A. 18444. In discussing the

preparation of these capsules and tablets for oral admin-

istration of fingolimod, Chiba teaches that “pharmaceuti-

cally or physiologically acceptable carriers or excipients

for use with the . . . compounds noted herein are known in

the art or can be readily found by methods and tests

known in the art.” J.A. 18446. In other words, Chiba

teaches a solid oral composition of fingolimod combined

with a generic excipient.

Aulton teaches the use of tablets and capsules to ad-

minister drugs orally. J.A. 19041. It specifically teaches

that “[t]he successful formulation of a stable and effective

solid dosage form depends on the careful selection of the

excipients which are added to facilitate administration,

promote the consistent release and bioavailability of the

drug and protect it from degradation.” J.A. 19066–167.

Aulton recommends mannitol as a common diluent used

in “[t]ableting by the wet granulation process,” which

Aulton describes as “the most widely used method for

pharmaceutical materials.” J.A. 19074–77. Aulton de-

scribes mannitol as “expensive,” but “commonly used” as

an excipient in solid oral compositions. J.A. 19077.

1 Citations to “J.A. ____” refer to the Joint Appendix

filed by the parties.

NOVARTIS AG v. TORRENT PHARMACEUTICALS 7

Sakai describes a pharmaceutical composition con-

taining fingolimod as an active ingredient. J.A. 18421.

More particularly, Sakai discloses that the composition

can be formulated into a liquid preparation, or can be a

solid lyophilized (freeze-dried) product. J.A. 18421–22.

Sakai further discloses that the addition of a saccharide,

such as sugar alcohol, to the composition can result in a

less irritating resulting liquid solution. J.A. 18421. Sakai

discloses a list of eight exemplary saccharides, including

mannitol. J.A. 18422. The saccharide, such as mannitol,

can be dissolved in the liquid for dissolution, or alterna-

tively may be contained in the lyophilized product along

with the active ingredient. J.A. 18422. Sakai teaches

that this liquid pharmaceutical composition can be used

for immunosuppression in connection with organ or bone

marrow transplantation, autoimmune diseases, or allergic

diseases. Id. In short, Sakai teaches the specific combi-

nation of fingolimod and mannitol for a liquid formula-

tion.

III.

On December 1, 2014, the Board granted in part Tor-

rent’s petition and instituted trial to review patentability

of the challenged claims in IPR2014-00784. Specifically,

the Board instituted on the first ground, the combination

of Chiba and Aulton, but declined to institute on grounds

two or three. The Board found that Chiba discloses the

use of fingolimod in a solid formulation for oral admin-

istration when combined with conventional excipients. It

then found that Aulton teaches the use of mannitol as a

conventional excipient that a person of skill in the art

would have looked to when formulating a solid composi-

tion with fingolimod.

The Board found Sakai to be an improper anticipatory

reference because the reference does not describe a solid

composition suitable for oral administration. It then

rejected the grounds predicated on the combination of

8 NOVARTIS AG v. TORRENT PHARMACEUTICALS

Chiba and Sakai for similar reasons, noting that, unlike

Aulton, “Sakai does not identify mannitol as a ‘conven-

tional excipient’ in solid pharmaceutical compositions,

and Sakai’s stated reasons for using mannitol in liquid

pharmaceutical compositions are inapplicable to its

potential use in connection with solid pharmaceutical

compositions.” J.A. 72.

Apotex and Mylan thereafter filed a separate petition

seeking to institute an IPR of claims 1–32 of the ’283

patent based on the already-instituted Chiba/Aulton

grounds and requested joinder with the Torrent proceed-

ings. On February 17, 2015, the Board instituted trial in

this follow-on proceeding in IPR2015-00518 and joined it

with the Torrent proceeding.

After briefing and oral argument, the Board issued its

Final Written Decision in the consolidated proceeding.

The Board concluded that Chiba and Aulton collectively

teach each limitation of claims 1–32 of the ’283 patent. It

first addressed claim 19, directed towards the specific

combination of fingolimod and mannitol. The Board

found that Chiba and Aulton together strongly suggested

the claimed two-ingredient combination:

First, Chiba teaches that a person of ordinary

skill in the art would have been able to identify or

easily determine excipients that would have been

compatible with fingolimod . . . (“pharmaceutically

or physiologically acceptable carriers or excipients

for use with the . . . compounds noted herein are

known in the art or can be readily found by meth-

ods and tests known in the art”). Second, Aulton

teaches that mannitol is not only a known diluent

for direct compression manufacturing, but also

commonly used in wet granulation, which Aulton

teaches is “the most widely used method for

pharmaceutical materials.” . . . This combination

of teachings already strongly suggests that man-

NOVARTIS AG v. TORRENT PHARMACEUTICALS 9

nitol likely would have been a target of investiga-

tion for a person of ordinary skill in the art inter-

ested in finding an excipient compatible with

fingolimod . . . .

Final Written Decision, 2015 WL 5719630, at *8. After

finding that the two references themselves strongly

suggested the claimed invention, the Board expressly

found “additional evidence of the reason to combine

fingolimod and mannitol.” Id. First, the Board noted that

Sakai “directly instructs that the two ingredients should

be combined.” Id. Although Novartis argued in its briefs

below that Sakai’s teaching is narrowly limited to liquid-

phase pharmaceutical compositions, as opposed to the

claimed solid oral dosage forms, the Board observed that

Novartis’ own expert, Dr. Stephen Byrn, had written an

article describing how “solution studies can be very help-

ful” in understanding drug degradations in the solid state.

Id. Despite Novartis’ attempt to minimize the article’s

meaning, the Board concluded that “a suggestion to

combine ingredients in the liquid phase would have been

relevant to the determination of a person of ordinary skill

in the art to combine the same ingredients in the solid

phase.” Id.

Acknowledging that it had denied instituting the IPR

based on Sakai alone (per § 102) or in combination with

Chiba (per § 103), the Board distinguished its final deci-

sion’s usage of Sakai, explaining that its final decision

simply relied on Sakai as a background reference that

offered additional motivation evidence to combine Chiba

with Aulton. The Board explained that even though

Sakai did not “teach that mannitol is a conventional

excipient for use in solid pharmaceutical compositions,”

id., the record evidence relating to Dr. Byrn’s article,

which was debated by the parties, supported a finding

that “Sakai’s teaching would have been relevant to the

decision on which excipient to use in formulating a solid

oral dosage form of fingolimod.” Id.

10 NOVARTIS AG v. TORRENT PHARMACEUTICALS

The Board went on to find that several additional

background references in the proceeding demonstrate

that mannitol provides advantages when used as a dilu-

ent in tableting, further supporting a reason to combine.

The Board concluded its motivation to combine analysis:

Given (1) the knowledge in the art that mannitol

provided advantages in formulating tablets gen-

erally, (2) Chiba’s teaching that a person of ordi-

nary skill in the art would have been able to

identify or easily determine excipients that would

have been compatible with fingolimod, (3) Aulton’s

teaching that mannitol was a diluent commonly

used in the most common form of pharmaceutical

manufacture, (4) Sakai’s teaching that mannitol

and fingolimod should be combined in the liquid

phase, and (5) Dr. Byrn’s statement that liquid-

phase compatibility was relevant to the prediction

of solid-phase compatibility, we conclude that Pe-

titioners have shown a reason to combine the

teachings of Chiba and Aulton.

Id. at *9.

The Board next turned to the objective indicia of non-

obviousness. First, it found that independent claims 1

and 19 were “not commensurate in scope” with the pur-

ported unexpected result of fingolimod’s low concentration

stability when combined with mannitol, because the

independent claims are “not limited to any particular dose

or dose range of fingolimod.” Id. at *10. Therefore, the

Board concluded, “even if the stability of the mannitol-

fingolimod combination at low doses was unexpected, it is

insufficient to support a legally significant finding of

unexpected results.” Id. at *11. The Board also rejected

Novartis’ long-felt but unsolved need, industry praise, and

commercial success arguments because all of Novartis’

proffered evidence was directed solely toward the fact that

NOVARTIS AG v. TORRENT PHARMACEUTICALS 11

Gilenya was a solid oral multiple sclerosis treatment,

which was already known in the prior art.

The Board then analyzed the dependent claims in

turn. Relevant to this appeal, the Board turned to de-

pendent claims 8, 10, 22, and 23, and proposed amended

claims 40, 42, 54, and 55, directed towards concentrations

with low percentages of fingolimod by weight. The Board

found that “Petitioners provide evidence that the selection

of the relative amounts of the constituents of the claimed

formulation is the result of routine optimization.” Id. at

*16. It further noted, “[w]e have not been directed to any

evidence of record contradicting this evidence, so we find

that a person of ordinary skill in the art familiar with

Chiba and Aulton would have been able and motivated to

optimize the amount of fingolimod . . .” Id.

In conclusion, the Board held every claim unpatenta-

ble as obvious and denied Novartis’ motion to amend for

essentially the same reasons it rejected the original

claims. Appellants timely appealed. We have jurisdiction

under 28 U.S.C. § 1295(a)(4)(A).

DISCUSSION

We review Board decisions using the standard set

forth in the Administrative Procedure Act (APA), 5 U.S.C.

§ 706. In re Sullivan, 362 F.3d 1324, 1326 (Fed. Cir.

2004) (citing Dickinson v. Zurko, 527 U.S. 150, 154

(1999)); see also Belden Inc. v. Berk–Tek LLC, 805 F.3d

1064, 1080 (Fed. Cir. 2015). Under the APA, we must

“hold unlawful and set aside agency action . . . not in

accordance with law [or] . . . without observance of proce-

dure required by law.” 5 U.S.C. § 706.

We review the Board’s legal conclusions de novo but

review for substantial evidence any underlying factual

determinations. See Nike, Inc. v. Adidas AG, 812 F.3d

1326, 1332 (Fed. Cir. 2016); In re Giannelli, 739 F.3d

1375, 1378–79 (Fed. Cir. 2014). Substantial evidence is

12 NOVARTIS AG v. TORRENT PHARMACEUTICALS

“such relevant evidence as a reasonable mind might

accept as adequate to support a conclusion.” Consol.

Edison Co. v. NLRB, 305 U.S. 197, 229 (1938); see In re

Applied Materials, Inc., 692 F.3d 1289, 1294 (Fed. Cir.

2012).

On appeal, Novartis first contends that the Board vio-

lated the APA when it relied on Sakai in the Final Writ-

ten Decision without affording Novartis proper notice and

a chance to be heard. Novartis goes on to argue that the

Board also erred on the merits, specifically in its analysis

of the motivation to combine evidence and in its treat-

ment of the alleged objective indicia of nonobviousness.

I. APA Due Process

We first turn to Novartis’ argument that the Board

violated the requirements of notice and an opportunity to

respond found in the APA when it used the Sakai refer-

ence as part of its motivation to combine analysis in the

Final Written Decision. According to Novartis, the Board

ruled Sakai entirely out of the case in the Institution

Decision, and on that basis, denied institution of the two

proposed grounds based on Sakai. Novartis contends that

it relied on that ruling and consequently submitted a

“vastly different” record than it would have if it had

known Sakai was still a live issue.

In a formal adjudication, such as an IPR, the APA im-

poses certain procedural requirements on the agency.

The Patent and Trademark Office, including the Board,

must provide the patent owner with timely notice of “the

matters of fact and law asserted,” and an opportunity to

submit facts and argument. 5 U.S.C. §§ 554(b)–(c), 557(c);

Dell Inc. v. Acceleron, LLC, 818 F.3d 1293, 1301 (Fed. Cir.

2016). The notice and opportunity to be heard provisions

of the APA have been applied “to mean that ‘an agency

may not change theories in midstream without giving

respondents reasonable notice of the change’ and ‘the

opportunity to present argument under the new theory.’”

NOVARTIS AG v. TORRENT PHARMACEUTICALS 13

Belden, 805 F.3d at 1080 (quoting Rodale Press, Inc. v.

FTC, 407 F.2d 1252, 1256–57 (D.C. Cir. 1968)). In this

case we conclude that the relevant APA provisions were

satisfied.

A.

We first disagree with Novartis that the Board ruled

Sakai out of the case entirely in the Institution Decision.

In the Institution Decision, the Board declined to read

Sakai as an anticipatory reference or primary obviousness

reference because Sakai does not disclose “mannitol as a

‘conventional excipient’ in solid pharmaceutical composi-

tions, and Sakai’s stated reasons for using mannitol in

liquid pharmaceutical compositions are inapplicable to its

potential use in connection with solid pharmaceutical

compositions.” J.A. 72. In other words, although Sakai

discloses the combination of fingolimod and mannitol, it

does not expressly disclose the combination in a solid

pharmaceutical composition nor does its teaching of a

liquid composition necessarily translate to a solid oral

composition.

This conclusion, however, is not contrary to the

Board’s discussion of Sakai in the Final Written Decision

that Sakai’s teachings would have nevertheless been

relevant to one of skill in the art in deciding which excipi-

ents to use in formulating a solid oral dosage form of

fingolimod. Having already found that Chiba and Aulton

strongly suggest the combination of fingolimod and man-

nitol in a solid oral composition, the Board found that

Sakai merely reinforced its finding that the person of

ordinary skill in the art would have expected mannitol to

be compatible with fingolimod because Sakai discloses a

stable combination of these two ingredients suitable for

long-term preservation. The Board’s discussion of Sakai

in the Final Written Decision was not inconsistent with

its review of Sakai in the Institution Decision.

14 NOVARTIS AG v. TORRENT PHARMACEUTICALS

B.

We also reject as unfounded Novartis’ complaints of

“surprise” and contention that, following the Institution

Decision, the parties “paid Sakai scant attention in sub-

sequent proceedings.” The parties debated Sakai at

length throughout the proceeding and in the same context

that it was discussed by the Board in the Final Written

Decision.

As an initial matter, we note that in addition to as-

serting Sakai as a primary reference, Torrent’s petition

also argued that several references, including Sakai,

further support the motivation to combine the teachings

of Chiba and Aulton. Specifically, Torrent argued in

connection with the combination of Chiba and Aulton that

“Sakai (Ex. 1005) reinforced the expectation to the ordi-

narily-skilled artisan that mannitol would have been

compatible with FTY720 [fingolimod] because Sakai

discloses pharmaceutical injectable compositions contain-

ing FTY720 [fingolimod] and mannitol in solution, as well

as lyophilized product meant for long-term preservation

in vials containing FTY720 [fingolimod] and mannitol.”

J.A. 6832. And in support of their petition, Apotex and

Mylan also explained that Sakai would direct the person

of ordinary skill in the art to the combined teachings of

Chiba with Aulton. It reiterated the argument raised in

the Torrent petition that the ordinarily skilled artisan

would have naturally considered mannitol because of its

known compatibility with fingolimod, again citing Sakai’s

disclosure of a stable composition comprised of these two

ingredients.

Following institution of the Apotex/Mylan proceeding

and joinder with the Torrent proceeding, the relevance of

Sakai’s compatibility-disclosure to support a motivation to

combine Chiba and Aulton was an ongoing, debated issue

that Novartis addressed directly, on multiple occasions.

In its Patent Owner’s Response, Novartis specifically

NOVARTIS AG v. TORRENT PHARMACEUTICALS 15

argued that Petitioners’ reliance on Sakai’s stability-

disclosure in connection with the motivation to combine

inquiry lacked merit because Sakai is relevant only to

liquid compositions. Petitioners continued to press the

issue in their Reply, contending that Sakai “would have

provided a [person of skill in the art] with a reasonable

expectation that mannitol is compatible with fingolimod.”

J.A. 7782.

Furthermore, both Petitioners’ expert and Novartis’

expert went into significant detail in their post-institution

declarations discussing Sakai and its applicability to the

motivation to combine inquiry. Novartis’ counsel then

questioned Petitioners’ expert at length about Sakai. And

Novartis’ submitted Observations on Cross Examination

repeatedly explained why Sakai did not support Petition-

ers’ motivation to combine argument. At the hearing,

both parties submitted demonstrative slides dedicated to

Sakai and spent considerable attention discussing Sakai’s

relevance as a background reference supporting the

motivation to utilize mannitol with fingolimod in an oral

formulation. Based on this record, it is quite clear that

Novartis had more than sufficient notice and opportunity

to be heard on Sakai’s potential relevance, and in fact

actively and repeatedly attempted to distinguish Sakai to

defeat the very argument relied on by the Board in the

Final Written Decision.

In sum, we reject Novartis’ contention to this court

that it believed Sakai was not at issue in the proceeding. 2

For this reason we reject Novartis’ APA challenge.

2 Indeed, had Novartis believed the Board eliminat-

ed Sakai from the proceeding, it had various procedural

mechanisms at its disposal to respond to any perceived

impropriety with Petitioners’ continued reliance on the

reference. In particular, Novartis could have moved to

16 NOVARTIS AG v. TORRENT PHARMACEUTICALS

C.

Finding no APA violation for the reasons discussed

above, we nevertheless also reject Novartis’ characteriza-

tion of Sakai as the “missing link” in the Board’s obvious-

ness analysis. Contrary to Novartis’ contention, Sakai

was discussed by the Board as one of several independent

grounds supporting the motivation to combine fingolimod

and mannitol in a solid oral composition. In finding a

motivation to combine, the Board explained that the

teachings of Chiba and Aulton alone “already strongly

exclude the Sakai reference. See Genzyme Therapeutic

Prods. v. Biomarin Pharm. Inc., 825 F.3d 1360, 1368 (Fed.

Cir. 2016). We find meritless Novartis’ argument that it

did in fact move to exclude Sakai from the proceeding.

See Oral Arg. at 53:30–53:51: available at

http://oralarguments.cafc.uscourts.gov/default.aspx?fl=20

16-1352.mp3. Although not provided in the Joint Appen-

dix, Novartis’ counsel invited the court to review its

motion to exclude. That invitation, unfortunately, led the

court on a road to nowhere. In its motion, Novartis moved

to exclude over fifty exhibits, including Sakai, all identi-

fied by exhibit number only and listed in one long string

cite, based on one conclusory sentence: “Petitioners rely

on numerous exhibits that are incomplete and/or irrele-

vant to the sole issue for review identified by the Board –

i.e., (non)obviousness of the ’283 Patent in light of Chiba

over Aulton).” Patent Owner’s Motion to Exclude at 20,

Paper No. 73. This superficial treatment amounts to little

more than a request that the Board peruse the cited

evidence and piece together a coherent argument on

Novartis’ behalf. It is far from sufficient to raise a mean-

ingful challenge to any of the several dozen exhibits, let

alone to sensitize the Board to the complained-of use of

Sakai in particular.

NOVARTIS AG v. TORRENT PHARMACEUTICALS 17

suggests that mannitol likely would have been a target of

investigation for a person of ordinary skill in the art

interested in finding an excipient compatible with fin-

golimod.” Final Written Decision, 2015 WL 5719630, at

*8.

Nevertheless, the Board continued to bolster its anal-

ysis with “additional evidence of the reason to combine

fingolimod and mannitol.” Id. And Sakai’s teaching to

combine fingolimod and mannitol was just one of those

additional reasons. The Board further explained that

“[i]n addition to the direct teaching in Sakai that manni-

tol and fingolimod should be combined, several documents

that would have been known to a person of ordinary skill

in the art teach that mannitol provides advantages when

used as a diluent in tableting.” Id. at *9. The Board went

on to explain that these references—all unchallenged on

appeal—describe known advantages of using mannitol as

an excipient in solid oral compositions that “provide a

strong reason to combine Chiba’s teaching of a solid oral

dosage form of fingolimod and Aulton’s teaching of manni-

tol as an excipient for making solid oral dosage forms.”

Id. These additional references are also substantial

evidence supporting the Board’s motivation to combine

conclusion, independent of Sakai. This is not a case

where Sakai provided the linchpin of the Board’s analysis,

as Novartis contends.

For all these reasons, we find no violation of the APA

with respect to the Board’s discussion of Sakai in the

Final Written Decision.

II. OBVIOUSNESS

We turn to Novartis’ remaining challenges to the

Board’s obviousness analysis.

Obviousness is a mixed question of fact and law. The

Board’s ultimate conclusion that the claims are not obvi-

ous is a legal determination subject to de novo review,

18 NOVARTIS AG v. TORRENT PHARMACEUTICALS

however, the subsidiary factual findings are reviewed for

substantial evidence. In re Gartside, 203 F.3d 1305, 1316

(Fed. Cir. 2000). Motivation to combine is one of those

underlying factual issues. Id. (“The presence or absence

of a motivation to combine references in an obviousness

determination is a pure question of fact.”). Whether

objective indicia support a finding of nonobviousness is

also a factual question. Merck & Cie v. Gnosis S.P.A., 808

F.3d 829, 833 (Fed. Cir. 2015).

A. Motivation to Combine

Novartis argues that the Board further erred in its

motivation to combine analysis because it failed to read

the prior art as a whole and overlooked critical evidence of

mannitol’s known disadvantages as an excipient for solid

compositions. In particular, Novartis argues that it

pointed out mannitol’s negative properties, including

difficulty to manufacture, the existence of impurities, and

expense. Because the Board did not expressly state that

it was weighing all of these negatives against mannitol’s

positives, Novartis contends that the Board’s motivation

to combine analysis was legally flawed. In support of its

contention, Novartis directs the court to Medichem, S.A.

v. Rolabo, S.L., 437 F.3d 1157 (Fed. Cir. 2006). In Medi-

chem, this court explained that “[w]here the prior art

contains ‘apparently conflicting’ teachings (i.e., where

some references teach the combination and others teach

away from it) each reference must be considered ‘for its

power to suggest solutions to an artisan of ordinary skill .

. . consider[ing] the degree to which one reference might

accurately discredit another.’” Id. at 1165 (quoting In re

Young, 927 F.2d 588, 591 (Fed. Cir. 1991)).

Contrary to Novartis’ contention, the record reflects

that the Board considered Novartis’ arguments regarding

motivation to combine, weighed them against the compet-

ing evidence and argument, and concluded that despite

Novartis’ contentions, one of skill in the art would have

NOVARTIS AG v. TORRENT PHARMACEUTICALS 19

been motivated to combine fingolimod with mannitol in a

solid composition. Indeed, the Board expressly discussed

one of mannitol’s negative properties in the Final Written

Decision—its expense—but noted that, despite this poten-

tially discouraging characteristic, it was still “commonly

used.” Final Written Decision, 2015 WL 5719630, at *5.

And it went on to cite the portion of the Patent Owner’s

Response discussing the arguments Novartis highlights

on appeal when rejecting Novartis’ teaching-away argu-

ment.

Moreover, the Board’s consideration of mannitol’s

negative properties in the Final Written Decision was at

least commensurate with Novartis’ presentation of those

issues to the Board in its Patent Owner Response. In a

lengthy brief, Novartis’ discussion was relegated to one

passing, unsupported sentence, stating that “[w]hile

mannitol has some positive properties, it also has nega-

tive ones, including expense, poor machinability and

possible impurities.” J.A. 7354. Novartis did not direct

the Board to the expert declarations it now highlights on

appeal, nor did it direct the Board to any record evidence

at all. And there is no indication in the record that No-

vartis elsewhere meaningfully advanced these suggested

negatives or developed them in such a fashion as to

necessarily overcome the numerous advantages of manni-

tol identified by Petitioners and discussed in the Final

Written Decision. Thus, we are not persuaded that No-

vartis presented its arguments against the use of manni-

tol in such a way that it would be appropriate to find fault

in the Board’s arguably limited treatment of those argu-

ments in the Final Written Decision.

This court’s discussion in Medichem does not change

our conclusion. Although the court there stated that prior

art must be considered as a whole and the disadvantages

of a reference must be considered in addition to the bene-

fits, 437 F.3d at 1165, there is no requirement that the

Board expressly discuss each and every negative and

20 NOVARTIS AG v. TORRENT PHARMACEUTICALS

positive piece of evidence lurking in the record to evaluate

a cursory argument. In addition, this court has said on

multiple occasions that failure to explicitly discuss every

issue or every piece of evidence does not alone establish

that the tribunal did not consider it. See, e.g., Carolina

Tobacco Co. v. Bureau of Customs & Border Prot., 402

F.3d 1345, 1350 (Fed. Cir. 2005) (“[T]he failure of Cus-

toms to explicitly discuss the six factors when it initially

increased Carolina’s bond does not establish that it did

not consider them.”); Lab. Corp. of Am. Holdings v. Chi-

ron Corp., 384 F.3d 1326, 1332 (Fed. Cir. 2004) (stating

that a district court’s failure to discuss an issue does not

necessarily establish that the court did not consider it);

Charles G. Williams Const., Inc. v. White, 326 F.3d, 1376,

1380 (Fed. Cir. 2003) (“The Board’s failure to discuss the

evidence upon which Williams relies does not mean that it

did not consider it”). The Board is “not require[d] . . . to

address every argument raised by a party or explain

every possible reason supporting its conclusion.” Synop-

sys, Inc. v. Mentor Graphics Corp., 814 F.3d 1309, 1322

(Fed. Cir. 2016). Here, given that the Board cited to the

relevant pages of Novartis’ Patent Owner Response, we

find no reason to assume the Board failed to consider

mannitol’s cited negatives simply because they were not

recited at length in the Board’s Final Written Decision.

Having dispatched Novartis’ numerous procedural ar-

guments, we ask finally whether substantial evidence

supports the Board’s finding on the motivation to combine

Chiba and Aulton. Here, we conclude that substantial

evidence supports the Board’s finding that, despite man-

nitol’s potentially negative characteristics, it was never-

theless a valid consideration as an excipient for solid oral

pharmaceuticals and a person of skill in the art would

have been motivated to combine fingolimod and mannitol

in the manner claimed by the ’283 patent. Indeed, the

Board cites to multiple pieces of evidence establishing

mannitol as one of a handful of excipients used in solid

NOVARTIS AG v. TORRENT PHARMACEUTICALS 21

oral compositions and its primacy as a non-hygroscopic

and compressible diluent which makes it particularly

valuable in tableting.

We, therefore, find no legal error in the Board’s

treatment of the motivation to combine evidence nor do

we find a lack of substantial evidence supporting its

conclusion.

B. Objective Indicia of Nonboviousness

Novartis next argues that the Board erred in its as-

sessment of the various objective indicia of nonobvious-

ness. We address each argument in turn.

i. Unexpected Results

According to Novartis, the Board erred when it

grouped several dependent claims with their independent

claims when considering Novartis’ unexpected results

evidence. Novartis argues that it presented persuasive

evidence to the Board that the combination of fingolimod

and mannitol solved the problem of fingolimod’s unex-

pected low dose instability. The Board rejected that

argument with respect to independent claims 1 and 19

because those claims contain no dosage limitation, and

therefore, the unexpected results evidence was not com-

mensurate in scope with the claims. Novartis does not

appeal that Board finding as it relates to claims 1 and 19.

Instead, Novartis argues that the Board should have

reassessed the unexpected results argument when it

found unpatentable dependent claims 8, 10, 22, and 23,

and proposed amended claims 40, 42, 54, and 55. 3 In

Novartis’ view, these claims recite the “low dosage” limi-

tation lacking in claims 1 and 19.

3 Proposed amended claims 40, 42, 54, and 55 are

identical to, and would have replaced, original claims 8,

10, 22, and 23, respectively.

22 NOVARTIS AG v. TORRENT PHARMACEUTICALS

At the outset, we note that the argument raised to the

Board below was quite different than Novartis’ character-

ization of that argument on appeal. In appeals from the

Board, “we have before us a comprehensive record that

contains the arguments and evidence presented by the

parties and our review of the Board’s decision is confined

to the four corners of that record.” In re Watts, 354 F.3d

1362, 1367 (Fed. Cir. 2004) (internal quotation marks and

citation omitted). Thus, we must first determine whether

Novartis preserved this argument for appeal. While the

court “retains case-by-case discretion over whether to

apply waiver,” Harris Corp. v. Ericsson Inc., 417 F.3d

1241, 1251 (Fed. Cir. 2005) (citations omitted), we have

held that a party waives an argument that it “failed to

present to the Board” because it deprives the court of “the

benefit of the Board’s informed judgment,” Watts, 354

F.3d at 1367–68. We turn our attention to the unexpected

results argument Novartis actually presented to the

Board.

The undeniable focus of Novartis’ arguments

throughout the proceeding, including its Patent Owner’s

Response, was the patentability of the combination of

fingolimod and mannitol, as broadly recited in claim 19.

The Argument section of the Patent Owner’s Response

alerts the Board in the very first paragraph that Novartis’

arguments are directed to claim 19. What follows in the

Patent Owners’ Response is Novartis’ explanation for why

there was no ex ante reason to combine fingolimod with

mannitol or to reasonably expect success in the combina-

tion—untethered from any specific dosage or concentra-

tion limitation and with no discussion of any dependent

claims.

Novartis’ objective indicia argument under the head-

ing “Objective Indicia Prove the Fingolimod-Mannitol

Invention” is similarly generic. Novartis there contends

that the objective indicia “overwhelmingly prove the

patentability of the fingolimod-mannitol formula.” J.A.

NOVARTIS AG v. TORRENT PHARMACEUTICALS 23

7362. Given the reference to the fingolimod-mannitol

formula only and the failure to identify any specific claim,

this section is simply a continuation of Novartis’ defense

of claim 19. And turning to the unexpected results section

in particular, we see the entirety of Novartis’ argument in

a few short sentences:

First, mannitol unexpectedly is stable with fin-

golimod throughout the full dosage range. Excipi-

ent stability normally does not vary with dose

proportions. Drs. Kent and Kibbe each confirmed

that fact in their depositions. [Citing deposition

transcripts]. So do Dr. Byrn, Dr. Pudipeddi, and

Mr. Oomura. [Citing declarations]. Petitioners do

not address this unexpected result at all.

J.A. 7363 (emphasis added). Novartis did not identify the

dependent claims at issue now or discuss specific dosages

or concentrations at all. Nor do any of the supporting

citations. The only fair characterization of Novartis’

argument is that the combination of fingolimod and

mannitol was unexpectedly stable irrespective of concen-

tration, i.e., “throughout the dosage range.” Id. Novartis’

Motion to Amend likewise fails to present any separate

argument for proposed amended claims 40, 42, 54, and 55.

We thus find no fault in the Board’s observation that

Novartis offered no separate argument with respect to

dependent claims 8, 10, 22, and 23, or proposed amended

claims 40, 42, 54, and 55. The sole focus of the proceed-

ing, including Novartis’ unexpected results argument, was

on claim 19. Novartis even conceded at oral argument

that the focus of its unexpected results argument was

that fingolimod was unexpectedly stable across the entire

dosage range. Oral Arg. at 48:40–48:49 and 53:00–53:08,

available at

http://oralarguments.cafc.uscourts.gov/default.aspx?fl=20

16-1352.mp3. We find no evidence that Novartis distinct-

ly argued an unexpected result specific to the dependent

24 NOVARTIS AG v. TORRENT PHARMACEUTICALS

claims Novartis now raises on appeal. That argument is

therefore waived.

ii. Nexus

Novartis next argues that the Board erred as a matter

of law in its analysis of the “nexus” requirement with

respect to the objective evidence of nonobviousness.

Before the Board, Novartis contended that its drug,

Gilenya, enjoyed commercial success, industry praise, and

met a long-felt but previously unsolved need, due to

Gilenya being the first commercially available solid oral

treatment for multiple sclerosis. On appeal, Novartis

complains that the Board wrongly discounted this evi-

dence in light of the disclosure of solid oral multiple

sclerosis drug formulas in prior art references, which, in

the Board’s view, precluded Novartis’ argued for basis for

a nexus between the ’283 patent’s invention and the

objective indicia—even though none of those drugs were

available to the market until after the ’283 patent was

filed. Distilled down, Novartis argues that, as a matter of

law, a feature that is known in the art but not actually

available to the market—i.e., in commerce—cannot be

used to disprove Novartis’ attempts to establish a nexus

based on that claimed feature.

We disagree. None of the cases cited by Novartis, or

any that we are aware of, stand for such a sweeping

proposition. For objective indicia evidence to be accorded

substantial weight, we require that a nexus must exist

“between the evidence and the merits of the claimed

invention.” Wyers v. Master Lock Co., 616 F.3d 1231,

1246 (Fed. Cir. 2010). 4 “Where the offered secondary

4 This is not a case where Novartis argues that the

required nexus may be presumed and the presumption

was disregarded by the Board. Moreover, any presump-

tion of nexus is nevertheless rebuttable by evidence that

NOVARTIS AG v. TORRENT PHARMACEUTICALS 25

consideration actually results from something other than

what is both claimed and novel in the claim, there is no

nexus to the merits of the claimed invention.” In re Kao,

639 F.3d 1057, 1068 (Fed. Cir. 2011); see also Tokai Corp.

v. Easton Enters., Inc., 632 F.3d 1358, 1369 (Fed. Cir.

2011) (“If commercial success is due to an element in the

prior art, no nexus exists.”); Ormco Corp. v. Align Tech.,

Inc., 463 F.3d 1299, 1312 (2006) (“[I]f the feature that

creates the commercial success was known in the prior

art, the success is not pertinent.”).

In evaluating whether the requisite nexus exists, the

identified objective indicia must be directed to what was

not known in the prior art—including patents and publi-

cations—which may well be the novel combination or

arrangement of known individual elements. See KSR Int’l

Co. v. Teleflex Inc., 550 U.S. 398, 418–19 (2007); Veritas

Techs. LLC v. Veeam Software Corp., 835 F.3d 1406,

1414-15 (Fed. Cir. 2016). Our opinion in Asyst Technolo-

gies, Inc. v. Emtrak, Inc., 544 F.3d 1310 (Fed. Cir. 2008),

is instructive. In Asyst, the trial court concluded that

patent owner Asyst failed to link the objective indicia to

the claimed invention because the proffered evidence

lacked a nexus to any feature of the invention’s commer-

cial embodiments that was not already disclosed in a prior

art patent—the Hesser patent. 544 F.3d at 1316. The

court found “even though commercial embodiments of

[Asyst’s] ’421 invention may have enjoyed commercial

success, Asyst’s failure to link that commercial success to

the features of its invention that were not disclosed in

[the] Hesser [patent] undermines the probative force of

the evidence pertaining to the success of Asyst’s [] prod-

the proffered objective evidence was due to extraneous

factors other than the merits of the claimed invention.

See, e.g., WBIP, LLC v. Kohler Co., 829 F.3d 1317, 1329

(Fed. Cir. 2016).

26 NOVARTIS AG v. TORRENT PHARMACEUTICALS

ucts.” Id. We explained, “[w]hile the evidence shows that

the overall system drew praise as a solution to a felt need,

there was no evidence that the success of the commercial

embodiment of the ’421 patent was attributable to the

substitution of a multiplexer for a bus, which was the only

material difference between [the] Hesser [patent] and the

patented invention.” Id.

Here, Novartis’ nexus argument for its objective indi-

cia evidence is based solely on a single premise—Gilenya

being the first commercially-available solid oral multiple

sclerosis treatment. The treatment of multiple sclerosis

with a solid oral composition, however, was indisputably

known in the prior art. The Board found evidence that

Chiba itself suggested treating multiple sclerosis using a

solid oral form of fingolimod. And at least two other solid

oral multiple sclerosis treatments were disclosed in the

prior art literature before the ’283 patent’s priority date.

The fact that Gilenya was the first to receive FDA ap-

proval for commercial marketing does not overcome the

fact that solid multiple sclerosis compositions were al-

ready known. Thus, we agree with the Board that Novar-

tis’ proffered evidence is not probative of the

nonobviousness inquiry. 5

5 Novartis raises an additional challenge to the

Board’s analysis of Novartis’ commercial success evidence

in particular. Having concluded that Novartis failed to

establish a sufficient nexus between its proffered com-

mercial success and the claims, the Board continued,

“setting aside the issue of whether commercial success of

Gilenya should be probative of nonobviousness, we are not

convinced that Patent Owners have carried their thresh-

old burden to show ‘significant sales in a relevant mar-

ket.’” Final Written Decision, 2015 WL 5719630, at *14.

The Board then proceeded to dismiss Novartis’ commer-

cial success argument due to its concerns with Novartis’

NOVARTIS AG v. TORRENT PHARMACEUTICALS 27

CONCLUSION

For the foregoing reasons, we affirm the Board’s deci-

sion. We have considered all of Novartis’ remaining

arguments but conclude that they are without merit.

AFFIRMED

COSTS

No costs.

assessment of the relevant market and the completeness

of its market share data. Because we agree with the

Board that Novartis failed to establish a nexus between

the claims and all purported evidence of nonobvious-

ness—including commercial success—we need not and do

not reach the Board’s additional grounds for rejecting this

evidence.