Endo Pharmaceuticals Inc., Strakan International S.a.r.l. v. Actavis Laboratories Ut, Inc.

N OTE: This disposition is nonprecedential.
United States Court of Appeals
for the Federal Circuit
______________________
ENDO PHARMACEUTICALS INC., STRAKAN
INTERNATIONAL S.A.R.L.,
Plaintiffs-Appellees
v.
ACTAVIS LABORATORIES UT, INC.,
Defendant-Appellant
______________________
2016-1146
______________________
Appeal from the United States District Court for the
Eastern District of Texas in No. 2:13-cv-00192-JRG,
Judge J. Rodney Gilstrap.
______________________
Decided: October 14, 2016
______________________
BARRY P. G OLOB, Cozen O'Connor, Washington, DC,
argued for plaintiffs-appellees. Also represented by K ERRY
BRENDAN MCT IGUE, WILLIAM BLAKE COBLENTZ, AARON S.
L UKAS , D ONALD R. M C P HAIL .
G EORGE C. L OMBARDI, Winston & Strawn LLP, Chica-
go, IL, argued for defendant-appellant. Also represented
by MICHAEL K EENAN N UTTER, I VAN MICHAEL P OULLAOS ,

-- 1 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
2
K URT A. MATHAS , J OHN REYNOLDS M CN AIR ; G EOFFREY P.
EATON, Washington, DC.
______________________
Before M OORE, T ARANTO, and HUGHES , Circuit Judges.
T ARANTO, Circuit Judge.
Strakan International S.à r.L. owns U.S. Patent Nos.
6,579,865 (issued in 2003) and 6,319,913 (issued in 2001),
with priority dating to 1997. Endo Pharmaceuticals Inc.
is the exclusive United States licensee of those patents.
In December 2010, Endo obtained final approval from the
Food and Drug Administration to market its testosterone
gel product, called Fortesta®, which is used by applying it
to the skin to deliver testosterone transdermally.
In 2013, Watson Laboratories filed an Abbreviated
New Drug Application with the FDA, seeking to market a
generic version of Fortesta. Upon receiving Watson’s
notification under 21 U.S.C. § 355(j)(2)(A)(vii)(IV), Endo
and Strakan (collectively, Endo) filed suit against Watson
under 35 U.S.C. § 271(e)(2), alleging that Watson’s mar-
keting of its proposed generic product would infringe
claims 1, 3, 4, 6, 11, 18, 22, and 28 of the ’865 patent and
claims 19 and 20 of the ’913 patent. Soon after, Watson
transferred its relevant interests to Actavis Laboratories
UT, Inc., which was substituted for Watson in this case.
For convenience, we refer to Actavis as if it, rather than
Watson, had always been the named defendant and had
filed the application for FDA approval of a generic version
of Fortesta.
In this action, Actavis alleged that all of the Endo-
asserted claims of both patents are invalid based on
anticipation and obviousness. Actavis also alleged that
the product described in its FDA application does not
meet a limitation of claims 1, 3, 4, 6, 11, 18, and 22 of the
’865 patent and, therefore, its marketing would not in-
fringe those claims. Actavis stipulated to infringement of

-- 2 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
3
the other three claims asserted by Endo. After a bench
trial, the district court ruled that the asserted patent
claims are not invalid for either anticipation, J.A. 30–35,
or obviousness, J.A. 46–54, and that Actavis’s marketing
of its product would infringe all of the asserted claims,
J.A. 65–68.
Actavis appeals the district court’s decision regarding
obviousness and infringement. We have jurisdiction
under 28 U.S.C. § 1295(a)(1). We affirm.
I
The claims at issue involve combinations of testos-
terone (or a derivative) and penetration-enhancing sys-
tems designed to allow testosterone to enter the body
through the skin. Even before the 1997 priority date for
the patents here, there was interest in producing a suc-
cessful transdermal delivery system for testosterone
because “there had been issues with delivery of testos-
terone through pill form (e.g. testosterone was largely
destroyed by the digestive system) and through injections
(e.g. patient compliance and dosage issues).” J.A. 46; see
J.A. 539–42. In 1993 and 1995, the FDA approved trans-
dermal testosterone patches—Testoderm and Androderm.
J.A. 507. The district court found that patients found the
first problematic because of the location where the patch
had to be worn and that the second “used a penetration
enhancing system,” in order for the testosterone to cross
the outer skin barrier in less sensitive body locations, “but
had significant side effects of irritation.” J.A. 10–11. The
’913 and ’865 patents resulted from attempts to find a
combination of a penetration enhancer and testosterone
that was effective in delivering the testosterone while
keeping irritation to acceptable levels. The claims at
issue, as relevant here, claim combinations of specified
penetration enhancers (sometimes in specified concentra-
tions) with specified concentrations of testosterone or its
derivatives.

-- 3 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
4
The makeup of the claimed penetration enhancers dif-
fers in certain respects from claim to claim, though much
remains constant. The claim-by-claim variations are
significant for the obviousness analysis, but not for the
usual reason that the challenger argues for invalidity on a
claim-by-claim basis. Here, Actavis makes no argument
that some claims are invalid even if others stand. Pre-
sumably reflecting the content of its generic-Fortesta
application to the FDA, Actavis makes a single, all-or-
nothing, across-the-board obviousness argument: as
Actavis argues its case, if any one of the asserted claims
survives, all do. As a result, the prior-art analysis called
for by Actavis’s approach properly considers differences
between prior art and any of the asserted claims.
Oleic acid is required in all of the claimed penetration
enhancers, as are a glycol and an alcohol, but the specific
alcohol and glycol used varies in the claims. Claim 1 of
the ’865 patent claims a composition “consisting essential-
ly of” (a) testosterone (or a derivative) having a concentra-
tion of about 0.1% to about 2% and (b) a penetration
enhancing system consisting of oleic acid, a C1-C4 alcohol,
and a glycol. ’865 patent, col. 14, lines 34–44. C1-C4
alcohols include ethanol, isopropanol, and propanol. J.A.
424. The claimed composition also includes a gelling
agent.
Claims 3, 4, 6, 11, 18, and 22 of the ’865 patent de-
pend on claim 1. Claim 3 limits the active agent to testos-
terone (rather than a derivative). Claim 4 limits the
concentration of testosterone to a range of “about 1% to
about 2%” (only a part of the claim 1 range). Claim 6
limits the concentration of oleic acid to a range of “about
0.1% to about 5%.” Claim 11 limits the “C1-C4 alcohol” to
a mixture of ethanol and isopropanol. Claim 18 limits the
“glycol” to propylene glycol. Claim 22 limits the composi-
tion to one in which the claim 1 “glycol is present from
about 30% to about 40%” of the composition’s weight.

-- 4 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
5
Independent claim 28 is similar to claim 1 but also re-
quires “inert carriers.”
As to the ’913 patent, independent claim 19 claims a
method of topically administering a composition “compris-
ing” (a) testosterone at a concentration “of about 0.1% to
about 2%” and (b) “a penetration-enhancing system
comprising: (i) oleic acid; (ii) a C3 alcohol; and (iii) propyl-
ene glycol.” ’913 patent, col. 15, line 31, through col. 16,
line 4. C3 alcohols include propanol and isopropanol. J.A.
434. Claim 20 depends on claim 19 and differs only in
that it includes specific concentrations of testosterone.
’913 patent, col. 16, lines 5–9.
II
Having decided not to appeal the district court’s find-
ing of no anticipation, Actavis no longer disputes the
novelty of the claimed compositions (’865 patent) and
methods (’913 patent). It argues, however, that it proved
by clear and convincing evidence that the inventions
defined by all of the asserted claims would have been
obvious in 1997. It does so by pointing to multiple pieces
of prior art.
The evidence and arguments presented to us permit
findings that, for each reference, there is a gap between
its teaching and at least one of the asserted claims—
something in at least one claim not disclosed in the refer-
ence. That is significant because of Actavis’s all-or-
nothing approach to arguing obviousness. We describe
examples of the gaps here, before discussing, in the next
section, the findings and evidence regarding whether a
relevant skilled artisan would have bridged those gaps.
Cooper ’872. This reference teaches the delivery of
corticosteroids, not testosterone (which is not a cortico-
steroid). The district court found that the reference
teaches that an alcohol such as ethanol should be used,
but not as part of the penetration enhancer, only as a

-- 5 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
6
solvent. J.A. 32. The court relied in part on the refer-
ence’s statement that the alcohol, “if used, should not
significantly interfere with the penetration action of the
binary combination” of “a diol and a cell-envelope disor-
dering compound.” J.A. 578, 577, quoted at J.A. 20, 22.
The court also found that there is no teaching of the
combination of isopropanol and ethanol required by claim
11 of the ’865 patent. J.A. 32.
Cooper ’934. The advance described and claimed in
this reference is a penetration-enhancing vehicle contain-
ing a “binary” mixture of 1-dodecylazacycloheptan-2-one
(azone) and either “a C3-C4 diol, such as propylene glycol,
or a second N-substituted azacycloalkyl-2-one.” J.A. 590.
Although there was evidence that azone and oleic acid
have some similar properties, it is undisputed that they
are not the same compound. The points made above
about Cooper ’872, concerning the role of relevant alcohols
and the absence of an isopropanol-ethanol combination,
apply as well to Cooper ’934. J.A. 34. The district court
found that Cooper ’934 does not specifically identify
testosterone, including in any of its Examples, which
discuss a significant number of differently constituted
penetration vehicles. J.A. 33; J.A. 600–05. The court
noted that Cooper ’934 does include “male sex hormones”
among the potential active ingredients listed in a multi-
page, unelaborated recitation of conditions and active
ingredients—which the district court found “enumerates
‘without limitation’ a list of what appears to be every
condition or ailment one might seek to apply a pharma-
ceutical agent to.” J.A. 33; see J.A. 28; J.A. 505 (testify-
ing to Cooper ’934’s “very long or comprehensive list of
different drug classes, and . . . pretty much every known
drug within those classes at the time of the invention”).
Patel ’970. This reference mentions testosterone
among many other potential active ingredients, but
Actavis has not pointed to evidence showing that it dis-
closes the three-part penetration enhancers of all of the

-- 6 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
7
asserted claims here, as required for Actavis’s argument.
The penetration-enhancing vehicle of the Patel ’970
patent, as described in the Abstract, has three compo-
nents: (i) a cell-disordering compound such as oleic acid;
(ii) specific lower alkanols, namely, ethanol, propanol,
isopropanol, or mixtures of them; and (iii) an optional
“inert diluent.” J.A. 615. But Actavis has cited to no
evidence that the patent discloses a three-part penetra-
tion enhancer of oleic acid, a C1-C4 alcohol such as etha-
nol, and propylene glycol, as required by claim 18 of the
’865 patent and claims 19 and 20 of the ’913 patent.
Actavis points to certain testimony of its expert, Dr.
Potts, but neither that nor other cited testimony says that
Patel ’970 teaches the three-part enhancer with propylene
glycol. Thus, Dr. Potts noted that the Abstract of Patel
’970 discloses a three-part composition including oleic acid
and ethanol and, also, polypropylene glycol (PPG) and
polyethylene glycol (PEG)—which Patel ’970 identifies as
among the “inert diluents” required by component (iii).
J.A. 431; see J.A. 615, 624, 625. Dr. Potts noted that PPG
and PEG are “glycols” under the ’865 and ’913 pa-
tents, J.A. 431, but he did not say, and Actavis has identi-
fied no testimony stating, that the disclosed PPG or PEG
is “propylene glycol,” as required by the above-identified
claims. When Patel ’970 discusses propylene glycol (PG),
it is not as an “inert diluent” in the claimed compound, as
with PPG and PEG, but only in comparing its own combi-
nation of oleic acid and component (ii)’s lower alkanols
favorably, with respect to skin irritation, to combinations
of oleic acid and PG. J.A. 621 (“[T]his combination of oleic
acid and propylene glycol causes severe skin irritation. . . .
[T]he combinations of 80% glycerol dioleate and/or oleic
acid with 20% ethanol provide penetration enhancement
similar to that obtained with propylene glycol and oleic
acid and, as will subsequently be demonstrated, does not
possess the skin irritation properties of propylene glycol-
oleic acid combinations.”); see J.A. 615, 617, 618, 624, 625.

-- 7 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
8
Patel ’190. This reference discloses a combination of
testosterone, glycerin (which the ’913 and ’865 patents
define as a glycol), oleic acid, and ethanol. J.A. 613. But
Actavis points to no such combination that does not also
include methyl laurate, and it neither identifies evidence,
nor even argues, that Patel ’190 discloses a combination
“consisting essentially” of glycerin, oleic acid, and ethanol,
i.e., with no other material components. J.A. 612. Actavis
likewise points to no evidence that Patel ’190 discloses the
use of propylene glycol, as required by claim 18 of the ’865
patent and claims 19 and 20 of the ’913 patent, or the use
of an ethanol-isopropanol mixture, as required by claim
11 of the ’865 patent.
Aungst. Both the 1989 and 1995 Aungst references
teach the combination of fatty acids, such as oleic acid,
with propylene glycol, but Actavis does not show that they
disclose the combination of those ingredients with an
alcohol. See J.A. 50. And while the 1995 Aungst refer-
ence mentions testosterone with a different fatty acid and
propylene glycol, Actavis does not show that either refer-
ence discusses the use of testosterone with an oleic-acid
penetration-enhancing system. See J.A. 1333–34, 1357.
Other References. The Santus reference, not specifi-
cally addressed by the district court, discusses the use of
oleic acid, propylene glycol, and “ethyl alcohol,” another
name for ethanol, as one example among many different
penetration enhancers shown in patents. J.A. 750.
Actavis identifies in Santus no mention of testosterone,
for this or any other enhancer, or of a mixture of ethanol
and isopropanol. The district court found, and it is not
disputed, that the Francoeur and Touitou references do
not disclose the use of testosterone. J.A. 51. Cormier
teaches the use of cytotoxic agents such as 5-flourouracil
as the active ingredient, and the district court found no
evidence that testosterone is a cytotoxic agent or substan-
tially similar to one. J.A. 51. And Actavis does not
demonstrate, or even clearly argue, that any of those

-- 8 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
9
references discloses the subject matter of all of the Endo-
asserted claims—not just some claimed combinations, not
just elements of claimed combinations, but all the specific
combinations claimed, together with “consisting essential-
ly of” for the ’865 patent. In particular, the Actavis-cited
testimony of Endo’s expert is fairly read as making only
more limited points than that. See J.A. 519–20, 522–25.
III
The district court concluded that a relevant skilled
artisan would not have considered it obvious to bridge the
gaps that separate the prior art from (at least one of) the
claimed compositions and methods. The only relevant
problem identified in the record was balancing delivery
effectiveness with acceptable irritation for testosterone.
J.A. 36. And while it would have been obvious to pursue a
transdermal delivery mechanism for testosterone, and to
consider using a penetration enhancer, the district court
determined, it would not have been “obvious how to select
and configure a particular penetration enhancer to com-
bine with a particular level of testosterone.” J.A. 47–48.
Actavis “has not established that it was obvious that a
person of ordinary skill in the art, which in this case is a
highly and specifically educated person, would have,
considering the art of the time, including the art present-
ed to the Court, found the inventions of the patents-in-
suit to be obvious either alone or in combinations of the
numerous various references [Actavis] puts forward in the
specific way that would yield the inventions of the pa-
tents-in-suit.” J.A. 52–53.
Actavis has not presented a persuasive showing of
prejudicial factual or legal error in the district court’s
determination. Notably, sufficient evidence supports the
finding that “there were tremendous numbers of penetra-
tion enhancers that were known in the relevant time
period and that one of ordinary skill, in this case, could
have combined with testosterone.” J.A. 52; see, e.g., J.A.

-- 9 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
10
40, 47; J.A. 441–43, 739–58, 1606–10.1 So too for the
finding that the desired balance of effects varies “unpre-
dictabl[y]” according to the “specific” makeup of the
particular enhancer and the choice of active ingredient
with which it was to be combined. J.A. 33; see, e.g., J.A.
501 (quoted at J.A. 16) (“quite difficult to predict”), 31, 33
(noting “the extensive record describing how specific and
unpredictable mixtures are in the context of transdermal
agents”), 36, 47, 52, 1292–96, 1333. And it is relevant,
too, in determining what specific course a relevant skilled
artisan would have had reason to pursue, among a large
number of possibilities, that some prior art taught that a
combination of oleic acid and propylene glycol could cause
“severe skin irritation” whereas certain other enhancers
would not. J.A. 621, 1355–57.2
That evidence is sufficient to uphold the district
court’s determination against the arguments Actavis has
presented for reversal, and we need not go on to review
1 Just as to what is reflected in patents (not other
publications), the Santus article, from 1993, reviews more
than 150 enhancer patents, categorizing them into twelve
groups of different types of enhancers, each type covering
different individual enhancers—“(a) alcohol enhancers; (b)
amide enhancers; (c) amino acid enhancers; (d) azone and
azone-like enhancers; (e) essential oil enhancers; (f) fatty
acid and fatty acid ester enhancers; (g) Macrocyclic en-
hancers; (h) phospholipid and phosphate enhancers; (i) 2-
pyrrolidone derivatives; (j) soft penetration enhancers; (k)
sulphoxide enhancers; (l) miscellaneous enhancers.” J.A.
741; see J.A. 839–51.
2 When Actavis argues that the irritability of the
claimed composition is irrelevant because Fortesta is the
most irritating product on the market, it is impermissibly
comparing the composition’s irritability to that of prod-
ucts that entered the market after the priority date.

-- 10 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
11
certain other determinations made by the district court
that, if sound, could only further weaken, not aid, Ac-
tavis’s argument for obviousness. Thus, we need not
consider whether the prior art’s discussion of irritation
effects of certain enhancer compositions, beyond support-
ing the finding of insufficient reason of a relevant skilled
artisan to pursue the combinations at issue here, actually
“teach away” from a path such an artisan otherwise would
pursue with the required expectation of success. Nor need
we consider whether any otherwise-persuasive showing of
obviousness could be overcome in this case by objective
indicia such as unexpected results, long-felt need, or
others’ failure to arrive at the inventions.
Finally, we see no basis for reversing the district
court’s judgment in the court’s statement that “the effec-
tiveness and side effects of using a combination of a
particular penetration enhancer to deliver a particular
compound were not readily predictable.” J.A. 47. Unlike
Actavis, we do not read that statement as applying an
improperly high threshold for proving obviousness—ready
predictability rather than a reasonable expectation of
success—where the requisite motivation to combine is
otherwise proved. Here, we read the district court’s
opinion as finding no such motivation to combine, a
question to which the sheer number of possible combina-
tions and degree of uncertainty are both relevant.
Moreover, the district court recited the expert testi-
mony that the relevant effects are “quite difficult to
predict,” J.A. 18 (internal quotation marks omitted), and
that “a person of ordinary skill would not have expected
success using the penetration enhancing system,” J.A. 36.
And far from requiring ready predictability, the passage
Actavis quotes immediately goes on to explain that “ex-
periments were depended upon for the characterization of
the properties of such a combination” and to credit the
evidence “that extensive work, including a progression of
studies, was generally required to characterize such a

-- 11 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
12
combination on human subjects.” J.A. 47. In the end, we
do not read the district court’s passage as raising the
threshold for proof of obviousness the way Actavis alleges.
We read the passage as addressing and rejecting one of
Actavis’s contentions in the very terms of “predictable
results” and “predictable use” from KSR Int’l Co. v. Tele-
flex Inc., 550 U.S. 398, 416, 417 (2007), that Actavis itself
invoked. Watson’s Proposed Findings of Fact and Conclu-
sions of Law 74–75, ECF No. 130.
IV
Actavis also argues that the district court’s finding of
infringement was clearly erroneous as to all of the assert-
ed claims of the ’865 patent except claim 28. Its theory is
that in its accused product water plays a material role as
part of the enhancer, so that its penetration enhancer
does not “consist[ ] essentially” of the claim-listed compo-
nents. Actavis agreed at oral argument in this court that
the infringement issue need not be decided if we affirm
the district court’s validity determination, because Ac-
tavis stipulated to infringement of claim 28 of the ’865
patent and claims 19 and 20 of the ’913 patent, and it
needs to succeed on all claims to be permitted to market
its generic product. Oral Arg. at 2:00–2:55. In any event,
we do not see clear error in the district court’s finding
that Endo proved infringement. Actavis, in its filings
with the FDA, listed only one function for water in its
product—serving as a solvent—even while it listed dual
solvent/enhancer functions for other components. It is a
legitimate inference in the circumstances here that water
in Actavis’s product does not play a material role in the
penetration enhancer. See J.A. 17, 66–67. Therefore, we
refuse to disturb the court’s infringement finding.
CONCLUSION
For the foregoing reasons, we affirm the judgment of
the district court.

-- 12 of 13 --

ENDO PHARMACEUTICALS INC. v. ACTAVIS LABORATORIES UT,
INC.
13
AFFIRMED

-- 13 of 13 --