IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`COALITION FOR AFFORDABLE DRUGS X LLC,
`Petitioner,
`
`v.
`
`ANACOR PHARMACEUTICALS, INC.,
`Patent Owner.
`
`Case No. IPR2015-01776
`Patent No. 7,582,621
`
`PATENT OWNER’S RESPONSE PURSUANT TO 37 C.F.R. § 42.120
`
`
`
`
`
`
`
`
`
`DC: 6086399-6
`
`
`
`

`
`TABLE OF CONTENTS
`
`
`IPR2015-01776
`
`Page
`
`I.
`
`II.
`
`
`INTRODUCTION .......................................................................................... 1
`
`BACKGROUND OF THE INVENTION .................................................... 10
`
`A.
`
`B.
`
`C.
`
`Treating onychomycosis requires a compound with selective
`toxicity ................................................................................................ 10
`
`A POSA would have been concerned about the toxicity of
`tavaborole, a boron-containing compound ......................................... 11
`
`Anacor defied the conventional wisdom in developing
`KERYDIN .......................................................................................... 18
`
`D.
`
`The ’621 Patent .................................................................................. 20
`
`III.
`
`PETITIONER HAS NOT OFFERED TESTIMONY PROBATIVE
`OF THE VIEWS OF A PERSON OF ORDINARY SKILL IN THE
`ART AT THE TIME OF THE INVENTION .............................................. 21
`
`IV. CLAIM CONSTRUCTION OF “THERAPEUTICALLY
`EFFECTIVE AMOUNT” ............................................................................. 24
`
`V.
`
`PETITIONER HAS FAILED TO MEET ITS BURDEN OF
`PROVING THAT THE CLAIMS OF THE ’621 PATENT ARE
`UNPATENTABLE ....................................................................................... 26
`
`A.
`
`Both Grounds fail because Austin is not analogous art ...................... 27
`
`1.
`
`2.
`
`3.
`
`Austin and the ’621 Patent are not from the same field of
`endeavor ................................................................................... 28
`
`Austin is not reasonably pertinent to the problem solved
`by the ’621 Patent .................................................................... 31
`
`Even if Austin were analogous art, there is no basis to
`conclude that a POSA would have selected tavaborole
`from among Austin’s multitude of compounds ....................... 33
`
`- i -
`
`

`
`B.
`
`Petitioner has not established that Claims 1-12 are unpatentable
`over Austin and Brehove (Ground 1) ................................................. 35
`
`IPR2015-01776
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`Even if Austin and Brehove were combined, that
`combination would not disclose every limitation of any
`claim ......................................................................................... 35
`
`A POSA in 2005 would understand that Brehove is not
`credible ..................................................................................... 36
`
`Petitioner has not shown that there would have been
`reason to combine Austin and Brehove .................................... 41
`
`a)
`
`b)
`
`c)
`
`Austin and Brehove concern structurally different
`compounds ..................................................................... 41
`
`Neither Austin nor Brehove discloses any activity
`against dermatophytes ................................................... 43
`
`The goal of transungual delivery does not establish
`a reason to combine Austin and Brehove with a
`reasonable expectation of success in achieving
`transungual delivery ...................................................... 47
`
`Long-standing concerns about tavaborole’s toxicity
`preclude a reasonable expectation of success in
`combining Austin and Brehove ................................................ 50
`
`Under Petitioner’s theory that boron-containing
`compounds share similar functional features, Freeman
`establishes that a POSA would have expected Brehove’s
`dioxaborinanes to be unsafe and ineffective for
`therapeutic antifungal use ........................................................ 53
`
`C.
`
`Petitioner has not established that Claims 1-12 are unpatentable
`over Austin and Freeman (Ground 2) ................................................ 54
`
`1.
`
`2.
`
`Ground 2 fails for many of the same reasons as Ground 1 ...... 54
`
`Freeman negates a reasonable expectation of success
`because PBAs were shown to be toxic and appeared to be
`ineffective ................................................................................. 55
`
`- ii -
`
`

`
`IPR2015-01776
`
`VI. UNEXPECTED RESULTS COMPEL A FINDING OF NON-
`OBVIOUSNESS ........................................................................................... 60
`
`VII. OBJECTIVE INDICIA CONFIRM NON-OBVIOUSNESS OF
`CLAIMS 1-12 ............................................................................................... 61
`
`A. KERYDIN succeeded—where others failed—in meeting the
`long-felt need for a safe and effective topical onychomycosis
`treatment ............................................................................................. 61
`
`B.
`
`KERYDIN has received praise in the industry .................................. 63
`
`VIII. CONCLUSION ............................................................................................. 64
`
`
`
`- iii -
`
`

`
`TABLE OF AUTHORITIES
`
`IPR2015-01776
`
` Page(s)
`
`Cases
`In re Bigio,
`381 F.3d 1320 (Fed. Cir. 2004) .......................................................................... 27
`
`In re Clay,
`966 F.2d 656 (Fed. Cir. 1992) ...................................................................... 27, 30
`
`Coalition for Affordable Drugs VI LLC v. Celgene Corp.,
`IPR2015-01092, Paper No. 18 (P.T.A.B. Sept. 25, 2015) .................................. 64
`
`Coalition for Affordable Drugs VI LLC v. Celgene Corp.,
`IPR2015-01169, Paper No. 22 (P.T.A.B. Nov. 16, 2015) .................................. 42
`
`In re Cuozzo Speed Techs., LLC,
`793 F.3d 1268 (Fed. Cir. 2015), reh’g en banc denied, 793 F.3d
`1297 (Fed. Cir. 2015), cert. granted, 136 S. Ct. 890 .......................................... 24
`
`In re Cyclobenzaprine Hydrochloride Extended-Release Capsule
`Patent Litig.,
`676 F.3d 1063 (Fed. Cir. 2012) .......................................................................... 26
`
`Eli Lilly & Co. v. Teva Pharm. USA, Inc.,
`619 F.3d 1329 (Fed. Cir. 2010) .......................................................................... 50
`
`Freund Indus. Co. v. Driam Metallprodukt GmbH & Co.,
`No. 88 Civ. 3605, 1989 WL 88704 (S.D.N.Y. July 31, 1989) ........................... 30
`
`In re Hedges,
`783 F.2d 1038 (Fed. Cir. 1986) .......................................................................... 36
`
`Hughes Network Sys., LLC v. Elbit Sys. Land and C4I Ltd.,
`IPR2016-00135, Paper No. 8 (P.T.A.B. Apr. 27, 2016) ...................................... 4
`
`In re ICON Health and Fitness, Inc.,
`496 F.3d 1374 (Fed. Cir. 2007) .......................................................................... 25
`
`- iv -
`
`

`
`IPR2015-01776
`
`InSite Vision Inc. v. Sandoz Inc.,
`No. CIV.A. 11-3080 MLC, 2013 WL 5975015 (D.N.J. Oct. 4,
`2013) ................................................................................................................... 40
`
`InTouch Techs., Inc. v. VGo Commc’ns,
`751 F.3d 1327 (Fed. Cir. 2014) .......................................................................... 26
`
`In re Kao,
`639 F.3d 1057 (Fed. Cir. 2011) .......................................................................... 61
`
`In re Klein,
`647 F.3d 1343 (Fed. Cir. 2011) .......................................................................... 27
`
`KSR Int’l Co. v. Teleflex Inc.,
`550 U.S. 398 (2007) ............................................................................................ 26
`
`Leo Pharm. Prods., Ltd. v. Rea,
`726 F.3d 1346 (Fed. Cir. 2013) .......................................................................... 61
`
`Medtronic, Inc. et al. v. Lifeport Sci. LLC,
`IPR2014-00288, Paper 34 (P.T.A.B. April 21, 2015) ........................................ 24
`
`Panduit Corp. v. Dennison Mfg. Co.,
`810 F.2d 1561 (Fed. Cir. 1987) .......................................................................... 33
`
`Parrot S.A. v. Drone Techs., Inc.,
`IPR2014-00732, Paper No. 29 (P.T.A.B. Oct. 20, 2015) ................................... 30
`
`Pfizer Inc. v. Teva Pharm. USA, Inc.,
`555 F. App’x 961 (Fed. Cir. 2014) ....................................................................... 6
`
`Procter & Gamble v. Teva Pharm.,
`566 F.3d 989 (Fed. Cir. 2009) ............................................................................ 35
`
`SATA GmbH & Co. KG v. Anest Iwata Corp.,
`IPR2013-00111, Paper No. 17 (P.T.A.B. June 25, 2013) .................................... 4
`
`Schering Corp. v. Geneva Pharm.,
`339 F.3d 1373 (Fed. Cir. 2003) .......................................................................... 39
`
`Schott Gemtron Corp. v. SSW Holding Co.,
`IPR2014-00367, Paper No. 62 (P.T.A.B. May 26, 2015) .................................. 27
`
`- v -
`
`

`
`IPR2015-01776
`
`Takeda Pharm. Co. v. Mylan Inc.,
`No. 13-CV-04001-LHK, 2014 WL 5862134 (N.D. Cal. Nov. 11,
`2014) ................................................................................................................... 22
`
`Statutes & Regulations
`
`35 U.S.C. § 316(e) ................................................................................................... 26
`
`37 C.F.R. § 42.100(b) .............................................................................................. 24
`
`
`
`- vi -
`
`

`
`IPR2015-01776
`
`Patent Owner Anacor Pharmaceuticals, Inc. (“Anacor”) provides this
`
`Response to the Petition (“Pet.”) filed by Coalition for Affordable Drugs X LLC
`
`(“Petitioner”), which requested inter partes review of claims 1-12 of U.S. Patent
`
`No. 7,582,621 (“the ’621 Patent,” Ex. 1001). This Response is timely. (Paper No.
`
`31). For the reasons set forth below and the accompanying declarations of four
`
`highly-credentialed experts, the Board should confirm the patentability of Claims
`
`1-12.
`
`I.
`
`INTRODUCTION
`
`The ’621 Patent claims an important new treatment option for a disease
`
`called onychomycosis. This is a fungal infection of the nail plate or nail bed, from
`
`which millions of Americans suffer, that can cause a range of physical and
`
`psychological problems. Before the 2005 priority date, the existing treatment
`
`options for onychomycosis were wholly unsatisfactory. Because no existing
`
`treatment was capable of effective topical delivery of an active ingredient through
`
`the nail, physicians and patients resorted to oral formulations that caused serious
`
`side effects. There was a long-felt, unmet need for topical treatments, and the field
`
`is littered with failed attempts to develop a safe and effective topical
`
`onychomycosis treatment.
`
`Anacor satisfied this unmet need with an approach that ran counter to the
`
`conventional wisdom in the field. Every claim of the ’621 Patent is drawn to
`
`1
`
`

`
`IPR2015-01776
`
`methods of treatment using a compound called tavaborole, which contains a boron
`
`atom. The record reflects a consensus in 2005 that using boron-containing
`
`compounds in pharmaceuticals raised serious toxicity concerns. From the
`
`standpoint of 2005, tavaborole unexpectedly turned out to be both highly effective
`
`against the fungus that causes onychomycosis, and safe for use in animals. And
`
`tavaborole’s unexpected safety and potency results from a property that would
`
`have been unexpected in 2005—a unique mechanism of action that avoids the
`
`toxicity problems that plagued prior attempts to use boron in pharmaceutical
`
`applications. (Ex. 2034 (Reider) ¶¶ 201 & 206-08.)
`
`Petitioner’s challenge to the ’621 Patent consists of an attorney-driven and
`
`hindsight-based analysis of the prior art, accompanied by declarations from experts
`
`lacking needed expertise. This is shown by the substance of Petitioner’s arguments
`
`and the process by which those arguments were formulated.
`
`Discovery revealed that the expert declarations purporting to support the
`
`Petition were first drafted by attorneys and left largely unchanged by the expert
`
`declarants, who later were often not familiar with, and sometimes even disavowed,
`
`the arguments drafted by the attorneys.1 One of Petitioner’s experts testified that
`
`
`1 See Ex. 2033 at 39:4-48:5 (Kahl); Ex. 2032 at 23:23-24:1, 32:3-36:6, 336:10-13,
`
`345:2-346:8 (Murthy) (conceding that his declaration in a related proceeding
`
`
`
`2
`
`

`
`IPR2015-01776
`
`the petition itself was already drafted before Petitioner’s attorneys sought his input.
`
`(Ex. 2032 at 27:2-23 (Murthy).) To the extent the experts had input into
`
`Petitioner’s validity arguments, it was given after they received the ’621 Patent.
`
`(Ex. 2032 at 28:5-29:8 (Murthy); Ex. 2033 at 27:2-24 (Kahl); see also Ex. 2032 at
`
`581:16-24 (Murthy) (“objective” of the exercise was to look at the three references
`
`provided by the attorneys and figure out how a POSA “would find tavaborole to be
`
`obvious”).) After receiving the ’621 Patent, nearly all of the prior art references
`
`the experts reviewed were provided by attorneys. (Ex. 2032 at 25:17-26:20
`
`(Murthy); Ex. 2033 at 26:3-23 (Kahl).) This is the paradigmatic hindsight-based
`
`approach.
`
`Petitioner’s experts also repeatedly give opinions they are not qualified to
`
`give. The claims of the ’621 Patent are directed to using tavaborole to treat fungal
`
`infections, especially including onychomycosis, which is overwhelmingly caused
`
`by fungi known as dermatophytes. In particular, dependent Claim 6 is limited to
`
`methods of using tavaborole to treat tinea unguium, defined as “onychomycosis
`
`caused by a dermatophyte.” (Ex. 1001 at 28:24-25; Pet. at 12-13.) Petitioner has
`
`identified no prior art suggesting that tavaborole would be effective to treat
`
`
`falsely asserted that Freeman disclosed human treatment, an assertion which
`
`similarly appears in Ex. 1008 (Murthy) ¶¶ 77, 124, 126, 130).
`
`3
`
`

`
`IPR2015-01776
`
`infections caused by a dermatophyte. And Anacor has provided testimony from an
`
`expert in mycology (the study of fungi), showing that, in the absence of such data,
`
`a POSA would have no reason to combine Austin with Brehove or Freeman to
`
`arrive at a method of treating dermatophyte infections, including onychomycosis,
`
`nor a reasonable expectation of success in doing so.
`
`To counter this evidence, Petitioner offers the testimony of two experts who
`
`each admitted that they have no expertise in mycology, (Ex. 2033 at 392:10-11
`
`(Kahl)), and Dr. Murthy also admitted that he is not a clinician or a chemist, let
`
`alone a medicinal chemist. (Ex. 2032 at 272:5-15, 290:8-19, 305:15-24 (Murthy).)
`
`The failure of proof this represents is illustrated by Ground 2, which is based on
`
`the allegation that the Freeman reference discloses the in vivo treatment of
`
`dermatophytes in humans. (Pet. at 44-45.) Dr. Murthy admitted that Freeman
`
`does not disclose the results of any in vivo tests, and that statements to this effect in
`
`his declaration are false. (Ex. 2032 at 346:5-8 (Murthy).)
`
`The manifest lack of reliability of Petitioner’s expert declarations, by itself,
`
`should doom Petitioner’s challenge. See SATA GmbH & Co. KG v. Anest Iwata
`
`Corp., IPR2013-00111, Paper No. 17 at 4 (P.T.A.B. June 25, 2013) (holding that
`
`the PTAB is “is not required to give [] testimony weight” where the testimony “is
`
`not credible given the absence of supporting facts and objective evidence”);
`
`Hughes Network Sys., LLC v. Elbit Sys. Land and C4I Ltd., IPR2016-00135, Paper
`
`4
`
`

`
`IPR2015-01776
`
`No. 8 at 9 n.2 (P.T.A.B. Apr. 27, 2016) (Expert’s testimony found to be unhelpful
`
`“because it merely mimics Petitioner’s argument.”).
`
`Petitioner’s flawed process and testimony from declarants lacking relevant
`
`expertise led to flawed obviousness arguments that leave the Petition far short of
`
`the preponderance of the evidence required to sustain Petitioner’s burden.
`
`The Petition fails at the outset because the reference common to both
`
`Grounds, Austin (Ex. 1002), is not analogous art. (See V.A.1. & 2., infra.) Austin
`
`is the only prior art reference disclosing tavaborole, the only compound recited in
`
`the claims. But Austin relates to industrial biocides, not medicinal antifungals for
`
`use in animals. This matters because these types of compositions have very
`
`different objectives. A medicinal antifungal must be selectively toxic, that is, it
`
`must act only on the fungus, and not the host. By contrast, biocides used to kill
`
`fungi in industrial applications (for example, paint) do not involve any mammalian
`
`or plant host, and thus there is little if any need to worry about selective toxicity.
`
`Indeed, industrial media do not even contain living cells. For these reasons, a
`
`POSA would consider Austin outside the relevant field of endeavor, and not
`
`pertinent to the problem solved by the ’621 Patent. Petitioner’s contrary argument
`
`implies that a medicinal chemist would expect the bleach used to kill a wide
`
`variety of bacteria and fungi on hospital floors to be a safe treatment for the human
`
`patients in those same hospitals suffering from an infection caused by a particular
`
`5
`
`

`
`IPR2015-01776
`
`bacteria or fungus. Needless to say, the fact that Petitioner had to go so far afield
`
`to find a reference disclosing tavaborole shows Petitioner’s improper “hindsight-
`
`driven search.” Pfizer Inc. v. Teva Pharm. USA, Inc., 555 F. App’x 961, 970 (Fed.
`
`Cir. 2014).
`
`Petitioner’s reliance on Austin is also misguided because Petitioner cherry-
`
`picked Austin’s data to try to show that a POSA would have concluded that
`
`“tavaborole is one of three preferred compounds tested that effectively inhibits
`
`Candida albicans, which is a cause of onychomycosis.” (Paper No. 24,
`
`“Decision,” at 10 (citing Pet. 31-32, Ex. 1006 ¶¶ 34, 38; Ex. 1008 ¶¶ 61, 64, 67-71,
`
`90).) Austin shows nothing of the sort. If anything, Austin’s data would have led a
`
`POSA to a different class of compounds than the one including tavaborole. (See
`
`V.A.3., infra.) This conclusion is supported by an expert with decades of
`
`medicinal chemistry experience, including with boron-containing compounds.
`
`(Ex. 2034 (Reider) ¶¶ 4-13.) Petitioner’s contrary conclusion, upon which the
`
`Board relied in instituting this proceeding (Decision at 10), is supported by
`
`someone who conceded that he was not a chemist. (Ex. 2032 at 305:15-24
`
`(Murthy).)
`
`Petitioner’s remaining references exacerbate the weakness of its arguments
`
`based on Austin. In Ground 1, Petitioner seeks to combine Austin with Brehove,
`
`which Petitioner touts as showing successful treatment of onychomycosis with
`
`6
`
`

`
`IPR2015-01776
`
`boron-containing compounds used as fungicides in aviation fuel. But Brehove
`
`simply underscores the fundamental differences between industrial biocides and
`
`medicinal antifungals. Austin, like industrial biocide literature generally, contains
`
`no information about human use. Brehove purports to recite the safety information
`
`about the compounds in the fuel additive, stating that they can lead to “mild” skin
`
`irritation, but the safety sheet for those fuel additive compounds warns of serious
`
`consequences that may result from human exposure to them. Petitioner is simply
`
`wrong that Brehove evidences that a POSA would have considered industrial
`
`biocide art in developing a safe topical treatment of fungal infections.
`
`Brehove’s disclosures are also belied by the prior art. A POSA would have
`
`concluded, based on basic organic chemistry and the manufacturer’s technical
`
`guide for the Brehove aviation fuel additive, that Brehove’s purported examples of
`
`administering dioxaborinane compounds to treat onychomycosis were prophetic
`
`examples and never carried out. (See V.B.2., infra.)
`
`But even if the Board disagreed with all of the above, Petitioner’s own
`
`arguments concerning the third asserted reference, Freeman (Ex. 1004), are fatal to
`
`both Grounds in this proceeding, because accepting those arguments would
`
`necessarily lead to the conclusion that a POSA in 2005 would not have sought to
`
`administer tavaborole to treat fungal infections. (See V.B.5., V.C.2., infra.)
`
`7
`
`

`
`IPR2015-01776
`
`Freeman discloses experiments with a class of boron-containing compounds
`
`called phenyl boronic acids, or PBAs. Like Brehove’s dioxaborinanes, PBAs are
`
`structurally different from tavaborole, meaning that Freeman would give a POSA
`
`no basis to expect that Austin’s tavaborole would be safe and effective in treating
`
`onychomycosis and other fungal infections in animals. (See V.C.2., infra.)
`
`Petitioner and its experts nevertheless assert—as they must, to have any
`
`chance of establishing a reason to combine these references—that tavaborole is
`
`sufficiently similar to Freeman’s PBAs, and para-fluoro PBA in particular, that a
`
`POSA would expect tavaborole to have similar properties with respect to both
`
`spectrum of activity and toxicity. (Ex. 2033 at 377:13-378:8, 380:6-9 (Kahl); Ex.
`
`2032 at 194:5-11, 212:15-213:3 (Murthy).) But if this is so, Petitioner’s case is
`
`finished. If tavaborole performed similarly to Freeman’s compounds—as
`
`Petitioner argues—a POSA would have expected the methods of using tavaborole
`
`claimed in the ’621 Patent to be unsafe and ineffective, causing serious toxicity
`
`while achieving little antifungal effect. (See V.B.5., infra.) As to efficacy,
`
`Freeman shows that PBAs have little to no activity against the tested
`
`microorganisms, including the dermatophyte (T. rubrum) that is overwhelmingly
`
`responsible for onychomycosis. As to safety, the prior art confirms the high
`
`toxicity of PBAs, specifically including the para-fluoro PBA derivative that
`
`Petitioner’s expert testified was structurally similar, and would be functionally
`
`8
`
`

`
`IPR2015-01776
`
`similar, to tavaborole. (Ex. 2033 at 377:13-378:8 (Kahl); Ex. 2202 at 56 (Kahl
`
`Dep. Ex. 70 is tavaborole).)
`
`Petitioner and its experts have also argued that PBAs are similar to
`
`Brehove’s dioxaborinanes. If that is true, then a POSA would also avoid that class
`
`of compounds based on the PBA data discussed above. That conclusion, in turn, is
`
`fatal to Ground 1. (See V.B.5., infra.)
`
`*
`
`*
`
`*
`
`Anacor overcame
`
`the perceived challenges facing boron-containing
`
`compounds and brought to market KERYDIN, an embodiment of the ’621 Patent’s
`
`claims. This met a long-felt need and succeeded where others had failed. Since its
`
`approval in 2014, KERYDIN has earned praise from doctors as an important new
`
`treatment option for onychomycosis. In contrast to this real-world evidence of
`
`nonobviousness, Petitioner’s attempt to invalidate the patent covering this
`
`treatment option is based on hindsight and ignores prior art that undermines key
`
`pillars of its obviousness arguments.
`
`9
`
`

`
`IPR2015-01776
`
`II. BACKGROUND OF THE INVENTION
`A. Treating onychomycosis requires a compound with selective
`toxicity
`
`Claims 4, 6, and 11 cover methods of treating onychomycosis, the FDA
`
`approved indication for KERYDIN.2 Onychomycosis is a fungal infection of the
`
`nail plate or nail bed, which can cause a wide range of physical and psychological
`
`effects, including nail pain, depression, and decreased quality of life. (Ex. 2035
`
`(Ghannoum) ¶ 24; Ex. 2037 (Maibach) ¶¶ 18-21.)
`
`Well over 90% of onychomycosis cases are caused by
`
`fungal
`
`microorganisms called dermatophytes. (Ex. 2035 (Ghannoum) ¶ 22.) No more
`
`than 5% of onychomycosis cases are caused by different organisms, yeasts. (Id. at
`
`¶ 29.) Dermatophytes and yeasts diverge at the taxonomic level of class, and differ
`
`in morphologies, appearance, growth rate, and biochemical characteristics. (Id. at
`
`¶¶ 37-38.) Another difference is that dermatophytes need keratin as a nutrient, a
`
`basic and prevalent component of nail tissue. (Ex. 2036 (Lane) ¶ 17.) To consume
`
`keratin, dermatophytes produce an enzyme which degrades it, while yeasts cannot
`
`degrade keratin. (Ex. 2035 (Ghannoum) ¶¶ 21, 39-40.)
`
`2 KERYDIN is “indicated for the topical treatment of onychomycosis of the
`
`toenails due
`
`to Trichophyton rubrum or Trichophyton mentagrophytes.”
`
`(Ex. 2001.)
`
`10
`
`

`
`IPR2015-01776
`
`In view of these differences, a POSA understood that a compound’s activity
`
`data generated against a yeast does not predict whether that compound would also
`
`be active against a dermatophyte, and vice versa. (Id. at ¶¶ 63-64.) Since over
`
`90% of onychomycosis cases are caused by dermatophytes, a POSA developing an
`
`onychomycosis treatment would naturally have been focused on data concerning
`
`dermatophytes. (Id. at ¶¶ 35, 64.)
`
`An important design goal for an antifungal is selective toxicity, which means
`
`killing the fungus without harming the host. (Id. at ¶¶ 42-43.) This goal is driven
`
`by the fact that fungal cells and animal cells are both eukaryotic. (Id. at ¶ 42.)
`
`This similarity means a high risk of causing toxic effects by harming the patient’s
`
`cells. (Id. at ¶¶ 42-43.) Targeting bacteria does not present this challenge because
`
`bacteria are prokaryotic cells, which are dissimilar to the surrounding host cells. In
`
`the context of antifungals, a compound that does not achieve selective toxicity is
`
`not considered safe and effective for use in animals. (Id. at ¶¶ 42-45.)
`
`B. A POSA would have been concerned about the toxicity of
`tavaborole, a boron-containing compound
`Each challenged claim is limited to methods of treatment using tavaborole.3
`
`Only one of Petitioner’s three asserted references, Austin, discloses tavaborole, and
`
`3 The Reider Declaration (Ex. 2034) ¶¶ 18-102 provides an overview of the
`
`chemistry concepts most relevant to this proceeding.
`
`11
`
`

`
`IPR2015-01776
`
`only in the context of “biocides” for industrial applications. The other asserted
`
`references, Brehove and Freeman, disclose different compounds that happen to
`
`contain a boron atom.
`
`To stitch together Austin, Brehove, and Freeman into purported obviousness
`
`combinations, Petitioner refers to each reference as disclosing “boron-based”
`
`compounds that would be expected to exhibit similar properties. (Pet. at 1-3, 9-10,
`
`17, 19-21, 29-34, 36-37, 43, 46-48, 57.) This characterization is misleadingly
`
`simplistic because, as both sides’ experts agree, there are many distinct classes of
`
`boron-containing compounds.4 And a POSA would not have expected two
`
`structurally different compounds to have similar functional properties. (Ex. 2034
`
`(Reider) ¶¶ 90-93, 172-75, 195-201.) Rather, small structural changes—even
`
`within a single class of compounds—can change the chemical, biological,
`
`pharmacokinetic, and pharmacodynamic properties of the compound in significant
`
`and unpredictable ways. (Id. at ¶ 90; Ex. 2033 at 143:25-144:8 & 149:9-11
`
`(Kahl).) Research by Petitioner’s expert Dr. Kahl provides one example of this
`
`phenomenon involving boron-containing compounds. His research with boron
`
`
`4 Ex. 1006 (Kahl) ¶ 30; Ex. 1008 (Murthy) ¶¶ 100-01; see also Ex. 2033 at 250:24-
`
`251:9 (Kahl) (admitting there are 10-20 classes of organic boron-containing
`
`compounds).
`
`12
`
`

`
`IPR2015-01776
`
`porphyrins shows that simply changing the position of a boron atom in a carborane
`
`isomer had substantial effects on binding affinity and consequent effectiveness
`
`(Ex. 2128 at 3427); and that changing a boron-containing closo-tetraphenyl
`
`porphyrin to a boron-containing nido-tetra porphyrin resulted in a compound that
`
`was “severely hepatoxic.” (See Ex. 2042 at 825.)
`
`At the same time, there was a well-founded consensus in 2005 that boron-
`
`containing compounds raised serious toxicity concerns. A 1975 study published
`
`by EPA concluded that external application of boron-containing compounds should
`
`be discouraged because of toxicity. (Ex. 2043 at I-91.) One group of researchers
`
`in 1998 referred to the “inherent toxicity of boron-containing compounds.” (Ex.
`
`2009 at 1541.) A 2004 reference studied a class of boron-containing compounds
`
`which had good activity but were “toxic due to the presence of boron atoms.” (Ex.
`
`2004 at 146.)5 Scientists up to 2005 had failed for decades in their efforts to
`
`
`5 See also Ex. 1003 at ¶¶ [0013]-[0014]; Ex. 2005 at 2109 (disclosing boron’s
`
`“inherent toxic potential”); Ex. 2026 at 2-3 (boric acid, first registered as a
`
`pesticide in 1948, has “moderate acute toxicity,” including “dermal toxicity”).
`
`13
`
`

`
`IPR2015-01776
`
`develop boron-containing pharmaceuticals, because of these toxicity issues.6 (Ex.
`
`2034 (Reider) ¶¶ 96-99.)
`
`The one pre-2005 FDA approval of a boron-containing compound validates
`
`these toxicity concerns. Bortezomib (VELCADE) causes serious adverse effects
`
`that are only acceptable because it is indicated for end-stage cancer. (Ex. 2011 at
`
`36.) As Dr. Kahl acknowledged, “[y]ou want an anticancer drug to be toxic.” (Ex.
`
`2033 at 199:19-200:1 (Kahl).) And as one researcher observed in 2004,
`
`“[b]ortezomib was poorly tolerated when administered daily, even at very low
`
`doses.” (Ex. 2012 at 3956-57; see also Ex. 2011 at 35.)
`
`Given this backdrop, a POSA would not have considered any boron-
`
`containing compound to be safe, at least in the absence of in vivo toxicity studies
`
`demonstrating safety. (Ex. 2034 (Reider) ¶¶ 68 & 96.) Dr. Kahl agreed that
`
`“ultimately toxicity must be evaluated in animals.” (Ex. 2033 at 462:9-17 (Kahl).)
`
`
`6 See, e.g., Ex. 2007 at 1525 (“adverse liver toxicity, believed to be related to the
`
`boron constituent” in a series of thrombin inhibitors); Ex. 2008 at 952 (disclosing
`
`toxicity of diazaborines); Ex. 2004 at 146 (disclosing that diazaborines “are toxic
`
`due to the presence of boron atoms”); Ex. 2010 at 107 (disclosing that boric acid is
`
`a “developmental toxicant”); Ex. 2017 at 720 (topical application of boric acid for
`
`as little as 14 days can be fatal).
`
`14
`
`

`
`IPR2015-01776
`
`Dr. Kahl nevertheless contends that “[b]oron-containing compounds are
`
`generally considered safe.” (Ex. 1006 (Kahl) ¶ 30.) This assertion—upon which
`
`the Board expressly relied in its Decision—is difficult to square with his own
`
`research, writings and testimony.7 For example, an article he co-authored in 1990
`
`about the use of boron-containing compounds for neutron capture therapy states
`
`that: “it should be noted that the majority of [boronated] compounds synthesized
`
`for possible use in NCT cannot be successfully administered because of overt
`
`toxicity.” (Ex. 2002 at 4864; see also Ex. 2042 at 826.)
`
`Petitioner and Dr. Kahl also misstate the findings of a 2001 article by
`
`Groziak (Ex. 1027). Petitioner fails to report Groziak’s observations that “[n]o
`
`pharmaceutical based on boron has yet made it to market” and that one of the
`
`reasons boron has not been used is because it often forms complexes that are
`
`“highly toxic to both bacteria and mammalian cells.” (Ex. 1027 at Abstract, 321.)
`
`Of all of the potential boron therapeutics that Groziak saw on the horizon—which
`
`did not include tavaborole—the only one to achieve FDA approval was bortezomib
`
`(VELCADE), which includes a boronic acid substituent, characterized by Groziak
`
`
`7 Decision at 10 n.5 (while relying on Dr. Kahl’s testimony for purposes of the
`
`institution decision, acknowledging Anacor’s “argument challenging Dr. Kahl’s
`
`credibility regarding the toxicity of boron-containing compounds”).
`
`15
`
`

`
`IPR2015-01776
`
`as not “unusually toxic.” (Id. at 322.) And even that assertion is undercut by the
`
`well-known severe toxicity of the drug. (Ex. 2034 (Reider) ¶¶ 101-02.)
`
`Petitioner’s asserted references themselves support the consensus recounted
`
`above. For example, Brehove states that “[t]he safety and non-toxicity of organo-
`
`boron compounds has been questioned.” (Ex. 1003 ¶ [0013].) Brehove then refers
`
`to the 2000 edition of Sax’s Dangerous Properties of Industrial Materials, which
`
`“reports that the ‘Safety Profile’ of ‘Boron Compounds’ is ‘very toxic’, and that
`
`‘Boron Compounds are therefore considered an industrial poi